The Larval and Reproductive Biology of the Giant Crab Pseudocarcinus gigas

Caleb Gardner

The Australian southern giant crab, Pseudocarcinus gigas (Lamarck, 1818), has traditionally been placed in the families Oziidae Dana, 1851 or Menippidae Ortmann, 1893 (superfamily Eriphioidea MacLeay, 1838). Previous genetic studies questioned the validity of this classification as the molecular markers placed it far from all eriphioid lineages. A morphological re-examination of P. gigas shows that it possesses a suite of unusual characters in the structure of the antennae, antennules, epistome, carapace, pereopods, thoracic sternum, and vulvae that show it is different from other Eriphioidea as well as all other extant heterotremes, confirming the genetic results. A new family, Pseudocarcinidae, and a new superfamily, Pseudocarcinoidea, both monotypic, are here established for P. gigas.

The complete larval development of the oziid crab Epixanthus frontalis (H. Milne Edwards, 1834) hatched from ovigerous specimens collected from Saso Island, southern Red Sea, was obtained under laboratory conditions. Four zoeae, one additional zoea and a megalopa were obtained and these are described and illustrated in detail for the first time. Larvae of this species can be differentiated from those of other oziid species based on a combination of characters such as the number of aesthetascs and setae of the antennule, and the coxal and basial setal numbers and patterns of the maxilla and maxillule.

The Larval and Reproductive Biology of the Giant Crab Pseudocarcinus gigas Caleb Gardner B. Sc. (hons), Grad. Dip. App. Sc. (Aqua.) Submitted in fulfilment of the requirements for the degree of Doctor of Philosophy University of Tasmania September 1998 1 This thesis contains no material which has been accepted for a degree or diploma by the University or any other institution, except by way of background information and duly acknowledged in the thesis, and to the best of the candidate's knowledge and belief, no material previously published or written by another person except where due acknowledgment is made in the text of the thesis. This thesis may be made available for loan and limited copying in accordance with the Copyright Act 1968. 2 Abstract This thesis documents research on two aspects of the biology of the giant crab Pseudocarcinus gigas: the development, behaviour, and rearing of the larvae; and the reproductive biology of both sexes. Larvae were reared from hatch to juvenile crabs. The larval development of 5 zoeal and one megalopal stages were described which permitted identification of P. gigas larvae from plankton samples. Samples from different depths were sorted to obtain information on vertical migration patterns, although few P. gigas larvae were collected. Vertical migration was further investigated in experiments which analysed the swimming response to gravity, light intensity, change in light intensity, light wavelength, change in pressure, current, temperature, and thermoclines. Response to temperature involved a feed-back mechanism that positioned larvae at temperatures optimal for growth, survival, and metamorphosis to megalopa (14-16°C). Light intensity and photoperiod had little effect on survival although larvae reached megalopa most rapidly with long photoperiod and high intensity and were smaller in continuous darkness treatments. Cannibalism of stage 1 and 2 zoeas was highest with long photoperiod and low intensity. Mycosis and epibiotic fouling of larvae necessitated trials with prophylactic treatments. Survival was highest with a broad spectrum antibiotic (oxytetracyline) while promising results were obtained with carbendazim and copper oxychloride. Suitable concentrations for indefinite baths were established by monitoring toxic effects as increased mortality, deformity, prolonged intermoult, or death during moulting. The male reproductive tract is typical of brachyurans with ovoid, enveloped, spermatophores stored in the mid vas deferens (MVD). Males pass through three morphological stages (of chela development) and individuals from all three stages had spermatophores in the MVD. Mating pairs were never observed but patterns of limb loss indicate that mating involves female-centred competition. Females appear to mate while soft-shelled with stored sperm remaining viable for at least four years. Broods are produced annually although females occasionally skip a reproductive season, which may be associated with moulting. Several broods may be produced between moults 3 although fecundity declines with successive broods. The hepatopancreas underwent little change in composition during gonadogenesis. Fecundity increased with female size, although not in a simple cubic (volumetric) relationship as larger females produced larger eggs. This increase in egg size was associated with a significant, albeit small, decrease in protein and carotenoid, and an increase in moisture, while lipid appeared unaffected. Protein was used preferentially to lipid during embryogenesis. Techniques for immobilising, humanely killing, and internal imaging of crabs were employed for research on reproduction and are described. 4 Acknowledgments Advice provided by my supervisors, Dr Greg Maguire and Dr Howel Williams, was invaluable throughout the project for improving experimental design and manuscripts. Assistance in obtaining crabs was provided by several skippers and processors including Theo Hairon (Galaxy Fishing Co.), Ian Coatsworth and Richard Pugh (St Helens Aquaculture), Justin Cohen (Tas Crays), Michael Hardy and Barry (Stanley Fish), and Sam Gregg (Stormboy). Additional crabs from Victoria were collected with the help of Andrew Levings and with support from the Victorian Department of Conservation and Natural Resources, especially from Ari Vlassopoulos. Description of the larval stages was possible through the guidance of Dr Rudolfo Quintana (Aquatas), the help of Dr K. Konishi who produced computer images of the first larval stage, and the loan of equipment from Ass. Prof. Alistair Richardson. Larval rearing was achieved with the advice and assistance of Debbie Gardner, Michel Bermudes, Polly Butler, Alan Beech, Cameron Johns, Michael Northam, Sid Saxby, Robert Browne, Danny Roden, Peter Farrell and Greg Commerford. Larval rearing containers were donated by Pacific World Packaging through Jim Foote. Plankton samples were made available by Mark Lewis and Barry Bruce (CSIRO) and assistance with sorting was provided by Erica Aheimer, Andrew Trotter and Josephine Walker. Samples of eggs, ovaries and hepatopancreas were counted and analysed with the help of Peter Machin, Justin Guest, Dean Thompson, Ben Verbeeten, Kris McKinley and Maggie Muttius. X-Ray imaging of female crabs was by Phillip Thomas and Michael Eland, while Martin Rush and Tim Bevilacqua of the Royal Hobart Hospital assisted with CT and MRI scanning. Many people assisted with improving drafts of individual chapters including Stewart Frusher, Bob Kennedy, Dr Don Fielder, Dr Paul Clark, Dr Barry Munday, Dr Barbara Nowak, Jenni Bruce, Dr Peter Beninger, and Prof. Nigel Forteath. Staff at the Taroona Marine Research Laboratories were a pleasure to work around and this project benefited from the positive atmosphere. 5 Lastly, the contribution of my partner Sha-sha Kwa was enormous. She helped with much of the research work, has improved the thesis by proof reading and formatting, tolerated a messy house etc. etc. I'm very grateful. Facilities and financial support was provided by the Aquaculture Department, University of Tasmania, the Australian Postgraduate Awards, and the Tasmanian Department of Primary Industry and Fisheries 6 Contents Abstract ................................................................................................................................. i Acknowledgments ............................................................................................................. iii ABSTRACT ............................................................................................................................................. 3 ACKNOWLEDGMENTS ............................................................................................................................ 5 CONTENTS.............................................................................................................................................. 7 PREFACE: THE HISTORY OF GIANT CRAB EXPLOITATION AND RESEARCH .............. 10 Taxonomy ....................................................................................................................................... 18 Distribution .................................................................................................................................... 18 History of harvest with emphasis on Tasmania.............................................................................. 20 Post-Colonisation History .............................................................................................................. 21 History of research ......................................................................................................................... 23 References ...................................................................................................................................... 28 GENERAL INTRODUCTION: LARVAL BIOLOGY OF THE GIANT CRAB PSEUDOCARCINUS GIGAS .............................................................................................................. 30 References ...................................................................................................................................... 24 LARVAL DEVELOPMENT OF THE GIANT CRAB PSEUDOCARCINUS GIGAS (LAMARCK, 1818)(DECAPODA: ERIPHIIDAE) REARED IN THE LABORATORY ...................................... 26 Abstract .......................................................................................................................................... 50 Introduction .................................................................................................................................... 50 Materials and methods ................................................................................................................... 50 Results ............................................................................................................................................ 51 Discussion ...................................................................................................................................... 68 References ...................................................................................................................................... 69 A SMALL SAMPLE OF GIANT CRAB PSEUDOCARCINUS GIGAS LARVAE COLLECTED FROM SOUTHERN TASMANIA ....................................................................................................... 71 References ...................................................................................................................................... 56 BEHAVIOURAL BASIS OF DEPTH REGULATION IN THE FIRST ZOEAL STAGE ............ 58 Abstract .......................................................................................................................................... 83 Introduction .................................................................................................................................... 83 Materials and methods ................................................................................................................... 84 Results ............................................................................................................................................ 91 Discussion .................................................................................................................................... 102 Conclusions .................................................................................................................................. 105 References .................................................................................................................................... 105 EFFECTS OF TEMPERATURE AND THERMOCLINES ON LARVAL BEHAVIOUR AND DEVELOPMENT ................................................................................................................................ 107 Abstract ........................................................................................................................................ 106 Introduction .................................................................................................................................. 106 Materials and methods ................................................................................................................. 107 Results .......................................................................................................................................... 110 Discussion .................................................................................................................................... 116 References .................................................................................................................................... 119 EFFECT OF PHOTOPERIOD AND LIGHT INTENSITY ON LARVAL SURVIVAL, DEVELOPMENT AND CANNIBALISM ......................................................................................... 123 Abstract ........................................................................................................................................ 124 7 Introduction .................................................................................................................................. 124 Materials and methods ................................................................................................................. 124 Results .......................................................................................................................................... 128 Discussion .................................................................................................................................... 132 References .................................................................................................................................... 135 USE OF PROPHYLACTIC TREATMENTS FOR LARVAL REARING................................... 138 Abstract ........................................................................................................................................ 142 Introduction .................................................................................................................................. 142 Materials and methods ................................................................................................................. 143 Results .......................................................................................................................................... 145 Detrimental effects ....................................................................................................................... 146 Enhanced survival ........................................................................................................................ 146 Discussion .................................................................................................................................... 153 Conclusion .................................................................................................................................... 154 References .................................................................................................................................... 155 GENERAL DISCUSSION: LARVAL BIOLOGY OF THE GIANT CRAB PSEUDOCARCINUS GIGAS ................................................................................................................................................... 156 References .................................................................................................................................... 161 GENERAL INTRODUCTION: REPRODUCTIVE BIOLOGY OF THE GIANT CRAB PSEUDOCARCINUS GIGAS ............................................................................................................ 164 References .................................................................................................................................... 164 OPTIONS FOR HUMANELY IMMOBILISING AND KILLING CRABS ................................. 165 Abstract ........................................................................................................................................ 175 Introduction .................................................................................................................................. 175 Materials and methods ................................................................................................................. 176 Results .......................................................................................................................................... 177 Discussion .................................................................................................................................... 180 Conclusions .................................................................................................................................. 182 References .................................................................................................................................... 183 NON-LETHAL IMAGING TECHNIQUES FOR CRAB SPERMATHECAE............................ 185 Abstract ........................................................................................................................................ 184 Introduction .................................................................................................................................. 184 Materials and methods ................................................................................................................. 184 Results .......................................................................................................................................... 185 Discussion .................................................................................................................................... 188 References .................................................................................................................................... 189 SPERMATOGENESIS AND THE REPRODUCTIVE TRACT OF THE MALE GIANT CRAB PSEUDOCARCINUS GIGAS ............................................................................................................ 191 Abstract ........................................................................................................................................ 197 Introduction .................................................................................................................................. 197 Materials and methods ................................................................................................................. 197 Results .......................................................................................................................................... 197 HISTOLOGY OF THE REPRODUCTIVE TRACT ................................................................................... 199 MATURATION OF MALE GIANT CRAB PSEUDOCARCINUS GIGAS AND THE POTENTIAL FOR SPERM LIMITATION IN THE TASMANIAN FISHERY .......................... 208 Abstract ........................................................................................................................................ 216 Introduction .................................................................................................................................. 216 Materials and methods ................................................................................................................. 217 Results .......................................................................................................................................... 219 Discussion .................................................................................................................................... 224 Summary ....................................................................................................................................... 226 References .................................................................................................................................... 226 OVARIAN DEVELOPMENT AND FEMALE REPRODUCTIVE BIOLOGY ........................... 229 8 Abstract ........................................................................................................................................ 240 Introduction .................................................................................................................................. 240 Materials and methods ................................................................................................................. 241 Results .......................................................................................................................................... 243 Discussion .................................................................................................................................... 249 References .................................................................................................................................... 253 EFFECT OF SIZE ON REPRODUCTIVE OUTPUT OF FEMALE GIANT CRABS ................ 257 Abstract ........................................................................................................................................ 262 Introduction .................................................................................................................................. 262 Materials and methods ................................................................................................................. 263 Results .......................................................................................................................................... 267 Discussion .................................................................................................................................... 273 References .................................................................................................................................... 276 COMPOSITION OF EGGS IN RELATION TO EMBRYONIC DEVELOPMENT AND FEMALE SIZE ...................................................................................................................................................... 278 Abstract ........................................................................................................................................ 277 Introduction .................................................................................................................................. 277 Materials and methods ................................................................................................................. 278 Results .......................................................................................................................................... 280 Discussion .................................................................................................................................... 283 References .................................................................................................................................... 285 GENERAL DISCUSSION: REPRODUCTIVE BIOLOGY OF THE GIANT CRAB PSEUDOCARCINUS GIGAS ............................................................................................................ 287 9 Preface: The History of Giant Crab Exploitation and Research 1 10 Taxonomy The current taxonomy of Pseudocarcinus gigas is: Section HETEROTREMATA Superfamily Xanthoidea MacLeay 1838 Family Eriphiidae1 MacLeay, 1838 Subfamily Oziinae Alcock 1898 G & Sp.Pseudocarcinus gigas (Lamarck, 1818). Pseudocarcinus gigas was originally described as Cancer gigas by Jean-Baptiste Lamarck in 1818 and later placed into Pseudocarcinus by Henri Milne Edwards after he created the genus in 1834. The family and subfamily status of the genus is unclear as Pseudocarcinus has not been included in recent reviews of the Xanthoidea by Guinot (1977, 1978, 1979) and Holthuis (1993) due to inadequate published descriptions. Critically, the male pleopods and sternum have not been described or illustrated and few specimens are held by international museums. Specimens and photographs were sent to xanthoid taxonomists Mr Peter Davie (Queensland Museum) and Dr Danièle Guinot (Museum National d’Histoire Naturelle, Paris) and their opinion was that Pseudocarcinus lies in the taxonomic cascade listed above. Further work on the taxonomic relationships of Pseudocarcinus is underway using sperm morphology (Prof. Barrie Jamieson, Queensland University and Dr Danièle Guinot) so the status should soon be resolved. The common name for Pseudocarcinus gigas was traditionally “giant crab” although the name “king crab” was widely adopted after 1980. Use of “king crab” seems to have resulted from confusion with the commercially important lithodid crabs which are generically termed “king crabs” (Dawson, 1989). In 1995, the Federal Department of Primary Industries and Energy published a report on marketing names for Australian seafood and it was concluded that “king crab” was misleading so “giant crab” was adopted as the official marketing name (MNFSA, 1995). Distribution In temperate regions of the world, the coastal waters of the continental shelf tend to be inhabited by at least one very large species of crab with large crab species seldom found in tropical waters. This trend may be due to a female defence polygynous mating system in open habitat, which favours larger, stronger males (Christy, 1987; Orensanz et al., 1995). The trend is worth noting as it may be of value in understanding the origins and niche of P. gigas. This position of large crab is filled by different taxa in various temperate regions and examples of genera are: Macrocheira, Chionoecetes (Majidae) and Erimacrus (Atecyclidae ) in the north-west Pacific; Chionoecetes (Majidae) and Paralithodes (Lithodidae; Anomura) in the north-eastern Pacific; Chionoecetes (Majidae) in the north-west Atlantic; Cancer (Cancridae) and Maja (Majidae) in the north-east Atlantic; Geryon (Geryonidae) around southern Africa; and Lithodes, Paralomis (Lithodidae, Anomura), and Geryon (Geryonidae) around southern America. While Pseudocarcinus gigas is the most abundant large crab in Australian temperate waters, other relatively large species are also found: Hypothalassia armata (Xanthoidea), Chaceon bicolor (Geryonidae), and deepwater lithodid species. 1 Eriphiidae is synonymous with Oziidae and Menippidae (Holthuis 1993). 18 Pseudocarcinus is an endemic, monospecific Australian genus belonging to the superfamily Xanthoidea — an extremely diverse taxon with approximately 166 species and 47 genera in Australia alone (Griffin and Yaldwyn, 1968). Xanthoids are found world-wide although they appear to be slightly more diverse through the Indo-West Pacific. Despite their wide distribution and diversification, relatively few xanthoids have attained large size with notable exceptions being three commercial genera Pseudocarcinus, Menippe (Americas), and Hypothalassia (north-west Pacific and southwest Australia). It is interesting that during the mid-Tertiary, the dominant large crab of the southern coastal shelf of nearby Tasmantis — the land mass bearing New Zealand and Lord Howe Island — was also a xanthoid. This species, Tumidocarcinus giganteus, is common in New Zealand middle and upper Miocene sediments laid down in deeper coastal waters (Fleming, 1962). As with P. gigas, the males had exceptionally large right chelae although overall body size appears to be somewhat smaller (Fleming, 1962; Fig. 1). This species may have been displaced by Cancer novaezelandiae (McLay, 1988), the only cancrid crab to reach Tasmantis or Meganesia (Australia and New Guinea), from South America via Antarctica in the mid-Tertiary (Nations, 1975). Figure 1. Frontal views of the extinct xanthoid crab, Tumidocarcinus giganteus, from New Zealand (left: after Fleming, 1962) and Pseudocarcinus gigas showing development of large right chelae. The present distribution of P. gigas stretches from south-west Western Australia, across southern Australia, to mid-New South Wales, with the most northerly record on the east coast from Five Islands off the coast of Wollongong NSW (McNeill, 1920; Fig. 2). The southern extent of their distribution appears to be the continental shelf around Tasmania and specimens have been obtained by rock lobster fishers from around this entire region. Although specimens have been collected off southern Tasmania, crabs are not common and commercial fishers seldom work below 42°30’ S. Bass Strait is popularly considered the major region inhabited by P. gigas (e.g. Griffin, 1970), possibly from repeated citation of early collections by Haswell (1882) and Rathbun (1926) in this area. However, most P. gigas are found at depths between 120 and 370 m (Levings et al., 1996) which effectively excludes Bass Strait as it is generally shallower than 100 m. 19 Figure 2. Distribution of Pseudocarcinus gigas (shaded area) and important Australian cities. History of harvest with emphasis on Tasmania Aboriginal history Although giant crabs are normally found in deep water out of reach of divers, large individuals are occasionally collected in shallow areas of less than 10 m (Hale, 1927). Francois Péron visited Maria Island on the east coast of Tasmania in 1802 as the zoologist for the Baudin expedition and his diary reports that giant crabs formed part of the diet of local Aborigines. Péron was one of the more enlightened explorers of his time, he had a great personal interest in human diversity and was one of the first people to coin the term “anthropology” (Plomley et al., 1990; Flannery, 1994). His diary record of the expedition’s visit to Bruni (now Bruny) Island shows great humanity (Wallace, 1984). As a consequence of his interest, Péron’s records of aboriginal lifestyle in Tasmania are particularly detailed and are in his words “minutely exact” or accurate; they provide one of the best records now available. They are also full of boundless enthusiasm. Péron describes his discovery of a giant crab cheliped on Maria Island and his reactions to the capture of these crabs by Aborigines thus: “Upon the beach at the head of the eastern bay, I, myself came upon the monstrous claw of a crab; the individual to which this redoubtable weapon had belonged must not have weighed less than 30 or 40 pounds. Moreover, these large species supply the natives of these regions with part of their diet. It is the women who dive to great depths for them, and I confess that I can scarcely imagine how they manage to pull from their rocky dwellings creatures so big and frightfully armed” (Plomley et al., 1990). 20 The “monstrous claw” collected by Péron was transported back to France and it is mentioned in the original description by Lamarck (1818). As Péron notes, it is difficult to imagine how the women were able to collect giant crabs. Maria Island is close to the edge of the continental shelf so giant crabs will be found closer to land than in most other areas of Tasmania and occasional animals may have wandered in close to land. These may have been washed ashore as with the specimen collected by Péron or they may have been actually collected by diving as Péron asserts. Oysters (Ostrea angasi), abalone (Haliotis rubra), and southern rock lobsters (Jasus edwardsii) were collected by diving so it is conceivable that giant crabs may have been encountered. Colonisation of Tasmania by Europeans occurred in 1802 and resulted in the near extermination of Aborigines so that by 1847 none were left in Tasmania save small populations on the Bass Strait Islands (Flannery, 1994). As a result, subsequent fishing for giant crabs was mainly by Europeans. Post-Colonisation History Harvest of rock lobsters appears to have commenced soon after colonisation and was clearly productive in inshore areas. In 1802, the Baudin expedition assigned a few crew members to catching rock lobsters with lines, a fairly ineffective method, yet they were able to catch enough within a few hours to feed the entire crew (Plomley et al., 1990). By 1884, William Saville-Kent, the colony of Tasmania’s superintendent and inspector of fisheries, had already noticed effects of fishing on the east coast with declines in size and abundance of rock lobsters (Saville-Kent, 1884). Giant crabs were occasionally collected as bycatch of this rock lobster fishery although a royal commission into the state of the fisheries in Tasmania concluded that though they were a splendid animal, they were only brought to market occasionally and were not of much commercial importance (Royal Commission Report, 1882). At around the same time in Victoria, McCoy (1889) reported that giant crabs, especially females, were occasionally brought to market and were especially common along the Victorian coast near Portland. This is the region where most of the Victorian catch is harvested today. William Saville-Kent appears to have been appointed following presentation of the Tasmanian royal commission report in 1882 and was keen to see the harvest of crustaceans diversified to provide greater variety of seafood to the Tasmanian public. He seemed to be exasperated by the state of Tasmanian technology and wrote: “The use of crab pots, as utilised in almost every other country on the face of the globe, might be advantageously recommended to the fishermen of Tasmania” (Saville-Kent, 1884). Saville-Kent was one of the visionaries of Australian marine science in many respects, notably in aquaculture where he: published methodology for rearing of the European lobster (Homarus gamarus; Saville-Kent, 1883); developed the Australian pearl industry in northern Western Australia; proposed oyster farming in Tasmanian Bays, Pipeclay Lagoon and Pittwater (which occurred almost 100 years later); and proposed the construction of a state hatchery for aquaculture trials with the giant Tasmanian crayfish (Astacopsis gouldii) and the striped trumpeter (Latris lineata) (which occurred around 100 years later). As with most of these other projects, Saville-Kent’s ideas on crab fishing did not develop further until the 1970’s when Tasmanian and Victorian crab fisheries were investigated in two projects: “Development of Small Scale Invertebrate Fisheries in Tasmanian Waters” (by the Tasmanian Fisheries Development Authority; Sumner and Dix, 1980); and “Experimental Trapping of the Giant Crab 21 Pseudocarcinus gigas” (by the Fisheries and Wildlife Division, Victoria; Winstanley, 1979). The Tasmanian project was headed by Colin Sumner and Trevor Dix and addressed fisheries for a variety of invertebrates including three crab species, the sand crab (Ovalipes australiensis; Portunidae), the spider crab (Leptomithrax gaimardii; Majidae), and giant crabs. The survey concluded that there was potential for development of a fishery for giant crabs, primarily as bycatch from the rock lobster (Jasus edwardsii) fishery, rather than as a targeted species. The development of a fishery was considered to rest on marketing as prices were generally too low to warrant fishers bothering with crabs (20c/kg, 1977/78). Fishing for giant crabs involves greater expense than for rock lobsters as giant crabs inhabit deeper, offshore areas; vessels need to be large to withstand high seas and gear is more expensive with larger pot haulers and more rope required. Crabs were processed by hand picking the meat and marketing trials were also made with whole cooked crabs. There appeared to be promise for the development of a fishery and further marketing trials were recommended (Sumner and Dix, 1980). Further market research was not undertaken and market prices did not improve for several years so most giant crab bycatch remained under utilised. Although most bycatch was taken by rock lobster fishers, trawlers also captured some giant crabs on muddy substrates and these were generally discarded. Crabs collected by rock lobster fishers were usually smashed so that they could be removed more easily from the wicker lobster pots and also because they were considered to interfere with the entry of rock lobsters (Sumner and Dix, 1980). Research conducted to improve live transport of rock lobsters in the late 1980’s allowed processors to gain higher prices for live rock lobster exports to Asia. The improved methods also enhanced survival of giant crabs so that they could be sold for far higher prices than was previously possible (VDCNR, 1995a). This opened the way for the development of a giant crab fishery as processors began to offer prices that were high enough to allow large vessels to profitably fish the deeper waters at the edge of the continental shelf (Yasuhara, 1995; Fig. 3). Tasmanian catches grew from 133 kg in 1990 to 243 tonnes in 1995 and Victorian catches also increased dramatically (Fig. 4; VDCNR, 1995b; TDPIF, 1995). As the market has become increasing aware of giant crabs, prices have steadily climbed so that beach price was over $50/kg for small crabs in 1997. This high price is based on the colour of crabs, not their unique size, as small crabs of less than 3 kg receive around double that of crabs greater than 5 kg (on a per kg basis). The giant crab fishery was initially subject to only two levels of input restriction, seasonal closures and pot limit — artefacts of the rock lobster fishery as similar gear was used for both species. The fishery grew at such a rapid rate that a minimum size restriction of 150 mm carapace length was introduced in 1994 across southern Australia which resulted in a decline in catch in Victoria (Fig. 4). This size limit was introduced as an interim measure until valid biological information could be collected and it is still in place. A more detailed discussion of changes in fishery management is documented elsewhere (Gardner, 1998). 22 Figure 3. Commercial fishing operation targeting giant crabs with purpose-built traps. Figure 4. Historical patterns of catch of giant crabs in Tasmania and Victoria since 1989 (data from the Victorian Department of Conservation and Natural Resources and the Tasmanian Department of Primary Industries and Fisheries). 250 Tasmania 200 Victoria Tonnes 150 100 50 0 1989 1990 1991 1992 1993 1994 1995 1996 1997 Year History of research Figure 3. Catch history of giant crabs in Tasmania and Victoria The oldest written record of giant crabs exists in Francois Péron’s accounts of his journey around Tasmania (1802), as discussed earlier. Péron was interested in the relationship between organisms and the environment and his ideas contributed to the formation of the science of ecology some 50 years later. He developed the idea of habitat preference in relation to giant crabs, and other crustaceans, by first discussing the terrain and then the interaction with crabs and lobsters2. 2 As an aside, it is also interesting to note his views on the earth’s age as the day of creation had been fixed in Péron’s era at 23 October 4004 BC by Archbishop Ussher. This view was beginning to be questioned in the early nineteenth century and was finally shattered by Charles Lyell’s “Principles of Geology, 1830-33”(Carey, 1995). 23 “... everything here [on Maria Island] bears the marks of the worlds upheavals, everything here attests to its great antiquity, everything recalls the painful struggle that it had to carry on against the fury of the waves, everything speaks of their ancient dominion over the land... the traces of their gradual recession are to be seen everywhere and in the shape and distribution of the rocks as well as their nature. There, ramparts of granite seem to present an insurmountable barrier to the ocean. Steep and sheer they rise to a height of two or three hundred feet. In their sides there are sometimes more or less large caves, in which the waters, as they surge tumultuously in produce dull boomings like the sound of distant thunder. ..it remains for me to say a few words about the genus Cancer, using the name in all its Linnean generality. Among the rocks that I have described and in the furthest depths of those caves which I have spoken, it is easy to conceive that the largest species of this genus [, the giant crab,] must not only multiply freely, but also reach a gigantic size.... Be that as it may, one further finds in connection with this genus of animals, a new and striking proof of the influence of nature of the seabed upon the existence of such and such a species in preference to all others. Lobsters, which seek out holes in rocks and their debris, exist in prodigious numbers around Maria Island, they were generally rare in D’Entrecasteaux Channel....On the other hand, spider crabs, which delight in filth and mud, abounded to excess on every point in the Channel and yet apparently did not exist around Maria Island. The different nature of the terrain must in fact repel them...”(Plomley et al., 1990). Following on from Péron’s report, several references to giant crabs have been made, generally of a taxonomic nature, although a small amount of biological information was also collected. The first taxonomic account was by Jean-Baptiste Lamarck, who is famous for his early contribution to evolution, published in “Zoological Philosophy” (1809). Although he invented the term “biology”, Lamarck’s original career plans had been with the army until a neck injury forced him to seek other directions (Elliot, 1914). At age 50, he had spent 25 years studying botany when three chairs were created at the Museum National d’Histoire Naturelle in Paris. The botany chair had been filled so Lamarck was appointed professor of zoology, insects, worms and microscopic animals. He practically abandoned botany and launched into the epic task of classifying the world’s invertebrate animals; the results of his research were then published in Histoire Naturelle des Animaux sans vertèbres, a seven volume work published from 1815-1822. The giant crab was included in the 1818 volume and it is the last crab that Lamarck described before moving onto another invertebrate group. His description is very brief and sounds almost tired: “Giant Crab (Cancer gigas) Inhabits the waters of New Holland, in Port Jackson. Péron and Lesueur [collectors]. The shell of an entire individual is 10 inches width; however according to a found anterior leg which is of human arm size, it can be of a huge size. The front of the shell bears four small teeth. Its posterior sides bear small sparse tubercles. The end articulation of the legs are slightly spiky. Etc. It is in the museum collection, which owns plenty of other species still undescribed”(Lamarck, 1818). 24 The holotype is male of around 27-28 cm carapace width and is still held at the French Museum National d’Histoire Naturelle although it is dry and some legs are detached (MNHN-B 13171; pers. comm., Danièle Guinot). Lamarck lists the collecting location of this holotype as Port Jackson, generally known as Sydney Harbour although the apparently old label on the holotype indicates “Tasmanie” as the collecting location. The location of Sydney is unlikely given the current knowledge of the distribution and appears to be an error, possibly introduced by Péron or Lesueur. The large anterior leg that Lamarck mentions is almost certainly that found by Péron on Maria Island, and the whole specimen may have been one collected by Aborigines from the same area. The large leg can no longer be located in the collection of the Museum National d’Histoire Naturelle. Thomas Whitelegge (1889) continued the probable error in the collecting location of the holotype by including the giant crab in a species list for Port Jackson and he lists the collecting location as Lane Cove River, a sheltered brackish estuary. Whitelegge notes that his task was difficult given the paucity of books in the colony and in many cases his quotations are second hand. This difficulty seems to have affected his inclusion of the giant crab, he cited Lamarck’s description incorrectly and included Lane Cove River as a collecting site when it was not previously listed. Taxonomic accounts of the giant crab included those by Milne Edwards (1834), Haswell (1882), McCoy (1889), Rathbun (1926), and Hale (1927-29). The description by McCoy is by far the most accurate and detailed and includes the earliest known illustrations of the species including the dissected mouth-parts and abdomens of both sexes (Fig. 5). McCoy completed his description as part of a massive two volume series on the natural history of Victoria, “A Prodromus of the Natural History of Victoria”. Only a few copies of this publication have survived so his description is reproduced in Appendix 1. Figure 5 (proceeding pages). Illustrations of female (first) and male (second) P. gigas by McCoy (1889) including dissected mouthparts. The plates were produced by lithography which reverses the image – the original crabs depicted both bore the larger molariform cheliped on the right hand side which is typical. The original figure caption was: EXPLANATION OF FIGURES. PLATE 179.-Fig. 1, female, about one third natural size. Fig la, abdomen of female, one third the natural size. Fig.1b, antennules, or inner antennae, movable portion without great fixed base, twice the natural size. Fig lc, antennae, or outer antennae, without the small basal joint, twice the natural size. Fig. lg, mandible and first and second maxillipedes, natural size. Fig. 1d, third or external maxillipede, natural size. Fig. 2, abdomen of male, one-third natural size. PLATE 180 -Fig. 1, male, about one-third natural size. (For abdomen, see pl. 179, f. 2). 25 26 Most of the descriptive texts also mention some biological information about giant crabs. Haswell (1882) states that the carapace is sometimes 2 feet in breadth; however, using regressions derived for male crabs in the submitted study, this would equate to the unlikely weight of around 60 kg. McNeill (1920) also discusses the size of giant crabs and he gives more accurate measures from a specimen at the Tasmanian Museum and 27 Art Gallery (still on display) of 330 mm carapace width and hand length of 438 mm. Several authors have noted the beautiful colouring which seems to vary for each animal in an individual pattern (McCoy, 1889; McNeill, 1920; Rathbun, 1926; Hale, 1927). McCoy (1889) and Rathbun (1926) observed increased dimorphism of the chelae with increased body size and that most animals were right handed. Rathbun’s (1926) work was based on collections made by the federal trawler, “The Endeavour”, from 19091914 and McNeill (1920) presented information recorded by a member of the staff aboard the vessel for these expeditions, Mr A.R. McCulloch. He noted that several juvenile specimens, as little as 1 inch carapace width, were found in sponge cavities on different occasions. This remains the only reliable information published on the habitat occupied by juvenile giant crabs. Very little research of any nature was conducted on giant crabs until the fisheries research by Sumner and Dix in the late 1970’s. Trevor Dix (1980) also suggested that giant crabs might be cultured although there was insufficient knowledge to assess their potential. It has not been until the 1990’s, after the development of the fishery, that the biology of the giant crab has begun to be studied in any detail; among the information collected is that reported in this thesis. References Carey, J. 1995. The Faber Book of Science. Faber and Faber, London. p. 71. Christy, J.H. 1987. Competitive mating, mate choice and mating associations of brachyuran crabs. Bull. Mar. Sci. 41:177-191. Dawson, E.W. 1989. King crabs of the world (Crustacea: Lithodidae) and their fisheries: a comprehensive bibliography. New Zealand Oceanographic Institute, Misc. Pub. 101. Dix, T. 1980. Aquaculture, what’s the future. Fintas, 2: 33-35. Elliot, H. 1914. Introduction to translation of [Lamarck, J.B., 1809. Zoological Philosophy], Macmillan, London.. Flannery, T. 1994. The Future Eaters. Reed Books, Sydney. 423 pp. Fleming, C.A. 1962. A Miocene crab-bed in Wairarapa District, New Zealand, and notes on allometry in Tumidocarcinus giganteus Glaessner. Trans. Royal. Soc. N.Z. (Geol.), 1(14): 207-213. Gardner, C. 1998. The Tasmanian giant crab fishery: a synopsis of biological and fisheries information. Tasmanian Department of Primary Industry and Fisheries, Internal Report 43, 40 pp. Griffin, D.J.G. 1970. Australian crabs. Aust. Nat. Hist., 16(9): 304-308. Griffin, D.J.G. and Yaldwyn, J.C. 1968. The constitution, distribution and relationships of the Australian decapod crustacea. Proc. Linn. Soc. N.S.W., 93: 164-183. Hale, H.M. 1927-29. The Crustaceans of South Australia. Government Printer, South Australia. 380 pp. Haswell, W.A. 1882. Catalogue of the Australian stalk and sessile-eyed crustacea. Sydney: Australian Museum, 326 pp, 4 pls. Holthuis, L.B. 1993. The non-Japanese new species established by Wide Haan in the Crustacea volume of Fauna Japonica (1833-1850). In: T. and K. Baba, 1993, Ph.F. von Siebold and Natural History in Japan. Crustacea. T. Yamaguchi (Ed.). Carcinological Society of Japan, Tokyo, 599-646. Lamarck, J.B. 1818. Histoire Naturelle des Animaux Sans Vertèbres. p 272 (in French). Levings, A., Mitchell, B.D., Heeren, T., Austin, C. and Matheson, J. 1996. Fisheries biology of the giant crab (Pseudocarcinus gigas, Brachyura, Oziidae) in southern Australia. High Latitude Crabs: Biology, Management, and Economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska Fairbanks, 125-151. McCoy, F. 1889. A Prodromus of the Natural History of Victoria (Zoology). p 293. McLay, C.L. 1988. Brachyura and crab-like Anomura of New Zealand. Leigh Lab. Bull., 22: i-iv, 1-463. McNeill, F.A. 1920. Studies in Australian carcinology, no. 1. Records of the Australian Museum, 13: 108-109. Milne Edwards, H. 1834. Histoire Naturelle des Crustacea. 1: 409 (in French). MNFSA 1995. Marketing names for fish and seafood in Australia. Fisheries Policy Branch, Department of Primary Industries and Energy, Canberra. 170 pp. 28 Nations, D. 1975. The genus Cancer (Crustacea: Brachyura): systematics, biogeography and fossil record. Nat. Hist. Mus. L.A. Science Bulletin, 23:1-104. Orensanz, J.M., Parma, A.M., Armstrong, D.A., Armstrong, J. and Wardrup, P. 1995. The breeding ecology of Cancer gracilis (Crustacea: Decapoda: Cancridae) and the mating systems of cancrid crabs. Zool. Lond., 235: 411-437. Plomley, B., Cornell, C. and Banks, M. 1990. Francois Péron’s natural history of Maria Island, Tasmania. Records of the Queen Victoria Museum, no. 99. 50 pp. Rathbun, M.J. 1926. Report on the crabs obtained by the F.I.S. “Endeavour” on the coasts of Queensland, New South Wales, Victoria, South Australia and Tasmania. Report on the Oxyrhyncha, Oxystomata, and Dromiacea. Biol. Res. Fish. Exp. Aust. 5(3): 93-156. Royal Commission Report. 1882. Fisheries of the Colony of Tasmania, presented to the governor, Sir George Strahan. Saville-Kent, W. 1883. The artificial culture of lobster. International Fisheries Expedition, London (unseen, cited by Saville-Kent, 1884). Saville-Kent, W. 1884. Report to the Governor Addressing Directions for the Fishing Industries of Tasmania. 6 pp. Sumner, C.E. and Dix, T.G. 1980. Development of small scale invertebrate fisheries in Tasmanian waters. Fishing Industry Research Trust Account, Final Report, 78/17. TDPIF (Tasmanian Department of Primary Industry and Fisheries). 1995. Interim access options for the king crab fishery. Fishing Today, 8(5): 20-21. VDCNR (Victorian Department of Conservation and Natural Resources). 1995a. Regulatory impact statement: Fisheries (king crab) regulations. Victorian Department of Conservation and Natural Resources. 11 pp. VDCNR. 1995b. Victorian Fisheries Catch and Effort Information Bulletin 1995. Victorian Department of Conservation and Natural Resources. Wallace, C. 1984. The Lost Australia of Francois Péron. Nottingham Court Press. Whitelegge, T. 1889. List of the marine and fresh-water invertebrate fauna of Port Jackson and the neighbourhood. Proc. Roy. Soc. N.S. Wales, 13: 227. Winstanley, R.H. 1979. Experimental trapping of the giant crab Pseudocarcinus gigas off western Victoria. Fisheries and Wildlife Paper No. 22, Fisheries and Wildlife Division, Victoria. 7 pp. Yasuhara, T. 1995. Notes on the Tasmanian giant crab, Pseudocarcinus gigas, (Xanthidae, Crustacea) in Tasmania. Cancer, 4: 27-29 (in Japanese). 29 General Introduction: Larval Biology of the Giant Crab Pseudocarcinus gigas 2 30 Research on the larval biology of the giant crab Pseudocarcinus gigas (Lamarck) was directed towards two aims: to provide biological information to assist research on recruitment processes; and to conduct a preliminary assessment of the potential for hatchery production of P. gigas. The rapid increase in the value of giant crabs from 1992 prompted high levels of fishing exploitation in some regions leading to concerns that crabs may be depleted from some areas, particularly in north-western Tasmania and western Victoria (see Chapter 16). The impact of depletion of crabs through fishing is especially severe where dispersal is limited, as localised recruitment failure may result. Although crabs may walk large distances as adults, dispersal is usually during planktonic larval stages (excepting among rafting species) making information on the larvae vital for assessing the potential for dispersal (Hitchcock, 1941; Kingsford and Choat, 1985; Havenhand, 1995). At the commencement of this project, commercial fishers reported that P. gigas larvae had direct development which is unusual, but does occur in other xanthoid crabs in the Australasian region (e.g. Pilumnus novaezelandiae and P. vestitus; McLay, 1988). This would imply that there was high potential for recruitment failure following localised depletion. However, shortly afterwards, an ovigerous female released free swimming zoeas3 in a tank which demonstrated that development was not direct (A. Levings, Deakin University, Pers. Comm.). Beyond this, nothing was known of the larvae and more information was required on the duration of the planktonic stage. Detailed analysis of dispersal of larvae also requires information on vertical migration behaviour. The larvae of P. gigas had not been described previously and this was undertaken to permit identification from plankton samples. Description of the larvae was also useful taxonomically as P. gigas has not been included in recent reviews of xanthoid crabs (see preface, Chapter 1). Plankton archives in Tasmanian marine research facilities were then searched for material collected in oceanic areas during late spring and summer. This material was sorted for P. gigas larvae to assess vertical distribution in natural conditions, and to find evidence of vertical migration stimuli such as time of day (i.e. light) and temperature (Chapter 4). Unfortunately, few larvae were obtained from field samples. The investigation of larval distribution in the field was complemented by laboratorybased experimental projects. Vertical migration of crustacean larvae is influenced by a range of stimuli including pressure, polarisation of light, gravity, absolute light intensity, salinity, change in light intensity, predator fields, prey fields, and temperature (Knight-Jones and Morgan, 1966; Umminger, 1969; Latz and Forward, 1977; Forward et al., 1984; Gliwicz and Pijanowska, 1988; and Forward, 1990). In a seminal paper, Sulkin (1984) considered that vertical migration of crab larvae results from a combination of orientating cues, principally gravity, combined with changes in upward swimming speed in response to various environmental cues, principally light. He also considered that change in pressure was important in regulating swimming speed, particularly in the absence of light (Sulkin, 1973). Consequently, this laboratory study focused on the response of P. gigas larvae to gravity, changes in light, and changes in pressure. Most research on the vertical migration behaviour of crab larvae has been with inshore, coastal species. Pseudocarcinus gigas are fished around the continental shelf so there is potential for the larvae to migrate through depths of over 300 m. It was considered that change in temperature may be an important migration cue in this 3 Zoea and megalopa are pluralised as zoeas and megalopas throughout this thesis as recommended by Martin (1984) and P. Clark (Natural History Museum, London, Pers. Comm., 1998). 23 environment so additional trials were conducted to assess the effect of temperature and thermoclines. Research on the potential for hatchery production was intended to provide preliminary information as it was recognised that a considerable research effort is required to achieve commercial production of a new species. Nonetheless, commercial production of giant crabs, or any other species, is highly dependent on hatchery techniques so this was a critical area for initial research. Giant crabs are highly valued with beach prices in 1997 fluctuating seasonally between $30/kg and $50/kg for small crabs under 3 kg. Markets would prefer crabs of around 100 mm carapace length which is smaller than the current minimum legal size of 150 mm carapace length, introduced in 1994. This market demand can only be filled by aquaculture and processors have indicated that prices would be higher than for legal sized crabs. At the commencement of this project, the only growth information available on giant crabs was from the moult of a juvenile giant crab captured in October 1993. This animal moulted from 70 mm to 94 mm carapace length which was an impressive 34% increase in length and indicated some potential for rapid growth (Frusher, 1994). Crab culture worldwide is small and has received relatively little research effort, with the exception of enhancement operations in Japan, so research on the culture of P. gigas larvae is of general interest for crab culture (Cowen, 1982). Much of the research on behaviour of P. gigas larvae overlapped with the second aim of assessing the potential for hatchery production of P. gigas. Unlike finfish, crab larvae are negatively buoyant and must swim to avoid sinking. Culture conditions that initiate upward swimming will exhaust energy reserves, while cues that initiate sinking may cause larvae to accumulate at the base of culture tanks, thus increasing contact with detritus. Research on optimising larval rearing investigated optimal lighting and temperature as these parameters affect swimming activity, intermoult duration, metabolism, rate of cannibalism, larval size, utilisation of energy reserves, feeding rate, and metamorphosis (Eagles et al., 1986; Waddy and Aiken, 1991; Hecht and Pienaar, 1993; Minagawa, 1994). The final aspect of research on assessing hatchery production of P. gigas was to control disease. In preliminary trials, larval mycosis caused high mortality so additional research was undertaken to refine the use of prophylactic treatments for this species. References Cowen, L. 1982. Crabs by the million in Japan’s ranching programme. Fish Farming International, 9: 16. Eagles, M.D., Aiken, D.E. and Waddy, S.L. 1986. Influence of light and food on larval American Lobsters, Homarus americanus. Canadian Journal of Fisheries and Aquatic Sciences, 43: 2303-2310. Forward, R.B.Jr. 1990. Behavioural responses of crustacean larvae to rates of temperature change. Biological Bulletin, 178: 195-204. Forward, R.B.Jr., Cronin, T.W. and Stearns, D.E. 1984. Control of diel migration: photoresponses of a larval crustacean. Limnology and Oceanography, 29: 146-154. Frusher, S. 1994. Giant crab moult: a promise of things to come ? Fishing Today 7: 2. Gliwicz, M.Z. and Pijanowska, J. 1988. Effect of predation and resource depth distribution on vertical migration of zooplankton. Bulletin of Marine Science, 43: 695-709. Havenhand, J.N. 1995. Evolutionary ecology of larval types. In: McEdward, L. (ed.), Ecology of Marine Invertebrate Larvae. CRC Press, New York, pp. 79-122. 24 Hecht, T. and Pienaar, A.G. 1993. A review of cannibalism and its implications in fish larviculture. Journal of the World Aquaculture Society, 24: 246-261. Hitchcock, H.B. 1941. The colouration and colour changes of the gulf weed crab, Planes minutus. Biological Bulletin, 80: 26-30. Kingsford, M.J. and Choat, J.H. 1985. The fauna associated with drift algae captured with a plankton-mesh purse seine net. Limnology and Oceanography, 30: 618-630. Knight-Jones, E.W. and Morgan, E. 1966. Responses of marine animals to changes in hydrostatic pressure. Oceanography and Marine Biology Annual Review, 4: 267-299. Latz, M.I. and Forward, R.B. Jr. 1977. The effect of salinity upon phototaxis and geotaxis in a larval crustacean. Biological Bulletin, 153: 163-179. McLay, C.L. 1988. Brachyura and crab-like Anomura of New Zealand. Leigh Laboratory Bulletin, 22: i-iv, 1463. Martin, J.W. 1984. Notes and bibliography on the larvae of xanthid crabs, with a key to the known xanthid zoeas of the western Atlantic and Gulf of Mexico. Bulletin of Marine Science, 34: 220-239. Minagawa, M. 1994. Effects of photoperiod on survival, feeding and development of larvae of the red frog crab, Ranina ranina. Aquaculture, 120: 105-114. Sulkin, S.D. 1973. Depth regulation of crab larvae in the absence of light. Journal of Experimental Marine Biology and Ecology, 13: 73-82. Sulkin, S.D. 1984. Behavioural basis of depth regulation in the larvae of brachyuran crabs. Marine Ecology Progress Series, 15: 181-205. Umminger, B.I. 1969. Polarotaxis in copepods, III. A light contrast reaction in Diaptomus Crustaceana, 16: 202-204. shoshone Forbes. Waddy, S.L. and Aiken, D.E. 1991. Scotophase regulation of the diel timing of the metamorphic moult in larval American lobsters, Homarus americanus. Journal of Shellfish Research, 10: 287. 25 Larval Development of the Giant Crab Pseudocarcinus gigas (Lamarck, 1818)(Decapoda: Eriphiidae) Reared in the Laboratory 3 Research for this chapter has been previously published as: Gardner, C. and Quintana, R., 1998. Larval development of the Australian giant crab Pseudocarcinus gigas (Lamarck, 1818)(Decapoda: Oziidae) reared in the laboratory. Journal of Plankton Research, 20(6): 1169-1188. 26 Abstract The larvae of Pseudocarcinus gigas, obtained from females collected from eastern Tasmania, were reared from hatching to metamorphosis. The larval series has five zoeal and one megalopal stages. This chapter presents a description of these stages and a comparison with other members of the sub-family Oziinae. Introduction Pseudocarcinus is a monospecific genus and has not been included in recent reviews of Xanthoidea (e.g. Guinot, 1978; Serene, 1984) although it appears to belong to the subfamily Oziinae within the family Eriphiidae 4 (P. Davie, Queensland Museum, personal communication). Larval information is available for several other species within the Oziinae including Ozius rugulosus (Kakati and Nayak, 1977); O. truncatus (Wear, 1968; Wear and Fielder, 1985); Baptozius vinosus (Saba et al., 1978a); Menippe adina (Martin, 1988; Martin et al., 1988); M. nodifrons (Scotto, 1979); M. rumphii (Kakati, 1977); M. mercenaria (Porter, 1960); and Epixanthus dentatus (Saba et al., 1978b). This chapter expands upon a preliminary description of the first stage zoea of P. gigas (Quintana et al., 1996; Appendix 2) to fully describe the zoeal stages and megalopa of P. gigas. The implications for taxonomic affinities are discussed. Materials and methods Ovigerous crabs were collected from depths in the range of 300 – 380 m off the east coast of Tasmania (41°17'S; 148°40'E) in June 1995. Females ranged in size from 2.2 – 3.5 kg and were held communally in 4 m3 tanks with flow through, unfiltered, seawater. Larvae were collected at dusk on the same day from two tanks to ensure that larvae were not from a single parent; further mixing probably occurred as several females were releasing larvae in each tank. Newly hatched larvae were rinsed in 0.2 µm filtered seawater (32‰ salinity) then transferred to 1.8 l, black, rectangular culture vessels. One hundred larvae were placed in each vessel and maintained in a temperature control room at 15.5°C. Cultures were not aerated (Appendix 3). Zoeal stages were fed a mix of Protein Selco™ (INVE aquaculture, Oeverstraat 7, Baasrode, Belgium) enriched rotifers (Brachionus plicatilis) and artemia5 nauplii for the first two instars and enriched artemia only thereafter. Ongrown artemia (7 d) were supplied to megalopa (Appendix 4). Larvae were transferred by pipette into cleaned sterile containers (2 l) with fresh, 0.2 µm filtered seawater daily (32-34‰ salinity). Larvae and exuvia from each stage were preserved for description in buffered 5% formalin. Voucher specimens have been deposited with the Tasmanian State Museum, the Laboratoire de Zoologie-Arthropodes, Muséum National d’Histoire naturelle and the Natural History Museum, London (registration number 1996.1195). 4 Eriphiidae is synonymous with Oziidae and Menippidae (Holthuis 1993). 5 Artemia is not italicized throughout this thesis as it has developed into a common name for a well known organism, as with eucalyptus, melaleuca or gorilla. 50 Measurements and descriptions were usually made from ten larvae for each stage. Measurements taken were: for the zoeal stages, (TT) the distance between tips of dorsal and rostral spines, (CW) between the tips of the lateral spines, and (CL) from the base of the rostral spine to the posterior margin of the carapace; for the megalopa, (CW) the maximum distance across the carapace, and (CL) the maximum distance along the carapace. Dissected appendages were first stained in Lees methylene blue before preparation as wet mounts. Drawings were made using a Wild M-5™ stereo microscope and an Nikon Optiphot-2™ compound microscope, both equipped with a camera lucida. Drawings of the first zoeal stage were then retraced and edited by computer using Adobe Illustrator™. Measurements were made by image analysis using NIH Image™ software. Setal counts are proximal to distal. Setae were classed as simple (S), sparsely plumose (SP), plumose (P), highly plumose (HP), or plumodenticulate (PD) (Greenwood and Fielder, 1984). Several other larval rearing trials were conducted to investigate the effects of temperature, light intensity, photoperiod, and prophylactic disease treatment. In some of these trials, larvae developed to an intermediate stage after zoea 5, particularly following prolonged exposure to high doses of oxytetracycline (Gardner and Northam, 1997; Chapter 8). None of these larvae developed further. This stage was considered abnormal so it is not included in the present description. Results Under rearing conditions used in this trial, mean time taken to reach megalopa was 54.0 d (SD=3.2; n=180) and mean time to complete larval development (crab 1) was 91.8 d (SD=3.93; n=26). The rate of development of Pseudocarcinus gigas larvae is discussed in more detail elsewhere (Gardner and Northam, 1997; Chapter 8). Prezoea (Figure 1A) Prezoea larvae were obtained from egg masses of females where hatching was in progress and were never collected with free swimming larvae taken from tanks immediately after hatching. As with prezoea larvae of other decapods (Quintana and Konishi, 1986), the prezoeal cuticle enveloped setae so that details of structure could not be seen. Carapace spines were depressed although the abdominal lateroventral spines were clearly apparent. Zoea I Carapace (Figs. 1B, 3A; Table 1): eyes immobile; all carapace spines well developed and prominent (frontal spine = 0.63 mm; dorsal spine = 1.36 mm; lateral spines = 0.43 mm); dorsal spine slightly curved backwards and pointed; lateral spines conspicuous and with slight ventral curvature; frontal spine smooth, distally pointed, and with slight anterior flexure; anterodorsal setae absent; 1 pair of posterodorsal setae; the ventral margins of the carapace without setae. Antennule (Fig. 3C): uniramous, endopod absent; exopod unsegmented with 3 aesthetascs and 2 simple (S) setae. Antenna (Fig. 3D): protopodal process less than half the length of the frontal spine and bearing around 25 spinules in rows along distal half; endopod absent; exopod rudimentary, unsegmented with 3 unequal setae (S). 51 Mandible (Fig. 3E): solid with molar and incisor processes well developed (clearly observed in well preserved exuviae); incisor process with small marginal teeth; endopod palp absent. Maxillule (Fig. 3F): coxa without setae; coxal endite with 7 plumodenticulate (PD) setae and 1 seta (S); basial endite with 5 setae (PD); endopod bisegmented, proximal segment with 1 sparsely plumose (SP) seta, distal segment with 2+2+2 setae (SP); exopod seta absent. Maxilla (Fig. 3G): coxal endite bilobed with 6+4 setae (PD); basial endite bilobed with 6+5 setae (PD); endopod bilobed with 3+5 setae (PD); exopod (scaphognathite) margin with 4 highly plumose (HP) setae and 1 long, plumose distal stout process. First maxilliped (Figs. 3H, 3H’): coxa without setae; basis with 2+3+3+3 setae (S); endopod 5-segmented with 2, 2,1,2,5-6 setae (PD except 1-2 simple setae on terminal); exopod 2-segmented, distal segment with 4 terminal setae (HP). Second maxilliped (Figs. 3I, 3I’): coxa without setae; basis with 4 setae (S); endopod 3segmented with 1,1,4 setae (PD except 2 simple setae on terminal); exopod 2-segmented, distal segment with 4 terminal setae (HP). Third maxilliped: absent. Pereiopods (Fig. 3J): uniramous and rudimentary. Abdomen (Figs. 3B, 3B’): five somites; somite 1 with mid-posterodorsal spine and no dorsomedial setae; somites 2 and 3 with anterolateral knobs, in addition, the later with sharp lateroventral spines as in somites 4 and 5; somites 2-5 with a posterodorsal prominence bearing 2 minute dorsal setae; somite 2 with a minute mid posteroventral seta; pleopod buds absent. Telson (Figs. 3B, 3B’): broad, bifurcated, curved backwards; each furca distally pointed, with a well developed, smooth lateral spine with a small spine on the inner margin; inner margin of furcae with 3+3 biplumose setae lateral to acute median notch; small dorsomedial spine present on each telson furca, arising well posterior to setae bases. Zoea II Carapace (Figs. 1C, 2A; Table 1): eyes now mobile; carapace larger, with two pairs of anterodorsal setae and each ventral margin with 3 setae (1 plumose (P) anterior seta and 2 sparsely plumose (SP) posterior setae), posterodorsal setation unchanged; dorsal spine more robust. Antennule (Fig. 4A): exopod with 4 aesthetascs, otherwise unchanged. Antenna (Fig. 4F): unchanged. Mandible (Fig. 5A): unchanged. Maxillule (Fig. 6A): coxa now with 1 seta (HP); coxal endite with 9-10 setae (PD); basial endite with 7-8 setae (PD); otherwise unchanged. Maxilla (Fig. 7A): coxal endite with 7+4 setae (PD); basial endite with 7+6 setae (PD); endopod unchanged; exopod (scaphognathite) margin with 17-18 setae (HP) and rounded distally. First maxilliped (8A): setae on basis unchanged in number but now plumodenticulate; terminal segment of endopod with 5 setae (PD); exopod distal segment with 6 terminal setae (HP); otherwise unchanged. Second maxilliped (8F): setae on basis unchanged in number but now plumodenticulate; exopod distal segment with 6 terminal setae (HP); otherwise unchanged. Third maxilliped: absent. Pereiopods (Fig. 9A): buds more developed and cheliped now chelate. 52 Abdomen (Figs. 10A, 10B): first abdominal somite with 2 dorsomedial setae; lateroventral spines of abdominal somites 3-5 of similar length, overlapping next somite by approximately three-quarter length; inner margin of furcae with additional biplumose setae lateral to acute median notch giving formula of 4+4; otherwise unchanged. Zoea III Carapace (Figs. 1D, 2B; Table 1): larger than in previous stages with ventral margin developed into two notches; 6 pairs of anterodorsal setae; each ventral margin with 9 setae (1 plumose anterior seta and 8 sparsely plumose posterior setae); otherwise unchanged. Antennule (Fig. 4B): endopod with additional aesthetasc; otherwise unchanged. Antenna (Fig. 4G): endopod bud present; otherwise unchanged. Mandible (Fig. 5B): unchanged. Maxillule (Fig. 6B): coxa with 2-3 setae (HP); coxal endite with 11-12 setae (PD); basial endite with 11-13 setae (PD); endopod unchanged. Maxilla (Fig. 7B): coxal endite with 8-10+5 setae (PD); basial endite with 8+7 setae (PD); endopod unchanged; exopod (scaphognathite) margin with 28-31 setae (HP). First maxilliped (Fig. 8B): basis unchanged; terminal segment of endopod with 6 setae (PD); exopod distal segment with 10 setae (HP); otherwise unchanged. Second maxilliped (Fig. 8G): number of setae on basis unchanged but now plumodenticulate; endopod unchanged; exopod distal segment with 10 setae (HP). Third maxilliped (Fig. 9B): present, rudimentary and biramous with endopod slightly longer than exopod. Pereiopods (Fig. 9B): developing with differentiation of segments, segments without setae. Abdomen (Fig. 10C): now 6-segmented; posterodorsal surface of somite 1 with 6 setae (S); other somites as previously but with ventral swelling in presumptive pleopod region; inner margin of furcae with additional biplumose setae lateral to acute median notch giving formula of 5+5; otherwise unchanged. Zoea IV Carapace (Figs. 1E, 2C; Table 1): ridge along the posterior margin is well defined; carapace now with 8 pairs of anterodorsal setae and each ventral margin with 15 setae (1 plumose anterior seta and 14 sparsely plumose posterior setae). Antennule (Fig. 4C): endopod bud absent; exopod now with 6 aesthetascs and 1 setae (S) terminally and with 1 aesthetasc and 1 setae (S) sub-terminally. Antennae (Fig. 4H): endopod bud on antenna approximately one-half the length of exopod; otherwise unchanged. Mandible (Fig. 5C): now with mandibular palp bud. Maxillule (Fig. 6C): coxa unchanged; coxal endite with 12-15 setae (PD); basial endite with 17-20 setae (PD); endopod unchanged. Maxilla (Fig. 7C): coxal endite with 10-11+6 setae (PD); basial endite with 9-10+8-10 setae (PD); endopod unchanged; exopod (scaphognathite) margin with 39-43 setae (HP). First maxilliped (Fig. 8C): basis now with 1+3+3+4 setae (PD); endopod unchanged; exopod distal segment with 14 setae (HP). Second maxilliped (Fig. 8H): basis and endopod unchanged; exopod distal segment with 14 setae (HP). Third maxilliped (Fig. 9C): further developed; now with epipod. 53 Pereiopods (Fig. 9C): further enlarged and differentiated into segments; no setae present. Abdomen (Fig. 10D): mid-posterodorsal spine on abdominal somite 1 reduced; pleopod buds on abdominal somites 2-6 uniramous with endopod absent; second lateral spine and dorsomedial spine on telson reduced; inner margin of furcae with additional biplumose setae lateral to acute median notch giving formula of 6+6; otherwise unchanged. Zoea V Carapace (Fig. 1F, 2D; Table 1): now with 14 pairs of anterodorsal setae and each ventral margin with 24 setae (1 plumose anterior seta and 23 sparsely plumose posterior setae); otherwise unchanged. Antennule (Fig. 4D): endopod bud present; exopod now with 11-13 aesthetascs and 2 setae (S) terminally, and 2 subterminal aesthetascs. Antenna (Fig. 4I): endopod bud partially segmented and approximately as long as exopod; otherwise unchanged. Mandible (Fig. 5D): endopod bud (palp) more developed but unsegmented and unarmed. Maxillule (Fig. 6D): coxa unchanged; coxal endite with 21 setae (PD); basial endite with 23-25 setae (PD); endopod unchanged. Maxilla (Fig. 7D): coxal endite with 13-14+8-9 setae (PD); basial endite with 13-15+11 setae (PD); endopod unchanged; exopod (scaphognathite) margin with 51-56 setae (HP). First maxilliped (Fig. 8D): exopod distal segment with 15 long terminal setae (HP); otherwise unchanged. Second maxilliped (Fig. 8I): number of setae on terminal segment of endopod unchanged but all now plumodenticulate; exopod distal segment with 17 long terminal setae (HP); otherwise unchanged. Third maxilliped (Fig. 9D): further developed with partial segmentation; otherwise unchanged. Pereiopods (Fig. 9D): further enlarged with partial segmentation. Abdomen (Fig. 10E): somite 1 now with mid-posterodorsal spine further reduced and 10 posterodorsal setae; pleopod buds on abdominal somites 2-6 further developed and elongate, now biramous with endopods present except on pleopod 5 (somite 6); the dorsomedial spine on the telson is further reduced or absent; otherwise unchanged. Megalopa Carapace (Fig. 11A): dimensions, CW- 2.92±0.18 mm, CL- 3.57±0.22 mm; carapace roughly quadrangular, broadest immediately below orbits and narrowing posteriorly; surface has a covering of setules, denser towards margins as illustrated; intra-orbital plate simple, over half width of carapace, lateral margins almost parallel, slightly bilobed with medial depression, and directed obliquely downward. Antennule (Fig. 4E): peduncle 3-segmented with 3, 5, 0 setae (S) respectively; endopod 2-segmented with 2,6 setae (S) respectively; exopod 4-segmented with 0, 5+6, 5+4, 4+3 subterminal aesthetascs respectively, segment 4 with 3 setae (S). Antenna (Fig. 4J): peduncle 3-segmented with 7 (2 S, 5 PD), 6 (PD), 6 (S) setae respectively; flagellum 8-segmented with 0, 2, 4, 1-2, 5, 2, 4, 5 setae (S) respectively. Mandible (Fig. 11B): endopod palp 3-segmented although divisions between segments 1 and 2 sometimes unclear, terminal segment with 19-23 marginal setae (PD). 54 Maxillule (Fig. 6E): coxa with 6 setae (3 P, 3PD); coxal endite with 33-39 setae (PD); basial endite with 35-37 setae (PD); endopod now unsegmented with 2+2+2+2 setae (PD). Maxilla (Fig. 7E): coxal endite bilobed with 23-26+11-13 setae (PD); basial endite bilobed with 17-21+16-18 setae (PD); reduced endopod which is no longer bilobate, with 8-14 setae (7-11 HP, 1-3 PD); exopod (scaphognathite) margin with 80-88 setae (HP) and 3 lateral setae (P) on each side. First maxilliped (Fig. 8E): well developed triangular epipod with 20-26 long setae (S); coxal endite with 22-27 setae (PD); basial endite with 53-61 setae (PD); endopod unsegmented with 5-7 spines, 3-6 subterminal setae (PD) and 2-5 terminal setae (PD); exopod 2-segmented with 5-8 (P) and 6-8 (HP) setae respectively. Second maxilliped (Fig. 8J): epipod bilobate with 21-22 long setae (HP); coxa with 2 setae (PD); basis with 6 setae (PD); endopod 5-segmented with 4, 5-8, 6, 12-15, 10-12 setae (PD) respectively; exopod 2-segmented, first segment with 4-7 spines and 1 setae (S), terminal segment with 8 setae (HP). Third maxilliped (Fig. 11C): epipod with 39-53 long setae (P) and arthrobranch gill; coxa and basis not differentiated with 16-28 setae (8-16 S, 8-12 PD); endopod 5segmented with 57-63, >40, >30, >30, 17-21 setae (PD) respectively; exopod 2segmented with 6-8 (S), 7-8 (1S, 6-7 HP) setae respectively. Pereiopods (Figs. 11D-F): cheliped with 2 curved spines on ischium, one more prominent; pereiopods 2-5 thin and setose; dactylus of pereiopods 2-4 with a strong, inwardly flexed terminal spine and with 9-10, 8-9, and 7-8 spines on inner margin respectively; strong spine on distal inner margin of pereiopods 2-4, strongest in pereiopod 2; pereiopod 5 with no spines along inner margin of dactylus, 4-6 small terminal spines, and 3-4 subterminal long setae. Sternal plates: plates anterior of 2nd pereiopods (plates 1-3) fused with row of 8 setae along anterior third; behind this row of setae is a minute medial spine flanked by 8-10 pairs of setae. Remaining sternal plates unarmed. Abdomen (Figs. 10F-I): 6 somites present plus telson with dorsal setation as figured; exopods of pleopods 1-4 with respectively 32-34, 33-35, 29-31, and 26-30 natatory setae; endopods with 6-7, 5, 5, 4-5 coupling hooks on the inner margin; uropods without endopod, with 19-20 natatory setae on distal and 1 seta on proximal segments; telson rounded, posterior margin with 5 setae. 55 Table I. Dimensions of zoeas of Pseudocarcinus gigas (in mm; mean of 10 individuals per zoeal stage, standard deviation in brackets). Zoeal Stage Feature 1 2 3 4 5 Dorsal to rostral spine (TT) 2.59 (0.10) 2.99 (0.23) 3.61(0.21) 4.44 (0.21) 5.56 (0.33) Lateral spine range (CW) 1.65 (0.13) 1.72 (0.06) 2.11 (0.09) 2.50 (0.15) 3.32 (0.16) Base of rostral spine to 0.93 (0.07) posterior margin of carapace (CL) 1.29 (0.07) 1.57 (0.06) 2.05 (0.11) 2.54 (0.12) Ratio TT/CW 1.6 1.7 1.7 1.8 1.7 Ratio CW/CL 1.8 1.3 1.3 1.2 1.3 56 Figure 1. Pseudocarcinus gigas: A, prezoea, lateral view; B, first zoea; C, second zoea; D, third zoea; E, fourth zoea; F, fifth zoea. 57 Figure 2. Pseudocarcinus gigas: Carapace ventral margin: A, second zoea; B, third zoea; C, fourth zoea; D, fifth zoea. 58 Figure 3. Pseudocarcinus gigas: First zoeal stage: A, frontal view; B and B’, abdomen, dorsal view and lateral view; C, antennule; D, antenna; E, mandible; F, maxillule; G, maxilla; H and H’, first maxilliped and endopod detail; I and I’, second maxilliped and endopod detail; J, pereiopod buds. Scale bars are 0.5 mm for A and B; and 0.1 mm for C to J. 59 Figure 4. Pseudocarcinus gigas: Antennule: A, second zoea; B, third zoea; C, fourth zoea; D, fifth zoea; E, megalopa. Antenna: F, second zoea; G, third zoea; H, fourth zoea; I, fifth zoea; J, megalopa. 60 Figure 5. Pseudocarcinus gigas: Mandible: A, B, C, D, second to fifth zoea. Posterior view (left), frontal view (right). 61 Figure 6. Pseudocarcinus gigas: Maxillule: A, B, C, D, second to fifth zoea; E, megalopa. 62 Figure 7. Pseudocarcinus gigas: Maxilla: A, B, C, D, second to fifth zoea; E, megalopa. 63 Figure 8. Pseudocarcinus gigas: First maxilliped: A, B, C, D, second to fifth zoea; E, megalopa. Second maxilliped: F, G, H, I, second to fifth zoea; J, megalopa. 64 Figure 9. Pseudocarcinus gigas: Rudimentary third maxilliped and pereiopods: A, B, C, D, second to fifth zoea. The third maxilliped is absent at the second zoeal stage. 65 Figure 10. Pseudocarcinus gigas: Abdomen: A, lateral view, second zoea; B, C, D, E, dorsal view, second to fifth zoea; F, megalopa; G, pleopod 1; H, endopod of pleopod one enlarged; I, telson and uropods. 66 Figure 11. Pseudocarcinus gigas: Megalopa: A, dorsal view; B, mandible; C, third maxilliped; D, cheliped; E, fourth pereiopod; F, fifth pereiopod. 67 Discussion Phylogenetic significance of larval characters Several authors have attempted to explain phylogenetic relationships of xanthoid crabs based on larval characters. Wear (1970) considered that the most important single zoeal character for distinguishing major groups is the length of the antennal exopod in relation to that of the protopodal process. Rice (1980) divided xanthoid zoea larvae into four groups using several characters with emphasis on the antennal exopod. This scheme was expanded further by Martin (1984) to cover two additional groups although the original groups proposed by Rice (1980) remained unaltered. Pseudocarcinus gigas falls into group III of these classifications as the zoea larvae have a robust antennal exopod with three unequal terminal setae. In addition, the other characters used by Rice (1980) and Martin (1984) to define group III are also present although there are five zoeal stages, a character more typical of group IV. The zoeal groups proposed by Rice (1980) and Martin (1984) correspond only loosely with the classification of Guinot (1978) which was based on adult characters. Zoeal group III includes genera from Eriphiidae, Trapeziidae, and Platyxanthidae while the genera Menippe and Sphaerozius are in zoeal group IV although these also lie within Guinot’s Eriphiidae (termed Menippidae by Guinot, 1978; see Holthuis, 1993). Although the zoeal groupings are broad, they do not challenge the placement of P. gigas within the Eriphiidae. Some of the larval characters of P. gigas indicate affiliation with the genus Ozius which is also present in southern Australia. The first abdominal somites of P. gigas zoea larvae bear a single dorsal spine which is an unusual larval character and has been suggested as a generic feature of Ozius based on O. truncatus and O. rugulosus rugulosus (Wear, 1968; Kakati and Nayak, 1977; Wear and Fielder, 1985). The cheliped ischium of the P. gigas megalopa bears a prominent recurved spine which is also present on the two described Ozius species, but not for any other member of the Oziinae with described larvae (listed in Introduction). Martin (1988) presented a detailed analysis of phylogenetic relationships of xanthoid crabs based on 16 megalopal characters; when megalopas of Ozius spp. and P. gigas are compared with this more extensive list of criteria, any relationship appears to be less clear (see Appendix 5). Using Martin’s (1988) criteria for megalopas, P. gigas appears to be more closely affiliated with the genus Menippe which also belongs to Oziinae yet falls into group IV of Martin’s (1984) grouping of xanthoid zoea larvae. This apparent affiliation of megalopas is largely due to higher setal counts on appendages. Setal counts on mouthpart segments tend to increase with each zoeal stage so it is not surprising that P. gigas and Menippe spp., each with five zoeal stages, tend to have more setae at megalopa than other members of Oziinae which only have four zoeal stages. Number of zoeal stages Most xanthoid crabs have four zoeal stages although this number is variable, even within genera (Wear, 1970); for example, Pilumnus lumpinus has only a single non-planktonic larval stage (Wear and Fielder, 1985) while most other Pilumnus species have 4 zoeal stages. Martin (1984) discussed trends in the number of larval instars and noted that abbreviated development has been attributed to development within restricted estuarine habitats (Rice, 1980), although he considered this theory untenable given that numerous exceptions exist. Advanced development (greater than four zoeal stages in xanthoids) 68 was discussed by Scotto (1979) who attributed the prolonged development of Menippe species to a retained primitive feature, being similar to cancrids. The explanation of Scotto (1979) implies that retention of 5 larval stages is a primitive trait and that Menippe is a primitive genus. However, other authors have considered the genus derived (Rice, 1980). The example already listed of variation in the number of larval stages in Pilumnus implies that number of zoeal stages is relatively plastic and may be a poor indicator of phylogeny. Dispersal will be influenced by the number of larval stages although Havenhand (1995) noted that dispersal of adults by rafting or drifting may be more widespread, and more important, than is generally appreciated. Clearly, chances of dispersal of adults by drifting is greatest in small species. The advanced larval duration of Menippe and Pseudocarcinus may simply be a response to the limited potential for dispersal of the adults as these are among the physically largest genera of the Xanthoidea. References Gardner, C. and Northam, M. 1997. Use of prophylactic treatments for larval rearing of giant crabs Pseudocarcinus gigas (Lamarck). Aquaculture, 158, 203-214. Greenwood, J.G. and Fielder, D.R. 1984. The complete larval development, under laboratory conditions, of Heteropanope glabra Stimpson 1858 (Brachyura, Xanthidae), from Australia. Aust. Zool., 21, 291-303. Guinot, D. 1978. Principes d’une classification évolutive des crustacés décapodes. Bull. biol. Fr. Belg., 112, 211-292. Hale, H.M. 1927-29. The Crustaceans of South Australia. Government Printer, South Australia. Havenhand, J.N. 1995. Evolutionary ecology of larval types. In McEdward, L. (ed.), Ecology of Marine Invertebrate Larvae. CRC Press, New York, pp. 79-122. Holthuis, L.B. 1993. The non-Japanese new species established by Wide Haan in the Crustacea volume of Fauna Japonica (1833-1850). In Baba, T. and Baba, K. (eds), Ph.F. von Siebold and Natural History in Japan, Crustacea. Carcinological Society of Japan, Tokyo, pp. 599-646. Kakati, V.S. 1977. Larval development of the crab Menippe rumphii (Fabricius) as observed in the Laboratory. Proc. Symp. Warm Water Plankton, Spec. Publ. Nat. Inst. Oceanogr., 60a, 634-641. Kakati, V.S. and Nayak, V.N. 1977. Larval development of the xanthid crab, Ozius rugulosus rugulosus Stimpson (Decapoda, Brachyura) under laboratory conditions. Indian J. Mar. Sci., 6, 26-30. Levings, A., Mitchell, B.D., Heeren, T., Austin, C. and Matheson, J. 1996. Fisheries biology of the giant crab (Pseudocarcinus gigas, Brachyura, Oziidae) in southern Australia. High Latitude Crabs: Biology, Management, and Economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska, Fairbanks, pp. 125-151. Martin, J.W. 1984. Notes and bibliography on the larvae of xanthid crabs, with a key to the known xanthid zoeas of the western Atlantic and Gulf of Mexico. Bull. Mar. Sci., 34, 220-239. Martin, J.W. 1988. Phylogenetic significance of the brachyuran megalopa: evidence from the Xanthidae. Symp. Zool. Soc. Lond., 59, 69-102. Martin, J.W., Truesdale, F.M., and Felder, D.L. 1988. The megalopa stage of the gulf stone crab, Menippe adina Williams and Felder, 1986, with a comparison of megalopae in the genus Menippe. Fish. Bull., 86, 289-297. Porter, H.J. 1960. Zoeal stages of the stone crab, Menippe mercenaria Say, Chesapeake Sci., 1, 168-177. Quintana, R. and Konishi, K. 1986. On the prezoeal stage: observations on three Pagurus species (Decapoda, Anomura). J. Nat. Hist., 20, 837-844. Quintana, R., Gardner, N.C. and Konishi, K. 1996. On the larvae of the giant crab, Pseudocarcinus gigas (Lamarck): a preliminary report. Abstracts for the meeting of the Jap. Soc. Fish. Sci., Tokyo, April, 1996. pp. 93. Rice, A.L. 1980. Crab zoeal morphology and its bearing on the classification of the Brachyura. Trans. Zool. Soc. London, 35, 271-424. Saba, M., Takeda, M., and Nakasone, Y. 1978a. Larval development of Baptozius vinosus (H. Milne Edwards). Proc. Jap. Soc. Syst. Zool., 14, 25-38. Saba, M., Takeda, M., and Nakasone, Y. 1978b. Larval development of Epixanthus dentatus (White)(Brachyura, Xanthidae). Bull. Nat. Sci. Mus. Tokyo (Zool.), 4, 151-161. 69 Scotto, L.E. 1979. Larval development of the Cuban stone crab, Menippe nodifrons (Brachyura, Xanthidae) under laboratory conditions with notes on the status of the family Menippidae. Fish. Bull. U.S., 77, 359-386. Serene, R. 1984. Crustacés décapodes brachyoures de l’Océan Indien Occidental et de la Mer Rouge. Xanthoidea: Xanthidae et Trapeziidae, avec addendum: Carpiliidae et Menippidae, par A. Crosnier. Faune trop., 24, 1-349. Wear, R.G. 1968. Life history studies on New Zealand Brachyura. 2. Family Xanthidae. Larvae of Heterozius rotundifrons A. Milne Edwards, 1867, Ozius truncatus H. Milne Edwards, 1834, and Heteropanope (Pilumnopeus) serratifrons (Kinahan, 1856). N.Z. J. Mar. Freshwat. Res., 2, 293-332. Wear, R.G., 1970. Notes and bibliography on the larvae of xanthid crabs. Pac. Sci., 24, 84-89. Wear, R.G. and Fielder, D.R. 1985. The marine fauna of New Zealand: larvae of the Brachyura (Crustacea, Decapoda). N. Z. Oceanogr. Inst. Mem., Wellington, 92. 70 A Small Sample of Giant Crab Pseudocarcinus gigas Larvae Collected from Southern Tasmania 4 Research for this chapter has been previously published as: Gardner, C. First record of larvae of the giant crab Pseudocarcinus gigas in the plankton. Papers and Proceedings of the Royal Society of Tasmania, 132: 47-48. 71 An important aspect of the biology of giant crabs Pseudocarcinus gigas for management is larval development and its influence on dispersal. Although several laboratory studies conducted on the larvae of the giant crab are described in this thesis, no larvae or recently settled juveniles have been collected from the wild previously. This chapter documents the collection of three stage II Pseudocarcinus gigas zoeas from oceanic waters in the vicinity of Pedra Branca off southern Tasmania (within the region: longitude 147°09'32"-147°28'30": latitude 44°11'23"–44°12'30"), an area at the southern limit of the range of P. gigas. The diel vertical distribution of brachyuran larvae was determined from plankton tows at 10 sampling depths, from 10 to 900 m, collected at 9 periods over 48 hours (Fig. 1). Different depths were sampled in a continuous tow using an EZ plankton net (1 m2 mouth) deployed from the Southern Surveyor, CSIRO Fisheries Research Vessel (modified Tucker trawl, see Harding et al., 1987). Sampling was conducted in November 1992 near the edge of the continental shelf. Bottom depth ranged from 965 to 1584 m and sampling depth of plankton tows was at 100 m intervals, from 10 to 900 m. The volume of water filtered at each sampling depth ranged between 1650 to 550 m3 and averaged 1140 m3. Sampling was conducted almost continually over 14 and 15 November 1992. Only higher brachyuran larvae were sorted from plankton samples so counts for total Brachyura probably exclude larvae of crabs from the families Homolidae and Dromiidae, both of which are present in the region. A total of 342 brachyuran larvae were collected and of these, only 3 were identified as Pseudocarcinus gigas. Identification was based on form of the telson, presence of a dorsal spine on the first abdominal somite, and on the setation patterns of the maxillule and maxillae (Gardner and Quintana, 1998; Chapter 3). All three P. gigas larvae were at the second of the five zoeal stages and were found in the upper 100 m; this depth was also where most of the brachyuran zoeas were collected. Total brachyuran larvae were distributed predominantly in the surface waters, above 100 m, and this distribution did not appear to be affected by time of sampling. Only one sample appeared to have a different pattern of larval distribution with most brachyuran larvae at >800 m (midday sample: 11.34-14.30; Fig. 1). The presence of zoeas at this depth is rare (Rice, 1979) although very few larvae were captured in this tow (n=13) so the unusual depth distribution may be spurious. The temperature gradient was relatively constant with depth and there were no indications of thermoclines. The presence of very few P. gigas larvae in samples is probably due to the southern latitude of the sampling program. The date of sampling appears to be appropriate as P. gigas larvae should have been released prior to the sampling trip. Hatching usually occurs in late October/early November and the presence of stage 2 zoeas also suggests that the plankton sampling was after the peak period of hatch. Likewise, the presence of stage 2 larvae suggests that settlement would not have occurred by this date. Further, the larval duration of laboratory reared larvae is around 50 days which suggests that settlement would not occur before late December (Gardner and Northam, 1997; Chapter 8). Sampling was conducted near the edge of the continental shelf, an appropriate region as it suspected to be where larval release occurs (Levings et al., 1996). Consequently, a possible explanation for the low capture rate of P. gigas larvae is the southerly latitude. Although commercial fishers occasionally capture adult P. gigas in this region as bycatch, high densities and targeted fishing generally only occurs further north above 42°30'. 55 Figure 1. Effect of time of day on brachyuran zoeal density (bars) in relation to water depth and temperature (lines) at Pedra Branca on 14th and 15th November 1992. A total of only 3 Pseudocarcinus gigas larvae were collected; these samples are marked with an asterisk (*). Time of sample 1.52-4.37 (24 h clock) 8.52-11.49 14.20-17.15 5 7.5 10 12.5 5 7.5 10 12.5 0 10 20 30 40 0 1 2 3 4 5 6 5 7.5 10 12.5 Temperature (°C) depth (m) 0 200 400 600 800 Time of sample 21.43-0.33 (24 h clock) 2.17-4.37 5 7.5 10 12.5 0 0.5 1 1.5 2 2.5 7.09-9.12 5 7.5 10 12.5 0 5 7.5 10 12.5 Temperature (°C) * * depth (m) 200 400 600 800 0 10 20 30 40 Time of sample 11.34-14.30 (24 h clock) 0 50 100 150 200 16.01-18.55 5 7.5 10 12.5 5 7.5 10 12.5 0 10 20 30 40 50 20.59-23.55 5 7.5 10 12.5 Temperature (°C) depth (m) 0 * 200 400 600 800 0 2.5 5 7.5 10 0 25 50 75 100 0 1 2 3 4 5 density of larvae (larvae/1000m3) References Gardner, C. and Northam, M. 1997. Use of Prophylactic Treatments for Larval Rearing of Giant Crabs Pseudocarcinus gigas (Lamarck). Aquaculture, 158: 203-214. Gardner, C. and Quintana, R. Larval development of the Australian giant crab Pseudocarcinus gigas (Lamarck 1818)(Decapoda: Oziidae) reared in the laboratory. J. Plank. Res., 20(6): 1169-1188. Harding, G.C., Pringle, J.D., Vass, W.P., Pearre, S. Jr., and Smith, S.J. 1987. Vertical distribution and daily movements of larval lobsters Homarus americanus over Browns Bank, Nova Scotia. Mar. Ecol. Prog. Ser., 41: 29-41. Levings, A., Mitchell, B.D., Heeren, T., Austin, C. and Matheson, J. 1996. Fisheries biology of the giant crab (Pseudocarcinus gigas, Brachyura, Oziidae) in southern Australia. High Latitude Crabs: Biology, 56 Management, and Economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska Fairbanks, 125-151. Rice, A.L. 1979. A remarkable benthic catch of Portunid crab zoeae (Decapoda, Brachyura). Crustaceana, Suppl. 5, 17-21. 57 Behavioural basis of depth regulation in the first zoeal stage 5 Research for this chapter has been previously published as: Gardner, N.C. 1996. Behavioural basis of depth regulation in the first zoeal stage of the giant crab Pseudocarcinus gigas (Brachyura: Oziidae). Proceedings of the International Symposium on Biology, Management, and Economics of Crabs from High Latitude Waters. Anchorage, Alaska, Oct. 1995. Pp. 229-253. 58 Abstract The responses of the first zoeal stage of Pseudocarcinus gigas (Lamarck) to stimuli affecting vertical migration were determined. Larvae are negatively buoyant and sink passively at 0.61 cm/s at 35 ppt salinity. Negative geotactic response was weak in larvae immediately after hatching, however, larvae exhibited strong negative geotaxis in tests at 15 and 20 hours; this pattern was less evident in subsequent samples up to day 13. It is suggested that a strong negative geotactic response is important initially to induce upward migration of larvae from the site of release, at approximately 160-250 m depth, to the surface. Light-adapted larvae were sensitive to change in light intensity from overhead lighting with both decreases and increases in intensity inducing downward migration; this negative phototactic response was only induced at low light intensities (below 230 lux). It is suggested that this may be a shadow response which has not been previously reported for increases in intensity. Distributions of larvae in angled testing columns suggest that the larvae are actively swimming away from the light source when negative phototaxis occurs. Where positive phototaxis occurred, larvae appeared to orientate with gravity while light provided the stimulus for increased locomotory activity. Larvae did not respond to small changes in pressure (0.24-2.70 cm water/s, at surface) of a similar magnitude to that which they would experience by vertical swimming. They were able to detect currents and most larvae could maintain position in currents of 1.12 cm/s. Introduction Development of the larvae of giant crabs consists of a prezoea, 5 zoeal stages and a megalopa (Gardner and Quintana, 1998; Chapter 3). Vertical migration of these larvae, and their response to currents, is likely to influence dispersal, survival and growth. Mechanisms controlling migration in planktonic crustaceans are highly complex and can be influenced by a range of stimuli including pressure, polarisation of light, gravity, absolute light intensity, salinity, change in light intensity, predator fields, prey fields, and temperature (Knight-Jones and Morgan, 1966; Umminger, 1969; Latz and Forward, 1977; Forward et al., 1984; Gliwicz and Pijanowska, 1988; and Forward, 1990). While many stimuli may be perceived by brachyuran larvae, and contribute to vertical migration, it appears that migration is predominantly controlled by phototaxis in the presence of light (Thorson, 1964), and geotaxis (gravity) and barokinesis (pressure) in the absence of light (Sulkin, 1973). The response to external stimuli varies between species so that widely different migration strategies exist in relation to depth and diel cycle. Plankton, including decapod larvae, do not show a clear general diel pattern in oceanic water off Tasmania; some species exhibit nocturnal upward migration while others are found on the surface during daylight hours (Pers. comm., Barry Bruce, Div. Mar. Sci., CSIRO, Hobart, Australia, Sept. 1995). Consequently, no assumptions can be made on the general pattern of larval movement in P. gigas. Further, it is not possible to predict larval movements of the giant crab based on other studies as deep-water crabs, or those with oceanic larval stages, have received scant attention in behavioural research. Most other studies on brachyuran larval behaviour have focused on estuarine species with only one other deep water species, Geryon quinquedens, having been studied (Kelly et al., 1982). The aim of this research was to provide behavioural information to assist with understanding movement of giant crab larvae and to contribute to research on deep 83 water crabs in general. An additional aim was to identify larval environmental preferences to assist with small scale production of juveniles. The responses of stage 1 larvae to several stimuli influencing vertical migration were assessed; gravity, spectral sensitivity, absolute light intensity, change in light intensity, orientation of light and change in pressure. Larvae were also tested for their response to lateral current movement. Materials and methods Source of larvae Thirty ovigerous females were collected from depths in the range of 300 – 380 m off the east coast of Tasmania (41°,15'S;148°40'E) in May 1994 by a commercial fisher. These females ranged in size from 2.2 – 3.5 kg and were held in two 4 m3 tanks with flow through, unfiltered, water supply. Crabs were fed twice weekly with abalone (Haliotis spp.) or mackerel (Trachurus sp.) and food remains were removed after 48 hours. Hatching of larvae commenced in November 1994 and continued for three weeks. Stage one zoeas were collected for behavioural experiments by first flushing the system of any zoeas present in the tanks, and then reducing flow so that only newly hatched larvae could be drawn from the tanks. Zoeas were mixed by drawing samples from each of the two holding tanks, so that no female contributed more than 50% of the larvae in any trial. Further mixing of larvae was achieved within tanks, as daily monitoring of the egg masses indicated that on any day where larvae were collected, hatching occurred in the egg masses of at least three females. Sinking and swimming rates Sinking rates were determined for 30 stage one zoeas collected from each of the two holding tanks and narcotised until immobile in a solution of 0.05% 2-phenoxyethanol in sea water. The larvae were then allowed to sink through a sea water filled plexiglas column and their rate of descent measured for 25 cm after an initial descent of 25 cm. Swimming rate was measured for 30 larvae which were introduced to a clear, horizontal plexiglas column with illumination from one end (500 lux). The time required for larvae to swim 15 cm without stopping or turning was measured. General experimental procedures for geotaxis, phototaxis and barokinesis experiments Zoeas were discarded after use in an experiment and were replaced if held for more than four hours after collection before use in a behavioural trial (with the exception of geotaxis experiments where zoeas up to 13 days old were used). All experiments were replicated four times with each replicate staggered between other treatment levels. For instance, the behavioural responses of zoeas to each level of light intensity was tested with a single trial at each level and then the entire set of trials was repeated to a total of four times. Trials were conducted at 13°C in a temperature controlled room. All water used was 0.2 µm filtered seawater of consistently 35 ppt salinity. The movement of larvae was measured by placing approximately 30 larvae into a clear plexiglas column divided into eleven 5 cm segments (30 mm internal diameter x 550 84 mm length) and recording their position after subjecting to a stimulus. Larvae were introduced into the middle of the column with a transparent 60 ml syringe. The light intensity that larvae are adapted to has been shown to influence their behaviour in response to light stimulus (Forward, 1974). To prevent confounding of experiments from previous light exposure, larvae were acclimatised to the lighting for 10 min within the transparent syringe. Preliminary trials established suitable duration of trials to be 2 min. Longer periods than this resulted in all the larvae gathering at either end of the testing chamber. Also, no change in the nature of the larval response occurred in trials of 15 min compared with 2 min trials. Barokinesis trials were reduced to 1.5 min due to constraints on the apparatus used to alter pressure. Injection of larvae into the testing column caused currents which tended to move the larvae vertically upwards. To compensate for this effect, the initial position of the larvae was determined by repeatedly introducing larvae into the testing column and recording their position immediately (433 zoeas in the angled column, 394 in the vertical column). The mean column position of these larvae was used as the point of origin in all trials. Significant difference between treatments was determined using the method outlined by Sulkin et al. (1980). "Mean position value" was calculated by assigning weights from 1-11 for each of the sections along the testing chamber, multiplying the weights by the number of larvae in each section and dividing the product by the total number of larvae. The mean position value was calculated for each replicate and these values were then used to compare treatments with the non-parametric Mann-Whitney U test (Zar, 1974). Differences in means were considered significant at P<0.05. Geotaxis Significant upwards movement of the larvae in the absence of light or pressure changes was attributed to negative geotaxis. The apparatus used to study geotactic response is illustrated in Fig. 1. Geotactic response of larvae was tested at: immediately post-hatch, 15 h, 20 h, 2 d, 6 d, 9 d, and 13 d. Larvae were maintained in a 1000 l tank on a recirculating water system with UV sterilisation and biofiltration. Larvae were fed 2nd instar artemia nauplii enriched with Protein Selco™ and they moulted to second stage zoeas after 7 days. At the culmination of the geotaxis trials (13 d), larvae were still at second stage zoea. 85 Figure 1. Detail of testing chamber and experimental apparatus used to measure larval response in darkness and also to light of 617 nm and 478 nm. Wavelength and light intensity was altered with optical filters held in a rack between the light source (quartz halogen flood light) and the testing chamber. Larvae were introduced through the entry port and water and air bubbles displaced through the exit port. DRAIN EXIT PORT BALL VALVE ENTRY PORT TESTING CHAMBER OPTICAL FILTERS LIGHT SOURCE Phototaxis Several aspects of phototactic response were investigated: spectral sensitivity, response to constant light intensity, response to change in light intensity, and the effect of incident angle of light source. All lighting was from quartz halogen globes which was diffused through neutral density optical filters. Generally, intensity of lighting was altered between treatments by changing the wattage of the globe or the neutral density filter although, for change in light intensity trials, the intensity was altered by moving the light source away or towards the testing chamber. Light intensity was measured in lux with a Gossen Profisix™ plan-diffuser light meter. The apparatus used to study spectral sensitivity is illustrated in Fig. 1. Lighting was from beneath and wavelength was altered with red (Kodak #25, dominant wavelength = 617 nm) and blue (Kodak #47A, dominant wavelength = 478 nm) gelatine filters. Response to constant light intensity, response to change in light intensity, and the effect of incident angle of light source were determined with the apparatus illustrated in Fig 2. 86 Natural underwater distribution of light was approximated by submerging the testing chamber in a 400 l tank, using angled light and diffusing the light source with neutral density filters. The walls of the outer tank were blackened walls and the tank was filled with 0.2 µm filtered seawater. All trials with constant light intensity were conducted with the light source in the lower position. Test intensities ranged from 3 lux to 40000 lux, recorded from the top of the testing chamber. To test the effect of change in light intensity, the light source was moved towards or away from the testing chamber by a variable speed 12 V electric motor (Fig. 2). The change in intensity commenced as the larvae were introduced to the column, and continued for the duration of each trial. The effect of change in light intensity was examined for both increasing and decreasing intensities for intensities between 6 and 2000 lux. The range of intensities experienced by larvae for each treatment is given in Fig. 8. Rates of change in light intensity, under natural conditions at sunset, were determined by measuring decline in light intensity on two days in September. Readings were taken every two minutes and rates of change averaged for the two days. Simulated declines in intensity were considerably faster than that which occurs at sunset (Table 1). 87 Figure 2. Experimental apparatus used to measure larval response to fixed intensity white light and change in light intensity. The light source was moved up or down the track to adjust intensity with a variable speed, 12 V, electric motor connected to a gearbox so as to reduce speed of revolution and increase torque. Initial light intensity was adjusted with neutral density filters or by changing the wattage of the quartz halogen globe. GEARBOX AND SPOOL MOVEABLE LIGHT SOURCE TESTING CHAMBERS EXIT PORT SPEED ADJUSTMENT ENTRY PORTS The effect of incident angle of light source was tested by comparing larval distributions in a testing chamber angled directly towards the angled light source with a testing chamber oriented vertically (Fig. 2). The angle of light incident on the testing chambers was 45° to vertical after refraction through the water surface. Trials conducted to compare the effect of incident angle of light source were conducted simultaneously for the vertical and angled testing chambers. 88 Table 1. Change of light intensity during natural sunset compared with experimental rates of intensity decline. Simulated change in intensity Initial Intensity (lux) Naturala (lux/min) Slow (lux/min) Rapid (lux/min) 15 2.8 3 4.5 230 20.9 60 95 900 57.5 175 395 2000 103.9 450 875 a Values for natural change in intensity at sunset are derived from a regression fitting recorded intensity changes. Barokinesis As with phototaxis experiments, the testing chamber was submerged in a water filled, blackened tank (Fig. 3). The testing chamber was orientated vertically and the response of larvae to change in pressure was measured in darkness and also with 800 lux lighting, angled at 45°. Larvae were introduced to the testing chamber through a ball valve which could be closed to seal the chamber. Silicon tubing was connected to the testing chamber and filled with seawater so that the water was continuous with that in the testing chamber. Pressure was then altered by raising or lowering this tubing with a variable speed electric motor and the rate of pressure change recorded as vertical cm per second. This method of recording change in pressure allowed pressure change to be directly related to potential larval movement in surface waters. 89 Figure 3. Experimental apparatus used to measure larval response to rate of change in pressure. Pressure in the testing chamber was regulated by the height of the silicon tubing, as the water in the tubing was continuous with that in the testing chamber. Change in pressure was achieved by raising or lowering the silicon tubing at different rates with a variable speed, 12 V, electric motor. Lighting was with a 50 W, quartz halogen floodlight. The ball valve was opened to introduce larvae into the testing chamber, with the syringe, and then sealed for pressure trials. The syringe and testing chamber supports were orientated away from the light source to prevent shadowing. LIGHT SOURCE, 800 LUX AT TESTING CHAMBER CLEAR SILICON TUBING TESTING CHAMBER GEARBOX AND SPOOL SYRINGE BALL VALVE . Rheotaxis experimental method Rheotactic responses of individual zoeas were measured within a 10 mm internal diameter glass tube connected to a peristaltic pump to provide current (Fig. 4). Larvae were introduced to the apparatus with a 60 ml syringe and pulsation was largely removed by constricting the 1.5 m length of expandable silicon tubing, feeding from the 90 pump, with a screw valve. Current speed was adjusted with the peristaltic pump and measured by recording the speed of passage of bubbles through the apparatus. The rheotactic responses of larvae were tested for current speeds from 0.35 – 1.87 cm/s. Lighting was at 90° to the current flow and oriented horizontally to produce an intensity of 80 lux incident on the testing chamber. A positive rheotaxis response was recorded when larvae actively oriented themselves and swam into the current or maintained position; a negative response was recorded if the larvae were swept along the testing chamber or swam indifferently to the current. At least thirty larvae were used for each current speed tested. Figure 4. Apparatus used for rheotaxis experiments. Pulsation in flow from the peristaltic pump was reduced by including 1.5 m of silicon tubing between the pump and the glass testing chamber and then constricting tubing immediately before the testing vessel. This caused the silicon tubing to expand and contract, which removed pulsation. Lighting was from a 50 W quartz halogen floodlight, 2.5 m distant and angled at 90° to the testing chamber to produce 80 lux incident on testing chamber. LIGHT SOURCE CLEAR SYRINGE TESTING CHAMBER SCREW VALVE SILICON RUBBER TUBING DRAIN PERISTALTIC PUMP RESERVOIR Results 91 Swimming and sinking speeds The mean vertical upwards swimming speed of larvae, without pausing, was 1.61 (±0.38 s.d., n=30) cm/s. Assuming that larvae did not pause in swimming and also chose to swim vertically upwards, this rate of swimming would enable larvae released at 350 m depth to reach the surface waters in about 6 hours. Larvae swam with the dorsal spine foremost and sank in the opposite manner, with the dorsal spine trailing. The average sinking rate for anaesthetised larvae was 0.61 (±0.084 s.d., n=30) cm/s. Geotaxis There was a tendency for stage 1 zoeas to be negatively geotactic but this appeared to be influenced by the age of the larvae, or possibly time of day (Fig. 5). Immediately after hatching the larvae did not exhibit any clear geotactic response. Larvae exhibited strong negative geotaxis in tests at 15 h and 20 h but the strength of response declined in older larvae. The mean response of larvae tested at 2 d and 13 d was positively geotactic although the pattern of movement for all larvae tested was not clear with some larvae moving upwards in the column and others downwards. 92 Figure 5. Ontogenic change in geotactic response of stage 1 and 2 zoeas. “Int. Pos.” is the distribution of larvae immediately after introduction to the testing column; the mean of this distribution was used to define the zero position of column height in the subsequent geotactic response plots. "Hatch" plot is for larvae collected and tested within 10 minutes of release. INT.POS. 225 HATCH 15 HOUR 20 HOUR 225 225 225 125 125 125 25 25 25 -75 -75 -75 -175 -175 -175 -275 -275 -275 Column Position (mm) 175 125 75 25 -25 -75 -125 -175 -225 -275 0 25 50 75 0 10 20 30 40 2 DAYS 225 0 25 50 75 100 6 DAYS 0 25 50 75 100 9 DAYS 13 DAYS 225 225 225 125 125 125 25 25 25 -75 -75 -75 -175 -175 -175 -275 -275 -275 Column Position (mm) 175 125 75 25 -25 -75 -125 -175 -225 -275 0 10 20 30 0 5 10 15 2025 0 5 10 15 20 25 0 10 20 30 40 Phototaxis Fixed intensity In the experimental situation shown in Figure 1 (results, Fig. 6.), downward movement is positive phototaxis. There was strong phototaxis to blue light of 2 lux while larvae failed to exhibit a strong phototactic response when exposed to red light of 11 lux (Fig. 6). In the experimental situation shown in Figure 2 (results Fig. 7), a positive phototactic response will result in upwards movement. Larvae were phototactic to all intensities tested for white light (3-40000 lux, Fig. 7) and at no intensity were larvae induced to swim or sink downwards. There were no significant differences between 93 larval responses to any intensities tested (P<0.05), in both the angled and the vertical columns. Figure 6. Vertical migration of Z1 larvae in response to white, red (617 nm) and blue (478 nm) light directed from below the vertical testing column. The response of larvae in darkness was measured and has been plotted as the lowest intensity reading for the white light series. This point has been artificially allocated an intensity of 0.5 lux so that light intensity could be plotted on a logarithmic scale. Negative values indicate positive phototaxis ie. movement towards the light below the testing chamber (see Fig. 1). 104 45 79 258 121 158 113 184 121 mean column position + std dev. 102 104 n, white 154 90 n, red 121 n, blue 250 200 150 100 50 0 -50 -100 -150 -200 -250 -300 1 white light 10 100 red light (617 nm) 1000 intensity (lux) blue light (478 nm) 94 Mean larval position in testing column ± std dev. Figure 7. Migration response of Z1 larvae, in vertical and angled testing columns, to light of different intensities. Means have been artificially displaced sideways to prevent overlap of error bars. Significant differences between distributions of angled and vertical tests are denoted by * at P<0.05. Squares -angled column, diamonds – vertically orientated column. Positive values indicate positive phototaxis i.e. movement towards the light above the testing chamber (see Fig. 2). 183 144 147 137 128 158 254 168 * * * 227 251 279 n, angled 172 n, vertical 2500 40000 300 200 100 0 -100 Dark 3 18 800 Light intensity (lux) 95 Effect of change in light intensity There was a significant (P<0.01) effect of change in light intensity on column position of larvae (Table 2). Downward movement (negative phototaxis, Fig. 2) was most evident in larvae exposed to changes in intensity between 9-15 lux and 6-15 lux (Figs. 8 to 10). In the vertical column, there was a significantly greater movement away from the light source at slow rates of changes in intensity than at faster rates (P<0.05; Fig. 8). There appeared to be no effect of direction of light change (increasing or decreasing intensity) on larval movement (Fig. 9). Table 2. Statistical comparisons of the effect of different light intensity changes under different regimes of: rate of change; increasing or decreasing intensity; and angled or vertical testing column. Light intensity changes are ranked from lowest to most positive phototactic response, left to right respectively. Bars beneath intensity ranges denote significance by joining non-significant tests (P<0.01). See Fig. 2 for experimental apparatus. "n" = number of larvae per treatment level. Treatment Intensity range (lux) n Rapidly increasing intensity (angled) 6-15 40-230 250-2000 110-900 1098 Rapidly decreasing intensity (angled) 15-6 230-40 900-110 2000-250 728 Slowly increasing intensity (angled) 9-15 130-230 550-900 1100-2000 838 Slowly decreasing intensity (angled) 15-9 230-130 2000-1100 Rapidly increasing intensity (vertical) 6-15 40-230 110-900 250-2000 1001 Rapidly decreasing intensity (vertical) 15-6 230-40 2000-250 900-110 1012 Slowly increasing intensity (vertical) 9-15 130-230 550-900 1100-2000 985 Slowly decreasing intensity (vertical) 15-9 230-130 2000-1100 900-550 900-550 938 860 96 Figure 8, a-d. Effect of rate of change of light intensity on vertical migration in Z1 larvae exposed to different varying light intensities. The effect of fast and slow rates of change were tested in combinations of decreasing/increasing intensity and with the testing column orientated directly towards the angled light source (angled) or else orientated vertically (vertical). Plot symbols have been artificially displaced sideways to prevent overlap of error bars. Positive values indicate positive phototaxis (see Fig. 2). * indicates significance at P<0.05. a. Increasing intensity, angled column Mean Column Position (mm) ± std dev 200 100 0 -100 -200 Rapid -300 Slow -400 6-15/9-15 light intensity (lux) 110-900/550-900 250-2000/1100-2000 40-230/130-230 Mean Column Position (mm) ± std dev b. Increasing intensity, vertical column 200 100 0 -100 -200 * * Rapid Slow -300 6-15/9-15 40-230/130-230 light 110-900/550-900 250-2000/1100-2000 intensity (lux) Mean Column Position (mm) ± std dev c. Decreasing intensity, angled column 200 100 0 -100 -200 -300 15-6/15-9 Mean Column Position (mm) ± std dev Rapid Slow * 230-40/230-130 900-110/900-550 light 2000-250/2000-1100intensity (lux) d. Decreasing intensity, vertical column 300 200 * 100 0 -100 -200 * Rapid Slow -300 15-6/15-9 230-40/230-130 900-110/900-550 light 2000-250/2000-1100intensity (lux) Figure 8, a-d. Effect of rate of change of light intensity on vertical migration in Z1 97 Figure 9, a-d. Effect of direction of change (increasing or decreasing) of light intensity on vertical migration in Z1 larvae exposed to different varying light intensities. The effect of direction of change was tested in combinations of rapid/slow change in intensity and with the testing column orientated directly towards the angled light source (angled) or else orientated vertically (vertical). Plot symbols have been artificially displaced sideways to prevent overlap of error bars. Positive values indicate positive phototaxis (see Fig. 2). * indicates significance at P<0.05. Mean Column Position (mm) ± std dev. a. Angled column, rapid change in intensity 200 100 0 -100 -200 Increasing Decreasing -300 Mean Column Position (mm) ± std dev. 6-15/15-6 40-230/230-40 110-900/900-110 light 250-2000/2000-250 intensity (lux) b. Angled column, slow change in intensity 200 100 * 0 -100 -200 Increasing Decreasing -300 9-15/15-9 130-230/230-130 light 550-900/900-550 1100-2000/2000-1100 intensity (lux) Mean Column Position (mm) ± std dev. c. Vertical column, rapid change in intensity 200 100 0 -100 Increasing Decreasing -200 -300 6-15/15-6 40-230/230-40 110-900/900-110 light 250-2000/2000-250intensity (lux) Mean Column Position (mm) ± std dev. d. Vertical column, slow change in intensity 200 * 100 0 -100 -200 Increasing Decreasing -300 9-15/15-9 130-230/230-130 light 550-900/900-550 1100-2000/2000-1100intensity (lux) Figure 9, a-d. Effect of direction of change (increasing or decreasing) of light 98 Figure 10, a-d. Effect of orientation of testing column to light source on vertical migration in Z1 larvae exposed to different varying light intensities. The effect of orientation was tested in combinations of decreasing/increasing intensity, at rapid/slow rates. Plot symbols have been artificially displaced sideways to prevent overlap of error bars. Positive values indicate positive phototaxis (see Fig. 2). * indicates significance at P<0.05. Mean Column Position (mm) ± std dev. a. Slowly increasing intensity 200 100 0 -100 -200 Angled Vertical -300 9-15 110-230 550-900 light 1100-2000 intensity (lux) Mean Column Position (mm) ± std dev. b. Rapidly increasing intensity 200 100 0 -100 * -200 Angled Vertical -300 6-15 40-230 110-900 light 250-2000 intensity (lux) Mean Column Position (mm) ± std dev. c. Slowly decreasing intensity 200 100 0 -100 -200 Angled Vertical -300 15-9 230-110 900-550 light 2000-1100 intensity (lux) Mean Column Position (mm) ± std dev. d. Rapidly decreasing intensity 200 100 0 -100 -200 * Angled Vertical -300 15-6 230-40 900-110 light 2000-250 intensity (lux) Figure 10, Effect of orientation of testing column to light source on vertical Effect of a-d. orientation of light source. The effect of orientation of light source was investigated by exposing the larvae to an angled light source and comparing the distribution of larvae in a vertical and an angled testing chamber (Fig. 2). This was trialed with larvae exposed to white light of fixed intensity (Fig. 7) and also with varying intensity (Fig. 10). In the fixed light intensity experiments, there was significantly greater movement of larvae along the column in the 99 vertically oriented column, than in the column angled directly towards the light source (P<0.05; Fig. 7). This trend was not significant at the higher light intensities measured, 2500 and 40000 lux or in darkness (P>0.05). In the varying light intensity experiments, significantly different movement, between the vertical and angled columns, only occurred when larvae exhibited negative geotaxis in response to rapidly changing light intensity (P<0.05; Fig. 10). Barokinesis Larvae did not clearly respond differently to increasing or decreasing pressure under conditions of total darkness or with 800 lux overhead lighting (Fig. 11). Although significant differences were observed, these were considered inconclusive as there was no consistent effect of the rate of change of pressure on larval movement. However, it is noteworthy that the mean upward movement of larvae when exposed to overhead lighting was smallest, although still upwards, when the water column was lowered (decreasing pressure) at 1.85 cm/s (Fig. 11b). This rate of change was the rate closest to the observed swimming speed of stage 1 zoeas, 1.61 cm/s. 100 Figure 11, ab. Comparison of the effect of increasing and decreasing pressure of different rates on the vertical migration of Z1 larvae held in darkness (a) or with 800 lux overhead lighting (b). Means have been displaced sideways to prevent overlap of error bars. Increasing pressure - squares, decreasing pressure - diamonds. a. 153 126 141 113 209 186 129 n increasing 165 n decreasing 250 200 Mean position in testing column ± std dev. 150 100 * 50 0 -50 -100 0.24 cm/s b. 145 127 1.1 cm/s 109 155 1.85 cm/s 230 135 2.7 cm/s 169 n increasing 124 n decreasing 250 200 150 * 100 50 0 -50 0.24 cm/s 1.1 cm/s 1.85 cm/s 2.7 cm/s Rate of change of pressure 101 Rheotaxis Larvae exhibited positive rheotaxis with the greatest proportion of larvae responding at a current speed of 1.12 cm/s (Fig. 12). At lower current speeds (0.35 & 0.59 cm/s), many larvae appeared to be unaware of the current and swam back and forth along the testing column apparently in response to the lighting. At current speeds greater than 1.12 cm/s (1.4 & 1.87 cm/s) larvae tended to swim into the current but failed to maintain position. Figure 12. Rheotactic response of stage 1 larvae to current velocities of 0.35-1.87 cm/s. Values represent percentage of total larvae responding to stimuli so no measure of error is presented. error is presented. 30 32 30 33 37 n % Maintaining Position 100 75 50 25 0 0.35 0.59 1.12 1.4 Current Speed, cm / s 1.87 Discussion The pattern of vertical swimming resulting in upward movement was the same as that described by Sulkin (1984); larvae are negatively buoyant which is countered by upwards orientation and locomotion to produce upwards swimming. Passive sinking rates of P. gigas (x=0.61 cm/s) were similar to that of stage one zoeas of several other brachyuran species: Cancer magister, 0.64 cm/s (Jacoby, 1982); Ebalia tuberosa, 0.60 cm/s (Schembri, 1982); and Hemigrapsus oregonensis, 0.67 cm/s (Arana and Sulkin, 1993). The upwards locomotory force of these species is sufficient to counter sinking and produces similar mean upwards swimming speed to that of P. gigas (1.61 cm/s): Cancer magister, 0.95 cm/s (Jacoby, 1982); Ebalia tuberosa, 0.96 cm/s (Schembri, 1982); and Hemigrapsus oregonensis, 1.78 cm/s (Arana and Sulkin, 1993). As discussed by Sulkin (1984), depth regulation relies on the interaction between passive sinking, active swimming, and orientation. Orientation and the speed of swimming are then adjusted in response to external stimuli, such as gravity, light and pressure, to induce depth regulatory response. Geotaxis The observed pattern of geotaxis suggests that larval swimming is not clearly affected by gravity immediately after hatching. After this initial period the larvae exhibit strong 102 negative geotaxis which suggests that larvae migrate upwards to the surface waters. The speed of larval swimming suggests that this would be accomplished in a period of around 6 hours if the larvae were released at 350 m and did not rest. A negative geotactic response was also observed in the first zoeal stage of the deep sea crab, Geryon quinquedens (Kelly et al., 1982). The strong negative geotactic response observed for P. gigas decreased in older larvae and it may be that the primary function of the geotactic response is to induce upward migration from the deep-water site of release to the prey-rich surface layers. An ontogenic change in geotactic behaviour between instars has also been reported for shallow water species Leptodius floridanus and Panopeus herbstii (Sulkin, 1973), Callinectes sapidus (Sulkin et al., 1980), and Ebalia tuberosa (Schembri, 1982). However, the changes in geotaxis observed in P. gigas differ from these reports as there was a decline in geotactic response within stage 1, rather than between instars. Although the pattern of geotactic response up to 20 hours appears clear, the results of trials from 2 to 13 days are more difficult to interpret and they may be confounded by at least two factors. Diel cycles have been shown to persist in the laboratory which may have influenced results from these trials (Cronin and Forward, 1986). Attempts were made to only test healthy larvae by selecting active larvae, however, less fit larvae may have been selected for testing and may have compromised results. Phototaxis The spectral sensitivity of larvae in this study was essentially the same as that described for several other species in more detail by Forward (1987), and Forward and Cronin (1979). Larvae were more sensitive to shorter blue wavelengths, which have better penetration, than red wavelengths. Light adapted larvae were not induced to descend in response to overhead white light of any intensity tested. The experimental design used in this study was intended to demonstrate the approximate field conditions where sinking may have been induced by light. Consequently, the testing chamber was oriented vertically so that negative phototaxis would be required to counter negative geotaxis in order to produce net downward movement (light-induced positive geotaxis). Comparisons between other studies where phototactic response was examined with horizontally oriented columns should be made cautiously. Where phototaxis has been examined independently of geotaxis by the use of horizontal testing columns, larvae tend to exhibit negative geotaxis at low intensities and positive phototaxis at high intensities. This pattern has been observed in: Rhithropanopeus harrisii (Forward et al., 1984); Cancer gracilis, Lophopanopeus bellus bellus, Hemigrapsus oregonensis (Forward, 1987); and Paralithodes camtschatica (Shirley and Shirley, 1988). Forward (1987) attributed this pattern to predator avoidance. Avoidance behaviour, or negative phototaxis at low light intensity, is not clearly demonstrated in vertically oriented columns where the natural behaviour of negative geotaxis is incorporated. Both Schembri (1982, Ebalia tuberosa) and Jacoby (1982, Cancer magister) tested the response of crab larvae to different light intensities in vertical columns and observed only upward movement. It is tempting to infer that the positive phototactic response of P. gigas (Fig. 7) indicates that the larvae congregate at the surface during the day. However, Forward (1985 & 1988) considers the absence of negative phototaxis in vertically oriented columns to be a laboratory artefact in most studies. For example, Stearns and Forward (1984) found that the copepod Acartia tonsa was positively phototactic to all light intensities although the natural migration pattern is nocturnal. 103 Simulated natural underwater lighting distribution is difficult to achieve, so the observed response of P. gigas may be nothing more than a laboratory artefact despite attempted simulation of natural light distribution. Larvae exposed to change in intensity at low light levels responded by downward movement. Conversely, larvae exposed to change in intensity at higher light levels were unaffected. This response was more pronounced at slower rates of change in intensity (in the vertical column only) but was not affected by the sign of intensity change (increasing or decreasing). Light-induced downward movement in response to change in light intensity, regardless of whether intensity is increasing or decreasing, has not been previously reported. This response may be a variation of the predator avoidance or shadow response proposed by Forward (1986) where negative phototaxis was induced by a rapid decrease in intensity. The shadow response proposed by Forward (1986) was only initiated by rapid decreases in intensity and not increases as was observed in this study. Forward (1986) noted that the change in intensities which resulted in negative phototaxis were too rapid to simulate dusk or dawn. Because of this, he believed that the larval responses did not represent a typical behaviour relevant to diel vertical migration. The simulated rates of intensity changes in this study were also greater than that at dawn or dusk (Table 1), suggesting the response in P. gigas was a predator avoidance, shadow response. The response of stage 1 P. gigas zoeas to change in light intensity, only at low light levels, suggests that the larvae are adapted to respond to low levels; this supports the dismissal of the results of fixed light intensity trials as a laboratory artefact. There was significantly greater (P<0.05) upward movement of larvae in the vertically oriented column exposed to fixed intensity of light, compared with the column angled directly towards the light source (45°). This suggests that geotaxis is the orienting cue while photokinesis controls locomotory activity. Forward (1985) observed this same interaction in larvae of the xanthid crab Rhithropanopeus harrisii. Geotaxis appeared to be less important in the larval response to changes in light intensity at low light levels. When downward movement occurred in response to rapidly changing intensity, the larval distribution was further from the origin in the angled column. This suggests that larvae were actively swimming from the light, rather than passively sinking. In this case, the response is a true negative geotaxis rather than light-induced, positive geotaxis. Barokinesis Larval detection and response to small changes in pressure have been used to explain vertical migration patterns, as it is considered that pressure response provides a negative feedback on vertical movement (Knight-Jones and Morgan, 1966). Stage 1 zoeas of P. gigas did not appear to respond to small pressure changes; this has been reported elsewhere for species where the larvae occupy water of considerable depth: Callinectes sapidus (Sulkin et al., 1980); Geryon quinquedens (Kelly et al., 1982); and Hemigrapsus oregonensis (Arana and Sulkin, 1993). Barokinesis has been shown to change dramatically with ontogeny (Bentley and Sulkin, 1977; Wheeler and Epifanio, 1978) so older larvae of P. gigas may possess greater pressure sensitivity. Rheotaxis Stage 1 zoeas of P. gigas were able to detect currents and actively swam against them. Larvae appeared to be unable to detect slight currents below 1.12 cm/s and they were 104 swept along by currents slower than their maximum swimming speed (1.4 and 1.61 cm/s respectively). The combination of the ability of larvae to detect currents, and then to swim against them, resulted in a narrow window within which larvae could maintain position. This suggests that rheotaxis may not be important in larval dispersal. Rheotaxis has also been observed in estuarine species such as the megalopas of Cancer magister (Fernandez et al., 1994) and is thought to assist in movement to and from the estuary. With open ocean species, the function of rheotaxis is less obvious as the environment is more homogeneous. Shirley and Shirley (1988) also observed rheotaxis in an oceanic species, Paralithodes camtschatica, and suggested that the function of rheotaxis in the oceanic environment may be important for zoeal feeding and predator avoidance. Conclusions Based on observed negative geotaxis, first stage zoeas of giant crab probably ascend to the surface waters after hatching. This initial upwards migration is probably the main function of the initially strong geotactic response of larvae. As the gravity-initiated strong upward swimming declines with age, the role of geotaxis may be to orientate larvae rather than to induce migration. This would produce vertical movement in response to angled light stimuli as appeared to occur in this study. The observed negative phototaxis of larvae to increases and decreases in low light intensities has not been observed in estuarine species and may be a variation of the previously reported shadow response. However, assigning function to this unusual response is speculative and further research is required if the function is to be clarified. Sensitivity of larvae to intensity changes at only low light levels suggests that larvae are adapted for low light conditions. The presence of a rheotactic response is surprising in an oceanic species and may affect the dispersion of larvae by currents. Although rheotaxis may affect dispersal, the biological function of the response is likely to be otherwise in an oceanic environment, perhaps predator avoidance or prey capture. Behavioural responses of oceanic, deep-water species have been investigated in only a few studies. Understanding the nature of these responses to oceanic conditions may assist in understanding other mechanisms such as survival and dispersal. References Arana, M. and Sulkin, S. 1993. Behavioural basis of depth regulation in the first zoeal stage of the Pacific shore crab, Hemigrapsus oregonensis (Brachyura: Grapsidae). Pacif. Sci. 47:256-262. Bentley, E. and Sulkin, S.D. 1977. The ontogeny of barokinesis during the zoeal development of the Xanthid crab Rhithropanopeus harrisii (Gould). Mar. Behav. Physiol. 4: 275-282. Cronin, T.W. and Forward, R.B.Jr. 1986. Vertical migration cycles of crab larvae and their role in larval dispersal. Bull. Mar. Sci., 39(2): 192-201. Fernandez, M., Iribarne, O. and Armstrong, D. 1994. Swimming behaviour of Dungeness crab, Cancer magister Dana, megalopae in still and moving water. Estuaries 17: 271-275. Forward, R.B.Jr. 1974. Negative phototaxis in crustacean larvae: possible functional significance. J. Exp. Mar. Biol. Ecol. 16: 11-17. Forward, R.B.Jr. 1985. Behavioural responses of larvae of the crab Rhithropanopeus harrisii (Brachyura: Xanthidae) during diel vertical migration. Mar. Biol. 90: 9-18. Forward, R.B.Jr. 1986. A reconsideration of the shadow response of a larval crustacean. Mar. Behav. Physiol. 12: 99-113. Forward, R.B.Jr. 1987. Comparative study of crustacean larval photoresponses. Mar. Biol. 94: 589-595. Forward, R.B.Jr. 1988. Diel vertical migration: zooplankton photobiology and behaviour. Oceanogr. Mar. Biol. Ann. Rev. 26: 361-393. Forward, R.B.Jr. 1990. Behavioural responses of crustacean larvae to rates of temperature change. Biol. Bull. 178: 195-204. 105 Forward, R.B.Jr. and Cronin, T.W. 1979. Spectral sensitivity of larvae from intertidal crustaceans. J. Comp. Physiol. 133: 311-315. Forward, R.B.Jr., Cronin, T.W. and Stearns, D.E. 1984. Control of diel migration: photoresponses of a larval crustacean. Limnol. Oceanogr. 29: 146-154. Gardner, C. and Quintana, R. 1998. Larval development of the Australian giant crab Pseudocarcinus gigas (Lamarck 1818)(Decapoda: Oziidae) reared in the laboratory. J. Plank. Res. 20: 1169-1188. Gliwicz, M.Z. and Pijanowska, J. 1988. Effect of predation and resource depth distribution on vertical migration of zooplankton. Bull. Mar. Sci. 43(3): 695-709. Jacoby, C.A. 1982. Behavioural responses of the larvae of Cancer magister Dana (1852) to light, pressure, and gravity. Mar. Behav. Physiol. 8: 267-283. Kelly, P., Sulkin, S.D. and Van Heukelem, W.F. 1982. A dispersal model for larvae of the deep sea red crab Geryon quinquedens based upon behavioural regulation of vertical migration in the hatching stage. Mar. Biol. 72: 35-43. Knight-Jones, E.W. and Morgan, E. 1966. Responses of marine animals to changes in hydrostatic pressure. Oceanogr. Mar. Biol. Ann. Rev. 4: 267-299. Latz, M.I. and Forward, R.B. Jr. 1977. The effect of salinity upon phototaxis and geotaxis in a larval crustacean. Biol. Bull. 153: 163-179. Schembri, P.J. 1982. Locomotion, feeding, grooming and the behavioural responses to gravity, light and hydrostatic pressure in the stage 1 zoea larvae of Ebalia tuberosa (Crustacea: Decapoda: Leucosiidae). Mar. Biol. 72: 125-134. Shirley, S.M. and Shirley, T.C. 1988. Behaviour of red king crab larvae: phototaxis, geotaxis and rheotaxis. Mar. Behav. Physiol. 13: 369-388. Stearns, D.E. and Forward, R.B.Jr. 1984. Photosensitivity of the calanoid copepod Acartia tonsa. Mar. Biol. 82: 85-89. Sulkin, S.D. 1973. Depth regulation of crab larvae in the absence of light. J. Exp. Mar. Biol. Ecol. 13: 73-82. Sulkin, S.D. 1984. Behavioural basis of depth regulation in the larvae of brachyuran crabs. Mar. Ecol. Prog. Ser. 15: 181-205. Sulkin, S.D., Van Heukelem, W.F., Kelly, P. and Van Heukelem, L. 1980. The behavioural basis of larval recruitment in the crab Callinectes sapidus Rathburn: a laboratory investigation of ontogenic changes in geotaxis and barokinesis. Biol. Bull. 159: 402-417. Thorson, G. 1964. Light as an ecological factor in the dispersal and settlement of larvae of marine bottom invertebrates. Ophelia 1(1): 167-208. Umminger, B.I. 1969. Polarotaxis in copepods, III. A light contrast reaction in Diaptomus shoshone Forbes. Crustaceana 16: 202-204. Wheeler, D.E. and Epifanio, C.E. 1978. Behavioural response to hydrostatic pressure in larvae of two species of Xanthid crabs. Mar. Biol. 46: 167-174. Zar, J.H. 1974. Biostatistical Analysis. Prentice-Hall, Inc., Englewood Cliffs, N.J. 620 pp. 106 Effects of temperature and thermoclines on larval behaviour and development 6 107 Abstract The behavioural responses of giant crab Pseudocarcinus gigas larvae to temperature in the absence of light were analysed in experimental water columns. In addition, larvae were reared at 12 temperatures (in 1°C increments between 10.5 and 21.1°C) to determine the effects of temperature on instar duration, somatic growth and survival. Behavioural experiments were conducted with first and second instar zoeas and these readily penetrated experimental thermoclines of approximately 2, 5 and 10°C. Both stages had similar behaviour with upward swimming induced in water temperatures ≤12.7°C and downward swimming induced by temperatures ò16.2°C. Behavioural response to temperature appeared to constitute a negative feedback system which could contribute to depth regulation. Intermoult duration decreased with increasing temperature so that most rapid development through the five zoeal stages to megalopa was 41 d at 20.2°C. More rapid growth at higher temperatures was at the expense of somatic growth with smaller megalopas occurring in warmer treatments. Best overall survival was in treatments between 15.8 and 20.2°C with no larvae surviving to megalopa in those treatments below the threshold temperature where upward swimming was induced in behaviour trials (10.5 and 11.7°C). Many megalopas died shortly after moulting from zoea 5, particularly in treatments with rapid growth (>16.8°C). Optimal survival of megalopas was in treatments below temperatures where downward migration of stage 1 and 2 zoeas was induced. These results suggest that temperature is an important factor in determining larval vertical migration, in the absence of light, and larval rearing studies indicate that this provides metabolic advantages. Introduction Temperature profoundly influences development of decapod larvae and numerous studies have demonstrated effects on instar duration, morphology, feeding rate, size, incidence of deformity, and survival (Johns, 1981; Shirley et al., 1987; Minagawa, 1990). These effects are species specific and this chapter presents results of trials to assess the effect of temperature on larval development of the giant crab Pseudocarcinus gigas (Lamarck, 1818). Crustacean larvae do not experience temperature passively but regulate their vertical migration behaviour, and thus depth, in response to both absolute temperature and rates of temperature change (Forward, 1990). Sulkin (1984) reviewed the effect of temperature on vertical migration and noted that in most cases there is a direct relationship with swimming speed. The orientation of swimming responses may also be affected so that the geotaxis response is reversed (Forward, 1990). Forward (1990) considered that larval response to temperature constituted a negative feedback system so that depth was regulated relative to temperature. In nature, these responses of organisms to temperature are compounded by thermoclines which influence the vertical distribution of a range of planktonic animals (Harder, 1968). Evidence from laboratory studies with crab larvae indicates that thermoclines do not have an inhibitory effect on vertical migration although results from few species have been reported (Geryon quinquedens, Kelly et al., 1982; Eurypanopeus depressus, Sulkin et al., 1983; Callinectes sapidus, McConnaughey and Sulkin, 1984). Unlike most species whose larval behaviour has been studied, P. gigas is a relatively deep water, oceanic species. This chapter presents results of studies on the behavioural response of P. gigas to absolute temperature and temperature change (thermoclines). 106 Despite improved understanding of the mechanisms of the behavioural response of crab larvae to temperature, understanding of the function of the response remains vague. Haney (1988) discussed a range of hypotheses to explain the adaptive advantage of diel migration of planktonic organisms and these include predator avoidance, avoidance of damaging solar radiation, tracking of prey items, and metabolic advantages. As Haney (1988) noted, there has been relatively little experimental support for proposed metabolic advantages. Comparisons between the behavioural and developmental responses of P. gigas larvae were intended to provide a method of assessing if the behavioural response to temperature had metabolic advantages. Materials and methods Trials were conducted to investigate the effect of temperature on behaviour and on development. Behavioural responses were monitored for the first two zoeal stages while larvae in development trials were reared through to megalopa. Source of larvae Ovigerous females for all trials were collected from depths of 300-380 m off the east coast of Tasmania, Australia (41°17'S; 148°40'E) in June 1995 (development trials) and July 1996 (behavioural trials). Females ranged in size from 2.2-4.5 kg and were held communally in 4 m3 flow-through tanks receiving unfiltered seawater. Temperature in broodstock tanks ranged from 8 to 14°C and the lighting regime was approximately 10 h light. For development trials, larvae were collected from two tanks to ensure that larvae were not from a single parent; further mixing probably occurred as several females were releasing larvae in each tank. For behaviour trials, 8 females were separated into 4 tanks before larval release so that larvae could be collected separately. Each of these tanks yielded a replicate group of larvae which were reared as 200 l upwelling cultures at 14°C (range ±0.3°C) for 18 d, through to the second zoeal stage. The cultures were maintained in a reverse circadian cycle with 12 h light (i.e. light phase from 7 pm to 7 am) as experimentation was conducted in darkness. Water for these 200 l cultures was recirculated through a shared sump and biofilter to minimise variation from tank effects. Zoea larvae for both trials were fed a mix of Protein Selco™ enriched rotifers (Brachionus plicatilis) and artemia instar II nauplii for the first two zoeal stages (after which behavioural trials were terminated) and enriched artemia only thereafter. System design for behavioural experiments The response of Pseudocarcinus gigas larvae to thermoclines was investigated with an experimental system modified from McConnaughey and Sulkin (1984) to produce thermoclines in vertical columns (Figs. 1 and 2). Testing columns (450 x 50 x 50 mm) were surrounded by heated or chilled, upper and lower water baths, separated by a 10 mm insulated layer. Water in the lower bath was recirculated through a sump with a heat-chill unit while water in the upper baths was heated with aquarium heaters and circulated by aeration. Temperature of the upper water bath was increased relative to the lower water bath to produce thermoclines in the testing columns of approximately 10°, 5°, and 2°C. Trials to assess the preferred temperature range of stage 1 and 2 zoeas were conducted with 5 different temperatures in the lower water bath which were intended to increase in 107 2°C steps (actual values approximately 11°, 12.7°, 14.5°, 16.0°, and 18.4°C) although there was slight variation between tests on stage 1 and stage 2 zoeas (for precise values see Fig. 3 and 4 results). In control chambers at the above temperatures, the regime was the same in both the upper and lower water bath so that no thermocline was generated. Trials were run in darkness for 15 min and columns were then illuminated, to record the position of larvae, with red light of 617 nm wavelength (Kodak™ gelatine filter #25) at 10 lux which does not induce phototaxis (Forward, 1990; Gardner, 1996; Chapter 5). Light was directed perpendicular to the testing chambers so any phototaxis of larvae would not result in vertical movement along the column. No trials were run with simulated natural lighting due to the difficulty of avoiding laboratory artefacts (Haney, 1988). Figure 1. Experimental apparatus used to generate thermoclines in testing columns (T) measuring 50 x 50 x 450 mm. The testing column was surrounded by two water baths separated by 10 mm; the upper (U) was maintained at a higher temperature than the lower (L). Larvae were introduced to the testing column at the top or through a ball valve (V) at the base of the column. In the upper chamber, experimental temperatures were maintained with a heater (H) and circulation was achieved with aeration. Experimental temperatures were maintained in the lower chamber by pumping water (P) from a sump (S) connected to a heat/chill unit (H/C). Water was returned to the sump by gravity via a standpipe to regulate pressure. In the control testing chamber (right), water was circulated from the lower chamber to the upper so that no temperature gradient was created. T H U L V S H/C P Figure X. Experimental apperatus used to generate thermoclines in .... 108 Figure 2. Vertical temperature profiles generated with the experimental system. Temperature profiles were recorded after introducing 15 ml of seawater at the base of the column to simulate the introduction of larvae. Experimental protocol for behaviour experiments In each trial, between 20 and 35 larvae from each of 4 replicates were introduced to the testing column by syringe after acclimatising to the experimental temperature for 5 minutes. Each trial involved the use of one control column and up to three experimental columns operated simultaneously. At the conclusion of each experiment (15 min), the position of larvae in each of 4 divisions along the testing column was recorded. Larvae were never reused in any behaviour trial. The ability of larvae to penetrate thermoclines was analysed by comparing the larval distribution following introduction through the top or the bottom of the testing chamber. Larvae were introduced to the chamber with a syringe and this was attached to a ball valve to introduce them at the base. Experiments to test response to thermoclines were conducted with the coldest lower bath temperature configuration (11.1° and 10.8°C for Z1 and Z2 larvae respectively) as this induced upward swimming. The proportion of larvae in the upper half of the chamber was arc-sine square root transformed and the difference between the distribution of larvae introduced to the column at the base or the top compared by one tailed, paired student’s t-tests (Sokal and Rohlf, 1981). A significant difference indicated inhibition of larval movement by the thermocline. Additional experiments were conducted to determine the preferred temperature as indicated by vertical distribution. These trials used the same apparatus as described for testing the response to thermoclines, except that larvae were introduced at the top of the column only and a range of lower water-bath temperatures was tested. Several studies have demonstrated that decapod larvae will sink passively to avoid temperatures above a preferred range (Ott and Forward, 1976; Yule, 1984). Consequently, the accumulation of larvae in the lower half of the testing chamber indicated that the temperature above 109 the thermocline was above the preferred range. Statistical analyses were by students ttest, as described for thermocline experiments, to test the null hypothesis that larvae were distributed evenly in the testing column, that is, half were in the upper chamber. Both sets of experiments were conducted with zoeas 1 and 2 to assess ontogenic changes. System design and protocol for development experiments The effect of temperature on larval development was assessed by culturing larvae in 12 temperature regimes separated by approximately 1°C increments (10.5, 11.7, 12.8, 13.8, 14.8, 15.8, 16.8, 17.8, 18.6, 19.4, 20.2, and 21.1°C). The 12 temperatures treatments were created with an aluminium temperature gradient plate constructed from a large aluminium block (800 x 400 x 50 mm) with channels at each end for circulating heated or chilled water (Edwards and Van Baalen, 1970). This created an even temperature gradient along the aluminium block and 50 ml culture vessels were placed into 12 rows of holes bored into the block. Six vessels were maintained at each temperature; three replicate larval cultures and three vessels for preheating water prior to daily transfer of larvae (n=10 per replicate) into new water (0.2 µm filtered seawater). Larval development in each of the 12 temperature treatments was monitored by recording the number and stage of dead larvae and exuviae. Many larvae moulted through to megalopa but died shortly afterwards. Those larvae which survived 24 h after moulting to megalopa were scored as viable. All megalopas were sacrificed at this stage then rinsed with distilled water and dried at 80°C for 24 h to determine dry weight. Statistical analysis of development data The effect of temperature on the timing of moults was tested by one way ANOVA at each zoeal stage using log transformed time (days; Hayes, 1949). Changes in survival due to temperature was tested by Kaplan-Meier survival analysis (Miller, 1981) to enable censoring of viable megalopas. Few larvae survived through to megalopa (n=81) so it was not possible to retain replicates for analysis of the effect of temperature on weight and viability of megalopas. Consequently, standard least squares regression of weight against temperature was used to test the null hypothesis that the slope = 0. The effect of temperature on the number of viable megalopas in each treatment was assessed with a Kolmogorov-Smirnov test (Sokal and Rohlf, 1981) by comparing the observed number of viable megalopas against the predicted number of viable megalopas if there was no temperature effect (that is, predicted frequency = initial number of megalopas in treatment x (total viable megalopas/total initial number of megalopas)). Results Response of larvae to thermoclines There was no indication of inhibition of larval movement by the experimental thermoclines as larvae introduced at the base of the column had distributions that were not significantly different from those introduced at the top (P>0.05; Figs. 3 and 4, first two columns). This pattern was observed with both stage 1 and 2 zoeas. Two 110 comparisons were almost statistically significant: the 8.5°C thermocline with stage 1 zoeas (11.1 and 19.6°C; P=0.09), and the 4.3°C thermocline with stage 2 zoeas (15.1 and 10.8°C; P=0.08). However, these were caused by trends in vertical movement which were opposite to that which would occur if the thermocline had inhibited larval movement. Vertical migration Larvae in control treatments tended to swim actively upwards at temperatures ó12.6°C and 12.7°C, while sinking was induced at ò 16.2°C and 18.3°C for stage 1 and 2 zoeas respectively (Figs. 3 and 4, upper row). A similar pattern was detected in treatments where two alternative temperatures were separated by a thermocline. The response of stage 2 zoeas to the 16.3°C / 14.0°C treatment indicates that stage 2 zoeas will also descend at temperatures ò 16.3°C. Although a significantly higher proportion of zoea 1 larvae were found in the upper half of the 16.7°C / 12.6°C treatment, this appeared to be due to larvae swimming upwards from the lower temperature as most larvae were found immediately above the thermocline. Larvae placed in treatments where both the upper and lower temperature alternatives were outside the apparent preferred range tended to congregate around the thermocline (19.6°C / 11.1°C and 21.3°C / 12.6°C in zoea 1, and 17.9°C / 12.7°C in zoea 2). 111 Figure 3. Response of stage 1 zoeas to experimental thermoclines. Bars represent mean percentage of larvae (± SE, n=4) in each of the 4 divisions along the length of the testing columns. Values are the temperature (°C) in the upper and lower halves of testing columns. Upper row represents response of larvae in control experiments without a thermal gradient. Significant deviation from 50% of larvae in the upper half denoted by * (P<0.05). Placement of zoeas in testing column: bottom top top top 10.8 10.8 * 12.7 * * top 15.9 14.0 13.6 14.6 18.3 * * 16.3 13.6 top 19.0 * 20.1 * 18.3 10.8 10.8 12.7 * * * 15.1 15.1 14.0 15.9 17.9 19.5 21.6 23.5 * * * 14.0 15.9 18.3 10.8 * 10.8 12.7 0 19.4 19.4 22.5 23.5 * * * * 10.8 50 100 0 % 50 100 14.0 12.7 10.8 0 50 100 0 50 100 0 50 100 0 50 100 % Figure X. Response of stage 2 zo.. 112 Figure 4. Response of stage 2 zoeas to experimental thermoclines. Bars represent mean percentage of larvae (± SE, n=4) in each of the 4 divisions along the length of the testing columns. Values are the temperature (°C) in the upper and lower halves of testing columns. Upper row represents response of larvae in control experiments without a thermal gradient. Significant deviation from 50% of larvae in the upper half denoted by * (P<0.05). Placement of zoeas in testing column: bottom top top top top top * 11.1 11.1 * * 12.6 15.0 * 16.2 18.4 17.7 20.1 * 14.7 13.7 11.1 * * 16.2 18.4 12.6* * 16.7 15.0 * 15.0 11.1 * 17.0 13.7 15.0 11.1 11.1 * * 20.7 12.6 23.7 20.7 * * * 16.2 18.4 15.0 0 19.6 19.6 23.5 21.3 25.6 * 100 * 15.0 11.1 50 % 16.2 11.1 12.6 0 50 100 0 50 100 0 50 100 0 50 100 0 50 100 % Figure X. Response of stage 1 zoe... 113 Growth Intermoult duration decreased with increasing temperature at each instar (N=3; P<0.0001; Fig. 5). Larvae reared at 10.5°C failed to develop past zoea 2 and mean duration of the first zoeal stage was 32 days (± 0.33; N=3). Most rapid development through to megalopa was 41 days (± 0.41; N=3) at 20.2°C. The plot of log time against temperature was not linear which indicates that the decrease in intermoult duration with increased temperature is not simply exponential as occurs in chemical systems by van’t Hoff’s rule (Hayes, 1949; Garside, 1966). The more rapid development of larvae at higher temperature appeared to be at the expense of somatic growth as dry weight of megalopas declined significantly with increasing temperature (P<0.01; Fig. 6). Figure 5. Effect of temperature on the timing of moulting of the planktonic larval stages of Pseudocarcinus gigas. Zoeal stages listed are the stages that zoeas moult into. Shift upwards indicates delay in the development of larvae. Missing points are due to complete mortality. Values are the means of three replicates. 114 Figure 6. Effect of temperature on dry weight of megalopas. Values are means of individual larvae pooled from all replicates. Labels next to the means are N. The regression was derived from raw data rather than plotted means. Mean weight (mg ± SD) 2.25 2 3 1.75 3 8 12 8 1.5 14 7 15 1.25 6 5 1 0.75 12 14 16 18 Temperature °C 20 22 FIGURE X. Effect ..n dry weight of megalopa.. Survival Survival increased with higher temperatures (P<0.001) so that best survival was obtained with treatments 15.8-20.2°C (Fig. 7). Most mortality occurred at zoea 1 and at megalopa (Fig. 8). Although survival to megalopa was low (24%), the proportion of viable megalopas (those alive after 24 h) appeared to be affected significantly by temperature. The largest unsigned difference in the Kolmogorov-Smirnov test was at 15.8°C which indicates that viability of megalopas was highest at this temperature (Fig. 9). Mean Survival (instars ± SE, n=3) Figure 7. Effect of temperature on survival of larvae, measured by survival to each instar with censoring of viable megalopas. Replicates in temperatures 14.8 and 19.4 suffered atypical mass mortality so the values plotted are drawn from single replicates. The regression was derived from raw data, rather than plotted means. 5 4 3 2 1 y = -13.750 + 1.927x - 0.052x 2 r2 = 0.7320 0 10 12 14 16 18 Temperature °C 20 22 FIGURE X. Ef... on survival ... 115 Figure 8. Patterns of mortality in larval rearing at different temperatures; temperatures were paired and survival averaged to show general trends. "M" is number of stage 5 zoeas moulting to megalopa, "M+" is megalopas active after 1 day. Figure 9. Viability of larvae reared at different temperatures. Megalopas often died after moulting, those that moulted successfully and were active after 1 day were scored as viable. Larvae did not survive to megalopa in all treatments (n total megalopas per treatment at base of column). 8 100 6 75 4 50 2 25 0 0 0 2 8 4 12 7 12 14 6 15 5 Percentage Frequency Frequency % Viable 0 10.5 11.7 12.8 13.8 14.8 15.8 16.8 17.8 18.6 19.4 20.2 21.1 Temperature °C Discussion FIGURE X. Effect of temperature on viability Behavioural responses to temperature Sulkin (1984) described the mechanism for vertical migration of brachyuran larvae as a combination between cues to orientate swimming, and cues to influence locomotor responses, either through passive sinking or active swimming. The principal factor 116 influencing locomotion and orientation during daylight hours is light and there is potential for this to be important to Pseudocarcinus gigas larvae even at the depth of hatching, around 300 m (Beebe, 19346; Clarke, 1970; Levings et al., 1996). However, phototactic responses do not account for larval migration patterns during the night. In an early paper, Sulkin (1973) proposed that decapod larvae regulate depth in the absence of light by a combination of responses to hydrostatic pressure and gravity. Since then, the response of larvae to temperature has been studied in greater detail and it appears that thermokinesis is also critical in maintaining depth (Ott and Forward, 1976; Kelly et al., 1982). Pseudocarcinus gigas zoeas do not appear to respond to change in hydrostatic pressure (Gardner, 1996; Chapter 5) so the thermokinesis observed in these trials may be an alternative mechanism for regulating locomotion. Although it is not possible to speculate on the effect of temperature on the response of P. gigas larvae during daylight hours, the observed temperature response indicates that the vertical position of stage 1 and 2 zoeas at night would be maintained within water of 12.6 and 16.3°C. This implies that where the surface temperature is less than 16.3°C, P. gigas larvae will migrate to the surface. Behavioural responses to the thermocline The distribution of planktonic organisms in nature is often strongly influenced by the presence of thermoclines with organisms restricted to one side, or found in greatest abundance around this layer (Keifer and Kremer, 1981; Southward and Barrett, 1983; Harding et al., 1987). Various mechanisms have been proposed for this phenomenon including: aggregation around areas of greatest prey density associated with the nitrite maxima at the thermocline (Keifer and Kremer, 1981), an inability to penetrate density gradients (Harder, 1968), and a negative feedback system of depth regulation based on rates of temperature change (Forward, 1990; Boudreau et al., 1991). It is important to note that the last of these proposed mechanisms, response to rates of temperature change, involves larvae detecting changes in temperature in relation to time, rather than responses to absolute temperatures. Forward (1990) demonstrated that this response was exhibited by Rhithropanopeus harrisii, although no response was detected with stage IV zoeas of Neopanope sayi. The thermoclines resulting from the system designed by McConnaughey and Sulkin (1984), and reproduced here, are very abrupt and more extreme than that observed in the field where thermoclines of 5°C may occur through a depth of 5 meters (eg. Gray, 1996). Consequently, extrapolation from crab larval response in the experimental system to the natural environment may be flawed, especially where larval movement appears inhibited by the experimental thermocline. Experimental results indicating no effect are more readily transferred to modelling of the natural system - if the larvae do not respond to the abrupt temperature changes generated experimentally, then they are unlikely to be influenced by the more gentle changes occurring in nature. Stage 1 and 2 P. gigas zoeas were not influenced by the experimental thermoclines produced in this trial and this has also been reported for three other crab species, Geryon 6 The light at 230 m depth was described by William Beebe following the first bathysphere descent, 6th June, 1930 “We were the first living men to look out at the strange illumination: and it was stranger than any imagination could have conceived. It was of an indefinable transclucent blue quite unlike anything I have ever seen in the upper world, and it excited our optic nerves in a most confusing manner. We kept thinking and calling it brilliant, and again and again I picked up a book to read the type, only to find I could not tell the difference between a blank page and a coloured plate.... It actually seemed to me to have a brilliance and intensity which the sunshine lacked”. 117 quinquedens (Kelly et al., 1982), Eurypanopeus depressus (Sulkin et al., 1983), and Callinectes sapidus (McConnaughey and Sulkin, 1984). This absence of an inhibitory effect of thermoclines on zoeal movement suggests that there is no physical barrier to restrict larval movement. Rather, P. gigas larvae appeared to migrate vertically in response to absolute temperature so that they accumulated either above or below the thermocline. A similar response was observed with Callinectes sapidus where larvae failed to penetrate a thermocline only when the upper absolute temperature was extreme (McConnaughey and Sulkin, 1984). These observations indicate that the primary factor influencing crab larval accumulation around a thermocline is depth maintenance to avoid temperature extremes, while the magnitude of natural temperature gradients across thermoclines would not impede migration. Response to absolute temperatures may still result in accumulation of larvae on one side of a thermocline as reported with Homarus americanus (Harding et al., 1987). Grey (1996) reported that thermoclines formed in waters at the northern end of the range of P. gigas in New South Wales (McNeill, 1920) did not appear to influence the broad scale distribution of larval fishes. Similar results would be expected with P. gigas although absolute temperature profiles are likely to be critical for modelling larval distribution. Comparison between the effect of temperature on development and behaviour Depth maintenance is used by estuarine species to regulate dispersal (Epifanio et al., 1984; Cronin and Forward, 1986) but this role is unlikely to be important for oceanic species where suitable sites for settling are more widespread and current flow is not bidirectional with tides. In P. gigas, vertical migration behaviour in response to temperature is likely to be directed towards optimising physiology and this was supported by observations on larval development at different temperatures. The general pattern of a retarding effect of low temperature and accelerating effect of high temperature on intermoult duration was the same as that reported for numerous other decapod larvae (Anger, 1983; Sulkin and McKeen, 1994; Goncalves et al., 1995). The relationship between log time and larval instar was not linear so van’t Hoff’s rule for chemical reactions was not met (Johansen and Krough, 1914; Garside, 1966) and larval development almost ceased in the 10.5°C treatment with no larvae surviving beyond zoea II. This implies that the thermal requirements for growth of P. gigas larvae have an optimal range and are not entirely cumulative (degree days), but involve some component of threshold phenomena (Waddy and Aiken, 1996). No larvae survived to megalopa in temperatures <12.8°C which is in the order of the absolute temperature initiating upward migration of stage 1 and 2 zoeas (12.6 and 12.7°C respectively). While there appeared to be close similarity between vertical migration patterns and larval survival at lower temperatures, overall survival was relatively high in treatments above temperatures where downward vertical migration behaviour was induced (>16.2°C). Best overall survival, averaged across instars, appeared to be between 14.8 and 20.2°C (inclusive) although there is potential for this to be confounded by ontogenetic changes in thermal tolerance (Sulkin and McKeen, 1989; Rasmussen and Tande, 1995) which was not possible to assess in the trial with P. gigas. Despite the high overall survival (cumulative for all instars) in treatments greater than 16.8°C, few larvae moulted successfully to viable megalopas and this is similar to the 118 situation where downward vertical migration of stage 1 and 2 zoeas was initiated (16.2 and 16.3°C respectively). Larval response to temperature is known to change during development in other species (Forward, 1990) so the behavioural response of stage 1 and 2 larvae is not directly applicable to the final zoeal stage. Nonetheless, it is noteworthy that the optimal treatment for viability of stage 5 zoeas moulting to megalopa (15.8°C) was within the range that stage 1 and 2 larvae would migrate to in the absence of light and was the control temperature with uniform larval distribution. Sulkin and McKeen (1989) considered the final zoeal stage of Cancer magister to be the most sensitive to temperature stress and P. gigas appears to be similar as mortality was highest at this stage. Incidence of deformity increases at extreme high temperature in most organisms (Battle, 1930) and this may have accounted for the low survival to megalopa which is typically reduced by high temperatures (Minagawa; 1990; Chaoshu and Shaojing, 1992; Okamoto, 1993). Johns (1981) noted that larval size of Cancer irroratus was greatest in the mid-range of thermal tolerance although larval weight more typically declines with increased temperature as was observed with P. gigas (Shirley et al., 1987; Minagawa, 1990; Sulkin and McKeen, 1994). Low weight of larvae reared at higher temperatures may be indicative of reduced energy stores (Minagawa, 1990) and this could also have contributed to the poor survival to megalopa in treatments greater than 16.8°C. Implications of temperature on distribution of Pseudocarcinus gigas Very few P. gigas larvae have been obtained from plankton tows (Chapter 4) so there is insufficient information to assess distribution and development of larvae in nature. Consequently, studies of larvae in the laboratory provide the best indications of the effect of temperature on development of P. gigas although compounding factors such egg incubation temperature will inevitably influence results (Laughlin and French, 1989). Several studies on brachyurans have demonstrated close relationships between laboratory results and observations from plankton sampling as temperature has a profound effect on development (Anger, 1983; Shirley et al., 1987). Factors limiting the distribution of a species fall into three groups: the introduction or historical presence of a species in a location; abiotic factors; and biotic factors. Kinne (1963) considered that temperature was the principal abiotic factor influencing distribution and it also influences biotic factors by increasing exposure to predation when instar duration is extended (Jamieson and Armstrong, 1991). Given the low survival of P. gigas larvae in treatments less than 14°C, it is likely that zoeal larvae have a plankton period of less than 3 months and hence there may be little potential for long distance dispersal (Thorson, 1961). P. gigas has not colonised nearby New Zealand (McLay,1988) while other species with longer larval duration, such as Jasus edwardsii (1224 month larval duration; Phillips and Sastry, 1980), are found in both locations. Low survival of P. gigas larvae in treatments below 14°C would also indicate low survival in southern Tasmania which is confirmed by the fishery, as only occasional crabs are caught in this region. Likewise, poor survival to megalopa in treatments above 16.8°C would limit the northern range of the species explaining the rarity of specimens from the state of New South Wales (most northerly record, 34°25'S; McNeill, 1920). References 119 Anger, K. 1983. Temperature and the larval development of Hyas araneus L. (Decapoda: Majidae); extrapolation of laboratory data to field conditions. J. Exp. Mar. Biol. Ecol. 69(3), 203-215. Battle, H.I. 1930. Effects of extreme temperatures and salinities on the development of Enchelyopus cimbrius (L.). Contrib. Can. Biol. Fisheries 5, 109-192. Beebe, W., 1934. Half Mile Down, Harcourt Brace, New York. Boudreau, B., Simard, Y. and Bourget, E. 1991. Behavioural responses of the planktonic stages of the American lobster Homarus americanus to thermal gradients, and ecological implications. Mar. Ecol. Prog. Ser. 76(1), 13-23. Chaoshu, Z. and Shaojing, L. 1992. Effects of temperature on survival and development of the larvae of Scylla serrata. J. Fish. China 16(3), 58-66. Clarke, G.L. 1970. Light conditions in the sea in relation to the diurnal vertical migration patterns of animals. In: G. B. Farquhar (Ed.), Proceedings of an international symposium on biological sound scattering in the ocean. Maury Center for Ocean Science, Department of the Navy, Washington D.C., pp. 41-50. Cronin, T.W. and Forward, R.B.Jr. 1986. Vertical migration cycles of crab larvae and their role in larval dispersal. Bull. Mar. Sci. 39(2), 192-201. Edwards, P. and Van Baalen, C. 1970. An apparatus for the culture of benthic marine algae under varying regimes of temperature and light intensity. Bot. Mar. 13, 42-43. Epifanio, C.E., Valenti, C.C. and Pembroke, A.E. 1984. Dispersal and recruitment of blue crab larvae in Delaware Bay, U.S.A. Estuar. Coast. Shelf Sci. 18, 1-12. Forward, R.B.Jr. 1990. Behavioural responses of crustacean larvae to rates of temperature change. Biol. Bull. 178(3), 195-204. Gardner, N.C. 1996. Behavioural basis of depth regulation in the first zoeal stage of the giant crab (Pseudocarcinus gigas, Brachyura, Xanthoidea, Oziidae). High latitude crabs: biology, management, and economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska, Fairbanks, pp. 229253. Garside, E.T. 1966. Effects of oxygen in relation to temperature on the development of embryos of brook trout and rainbow trout. J. Fisheries Res. Board Can. 23(8), 1121-1134. Goncalves, F., Ribeiro, R. and Soares, A.M.V.M. 1995. Laboratory study of effects of temperature and salinity on survival and larval development of a population of Rhithropanopeus harrisii from the Mondego River estuary, Portugal. Mar. Biol. 121(4), 639-645. Gray, C.A. 1996. Do thermoclines explain the vertical distributions of larval fishes in the dynamic coastal waters of south-eastern Australia? Mar. Freshwater Res. 47(2), 183-190. Haney, J.F. 1988. Diel patterns of zooplankton behaviour. Bull. Mar. Sci. 43(3), 583-603. Harder, W. 1968. Reactions of planktonic organisms to water stratification. Limnol. Oceanogr. 13, 156-168. Harding, G.C., Pringle, J.D., Vass, W.P., Pearre, S.Jr. and Smith, S.J. 1987. Vertical distribution and daily movements of larval lobsters Homarus americanus over Browns Bank, Nova Scotia. Mar. Ecol. Prog. Ser. 41(1), 29-41. Hayes, F.R. 1949. The growth, general chemistry, and temperature relations of salmonid eggs. Quart. Rev. Biol. 24(4), 281-308. Jamieson, G.S. and Armstrong, D.A. 1991. Spatial and temporal recruitment patterns of Dungeness crab in the northeast Pacific. Mem. Qld. Mus. 31, 365-381. 120 Johansen, A.C. and Krough, A. 1914. The influence of temperature and certain other factors upon the rate of development of the eggs of fishes. Conseil Perm. Int. Exploration de la Mer, Public. de Circonstance 68, 144. Johns, D.M. 1981. Physiological studies on Cancer irroratus larvae. I. Effects of temperature and salinity on survival, development rate and size. Mar. Ecol. Prog. Ser. 5(1), 75-83. Kelly, P., Sulkin, S.D. and Van Heukelem, W.F. 1982. A dispersal model for larvae of the deep sea red crab Geryon quinquedens based upon behavioural regulation of vertical migration in the hatching stage. Mar. Biol. 72, 35-43. Kiefer, D.A.and Kremer, J.N. 1981. Origins of vertical patterns of phytoplankton and nutrients in the temperate, open ocean: a stratigraphic hypothesis. Deep Sea Res. 28A(10), 1087-1105. Kinne, O. 1963. The effects of temperature and salinity on marine and brackish water animals. I. Temperature. Oceanogr. Mar. Biol. Ann. Rev. 1, 301-340. Laughlin, R.B.Jr. and French, W. 1989. Interactions between temperature and salinity during brooding on subsequent zoeal development of the mud crab Rhithropanopeus harrisii. Mar. Biol. 102(3), 377-386. Levings, A., Mitchell, B.D., Heeren, T., Austin, C. and Matheson, J. 1996. Fisheries biology of the giant crab (Pseudocarcinus gigas, Brachyura, Oziidae) in southern Australia. High latitude crabs: biology, management, and economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska Fairbanks, pp. 125-151. McConnaughey, R.A. and Sulkin, S.D. 1984. Measuring the effects of thermoclines on the vertical migration of larvae of Callinectes sapidus (Brachyura: Portunidae) in the laboratory. Mar. Biol. 81(2), 139-145. McLay, C.L. 1988. Brachyura and crab-like Anomura of New Zealand. Leigh Lab. Bull. 22 (1-4), 1-463. McNeill, F.A. 1920. Studies in Australian carcinology, no. 1. Rec. Aust. Mus. 13, 108-109. Miller, R.G.Jr. 1981. Survival Analysis. John Wiley and Sons, N.Y., pp. 238. Minagawa, M. 1990. Influence of temperature on survival, feeding and development of larvae of the red frog crab, Ranina ranina (Crustacea, Decapoda, Raninidae). Nippon Suisan Gakkaishi 56(5), 755-760. Okamoto, K. 1993. Influence of temperature on survival and growth of larvae of the giant spider crab, Macrocheira kaempferi (Crustacea, Decapoda, Majidae). Nippon Suisan Gakkaishi 59(3), 419-424. Ott, F.S. and Forward, R.B.Jr. 1976. The effect of temperature on phototaxis and geotaxis by larvae of the crab Rhithropanopeus harrisii (Gould). J. Exp. Mar. Biol. Ecol. 23, 97-107. Phillips, B.F. and Sastry, A.N. 1980. Larval ecology. In: J.S. Cobb and B.F. Phillips (Eds.), The biology and management of lobsters, Vol. II. ecology and management. Academic Press, New York, pp. 11-57. Rasmussen, T. and Tande, K. 1995. Temperature-dependent development, growth and mortality in larvae of the deep-water prawn Pandalus borealis reared in the laboratory. Mar. Ecol. Prog. Ser. 118(1-3), 149-157. Shirley, S.M., Shirley, T.C. and Rice, S.D. 1987. Latitudinal variation in the Dungeness crab, Cancer zoeal morphology explained by incubation temperature. Mar. Biol. 95(3), 371-376. magister: Sokal, R.R. and Rohlf, F.J. 1981. Biometry, second ed. W.H. Freeman, New York, pp. 859. Southward, A.I. and Barrett, R.L. 1983. Observations on the vertical distribution of zooplankton including post-larval teleosts off Plymouth in the presence of a thermocline and a chlorophyll-dense layer. J. Plank. Res. 5(4), 599-618. 121 Sulkin, S.D. 1973. Depth regulation of crab larvae in the absence of light. J. Exp. Mar. Biol. Ecol. 13, 73-82. Sulkin, S.D. 1984. Behavioural basis of depth regulation in the larvae of brachyuran crabs. Mar. Ecol. Prog. Ser. 15(1-2), 181-205. Sulkin, S.D. and McKeen, G.L. 1989. Laboratory study of survival and duration of individual zoeal stages as a function of temperature in the brachyuran crab Cancer magister. Mar. Biol. 103(1), 31-37. Sulkin, S.D. and McKeen, G. 1994. Influence of temperature on larval development of four co-occurring species of the brachyuran genus Cancer. Mar. Biol. 118(4), 593-600. Sulkin, S.D., Van Heukelem, W. and Kelly, W. 1983. Behavioural basis of depth regulation in the hatching and post-larval stage of the mud crab Eurypanopeus depressus. Mar. Ecol. Prog. Ser. 11(2), 157-164. Thorson, G. 1961. Length of pelagic larval life in marine bottom invertebrates as related to larval transport by ocean currents. V. Pbl. Amer. Assoc. Advancement Sci. 67, 455-474. Waddy, S.L. and Aiken, D.E. 1996. Temperature control of recruitment in the American lobster. J. Shellfish Res. 15(2), 495. Yule, A.B. 1984. The effect of temperature on the swimming activity of barnacle nauplii. Mar. Biol. Lett. 5(1), 1-11. 122 Effect of photoperiod and light intensity on larval survival, development and cannibalism 7 Research for this chapter has been previously published as: Gardner, C. and Maguire, G.B., in press. Effect of photoperiod and light intensity on survival, development and cannibalism of larvae of the Australian giant crab Pseudocarcinus gigas (Lamarck). Aquaculture. 123 Abstract Pseudocarcinus gigas larvae were reared to megalopa under two light intensities (2 and 500 lux) in five photoperiod regimes (0, 6, 12, 18, and 24 h light). Survival was not significantly affected by photoperiod or light intensity (P>0.05) although other effects were observed which are discussed in relation to swimming activity and feeding. Larvae had shorter intermoult duration in treatments with longer photoperiods and brighter light with most rapid development to megalopa in the continuous light, 500 lux treatment (49.2 d). Size (measured as telson width) of stage 4 zoeas was affected by photoperiod with smallest zoeas in the continuously dark treatment, whereas all other treatments were similar. Cannibalism was strongly influenced by lighting with greater damage to the dorsal spine occurring with increasing photoperiod and also in dimmer (2 lux) treatments. Lowest incidence of cannibalism was observed in continuous darkness. Viability of larvae after metamorphosis to megalopa was variable and no treatment effect was observed although viability was lowest in the two continuous light treatments. Results of this trial indicate that continuous light or dark regimes should be avoided. Optimal light intensity for culture was less clear and is discussed in relation to intermoult duration and cannibalism. Introduction In an attempt to optimise larval rearing of Pseudocarcinus gigas (Lamarck) the effects of photoperiod and light intensity were assessed. Light is one of the major factors influencing swimming and feeding behaviour in decapod larvae (Sulkin, 1984; Minagawa and Murano, 1993) and the effect of photoperiod on larval survival has been studied in several species (e.g. Sandoz and Rogers, 1944; Dalley, 1980; Radhakrishnan and Vijayakumaran, 1986; Nakanishi, 1987; Minagawa, 1994). Changes in survival and development of decapod larvae in different lighting regimes are often simply attributed to effects on feeding. However, other aspects of behaviour and physiology may be influenced by lighting including: swimming activity and thus metabolism (Gardner, 1996); cannibalism (Hecht and Pienaar, 1993); entrainment of physiological processes to circadian cycles (Dalley, 1980); initiation of ecdysis (Waddy and Aiken, 1991); and endocrine control of metamorphosis (Eagles et al., 1986). Due to the potentially complex nature of effects of light on larval development, research on this topic tends to involve measurement of a wide range of responses (Aiken et al., 1981; Minagawa, 1994). This study examined the effect of photoperiod and light intensity on larval survival, intermoult duration, size, cannibalism, and metamorphosis to megalopa. Results are directly applicable to culture of P. gigas but information of potential relevance to decapod culture in general was also gained. As noted by Minagawa (1994), the effects of continuous light treatments has seldom been examined. The effect of light intensity on larval development has also received little attention despite the importance of intensity on swimming behaviour (Sulkin, 1984). Materials and methods Source of larvae Ovigerous females were collected from depths of 300 – 380 m off the east coast of Tasmania, Australia (41°17'S; 148°40'E) in June 1995. Females ranged in size from 2.2 124 – 3.5 kg and were held communally in 4 m3 tanks with flow through, unfiltered, seawater. Temperature in broodstock tanks ranged from 8 to 14°C and the lighting regime was approximately 10 hours light per day. Larvae were collected at dusk from two tanks to ensure that larvae were not from a single parent; further mixing probably occurred as several females were releasing larvae in each tank. Culture methods and experimental design Newly hatched larvae were rinsed in 0.2 µm filtered seawater (32‰ salinity) and 100 were transferred to each of 36 (9 treatments x 4 replicates), 1.8 l, black, rectangular static culture vessels. Zoeas were maintained in a temperature control room at 15.5°C and were fed a mix of Protein Selco™ enriched rotifers (Brachionus plicatilis) and instar II artemia nauplii for the first two instars and artemia only thereafter. Larvae were pipetted into fresh, 0.2 µm filtered seawater daily (32-34‰ salinity). Although four replicates were used for all treatments initially, this was reduced to three in some treatments by high mortality not associated with treatment effects (chlorine contamination of one replicate in both the 24 L, 500 lux and 18 L, 500 lux treatments). Consequently, 1 replicate was randomly deleted from the remaining treatments to produce 3 replicates for all treatments. Incidence of cannibalism was assessed on stage 1 and 2 zoeas in all four replicates. The trial ran for 99 days after which all live animals were counted and censored in survival analyses. Larvae were cultured under five photoperiod regimes with light phases of 0, 6, 12, 18, and 24 h per day (hereafter: 0, 6, 12, 18, and 24 L). Daily transfer of larvae to new culture vessels took around 10 min per container; therefore 0 L treatments actually received a short light phase. With the exception of the 0 L treatment, the photoperiod regimes were further divided into 2 light intensity treatments: 2 and 500 lux. Lighting in the 500 lux treatments was by 12 V, 20 W quartz halogen globes suspended 1.5 m over the culture vessels. The 2 lux treatments were conducted in separate compartments next to the 500 lux treatments with small gaps in the adjoining wall. These gaps allowed light to pass through which was then reflected onto the culture vessels. Gaps in the adjoining walls also allowed circulation of air to reduce potential temperature difference between treatments. Temperature was monitored between different treatments and no effect of warming from the light source was detectable at a resolution of 0.1°C. Light intensity was recorded with a Profisix™ lux meter. Response data collected The effects of photoperiod and light intensity were monitored daily by recording the number and instar of exuviae and mortalities to calculate survival and time to each moult. Larval size was measured on exuvia of stage 4 zoeas as treatment effects should be more readily apparent than in earlier stages. Only exuvia were measured to ensure the sample included only healthy, actively moulting zoeas. The carapace splits during ecdysis so it was impossible to measure carapace length on exuvia. Consequently, size was measured as telson width between the inner angles of the lateral spines on exuvia (Fig. 1). All measurements were made by image analysis using NIH-Image™ 1.6 software. Extent of cannibalism was measured by a semi-quantitative scale modified from Minagawa (1994) by ranking damage to the dorsal spine into 4 categories (Fig. 2). Only stage 1 and 2 zoeas were assessed for cannibalism as spine regeneration by moulting may have obscured patterns in later stage larvae. 125 Most of the mortality of P. gigas larvae occurred within a short period after the moult to megalopa. The effect of treatments on megalopa viability was tested by rating any megalopas which survived for a period of longer than 1 day as viable. Figure 1. Larval size was measured on exuvia as telson width, measured between the inner angles of the lateral spines. 126 Figure 2. Cannibalism was measured by a semi-quantitative scale based on damage to the dorsal spine. Undamaged spines were ranked 0 and increasing damage was ranked 1-3 with increasing damage. Statistical analysis Treatment effects were analysed in three ways to incorporate the 0 L treatment which could be classed as neither bright (500 lux) nor dim (2 lux) intensity. First, the 0 L treatment was excluded and the effects of photoperiod, intensity and interactions were assessed for the 6, 12, 18 and 24 L treatments. The effect of photoperiod only was then assessed by including the 0 L treatment and analysing data from the bright and dim intensity treatments separately. All statistical analyses were performed with JMP 3.0™ software (SAS Institute). Development of P. gigas megalopas to crab 1 has been accomplished when ongrown artemia were supplied (Gardner and Gardner, 1996). No ongrown artemia were supplied in this trial which may have affected megalopa survival, so only mortalities up to the final zoeal stage (5) were included in survival analyses. The significance of the effects of photoperiod and light intensity on survival were tested with the semiparametric, Cox’s proportional-hazards model (Miller, 1981) as it was not possible to model accurately the hazard function parametrically. Data for survival analyses were obtained by recording the instar of all mortalities throughout the experiment. Survival to each zoeal stage, rather than through time, was then assessed by Cox’s proportionalhazards model with censoring of megalopa. The effects of light intensity and photoperiod on moult timing were tested by repeated measures analyses with significance determined with Wilk’s lambda (Mardia et al., 1979). Effects of treatments on size of stage 4 zoeas, cannibalism of stage 1 and 2 zoeas, the proportion of megalopas that were viable, and survival to stage 5 zoea were 127 tested by two-way ANOVA. Where a treatment effect was observed, means were compared by Tukey Kramer HSD (Sokal and Rohlf, 1995). Results Survival rate Survival was assessed in two ways: by Cox’s proportional-hazards to assess the survival of larvae throughout development (Fig. 3); and by two way ANOVA to test the effects of treatment on survival to the final zoeal stage (stage 5; Table 1). When survival was assessed throughout development, there was considerable variation between treatments although there was no significant effect of photoperiod or light intensity (Fig. 3; P>0.05; Table 1). Survival of larvae throughout development was not significantly different (P>0.05) between the two extreme photoperiods: 0 and 24 L. Survival of larvae through to the final zoeal stage is presented in Table 1 and there was no significant effect of photoperiod or intensity on survival although survival tended to be higher in low intensity treatments (P>0.05). Survival to megalopa was low so information presented in Table 1 is intended to provide general information on survival and no statistical analysis is presented. Table 1. Percentage survival (± SE; n=3) of larvae to the final zoeal stage (Z5) and to megalopa in each of the lighting treatments. Survival to the final zoeal stage was not significantly affected by either photoperiod or light intensity (P>0.05). Photoperiod (h) Intensity (lux) Z5 Megalopa 50.7 (13.7) 5.0 (1) 2 52.3 (25.7) 7.7 (7.2) 500 25.3 (18.4) 3.3 (1.7) 2 70.0 (4.2) 9.3 (8.3) 500 2 .0 (1.5) 1.3 (0.9) 2 44.3 (7.5) 9.0 (3.5) 500 64.7 (15.2) 10.7 (5.3) 2 74.3 (1.5) 10.3 (6.0) 500 47.7 (11.3) 7.0 (3.1) 0 6 12 18 24 128 Survival (mean instars) Figure 3. Mean survival (instars ± s.e., n=3) of Pseudocarcinus gigas larvae cultured under different photoperiods and light intensities. Larval development involves 5 zoeal stages, therefore, a mean survival of 5 instars implies that all larvae survived through the zoeal stages. 5 4 0 lx 3 2 lx 2 500 lx 1 0 0 6 12 18 Photoperiod (h) 24 Intermoult period and accumulated zoeal duration Intermoult period was significantly affected by both photoperiod and light intensity (P<0.05; Fig. 4). Larvae cultured under 2 lux lighting tended to have longer intermoult period than 500 lux treatments and there was a general trend of increasing intermoult period with decreasing photoperiod. The interaction term between photoperiod and intensity was also significant (P<0.05). This relationship was influenced by the 12 L treatment and was no longer apparent when this treatment was excluded (survival was lower in the 12 L 500 lux treatment than in any other treatment). The most rapid development through the zoeal phases to megalopa was in the 500 lux 24 L treatment (mean = 49.2 d). Relative to larvae in the 500 lux 24 L treatment, dimmer light intensity appeared to delay development by around 4 days (mean = 53.1 d; 2 lux, 24 L) and shorter photoperiod delayed development by about 10 days (mean = 59.6 d; 0 L). Figure 4. Effect of photoperiod and light intensity on duration of intermoult period in Pseudocarcinus gigas larvae. Means from larvae reared at high light intensity (500 lux) are joined by solid lines; low light (2 lux) by dotted lines. A shift towards the right implies an increase in intermoult period. Paste picture here (printed at end of document) 129 Size of stage 4 zoeas Larval size did not appear to be affected by lighting intensity (P>0.05). There was a significant effect of photoperiod on larval size (P<0.05; Fig. 5) with larvae cultured under 0 L being significantly smaller than in all other treatments. Mean size of larvae in the 6, 12, 18, and 24 L treatments were similar and not significantly different (P>0.05). There was no significant interaction between intensity and photoperiod (P>0.05). Mean telson width (µm± s.e.) Figure 5. Effect of photoperiod on size of stage 4 Pseudocarcinus gigas zoeas, measured by width of telson. Light intensity appeared to have no effect (P<0.05) so treatments were combined. Consequently, the mean was derived from 3 replicates for the complete darkness treatment (0 h light/d) and from 6 replicates for all other treatments. 840 830 820 810 800 790 0 6 12 18 Photoperiod (h/d) 24 130 Cannibalism Patterns of cannibalism were the same for both stage 1 and 2 zoeas although incidence increased with development (Fig. 6). Evidence of cannibalism was rarely observed in the 0 L treatment but increased significantly and proportionally with photoperiod (P<0.001). Larvae cultured under 2 lux lighting suffered dramatically higher cannibalism damage than those cultured at 500 lux (P<0.0001). There was no significant interaction between photoperiod and intensity (P>0.05). Figure 6. Effect of photoperiod and light intensity on extent of cannibalism in Pseudocarcinus gigas zoeas, measured by the extent of dorsal spine damage as "cannibalism index" (+s.e., n=4 replicates per treatment). Cannibalism index was measured on exuvia of stage 1 zoeas (upper) and stage 2 zoeas (lower). Cannibalism Index 1.5 Z1 1 0.5 0 Cannibalism Index 2.5 2 0 6 12 18 24 12 18 Photoperiod (h) 24 Z2 1.5 1 0.5 0 0 6 dark 2 lux 500 lux 131 Viability of megalopas Very few larvae reached megalopa in the 12 L, 500 lux treatment so these results were not included in analyses; in all other treatments at least 15 larvae moulted from the last zoeal phase in each replicate. There was large variation between treatments (range 1624%; Fig. 7) and there was no significant effect on the proportion of megalopas which were viable (P>0.05). Figure 7. Effect of photoperiod and light intensity on the viability of Pseudocarcinus gigas megalopas. Larvae which died during moulting to megalopa or which failed to survive for longer than 1 day were classed as unviable. All replicates include data from at least 15 individuals. No information is presented for the 12 h, 500 lux treatment as survival was too low to allow meaningful analysis. % of megalopae viable (+s.e., n =3) 60 50 40 30 20 10 0 0 6 dark 12 18 Photoperiod (h) 2 lux 24 500 lux Discussion Survival throughout the zoeal phases was not significantly affected by photoperiod yet within treatment variation was relatively small. This indicates that Pseudocarcinus gigas larvae are tolerant of various photoperiod regimes in culture. There was no significant effect of light intensity on survival although there appeared to be a trend of improved survival at low intensity. Only two treatments were examined and further research is required to clarify the effect of light intensity on larval survival, especially at higher intensity than that used in this trial. Although the effects of light on overall survival appeared to be minor, light intensity and photoperiod were observed to affect both growth of zoeas and incidence of cannibalism. It is probable that these effects resulted from changes in either feeding or swimming behaviours which are strongly influenced by light in P. gigas (Gardner, 1996). Effects of photoperiod due to behaviour Brachyuran larvae tend to migrate vertically in response to light by active upward swimming and passive sinking (Sulkin, 1984). In culture situations, this may result in larvae attempting to rise by actively swimming at the surface, or passively resting on the base in an attempt to migrate to greater depths. These behaviours become less apparent when larvae are mixed in a culture system such as plankton-kreisels (CharmantierDaures and Charmantier, 1991), although energetic costs of increased swimming may still affect growth. This increased activity in response to lighting is known to retard development in teleost larvae (Bolla and Holmefjord, 1988; Liu et al., 1994). Pseudocarcinus gigas larvae in this trial were considerably more active during dark 132 phases, as are Callinectes sapidus, due to negative geotaxis (Sulkin et al., 1979); this increased energy expenditure may have resulted in the smaller larvae in the 0 L treatment. Although there was no trend of increased size with longer photoperiod, the rate of development did tend to proceed faster with increased photoperiod which may be attributable to the expenditure of less energy during the dark phase. Passive sinking in response to light will cause larvae to accumulate at the base of culture vessels in relatively high density, which tends to cause higher cannibalism (Hecht and Pienaar, 1993). Larvae of P. gigas have been shown to exhibit different swimming patterns with changes between light of low (6-15 lux) and high (550-900 lux) intensity (Gardner, 1996) which may have contributed to the observed effect of intensity on cannibalism. Likewise, increased swimming during dark phases should result in less cannibalism in treatments with shorter photoperiods. Effects of photoperiod due to feeding An alternative hypothesis for the observed effects of lighting is that intensity and photoperiod influenced feeding activity. While total reliance on visual cues for feeding is common in teleosts (Miner and Stein, 1993; Hart et al., 1996), it has seldom been reported in decapod crustaceans. In one of the first studies of decapod rearing, Sandoz and Rogers (1944) found zoeal stages of C. sapidus starve when reared in 0 L. In most other decapod species, larvae are less severely affected by 0 L and are able to capture prey at a reduced rate (for example: Panulirus homarus (Radhakrishnan and Vijayakumaran, 1986), and Ranina ranina (Minagawa, 1994)), or they may actually feed at a rate equal to or greater than in light (for example: Homarus americanus (Templeman, 1936; Eagles et al., 1986), Pandalus borealis (Wienberg, 1982), and Paralithodes camtschaticus (Nakanishi, 1987)). Pseudocarcinus gigas zoeas appear to be able to feed in the absence of light as larvae in the 0 L treatment developed through to megalopa, albeit at a slower rate than larvae exposed to light. This implies a degree of feeding by either chemosensory detection or by random encounter, as starved controls of other experiments died at the first zoeal stage (Gardner and Northam, 1997). A pattern of feeding by random encounter in brachyuran larvae, akin to filter feeding, has been reported in other brachyurans (McConaugha et al., 1991; Minagawa and Takashima, 1994). While P. gigas are not obligate visual feeders, feeding appears to be enhanced by light; larvae cultured with light periods were larger than those cultured in 0 L and duration of intermoult tended to decrease with longer photoperiods. This is similar to patterns in P. homarus and R. ranina where consumption of artemia and growth was enhanced by lighting (Radhakrishnan and Vijayakumaran, 1986; Minagawa, 1994). Effect of light intensity on growth Light intensity affected duration of P. gigas larval intermoult with shorter intermoult periods at brighter intensity, possibly caused by increased feeding. However, the dimmer light intensity used in this study (2 lux) is at the extreme lower limit of the visual capacity of P. gigas (Gardner, 1996) so bright light (e.g. 500 lux or more) is not necessarily optimal for feeding. The effect of light intensity on larval feeding and development has seldom been examined in decapod crustaceans. Survival of larval Paralithodes camtschaticus is enhanced at intensities as high as 2000 lux (Nakanishi, 1987), which would be expected given that P. camtschaticus are known to migrate towards the surface during the day (Shirley and Shirley, 1987). In P. gigas, the diel 133 migration pattern and associated natural light intensities is less clear although it is likely that they migrate from the surface at day (Gardner, 1996). Cannibalism Use of visual cues for feeding may not only affect food consumption but also incidence of cannibalism. Minagawa (1994) found that increased cannibalism occurred with increased food consumption in R. ranina. Both photoperiod and light intensity are considered amongst the primary factors affecting cannibalism (Hecht and Pienaar, 1993) and they appeared to influence cannibalism of P. gigas. The observed effect of increased photoperiod on increased cannibalism in P. gigas is readily related to feeding: if larvae utilise visual cues to feed, longer photoperiods will provide greater opportunity to consume, or cannibalise, prey items. Two feeding-related hypotheses are suggested to explain the observed higher cannibalism at lower light intensity. First, feeding rate may be enhanced by dim lighting (2 lux) resulting in increased capture of artemia and also increased cannibalism. Running contrary to this hypothesis is the observed longer intermoult duration of zoeas reared in dim light than in bright light. However, slower growth may be expected as dorsal spine damage was greater and this is known to increase incidence of disease (Armstrong et al., 1976). Under this hypothesis, optimal lighting for P. gigas culture would be long photoperiods at dim intensity (e.g. 2 lux) to optimise food consumption with some form of mixing, such as aeration, to reduce cannibalism. A second hypothesis to explain the higher cannibalism in the 2 lux treatment is that a shift to larger prey may occur at dimmer light intensity. As the 2 lux treatment was at the extreme lower limit of the visual capacity of P. gigas (Gardner, 1996), visual detection of small prey items may have been impaired. Miner and Stein (1993) reported a shift of this nature in predation by larval finfish (Lepomis macrochirus) where larger prey were consumed at dimmer light intensities. Also, avoidance of predating zoeas may have been impaired. Under this hypothesis, optimal lighting for P. gigas culture would be long photoperiods at bright intensity (e.g. 500 lux) to optimise predation on artemia, rather than on zoeas. Performance of larvae in continuous light or dark regimes Development of P. gigas larvae in 24 L was relatively rapid but otherwise similar to larvae reared in light-dark regimes. This is similar to Paralithodes camtschaticus (Nakanishi, 1987) but contrasts to most other planktonic decapods where continuous lighting retards growth (Starkweather, 1976; Dalley, 1980; Radhakrishnan and Vijayakumaran, 1986; Minagawa, 1994). The hormonal pathways underlying metamorphosis to megalopa are not fully understood although a low titre of juvenile hormone is thought to be necessary (Christiansen, 1988). Low rates of metamorphosis have been observed in larval H. americanus and R. ranina reared in continuous darkness, despite relatively high growth, which suggests a neuroendocrine effect of continuous darkness (Eagles et al., 1986; Minagawa, 1994). Metamorphosis of P. gigas larvae did not appear to be affected by 24 or 0 L treatments, however this should be accepted cautiously as data were variable and lacked statistical power (Fig. 7). Given that mean metamorphosis was lowest in the two 24 L treatments, a cautious approach of avoiding continuous lighting is suggested. 134 Conclusion Although lighting did not appear to influence survival of P. gigas in this trial, there does appear to be potential to influence growth and cannibalism. Larvae were able to feed in continuous darkness and had low incidence of cannibalism, however, growth was slow and larvae were smaller than in treatments with light periods. Longer photoperiods resulted in more rapid growth and are recommended, although 24 L should be avoided as viability of megalopa may be affected. Larvae grew more rapidly and suffered less cannibalism at brighter intensity (500 lux). Consequently, brighter light appears preferable for culture although research is needed to clarify the effect of light intensity on feeding and feeding-related cannibalism. References Aiken, D.E., Martin, D.J., Meisner, J.D., Sochasky, J.B. 1981. Influence of photoperiod on survival and growth of larval American lobsters (Homarus americanus). Journal of the World Mariculture Society 12. 225-230. Armstrong, D.A., Buchanan, D.V., Caldwell, R.S. 1976. A mycosis caused by Langenidium sp. in laboratory reared larvae of the Dungeness crab, Cancer magister, and possible chemical treatments. Journal of Invertebrate Pathology 28. 329-336. Bolla, S., Holmefjord, I. 1988. Effect of temperature and light on development of Atlantic halibut larvae. Aquaculture 74. 355-358. Charmantier-Daures, M., Charmantier, G. 1991. Mass-culture of Cancer irroratus larvae (Crustacea, Decapoda): Adaptation of a flow-through sea-water system. Aquaculture 97. 25-39. Christiansen, M.E. 1988. Hormonal processes in decapod crustacean larvae. In: Fincham, A.A., Rainbow, P.S. (Eds.), Aspects of Decapod Crustacean Biology. Clarendon Press, Oxford, pp. 47-68. Dalley, R. 1980. The survival and development of the shrimp Crangon crangon (L.), reared in the laboratory under non-circadian light-dark cycles. Journal of Experimental Marine Biology and Ecology 47. 101-112. Eagles, M.D., Aiken, D.E., Waddy, S.L. 1986. Influence of light and food on larval American Lobsters, Homarus americanus. Canadian Journal of Fisheries and Aquatic Sciences 43. 2303-2310. Gardner, N.C. 1996. Behavioural basis of depth regulation in the first zoeal stage of the giant crab (Pseudocarcinus gigas, Brachyura, Xanthoidea, Oziidae). High Latitude Crabs: Biology, Management, and Economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska, Fairbanks, pp. 229-253. Gardner, C., Gardner, D. 1996. An improved method for on-growing artemia in batch culture. I. System design. Austasia Aquaculture 10(4). 62-63. Gardner, C., Northam, M. 1997. Prophylactic chemical treatments for mycosis of giant crab Pseudocarcinus gigas larvae in intensive culture. Aquaculture 158. 203-214. Hart, P.R., Hutchinson, W.G., Purser, G.J. 1996. Effects of photoperiod, temperature and salinity on hatchery-reared larvae of the greenback flounder (Rhombosolea tapirina Gunther, 1862). Aquaculture 144. 303-311. Hecht, T., Pienaar, A.G. 1993. A review of cannibalism and its implications in fish larviculture. Journal of the World Aquaculture Society 24. 246-261. Liu, H.W., Stickney, R.R., Dickhoff, W.W., McCaughran, D.A. 1994. Effects of environmental factors on egg development and hatching of Pacific Halibut Hippoglossus stenolepis. Journal of the World Aquaculture Society 25. 317321. Mardia, K.V., Kent, J.T., Bibby, J.M. 1979. Multivariate Analysis. Academic Press, New York. McConaugha, J.R., Tester, P.A., McConaugha, C.S. 1991. Feeding and growth in meroplanktonic larvae of Callinectes sapidus (Crustacea: Portunidae). In: Davie, P., Quinn, R.H. (Eds.), Proceedings of the 1990 International Crustacean Conference, September, 1991, Memoirs of the Queensland Museum 31. 320. Miller, R.G. Jr. 1981. Survival Analysis. John Wiley and Sons, New York. Minagawa, M. 1994. Effects of photoperiod on survival, feeding and development of larvae of the red frog crab, Ranina ranina. Aquaculture 120. 105-114. Minagawa, M., Murano, M. 1993. Larval feeding rhythms and food consumption by the red frog crab Ranina ranina (Decapoda, Raninidae) under laboratory conditions. Aquaculture 113. 251-260. Minagawa, M., Takashima, F. 1994. Developmental changes in larval mouthparts and foregut in the red frog crab, Ranina ranina (Decapoda: Raninidae). Aquaculture 126. 61-71. Miner, J.G., Stein, R.A. 1993. Interactive influence of turbidity and light on larval bluegill (Lepomis macrochirus) foraging. Canadian Journal of Fisheries and Aquatic Sciences 50. 781-788. Nakanishi, T. 1987. Rearing conditions of eggs, larvae and post-larvae of king crab. Bulletin Japan Sea Regional Fisheries Research Laboratory 37. 57-161 (in Japanese). Radhakrishnan, E.V., Vijayakumaran, M. 1986. Observations on the feeding and moulting of laboratory reared phyllosoma larvae of the spiny Lobster Panulirus homarus (Linnaeus) under different light regimes. Proceedings of the Symposium on Coastal Aquaculture, Cochin (4) 1261-1266. Sandoz, M., Rogers, R. 1944. The effect of environmental factors on hatching, moulting, and survival of zoea larvae of the blue crab Callinectes sapidus Rathbun. Ecology 25 (2). 216-228. 135 Shirley, T.C., Shirley, S.M. 1987. Diel vertical movements of Alaskan red king crab zoeae. American Zoologist 27(4). 103A. Sokal, R.R., Rohlf, F.J. 1995. Biometry, the Principles and Practice of Statistics in Biological Research. W.H. Freeman, New York. Starkweather, P.L. 1976. Influences of light regime on postembryonic development in two strains of Daphnia pulex. Limnology and Oceanography 21 (6). 830-837. Sulkin, S.D. 1984. Behavioural basis of depth regulation in the larvae of brachyuran crabs. Marine Ecology Progress Series 15. 181-205. Sulkin, S.D. Phillips, I., Van Heuklem, W., 1979. On the locomotory rhythm of brachyuran crab larvae and its significance in vertical migration. Marine Ecology Progress Series 1. 331-335. Templeman, W. 1936. The influence of temperature, salinity, light and food conditions on the survival and growth of the larvae of the lobster (Homarus americanus). Journal of the Biology Board of Canada 2 (5). 485-497. Waddy, S.L., Aiken, D.E. 1991. Scotophase regulation of the diel timing of the metamorphic moult in larval American lobsters, Homarus americanus. Journal of Shellfish Research 10. 287. Wienberg, R. 1982. Studies on the influence of temperature, salinity, light, and feeding rate on laboratory reared larvae of deep sea shrimp, Pandalus borealis Kroyer 1838. Meeresforschung 29. 136-153. 136 Z1-Z2 Z2-Z3 Z3-Z4 Z4-Z5 Z5-M Photoperiod (h) 0 6 12 18 24 0 10 20 30 40 50 Mean Days ± s.e. (n =3) 60 137 Use of Prophylactic Treatments for Larval Rearing 8 Research for this chapter has been previously published as: Gardner, C. and Northam, M., 1997. Use of Prophylactic Treatments for Larval Rearing of Giant Crabs Pseudocarcinus gigas (Lamarck). Aquaculture, 158: 203-214. 138 Abstract Chemicals were screened for prophylactic treatment of epibiotic fouling and fungal mycosis in larvae of giant crabs Pseudocarcinus gigas. The following treatments were applied as indefinite baths: oxytetracycline, trifluralin, carbendazim, copper oxychloride, malachite green, and formalin. Most effective treatments for improving survival were oxytetracycline (25 mg l-1, despite increased deformity), trifluralin (0.01 mg l-1), carbendazim (0.001 mg l-1), and copper oxychloride (0.05 mg l-1). Three of these treatments affected size and shape of the megalopa carapace with relatively smaller megalopas developing in carbendazim and trifluralin, and relatively broader megalopas in copper oxychloride. Toxic effects, measured by increased mortality, deformity, death during ecdysis, or delayed ecdysis, were recorded with oxytetracycline (≥25 mg l-1), trifluralin (≥0.03 mg l-1), malachite green (≥0.1 mg l-1), and formalin (for all concentrations tested: ≥2.5 mg l-1). Introduction Optimising survival during larval rearing of crustaceans is important in aquaculture and is also of benefit in research of crustacean taxonomy, physiology, and fisheries biology. Laboratory cultures of crab larvae often suffer severe mortality from disease, particularly from epibiotic bacteria and larval mycosis (Armstrong et al., 1976; Ebert et al., 1983; Hamasaki and Hatai, 1993). Larval mycosis occurs when fungal hyphae, commonly Lagenidium and Sirolpidium species (Brock and LeaMaster, 1992), invade body tissues, developing zoospore discharge tubes which protrude through the crustacean’s cuticle in the later stages (Fig. 1; Paynter, 1989). Infected larvae become immobilised and their surface develops a fouled appearance from the external processes of the fungi. Figure 1. Fouling diseases on giant crab zoeas. A stalked fungal sporangium is present in the center and filamentous bacterial fouling is visible as fine threads. In research involving culture of giant crab larvae, larval mycosis has consistently caused high mortality which prompted assessment of the various chemical controls available. The pelagic zoeal phase of P. gigas is relatively long, around 60 days, and larval mycosis occurs throughout this period, rather than just in early stages as is more typical in crustaceans (Brock and LeaMaster, 1992; Lightner, 1993). Consequently, prophylactic treatments must be administered for prolonged periods which can produce toxic effects, not evident in short term trials, such as delay in moulting (Caldwell et al., 1978) and deformity at megalopa (Ebert et al., 1983). 142 Prophylactic treatments are often assessed by separately establishing toxic levels of a given chemical for fungal zoospores and for the larvae to be treated (Armstrong et al., 1976; Lio-Po et al., 1982; Lio-Po and Sanvictores, 1986). While this is an invaluable technique for evaluating treatments rapidly, it can fail to assess the value of a treatment in culture situations. Adverse effects of the treatment may be underestimated because chronic toxicity can become apparent late in development and there may be interaction with bacterial species (Gil-Turnes and Fenical, 1992). In this chapter results are presented from prolonged clinical trials with giant crab larvae comparing several chemicals reported to immobilise fungal zoospores: malachite green (Armstrong et al., 1976); trifluralin (Armstrong et al., 1976); and formalin (Hamasaki and Hatai, 1993). In addition, trials were conducted with two horticultural fungicides with low toxicity to insects: carbendazim and copper oxychloride. Larvae were also treated with the antibiotic, oxytetracycline, to control pathogenic bacteria. Materials and methods Source of larvae Ovigerous females were collected from depths in the range of 300 to 380 m off the east coast of Tasmania (41°17'S; 148°40'E) in June 1995. Females ranged in size from 2.2 to 3.5 kg and were held communally in 4 m3 tanks with flow through, unfiltered water supply. To ensure that only freshly hatched larvae were used, the tanks were thoroughly flushed prior to collecting larvae for the trial. Although attempts were made to collect larvae from several females, most of the larvae used for this trial appeared to be from one female weighing 2.6 kg. Culture methods and experimental design Newly hatched larvae were rinsed in 0.2 µm filtered seawater (34‰ salinity) then transferred to 1.8 l vessels. Fed controls, starved controls, and chemical treatments were randomly allocated to vessels. Fifty larvae were placed in each vessel and were maintained in a temperature control room at 16°C with 700 lux, 8 l :16 d photoperiod. Zoeas were fed Protein Selco™ enriched artemia nauplii but megalopas required larger, 10 day old artemia. Larvae were pipetted into fresh, 0.2 µm filtered seawater every 2 days. The concentrations for each chemical tested to control fouling were: carbendazim (Hoechst and Schering™), malachite green, and trifluralin (DowElanco™) at 0.001, 0.003, 0.01, 0.03, and 0.1 mg kg-1; copper oxychloride (ChemSpray™) at 0.025, 0.05, 0.1, 0.2, and 0.4 mg kg-1; formalin at 2.5, 5, 10, 20, and 40 mg kg-1; and oxytetracycline (Norbrook™) at 10, 25, 50, 100, and 200 mg kg-1. Four replicates were used for each concentration. Treatments were run as indefinite baths for 115 days after which all live animals were censored in survival analyses. Response data collected The effect of different concentrations was monitored by recording the number and instar of exuvia and mortalities which allowed calculation of survival and time to each moult. All mortalities were also classed into possible causes of death: fouled (severe fouling of exoskeleton), moulting (died during ecdysis), deformed, and normal (where no cause was apparent). Larvae were classed as fouled when at least 50% of the external surface was covered. Epibiotic bacterial fouling increased rapidly after death, so the fouling 143 index only gives a general indication of disease. When dead larvae could be ascribed to more than one possible cause of mortality, such as moulting and fouled, each cause was counted and treated as independent. Although wet preparations and microbiology were performed, the cause of fouling was not determined in all cases; animals classed as fouled may have had fungal or bacterial infections. Microbiological cultures were made on TCBS, blood, and Ordal’s medium. In normal development, giant crab larvae pass through 5 zoeal stages (Gardner and Quintana, 1998; Chapter 3). In this study, many larvae developed to a sixth zoeal stage, which was intermediate between zoea 5 and megalopa (termed 5-a-lopae by Ebert et al., 1983; Fig. 2). Chelae were present and the pleopods bore setae, while the carapace retained the form of the zoea. None of these larvae survived and they were classed as deformed. The effect of treatments on size of dead megalopas or megalopa exuvia was assessed by measuring carapace length, carapace width, and rostrum width (Fig. 3). The shape of the rostrum varied between individuals so carapace length measurements were made slightly off centre. All measurements of megalopas were made by image analysis using NIH-Image™ 1.6 software. Figure 2. Giant crab larva intermediate between zoea 5 and megalopa. As with a megalopa, the chelae are differentiated and the pleopods are setose, however zoeal characters such as the dorsal carapace spine and bifurcated telson have been retained. These larvae were classed as deformed. 144 Figure 3. Morphological measurements taken to assess the effect of treatment on megalopa size. RW = rostrum width, CW = carapace width, CL = carapace length. Note that there was considerable variation in the form of the rostrum so CL was measured slightly to the left of center. Statistical analysis Statistical analysis was performed with JMP 3.0™ software (SAS Institute). Where comparisons between treatments and controls are presented, the controls are the fed group. The effect of concentration on moult timing was tested with repeated measures analysis. As survival was poor in some treatment replicates, analysis was restricted to data collected from the start of the trial to the moult from zoea 3 to zoea 4. Data were arranged for multivariate analysis and significance determined for between effects (concentration) with Wilk’s lambda (Mardia et al., 1979). Where a significant effect of concentration was found, comparisons were made between concentrations by analysis of the canonicals (Mardia et al., 1979). Survival data were analysed by the Kaplan-Meier method with significance between groups determined by Wilcoxon’s test (Miller, 1981). To prevent an increase in type 1 errors, a Bonferroni adjustment was made to alpha for comparisons between concentrations, so that comparisons were only treated as significant where P<0.003 (Sokal and Rohlf, 1995). The effect of treatments on size of megalopas and cause of mortality was assessed by one way ANOVA. Cause of mortality data were arc-sine square-root transformed to produce normality and remove heteroscedasticity. Where a treatment effect was observed, means were compared by Tukey Kramer HSD (Sokal and Rohlf, 1995). Percentage survival to crab 1 data are presented although low survival prevented meaningful statistical analysis. Results Larvae classed as fouled appeared to be afflicted with several types of infection. Peritrich ciliates were present in low numbers and were not regarded as pathogenic. Apart from larvae cultured in oxytetracycline, mixed bacterial flora consisting predominantly of Vibrio spp were isolated from all cases of fouling. Fouling appeared 145 to be predominantly from two causes: the terminal vesicles of fungal infections (tentatively identified as Lagenidium spp based on method of sporogenesis; Lightner, 1993) and filamentous bacterial infection (colourless, gram-negative and short cylindrical bacteria forming unbranched filaments, tentatively identified as family Leucothricheae; Lightner, 1993). In all chemical treatments and controls, some of the mortalities classed as fouled had larval mycosis. Larvae in starved controls remained alive for a mean time of 18.8 days although all were dead by day 20 (survival data from starved and other treatments are presented in Appendix 8). A small number of these successfully moulted to zoea 2 (1.5%). Larvae from both starved and fed controls had low level surface infections of mixed Vibrio spp. The proportion of larvae which died with high levels of fouling were significantly higher (P<0.001) in fed controls than in starved controls: 45.5% compared with 10.5%. Detrimental effects Some treatment concentrations had significantly lower survival than fed controls which is attributed to toxicity: 100 and 200 mg l-1 oxytetracycline, and 20 and 40 mg l-1 formalin (P<0.003; Figs. 4 and 5). Chronic toxicity was evident in 25 and 50 mg l-1 oxytetracycline treatments with significantly increased incidence of deformity (P<0.001). Both of these oxytetracycline treatments had significantly higher moulting mortality (P<0.001) as did treatments 0.03 and 0.1 mg l-1 trifluralin (P<0.01), and 10 mg l-1 formalin (P<0.05). Survival was also reduced in the 10 mg l-1 oxytetracycline treatment which was associated with significantly higher fouling relative to controls (P<0.0001; Figs. 4 and 5). This fouling was gelatinous, rather than filamentous as in other treatments, and was caused by a single Vibrio species which produced pale yellow colonies in TCBS media. Sub-lethal toxic effects were observed in the 100 mg l-1 oxytetracycline and 0.1 mg l-1 malachite green treatments as moulting was significantly delayed (Fig. 6; P<0.001). Consequently, the mean time to reach megalopa was delayed by 51 days and 13 days respectively. Enhanced survival Although toxic effects were observed at some concentrations, survival was enhanced by the best concentrations of all treatments relative to controls (P<0.003). This improved survival was associated with significantly lower fouling only in oxytetracycline, for concentrations ≥ 25 mg l-1 (Fig. 4; P<0.0001). In no other treatment was there a significant reduction in fouling to account for the improved survival although trends were apparent, such as in copper oxychloride and malachite green treatments. Because there was large variation in the incidence of fouling in controls, the experiment had low power to discern differences; as percentages, the mean incidence of fouling in controls was 45% while standard error was 20.1%. Ranking of the “best” concentrations of each treatment in increasing survival (quantified by the Kaplan-Meier method) was: oxytetracycline (50 mg l-1) > carbendazim (0.001 mg l-1) > trifluralin (0.1 mg l-1) > malachite green (0.03 mg l-1) > formalin (10 mg l-1) > copper oxychloride (0.05 mg l-1). There was no significant difference between treatments except between oxytetracycline and all other treatments (P<0.0001). Larvae survived to crab 1 in several treatments. Ranking of treatments for survival to crab 1 was: 25 mg l-1 oxytetracycline (7%) > 50 mg l-1 oxytetracycline (3.5%) > 0.4 mg 146 l-1 copper oxychloride (1.5%) > 0.05 mg l-1 copper oxychloride (1.0%) > 0.025 mg l-1 copper oxychloride, 0.001 and 0.003 mg l-1 carbendazim (all 0.5%). No larvae survived to crab 1 in controls or trifluralin, malachite green, and formalin treatments. 147 Figure 4. Effect of treatment concentration (mg l-1) on larval survival and cause of mortality. Mean survival for each concentration is the mean number of instars to which larvae survived, eg. mean survival of 2.9 implies that the mean survival was almost to Z3, while a mean of 1.1 implies that most larvae died at Z1 before moulting. Letters denote significantly different pairs of means (P<0.003). Symbols indicate major causes of mortality (accounting for >20% of total during moulting, >30% other causes) for each concentration: ♣ - moulting; ∇ - fouling; ♠ - deformity; ♦- normal. 148 Oxytetracycline 5 Mean Survival + S.E. 5 4 3 2 1 Mean Survival + S.E. 3 ♦ c a d d b ♦ a 0 10 25 50 100 200 Trifluralin ♦ ∇ a 0 5 ♦ ab ♦ ∇ b ♦ ∇ b ♣ ♦ ∇ ♣ ♦ a b 0 ♦ ∇ ♦ ♦ a a a a 0.001 0.003 0.01 0.03 0.1 Malachite Green b b 0.001 0.003 0.01 0.03 3 ♦ ∇ 2 1 a 0 0.1 Formalin 5 ♦ ∇ ♦ ♦ b a a ♦ ♦ b b 0.001 0.003 0.01 0.03 0.1 Copper Oxychloride 4 ♦ ∇ ♦ ∇ c d c ♦ ∇ ♣ d 0 2.5 5 10 2 1 1 5 4 3 ♦ ∇ ♣ 4 2 1 ♦ ∇ 2 4 3 ♦ ∇ 4 ∇ 5 Mean Survival+ S.E. ♠ ♦ ♣ ♦ ∇ ♠ ♣ Carbendazim ♦ ♣ 3 ♦ ∇ ♦ 2 ♦ b ♦ a 20 40 1 a 0 c ♦ ∇ ♦ c ab abc bc 0.025 0.05 0.1 ♦ 0.2 ♦ 0.4 149 Figure 5. Survival of larvae in oxytetracycline treatments to each larval stage. Z1...Z5 = zoea 1...zoea 5, M = megalopa, and C = crab 1. Mean survival is the mean number of instars that larvae survived to, eg. mean survival of 2.9 implies that the mean survival was almost to Z3. Superscripts denote significantly different pairs of means (P<0.003). Conc. Survival (% ± SE) 100 • xΘ∇♦ 80 • Θ 10 a 200a 100 b 0c 25d 50 d Mean Surv. 1.0 1.0 • Θ 1.5 2.1 4.1 4.9 • • Θ Θ 60 40 ∇ 20 • Θ ∇ ∇ 0 Z1 x♦ Z2 Z3 Z4 ∇ ∇ Z5 x 10 mg/l • 50 mg/l Θ 25 mg/l ∇ 100 mg/l M ♦ •∇ Θ C 200 mg/l 150 80 ♦ 20 • • • • • • ♦ ♦ ⊕ ⊕ ⊕ ⊕ • • • • 0 10 25 50 100 200 0 0.001 0.003 0.01 0.03 0.1 60 Days (± S.E.) ∗∗ 0 ⊕ ⊕ ⊕ ⊕ ⊕ 10 ⊕ ∇ ♦ ∇ - Z5 to megalopa × - megalopa to crab 1 ♦ ♦ ∗ ∇ 40 ♦ ∇ 20 ♦ ♦ ♦ ∗ 0 ∗ ∗∗ 40 ∗∗ ∗ 50 ∇ ∇ ∇ ∇ b. Oxytetracycline 60 ∇ 30 a. Malachite Green 70 × × • - Z1 to Z2 ⊕ - Z2 to Z3 ♦ - Z3 to Z4 ∗ - Z4 to Z5 80 ∇ 100 90 100 Figure 6, a & b. Effect of treatments on the timing of moults. Shift towards the right suggests that the treatment has delayed the development of the larvae. Points plotted for moult stages more advanced than the moult to Z4 should be interpreted with caution, as replication was frequently reduced in these later samples to less than 4. Treatment concentrations given on the X axes are in mg l-1. Missing points are due to complete mortality. Effects on megalopas Megalopa size and morphology was affected by treatment (Fig. 7; P<0.05). Megalopas from copper oxychloride, malachite green, and oxytetracycline treatments were larger than those from carbendazim and trifluralin treatments as measured by carapace area, carapace width, and carapace length. Opposite trends between the carapace width/carapace length and rostrum width/carapace width ratios were observed. This suggests that treatment type influences the shape of the carapace by increasing curvature or flexion of the lateral margins of the carapace. Megalopas from copper oxychloride and malachite green treatments were relatively broader. 151 "C Area" (mm2) ± SE Figure 7. Effect of treatments on size of megalopas. C = copper oxychloride, M = malachite green, Ca = carbendazim, O = oxytetracycline, T = trifluralin. Treatment had a significant effect (P<0.05) on size for all measures tested: "carapace area" (C Area = CW x CL); carapace width (CW); carapace length (CL); and the ratios between carapace width/carapace length (CW/CL) and rostrum width/carapace width (RW/CW). Higher mean values for CW/CL implies a relatively broader carapace and higher values for RW/CW implies a relatively broader rostrum. Labels next to means indicate significance of means comparisons. Formalin and control measurements were not included due to poor survival. Mean values are from all concentrations of each treatments. Larvae did not develop to megalopa in most replicates so the number of replicates contributing to the mean varied: C, n= 6; M, n=5; Ca, n=6; O, n=7; T, n=8. 8.5 •c 8 •c • bc 7.5 •ab 7 •a 6.5 C M Ca O T 3350 2550 •c •c CL (µm) ± SE CW (µm) ± SE 2650 • bc 2450 • ab 2350 •a 2250 C M Ca O • ab • ab 3050 •a 2950 T C •a M Ca O T 0.61 • c RW/CW ± SE CW/CL ± SE 0.82 3150 •b 2850 0.86 0.84 3250 •bc 0.8 • ab •a 0.78 0.76 C M Ca O •a T 0.59 •c • bc 0.57 0.55 • ab 0.53 0.51 •a •a C M 0.49 Ca O 152 T Discussion General observations on prophylactic treatments Larval P. gigas were successfully reared to juvenile crabs with prophylactic treatment. This has not been achieved in numerous other trials, even with high standards of hatchery hygiene and water quality. Both larval mycosis and filamentous bacteria disease were associated with mortality of larvae in this trial. Possible sources of fungal infection were split into three options by Ebert et al. (1983): parents; food source (artemia); and incoming water. No spores or bacteria should have entered this trial via the incoming water as filtration was absolute to 0.2 µm and Lagenidium spp. spores are 7 by 5 µm in size (Armstrong et al., 1976). The higher incidence of fouling in fed controls than in starved controls suggests that artemia contributed to disease, by acting as a source of infection or by contributing to water quality deterioration. Infection risk from artemia could be lowered by avoiding overfeeding and by prophylactic chemical treatment of artemia. Oxytetracycline The good performance of larvae cultured with oxytetracycline indicates the importance of bacterial disease on larval mortality. There are indications that oxytetracycline also reduced fungal infection because no cases of larval mycosis were found in 50, 100, and 200 mg l-1 oxytetracycline treatments. Possible explanations for this are: disease resistance improved as a result of general improvement in larval health (Anderson, 1989); possible deactivation of fungal zoospores; prevention of bacterial septicaemia following dorsal spine damage, which is known to increase risk of fungal infection (Ebert et al., 1983; Anderson, 1989); and an interaction between bacteria and fungal populations so that reduction of bacterial fauna resulted in a reduction in the incidence of fungal infection. In regards to the last hypothesis, the opposite trend is usually observed with bacterial competition preventing fungal infections (Gil-Turnes and Fenical, 1992) so that antibiotics, such as oxytetracycline, increase the risk of larval mycosis (Nogami and Maeda, 1992). Lightner (1993) suggested oxytetracycline in concentrations of 10 to 90 mg l-1 for use as indefinite baths to control Leucothrix mucor. For rearing giant crab larvae, this range is too broad. Bacteria proliferated at 10 mg l-1 and chronic toxicity resulted in high levels of deformity, similar to those reported by Ebert et al. (1983) with streptomycin sulphate, at all concentrations where larvae survived better than controls (≥25 mg l-1). Of the oxytetracycline concentrations tested, 25 mg l-1 would be most suitable due to lower chronic toxicity. The prophylactic use of oxytetracycline is clearly effective in culturing giant crab larvae although the widespread use of antibiotics is discouraged to prevent development of resistant strains (Anderson, 1989). Carbendazim and trifluralin Highest survival in prolonged fungicide treatments, without chronic toxic effects, was observed in carbendazim (0.001 mg l-1) and trifluralin (0.003 mg l-1). Both of these chemicals would be expected to degrade between water changes. Williams et al. (1986) reported the half-life of trifluralin in water could be as short as 30 minutes although the manufacturers, DowElanco, list the half-life as 6 days in aquatic systems. Survival of larvae may have been improved further by more frequent application of chemicals. 153 Despite the apparent lack of toxic effects at higher concentrations, highest survival in carbendazim treatments was at only 0.001 mg l-1. The apparent absence of toxic effects at all concentrations tested (up to 0.1 mg l-1) for carbendazim suggests that further trials are warranted. Although the implications on survival and growth are unclear, it is worth noting that megalopas from both carbendazim and trifluralin treatments were smaller than in other treatments. Toxic effects of trifluralin on larvae were observed at only 0.03 mg l-1 which is within the range suggested for prophylactic treatment of crustacean larvae by Lio-Po and Sanvictores (1986) and Lightner (1993). This discrepancy is likely to be due to the prolonged period of treatment needed with giant crab larvae. The maximum acceptable trifluralin concentration determined in this study was 0.01 mg l-1 as no toxic effects were observed. This concentration leaves some margin for toxicity to fungal zoospores as there was a significant improvement in survival at 0.003 mg l-1. Also, Armstrong et al. (1976) reported effective control of mycosis with trifluralin at 0.0015 mg l-1. Copper oxychloride, malachite green and formalin Copper oxychloride, malachite green and formalin treatments had lower survival than optimal concentrations of carbendazim and trifluralin although all significantly improved survival relative to controls. The action of copper oxychloride, malachite green and formalin is broad as they can also control epibiotic fouling bacteria and protozoa (Lightner, 1993). Consequently, this trial evaluated appropriate concentrations for prolonged treatment rather than the effect of treatments on specific pathogens. Copper oxychloride appears useful as a prophylactic as it improved survival and no toxic effects were observed within the range tested, up to 0.4 mg l-1. Although survival was less than in optimal concentrations of trifluralin or carbendazim, this difference was not significant (P>0.05). It is also worth noting that relatively large megalopas were produced in the copper oxychloride treatment, several of which developed to crab 1. Malachite green at 0.03 mg l-1 improved survival of giant crab larvae while higher concentrations delayed moulting. Malachite green has been used with crab eggs (Fisher, 1976) but was considered too toxic to zoeas to be a useful prophylactic treatment by Armstrong et al. (1976) and Fisher and Nelson (1977). Armstrong et al. (1976) found a Lagenidium sp. was resilient to concentrations of malachite green below 1 mg l-1 which is far greater than the concentration which produced chronic toxic effects in this study (0.1 mg l-1). Formalin appeared to be unsuitable for prolonged prophylactic treatment as moulting mortality was significantly higher (P<0.005) where survival was improved, indicating poorer health (Fisher, 1983). Short term formalin baths as used by Hamasaki and Hatai (1993) or Kaji et al. (1991) may be beneficial when infection occurs for only a short period. Conclusion The effect of chemicals on toxicity and survival were evaluated in this trial as prolonged prophylactic treatments. Best results were obtained with oxytetracycline (25 mg l-1) although use of antibiotics for commercial larval rearing is generally discouraged. Trifluralin (0.003 mg l-1) and carbendazim (0.001 mg l-1) improved survival and appear to be useful alternatives. Further trials with copper oxychloride and carbendazim may be warranted as survival was improved with no apparent toxic effects, even at high doses. Concentrating prophylactic treatment around critical periods, such as moulting, 154 may allow stronger concentrations to be used without causing toxic effects (Armstrong et al., 1976). References Anderson, I.G. 1989. Hatchery health problems and hygiene management. In: Invertebrates in Aquaculture. Proceedings of Refresher Course for Veterinarians, 19-21 May 1989, Brisbane. Postgraduate committee in veterinary science, University of Sydney, 117: 109-120. Armstrong, D.A., Buchanan, D.V. and Caldwell, R.S. 1976. A mycosis caused by Lagenidium sp. in laboratory reared larvae of the Dungeness crab, Cancer magister, and possible chemical treatments. J. Invertebr. Pathol., 28: 329-336. Brock, J.A. and LeaMaster, B. 1992. A look at the principal bacterial, fungal and parasitic diseases of farmed shrimp. In: J. Wyban (Editor), Proceedings of the Special Session on Shrimp Farming, AQ92. World Aquaculture Society, Baton Rouge, LA USA, 212-226. Caldwell, R.S., Armstrong, D.A., Buchanan, D.V., Mallon, M.H. and Millemann, R.E. 1978. Toxicity of the fungicide Captan to the Dungeness crab Cancer magister. Mar. Biol., 48: 11-17. Ebert, E.E., Haseltine, A.W., Houk, J.L. and Kelly, R.O. 1983. Laboratory cultivation of the Dungeness crab, Cancer magister. In: P.W. Wild and R.N. Tasto (Editors), Life history, environment, and mariculture studies of the Dungeness crab, Cancer magister, with emphasis on the central California fishery resource. State of California Dept. of Fish and Game, Fish Bulletin, 172: 259-309. Fisher, W.S. 1976. Relationships of epibiotic fouling and mortalities of the eggs of the Dungeness crab (Cancer magister). Can. J. Fish. Aquat. Sci., 33: 2849-2853. Fisher, W.S. 1983. Epibiotic microbial infestations of cultured crustaceans. In: C.J. Berg jr. (Editor), Culture of marine invertebrates, selected readings. Hutchison Ross Publishing Company, Pennsylvania, pp. 257-265. Fisher, W.S. and Nelson, R.T. 1977. Therapeutic treatment for epibiotic fouling on Dungeness crab larvae (Cancer magister) reared in the laboratory. Can. J. Fish. Aquat. Sci., 34: 432-436. Gardner, C. and Quintana, R. 1998. Larval development of the Australian giant crab Pseudocarcinus gigas (Lamarck 1818)(Decapoda: Oziidae) reared in the laboratory. J. Plank. Res., 20: 1169-1188. Gil-Turnes, M.S. and Fenical, W. 1992. Embryos of Homarus americanus are protected by epibiotic bacteria. Biol. Bull., 182: 105-108. Hamasaki, K. and Hatai, K. 1993. Prevention of fungal infection in the eggs and larvae of the swimming crab Portunus trituberculatus and the mud crab Scylla serrata by bath treatment with formalin. Nippon Suisan Gakkaishi, 59: 1067-1072. Kaji, S., Kanematsu, M., Tezuka, N., Fushimi, H. and Hatai, K. 1991. Effects of formalin bath of Haliphthoros infection on ova and larvae of the mangrove crab Scylla serrata. Nippon Suisan Gakkaishi, 57: 51-55. Lightner, D.V. 1993. Diseases of cultured Penaeid shrimp. In: J. McVey (Editor), CRC Handbook of Mariculture, Crustacean Aquaculture. CRC Press, Tokyo, pp. 393-486. Lio-Po, G.D. and Sanvictores, E.G. 1986. Tolerance of Penaeus monodon eggs and larvae to fungicides against Lagenidium sp. and Haliphthoros philippinensis. Aquaculture, 51: 161-168. Lio-Po, G.D., Sanvictores, M.E.G., Baticados, M.C.L. and Lavilla, C.R. 1982. In-vitro effect of fungicides on hyphal growth and sporogenesis of Lagenidium spp. isolated from Penaeus monodon larvae and Scylla serrata eggs. J. Fish. Dis., 5: 97- 112. Mardia, K.V., Kent, J.T. and Bibby, J.M. 1979. Multivariate Analysis. Academic Press, N.Y., 521 pp. Miller, R.G.Jr. 1981. Survival Analysis. John Wiley and Sons, N.Y., 238 pp. Nogami, K. and Maeda, M. 1992. Bacteria as biocontrol agents for rearing larvae of the crab Portunus trituberculatus. Can. J. Fish. Aquat. Sci., 49: 2373-2376. Paynter, J.L. 1989. Penaeid prawn diseases. In: Invertebrates in Aquaculture. Proceedings of Refresher Course for Veterinarians, 19-21 May 1989, Brisbane. Postgraduate committee in veterinary science, University of Sydney, 117: 145-190. Sokal, R.R. and Rohlf, F.J. 1995. Biometry, the Principles and Practice of Statistics in Biological Research. W.H. Freeman, N.Y., 887 pp. Williams, R.R., Bell, T.A. and Lightner, D.V. 1986. Degradation of trifluralin in seawater when used to control larval mycosis in Penaeid shrimp culture. J. World Aquacult. Soc., 17: 8-12. 155 General Discussion: Larval Biology of the Giant Crab Pseudocarcinus gigas 9 156 Taxonomy The anatomy of the larval stages of Pseudocarcinus gigas was consistent with other members of the family Eriphiidae. Some unusual characters indicated affinity with crabs of the genus Ozius, such as a spine on the dorsal surface of the first abdominal somite of zoeas, and a spine on the cheliped ischium of the megalopa. Ozius truncatus and O. deplanatus are found across southern Australia in the same region as P. gigas which strengthens the proposed affinity (Edgar, 1997). Many life history traits of crab species, such as mating behaviour, are consistent within family taxa so comparison with similar species can be useful (Hartnoll, 1969). McLay (1988) reviewed published information on O. truncatus which is limited to observations on feeding and defence behaviour of adults, diet studies (Chilton and Bull, 1984; Skilleter and Anderson, 1986), and description of the larvae (Wear, 1968; Wear and Fielder, 1985). Unfortunately, no information has been published on the reproductive biology of this species although it would be useful for predicting strategies in P. gigas due to the similarity between the genera evident from larval morphology. Taxonomically conservative aspects of the reproductive biology discussed in the next section of this thesis, such as soft/hard shelled mating of females, could be studied more easily with a small intertidal species like O. truncatus. The megalopas of Menippe species are similar to P. gigas based on the characters used by Martin (1988), although this was considered to be mainly an artefact of the presence of 5 zoeal stages, rather than 4 as in other members of Eriphiidae. Nonetheless, the zoeas of Menippe and P. gigas were grouped together using a range of characters described by Martin (1984) for xanthoid crabs and this provides more legitimate evidence of taxonomic affinity. Menippe species are harvested commercially in the West Atlantic and the Caribbean and there have been several publications on the reproductive biology. These studies provide a guide to aspects of the reproductive strategies of P. gigas and are discussed in the next section. Larval dispersal Modern, three-dimensional hydrodynamic models permit the larval dispersal of species to be predicted provided accurate biological and oceanographic information is available (Keough and Black, 1996). These authors listed categories of biological data that they considered critical for modelling and these were the duration of the planktonic period, buoyancy, and larval taxes and swimming speeds. An aim of the research on larval behaviour presented in this thesis was to collect this information so that dispersal modelling could be conducted in the future. The construction of three-dimensional hydrodynamic models of southern Australia is underway to investigate dispersal of southern rock lobsters Jasus edwardsii (CSIRO Australia) and it is anticipated that these models will be suitable for P. gigas. Model outputs of dispersal of P. gigas larvae should have higher precision than for J. edwardsii as rock lobsters have an extremely protracted planktonic stage of up to 24 months (Cobb et al., 1997) and are able to delay metamorphosis by mark-time moulting until conditions are suitable (Pollock and Melville-Smith, 1993; Baisre, 1994). Planktonic period The planktonic period of P. gigas is relatively long compared with other eriphiids, due to development through 5, rather than 4 zoeal stages (in eriphiids other than Menippe 155 species), and the more temperate environment. Ozius truncatus and O. verreauxii take 24-28 and 15 days to reach megalopa respectively, while M. mercenaria takes 12-22 days (Wear, 1968; Ong and Costlow, 1970; Dittel and Epifanio, 1985; McConnaughey and Krantz, 1992). Larval duration of P. gigas was variable and became longer at lower temperature, lower light intensity, shorter photoperiod, and in response to toxicity at high concentrations of prophylactic chemical treatments. Feeding and nutrition was not researched but this can also influence larval duration, although larvae tend to grow larger, rather than faster, with improved nutrition (Minagawa and Murano, 1993; Tong, et al., 1997). Temperature had a profound effect on planktonic period and the duration from hatch to megalopa ranged from 41 days at 21°C to 73 days at 13°C. As is often demonstrated in crustacean aquaculture trials, rapid growth is not necessarily optimal growth, as larvae may fail to accumulate sufficient reserves to metamorphose successfully (Minagawa; 1990; Chaoshu and Shaojing, 1992; Okamoto, 1993). The size of P. gigas larvae decreased with increasing temperature which appeared to affect metamorphosis in larvae cultured at temperatures greater than 16.8°C. Results from larval behaviour trials indicated that larvae would sink to avoid these temperatures so that the typical temperature range of P. gigas larvae is likely to lie between 14°C and 16°C (established for stage 1 and 2 zoeas only). This is similar to water temperature in the upper 50 m along southern Australia during late Spring and early Summer (Fig. 1). The mean duration of zoeal stages at these temperatures ranged from 48 to 62 days. The megalopa stage of brachyurans is also a planktonic dispersal stage although they regularly descend to the substrate (Lochmann et al., 1995). The duration of the megalopa stage in P. gigas was only determined in trials with prophylactic treatments for disease at 16°C where the mean duration was 38 days. Keough and Black (1996) considered that estimates of larval duration from laboratory studies were upper limits as most planktonic marine larvae require specific cues to settle, which are not usually present in the laboratory. This is unlikely to apply to brachyuran zoeas (which do not moult to a settling stage), although the duration of the megalopa stage in laboratory studies may be longer than in nature. There was some evidence of this in rearing of P. gigas as the megalopas ceased swimming around 10 days before metamorphosing to juveniles and moved entirely by walking. The abdomen was flexed under the cephalothorax which suggested that the musculature of the abdomen had altered, and that locomotion by swimming with the pleopods was no longer possible. The natural location of settlement of this species is likely to be at considerable depth so it is quite probable that appropriate settlement cues were not present and that metamorphosis was delayed. 156 Figure 1. Temperature maps of the upper 50 m of oceanic water across southern Australia in November, December and January (averaged for 7 years preceding 1996; maps supplied by CSIRO). 157 Buoyancy All P. gigas larval stages were negatively buoyant and stage 1 zoeas sank at 0.61 cm s-1. However, the mean upward swimming speed of stage 1 zoeas was 1.61 cm s-1 so larvae were readily able to maintain vertical position and some larvae maintained position in horizontal currents of up to 1.87 cm s-1. Except in conditions of upwelling or downwelling, their ability to maintain position and vertical distribution would be determined by their response to environmental stimuli. Larval taxis Behavioural responses of zoea 1, P. gigas larvae to a range of environmental stimuli suggests that they migrate towards the surface after hatching, then live in relatively low light environments (< 230 lux). The conclusion of upward swimming to the surface immediately after hatch is based on the strong, negative geotactic swimming behaviour of larvae during the first 24 h after hatch. Although only three specimens were captured, this is supported by the depth of P. gigas larvae obtained from plankton samples which were all in the upper 100 m of water. An indication that larvae are adapted for low light environments is that the negative phototactic response of larvae to change in light intensity only occurred at low intensity below 230 lux (roughly similar to the light level in a forest on an overcast day). This was an unusual response as it involved active negative phototaxis, rather than merely slow swimming speed resulting in sinking. The sensitivity of larvae to changes in light at low intensity, but not at bright intensity, suggests that this may be the natural light environment of P. gigas larvae. Although positive phototaxis was observed at higher intensities, this may be a laboratory artefact which is difficult to avoid, even with the apparatus design described in Chapter 5 which was intended to simulate natural underwater light distribution (Stearns and Forward, 1984; Forward, 1985; Forward, 1988). Larvae were able to feed and grow through to megalopa in low intensity lighting (2 lux) or complete darkness which demonstrates that they would not need to be present in surface waters during the day to capture prey. The proposed diel migration pattern of P. gigas larvae is deeper distribution during the day and shallower distribution during the night. Results of trials on the effect of temperature on larval behaviour indicate that temperature regulates depth distribution in the absence of light. As noted earlier, the typical temperature range of P. gigas larvae is likely to lie between 14°C and 16°C. This indicates that at night, larvae will accumulate at the surface in water of less than 14°C and at deeper depths in water of more than 16°C. Larvae readily penetrated thermoclines but their distribution may still be influenced by the absolute temperatures on either side of the boundary layer. For instance, it is predicted that larvae would accumulate at a thermocline with an upper temperature of 17°C and a lower temperature of 13°C. Results from behavioural experiments were limited for several reasons and it is unrealistic to expect precise prediction of the natural distribution of larvae. Rather, these trials provide experimental evidence of mechanisms to explain larval distribution. Most experimentation was based on first stage zoeas and there is potential for ontogenetic change in the response of larvae to environmental stimuli (Forward, 1990). Also, larval behaviour is extremely complex and an attempt was made to isolate specific stimuli in most of the experiments reported here. Natural behavioural responses of brachyuran larvae involve complex interactions between separate stimuli, such as light 158 and temperature (Anger, 1983; Sulkin, 1984). Accurate information on natural distribution could be obtained by plankton sampling, as described in Chapter 4, at different depths and in regions with higher density of adult crabs, such as in the northwest and north-east of Tasmania during October, November and December. This information would allow testing of the hypotheses on larval vertical migration presented in this thesis. Aquaculture potential Demand for small giant crabs of less than 3 kg has continued to grow since the inception of this project with an associated increase in beach price. The fishery appears to have peaked and be in decline after an initial fish-down of virgin stock, so the current high beach prices of $(Aust)50/kg should be sustained and may increase. Most product is exported to Asian markets and prices have not been affected by recent economic problems in this region, which also indicates that current prices are robust. There is clearly market opportunity for small, aquaculture product. Hatchery production Research conducted on prophylactic treatments for larval disease demonstrated that hatchery production of juvenile giant crabs is possible (Fig. 2). Best performance was in treatments with oxytetracycline with survival in some tanks of 10% through to crab 1. Although oxytetracycline enhanced survival of P. gigas larvae, the prophylactic use of antibiotics in crustacean hatcheries is inappropriate as it promotes the development of resistant strains of bacteria (Anderson, 1989). Two other chemicals enhanced survival of P. gigas larvae and appear to be promising alternatives to antibiotics: copper oxychloride and carbendazim. Figure 2. Hatchery-reared, first instar giant crab on an Australian 5 cent coin (around the size of a US 1 cent coin). 159 Survival through to the last zoeal stage was generally above 50% with most mortality at the moult to megalopa. This may indicate that conditions were inappropriate for the last zoeal stage, or more probably, that the larvae were lacking sufficient reserves from the zoeal stages to successfully metamorphose. This is a common problem in larval rearing of decapods and further research is required with P. gigas to improve survival to megalopa. Two areas are suggested for further trials in hatchery production of P. gigas. First, cannibalism of spines and appendages appeared to cause mortality, primarily through weakening the larvae and exposing the larvae to bacterial septicaemia. Although this was reduced by culturing larvae in complete darkness, the size and growth rate of larvae was affected and continuous darkness is not recommended. A more effective option may be improved tank design with greater mixing of larvae. High survival of over 70% to megalopa has been achieved with Menippe mercenaria cultured in Kreisel tanks which produce thorough mixing to minimise contact between larvae (Hughes et al., 1974; McConnaughey and Kranz, 1992). A second area for further research on hatchery production is to adjust environmental conditions through development to suit ontogenetic changes in environmental preferences. Temperature is likely to be especially important as behavioural tolerance of first and second stage zoeas was restricted to a narrow temperature window. Ontogenetic changes in temperature tolerance could be assessed by varying temperature through development experimentally, or by assessing instantaneous mortality rates of different stages maintained in static temperature systems. This was not possible in the trial presented in Chapter 6 due to low sample size. Larval duration of P. gigas was extremely protracted relative to other species which have been considered as aquaculture candidates. For instance, both Scylla serrata and M. mercenaria take around 12 days to reach megalopa compared with 48 to 62 days in P. gigas (Chaoshu and Shaojing, 1992; McConnaughey and Kranz, 1992). This extended larval duration may prohibit production of P. gigas juveniles at a realistic cost and an economic assessment would be wise prior to any additional research. Estimates of survival and density could be based on those obtained with M. mercenaria in pilot scale hatchery production (72.2% and 30.7 megalopas/l respectively; McConnaughey and Kranz, 1992). 160 Grow-out of juvenile crabs No trials on the grow-out of P. gigas juveniles are described in this thesis although some observations are relevant. Cannibalism has been a problem with the grow-out phase in culture of portunids, especially during ecdysis (Bardach et al., 1972), but was not observed with P. gigas juveniles or adults. Growth of juvenile P gigas was very slow for an aquaculture species with juveniles reaching a mean of 27 mm carapace width in 12 months in ambient flow-through water in Hobart (Fig. 3; Gardner, 1998). This is similar to M. mercenaria which take 12 months to reach 10 mm carapace width under normal conditions for Florida (Tweedale et al., 1993) and 2.5 years to reach marketable size of 100 mm carapace width (Bardach et al., 1972). An increase in temperature typically leads to shorter intermoult up to an optimal temperature for fastest growth (Kondzela and Shirley, 1993). Water temperature in the grow-out of P. gigas juveniles ranged from 7° to 17°C and faster growth of juveniles may be possible at warmer temperatures. However, growth is unlikely to compare with that of other aquaculture candidate species such as Scylla serrata which can reach marketable size within 9 months (C. Keenan, Pers. Comm., Queensland DPIF, 1997) or Callinectes sapidus which can reach marketable size in 4 months (Bardach et al., 1972). Although S. serrata and C. sapidus are worth far less than P. gigas, the slow growth of P. gigas juveniles remains a problem for aquaculture. Grow-out of P. gigas juveniles may not be viable except in combination with existing operations, such as abalone culture, or for stock enhancement. Figure 3. Hatchery-reared juvenile giant crab, 12 months after hatching. References Anderson, I.G. 1989. Hatchery health problems and hygiene management. In: Invertebrates in Aquaculture. Proceedings of Refresher Course for Veterinarians, 19-21 May 1989, Brisbane. Postgraduate committee in veterinary science, University of Sydney, 117: 109-120. Anger, K. 1983. Temperature and the larval development of Hyas araneus L. (Decapoda: Majidae); extrapolation of laboratory data to field conditions. Journal of Experimental Marine Biology and Ecology, 69: 203-215. 161 Baisre, J.A. 1994. Phyllosoma larvae and the phylogeny of Palinuroidea (Crustacea: Decapoda): a review. Australian Journal of Marine and Freshwater Research, 45: 925-944. Bardach, J.E., Ryther, J.H. and McLarney, W.O. 1972. Aquaculture, the Farming and Husbandry of Freshwater and Marine Organisms. John Wiley and Sons, Brisbane. 868 pp. Chaoshu, Z. and Shaojing, L. 1992. Effects of temperature on survival and development of the larvae of Scylla serrata. Journal of Fisheries China, 16: 58-66. Chilton, N.B. and Bull, C.M. 1986. Size related selection of two intertidal gastropods by the reef crab Ozius truncatus. Marine Biology, 93: 475-480. Cobb, J.S., Booth, J.D. and Clancy, M. 1997. Recruitment strategies in lobsters and crabs: a comparison. Marine and Freshwater Research, 48: 797-806. Dittel, A. and Epifanio, C.E. 1985. Desarrollo larval de Ozius el laboratorio. Revista de Biologia Tropical, 32: 171-172. verreauxii Saussure (Brachyura: Xanthidae) en Edgar, G. 1997. Australian Marine Life: the Plants and Animals of Temperate Waters. Reed Books, Kew, Victoria. 544 pp. Forward, R.B.Jr. 1985. Behavioural responses of larvae of the crab Rhithropanopeus Xanthidae) during diel vertical migration. Marine Biology, 90: 9-18. harrisii (Brachyura: Forward, R.B.Jr. 1986. A reconsideration of the shadow response of a larval crustacean. Marine Behaviour and Physiology, 12: 99-113. Forward, R.B.Jr. 1988. Diel vertical migration: zooplankton photobiology and behaviour. Oceanography and Marine Biology Annual Review, 26: 361-393. Forward, R.B.Jr. 1990. Behavioural responses of crustacean larvae to rates of temperature change. Biological Bulletin, 178: 195-204. Gardner, C. 1998. The Tasmanian giant crab fishery: a synopsis of biological and fisheries information. Tasmanian Department of Primary Industry and Fisheries, Internal Report 43, 40 pp. Hartnoll, R.G. 1969. Mating in the Brachyura. Crustaceana, 16: 161-181. Hughes, J.T., Shleser, R.A. and Tchobanoglous, G. 1974. A rearing tank for lobster larvae and other species. Progressive Fish Culturist, 36: 129-132. Keough, M.J. and Black, K.P. 1996. Predicting the scale of marine impacts: understanding planktonic links between populations. In: Detecting Ecological Impacts, Concepts and Applications in Coastal Habitats (Eds. R.J. Schmitt and C.W. Osenberg). Academic Press, Sydney. Pp. 199-234. Kondzela, C.M. and Shirley, T.C. 1993. Survival, feeding, and growth of juvenile Dungeness crabs from southeastern Alaska reared at different temperatures. Journal of Crustacean Biology, 13: 25-35. Lochmann, S.E., Darnell, R.M. and McEachran, J.D. 1995. Temporal and vertical distribution of crab larvae in a tidal pass. Estuaries, 18: 255-263. McConnaughey, R.A. and Krantz, G.E. 1992. Hatchery production of stone crab, Menippe mercenaria (Say), megalopae. Proceedings of a Symposium on Stone Crab (Genus Menippe) Biology and Fisheries, Florida Marine Research Publications, 50: 60-66. McLay, C.L. 1988. Brachyura and crab-like Anomura of New Zealand. Leigh Laboratory Bulletin, 22: i-iv, 1463. 162 Martin, J.W. 1984. Notes and bibliography on the larvae of xanthid crabs, with a key to the known xanthid zoeas of the western Atlantic and Gulf of Mexico. Bulletin of Marine Science, 34: 220-239. Martin, J.W. 1988. Phylogenetic significance of the brachyuran megalopa: evidence from the Xanthidae. Symposium of the Zoological Society of London, 59: 69-102. Minagawa, M. 1990. Influence of temperature on survival, feeding and development of larvae of the red frog crab, Ranina ranina (Crustacea, Decapoda, Raninidae). Nippon Suisan Gakkaishi, 56: 755-760. Minagawa, M. and Murano, M. 1993. Larval feeding rhythms and food consumption by the red frog crab Ranina ranina (Decapoda, Raninidae) under laboratory conditions. Aquaculture, 113: 251-260. Okamoto, K. 1993. Influence of temperature on survival and growth of larvae of the giant spider crab, Macrocheira kaempferi (Crustacea, Decapoda, Majidae). Nippon Suisan Gakkaishi, 59: 419-424. Ong, K.S. and Costlow, J.D.Jr. 1970. The effect of salinity and temperature on the larval development of the stone crab, Menippe mercenaria (Say), reared in the laboratory. Chesapeake Science, 11: 16-29. Pollock, D.E. and Melville-Smith, R. 1993. Decapod life histories and reproductive dynamics in relation to oceanography off southern Africa. South African Journal of Marine Science, 13: 205-212. Shirley, S.M. and Shirley, T.C. 1988. Behaviour of red king crab larvae: phototaxis, geotaxis and rheotaxis. Marine Behaviour and Physiology, 13: 369-388. Skilleter, G.A. and Anderson, D.T. 1986. Functional morphology of the chelipeds, mouthparts and gastric mill of Ozius truncatus (Milne Edwards)(Xanthidae) and Leptograpsus variegatus (Fabricius)(Grapsidae)(Brachyura). Australian Journal of Marine and Freshwater Research, 37: 67-79. Stearns, D.E. and Forward, R.B.Jr. 1984. Photosensitivity of the calanoid copepod Acartia Biology, 82: 85-89. tonsa. Marine Sulkin, S.D. 1984. Behavioural basis of depth regulation in the larvae of brachyuran crabs. Marine Ecology Progress Series, 15: 181-205. Tong, L.J., Moss, G.A. Paewai, M.M. and Pickering, T.D. 1997. Effect of brine shrimp numbers on growth and survival of early stage phyllosoma larvae of the rock lobster Jasus edwardsii. Marine and Freshwater Research, 48: 935-940. Tweedale, W.A., Bert, T.M. and Brown, S.D. 1993. Growth of postsettlement juveniles of the Florida stone crab, Menippe mercenaria (Say) (Decapoda: Xanthidae), in the laboratory. Bulletin of Marine Science, 52: 873-885. Wear, R.G. 1968. Life history studies on New Zealand Brachyura. 2. Family Xanthidae. Larvae of Heterozius rotundifrons A. Milne Edwards, 1867, Ozius truncatus H. Milne Edwards, 1834, and Heteropanope (Pilumnopeus) serratifrons (Kinahan, 1856). New Zealand Journal of Marine and Freshwater Research, 2: 293-332. Wear, R.G. and Fielder, D.R. 1985. The marine fauna of New Zealand: larvae of the Brachyura (Crustacea, Decapoda). New Zealand Oceanography Institute Memoirs, Wellington, 92 pp. 163 General Introduction: Reproductive Biology of the Giant Crab Pseudocarcinus gigas 10 164 The need for research The giant crab fishery grew rapidly after 1991 and fisheries managers initially had no biological information to assist in formulating management strategies. During this development period, the fishery was managed by a federal authority (Australian Fisheries Management Authority) and licences were issued with little attempt to restrain increase in effort. The lack of caution exercised by this authority was an anathema to most State fisheries management authorities and it was criticised in the wider community. At this time, the orange roughy Hoplostethus atlanticus fishery had just collapsed as information on growth and biomass had been collected only after intense fishing pressure was applied to stocks, rather than before. Scientists such as Tim Flannery (1994) considered that the orange roughy collapse could be repeated with giant crabs as they were also from deep water, were probably slow growing, and were subjected to high fishing pressure before research was conducted. Tsaamenyi and McIlgorm (1995) discussed the draft "FAO Code of Conduct for Responsible Fishing" and illustrated problems with current practice in Australia using the Tasmanian giant crab fishery as an example. This code of conduct suggests that exploitation of a previously unfished stock should not increase until a plan for rational exploitation has been agreed. This illustrates the need for information on the biology of the giant crab which prompted research including that presented in this thesis. Information on larval biology was initially required to address concerns that dispersal may be limited. After having established that development was not abbreviated, research on the larval biology was directed towards providing input data for dispersal modelling. While this research can contribute to the long term viability of a giant crab fishery across southern Australia, information on the reproductive biology was required more immediately for establishing basic management rules, such as size limits and closed seasons. Commercial management arrangements Current management arrangements for the Tasmanian commercial fishery are listed in Table 1. There are also restrictions of the recreational harvest of giant crabs with a maximum daily catch limit of two crabs. This recreational limit is largely unnecessary as no recreational catch of giant crabs was recorded in a recent survey of recreational fishing in Tasmania (Lyle and Smith, 1998). Note that the commercial management arrangements are based on input controls to constrain effort and most of these are based on the rock lobster fishery for ease of enforcement (e.g. escape gaps and closed seasons). Of the existing controls, only the minimum size limit is linked to biological information and this was introduced as an interim measure until more extensive data became available. The management system is currently under review and various alternatives have been proposed including the introduction of a total allowable catch, reduction of licenses, and the introduction of a maximum size limit. Information on the reproductive biology is clearly useful in this process. 161 Table 1. Current management arrangements for the Tasmanian giant crab fishery Management zone: one management zone for the State (since January 1997). Limited entry: 106 licences (approximately 1/3 of the 321 rock lobster licences in the State). Limited seasons: 18th November - 21st December; 3rd January - 14th February; 1st March - 31st August. Limits of pots on vessels: minimum of 15 pots, maximum of 50 pots. Restrictions on setting pots: pots cannot be set, or pulled, between two hours after sunset and two hours before sunrise. Pots must be hauled no longer than 5 days after being set. Pots may be deployed in long-lines of up to 10 pots. Restrictions on pot size: maximum size of 1250 mm x 1250 mm x 750 mm. Escape gaps: one escape gap at least 57 mm high and 400 mm wide and not more than 150 mm from the inside lower edge of the pot, or two escape gaps at least 57 mm high and 200 mm wide and not more than 150 mm from the inside lower edge of the pot (as designed for rock lobster). Minimum size limits: 150 mm carapace length for both sexes (since 1993). Ovigerous females: taking of ovigerous females prohibited (since 1993). Development of techniques for research Two chapters in this section describe techniques developed for use in subsequent research on reproductive biology. The first reports the effect of various treatments tested for humanely killing or immobilising giant crabs. Crabs were difficult to subdue or kill using conventional methods such as chilling and they would often damage themselves struggling against claw ties. Aside from ethical considerations, the ability to immobilise crabs during procedures such as removing eggs was beneficial as crabs were easier to work with. Giant crabs are extremely strong, and sometimes fast, so techniques to immobilise them reduced the risk of human injury. The second chapter on technique describes options for the non-lethal imaging of spermathecae. This was investigated for research on sperm storage in females held for several years in tanks (Chapter 14). It was intended that scanning spermathecae would allow images to be collected of the spermathecae without killing the crabs, should infertile clutches be produced. This never eventuated despite holding the female crabs isolated from males for the duration of the project, so the techniques described were never employed. However, the imaging techniques are useful as crab research methods and have been applied in research that is on-going from that described in this thesis. This ongoing research utilises computerised tomography scanning (CT) to assess gonad development and thus onset of sexual maturity. Important benefits from non-lethal imaging are that crabs can be sold after the images are collected (the market for giant crab is based on live animals with beach price ranging from $50 to $200 each), and crabs can be processed more rapidly than would be possible by dissection. 162 Objectives of research on giant crab reproductive biology Research was conducted on the reproductive biology of both male and female giant crabs as both are available for harvest. Although both sexes are harvested, males may be subject to higher exploitation as they appear to grow faster and the same minimum legal size limit is applied to both sexes (Pers. Comm., R. McGarvey, South Australian Research and Development Institute). Also, the open season is effectively reduced for females as they cannot be retained while ovigerous and egg extrusion occurs in May or June, while the season closes at the end of August. Initial research on the reproductive biology of male crabs identified morphological stages in development which have been observed in numerous other species (Paul, 1992). Three morphological stages of development in male giant crabs were defined by development of the molariform chelae relative to carapace size (Figure 1). The reproductive maturity of these stages was the subject of additional research which involved a preliminary histological study on the formation of spermatophores. Results from this research are discussed in relation to current management of the resource. Several aspects of the reproductive biology of female crabs was studied including the mating system, sperm storage, and cycles of ovarian development and interactions with the hepatopancreas. Egg production was studied in detail to assess the effect of female size on fecundity, individual egg size and egg composition. This information is required for modelling the impact of fishing on egg production of the population. Although information on the onset of sexual maturity in females is critical for management of the resource, this was not investigated in the current study as it was beyond the scope of available resources. However, where possible, data were collected from different regions of the fishery to provide some information on the extent of spatial effects. Taxonomic affinity between Pseudocarcinus and Menippe was demonstrated in Chapter 3 and this was useful for evaluating results of research on reproductive biology. 163 Figure 1. Male (upper) and female (lower) giant crabs Pseudocarcinus gigas showing sexual dimorphism in the development of the chelae. The male is large (11 kg) and is in the final morphological stage identified in Chapter 14, termed morphological maturity (photos by Karen Gowlett-Holmes). References Flannery, T. 1994. The Future Eaters. Reed Books, Sydney. 423 pp. Lyle, J.M. and Smith, J.T. 1998. Pilot survey of licensed recreational sea fishing in Tasmania - 1995/96. Department of Primary Industry and Fisheries Tasmania, Technical Report, 51. Paul, A.J. 1992. A review of size at maturity in male tanner (Chionoecetes bairdi) and King (Paralithodes camtschaticus) crabs and the methods used to determine maturity. American Zoologist, 32: 534-540. Tsaamenyi, M. and McIlgorm, A. 1995. International Environmental Instruments- Their Effect on the Fishing Industry. Australian Fisheries and Research Development Corporation Final Report. 164 Options for humanely immobilising and killing crabs 11 165 Abstract Trials were conducted on the Australian giant crab Pseudocarcinus gigas (Lamarck) to evaluate methods to: paralyse by injection (so that no muscular response is observed); paralyse by bath; humanely kill for scientific purposes; and humanely kill for human consumption. Treatments tested were: freshwater bath, chilling, heating, prolonged exposure to air, hypercapnic seawater bath (carbon dioxide addition), 2-phenoxy ethanol bath, magnesium sulphate bath, benzocaine bath, MS 222 bath, chloroform bath, clove oil bath, AQUI-S™ bath, xylazine-HCl by injection, and ketamine-HCl by injection. Xylazine-HCl (16 or 22 mg/kg) and ketamine-HCl (0.025-0.1 mg/kg), administered by injection, appear to be the best techniques for paralysing crabs for short periods. Where injection is impractical, crabs may be successfully paralysed within 30 min by a bath treatment of clove oil (≥0.125 ml/l) or AQUI-S™ (≥0.5ml/l). Chloroform (1.25 ml/l; 1.5 h) and clove oil (≥0.125 ml/l; ≤60 min) baths appeared to kill crabs humanely and are useful options for scientific use, however, clove oil is preferred as chloroform poses a human health risk. Of the methods tested, only clove oil and AQUI-S™ appear promising as treatments for the humane killing of crabs for human consumption. Introduction Methods of paralysing crabs can benefit many research situations involving live crabs; procedures may be conducted more efficiently and trauma to the crab is reduced (Oswald, 1977). Where the application is prolonged or the dosage increased, humane killing may result which is desirable for research and commercial uses of crabs. In commercial situations, it is important that quality is not harmed by effects such as autotomy, and toxic chemicals cannot be used. Recent changes to Australian animal cruelty legislation have added another consideration to the commercial killing of crabs: that the crab be killed humanely. Humane killing involves attempting to inflict as little pain as possible while killing the crab. Pain is a difficult, or perhaps impossible, aspect to measure in animals other than humans, so it is usually inferred from changes in behaviour which seem to indicate distress (Chapman, 1992; Cook, 1996). These behavioural changes are not apparent when the muscular response is blocked by induced paralysis, so anaesthesia (blockage of pain) is not assured, despite an apparent lack of distress. Likewise, the absence of behavioural indications of distress does not necessarily indicate that killing is painless. Nonetheless, in the absence of methods to quantitatively measure pain, techniques for killing or immobilising animals where distress is apparently reduced are preferred to techniques which produce obvious distress. This study reports the results of trials in temporarily paralysing and killing the Australian giant crab Pseudocarcinus gigas (Lamarck) which is large and potentially dangerous, as large males are capable of fracturing a human wrist. Thus, techniques for immobilising giant crabs are also beneficial as they improve handling. Although numerous methods of temporarily paralysing and killing crustaceans have been documented, many are slow, inconsistent, and appear to cause trauma (Brown et al., 1996). A range of physical and chemical treatments was tested on giant crabs to establish which treatments were effective and economical for this large species, and also to note apparent trauma from treatments. Treatments were evaluated for the following applications: paralysing by injection (appropriate for large crabs); paralysing by bath 175 (appropriate for small crabs); killing for research (toxic chemicals acceptable); and killing for commercial use (safe for consumption). Materials and methods Adult giant crabs (Pseudocarcinus gigas) were collected from western Tasmania by commercial fishers and ranged from 1 to 7 kg with most between 2.5 and 3.5 kg. Crabs were held in 4 m3 tanks with flow-through seawater and were only used if they exhibited normal avoidance of capture. Treatments were first tested in producing paralysis; crabs were then allowed to recover in tanks with flow-through seawater and were monitored for two days to assess any ill effects. Where the treatment was effective and did not appear to cause pain (see below), further trials were undertaken to establish appropriate dosages for producing temporary paralysis and to assess the treatment for humane killing. In some treatments, the crabs appeared to be severely harmed by the paralysis trial and recovery was not assessed. Criteria for assessing pain, paralysis, and death Although pain is impossible to quantify, changes in behaviours of experimental animals have been used to infer perception of pain (Chapman, 1992; Cook,1996). In these trials, the treatment was considered to have caused pain where crabs dropped limbs (autotomy), tore at their appendages or abdomens, became tensed and rigid, or appeared to have muscle spasms. Paralysis was considered complete when the abdomen could be easily lifted and chelae (claws) could not be used defensively. Where recovery was to be assessed, crabs were removed from bath treatments before circulation of water over the gills ceased (externally observed by flow of water). Nervous systems of crabs have two centres, the cerebral and the posterior ganglia. Baker (1955) devised a simple system of testing if these centres were functioning, and thus if the crab was alive, by observing the response to stimuli applied to different appendages. Where no response was observed the crab was classed dead. Baker’s (1955) system was modified in this study to avoid the use of optical stimuli as giant crabs are deep sea animals and their spectral sensitivity may have been impaired by surface level sunlight after capture (Cronin and Forward,1988). Consequently, the following tests were used to assess if crabs were dead: Antennal reaction: The crab does not retract the first antennae when the distal end is touched (cerebral ganglion). Maxilliped reaction: The third maxilliped (mouth frame) can be moved outwards from the body and is not drawn back (posterior ganglion). Treatment strategy Bath treatments were conducted in individual tanks of 20 l with continuous aeration (except hypercapnic seawater treatment). These tanks were filled with water from the larger holding tanks so that salinity (35 ppt) and temperature (range 9-13°C) were not altered. Injections were made intravascularly through the coxal arthropodial membrane of a cheliped (Fig. 1). Doses by injection were made up to a maximum of 2 ml as volumes greater than this were considered difficult to administer. 176 Figure 1. Ventral surface of a giant crab showing the site of intravascular injections (i). Treatments were introduced with the needle tip only slightly below the joint membrane to avoid penetrating muscle tissue. Physical methods tested for paralysing were: freshwater bath; chilling (5°, 2°, and 1.5°C); heating (17°, 18°, 20°, and 24°C); and prolonged exposure to air. Chemical methods tested as baths were: hypercapnic sea water (CO2 bubbled into bath through a graphite airstone); 2-phenoxy ethanol (maximum 1 ml/l); magnesium sulphate (35 g/l); benzocaine (0.08 and 0.24 g/l, stock solution of 40 g/l benzocaine in acetone); MS 222 (tricaine methane sulphonate; 0.5 g/l); chloroform (1.25 and 2.5 ml/l in water and agitated); clove oil (0.015-1.0 ml/l, dissolved in ethanol); and AQUI-S™ (Fish Transport Systems™ New Zealand)(0.015-1.0 ml/l). Chemical methods tested by intravascular injection were: xylazine-HCl (0.6-22.0 mg/kg; as 2% solution, Rompun-Bayer™); and ketamine-HCl (0.01-0.05 mg/kg; as 10% solution, Ilium-Troy™). Of these treatments, four were tried for humane killing: freshwater bath; chilling; chloroform; and clove oil. Chilling was achieved by the addition of ice slurry to 100 l tanks held in a refrigerated room. Heating was achieved by placing immersion heaters in 100 l tanks. The number of crabs used for experiments varied (Table 1) as the response of individual crabs to some treatments was so poor at very high doses that further trials were not warranted. Other trials were conducted opportunistically with industry so large numbers were used, such as with prolonged exposure to air where 55 animals were monitored. The opportunistic nature of the trials prevented concurrent experimentation. Results None of the physical methods appeared to be suitable for producing temporary paralysis as they were either ineffective or they appeared to distress the crab (Table 1). There were also practical problems with the physical methods that rendered them unsuitable. Crabs were only affected by cold water temperatures close to freezing. Consequently, regular monitoring was required for the entire 2 h period needed to partially paralyse crabs, to ensure the water did not freeze. Also, crabs revived when the temperature rose so they recovered rapidly during experimental procedures. 177 Table 1. Results of trials to assess the use of treatments for paralysing. Method Time to paralyse Indication of stress / revival Fresh water bath (n=10) Immediately became rigid and easily handled Motionless and rigid for 10 min then became very active. Autotomy occurred and crabs tore at their abdomens and walking legs. No revival was attempted. Chilling: 5°, 2°, and -1.5°C (n=10, 10, and 60 respectively) Unaffected after 14 h at 5° and 2°C. Mild paralysis in 2 h at 1.5°C (retained antennal, maxilliped, and limb movement). Active at 5° and 2°C. Ice formed at -1.5°C so the last segments (propodus and dactylus) of the limbs to became frozen. All recovered within 45 min on return to 10°C and appeared healthy after 48 h. No effects of freezing were seen although tissue damage is likely. Heating: 17°, 18°, 20°, and 24°C (n=3 for all treatments) Appeared unaffected at all temperatures tested except 24°C. Mild paralysis at 24°C in 2 h. Appeared uncomfortable and attempted to climb from the container as temperature rose. Although apparently paralysed at 24°C, limbs constantly twitched. Recovery was rapid and crabs appeared healthy after 48 h. Prolonged exposure to air (n=55). No effect at 4 or 8 hours. Less active after 14 h (8-12°C). Crabs were vigorous after 14 h. Appeared healthy after 48 h in seawater. Hypercapnic seawater (n=3) Mean = 44 min (range 33-60 min). Thrashed and crushed limbs. Although immobile, they were tensed and became rigid when returned to fresh seawater to recover. Some autotomy. Slow recovery, incomplete after 48 h. 2-phenoxy ethanol: 1 ml/l (n=1) No effect after 14 h in saturated solution. No apparent effect. Healthy 48 h after return to seawater. MgSO4: 35 g/l in fresh water; 35 g/l in sea water (n=6) No effect at 4 h Active 48 h after return to seawater. Cost of chemicals made trials with higher doses unviable. Benzocaine: 0.08 g/l (n=1), and 0.24 g/l (n=3). 2 h at 0.08 g/l, Mean = 45 min, range 2055 min at 0.24 g/l Apparent distress, tensed and rigid when immobilised. Autotomy occurred. Rapid recovery, crabs mobile within 10 min and healthy after 48 h. MS 222: 0.5 g/l (n=1) No effect after 4 h. One tenth this dose (20 min) is used for killing finfish (Clark,1990). Trials at higher doses were unviable due to chemical costs. Chloroform: 1.25 ml/l (n=3), and 2.5 ml/l (n=3) 60 min for all crabs No apparent distress. Slow recovery, crabs still sedated after 24 h although apparently normal after 48 h. Clove oil: 0.015 ml/l - 1.0 ml/l (Fig. 2; n=18). Ineffective at 0.015 ml/l. Time at higher doses (≥0.03 ml/l) ranged from 85-16 min. No apparent distress. Rapid recovery (at 0.125 ml/l) and active 2.5 h after return to seawater. Appeared healthy after 48 h. 178 Method Time to paralyse Indication of stress / revival AQUI-S™: 0.015 ml/l - 1.0 ml/l (Fig. 2; n=14). Ineffective at ≤ 0.06 ml/l. Effective at ≥ 0.125 ml/l in 7020 min. No apparent distress. Rapid recovery and active 2.5 h after return to seawater. Appeared healthy after 48 h. Xylazine-HCl: 0.6, 1.2, 5.6, 11.2, 16, and 22 mg/kg. (n=6) Ineffective ≤11.2 mg/kg. Effective in 3-5 min at 16 and 22 mg/kg. No apparent distress. Rapid recovery: 25 min at 16 mg/kg, and 45 min at 22 mg/kg. Appeared healthy at 48 h. Ketamine-HCl: 0.01, 0.025, 0.05, and 0.1 mg/kg. (n=8) Ineffective at 0.01 mg/kg; effective at 0.025, 0.05, and 0.1 mg/kg in 15-45 s at all concentrations. Cheliped became rigid immediately after injection, the other cheliped was thrashed, then relaxed. Recovery took 8, 15, 25, and 40 min at 0.01, 0.025, 0.05, and 0.1 mg/kg respectively. Apparently healthy after 48 h. None of the physical methods appeared suitable for humane killing (Table 2). Aside from ethical problems, prolonged exposure to air would take longer than 48 h and was not attempted. Heating appeared to cause distress to the crabs and killing by a gradual increase in temperature was not attempted. While a fresh water bath killed crabs, it did not appear to be a humane method. Chilling was ineffective. Table 2. Results of trials to assess the use of treatments for humane killing. Method Time to death Comments Fresh water bath (n=10) Mean = 4.6 h (range = 3-5 h) Apparent distress (see Table 1). Chilling: 2° and 1.5°C (n=10 and 60 respectively). 2°C group alive at 24 h. -1.5°C alive at 6 h. Appendage reactions occurred at -1.5°C and limb movement was retained. Activity increased rapidly on warming. Chloroform: 2.5 ml/l (n=3) 1.5 h in all crabs. No apparent distress. Clove oil: 0.06 ml/l - 1.0 ml/l (Fig. 2; n=17). 180-28 min, varying with concentration. No apparent distress. Of the chemical bath treatments tested, only chloroform, clove oil and AQUI-S™ produced what appeared to be relaxed temporary paralysis. Crabs treated with chloroform had poor recovery after paralysis and took longer to die compared with clove oil treatments. Chloroform solutions become saturated at approximately 6.17 ml/l at 10°C which is considerably higher than the concentrations used in this study (1.25 and 2.5 ml/l). Consequently, the similar times for crabs to become paralysed in the two concentrations of chloroform cannot be attributed to saturation of solution. The optimal concentration of clove oil for both paralysing and killing was 0.125 ml/l as stronger doses did not produce faster effects (Fig. 2; Tables 1 and 2). Optimal concentration of AQUI-S™ for paralysing was 0.5 ml/l (Fig. 2). Both of the temporary paralysis treatments administered by injection were effective and acted rapidly. Unlike xylazine-HCl, where paralysis appeared to be painless, ketamineHCl appeared to produce distress in the crabs although this was only momentary as paralysis occurred within 45 seconds. 179 Figure 2. Effect of concentration of clove oil (circles) and AQUI-S™ (diamonds) on time taken to paralyse (upper) and to kill (lower) giant crabs Pseudocarcinus gigas. Solid symbols represent trials terminated before paralysis or death occurred. Value labels next to means are numbers of crabs used. Tests with AQUI-S™ used only two crabs at each concentration so no error values are presented. Mean time ± SD (min) 250 1 a. Anaesthetised 200 150 100 2 50 2 3 3 3 0 0.015 0.03 0.06 0.125 0.25 0.5 Concentration (ml/l) 4 1.0 350 Mean time ± SD (min) 300 b. Dead 2 250 200 2 150 100 3 50 3 3 4 0 0.015 0.03 0.06 0.125 0.25 0.5 Concentration (ml/l) 1.0 Discussion Several of the treatments tested in producing paralysis were rejected as they were ineffective or because the dose required was too large for practical purposes: prolonged exposure to air, 2-phenoxy ethanol, magnesium sulphate, and MS 222. MS 222 is used widely in paralysing finfish (Clark, 1990) and was recommended by Ahmad (1969) for amphipods although several other studies have confirmed that it is ineffective in decapods (Foley et al., 1966; Oswald, 1977; Brown et al., 1996). MS 222 is believed to act at the nerve membrane affecting sodium conductance in finfish (Ryan, 1992) and the ineffectiveness of MS 222 in decapods may be related to the absence of acetylcholine at these terminals (Oswald, 1977). The large amount of magnesium sulphate required to paralyse large decapods was considered impractical in this study and the same conclusion was drawn by Foley et al. (1966). For smaller animals, and thus smaller bath volumes, the technique may still have value (Gohar, 1937). Temporary paralysis Although many of the methods tested in this trial produced paralysis, some were only partially effective and others appeared to be unsuitable due to evidence of pain or 180 distress during relaxation. Hypercapnic seawater has been recommended for paralysing reptantian decapods (Smaldon and Lee, 1979) and was also effective with Pseudocarcinus gigas. However, the technique resulted in autotomy and thrashing of limbs in P. gigas which indicates paralysis was not painless. Smaldon and Lee (1979) report that Crangon spp. and Palaemon spp. also exhibit distress when placed in hypercapnic seawater. Oswald (1977) assessed the use of benzocaine to temporarily paralyse Cancer pagurus L. and Carcinus maenas (L.) by injection and observed no effect. Benzocaine is widely used to produce paralysis in finfish and abalone research where it is administered as a bath, as was done in this study with P. gigas. The bath solution of 0.24 g/l benzocaine produced paralysis in P. gigas although there was some indication of pain, as with hypercapnic sea water. None of the physical methods tested produced relaxed paralysis in Pseudocarcinus gigas. A gradual increase in temperature was described as an effective and humane method of anaesthetising and killing large crustaceans by Gunter (1961) and was subsequently recommended by Smaldon and Lee (1979). This method was effective at paralysing P. gigas although animals showed signs of distress, contrary to the observations of Gunter (1961). Baker (1955) also tested the response of crabs to gradual increase in temperature and concluded that the method was unacceptable on humanitarian grounds as indications of distress, such as autotomy, occurred unless the crab was already in poor health. Following publication of Gunter’s (1961) conclusion on the use of gradual heating, objections were raised to the method on the basis that there was no evidence of an anaesthetic effect (Baker, 1962; Schmidt-Nielsen, 1962). Current Australian guidelines for the killing of crabs for scientific purposes recommend chilling as a humane method for paralysing crabs, which can then be killed by sectioning to destroy ganglia (Reilly, 1993). While chilling may be useful for reducing activity in tropical or warmer water species, it was ineffective as a paralysing technique for the temperate Pseudocarcinus gigas. Freezing inevitably results in death in P. gigas but is of limited use in research as tissues are no longer suited for many applications, such as histology. Chilling has drawbacks which affect its use in all species, it is generally a slow and inconsistent technique (Brown et al. 1996) and it is ethically dubious as it involves subjecting the crab to conditions which it would normally avoid (SchmidtNielsen, 1962). Killing crabs by freshwater bath is one of the most widely used methods in Australia; it is termed “drowning” and is popularly considered a humane technique. Of all the treatments tested for producing paralysis, “drowning” in a freshwater bath appeared to cause greatest trauma as crabs dropped most limbs. Similar conclusions were drawn by Baker (1955) for Cancer pagurus. Both xylazine-HCl and ketamine-HCl were particularly effective and produced paralysis in less than 5 min in Pseudocarcinus gigas. Xylazine-HCl produces relaxation by central blockade of interneurones in the mammal (Oswald, 1977) but the mode of action in crustaceans is unknown. Although injection of ketamine-HCl appeared to cause localised excitation, the apparent distress was only momentary as most crabs became paralysed within 45 s. Ketamine-HCl is effective in paralysing crayfish Orconectes virilis (Hagen) although the reported dose rate by intramuscular injection (90 mg/kg body weight; Brown et al., 1996) was considerably higher than that required by P. gigas by intravascular injection (0.025 mg/kg). The duration of paralysis in P. gigas treated with ketamine-HCl (8-40 min) also differed from O. virilis (>1 h; Brown et al., 1996). Dose rates of xylazine-HCl required to temporarily paralyse P. gigas (22 mg/kg) were less than reported values for Cancer pagurus and Carcinus maenas (70 mg/kg; Oswald, 1977), although 181 duration of paralysis was similar (around 45 min for all species). Two other chemicals, reportedly effective in other decapods, were not tested on P. gigas but warrant mention: procaine-HCl is reported to produce prolonged paralysis of 60 min in C. pagurus and C. maenas (Oswald, 1977); and lidocaine-HCl is reported to produce shorter-duration paralysis of 20-25 min in O. virilis (Brown et al., 1996). Where the experimental animals are very small, injection is less practical than bath treatments; chloroform (>1.25 ml/l), clove oil (>0.125 ml/l), and AQUI-S™(>0.5 ml/l) produced relaxed paralysis by this method. Chloroform has been used for many decades to kill decapods that subsequently remain relaxed for museum storage (Gohar, 1937; Mahoney, 1966), but the use of chloroform to produce temporary paralysis has been less well documented. Foley et al., (1966) attempted to paralyse Homarus americanus Milne Edwards, by chloroform bath but concluded that too large a dose was required for practical purposes. The chloroform bath treatment was effective and inexpensive with P. gigas although there are important limitations: the time to onset of paralysis (60 min) and recovery from paralysis (>24 h) was protracted; and chloroform poses a serious health risk to humans due to its hepatotoxicity. Clove oil was the superior bath treatment in respect of both time to onset of paralysis (as rapid as 16 min) and recovery (2.5 h). Clove oil is inexpensive and is likely to be effective over a wide range of species given that it also produces paralysis in rabbitfish Siganus lineatus (Cuvier and Valenciennes)(Soto and Burhanuddin, 1995). AQUI-S™ produced similar results to clove oil but may have limited application in paralysing crabs for scientific purposes as higher doses were required. A potentially useful observation from the clove oil trials is that embryos of ovigerous giant crab females did not appear to be harmed by the treatment and continued through development to hatch. Bath treatments of clove oil and AQUI-S™ may have commercial application to improve seafood quality and to reduce mortality during live transport. Reduction of stress during transport and prior to harvest is known to increase quality of seafoods (Lowe et al. 1993) and to decrease transport mortality (Paterson et al., 1994). Humane killing Two of the treatments widely used in Australia for killing crabs were either ineffective or appeared to cause suffering: chilling; and freshwater bath. Both chloroform and clove oil were effective and crabs did not appear distressed by the treatments. As discussed earlier, chloroform has long been used to kill crustaceans for museum collections (Gohar, 1937; Mahoney, 1966), where it is important that the animal does not autotomise limbs. Chloroform is hazardous to humans due to its hepatotoxicity so it should only be used where all fumes can be removed. Unlike chloroform, clove oil has potential to be used for killing animals destined for human consumption although the long term chronic effects on humans are not yet known (Soto and Burhanuddin, 1995). Cloves have been shown to delay rancidity of seafood (Joseph et al., 1989) although the oil has a strong smell which can alter the taste of the meat. AQUI-S™ is approved for use with food fish in New Zealand with zero withholding time; it produced paralysis in giant crabs although higher doses were required than with clove oil. Unlike clove oil, AQUI-S™ does not have a strong odour so is less likely to affect the taste of the meat. Further trials are warranted to assess the use of AQUI-S™ in the killing of crabs and to assess the effect on meat quality utilising human sensory evaluation. Conclusions 182 Several methods of paralysing and killing crabs are clearly suitable for research situations. The two injectable treatments, xylazine-HCl (16 or 22 mg/kg) and ketamineHCl (0.025-0.1 mg/kg), have much potential in research to reduce trauma to the crab, increase work efficiency, and reduce risk to humans from the chelae. Xylazine-HCl and ketamine-HCl act rapidly so they can be readily applied in most research situations. Where injection is impractical, clove oil (≥0.125 ml/l) or AQUI-S™ (≥0.5ml/l) baths were effective in paralysing crabs although they both required around 20 min to act at optimal doses. A clove oil (≥0.125 ml/l) bath appeared to kill crabs humanely and is a useful option for research; crabs did not appear to experience trauma by this method and there was no limb loss or other damage. Of the methods tested, only clove oil and AQUI-S™ appear promising as treatments for the humane killing of crabs for human consumption, however both required long periods (≥28 min) to act which may limit their commercial application. Baker (1955) described a method for killing crabs for human consumption by sticking, which involves piercing the nerve ganglia with an awl. This was not attempted with P. gigas as the sternum is exceptionally thick and difficult to pierce. However, in other species sticking is likely to be a useful, and rapid, technique. References Ahmad, M.F. 1969. Anaesthetic effects of tricaine methane sulphonate (MS 222 Sandoz) on Gammarus pulex (L.)(Amphipoda). Crustaceana 16: 197-201. Baker, J.R. 1955. Experiments on the humane killing of crabs. J. Mar. Biol. Ass. U.K. 34: 15-24. Baker, J.R. 1962. Humane killing of crustaceans. Science 135: 589-593. Brown, P.B., White, M.R., Chaille, J., Russell, M. and Oseto, C. 1996. Evaluation of three anaesthetic agents for crayfish (Orconectes virilis). J. Shellfish Res. 15: 433-435. Chapman, C.R. 1992. Suffering in animals: towards comprehensive definition and measurement for animal care. pp. 19-25. In: T.R. Kuchel, M. Rose and J. Burrell (eds.). Animal Pain: Ethical and Scientific Perspectives. ANZCCART, Adelaide, South Australia. Clark, A. 1990. Gross examination in fish health work. pp. 9-45. In: B. Munday (ed.). Fin Fish Diseases. Proceedings 128, Post Graduate Committee in Veterinary Science, University of Sydney, Australia. Cook, C. 1996. Pain in farmed animals, awareness, recognition and treatment. pp. 75-81. In: R. Baker, R. Einstein and D. Mellor (eds.). Farm Animals in Biomedical and Agricultural Research. ANZCCART, Adelaide, South Australia. Cronin, T.W. and Forward, R.B.Jr. 1988. The visual pigments of crabs. II. Environmental adaptations. J. Comp. Physiol. A. 162: 479-490. Foley, D.M., Stewart, J.E. and Holley, R.A. 1966. Isobutyl alcohol and methyl pentynol as general anaesthetics for the lobster, Homarus americanus Milne-Edwards. Can. J. Zool. 44: 141-143. Gohar, H.A.F. 1937. The preservation of contractile marine animals in an expanded condition. J. Mar. Biol. Ass. U.K. 22: 295-299. Gunter, G. 1961. Painless killing of crabs and other crustaceans. Science 133: 327. Joseph, J., George, C. and Perigreen, P.A. 1989. Studies on minced fish-storage and quality improvement. J. Mar. Biol. Ass. India 31: 247-251. Lowe, T.E., Ryder, J.M., Carragher, J.F. and Wells, R.M.G. 1993. Flesh quality in snapper, Pagrus auratus, affected by capture stress. J. Food Sci. 58: 770-773. Mahoney, R. 1966. Laboratory Techniques in Zoology. Butterworths, London. 404 pp. Oswald, R.L. 1977. Immobilisation of decapod crustaceans for experimental purposes. J. Mar. Biol. Ass. U.K. 57: 715-721. Paterson, B.D., Exley, P.S. and Smith, R.A. 1994. Live Transport of Crustaceans in Air - Prolonging the Survival of Crabs. Fisheries Research and Development Corporation, and Queensland Department of Primary Industries (QDPI) Report Q094035, QDPI, Brisbane. 55 pp. Reilly, J.S. 1993. Euthanasia of animals used for scientific purposes. ANZCCART, Adelaide, South Australia. 73 pp. Ryan, S. 1992. The dynamics of MS-222 anaesthesia in a marine teleost (Pagrus auratus: Sparidae). Comp. Biochem. Physiol. 101C: 593-600. Schmidt-Neilsen, K. 1962. Humane killing of crustaceans. Science 135: 587-588. 183 Smaldon, G. and Lee, E.W. 1979. A Synopsis of Methods for the Narcotisation of Marine Invertebrates. Royal Scottish Museum Information Series, Natural History 6. Edinburgh, Scotland. 96 pp. Soto, C.G. and Burhanuddin. 1995. Clove oil as a fish anaesthetic for measuring length and weight of rabbitfish (Siganus lineatus). Aquaculture 136: 149-152. 184 Non-lethal Imaging Techniques For Crab Spermathecae 12 Research for this chapter has been published previously as: Gardner, C., Rush, M. and Bevilacqua, T. 1998. Non-lethal imaging techniques for crab spermathecae. Journal of Crustacean Biology, 18: 6469. 185 Abstract Techniques for collecting information on spermathecae without dissection were evaluated on the giant crab Pseudocarcinus gigas. Techniques tested were: biopsy, ultrasound, conventional x-radiography, computerised tomography (CT) scans, and magnetic resonance imaging (MRI). Attempts at biopsy and ultrasound imaging were unsuccessful. Spermathecae were imaged by x-radiography, although resolution was poor, suggesting that it can be applied only to gain general information, such as determining whether mating has occurred. High resolution images were produced with CT and MRI. Resolution by MRI is of such detail that internal structure of spermathecae is imaged. Nonlethal techniques allow animals to be used repeatedly, which permits monitoring of changes during mating, sperm storage, and extrusion. Introduction The more advanced brachyuran crabs possess paired spermathecae which store sperm between mating and fertilisation. The function of these organs is of biological interest as layered storage of separate ejaculates can affect paternity of offspring (Koga et al., 1993; Sévigny and Sainte-Marie, 1996). Spermathecae have also been the subject of fisheries-oriented research where they provide a means of measuring the occurrence of copulation, which may be affected when males are harvested. The ability of females to utilise stored sperm to fertilise separate broods may buffer the impact of male-only fisheries (Paul and Paul, 1992). Examination of spermathecae for this range of research has traditionally involved dissection and analysis (e.g. histology) of the spermathecae, or assessment of the fertilisation rate after eggs are extruded. This paper reports results of trials on a range of nonlethal techniques tested on the Australian giant crab Pseudocarcinus gigas (Lamarck). The objective was to develop techniques to examine spermathecae before ovulation, so infertility could be linked to either unviable sperm if spermathecae were full, or depleted reserves if spermathecae were empty. Nonlethal techniques allow for the repeated “sampling” of the spermathecae from the same individuals. In this way, changes can be tracked from mating, through storage, to extrusion. Nonlethal techniques tested were: biopsy, ultrasound, conventional x-radiography, computerised tomography scan (CTS), and magnetic resonance imaging (MRI). Materials and methods Female giant crabs Pseudocarcinus gigas, greater than 3.0 kg, were collected by commercial fishers from depths in the range of 300–380 m off the east coast of Tasmania (41°,15'S;148°40'E) in May 1994. Crabs used for trials with conventional x-radiography were alive and restrained by claw ties; in all other trials, crabs were killed in baths of clove oil in sea water (0.125 ml/l; Gardner, 1997). Following imaging, crabs were dissected to relate imaged structures with actual tissues. In the case of CT and MRI, crabs were frozen and sliced by band-saw (5 mm sections) to duplicate the orientation of the images which were collected as transverse slices. Attempts were made to biopsy contents of spermathecae using Pipelle de Cornier™ human endometrial biopsy catheters on dissected females with the gonoduct exposed. This allowed the biopsy tube to be observed as it was passed through the gonopore and along the gonoduct. Conventional x-radiography images were taken with standard veterinary equipment (AtomScope™ 100P). Ultrasound imaging was attempted with an 184 ATL-8™ ultrasound using 7, 5, and 3.5 Mhz transducers. Computerised tomography scanning was performed with a GE™ scanner and MRI was performed with a 1.5 Tesla, Picker™ magnetic resonance imager. Results The collection of stored sperm by biopsy was not possible without injuring the crab. Although the gonopore of giant crabs is large, it is calcified during intermoult and force was required to introduce a biopsy tube. Once within the lumen of the gonoduct, the tube tore through the delicate walls rather than bending with the curvature of the duct. Ultrasound penetrated the exoskeleton, but there was insufficient definition to discern the spermathecae. Factors contributing to the poor results with ultrasound appeared to be the echoing from internal calcified plates and the lack of sufficient acoustic difference between the spermathecae and surrounding tissue. Spermathecae were successfully imaged using conventional x-radiography, although images were very unclear (Fig. 1). Optimal exposure was relatively high in order to penetrate the carapace (3 LV, 0.1 SIC). Consequently, most tissue definition was lost and the boundary layer between the spermathecae and surrounding tissue became blurred. Figure 1. Image of female Pseudocarcinus gigas produced by conventional x-radiography showing spermathecae (S) as a pale region. High quality images of the spermathecae were made using both MRI and CT scans so that details of substructure could be detected, especially with MRI (Fig. 2). Spermathecae were dissected after imaging. The dark bands detected by imaging related to separate areas of seminal plasma which bounded areas of spermatophore deposit, possibly from separate ejaculates. The optimal MRI setting for giant crabs was considered to be scan protocol 14. Images produced by CTS were not as defined as those by MRI, but additional information can be gained by this technique, since internal calcified plates are detected (Fig. 3). Both MRI- and CTS- scanned images are collected as a series of “slices” through the specimen. Thus, 3-dimensional surface profiles can be composed and the resulting 3-dimensional images can be reconstructed, and cored or sliced along any plane (Fig. 4). 185 Figure 2. Dorsal (upper) and longitudinal (mid) images of a female Pseudocarcinus gigas produced by magnetic resonance imaging (MRI) showing high resolution of spermathecae (S). Schematic (lower) is of the longitudinal image produced by MRI. Separate regions (EJ1, EJ2, and EJ3) can be seen within the spermatheca which may relate to different ejaculates. 186 Figure 3. Dorsal (upper) and longitudinal (lower) images of a female Pseudocarcinus gigas produced by computerised tomography (CT) showing high resolution of spermathecae (S). Unlike MRI, CT is sensitive to calcified structures and the carapace is seen as a white band. 187 Figure 4. Three dimensional surface image of female Pseudocarcinus gigas reconstructed from CT scans. Both MRI and CT scans can be processed into three dimensional images and “sliced” along any plane. Discussion The thick exoskeleton of P. gigas impaired the use of both biopsy sampling and ultrasound. No entry location for the biopsy could be found other than via the gonopore, which was unsuitable, since the gonoduct tore when a biopsy tube was introduced. Ultrasound has been used in research of organisms with soft bodies, such as marine mammals and fish (Gales and Burton, 1987; Bonar et al., 1989), but it appears to be unsuited to crustaceans. Conventional x-radiography produced relatively poor images, since the boundaries between the spermathecae and surrounding tissue were blurred. Consequently, the exact size and form of the spermathecae could not be determined. Pseudocarcinus gigas have particularly thick carapaces, around 3.0 mm in relatively x-ray opaque larger females. The definition of spermathecae by conventional x-radiography may be greater in species with a thinner carapace. If higher resolution can be achieved, x-radiography may have application where the objective is simply to determine whether insemination has occurred. This can be important in experiments such as those reported by Koga et al. (1993) and Sainte-Marie and Lovrich (1994), where female crabs were dissected to confirm mating. The technique may also have application in studies assessing repeated spawning, using stored sperm where it is necessary to determine whether sperm reserves have been depleted. Conventional x-radiography is relatively inexpensive (around $US10 per exposure and several crabs can be x-rayed on each plate), and more readily available than the other feasible techniques, CT and MRI scans. Of the methods tested, only CT and MRI produced clear images of the spermathecae so that size could be measured and some internal structure viewed. The calcified exoskeleton did not interfere with imaging in either method. The shell was effectively transparent to MRI, since this technique relies on resonance of hydrogen nuclei (protons) which tend to be in low concentration within calcified tissues (Young, 1984). Resolution of internal structure was greater with MRI than with CT, and separate regions could be discerned which may relate to separate sperm deposits. Several authors have made observations based on separate sperm deposits in work investigating 188 sperm storage and competition in brachyurans (e.g., Paul, 1984; Diesel, 1989; Sévigny and Sainte-Marie, 1996). The technique of MRI provides the option of measuring or viewing contents of the spermathecae before, or perhaps even during, mating. Although logistically difficult, MRI may also have application in observing changes to spermathecae during the process of extrusion and fertilisation. An additional benefit from computerised scanning methods is that 3-dimensional plots may be generated, either of the external surface structure (CT), or of the internal boundary between shell and soft tissues (MRI). Although CT scans were of lower resolution than MRI, additional information is gained as calcified structures are detected. Where calcified structures are the subject of research, such as in the examination of skeletal growth of corals, CT scans are valuable (Logan and Anderson, 1991). In crustaceans, a benefit of imaging internal calcified structures is in the location of tissues, as calcified plates in the stomach and between limb muscles can serve as landmarks. Practical considerations in the use of MRI and CT imagery include the costs of the procedures and the restraint of the crabs to prevent movement during imaging. Both procedures are usually available from medical imaging facilities and fees for research use vary widely. Pseudocarcinus gigas is a highly valued crustacean, with individuals wholesaling at around $US100. With P. gigas, and other highly valued crustaceans, costs of imaging are offset by the reduced purchase of specimens, since sampling is nonlethal and the same individuals may be used repeatedly. While crabs were dissected after imaging in this study, this would not usually be necessary. Further reduction in costs can be made by scanning more than one individual simultaneously and this also increases the rate of specimen processing. In CT-scanning undertaken subsequently to this project, groups of 3 giant crabs were scanned simultaneously to examine ovarian development, and it was possible to process over 50 crabs per hour. Crabs must be motionless during scanning so paralysing agents such as xylazine-HCl, ketamine-HCL, procaine-HCl, or clove oil should be used (Oswald, 1977; Gardner, 1997). References Bonar, S.A., Thomas, G.L., Pauley, G.B. and Martin, R.W. 1989. Use of ultrasonic images for rapid nonlethal determination of sex and maturity of Pacific herring. North American Journal of Fisheries Management 9: 364-366. Diesel, R. 1989. Structure and function of the reproductive system of the symbiotic spider crab Inachus phalangium (Decapoda: Majidae): observations on sperm transfer, sperm storage, and spawning. Journal of Crustacean Biology 9: 266-277. Gales, N.J. and Burton H.R. 1987. Ultrasonic measurement of blubber thickness of the southern elephant seal, Mirounga leonina (Linn.). Australian Journal of Zoology 35: 207-217. Gardner, C. 1997. Options for humanely immobilising and killing crabs. Journal of Shellfish Research. 16: 219-224. Koga, T., Henmi, Y. and Murai, M. 1993. Sperm competition and the assurance of underground copulation in the sand-bubbler crab Scopimera globosa (Brachyura: Ocypodidae). Journal of Crustacean Biology 13: 134-137. Logan, A. and Anderson, I.H. 1991. Skeletal extension growth rate assessment in corals, using CT scan imagery. Bulletin of Marine Science 49: 847-850. Oswald, R.L. 1977. Immobilisation of decapod crustaceans for experimental purposes. Journal of the Marine Biological Association of the United Kingdom 57: 715-721. Paul, A.J. 1984. Mating frequency and viability of stored sperm in the tanner crab Chionoecetes bairdi (Decapoda, Majidae . Journal of Crustacean Biology 4: 375-381. Paul, A.J. and Paul, J.M. 1992. Second clutch viability of Chionoecetes bairdi Rathbun (Decapoda: Majidae) inseminated only at the maturity moult. Journal of Crustacean Biology 12: 438-441. 189 Sainte-Marie, B. and Lovrich, G.A. 1994. Delivery and storage of sperm at first mating of female Chionoecetes opilio (Brachyura: Majidae) in relation to size and morphometric maturity of male parent. Journal of Crustacean Biology 14: 508-521. Sévigny, J. and Sainte-Marie, B. 1996. Electrophoretic data support the last-male sperm precedence hypothesis in the snow crab, Chionoecetes opilio (Brachyura: Majidae). Journal of Shellfish Research 15: 437-440. Young, S.W. 1984. Magnetic Resonance Imaging: Basic Principles. Raven Press, New York, New York. Pp. 1-282. 190 Spermatogenesis and the Reproductive Tract of the Male Giant Crab Pseudocarcinus gigas 13 191 Abstract The reproductive tract of the male giant crab consists of paired elongate testes, located immediately below the hypodermis of the carapace, and convoluted vas deferens. The vas deferens consists of the anterior vas deferens (AVD; white), the mid-vas deferens (MVD; white), the posterior vas deferens (PVD; transparent), and the ejaculatory duct (ED). The ejaculatory duct exits at the ventral base of the fifth pereiopod. Sperm cells are produced in the testis and are carried by the seminiferous tubule to the AVD. Luminal secretions produced in the anterior portion of the AVD condense around clumps of sperm cells to produce a single, thin, envelope layer around the spermatophore. Spermatophores are ovoid, non-pedunculate, and contain numerous, closely packed sperm cells. A granular luminal substance is secreted in posterior portions of the AVD and this secretion, plus spermatophores, are stored in the MVD. The PVD and ED contain an agranular secretion which contains few, or no, spermatophores. This luminal substance in the PVD is likely to serve as the sperm plug in the spermathecae of females. Investigation of physiological maturity of giant crabs should focus on the presence of spermatophores in the MVD as this is the main site of spermatophore storage. Introduction Current management arrangements for the giant crab fishery include a minimum size limit, which is the same for both sexes, and restrictions on the taking of ovigerous females. Both these practices promote the harvest of males rather than females so there is concern that sperm limitation may occur in the residual population. In response to this concern, research has been undertaken on the sexual maturation of male giant crabs. This chapter contributes to that research and describes the anatomical and histological structure of the reproductive system. Numerous other papers have described the histology of the male reproductive tract of various crab species. However, there is species-specific variation in the gross anatomy of the tract, form of the spermatophore, luminal secretions, and storage location of spermatophores within the vas deferens (Uma and Subramoniam, 1984). This information is required to assess accurately the onset of physiological maturity, as determined by the production of spermatophores (Paul, 1992). Materials and methods Male giant crabs Pseudocarcinus gigas were obtained in March and July 1996 from commercial fishers who had captured the crabs at approximately 350 m depth off eastern Tasmania. Fifty-eight specimens were examined and these ranged in size from 81 to 208 mm carapace length. Crabs were killed in a bath of clove oil in seawater at 0.125 ml/l (Gardner, 1997). The reproductive tracts were carefully removed and examined for their gross anatomy and morphology. Tissue was preserved in 10% neutral buffered formalin for at least a week before processing and sectioning by standard paraffin histology. Sections were cut at 5 µm and stained with haematoxylin and eosin. As luminal secretions in the vas deferens tended to wash off during staining, slides were coated with albumin. Results 197 General morphology The male reproductive tract consists of paired testes and elongate, convoluted vas deferens (Figure 1). The paired testes are white and are interconnected medially by a commissure. They lie dorsal to the hepatopancreas and immediately below the hypodermis of the carapace, extending forwards from the middle of the antero-lateral border of the carapace, lateral to the stomach, and ending medially to the anterior vas deferens. The vas deferens is divided into 4 regions: the anterior vas deferens (AVD); mid vas deferens (MVD); posterior vas deferens (PVD); and ejaculatory duct (ED). The AVD is highly coiled into a roughly spherical mass and is an opaque dull white. It is approximately 1.5-2 mm in diameter. The MVD is more loosely coiled and of greater diameter (approximately 5 mm) than the AVD. This part is white and is the largest portion of the system, being marginally larger than the PVD. The PVD is transparent rather than white, clearly differentiating it from the MVD. It is swollen, highly convoluted, and around 3 mm in diameter. It is easily ruptured and extends directly to the ejaculatory duct through a crescent shaped aperture in the internal calcified plates of the fifth pereiopod. The ejaculatory duct passes through the musculature of the coxa to the penis which is located on the ventro-medial border of the coxopodite of the fifth pereiopod. The ejaculatory duct is a simple uncoiled tube and the diameter varies from 2 mm in the proximal region to 1 mm distally. Figure 1. Gross anatomy of the Pseudocarcinus gigas male reproductive tract, viewed dorsally. T - testis; AVD - anterior vas deferens; MVD - mid vas deferens; PVD - posterior vas deferens; ED - ejaculatory duct. The testis is dorsal to the hepatopancreas and stomach which have been removed to expose the vas deferens. 198 Histology of the reproductive tract Testis The testis consists of blind lobes, which may be divided into lobules, converging on a seminiferous tubule. Lobules have an outer layer of thin connective tissue tunica and an inner layer of squamous epithelium. The lumen of lobules contain cells in spermatogenesis; sections through distal regions of lobules are filled with spermatogonia while sections through more proximal regions are partially filled with sperm cells (Figure 2, upper). Sperm cells are not generally located within the centre of the lumen, rather, they tend to collect along one wall forming a lens-shaped area in transverse cross-section. The seminiferous duct has a cuboidal epithelium and the lumen is entirely filled with sperm cells (Figure 2, lower). Anterior vas deferens The histological structure of the AVD changes along its length although it is composed of three basic layers: an outer covering of connective tissue, a middle muscular layer, and an inner epithelial lining. These layers are also present in the MVD and the PVD. In the most anterior segment of the AVD, the epithelial lining is cuboidal with secretory globules in the cytoplasm. The lumen is filled with sperm mass, which is not always fully divided into spermatophore packets, and a luminal substance. The luminal substance appears to encapsulate the sperm masses by condensation (Figure 3, upper). In the mid-AVD, the epithelium becomes multinucleate columnar (25-30 µm high) with considerable glandular activity and sperm masses are more dispersed due to greater proportion of luminal substance. All sperm cells are massed into ovoid spermatophores with a distinctive envelope layer (110-230 µm on longest axis) (Figure 3, lower). The posterior AVD is similar with a columnar epithelium although this is more elongated (30-40 µm), also with glandular activity. Luminal secretions in the mid- and posterior AVD are granular and more eosinophilic than secretions in the anterior AVD. 199 Figure 2. Histology of the testis. Upper figure shows meiotic spermatogonia and spermatids (bar=35 µm). Lower figure shows a lobule with a seminiferous duct along one margin (bar=80 µm). 200 Figure 3. Histology of the anterior vas deferens (AVD). Upper - anterior segment of AVD with lumen filled with spermatids which are not yet formed into spermatophores. Luminal secretions have begun to encapsulate sperm masses (bar=80 µm). Lower - Mid section of the AVD. Sperm cells are massed into spermatophores although these remain closely packed (bar=80 µm). 201 Mid-vas deferens The anterior region of the MVD is initially similar to the AVD although the epithelium gradually becomes less elongated (from 30-40 µm to 25 µm high) so that it approaches a cuboidal type epithelium (Figure 4, upper). There appears to be little glandular activity in the mid-region of the MVD. The muscular layer is also thinner while the lumen is larger than in the AVD. The lumen is filled with a granular luminal substance and scattered spermatophores. Granules range in size from 2 to 100 µm although most are 10 µm (Figure 4, lower). Posterior vas deferens The connective tissue layer is thicker than in the MVD (20 µm). In the anterior third, the epithelial layer is low columnar with rounded infoldings into the lumen and appears to be secretory (Figure 5, upper). The epithelium gradually changes to a cuboidal epithelium with no secretory activity. Basophilic, simple branched acinar glands open into the lumen of the duct and are particularly numerous in the anterior third of the PVD. Luminal contents change along the length of the PVD with anterior portions filled with: a granular substance - apparently the same as that in the MVD; an agranular less eosinophilic substance; and occasional spermatophores. The agranular luminal substance appears to be secreted in the PVD and this displaces other luminal contents. After the first third of the PVD, spermatophores are no longer present within the lumen which is filled with the agranular substance (Figure 5, middle). Ejaculatory duct The ejaculatory duct (ED) is lined by a simple columnar epithelium (30-40 µm) which is produced into folds, more pronounced distally (Figure 5, lower). The connective tissue layer and muscle layer are thicker than in the PVD. The lumen contains a substance which is predominantly agranular and similar to that in the PVD. 202 Figure 4. Histology of the mid vas deferens (MVD). Upper- Spermatophores in the anterior MVD are well formed with a distinct envelope layer and are surrounded by luminal secretions (bar=80 µm). Lower - the lumen of the MVD is filled with granular secretions and spermatophores are now more widely separated than in the posterior region of the AVD (bar=80 µm). 203 Figure 5. Histology of the posterior vas deferens (PVD) and ejaculatory duct (ED). Upper - section through the anteriour PVD. Note that spermatophores are seldom seen in the lumen and that the epithelial cells are rounded with infoldings protruding into the lumen. The lumen is mainly filled with an agranular substance although there are occasional granules, similar to those in the MVD (bar=35 µm). Middle - mid PVD showing change in the epithelium to a low cuboidal epithelium (bar=35 µm). Lower ED with the lumen produced into folds and surrounded by a thick muscular layer (bar=450 µm). 204 Discussion The spermatophores of Pseudocarcinus gigas are typical of brachyuran crabs as they are non-pedunculate and vesicular, unlike anomuran spermatophores (Uma and Subramoniam, 1979; Subramoniam, 1993). A distinctive, single, envelope layer was present around the spermatophore which appeared to be deposited in the initial section of the anterior vas deferens. This envelope layer has been reported in Portunus sanguinolentus (Ryan, 1967), Geryon fenneri (Hinsch, 1988), Scylla serrata (Subramoniam, 1993), and Chionoecetes opilio (Beninger et al., 1988). An aggregation of sperm cells without an envelope layer has been reported in several species including the spanner crab, Ranina ranina (Ryan, 1984). There is also variation within the Brachyura in the number and density of sperm held in the spermatophore; P. gigas has a spermatophore similar to Ovalipes ocellatus, where numerous sperm cells are closely packed with little fluid between sperm (Hinsch, 1986). Luminal secretions of P. gigas appear to be similar to those of Scylla serrata, in both function and site of production. Uma and Subramoniam (1984) classified luminal secretions of S. serrata into four groups: A,B,C and D. These were produced in the most anterior portion of the AVD, the posterior portion of the AVD and the MVD, the PVD, and the ED respectively. The spermatophore envelope of P. gigas was secreted in the anterior portion of the AVD and appears to correspond to luminal substance A. The granular substance secreted in the posterior portion of the AVD appears to be luminal substance B, and this is considered to have a nutritive function (Jeyalectumie and Subramoniam, 1991). The PVD of P. gigas is filled with an agranular substance, which appears to be the luminal substance C described by Uma and Subramoniam (1984). Given the absence of spermatophores in luminal substance C, and the proximity to the ED, it is probable that this substance forms the sperm plug in the spermatheca of the female (Diesel, 1989). Uma and Subramoniam (1984) identified a fourth luminal secretion in the ED, which they labelled "D". In P. gigas, the luminal substance in the ED appeared to be the same as that in PVD ("C") and there appeared to be no justification to classify them separately based on histology alone. Observations on the presence or absence of spermatophores in the male reproductive tract are useful for fisheries management in establishing the onset of physiological maturity. For instance, Comeau and Conan (1986 and 1992) examined the vas deferens of tanner crabs Chionoecetes opilio to determine if spermatophores were produced, although they also proposed that the presence of spermatophores may not necessarily mean that the crab is functionally able to mate. Paul (1992) asserted that it was improbable that a crab would expend energy producing sperm unless there was some chance of using it. This discussion over the use of gonad maturity to determine the potential of males to mate is fuelled by observed differences in physiological and morphological maturity. That is, male crabs tend to produce spermatophores before they fully develop morphological secondary sexual characteristics such as enlarged chelae. There is potential for this debate to be obscured by the false classification of a crab as physiologically mature when only low numbers of spermatophores are produced, especially if the anterior vas deferens is examined rather than the main storage region. For instance, Van Engel (1990) determined physiological maturity of Callinectes sapidus by the presence of spermatophores in the AVD although seminal products are stored in the MVD in this species (Johnson, 1980). Although most brachyurans store spermatophores in the MVD, there is interspecific variation with some species storing spermatophores in the PVD (e.g. Libinia emarginata Hinsch and Walker, 1974; and 205 Uca lacteus Uma, 1978). Thus, it is clearly important to establish the site of spermatophore storage, before conducting sampling to determine physiological maturity. In P. gigas, spermatophores are formed in the AVD and stored in the MVD as evidenced by the high density of numerous, well formed spermatophores in this large region of the tract. References Beninger, P.G., Elner, R.W., Foyle, T.P. and Odense, H.P. 1988. Functional anatomy of snow crab (Chionoecetes opilio) reproductive systems, and a hypothesis for fertilisation. Journal of Shellfish Research 7, 196. Conan, G.Y., and Comeau, M. 1986. Functional maturity and terminal moult of male snow crab, Chionoecetes opilio. Canadian Journal of Fisheries and Aquatic Science 43, 1710-1719. Comeau, M., and Conan, G.Y. 1992. Morphometry and gonad maturity of male snow crab, Chionoecetes opilio. Canadian Journal of Fisheries and Aquatic Science 49, 2460-2468. Diesel, R. 1989. Structure and function of the reproductive system of the symbiotic spider crab Inachus phalangium (Decapoda: Majidae): observations on sperm transfer, sperm storage, and spawning. Journal of Crustacean Biology 9, 266-277. Gardner, C. 1997. Options for humanely immobilising and killing crabs. Journal of Shellfish Research 16, 219-224. Hinsch, G.W. 1986. A comparison of sperm morphologies, transfer and sperm mass storage between two species of crabs, Ovalipes occelatus and Libinia emarginata. International Journal of Invertebrate Reproduction and Development 10, 79-87. Hinsch, G.W. 1988. Morphology of the reproductive tract and seasonality of reproduction in the golden crab Geryon fenneri from the eastern Gulf of Mexico. Journal of Crustacean Biology 8, 254-268. Hinsch, G.W. and Walker, M.H. 1974. The vas deferens of the spider crab Libinia emarginata. Journal of Morphology 143: 1-20. Jeyalectumie, C., and Subramoniam, T. 1991. Biochemistry of seminal secretions of the crab Scylla serrata with reference to sperm metabolism and storage in the female. Molecular Reproduction and Development 30, 4455. Johnson, P.T. 1980. ‘Histology of the Blue Crab, Callinectes sapidus: A Model for the Decapoda’. pp. 440. (Praeger, New York.) Paul, A.J. 1992. A review of size at maturity in male tanner (Chionoecetes bairdi) and king (Paralithodes camtschaticus) crabs and the methods used to determine maturity. American Zoologist 32, 534-540. Ryan, E.P. 1967. Structure and function of the reproductive system of the crab Portunus sanguinolentus (Herbst). (Brachyura, Portunidae). I. The male reproductive system. Marine Biological Association of India, Symposium Series 2, 506-521. Ryan, E.P. 1984. Spermatogenesis and male reproductive system in the Hawaiian Crab Ranina ranina (L.). In ‘Advances in Invertebrate Reproduction’ (Ed W. Engels.) pp 629. (Elsevier, New York.) Subramoniam, T. 1993. Spermatophores and sperm transfer in marine crustaceans. Advances in Marine Biology 29, 129-214. Uma, K., 1978. Studies on comparative sperm morphology and spermatophores of crustaceans. Master of Philosophy Thesis, Madras University (unseen, cited by Uma and Subramoniam, 1984). 206 Uma, K., and Subramoniam, T. 1979. Histochemical characteristics of spermatophore layers of Scylla serrata (Forskal)(Decapoda: Portunidae). International Journal of Invertebrate Reproduction 1, 31-40. Uma, K., and Subramoniam, T. 1984. A comparative study on the spermatophore in Scylla serrata (Forskal) (Decapoda: Brachyura) and Clibanarius longitarsus (De Haan) (Decapoda: Anomura). Journal of the Marine Biological Association of India 26, 103-108. Van Engel, W.A. 1990. Development of the reproductively functional form in the male blue crab, Callinectes sapidus. Bulletin of Marine Science 46, 13-22. 207 Maturation of Male Giant Crab Pseudocarcinus gigas and the Potential for Sperm Limitation in the Tasmanian Fishery 14 208 Abstract Onset of sexual maturity of the male giant crab (Pseudocarcinus gigas) from Tasmania was assessed to determine whether protection is provided by the current minimum legal size of 150 mm carapace length (CL). Maturity was assessed by morphometric analysis of chela development, change in vaso-somatic index (VSI) with size, and production of spermatophores. Males were observed to develop through three morphological groups based on development of the molariform chela relative to CL: morphologically immature, morphologically adolescent, and morphologically adult. Onset of morphological adolescence occurs at 134 mm CL and this was not influenced by sample site (east and west Tasmania). Males produce spermatophores while they are morphologically immature and the mid vas deferens of all animals sampled greater than 90 mm CL contained numerous well-formed spermatophores. VSI increases with CL for morphologically immature and adolescent crabs. Male crabs appear to suffer higher incidence of cheliped loss than females, which may be due to competition for mates. Tank trials demonstrated that female giant crabs carry broods in successive years without moulting and are able to fertilise eggs using sperm stored for at least four years. Information on functional maturity of males, from observations of mating pairs, is required to fully assess the potential for sperm limitation in the fishery. Nonetheless, there are factors that reduce the risk of sperm limitation: both females and males are harvested; females do not moult every year thereby raising the operational sex ratio; and females can store sperm for extended periods. Introduction Crab fisheries are frequently managed by restricting catch to males as this was traditionally considered to have relatively little impact on the reproductive output of the remaining stock. More recently, concern for the effects of reduction of males has prompted research on male maturity in several established fisheries including those for Chionoecetes bairdi (Brown and Powell, 1972), C. opilio (Conan and Comeau, 1986; Sainte-Marie et al., 1995), Scylla serrata (Knuckey, 1996), Callinectes sapidus (Wenner, 1989; Van Engel, 1990), and Paralithodes camtschaticus (Paul and Paul, 1990). Indications of high-exploitation of male giant crab Pseudocarcinus gigas (Lamarck, 1818) has prompted similar research for this fishery. A critical step in estimating the risk of sperm depletion due to fishing is to determine the size at which male crabs are able to reproduce. Several methods have been used and the most direct is the observation of mating pairs, either in the field (Ennis et al., 1988; Oransanz et al., 1995), in tank based trials (Adams and Paul, 1983; Paul and Paul, 1996a), or from the presence of mating scars (Knuckey, 1996). These approaches are impractical with P. gigas for several reasons: they are seldom found in water shallower than 100 m; commercial fishers do not catch mating pairs in traps; no mating has been observed with males held in tanks with females for periods of 12 months; and no evidence of mating scars have been detected on males. Alternative methods for estimating size at which males are able to reproduce (i.e. functionally mature) were summarised by Paul (1992). They include examining the vas deferens for the presence of spermatophores (physiological maturity) and morphometric techniques such as ratios of reproductive tract weights and chela size to body size (morphometric maturity). Due to difficulty in determining functional maturity of P. gigas males, these alternative approaches were applied to assess the onset of maturity, 216 and also to investigate regional differences between the two main areas of the Tasmanian fishery. Aspects of the female reproductive biology, such as sperm storage in the spermatheca, also influence the potential for sperm limitation (Wenner, 1989; Paul and Paul, 1992; Sainte-Marie and Carriere, 1995). An additional objective of this study was to provide information on the viability of stored sperm in P. gigas and the cycles of egg extrusion and moulting. Materials and methods All crabs were collected by commercial fishers fishing for giant crabs in 200–350 m depth, along the rim of the continental shelf. Crabs were captured in traps from two sites (Fig. 1) off eastern and western Tasmania. Populations of adult P. gigas from these regions are relatively separate as few crabs are found in Bass Strait or around the southern region of Tasmania. Figure 1. Sampling locations in eastern and western Tasmania, Australia. Relatively few crabs are found in Bass Strait or around southern Tasmania indicating these populations are not continuous. 217 Morphological maturation Preliminary analyses were conducted on measurements of 296 male giant crabs to determine suitable parameters for assessing morphometric maturity. Parameters measured were: abdomen length; width of abdominal segment 1; width of abdominal segment 5; chela propodus width; chela propodus height; chela propodus length; and length of merus of the 5th pereiopod. The most suitable parameter for assessing morphometric development appeared to be chela propodus length (CPL) as other parameters either developed in a simple linear relationship with carapace length (CL), or showed the same pattern as CPL but with greater scatter (i.e., lower correlation with CL). CPL was measured on the molariform chela from the proximal edge of the lower articulation knob to the distal tip of the immovable finger. Sampling of CPL was then extended to a total of 533 crabs from eastern (N=186) and western (N=347) Tasmania ranging in size from 46 to 235 mm CL. All morphological sampling was conducted from March 1996 to June 1997. In addition to morphological measurements, missing pereiopods or left handed molariform chelae were recorded. Missing limbs without a blackened plaque over the wound were not recorded as these had occurred recently, probably during handling after capture. Female crabs were also sampled from the same location for limb loss and incidence of left handed molariform chelae (N= 114 and 319 respectively). Physiological maturation To determine the weight and maturity of gonads, 61 male crabs were obtained from eastern Tasmania in March and July 1996 ranging in size from 81 to 208 mm CL. Based on the analysis of morphological groups described below, this sample included 22 morphologically immature, 29 morphologically adolescent, and 10 morphologically mature crabs. Crabs were killed in a bath of clove oil in seawater at 0.125 ml l-1 (Chapter 11; Gardner, 1997) and immediately dissected to remove the reproductive tract. The paired testis and vas deferens were gently blotted and weighed separately to the nearest 0.1 g. Vas deferens weight was divided by CL to provide an index of vas deferens development: vaso-somatic index (VSI). Carapace length (CL), rather than whole weight, was used to scale the index of vas deferens weight as it is independent of development of the molariform chela. One vas deferens and one testis were preserved in 10% phosphate buffered saline for histological processing while smears from the other mid-vas deferens were examined microscopically for the presence of spermatophores, after staining with 5% neutral red. Specimens preserved in formalin were processed, sectioned, and stained by standard haematoxylin-and-eosin paraffin histology then examined for spermatogenesis. Males were classed as physiologically mature when the mid-vas deferens, the site of spermatophore storage in P. gigas, contained numerous well formed spermatophores, i.e. ovoid spermatophores with a complete, single envelope layer enclosing densely packed sperm cells, 110-230 µm on the longest axis (see Uma and Subramoniam (1984) for similar histology). The two methods used to prepare specimens for determining physiological maturity (smears and histology) provided confirmation of classification and there was no disagreement between readings with the two techniques. 218 Sperm storage by females To examine the viability of sperm stored for extended periods, 31 ovigerous females were captured in traps from depths in the range of 300-380 m off the east coast of Tasmania (41°15'S;148°40'E) in May 1994. Ovigerous females had only just begun to be observed by fishers so these egg masses were regarded as recently extruded. Females ranged from 2.2-3.5 kg and were maintained in two 4 m3 tanks with flow through water supply and fed twice weekly. Substrate was placed in tanks during ovipositioning to ensure adherence of eggs to pleopods (Shields, 1991). Females were individually tagged (see Levings et al., 1996) and monitored until September 1997 to record the viability of broods and patterns in reproductive biology. It was found that broods were produced annually and were held from around May to November; consequently, females were monitored for four reproductive cycles. Females were held without males for the duration of the experiment. To examine changes in the weight of spermathecae in relation to the size of the female, 90 females were obtained from eastern Tasmania between June 1994 and August 1995 (samples of around ten individuals per month except during seasonal closures in the fishery: September to December). These females were killed as described for males and dissected to remove the paired spermathecae which were blotted and weighed separately; the mean weight of the two spermathecae was used for analyses. A small proportion of female giant crabs do not produce eggs each year so females were classed as reproductively active or inactive based on the presence of eggs and the development of the gonad. This classification could only be done during months where crabs were ovigerous or in advanced gonadogenesis (i.e. excluding January and February when gonadogenesis was commencing). Analysis of data Morphological groups based on chela allometry were defined by iterative K-means cluster analysis (Everitt, 1974; Massart and Kaufman, 1983). Chela allometry of these groups was then defined by linear regression of CPL on CL (Sokal and Rohlf, 1981). The effect of site on morphometric development of chela was assessed by classic linear regression (Myers, 1990). Chela morphometric data have been log transformed in similar analyses with other brachyuran species (Comeau and Conan, 1992; Sainte-Marie et al., 1995). However, analysis of normal quantile plots indicated that regressions were best conducted with untransformed data in P. gigas. Slopes of regressions were compared by analysis of variance (ANOVA; Sokal and Rohlf, 1981) and the abscissa value for the point of intersection (I) between regression lines was calculated as I = (b2b1)(a1-a2)-1 , where a1 and a2 are the slopes and b1 and b2 are the Y-intercept values of the two regressions (Sainte-Marie et al., 1995). This procedure was also applied to VSI data, which were log transformed. The significance of differences between sexes in limb loss and incidence of left handed molariform chela were assessed by contingency table analysis using G-tests (Sokal and Rohlf, 1981). Results 219 Maturity in males and chela allometry Large numbers of spermatophores were present in the mid-vas deferens of all males greater than 90 mm carapace length (CL; N=59) and no spermatophores were detected in either animal sampled below 82 mm CL (N=2). The sample size of crabs around this CL is insufficient to pinpoint the size at onset of gametogenesis. However, onset of gametogenesis clearly occurs at a size considerably less than the minimum legal size of 150 mm CL (Fig. 2) and probably less than 110 mm CL for most males. Figure 2. Sampling to determine onset of physiological maturity (spermatophore production) in male giant crabs Pseudocarcinus gigas. Minimum legal size is 150 mm (dashed line) carapace length. Solid columns represent physiologically mature crabs, and shaded columns represent physiologically immature crabs. 6 4 2 0 150 mm - Carapace length (mm) Cluster analysis of points in the scattergram of chela propodus length (CPL) on CL defined three clear clusters which are termed morphologically immature, morphologically adolescent, and morphologically adult (Fig. 3). Site appeared to have no significant effect (P>0.70) on size of onset of adolescent chela morphology so data were combined for subsequent analyses. Slopes of morphologically adolescent and morphologically immature crabs were significantly different (P<0.0001) and intersected at 133.5 mm CL. There was considerable overlap of morphologically adolescent and adult males with the largest morphologically adolescent male 207 mm CL and the smallest morphologically adult crab 174 mm CL. Mean CL of morphologically adolescent and adult crabs were 172 mm and 204 mm respectively; that is, CL of morphologically adult crabs were around 19% larger than adolescents. 220 Figure 3. Scattergram of chela propodus length (CPL) on carapace length (CL) for samples of morphologically immature, adolescent, and adult male Pseudocarcinus gigas taken from May 1995 to June 1997. Data for crabs collected on eastern and western coasts of Tasmania are combined. Males were classified into morphological maturity groups by cluster analysis as described in text. Equations of regressions are: CPL = 1.245CL - 22.895 for immature males (lower scatter points, N = 32, r2 = 0.97, P < 0.0001); CPL = 2.318CL - 166.026 for adolescent males (middle scatter points, N = 291, r2 = 0.62, P < 0.0001); and CPL = 2.285CL - 86.762 for adult males (upper scatter points, N = 210, r2 = 0.544, P<0.0001). "I" represents the abscissa value at which the regressions of morphologically immature and adolescent males intersect. 500 Minimum legal size: 150 mm CL 450 400 350 300 1 1 11 1 11 11 11 1 1 1 1 111 11111 1 1 111 11 1 1 1 11 111 11111 1 1 1 1 1 11 1 11 1 1 1 11 1 1 1 11 1 1 1 11 11 1 111 111 11 1 1 1111 1 1 1 1 1 11 11 111 1 1 1 111 1 1 111 11 1 11 1 1 11 1 1 1 11 11 11 1 1 1 1 1111 1 111 11 111 1 1 11 1 111 1 11 1 1 111111 1 11 11 1 111 1 1111 11 1 11 1 11 1 11 1 111 1 1 111 1111 11 11 111 1 1 1 11 1 1111 11111 1 1 11 1 11 1 1 1 11 1 1 1 250 200 150 XXX X 100 50 X X XXX X X XX X X X X X XX XX XX X X X X X XX X X X X XXXX XX X X XXX X XX X XXX X X XXX X XX X X X XX X X XXX XX X XX XX X X X XXX XX X X XXX XX XX X X XXXX XX XX XX X X XX X XX X X XXXX XX X X XX XX XX X X XXX X XX XXXX X X X XX X XX X X X XXXXX X X XX X XX XXX XX X 1 11X 1 1X1 1 11 1 X X XX X 1 1 X X XX X XXX XXX X XXX X X X I=133.5 mm CL 0 40 60 80 100 120 140 160 180 200 220 240 Carapace Length (mm) VSI increased with male size for both morphologically immature and adolescent crabs (P<0.001; Fig 4). No effect of carapace size on VSI was detected for morphologically adult crabs although most of these crabs were of similar size and sample size was small. The slopes of regressions of log vaso-somatic index (VSI) against log CL were significantly different for morphologically immature and adolescent crabs (P<0.01) and they intersected at 140 mm CL (Fig. 4). Note that this abscissa value of 140 mm CL should be interpreted cautiously as it is largely an artefact of the morphological grouping variable which had an abscissa value of 133.5 mm CL. 221 Figure 4. Relationship between log vaso-somatic index (VSI; wet weight of the vas deferens / CL) and log carapace length (CL) of morphologically immature (E), adolescent (J), and adult (H) male giant crabs Pseudocarcinus gigas. Note that weight of vas deferens was scaled against CL, rather than body weight, to reduce confounding effects from chela development. Equations of regressions are: log (VSI) = 3.672logCL - 8.858 (N = 20, r2 = 0.69, P<0.0001) for morphologically immature males; and log (VSI) = 1.213logCL - 3.578 (N = 29, r2 = 0.23, P<0.01) for morphologically adolescent males. No significant effect of CL on VSI was detected for morphologically mature males (N=10; P>0.50). "I" represents the abscissa value at which the two regressions intersect. -0.4 Minimum legal size: 150 mm CL -0.6 -0.8 E E J JJ E E J EJ J E JJ J J J E J J J J E EE J J E J E E E -1 E -1.2 J J HJ J HH J J J JJ H HJ J H H H H H E -1.4 E -1.6 E -1.8 1.9 1.95 I=140 mm CL 2 2.05 2.1 2.15 2.2 2.25 Log Carapace Length (mm) 2.3 2.35 Limb loss and incidence of left handed molariform chelae Male crabs had significantly higher incidence of left-handed molariform chelae than females (P<0.05; Fig. 5). Although there appeared to be a trend of greater limb loss in male crabs, this was not significant (P>0.05). Figure 5. Proportion of male and female giant crabs Pseudocarcinus gigas with missing pereiopods (left) and with left handed, molariform chela (right). 0 2 1 3 4 L handed R handed 1 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 F (n=114) M (n=118) F (n=319) M (n=554) 222 Sperm storage by females and annual reproductive patterns Females held in tanks had an annual reproductive cycle with most females producing broods in successive years (Fig. 6). Around one quarter of females did not produce broods in the three seasons after capture and these individuals went on to moult in 1995/96. However, this pattern did not occur in 1996/97 when females that had not produced broods in 1996 were able to produce broods in 1997 without moulting. Sixteen of the seventeen crabs remaining alive at the end of the trial in September 1997 were able to produce fertile broods although there had been no contact with males since capture in May 1994. No attempt was made to measure fecundity as nutrition affects fecundity in crustaceans (Harrison, 1990) and the supplied diet of mackerel (Trachurus sp.) was unlike the natural diet which is primarily echinoderms, anomurans, and gastropods (Heeren and Mitchell, 1997). Nonetheless, in the final season, many females produced large broods that appeared normal. No females survived moulting; the spermathecae of these females remained full although it was clearly not possible to assess the viability of stored sperm retained through a moult. Figure 6. Reproductive history of 31 female giant crabs Pseudocarcinus gigas held in tanks without males from May 1994 to September 1997. Females were captured with eggs. Moulting occurred in January and February and all moulting individuals died. Embryogenesis was observed in all egg masses from females classed as ovigerous. 30 ovigerous 25 non-ovigerous 20 15 mortality 10 moult (mortality) 5 survived 0 1994 . 1995 . 1996 . 1997 Several of the dissected females appeared to be skipping a reproductive season as they did not have developing ovaries or broods during the brooding season (12.8% of females with sperm deposits; Fig. 7). Levings et al. (1996) stated that the size of onset of maturity of female crabs in Tasmania is around 115 mm CL and most of these reproductively inactive females were considerably larger. All females greater than 124 mm CL had deposits in their spermathecae. Weight of spermathecae increased with female size (P<0.0001) indicating that larger females have mated more often than smaller females and that they continue to accrue sperm reserves (Fig. 7). 223 Figure 7. Scattergram of average wet weight of spermathecae (SW) on carapace length (CL) of female giant crabs Pseudocarcinus gigas. Females with eggs or developing ovaries are represented by open circles (E) and reproductively inactive females are represented by crosses (1). The regression excludes the four females without sperm deposits and the equation is: SW = -5.712 + 0.0624 CL (N = 90, r2 = 0.41, P<0.0001). 8 ο 7 1 ο 1ο 6 ο ο ο ο ο ο ο ο οο ο ο ο ο 1 ο οο ο 1 ο ο 1 ο ο ο ο ο 1 ο ο ο1 ο οοοο1 ο ο ο ο ο1 οοοοο ο ο ο 1ο ο οο οο ο ο ο ο ο οο ο ο ο οοο οο1 ο οο οο ο ο 5 4 3 2 1 1 0 80 1 1 100 1 120 140 160 180 Carapace length (mm) 200 220 Discussion Maturation of males Male Pseudocarcinus gigas pass through three distinct morphological stages based on allometry of the chela which have been termed morphologically immature, morphologically adolescent, and morphologically adult. While the use of morphological indices for assessing maturity is controversial (see Paul, 1992; Comeau and Conan, 1992; Sainte-Marie et al., 1995), it is a useful method for determining whether there are regional differences in size of onset of maturity. Also, in the absence of information on functional maturity, morphological indices can provide the best estimate of size of maturity in some species (Somerton and Macintosh, 1983). There did not appear to be any regional effect (between east and west Tasmania) on size of onset of morphological adolescence in P. gigas. This suggests the information on physiological maturity of P. gigas obtained from specimens collected on the east coast may be equally applicable to west coast populations. To use morphological information for fisheries management, it is necessary to relate changes to moult increments and these relationships are not necessarily simple (Paul and Paul, 1995a). The mean CL of morphologically adult P. gigas was 19% larger than the mean CL of adolescent crabs. This is around the same size as the moult increment of male crabs reported by Levings et al. (22%; 1996) although it should be noted their estimate was drawn from only four individuals between 115 and 140 mm CL. Nonetheless, this indicates that the distinct stages of chela development correspond to different instars and similar patterns have been shown in numerous brachyurans, including other xanthids (Finney and Abele, 1981). The data presented indicate that the morphologically adult stage of P. gigas may be terminal as with Chionoecetes opilio (Sainte-Marie et al., 1995) although it is possible that additional moults may occur as with Chionoecetes bairdi (Paul and Paul, 1995b). These could result in morphological stages beyond the upper size limit sampled which is limited at around 225 mm CL by the legal maximum neck diameter of pots. 224 Unlike Chionoecetes spp., the morphological stages of P. gigas do not correspond to distinct stages in the onset of gametogenesis as crabs of all morphological stages had spermatophores in the mid vas deferens and possessed well developed gonads (Brown and Powell, 1972). Several authors have regarded the presence of spermatophores as indication of functional maturity (Hartnoll, 1969; Paul, 1992) although this relationship is debated in other species where morphological development seems to be an important determinant of functional maturity (Van Engel, 1990; Comeau and Conan, 1992; SainteMarie et al., 1995; Knuckey, 1996). In regards to P. gigas, the morphological stages of chela development represent three possibilities for fisheries management. First, physiologically mature males of immature morphology may be functionally mature. These animals are generally protected by the current minimum size (150 mm) so impacts of fishing will be slight. Secondly, males may not become functionally mature until morphological adolescence, despite the presence of spermatophores at much smaller sizes. Male P. gigas begin to exhibit morphological adolescence at around 135 mm CL with most morphological adolescents being larger than legal size, so little protection would be provided by current legislation. The third alternative is that only morphologically adult crabs are functionally mature. Current minimum size limits would be ineffective for protecting breeding male crabs in this alternative and most could be harvested as morphological adolescents well before reaching functional maturity. Without direct measures of mating, it is not possible to ascribe definitively the maturation of P. gigas to any of the options listed above, however, there are clues to their mating behaviour. Species with larger males relative to females, as in P. gigas, tend to have a polygamous, female-centred competition system (Martin and May, 1981; Christy, 1987). The large molariform chela of morphologically adult male P. gigas is of limited use for feeding (Heeren and Mitchell, 1997) and is used aggressively in tanks so that opposing males are killed although females have never been observed to damage each other (similarly with C. bairdi; Paul and Paul, 1996b). Regenerated molariform chela often regrow as incisiform chela so the greater incidence of "left-handed" molariform chela in male P. gigas may be a result of greater limb loss due to inter-male aggression (Cheung, 1976; Smith, 1990; Simonson and Hochberg, 1992; Abello et al., 1994). On the basis of these indirect observations on behaviour, it is probable that the morphological stages of development of the chelae have functional significance in mating and that only morphologically adolescent or adult crabs are important for fertilisation. Production of spermatophores before males are functionally mature appears to be common in brachyurans (Oransanz et al., 1995; Knuckey, 1996) so this is unlikely to indicate functional maturity in morphologically immature P. gigas. Spermatophore production at this stage may simply indicate the onset of gonadogenesis in preparation for mating at subsequent stages, as VSI increases sharply in relation to CL for morphologically immature crabs (note VSI was calculated using CL as the somatic scale so the relationship is not merely volumetric). The implications of removing large males from the population have caused concern in some crab fisheries as they are more successful in restraining females during the precopulatory embrace (Comeau and Conan, 1992; Hankin et al., 1996). The increase in P. gigas spermatheca weight with CL indicates that females mate on several occasions as they grow (Fig. 7) and spermathecae of large females contain separate regions of sperm deposit that appear to correspond to separate ejaculates (Gardner et al., 1998). Consequently, there is potential for large females to avoid mating if large males are removed by the fishery and if male size influences mating success (Christy, 1987). 225 Removal of large males by fishing may also lead directly to sperm limitation where small males replace larger males after competition for possession of females is reduced (Ennis et al., 1988). For instance, males in exploited populations of C. opilio have reduced sperm reserves, presumably due to more frequent mating (Sainte-Marie et al., 1995). More frequent mating is of limited concern where small males are able to inseminate several females (Sainte-Marie and Lovrich, 1994), however, this is not the case for all species. Paul (1992) noted that large male Paralithodes camtschaticus are capable of inseminating several females while smaller males may be limited to a single mating. The implications of the removal of large males from breeding populations of P. gigas are unknown so future monitoring of VSI as the fishery develops is advisable. Sperm storage Irregular moulting of female crabs and associated reproductive asynchrony have the potential to inflate the operational sex ratio and to increase competition for females (Wenner, 1989; Orensanz, et al., 1995). Inflation of the operational sex ratio occurs in Cancer gracilis where sperm can be retained through moults and is used to fertilise several broods (Orensanz, et al., 1995). P. gigas are also able to retain sperm through moults and stored sperm remains viable for extended periods, so that at least four annual broods can be fertilised. This prolonged viability of stored sperm in P. gigas, combined with the production of successive broods without moulting suggests that the operational sex ratio is inflated. This acts to reduce the risk of sperm limitation. Summary The onset of morphological adolescence occurred at around 135 mm CL in male crabs from both east and west Tasmania, which is slightly below the legal minimum size limit (150 mm CL). Although spermatophores are produced by morphologically immature P. gigas, it is unlikely that these crabs are functionally mature as VSI is still increasing at this stage. Also, spermatophore production by functionally immature males appears to be common among the Brachyura. Morphological adolescents and morphological adults are likely to be distinct moult stages with most animals at this stage larger than the legal minimum size. Aspects of the reproductive biology of females which reduce the risk of sperm limitation are: 1) successive annual broods are produced without moulting; and 2) sperm can be stored in the paired spermathecae, retained through moults, and remains viable for at least four years. References Abello, P., Warman, C.G., Reid, D.G. and Naylor, E. 1994. Chela loss in the shore crab Carcinus maenas (Crustacea: Brachyura) and its effect on mating success. Marine Biology, 121: 247-252. Adams, A.E. and Paul, A.J. 1983. Male parent size, sperm storage and egg production in the crab Chionoecetes bairdi (Decapoda, Majidae). International Journal of Invertebrate Reproduction, 6: 181187. Brown, R.B. and Powell, G.C. 1972. Size at maturity in the male Alaskan Tanner crab, Chionoecetes bairdi, as determined by chela allometry, reproductive tract weights and size of precopulatory males. Journal of the Fisheries Research Board of Canada, 29: 423-427. Cheung, T.S. 1976. A biostatistical study of the functional consistency in the reversed claws of the adult male stone crab, Menippe mercenaria (Say). Crustaceana, 31: 137-144. Christy, J.H. 1987. Competitive mating, mate choice and mating associations of brachyuran crabs. Bulletin of Marine Science, 41: 177-191. Comeau, M. and Conan, G.Y. 1992. Morphometry and gonad maturity of male snow crab, Chionoecetes opilio. Canadian Journal of Fisheries and Aquatic Sciences, 49: 2460-2468. 226 Conan, G.Y. and Comeau, M. 1986. Functional maturity and terminal moult of male snow crab, Chionoecetes opilio. Canadian Journal of Fisheries and Aquatic Sciences, 43: 1710-1719. Ennis, G.P., Hooper, R.G. and Taylor, D.M. 1988. Functional maturity in small male snow crabs (Chionoecetes opilio). Canadian Journal of Fisheries and Aquatic Sciences, 45: 2106-2109. Everitt, B. 1974. Cluster Analysis. Heinemann Educational Books, London. 122 pp. Finney, W.C. and Abele, L.G. 1981. Allometric variation and sexual maturity in the obligate coral commensal Trapezia ferruginea Latreille (Decapoda, Xanthidae). Crustaceana, 41: 113-130. Gardner, C. 1997. Options for humanely immobilising and killing crabs. Journal of Shellfish Research, 16: 219-224. Gardner, C., Rush, M. and Bevilacqua, T. 1998. Non-lethal imaging techniques for crab spermathecae. Journal of Crustacean Biology, 18: 64-69. Hankin, D.G., Butler, T.H., Wild, P.W. and Xue, Q.L. 1996. Does intense male harvest limit egg production of protected female stocks of Dungeness crabs ? High Latitude Crabs: Biology, Management, and Economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska Fairbanks, 507-510. Harrison, K.E. 1990. The role of nutrition in maturation, reproduction and embryonic development of decapod crustaceans: a review. Journal of Shellfish Research, 9: 1-28. Hartnoll, R.G. 1969. Mating in the Brachyura. Crustaceana, 16: 161-181. Heeren, T. and Mitchell, B.D. 1997. Morphology of the mouthparts, gastric mill and digestive tract of the giant crab, Pseudocarcinus gigas (Milne Edwards)(Decapoda: Oziidae). Marine and Freshwater Research, 48: 7-18. Knuckey, I.A. 1996. Maturity in male mud crabs, Scylla serrata, and the use of mating scars as a functional indicator. Journal of Crustacean Biology, 16: 487-495. Levings, A., Mitchell, B.D., Heeren, T., Austin, C. and Matheson, J. 1996. Fisheries biology of the giant crab (Pseudocarcinus gigas, Brachyura, Oziidae) in southern Australia. High Latitude Crabs: Biology, Management, and Economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska Fairbanks, 125-151. Martin, R.D. and May, R.M. 1981. Outward signs of breeding. Nature, 293: 7-8. Massart, D.L. and Kaufman, L., 1983. The Interpretation of Analytical Chemical Data by the Use of Cluster Analysis. John Wiley and Sons, New York. 237 pp. Myers, R.H. 1990. Classical and Modern Regression with Applications. 2nd Edition. Duxbury Press, Belmont, California. 488 pp. Orensanz, J.M., Parma, A.M., Armstrong, D.A., Armstrong, J. and Wardrup, P. 1995. The breeding ecology of Cancer gracilis (Crustacea: Decapoda: Cancridae) and the mating systems of cancrid crabs. Journal of Zoology, London, 235: 411-437. Paul, A.J. 1992. A review of size at maturity in male tanner (Chionoecetes bairdi) and King (Paralithodes camtschaticus) crabs and the methods used to determine maturity. American Zoologist, 32: 534-540. Paul, A.J. and Paul, J.M. 1992. Second clutch viability of Chionoecetes bairdi Rathburn (Decapoda: Majidae) inseminated only at the maturity moult. Journal of Crustacean Biology, 12: 438-441. Paul, A.J. and Paul, J.M. 1995a. Changes in chela heights and carapace lengths in mature male red king crabs Paralithodes camtschaticus after moulting in the laboratory. Alaska Fisheries Research Bulletin, 2: 164-167. Paul, A.J. and Paul, J.M. 1995b. Moulting of functionally mature male Chionoecetes bairdi Rathburn (Decapoda: Majidae) and changes in carapace and chela measurements. Journal of Crustacean Biology, 15: 686-692. Paul, A.J. and Paul, J.M. 1996a. Observations on mating of multiparous Chionoecetes bairdi Rathburn (Decapoda: Majidae) held with different sizes of males and one clawed males. Journal of Crustacean Biology, 16: 295-299. Paul, J.M. and Paul, A.J. 1990. Breeding success of legal and sublegal size male red king crab, Paralithodes camtschatica (Tilesius, 1815)(Decapoda, Lithodidae). Journal of Shellfish Research, 9: 2932. Paul, J.M. and Paul, A.J. 1996b. A note on mortality and injury rates of male Chionoecetes bairdi (Decapoda, Majidae) competing for multiparous mates. High Latitude Crabs: Biology, Management, and Economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska Fairbanks, 343347. Sainte-Marie, B. and Carriere, C. 1995. Fertization of the second clutch of eggs of snow crab, Chionoecetes opilio, from females mated once or twice after their moult to maturity. Fishery Bulletin, 93: 759-764. 227 Sainte-Marie, B. and Lovrich, G.A. 1994. Delivery and storage of sperm at first mating of female Chionoecetes opilio (Brachyura, Majidae) in relation to size and morphometric maturity of male parent. Journal of Crustacean Biology, 14: 508-521. Sainte-Marie, B., Raymond, S. and Brêthes, J.C. 1995. Growth and maturation of the benthic stages of male snow crab, Chionoecetes opilio (Brachyura: Majidae). Canadian Journal of Fisheries and Aquatic Sciences, 52: 903-924. Shields, J.D. 1991. The reproductive ecology and fecundity of Cancer crabs. In Crustacean Egg Production, pp. 193-213. Ed. by A. Wenner and A. Kuris. A.A. Balkema, Rotterdam. Simonson, J.L. and Hochberg, R.J. 1992. An analysis of stone crab (genus Menippe) claws and their use in interpreting landings on Florida's west coast. In Proceedings of a Symposium on Stone Crab (Genus Menippe) Biology and Fisheries, pp. 26-50. Ed. by T.M. Burt. Florida Marine Research Institute no. 50. 118 pp. Smith, L.D. 1990. Patterns of limb loss in the blue crab, Callinectes sapidus Rathburn, and the effects of autotomy on growth. Bulletin of Marine Science, 46: 23-36. Sokal, R.R. and Rohlf, F.J. 1981. Biometry. 2nd ed. W.H. Freeman, New York. 859 pp. Somerton, D.A. and MacIntosh, R.A. 1983. The size at sexual maturity of blue king crab, Paralithoides platypus, in Alaska. Fishery Bulletin, 81: 621-628. Uma, K. and Subramoniam, T. 1984. A comparative study on the spermatophore in Scylla serrata (Forskal) (Decapoda: Brachyura) and Clibanarius longitarsus (De Haan) (Decapoda: Anomura). Journal of the Marine Biological Association of India, 26: 103-108. Van Engel, W.A. 1990. Development of the reproductively functional form in the male blue crab, Callinectes sapidus. Bulletin of Marine Science, 46: 13-22. Wenner, E.L. 1989. Incidence of insemination in female blue crabs, Callinectes sapidus. Journal of Crustacean Biology, 9: 587-594. 228 Ovarian Development and Female Reproductive Biology 15 229 Abstract The gross anatomy of the female reproductive tract of the giant crab Pseudocarcinus gigas is typical of brachyurans. The mating system was investigated and although copulation was never observed, indirect evidence from the anatomy of the vagina and apparent mating scars indicates that mating occurs while the female is soft-shelled, which is consistent with other members of the family Eriphiidae. Females produce broods at more than one instar and retain ejaculates through moults. Ejaculates are stored in paired ventral-type spermathecae. Separate regions of sperm cells were observed in the spermathecae and these were presumed to be separate ejaculates. The number of these presumptive ejaculates increased with female size with up to three regions identified in some individuals. Internal sperm plugs were apparent although external sperm plugs were never observed. However, this may have been an artefact of sampling as all stored spermatophores had dehisced (released spermatids), indicating that mating had not occurred recently. Patterns of ovarian development and observations of crabs held in tanks demonstrated that the reproductive cycle is annual, although females may fail to produce a brood each reproductive season. Hepatosomatic index (weight of hepatopancreas relative to total weight) was reduced in response to ovarian development although there was no evidence of depletion of reserves in the hepatopancreas during gonad development. Moisture content of the hepatopancreas declined as gonad development proceeded and protein concentration (dry weight) remained stable. Carotenoid concentration in the hepatopancreas showed strong seasonal trends although these appeared unrelated to vitellogenesis as trends were similar between reproductively active and inactive crabs. Ovarian carotenoid concentration increased during vitellogenesis and peaked immediately prior to extrusion. Introduction Information on reproductive biology is essential for management of exploited crab species as it forms the basis of biologically rational controls on fishing effort, especially for input controls such as closed seasons and size limits. The only published information on female reproduction in Pseudocarcinus gigas (Lamarck, 1818) is by Levings et al. (1996) who reported seasonal patterns of capture of ovigerous females, determined a regression of egg mass weight against carapace length, and noted that not all females become ovigerous during the incubation period (stated as June/July to October/November). They also stated that females migrate from 270 m depth to 210140 m to release larvae, and that they mate while soft-shelled in June and July, although no data were presented in support. The genus Menippe is in the same family as P. gigas and they are the most taxonomically similar exploited crab species (Family Eriphiidae7). A considerable amount of information has been published on Menippe species8 which serves as a useful guide to probable reproductive patterns in P. gigas. However, while there is similarity 7 Eriphiidae is synonymous with Oziidae and Menippidae (Holthuis 1993). 8 Menippe adina was separated from Menippe mercenaria in 1986 so earlier studies referring to M. mercenaria may have been based on M. adina, or on hybrid material (Williams and Felder, 1986; Wilber, 1989a). 240 in reproductive biology within taxa, aspects such as mating behaviour may vary within brachyuran families (Norman, 1996) and even within genera (Orensanz et al., 1995). Given the limited biological information on reproduction in female P. gigas, it was considered useful to investigate fundamental aspects, as well as issues relating directly to fisheries management. The anatomy of the reproductive tract is described with emphasis on the spermathecae as the structure of this organ reflects aspects of the mating system (Jensen et al., 1996). Additional research on the mating system was conducted by maintaining crabs in tanks for several years in an attempt to observe mating. Information on the mating system is also useful for managing the harvest of male crabs as males tend to suffer higher exploitation and there is concern that sperm limitation may result (Chapter 14). The operational sex ratio can be lowered, and thus the risk of sperm limitation increased, by factors such as mating while the female is soft-shelled (during short seasonal periods) and by prolonged mate guarding behaviours (Christy, 1987). Seasonal patterns of development were investigated by holding animals for prolonged periods in tanks and from changes in the size, composition and histology of dissected ovaries. Seasonal changes in the hepatopancreas were also monitored as this organ is considered to be an important organ for storage of nutrients prior to gonadogenesis, and its composition tends to follow opposite trends to the ovary (Kulkarni and Nagabhushanam, 1979; Castille and Lawrence, 1989; Dy-Penaflorida and Millamena, 1990). Materials and methods Collection of specimens for dissection A total of 101 female crabs were captured in traps set at 150-300 m by a commercial fisher. All samples were collected from the east coast of Tasmania (41°10'-42°00'S; 148°30'-149°00'E) between May 1994 and August 1995. It was intended to collect at least 10 crabs monthly although this was not always possible due to poor weather or closed seasons (Table 1). Specimens ranged from 83 to 203 mm carapace length (CL). Commercial fishers reported that not all crabs became ovigerous during the winter months. To improve clarity of results in gonad analyses, females were classed as reproductively active or inactive, on the basis of gonad development. As this classification discriminated on the basis of gonad development, it was theoretically possible for a female crab to be ovigerous, yet classed as having inactive gonads - for the next spawning season. This distinction was not clear in samples collected in June and August as gonadogenesis was not advanced. Consequently, no crabs sampled in these periods were classed as inactive. Overall, 15 of the total 101 female crabs were classed as inactive. Table 1. Number of crabs sampled in each month from May 1994 to August 1995. 1994 Active Inactive M J 3 9 J A 9 S 1995 O 8 N D J F M A M J 12 12 3 9 4 12 1 2 2 1 5 J A 9 241 Total 3 9* 9* 8 (6*) 13 14 5 10 9 12 (10*) 9* * number of ovigerous crabs in sample Assessing maturity Crabs were killed in a bath of clove oil in seawater at 0.125 ml l-1 (Chapter 11; Gardner, 1997a), measured (carapace length), weighed, and then dissected to remove the hepatopancreas, ovary, and spermathecae. Where crabs were ovigerous, the egg mass was removed before the female was weighed to obtain whole weight. The hepatopancreas and ovary were then weighed and subsamples were either preserved in 10% phosphate buffered formalin for histology, or frozen (-60°C) for biochemical analyses. Indices of gonad (GSI) and hepatopancreas (HSI) size were calculated as: (wet weight of organ / total body weight) x 100. Ovarian tissue and bisected spermathecae were processed by standard haematoxylin and eosin, paraffin histology and sections were cut at 7 µm. The development of oocytes was quantified from histological sections by measuring the area of 50 large oocytes within the ovary by image analysis using NIH-Image™ 1.6 software. As the diameter of oocytes was often greater than the thickness of sections, it was inevitable that many oocytes would be not be sectioned through their medial plane. This problem was partially overcome by only selecting large oocytes for measurement. Spermathecae were preserved in Davidson’s fixative. Distinct regions of sperm deposits were apparent in histological sections of the spermathecae and these may have been separate ejaculates although biochemical techniques, such as electrophoretic analysis of enzymes, is required to confirm this (Sévigny and Sainte-Marie, 1996). The number of these presumptive ejaculates was counted for each female. Composition of ovary and hepatopancreas All analyses were duplicated and averaged. Where the duplicate analyses differed by more than 10% the analysis was repeated. To determine water content, samples were dried at 80°C for 24 h, with a final 1 h vacuum period before weighing. Samples of around 2 g were ashed at 450°C for 2 h. Samples for protein and carotenoid analyses were homogenised with a mortar and pestle. Protein was assayed by a modified Lowry procedure (Peterson 1977; Sigma Diagnostics™ #5656; Appendix 10). Carotenoids were extracted from tissue with acetone (Appendix 10). The acetone extract was partitioned with diethyl ether which was then washed with 20 volumes of 10% NaCl to remove residual acetone. Six samples of carotenoid extract from well developed ovaries collected in May 1994 and 1995 were chromatographed by thin layer chromatography (TLC) on C8 octyl silica plates (Merck™) using a solvent mixture of 95 : 5 petroleum ether : methanol. Tissue extracts were run alongside saponified extracts (5% ethanolic KOH for 24 h at room temperature) and astaxanthin standard (Roche Pharmaceuticals™). Carotenoids present in extracts were identified as predominantly astaxanthin or astaxanthin esters. Consequently, total carotenoids were estimated from the extracts as astaxanthin by measuring their absorption in diethyl ether at 472 nm assuming an E1%1 cm of 2099 (Clarke, 1977). Although lipids were to be analysed, this was not possible due to the destruction of specimens during a freezer malfunction. 242 Reproductive behaviour Three large, morphometrically mature (see Chapter 14) male crabs were placed with three female crabs in a 4 m3 tank from January 1995 to August 1995 and monitored daily for signs of precopulatory mate guarding. In addition, males were placed into 4 m3 tanks with females undergoing ecdysis in February 1995 and their behaviour monitored. To permit cycles of female reproduction to be observed in tanks, 31 ovigerous females were captured in traps from depths in the range of 300 – 380 m off the east coast of Tasmania (41°15'S;148°40'E) in May 1994 by a commercial fisher. These females ranged from 2.2 – 3.5 kg and were maintained in two 4 m3 tanks with flow through water supply and fed twice weekly. Females were individually tagged. These crabs were maintained in isolation from males until September 1997 (3 years, 4 months). Statistical analysis The effect of female size on oocyte area (assumed to infer timing of extrusion) was assessed by standard least squares regression for two separate samples: February and April, 1995. These were also tested for normality with a Shapiro Wilk w-test (Tietjen, 1986). Non-normal distribution would provide evidence of individuals at separate development stages, possibly due to a biannual cycle. ANOVA was used to test if female size differed between crabs with 1, 2 or 3 presumptive ejaculates. Results General morphology of the female reproductive tract The reproductive tract of female Pseudocarcinus gigas consists of the ovary, paired oviducts, spermathecae, vaginas and gonopores (Fig. 1). The ovary is surrounded by fibrous connective tissue and is located beneath the endocuticle, and dorsal to the hepatopancreas. The anterior lobes of the ovary extend to both lateral sides of the hepatic regions of the carapace. The posterior lobes extend laterally past the cardiac region to the intestinal region where they lie immediately lateral to the intestine. There is a medial commissure anteroventral to the heart. The ovary extends ventrally to the spermathecae so that the oviduct forms only a short tube of less than 10 mm passing along one side of the spermathecae. Spermathecae were ovoid, and were 41 mm on the longest axis in the largest specimen (203 mm CL). The oviducts join the spermathecae ventrally, thus the spermathecae are of the ventral type. The cuticular vaginas extend from the ventral surface of the spermathecae, through muscular tissue to the gonopores which are located on the sternal plastron of the 2nd walking leg. An external sperm plug was never observed in the vaginas of any specimen dissected here, or occluding the vulva in later sampling of 162 ovigerous female crabs (Gardner, 1997b). 243 Figure 1. Drawing of the gross anatomy of the female giant crab Pseudocarcinus gigas. The carapace, heart, and pericardiac sinuses were removed to expose the ovary. The spermathecae are largely obscured by the ovary but are located below the posterior lobes of the ovary, and lateral to the intestine. Histology of the spermathecae The spermathecae of all females greater than 124 mm CL contained sperm cells indicating that females had copulated at least once. Only 5 females were sampled below this size so it is not possible to assess the size at first mating. In all mated specimens, spermatophores had already dehisced and sperm cells were observed as a closely packed mass (Fig. 2). This suggests that either the spermatophores dehisce soon after copulation, or that none of the females sampled had mated recently. Masses of sperm cells were divided by basophilic proteinaceous matrix and separate sperm deposits appeared to be present, possibly as a result of separate copulations (Fig. 3). The number of these presumptive ejaculates ranged from 1 to 3 with larger females tending to have more presumptive ejaculates (P=0.062; Fig. 4). Females which moulted in isolation from males were dissected and found to have retained the contents of the spermathecae. 244 Figure 2. Sperm cells within the spermathecae. Note that the sperm cells are not bound into spermatophores. Figure 3. Spermathecae of giant crab Pseudocarcinus gigas bisected along the longitudinal axis. Note that three distinct deposits are present within the spermathecae, these are termed presumptive ejaculates. Mean carapace length (mm±s.d.) Figure 4. Relationship between size of female (carapace length) and the number of distinct regions of sperm deposit within the spermathecae. Distinct sperm deposits may be separate ejaculates. Number of females was 64, 24 and 8 for groups 1,2 and 3 respectively. 160 155 24 8 150 145 64 140 135 130 1 2 3 Presumptive ejaculates 245 Ovarian maturation The proportion of reproductively inactive crabs included in samples is influenced by feeding behaviour as ovigerous giant crabs are less likely to enter traps, as with many other species (Howard, 1982; Heasman et al., 1985; Schultz et al., 1996). Consequently, the ratio of reproductively inactive to reproductively active giant crabs is likely to biased and is of limited value. Nonetheless, it is noteworthy that reproductively inactive crabs spanned a range of sizes which suggests that some of these females were not simply immature, but were “skipping” a reproductive season (Fig. 5). Results from these females were kept separate in subsequent analyses (Figs. 7 and 8). Patterns in the gonadosomatic index (GSI) and oocyte area indicated that ovarian maturation is seasonal with increase in ovary size commencing shortly after extrusion (Fig. 7). Histological examination of the ovary indicated that oocytes separated from the germinal layer developed synchronously which indicates that only a single clutch is produced each spawning cycle (Fig. 6). For both years sampled, female giant crabs extruded eggs in late Autumn, during the months May to June. All females captured in May had highly developed ovaries while females captured in June were ovigerous. The hepatosomatic index (HSI: (wet weight of hepatopancreas/whole weight)x100) increased during spring (September to November) then slowly declined in summer (December to February). This decline in HSI occurred concurrently with a sharp increase in the gonadosomatic index (GSI) in summer (Fig. 7). The HSI of reproductively inactive females also declined at this time, but tended to be larger than the HSI of reproductively active females, presumably due to greater space available within the carapace in the absence of developed gonads. Figure 5. Weight of females in relation to carapace length (CL). Hollow symbols represent crabs with developing ovaries (n=86); solid symbols represent crabs classed as reproductively inactive (n=15). Whole weights of ovigerous crabs are with eggs removed. The regression is for both groups combined. 6000 y = 0.246x 2 - 31.816x + 1407.949 weight (g) 5000 4000 3000 2000 1000 0 75 100 125 150 175 200 carapace length (mm) 246 Figure 6. Oocytes from ovary in advanced development (April 1995). All oocytes are at a similar stage of development indicating that only a single clutch is produced each spawning cycle. Figure 7. Seasonal changes in gonado-somatic index (GSI), oocyte size (area in cross section), and hepato-somatic index (HSI) of female giant crabs Pseudocarcinus gigas. Hollow symbols represent crabs with developing ovaries or brooding egg masses; solid symbols represent crabs classed as reproductively inactive. N of each sample was variable and is listed in Table 1. mean GSI ± sd 10 7.5 5 2.5 0 200000 150000 100000 50000 0 M J J A SO N D J FMA M J J A 15 mean HSI±sd mean oocyte area (µm2 +sd) M J J A SO N D J FMA M J J A 250000 12.5 10 7.5 5 2.5 M J J A SO N D J FMA M J J A 247 There was no evidence of an effect of female size on timing of the reproductive cycle (assessed by oocyte area) for either February (n=12) or April (n=9) samples (P>0.4). However, it should be noted that sampling was not specifically designed to assess the effect of female size on ovarian development, so the sample sizes were small and the results are not definitive. Increase in ovary size, due to vitellogenesis, initially resulted in a decline in the moisture content of the ovaries and an increase in protein content (Fig. 8). Protein content peaked in December and then declined until ovaries were fully ripe in May. Total carotenoid content of the ovaries increased throughout gametogenesis. The hepatopancreas composition of reproductively active and inactive females were similar which suggests composition was not affected by vitellogenesis. Moisture content of the hepatopancreas was lowest during summer (December-February; Fig. 8) when HSI was relatively high. Total carotenoid composition of the hepatopancreas rose in spring (September-November; Fig. 8) in a similar pattern to the seasonal trend in HSI (Fig. 7). Figure 8. Seasonal changes in composition of giant crab Pseudocarcinus gigas ovaries and hepatopancreas. Protein and carotenoid composition are expressed as ash free dry weight. Hollow symbols represent crabs with developing ovaries or brooding egg masses; solid symbols represent crabs classed as reproductively inactive. N of each sample was variable and is listed in Table 1. non gon % moist GONAD HEPATOPANCREAS 90 80 80 70 70 60 60 50 50 40 M J J A SO N D J FMA M J J A M J J A SO N D J FMA M J J A non gon % prot 10 15 7.5 10 5 5 2.5 0 0 non gon carot M J J A SO N D J FMA M J J A M J J A SO N D J FMA M J J A 200 200 150 150 100 100 50 50 0 0 M J J A SO N D J FMA M J J A M J J A SO N D J FMA M J J A Reproductive behaviour Neither mating nor mate-guarding were observed in tank trials, and males did not appear to respond to females undergoing moult. Consequently, it was not possible to ascribe their mating behaviour to being either soft- or hard-shelled. However, in 248 additional research of fecundity of giant crabs (Gardner, 1997b), females were observed with melanised wounds around the gonopore (Fig. 9). These appear to be due to damage from the male’s pleopods as melanised lesions in the muscle beneath these wounds had histopathology consistent with spermatiferous granulomas, as occur in female sheep (Pers. Comm., Barry Munday, University of Tasmania). The sternum of females in intermoult is extremely thick so damage to the female sternum from the male’s pleopods could only occur when the female was soft-shelled (although this does not exclude the potential for hard-shelled mating). Figure 9. Apparent mating scars near the gonopores of a female giant crab Pseudocarcinus gigas. The abdomen has been pulled downwards (4 fingers are out of focus at the base of the figure) to expose the sternum. Two black lesions are visible immediately below the gonopores. Tank observations of females and evidence for annual reproductive cycle Detailed reproductive history of individual P. gigas females held for extended periods without males was presented previously in Chapter 14. In brief, most of the females held in tanks were observed to produce clutches of eggs annually, although females occasionally skipped a season. Skipping a reproductive season appeared to be associated with moulting as 6 of the 31 crabs moulted in January/February (late summer) after failing to produce a brood in the previous winter. General pattern of egg extrusion and hatching the same as observed in wild stocks, that is, eggs were extruded in May/June and hatched in October/November. Twelve of the 31 crabs produced 4 viable broods in successive years without moulting. The distribution of oocyte area data in samples from February and April 1995 provided additional evidence for an annual cycle of ovarian development. These data were distributed normally (after exclusion of reproductively inactive females), which would not occur if there was a bimodal distribution due to a biennial reproductive cycle. The small error bars for mean oocyte area (Figure 7) for most months are also consistent with a unimodal distribution. Also, all females with broods dissected in October 1994 (late Spring; n=8) had well developed ovaries indicating extrusion would occur the following Autumn. These results confirmed that females can produce broods annually, although they will occasionally skip a reproductive season. Discussion 249 Anatomical features The gross anatomy of the reproductive tract of female Pseudocarcinus gigas conforms to that of other brachyurans (Ryan, 1967; Diesel, 1989; Beninger et al., 1993; Jensen et al., 1996; Nagao et al., 1996). The ovary is bilobed and these lobes are connected by a central commissure, with the posterior lobes connecting to paired spermathecae which are linked to the gonopore by a vagina. The vagina is simple and without a bursa or other accessory structures. Mating system Due to the absence of direct observations of mating in P. gigas, it is not possible to ascribe conclusively the mating system to either hard or soft-shelled female (i.e. during inter-moult or at post-moult; Hartnoll, 1969). However, indirect evidence indicates that P. gigas may copulate while the female is at the soft post-moult stage: the vagina is of the simple type (Hartnoll, 1969); and the scars observed on the sternum of female P. gigas were consistent with soft-shelled mating. The apparent lack of a response by males to moulting females does not invalidate soft-shelled mating in P. gigas as the males were disturbed by moving them from a different tank, and the females had already commenced moulting before the male was introduced. Also, tank conditions in behavioural trials were far from natural as P. gigas inhabits deep water with most of the fishery based between 150 and 200 m. Although there are exceptions (e.g. hard shelled mating of the portunid crab, Thalamita sima; Norman, 1996), the mating system of crabs is generally consistent across family taxa so the system of P. gigas is likely to be the same as that of Eriphia smithii and Menippe species, the only crabs of the family Eriphiidae where the mating behaviour has been reported. Eriphia smithii and Menippe species (M. adina, M. mercenaria and hybrids) mate exclusively during the female’s soft post-moult phase (Binford, 1913; Porter, 1960; Savage, 1971; Tomikawa and Watanabe, 1992) and these species are not only in the same family as P. gigas, but also the same subfamily (Oziinae). Phylogenetic similarity between P. gigas and M. adina, M. nodifrons, M. rumphii and M. mercenaria was evident from studies of larval development (Chapter 3; Gardner and Quintana, 1998). Menippe species have mating behaviour typical of crabs which mate while the female is soft-shelled, such as protacted pre- and post-copulatory mate guarding, and lack of elaborate courtship behaviour (Hartnoll, 1969; Wilber, 1989a). The duration of precopulatory mate guarding appears to be scaled to size within the Cancridae (Orensanz et al., 1995) which suggests mate guarding may be protracted in the massive P. gigas. If male P. gigas guard females prior to copulation, the duration of this attendance is likely to be longer than for Menippe spp. (15 days; Wilber, 1989b; Wilber 1992), and possibly up to 21 days or more as this has been reported for the largest Cancer species, Cancer pagurus which inhabits similar open habitat in temperate regions (Edwards, 1966). While it appears likely that P. gigas mate while the female is soft-shelled, electrical stimulation of the gonopore of intermoult females causes the gonopore to dilate (unpublished research by Rudolf Diesel and the author). This demonstrates that control of the gonopore is muscular and that hard shelled mating is theoretically possible, although this has also been observed with species which mate exclusively while the female is soft-shelled (R. Diesel, Pers. Comm.). 250 Sperm competition and the functional anatomy of the spermathecae Diesel (1991) divided the spermathecae of higher brachyuran crabs into two groups, ventral and dorsal, based on the position of the attachment of the ovary and vagina. Both types of spermathecae are found in the super family Xanthoidea, which includes P. gigas, with dorsal spermathecae in the Pilumnidae (Diesel, 1991) and ventral type spermathecae in Eriphiidae (e.g. Menippe mercenaria; Wilber, 1989b). The presence of ventral type spermathecae in both P. gigas and M. mercenaria is consistent with their phylogenetic similarity indicated by larval development as mentioned previously (Chapter 3; Gardner and Quintana, 1998). Female xanthoid crabs typically have several post-pubertal instars (Tomikawa and Watanabe, 1992) and this occurs in P. gigas as moulting was observed of females which had previously produced broods. The progression through several post-pubertal moults by female P. gigas was also indicated by the broad size range of individuals with developing ovaries. Females which moulted were found to have retained the spermathecal contents as reported in M. mercenaria (Cheung, 1968). The trend towards an increase in number of presumptive ejaculates with female size may result from additional inseminations at each moult and the position of these presumptive ejaculates in the spermathecae indicates that sperm is displaced dorsally, which would result in last male sperm precedence. Diesel (1991) described two kinds of sperm plug: internal plugs which seal off previous ejaculates within the spermathecae; and external sperm plugs which extend up the vagina and out the gonopore. These plugs are generally considered to function in preventing paternity by other males (Jensen et al., 1996) although it has also been proposed that the plug is produced by the female (Bawab and El-Sherief, 1989) and that its role is to reduce damage to the female by repeated mating (Diesel, 1991). A basophilic, proteinaceous matrix without sperm cells or spermatophores was observed within the spermathecae of P. gigas females and this appeared to act as an internal sperm plug, sealing off separate ejaculates. However, no external sperm plug was observed despite the presence of an external sperm plug in other crabs in the family Eriphiidae (Tomikawa and Watanabe, 1990). External sperm plugs are short lived and function only within a single receptive period (Orensanz et al., 1995) so it is possible that they are present in P. gigas, despite the fact that none were observed in this research. As with Menippe species, P. gigas may produce 4 broods between moults fertilised by stored sperm so recently mated females may be encountered only occasionally (Porter, 1960; Chapter 14). This is consistent with unpublished analyses of tag recapture data of mature size female P. gigas which indicate that moulting may occur only every 4 years or more (R. McGarvey, Pers. Comm.). Consequently, if an external sperm plug is produced during mating, it may only be present in a small proportion of the population for a brief period each year. It is also noteworthy that no spermatophores were observed within the spermathecae, that is, they had already dehisced prior to sampling. This may be evidence of a prolonged delay between mating and sampling as spermatophores remain intact a considerable time after mating in Chionoecetes opilio (Beninger et al., 1988; Beninger et al., 1993), and until after dissolution of the sperm plug in Portunus pelagicus (Ryan, 1967). 251 Female reproductive cycle Ovigerous giant crabs are found from May/June to October/November (winter and spring) but around 50% of the females captured by commercial fishers during this period are non-ovigerous (Levings et al., 1996). Until 1995, this was interpreted as evidence for a biennial reproductive cycle until females maintained in tanks for the present study produced eggs in successive years, and ovigerous females were found to have developing ovaries (as cited by Levings et al., 1996). Additional evidence of an annual cycle is presented here with distinct annual trends in oocyte diameter and GSI, and no bimodal distribution in oocyte size (as occurs in biennial species; Jensen and Armstrong, 1989). The presence of non-ovigerous P. gigas females has been reported in populations of other species with strongly seasonal, synchronised reproduction (e.g. Cancer magister, Hankin et al., 1989) and in P. gigas it appears to be associated with moulting (Chapter 14). The proportion of non-ovigerous females in the population is likely to be less than the 50% determined by trapping surveys as ovigerous females tend to avoid traps (Howard, 1982). Seasonal synchronisation of reproduction patterns is typical of temperate decapods (less so in deep water; Haefner, 1978) although there is considerable variation in the duration of the reproductive cycle (Sastry, 1983). Smaller temperate crabs often produce several broods within a seasonally restricted period (Griffin, 1971) while larger crabs may have biannual cycles (e.g. Chionoecetes opilio, Sainte Marie, 1993; and Paralithodes platypus, Jensen and Armstrong, 1989), and even triannual cycles (Erimacrus isenbeckii, Nagao et al., 1996). An annual cycle, as in P. gigas, is typical of larger crabs found at similar latitudes (e.g. Ovalipes catharus, Armstrong, 1988). The timing of the female reproductive cycle may also vary with female size so that timing of larval release varies (Attard and Hudon, 1987; Bakir and Healy, 1995). From the analyses of oocyte size conducted in this study, there was no evidence that timing of oviposition in P. gigas was affected by female size (as in Scylla serrata; Heasman et al., 1985). However, this result is only a general indication as the sample size was small, and the size of extruded eggs is known to vary with female size which could confound analyses (Chapter 16; Gardner, 1997b). Ovary and hepatopancreas interactions Both HSI and GSI began to increase in spring, possibly in response to increased foraging behaviour and food supply associated with warmer water. This is supported by the observed decline in moisture content of the hepatopancreas during this period. Dichotomous trends in moisture content of the hepatopancreas and gonad due to transfer of nutrients during vitellogenesis have been observed in Scylla serrata, a tropical species (Nagabhushanam and Farooqui, 1982), but the trends in moisture content of the two organs are similar in Crangon crangon and P. gigas which are both temperate species (Haefner and Spaargaren, 1993). The effect of gonad development on moisture content of the hepatopancreas may be overshadowed in temperate species by seasonal change in food availability and consumption rate. HSI peaked in late Summer (February 1995 sample) and then declined while GSI continued to increase. It is tempting to attribute the decline in HSI to the transfer of nutrient reserves from the hepatopancreas to the gonad, and physical displacement by the enlarging gonad, but HSI of reproductively inactive crabs followed the same trend. Overall, HSI was lower in reproductively active crabs than in reproductively inactive 252 crabs, which is expected given the limited volume available within the body of a crab for ovarian development. Despite large fluctuations in the protein composition of the gonad, there was no evidence of transfer of protein from the hepatopancreas. Protein in the hepatopancreas remained relatively stable, which has been reported in other crabs, and there appeared to be no difference between reproductively active and inactive females (Heath and Barnes, 1970; Nagabhushanam and Farooqui, 1982; Mourente et al., 1994). Percentage protein content of the gonad began to decline several months before extrusion which has been reported in other decapods during late ovarian maturation (Dy-Penaflorida and Millamena, 1990). This relative decline in protein may be due to the deposition of lipid-rich lipovitellin in oocytes (Fyffe and O’Connor, 1974; Clarke, 1979; Castille and Lawrence, 1989). The absence of lipid analysis in the present study makes comment on the stages of vitellogenesis in P. gigas difficult, although the pattern of carotenoid and lipid concentration in the ovary may be linked. This is because carotenoids in decapod ovaries are usually incorporated into vitellogenin molecules and increase in concentration during ovarian development in proportion to the deposition of vitellogenin (Wallace et al., 1967; Fyffe and O’Connor, 1974; Komatsu and Ando, 1992; Dall et al., 1995). This deposition of carotenoid during vitellogenesis causes the ovary to change in colour which has been used as a guide to ovarian development by numerous authors (e.g. Meredith, 1952; Dayakar and Rao, 1992), although there is no consistent pattern in some species (eg. Scylla serrata: Heasman et al., 1985). The concentration of carotenoids in the gonad of P. gigas females increased rapidly from January until oviposition, presumably due to secondary vitellogenesis. Secondary vitellogenesis did not appear to result in depletion of carotenoids in the hepatopancreas, as concentration in reproductively active females was similar to that of reproductively inactive females. This indicates that carotenoid was not depleted from the hepatopancreas during vitellogenesis as has been reported in other decapods (Dall et al., 1995). References Armstrong, J.H. 1988. Reproduction in the paddle crab Ovalipes catharus (Decapoda: Portunidae) from Blueskin Bay, Otago, New Zealand. New Zealand Journal of Marine and Freshwater Research, 22: 529536. Attard, J. and Hudon, C. 1987. Embryonic development and energetic investment in egg production in relation to size of female lobster (Homarus americanus). Canadian Journal of Fisheries and Aquatic Sciences, 44: 1157-1164. Bakir, W.M.A. and Healy, B. 1995. Reproductive cycle of the velvet swimming crab Necora puber (L.) (Decapoda, Brachyura, Portunidae) on the east coast of Ireland. Irish Fisheries Investigations, Series B, 43. 13 pp. Bawab, F.M. and El-Sherief, S.S. 1989. Contributions to the study of the origin, nature and formation of the plug in the spermatheca of the female crab Portunus pelagicus (Linnaeus, 1766) (Decapoda, Brachyura). Crustaceana, 57: 9-24. Beninger, P.G., Elner, R.W., Foyle, T.P. and Odense, P.H. 1988. Functional anatomy of the male reproductive system and the female spermathecae in the snow crab Chionoecetes opilio (Fabricius) (Decapoda: Majidae) and a hypothesis for fertilisation. Journal of Crustacean Biology, 8: 322-332. Beninger, P.G., Lanteigne, C. and Elner, R.W. 1993. Reproductive processes revealed by spermatophore dehiscence experiments and by histology, ultrastructure, and histochemistry of the female reproductive system in the snow crab Chionoecetes opilio (Fabricius). Journal of Crustacean Biology, 13: 1-16. Binford, R. 1913. The germ cells and the process of fertilisation in the crab, Menippe mercenaria. Journal of Morphology, 24: 147-210. 253 Castille, F.L. and Lawrence, A.L. 1989. Relationship between maturation and biochemical composition of the gonads and digestive glands of the shrimps Penaeus aztecus ives and Penaeus setiferus (L.). Journal of Crustacean Biology, 9: 202-211. Cheung, T.S. 1968. Trans-moult retention of sperm in the female stone crab, Menippe mercenaria (Say). Crustaceana, 15: 117-120. Christy, J.H. 1987. Competitive mating, mate choice and mating associations of brachyuran crabs. Bulletin of Marine Science, 41: 177-191. Clarke, A. 1977. Lipid class and fatty acid composition of Chorimus antarcticus (Pfeffer) (Crustacea: Decapoda) at South Georgia. Journal of Experimental Marine Biology and Ecology, 28: 297-314. Clarke, A. 1979. Lipid content and composition of the pink shrimp, Pandalus montagui (Leach) (Crustacea: Decapoda). Journal of Experimental Marine Biology and Ecology, 38: 1-17. Dall, W., Smith, D.M. and Moore, L.E. 1995. Carotenoids in the tiger prawn Penaeus esculentus during ovarian maturation. Marine Biology, 123: 435-441. Dayakar, Y. and Rao, K.V.R. 1992. Breeding periodicity of the paddy field crab Oziotelphusa senex senex (Fabricius)(Decapoda: Brachyura), a field study. Journal of Crustacean Biology, 12: 655-660. Diesel, R. 1989. Structure and function of the reproductive system of the symbiotic spider crab Inachus phalangium (Decapoda: Majidae): observations on sperm transfer, sperm storage, and spawning. Journal of Crustacean Biology, 9: 266-277. Diesel, R. 1991. Sperm competition and the evolution of behaviour in Brachyura, with special reference to spider crabs (Decapoda, Majidae). In (Eds. R.T. Bauer and J.W. Martin) Crustacean Sexual Biology, Columbia University Press, New York. pp 145-163. Dy-Penaflorida, V. and Millamena, O.M. 1990. Variation in the biochemical composition of Penaeus monodon tissues during the reproductive cycle. The Israeli Journal of Aquaculture-Bamidgeh, 42: 84-90. Edwards, E. 1966. Mating behaviour in the European edible crab (Cancer pagurus). Crustaceana, 10: 23-30. Fyffe, W.E. and O’Connor, J.D., 1974. Characterisation and quantification of a crustacean lipovitellin. Comparative Biochemistry and Physiology, 47B: 851-867. Gardner, C. 1997a. Options for humanely immobilising and killing crabs. Journal of Shellfish Research, 16: 219-224. Gardner, C. 1997b. Effect of size on reproductive output of giant crabs Pseudocarcinus gigas (Lamarck) (Oziidae). Marine and Freshwater Research, 48: 581-587. Gardner, C. and Quintana, R. 1998. Larval development of the Australian giant crab Pseudocarcinus gigas (Lamarck, 1818)(Decapoda: Oziidae) reared in the laboratory. Journal of Plankton Research, 20(6), 1169-1188. Griffin, D.J.G. 1971. The ecological distribution of grapsid and ocypodid shore crabs in Tasmania. Journal of Animal Ecology, 40: 597-621. Haefner, P.A.Jr. 1978. Seasonal aspects of the biology, distribution and relative abundance of the deepsea red crab Geryon quinquedens Smith, in the vicinity of the Norfolk Canyon, Western North Atlantic. Proceedings of the National Shellfisheries Association, 69: 49-61. Haefner, P.A.Jr. and Spaargaren, D.H. 1993. Interactions of the ovary and hepatopancreas during the reproductive cycle of Crangon crangon (L.). I. Weight and volume relationships. Journal of Crustacean Biology, 13: 523-531. Hankin, D.G., Diamond, N., Mohr, M.S. and Ianelli, J. 1989. Growth and reproductive dynamics of adult female Dungeness crabs (Cancer magister) in northern California. Journal du Conseil International pour l’Exploration de la Mer, 46: 94-108. Hartnoll, R.G. 1969. Mating in the Brachyura. Crustaceana, 16: 161-181. Heasman, M.P., Fielder, D.R. and Shepherd, R.K. 1985. Mating and spawning in the mud crab, Scylla serrata (Forskål) (Decapoda: Portunidae), in Morton Bay, Queensland. Australian Journal of Marine and Freshwater Research, 36: 773-783. Heath, J.R. and Barnes, H. 1970. Some changes in the biochemical composition with season and during the moulting cycle of the common shore crab, Carcinus maenus (L.). Journal of Experimental Marine Biology and Ecology, 5: 199-233. Holthuis, L.B. 1993. The non-Japanese new species established by Wide Haan in the Crustacea volume of Fauna Japonica (1833-1850). In: T. and K. Baba, 1993, Ph.F. von Siebold and Natural History in Japan. Crustacea. T. Yamaguchi (Ed.). Carcinological Society of Japan, Tokyo, pp 599-646. Howard, A.E. 1982. The distribution and behaviour of ovigerous edible crabs (Cancer pagurus), and consequent sampling bias. Journal du Conseil International pour l’Exploration de la Mer, 40: 259-261. Jensen, G.C. and Armstrong, D.A. 1989. Biennial reproductive cycle of blue king crab, Paralithodes platypus, at the Pribilof Islands, Alaska and comparison to a congener, P. camtschatica. Canadian Journal of Fisheries and Aquatic Sciences, 46: 932-940. 254 Jensen, P.C., Orensanz, J.M. and Armstrong, D.A. 1996. Structure of the female reproductive tract in the Dungeness Crab (Cancer magister) and implications for the mating system. Biology Bulletin, 190: 336349. Komatsu, M. and Ando, S. 1992. Isolation of crustacean egg yolk lipoproteins by differential density gradient ultracentrifugation. Comparative Biochemistry and Physiology, 103B: 363-368. Kulkarni, G.K. and Nagabhushanam, R. 1979. Mobilisation of organic reserves during ovarian development in a marine penaeid prawn, Parapanaeopsis hardwickii (Miers) (Crustacea, Decapoda, Penaeidae). Aquaculture, 18: 373-377. Levings, A., Mitchell, B.D., Heeren, T., Austin, C., and Matheson, J. 1996. Fisheries biology of the giant crab (Pseudocarcinus gigas, Brachyura, Oziidae) in southern Australia. High latitude crabs: biology, management, and economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska, Fairbanks, pp. 125-151. Meredith, S.S. 1952. A study of Crangon vulgaris in the Liverpool Bay area. Proceedings and Transactions of the Liverpool Biological Society, 58: 75-109. Mourente, G., Medina, A., Gonzalez, S. and Rodriguez, A. 1994. Changes in lipid class and fatty acid contents in the ovary and midgut gland of the female fiddler crab Uca tangeri (Decapoda, Ocypodidae) during maturation. Marine Biology, 121: 187-197. Nagabhushanam, R. and Farooqui, U.M. 1982. Mobilisation of protein, glycogen and lipid during ovarian maturation in marine crab, Scylla serrata Forskal. Indian Journal of Marine Sciences, 11: 184186. Nagao, J., Munehara, H. and Shimazaki, K. 1996. Spawning cycle of horsehair crab (Erimacrus isenbeckii) in Funka Bay, Southern Hokkaido, Japan. High latitude crabs: biology, management, and economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska, Fairbanks, pp 315-331. Norman, C. 1996. Reproductive biology and evidence for hard-female mating in the brachyuran crab Thalamita sima (Portunidae). Journal of Crustacean Biology, 16: 656-662. Orensanz, J.M., Parma, A.M., Armstrong, D.A., Armstrong, J. and Wardrup, P. 1995. The breeding ecology of Cancer gracilis (Crustacea: Decapoda: Cancridae) and the mating systems of cancrid crabs. Journal of Zoology, London, 235: 411-437. Peterson, G.L. 1977. A simplification of the protein assay method of Lowry et al. which is more generally applicable. Analytical Biochemistry, 83: 346-356. Porter, H.J. 1960. Zoeal stage of the stone crab, Menippe mercenaria Say. Chesapeake Science, 1: 168177. Ryan, E.P. 1967. Structure and function of the reproductive system of the crab Portunus sanguinolentus (Herbst) (Brachyura: Portunidae). II. The female system. Proceedings of a Symposium on Crustacea, Marine Biological Association of India, Jan. 12-15, Ernakulum, Pt 2, pp 522-544. Sainte-Marie, B. 1993. Reproductive cycle and fecundity of primiparous and multiparous female snow crab, Chionoecetes opilio, in the Northwest Gulf of Saint Lawrence. Canadian Journal of Fisheries and Aquatic Sciences, 50: 2147-2156. Sastry, A.N. 1983. Ecological aspects of reproduction. In (Eds. F.J. Vernberg and W.B. Vernberg) The Biology of Crustacea, Academic Press, New York. Pp 179-270. Savage, T. 1971. Mating of the stone crab, Menippe mercenaria (Say)(Decapoda, Brachyura). Crustaceana, 20: 315-316. Schultz, D.A., Shirley, T.C., O’Clair, C.E. and Taggart, S.J. 1996. Activity and feeding of ovigerous Dungeness crabs in Glacier Bay, Alaska. High latitude crabs: biology, management, and economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska, Fairbanks, pp. 411-424. Sévigny, J.M. and Sainte-Marie, B. 1996. Electrophoretic data support the last-male sperm precedence hypothesis in the snow crab, Chionoecetes opilio (Brachyura: Majidae). Journal of Shellfish Research, 15: 437-440. Tietjen, G.L. 1986. A Topical Dictionary Of Statistics. Chapman and Hall, New York, 148 pp. Tomikawa, N. and Watanabe, S. 1990. Occurence of the sperm plugs of Eriphia smithii MacLeay. Researches on Crustacea, 18: 19-21. Carcinological Society of Japan. Odawara Carcinological Museum. Tomikawa, N. and Watanabe, S. 1992. Reproductive ecology of the xanthid crab Eriphia smithii McLeay. Journal of Crustacean Biology, 12: 57-67. Wallace, R.A., Walker, S.L. and Hauschka, P.V. 1967. Crustacean lipovitellin. Isolation and characterisation of the major high density lipoprotein from eggs of Decapoda. Biochemistry, 6: 15821590. Wilber, D.H. 1989a. Reproductive biology and distribution of stone crabs (Xanthidae, Menippe) in the hybrid zone on the northeastern Gulf of Mexico. Marine Ecology Progress Series, 52: 235-244. 255 Wilber, D.H. 1989b. The influence of sexual selection and predation on the mating and postcopulatory guarding behaviour of stone crabs (Xanthidae, Menippe). Behavioural and Ecological Sociobiology, 24: 445-451. Wilber, D.H. 1992. Observations on the distribution and mating patterns of adult stone crabs (Genus Menippe) on the northern gulf coast of Florida. Proceedings of a Symposium on Stone Crab (Genus Menippe) Biology and Fisheries, Florida Marine Research Publications, 50: 10-16. Williams, A.B. and Felder, D. 1986. Analysis of stone crabs: Menippe mercenaria (Say), restricted, and a previously unrecognised species described (Decapoda: Xanthidae). Proceedings of the Biological Society of Washington, 99: 517-543. 256 Effect of Size on Reproductive Output of Female Giant Crabs 16 Research for this chapter has been previously published as: Gardner, C. 1997. Effect of size on reproductive output of giant crabs Pseudocarcinus gigas (Lamarck) (Oziidae). Marine and Freshwater Research, 48: 581-587. 257 Abstract Fecundity and egg size of giant crabs (Pseudocarcinus gigas) were determined from egg masses of 162 crabs sampled from three sites in south-eastern Australia: western Victoria; western Tasmania; and eastern Tasmania. Crabs ranged in carapace length from 126 to 220 mm and egg number ranged from 830,000 to 2,500,000. Egg number and egg size increased with size of female. There appeared to be a decline in number of eggs and size of eggs with successive broods (inferred from carapace-condition) produced between moults. Sampling locality appeared to have little effect on reproductive output. Comparison of regressions of oocyte number in ovaries and number of eggs beneath the abdomen indicated there was no detectable loss of eggs during oviposition. Regression of an allometric model of log egg number to log crab size had a slope of 1.76 which was significantly less than 3.0. This indicates there is not a simple volumetric relationship between the variables, which would tend to occur if increasing fecundity with female size was a simple function of increased body space available for ovarian development. This pattern appeared to be a function of decreasing egg number and size with successive broods, and the trend of increasing egg size with female size. Ovigerous giant crab Pseudocarcinus gigas (photo by Karen Gowlett-Holmes) Introduction Female giant crabs produce broods annually (see previous chapter) and fishery restrictions currently exclude them from harvest while ovigerous. However, unlike many crab fisheries9, females may be harvested when they are not carrying broods. As most fishing effort for giant crabs is in February and March, prior to oviposition, exploitation rate of females may be high in some areas. High exploitation of females can affect recruitment so information on fecundity is critical for management. 9 e.g. Cancer magister in USA (Hankin et al., 1996); Lithodes santolla and Paralomis granulosa in Chile and Argentina (Vinuesa et al., 1996); Chionoecetes japonicus in Japan (Kon, 1996); and Chionoecetes opilio in Canada (Sainte-Marie et al., 1996). 262 Information on the fecundity-size relationship can be combined with growth information, size frequency, onset of maturity, and the proportion mature over the size range to develop fundamental management models such as egg per recruit and relative population fecundity (Annala and Bycroft, 1987). The objectives of this study were to assess the effect of the following factors on fecundity in giant crabs: (1) size of the female, (2) regional variation between three widespread sites around south-east Australia (Fig. 1) representative of the major fishing areas, and (3) successive broods produced between moults (moult frequency remains poorly understood). In addition, the effect of these same variables on egg size was examined. Figure 1. Location of sampling sites in western Victoria, western Tasmania, and eastern Tasmania. Latitude 42°30’S approximates the southern limit of the fishery for P. gigas. Materials and methods Sampling Ovigerous female crabs (n=166) were collected by commercial fishers using traps during the period 10 August to 15 September 1995 from three regions: eastern Tasmania (n=30), western Tasmania (n=121), and western Victoria (n=11) (Fig. 1). As Howard (1982) noted with Cancer pagurus, female P. gigas feed less when ovigerous and are less likely to enter pots. This behaviour appears to have affected sampling from western Victoria where only 11 crabs were collected despite considerable effort. In addition to sampling of ovigerous females, 13 females in late ovarian development but before extrusion, were collected for oocyte counts, in April and May 1995, from north-east Tasmania. 263 Preliminary research in 1994 investigated changes in eggs in relation to development which was categorised by the scheme of Subramoniam (1991). This preliminary study showed that although the size and mass of individual P. gigas eggs changed with development, relatively little change occurred in the first 2 stages of embryonic development in: dry weight, 3.4% compared with 11.0% throughout development; diameter, 1.8% compared with 7.5% throughout development; and composition, e.g. moisture, 2.2 % compared with 8.9 % throughout development. Consequently, sampling was restricted to eggs up to and including development stage 2 (Subramoniam, 1991). The following measures of size were recorded for all specimens: whole weight (with egg mass removed), carapace length (CL, mm), abdomen width, chela length, and chela height. The sample included 21 crabs missing 1 leg and 2 crabs missing 2 legs; the whole weight (W, g) of these individuals was adjusted by correcting for missing limbs with a regression formula, W = -4076.3 + 37.35[CL] + 0.0325[CL]2 (n=138, r2=0.95). Female P. gigas appear to produce clutches annually, and several clutches of eggs may be produced between moults (see chapters on sperm limitation and female reproduction). It was considered that fecundity may be altered in these successive clutches so a measure of shell wear was used to quantify time since the previous moult. Shell wear, or “carapace-condition”, was recorded using the following scheme: Carapace- clean bright shell; little to no fouling. If gooseneck barnacles present, then < condition 1 5 mm across longest axis, little to no wear apparent on the dactylus of pereiopods. Carapace- bright shell; fouling often heavy but composed almost entirely of gooseneck condition 2 barnacles; wear apparent on the dactylus of pereiopods with the bristles completely removed in places. Carapace- shell is often faded; fouling is heavy and is composed of many organisms condition 3 besides gooseneck barnacles, especially colonial ascidians and bryozoans; the dactylus of pereiopods is heavily worn with abrasion on the shell surface. It is important to note that the carapace-condition grade does not provide a measure of the actual number of clutches produced by a female since moulting. For instance, it is assumed that a carapace-condition 3 (heavy wear) female is likely to have had more clutches of eggs since moulting than a carapace-condition 1 female (clean shell), but it does not imply that the crab has produced 3 broods. Derivation of egg counts Half of each egg mass was removed from the crabs by severing four of the eight pleopods at their base in an alternate fashion, to remove possible bias from uneven distribution of eggs between pleopods. Eggs were then removed from the pleopods by severing setae at the point of junction with the pleopods and pooled. Mean individual egg dry weight was calculated based on the weight of two sub-samples of at least 250 eggs, counted and then weighed collectively (±10µg). Samples were dried at 80°C for 24 h, with a final 1 h vacuum period. Dry weight of the whole egg mass was estimated by weighing the blotted egg mass and then calculating mean moisture content from two sub-samples of 1.5 g. Total number of eggs per female was 264 then estimated from the values of mean individual egg dry weight and the dry weight of the whole egg mass. Derivation of oocyte counts The number of oocytes was compared with the number of eggs in abdominal clutches (after correcting for female size) to assess realised reproduction. That is, what proportion of the oocytes were lost during attachment to pleopod setae. Whole ovaries were removed by dissection and weighed to obtain whole wet weight. Number of oocytes in ovary was estimated for each sample as described for abdominal egg counts except that portions of the ovary were first fixed in Davidson’s fixative. This hardened the yolk so that oocytes could be teased apart and counted. Fixation resulted in leaching of some components from the oocytes, so it was necessary to calculate a correction factor by weighing the sub-sample before and after fixation. Egg diameter Sub-samples were taken from each abdominal egg mass and teased apart in sea water. The diameter of 50 eggs were then recorded by image analysis using NIH-Image™ 1.60 software. Only those eggs which were round and appeared normal were measured. Egg volume was calculated from egg diameter. Statistical analyses Four crabs, from western Tasmania, were excluded from analyses as they had few eggs and were not representative of the population. Three had a single small chela which appeared to be partially regenerated; chelae loss before extrusion has been shown to affect brood size so these individuals were excluded (Norman and Jones, 1993). The fourth crab had eggs attached to only a small proportion of available pleopod setae which suggested that extrusion may have taken place after capture and without a natural substrate present (Shields, 1991). The relationship between egg number and size was assessed with an allometric model of log fecundity (abdominal egg number) and log size (carapace length) as recommended by Somers (1991). Somers (1991) noted that the theoretical slope of linear regressions of log fecundity–log crab size should equal 3.0 as the relationship should be simply volumetric. Analysis of factors affecting the slope can provide additional information on the reproductive biology and this technique was applied to P. gigas. Analysis of factors affecting the slope of this model was by t-tests. In attempt to explain patterns of fecundity for fisheries management, the log fecunditylog size model was compared with models which included additional effects of carapace-condition, site, and interactions, using F-tests on the residual sum of squares (Draper and Smith, 1981). All additions to the model were tested using the residual mean square from a model incorporating all possible terms as the scale factor. This stepwise multiple regression is the same as that used for developing fecundity models currently in use for managing the Southern Rock Lobster Jasus edwardsii fishery (R. Kennedy, Tasmanian Department of Primary Industry and Fisheries, Pers. Comm., 1997). The effect of female size, site, carapace-condition, and interactions on individual egg dry weight and diameter were also assessed by stepwise multiple regression. As with egg number, data were log transformed to account for allometric relationships between reproductive traits. Additional curvature in regressions, after transformation, was tested 265 by including polynomial functions in stepwise multiple regression (Sokal and Rohlf, 1981). Where the categorical effects of site and carapace-condition were found to reduce error significantly (P<0.05), comparisons within the effects were made by correcting for size effects with analysis of covariance (ANCOVA). Comparisons could then be made between separate sites or carapace-condition by contrasts. Effect of size of crabs on carapace-condition was assessed by one way analysis of variance (ANOVA); where crab size had a significant effect the individual means were compared by Tukey-Kramer HSD. Where appropriate, power details (as β, PC type II error) are supplied for non-significant results (Searcy-Bernal, 1994). Comparison between oocyte number and egg number were made by comparing regression elevation differences with ANCOVA to correct for size of female (Sokal and Rohlf, 1981). All analyses were performed with JMP™ 3.1 software. 266 Results Fecundity model Egg number increased with adult size for all size measures used: whole weight, carapace length, abdomen width, chela length, and chela height (Appendix 11). As variation was similar for each measure of size, carapace length was selected for use in subsequent analyses as this measure of size is already used in the fishery for describing size limits (Fig. 2). Figure 2. Effect of body size of female giant crabs, measured as carapace length (mm), on fecundity (number of eggs). Trends were similar for all other measures of crab size: weight, abdomen width, chela length, and chela height. Ln Fecundity 15.0 14.5 14.0 13.5 4.8 4.9 5.0 5.1 5.2 5.3 5.4 Ln Carapace length The number of eggs produced by each female ranged from 706,000 (carapace length = 131 mm) to 2,545,000 (carapace length = 203 mm) with a mean of 1,575,000. Sample data from the different sites are summarised in Table 1. The hypothesis that increase in size leads to no increase in egg number was tested and rejected (P<0.001). Additional terms were then added to the model to attempt to explain variation. Inclusion of the effect of site in the model did not significantly improve model fit (P>0.1, β = 0.79) so site was not included in subsequent analyses. Inclusion of carapace-condition in the model significantly reduced the unexplained error compared to the model with carapace length only (P<0.001). This was due to a trend of decreasing fecundity with deteriorating carapace-condition (Table 2; Fig. 3). The effect of including a term for extra curvature ((ln carapace length)2) was tested but did not significantly improve model fit (P>0.3, β=0.13). Table 1. Characteristics of female giant crabs from three sites in southern Australia. Data (mean ± s.d.) for three grades of carapace condition pooled. Significantly different sites denoted by superscripts in ascending order of intercept on regression (P<0.05). Significance of difference in female size between sites not tested; female size used only as covariant to establish the effect of site on reproductive output. NS, not significant; β, power. Site P 267 East Tasmania (n=30) West Tasmania (n=117) West Victoria (n=11) Carapace length, mean ± SD (mm) 156 ± 13.7 183 ± 17.0 147 ± 16.6 - Weight, mean ± SD (g) 2488 ± 617.8 3859 ± 915.4 2072 ± 644.4 - Egg number, mean ± SD (millions) 1.35 ± 0.294 1.67 ± 0.435 1.07 ± 0.407 NS β=0.26 Egg number, coefficient of variation 31.94 26.93 34.20 - Egg mass dry weight, mean ± SD (g) 76.5 ± 26.2 99.1 ± 26.2 69.6 ± 28.7 NS β=0.45 Individual egg dry weight, mean ± SD (µg) 56.37 ± 5.03 58.17 ± 4.55 56.64 ± 5.88 NS β=0.06 Egg diameter, mean ± SD (µm) 562.6 ± 22.5a 600.0 ± 26.2b 579.2 ± 23.1b <0.001 β is power Table 2. Summary of characteristics of females of carapace-condition grades 1, 2 and 3 Data (mean ± s.d.) for three sites pooled. For qualitative carapace-condition grades, see text. Significantly different carapace conditions denoted by superscripts in ascending order (P<0.05). Analysis of the effect of carapace-condition on egg measures by ANCOVA to correct for the significantly different carapace length of females. For egg mass dry weight, the pattern of ascending superscripts relates to intercepts of the model, rather than to mean values, because this corrects for carapace length. Carapace-condition P 1 (n=53) 2 (n=78) 3 (n=27) Carapace length, mean ± SD (mm) 174 ± 17a 173 ± 22a 190 ± 10b <0.001 Weight, mean ± SD (g) 3462 ± 982a 3422 ± 1110a 4194 ± 609b <0.01 Egg number, mean ± SD (millions) 1.64 ± 0.40b 1.53 ± 0.52b 1.57 ± 0.43a <0.0001 Egg number, coefficient of variation 24.42 33.71 27.15 - Egg mass dry weight, mean ± SD (g) 99.7 ± 26.6c 89.1 ± 29.6b 92.4 ± 26.3a <0.001 Individual egg dry weight, mean ± SD (µg) 59.89 ± 3.56c 56.72 ± 4.69b 56.61 ± 5.74a <0.0001 Egg diameter, mean ± SD (µm) 597.2 ± 21.49 586.8 ± 33.78 594.0 ± 26.32 NS β=0.66 β is power. 268 Table 3. Parameter estimates of fecundity models for female giant crabs. My.x = βo + [CC]i + β1 ln [CL], where CC is carapace condition, CL is carapace length and i is 1, 2, 3 using ln transformed fecundity. Correction factor for converting the results of the model back to the original scale of measurement is 1.0045819 for the full model, and 1.0119281 for the model with length alone, i.e. predicted values (as numbers of eggs) are (eMy.x) 1.0045819. Parameter Full model ln length Estimated value SE Estimated value SE βo (intercept) 4.598535 0.56149 5.202213 0.56429 β1 (slope) 1.868524 0.10831 1.754983 0.10915 Carapace-cond.`offset: grade 1; 0.070893 0.01841 grade 2; 0.017130 0.01713 grade 3 0 269 Figure 3. Relationship between Ln fecundity and Ln carapace length for each carapace-condition in female giant crabs. Ln Fecundity (eggs) 15 Carapace-condition 1 14.5 Ο 14 ΟΟ Ο 13.5 13 15 Ln Fecundity (eggs) ΟΟ Ο Ο Ο ΟΟΟ ΟΟ Ο Ο Ο ΟΟ ΟΟ Ο Ο Ο Ο Ο Ο ΟΟΟΟΟ Ο ΟΟ ΟΟ Ο Ο ΟΟ Ο Ο Ο Ο Ο Ο Ο Carapace-condition 2 14.5 14 Ο 13.5 Ο Ο Ο Ο Ο ΟΟ Ο ΟΟΟ Ο ΟΟ Ο Ο Ο ΟΟΟΟΟΟ ΟΟ ΟΟ Ο Ο Ο Ο Ο Ο Ο ΟΟΟ Ο Ο ΟΟ Ο Ο ΟΟ Ο Ο Ο Ο ΟΟ Ο ΟΟ ΟΟ Ο Ο Ο Ο Ο Ο Ο Ο ΟΟΟΟΟ Ο Ο Ο Ln Fecundity (eggs) 13 15 Carapace-condition 3 14.5 Ο Ο 14 Ο Ο ΟΟ Ο ΟΟ ΟΟ Ο ΟΟ ΟΟ Ο ΟΟΟ ΟΟ ΟΟ Ο Ο Ο 13.5 13 4.8 4.9 5 5.1 5.2 5.3 Ln Carapace Length (mm) 5.4 In summary, the most accurate model for describing the egg number-size relationship incorporates the effects of log carapace length and carapace-condition (Table 3). The slope of this model was significantly less than 3.0, based on the associated 95% confidence limits (Table 4). Most of the model error was explained by inclusion of size with only a relatively slight improvement in fit achieved by incorporating carapacecondition (4%). As improvement in fit with carapace-condition was relatively small, albeit significant, model parameters are also presented for the model with carapace length alone (Table 3). A summary of the effect of carapace-condition on reproductive output is outlined in Table 2. Effect of female size on egg size Although there was a great deal of variation, the dry weight of individual eggs was significantly affected by female size (P<0.0001; Fig. 4). Incorporation of carapacecondition into the model for mean egg weight significantly improved fit (P<0.0001) while site did not (P>0.05, b = 0.12). All grades of carapace-condition were significantly different and there was a trend of declining egg size with a deterioration in carapace-condition (Table 2). Including a term for extra curvature ((ln carapace 270 length)2) significantly improved fit (P<0.05), suggesting the increase in egg size with female size levels out as the females become larger. Figure 4. Effect of size of female giant crab (carapace length) on egg size. (a) Effect of maternal size on egg dry weight: upper regression, carapace-condition 1 (r2=0.17), mid regression, carapace-condition 2 (r2=0.18), lower regression, carapace-condition 3 (r2=0.10). (b) Effect of maternal size on mean egg diameter: upper regression, western Tasmania (r2=0.09), middle regression, western Victoria (r2=0.85), lower regression, eastern Tasmania (r2=0.08). Ln Mean Egg Weight (µg) 4.3 4.2 4.1 4 3.9 3.8 Ln Mean Egg Diameter (µm) 3.7 6.6 6.5 6.4 6.3 6.2 4.8 4.9 5 5.1 5.2 5.3 Ln Carapace Length (mm) 5.4 Mean egg diameter increased significantly with size of the female (P<0.05; Fig. 4), and terms for extra curvature (β = 0.29) or carapace-condition (β = 0.56) did not significantly improve fit (P>0.05). The model was significantly improved by including site (P<0.001). This resulted from significantly smaller diameter eggs in the eastern Tasmania samples while eggs from western Victoria and western Tasmania were not significantly different (P>0.05). Realised reproduction There was no significant difference between the number of oocytes in the ovary and the number of eggs held under the abdomen which suggests that there is no egg loss during oviposition (Fig. 5; P>0.6, β = 0.16). 271 Figure 5. Comparison between fecundity estimated by the number of mature oocytes in the ovary, and the number of eggs held under the tail after oviposition. Linear regressions against ln carapace length are shown: the lower regression is for oocytes. Ln FECUNDITY 15.0 14.5 14.0 Eggs Oocytes 13.5 13.0 4.8 4.9 5.0 5.1 5.2 5.3 5.4 Ln CARAPACE LENGTH Analysis of factors affecting slope Figure X. Comparison between fecundity estimated by number of The slope of the allometric model of log carapace length–log fecundity, was 1.76 and significantly less (P<0.05) than the theoretical value of 3.0 (Somers, 1991). This appeared to be due to the observed decline in egg number with deterioration in carapace-condition and the observed increase in egg size with female size (Table 4). Both of these factors were reassessed by including carapace-condition and total egg volume (egg number by egg volume) in the model as independent and dependent variables respectively. Although inclusion of these factors produced a slope of 2.63, this was still significantly less (P<0.05) than the theoretical value of 3.0. Table 4. Regression statistics of reproductive output to evaluate whether the relationship between carapace length (CL) and fecundity (F) is volumetric with a slope of 3.0. Model 1a: simple allometric model of F predicted by CL; model 1b predicts total eggmass dry weight (W) from CL. Models 2a and 2b: as 1a and 1b, but with the addition of carapace-condition (CC) as an independent variable. Model 3 predicts volume of the whole egg-mass as the product of individual egg volume (Ve) and F from CL and CC. All continuous variables were log transformed. All models were tested against the null hypothesis (Ho) that the slope (β1) was not different from 3.0. Model Slope 95% Conf. Limits H0; β1=3.0 (t-test) r2 value (β1) on slope (L1-L2) 1a) y = ln F; x = ln CL 1.76 1.55-1.97 Reject 0.635 1b) y = ln W; x = ln CL 2.25 1.95-2.56 Reject 0.581 2a) y = ln F; x = ln CL & CC 1.87 1.66-2.08 Reject 0.675 2b) y = ln W; x = ln CL & CC 2.40 2.10-2.69 Reject 0.638 3) y = ln (Ve x F); x = ln CL & CC 2.63 2.31-2.95 Reject 0.647 272 Discussion Site The absence of a site effect on number of eggs and egg dry weight suggest that the models listed (Table 3) are appropriate for stocks throughout Tasmania and through to western Victoria. Fishing grounds for P. gigas in Tasmania are roughly split into western and eastern regions with few animals captured in Bass Straight or below 42°30'S on the east coast of Tasmania. Consequently, sample sites were from the extreme ranges of the fishery for these States. Site had a significant effect on egg diameter (P<0.001) with females from eastern Tasmania (n=30) producing smaller eggs than those from the other two sites (n= 132). This observed effect of site on egg diameter is surprising as individual dry egg weight did not appear to be affected by site. It is important to note that the effect of site on diameter was very weak, although significant, so there is unlikely to be any biological implication. Effect of carapace condition on reproductive output The significant improvement in model fit of log length-log fecundity by incorporating carapace-condition indicates that there is a decline in fecundity with successive broods within an instar. The effect of successive broods between moults on fecundity has been examined previously in snow crabs, Chionoecetes opilio (Majidae) which terminally moult into maturity and produce only two broods. Sainte-Marie (1993) observed an increase with consecutive clutches (opposed to a decline in P. gigas) in the order of 20% per brood. The relationship in P. gigas is more complex as there does not appear to be terminal moult into maturity, so broods are produced at several instars. This pattern is similar to most Cancer species where viable sperm may be retained for greater than 2 years and used to fertilise successive broods without moulting. A pattern of declining fecundity for successive broods produced between moults, as in P. gigas, also occurs with Cancer magister and Cancer anthonyi (Hankin et al., 1989; Shields et al., 1991). Some females in carapace-condition groups 2 and 3 appeared to have lower fecundity than those in group 1. Biological causes for this observed effect of carapace-condition on fecundity may include: damage to pleopods that may physically prevent a crab from carrying a full egg mass (Hankin et al., 1989); reduced ovarian development in response to depletion of spermatophore reserves from previous extrusions (Hankin et al., 1989); and senescence of older females. The potential for damage to pleopods to impair attachment of eggs in carapace-condition 3 crabs could not be assessed from information collected in the present study, but Hankin et al. (1989) did observe short pleopods on several non-moulted C. magister females. Large, carapace-condition 3, P. gigas females have been held in tanks to extrude their broods and no egg loss appeared to occur (Gardner, unpublished). Heavily fouled, grade-3 P. gigas females were significantly larger than other grades (Table 2) and are unlikely to have natural predators except during ecdysis. Consequently, it is feasible that this group does contain senescent females. Large female European edible crabs (Cancer pagurus) are also often heavily fouled and termed “granny crabs” (Pearson, 1908) which has been attributed to both prolonged intermoult duration and senescence (Edwards 1979). However, as with C. magister and C. anthonyi, decline in fecundity of P. gigas with successive batches of eggs produced between moults confounds detection of senescence (Hankin et al., 1989; Shields, 1991). 273 While the decline in fecundity of grade 3 females may be an effect of senescence, this is unlikely to be the case with grade 2 females. The latter were well represented across the size range collected, but many had significantly smaller clutches than did grade 1 females (P<0.05). If there is a limiting factor on reproductive output, such as remaining sperm reserves, there are advantages to a strategy of reducing clutch size to optimise future survival, rather than risking mortality during moulting to re-mate (Begon and Parker, 1986). Mating has never been observed in P. gigas, but there is indirect evidence that it occurs while the female is soft-shelled (see previous chapter). The effect of carapace-condition on fecundity is important when modelling fecundity for fishery management as it suggests that changes to stocks, other than those detected by monitoring size of individuals, have the ability to affect egg production. Egg size Size of individual eggs did not remain constant but declined significantly, albeit slightly, with deterioration in carapace-condition and increased with carapace length (P<0.0001). Change in individual egg size with deterioration in carapace condition and increase in CL has also been reported in Chionoecetes opilio in which older, multiparous females produced smaller eggs (Sainte-Marie, 1993). Change in egg size with female size is important if the amount of nutrient reserves available to the embryo is increased (Clarke, 1993). This has been demonstrated in the lobster Homarus americanus where larger females produce eggs with greater energy content (Attard and Hudon, 1987). Although significant, the effect of female size on egg size in Pseudocarcinus gigas was weak and there was considerable variation between individuals; consequently, female size on egg size may have little effect on nutrient reserves. Differences in egg size may affect not only the nutrient reserves available to the larvae, but may affect timing of larval release (Wear, 1974). It appears that even small differences in egg size can have profound effects on egg development; Sainte-Marie (1993) noted that development takes around 3 months longer in primiparous C. opilio, with larger eggs, than in multiparous snow crab females, even though egg diameter differed by only 1.4-2.7%. This suggests that the observed effect of female size on egg size in P. gigas may influence larval release, although it was not possible to assess this in the current study. Realised reproduction Corey (1991) listed factors that may cause loss of eggs at oviposition in crustaceans: incomplete extrusion of eggs; infertile eggs; lack of proper attachment of eggs to pleopods; and the extrusion of too many eggs for attachment. Comparison between counts of oocytes and eggs held under the tail indicated that none of these factors contribute to significant egg loss in P. gigas. Corey (1991) speculated that the most important factor affecting attachment of eggs in the freshwater crayfish, Orconectes spp, was rapid water movement. Female P. gigas held for long term sperm retention trials (see Chapter 14) were held in a 200 m3 tank with a 15 cm deep sandy substrate and observed during oviposition. Females buried their abdomens in substrate during oviposition and remained in this position for several weeks (Fig. 6). This behaviour has been observed in other brachyurans and is known to reduce loss of eggs during oviposition (Crothers, 1969; Wear, 1974; Edwards, 1979; Wild, 1983; Shields, 1991; Shields et al., 1991). 274 Figure 6. Female giant crabs Pseudocarcinus gigas during oviposition with their abdomens buried into pits in the substrate. Relative reproductive expenditure with increase in female size As would be expected, egg number increased with size of the female. With log transformed data, the slope of this relationship has a theoretical value of 3.0 as egg number is a volumetric measure while size is linear (Somers, 1991); however, in Pseudocarcinus gigas, the slope was 1.76 and significantly less than 3.0 (P<0.05). Somers (1991) postulated that values less than 3.0 could be caused by: separate age classes; changes in the proportionate size of the ovaries relative to the female; change in egg size relative to female size; and senescence. Additional factors which may have been important in this study were the effects of site, and successive clutches between moults. Of these hypotheses, it was possible to assess the effect of change in egg size relative to female size, site, successive clutches between moults, and to some extent the effect of separate age classes. There was no evidence that site affected fecundity as it did not improve the regression fit. Likewise the inclusion of a term for extra curvature did not improve fit, suggesting that there was no distinct trend of separate age classes as shown by Wenner et al. (1987) in Emerita analoga. However, by adjusting for carapacecondition (or possibly senescence as discussed above) the slope was raised (see effect of carapace-condition, Table 4). Also, there were indications that changes in the size of eggs with size of females was affecting the slope, based on comparisons between regressions for log fecundity and log total egg mass dry weight, the latter having higher slopes. This would be expected given the observed relationship between female size and egg size in P. gigas (Fig. 4); if a female crab is producing larger eggs then fewer could be contained within the space available to the ovary within the body cavity. This hypothesis was supported as slope of the regression for total egg volume (number of eggs x egg volume) was 2.63, and far closer to the theoretical value of 3.0, although still significantly less (P<0.05). In summary, the fecundity of female giant crabs did not follow a simple volumetric relationship; the relationship is affected by declining fecundity between successive clutches, and also by an increase in egg size with female size. Despite correcting for both these factors, the slope was still less than 3.0 which may be due to error from the coarseness of the measure of time since moulting by carapace-condition, or there may be an additional biological process involved. If moulting occurs in periods of greater than 3 years then carapace-condition may mask patterns as this was only divided into 3 groups. Also, if senescent females were present, they were likely lumped with fouled, but healthy, females into carapace-condition 3. While there is a general assumption that clutch size should increase with female size, optimising reproductive output by optimising future growth and survival can lead to a declining clutch size with age 275 (Charlesworth and Leon, 1976; Begon and Parker, 1986); such a biological pattern may be operating with P. gigas females. Additional factors affecting fecundity All samples for this study were collected in the early stage of embryonic development, as preliminary trials had shown that variation in individual egg weight was least during that period. This prevented analysis of possible decline in brood size through development which has been shown to be important in other crustacean species (Annala and Bycroft 1987; Kuris and Wickham 1987). Nemerteans and amphipods were observed in eggs sampled from P. gigas so a decline in brood size during development from predation is possible. References Annala, J.H. and Bycroft, B.L. 1987. Fecundity of the New Zealand red rock lobster, Jasus edwardsii. New Zealand Journal of Marine and Freshwater Research 21, 591-597. Attard, J. and Hudon, C. 1987. Embryonic development and energetic investment in egg production in relation to size of female lobster (Homarus americanus). Canadian Journal of Fisheries and Aquatic Sciences 44, 1157-1164. Begon, M. and Parker, G.A. 1986. Should egg size and clutch size decrease with age ? Oikos 47, 293302. Charlesworth, B. and Leon, J.A. 1976. The relation of reproductive effort to age. American Naturalist 110, 449-459. Clarke, A. 1993. Egg size and egg composition in polar shrimps (Caridea: Decapoda). Journal of Experimental Marine Biology and Ecology 168, 189-203. Crothers, J.H. 1969. The distribution of crabs in Dale Road (Milford Haven, Pembrokeshire) during summer. Field Studies 3, 109-124. Draper, N.R. and Smith, H. 1981. Applied regression and analysis. John Wiley and Sons, Toronto, Canada. 709 pp. Edwards, E. 1979. The edible crab and it’s fishery in British waters. Fishing News Books, Farnham, England. 142 pp. Hankin, D.G., Diamond, N., Mohr, M.S. and Ianelli, J. 1989. Growth and reproductive dynamics of adult female Dungeness crabs (Cancer magister) in northern California. Journal du Conseil International pour l’Exploration de la Mer 46, 94-108. Hankin, D.G., Butler, T.H., Wild, P.W. and Xue, Q.L. 1996. Does intense male harvest limit egg production of protected female stocks of Dungeness crabs ? High Latitude Crabs: Biology, Management, and Economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska Fairbanks, 507-510. Howard, A.E. 1982. The distribution and behavior of ovigerous edible crabs (Cancer pagurus), and consequent sampling bias. Journal du Conseil International pour l’Exploration de la Mer 40, 259-261. Kon, T. 1996. Overview of tanner crab fisheries around the Japanese Archipelago. High Latitude Crabs: Biology, Management, and Economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska Fairbanks, 13-24. Kuris, A.M. and Wickham, D.E. 1987. Effect of nemertian egg predators on crustaceans. Bulletin of Marine Science 41, 151-164. Norman, C.P. and Jones, M.B. 1993. Reproductive ecology of the velvet swimming crab, Necora puber (Brachyura: Portunidae), at Plymouth. Journal of the Marine Biological Association of the United Kingdom 73, 379-389. Pearson, J. 1908. Cancer (the edible crab). Memoirs of the Liverpool Marine Biology Committee 16, 263 p. Sainte-Marie, B. 1993. Reproductive cycle and fecundity of primiparous and multiparous female snow crab, Chionoecetes opilio, in the Northwest Gulf of Saint Lawrence. Canadian Journal of Fisheries and Aquatic Sciences 50, 2147-2156. Sainte-Marie, B., Sevigny, J.M., Smith, B.D. and Lovrich, G.A. 1996. Recruitment variability in snow crab, Chionoecetes opilio: pattern, possible causes, and implications for fisheries management. High Latitude Crabs: Biology, Management, and Economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska Fairbanks, 451-478. 276 Searcy-Bernal, R. 1994. Statistical power and aquacultural research. Aquaculture 127, 371-388. Shields, J.D. 1991. The reproductive ecology and fecundity of Cancer crabs. In ‘Crustacean egg production’. (Eds A. Wenner and A. Kuris) pp. 193-213. (A.A. Balkema, Rotterdam.) Shields, J.D., Okazaki, R.K. and Kuris, A.M. 1991. Fecundity and the reproductive potential of the yellow rock crab Cancer anthonyi. Fishery Bulletin 89, 299-305. Sokal, R.R. and Rohlf, F.J. 1981. Biometry. 2nd ed. W.H. Freeman, New York, N.Y. 859 p. Somers, K.M. 1991. Characterising size-specific fecundity in crustaceans. In ‘Crustacean egg production’. (Eds A. Wenner and A. Kuris) pp. 357-378. (A.A. Balkema, Rotterdam.) Strathmann, R.R. 1990. Why life histories evolve differently in the sea. American Zoologist 30, 197207. Strathmann, R.R. and Chaffee, C. 1984. Constraints on egg masses. II. Effect of spacing, size, and number of eggs on ventilation of masses of embryos in jelly, adherent groups, or thin-walled capsules. Journal of Experimental Marine Biology and Ecology 84, 85-93. Subramoniam, T. 1991. Yolk utilization and esterase activity in the mole crab Emerita asiatica (Milne Edwards). In ‘Crustacean egg production’. (Eds A. Wenner and A. Kuris) pp. 19-29. (A.A. Balkema, Rotterdam.) Vinuesa, J. H., Guzman, L., and Gonzalez, R. 1996. Overviews of southern king crab and false king crab fisheries in the Magellanic region. High Latitude Crabs: Biology, Management, and Economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska Fairbanks, 3-12. Wear, R.G. 1974. Incubation in British decapod Crustacea, and the effects of temperature on the rate of success of embryonic development. Journal of the Marine Biological Association of the United Kingdom 54, 745-762. Wenner, A.M., Hubbard, D.M., Dugan, J., Shoffner, J. and Jellison, K. 1987. Egg production by sand crabs (Emerita analoga) as a function of size and year class (Decapoda, Hippidae). Biological Bulletin 172, 225-235. Wild, P.W. 1983. Influence of seawater temperature on spawning, egg development, and hatching success of the Dungeness crab, Cancer magister. In (Eds.P. Wild and R. Tasto) Life History, Environment, and Mariculture Studies of the Dungeness Crab, Cancer magister, with Emphasis on the Central Californian Fishery Resource. Californian Department of Fish and Game Fisheries Bulletin 175, 197-205. 277 Composition of Eggs in Relation to Embryonic Development and Female Size 17 Research for this chapter has been previously published as: Gardner, C. Composition of eggs in relation to embryonic development and female size in giant crabs Pseudocarcinus gigas (Lamarck) Marine and Freshwater Research, in press. 278 Abstract The size and composition of eggs from 22 giant crabs Pseudocarcinus gigas were monitored over 165 days to determine trends through embryogenesis. Egg composition was most stable during the early stages of embryogenesis so additional sampling (n=143) was conducted during this period to assess the effect of female size, sampling location (east and west Tasmania), and successive broods between moults, on egg composition. During embryogenesis, eggs increased in diameter and moisture content while organic dry weight declined. Total carotenoid content did not change significantly while protein declined and lipid increased (as proportion of ash-free dry weight). This indicates that protein was used preferentially to lipid which is atypical of most decapods and may be an adaptation to the deeper water habitat of P. gigas. Females with heavy and intermediate carapace wear were considered more likely to have produced previous clutches and they produced eggs with significantly less carotenoid. The eggs of larger females contained significantly more water, less protein, and less carotenoid while there was no effect on total lipid (P<0.05). Although the effects of female size on egg composition were significant, the magnitude of the effect was small (r2ó0.17). Consequently, it is unlikely that larval viability is affected, or that larger females contribute more to recruitment than predicted by fecundity. Introduction Yolk reserves of crustacean eggs are critical to the development of the embryo and subsequent larvae. The pattern of utilisation of reserves varies between species and there is evidence that this inter-specific variation is an effect of environmental adaptation (Pandian, 1970; Pillai and Subramoniam, 1985). Ecophysiological research on yolk utilisation in decapods has been based on freshwater, inshore coastal, and terrestrial species while this chapter describes egg development and yolk utilisation in the deeper water, oceanic decapod species, Pseudocarcinus gigas. Given that the demands on an emerging larva in deep, oceanic water are likely to be different to those of a larva hatching in inshore areas, it was anticipated that yolk utilisation may be affected. In commercial crustacean species, understanding factors affecting egg production is important as these may influence recruitment. Egg size has been shown to vary within species in response to a range of factors including depth (Thessalou-Legaki, 1992), region (Collart and Rabelo, 1996), female size (Attard and Hudon, 1987), and production of successive broods (Sainte-Marie, 1993). Earlier work on fecundity of giant crabs demonstrated that egg size increased with female size (previous chapter; Gardner, 1997). In addition, egg size of P. gigas appeared to decline with successive clutches of eggs produced between moults. It was hypothesised that these differences in egg size may influence larval viability and subsequent larval recruitment. Although these observations on egg size of P. gigas indicated effects of female size and successive broods on reproductive output, they are not definitive. As noted by Clarke (1993), the assumption that larger eggs contain more nutrient is usually intuitive and is rarely tested. While egg size has been shown to reflect real differences in investment per egg by polar shrimps (Caridea; Clarke, 1993), this relationship is not present in all invertebrates and cannot be assumed (McEdward and Coulter, 1987). The most direct measure of the “quality” of eggs, and the effect of female size or successive broods, is a measure of larval size or vitality. This can be difficult to 277 manage, especially in larger crustaceans like the giant crab, as females must be housed separately until larval release. An indirect technique to assess egg quality was employed by Attard and Hudon (1987) in an investigation of the effect of female lobster Homarus americanus size on egg quality. They assessed egg composition as energy content and showed that the expenditure per egg was greater by larger females than smaller females. This suggests that larger female lobsters may contribute more to recruitment than would be predicted by fecundity alone. Similar research was conducted on giant crabs and is described in this chapter. The composition of eggs of giant crabs was analysed to test if the effects of female size and successive clutches on egg size represent real differences in parental contribution (in other words, do bigger female giant crabs produce better eggs?). Materials and methods Collection of samples Two sets of samples were collected: the first was for the assessment of changes in egg composition during embryogenesis, and the second was for assessment of female size on egg composition. For the first, 22 ovigerous females were captured in traps from depths in the range of 300 – 380 m off the east coast of Tasmania (41°15'S;148°40'E) in May 1994 by a commercial fisher. Ovigerous females had only just begun to be observed by fishers so these egg masses were regarded as recently extruded. Females ranged from 2.2 – 3.5 kg and were maintained in two 4 m3 tanks with flow through water supply and fed twice weekly. Hatching occurred over a period of 2 weeks in November 1994, during which females were checked every two days to allow the date of egg sample collection to be back-calculated, relative to hatching. Females were individually tagged and samples of eggs (around 30 g) were removed at 165, 125, 75, 50, and 20 d before hatching (averaged across sample). This first set of samples delineated a period during development when egg composition was relatively stable; sampling of females to determine the effect of female size was then conducted during this more stable period. Development of eggs was staged by a qualitative scheme (Table 1). For the second set of samples, ovigerous female crabs (n=143) were collected during the period 10 August to 15 September 1995 from eastern Tasmania (n=30), and western Tasmania (n=113). These were the same samples used previously to determine fecundity (see Fig. 1 in Gardner, 1997). Half of the egg mass was removed from each female for analysis of composition (development of individual eggs was homogeneous within the egg mass). Four of these females were captured with more advanced eggs but all eggs retained for analysis were at development stage 2 (Table 1). Female Pseudocarcinus gigas produce broods annually and appear to produce several broods between moults (Chapter 14). The time since moulting was roughly quantified by assigning a carapace condition grade ranging from grade 1, clean shelled, to grade 3, heavily worn and fouled (Gardner, 1997). This carapace condition grade was not intended to provide a direct scale of years since the previous moult, rather, it simply assumes that a heavily fouled female is more likely to have produced previous broods than a clean shelled female. Table 1. Classification of egg development in Pseudocarcinus gigas Stage Description 278 I: recently extruded Eggs bright orange with no embryo pigmentation. The egg is translucent and contains evenly distributed yolk granules. II: early developmental Eggs bright orange; a clear, yolk-free streak evident at one pole; yolk granules evenly distributed throughout. III: intermediate I Eggs dull orange; a quarter of the yolk mass is cleared; some structural development of embryo; no black pigmentation; some faint red eye pigmentation. IV: intermediate II Eggs dull orange; embryo eye pigmentation visible as small, grey/black patches; the inner yolk sac appears slightly detached from the surrounding capsule. V: late stage Eggs brownish orange; eye pigmentation is black; embryo well formed; heart beat is obvious and scattered pigmentation is visible; some yolk remains. VI: pre-hatch Eggs brownish orange to burgundy; embryo fully formed; pigmentation more defined with lines on the abdomen; small amount of pale yolk; frequent movement. Analysis of egg size and composition For measuring diameter, sub-samples were teased apart in sea water and the diameter of 50 eggs was then recorded by image analysis using NIH-Image™ 1.60 software. Pseudocarcinus gigas eggs are round and only those eggs which appeared normal were measured. All analyses were duplicated. Eggs were blot dried before weighing to obtain an initial wet weight. To determine water content, eggs were dried at 80°C for 24 h, and cooled for 1 h under vacuum before weighing. Samples of around 2 g were ashed at 450°C for 2 h. Mean individual egg dry weight for each brood was obtained by counting at least 250 eggs which were then rinsed in distilled water, dried, and weighed. Samples for biochemical analyses were stored at -60°C then thawed and ground in a mortar to a homogeneous paste. Protein was assayed by a modified Lowry procedure (Peterson, 1977; Sigma Diagnostics™ #5656; Appendix 9). Total lipid was measured by the gravimetric method of Folch et al. (1957) using chloroform and methanol as solvents (Appendix 9). Carotenoids were extracted from tissue with acetone (Appendix 9). The acetone extract was partitioned with diethyl ether which was washed with 20 volumes of 10% NaCl to remove residual acetone. Four samples of carotenoid extract from both early and late development stage eggs of the same females were chromatographed by thin layer chromatography (TLC) on C8 octyl silica plates (Merck™) using a solvent mixture of 95 : 5 petroleum ether : methanol. Tissue extracts were run alongside saponified extracts (5% ethanolic KOH for 24 h at room temperature) and astaxanthin standard (Roche Pharmaceuticals™). Carotenoids present in extracts were identified as predominantly astaxanthin or astaxanthin esters with low levels of a rapidly eluting red pigment, possibly ß carotene, and an unidentified yellow pigment. Consequently, total carotenoids were estimated from the extracts as astaxanthin by measuring their absorption in diethyl ether at 472 nm assuming an E1%1cmof2099(Clarke,1977). 279 Data analysis The effect of embryonic development on egg size and composition was tested by repeated measures analysis as the same females were sampled throughout the trial (Mardia et al., 1979). Significance was tested by Wilk’s lambda (Mardia et al., 1979). Effect of female size on egg size and composition was initially tested by simple linear regression. Additional analyses were then conducted to assess the effects of site and carapace-condition by analysis of covariance (ANCOVA), after first establishing equality of slopes by testing for interaction with female size (Sokal and Rohlf, 1981). Where the results of ANCOVA were significant, elevations of separate carapacecondition classes were compared by t-tests. All analyses were performed with JMP™ 3.1 software (SAS Institute). Results Changes in egg size, mass, and composition during embryogenesis Relatively little embryonic development occurred during the first 90 d of incubation with the majority of egg masses sampled at 75 d before hatch being at development stage 2 (Fig. 1). By 50 d prior to hatch, most egg masses had progressed to stage 3 with a quarter of the yolk cleared. Embryogenesis appeared to proceed more rapidly in the last 50 d with most yolk utilised by 20 d. Embryogenesis resulted in a significant change in egg diameter (P<0.001); individual egg dry weight (P<0.01); moisture (P<0.001); protein (P<0.001); and lipid (P<0.01; Fig. 2). Greatest rate of change in the water, protein, and lipid composition appeared to occur during the period of most rapid yolk depletion: between 50 d and hatch (Fig. 2). Mean egg dry weight declined during development while diameter increased. Total carotenoid (as ash free dry weight) did not change significantly during embryogenesis (P=0.22). 280 Figure 1. Effect of embryogenesis on egg diameter, weight and composition (moisture, protein, lipid, and carotenoid; n=22). Moisture content is presented as percentage of wet weight while protein, lipid, and carotenoid are presented as proportion of ash-free dry weight. Days prior to hatch was averaged across each sample. diameter (µm) 650 mean egg diameter ± s.d. 600 550 65 weight (µg) mean egg dry weight ± s.d. 60 55 50 45 70 water (%) mean % water ± s.d. 65 60 protein (mg/g) 55 600 500 400 300 mean protein ± s.d. lipid (mg/g) 200 mean lipid ± s.d. 175 150 125 carotenoid (ug/g) 100 75 500 mean carotenoid ± s.d. 400 300 200 100 150 100 50 0 days before hatch 281 Percentage of egg clutches (n=22) Figure 2. Developmental stages of egg masses collected at each sample period. Days prior to hatch was averaged across each sample. 100 %I 75 %II %III 50 %IV %V 25 %VI 0 170 145 120 95 70 45 20 days prior to hatch Influence of female size and carapace condition on egg composition The eggs of larger females tended to have significantly higher water content (P<0.01), less egg protein (P<0.05), and less total carotenoid (P<0.0001; Table 2). Although significant, the effect of female size appeared to be relatively slight, especially for water and protein content (r2=0.049). Lipid content did not appear to be influenced by female size (P>0.2). Additional analyses by ANCOVA indicated site and carapace condition had no significant effect on water, protein or lipid content (P>0.2; both as a proportion of ash free dry weight and as a total mass per egg). Total carotenoid did not appear to be influenced by site although it was affected by carapace condition (P<0.05; Table 3). The intercepts of each carapace condition grade were compared by t-tests; these tests indicated that the significant effect of carapace condition on total carotenoid content was attributable to higher levels of carotenoid in eggs from grade 1 (clean shelled) females while grades 2 and 3 (intermediate and heavy wear) were not significantly different (Fig. 3). Table 2. Regressions of egg composition on female size, measured as carapace length (CL; mm). Results from 143 females were used in each analysis. Protein and carotenoid concentrations are as ash-free dry weight per gram of egg mass and per individual egg. NS, P>0.2; * P<0.05; **P<0.01; ***P<0.0001. Regression equation r2 F-ratio water (mg/g) = 0.673CL + 502.39 0.049 7.420** protein (mg/g) = -1.332CL + 750.10 0.028 4.129* protein per egg (µg) = -0.030CL + 15.14 0.037 5.564* lipid (mg/g)= -0.190CL + 214.10 0.009 1.119 NS lipid per egg (µg)= 0.00001CL + 1.04 <0.001 0.0001NS carotenoid (µg/g) = -2.176CL + 603.11 0.153 25.753*** carotenoid per egg (ng) = -0.046CL + 12.75 0.170 29.216*** 282 Table 3. Results of analyses of variance examining the effect of carapace-condition on carotenoid content. Analyses were conducted on carotenoid content as ash-free dry weight per gram of egg mass and per individual egg. Carapace length (mm) was included in the analyses to correct for the effect of female size on carotenoid content of eggs. * P<0.05; **P<0.01; ***P<0.0001. Carotenoid (µg/g) Source of variation Carotenoid per egg (ng) D.F. M.S. F-ratio D.F. M.S. F-ratio Carapace length (covariate) 1 170254 21.01*** 1 86.54 22.48*** Carapace condition 2 28766 3.55* 2 19.34 5.02** 140 8103 140 3.85 error Total carotenoid per egg (ng) Figure 3. Regressions of total carotenoid per egg (as ash-free dry weight) on carapace length, for each carapace-condition class (denoted by numerals). Carapacecondition class 1 - solid dots; class 2- hollow dots; class 3-diamonds. 12.5 10 7.5 1 2 5 3 2.5 0 125 150 175 200 Carapace length (mm) 225 Discussion Changes in egg size, mass, and composition during embryogenesis General trends during embryogenesis in Pseudocarcinus gigas were typical of decapods; eggs took up water during development with a resultant increase in diameter while organic dry weight declined (Clarke et al., 1990; Subramoniam, 1991; Lardies and Wehrtmann, 1996; Wild, 1983). However, utilisation of protein and lipid reserves differed from the pattern of most decapods where lipid tends to decline during development with protein remaining stable or proportionally increasing (Pandian, 1970; Rao et al., 1981; Pillai and Subramoniam, 1985; Subramoniam, 1991). In P. gigas, protein appeared to be utilised in preference to lipid with lipid reserves appearing to decline slightly initially then increase later in embryogenesis. The apparent increase in 283 lipid content of eggs is confounded to some extent by concurrent declines in other components but the results clearly indicate that lipid catabolism is relatively low. An increase in lipid content during embryogenesis is unusual in decapods although it has been reported previously in Ovalipes punctatus (Portunidae; DuPreez and McLachlan, 1984) and Macrobrachium rosenbergii (Palaemonidae; Clarke et al., 1990). Enhanced lipid metabolism is a feature of cleidoic eggs (which do not exchange material other than gasses with the environment) and terrestrial crabs have been shown to produce eggs with large lipid reserves which are utilised throughout embryogenesis (Pillai and Subramoniam, 1985). Eggs of marine species have greater opportunity for release of waste products and greater reliance on protein reserves has been reported; eggs of the mole crab Emerita asiatica are intermediate between cleidoic and noncleidoic as protein reserves are used continuously through development, although lipid is also depleted (Subramoniam, 1991). Eggs of Pseudocarcinus gigas appear to be further developed towards non-cleidoic type metabolism with greater reliance on protein reserves in a similar pattern to bony fish (Lasker, 1962). Unlike most decapod species where the effects of embryogenesis have been investigated, Pseudocarcinus gigas inhabits relatively deep water of around 300 m. Habitat is known to influence yolk utilisation (Pillai and Subramoniam, 1985) and the low level of lipid catabolism by P. gigas embryos may be an adaptive strategy for deeper water. Subramoniam (1991) suggested that the lower level of lipid utilisation in the shallow-water mole crab Emerita asiatica may be a strategy to retain higher lipid content in zoeas to increase buoyancy and provide a buffer against starvation. Buoyancy may be especially important in deeper water decapods such as P. gigas where newly hatched larvae appear to swim upwards in response to negative geotaxis (Gardner, 1996). Likewise, retaining lipid reserves may be of advantage in delaying starvation of zoeas in oceanic waters (19 d at 16°C in P. gigas; Gardner and Northam, 1997). The predominant carotenoid in Pseudocarcinus gigas eggs is unesterified astaxanthin, as in Penaeus semisulcatus (Dall et al., 1995). Concentration of total carotenoid remained constant during embryogenesis in Pseudocarcinus gigas although a decline has been reported in other decapods (Dersan-Kour and Subramoniam, 1992; Dall, 1995). The apparently low utilisation rate of carotenoid reserves in P. gigas may be due to delay of development of carotenoid oxidation and esterification pathways until the larval stages as has been observed in Penaeus japonicus (Petit et al., 1991). Effect of female size on egg composition The composition of Pseudocarcinus gigas eggs was most stable during the early stages of embryogenesis, so further sampling to determine the effect of female size on egg composition was conducted during this period (egg development stage 2). Female size appeared to have a significant effect on egg composition with larger females producing eggs with more water, less protein, and less total carotenoid while there was no effect on total lipid. Although significant, the effect of female size on water and protein content was small and correlation was weak, even when analysed as mass per individual egg (r2=0.049; Table 2). Based on these analyses, the effect of female size on egg size reported in Gardner (1997) is likely to have negligible effect on the viability of larvae. In a similar study by Attard and Hudon (1987), size of female lobsters Homarus americanus was shown to influence energy content of eggs and it was speculated that this effect was large enough to influence larval growth and survival. Attard and Hudon (1987) concluded that the effect of female size on egg composition is likely to enhance 284 the contribution of larger females to recruitment. In Pseudocarcinus gigas, it appears that the effect of female size on egg composition is small so reproductive output of different sized females may be effectively modelled by analysis of fecundity alone. Carotenoid content declined with female size and was also influenced by carapace condition, a qualitative measure of time since moulting. Extremely low levels of dietary carotenoids have been shown to reduce survival of crustaceans (Penaeus japonicus; Chein and Jeng, 1992) although in Pseudocarcinus gigas, there was only a weak correlation between egg carotenoid content and female size (r2=0.17). Consequently, the effects on larval vitality may not be important. Carotenoids cannot be synthesised de novo by crustaceans so the diet of females during gametogenesis will influence carotenoid content of eggs (Harrison, 1990). Levings et al. (1996) stated that Pseudocarcinus gigas show depth stratification of different size and carapace condition classes with most females found between 120 and 270 m. Although all females used in this study were captured at similar depth, they may have moved from different depths before capture. Consequently, the observed effect of female size and carapace condition on carotenoid composition of eggs may simply reflect different prey items encountered. An effect of female size on composition has been recorded in other decapod species including Chionoecetes opilio and Homarus americanus (Attard and Hudon, 1987; Sainte-Marie, 1993). It has been speculated that this effect of female size may mean that larger females have greater contribution towards recruitment than would be predicted by fecundity alone. This clearly has important implications for fisheries management of P. gigas where minimum size limits selectively target large females for harvest. This study shows that biochemical composition of eggs was not largely affected by female size and that management by minimum size limit may be appropriate. Nonetheless, analysis of biochemical composition does not necessarily provide a measure of larval vitality which is critical in assessing the effect of female size on egg quality. Further research on the effect of female size on larval vitality is needed to fully address the issue. References Attard, J. and Hudon, C. 1987. Embryonic development and energetic investment in egg production in relation to size of female lobster (Homarus americanus). Canadian Journal of Fisheries and Aquatic Sciences 44, 1157-1164. Chien, Y-H. and Jeng, S-C. 1992. Pigmentation of kuruma prawn Penaeus japonicus Bate, by various pigment sources and levels and feeding regimes. Aquaculture 102, 333-346. Clarke, A. 1977. Lipid class and fatty acid composition of Chorimus antarcticus (Pfeffer) (Crustacea: Decapoda) at South Georgia. Journal of Experimental Marine Biology and Ecology 28, 297-314. Clarke, A. 1993. Egg size and composition in polar shrimps (Caridea; Decapoda). Journal of Experimental Marine Biology and Ecology 168, 189-203. Clarke, A., Brown, J.H. and Holmes, L.J. 1990. The biochemical composition of eggs from Macrobrachium rosenbergii in relation to embryonic development. Comparative Biochemistry and Physiology 96B, 505-511. Collart, O.O. and Rabelo, H. 1996. Variation in egg size of the fresh-water prawn Macrobrachium amazonicum (Decapoda: Palaemonidae). Journal of Crustacean Biology 16, 684-688. Dall, W. 1995. Carotenoids versus retinoids (Vitamin A) as essential growth factors in penaeid prawns (Penaeus semisulcatus). Marine Biology 124, 209-213. Dall, W., Smith, D.M. and Moore, L.E. 1995. Carotenoids in the tiger prawn Penaeus esculentus during ovarian maturation. Marine Biology 123, 435-441. Dersan-Kour, V.R. and Subramoniam, T. 1992. Carotenoid metabolism during embryonic development of a marine crab, Emerita asiatica (Milne Edwards). Invertebrate Reproduction and Development 21, 99106. DuPreez, H.H. and McLachlan, A. 1984. Biology of the three spot swimming crab, Ovalipes punctatus (De Haan). III. Reproduction, fecundity and egg development. Crustaceana 47, 285-297. 285 Folch, J., Lees, M. and Bloane-Stanley, G.H. 1957. Purification of total lipids from animal tissues. Journal of Biological Chemistry 266, 497-509. Gardner, N.C. 1996. Behavioural basis of depth regulation in the first zoeal stage of the giant crab (Pseudocarcinus gigas, Brachyura, Xanthoidea, Oziidae). In ‘High Latitude Crabs: Biology, Management, and Economics’. pp. 229-53. (Alaska Sea Grant College Program Report No. 96-02, University of Alaska, Fairbanks). Gardner, C. 1997. Effect of size on reproductive output of giant crabs Pseudocarcinus gigas (Lamarck) (Oziidae). Marine and Freshwater Research 48, 581-587. Gardner, C. and Northam, M. 1997. Use of prophylactic treatments for larval rearing of giant crabs Pseudocarcinus gigas (Lamarck). Aquaculture 158, 203-214. Harrison, K.E. 1990. The role of nutrition in maturation, reproduction and embryonic development of decapod crustaceans: a review. Journal of Shellfish Research 9, 1-28. Lardies, M.A. and Wehrtmann, I.S. 1996. Aspects of the reproductive biology of Petrolisthes laevigatus (Guérin, 1835) (Decapoda, Anomura, Porcellanidae). Part I: Reproductive output and chemical composition of eggs during embryonic development. Archive of Fishery and Marine Research 43, 121135. Lasker, R. 1962. Efficiency and rate of yolk utilisation by developing embryos and larvae of the Pacific sardine Sardinops caerula (Girard). Journal of the Fisheries Research Board of Canada 19, 867-875. Levings, A., Mitchell, B.D., Heeren, T., Austin, C. and Matheson, J. 1996. Fisheries biology of the giant crab (Pseudocarcinus gigas, Brachyura, Oziidae) in southern Australia. In ‘High Latitude Crabs: Biology, Management, and Economics’. pp. 125-51. (Alaska Sea Grant College Program Report No. 9602, University of Alaska, Fairbanks). Mardia, K.V., Kent, J.T. and Bibby, J.M. 1979. ‘Multivariate Analysis.’ (Academic Press, N.Y.) McEdward, L.R. and Coulter, L.K. 1987. Egg volume and energetic content are not correlated among sibling offspring of starfish: implications for life history theory. Evolution 41, 914-917. Pandian, T.J. 1970. Ecophysiological studies on the developing eggs and embryos of the European lobster Homarus gammarus. Marine Biology 5, 154-167. Peterson, G.L. 1977. A simplification of the protein assay method of Lowry et al. which is more generally applicable. Analytical Biochemistry 83, 346-356. Petit, H., Sance, S., Negre-Sadargues, G., Castillo, R. and Trilles, J.P. 1991. Ontogeny of carotenoid metabolism in the prawn Penaeus japonicus Bate (1888)(Crustacea: Penaeidea). A qualitative approach. Comparative Biochemistry and Physiology 99B, 667-671. Pillai, C.K. and Subramoniam, T. 1985. Yolk utilisation as an adaptive strategy of terrestialisation in the freshwater crab Paratelphusa hydrodromous (Herbst). Physiological Zoology 58, 445-457. Rao, N.C.H., Ponnuchamy, C., Shakuntala, K. and Reddy, S.R. 1981. Fecundity and energetics of embryonic metabolism of Caridina weberi (de Mann) (Decapoda: Atydae). International Journal of Invertebrate Reproduction 3, 75-85. Sainte-Marie, B. 1993. Reproductive cycle and fecundity of primiparous and multiparous female snow crab, Chionoecetes opilio, in the Northwest Gulf of Saint Lawrence. Canadian Journal of Fisheries and Aquatic Sciences 50, 2147-2156. Sokal, R.R. and Rohlf, F.J. 1981. ‘Biometry. 2nd ed.’ (W.H. Freeman, New York, N.Y.) Subramoniam, T. 1991. Yolk utilisation and esterase activity in the mole crab Emerita asiatica (Milne Edwards). In ‘Crustacean Egg Production’. (Eds A. Wenner and A. Kuris.) pp. 19-29. (A.A. Balkema: Rotterdam.) Thessalou-Legaki, M. 1992. Reproductive variability of Parapandalus narval (Crustacea: Decapoda) along a depth gradient. Estuarine, Coastal and Shelf Science 35, 593-603. Wild, P.W. 1983. The influence of seawater temperature on spawning, egg development, and hatching success of the Dungeness crab, Cancer magister. In ‘Life History, Environment, and Mariculture Studies of the Dungeness Crab, Cancer magister, with Emphasis on the Central California Fishery Resource’. (Eds P. Wild and R. Tasto.). State of California Department of Fish and Game, Fish Bulletin 172, 197213. 286 General Discussion: Reproductive Biology of the Giant Crab Pseudocarcinus gigas 18 287 Research conducted on the reproductive biology of the giant crab was primarily intended to provide fundamental information for fisheries management. For effective fisheries management, information is required on growth, behaviour (selectivity), and population structure and detailed information on all these factors is not available. Nonetheless, information on the reproductive biology provides a guide for basic fisheries management tools such as seasonal restrictions and size limits. Harvest of males Despite research on morphological and physiological maturity of male giant crabs, the available data do not clearly indicate the size of functional maturity (Chapter 14). The current minimum size limit does not protect morphologically adolescent or mature male crabs although the relationship between these morphological groupings and functional maturity is not known. Given that exploitation rate appears to be high, and that males appear to spend at least 2 years between moults (Pers. Comm., R. McGarvey, South Australian Research and Development Institute), it is probable that very few males will live through morphological adolescence to reach morphological maturity. If morphologically adolescent crabs are not functionally mature, sperm limitation would seem inevitable. Further research on the onset of functional maturity is important but is constrained by logistic difficulties in working with these relatively deep water crabs (as discussed in Chapter 14). A technique developed for recording mating of intertidal xanthoid crabs may be suitable. This involves immobilising crabs (Chapter 11) and attaching a trigger across the abdomen to determine if recaptured males have lifted their abdomen, and thus may have mated (Pers. Comm., P. Jivoff, Smithsonian Marine Station, Florida; Fig. 1). Male crabs are being fitted with these triggers in ongoing research but until data become available on functional maturity, management must be conservative and should aim to protect a portion of morphologically adult males. This could be achieved with a conservative quota (TAC) or by implementing a maximum size limit so that a proportion of crabs could moult directly from below the minimum size limit to above the maximum size limit. 288 Figure 1. Male giant crab with trigger attached to the abdomen to detect if the abdomen has been lifted since the crab was released. The trigger is mono-filament nylon and is glued to the sternum. It runs backwards to the abdomen and through a tube so that the abdomen can be lifted freely, but the nylon will slip from the tube. Female giant crabs are harvested in Tasmania (Chapter 10) so the sex ratio is less likely to be biased by fishing than in a male-only fishery, although male giant crabs may be harvested preferentially. Males grow larger than females so less protection is offered by the minimum legal size, and restrictions on the taking of ovigerous females prevent harvest of females during 2 or 3 months of the open season. While it would be useful to monitor changes in sex ratio within the population by market measuring or by field sampling, it is a difficult goal as crabs are not distributed randomly. Giant crabs appear to live at different depths depending on sex and size (Levings et al., 1996) and commercial fishers aim to catch the smaller sized crabs which are favoured by exporters. Fisheries independent surveys conducted at fixed stations are an option for obtaining more reliable data on changes in sex ratio. In crab fisheries where harvest of females is prohibited, research on the development of males has taken priority with the aim of limiting exploitation to sexually mature males (Hankin et al., 1996). Even with good information on the size of males in mating pairs, and thus functional maturity, there is concern about the potential for sperm limitation in several crab fisheries. Important aspects in considering the potential for sperm limitation are: ¤ male-only fisheries inevitably bias the operational sex ratio (Smith and Jamieson, 1991); 289 ¤ the male is always larger than the female in mating pairs of many species (e.g. Cancer magister), so fertility of large females will be affected by intense exploitation of larger males10 (Smith and Jamieson, 1991); ¤ large interannual fluctuations in recruitment may lower the operational sex ratio, due to size/age dimorphism between the sexes (i.e. males need to be older than females to mate1)(Sainte-Marie and Sevigny, 1997); ¤ the number of females that a male can mate with is restricted by pre- and post-copulatory mate guarding behaviour and the seasonal window of moulting by females (where mating is with soft-shelled females) (Armstrong and Jamieson, 1997); and ¤ the weight of ejaculate delivered to females is directly proportional to the operational sex ratio, so ejaculate reserves do not recover rapidly between copulations. This subsequently results in lowered fertilisation rate (in Chionoecetes opilio; Sainte-Marie and Sevigny, 1997). Several of these issues are relevant to giant crab. Despite the lack of direct information on the mating system of P. gigas, indirect evidence from the anatomy of the vagina, mating scars, and taxonomic affinities (Menippe spp.) indicates that mating probably occurs while the female is soft-shelled (Chapter 15). Species which mate while the female is soft-shelled typically have protacted pre- and post-copulatory mate guarding (Hartnoll, 1969; Wilber, 1989) which lowers the operational sex ratio and raises the risk of sperm limitation. It is also likely that males need to be larger than females for mating to be successful as indicated by their habitat (open type), dimorphism of the chelae, large differences in body size between sexes, greater limb loss in males (indirect evidence of male rivalry), and taxonomic affinity (Menippe spp.) (Christy, 1987; Wilber, 1992; Orensanz et al., 1995). As with Cancer magister, this suggests that sperm limitation may be a problem for larger P. gigas females if there is intense exploitation of males (Smith and Jamieson, 199111). If female giant crabs can only mate when soft-shelled, then mating may occur only every 4 years or more, as preliminary analysis of recapture information from South Australia indicates that this is a typical intermoult period (Pers. 10 If growth rate of both sexes is similar until sexual maturity (typical for crabs), and if males need to be larger than their female partner, then most males involved in mating will be older than females. Consequently, they will have been subject to an additional period of natural mortality and the sex ratio will be skewed towards females (Armstrong and Jamieson, 1997). 11 Note that Hankin et al. (1996) consider that there is no field evidence in support of the conclusions drawn by Smith and Jamieson (1991). 290 Comm., R. McGarvey, South Australian Research and Development Institute). Although females were able to store sperm for at least 4 years, and produce broods with viable eggs (Chapter 14), the fecundity of heavily fouled, presumably late intermoult, females was lower than clean shelled females. If this reduced fecundity was due to depletion of sperm reserves, then fecundity of late intermoult females would be more severely reduced where the amount of sperm delivered was reduced. This has been reported in other species where the abundance of functional males has been reduced (Jivoff, 1997; Sainte Marie et al., 1997). If males need to be of a different age to females, natural variation in recruitment will alter the sex ratio and may compound the effect of fishing mortality. Managing variation in recruitment Classification of the life history strategies of marine organisms as r- or Kselected provided a useful guide of the potential of species to increase in population (Pianka, 1970). However, this simple continuum was limited and failed to describe some strategies, so a triangular continuum was proposed by Winemiller and Rose (1992) with three end points of opportunistic-periodicequilibrium type strategies. As with most crabs, clawed-lobsters and spinylobsters, Pseudocarcinus gigas may be considered a periodic strategist since it is large, highly fecund and long lived. Periodic strategists are typified by fluctuating recruitment (Winemiller and Rose, 1992). Cobb et al. (1997) considered that Cancer species are likely to have recruitment variability equal to or greater than that of spiny lobsters, based on their relative fecundity and larval duration. This is likely to apply to P. gigas also as the life history strategy is similar to that of Cancer species with similar fecundity (Hines, 1991; Bennett, 1995) and larval duration (Sulkin and McKeen, 1994; Bennett, 1995). Recruitment variation of spiny lobsters Jasus edwardsii in Tasmania can vary inter-annually by a factor of 5 (Gardner et al., 1998) so similar or greater variation may occur in P. gigas. The potential for large inter-annual variation in recruitment suggested by the fecundity and larval duration of P. gigas may be compounded by cannibalism as fragments of giant crab exoskeleton have been found in the stomachs of giant crabs (Heeren and Mitchell, 1997). Although these shell fragments may have been from exuviae rather than live animals, cannibalism has been reported in other crabs and it is quite possible in P. gigas (Hines et al., 1987; Smith, 1995; Botsford and Hobbs, 1995). Cannibalism in crab populations is important to fisheries management as it can compound fluctuation in recruitment by cannibalism of smaller crabs by cohorts from high recruitment years. Several authors have considered that the inherent fluctuations in Canadian snow crab Chionoecetes opilio and Dungeness crab Cancer magister fisheries may be due to cannibalism (Comeau and Conan, 1992; Sainte-Marie et al., 1995, 1996; Botsford and Hobbs, 1995; Lovrich and Sainte-Marie, 1997). Large fluctuations in recruitment is problematic for traditional single-species fisheries management as it becomes difficult to interpret the cause of declines in catch and catch rates (is it an effect of cyclical low recruitment or intensive exploitation ?), and it removes economic stability for participants in the fishery. Understanding these patterns will only become possible in the P. gigas fishery by the development of pre-recruit indices (such as fishery-independent 291 surveys of sub-legal crabs) combined with long term monitoring of at least ten years (Thresher, 1997). Managers must consider that the exploitable biomass will fluctuate between years so that fishing mortality will increase when a recruitment trough enters the fishery, and decline when a recruitment pulse enters the fishery. This is especially difficult to manage in majid crabs as they often reach legal size at the terminal moult. In periods of high recruitment, crabs will grow older with associated increase in exoskeleton wear and a decline in value (Sainte-Marie et al., 1996). Giant crabs do not appear to have a terminal moult and their beach price is not affected by exoskeleton condition, so unlike majid crabs, it should be possible to “bank” giant crabs during periods of high recruitment into the fishery. That is, periods of high recruitment can be exploited over several years to bridge the gap to the next period of high recruitment. Management decisions which aim to dampen variation in recruitment will be enhanced by the ability to forecast abundance trends several years in advance as has been achieved with snow crab Chionoecetes opilio in eastern Canada (Sainte-Marie, 1997). In the absence of this information, restrictions on effort such as total allowable catches (TAC’s) should be set conservatively to permit “banking” of crabs from high recruitment years. This assumes that population cycling will continue at regular duration and that fishing mortality does not destabilise cycles. Destabilisation of recruitment cycles through exploitation has been predicted in Cancer magister (Botsford, 1995) and is conceivably possible in P. gigas as egg production is reduced. Egg production Information on egg production is essential for formulating fisheries management tools, such as minimum legal size, and in modelling the effect of management strategies on egg production. Simple static models such as egg per recruit analyses incorporate fecundity, growth, and natural mortality to estimate suitable size limits to protect egg production above a certain proportion of the virgin stock. This proportion is usually set arbitrarily at 20% to 30%12, although some species are resilient to far higher exploitation (e.g. 90% in Spanish fisheries for Maja squinado; Freire et al., 1997) while other crab fisheries have highly conservative management which aims to protect 100% of egg production by restricting landings to male crabs (eg in the Californian fishery for Cancer magister, Hankin et al., 1989). Female giant crabs are harvested commercially so the egg production of fished populations will be affected. An interim size limit of 150 mm carapace length was introduced by Tasmanian fisheries management in 1994 and this was intended to protect approximately 60% of virgin egg production in NorthWestern Tasmania and Western Victoria. This minimum size limit was based 12 These limits are generally “plucked out the air” as precise stock-recruitment-relationships (SRR) are extremely variable between species and accurate values can only be obtained through collapse of the fishery. Modeling the SRR is generally difficult and has low precision. For instance, Mace and Sissenwine (1994) conducted an extensive review of the SRR of 90 species and stocks and concluded that species such as cod would be resilient to recruitment overfishing. Unfortunately, their timing was extremely bad as cod stocks collapsed almost immediately after this work was published. Very little is known about the limits of recruitment overfishing in crab stocks. The Hawaiian spanner crab fishery collapsed but no precise data was collected (see discussions in “Workshop on Stock Recruitment Relationships in Australian Crustacean Fisheries”). 292 on an analysis combining the proportion of females mature at each size interval (mm), the relationship between female size and egg mass weight, and a lengthfrequency distribution of female crabs obtained during months when females were ovigerous. As an interim restriction on fishing effort, the implementation of a size limit provided an effective input control of effort until more data became available. However, there were flaws with the original analysis. Firstly, there was no measure of gear selectivity so the analysis assumed that the natural population distribution was equivalent to the distribution of catch taken by trapping. Trapping inevitably biases population surveys against smaller animals (Myers and Hoenig, 1997). Estimates of population structure were also biased by sampling when females were ovigerous, which affects their catchability (Edwards, 1978; Howard, 1982; Levings et al., 1996; Schultz et al., 1996). It is likely that this bias would also have affected estimates of the onset of sexual maturity. To emphasise the problems with seasonal bias in population surveys of giant crabs, the analysis of egg production was repeated using data collected in Autumn 1998, which is before females become ovigerous (Fig. 2). This analysis indicates that the current size limit does little to protect egg production with only around 10% of egg production affected. However, there are problems with this analysis as there is no measure of onset of maturity (it assumes all females are mature), the population had been fished prior to the population survey (so large females will be under-represented, relative to virgin stocks), and not all females produce eggs every year (especially when they are small). All these factors cause the analysis to overestimate the egg production by females under the size limit (that is, egg production of females less than 150 mm is probably less than the 10% shown). Also, the effect of gear selectivity has not been incorporated and this is critical for estimating true population distribution. 293 Figure 2. Estimation of the proportion of egg production that would be protected by the minimum size limits shown on the horizontal axis. This analysis combines catch data from Tasmanian Department of Primary Industry and Fisheries research sampling (1998; n=659) with a simple fecundity model (Gardner, 1997). The resulting plot is an estimate of the cumulative contribution of different sized females to the egg production of the population. This analysis indicates that around 10% of egg production is protected by the current minimum size limit of 150 mm, and that the minimum size limit would need to be around 159 mm to protect around 30%. Two paths can be taken to obtain useful and meaningful minimum size limits for the Tasmanian fishery. First, onset of maturity and gear selectivity effects could be estimated to improve the analysis shown in Figure 2. Onset of sexual maturity should be assessed by ovarian development to prevent bias from reduced catchability of ovigerous crabs (e.g. by CT imaging, Chapter 12). Gear selectivity could be estimated by analysis of the effect of size on recapture rates of tagged animals (provided there is only a short period between samples of 1 or 2 weeks) (Myers and Hoenig, 1997). Alternatively, minimum size limits could be formulated by an egg per recruit analysis although this requires precise growth information. A large scale tagging project is underway to obtain growth data but this is unlikely to provide enough data for a basic growth model for several years due to low recapture rates, long intermoult periods, and the narrow size range available for tagging. Either method requires fecundity information as an input and this information is presented in Chapter 16. The models presented appear to be applicable for all Tasmanian waters as there was no effect of site, although estimates of egg production of the population would be enhanced by recording the extent of fouling on females sampled in population surveys (as carapace condition). Attard and Hudon (1987) demonstrated that the contribution of female American lobsters Homarus americanus towards recruitment could not be measured by fecundity alone, as egg quality was affected by female size. Modeling the effect of size limits on giant crab egg production is more straight forward as the composition of eggs was not affected by female size in this species (Chapter 17). 294 References Armstrong, D. and Jamieson, G. 1997. Evidence for fishery induced sperm limitation in Dungeness crab, Cancer magister. Abstracts, ICES International Symposium – Recruitment Dynamics of Exploited Marine Populations: Physical-Biological Interactions, Baltimore, USA, September, 1997. pp 62-63. Attard, J. and Hudon, C. 1987. Embryonic development and energetic investment in egg production in relation to size of female lobster (Homarus americanus). Canadian Journal of Fisheries and Aquatic Sciences, 44: 1157-1164. Bennett, D.B. 1995. Factors in the life history of the edible crab (Cancer pagurus L.) that influence modelling and management. International Council for the Exploration of the Sea Marine Science Symposia, 199: 89-98. Botsford, L.W. 1995. Population dynamics of spatially distributed, meroplanktonic, exploited marine invertebrates. International Council for the Exploration of the Sea Marine Science Symposia, 199: 157-166. Botsford, L.W. and Hobbs, R.C. 1995. Recent advances in the understanding of cyclic behaviour of Dungeness crab (Cancer magister) populations. International Council for the Exploration of the Sea Marine Science Symposia, 199: 157-166. Christy, J.H. 1987. Competitive mating, mate choice and mating associations of brachyuran crabs. Bulletin of Marine Science, 41: 177-191. Cobb, J.S., Booth, J.D. and Clancy, M. 1997. Recruitment strategies in lobsters and crabs: a comparison. Marine and Freshwater Research, 48: 797-806. Comeau, M. and Conan, G.Y. 1992. Morphometry and gonad maturity of male snow crab, Chionoecetes opilio. Canadian Journal of Fisheries and Aquatic Sciences, 49: 2460-2468. Edwards, E., 1978. The Edible Crab and its Fishery in British Waters. Fishing News Books, Farnham, England. 142 pp. Freire, J., Fernández, L. and González-Gurriarán, E. 1997. Interactions of the fishery of the spider crab Maja squinado with mating, reproductive biology and migrations. Abstracts, ICES International Symposium – Recruitment Dynamics of Exploited Marine Populations: PhysicalBiological Interactions, Baltimore, USA, September, 1997. p 67. Gardner, C. 1997. Effect of size on reproductive output of giant crabs Pseudocarcinus gigas (Lamarck): Oziidae. Marine and Freshwater Research, 48: 581-587. Gardner, C., Cawthorn, A., Gibson, I., Frusher, S., Kennedy, R.B. and Pearn, R.M. 1998. Review of the Southern Rock Lobster Jasus edwardsii Puerulus Monitoring Program: 1991-1997. Tasmanian Department of Primary Industry and Fisheries Technical Report, 52. 40 pp. Hankin, D.G., Diamond, N., Mohr, M.S. and Ianelli, J. 1989. Growth and reproductive dynamics of adult female Dungeness crabs (Cancer magister) in northern California. Journal du Conseil International pour l’Exploration de la Mer 46, 94-108. Hankin, D.G., Butler, T.H., Wild, P.W. and Xue, Q.L. 1996. Does intense male harvest limit egg production of protected female stocks of Dungeness crabs ? High Latitude Crabs: Biology, Management, and Economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska Fairbanks, 507-510. Hartnoll, R.G. 1969. Mating in the Brachyura. Crustaceana, 16: 161-181. 295 Heeren, T. and Mitchell, B.D. 1997. Morphology of the mouthparts, gastric mill and digestive tract of the giant crab, Pseudocarcinus gigas (Milne Edwards) (Decapoda: Oziidae). Marine and Freshwater Research, 48: 7-18. Hines, A.H., Lipcius, R.N. and Haddon, A.M. 1987. Population dynamics and habitat partitioning by size, sex, and moult stage of blue crabs Callinectes sapidus in a subestuary of central Chesapeake Bay. Mar. Ecol. Prog. Ser., 36: 55-64. Hines, A.H. 1991. Fecundity and reproductive output in nine species of Cancer crabs (Crustacea, Brachyura, Cancridae). Canadian Journal of Fisheries and Aquatic Sciences, 48: 265-275. Howard, A.E. 1982. The distribution and behaviour of ovigerous edible crabs (Cancer pagurus), and consequent sampling bias. Journal du Conseil International pour l’Exploration de la Mer 40: 259261. Jivoff, P., 1997. The relative roles of predation and sperm competition on the duration of the post-copulatory association between the sexes in the blue crab, Callinectes sapidus. Behavioural and Ecological Sociobiology, 40: 175-185. Levings, A., Mitchell, B.D., Heeren, T., Austin, C. and Matheson, J. 1996. Fisheries biology of the giant crab (Pseudocarcinus gigas, Brachyura, Oziidae) in southern Australia. High Latitude Crabs: Biology, Management, and Economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska Fairbanks, 125-151. Lovrich, G.A. and Sainte-Marie, B. 1997. Cannibalism in the snow crab, Chionoecetes opilio (O. Fabricius) (Brachyura: Majidae), and its potential importance to recruitment. Journal of Experimental Marine Biology and Ecology, 211: 225-245. Mace, P. and Sissenwine, M.P. 1994. How much spawning per recruit is enough ? In: Risk evaluation and biological reference points for fisheries management. (Eds. S.J. Smith, J.J. Hunt and D. Rivard. Special Publication Canadian Journal of Fisheries and Aquatic Science, 120: 101118. Myers, R.A. and Hoenig, J.M. 1997. Direct estimates of gear selectivity from multiple tagging experiments. Canadian Journal of Fisheries and Aquatic Sciences, 54: 1-9. Orensanz, J.M., Parma, A.M., Armstrong, D.A., Armstrong, J. and Wardrup, P. 1995. The breeding ecology of Cancer gracilis (Crustacea: Decapoda: Cancridae) and the mating systems of cancrid crabs. Journal of Zoology, London, 235: 411-437. Pianka, E.R., 1970. On r- and K- selection. The American Naturalist, 104: 592-597. Sainte-Marie, B., 1997. An improved link between industry, management and science: review of case history of the southwestern Gulf of St Lawrence snow crab fishery. Canadian Journal of Fisheries and Aquatic Sciences, 54: 496-500. Sainte-Marie, B., Sevigny, J.M. 1997. Evidence for sperm limitation in the snow crab (Chionoecetes opilio). Abstracts, ICES International Symposium – Recruitment Dynamics of Exploited Marine Populations: Physical-Biological Interactions, Baltimore, USA, September, 1997. p 61. Sainte-Marie, B., Raymond, S. and Brêthes, J.C. 1995. Growth and maturation of the benthic stages of male snow crab, Chionoecetes opilio (Brachyura: Majidae). Canadian Journal of Fisheries and Aquatic Sciences, 52: 903-924. Sainte-Marie, B., Sevigny, J.M., Smith, B.D. and Lovrich, G.A. 1996. Recruitment variability in snow crab, Chionoecetes opilio: pattern, possible causes, and implications for fisheries management. High Latitude Crabs: Biology, Management, and Economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska Fairbanks, 451-478. 296 Sainte-Marie, B., Sevigny, J.M. and Gauthier, Y. 1997. Laboratory behaviour of adolescent and adult males of the snow crab (Chionoecetes opilio)(Brachyura: Majidae) mated noncompetitively and competitively with primiparous females. Canadian Journal of Fisheries and Aquatic Sciences, 54: 239-248. Schultz, D.A., Shirley, T.C., O’Clair, C.E. and Taggart, S.J. 1996. Activity and feeding of ovigerous Dungeness Crabs in Glacier Bay, Alaska. High Latitude Crabs: Biology, Management, and Economics. Alaska Sea Grant College Program Report No. 96-02, University of Alaska Fairbanks, 411-424. Smith, L.D. 1995. Effects of limb autotomy on juvenile blue crab survival from cannibalism. Marine Ecology Progress Series, 116: 65-74. Smith, B.D. and Jamieson, G.S. 1991. Possible consequences of intensive fishing for males on the mating opportunities for Dungeness crabs. Transactions of the American Fisheries Society, 120: 650-653. Sulkin, S.D. and McKeen, G. 1994. Influence of temperature on larval development of four cooccurring species of the Brachyuran genus Cancer. Marine Biology, 118: 593-600. Thresher, R.E. 1997. Evidence of global scale climate forcing of quasi-decadal recruitment cycles in marine and terrestrial populations, with comments on likely forcing mechanisms. Abstracts, ICES International Symposium – Recruitment Dynamics of Exploited Marine Populations: Physical-Biological Interactions, Baltimore, USA, September, 1997. Wilber, D.H. 1989. Reproductive biology and distribution of stone crabs (Xanthidae, Menippe) in the hybrid zone on the northeastern Gulf of Mexico. Marine Ecology Progress Series, 52: 235244. Wilber, D.H. 1992. Observations on the distribution and mating patterns of adult stone crabs (Genus Menippe) on the northern gulf coast of Florida. Proceedings of a Symposium on Stone Crab (Genus Menippe) Biology and Fisheries, Florida Marine Research Publications, 50: 1016. Winemiller, K.O. and Rose, K.A. 1992. Patterns of life-history diversification in North American fishes: implications for population regulation. Canadian Journal of Fisheries and Aquatic Sciences, 49: 2196-2218. Workshop on Stock Recruitment Relationships in Australian Crustacean Fisheries, Proceedings, 1994. (Eds. A.J. Courtney and M.G. Cosgrove). Joondoburri, Queensland Department of Primary Industries and Fisheries, Brisbane. 38-40. 297

Log In

The Larval and Reproductive Biology of the Giant Crab Pseudocarcinus gigas

The Larval and Reproductive Biology of the Giant Crab Pseudocarcinus gigas

The Larval and Reproductive Biology of the Giant Crab Pseudocarcinus gigas

The Larval and Reproductive Biology of the Giant Crab Pseudocarcinus gigas

The Larval and Reproductive Biology of the Giant Crab Pseudocarcinus gigas

Related Papers

RELATED PAPERS