Groundwater copepods: diversity patterns over ecological and evolutionary scales

Diana Galassi

206-210. In the last decade, the biodiversity of groundwaters, including karstic systems, has become intensively studied in Slovenia. We have especially focused on stygobitic copepods which inhabit small fissures and cracks in the so-called vadose (or unsaturated) zone, i.e. the space between fossil cave galleries and the surface. The upper layer of this zone is called the epikarstic zone. Epikarstic species can be collected in trickles of percolat-ing water as well as in the pools filled with it. Faunal samples, along with water for physical and chemical analyses, were collected once per week over a period of 1 yr. Twenty-three taxa of copepods were recorded from the trickles and pools in 3 caves of the Postojnska Jama Cave System. Four taxa of Cyclopoida and 19 taxa of Harpacticoida were recognized. Eight harpacticoid species are probably new to science and belong to 7 genera. We collected 13 species in Postojnska jama, 14 species in Pivka jama, and 11 species in rna jama. The sur...

Hydrobiologia 453/454: 227–253, 2001. R.M. Lopes, J.W. Reid & C.E.F. Rocha (eds), Copepoda: Developments in Ecology, Biology and Systematics. © 2001 Kluwer Academic Publishers. Printed in the Netherlands. 227 Groundwater copepods: diversity patterns over ecological and evolutionary scales Diana M. P. Galassi Dipartimento di Scienze Ambientali, University of L’Aquila, via Vetoio, Coppito, I-67100 L’Aquila, Italy E-mail: diana.galassi@aquila.infn.it Key words: Copepoda, groundwater, stygobiont, diversity, ecology, biogeography Abstract Copepods are common components of the groundwater fauna, and greatly increase the diversity of groundwater communities. With more than 900 species/subspecies known from continental groundwaters, stygobiont copepods inhabit all kinds of aquifers (karstic, fissured, porous), as well as surface/subsurface ecotones (land/water and water/water). The polyhedral and varied structure of the stygohabitats is reflected in the surprising mixture of functional morphologies and habitat exploitations experienced by groundwater copepods. Morphological adaptations and specializations are discussed, as well as the chronology of their appearance in the evolutionary history of several taxa. Diversity patterns of copepod assemblages in groundwater are examined under both structural and functional profiles, as well as across a range of scales. Structure and function operate in an interactive, sometimes hierarchical ways, as well as scales. On the ecological scale, local heterogeneity and patchiness in geomorphic and hydrologic characteristics, as well as biotic interactions, are to be considered causal factors affecting the diversity patterns over a range of spatial and temporal scales. On the evolutionary scale, it is widely accepted that stygobiont copepods evolved from ancestors living in marine, freshwater and semiterrestrial environments. They gained access to the groundwater through major highways represented by the interstitial and the crevicular/karstic corridors. ‘Phylogenetic diversity’ in groundwater copepod taxocoenoses is viewed as a heterogeneous assemblage of species belonging to different phylogenetic lineages, which entered groundwater at different times and by different ways. Introduction In spite of the past view of the groundwater as a relict environment harbouring rare exceptions of the living world, it is now evident that its biodiversity is greater than previously known. Ground water is of vast extent, representing about 40% of the global inland water (Castany, 1982), and is composed of many distinct habitats, more or less interdependent and/or intergrading with each other (Botosaneanu, 1986; Gibert et al., 1994a, 1997; Palmer et al., 1997). Stygohabitats may be found on continents and islands. Moreover, beneath the sea, sediments provide interstitial and crevicular habitats with many features in common with continental and insular stygohabitats. In this review, the terms ‘stygohabitat’ and ‘stygofauna’ are limited to the groundwater beneath the land surface, with particular reference to fresh groundwater. Groundwater habitats in inland waters are developed in three main kinds of aquifers: karstic, fissured and porous, although these categories are not always strictly separate in natural conditions. Moreover, lentic water bodies, stream channels and springs are intimately connected to groundwater through subsurface waters beneath and lateral to the surface open waters. These ‘dynamic transition zones’ (Stanford & Ward, 1993; Gibert et al., 1994a, 1997) represent sites of interactions between surface and groundwater systems (Ward & Palmer, 1994; Vervier et al., 1997). On the basis of Fisher’s (1997) dialectic on lotic ecosystems, questions related to how groundwater is structured and how it functions are of general application, and answers could be sought just as well in other environments. The structure of groundwater can be defined by several aspects, such as geomorphology, hydrology, chemistry, substrate composition, food re- 228 sources and biota. The role of different descriptive elements also varies as a function of scale. The function of groundwater is defined by the role it plays in respect to the environmental context in which it is placed, and by the organisms, with their morphology, behaviour, adaptation and evolutionary history, all attributes defining the fundamental niche of any species. The copepods, one of the major components of the groundwater fauna, successfully invaded subsurface habitats in both marine and continental waters, at different times and in different ways. The polyhedral and variegate structure of stygohabitats is reflected in the surprising mixture of functional morphologies and habitat exploitations developed by groundwater copepods. Of the 10 known orders of Copepoda, 6, namely Platycopioida, Calanoida, Misophrioida, Cyclopoida, Harpacticoida and Gelyelloida, possess stygobiont representatives (Fig. 1), although these are not evenly distributed among and within orders (Figs 2 and 3). Stygobiont platycopioids are known exclusively from anchialine caves, whereas subterranean misophrioids are known from both littoral and anchialine stygohabitats. Neither platycopioids nor misophrioids ever entered fresh groundwater. On the other hand, the Gelyelloida are known only from continental fresh groundwater. The Calanoida are represented in fresh groundwater with a few species belonging to the Diaptomidae, although in this order a comparatively wider diversification occurred in marine and anchialine stygohabitats, where at least 7 families are represented (Fig. 3). The Cyclopoida and the Harpacticoida most successfully invaded inland groundwater, with some 897 species and subspecies, mostly belonging to the harpacticoid families Canthocamptidae, Parastenocarididae, and Ameiridae (Fig. 4) and the cyclopoid family Cyclopidae (Fig. 5). Adaptations to groundwater life According to the degree of adaptation to groundwater life, the copepods can be classified as stygobionts, stygophiles and stygoxenes. Stygobionts are strictly linked to the groundwater environment during their entire life cycle, and frequently show adaptations to the biotic and abiotic conditions of subterranean waters. Stygophiles can live and reproduce in subterranean habitats, as well as in some epigean marginal habitats, such as springs, edaphic habitats, near-surface sediments of running waters, and lentic water bodies; moreover, they may or may not possess incipi- ent troglomorphic features. Moreover, the stygophile condition is not always to be viewed as an intermediate evolutionary step in the ‘stygobization’ process (Stoch, 1995). The ‘stygoxene’ concept, defined as the accidental or occasional presence of species in subterranean waters, should be reconsidered in the light of recent extensive research in groundwater ecology, since the stygoxene concept merges into the stygophile one. Sometimes it is difficult to define the real ecology of the copepods found in groundwater. Many cyclopoids, such as Paracyclops fimbriatus (Fischer, 1853), Eucyclops serrulatus (Fischer, 1851), Megacyclops viridis (Jurine, 1820), and harpacticoids such as Bryocamptus (B.) pygmaeus (Sars, 1863), and B. (R.) zschokkei (Schmeil, 1893), are frequently found in groundwater, sometimes with egg-sacs and copepodids, but this observation is insufficient to assess the ecological affinity of the species with the groundwater environment. For instance, P. fimbriatus is able to live in ponds, moss, different benthic habitats of lakes, streams and springs, irrespective of current speed, temperature, organic matter and granulometric composition of sediments; and it can also live in groundwater. Vagueness and misinterpretation in the ecological categorisation of groundwater fauna may be reflected in ecological studies, where the biota is used (Danielopol, 1989; Williams, 1991; Malard et al., 1994a–c, 1998; Särkkä et al., 1997) to monitor environmental situations, such as the isolation or the contamination of individual aquifers and springs. In particular, stygophile species are generally counted together with stygobionts, and opposed to stygoxenes, in the estimation of the isolation of groundwater habitats in respect to the physically connected epigean ones. As a matter of fact, many so-called stygophiles are generalist species, which continuously enter groundwater through subsurface habitats, and they should be counted together with stygoxenes. The ‘stygophile’ definition should be limited to the eustygophile species, that is, to species which show an ecological preference for the groundwater habitats, and a higher frequency and occurrence in groundwater than in surface habitats. A wide range of copepod body morphologies has found success in groundwater, as a result of different selective pressures and different evolutionary pathways. Whereas adaptation consists of the basic solution to the major environmental descriptors (i.e. darkness, oligotrophy), specialisations represent peculiar and somewhat original solutions of the different groundwater organisms, depending on the 229 Figure 1. Stygobiont representatives among copepod orders. (A) Platycopioida: Antrisocopia prehensilis (after Huys & Boxshall, 1991); (B) Calanoida: Stygodiaptomus petkovskii (after Brancelj, 1991a); (C) Misophrioida: Boxshallia bulbantennulata (after Huys, 1988); (D) Gelyelloida: Gelyella droguei (after Huys & Boxshall, 1991); (E) Harpacticoida: Parastenocaris hispanica (after Martı́nez Arbizu, 1997); (F) Cyclopoida: Graeteriella unisetigera (after Kiefer, 1937). Figure 2. Distribution of stygobiont species/subspecies among copepod orders. genetic background of the species and habitat selection. In spite of some strange or fascinating body structures, such as those of Caspicyclops mirabilis Monchenko, 1986, and Teratocyclops cubensis Plesa, 1981, among cyclopids, Stygepactophanes jurassicus Moeschler & Rouch, 1984 and Psammonitocrella boultoni Rouch, 1992, among harpacticoids, and the entire order Gelyelloida, the body shapes of many groundwater copepods closely conform to those of their closest epigean relatives. As a consequence, body morphology per se is not always indicative of adaptation to groundwater life. For instance, a direct morphological comparison between the common epigean Diacyclops bicuspidatus (Claus, 1857) and the true stygobiont Diacyclops charon (Kiefer, 1931), apart from depigmentation and anophthalmy, reveals no marked differences in body form, size and other features that are considered adaptive to groundwater life, such as the relative length of legs 1–4, antennules and caudal rami. This condition comes from the fact that D. charon lives in karstic pools and lakes: this kind of stygohabitat poses no marked spatial constraint to the fauna that live in it. A similar situation has been observed in some cyclopoids collected from Yucatán cenotes (Fiers et al., 1996) and in Kieferella delamarei (Lescher-Moutoué, 1971), known from a French karstic system. Likewise the stygobiont calanoids, apart from a tendency to reduced body size, anophthalmy and cuticular depigmentation, do not possess peculiar morphological adaptations to the subterranean life, and the observed range of their body morphologies is small. Body specialisation in 230 Figure 3. Distribution of stygobiont representatives within families. (A) Harpacticoida, (B) Cyclopoida, (C) Calanoida, (D) Misophrioida, (E) Platycopioida, (F) Gelyelloida (species-rich families in groundwater marked with arrows). 231 Figure 4. Representativeness of genera within and among harpacticoid families in groundwater (true marine excluded). calanoids is predominantly related to the dimension and configuration of their spatial niche: from the short antennules and the compact body of the hyperbenthic Pseudocyclopiidae, to the long antennules and the slender body of the planktonic Diaptomidae and the pelagic Fosshageniidae. Within calanoids, it seems that diversification in the mouthparts may be the key factor in radiation into groundwater, as suggested by the great diversity observed among Ridgewayiidae and Epacteriscidae, which implies different ways of feeding and swimming (Fosshagen & Iliffe, 1998). An opposite condition is observable in the minute voids of unconsolidated sediments, almost exclusively colonised by members of the Cyclopoida and Harpacticoida. The interstitial environment poses the major constraint of the reduced living space. Together with body depigmentation and anophthalmy/microphthalmy, miniaturisation is the most common feature shared by intersti- tial copepods (Lescher-Moutoué, 1986; Rouch, 1986). Nevertheless, Reid & Strayer (1994) observed that reduction in size does not occur in the reproductive apparatus, and in the female genital segment, most probably because eggs and spermatophores do not follow the reductional trend of the body size. The reduced body size seems often related to a kind of heterochrony, the progenetic paedomorphosis, occurring in some groundwater taxa, of different phylogenetic lineages. This is the case for the entire order Gelyelloida and some cyclopids and harpacticoids, which exhibit a somewhat immature morphological aspect, accompanied by sexual maturity. The recurrent separate genital double somite in some interstitial, inbenthic, and muscicolous copepods belonging to different phylogenetic lineages (i.e. some species of Speocyclops Kiefer, and Bryocyclops Kiefer among cyclopids, some Nitocrella Chappuis, Moraria T. & A. Scott, and 232 Figure 5. Representativeness of genera among and within cyclopoid families in inland groundwater (true marine excluded). Bryocamptus among harpacticoids) may have a similar significance (Galassi et al., 1999a; Dole-Olivier et al., 2000). The boundaries among segments of swimming legs are similarly more or less incomplete in some groundwater cyclopoids, such as Acanthocyclops agamus Kiefer, 1938 and Reidcyclops dimorphus (Reid & Strayer, 1994), and also in some species of Bryocyclops and Haplocyclops Kiefer, found in both moss and subterranean habitats. Copepods show certain other peculiar traits adaptive to the interstitial life. A few cyclopids (C. mirabilis, G. unisetigera) possess a cylindrical habitus, more like the successful harpacticoid body form. In some harpacticoids, such as the marine interstitial Apodopsyllus Kunz, the urosomal boundaries are poorly defined; this worm-like habitus enhances the ability to move within sediments. On the other hand, the additional pseudosomite, anterior to the genital somite, found in other interstitial species apparently optimizes the flexibility of the urosome (Huys & Boxshall, 1991; Boxshall et al., 1993). The recent discovery of the additional pseudosomite in Stolonicyclops heggiensis Reid & Spooner, 1998, collected from ephemeral seepage areas (Georgia, U.S.A.), and in Bryocyclops muscicola (Menzel, 1926), as redescribed by Reid (1999), suggests that typological affinities occur among some surface and interstitial habitats, resulting in the appearance of common features in taxa of different phylogenetic lineages. Another modification in interstitial copepods consists of shortening of the swimming legs; often together with fusion or loss of their relative segments (oligomerization). Interstitial cyclopoids move among sediment particles by pushing their body, wormlike, along the grains. Nevertheless, some stygobiont harpacticoids have relatively longer swimming legs in respect to body diameter (viz. species of Parastenocaris Kessler, Chappuisius Kiefer, and Pseudectinosoma Kunz), most probably because of an alternative locomotion by jerking, crawling, and winding (Glatzel, 1990 for Parastenocaris phyllura Kiefer, 1938). Reduction in setation of the cephalic appendages is another common, although not exclusive, feature in stygobiont cyclopoids. For instance, Pesce & Galassi (1985), Boxshall et al. (1993) and Jaume & Boxshall (1996a) reported the lack of the lateral basipodal seta (vestigial exopod) on the antenna of several stygobiont copepods, though Reid (1991) observed 233 that this seta is also lacking in some epigean benthic species. Most probably, the loss of some setae from the anteriormost part of the body serves to facilitate movement in sandy, muddy and epiphytic environments, and/or to avoid adhesion of particles on the body surface. In regard to current velocity, no peculiar morphological adaptations occur in interstitial hyporheic copepods. The habitat of most species of Rheocyclops Reid & Strayer, 1999 is highly permeable; they may be rheophilic or tolerate a higher current velocity in coarse-grained sediments. The pseudosomite of some Rheocyclops may facilitate better adhesion to sand grains. Interestingly, Ahnert (1998) observed a peculiar behaviour in the parastenocaridid harpacticoid Potamocaris bidens (Noodt, 1955): when exposed to water currents, this animal quickly adheres to sand grains with its caudal rami, and rolls up its body ventrally to avoid displacement. Whereas some data exist on morphological adaptations to groundwater life, very little is known about physiological and behavioural strategies in stygobiont copepods (Glatzel & Schminke, 1996). Stygobiont cyclopids often lack true egg sacs (Lescher-Moutoué, 1973). Females with egg sacs have never been found in the ridgewayiid calanoid Exumella mediterranea Jaume & Boxshall, 1995, in the ectinosomatid harpacticoid P. janineae Galassi Dole-Olivier & De Laurentiis, 1999, as well as in some Parastenocaris, in which two eggs at a time are released (Glatzel, 1990). No egg sac has ever been observed in Nitocrellopsis rouchi Galassi et al., 1999, although numerous specimens were collected in the type locality over a full year; nor in the marine endobenthic harpacticoid Metahuntemannia Smirnov (see Dahms & Pottek, 1992). Both Nitocrella kunzi Galassi & De Laurentiis, 1997, and N. pescei Galassi & De Laurentiis, 1997, have never been observed to carry the egg sac, although 550 specimens were collected over a 1-year period (Galassi & De Laurentiis, unpublished). The number of broods from a single copulation is variable, as well as the periodicity of broods, but stygobiont copepods produce fewer (though larger) eggs than epigean species, to secure an endogenous food supply for the nauplii in the oligotrophic subterranean environment. In regard to the post-embryonic development, from the sparse data available, the C5 female has never been observed in the stygobiont species Parastenocaris phyllura (Glatzel, 1990, 1991). Stygobiont cyclopids require a longer time for egg development, as pointed out by Lescher-Moutoué (1973) and indirectly confirmed by Bjornberg & Por (1986), in a comparative study of the epigean Bryocyclops and stygobiont Speocyclops. Stygobiont harpacticoids show a general trend toward prolongation of the post-embryonic development: from 3 to 9 weeks in epigean species vs. 13 to 16 weeks in stygobiont species (Rouch, 1968; Glatzel, 1990). Subterranean cyclopoids and harpacticoids develop more slowly (although they live longer) than surface relatives, with one or more generations per year. LescherMoutoué (1973) found peaks in spring and autumn for cyclopids. Harpacticoids have life cycles comparable with that observed by Sarvala (1990) in a lacustrine profundal population of P. schmeili, although their continuous reproduction makes it difficult to separate successive generations by sampling wild populations (Rouch, 1961, 1968). Copepod diversity in groundwater In spite of the general impression in the published literature that groundwater communities have low biodiversity, Rouch & Danielopol (1997) demonstrated that subterranean fresh waters are richer in species than was previously accepted. Comparative evaluations of the species richness from different subsurface habitats revealed that the total number of copepod species in hyporheic situations (Rouch, 1988, 1991, 1995; Rouch & Lescher-Moutoué, 1992) is higher or comparable to that in benthic surface habitats (Amoros & Mathieu, 1984; Rundle & Ormerod, 1991; Rouch & Lescher-Moutoué, 1992; Robertson et al., 1995; Rouch & Danielopol, 1997). Some exceptions are the copepod assemblage in a Brazilian ‘wet campo’, (Reid, 1984, 1993a), and the benthic copepod taxa found in the Lake Pääjärvi (Sarvala, 1986). Species richness values decrease in the phreatic zone, because of the rarefaction of the surface components of the taxocoenoses. Structural level After the last estimation (Botosaneanu, 1986), the number of stygobiont copepods in both fresh and anchialine habitats increases at different hierarchical levels (Table 1). Many new taxa and new records of known species are now available from unexpected zoogeographical regions, offering hints for more profound speculation and hypotheses on the evolutionary pathways of different copepod lineages, and on the different routes of colonisation followed by surface ancestors to enter the groundwater. 234 Table 1. Stygobiont copepods at ordinal, familial, and generic levels described after Botosaneanu (1986)’s publication of Stygofauna Mundi Platycopioida Fosshagen, 1985 (former calanoid Platycopiidae) Antrisocopia Fosshagen (in Fosshagen & Iliffe, 1985) Nanocopia Fosshagen (in Fosshagen & Iliffe, 1988) Calanoida Arietellidae Sars, 1902 Paramisophria T. Scott, 1897 (first stygobiont representatives in Ohtsuka et al., 1993) Boholiniidae Fosshagen (in Fosshagen & Iliffe, 1989) Boholina Fosshagen, 1990 Epacteriscidae Fosshagen, 1973 Erebonectes Fosshagen (in Fosshagen & Iliffe, 1985) Fosshageniidae Suárez-Morales & Iliffe, 1996 Fosshagenia Suárez-Morales & Iliffe, 1996 Pseudocyclopiidae T. Scott, 1892 Paracyclopia Fosshagen (in Fosshagen & Iliffe, 1985) Stygocyclopia Jaume & Boxshall, 1995 Thompsonopia Jaume, Fosshagen & Iliffe, 1999 Ridgewayiidae M. S. Wilson, 1958 Exumella Fosshagen, 1970 (first stygobiont representative in Jaume & Boxshall, 1995) Exumellina Fosshagen (in Fosshagen & Iliffe, 1998) Battstromia Fosshagen (in Fosshagen & Iliffe, 1991) Misophrioida Misophriidae Brady, 1878 Stygomisophria Ohtsuka, Huys, Boxshall & Itö, 1992 Palpophriidae Boxshall & Jaume, 2000 Palpophria Boxshall & Iliffe, 1987 Speleophriidae Boxshall & Jaume, 2000 Boxshallia Huys, 1988 Dimisophria Boxshall & Iliffe, 1987 Expansophria Boxshall & Iliffe, 1987 Huysia Jaume, Boxshall & Iliffe, 1998 Protospeleophria Jaume, Boxshall & Iliffe, 1998 Speleophria Boxshall & Iliffe, 1986 Speleophriopsis Jaume & Boxshall, 1996 Harpacticoida Rotundiclipeidae Huys, 1988 Rotundiclipeus Huys, 1988 Superornatiremidae Huys, 1996 Intercrusia Huys, 1996 Neoechinophora Huys, 1996 Superornatiremis Huys, 1996 Novocriniidae Huys & Iliffe, 1998 Novocrinia Huys & Iliffe, 1998 Ameiridae Monard, 1927 Psammonitocrella Rouch, 1992 Canthocamptidae Brady, 1880 Epactophanes Mrázek, 1893 (first stygobiont representative in Bruno & Cottarelli, 1999) Paramorariopsis Brancelj, 1991 Lessinocamptus Stoch, 1997 Parastenocarididae Chappuis, 1940 Murunducaris Reid, 1994 Continued on p. 235 235 Table 1. contd. Cyclopoida Speleoithonidae Rocha & Iliffe, 1991 Speleoithona Rocha & Iliffe, 1991 Cyclopidae Burmeister, 1834 Troglocyclops Rocha & Iliffe, 1994 Prehendocyclops Rocha (in Rocha et al., 2000) Caspicyclops Monchenko, 1986 Hesperocyclops Herbst, 1984 Idiocyclops Herbst, 1987 Rheocyclops Reid & Strayer, 1999 Reidcyclops Karanovic, 2000 Itocyclops Reid & Ishida, 2000 Cyclopinidae Sars, 1913 Troglocyclopina Jaume & Boxshall, 1996 Muceddina Jaume & Boxshall, 1996 Oromiina Jaume & Boxshall, 1997 Ginesia Jaume & Boxshall, 1997 Gelyelloida Huys, 1988 (former harpacticoid Gelyellidae) The structural knowledge (The What? Stage, Wiegert, 1988) of copepod diversity in groundwater is related to faunistic inventories and is complicated by some problems related to taxonomic precision. Since the species level of taxa identification is the most important condition for accurate and more precise ecological studies, taxonomic identification (Reid, 1988; Robertson et al., 1997) requires that names of taxa should be correct and up-to-date. Problems of synonymy affect many faunistic lists. For instance, the common harpacticoid canthocamptid genus Elaphoidella Chappuis poses severe problems in the definition of its generic limits, as well as at the specific level (Apostolov, 1985; Rouch, 1986; Reid, 1990; Janetzky et al., 1996; Galassi, 1997a). The cyclopoid Diacyclops Kiefer-Acanthocyclops Kiefer complex is the major ‘hot spot’ in the systematics of the Cyclopidae (Petkovski, 1984; Pesce & Galassi, 1987; Reid & Strayer, 1994; Reid, 1998). The generic status of many taxa, such as the harpacticoid Nitokra and Parapseudoleptomesochra Lang, and most probably Nitocrella and Stygonitocrella Petkovski, is in urgent need of revision. Some stygobiont genera, such as the cyclopoid Austriocyclops Kiefer and the harpacticoid Psammonitocrella Rouch, have controversial positions within subfamilies (Pospisil & Stoch, 1997) and families (Martínez Arbizu & Moura, 1994), respectively. Since the groundwater environment is considered more conservative than surface habitats, frequently the species differ only in a few apparent morphological characters (see Reid, 1998, and Pospisil, 1999, for the taxonomic distinction of the European Acanthocyclops sensitivus (Graeter & Chappuis, 1914) and the North American A. parasensitivus Reid, 1998). In some cases, however, no morphological differences, even in microcharacters, are detectable among isolated populations of the same species (e.g. the Balearic and Sardinian populations of the ridgewayiid Exumella mediterranea Jaume & Boxshall, 1995, and the Sardinian, Balearic and Canarian populations of the cyclopinid Muceddina multispinosa Jaume & Boxshall, 1996; or else the observed differences have controversial taxonomic value. Consequently, several stygobiont taxa have assumed broad, enigmatic distributions (e.g. Diacyclops clandestinus Kiefer, 1926, Graeteriella unisetigera, Elaphoidella elaphoides (Chappuis, 1923/24), and Nitocrella stammeri Chappuis, 1938. On the other hand, certain morphological microcharacters have been incorrectly estimated as stable. Some examples are the number of spines of the anal operculum in the genus Elaphoidella, in which intraspecific variation frequently bridges the gap between species (Galassi, 1997a); or the relative length of the caudal rami and the shape/length of the endopod 3 of leg 4 in the Diacyclops-Acanthocyclops complex, which vary greatly within populations (Stoch, 2001). Taxonomic confusion and misidentification have also occurred 236 with some stygophile and/or stygoxene taxa. This is the case for the cyclopoid species Acanthocyclops robustus (G.O. Sars, 1863) and A. vernalis (Fischer, 1853), both frequently encountered in groundwater; the Eucyclops serrulatus complex; the Paracyclops fimbriatus complex (Karaytug & Boxshall, 1998a, b); and some Bryocamptus groups. The high frequency of sympatry and syntopy of congeneric species represents another problem. As a rule, the existence of co-occurring congeners is overlooked by ecologists, who learned to associate a particular habitus with a given species. Surely this practice sometimes affects the calculations of both αand β-diversities. The most common taxa affected by taxonomic vagueness are species of the harpacticoid genera Elaphoidella, Parastenocaris and Nitocrella, and the cyclopids Diacyclops and Acanthocyclops. It is not uncommon that under a Diacyclops ‘species’ are hidden several different cryptic species (Reid, 1992; Pospisil & Stoch, 1999). Moreover, our recent finescale studies on the meiofauna of the Italian Presciano Spring system (Galassi & De Laurentiis, unpubl.) revealed three different Diacyclops of the languidoidesgroup, as well as two parastenocaridid species, belonging to Parastenocaris and an undescribed genus. Rouch (1995) found four species of Parastenocaris in a 25 m2 -area in the Nert Brook. These observations suggest also that diversity values may be strongly affected by the spatial and temporal scales of the sampling programmes. The functional level ‘How do copepods live in groundwater?’ ‘How do copepods partition habitats?’ Answers to these questions pertain to the functional aspect (The How? and the Why? Stages, Wiegert, 1988) of groundwater biodiversity. The distribution patterns of the stygobiont copepods can be examined across a range of spatial and temporal scales. Moreover, scales may integrate with each other to explain the present distribution of species. Some large scale (biogeographical) patterns can be often explained, or at least supported, by finer scale (ecological) events (Danielopol et al., 1994; Ward et al., 1999; Gibert et al., 2000) or vice versa (Strayer & Reid, 1999). The ecological scale What determines where a species lives is the result of a multifactorial frame of determinants, linked to each other in an interactive, often hierarchical, way. Habitat partitioning of stygobiont copepod taxocoenoses has been described in several studies, but how the patterns are formed is far from being understood. Until now, great emphasis has been given to the role of abiotic environmental parameters in explaining the spatial and temporal distribution of copepods in karstic and porous habitats and related ecotones. According to data from literature, hydrogeology and geomorphology are the prevailing factors affecting taxocoenoses at different scales (the floodplain scale, the reach scale and the gravel bar scale; Ward & Palmer, 1994). Moreover, these factors operate together with other covariables, such as O2 concentration, porosity, organic matter, and temperature. Many trends have been observed and predictive models formulated, but only a few studies have examined the distribution of copepods in groundwater habitats at well-defined spatial and temporal scales and at the species level of identification. At the floodplain scale, Ward & Voelz (1994) investigated the spatial distribution pattern of the interstitial fauna along the South Platte River (U.S.A.). They found no differences in hypogean community composition along a longitudinal gradient. Only a weak altitudinal pattern was observed in surface gravel habitats (hyporheic), as indicated by the presence of Acanthocyclops vernalis, which was never collected downstream. In the same aquifer, Ward & Voelz (1997) found a marked faunal gradient on a scale of meters across the groundwater/surface water ecotone, Parastenocaris sp. being the describer for the hypogean type of habitat distribution, Macrocyclops albidus (Jurine, 1820) for the epigean type, and Diacyclops crassicaudis (G.O. Sars, 1863) for the eurytopic type. On the microspatial scale (habitat), the most detailed knowledge has been developed by Rouch (1988, 1991, 1992), Rouch et al. (1989) and Rouch & Lescher-Moutoué (1992) in describing interstitial copepod assemblages over a 75 m2 area in the Lachein Brook. The microtopography of the copepod species was greatly affected by habitat patchiness. Indeed, Parapseudoleptomesochra subterranea (Chappuis, 1928) was sensitive to the hydrological pattern of the riverbed, preferring conductive sediments; Elaphoidella bouilloni Rouch, 1964 was restricted to a low-porosity, poorly oxygenated site in the same area (Fig. 6), and cyclopoid species richness varied with the fine-scale permeability pattern of the riverbed, the only exception being Eucyclops serrulatus. In the Presciano Spring system (Galassi 237 Figure 6. Microtopography of copepods in the Lachein brook, France (after Rouch, 1991) and in the Presciano Spring, Italy (preliminary data). The species show different fine-scale distributions depending on their own ecological requirements and habitat patchiness. & De Laurentiis, unpubl.), Nitocrella pescei occurred in fissured sites, living with Parapseudoleptomesochra italica Pesce & Petkovski, 1980; whereas N. kunzi preferred fine gravel sites. Diacyclops paolae was restricted to low-porosity sites, rich in organic matter (Fig. 6). Similarly, Dreher et al. (1997) found a positive correlation between O2 concentration and density of Diacyclops languidus s.l. and Diacyclops languidoides s.l. in an old arm of the Danube River at Lobau (Austria). Parastenocaris palmerae Reid, 1991, from Goose Creek (northern Virginia, U.S.A.) is a common inhabitant of the sandy mid-channel and coarse sediment debris, and is rarely found in fine debris patches. It is a sand-interstitial species and requires well-oxygenated sediments (Palmer et al., 1995). The distribution of P. janineae seems to be restricted to the floodplain margins (oxbow lakes and backwaters situated far from the active channel), or to deep zones in the sediments (1 m, 2 m, 2.5 m or 7 m below the surface of the substrate) and/or to upwelling areas of the main channel of the Rhône River (France). On a gravel bar scale in the Rhône, Dole-Olivier & Marmonier (1992a, b) found a clear gradient from the downwelling zone to the upwelling zone. In the upwelling zone, mainly stygobiont species were present, among which the copepod harpacticoid Parastenocaris fontinalis Schnitter & Chappuis, 1914, was dominant. Nitocrellopsis rouchi was found in phreatic waters at different depths (50–650 cm) along the alluvial plain of the river (Gran Gravier area), with many individuals at 550–650 cm, and few at shallower depths. Such a distribution is consistent with other biological observations, since these depths correspond to a transition zone from coarse to fine 238 sediments and a storage zone for organic matter and bacterial activity (Gibert, 1994). Parastenocaris orcina Chappuis, 1938, lives only in sandy sediments of small concretional pools of Castelcivita Cave (southern Italy) (vadose zone), and is never collected from epiphreatic and phreatic waterbodies in the cave. In contrast, Pseudectinosoma kunzi and Acanthocyclops agamus have been found only in the siphon lake of the cave (phreatic zone). On the other hand, several Parastenocaris species, P. phyllura, P. glacialis Noodt, 1954 and P. vicesima Klie, 1935 showed a wider tolerance for certain abiotic parameters, such as oxygen availability and grain size composition (Enckell, 1968, 1969). Moreover, the fact that a species is sampled from gravel or sand is not always indicative of its ‘interstitiality’: Pseudectinosoma reductum from the Presciano Spring system of the Tirino River (central Italy) has been collected from only 1 of 50 interstitial samples in a 2000 m2 sampling area (Fig. 6), most probably ‘trapped’ by the porous alluvial that overlies a deeper karstic conduit (Galassi & De Laurentiis, 1997a). The thermal regime sets limits to the latitudinal and altitudinal ranges of copepod species. Numerous coldstenotherm harpacticoid species, such as Maraenobiotus vejdovskyi Mrázek, 1893; Hypocamptus brehmi (Van Douwe, 1922); and Bryocamptus (A.) cuspidatus (Schmeil, 1893), are relicts of an ancient glacial fauna. During the postglacial period, these species confined to coldwater habitats, such as in highlands, springs and some stygal biotopes (Husmann, 1975; Pesce et al., 1995; Galassi, 1997a; Särkkä et al., 1998). Glaciations have strongly affected the composition of present-day communities of interstitial copepods, as recently observed by Strayer & Reid (1999) for the eastern United States. From this study, it appears that local ecological conditions have weaker effects on species composition. Large-scale climatic changes, during the Pleistocene, had a dominant role in structuring the fine-scale composition of copepod assemblages in shallow hyporheic habitats, to the extent that glaciated sites now contain fewer species of interstitial specialists, and fewer endemic species, than unglaciated sites. On the other hand, Stoch (2000) reported a high degree of endemism among stygobiont species in the Trentino caves (northern Italy), and postulated that deep karstic fissures may have served as refuges for the aquatic fauna during Pleistocene glaciations. The stygobionts recolonised successively the most superficial zones of the aquifer. A similar argument was proposed by Rouch (1986) to explain the survival of Antrocamptus species in different French karstic systems that were covered by ice during Pleistocene. The different conclusions reached by these authors may be explained by the different habitats studied, as the hyporheic interstitial is possibly less protective than the deeper ancient karstic areas. Oxygen concentration is a variable linked to the hydrologic pattern of groundwater sites, temperature and organic matter concentration. Oxygen content may influence the fine-scale distribution of copepod species, as observed by Pospisil (1994); Ward & Voelz (1997); Strayer (1994); Strayer et al. (1997); Dreher et al. (1997); and many others. Nevertheless, many stygobiont species may tolerate hypoxic conditions, and the influence of oxygen on biological distribution is equivocal (Malard & Hervant, 1998). The structure of copepod assemblages in karstic aquifers is less well documented than in porous ones. The great majority of contributions deals with taxonomic descriptions and faunistic inventories. Ecological analyses of copepod assemblages have been extensively performed by Rouch (1977, 1982); Chafiq et al. (1992); Gibert et al. (1994b); and Malard et al. (1998). Whatever the spatial scale considered in limestone aquifers, the overall heterogeneity of karst topography does not permit the recognition of subhabitats. The major functional sub-units are the conductive vs. the capacitive fractures (Gibert et al., 1994b). The low-water velocity habitats are colonised by free-swimming cyclopoids, which are able to live in the pools. Assemblages in conductive fractures are less specialised, and epigean copepods represent the dominant components of these assemblages. In the Baget karstic system (France), Rouch et al. (1993) found harpacticoid species in different locations: P. subterranea dominated in the main channel, while Ceuthonectes gallicus Chappuis, 1928 and Nitocrella gracilis Chappuis, 1955 were common in the lateral, less conductive subsystems. Similarly, Malard & Simon (1997) found that generalist stygophile species [Diacyclops languidoides (Lilljeborg, 1901), and Acanthocyclops venustus (Norman & Scott, 1906)] dominated in the conductive fractures of the Lez system (France). Habitat partitioning is also the result of preferences of individual species. These preferences are expressions of the physiology, morphology, behaviour and ancestry of any species and of species interactions, but the role of biotic factors in structuring groundwater communities is completely unknown. In habitats defined by multiple environmental gradients, centrifu- 239 gal organization can occur, in the sense of a combination of shared and distinct preferences (Wisheu, 1998) among species for each gradient. Only studies focused on differential distribution among age classes and sexes, food selection, predation and interspecific competition can provide information regarding the role of biotic interactions. Micro-scale food patchiness, often related to hydraulic variables, may explain differential densities and the spot distributions of some stygobiont species. Competition for food could magnify some enigmatic negative correlations in the abundance of syntopic species. Competitive exclusion is often invoked to explain the lower diversity in some freshwater meiofaunal communities (Rouch, 1988, 1992; Rundle, 1990; Rouch & Danielopol, 1997; Danielopol et al., 1997); or the disappearance of one species when another becomes abundant. Nevertheless, clear relationships among particulate organic matter (POM) and copepod assemblage densities have been rarely reported (Chafiq et al., 1992; Strayer et al., 1997). In general, POM concentration is weakly related to faunal densities in stygohabitats, generally as a covariable of hydraulic conductivity and aquifer recharge in hyporheic situations and springs. Moreover, as faunal densities are generally reported as total number of individuals irrespective of their body size, comparisons among copepod assemblages with different proportions of large and small species may underestimate the role of organic matter. In general, the high morphological variation in mouthparts among closely related taxa found in the same habitat suggests that trophic diversification may be important to avoid competition. A recent study by Boxshall & Evstigneeva (1994) on the copepod fauna of Lake Baikal demonstrated the role of trophic specialisation in adaptive radiation within the Acanthocyclops/Diacyclops species flocks. Rouch & Lescher-Moutoué (1992) invoked competition to explain the spatial pattern of cyclopoid diversity in Lachein Brook. Indeed, although the gravel bar that they studied was the favoured environment for cyclopoids, diversity was lower than expected, with Graeteriella sp. dominating. Nitocrella pescei and N. kunzi, if sympatric, have rarely been found syntopic in the Presciano Spring (Italy). These species differ in mouthpart morphology, also suggesting habitat partitioning by distinct trophic preferences. Nitocrella pescei and N. kunzi also are distributed differently in respect to the hydrologic pattern of the spring: N. pescei is found in the karstic discontinuities of the carbonate bedrock, and is located near the major openings of the spring; whereas N. kunzi lives deeper in the unconsolidated portion of the aquifer (Galassi & De Laurentiis, 1997b). Because several gradients are present, centrifugal habitat partitioning should be considered. The evolutionary scale It is widely accepted that stygobiont copepods originated from surface ancestors living in marine, freshwater and semiterrestrial environments. The transition from surface to subsurface habitats surely involved crossing the ecotonal boundaries between adjacent habitats. Copepods gained access to groundwater through major highways represented by the interstitial and the crevicular/karstic corridors (Fig. 7). General conclusions on the distribution patterns of the freshwater copepods are still problematical, since some taxa thought to be cosmopolitan or shared between continents, have turned out upon αtaxonomical re-examination, to be much more restricted in distribution (e.g. Reid, 1998). Moreover, lack of data obscures the true distribution patterns of many species. Groundwater copepods are widely distributed around the world, although individual species often show quite different distribution patterns. For instance, while some genera are cosmopolitan, others are endemic to more or less restricted sectors of different biogeographic regions (Lewis, 1986; Reid, 1994; Dussart & Defaye, 1995). An outstanding degree of endemism occurs in groundwater taxa, at familial, generic, and, most frequently, specific taxonomic levels. Most of these taxa are phylogenetic and/or distributional relicts (Figs. 8–10), with no close living relatives (i.e. the harpacticoid Chappuisiidae and Rotundiclipeidae; the cyclopoid Speleoithonidae; the calanoid Boholiniidae, Fosshageniidae and Microdiaptominae), or with close counterparts traceable in disjunct geographical areas, in groundwater (i.e. the Gelyelloida and the harpacticoid Superornatiremidae), or in different habitats (the harpacticoid Novocriniidae, the calanoid Exumella Fosshagen, the ectinosomatid Pseudectinosoma). Several species of Diacyclops, Speocyclops and Graeteriella among cyclopoids, and Nitocrella, Elaphoidella and Parastenocaris among harpacticoids, show ‘spot’ distributions, sometimes confined to a single habitat of the typelocality. These distributions are in sharp contrast with other ones, such as those of Parastenocaris glacialis, Parastenocaris fontinalis, Elaphoidella elaphoides (Fig. 11), Nitocrella stammeri, Speocyclops demetiensis (Scourfield, 1932) and Graeteriella unisetigera. 240 Figure 7. Stygobiont copepods originated from epigean ancestors living in marine, freshwater and semiterrestrial habitats. Figure 8. ‘Spot’ distributions of anchialine taxa at different hierarchical levels. Most of them are phylogenetic relicts, with no close living relatives in any other epigean and subterranean habitats. Assuming that the taxonomic identifications are correct, some wide distribution patterns are reasonably linked to dispersal by means of passive transport, as a consequence of the ecological plasticity of the species, sometimes able to live in different habitats such as moss, superficial inbenthic, krenal, and hyporheic interstitial (Reid, 1995; Fiers & Ghenne, 2000). Enckell (1995) hypothesised the presence of resting stages in the genus Parastenocaris to explain enigmatic aspects of the aptitude for dispersal of Parastenocaris brevipes Kessler, 1913. Stream capture may explain the wide distribution of E. elaphoides and G. unisetigera. Several anchialine stygobiont copepods with marine origins, such as the misophrioids Speleophriopsis Jaume & Boxshall and Expansophria Boxshall & Iliffe, the cyclopinid Muceddina Jaume & Boxshall, the cyclopoid Neocyclops Gurney, the calanoid Exumella, and the harpacticoid Superornatiremidae, exhibit disjunct distributions embracing the ancient limits of the Tertiary Tethys Sea (Fig. 12). On the basis of their low potential for dispersal, the present-day distribution of these taxa may be the result of continental drift by plate tectonics. Similar distributional trends are now emerging in freshwater stygobionts of marine origin, such as the ectinosomatid Pseudectinosoma and the Gelyelloida (Fig. 13). Similarly, Reid et al. (1999) considered the North-American cyclopid Rheocyclops as possibly the vicariant counterpart of the European Speocyclops, which may have diverged after the continents split. 241 Figure 9. Several freshwater stygobiont taxa, at different hierarchical levels, show restricted distributions in the Palaearctic Region, sometimes confined to a single habitat of the type-locality. Most of them are phylogenetic relicts, with no close living relatives in any other habitat. Figure 10. Among freshwater stygobiont copepods, several genera with more or less restricted distributions and no close living relatives in epigean habitats, are now known from the Nearctic Region. While it is widely accepted that vicariance, in the meaning of fragmentation of the original distributional range of a taxon, leads to isolated populations and possibly to the origin of stygobiont taxa, the chronology of vicariant events in respect to both colonisation and speciation processes in groundwater remains open to question. As pointed out by Danielopol et al. (1994), Figure 11. Distribution of Elaphoidella elaphoides (Chappuis, 1923/24). two mechanisms could have operated. In the first, the distribution range of the ancestor, already stygobiont, was fragmented by geological, climatic and geographic events that acted over different, large scales (continental separation, marine regressions, climatic fluctuations). As we can know the age of the barrier, as well as the phylogenetic relationships among taxa belonging to the same phylogenetic lineage and the degree of adaptation to groundwater life, we can in- 242 Figure 12. Several anchialine taxa, at different hierarchical levels, show distributions embracing the paleocoastlines of the Tethys Sea, fitting the ‘Tethyan track’ of distribution. fer the time since each taxon speciated. The second supposed mechanism is that the ancestor was an epigean species that vicariated when still living in surface environments. The vicariant taxa then separated from each other in surface habitats, and only after this event they could have independently colonised, or not, groundwater and possibly speciated there. In the latter case, the phylogenetic relationships of related taxa would be more difficult to ascertain, since disjunct surface species possibly underwent different selective pressures and entered groundwater at different times and independently from vicariant events. In this case, the age of subterranean taxa cannot be inferred from the age of the barrier, and the phylogenetic relationships among disjunct stygobiont taxa could be complicated by the independent evolutionary paths followed by their surface ancestors. Moreover, if a palaeostygobiont group is concerned, different selective pressures of different environments (i.e. surface marine vs. groundwater) could have fostered morphological differences between surface ancestors and stygobiont descendants, and the genealogical affinity with extant epigean relatives could be masked by morphological divergences. As Coineau (1986) observed for the peracarid isopod Microcharon, the most plesiomorphic species of a given monophyletic group is not always the species that lives in the plesiotypic habitat. So, we can expect also that a stygobiont species could be more plesiomorphic than (or at least as plesiomorphic as) its epigean relatives. This situation stems from the conservative nature of groundwater in respect to the more changeable coastal marine psammolittoral and epigean freshwater. In many cases, no close relatives exist in the plesiotypic epigean habitats. An example is the order Gelyelloida, with two known species from European karstic systems, and another species discovered in a stream interstitial in South Carolina, U.S.A. (J.W. Reid, pers. comm.). At lower hierarchical levels, several cyclopoid and harpacticoid genera are exclusively known from groundwater, with no congeners found in plesiotypic habitats (i.e. the canthocamptids Antrocamptus Chappuis, and Stygepactophanes Moeschler & Rouch; the ameirids Nitocrellopsis, and Psammonitocrella; the monotypic cyclopids Troglocyclops Rocha & Iliffe, Teratocyclops Plesa; the anchialine cyclopinids Oromiina Jaume & Boxshall, and many others). If convincing phylogenies must be based on characters not influenced by direct selection, we often have to deal with features whose adaptive significance is still debatable (Pesce & Galassi, 1985, 1986; Reid, 1991; Boxshall et al., 1993; Reid & Strayer, 1994; Jaume & Boxshall, 1996a; Galassi, 1997b; Galassi et al., 1999b) and, for many characters, unknown. As mentioned in the previous section, simplification of structures, loss of characters and reductions in morphological and physiological traits are common denominators of the stygobiont structural plan, since they define the ‘darkness syndrome’; and troglomorphy is frequently synonymous with ‘structural rudimentation’. One of the major aspects of the adaptation concept in groundwater is represented by the 243 Figure 13. Some freshwater paleo-stygobiont copepods, of ancient marine origin, show distributions related to the opening of the Atlantic Ocean, during the Late Cretaceous-Early Tertiary. origin and the chronology of the appearance of regressive features in stygobiont taxa. Closely related epigean and stygobiont taxa show a general trend towards reduction and/or character losses in their evolutionary history. This is the case for the harpacticoid Sigmatidium-Pseudectinosoma lineage (Galassi et al., 1999b), for the Diacyclops-Acanthocyclops complex among cyclopids, for cave Misophrioida (Boxshall & Jaume, 2000), and for the order Gelyelloida as a whole. The regressive nature of certain characters in stygobiont taxa could also be sometimes regarded as adaptative features to life in groundwater (Parzefall, 1986; Botosaneanu & Holsinger, 1991; Notenboom, 1991; Holsinger, 1992; Boutin, 1994; Culver et al., 1995). This dualism raises the question whether “... these adaptative features arose before (‘adoption’) or after (‘adaptation’) the ancestral populations invaded groundwaters” (Danielopol & Rouch, 1991). On the basis of a recent study (Galassi et al., 1999b) on the paleobiogeography of the ectinosomatid genus Pseudectinosoma, it seems that many regressive traits were already present in the closely related marine epigean Sigmatidium Giesbrecht and in the surface-water species P. minor (Kunz, 1935). The phenotype of P. minor tends to be maintained in its stygobiont relatives, supporting the hypothesis that certain regressive features shared by groundwater species of Pseudectinosoma are not the result of direct selection in groundwater, but were already present in the epigean ancestor and were ‘adopted’ by the new environment. In contrast with the widespread oligomerization in stygobiont copepods, primitive character states have been maintained in several stygobiont taxa of different lineages. In these cases, the more stable groundwater possibly ‘froze’ the plesiomorphic structural plan of the ancestors. Pseudectinosoma janineae, known from the alluvial plain of the Rhône River (France) is closer to the epibenthic, Baltic species P. minor than to P. vandeli from a French karstic system. The topology of the stygobiont misophrioids Speleophriopsis and Palpophria Boxshall & Jaume, which diverged earlier in respect to the epigean marine genera according to the cladistic analysis by Boxshall & Jaume (2000), could have a similar explanation. All the members of the family Ameiridae possess a genital double somite, until now considered the plesiomorphic condition in harpacticoids (Huys & Boxshall, 1991). Nevertheless, the genital somite and the first abdominal somite are distinct both dorsally and ventrally in several, but not all, stygobiont Nitocrella, Nitocrellopsis and Parapseudoleptomesochra. With present knowledge, the phylogenetic significance of the presence vs. absence of the genital double somite in harpacticoids is difficult to assess, and the existence of both conditions in congeners further compounds the problem. We could hypothesise that the most ancient stygobiont descendants, as a result of the conservative role of the subterranean environment, have maintained the ‘unfused’ condition (relict character?), present in the primitive marine ancestors and lost in the extant marine descendants. Alternatively, the ‘unfused’ condition 244 could be a secondary condition, regained by heterochrony, as Fiers (1990) observed in the Cancrincolidae. We have also further speculated on the ‘unfused’ condition as positively selected (adopted) by the environment to facilitate movements in the interstitial habitat, in the same way as the additional pseudosomite in some cyclopids. Its plesiomorphic character states, such as the still-distinct first pedigerous somite, the 15-segmented antennule, and the 2-segmented maxillary endopodite, indicate that the monotypic stygobiont Troglocyclops has an ancestral position in the Halicyclopinae and in the Cyclopidae lineage in general (Rocha & Iliffe, 1994). In the entire order Harpacticoida, the plesiomorphic condition of leg 5 is retained only in the freshwater stygobiont Chappuisiidae and in the anchialine Superornatiremidae (Huys & Boxshall, 1991; Huys, 1996; Jaume, 1997). Stygobiont Platycopioida and Misophrioida are considered living fossils in several aspects (Fosshagen & Iliffe, 1985; Boxshall & Jaume, 2000). The cyclopinid Muceddina multispinosa possesses many plesiomorphic characters, as does the monotypic Oromiina within the entire order Cyclopoida (Jaume & Boxshall, 1997). Groundwater copepods originating from surface freshwater ancestors cannot always be distinguished from those with marine origins, in particular when phylogenetic relationships are difficult to ascertain (i.e. Gelyelloida and the harpacticoid Chappuisiidae, Parastenocarididae and Phyllognathopodidae). The usual arguments for a direct marine origin state that a stygobiont group has a marine sister-group, or it has a distribution not far from present-day shorelines (thalassoid) or related to paleocoastlines of ancient epicontinental seas. The marine origin The case of the ectinosomatid Pseudectinosoma history Hypotheses on the origins of groundwater taxa from marine ancestors are intensively debated (Stock, 1980; Iliffe, 1986; Rouch & Danielopol, 1987; Holsinger, 1988, 1994; Boutin & Coineau, 1990; Botosaneanu & Holsinger, 1991; Notenboom, 1991; Coineau & Boutin, 1992). Harpacticoids, and copepods in general, are common representatives of the groundwater fauna, but only rarely (Sewell, 1956; Enckell, 1969; Rouch, 1986; Lewis, 1986; Reid, 1993b) has their distribution at different spatial scales been explained in a paleobiogeographical perspective. The genus Pseudectinosoma has a marine origin and, with the exception of the brackish-water species P. minor, the other species show ‘spot’ distributions, endemic to circum-Mediterranean stygohabitats. The amphiatlantic distribution of the genus could be related to the opening of the Atlantic Ocean in the Late Cretaceous or Early Tertiary, supporting an ancient origin of the genus (Galassi et al., 1997). Moreover, P. minor, linked to oligohaline littoral biotopes, could represent an intermediate level in the evolutionary history of the subterranean freshwater species of the genus from a more ancient marine ancestor (Galassi et al., 1999b). The marine littoral ancestor probably lived along the ancient shorelines of the Mediterranean, but, in contradiction to this assumption, no representatives of the genus are presently known from this basin. A recent cladistic analysis of the genus revealed a good correlation between the pattern of relationships among stygobiont species and the sequence of epicontinental sea regressions in the Mediterranean area during the Miocene. In particular, the phylogenetically older P. vandeli and P. janineae occur in areas flooded by the Tethys Sea at Burdigalian. The regression of this continuous arm of the sea may represent the primary event which led to the fragmentation of the marine ancestor’s range, and to the independent evolution of the isolated populations in the groundwater. The phylogenetically younger P. kunzi and P. reductum may have originated during the Messinian regression of the Mediterranean, since they are distributed in areas covered by the sea until the late Miocene (Messinian). Combining the assumption based on the ancient Tethyan origin of the genus with the fact that the stygobiont species of Pseudectinosoma appear to be restricted to circum-Mediterranean groundwater, we are led to believe that their ancestor invaded groundwaters before or during the Messinian salinity crisis (Hsü et al., 1973; Hsü, 1978). The Upper Tertiary was a period of drastic changes in the Mediterranean marine biota (Jaume & Boxshall, 1996b), which may have provided the suitable landscape that is reflected both in the increasing speciation in groundwater and in the extinction of the putative littoral ancestor from the hypersaline Mediterranean biotopes. The case of the ameirid history The marine origin of the stygobiont representatives of the ameirid harpacticoids is certain, but many doubts exist as to the colonisation pathways followed by their ancestors. The primarily marine family Ameiridae 245 at present contains some 300 species and subspecies (Conroy-Dalton & Huys, 1996). Ameirids sometimes became successful in freshwaters, in which they are especially linked to groundwater situations. Of the seven ameirid genera recorded from fresh groundwater, with some 97 species and subspecies, Nitokra, Parapseudoleptomesochra and Praeleptomesochra are still represented in the marine environment, while Nitocrella has been only sporadically recorded from littoral brackish waters, and is frequently found in both coastal and inland groundwater. Stygonitocrella, Psammonitocrella and Nitocrellopsis are exclusively continental, although some species are known from coastal (not littoral) ground waters, and some others from continental saline and brackish waters. The genus Nitokra is commonly found in limnic superficial habitats, and rarely in ground water, suggesting an intermediate step in the colonisation of continental waters via surface habitats (Rouch, 1986). In contrast, Nitocrella, Parapseudoleptomesochra, Nitocrellopsis, Stygonitocrella and Psammonitocrella lack such intermediate representatives, and possibly colonised groundwater via coastal interstitial and, less frequently, via karstic discontinuities of the carbonatic platform, bypassing the limnicoid condition, according to the “Two step model evolution” (Boutin & Coineau, 1990; Notenboom, 1991). Nitocrella, Parapseudoleptomesochra, Stygonitocrella and Nitocrellopsis have wide distributions at the generic level, and occur in both the Old and the New World. The genus Nitocrella is the largest taxon, with 56 species. The wide distributions of these genera, together with the fact that no epigean representatives are known for any genus (the only exception being Nitocrella unispinosus Shen & Tai, 1973), indicate an ancient generic differentiation in groundwater, which could have taken place before the opening of the Atlantic Ocean. With present knowledge it is difficult to assess whether the success of Nitocrella in fresh groundwater is linked to radiation or to multiple invasions (Fig. 14). The radiation hypothesis (monophyletic origin) implies that closely related species evolve from an immediate common thalassostygobiont ancestor by geographic speciation or by niche differentiation processes. In the radiation hypothesis, all the members of the genus should share the same degree of apomorphy and should possess the same degree of adaptation to groundwater. Nevertheless, if geographic isolation by fragmentation of the ancestor’s range (Notenboom, 1991) occurred prior to the regressive event, different patterns of shared apomorphies would be expected in the descendants (stygobiont species derived from peripheral populations of the ancestor). The multiple invasion hypothesis (polyphyletic origin) means that groundwater has been colonised by different thalassostygobiont ancestors, possibly originating from peripatric speciation of a more ancient, widely distributed population. Analysis of the origin of the Nitocrella species is further complicated by the fact that radiation and multiple invasion are not mutually exclusive, and a combination of both processes may have confused the observed phylogenetic and geographic patterns. In an ‘among genera approach’, there are two extreme hypotheses to test, although intermediate situations are also possible: 1. Nitocrella, Nitocrellopsis, Stygonitocrella and Parapseudoleptomesochra derived from a common psammolittoral ancestor, which entered fresh groundwater early, splitting into several lineages; 2. these genera originated from different ancestors, which entered fresh groundwater independently, though not necessarily at the same time. We can hypothesize that an ancient marine ancestor could have colonised the continental groundwater (the first invasion linked to an earlier regressive phase of the sea). Successively, additional invasions could have taken place, starting from the first marine ancestor’s descendants, more apomorphic in respect to the first invader. In this case: 1. the freshwater stygobiont stock originating by the first invasion should be phylogenetically older (more plesiomorphic as a result of the conservative role of the groundwater environment) than those originating from the subsequent invasions; 2. the degree of adaptation to groundwater should be different among consecutive invaders, since they spent different times in this environment. With present knowledge, it is difficult to set the sequence of troglobisation and speciation processes of freshwater stygobiont ameirids, as well as patterns of relationship among and within stygobiont genera, because of inadequate knowledge of the systematics and phylogeny of the entire family Ameiridae. Many features characterising the stygobiont ameirid genera are regressive, and this status may in some respects be the result of convergence. The segmental patterns of the swimming legs, which Petkovski (1976) used as the only characterization of the generic boundaries, should be analysed together with certain other features (i.e. the structural plan of the cephalic appendages, the urosomal segmentation in females, the structure of the genital field). The generic boundaries of the ‘Nitocrella-related genera’ are presently weakly defined, and possibly at least some 246 Figure 14. The radiation hypothesis and the multiple invasion hypothesis are tested in relation with the diversification of the ameirid genus Nitocrella Chappuis. The same argumentation is extendable to an ‘among genera’ approach (after Ward et al., 2000, modified). genera are polyphyletic. Mielke (1995) questioned the monophyletism of Parapseudoleptomesochra, as defined by Pesce & Petkovski (1980) and Martínez Arbizu & Moura (1994) questioned the position of Psammonitocrella in the Ameiridae. Galassi & De Laurentiis (1997b) and Galassi et al. (1999a) recognized morphological groups in both Nitocrella and Nitokra, which could be considered as distinct lineages. Only a clear assessment of the phylogenetic significance of regressive characters and the recognition of homoplasies will allow us to reconstruct a natural system of relationships among and within the genera. The segmental pattern of the swimming legs may have a lesser phylogenetic significance in respect to other more informative characters. The freshwater origin The so-called freshwater origin means that the closest relatives of some groundwater taxa lived in freshwater or semiterrestrial environments. Some limnicoid taxa possess more ancient marine ancestors, but the affinities with extant groups are in many cases questionable, because they are weak and difficult to reconstruct. Harpacticoids successfully invaded groundwater, with representatives of the Canthocamptidae and the Parastenocarididae. Among canthocamptids, the cosmopolitan genus Elaphoidella is the most frequent and abundant genus in groundwater, and shows different habitat preferences depending on the geo- graphic location of the species. They are mainly stygobionts in the Holarctic Region, and mainly epigean benthic/semiterrestrial inhabitants in the tropics. Of the 132 stygobiont species/subspecies known at present, some 36 come from non-Palaearctic Africa and from the Neotropical Region, whereas some 70 are epigean in the tropics. Conversely, a few species are found in surface waters in the Holartic Region, most probably as secondary post-glacial recolonisation [E. bidens (Schmeil, 1894), E. gracilis (Sars, 1863)]. It seems that certain marginal habitats were the plesiotypic habitats of Elaphoidella. The species continued in those habitats if climatic conditions allowed them to preserve the hydric hyperspace. In the Northern Hemisphere, Quaternary glaciations may have played an important role in extinction of the surface species and increasing colonisation and speciation in groundwater. Dry periods in the tropics may have played a similar but less drastic role, as superficial wet refugia were available only sparsely in the forests. Although in some respects too much emphasised, in others too much criticised, the Climatic Refugium Model (Jeannel, 1923) is still the best explanation of the different habitat exploitation in geographic areas influenced by different paleoclimatic vicissitudes. The process involved active and passive dispersion, and a preliminary part of the process of colonisation might have taken place before the appearance of the ‘environmental constraint’. Expansion of a population from surface to subterranean environments could also be 247 favoured by a drastic change in the original habitat. As the habitat changes, the population can gradually retreat into adjacent, similar habitats, perhaps preserving, in the initial phase, its fundamental niche, and adjusting its realized niche. Cold-stenotherm species, living among leaf litter particles and the tiny leaves of moss, in a physical and typological continuum between superficial ‘transitional habitats’ and the true subterranean ones, might have been good candidates to enter groundwater, since they were preadapted to certain environmental parameters shared between marginal habitats and true groundwater. The extinction of the conspecific individuals in the surface environment, although not extendable to all situations, seems to have happened several times (Elaphoidella, Speocyclops). The canthocamptids Moraria, Morariopsis Borutzky and Pseudomoraria Brancelj live in habitats similar to those of Elaphoidella, although they found less success in groundwater, as their preferred habitat remains mosses. Nevertheless, Paramorariopsis Brancelj and Morariopsis species are found in groundwater, in a very explanatory intermediate habitat. Brancelj (1991b) supposed that the preferred habitats of both taxa are crevices in the rock filled by percolating water, or wet moss pillows at the entrances of caves. The family Parastenocarididae requires special mention. At present, it includes 5 genera, 4 of which (Forficatocaris Jakobi, Paraforficatocaris Jakobi, Potamocaris Dussart and Murunducaris Reid) are found exclusively in the Neotropical Region. The highest diversification of the family in the neotropics could be misleading in tentative biogeographical and phylogenetic reconstructions within the family, since some new genera await description from other zoogeographical regions (Schminke, 1981), because of the polyphyletic nature of the genus Parastenocaris (Schminke, 1986). The cosmopolitan Parastenocaris is widely distributed in interstitial habitats of both lotic and lentic habitats, with more than 200 species/subspecies, most with local distributions. Schminke (1981) has argued the ancient colonisation of the fresh groundwater, but Martínez Arbizu & Moura (1994) and Bruno et al. (1998) have recently postulated close relationships with the marine Leptopontiidae. Cyclopoids are well represented in groundwater, although less so than harpacticoids. Diacyclops is surely the most diversified genus, with 71 species/subspecies and several new taxa in the process of being described (Pospisil & Stoch, 1999). The stygobiont species of Diacyclops and Acanthocyclops show a variety of adaptations to groundwater life, from taxa quite similar to the epigean relatives to transformed morphs. The origin of the groundwater species of both genera is difficult to assess. Different groups of species are recognisable as different evolutionary steps. Moreover, different degrees of adaptation to groundwater life are found within each group (Pospisil & Stoch, 1999). The phylogenetic relationships within Diacyclops as a whole are too confused, and multiple invasions and subsequent radiations may have played together to result in the present-day diversification of Diacyclops in groundwater (Stoch, 1995). The genus Speocyclops (Fig. 15) is known from different groundwater habitats in the Palaearctic Region, mostly in southern Europe, with 42 species/subspecies, with one exception in tropical West Africa (Speocyclops transsaharicus Lamoot, Dumont & Pensaert, 1981). The phylogenetic relationships of the genus Speocyclops are difficult to trace, since the systematics of the entire family Cyclopidae and relative subfamilies (Pospisil & Stoch, 1997; Reid et al., 1999) is in need of revision. Although related to Bryocyclops, the closest relationships may be with the genera Graeteriella and Rheocyclops, the latter occupying an intermediate position between Diacyclops and Speocyclops. The distribution pattern of Speocyclops, as well as the ecology of the species, allow us to hypothesise that the genus lived in the Palaearctic Region, originally in surface waters. Speocyclops demetiensis has wide distribution (Fig. 15), and is recorded from both subterranean and epigean habitats. The remaining species are true stygobionts, with the exception of S. transsaharicus, which lives in an epigean habitat in the Ivory Coast, suggesting a plesiotypic surface habitat for the genus. The critical climatic events of the Quaternary led to the extinction of the epigean representatives in the Palaearctic Region, where the distribution and ecology of the only epigean species, Speocyclops demetiensis, is considered a recent event, linked to the post-glacial phase. Conclusion This review has attempted to recognise the key factors structuring copepod assemblages in groundwater. The topic is perhaps too ambitious, but born under the perspective to define the ‘state of the art’ in our knowledge. Recent extensive research in groundwater has resulted in a ‘paradigm shift’, as the concepts of stability, simplicity and predictability of the groundwater environment are no longer sustainable. Distinct pat- 248 Figure 15. Distribution of the cyclopoid genus Speocyclops Kiefer. terns in the structure and composition of copepod assemblages are the result of habitat heterogeneity and patchiness at different scales. Gradients in environmental stability, primarily described by geomorphological and hydrological variations, correspond to gradients in groundwater communities. Disturbance by means of flooding in karst, and spatial and temporal shifts in ecotonal habitats, disrupts the communities’ structure, which recover only slowly after disturbance (Dole-Olivier et al., 1994). The perception of the subterranean realm as a ‘world beyond reality’ has vanished, since it and its biota can be described and approached as other aquatic habitats. As for all the epigean environments, for groundwater also: “Diversity is an assemblage, dynamically variable in time and space – a synthesis of evolutionary changes and real-time ecological constraints. Considering populations as basic elements of the assemblages, different sequences of species colonisation can translate into different community assembly trajectories” (Drake et al., 1996), and different levels of biodiversity. Overemphasis on the ‘darkness syndrome’ obscured attention to the elaborate features by which the stygobiont copepods perceive the environment without light. Notwithstanding the increasing number of taxonomic and faunistic studies dealing with copepods in groundwater, very little is known of their ecology and evolutionary history, and only sparse data are available on their life-cycles, reproductive behaviour, and taxocoenoses dynamics. So far, no attempt has been made to explain the role of copepods as functional components of groundwater communities, although they are often numerically dominant in groundwater situations. Six orders have radiated into groundwater. If platycopioids and misophrioids shared some means of colonisation and habitat, they were unable to pass the salinity boundary and never entered fresh groundwater. Gelyelloids are surrounded by mystery, and great confusion obscures the radiation of cyclopoids and harpacticoids, since their systematics vacillates and present families and genera represent taxonomic limbs, sometimes polyphyletic. Primitive and derived structural plans coexist in stygobiont copepods, since they entered groundwater at different times and by different ways. Several stygobionts, sparsely distributed in taxonomic space, represent phylogenetic relicts, present-day witnesses of very ancient groups, which disappeared from the epigean world. But if in many respects the groundwater is not simple, not stable and not predictable, what made the conservation of the ancient ‘Bauplan’ possible? In a comparative way, the groundwater is more conservative than surface habitats. Although over a geological scale, any karstic area, and on a regional scale, any cave system changed, perhaps, at microhabitat scales, some niches, as well as their inhabitants, were left more or less unaltered: it is as if the groundwater ‘froze’ in stygobionts the plesiomorphic body plan of the ancestors. Nevertheless, 249 this is not the rule among the groundwater copepods. Rudimentary structures are ubiquitous, but how many reductions and losses are products of direct selection in groundwater? How many of them were already present in ancestral epigean lines? What is the role played in character selection by plesiotypic habitats, with many typological affinities with the subterranean ones? Moreover, we lack adequate information on the relationship between some presumptively adaptive characters and the fitness of their possessors. Adaptation is a central question in subterranean biology, and clear statements on the adaptive significance of morphological and physiological features will come from studies on comparative morphology and physiology between closely related epigean and stygobiont taxa. This approach is impossible without better knowledge of the systematics and phylogeny of stygobiont copepods. To summarise and integrate these observations, is the groundwater a challenging environment? If we consider the high diversity observed in groundwater communities, the dogmatic perception of the underground as an inhospitable environment is no longer convincing. Acknowledgements I am particularly grateful to Janet Reid (Smithsonian Institution, Washington D.C., U.S.A.) and to Marie-José Dole-Olivier (University of Lyon, France) for the constructive criticism and editorial expertise on the manuscript. G. Boxshall, A. Brancelj, P. De Laurentiis, F. Fiers, A. Fosshagen, B. Humphreys, R. Huys, D. Jaume, F. Malard, P. Martínez-Arbizu, G. Moura, G. L. Pesce, C. Rocha, F. Stoch and D. L. Strayer, with improvements, suggestions and unpublished data, made this review possible. Nevertheless, responsibility for errors and misinterpretations falls only on me. This research was partially supported by the Italian “Ministero dell’Università e della Ricerca Scientifica e Tecnologica” Cluster 11-B and by “Cofinanziamento M.U.R.S.T.” References Ahnert, A., 1998. Has the main habitat of Potamocaris species been overlooked until now? (Harpacticoida, Parastenocarididae). J. mar. Syst. 15: 121–125. Amoros, C. & J. Mathieu, 1984. Structure et fonctionnement des écosystèmes du Haut-Rhône français. 35 – Relations entre les eaux interstitielles et les eaux superficiels: influence sur les peuplements de Copépodes Cyclopoïdes (Crustacés). Hydrobiologia 108: 273–280. Apostolov, A., 1985. Etude sur quelques Copépodes harpacticoides du genre Elaphoidella Chappuis, 1929 de Bulgarie avec une révision du genre. Acta Mus. Mac. Sci. nat. 17: 135–163. Bjornberg, M. H. G. C. & F. D. Por, 1986. Comparative notes on the development of two species of Bryocyclops (Copepoda, Cyclopoida). In Schriever, G., H. K. Schminke and C.-t. Shih (eds), Proceedings of the Second International Conference on Copepoda, Syllogeus 58: 229–231. Botosaneanu, L., 1986. General Introduction. In Botosaneanu, L. (ed.), Stygofauna Mundi – A Faunistic, Distributional and Ecological Synthesis of the World Fauna Inhabiting Subterranean Waters (Including the Marine Interstitial). E. J. Brill, Leiden: 1–4. Botosaneanu, L. & J. R. Holsinger, 1991. Some aspects concerning colonization of the subterranean realm – especially of subterranenan waters: a response to Rouch & Danielopol, 1987. Stygologia 6: 11–39. Boutin, C., 1994. Phylogeny and biogeography of metacrangonyctid amphipods in North Africa. Hydrobiologia 287: 49–64. Boutin, C. & N. Coineau, 1990. ‘Regression Model’, ‘Modèle Biphase’ d’évolution et origine des micro-organismes stygobies interstitiels continentaux. Rev. Micropaléontol. 33: 303–322. Boxshall, G. A. & T. D. Evstigneeva, 1994. The evolution of species flocks of copepods in Lake Baikal: a preliminary analysis. Ergebn. Limnol. 44: 235–245. Boxshall, G. A., T. D. Evstigneeva & P. F. Clark, 1993. A new interstitial cyclopoid copepod from a sandy beach on the western shore of Lake Baikal, Siberia. Hydrobiologia 268: 99–107. Boxshall, G. A. & D. Jaume, 2000. Discoveries of cave misophrioids (Crustacea: Copepoda) shed new light on the origin of anchihaline faunas. Zool. Anz. 239: 1–19. Brancelj, A., 1991a. Stygobitic Calanoida (Crustacea: Copepoda) from Yugoslavia with description of a new species – Stygodiaptomus petkovskii from Bosnia and Hercegovina. Stygologia 6: 165–176. Brancelj, A., 1991b. Paramorariopsis anae gen. n., sp. n. and the female of Ceuthonectes rouchi Petkovski, 1984 – two interesting harpacticoids (Copepoda: Crustacea) from caves in Slovenia (NW Yugoslavia). Stygologia 6: 193–200. Bruno, M. C. & V. Cottarelli, 1999. Harpacticoids from groundwaters in the Philippines: Parastenocaris mangyans, new species, Epactophanes philippinus, new species, and redescription of Phyllognathopus bassoti (Copepoda). J. crust. Biol. 19: 510– 529. Bruno, M. C., V. Cottarelli & R. Berera, 1998. Preliminary remarks on the cladistic systematics in some taxa of Leptopontiidae and Parastenocarididae (Copepoda, Harpacticoida). Mem. Mus. Civ. St. Nat. Verona 13: 69–79. Castany, G., 1982. Principes et méthodes de l’hydrogéologie. Dunod Université, Paris: 238 pp. Chafiq, M., J. Gibert, P. Marmonier, M. – J. Dole-Olivier & J. Juget, 1992. Spring ecotone and gradient study of interstitial fauna along two floodplain tributaries of the River Rhône, France. Reg. Riv. Res. Manage. 7: 103–115. Coineau, N., 1986. Isopoda: Asellota: Janiroidea. In Botosaneanu, L. (ed.), Stygofauna Mundi – A Faunistic, Distributional and Ecological Synthesis of the World Fauna Inhabiting Subterranean Waters (Including the Marine Interstitial). E. J. Brill, Leiden: 465–472. Coineau, N. & C. Boutin, 1992. Biological processes in space and time. Colonization, evolution and speciation in interstitial 250 stygobionts. In Camacho, A. I. (ed.), The Natural History of Biospeleology. Mus. Nac. Cienc.nat., CSIC Ed., Monografias, Madrid 7: 427–451. Conroy-Dalton, S. & R. Huys, 1996. Towards a revision of Ameira Boeck, 1865 (Harpacticoida, Ameiridae): re-examination of the A. tenella-group and the establishment of Filexilia gen. n. and Glabameira gen. n. Zool. Scr. 25: 317–339. Culver, D. C., T. C. Kane & D. W. Fong, 1995. Adaptation and Natural Selection in Caves. The Evolution of Gammarus minus. Harvard University Press, Cambridge: 223 pp. Dahms, H.-U & M. Pottek, 1992. Metahuntemmannia Smirnov, 1946 and Talpina gen. nov. (Copepoda, Harpacticoida) from the deep-sea of the high Antarctic Weddell Sea with a description of eight new species. Microfauna Marina 7: 7–68. Danielopol, D., 1989. Groundwater fauna associated with riverine aquifers. J. n. am. Benthol. Soc. 8: 18–35. Danielopol, D. L., P. Marmonier, A. J. Boulton & G. Bonaduce, 1994. World subterranean ostracod biogeography: dispersal or vicariance. Hydrobiologia 287: 119–129. Danielopol, D. & R. Rouch, 1991. L’adaptation des organismes au milieu aquatique souterrain. Réflexions sur l’apport des recherches écologiques récentes. Stygologia 6: 129–142. Danielopol, D. L., R. Rouch, P. Pospisil, P. Torreiter & F. Möszlacher, 1997. Ecotonal animal assemblages; their interest for groundwater studies. In Gibert, J., J. Mathieu & F. Fournier (eds), Groundwater/Surface Water Ecotones: Biological Interactions and Management Options International Hydrology Series. Cambridge University Press: 11–20. Dole-Olivier, M.-J., D. M. P. Galassi, P. Marmonier & M. Creuzé des Chatelliers, 2000. The biology and ecology of lotic microcrustaceans. Freshwat. Biol. 44: 63–92. Dole-Olivier, M.-J. & P. Marmonier, 1992a. Ecological requirements of stygofauna in an active channel of the Rhône River. Stygologia 7: 65–75. Dole-Olivier, M.-J. & P. Marmonier, 1992b. Patch distribution of interstitial communities: prevailing factors. Freshwat. Biol. 27: 177–191. Dole-Olivier M.-J., P. Marmonier, M. Creuzé des Chatelliers & D. Martin, 1994. Interstitial fauna associated with the alluvial floodplains of the Rhône River (France). In Gibert, J., D. L. Danielopol & J. A. Stanford (eds), Groundwater Ecology. Academic Press, Inc., San Diego: 313–346. Drake, J. A., C. L. Hewitt, G. R. Huxel & J. Kolasa, 1996. Diversity and higher levels of organization. In Gaston, K. J. (ed.), Biodiversity. A Biology of Numbers and Difference. Blackwell Science Ltd., Oxford: 149–166. Dreher, J. E., P. Pospisil & D. L. Danielopol, 1997. The role of hydrology in defining a groundwater ecosystem. In Gibert, J., J. Mathieu & F. Fournier (eds), Groundwater/Surface Water Ecotones: Biological Interactions and Management Options. International Hydrology Series, Cambridge University Press: 119–126. Dussart, B. H. & D. Defaye, 1995. Copepoda. Introduction to the Copepoda. In Dumont, H. J. (ed.), Guides to the Identification of the Microinvertebrates of the Continental Waters of the World. SPB Academic Publishing bv, Amsterdam 7: 277 pp. Enckell, P. H., 1968. Oxygen availability and microdistribution of interstitial mesofauna in Swedish fresh-water sandy beaches. Oikos 19: 271–291. Enckell, P. H., 1969. Distribution and dispersal of Parastenocarididae (Copepoda) in northern Europe. Oikos 20: 493–507. Enckell, P. H., 1995. Parastenocaris glacialis (Crustacea: Copepoda: Parastenocarididae) in the Faroe Islands. Fróoskaparrit 43. bók.: 101–105. Fiers, F., 1990. Abscondicola humesi n. gen. n. sp. from the gill chambers of land crabs and the definition of Cancrincolidae n. fam. (Copepoda, Harpacticoida). Bull. Inst. r. Sci. nat. Belg. (Biologie) 60: 69–103. Fiers, F. & V. Ghenne, 2000. Cryptozoic copepods from Belgium: diversity and biogeographic implications. Belg. J. Zool. 130: 11– 19. Fiers, F., J. W. Reid, T. M Iliffe & E. Suárez-Morales, 1996. New hypogean cyclopoid copepods (Crustacea) from the Yucatán Peninsula, Mexico. Contr. Zool. 66: 65–102. Fisher, S. G., 1997. Creativity, idea generation and the functional morphology of streams. J. n. am. Benthol. Soc. 16: 305–318. Fosshagen, A. & T. M. Iliffe, 1985. Two new genera of Calanoida and a new order of Copepoda, Platycopioida, from marine caves on Bermuda. Sarsia 70: 345–358. Fosshagen, A. & T. M. Iliffe, 1998. A new genus of the Ridgewayiidae (Copepoda, Calanoida) from an anchialine cave in the Bahamas. J. mar. Syst. 15: 373–380. Galassi, D. M. P., 1997a. Little known harpacticoid copepods from Italy, and description of Parastenocaris crenobia n. sp. (Copepoda, Harpacticoida). Crustaceana 70: 694–709. Galassi, D. M. P., 1997b. The genus Pseudectinosoma Kunz, 1935: an update, and description of Pseudectinosoma kunzi sp. n. from Italy (Crustacea: Copepoda: Ectinosomatidae). Arch. Hydrobiol. 139: 277–287. Galassi, D. M. P. & P. De Laurentiis, 1997a. Pseudectinosoma reductum, a new ectinosomatid harpacticoid from spring waters in Italy (Crustacea: Copepoda). Hydrobiologia 356: 81–86. Galassi, D. M. P. & P. De Laurentiis, 1997b. Two new species of Nitocrella from groundwaters of Italy (Crustacea, Copepoda, Harpacticoida). Ital. J. Zool. 64: 367–376. Galassi, D. M. P., P. De Laurentiis & M.-J. Dole-Olivier, 1997. The genus Pseudectinosoma Kunz, 1935 (Crustacea: Copepoda: Ectinosomatidae) in the Mediterranean Region: relict of an ancient Tethyan fauna? XIII International Symposium of Biospeleology, Marrakesh (20–27 April 1997): 40. Galassi, D. M. P., P. De Laurentiis & M.-J. Dole-Olivier, 1999a. Nitocrellopsis rouchi sp. n., a new ameirid harpacticoid from phreatic waters in France (Copepoda: Harpacticoida: Ameiridae). Hydrobiologia 412: 177–189. Galassi, D. M. P., M.-J. Dole-Olivier & P. De Laurentiis, 1999b. Phylogeny and biogeography of the genus Pseudectinosoma, and description of P. janineae sp. n. (Copepoda, Ectinosomatidae). Zool. Scr. 28: 289–303. Gibert, J., 1994. Ecologie et dynamique biogéochimique des systèmes souterrains. Programme Interdisciplinaire de Recherche sur l’Environnement (PIREN). Rapport d’Activités 1991–1994, Lyon: 171 pp. Gibert, J., F. Fournier & J. Mathieu, 1997. The groundwater/surface water ecotone perspective: state of the art. In Gibert, J., J. Mathieu & F. Fournier (eds), Groundwater/Surface Water Ecotones: Biological Interactions and Management Options. International Hydrology Series. Cambridge University Press: 3–8. Gibert, J., F. Malard, M. J. Turquin & R. Laurent, 2000. Karst ecosystems in the Rhône River Basin. In Wilkens, H., D. C. Culver & B. Humpreys (eds), Ecosystems of the World – Subterranean Ecosystems. Elsevier, Amsterdam: 533–558. Gibert, J., J. A. Stanford, M.-J. Dole-Olivier & J. V. Ward, 1994a. Basic attributes of groundwater ecosystems and prospects for research. In Gibert, J., D. L. Danielopol & J. A. Stanford (eds), Groundwater Ecology. Academic Press, Inc., San Diego: 7–40. Gibert, J., Ph. Vervier, F. Malard, R. Laurent, & J.- L. Reygrobellet, 1994b. Dynamics of communities and ecology of karst ecosystems: example of three karsts in eastern and southern France. In 251 J. Gibert, D. L. Danielopol & J. A. Stanford (eds), Groundwater Ecology. Academic Press, Inc., San Diego: 425–450. Glatzel, T., 1990. On the biology of Parastenocaris phyllura Kiefer (Copepoda, Harpacticoida). Stygologia 5: 131–136. Glatzel, T., 1991. Neue morphologische Aspekte und die Copepodid-Stadien von Parastenocaris phyllura Kiefer (Copepoda, Harpacticoida). Zool. Scr. 20: 375–393. Glatzel, T. & H. K. Schminke, 1996. Mating behaviour of the groundwater copepod Parastenocaris phyllura Kiefer, 1938 (Copepoda: Harpacticoida). Contr. Zool. 66: 103–108. Holsinger, J. R., 1988. Troglobites: the evolution of cave-dwelling organisms. Am. Sci. 76: 147–153. Holsinger, J. R., 1992. Two new species of the subterranean amphipod genus Bahadzia (Hadziidae) from the Yucatan Peninsula region of southern Mexico, with an analysis of phylogeny and biogeography of the genus. Stygologia 7: 85–105. Holsinger, J. R., 1994. Pattern and process in the biogeography of subterranean amphipods. Hydrobiologia 287: 131–145. Hsü, K. J., 1978. When the Black Sea was drained. Sci. am. 128: 53–63. Hsü, K. J., W. B. F. Ryan & M. B. Cita, 1973. Late Miocene desiccation of the Mediterranean. Nature 242: 240–244. Husmann, S., 1975. The boreoalpine distribution of groundwater organisms in Europe. Verh. int. Ver. Limnol. 19: 2983–2988. Huys, R., 1988. Boxshallia bulbantennulata gen. et spec. nov. (Copepoda: Misophrioida) from an anchihaline lava pool on Lanzarote, Canary Islands. Stygologia 4: 138–154. Huys, R., 1996. Superornatiremidae fam. nov. (Copepoda: Harpacticoida): an enigmatic family from North Atlantic anchihaline caves. Sci. mar. 60: 497–542. Huys, R. & G. A. Boxshall, 1991. Copepod Evolution. The Ray Society, London: 468 pp. Iliffe, T. M., 1986. The zonation model for the evolution of aquatic faunas in anchialine caves. Stygologia 2: 2–8. Janetzky, W., P. Martínez Arbizu & J. W. Reid, 1996. Attheyella (Canthosella) mervini sp. n. (Canthocamptidae, Harpacticoida) from Jamaican bromeliads. Hydrobiologia 339: 123–135. Jaume, D., 1997. First record of Superornatiremidae (Copepoda: Harpacticoida) from Mediterranean waters, with description of three new species from Balearic anchihaline caves. Sci. mar. 61: 131–152. Jaume, D. & G. Boxshall, 1995. A new species of Exumella (Copepoda: Calanoida: Ridgewayiidae) from anchihaline caves in the Mediterranean. Sarsia 80: 93–105. Jaume, D. & G. A. Boxshall, 1996a. Two new genera of cyclopinid copepods (Crustacea) from anchihaline caves on western Mediterranean and eastern Atlantic islands. Zool. J. linn. Soc. 117: 283–304. Jaume, D. & G. A. Boxshall, 1996b. The persistence of an ancient marine fauna in Mediterranean waters: new evidence from misophrioid copepods living in anchihaline caves. J. nat. Hist. 30: 1583–1595. Jaume, D. & G. A. Boxshall, 1997. Two new genera of cyclopinid copepods (Cyclopoida. Cyclopinidae) from anchihaline caves of the Canary and Balearic Islands, with a key to genera of the family. Zool. J. linn. Soc. 120: 79–101. Jeannel, R., 1923. Sur l’évolution des Coléoptères aveugles et le peuplement des grottes dan le mons du Bihor, en Transylvanie. C. r. Acad. Sci., Paris 176: 1670–1673. Karaytug, S. & G. A. Boxshall, 1998a. Partial revision of Paracyclops Claus, 1893 (Copepoda, Cyclopoida, Cyclopidae) with descriptions of four new species. Bull. nat. Hist. Mus. Lond. (Zool.) 64: 111–205. Karaytug, S. & G. A. Boxshall, 1998b. The Paracyclops fimbriatuscomplex (Copepoda, Cyclopoida): a revision. Zoosystema 20: 563–602. Kiefer, F., 1937. Ueber Systematik und geographische Verbreitung einiger Gruppe stark verkömmerter Cyclopiden (Crustacea, Copepoda). Zool. Jahrb. Syst. 70: 421–442. Lescher-Moutoué, F., 1973. Sur la biologie et l’écologie des copépodes cyclopides hypogés (crustacés). Ann. Spéléol. 28: 429–502. Lescher-Moutoué, F., 1986. Copepoda Cyclopoida Cyclopidae des eaux douces souterraines continentales. In Botosaneanu, L., (ed.), Stygofauna Mundi – A Faunistic, Distributional and Ecological Synthesis of the World Fauna Inhabiting Subterranean Waters (Including the Marine Interstitial). E. J. Brill, Leiden: 299–312. Lewis, M. H., 1986. Biogeographic trends within the freshwater Canthocamptidae (Harpacticoida). In Schriever, G., H. K. Schminke & C.-t. Shih (eds), Proceedings of the Second International Conference on Copepoda. Syllogeus 58: 115–125. Malard, F., G. Crague, M. -J. Turquin & Y. Bouvet, 1994a. Monitoring karstic ground water: the practical aspect of subterranean ecology. Theor. appl. Karstol. 7: 115–126. Malard, F. & F. Hervant, 1998. Oxygen supply and the adaptations of animals in groundwater. Freshwat. Biol. 40: 1–30. Malard, F., J. –L. Reygrobellet, J. Gibert, R. Chappuis, C. Drogue, T. Winiarsky & Y. Bouvet, 1994b. Sensitivity of underground karst ecosystems to human perturbation – Conceptual and methodological framework applied to the experimental site of Terrieu (Herault – France). Verh. int. Ver. Limnol. 24: 1414–1419. Malard F., J. –L. Reygrobellet, J. Mathieu & M. Lafont, 1994c. The use of invertebrate communities to describe groundwater flow and contaminant transport in a fractured rock aquifer. Arch. Hydrobiol. 131: 93–110. Malard, F., J.-L. Reygrobellet & R. Laurent, 1998. Spatial distribution of hypogean invertebrates in an alluvial aquifer polluted by iron and manganese, Rhône River, France. Verh. int. Ver. Limnol. 26: 1590–1594 Malard, F. & K. Simon, 1997. Sampling in wells for describing ecological patterns at a microscale in karst aquifers. In Sasowsky, I. D., D. W. Fong & E. L. White (eds), Conservation and Protection of the Biota of Karst. Karst Water Institute, Charlestown, West Virginia: 46–55. Martínez Arbizu, P., 1997. Parastenocaris hispanica n. sp. (Copepoda: Harpacticoida: Parastenocarididae) from hyporheic groundwaters in Spain and its phylogenetic position within the fontinalis-group of species. Contr. Zool. 66: 215–226. Martínez Arbizu, P. & G. Moura, 1994. The phylogenetic position of the Cylindropsyllinae Sars (Copepoda, Harpacticoida) and the systematic status of the Leptopontiinae Lang. Zool. Beitr. N. F. 35: 55–77. Mielke, W., 1995. Interstitial copepods (Crustacea) from Caribbean coast of Venezuela. Microfauna Marina 10: 41–65. Notenboom, J. 1991. Marine regressions and the evolution of groundwater dwelling amphipods (Crustacea). J. Biogeogr. 18: 437–454. Ohtsuka, S., A. Fosshagen & T. M. Iliffe, 1993. Two new species of Paramisophria (Copepoda, Calanoida, Arietellidae) from anchialine caves on the Canary and Galapagos Islands. Sarsia 78: 57–67. Palmer, M. A., P. Arensburger, P. S. Botts, C. C. Hakenkamp & J. W. Reid, 1995. Disturbance and the community structure of stream invertebrates: patch-specific effects and the role of refugia. Freshwat. Biol. 34: 343–356. 252 Palmer, M. A., A. P. Covich, B. J. Finlay, J. Gibert, K. D. Hyde, R. K. Johnson, T. Kairesalo, S. Lake, C. R. Lovell, R. J. Naiman, C. Ricci, F. Sabater & D. Strayer, 1997. Biodiversity and ecosystem processes in freshwater sediments. Ambio 26: 571–577. Parzefall, J., 1986. Behavioural preadaptations of marine species for the colonisation of caves. Stygologia 2: 144–155. Pesce, G. L. & D. M. P. Galassi, 1985. Due nuovi Diacyclops del complesso ‘languidoides’ (Copepoda: Cyclopidae) di acque sotterranee di Sardegna e considerazioni sul significato evolutivo dell’antenna nei copepodi stigobionti. Boll. Mus. Civ. St. Nat. Verona 12: 411–418. Pesce, G. L. & D. M. P. Galassi, 1986. Taxonomic and phylogenetic value of the armature of coxa and antenna in stygobiont cyclopoid copepods. Atti Convegno U. Z. I., Roma, 1986, Boll. Zool. Modena 53 (suppl.): 58. Pesce, G. L. & D. M. P. Galassi, 1987. New or rare species of Diacyclops Kiefer, 1927 (Copepoda, Cyclopoida) from different groundwater habitats in Italy. Hydrobiologia 148: 103–114. Pesce, G. L., D. M. P. Galassi & V. Cottarelli, 1995. Parastenocaris lorenzae n. sp., and first record of Parastenocaris glacialis Noodt (Copepoda, Harpacticoida) from Italy. Hydrobiologia 302: 97– 101. Pesce, G. L. & T. K. Petkovski, 1980. Parapseudoleptomesochra italica n. sp., a new harpacticoid from subterranean waters of Italy (Crustacea, Copepoda, Ameiridae). Frag. Balc., Mus. Mac. Sc. Nat. 11: 33–42. Petkovski, T. K., 1976. Drei neue Nitocrella - Arten von Kuba, zugleich eine Revision des Genus Nitocrella Chappuis (s. rest.) (Crustacea, Copepoda, Ameiridae). Acta Mus. Mac. Sci. nat. Skopje 15: 1–26. Petkovski, T. K., 1984. Bemerkenswerte Cyclopiden (Crustacea: Copepoda) aus den subterranen Gewässern Sloweniens. Acta Mus. Mac. Sci. nat. Skopje 17: 23–52. Pospisil, P., 1994. The groundwater fauna of a Danube aquifer in the ‘Lobau’ wetland in Vienna, Austria. In Gibert, J., D. L. Danielopol & J. A. Stanford (eds), Groundwater Ecology. Academic Press, Inc., San Diego: 347–366. Pospisil, P., 1999. Acanthocyclops sensitivus (Graeter & Chappuis, 1914) (Copepoda: Cyclopoida) in Austria. Ann. Limnol. 35: 49– 55. Pospisil, P. & F. Stoch, 1997. Rediscovery and redescription of Austriocyclops vindobonae Kiefer, 1964 (Copepoda, Cyclopoida) with remarks on the subfamily Eucyclopinae Kiefer. Crustaceana 70: 901–910. Pospisil, P. & F. Stoch, 1999. Two new species of the Diacyclops languidoides – group (Copepoda, Cyclopoida) from groundwaters of Austria. Hydrobiologia 412: 165–176. Reid, J. W., 1984. Semiterrestrial meiofauna inhabiting a wet campo in central Brazil, with special reference to the Copepoda (Crustacea). Hydrobiologia 118: 95–111. Reid, J. W., 1988. Copepoda (Crustacea) from a seasonal flooded marsh in Rock Creek Stream Valley Park, Maryland. Proc. biol. Soc. Wash. 101: 31–38. Reid, J. W., 1990. Canthocamptus (Elaphoidella) striblingi, new species (Copepoda: Harpacticoida) from Costa Rica. Proc. biol. Soc. Wash. 103: 336–340. Reid, J. W., 1991. Use of fine morphological structures in interpreting the taxonomy and ecology of continental cyclopoid copepods (Crustacea). Anais do IV Encontro Brasileiro de Plâncton, Recife 4: 261–282. Reid, J. W., 1992. Taxonomic problems: a serious impediment to groundwater ecological research in North America. In Stanford, J. A. & J. J. Simons (eds), Proceedings of the First Interna- tional Conference on Ground Water Ecology. American Water Research Association, Bethesda, Maryland: 133–142. Reid, J. W., 1993a. The harpacticoid and cyclopoid copepod fauna in the Cerrado region of central Brazil. 2. Community structures. Acta limnol. bras. 6: 69–81. Reid, J. W., 1993b. The harpacticoid and cyclopoid copepod fauna in the Cerrado region of central Brazil. 1. Species composition, habitats, and zoogeography. Acta limnol. bras. 6: 56–68. Reid, J. W., 1994. Murunducaris juneae, new genus, new species (Copepoda: Harpacticoida: Parastenocarididae) from a wet campo in central Brazil. J. crust. Biol. 14: 771–781. Reid, J. W., 1995. Redescription of Parastenocaris brevipes Kessler and description of a new species of Parastenocaris (Copepoda: Harpacticoida: Parastenocarididae) from the U.S.A. Can. J. Zool. 73: 173–187. Reid, J. W., 1998. How ‘cosmopolitan’ are the continental cyclopoid copepods? Comparison of the North America and Eurasian faunas, with description of Acanthocyclops parasensitivus sp. n. (Copepoda: Cyclopoida) from the U. S. A. Zool. Anz. 236: 109–118. Reid, J. W., 1999. New records of Bryocyclops from the continental U.S.A., Puerto Rico and Brazil (Copepoda: Cyclopoida: Cyclopidae). J. crust. Biol. 19: 84–92. Reid, J. W. & D. L. Strayer, 1994. Diacyclops dimorphus, a new species of copepod from Florida, with comments on morphology of interstitial cyclopine cyclopoids. J. n. am. Benthol. Soc. 13: 250–265. Reid, J. W., D. L. Strayer, J. V. McArthur, S. E. Stibbe & J. J. Lewis, 1999. Rheocyclops, a new genus of copepods from the southeastern and central U.S.A. (Copepoda: Cyclopoida: Cyclopidae). J. crust. Biol. 19: 384–396. Robertson, A. L., J. Lancaster, L. R. Belyea & A. G. Hildrew, 1997. Hydraulic habitat and the assemblage structure of stream benthic microcrustacea. J. n. am. Benthol. Soc. 16: 562–575. Robertson, A. L., J. Lancaster & A. G. Hildrew, 1995. Stream hydraulics and the distribution of microcrustacea: a role for refugia? Freshwat. Biol. 33: 469–484. Rocha, C. E. F. & T. M. Iliffe, 1994. Troglocyclops janstocki, new genus, new species, a very primitive cyclopid (Copepoda: Cyclopoida) from an anchialine cave in the Bahamas. Hydrobiologia 292/293: 105–11. Rouch, R., 1961. Le développement et la croissance des Copépodes Harpacticides cavernicoles (Crustacés). C. r. Acad. Sci., Paris 4: 4062–4064. Rouch, R., 1968. Contribution à la connaissance des harpacticides hypogés (Crustacés-Copépodes). Ann. Spéléol. 23: 5–167. Rouch, R., 1977. Considérations sur l’écosytème karstique. C. r. Acad. Sci. Paris 284: 1101–1103. Rouch, R., 1982. Le système karstique du Baget. XIII – Comparison de la dérive des Harpacticides à l’entrée et à la sortie de l’acquifère. Ann. Limnol. 18: 133–150. Rouch, R., 1986. Copepoda: les Harpacticoïdes souterrains des eaux douces continentales. In Botosaneanu, L. (ed.), Stygofauna Mundi – A Faunistic, Distributional and Ecological Synthesis of the World Fauna Inhabiting Subterranean Waters (Including the Marine Interstitial). E. J. Brill, Leiden: 321–355. Rouch, R, 1988. Sur la répartition spatiale des Crustacés dans le sous-écoulement d’un ruisseau des Pyrénées. Ann. Limnol. 24: 213–234. Rouch, R., 1991. Structure du peuplement des Harpacticides dans le milieu hyporhéique d’un ruisseau des Pyrénées. Ann. Limnol. 27: 227–241. 253 Rouch, R., 1992. Caractéristiques et conditions hydrodynamiques des écoulements dans les sédiments d’un ruisseau des Pyrénées. Implications écologiques. Stygologia 7: 13–25. Rouch, R., 1995. Peuplement des Crustacés dans la zone hyporhéique d’un ruisseau des Pyrénées. Ann. Limnol. 31: 9–28. Rouch, R., M. Bakalowicz, A. Mangin & D. D’Hulst, 1989. Sur les caractéristiques chimiques du sous-écoulement d’un ruisseau des Pyrénées. Ann. Limnol. 25: 3–16. Rouch, R. & D. L. Danielopol, 1987. L’origine de la faune aquatique souterraine entre le paradigme du réfuge et la modèle de la colonization active. Stygologia 3: 345–372. Rouch, R. & D. L. Danielopol, 1997. Species richness of microcrustacea in subterranean freshwater habitats. Comparative analysis and approximate evaluation. Int. Rev. ges. Hydrobiol. 82: 121–145. Rouch, R. & F. Lescher-Moutoué, 1992. Structure du peuplement des Cyclopides (Crustacea: Copepoda) dans le milieu hyporhéique d’un ruisseau des Pyrénées. Stygologia 7: 197–211. Rouch, R., A. Pitzalis & A. Descouens, 1993. Effects d’un pompage à gros débit sur le peuplement des Crustacés d’un acquifère karstique. Ann. Limnol. 29: 15–29. Rundle, S. D., 1990. Micro-arthropod seasonality in streams of varying pH. Freshwat. Biol. 24: 1–21. Rundle, S. D. & S. J. Ormerod, 1991. The influence of chemistry and habitat features on the microcrustacea of some upland Welsh streams. Freshwat. Biol. 26: 439–451. Särkkä, J., L. Levonen & J. Mäkela, 1997. Meiofauna of springs in relation to environmental factors. Hydrobiologia 347: 139–150. Särkkä, J., L. Levonen & J. Mäkela, 1998. Harpacticoid and cyclopoid fauna of groundwater and springs in southern Finland. J. mar. Syst. 15: 155–161. Sarvala, J., 1986. Patterns of benthic copepod assemblages in an oligotrophic lake. Ann. Zool. fenn. 23: 101–130. Sarvala, J., 1990. Complex and flexible life history of a freshwater benthic harpacticoid species. Freshwat. Biol. 23: 523–540. Schminke, H. K., 1981. Perspectives in the study of the zoogeography of interstitial Crustacea: Bathynellacea (Syncarida) and Parastenocarididae (Copepoda). Int. J. Speleol. 11: 83–89. Schminke, H. K., 1986. The systematic confusion within the family Parastenocarididae (Copepoda, Harpacticoida). In Schriever, G., H. K. Schminke and C.-t. Shih (eds), Proceedings of the Second International Conference on Copepoda, Syllogeus 58: 635. Sewell, R. B. S., 1956. The continental drift theory and the distribution of the Copepoda. Proc. linn. Soc. Lond. 166: 149–177. Stanford, J. A. & J. V. Ward, 1993. An ecosystem perspective of alluvial rivers: connectivity and the hyporheic corridor. J. n. am. Benthol. Soc. 12: 48–60. Stoch, F., 1995. The ecological and historical determinants of crustacean diversity in groundwaters, or: why are there so many species? Mém. Biospéol. 22: 139–160. Stoch, F., 2000. Indagini sulla fauna acquatica delle grotte del Trentino (Italia settentrionale). Studi Trent. Sci. Nat., Acta Biol. 74: 117–132. Stoch, F., 2000. How many species of Diacyclops? New taxonomic characters and species richness in a freshwater cyclopoid genus (Copepoda, Cyclopoida). Hydrobiologia (this volume). Stock, J. H., 1980. Regression model evolution as exemplified by the genus Pseudoniphargus (Amphipoda). Bijdr. Dierk. 50: 105– 144. Strayer D. L., 1994. Limits to biological distribution in groundwater. In J. Gibert, D. L. Danielopol & J. A. Stanford (eds), Groundwater Ecology. Academic Press, Inc., San Diego: 287–310. Strayer, D. L., S. E. May, P. Nielsen, W. Wollheim & S. Hausam, 1997. Oxygen, organic matter and sediment granulometry as controls on hyporheic animal communities. Arch. Hydrobiol. 140: 131–144. Strayer, D. L. & J. W. Reid, 1999. Distribution of hyporheic cyclopoids (Crustacea: Copepoda) in the eastern United States. Arch. Hydrobiol. 145: 79–92. Vervier, P., M. H. Valett, C. C. Hakenkamp & M.-J. Dole-Olivier, 1997. Contribution of the groundwater/surface water ecotone concept to our knowledge of river ecosystem functioning. In Gibert, J., J. Mathieu & F. Fournier (eds), Groundwater/Surface Water Ecotones: Biological Interactions and Management Options. International Hydrology Series. Cambridge University Press: 238–242. Ward, J. V., F. Malard, J. A. Stanford & T. Gonser, 2000. Interstitial aquatic fauna of shallow unconsolidated sediments, particularly hyporheic biotopes. In Wilkens, H., D. C. Culver & B. Humpreys (eds), Ecosystems of the World – Subterranean Ecosystems. Elsevier, Amsterdam: 41–58. Ward, J. V. & M. A. Palmer, 1994. Distribution patterns of interstitial freshwater meiofauna over a range of spatial scales, with emphasis on alluvial river-aquifer systems. Hydrobiologia 287: 147–156. Ward, J. V. & N. J. Voelz, 1994. Groundwater fauna of the South Platte River system, Colorado. In Gibert, J., D. L. Danielopol & J. A. Stanford (eds), Groundwater Ecology. Academic Press, Inc., San Diego: 391–423. Ward, J. V. & N. J. Voelz, 1997. Interstitial fauna along an epigeanhypogean gradient in a Rocky Mountain river. In Gibert, J., J. Mathieu & F. Fournier (eds), Groundwater/Surface Water Ecotones: Biological Interactions and Management Options. International Hydrology Series, Cambridge University Press: 37–41. Wiegert, R. G., 1988. Holism and reductionism in ecology: hypotheses, scale and systems. Oikos 53: 267–269. Williams, D. D., 1991. The springs as an interface between groundwater and lotic faunas and as a tool in assessing groundwater quality. Verh. int. Ver. Limnol. 24: 1621–1624. Wisheu, I. C., 1998. How organisms partition habitats: different types of community organization can produce identical patterns. Oikos 83: 246–258.

RELATED PAPERS

RELATED TOPICS

Log In

Groundwater copepods: diversity patterns over ecological and evolutionary scales

Groundwater copepods: diversity patterns over ecological and evolutionary scales

Related Papers

RELATED PAPERS

RELATED TOPICS