22 Rivulariaceae Brian A. Whitton and Pilar Mateo Contents 22.7 Geological Record and Environmental Change ............ 584 Summary...................................................................................... 561 22.8 Discussion .......................................................................... 585 22.1 Introduction ................................................................ 562 References .................................................................................... 586 22.2 22.2.1 22.2.2 22.2.2.1 22.2.2.2 22.2.2.3 22.2.2.4 22.2.2.5 Morphology................................................................. Morphology and Classical Taxonomy ......................... Range of Form ............................................................. Tapering and Hairs ....................................................... Heterocyst .................................................................... Hormogonia ................................................................. Akinete ......................................................................... Colony, Trichomes and Sheath..................................... 562 562 563 563 564 565 566 567 22.3 22.3.1 22.3.2 Molecular Perspective on Taxonomy........................ Status of Classical Genera............................................ Molecular Diversity Within Form-Genera ................... 568 568 569 22.4 22.4.1 22.4.2 22.4.3 Physiology ................................................................... Carbon and Nitrogen .................................................... Phosphorus ................................................................... Molecules with Possible Ecological Effects ................ 570 570 571 574 22.5 22.5.1 22.5.2 22.5.3 22.5.4 Overview of the Genera ............................................. Introduction .................................................................. Calothrix ...................................................................... Rivularia, Dichothrix and Isactis ................................. Gloeotrichia ................................................................. 574 574 575 576 578 22.6 22.6.1 22.6.2 22.6.3 22.6.3.1 22.6.3.2 22.6.3.3 22.6.3.4 22.6.4 Interactions with Other Organisms.......................... Free-Living Phototrophs .............................................. Symbiotic Associations ................................................ Animals ........................................................................ Introduction .................................................................. Destruction of Whole Filaments .................................. Partial Destruction of Filaments .................................. Avoiding the Grazers.................................................... Bacteria and Fungi ....................................................... 581 581 582 582 582 583 583 583 584 B.A. Whitton (*) School of Biological and Biomedical Sciences, Durham University, Durham DH1 3LE, UK e-mail: b.a.whitton@durham.ac.uk P. Mateo Departamento de Biologia, Facultad de Ciencias, Universidad Autonoma de Madrid, 28049 Madrid, Spain e-mail: pilar.mateo@uam.es Summary The Rivulariaceae are treated here as all the cyanobacteria which have trichomes with a marked taper and a basal heterocyst for much of their growth cycle. Molecular sequencing of Calothrix and Rivularia shows that these are heterogeneous and this probably also applies to Dichothrix and Gloeotrichia. The unispecific Isactis has received little study, but is morphologically close to some Calothrix. These should all be treated as form-genera for ecological descriptive purposes until more sequencing studies have been made. Sacconema and Gardnerula are not considered distinct enough to be treated as distinct genera. Colony formation occurs by aggregation of hormogonia and this may lead to the inclusion of more than one genotype. The group as a whole occurs in environments with highly variable P concentrations, usually short periods of relatively high ambient P followed by much longer periods of low P; a few possible exceptions are discussed. Typically organic P exceeds inorganic P and the Rivulariaceae as a whole are especially efficient at using organic phosphates. N2 fixation in Rivulariaceae is highest during the period of high P. In the case of Gloeotrichia echinulata, which sometimes forms blooms in lakes, P acquisition takes place mainly while the organism is growing near the sediments. Its competitive success is favoured by a combination of P-rich sediments and relatively low P concentrations in the water. Long colourless multicellular hairs are formed by many species, including all those with distinct hemispherical and spherical colonies. The hairs not only enhance the surface area for phosphatase activities, but also aid P acquisition from environments where the ambient P concentration is B.A. Whitton (ed.), Ecology of Cyanobacteria II: Their Diversity in Space and Time, DOI 10.1007/978-94-007-3855-3_22, © Springer Science+Business Media B.V. 2012 561 562 B.A. Whitton and P. Mateo mostly low, but with occasional much higher pulses. Although most phosphatase activities take place at the cell surface, some activity often occurs in sheaths and probably also inside the cytoplasmic membrane of older hair cells, suggesting that organic phosphates can sometimes pass through the membrane. Sufficient is known about the relationship between morphology and the environment in the Rivulariaceae to make them excellent environmental indicators. Many of the morphological characters used for classical taxonomic descriptions are expressed only when the organisms are P-limited, but some culture collections and many experimental studies have used media with very high P concentrations, making past interpretation of results doubtful, if not wrong. Fig. 22.1 Young planktonic colonies of Gloeotrichia echinulata (Photo P.V. York, with permission) 22.1 Introduction A number of general accounts of blue-green algae in the nineteenth century grouped genera with obviously tapered trichomes into the family Rivulariaceae, with Rivularia haematites the type species. Geitler (1932) summarized the earlier information, while Desikachary (1959) provided a succinct definition; “Trichomes with a single row of cells, apices generally attenuated or tapering in hair, unbranched or false-branched, sometimes with a distinct intercalary meristematic zone and trichothallic growth; hair with elongated more or less vacuolated cells; heterocysts present or absent, when present basal, intercalary heterocysts also present in some; hormogonia present, akinetes present or absent, when present single or in series”. The 12 genera listed by Geitler and Desikachary included several without heterocysts such as Homoeothrix, but, even before molecular data started to make it possible to assess relationships more critically, most researchers assumed that non-heterocystous species were different. The heterocystous Rivulariaceae include organisms found widely in nature and are in some cases visually conspicuous. Perhaps the best known is the planktonic Gloeotrichia echinulata (Sect. 22.5.4; Fig. 22.1), which can form blooms, but other species of Gloeotrichia, Dichothrix and Rivularia form conspicuous attached or floating colonies in shallow waters (Sect. 22.5.3). This chapter tries to establish to what extent the heterocystous Rivulariaceae are a uniform group phylogenetically and ecologically. At the same time it sets out to explain how phenotypic differences influence the success of a particular genus or species, and conversely, the extent to which one can interpret the environmental features of a site from the morphology of its Rivulariaceae. The reader needs to know some of the older literature in order to understand the argument. 22.2 Morphology 22.2.1 Morphology and Classical Taxonomy The most widely accepted heterocystous genera of Rivulariaceae in classical floras (i.e. those based on International Code of Botanical Nomenclature) have trichomes with a basal heterocyst, a distinct taper to the trichome and a sheath enclosing everything but the heterocyst. Calothrix includes all the forms growing as individual filaments or ill-defined colonies. Four genera form distinct colonies, with Gloeotrichia separated by its ability to form akinetes. The others are distinguished by colony morphology, which is determined by the pattern formed by hormogonia when they first aggregate and also by the division pattern during colony growth. Rivularia colonies have more or less parallel sheathed trichomes inside hemispherical, subspherical or spherical colonies; the trichomes often have false branches. Isactis forms spreading colonies in the marine intertidal, with sparsely branched trichomes perpendicular to underlying rock surface. Dichothrix forms small colonies of various shapes, mostly cushions or dense tufts; trichome branching is conspicuous and there are often many trichomes inside one sheath. Many, but not all, freshwater Dichothrix and Rivularia become partially calcified (Fig. 22.2). The distinctions between genera are not always clear-cut, such as between Dichothrix and Rivularia, but the names still prove useful in floristic accounts of natural communities. However, Rippka et al. (1979) included all heterocystous cyanobacteria with a low or high degree of tapering in Calothrix on the grounds that colony appearance in feral samples seemed to be a doubtful taxonomic trait. Only a brief justification for this statement was given and in any 22 Rivulariaceae 563 Fig. 22.2 (a, b) Dichothrix baueriana in River Caher, draining The Burren, Co. Clare, Ireland, June 2006: (a) Upstream view of calcified boulders; (b) close-up of the colonies during a period of low flow; showing growth on the boulder in the zone influenced by recent fluctuations in water level; (c) Rivularia colonies among developing calcareous crust on boulder in Loch Nam Bakgan, South Uist, Outer Hebrides (Scotland), September 2009; colonies up to 8 mm diameter, with obvious calcification inside them (Photos B.A.W.) case it is doubtful how many of the cultures had been observed closely when the organism was first sampled from nature. Concern about the reliability of names in culture collections using P-rich medium (Chap. 1) is especially relevant to the Rivulariaceae due to the need for the organisms to be at least moderately P-limited to express their characteristic morphological features (Whitton 1987, 1989, 2008, 2009; Berrendero et al. 2008; Sect. 22.4). Such concern extends to the names associated with sequence data deposited in GenBank. in species where the trichome becomes much wider close to the heterocyst, giving the appearance of a spindle-shaped swelling (Schwenender 1894; Friedmann 1956). Many species can form multicellular hairs, structures which were defined by Sinclair and Whitton (1977a) as “a region of the trichome where the cells are much narrower, elongated, highly vacuolated and usually apparently colourless”. This definition fits the way the term is used for descriptive purposes in most floras, including all but one of the species in Geitler (1932). However, other species occur with long tapered structures which resemble hairs in outline, but the cells retain their chlorophyll and do not elongate; these are quite common, at least in the subtropics. An example, Calothrix D764, is discussed in Sect. 25.2.2. The only exception in Geitler (1932), C. kossinskajae, is somewhat like this, but the figures show a scarcely tapered trichome which contracts over one or two cells to a very long extension of the main trichome, but only about one-third of its width. If (as seems likely) this gives rise to hormogonia, they would be little more than 1 mm wide. All the descriptions of species in the genera forming distinct colonies show hairs. In the case of Calothrix, 56 out of the 78 species (72%) reported in the taxonomic literature surveyed by Kirkby and Whitton (1976) formed hairs, while 22.2.2 Range of Form 22.2.2.1 Tapering and Hairs The tapered trichome has a terminal heterocyst at the wider end and growth is often localized in a region near, but not immediately adjacent, to the heterocyst. Although the term ‘meristematic’ is often used in descriptions of Rivulariaceae to indicate this region, its position and the extent to which it is localized varies during growth. It is suggested that the region corresponds to where longitudinal gradients of N and P in the trichome (see below) interact at any one time to provide optimum growth conditions. The region is most distinct 564 82 of 103 (81%, omitting a few uncertain species) did so in the names in CyanoDB.cz (July 2011). However, only 7 out of 29 Calothrix strains (24%) in culture did so, in spite of being tested under a range of environmental conditions known to lead to hair formation (Sinclair and Whitton 1977a). This raises the possibility that isolation procedures may be selecting against hair-forming strains. The hair cells of Gloeotrichia echinulata mostly retain their gas vacuoles (Smith and Peat 1967). In the study by Sinclair and Whitton (1977a), P limitation enhanced tapering in all 34 Rivulariaceae strains tested, but only 12 strains developed hairs. Even in the presence of combined N, P-limitation still led to the formation of hairs in these strains (Sinclair and Whitton 1977b). Three strains formed some hairs under P-rich conditions, although hair frequency increased with increasing P limitation (Sinclair and Whitton 1977a). Much shorter hairs were formed under Fe deficiency in seven strains and under Mg deficiency in one strain, but there was no response to Ca, Mo or SO4 deficiencies. Strains lacking the ability to form hairs had all been described as Calothrix in the culture collections from which they came. In addition to the uncertainty as to whether or not they possessed hairs when originally collected, there is also the possibility that they might have lost this ability during repeat subculture in a P-rich medium. However, when Gloeotrichia aff. pisum colonies with trichomes showing obvious hairs were removed from deepwater rice plants in Bangladesh, all attempts to isolate trichomes able to form hairs in culture failed (Aziz 1985). In addition to nutrient limitation, methods tested without success included changing the light and temperature regimes, spectral composition of the light and incubation in various organic phosphates as the sole P source. It remains uncertain whether an important factor was overlooked or whether the trichomes isolated were not representive of the main colony. Similarly, during isolation of three Calothrix species from an Iraqi marsh rice field, C. fusca and C. parietina both formed hairs in enrichment culture, but, once they were brought into axenic culture, C. parietina failed to do so (Al-Mousawi and Whitton 1983). Cultures grown under increasingly P-limited conditions cease to produce hormogonia. The apical cells then start to develop intrathylakoidal vacuoles, which remain small in species which do not form hairs, but continue to increase in size and lose their photosynthetic pigments in species forming hairs. Formation (where present) can continue for a long period. The hair cells consist largely of vacuoles originating from intrathylakoidal vacuoles and in very long hairs each cell is largely filled by a single vacuole. Rivularia and other colonial forms with hairs which persist for many months usually have a long transition zone from typical vegetative cell to cells with increasingly larger vacuoles and then to the colourless hair cells (Wood 1984; Wood et al. 1986). B.A. Whitton and P. Mateo Multicellular hairs which appear morphologically similar also occur in at least some species of most genera of Stigonematales, the similarity being especially marked in Nostochopsis and Brachytrichia. The ends of the trichomes of Aphanizomenon flos-aquae often taper into short hairlike lengths, while several species of the non-heterocystous Homoeothrix and Ammatoidea also form hairs. This indicates that multicellular hairs have evolved many times in cyanobacteria, just as they have in eukaryotic algae (Whitton 1988). 22.2.2.2 Heterocyst Although all Rivulariaceae can form terminal heterocysts, some also form intercalary ones. These have been observed in many species, but occur much more frequently in some than others. Polarity of the terminal cell is established before the hormogonium is released from the mother trichome. The heterocyst always develops in the cell adjacent to the main trichome i.e. basal, even though the cell furthest from it might be expected to be the most N-limited (Whitton 1987). In contrast to the terminal heterocyst, an intercalary heterocyst has polar nodules (cyanophycin) on both sides, so presumably fixed N can pass in both directions and thus permit further growth in both parts of the trichome. Particularly in Rivularia, a new heterocyst is sometimes developed above the current one and the old one starts to collapse, eventually showing with the light microscope little more than the wall. Typically the new heterocyst has only a single polar nodule and thus becomes the terminal heterocyst. Growth of Calothrix parietina D184 (= PCC 7713) in medium without Fe led to repeat formation of new single-pored heterocysts, with rapid degeneration of the old one (Douglas et al. 1986). The new heterocyst was often separated from the old one by necridial cells. Several illustrations in the literature suggest that the new heterocyst may in some cases be intercalary, though this has never been observed by the authors; if it does occur, this might permit higher rates of N2 fixation. Large Rivularia colonies, which have already grown for several years, sometimes have as many as eight remnant heterocysts beneath the normal terminal one. Presumably each new heterocyst forms as a response to a change in environmental conditions. One such possible response is a loss of the ability of the heterocyst to fix N2 under prolonged periods of high ambient combined N, resulting in the need for a new one to be formed the next time N2 fixation becomes important. All but one of 34 Rivulariaceae (representing all four genera) lost all heterocysts when cultured with 10 mM NaNO3 (Sinclair and Whitton 1977b). However, this concentration of combined N is much higher than likely to occur in most environments. No studies have yet been made on the effects of fluctuating concentrations of combined N supplied at more realistic concentrations. Another possible response is the formation of a new heterocyst due to the deficiency of an element important in 22 Rivulariaceae Fig. 22.3 Trichomes near the centre of a Gloeotrichia pisum colony. The bright coloured heterocysts often occur in members of the Rivulariaceae attached to submerged plants, sometimes green or blue-green (as shown here), but other times bright blue (Photo C. F. Carter, with permission) heterocyst functioning. The new terminal heterocyst is often smaller than the original one. Tests on the influence of Mg, Ca, Fe, Mo or S deficiencies in 13 Calothrix strains showed a particularly marked effect of Mo deficiency on heterocyst frequency, the increase being at least four-fold in each case (Sinclair and Whitton 1977a). Ca and Fe deficiencies also led to obvious increases in heterocyst frequency in most strains. The extra heterocysts developed in both basal and intercalary positions, but most of the latter formed only a single polar nodule, thus leading to a ‘terminal’ heterocyst in the middle of the trichome. The hypothesis is suggested that in these cases the replacement of one type of heterocyst by another may be due to a shift in the type of electron transport system. Presumably some of the element in short supply is transported to the rest of the trichome, thus aiding the formation of heterocysts elsewhere. It is over a century since Fritsch (1907a, b) first commented on the tendency for heterocysts in tropical Rivulariaceae to be blue and subsequent studies have shown blue heterocysts to be widespread, especially in ponds and rice fields (Whitton 1987). This also occurs in temperate regions, although less widely. In other cases they may be bright green (Fig. 22.3) rather than blue. Studies on Calothrix D603 isolated from the surface of a deepwater rice plant in Bangladesh showed that the blue colour increased as cultures became older and trichomes increased in length (Whitton et al. 1986). This is the exact opposite of the situation with some Anabaena, especially tropical strains, where the blue colour persists for some time after differentiation, but eventually disappears. Addition of NH4-N to young Calothrix D603 cultures 565 grown in the absence of combined N led to about 20% heterocysts turning an obvious blue. In shaded forest habitats Calothrix and Dichothrix with pink trichomes have bright blue heterocysts, indicating the absence of phycoerythrin in these heterocysts (B.A.W., unpublished). However, the heterocysts of a lichenized Calothrix form phycoerythrin (see Sect. 22.6.2) The heterocysts of several Rivulariaceae have been reported to germinate and produce short germlings. This was described in most detail for Gloeotrichia raciborskii and Rivularia manginii by Desikachary (1946) and a nonsporulating mutant of Gloeotrichia ghosei by Singh and Tiwari (1970). In all cases the cyanophycin granule by the cross-wall disappeared at an early stage. Even where there is not such a marked reversal of the initial differentiation to form a heterocyst, heterocysts of Rivulariaceae sometimes reform internal structures which have been lost in their initial differentiation (Whitton 1987), such as carboxysomes in Calothrix D603 (Whitton et al. 1986). It was suggested that the heterocyst may have switched back from N2 fixation to CO2 fixation. Another possible explanation for blue heterocysts is that the phycocyanin is behaving as a N store; perhaps it can be mobilized more rapidly than cyanophycin. Either or both features would be useful for an organism living in a highly variable environment. 22.2.2.3 Hormogonia Hormogonia typically develop at the end of the trichome or, in species with a hair, immediately below this; in the latter case the hair becomes detached and disintegrates. The main zone of cell division typically shifts from nearer the base of the trichome to nearer the apex. A necridial cell is formed at the base of each hormogonium in Calothrix D184 and D550 (Wood 1984; Wood et al. 1986) and this is apparently the general situation in Rivulariaceae. Perhaps formation of the necridial cell helps to determine that the adjacent cell will differentiate into a heterocyst. The rice-field isolate Gloeotrichia D613 (see Sect. 22.4.2) showed a distinctive pattern when grown under a relatively high P concentration (Aziz and Whitton 1989). The trichome developed as cell groups, each of which had originated from one mother cell, something previously reported by Schwendener (1894). Typically the cell group of Gloeotrichia D613 consisted of 8 cells; if so, cell 8 distant from the apex differentiated into a necridium, releasing the other 7 cells as a cyanophycin-rich hormogonium. Subsequent to its release, the lowermost cell of the hormogonium differentiated into a heterocyst about 1 day later. Seven is the typical cell number in the hormogonia of many Rivulariaceae, although in some strains there is considerable variation (Fig. 22.4). Hormogonia of Rivulariaceae usually contain all three types of storage body, polyglucoside, cyanophycin and polyphosphate. In C. parietina D184 cultured in N-free 566 Fig. 22.4 Hormogonia from Rivularia biasolettiana (B.A.W.) medium cyanophycin was the most rapid to decrease (Wood et al. 1986). Few, if any, divisions occur in a hormogonium between the time it is differentiated and the time it separates and often not until the first heterocyst is formed. However, subsequent to their release, Gloeotrichia echinulata hormogonia can elongate and then form necridia and fragment into short lengths (Maxwell 1974). Something similar was reported by Adamec et al. (2005) for Calothrix elenkinii, though their growth conditions make it hard to interpret the significance. When a culture grown in BG11 medium (high combined N and high P) was transferred to N-free medium and then exposed to green light, there was a transient development of a trichome cell with high fluorescence initiated by phycoerythrin absorption; this was followed by bleaching within about 20 min. These necridia permitted trichome fragmentation. The authors suggested that this allowed filaments to escape unfavourable environmental conditions; in this particular case, however, there was no nutrient shortage. Hormogonia can glide on surfaces, including other hormogonia and cyanobacterial trichomes, an important feature in colony formation (see below). Hormogonia of Calothrix in streams, ponds and rice fields often have gas-vacuoles, even though these are absent in the mature population; it is unclear whether the presence of gas vacuoles influences gliding. Even without gas vacuoles, hormogonia of pond populations can persist in the plankton for some time and it needs careful observation to separate them from Oscillatoria. The transient nature of the gas vacuoles was confirmed in the study of two strains by Campbell et al. (1993): 8 h after release of hormogonia, gas vesicle genes were expressed and phycobiliprotein genes repressed; by 24 h the converse was true. In addition, hormogonia can lose their ability to form gas vacuoles during prolonged subculture in the laboratory, as was shown for two Calothrix strains isolated from rice fields in Nepal (Vaidya 1989). The addition of phosphate to P-limited cultures led to hormogonia formation in all the Calothrix and Rivularia strains tested (Sinclair and Whitton 1977a; Whitton 1987); addition B.A. Whitton and P. Mateo of the missing element to cultures limited by other elements also led to hormogonia formation, though the response usually took longer than with P limitation. However, growth of Calothrix parietina D184 to Fe-limitation, followed by the addition of Fe, led to the release of gas-vacuolate hormogonia, though often with hair fragments still attached (Douglas et al. 1986). The study by Campbell et al. (1993) on Calothrix PCC 7601 and 7504 showed that transfer to fresh medium and incubation under red light led to hormogonia formation, but not green light. Use of inhibitors demonstrated that the opposing effects arose through differential excitation of photosystems I and II. Although the authors showed the importance of electron transport in regulating cell differentiation, assessing the ecological relevance of their results is more difficult. There is no mention of phosphate, but it seems likely that the cultures had been subcultured routinely under P-rich conditions and never encountered the alternation of P-limited and P-rich conditions characteristic of their natural environment. (Based on its name in the UTEX Culture Collection and published accounts, PCC 7601 is probably not Calothrix, but Microchaete, though it has also been called Fremyella and Tolypothrix elsewhere! PCC 7504 was reported to be quite similar.) Calothrix strains in culture subjected to P limitation and then left for long periods eventually start to lyse, though often not for many months. However, in at least two strains studied in Durham, short lengths in otherwise partially lysed cultures differentiated into healthy hormogonia (B.A.W., unpublished). Assuming the same behaviour occurs in some strains in nature, this would provide a means for a dying population to colonize new sites. 22.2.2.4 Akinete The akinete develops next to the terminal heterocyst in most cases, but intercalary akinetes are frequent in some species. The akinete may develop from a single cell (Desikachary 1959) or a number of cells with breakdown of the cross-walls (e.g. Gloeotrichia ghosei: Claassen 1973). Akinete formation is a characteristic feature of Gloeotrichia, but also occurs in several species of Calothrix and in Dichothrix gelatinosa (Böcher 1946). Two species of Calothrix have been reported to form akinetes. The original illustration of C. stagnalis by Gomont (1895) shows individual filaments epiphytic on Cladophora, so it is clearly not just a loosely organized Gloeotrichia colony. The description of C. santapaui (Gonsalves and Kamat 1960) is less convincing, because the filaments were in a mucilaginous mass with other algae. In addition, the spreading layers of the sheath are more like those of some Dichothrix. In G. ghosei, the only akineteforming strain of the 34 Rivulariaceae studied by Sinclair and Whitton (1977a), increased akinete formation occurred under P limitation. 22 Rivulariaceae Fig. 22.5 Growth of Calothrix parietina D550 in culture, showing arrangement of trichomes following an initial period of hormogonial aggregation and subsequent commencement to form multicellular hairs when P-limited (Photo B.A.W.) 22.2.2.5 Colony, Trichomes and Sheath Rivularia, Isactis, Gloeotrichia and most species of Dichothrix all form colonies, while many species of Calothrix also form a macroscopic thallus with a characteristic appearance. For instance, the hormogonia of C. pulvinata attach themselves to parent filaments, leading to a tufted, bushy structure. In many other strains of Calothrix grown in liquid culture the released hormogonia aggregate to form tight clumps or, more usually, ropes (Fig. 22.5); observations on C. parietina D550 isolated from a stream showed that the most recently released hormogonia moved to the end of the rope (Livingstone and Whitton 1983). As growth proceeds, the aggregated hormogonia eventually separate slightly (B.A.W., unpublished). Rivularia sp. IAM-M-261 showed a positive phototropic response to white light (Katayama et al. 2007), but it is not known whether this is widespread in the Rivulariaceae. De Bary (1863) was the first to report that aggregation of motile hormogonia into small non-motile groups was the initial sign of colony formation in Rivularia. Similar behaviour occurs in Gloeotrichia echinulata (Maxwell 1974) and Gloeotrichia aff. pisum (Aziz and Whitton, unpublished). After re-arrangement into a spherical shape, the trichomes start to differentiate (Fig. 22.6). Clonal cultures of colonyforming Rivulariaceae can sometimes form colonies, while others apparently do not (e.g. Chang 1979b), though in at least some cases this is merely due to excessively high phosphate concentration in the medium, as shown by Berrendero et al. (2008). However, other factors are probably involved in at least some cases. For instance, in a study of the behaviour of two marine Rivulariaceae strains transferred to laboratory culture, Darley (1967) found that aggregation of hormogonia only occurred if a large quantity was inoculated simultaneously. 567 Fig. 22.6 Very early stage in the formation of an epiphytic Gloeotrichia colony (perhaps G. pisum). The orange-brown colour of the heterocysts probably indicates a high carotenoid content (Photo C. F. Carter, with permission) Colony formation by a strain of Rivularia biasolettiana on agar involves an initial period of aggregation, followed by a period when no more trichomes are incorporated into the clump (B.A.W., unpublished). The trichomes then reorientate into a hemispherical or subspherical arrangement and heterocyst development in the basal cell of each trichome becomes visually obvious. If determination of the cell to become a heterocyst has already occurred before the hormogonium is released and occurs in the cell nearest the mother trichome, as observed for Calothrix D.256 (Sect. 22.2.2.2), rearrangement to form a colony would require this end of a moving trichome to be in front, the opposite of what happens when hormogonia are first released. Initially each trichome develops a sheath, which is attached to the lowermost vegetative cell; it never surrounds the heterocyst. Differences in colony structure, which vary markedly between species of Dichothrix, result from differences in the way hormogonia become aggregated with the original after release from the mother trichome . There are also marked differences in sheath structure, which are especially obvious in Dichothrix (Fig. 22.7). The sheath stays close to the trichome in D. orsiniana, with only the colourless parts of the hair projecting from the end; its sheath varies considerably in the extent to which it is laminated. The sheaths of another widespread species, D. gypsophila, are not only strongly laminated, but the laminations splay open at the ends. Sheath thickness increased in all 34 Rivulariaceae strains when grown to P limitation and in 13 strains became dark brown due to scytonemin (Sinclair and Whitton 1977a). Other element deficiencies also led to thicker sheaths and in some cases scytonemin formation. Comparison of two Calothrix populations differing markedly in their scytonemin content 568 B.A. Whitton and P. Mateo Fig. 22.7 Dichothrix (probably D. gypsophila) filaments in a mixed community developing into a stromatolitic crust on a boulder in Lough Corrib, Ireland; sample treated with dilute HCl. Other phototrophs include Schizothrix fasciculata and several pennate diatoms (Photo Bryan Kennedy, with permission). More views of this lake are shown in the attached online article by Kennedy, O’Grady and Whitton from two Yellowstone thermal spring outflows indicated that the difference was partly genetic and partly environmental (Dillon and Castenholz 2003, Dillon et al. 2003). However, the sheath composition (with about 50% neutral sugars and also including 5% amino acids, small amounts of glucosamine and galacturonic acid) of Calothrix parietina D550 and C. scopulorum D256 in culture was not influenced by changes in the P or Fe content of the medium (Weckesser et al. 1988). Three different morphological types of Rivulariaceae trichome were isolated from a single colony of Rivularia biasolettiana from a highly calcareous pond (Sinclair and Whitton 1977a). Molecular studies by Berrendero et al. (2008) also indicate that Rivularia colonies may include more than one genotype (see below). 22.3 Molecular Perspective on Taxonomy 22.3.1 Status of Classical Genera A considerable number of molecular studies have been reported to help understanding of the phylogenetic relationships within the cyanobacteria in general or in particular groups. Most have been based on 16S rRNA gene sequences, with the interpretation on results relying on the fact that no horizontal transfer has so far been detected for this gene (Thomazeau et al. 2010). A variety of terms have been introduced to help interpretation of molecular results, such as form-genus in the 2001 edition of Bergey’s Manual. As mentioned above, interpretation of such studies relies on the name given to a strain being that most suitable for the material when first removed from nature. It also relies on the trichome or clonal culture originating from that trichome being representative of the material isolated from nature. In the Rivulariaceae in particular there is a risk that either of these is wrong. Reliable identification of a P-rich culture based solely on morphological features is almost impossible and there is no way of knowing how representive a single trichome taken from a colony is of the whole colony. Two genome sequences have been completed: Calothrix desertica PCC 7102 and Calothrix PCC 7103 (DOE Joint Genome Project website, May 2011). Both strains have been in culture collections for very long periods – over 60 years for C. desertica. Both belong to the groups able to form hairs (Sinclair and Whitton 1977a). Calothrix PCC 7103 was originally maintained as Gloeotrichia sp. and has also been called Nodularia sphaerocarpa, presumably when cultured in high nutrient medium. Neither appears to have been used for ecologically orientated studies. Rippka et al. (2001a, b) considered the taxonomic diversity of Calothrix and Rivularia. Clusters within these genera were recognized by DNA-DNA hybridization studies using the methodology of Lachance (1981), supplemented in some cases by differences in pigment composition. Three clusters were distinguished within the 16 strains originally maintained as Calothrix (Rippka et al. 2001a). Cluster 1 (eight strains) showed a relatively high degree of genetic relatedness, although it could be divided into five subclusters. All but one strain (C. marchica PCC 7714 = D184) can form hairs. Cluster 2 (renamed from the cluster 3 of Rippka et al. 1979), which had only one strain, had been isolated from a sphagnum bog and was the only strain to form akinetes, when cultured in the absence of combined N. It conformed best, though not perfectly, to C. stagnalis Gomont. Subsequent phylogenetic analysis (Wilmotte and Herdman 2001) has shown that this strain (PCC 7507) fits in a different clade, where it shares a branch with Cylindrospermum stagnale 7417. The other cluster (seven strains), which were all marine isolates, was transferred to the form-genus Rivularia based on the fact that one of the strains had previously been identified by R. A. Lewin as Rivularia sp. Essentially this means that Rippka et al. (2001b) were treating freshwater tapered organisms with and without hairs as Calothrix and marine ones as Rivularia, irrespective of any colony structure. Because of uncertainty about strain names in the Pasteur Culture Collection, it is unclear whether any freshwater Rivularia (or Dichothrix) had been included in the study. The study of Sihvonen et al. (2007), which was based on 16S rRNA sequencing, compared 42 strains of Calothrix, Gloeotrichia and Tolypothrix. This study had the advantage over Rippka et al. (2001a, b) in that 34 of the strains had been collected from brackish (mostly c. 50/00) and freshwater sites around the Baltic Sea and only eight were taken from the Pasteur Culture Collection. The three genera showed a 22 Rivulariaceae high level of genetic diversity and, in the case of organisms identified morphologically as Calothrix, the sequence differences were sufficiently large to suggest at least five different genera. There was no correlation between the phylogenetic clusters and site or habitat from which the strain had been isolated, nor did the nine strains able to produce hairs group together. Several strains clustered with 16S rRNA genes from other genera e.g. Calothrix brevissima IAM-M249 with Tolypothrix strains. The strongly tapering Calothrix strain BECID 18 did not cluster with any other strain and had the closest similarity (94%) with Cyanospira rippkae PCC 9501. A cluster comprising Gloeotrichia echinulata strains shared 98% sequence similarity, but were distant from the strains of the other genera. The authors also reported that a number of Calothrix strains from the Baltic Sea differentiated solitary akinetes or chains of akinetes. Although it seems clear that several organisms considered as Calothrix can form akinetes (Sect. 22.2), the comment by Sihvonen et al. suggests that their study may have included strains corresponding to Gloeotrichia based on classical taxonomic criteria. Calothrix is also mentioned among various cyanobacteria in several 16S rRNA sequence studies dealing with a particular type of environment. Taton et al. (2006) isolated 59 cyanobacteria from 23 Antarctic lakes in order to characterize morphological and genotypic diversity (though sequence studies were only made on 56 strains). Based on morphology, 12 species were recognized, 4 of which are Antarctic endemics. Based on gene sequence, 21 OTUs (operational taxonomic units) were recognized, using the criterion of sequences having more than 97.5% similarity. These included 9 novel and 3 exclusively Antarctic OTUs. Calothrix was represented by one morphospecies and two molecular data analyses. A project (Cuzman et al. 2010) on the biodiversity of phototrophic biofilms on fountains associated with monuments in Italy and Spain led to a comparison of 16S rRNA for 31 isolates with sequences of strains from other sources in the GenBank database. A neighbour-joining phylogenetic tree showed all seven Calothrix and Rivularia strains grouping more closely together than with any other genera. These included two Calothrix and one Rivularia from their own isolates. Hongmei et al. (2005) reported a phylogenetic analysis of communities of mats of mostly filamentous cyanobacteria in the 42–53°C range from China, The Philippines and Thailand. Separation of 16S rRNA gene-defined genotypes from community DNA was achieved by DGGE, leading to the recovery of 36 sequences. Phylogenetic analyses indicated these formed novel thermophilic lineages distinct from their mesophilic counterparts in the case of seven genera, including the only Calothrix sampled. Srivastava et al. (2009) also used comparisons of DGGE bands from samples originating in (Uttar Pradesh) rice fields with GenBank database to characterize the cyanobacterial flora of low and high salinity sites. They included a list of the taxa observed 569 by light microscopy and both this and the DGGE comparison indicated one Rivularia and one Gloeotrichia. Phylogenetic analysis by Berrendero et al. (2008) used sequences from both 16S rRNA genes and an intergenic spacer (cpcBA-IGS) (see Sect. 22.3.2). A neighbour-joining tree based on 16S rRNA genes showed 35 Rivulariaceae grouped together, including all those isolated in their study. Data for two other strains listed as Calothrix in GenBank grouped with other genera. A further analysis based on 16S rRNA gene sequences (Berrendero et al. 2011) confirmed the diversity of Calothrix, but also their clear separation from eight Tolypothrix isolates and material of T. penicillata, all from rivers in Spain. The two groups could also be separated based on their morphology when cultured in a medium with 0.2 mg L−1 P, but not in standard BG-110 medium with 5 mg L−1 P. Domínguez-Escobar et al. (2011) made a comparison based on not only almost the complete 16S rRNA gene, but also intergenic transcribed spacers and part of the 23S rRNA gene. Strains of Calothrix, Rivularia and Tolypothrix were isolated from diverse geographical regions and habitats. Based on results for their newly isolated strains and data for Gloeotrichia from Sihvonen et al. (2007), the authors made molecular clock estimates on the origin of the various genera. The methodology for doing this was based on a number of previous studies (e.g. Sanderson 2002) and involves approximations and estimates, such as the node defining cyanobacteria being fixed at ~2,700 Ma (million years ago) and a minimum age for the heterocystous cyanobacteria at ~1,618 Ma (Falcón et al. 2010). The latter value in particular differs considerably from the ~2,400 Ma for Nostocales suggested in Chap. 2 (Sect. 2.4.1.2), so the values of ~1,500 Ma for Calothrix and Rivularia, and 400–300 Ma for Gloeotrichia and Tolypothrix should be treated with caution. Nevertheless they provide a stimulus for further critical study. 22.3.2 Molecular Diversity Within Form-Genera While studies described above include information on the form-genera Calothrix and Rivularia, several deal specifically with them. Shalini et al. (2008) using RAPD – PCR to assess the phylogenetic relatedness of 31 Calothrix strains from India and a reference strain (UTEX 379). A combination of 12 sets of primers generated 903 distinct polymorphic DNA fragments, indicating a wide range of variation among the strains. The highest correlation coefficient (0.821) was found for two strains which came from the same geographical location. The authors discussed the possible importance of geographical proximity in relatedness between strains, though it seems just as likely that environmental similarity is a key factor in this genus. 570 Berrendero et al. (2008) focussed on Rivularia, using colonies from five different rivers in Spain and one stream in the UK, all calcareous. They also used 13 strains isolated from calcareous rivers and one from a stream flowing over siliceous substrates. All were listed as ‘strongly tapering (Rivularia or Calothrix)’, but only Calothrix – no Rivularia – occurred in the siliceous stream. The methods included analysis of the phycocyanin operon (PC) and intervening intergenic spacer (cpcBA-IGS) and 16S rRNA gene sequences, together with molecular fingerprinting. The cpcBA-IGS analysis showed that all the sequences of environmental Rivularia colonies and rivulariacean-type isolates from calcareous rivers fell into one of three distinct genotype groups. Alignment of the sequences revealed 100% similarity of some isolated strains with the sequences in some field Rivularia colonies. Genotype I was found in Rivularia colonies from the Endrinales and Matarraña rivers and strains isolated from these rivers, together with two other strains. Genotype II was found in isolates from the Endrinales, Muga and the siliceous stream. Genotype III was found in Rivularia colonies from the Alharabe, Muga and Red Sike (UK) and one isolate from another river. All three genotypes were reported for the Muga, though this may simply reflect the fact that, in addition to the colonies, six isolates came from this river. Culture of Genotype I isolates in a medium with relatively low phosphate concentration (0.2 mg L−1 P) led to Rivularia-like morphological characteristics, including secondary trichomes remaining in the mother sheath, lamellated sheath, confluent trichomes and a tendency to form spherical colonies. However, when Genotype II isolates were cultured in the same medium, they maintained Calothrix-like morphological characteristics and only formed irregular clumps or bundles of trichomes. These results suggest that Genotype I isolated strains belong to traditional Rivularia and Genotype II strains to traditional Calothrix. Berrendero et al. cautioned that, because some cyanobacteria possess two or three PC operons (Golden 1995), their cpcBA-IGS analysis does not rule out the possibility of more than one PC being present, but only one sequenced. They therefore investigated this further using TGGE (temperature-gradient gel electrophoresis), which allows sequence-dependent separation of PCR products. The high sequence variability among the three genotypes, which gave rise to three clearly separated bands, allowed each genotype to be characterized by its corresponding fingerprint. There was only one band for each culture and this was consistent with the corresponding fingerprint. Colonies from the Alharabe and Red Sike, which best fitted Rivularia biasolettiana, corresponded to Genotype III. Colonies from the Blanco, Endrinales and Matarraña rivers all had bands corresponding to more than one genotype. Those from the Endrinales, which were typical R. haematites, had not only a band corresponding to Genotype I, but also one to Genotype II. B.A. Whitton and P. Mateo The Blanco and Muga also had colonies with Genotype II, suggesting the presence of Calothrix-like filaments in the Rivularia colonies. Some colonies from the Blanco showed bands corresponding to both Genotypes I and III, suggesting that a single colony may include filaments of two different types (“species”) of Rivularia. The phylogenetic sequencing analysis, which formed part of the study by Berrendero et al., showed that the 31 strains in the Rivulariaceae group (see Sect. 22.3.1) fell into three distinct genotype clusters. The Genotypes I, II and III from studies on their own colonies and isolates each fitted into one of the three broader genotype clusters. Apparently no similar studies have included Dichothrix, but the results would be of considerable interest, because some species are morphologically little more than complex forms of Calothrix, while others are hard to distinguish from Rivularia. A detailed analysis of Gloeotrichia would be even more interesting, because the planktonic G. echinulata differs considerably from other species in the genus, at least some of which seem little different from a Rivularia able to form akinetes (Sect. 22.5). 22.4 Physiology 22.4.1 Carbon and Nitrogen In laboratory culture most Rivulariaceae appear to grow more slowly than filamentous species with untapered trichomes, though a Gloeotrichia aff. pisum culture from a Bangladesh deepwater rice field had a doubling time of 12.6 h under optimum growth conditions (Aziz 1985). There are a number of reports of photoheterotrophic growth and other physiological responses to sugars. When Calothrix elenkinii was grown on a light-dark cycle, glucose enhanced pigment production, especially b-carotene and chlorophyll (Prasanna et al. 2004a, b). Lebedeva et al. (2005) found that glucose enhanced phycocyanin production of Calothrix PCC 7601 under red light, but not phycoerythrin under green light. Dark heterotrophic growth with sucrose was reported for the two Calothrix strains included in the study by Khoja and Whitton (1971); glucose did not support heterotrophic growth of Gloeotrichia echinulata (Chang 1979a). All (heterocystous) Rivulariaceae tested have proved capable of N2 fixation and, in batch culture under continuous light, the highest rates have been reported several days after subculture to fresh medium. The rates are usually much higher in the light than the dark, whether measured in the laboratory (Stewart 1967; Chang and Blauw 1980; Al-Mousawi 1984) or the field. The rate dropped rapidly when intertidal Rivularia atra was transferred from light to dark (Khoja et al. 1984), a useful feature for this population which is sometimes smothered in sand for part of the tidal 22 Rivulariaceae cycle. This contrasts with the only slight initial change in rate shown by Calothrix crustacea mat on sand and Rivularia mat on silt at the edge of the lagoon of Aldabra Atoll (Potts and Whitton 1977). Diel studies by Hübel and Hübel (1974) of C. scopulorum and Rivularia on the Baltic Sea coast found that rates were usually 6–10 times higher at midday than in the night, though the difference was less marked during July and August, presumably due to the greater daytime accumulation of carbohydrate. The ratios between midday and midnight N2 fixation in R. biasolettiana in Red Sike, Upper Teesdale, northern England, on two dates in August were similar (8–16% at night) (Livingstone et al. 1984). Overall about 400 mol CO2 to 1 mol N2 was fixed during daylight and the authors suggested that N2 fixation may supply only a small percentage of its N requirement. However, another possibility is that N2 fixation rates are higher in spring when ambient, mostly organic, P can be almost three orders of magnitude higher in this stream than in August (Livingstone and Whitton 1984). (Use of the term ‘organic’ for P fractions is explained by Whitton and Neal 2010). If so, then much of the C fixed in summer may be used in sheath formation. The upper intertidal organisms Calothrix scopulorum (Jones and Stewart 1969) and Rivularia atra (Reed and Stewart 1983) show greater inhibition under high salinity of nitrogenase than photosynthesis, and it was suggested that this may reflect channelling of fixed C into an osmoticum. A comparison of nitrogenase activity of Gloeotrichia pisum colonies from the aquatic roots of deepwater rice plants in Bangladesh with a laboratory isolate (Gloeotrichia D613) showed a rapid fall in activity in both cases with the onset of dark conditions (Aziz and Whitton 1988). Nitrogenase activity was much higher when material grown in the dark for 12 h was transferred to the light than the highest rate found under continuous illumination. Longer periods of dark pretreatment led to only moderate increases in the period before peak nitrogenase activity was reached again. On transfer from light to dark, nitrogenase activity fell rapidly to very low values. Comparison in a Bangladesh deepwater rice field of the% total nitrogenase activity at night for Gloeotrichia cf pisum and G. natans (each on two different days) compared with that of four other cyanobacteria showed that Gloeotrichia always had much lower % values (Rother et al. 1988). Nitrogenase activity of Calothrix D764, another deepwater rice-field isolate, showed similar responses (Islam and Whitton 1992b). Rapid and marked responses to changes in light are important during the monsoon season and may be a feature of deepwater rice-field cyanobacteria, as many were found to show peak activity 6 h after dawn (Rother et al. 1988). This response of nitrogenase activity on transfer to light parallels the high rates of phosphate accumulation when P-deficient cyanobacteria are presented with phosphate (Healey 1982). The early literature on N2 fixation rates by 571 Rivulariaceae in the field, mainly marine sites, was summarized by Whitton (1987). Marked changes in N composition were found during batch culture of Calothrix D764 (Islam and Whitton 1992b). The concentrations of chlorophyll, phycocyanin and phycoerythrin had all reached their highest values by the stage when the culture had reached 35% of its final yield. Soon afterwards there was a sharp decrease in nitrogenase activity, which subsequently persisted at the same rate for several weeks. 51% of the N present in pigments at the time they reached their peak values was transferred to other substances in old cultures, presumably cyanophycin and molecules associated with maximizing uptake of nutrients in limited supply. There are a number of reports of biochemical features which appear to characterize the Rivulariaceae. For instance, in a study of heterocyst glycolipids, those of four Calothrix strains were dominated by a C28 rather than by the C26 carbon chain of the 17 Nostocaceae studied (Bauersachs et al. 2009). The functional and possible ecological significance of such molecular differences awaits study. 22.4.2 Phosphorus The inhibitory effect of even moderate concentrations of inorganic phosphate noted by Fogg (1969) for Gloeotrichia echinulata seems to be widespread in Rivulariaceae; however, all the reported studies are for hair-forming species, so it is unclear whether this also applies to Calothrix strains incapable of forming hairs. The formation of hairs under P limitation (Sect. 22.2.2.3) was investigated further in Calothrix parietina D550 (Livingstone and Whitton 1983). Hairs are formed as the average value for cellular P falls to about 1% dry weight, when the filaments still appear healthy, indicating that the change occurs when the organism is only moderately P-limited. If phosphate is added to a culture with hairs, polyphosphate granules are formed rapidly in cells towards the base of the trichome and a few hours later several hormogonia start to differentiate at the apical end of the chlorophyllcontaining part of the trichome. Light is not essential for polyphosphate granule formation in Calothrix D550, though it is for hormogonia formation. Surface phosphomonoesterase (PMEase) and phosphodiesterase (PDEase) activities of C. parietina D550 during batch culture commence at about the same stage of P limitation as hair formation (Livingstone and Whitton 1983); release of extracellular PMEase commences at the same time (Grainger et al. 1989), though in some other strains it commences later. No soluble extracellular PDEase is formed in C. parietina D550, nor in most other cyanobacteria and eukaryotic algae (Whitton et al. 2005). Several staining techniques have been used to compare differences in distribution of PMEase activity between 572 B.A. Whitton and P. Mateo Fig. 22.8 (a–c) Examples of location of phosphomonoesterase (PMEase) activity following incubation in para-nitrophenyl phosphate (2 mg L−1 P) for 30 min with subsequent staining in 1% lead nitrate to deposit lead phosphate: (a) PMEase activity localized mainly on outer wall layer of the hair cells of Calothrix parietina D184 (= PCC 7713); (b) PMEase activity in sheath of C. parietina D184; (c) PMEase activity mostly inside the cytoplasmic membrane of hair cells of Rivularia biasolettiana from a stream in Middleton-in-Teesdale, UK; (d) Microtubule crossing middle of the cross-wall of a vegetative cell of Calothrix parietina D184: mi, microtubule (Photos A. Peat, P. Wood and B.A.W.) species, growth stage and to show the influence of the environment. Light microscopy using BCIP (bromo-4chloro-3-indolyl phosphate) and a simple phosphomonoester as substrate to generate a blue stain can provide much useful ecological information, but electron microscopy with precipitation of lead phosphate is needed to show the exact location of PMEase (Wood et al. 1986). PMEase activity usually occurs on the outer layer of the cell wall (Fig. 22.8a); it is often especially strong on the cross-walls. In C. parietina strains it develops on all cell walls (apart from the heterocyst) and also in the sheath (Fig. 22.8b). It is more restricted in some other Rivulariaceae, developing mainly on the hair cells and chlorophyll-containing cells near the hair and also the part of the sheath furthest from the heterocyst. However, the extracellular matrix of Rivularia colonies often shows marked PMEase activity. It is uncertain whether PDEase activity is also present, but in view of the absence of soluble extracellular PDEase, this seems unlikely. Electron micrographs of old cultures of Calothrix and field Rivularia indicate that some activity also occurs inside the cytoplasmic membrane of old hair cells (Fig. 22.8c), suggesting that organic P molecules sometimes pass through the cytoplasmic membrane of the hair cell. Hairs of Calothrix D550 formed in response to Fe limitation show no phosphatase activity (Douglas et al. 1986). In C. viguieri D253 (= PCC 7709) phosphatase activity is largely restricted to the hairs (Mahasneh et al. 1990). This strain, which had been isolated from a mangrove root, was able to grow in media of varying salinities ranging from freshwater to 20% seawater. However, when the culture was P-limited, hair formation only occurred in freshwater medium. During growth in saline medium the PMEase activity was 22 Rivulariaceae much lower and PDEase activity absent, while yields were less for most of the organic P substrates tested as the P source for growth. It would be interesting to establish whether such differences reflect adaptations to variations in the P status of water at its original location, such as might occur over a 14-day tidal cycle. Once phosphate ions are released in the hair, rapid transfer must occur to the base near of the trichome where polyphosphate granule formation starts. Electron-microscopy of several strains showed conspicuous granule formation here within 1 min, but much later further along the trichome (Wood 1984). Studies are needed to establish the exact mechanism by which phosphate moves from the hair to the base of the trichome, but presumably a diffusion gradient is important. Calothrix D184 has a conspicuous microtubule passing through the middle of the cross-walls of vegetative cells (Wood 1984). This extended for about 1 mm each side of each cross-wall, but apparently not the whole length of the cell (Fig. 22.8d). Several studies have permitted comparisons to be made between Rivulariaceae versus other cyanobacteria. In one of these (Whitton et al. 1991) this was based on the ability of 50 cyanobacterial strains (mostly tropical and subtropical) to grow in batch culture with various P sources (1 mg L−1 P) for 16 days; the study involved 30 Rivulariaceae and 20 non-Rivulariaceae (13 filamentous and 7 Synechococcus) and between hair-forming and non-hair-forming Rivulariaceae. Several differences were significant (p < 0.05). In general Nostoc (six strains) produced the highest yields with inorganic P, whereas Rivulariaceae produced higher yields than filamentous non-Rivulariaceae with b-glycerophosphate, pNPP and DNA. All Rivulariaceae grew well with a phosphomonoester (pNPP, para-nitrophenyl phosphate) and a diester (bis-pNPP, bis-para-nitrophenyl phosphate) and all but one (Calothrix D764 from a deepwater rice field) grew with DNA. Many also used phytic acid (myo-inositol hexakisphosphate), but seven did not (including Calothrix D764). Rivulariaceae forming hairs were more effective than those not forming hairs at utilizing phytic acid. It seems likely that P is sometimes released by nuclease as well as phosphatase activity, since 10 strains, all Rivulariaceae, produced a yield with DNA at least 1.5 times that with pNPP. This study also indicated that Rivulariaceae are especially sensitive to ATP. Eight Rivulariaceae (including six hairformers) failed to show any growth with ATP (1 mg L−1 P) and in some cases this was toxic. All the other filamentous forms grew well with ATP. A possible explanation is that ATP enters the hair cells rather than being hydrolyzed at the surface and this leads to internal concentrations which are inhibitory or toxic. The Rivulariaceae tended to show higher rates of surface and extracellular PMEase and surface PDEase activity and the highest rates all occurred in strains of Gloeotrichia; none of 573 the strains in the study formed extracellular PDEase. Within the Rivulariaceae, strains from calcareous environments show higher PMEase activity than strains from non-calcareous environments at pH 10.3 (p < 0.01), though not at pH 7.6. All strains in this study were assayed using 250 mM substrate, but a further study with 16 phototrophs showed that substrate concentration sometimes, but not always, influences the response to pH (Whitton et al. 2005; Whitton and Donaldson, unpublished data). The pH optimum for three Calothrix strains was 1.5–2 pH units lower using 1 mM substrate (methylumbelliferyl phosphate) compared with 250 mM; the lower pH value was close to the typical field pH All three strains were ones with hairs, as was the only eukaryote to show a similar response (Stigeoclonium D565). The strains not showing a response included all the Rivulariaceae not forming hairs. Perhaps the species showing two pH optima have more than one mechanism for mobilizing organic phosphate. Phytase activity has been reported for many Rivulariaceae, such as a Calothrix parietina strain from Upper Teesdale, UK, (Livingstone et al. 1983) and 35 of the 42 strains tested by Whitton et al. (1991: see above), but the influence of pH has yet to be studied. Further insight into differences between Rivulariaceae and non-Rivulariaceae was provided by comparisons between two strains isolated from Rio Alberche, Spain: Calothrix elenkinii (not hair-former) and Nostoc punctiforme (Mateo et al. 2006). When inorganic phosphate was supplied to a P-limited culture, P uptake kinetics was closely similar for both strains. For instance, the half-saturation constant (Km) was 2.98 mM for C. elenkinii and 2.99 mM for Nostoc punctiforme. This indicates that a higher affinity for a low inorganic phosphate concentration was not a factor likely to influence the relative success of C. elenkinii. The authors suggested that this was due to its greater ability to store P as polyphosphate granules and its higher surface PMEase activity. However, an in situ survey (Mateo et al. 2010) of phosphatase activities of Rivularia biasolettiana and three other cyanobacteria (Schizothrix coriacea, Tolypothrix distorta var. penicillata and Nostoc verrucosum) in the River Muga, found that Tolypothrix had the highest PMEase and PDEase activities (related to chlorophyll), Schizothrix the next highest and Rivularia the least on the four occasions measured. Surface phosphatase activities of all the populations obeyed apparent Michaelis-Mention kinetics, showing similar values for Km, but markedly different ones for Vmax, with Rivularia having the lowest values for both PMEase and PDEase activities. In March, when Rivularia not only showed the least phosphatase activities of any season, but also the greatest contrast with Schizothrix and Tolypothrix (Nostoc was absent), its trichomes had prominent polyphosphate granules. It was suggested that the contrasting results may be due to Schizothrix and Tolypothrix growing faster than Rivularia and already responding to the low ambient P found at the time. 574 Stream surveys provide records of Calothrix populations at zinc concentrations just below 10 mg L−1 (Shehata and Whitton 1981; Whitton et al. 1981), a value well above that for other heterocystous cyanobacteria. It is suggested that this is another example of an environment where an ability to use organic P effectively is important, because inorganic P is relatively insoluble under these conditions, especially as these streams also had elevated lead. Pb++ removal from wastewater by Calothrix marchica was higher in old than young material (Ruangsomboon et al. 2006), presumably because of increased sheath production. 22.4.3 Molecules with Possible Ecological Effects A range of organic molecules with biological effects on cells of other organisms have been isolated from cyanobacteria (Smith and Doan 1999; Skulberg 2000; Van Wagoner et al. 2007), including some Rivulariaceae, such as the carbolines from Dichothrix baueriana (Larsen et al. 1994). In some cases the circumstantial evidence suggests that these may be involved in protection from grazing (see Sect. 22.6.3.3). In other cases, even when there have been no ecological studies, features of the molecule, including its effects in bioassays, suggest that it be well worth doing so. Schlegel et al. (1998) found that, of 198 cyanobacterial strains, 7 (out of 13) Fischerella, 5 (out of 28) Nostoc and 3 (out of 9) Calothrix had bioactivity against green algae; the Calothrix affected all three species tested. As the cyanobacteria were cultured in a very P-rich medium, it would be worth repeating the survey with old, P-limited cultures, especially for Calothrix. An extract of a marine Rivularia sp. showed marked antibacterial activity against gram negative bacteria in comparison with ampicillin and streptomycin (Zarmouh 2010). Studies initiated by the research group of G. D. Smith at Canberra were especially promising, so it is worth summarizing how this research proceeded. This project started when Rickards et al. (1999) showed that cell extracts of two Calothrix isolates inhibited the growth in vitro of a chloroquine-resistant strain of the malaria parasite, Plasmodium falciparum, and of human HeLa cancer cells, both in a dose-dependent manner. Two pentacyclic metabolites (calothrixins A and B), with an indolo[3,2-j] phenanthridine ring system unique amongst natural products, were isolated and shown to have growth-inhibitory effects at nanomolar concentrations (Doan et al. 2000, 2001). Calothrixin A was one of two cyanobacterial molecules found to inhibit Escherichia coli RNA polymerase competitively with respect to ATP, and non-competitively with respect to UTP. Based on comparisons with the sensitivity of whole cells to these inhibitors, the authors concluded that other targets in addition to RNA polymerase may also be B.A. Whitton and P. Mateo implicated in their action. The antiproliferative effect of calothrixins on several human cancer cell lines may be related to their ring structure (Chen et al. 2003) and their ability to undergo redox cycling (Bernardo et al. 2004); further insight to the mechanism was obtained by synthesizing related quinones (Bernardo et al. 2007). Among various effects on human cell metabolism, Khan et al. (2009) found that calothrixins act as poisons of DNA topoisomerase I and do so reversibly. The two molecules have now been synthesized (McErlean et al. 2007; Abe et al. 2011). Among other molecules reported from Rivulariaceae, but few, if any, other cyanobacteria, are octadecatetraenoic acid (Kenyon et al. 1972), brominated phenolic substances from an axenic freshwater Calothrix and field material of marine Rivularia (Pedersen and DaSilva 1973) and six polybrominated biindole derivatives from R. firma (Norton and Wells 1982), the last being named rivularins by Maehr and Smallheer (1984). Several of these authors discussed possible biological effects, but apparently none has been tested thoroughly. In an account of gene products of cyanobacteria, Ehrenreich et al. (2005) listed a desert sand strain of Calothrix, which released a siderophore into the medium, as having some inhibitory effect on Synechococcus PCC 7002. Molassamide, a depsipeptide serine protease inhibitor was isolated from the marine Dichothrix utahensis (Gunasekera et al. 2010). The possible role of microcystins and other toxins in reducing grazing is considered in Sect. 22.6.3.3. 22.5 Overview of the Genera 22.5.1 Introduction The features of the five genera of Rivulariaceae were introduced in Sect. 22.2.1. The morphological features of these and other morphologically complex cyanobacteria frequently show an obvious relationship to features of their environment (Whitton 2008). It should therefore be possible to comment on the range of environments where Rivulariaceae occur from an understanding of their morphological diversity. In addition, the increasing realization of the molecular diversity of the genera (Sect. 22.3.2) raises the question as to how much it will be possible to recognize by morphological criteria the groups within the form-family Rivulariaceae which have close molecular similarity. If this proves possible, classical taxonomy could be revized to fit better with the molecular data. The presence of heterocysts indicates that N2 fixation is important at least some growth stage and thus there is probably a relative deficiency in combined N at that time. As the extent of tapering increases with increasing P limitation (Sect. 22.4.2), while hormogonia formation occurs when 22 Rivulariaceae phosphate is added to P-limited material, the hypothesis is suggested that the Rivulariaceae as a whole are adapted to environments showing alternating N and P limitation. The extent to which the available data supports this is discussed below. 22.5.2 Calothrix In addition to trichome width, whether or not mature trichomes end in multicellular hairs and whether intercalary heterocysts form under some conditions, there are a number of features which vary within the genus. Some, but not all, have been used as criteria for distinguishing species. Two species have been reported to form akinetes. The original illustration of C. stagnalis by Gomont (1895) shows individual filaments epiphytic on Cladophora, so it is clearly not just a loosely organized Gloeotrichia colony; the trichomes have hairs. The basal part of the trichome is swollen in some species (Sect. 22.2.2.1) and most, if not all, shown in Geitler (1932) are species forming hairs. The data by Sinclair and Whitton (1977a, b) together indicate that species with swollen bases tend to form hairs when P-limited. In the study of how Rivulariaceae strains respond to combined N (nitrate), all strains with swollen bases in N-free medium belonged to the group of strains where many trichomes retained obvious tapering (but no heterocyst) in the presence of N; none belonged to the other group where all the trichomes had parallel sides, much like Lyngbya. However, Rai et al. (1978) showed that the response of a C. brevissima strain differed according to N supply: slight tapering remained with NO3-N, but not with NH4-N. When a culture supplied with NH4-N had depleted this, the trichomes formed intercalary heterocysts, followed by the initiation of polarity once more with trichome breakage adjacent to the heterocyst.. The other characters which have been incorporated into taxonomic descriptions are whether the sheaths are brown (scytonemin), whether the sheaths are layered (e.g. C. parietina) and whether there is a distinctive colony structure (e.g. C. pulvinata). There are thus at least six features in addition to dimensions that could be considered in relation to molecular groupings. Batch culture studies on Calothrix D764, a strain isolated from a hormogonium in the plankton of a Bangladesh deepwater rice field provide insight on how trichome morphology, cell composition, nitrogenase and surface phosphatase activities interact (Islam and Whitton 1992a, b). Although this strain produced long, tapered filaments, which were ultimately very narrow and thus had the shape of a hair, the terminal cells remained photosynthetic and increased only slightly in length. Under P-rich conditions a hormogonium formed in the apical region of the trichome without any cell 575 loss. Chlorophyll a ranged from 0.25% to 2.8% (dry weight), with values >1% only occurring for a short period when cellular P was at its maximum; 0.5% dry weight is probably typical for healthy field material. Chlorophyll a and phycocyanin contents both increased in% values as light intensity was reduced from 85 to 10 mmol photon m−1 s−1. There was a shift of N from cyanophycin to phycocyanin formation with decreasing light intensity. When P was added to a P-limited culture of Calothrix D764 at the highest light value, the maximum cellular P content was reached by 1 h, while the maximum nitrogenase activity by 1 day. After a relatively short period of very high nitrogenase activity, the rate per unit mass fell rapidly, being 150 times higher on day 2 than at 5 weeks. However, each heterocyst had to support about five times as many cells, so the equivalent comparison based on heterocysts between day 2 and 5 weeks was 32 times. Formation of new trichomes had slowed down by the end of the first week, but some still formed for a few more days; the subsequent increase in biomass resulted entirely from the filaments growing longer. Addition of P during this stage led to rapid formation of further hormogonia, each one being initiated at the end of the filament. The P content of 2.9% at day 1 fell rapidly and surface PMEase first became detectable when it had fallen to 0.95%. PDEase was initiated at the same time as PMEase and the two activities ran closely parallel for the first 2 weeks, with values for PDEase being about one-third those of PMEase. The presence of combined N in the medium led to higher surface PMEase and PDEase activities. Light had no effect on PMEase or PDEase activities on this or other Calothrix strains tested over periods up to at least 1 h, nor on two rice-field Rivularia strains (Banerjee and John 2005) or Rivularia colonies from River Muga (Mateo et al. 2010). This species grew almost as rapidly with glucose-6-P as with inorganic P and quite rapidly with phosphomonesters, but not with DNA, in spite of the high PDEase activity and the fact that most other Calothrix strains grow well with DNA as the P source (Whitton et al. 1991). No doubt studies on other strains would all show slight differences in the sequence of events, but the morphological changes in C. parietina D550, isolated from a UK upland stream, were largely the same (Livingstone and Whitton 1983; Sect. 22.4.2), the main differences being that it formed colourless hairs, which involved cell wastage when hormogonium formation was initiated beneath it; in addition, hormogonium formation ceased at an earlier growth stage than in Calothrix D764. The key changes in batch culture of Calothrix seem clear. Nitrogenase activity is at its highest when the P content is highest; surface phosphatase activities are absent at this stage, but then start to increase quite rapidly. Shortly after this, hormogonia formation ceases, and only recommences if further P is added to the culture. Strains probably differ in that differentiation of hormogonia 576 Fig. 22.9 Rivularia bullata colonies on boulder taken from shallow water in the very slightly brackish Loch Àirdh a’ Mhuile, South Uist, Outer Hebrides (Scotland), in September 2009 (Photo B.A.W.) tends to cease earlier in strains forming hairs than in those which do not. As about 100 Calothrix species have been described, it is not surprizing they are recorded from a range of environments. However, quite a number occur in highly variable ones, such as the upper part of the marine intertidal zone. C. parietina often forms part of the microbial community in the zone where the water level fluctuates in about half the mature garden ponds in the UK with water above pH 7.0 (B.A.W., unpublished). This also applies to C. parietina D550, which was isolated from the side of one of the streams described by Livingstone and Whitton (1984), where the ambient P concentration is highly variable and mostly organic. Organic P formed well over half the mean value for filterable P (n = 53) in Bangladesh deepwater rice fields (Whitton et al. 1988a), where Calothrix and Gloeotrichia were frequent, and where Calothrix D764 (described above) was isolated. In the case of the Calothrix population in Hunter’s Hot Spring described by Castenholz (1973; Sect. 3.2.3.2; see also Fig. 3.2d, g), it seems probable that the intense grazing of the Oscillatoria mat immediately upstream releases organic P. 22.5.3 Rivularia, Dichothrix and Isactis Rivularia species in classical floras are separated into marine and freshwater, though records are often confused because of freshwater streams flowing into the intertidal zone. Conversely, waterbodies near the sea, but which are only very slightly brackish, may have marine species, as shown for R. bullata (Fig. 22.9). The characters used for separating species are mostly similar in marine and freshwater forms: trichome width, maximum colony size, soft or firm, whether the sheath is layered, whether the ends of the sheaths are B.A. Whitton and P. Mateo Fig. 22.10 Colonies of Rivularia cf R. biasolettiana and the red alga Chroothece growing in shallow water on the same boulder in a fastflowing stretch of the Rio Chícamo, S-E. Spain. The dark colour of the Rivularia colonies is due to highly pigmented brown sheaths, while the orange-brown of the Chroothece is due to intracellular carotenoids (M. Aboal, personal communication) (Photo B.A.W.) spread out in many layers; freshwater forms are also separated according to whether they show calcification. The taxonomic characters used for Dichothrix are similar to those for Rivularia, except that the extent to which hormogonia are arranged in bundles is important; there is a tendency for the trichomes of marine species to be wider than freshwater species. Isactis is unispecific; it is little more than a large group of vertical Calothrix filaments embedded in mucilage. Freshwater Calothrix and Dichothrix occur in both very soft (Heuff and Horkan 1984) and hard waters. The records for Czech and Slovak Republics listed by Skácelová (2006) confirm this for Dichothrix. Most records for Rivularia are for calcareous waters (Fig. 22.10) and it is often the dominant near the source of highly calcareous streams (see Whitton 1987; Sabater 1989; Sabater et al. 2000). However, in Jämtland, Sweden, and probably elsewhere in northern Scandinavia (Johansson 1979; Johansson, personal communication 1981) the genus is widespread in noncalcareous waters. In the marine environment Dichothrix is frequent on submerged angiosperms (Uku et al. 2007) and floating masses of larger algae. Floating masses of Sargassum and epiphytic Dichothrix fucicola contributed over 0.5% total primary production in the western Sargasso Sea (Carpenter and Cox 1974), while production in October by Dichothrix was about 15% macroalgal production in continental shelf waters, though only 1.2% in the northern Sargasso Sea. The greater abundance of Dichothrix in shelf waters than the Sargasso Sea probably reflects the availability of Fe. Calothrix is also frequent on marine angiosperms, but more especially on ones that are intermittently exposed, such as described by Webber (1967). 22 Rivulariaceae The cycle of morphological changes for Rivularia in relation to ambient P is similar to that for Calothrix, but the periodicity of hormogonia formation and release is much longer. Streams in Upper Teesdale, UK, show an annual cycle of elevated P in spring, followed by far lower concentrations for much of the rest of the year (Turner et al. 2003a, b). The annual cycle of changes was especially clear in 1981–1982 (Livingstone and Whitton 1984); most release of hormogonia by R. biasolettiana occurred in early spring following a period of high organic P concentrations in the water. However, in some years there are further short periods of sufficiently elevated P concentrations to lead to marked hormogonia release and further colony formation (Whitton et al. 1998). In addition to the release of hormogonia, trichomes arising from “false” branches occur in almost all Rivulariaceae; the lower part of the original trichome develops a tapered shape as it grows. However, in some cases true hormogonia are formed which are not released from the colony, but differentiate into typical filaments inside the colony. Although growth of an upper intertidal R. atra population at Tyne Sands, S-E. Scotland, occurred mostly in summer (Yelloly and Whitton 1996), periodicity of colony formation depended on a combination of extreme storm events and tides. Storms deposited large masses of detached, mostly sublittoral, seaweeds high on the supralittoral. These rotted and released high concentrations of nutrients next time there was a high spring tide; much of the filterable P was organic. All the trichomes in large colonies released hormogonia within a couple of days, leading to the formation of new colonies. Dichothrix shows a similar response to periods of elevated ambient P, with hormogonia release followed by colony formation, but the periodicity seems less clear-cut. Dichothrix shows a marked tendency to be more frequent than Rivularia in the west of the British Isles, while the reverse applies in the east (Whitton 2011). It was suggested that this may be favoured by the greater and more irregular precipitation and less marked seasonal difference between summer and winter that occur in the west. These may also explain why the dominant Rivularia species in streams on Clare Island off the west coast of Ireland is R. beccariana (Whitton 2007), which forms only very small colonies, and probably does so for much of the year rather than mainly in spring. Many studies have reported that Calothrix species and sometimes Rivulariaceae in general are associated with low levels of nutrients. For instance, Schneider and Lindstrøm (2011) presented an index based on non-diatomaceous benthic cyanobacteria and algae to characterize river trophic status in Norway based on total ambient P; all the Rivulariaceae showed a low indicator value. Although such indices are often required by water administration organizations, they neglect the importance of temporal variability in flowing waters draining semi-natural ecosystems and may hinder understanding of the processes occurring there. 577 Fig. 22.11 Rivularia cf R. biasolettiana in River Muga, N-E. Spain, showing range of colony size reflecting older colonies and recently formed ones (Photo P.M.) Many “oligotrophic” waters do not contain Rivulariaceae. However, the sites with Rivularia studied by Livingstone and Whitton (1984) and Yelloly and Whitton (1996) do at times have high concentrations of ambient P (in some cases >1 mg L−1 P for short periods), even though the annual mean is far lower. It is unclear how wide a range of P concentrations or N:P ratios is required to provide competitive success for Rivulariaceae and whether an alternation between low and extremely low values would be suitable. The River Muga, Spanish Pyrenees, never showed values for ambient total filterable P > 6.8 mg L−1 P (mostly organic) during measurements at five different times of year (Mateo et al. 2010), yet new colonies can form in spring (Fig. 22.9). The Rivularia population in the highly calcareous and slightly saline Rio Chicamo, S-E. Spain, presents a still unexplained anomaly (M. Aboal, personal communication). Although healthy colonies with typical heterocysts are abundant in some stretches of the river, often together with the red alga Chroothece (Fig. 22.11), the mean concentration of ambient combined N is high, hence no obvious advantage for a N2-fixer. However, the mean P concentration is low and mostly organic, features which are likely to favour Rivularia. The most probable explanation is that there is some season of the year or river condition when combined N is sufficiently low to give Rivularia a selective advantage in competition with Chroothece, but this has yet to be established. Other possible (though unlikely) explanations for formation of the hormogonia is that at some time of year Rivularia switches almost entirely to P accumulation, leading to a high internal concentration in spite of low ambient concentration, or that some other element shifts from scarcity to abundance for a short period and this triggers hormogonia formation. 578 Fig. 22.12 Rivularia haematites, from R. Guadiela, Spain. The layers of calcite obvious in field material have been dissolved with very dilute acid, but a banded structure is still obvious due to differences in sheath density and scytonemin formation (Photo P.M.) The seasonality of growth of Rivularia in many temperate region streams and lake margins suggests this may be an important factor leading to the laminated structure of many colonies, especially those treated as R. haematites (Fig. 22.12); evidence for this in a UK stream population was given by Pentecost (1987). Intermittent exposure to the air and evaporation during the warmer season led to some calcium carbonate deposition in this population. Rivularia EPS may also trap and bind calcite crystals (Pentecost and Riding 1986). The factors influencing calcite formation inside R. haematites colonies were studied at five sites in the Salzkammergut, Austria, by Obenlüneschloss and Schneider (1991). The sites provided a wide range of conditions for light, flow variability and current speed and calcite crystal size showed an obvious correlation with these. For instance, the largest and most regular crystals occurred at sunny sites with continuous flow and low current speed. The use of planar optodes and microelectrodes helped Pentecost and Franke (2010) to study the process of calcification in still more detail. A quantitative assessment of the factors involved at each of various depths within the colonies of two populations (R. haematites and R. biasolettiana) led them to conclude that about 14% total calcium carbonate was the result of photosynthesis. The calcite crystals grow parallel to the trichomes and nucleate on the fibrous outer layer of the sheath (Caudwell et al. 2001). The formation of laminae in R. haematites was monitored over a 7-year period by Caudwell et al. (1997, 2001) in central France. A micritic dark lamina formed in three stages: (i) initial development of a dark lamina in the zone where the filaments develop false branching during the wet season; (ii) calcification in this zone as microsparitic and sparitic B.A. Whitton and P. Mateo lamellae during dry spells; subsequent bacterial micritization of the lamellae during an extended warm dry season. In the particular climate experienced during the study period there was either a single dark lamina for 2 years growth or a dark lamina thicker than the annual growth rate. This suggests that the extent to which laminations reflect marked annual changes in climate and nutrient concentrations is likely to differ between regions and sometimes between years. Although carbonate deposition is rare among marine Rivularia, it has been reported for R. mesenterica and R. polyotis (Golubic and Campbell 1981), but apparently without laminations. In a comparison (Myshrall et al. 2010) of stromatolitic and thrombolitic (non-laminated) mats at Highbourne Cay, Bahamas, the most obvious difference was the button mats of calcifying Dichothrix in the latter. Dichothrix can also be an important component of freshwater calcifying cyanobacterial communities (Whitton et al. 1986), individual colonies are not large enough to form distinct layers. Its role in the marine environment is discussed in Sect. 22.7. Two other genera have been described. Sacconema Borzi (1882) ex Bornet et Flahault (1886) seems little more than a form of Gloeotrichia with very thick and partly confluent sheaths, while Gardnerula (de Toni 1936) resembles a large Dichothrix colony with subdichotomous branching. Both genera were mentioned briefly by Castenholz (1989), but only Gardnerula by Rippka et al. (2001c). However there seems little reason why these should be treated as separate genera rather than species of Gloeotrichia and Dichothrix, respectively. 22.5.4 Gloeotrichia Gloeotrichia includes all the colonial forms producing akinetes. Most are benthic in shallow lakes, ponds and ditches, but G. raciborskii and G. natans typically float at the surface for a considerable period, G. pisum and similar forms are epiphytic on older parts of submerged aquatic plants and G. echinulata is planktonic during part of its growth cycle. G. ghosei was reported as floating or benthic (Claassen 1973) and it seems likely that some other species are the same. Other characters used in species descriptions include size of colonies, whether the sheath is layered, whether the hair extends beyond the surface of the colony and the size and shape of akinetes. Akinete details are in some cases essential for species recognition, while long inspection of a pond population may be needed to decide whether or not it can form akinetes. If not, it would be recorded as Rivularia, as molecular data would be needed to exclude the possibility that it was a Gloeotrichia which had adapted for survival at that site by no longer forming akinetes.. Akinete formation often leads to disintegration of the trichome. However, sometimes several akinetes are formed, 22 Rivulariaceae whilst at other times the trichome breaks away and can then form another akinete. Many queries could be answered by careful observation. To what extent is the behaviour of the trichome subsequent to formation of its first akinete a species- or strain- specific character? To what extent can increase in colony number occur without akinete formation? Does the large size of the akinete permit a colony to develop from a single trichome and, even if so, does colony formation normally involves trichome aggregation. If trichome aggregation is the norm for colony initiation, what happens if the trichomes are too sparse for this to happen? The highly tapered and highly gas-vacuolate trichomes lacking a heterocyst which were occasional in the plankton of a shallow lakes on Clare Island in 2004 may have been such a case (Whitton 2007). These might have been a residual population of G. echinulata too sparse to permit trichome aggregation, as this species had been reported on the island by West (1912), but typical material could no longer be found there. Apart from G. echinulata, G. natans is the most widely reported species, though apparently more frequent in subtropical and tropical waters. It is common in rice fields. For instance, it is the most widespread cyanobacterium in rice fields in Chile (Pereira et al. 2005), frequent in rice fields around Los Baños, Philippines (Martinez-Goss and Whitton, unpublished data) and frequent in deepwater rice fields near Manikganj, Bangladesh (Rother et al. 1988). Populations usually start to develop as attached colonies, but then become detached; these floating colonies sometimes become intermingled with submerged macrophytes, but often they float to the surface. Young colonies may include trichomes with gas vacuoles (Geitler 1932), but old colonies without gas vacuoles may still float. A Philippines rice-field isolate was studied in the laboratory and outdoor raceway ponds by Querijero-Palacpac et al. (1990), but BG-110 medium was used, so the results are unlikely to reflect its behaviour in nature. In the laboratory the specific growth rate was 0.076 h−1. Using a stirring rate of 30 rpm in the raceway ponds, daily production of cultures harvested to maintain cell densities of 0.7, 1.15 and 1.5 g (d. wt) L−1 was 24.7, 17.1 and 18.1 g m−2 day−1, respectively. The phycobiliprotein content in the culture maintained at 1.5 g L−1 reached 14% of the biomass. The planktonic G. echinulata has received considerable study. Some of the first field experiments were by Spodniewska (1971) using suspended bottles during summer in the eutrophic, holomictic Lake Mikołasjskie, Poland. She found that respiration averaged about 30% gross primary production for G. echinulata. A suggestion by Rodhe (1949) that G. echinulata needed an unknown organic component for growth led Zehnder (1963) to obtain evidence that the only organic factor required for laboratory growth is a chelating agent. Lange (1974) reported that G. echinulata was one of four bacterized strains (species) of cyanobacteria which failed to grow 579 without added chelator, as opposed to six which did. However, even when an artificial chelator was included in the medium, Chang (1979a) found that soil extract and 5–40 mM glucose both aided growth of an axenic strain. It has still not been established clearly whether or not planktonic G. echinulata forms its own chelator. This also raises the question as to how important accumulation of elements besides P (see below) is for colonies developing on bottom sediments. Although many of the reports of conspicuous G. echinulata populations are for lakes which might otherwise be considered oligo- to mesotrophic, research was focussed initially on eutrophic lakes. Caution is needed in comparing results for a bloom-forming species at different sites, as has become clear in the study of Microcystis (Chap. 7). Formation of dense plankton populations of Gloeotrichia echinulata was first described for Ellesmere, a shallow English lake, by Phillips (1884), though he quotes an 1880 report which mentioned its period of hibernation at the bottom and then rising in summer. However, the first detailed account was by Roelofs and Oglesby (1970) for Green Lake, Washington, a shallow lake without a permanent thermocline in summer. Subsequent to G. echinulata disappearing from the plankton in September, a bottom sample taken in November showed akinetes, but no colonies. However, developing colonies were found on the bottom in the following January. The authors mentioned short filaments of 4–6 cells, though they did not explain whether these were organized into spherical colonies. Colonies were still absent from the plankton on 23 June, but were entering it on 3 July. Apparently colonies were developing on the bottom for 6 months, while the planktonic phase only lasted for about one-quarter of the year. In a 2-year study on the same eutrophic lake, Barbiero and Welch (1992) concluded that the plankton derived 40% G. echinulata colonies from the benthos and that these accounted for a significant part of the internal P loading of the lake. However, Istvánovics (2008) concluded from their results and those of Karlsson-Elfgren et al. (2003) that recruitment of colonies was insignificant in some years and so the internal P load is highly variable in Gloeotrichia lakes. The most detailed studies have been on the moderately eutrophic, stratified (in summer) Lake Erken in S-E. Sweden. Heavy blooms of G. echinulata often occur during July and August at the same time as the epilimnion deepens gradually to 10 m (Pettersson et al. 1990, 1993). Like other wind-exposed lakes, akinetes and colonies are found in large numbers on the bottom sediments (Istvánovics et al. 1993). Benthic colonies averaged 5 × 105 m−2 in the top 4 cm of sediments in areas of the lake shallower than 10 m in March 1991 (Pettersson et al. 1993). Istvánovics et al. (1990) and Pettersson et al. (1990) suggested that epilimnetic growth might be largely or solely based on the internal P pool obtained while the organism was associated with the sediments and a number of P uptake experiments led Istvánovics et al. (1993) 580 to conclude the colonies were unable to acquire any P in the epilimnion. The P uptake threshold exceeded the epilimnetic concentration of soluble (filterable) reactive P by an order of magnitude. An overview of the results for Lake Erken indicated that recruiting colonies translocated some two-thirds of the total net P load from the sediments to the epilimnion (Istvánovics 2008). It is not clear how this fits with the observations of Karlsson-Elfgren et al. (2003) made during a 2-year study that recruitment only contributed to <5% of the maximum G. echinulata abundance during late summer. However, the authors emphasized that variations in the measured abundance of G. echinulata could reflect measured rates of migration from the sediment, and variations in either pelagic colony division rate and pelagic residence time. The conclusion of Istvánovics et al. (1993) that epilimnetic G. echinulata did not use organic P in Lake Erken was based on “alkaline phosphatase” activity measurements using the method of Pettersson (1980). However, this methodology has limitations. Because of the high substrate concentration used, the results may be unreliable at the pH tested (Whitton et al. 2005). PMEase activity tends to be higher in Rivulariaceae than other cyanobacteria (Sect. 22.4.2) and when PDEase activity was compared, a Gloeotrichia strain showed the highest ratio to phosphomonoesterase activity to any strain, suggesting the importance of phosphodiesters as substrates (Whitton et al. 1991). In addition Gloeotrichia strains were the most successful genus at utilizing phytic acid. Although these strains did not include G. echinulata, they indicate the importance of checking its ability to use a range or organic P compounds under realistic assay conditions. If the hairs of planktonic Gloeotrichia do not have this role, are they little more than a relic of structures important at a late stage of benthic growth which helps to enlarge colony size? The possibility that iron and boron might influence growth of G. echinulata colonies in Lake Erken was considered by Hyenstrand et al. (2001) using an in situ experiment. Iron had the most effect, enhancing the effect of adding phosphate and nitrate. Addition of boron led to an even greater increase, but boron had no effect if there was not additional iron at the same time. Studies by Vuorio et al. (2009) on carbon isotope signatures in colonies in Erken and Pyhäjärvi (Finland) showed that there was a systematic increase in d13C with increasing colony size, in spite of substantial differences in the average d13C in the two lakes. The response to size was probably due to diffusion limitation of C availability. Carey et al. (2008, 2009) reported on recent outbreaks of G. echinulata in oligo- to mesotrophic lakes in N-E. USA, where at least 27 examples had occurred between 2002 and 2006 in Maine and New Hampshire. In one of these, Lake Sunapee, recruitment of colonies was 1.13 and 0.32 colonies cm−2 day−1 in 2005 and 2006, respectively. These values contrast with the 36 colonies cm−2 day−1 for Green B.A. Whitton and P. Mateo Lake (Barbiero 1993) and c. 1,520 colonies cm−2 day−1 for Lake Erken (Forsell and Pettersson 1995). Nevertheless Carey et al. (2009) concluded that increased sediment total dissolved phosphate may have influenced recruitment. A pulse of sediment was typically followed by an increase in G. echinulata 18–19 days later. The authors commented that this time-lag corresponds to the time akinetes need to germinate (1–7 days) and take up P from the sediment (2–3 weeks) before recruiting to the water column (Tymowski and Duthie 2000; Karlsson 2003). Carey et al. (2009) also raised the question as to whether the influence of P on germination is accelerated by a discrete pulse or because the concentration exceeds a particular threshold, as suggested by Pettersson et al. (1993). Temperature and light are also factors which can be involved (Karlsson-Elfgren et al. 2004). In spite of the many studies on G. echinulata the growth and division of colonies have not been described in detail, though partially separated, healthy colonies quite often occur in samples. Grazers can impact in various ways, including removal of a whole colony or disruption into fragments (Sect. 22.6.3.1). It is unclear whether single filaments or the small groups remaining after grazing can continue growth and perhaps eventually form akinetes. G. echinulata can also have a considerable impact on other phytoplankton grazers. Colonies and culture filtrate of Lake Erken G. echinulata both stimulated the growth of some species known to co-occur with it in Lake Sunapee and Lake Erken (Carey and Rengefors 2010). Of the seven species tested, none isolated from Lake Erken, five were stimulated by the presence of colonies and two by filtrate. The potential effects of zooplankton populations may depend on the duration and severity of a bloom and the tolerance of the zooplankton to potential toxins; zooplankton populations may be able to maintain themselves during prolonged blooms (Fey et al. 2010). However, interactions with G. echinulata and zooplankton might change with time due to the development of tolerance to toxins. Cyanobacteria and heterotrophic bacteria associated with G. echinulata colonies in Lake Erken, Sweden, may be a supplementary food source for the grazing zooplankton (Eiler et al. 2006), especially Daphnia pulex (Fey et al. 2010). The use of experimental ponds by Carey et al. (2011) showed that in nutrient-limited ponds the presence of G. echinulata led to an increase in the biomass of small-sized phytoplankton and that this stimulation was related to the zooplankton biomass. An increased zooplankton biomass led to increased grazing of colonies, which may have increased the rate of nutrient leakage to other phytoplankton, thereby intensifying the stimulatory effect. In contrast, G. echinulata had a negative effect on small-sized phytoplankton in eutrophic ponds. The authors concluded that in nutrient-limited systems, G. echinulata may subsidize plankton food webs in nutrient-limited systems through nutrient leakage and could thus accelerate eutrophication. 22 Rivulariaceae In several lakes studied in the UK, such as Talkin Tarn, Cumbria, akinetes in planktonic colonies are most frequent towards the end of the growth period, though occasionally occur earlier in the season (B.A.W., unpublished). The storage granules abundant in the akinetes are cyanophycin, suggesting the trichomes are P-limited by the time the akinetes form; presumably they are important as a N source when germinating on the sediments. P limitation is therefore one factor likely to be involved, but other factors should be investigated such as possible responses to grazer activity. In view of the many studies on G. echinulata we will summarize our own interpretation of the results. The species can occur in lakes or other waterbodies ranging from near oligotrophic to eutrophic, but obtains much of its phosphate from bottom sediments. It is therefore possible for moderate blooms to occur in lakes where water management organizations had not expected a problem, providing the sediments are rich enough in P. The presence of the organism may itself enhance eutrophication over a number of years. Although no evidence of P uptake could be found for epilimnetic G. echinulata in Lake Erken, further research is needed before it can be assumed that this applies to most lakes. The possibility that hairs have a role in mobilizing organic P in less nutrient-rich lakes should be investigated. Populations can persist in lakes ranging over at least three orders of magnitude of colony recruitment from the sediments. The species mostly occurs in stratified lakes, but populations can persist in relatively shallow unstratified lakes, at least for a few years. In temperate regions it persists on the bottom for the majority of the year, but little is known about how much growth occurs during this period. Although the species occurs in the tropics (e.g. Aldabra Atoll, 9° S: Donaldson and Whitton 1977; central Brazil, 15° S: Campos and Senna 1988), there are no accounts of its seasonality in large tropical waterbodies. 22.6 Interactions with Other Organisms 22.6.1 Free-Living Phototrophs There are many records of Calothrix and Gloeotrichia, and occasionally other Rivulariaceae, growing as epiphytes, but the extent to which this depends on features of the host plant or merely the overall chemical environment is usually unclear. However, the physical association with the host plant is especially close in some cases, such as the occurrence of Calothrix and Gloeotrichia (and several other cyanobacterial genera) on the lower epidermis and reproductive pockets of Lemna leaves, and are presumably responsible for the N2-fixing activity associated with some duckweed blooms (Duong and Tiedje 1985). A small colonial Gloeotrichia is frequent on the submerged stems of deepwater rice plants in 581 Bangladesh at situations where the underwater parts of the plant are well illuminated, such as next to channels (Whitton et al. 1988b). There are many studies on the effects of cyanobacteria on rice plants, but most have used strains which had not been growing in the vicinity of rice plants in the field. However, Karthikeyan et al. (2009) studied the effects of 8 axenic strains, including 3 Calothrix, which had been isolated from the rhizosphere of a wheat cultivar. C. ghosei showed the second highest nitrogenase activity in the light and the highest value in the dark. When wheat seedlings were co-cultured with C. ghosei, short filaments were found inside the root hairs and cortical region. Such strains were considered promising candidates for developing plant growth-promoting associations with wheat. In a comparison of the distribution of the epiphytes on the marine angiosperm Posidonia by Trautman and Borowitzka (1999), Calothrix, the only cyanobacterium included in the list, grew on both sides of the leaves of Posidonia australis, though restricted to the basal sections. Other records include on Cymodocea nodosa (Reyes and Sansón 1997) and on Thalassia testudinum, where N2-fixing activity was correlated with the abundance of Calothrix (Capone and Taylor 1977). A similar correlation was suggested for N2-fixing activity on the mangrove Avicennia marina (Hicks and Silvester 1985). Colonial Rivulariaceae almost always contain other cyanobacteria and sometimes eukaryotic algae as well, especially pennate diatoms. Perhaps the initial period of trichome aggregation leading to colony formation permits entry of some other motile species, though little is known about the extent to which invasion of mature healthy colonies occurs. Chang (1983) reported that Pseudanabaena catenata was frequent in Gloeotrichia echinulata colonies in Plöner See, Germany, in August and September, but not in G. echinulata colonies earlier in the year. Based on various observations, the author suggested that release of substances by G. echinulata may favour growth of Pseudanabaena catenata. Quantitative studies on the frequency of various types of diatoms in and on colonies of the marine Rivularia atra versus their frequency as epiphytes on macroalgae were made by Snoeijs and Murasi (2004); the former relationship was described as symbiosis, but this extends the definition well beyond that used by most authors. Among the 35 most frequent diatom species, the relative abundance of species with motile and epipsammic life-form was 2.5 times higher within the Rivularia colonies than those with the attached epiphytic life-form. The frequency of epipsammic forms in the colonies was due to the frequency with which sand grains were incorporated into the colonies. Diatom community diversity was higher in the Rivularia samples, probably because cells epiphytic on the colonies were included. The motile diatoms inside the colonies were ones that occurred in mucilage matrices when present on stones. Other reasons were suggested why the 582 relationship should be favourable for diatoms, including the fact that the colonies are tough and can resist wave action; the authors did not consider the possibility that Rivularia toxicity might deter grazers (see below), nor did they have any suggestion as to how Rivularia might gain from the interaction, other than some possible nutritional advantage. 22.6.2 Symbiotic Associations Extracellular growths of Calothrix have been reported from the thalli of several marine green algae, either entirely within the tissue, as in the Enteromorpha reported by Lami and Meslin (1959) and Codium decorticum from USA (Rosenberg and Paerl 1981), or partially epiphytic and partially inside the tissues, as a Codium from New Zealand (Dromgoole et al. 1978). In the case of C. decorticum, Calothrix was one of three cyanobacteria responsible for nitrogenase activity within a reducing microzone. The best known symbiotic associations between Calothrix and other organisms are cyanobiont-containing lichens (Adams 2000). Although Nostoc is much the most frequent cyanobiont in lichens, Calothrix is probably the next most frequent. According to Ahmadjian (1967), it is the phototroph in Lichina, Calotrichopsis and Porocyphus and Dichothrix in Placynthium, although other authors consider the phototroph of some Lichina corresponds better to Dichothrix. The morphological appearance inside the lichen usually has only slight resemblance to free-living Calothrix. Henssen (1969) found that the phototroph of Lichina rosulans consisted of short, contorted chains and basal heterocysts were rare. Basal heterocysts, but no tapering were found in L. polycarpa, while the photobiont of L. tasmanica and L. willeyi had normal tapering (Henssen 1973). An ultrastructural study of L. confinis by Janson et al. (1993) showed that heterocysts were present in the thallus, but differed from the rest of the Calothrix cells in that no haustoria were in contact with them. Glutamine synthetase levels in the cyanobiont were a great deal less in free-living (cultured) Calothrix sp. The decrease was greater in mature parts of the lichen thallus than in the apical region. Rubisco was mostly located in carboxysomes of the vegetative cells of the cyanobiont. In free-living (cultured) Calothrix sp., phycoerythrin was located along the thylakoid membranes in both vegetative cells and heterocysts. Although the cyanobiont cells showed a similar pattern, the levels were lower. The cyanobacterium in the coralloid roots of the cycad Encephalartos hildebrandtii showed no resemblance to any well known genus, but developed what the authors considered to be a typical Calothrix morphology in culture (Huang and Grobbelaar 1989). The micrographs of the association do not show unequivocally that the organism is Calothrix, perhaps because the study used BG-11o medium, so further studies are needed. Although the endophyte exhibited substantial B.A. Whitton and P. Mateo nitrogenase activity, no heterocysts could be distinguished, yet heterocysts were found during culture; akinetes were also formed, again raising doubt whether this symbiont really is Calothrix. Thajuddin et al. (2010) reported one strain of Calothrix (again based on culture with BG-11o medium) among 18 cyanobacteria from coralloid roots of Cycas revoluta; presumably they overlooked the previous study, as they claimed their record to be the first for Calothrix as a cycad symbiont. The importance of marine diatoms with symbiotic N2fixing cyanobacteria in various subtropical oceans, especially the North Pacific, has only become clear in recent years, although they were first described by Lemmermann (1905). The two intracellular symbionts, Richelia intracellularis and Calothrix rhizosoleniae, both have terminal heterocysts, but only the latter has a trichome with obvious tapering. Although Richelia intracellularis has not been isolated and brought into long-term culture, this has been done for Calothrix rhizosoleniae from Chaetoceros in the N. Pacific (Foster et al. 2010). The trichomes retained the terminal heterocyst, but 2 years after isolation a culture also showed intercalary heterocysts. The nifH, 16S rRNA and hetR sequences were amplified and cloned from this isolate and field populations of Richelia associated with Hemiaulus hauckii (N. Atlantic) and Rhizosolenia clevei (N. Pacific) (Foster and Zehr 2006). The results indicated that the symbionts in the three different hosts are distinct species or strains. Based on assays of nifH gene abundance and gene expression of plankton populations, there was no record for Calothrix rhizosoleniae in the western tropical Atlantic (Foster et al. 2007), but there was in the eastern equatorial Atlantic (Foster et al. 2009). This was probably the intracellular symbiont, but the possibility of free-living trichomes could not be ruled out. A comparison of the growth rates of cells in symbiotic association with estimates for ones which were not showed that the rates were higher for symbiotic cells (Foster et al. 2011). In the case of N2 fixation, the rates were similar among the symbioses and there was rapid transfer (within 30 min) of fixed N. The N2 fixation rates estimated for Calothrix and Richelia symbionts were 171–420 times higher when the cells were symbiotic compared with estimates for cells living freely. The cyanobacterial symbiont fixed more N than needed for its own growth and in the case of Richelia, up to 97.3% fixed N was transferred to the diatom. 22.6.3 Animals 22.6.3.1 Introduction The account of Gloeotrichia echinulata in Sect. 22.5.4 indicates that a lot needs to be known for successful application of a modelling approach to provide quantitative predictions. Nevertheless Benedetti-Cecchi et al. (2005) showed for a marine mid-littoral assemblage how a basic understanding of the mechanisms of interaction of the main species plus some 22 Rivulariaceae knowledge of their natural history can lead to correct predictions of the effects of one species on another. Filamentous algae monopolized the substratum when limpets were absent, but, when present Rivularia, the red alga Rissoella and barnacles colonized. Barnacles and Rissoella jointly reduced the coverage of Rivularia and the local density of limpets and this eventually led to colonization by filamentous algae late in succession. There was thus an intermediate stage in the succession when Rivularia was successful. The interactions are probably similarly complex in most ecosystems, but the following are simplified accounts of the main types of interaction between Rivulariaceae and animals.. 22.6.3.2 Destruction of Whole Filaments Rivulariaceae are of course frequently influenced by grazing and this is described for Gloeotrichia echinulata in Sect. 22.5.4. Sometimes there may be little selection for or against a particular cyanobacterium, as in the study by Theivandirrajah and Jeyaseelan (1977) on grazing of Calothrix and Anabaena by mosquito larvae in a Sri Lankan pond. However, Edmondson (1938) reported that the rotifer Lindia euchromatica in the littoral zone of Linley Pond, Connecticut, targeted Gloeotrichia. As it lived in the “slime” of the cyanobacterium, where it fed on the filaments and deposited its eggs, it is unclear whether this was G. echinulata or a species forming larger colonies. In the case of floating G. natans colonies in the Botanic Garden near Chiangmai, Thailand, in October 2011, the colonies contained three large species of testate amoebae, at least one of which was frequent and grazing Gloeotrichia trichomes by gradually engulfing and digesting them from the ends (B.A.W., unpublished data). Lindia euchromatica was the only grazer feeding on G. echinulata in Green Lake, Washington (Roelofs and Oglesby 1970). Although sometimes very abundant, it was noted on relatively few occasions and only in late summer. Colony size (up to 2 mm) was thought to restrict grazing by copepods and cladocerans. However, mesocosm experiments with Daphnia pulex, Holopedium gibberum, Ceriodaphnia quadrangula and Bosmina longirostris showed that they all increased the proportion of damaged colonies (Fey et al. 2010). Ceriodaphnia damaged the greatest proportion, but only Daphnia pulex fed on Gloeotrichia echinulata reproduced. Nothing is known about the impact of grazers on bottom sediments. G. echinulata was one of a number of filamentous cyanobacteria cultures grazed by the amoeba Naegleria isolated from Dianchi Lake, Kunming, Yunnan, China (Liu et al. 2006). Forms with aggregated filaments such as Aphanizomenon were not grazed, which suggests that the Gloeotrichia culture had lost its colonial structure. Unicellular Chroococcales were ingested, but subsequently excreted. Nematodes are sometimes frequent in old colonies of Rivularia biasolettiana, but their possible role in destroying colonies has not been studied. The possibility of putting grazing to practical use as food for the apple snail Pomacea patula catemacensis 583 which is endemic in Lake Catemaco, Mexico, was investigated by Ruiz-Ramírez et al. (2005). The survival of the snail, which is an important fishing resource, is threatened and hence there is a need to culture it. Calothrix sp. grown in 40-L containers proved to be better than pelleted carp food. 22.6.3.3 Partial Destruction of Filaments A number of studies are known where Rivulariaceae populations appear especially successful in resisting grazing pressure. Several reasons have been put forward to explain this. Mats of Calothrix embedded among Pleurocapsa persist in the lower parts of the temperature gradient of hot springs of the western USA, where grazing by ostracods eliminates other cyanobacteria capable of overgrowing these species (Wickstrom and Castenholz 1985). In the case of Hunter’s Hot Springs, Oregon, Calothrix and Pleurocapsa withstood a dense population of Potamocypris sp., which exerted heavy grazing pressure on Oscillatoria terebriformis, the cyanobacterium dominating immediately upstream in the temperature zone where the ostracod was unable to persist (Castenholz 1973). The hairs of the Calothrix, which protrude from the colonies, are, however, heavily grazed (Castenholz, personal communication 1978). An experimental study (Power et al. 1988) showed that the dominance of Calothrix in a stream in Ozark Mountains, Oklahoma, depended on the presence of grazing by fish such as minnows. When the grazers were removed, the Calothrix mats became overgrown by diatoms within 4–10 days, while re-exposure to grazers led to the mats developing again. The authors suggested that the ability to regenerate from basal parts of the trichome may contribute to their persistence under intense grazing. They also discussed possible ecosystem-level effects, such as whether the removal of loosely attached diatoms contributed to the water clarity of Ozark streams dominated by cyanobacteria felts. This was in spite of the fact that dominance by the N2-fixer brought about by the grazing might enhance N loading in the stream, as shown by Wilkinson et al. (1985) for the grazing of abundant N2-fixing cyanobacteria (including Calothrix) on coral reefs. 22.6.3.4 Avoiding the Grazers Cattaneo (1983) interpreted the success of Gloeotrichia pisum as an epiphyte in Lake Memphremagog, Québec-Vermont, as resulting from its ability to resist grazing by snails, which had a large impact on other algae. When snails were excluded, but not oligochaetes and chironomids, the control epiphyte community was replaced by green algae, mainly Mougeotia. It was suggested that the very tough sheath of Gloeotrichia may protect it from grazing. A laboratory experimental study (Osa-Afiana and Alexander 1981) showed that established populations of Calothrix and Tolypothrix were not grazed by the ostracod Cypris sp., whereas Aulosira and, to a lesser extent, Anabaena were. However, when the cyanobacterium and ostracod were inoculated together, the ostracod increased 584 in biomass just as much with Calothrix as with Anabaena. Presumably Calothrix and Tolypothrix formed sheaths in older cultures, though the authors did not discuss whether this was a factor reducing the effects of grazing. Rivularia colonies sometimes persist for very long periods, which can be as much as 10 years in the case of R. haematites (Pentecost and Whitton 2000). As the presence of calcite crystals does not appear to inhibit grazing, at least by molluscs, toxicity was suggested as a likely factor. Several studies have now reported studies indicating its probable importance for at least some Rivulariaceae. The need for colonies to survive for many months under P-limited conditions with only slow growth during that period (Sect. 22.5) would add to the value of any means of detering grazers. The toxicity of calcified cyanobacterial communities from Spanish streams to stream invertebrates was investigated by Aboal et al. (2000, 2002). Rivularia was abundant in these communities, though other cyanobacteria were also well represented. The latter study, which generated index values for routine monitoring assays based on macroinvertebrates and diatoms, showed a clear inverse relationship between the dominance of cyanobacteria and the values obtained for the indices. Evidence for microcystin production by Rivularia in populations from N-E. and S-E. Spain was reported by Aboal et al. (2005), Aboal and Puig (2009): MC-LR, MC-RR and MC-YR were predominant. Microcystin-LR was also reported for Gloeotrichia echinulata (Carey et al. 2007). Among 19 cyanobacteria isolated from the sediments of R. Nile and adjacent channels, all of which were toxic to Artemia, extracts from Calothrix parietina and Phormidium tenue caused neurotoxic symptoms to mice within 10 h; five other species showed hepatotoxicity (Mohamed et al. 2006). A different group of compounds was investigated by Höckelmann et al. (2009). This was the range of sesquiterpenes formed by the axenic geosmin-producing cyanobacterium, Calothrix PCC 7507, together with a comparison of their abundance in old versus very old standing cultures. Among the quantitatively important products, eremophilone differed from the others in being mainly excreted into the medium, so bioassays were conducted on its toxicity to three invertebrates. Acute toxicity to Chironomus riparius (insect) occurred at 29 mM and to Thamnocephalus platyurus (crustacean) at 22 mM, whereas no toxic effects were observed on Plectus cirratus (nematode). Neither 6.11-epoxyisodaucane nor isodihydroagarofuran exhibited toxicity to any of these at concentrations up to 100 mM. 22.6.4 Bacteria and Fungi Sequencing and phylogenetic analysis of 16S rRNA genes of bacteria attached to Gloeotrichia echinulata colonies in Lake Erken showed a diverse community different from that in the B.A. Whitton and P. Mateo main waterbody and included not only cyanobacteria, but populations affiliated with Proteobacteria, Bacteriodetes, Acidobacteria, Fusobacteria, Firmicutes and Verrucomicrobia (Eiler et al. 2006). Fan (1956) reported that the hyphae of certain fungi parasitize the cells of Calothrix and the cells or trichomes usually die. Although cyanophage activity is important for many other bloom-forming cyanobacteria, no studies have been reported for Gloeotrichia echinulata. 22.7 Geological Record and Environmental Change As it is increasingly becoming possible to relate genera and often also particular species of modern-day Rivulariaceae to particular types of environment, their presence in the fossil record is also becoming important in interpreting past environments. If the interpretation of molecular data by Domínguez-Escobar et al. (2011) is correct (see Sect. 22.3.1), then Calothrix and Rivularia originated about ~1,500 Ma (Mesoproterozoic) and Gloeotrichia about and 400–300 Ma. (Carboniferous) This means that the doubts of Golubic and Campbell (1981) about fossil Rivulariaceae (mostly concerning Palaeorivularia) may not be justified. Palaeocalothrix, described from the Precambrian by Zhao-Liang (1984a, b) has a basal heterocyst, large akinete, trichome tapered toward the apex and a sheath, all of which are features of Gloeotrichia and possibly a few species of Calothrix. Assuming structures had a similar physiological significance then as in modern Calothrix, then Palaeocalothrix lived in a well oxygenated environment with a marked variation in some nutrient, probably P. There are convincing records of Rivulariaceae for most geological periods from the Pleiocene onwards. These are a few examples. Dragastan et al. (1996) compared presumed Rivularia haematites from the Pleistocene and modern material. Based on tube dimensions and morphology they appear to be essentially the same. A 20-m core of Middle Pleistocene travertine showed encrustations of Phormidium, Dichothrix and Rivularia (Casanova 1984). Various Pliocene and Pleistocene deposits appear to include oncoids (see below) or similar structures formed by cyanobacteria (Casanova 1982, 1984). Celyphus rallus is almost certainly an Early Cretaceous Rivulariaceae (Batten and Van Geel 1985). Among the microfossils associated with Late Cretaceous freshwater stromatolites in Sonora, Mexico, Beraldi-Campesi et al. (2004) listed Calothrix estromatolitica. The use of cyanobacteria as biomarkers of hydrological changes in the Late Quaternary sediments of the South Kerala Sedimentary Basin in India, was assessed by Limaye et al. (2010). A study of three different types of modern Rivularia colony (see Sect. 22.5.3) led Obenlüneschloss and Schneider (1991) to several conclusions relevant to interpreting fossil material. Once the organic matter has decomposed, it is 22 Rivulariaceae impossible to make direct conclusions on the trichome and sheath morphology based on the calcification structures. As several different calcification mechanisms occurred in these colonies, it is doubtful whether different calcification structures in fossil analogues can be attributed to different calcification mechanisms. The possibility that carbonaceous meteorites might also contain fossil cyanobacteria has been presented in a number of papers by Hoover, with the latest (2011) arguing the case in detail. It includes the suggestion that several tapering structures with a smooth basal body in the Orgueil CI1 carbonaceous meteorite resemble the filaments with a heterocyst of a modern Calothrix, though the organism shown for comparison from White River, Washington, does not resemble typical Calothrix, including the fact that the filament is only 0.8 mm wide, apparently well below the limit for other records in the genus. The paper includes much of interest and led to rapid and vigorous internet discussion, some of it highly critical. It seems likely the debate will continue for some time before there is any scientific consensus about the possibility of these meteorites containing cyanobacteria, let alone that some of these resemble Rivulariaceae. The presence of Rivulariaceae in relatively recent calcareous deposits has been reported by many authors based on both molecular (Sigler et al. 2003) and morphological data. Calothrix was among the cyanobacteria of stromatolitic stalagmites at the entrance to a cave in Slovenia (Mulec et al. 2007). The calcified remains of Rivularia colonies, especially R. haematites, often persist in or around streams and lake margins when they become exposed. Spherical or subspherical oncoids in small calcareous streams usually consist predominantly of Rivularia (Rott 1991) or almost entirely so, such as those at Sunbiggin, Cumbria, UK (illustrated in Whitton and Potts 2000). Based on the living algae found in Little Conestoga Creek, Roddy (1915) concluded that the calcareous structures all the way down the stream were formed largely by Rivularia, and possibly also the calcareous structures forming deep deposits under soil in the catchment. Oncoids had disappeared from this stream by the time it was resurveyed by Golubic and Fischer (1975), mostly likely because of acidic industrial effluents. Most records of the decrease (Kann 1982) or loss of calcifying Rivularia are, however, probably due largely to P enrichment (Pentecost and Whitton 2000). It also seems possible that atmospheric N deposition leading to increases in stream water N might shift away from N2-fixers to eukaryotic algae adapted to environments with high N:P (Mateo et al. 2010). The Rivulariaceae are usually frequent together with narrow sheathed Oscillatoriaceae in calcareous deposits on submerged rocks in shallow, highly calcareous lakes, where the deposits are sufficiently thick and well-laminated to justify the term stromatolite. The calcified layers forming stromatolites in Lough Corrib, Ireland, provide striking examples, with 585 frequent Dichothrix being intermingled with abundant Schizothrix fasciculata (Fig. 22.10 ): see also article by B. Kennedy et al. (2012) in online supplement to this book. In larger calcareous lakes, such as the Bodensee (Lake Constance), benthic rocks can have a thick calcareous layer with a furrow or brain-like pattern (Müller-Stoll 1986), which is formed mainly by Schizothrix fasciculata and Rivularia haematites in the Bodensee; the author lists reports of other lakes with similar structures. Oncoids form where detached calcareous growths are moved around by weak currents, such as the floor of the Untersee of Lake Constance, where they consist of Phormidium and Calothrix and/or Dichothrix (Schäfer and Stapf 1978). A study (Van Geel et al. 1984) of Lateglacial deposits near Usselo, The Netherlands, gives detailed records of changes in “Gloeotrichia-type”, though the authors did not give details of the structures counted. The sequence, which started in an oligotrophic shallow pool with very low organic production in a barren sandy-landscape, had an early phase at about 12,600 B.P., but lasting several hundred years, characterized as the Gloeotrichia-type, where Characeae were also important. The authors suggested N2 fixation may have initiated nutrient availability. As mentioned in Sect. 22.5.3, Rivulariaceae sometimes calcify in the marine environment. This was reported for Dichothrix at Highbourne Cay, Bahamas (Planavsky et al. 2009), but the filaments were not preserved to form lithified microbialites. In contrast, adjacent cyanobacterial mats formed well-laminated stromatolites. However, Dichothrix contributed to one of the types of thrombolite described by Mobberley et al. (2011) and carbonate-trapping stromatolites in the marine environment can be formed by D. bornetiana (Monty 1967). 22.8 Discussion Although molecular data show that the Rivulariaceae are a heterogeneous group, field and experimental observations indicate that they are all well adapted to environments with marked variation in ambient P, often organic P in particular. In flowing waters and the marine intertidal this variation is temporal, but it is both temporal and spatial in Gloeotrichia echinulata. The periodicity of temporal variation in Calothrix is much shorter than in colonial forms and in Dichothrix probably shorter and more irregular than in most Rivularia. When ambient P concentrations drop, stored polyphosphate is metabolized and surface phosphatase activities increase, permitting efficient use of organic phosphates in the environment. Hormogonia formation ceases when internal P concentrations fall below a particular level. Taxa differ considerably in the relative extent of their P-rich and P-limited periods. G. echinulata in the eutrophic Lake Erken is probably P-rich for much of the time it is growing actively, but there 586 may be longer periods in other lakes when the species is at least moderately P-limited. Rivularia is at the other extreme and in temperate region streams it is P-limited for much of the year. In this case the long periods of presumably low metabolic activity and the persistence of colonies for many years at some sites makes it important for grazing activity to be minimized; their formation of microcystins needs to be studied in detail. The presence of more than one genotype in the colonies of at least some Rivularia populations may provide flexibility in responding to longer term fluctuations in the environment at the site. Presumably the genotype best suited to the current environment grows the most rapidly, but perhaps the relative proportions of different genotypes are also controlled at the stage when trichomes first aggregate to form colonies. The ability of some species to form long colourless hairs greatly increases the surface for phosphatase activities and probably also uptake of inorganic P. In colonial forms hairs also help to enhance the size of the colony. However, the production of hairs leads to the loss of cells each time the trichome becomes P-rich. There is no such wastage in species which do not form hairs. Hairs permit the trichome to use very low ambient P concentrations, yet also make efficient use of pulses of higher P concentration. It also seems possible that hair cell formation in some species leads to changes in the cytoplasmic membrane that permit passage of organic phosphates, which are then hydrolyzed inside the cell. The need to reduce entry of seawater might explain why Calothrix viguieri only forms hairs in freshwater (Sect. 22.4.2). Nevertheless, most marine Rivulariaceae do possess hairs. Once phosphate ions are released in the hair, rapid transfer must occur to the base near of the trichome where polyphosphate granule formation is first evident. Although this review has emphasized the importance of P in determining the growth cycle, the possibility that Fe may have a similar influence in some ecological situations should be borne in mind. The cycle of hair and hormogonia formation in response to Fe status reported for Calothrix strains in Sects. 22.2.2.1 and 22.2.2.3 is probably an indirect response to P limitation. Nevertheless, there may be environments where marked changes in Fe availability are the key factor and perhaps there were past geological periods when it was especially important. Apart from this uncertainty, understanding of how morphology relates to the environment is sufficiently advanced that it should be possible to characterize the environment of all distinctive taxa of Rivulariaceae. However, the fact that colonies may include more than one genotype indicates the likelihood of a continuum of forms in many types of environment. Molecular information about how different taxa respond to changes in internal P status would be a great help, as would establishing whether or not morphological changes are sometimes a response to external P concentration or types of molecule. Nevertheless sufficient is known about the B.A. Whitton and P. Mateo ecology of Rivulariaceae to comment on the environment of fossil material and old lake plankton records in some detail. Acknowledgements We thank Prof. Yuwadee Peerapornpisal and her students at the Department of Biology, Chiangmai University, Thailand, for helping B.A.W. to visit the Queen Sirikit Botanic Garden with its conspicuous floating cyanobacterial colonies. We much appreciate helpful comments on a draft of this chapter by Dr Allan Pentecost. References Abe T, Ikeda T, Yanada R, Ishikura M (2011) Concise total synthesis of calothrixins A and B. Org Lett 13(13):3356–3359 Aboal M, Puig MA (2009) Microcystin production in Rivularia colonies of calcareous streams from Mediterranean Spanish basins. Algol Stud 130:39–52 Aboal M, Puig MA, Rios H, López-Jiménez E (2000) Relationship between macroinvertebrate diversity and toxicity of cyanophyceae (Cyanobacteria) in some streams from eastern Spain. Verh Int Verein Limnol 27:555–559 Aboal M, Puig MA, Mateo P, Perona E (2002) Implications of cyanophyte toxicity on biological monitoring of calcareous streams in north-east Spain. J Appl Phycol 14:49–56 Aboal M, Puig MA, Asencio AD (2005) Production of microcystins in calcareous Mediterranean streams: the Alharabe River, Segura River basin in south-east Spain. J Appl Phycol 17:231–243 Adamec F, Kaftan D, Nedbal L (2005) Stress-induced filament fragmentation of Calothrix elenkinii (Cyanobacteria) is facilitated by death of high-fluorescence cells. J Phycol 41:835–839 Adams DG (2000) Symbiotic interactions. In: Whitton B, Potts M (eds) Ecology of cyanobacteria: their diversity in time and space. Kluwer Academic Publishers, Dordrecht, pp 523–561, 668 pp Ahmadjian V (1967) A guide to the algae occurring in lichen symbionts: isolation, culture, cultural physiology, and identification. Phycologia 6:127–160 Al-Mousawi AHA (1984) Biological studies on algae in ricefield soil from the Iraqi marshes. Ph.D. thesis, University of Durham, Durham Al-Mousawi AHA, Whitton BA (1983) Influence of environmental factors on algae in rice-field soil from the Iraqi marshes. Arab J Gulf Res 1:237–253 Aziz A (1985) Blue-green algal nitrogen fixation associated with deepwater rice in Bangladesh. Ph D thesis, University of Durham, Durham Aziz A, Whitton BA (1988) Influence of light flux on nitrogenase activity of the deepwater rice-field cyanobacterium (blue-green alga) Gloeotrichia pisum in field and laboratory. Microbios 53:7–19 Aziz A, Whitton BA (1989) Morphogenesis of blue-green algae. II. Hair differentiation in Gloeotrichia pisum. Bangladesh J Bot 16:69–81 Banerjee M, John J (2005) Phosphatase activity of non-hair forming cyanobacterium Rivularia and its role in phosphorus dynamics in deepwater rice-fields. Appl Ecol Environ Res 3(1):55–60 Barbiero RP (1993) A contribution to the life history of the planktonic cyanophyte, Gloeotrichia echinulata. Arch Hydrobiol 127:87–100 Barbiero RP, Welch EB (1992) Contribution of benthic blue-green algal recruitment to lake populations and phosphorus recruitment. Freshw Biol 27:249–260 Bary D (1863) Beitrag zur Kenntnis der Nostocaceen, insbesondere Rivulariaceen. Flora (Jena) 46:553–560 Batten DL, Van Geel B (1985) Celyphus rallus, probably early Cretaceous rivulariacean blue-green alga. Rev Palaeobot Palynol 44:233–241 Bauersachs T, Compaoré J, Hopmans EC, Stal LJ, Schouten S, Damsté JSS (2009) Distribution of heterocyst glycolipids in cyanobacteria. Phytochemistry 70:2034–2039 22 Rivulariaceae Benedetti-Cecchi L, Vaselli S, Maggi E, Bertocci I (2005) Interactive effects of spatial variance and mean intensity of grazing on algal cover in rock pools. Ecology 86:2212–2222 Beraldi-Campesi H, Cevallos-Ferriz SRS, Chacón-Baca E (2004) Microfossil algae associated with Cretaceous stromatolites in the Tarahumara Formation, Sonora, Mexico. Cretac Res 25:249–265 Bernardo PH, Chai CLL, Heath GA, Mahon PJ, Smith GD, Waring P, Wilkes BA (2004) Synthesis, electrochemistry, and bioactivity of the cyanobacterial calothrixins and related quinones. J Med Chem 47:4958–4963 Bernardo PH, Chai CLL, Le Guen M, Smith GD, Waring P (2007) Structure-activity delineation of quinones related to the biologically active calothrixin B. Bioorg Med Chem Lett 17(1):82–85 Berrendero E, Perona E, Mateo P (2008) Genetic and morphological characterization of Rivularia and Calothrix (Nostocales, Cyanobacteria) from running water. Int J Syst Evol Microbiol 58:447–460 Berrendero E, Perona E, Mateo P (2011) Phenotypic variability and phylogenetic relationships of Tolypothrix and Calothrix (Nostocales, Cyanobacteria) from running water. Int J Syst Evol Microbiol 61(12):3039–3051 Böcher TW (1946) Dichothrix gelatinosa sp. n., its structure and resting organs. K Danske Vidensk Selsk Skr IV(4):1–15 Bornet JBE, Flahault CGM (1886) Revision des Nostocacées hétérocystées contenues dans les principaux herbiers de France. Ann Sci Nat Bot Sér 7(3):323–381 Campbell D, Houmard J, Tandeau de Marsac N (1993) Electron transport regulates cellular differentiation in the filamentous cyanobacterium Calothrix. Plant Cell 5(4):451–463 Campos IFP, Senna PAC (1988) Nostocophyceae (Cyanophyceae) de Lagao Bonita, Distrito Federal, Brasil. Parto 1. Acta Bot Bras 2(1–2):7–30 Capone DG, Taylor BF (1977) Nitrogen fixation (acetylene reduction) in the phyllosphere of Thalassia testidinum. Mar Biol 40(1): 19–28 Carey CC, Rengefors K (2010) The cyanobacterium Gloeotrichia echinulata stimulates the growth of other phytoplankton. J Plankton Res 32:1349–1354 Carey CC, Haney JF, Cottingham KL (2007) First report of microcystinLR in the cyanobacterium Gloeotrichia echinulata. Environ Toxicol 22:337–339 Carey CC, Weathers KC, Cottingham KL (2008) Gloeotrichia echinulata blooms in an oligotrophic lake helpful insights from eutrophic lakes. J Plankton Res 30:893–904 Carey CC, Weathers KC, Cottingham KL (2009) Increases in phosphorus at the sediment-water interface may influence the initiation of cyanobacterial blooms in an oligotrophic lake. Verh Int Verein Limnol 30(8):1185–1188 Carey CC, Cottingham KL, Weathers KC, Brentrup JA, Ruppertsberger NM, Ewing HA, Hairston NG (2011) Cyanobacteria are not all bad: Gloeotrichia echinulata may stimulate plankton food webs in nutrient-limited freshwater ecosystems. In: Abstract, 96th Ecological Society of America annual meeting, Austin, TX, 7–12 Aug 2011 Carpenter EJ, Cox JL (1974) Production of pelagic Sargassum and a blue-green epiphyte in the western Sargasso Sea. Limnol Oceanogr 19:429–436 Casanova J (1982) Morphologie et biolithogenèse des barrages de travertins. In: Nicod J (ed) Formations Carbonatées -Externes, Tufs et Travertins. Institut de géographie, Aix-en-Provence, pp 45–54 Casanova J (1984) Genèse des carbonates d’un travertin pléistocène: interpretation paleoecologique du sondage Peyr I (Comprégnac, Aveyron). Geobios 17:219–229 Castenholz RW (1973) Ecology of the blue-green algae in hot springs. In: Carr NG, Whitton BA (eds) The biology of the blue-green algae. Blackwell, Oxford, pp 379–414, 655 pp 587 Castenholz RW (1989) Rivulariaceae. In: Bergey’s manual of systematic bacteriology 3. Williams & Wilkins, Baltimore, pp 1790–1793 Cattaneo A (1983) Grazing on epiphytes. Limnol Oceanogr 28:124–132 Caudwell C, Lang J, Pascal A (1997) Étude expérimentale de la lamination stromatolithes à Rivularia haematites mat en climat tempéré: édification des lamines microcritiques. C R Acad Sci Paris 324(IIa): 883–890 Caudwell C, Lang J, Pascal A (2001) Lamination of swampy-rivulets Rivularia haematites stromatolites in a temperate climate. Sediment Geol 143:125–147 (+ 148: 451) Chang TP (1979a) Growth and acetylene reduction by Gloeotrichia echinulata (Smith) Richter in axenic culture. Br Phycol J 14:207–210 Chang TP (1979b) Zur Morphologie von Gloeotrichia echinulata (Smith) Richter in axenischer Kultur. Nova Hedwigia Kryptogamenk 21:265–283 Chang T-P (1983) Interaction of water-blooming cyanophyte Gloeotrichia echinulata and its endophytic blue-green alga Pseudanabaena catenata. Arch Hydrobiol 97:320–328 Chang T-P, Blauw TS (1980) Nitrogen fixing capacity of two colonial types of Gloeotrichia echinulata and its endophytic blue-green alga Pseudanabaena catenata. Arch Hydrobiol 89:382–386 Chen X, Smith GD, Waring P (2003) Human cancer cell (Jurkat) killing by the cyanobacterial metabolite calothrixin A. J Appl Phycol 15:269–277 Claassen MI (1973) Freshwater algae of Southern Africa. I. Notes on Gloeotrichia ghosei R.N.Singh. Br Phycol J 8:325–331 Cuzman OA, Stefano Ventura S, Sili C, Mascalchi C, Tulio Turchetti T, D’Acqui LP, Tiano P (2010) Biodiversity of phototrophic biofilms dwelling on monumental fountains. Microb Ecol 60:81–95 Darley J (1967) Sur quelques résultats de la culture en laboratoire de deux espèces de Calothrix Agardh (Myxophycées-Rivulariacées). C R Acad Sci Ser 264D:1013–1015 de Toni G (1936) Noterelle di Nomencl. Algological 8:5 Desikachary TV (1946) Germination of the heterocyst in two members of the Rivulariaceae, Gloeotrichia raciboskii Wolosz. and Rivularia mangini Frémy. J Indian Bot Soc 25:11–17 Desikachary TV (1959) Cyanophyta. Indian Council of Agricultural Research, New Delhi, 686 pp Dillon JG, Castenholz RW (2003) The synthesis of the UV-screening pigment, scytonemin, and photosynthetic performance in isolates from closely related natural populations of cyanobacteria (Calothrix sp.). Environ Microbiol 5(6):484–491 Dillon JG, Miller SR, Castenholz RW (2003) UV-acclimation responses in natural populations of cyanobacteria (Calothrix sp.). Environ Microbiol 5:473–483 Doan NT, Rickards RW, Rothschild JM, Smith GD (2000) Allelopathic actions of the alkaloid 12-epi-hapalindole E isonitrile and calothrixin A from cyanobacteria of the genera Fischerella and Calothrix. J Appl Phycol 12:409–416 Doan NT, Stewart PR, Smith GD (2001) Inhibition of bacterial RNA polymerase by the cyanobacterial metabolites 12-epi-hapalindole E isonitrile and calothrixin A. FEMS Microbiol Lett 196:135–139 Domínguez-Escobar J, Beltrán Y, Bergman B, Díez B, Ininbergs K, Souza V, Falcón LI (2011) Phylogenetic and molecular clock inferences of cyanobacterial strains within Rivulariaceae from distant environments. FEMS Microbiol Lett 316:90–99 Donaldson A, Whitton BA (1977) Algal flora of freshwater habitats on Aldabra. Atoll Res Bull 215:1–26 Douglas D, Peat A, Whitton BA, Wood P (1986) Influence of iron status on structure of the cyanobacterium (blue-green alga) Calothrix parietina. Cytobios 47:155–165 Dragastan G, Golubic S, Richter DK (1996) Rivularia haematites: a case study of the recent versus fossil morphological. Taxonomical considerations. Rev Española Micropaleontol 28:43–73 Dromgoole FI, Silvester WB, Hicks BJ (1978) Nitrogenase associated with Codium species from New Zealand. N Z J Mar Freshw Res 12:17–22 588 Duong TP, Tiedje JM (1985) Nitrogen fixation by naturally occurring duckweed-cyanobacterial associations. Can J Microbiol 31: 327–330 Edmondson WT (1938) Three new species of Rotatoria. Trans Am Microsc Soc 57:153–157 Ehrenreich IM, Waterbury JB, Webb EA (2005) Distribution and diversity of natural product genes in marine and freshwater cyanobacterial cultures and genomes. Appl Environ Microbiol 71(11):7401–7413 Eiler A, Olsson JA, Bertilsson S (2006) Diurnal variations in the auto- and heterotrophic activity of cyanobacterial phycospheres (Gloeotrichia echinulata) and the identify of attached bacteria. Freshw Biol 51:298–311 Falcón LI, Magall Falcón S, Castillo A (2010) Dating the cyanobacterial ancestor of the chloroplast. ISME J 4:777–783 Fan KC (1956) Revision of Calothrix Ag. Rev Algol N.S. 2(3): 154–178 Fey SB, Mayer ZA, Davis SC, Cottingham KL (2010) Zooplankton grazing of Gloeotrichia echinulata and associated life history consequences. J Plankton Res 9:1337–1347 Fogg GE (1969) The physiology of an algal nuisance. Proc R Soc Lond B 173:175–189 Forsell L, Pettersson K (1995) On the seasonal migration of the cyanobacterium Gloeotrichia echinulata in Lake Erken, Sweden, and is influence on the pelagic population. Mar Freshw Res 46:287–293 Foster RA, Zehr JP (2006) Characterization of diatom-cyanobacteria symbioses on the basis of nifH, hetR and 16S rRNA sequences. Environ Microbiol 8:1913–1925 Foster RA, Capone DG, Carpenter EJ, Mahaffey C, Subramaniam A, Zehr JP (2007) Influence of the Amazon River Plume on distributions of free-living and symbiotic cyanobacteria in the Western Tropical North Atlantic Ocean. Limnol Oceanogr 52:517–532 Foster RA, Subramaniam A, Zehr JP (2009) Distribution and activity of diazotrophs in the Eastern Equatorial Atlantic. Environ Microbiol 11:741–750 Foster RA, Goebel NL, Zehr JP (2010) Isolation of Calothrix rhizosoleniae (cyanobacteria) strain sc01 from Chaetoceros (Bacillariophyta) spp. diatoms of the subtropical North Pacific Ocean. J Phycol 46:128–1037 Foster RA, Kuypers MMM, Vagner T, Paerl RW, Muzat N, Zehr JP (2011) Nitrogen fixation and transfer in open ocean diatomcyanobacterial symbioses. ISME J 5:1484–1493. doi:10.1038/ ismej.2011.26 Friedmann I (1956) Über der Blaualga Gardnerula corymbosa (Harvey) J. De Toni und ihr Vorkommen im Mittelmeer. Öst Bot Z 103:336–341 Fritsch FE (1907a) The subaerial and freshwater algal flora of the Tropics. Ann Bot 21:235–275 Fritsch FE (1907b) A general consideration of the freshwater algae and subaerial algae of Ceylon. Proc R Soc Lond B 79:197–254 Geitler L (1932) Cyanophyceae. In: Rabenhorst’s Kryptogamen-flora, vol 14. Akademische Verlagsgesellschaft, Leipzig, 1196 pp Golden SS (1995) Light-responsive gene expression in cyanobacteria. J Bacteriol 177:1651–1654 Golubic S, Campbell SE (1981) Biogenically formed aragonite concretions in marine Rivularia. In: Monty C (ed) Phanerozoic stromatolites. Springer, Berlin/Heidelberg, pp 209–229 Golubic S, Fischer AG (1975) Ecology of calcareous nodules forming in Little Connestoga Creek, near Lancaster, Pennsylvania. Verh Int Verein Limnol 19:2315–2323 Gomont M(A) (1895) Note sur un Calothrix sporifère (Calothrix stagnalis sp. n.). J Bot 9:1–6 Gonsalves EA, Kamat ND (1960) New species of Cyanophyceae from Mysore State – 1. J Bombay Nat Hist Soc 57(2):454–456 Grainger SLJ, Peat A, Tiwari DN, Whitton BA (1989) Phosphomonoesterase activity of the cyanobacterium (blue-green alga) Calothrix parietina. Microbios 59:7–17 B.A. Whitton and P. Mateo Gunasekera SP, Miller MW, Kwan JC, Luesch H, Paul VJ (2010) Molassamide, a depsipeptide serine protease inhibitor from the marine cyanobacterium Dichothrix utahensis. J Nat Prod 73:459–462 Healey FP (1982) Phosphate. In: Carr NG, Whitton BA (eds) The biology of cyanobacteria. Blackwell Scientific Publishers, Oxford, pp 105–124, 688 pp Henssen A (1969) Three non-marine species of the genus Lichina. Lichenologist 4:88–98 Henssen A (1973) New or interesting cyanophytic lichens. I. Lichenologist 5:444–451 Heuff H, Horkan K (1984) Caragh. In: Whitton BA (ed) Ecology of European rivers. Blackwell, Oxford, pp 363–384, 644 pp Hicks BJ, Silvester WB (1985) Nitrogen fixation associated with the New Zealand mangrove (Avicennia marina (Forsk.) Vierh. var. resinifera (Forst. f.) Bakh.). Appl Environ Microbiol 49:955–959 Höckelmann C, Becher PG, von Reuss SH, Jüttner F (2009) Sesquiterpenes of the geosmin-producing cyanobacterium Calothrix PCC 7507 and their toxicity to invertebrates. Z Naturforsh 64C:49–55 Hongmei J, Aitchison JC, Lacap DC, Peerapornpisal Y, Sompong U, Pointing SB (2005) Community phylogenetic analysis of moderately thermophilic cyanobacterial mats from China, the Philippines and Thailand. Extremophiles 9:325–332 Hoover RB (2011) Fossils of cyanobacteria in CI1 carbonaceous meteorites. J Cosmol 13:1–39 Huang TC, Grobbelaar N (1989) Isolation and characterization of endosymbiotic Calothrix. (Cyanophyceae) in Encephalartos hildebrandtii (Cycadales). Phycologia 28:464–468 Hübel H, Hübel M (1974) In-situ Messungen der diurnalen Stickstoff Fixierung an Mikrobenthos der OstseeKüste. Arch Hydrobiol Suppl 46:39–54 Hyenstrand P, Rydin E, Gunnerhed M, Linder J, Blomqvist P (2001) Response of the cyanobacterium Gloeotrichia echinulata to iron and boron additions – an experiment from Lake Erken. Freshw Biol 46:735–741 Islam MR, Whitton BA (1992a) Phosphorus content and phosphatase activity of the deepwater rice-field cyanobacterium (blue-green alga) Calothrix D764. Microbios 69:7–16 Islam MR, Whitton BA (1992b) Cell composition and nitrogen fixation by the deepwater rice -field cyanobacterium (blue-green alga) Calothrix D764. Microbios 69:77–88 Istvánovics V (2008) The role of biota in shaping the phosphorus cycle in lakes. Freshw Rev 1:143–174 Istvánovics V, Pettersson K, Pierson D (1990) Partitioning of phosphate uptake between different size groups of planktonic microorganisms in Lake Erken. Verh Int Verein Limnol 24:231–245 Istvánovics V, Pettersson K, Rodrigo MA, Pierson D, Padisák J, Colom W (1993) Gloeotrichia echinulata, a colonial cyanobacterium with a unique phosphorus uptake and life strategy. J Plankton Res 15:531–552 Janson S, Rai AN, Bergman B (1993) The marine lichen Lichina confinis (O. F. Müll.) C. Ag.: ultrastructure and localization of nitrogenase, glutamine synthetase, phycoerythrin and ribulose 1, 5-bisphosphate carboxylase/oxygenase in the cyanobiont. New Phytol 124:149–160 Johansson C (1979) Chlorophyll content and the periphytic algal vegetation in six streams in North Jämtland, Sweden, 1977. Medd Växtbiol Inst Uppsala 1979(2):1–28 Jones K, Stewart WDP (1969) Nitrogen turnover in marine and brackish habitats III. The production of extracellular nitrogen by Calothrix scopulorum. J Mar Biol Assoc UK 49:475–488 Kann E (1982) Qualitative Veränderungen der litoralen Algenbiocönose österreichischer Seen. Arch Hydrobiol Suppl 62:440–490 Karlsson I (2003) Benthic growth of Gloeotrichia echinulata cyanobacteria. Hydrobiologia 506:189–193 Karlsson-Elfgren I, Rydin E, Hyenstrand P, Pettersson K (2003) Recruitment and pelagic growth of Gloeotrichia echinulata in Lake Erken. J Phycol 39:1050–1056 22 Rivulariaceae Karlsson-Elfgren I, Rengefors K, Gustafson S (2004) Factors regulating recruitment from the sediment to the water column in the bloomforming cyanobacterium Gloeotrichia echinulata. Freshw Biol 49:265–273 Karthikeyan N, Prasanna R, Sood A, Jaiswal P, Nayak S, Kaushik BD (2009) Physiological characterization and electron microscopic investigation of cyanobacteria associated with wheat rhizosphere. Folia Microbiol 54:43–51 Katayama M, Kobayashi M, Ikeuchi M (2007) Phototropism observed in cyanobacterium Rivularia sp. Plant Cell Physiol 48:158 Kenyon CN, Rippka R, Stanier RY (1972) Fatty acid composition and physiological properties of some filamentous blue-green algae. Arch Mikrobiol 83:216–236 Khan QA, Lu J, Hecht SM (2009) Calothrixins, a new class of human DNA topoisomerase I poisons. J Nat Prod 72(3):438–442 Khoja TM, Whitton BA (1971) Heterotrophic growth of blue-green algae. Arch Mikrobiol 79:280–282 Khoja TM, Livingstone D, Whitton BA (1984) Ecology of a marine Rivularia population. Hydrobiologia 108:65–73 Kirkby SM, Whitton BA (1976) Uses of coded data in study of Calothrix and Rivularia. Br Phycol J 11:407–416 Lachance M-A (1981) Genetic relatedness of heterocystous cyanobacteria by deoxyribonucleic acid-deoxyribonucleic acid reassociation. Int J Syst Bacteriol 31:13–147 Lami R, Meslin R (1959) Sur une cyanophycée, Calothrix Chapmanii nom. nov., vivant à l’intérieur d’une Entéromorphe limicole. Bull Lab Marit Dinard 44:47–49 Lange W (1974) Chelating agents and blue-green algae. Can J Microbiol 20(10):1311–1320 Larsen LK, Re M, Patterson GML (1994) b-carbolines from the blue-green alga Dichothrix baueriana. J Nat Products 57(3): 419–421 Lebedeva NV, Boichenko VA, Semenova LR, Pronina NA, Stadnichuk IN (2005) Effects of glucose during photoheterotrophic growth of the cyanobacterium Calothrix sp PCC 7601 capable for chromatic adaptation. Russ J Plant Physiol 52:235–241 Lemmermann E (1905) Die Algenflora der Sandwich-Islen. Ergebnisse einer Reise nach dem Pacific. H. Schauinsland 1896/97. Bot Jahrbuch Systematik Pflanzengesch Pflanzengeogr 3:607–663 Limaye RB, Kumaran KPN, Nair KM, Padmalal D (2010) Cyanobacteria as potential biomarkers of hydrological changes in the Late Quaternary sediments of South Kerala Sedimentary Basin, India. Quaternary Int 213:79–90 Liu Xinyao, Shi Miao, Liao Yonghong, Gao Yin, Zhang Zhongkai, Wen Donghui, Wu Weizhong (2006) An Chencai Feeding characteristics of an amoeba (Lobosea: Naegleria) grazing upon cyanobacteria: food selection, ingestion and digestion progress. Microb Ecol 31:315–325 Livingstone D, Whitton BA (1983) Influence of phosphorus on morphology of Calothrix parietina (Cyanophyta) in culture. Br Phycol J 18:29–38 Livingstone D, Whitton BA (1984) Water chemistry and phosphatase activity of the blue-green alga Rivularia in Upper Teesdale streams. J Ecol 72:405–421 Livingstone D, Khoja TM, Whitton BA (1983) Influence of phosphorus on physiology of a hair-forming blue-green alga (Calothrix parietina) from an upland stream. Phycologia 22:345–350 Livingstone D, Pentecost A, Whitton BA (1984) Diel variations in nitrogen and carbon dioxide fixation by the blue-green alga Rivularia in an upland stream. Phycologia 23:125–133 Maehr H, Smallheer JM (1984) Rivularins. Preliminary synthetic studies. Org Chem 49:1549–1553 Mahasneh IA, Grainger SLJ, Whitton BA (1990) Influence of salinity on hair formation and phosphatase activities of the blue-green alga (cyanobacterium) Calothrix viguieri D253. Br Phycol J 25: 25–329 589 Mateo P, Douterelo I, Berrendero E, Perona E (2006) Physiological differences between two species of cyanobacteria in relation to phosphorus limitation. J Phycol 42:61–66 Mateo P, Berrendero E, Perona E, Loza V, Whitton BA (2010) Phosphatase activities of cyanobacteria as indicators of nutrient status in a Pyrenees river. Hydrobiologia 652:255–268 Maxwell TF (1974) Developmental morphology of the blue-green alga, Gloeotrichia echinulata. J Phycol 10 (Suppl):4 McErlean CSP, Sperry J, Blake AJ, Moody CJ (2007) Synthesis of the calothrixins, pentacyclic indolo[3,2-j]phenanthridine alkaloids, using a biomimetic approach. Tetrahedron 63(45):10963–10970 Mobberley JM, Ortega MC, Foster JS (2011) Comparative microbial diversity analyses of modern marine thrombolitic mats by barcoded pyrosequencing. Environ Microbiol. doi:10.1111/j.1462-2920. 2011.02509.x Mohamed ZA, El-Sharouny HM, Ali WSM (2006) Microcystin production in benthic mats of cyanobacteria in the Nile River and irrigation canals, Egypt. Toxicon 47:584–590 Monty CLV (1967) Distribution and structure of recent stromatolitic algal mats, eastern Andros Island, Bahamas. Ann Soc Geol Belg Bull 96:585–624 Mulec J, Kosi G, Vrhovsek D (2007) Algae promote growth of stalagmites and stalactites in karst caves (Skocjanske jame, Slovenia). Carbonates Evaporites 22:6–9 Müller-Stoll WR (1986) Der Cyanophyceen-Bewuchs der Furchenoder Hirnsteine des Bodensees. Carolinea 44:51–60 Myshrall KL, Mobberley JM, Green SJ, Visscher PT, Havemann SA, Reid RP, Foster JS (2010) Biogeochemical cycling and microbial diversity in the thrombolitic microbialites of Highborne Cay, Bahamas. Geobiology 8(4):337–354 Norton RS, Wells RJ (1982) A series of chiral polybrominated biindoles from the blue-green marine algae Rivularia firma. Applications of 13C NMR spin-lattice relaxation data and 123H1H coupling constants to structure elucidation. J Am Chem Soc 104:3628–3635 Obenlüneschloss J, Schneider J (1991) Ecology and calcification patterns of Rivularia (Cyanobacteria). Algol Stud 64:489–502 Osa-Afiana LO, Alexander M (1981) Factors affecting predation by a microcrustacean (Cypris sp.) on nitrogen-fixing blue-green algae. Soil Biol Biochem 13:27–32 Pedersen M, DaSilva EJ (1973) Simple brominated phenols in the blue-green alga Calothrix brevissima West. Planta 115:83–86 Pentecost A (1987) Growth and calcification of the freshwater cyanobacterium Rivularia haematites. Proc R Soc Lond B 232:125–136 Pentecost A, Franke U (2010) Photosynthesis and calcification of the stromatolitic freshwater cyanobacterium Rivularia. Eur J Phycol 45:345–353 Pentecost A, Riding R (1986) Calcification in cyanobacteria. In: Leadbeater BSC, Riding R (eds) Biomineralization of lower plants and animals. Clarendon, Oxford, pp 73–90 Pentecost A, Whitton BA (2000) Limestones. In: Whitton BA, Potts M (eds) Ecology of cyanobacteria: their diversity in time and space. Kluwer Academic Publishers, Dordrecht, pp 257–279, 668 pp Pereira I, Moya M, Reyes G, Kramm V (2005) A survey of heterocystous cyanobacteria in Chilean rice-fields. Gayana Bot 62(1):26–32 Pettersson K (1980) Alkaline phosphatase activity and algal surplus phosphorus as phosphorus deficiency indicators in Lake Erken. Arch Hydrobiol 89:54–87 Pettersson K, Istvánovics V, Pierson D (1990) Effects of vertical mixing on phytoplankton phosphorus supply during summer in Lake Erken. Verh Int Verein Limnol 24:236–241 Pettersson K, Herlitz E, Istvánovics V (1993) The role of Gloeotrichia echinulata in the transfer of phosphorus from sediments to water in Lake Erken. Hydrobiologia 253:123–129 Phillips W (1884) The breaking of the Shropshire meres. Trans Shropshire Archaeol Nat Hist Soc 7:277–300 590 Planavsky N, Reid RP, Lyons TW, Myshrall KL, Visscher PT (2009) Formation and diagenesis of modern marine calcified cyanobacteria. Geobiology 7:566–576 Potts M, Whitton BA (1977) Nitrogen fixation by blue-green algal communities in the intertidal zone of the Lagoon of Aldabra Atoll. Oecologia 27:275–283 Power ME, Stewart AJ, Matthews WJ (1988) Grazer control of algae in an Ozark mountain stream: effects of short-term exclusion. Ecology 69:1894–1898 Prasanna R, Pabby A, Saxena S, Singh PK (2004a) Modulation of pigment profiles of Calothrix elenkenii in response to environmental changes. J Plant Physiol 161:1125–1132 Prasanna R, Pabby A, Singh PK (2004b) Effect of glucose and light-dark environment on pigmentation profiles in the cyanobacterium Calothrix elenkenii. Folia Microbiol (Praha) 49:26–30 Querijero-Palacpac NM, Martinez MR, Boussiba S (1990) Mass cultivation of the nitrogen-fixing cyanobacterium Gloeotrichia natans, indigenous to rice-fields. J Appl Phycol 2:319–325 Rai AK, Pandey KD, Kashyap AK (1978) Heterocyst differentiation in Calothrix. New Phytol 81:647–651 Reed RH, Stewart WDP (1983) Physiological responses of Rivularia atra: osmotic adjustment to hypersaline media. New Phytol 95:595–603 Reyes J, Sansón M (1997) Temporal distribution and reproductive phenology of the epiphytes on Cymodocea nodosa leaves in the Canary Islands. Bot Mar 40:1–6 Rickards RW, Rothschild JM, Willis AC, de Chazal NM, Kirk J, Kirk K, Saliba KJ, Smith GD (1999) Calothrixins A and B, novel pentacyclic metabolites from Calothrix cyanobacteria with potent activity against malaria parasites and human cancer cells. Tetrahedron 55(47): 13513–13520 Rippka R, Deruelles J, Waterbury JB, Herdman M, Stanier RY (1979) Generic assignments, strain histories and properties of pure cultures of cyanobacteria. J Gen Microbiol 111:1–61 Rippka R, Castenholz RW, Herdman M (2001a) Subsection IV.II form-genus I. Calothrix Agardh 1824. In: Boone DR, Castenholz RW, Garrity GM (eds) Bergey’s manual of systematic bacteriology, vol 1, 2nd edn. Springer, New York, pp 582–585, 721 pp Rippka R, Castenholz RW, Herdman M (2001b) Subsection IV.II form-genus II. Rivularia Agardh 1824. In: Boone DR, Castenholz RW, Garrity GM (eds) Bergey’s manual of systematic bacteriology, vol 1, 2nd edn. Springer, New York, pp 586–587, 721 pp Rippka R, Castenholz RW, Herdman M (2001c) Subsection IV. (Formerly Nostocales Castenholz 1989b sensu Rippka, Deruelles, Waterbury, Herdman and Stanier 1979) In: Boone DR, Castenholz RW, Garrity GM (eds) Bergey’s manual of systematic bacteriology. vol 1, 2nd edn. Springer, New York, pp 562–599, 721 pp Roddy HJ (1915) Concretions in streams formed by the agency of blue green algae and related plants. Proc Am Phil Soc 54:246–258 Rodhe W (1949) Environmental requirements of fresh-water plankton algae. Symb Bot Ups X:149 Roelofs TD, Oglesby RT (1970) Ecological observations on the planktonic cyanophyte Gloeotrichia echinulata. Limnol Oceanogr 15:224–229 Rosenberg G, Paerl HW (1981) Nitrogen fixation by blue-green algae associated with the siphonous green seaweed Codium decorticum. Mar Biol 61:151–158 Rother JA, Aziz A, Hye Karim N, Whitton BA (1988) Ecology of deepwater rice-fields in Bangladesh 4. Nitrogen dixation by blue-green algal communities. Hydrobiologia 169:43–56 Rott E (1991) Oncoids from the summer-warm River Alz (Bavaria) – mophology and dominant cyanophytes. Algol Stud 64:469–482 Ruangsomboon S, Chidthaisong A, Bunnag B, Inthorn D, Harvey NW (2006) Lead (Pb2+) removal from wastewater by the cyanobacterium Calothrix marchica. Kasetsart J (Nat Sci) 40:784–794 Ruiz-Ramírez R, Espinosa-Chávez F, Martínez-Jerónimo F (2005) Growth and reproduction of Pomacea patula catemacensis B.A. Whitton and P. Mateo Baker, 1922 (Gastropoda: Ampullariidae) when fed Calothrix sp (Cyanobacteria). J World Aquacult Soc 36:87–95 Sabater S (1989) Encrusting algal communities in a Mediterranean river basin. Arch Hydrobiol 114:555–573 Sabater S, Guasch H, Romani A, Muñoz I (2000) Stromatolitic communities in Mediterranean streams: adaptations to a changing environment. Biodivers Conserv 9:379–392 Sanderson MJ (2002) Estimating absolute rates of molecular evolution and divergence times: a penalized likelihood approach. Mol Biol Evol 19:101–109 Schäfer A, Stapf KRG (1978) Permian Saar-Nahe basin and recent Lake Constance (Germany): two environments of lacustrine algal carbonates. Spec Publ Int Assoc Sediment 2:83–107 Schlegel I, Doan NT, de Chazal N, Smith GD (1998) Antibiotic sensitivity of cyanobacterial isolates from Australia and Asia against green algae and cyanobacteria. J Appl Phycol 10:471–479 Schneider SC, Lindstrøm E-A (2011) The periphyton index of trophic status PIT: a new eutrophication metric based on non-diatomaceous benthic algae in Nordic rivers. Hydrobiologia 685:143–156 Schwenender S (1894) Zur Wachstumgeschichte der Rivularien. Sber Preuss Akad Wissensch, Berlin, pp 951–961 Shalini IM, Dhar DW, Gupta RK (2008) Phylogenetic analysis of cyanobacterial strains of genus-Calothrix by single and multiplex randomly amplified polymorphic DNA-PCR. World J Microbiol Biotechnol 24:927–935 Shehata FHA, Whitton BA (1981) Field and laboratory studies on blue-green algae from aquatic sites wutg high zinc levels. Verh Int Verein Limnol 21:1466–1471 Sigler WV, Bachofen R, Zeyer J (2003) Molecular characterization of endolithic cyanobacteria inhabiting exposed dolomite in central Switzerland. Environ Microbiol 5:618–627 Sihvonen LM, Lyra C, Fewer DP, Rajaniemi-Wacklin P, Lehtimäki JM, Wahlsten M, Sivonen K (2007) Strains of the cyanobacterial genera Calothrix and Rivularia isolated from the Baltic Sea display cryptic diversity and are distantly related to Gloeotrichia and Tolypothrix. FEMS Microbiol Ecol 61:74–84 Sinclair C, Whitton BA (1977a) Influence of nutrient deficiency on hair formation in the Rivulariaceae. Br Phycol J 12:297–313 Sinclair C, Whitton BA (1977b) Influence of nitrogen source on morphology of Rivulariaceae (Cyanophyta). J Phycol 13:335–340 Singh RN, Tiwari DN (1970) Frequent heterocyst germination in the blue-green alga Gloeotrichia ghosei Dingh. J Phycol 6:172–176 Skácelová O (2006) Dichothrix ledereri sp. nova, a new cyanobacterium from old coal/mining deposits, and occurrence of the genus Dichothrix (Cyanobacteria, Nostocales) in the Czech Republic. Algol Stud 121:1–21 Skulberg OM (2000) Microalgae as a source of bioactive molecules – experience from cyanophyte research. J Appl Phycol 12:341–348 Smith GD, Doan NT (1999) Cyanobacterial metabolites with bioactivity against photosynthesis in cyanobacteria, algae and higher plants. J Appl Phycol 11:337–344 Smith RV, Peat A (1967) Comparative structure of the gas-vacuoles of blue-green algae. Arch Mikrobiol 57:111–122 Snoeijs P, Murasi LK (2004) Symbiosis between diatoms and cyanobacterial colonies. Vie Milieu 54:163–170 Spodniewska I (1971) The influence of experimental increase of biomass of the blue-green algae Gloeotrichia echinulata (Smith) Richter on phytoplankton production. Ekol Pol 19:475–483 Srivastava AK, Bhargava P, Kumar A, Rai LC, Neilan BA (2009) Molecular characterization and the effect of salinity on cyanobacterial diversity in the rice fields of Eastern Uttar Pradesh, India. Saline Syst 5/1/4:17. doi:10.1186/1746-1448-5-4 Stewart WDP (1967) Nitrogen turnover in fresh and marine habitats. II. Use of 15N in measuring nitrogen fixation in the field. Ann Bot N.S. 31:385–407 22 Rivulariaceae Taton A, Grusibic S, Ertz D, Hodgson DA, Piccardi R, Biondi N, Tredici MR, Mainini M, Losi D, Marinelli F, Wilmotte A (2006) Polyphasic study of Antarctic cyanobacterial strains. J Phycol 42:1257–1270 Thajuddin N, Muralitharan G, Sundaramoorthy M, Ramamoorthy R, Ramachandran S, Akbarsha MA, Gunasekaran M (2010) Morphological and genetic diversity of symbiotic cyanobacteria from cycads. J Basic Microbiol 50:254–265 Theivendirarajah K, Jeyaseelan K (1977) The ingestion of blue-green algae by mosquito larvae. Ceylon J Sci Biol Sci 12:156 Thomazeau S, Houdan-Fourmont A, Couté A, Duval C, Couloux A, Rousseau F, Bernard C (2010) The contribution of sub-Saharan African strains to the phylogeny of cyanobacteria: focusing on the Nostocaceae (Nostocales, Cyanobacteria). J Phycol 46:564–579 Trautman DA, Borowitzka MA (1999) Distribution of the epiphytic organisms on Posidonia australis and P. sinuosa, two seagrasses with differing leaf morphology. Mar Ecol Prog Ser 179:215–229 Turner BL, Baxter R, Ellwood NTW, Whitton BA (2003a) Seasonal phosphatase activities of mosses from Upper Teesdale, northern England. J Bryol 25:189–200 Turner BL, Baxter R, Whitton BA (2003b) Nitrogen and phosphorus in soil solutions and drainage streams in Upper Teesdale, northern England: implications of organic compounds for biological nutrient limitation. Sci Total Environ 314–316C:153–170 Tymowski RG, Duthie HC (2000) Life strategy and phosphorus relations of the cyanobacterium Gloeotrichia echinulata in an oligotrophic Precambrian Shield lake. Arch Hydrobiol 148:321–332 Uku J, Björk M, Bergman B, Diez B (2007) Characterization and comparison of prokaryotic epiphytes associated with three East African seagrasses. J Phycol 43:768–779 Vaidya HM (1989) Comparative studies on Calothrix isolates from Nepalese rice-fields. M.Sc. thesis, University of Durham, Durham Van Geel B, de Lange L, Wiegers J (1984) Reconstruction and interpretation of the local vegetational succession of a Lateglacial deposit from Usselo (The Netherlands), based on the analysis of micro- and macrofossils. Acta Bot Neerl 33(4):535–546 Van Wagoner RM, Drummond AK, Jeffrey LC, Wright JLC (2007) Biogenetic diversity of cyanobacterial metabolites. Adv Appl Microbiol 61:89–217 Vuorio K, Meili M, Sarvala J (2009) Natural isotopic composition of carbon(d13C) correklates with colony size in the planktonic cyanobacterium Gloeotrichia echinulata. Limnol Oceanogr 54(3):925–929 Webber EE (1967) Bluegreen algae from a Massachusetts salt marsh. Bull Torrey Bot Club 94:99–106 Weckesser J, Hofmann K, Jürgens EJ, Whitton BA, Raffelberger B (1988) Isolation and chemical analysis of the sheaths of the filamentous cyanobacteria Calothrix parietina and C. scopulorum. J Gen Microbiol 134:629–634 West W (1912) Fresh-water algae. In: Clare Island survey. Proc R Irish Acad XXXI(Pt 16):1–62 Whitton BA (1987) The biology of Rivulariaceae. In: Fay P, Van Baalen C (eds) The Cyanobacteria – a comprehensive review. Elsevier, Amsterdam, pp 513–534, 543 pp Whitton BA (1988) Hairs in eukaryotic algae. In: Round FE (ed) Algae and the aquatic environment contributions in honour of J.W.G.Lund. Biopress, Bristol, pp 446–460, 460 pp Whitton BA (1989) Genus I. Calothrix Agardh 1824. In: Staley JT, Bryant MP, Pfennig N, Holt JG (eds) Bergey’s manual of systematic bacteriology, vol 3, 1st edn. Williams & Wilkins, Baltimore, pp 1791–1794 Whitton BA (2007) The blue-green algae of Clare Island, Co. Mayo, Ireland. In: New Survey of Clare Island, vol 6: the freshwater and terrestrial algae. Royal Irish Academy, Dublin, pp +141–178, 254 pp Whitton BA (2008) Cyanobacterial diversity in relation to the environment. In: Evangelista V, Barsanti L, Frassanito AM, Gualtieri P, 591 Passarelli V (eds) Algal toxins, nature, occurrence, effect and detection. Springer, Dordrecht, pp 17–35, 399 pp Whitton BA (2009) Phosphorus status and the naming of freshwater algae. Algas 41:20–22 Whitton BA (2011) Phylum cyanobacteria (Cyanophyta). In: John DM, Whiton BA, Brook AJ (eds) The freshwater algal flora of the British Isles, 2nd edn. Cambridge University Press, Cambridge, pp 31–158, 878 pp Whitton BA, Neal C (2010) Organic phosphate in UK rivers and its relevance to algal and bryophyte surveys. Int J Limnol 47:1–8 Whitton BA, Potts M (eds) (2000) The ecology of cyanobacteria. Their diversity in time and space. Springer, Dordrecht, 669 pp Whitton BA, Gale NL, Wixson BG (1981) Chemistry and plant ecology of zinc-rich wastes dominated by blue-green algae. Hydrobiologia 83:331–341 Whitton BA, Grainger SLJ, Harris N (1986) Blue heterocysts in the Rivulariaceae. Br Phycol J 21(3):338–339 Whitton BA, Aziz A, Francis P, Rother JA, Simon JW, Tahmida ZN (1988a) Ecology of deepwater rice-fields in Bangladesh. 1. Physical and chemical environment. Hydrobiologia 169:3–22 Whitton BA, Aziz A, Rother JA (1988b) Ecology of deepwater ricefields in Bangladesh. 3. Associated algae and macrophytes. Hydrobiologia 169:31–42 Whitton BA, Grainger SLJ, Hawley GRW, Simon JW (1991) Cell-bound and extracellular phosphatase activities of cyanobacterial isolates. Microb Ecol 21:85–98 Whitton BA, Yelloly JM, Christmas M, Hernández I (1998) Surface phosphatase activity of benthic algal communities in a stream with highly variable ambient phosphate concentrations. Verh Int Verein Limnol 26:967–972 Whitton BA, Al-Shehri AH, Ellwood NTW, Turner BL (2005) Ecological aspects of phosphatase activity in cyanobacteria, eukaryotic algae and bryophytes. In: Turner BL, Frossard E, Baldwin DS (eds) Organic phosphorus in the environment. Commonwealth Agricultural Bureau, Wallingford, pp 205–241, 399 pp Wickstrom CE, Castenholz W (1985) Dynamics of cyanobacterial and ostracod interactions in an Oregon hot spring. Ecology 66:1024–1241 Wilkinson CR, Sammarco PW, Trott LA (1985) Seasonal and fish grazing effects on rates of nitrogen fixation on coral reefs (Great Barrier Reef, Australia). In: Proceedings of the 5th international coral reef congress, vol 4l, Tahiti, pp 61–65 Wilmotte A, Herdman M (2001) Phylogenetic relationships among the cyanobacteria based on 16S rRNA sequences. In: Boone DR, Castenholz RW, Garrity GM (eds) Bergey’s manual of systematic bacteriology, vol 1, 2nd edn. Springer, New York, pp 487–493, 721 pp Wood P (1984) Structural and physiological studies on the effects of mineral deficiencies on the Rivulariaceae (Cyanobacteria), vols I and II. Ph D thesis (CNAA), Sunderland Polytechnic (now Sunderland University), UK Wood P, Peat A, Whitton BA (1986) Influence of phosphorus status on fine structure of the cyanobacterium (blue-green alga) Calothrix parietina. Cytobios 47:89–99 Yelloly JM, Whitton BA (1996) Seasonal changes in ambient phosphate and phosphatase activities of the cyanobacterium Rivularia atra in intertidal pools at Tyne Sands, Scotland. Hydrobiologia 325:201–212 Zarmouh MM (2010) Antibacterial activity of Rivularia species and Oscillatoria salina collected from coastal region of Misurata, Libya. J Arab Soc Med Res 5(2):159–163 Zehnder A (1963) Kulturversuch mit Gloeotrichia echinulata (J. E. Smith) P. Richter. Schweiz Z Hydrol 25:65–83 Zhao-Liang X (1984) Investigation on the procaryotic microfossils from Gaoyuzhuang Formation, Jixian, North China. Acta Bot Sin 26(216–222):312–319