Protist Protist, Vol. 152, 265–300, December 2001 © Urban & Fischer Verlag http://www.urbanfischer.de/journals/protist ORIGINAL PAPER Molecular Phylogeny and Taxonomic Revision of Chlamydomonas (Chlorophyta). I. Emendation of Chlamydomonas Ehrenberg and Chloromonas Gobi, and Description of Oogamochlamys gen. nov. and Lobochlamys gen. nov.1 Thomas Pröscholda,2, Birger Marina, Uwe Gert Schlösserb, and Michael Melkoniana a Botanisches Institut, Lehrstuhl I, Universität zu Köln, Gyrhofstr. 15, D-50931 Köln, Germany Abteilung Experimentelle Phykologie und Sammlung von Algenkulturen, Albrecht-von-Haller-Institut für Pflanzenwissenschaften der Universität Göttingen, Untere Karspüle 2, D-37073 Göttingen, Germany b Submitted March 27, 2001; Accepted September 18, 2001 Monitoring Editor: Robert A. Andersen The genus Chlamydomonas (including Chloromonas) is one of the largest green algal genera comprising more than 600 species. To initiate a comprehensive analysis of the phylogeny and systematics of the genus, we determined nuclear-encoded SSU rRNA sequences from 32 strains of Chlamydomonas, Chloromonas and Chlorogonium with emphasis on oogamous taxa and related strains, and incorporated these into global molecular phylogenetic analyses of 132 strains of Chlorophyceae. In addition, we studied the morphology and reproduction of oogamous and related strains by light microscopy. We recognize and designate 18 monophyletic lineages (clades) within the Chlorophyceae, 11 of which are confined to the CW (basal bodies displaced clockwise) subgroup. The majority of clades recognized within the Chlorophyceae do not correspond to any of the traditional classification systems, which are still largely based on the organization level. Strains assigned to Chlamydomonas and Chloromonas were found in seven different clades confirming the polyphyly of the two genera as presently conceived. To initiate the taxonomic revision of Chlamydomonas, C. reinhardtii is proposed as the conserved type of the genus. In consequence, species in clades other than the clade containing C. reinhardtii must be transferred to other genera, a process initiated in this contribution. The oogamous strains studied represent a monophyletic lineage, which is described as Oogamochlamys gen. nov. comprising three species (O. gigantea, O. zimbabwiensis and O. ettlii spec. nov.). The sister clade to Oogamochlamys consists of isogamous strains characterized by chloroplasts with incisions and is described as Lobochlamys gen. nov. with two species (L. culleus and L. segnis). Another clade is characterized by asteroid or perforated, parietal chloroplasts and contains the type species of Chloromonas (C. reticulata). Thus, the polyphyletic Chloromonas (traditionally defined as “Chlamydomonas without pyrenoids”) can be legitimized as a monophyletic genus by restriction to this clade and is here emended on the basis of chloroplast characters (the clade contains strains with or without pyrenoids thus rejecting the character “absence of pyrenoids”). 1 In memory of Hanuš Ettl (1931–1997), promoter of the systematics of the genus Chlamydomonas Corresponding author; fax 49 221 470 5181 e-mail Thomas.Proeschold@uni-koeln.de 2 1434-4610/01/152/04-265 $ 15.00/0 266 T. Pröschold et al. Introduction Traditionally the genus Chlamydomonas Ehrenberg comprises all biflagellate green algae in which the two flagella are of equal length and emerge in close proximity to each other, and which contain a single chloroplast with pyrenoid(s) and a cell wall (Ettl 1976). However, these characters are also found in other taxa, and are not synapomorphic for Chlamydomonas. Ettl (1976) therefore considered the genus Chlamydomonas as an artificial taxon, which includes all green flagellates with the characteristics mentioned above that lack other, specialized characters. The circumscription of species in this genus is problematic because many original descriptions were based on light microscope observations of a few specimens from natural samples without consideration of the variability of morphological characters within the population or of the life history. The genetic identity of many of the Chlamydomonas spp. described remains to be determined. Species of Chlamydomonas have been classified into ten subgenera or sections, characterized by the number and position of pyrenoids within the chloroplast (Ettl 1976, 1983a; Pascher 1927). Based on a single character, i.e. the lack of pyrenoids, several species were transferred into a new genus, Chloromonas Gobi emend. Wille (Ettl 1970; Gerloff 1940; Gobi 1899/1900; Wille 1903); however, this classification has also been considered as artificial (Ettl 1970, 1979, 1983a). Chlamydomonas (including Chloromonas) is one of the largest green algal genera with more than 600 species described (Ettl 1976, 1983a). Fortunately, more than 300 strains including approximately 100 authentic strains (“type strains”) are available in culture and amenable to experimental, including molecular analyses. Schlösser (1976, 1984), for example, examined 65 strains of Chlamydomonas for the presence and cross-reactivity of sporangial cell wall-lysing enzymes (v-lysins according to Adair and Snell 1990) and established 15 vegetative lysin enzyme groups (VLE-groups) in the genus. Ettl and Schlösser (1992) showed, based on the example of Chlamydomonas applanata Pringsheim, that morphological and reproductive characters of a group of strains within an autolysin group correlated well, leading the authors to conclude that strains of a VLE-group “can be regarded as clones of one species”. These conclusions were corroborated by SSU rRNA sequence comparisons of two strains of the C. applanata group, which were identical (Gordon et al. 1995). The polyphyly of the genus Chlamydomonas as well as the genetic distinctness of some VLE-groups was established by Buchheim et al. (1990, 1996, 1997a, 1997b; see also Nakayama et al. 1996) based on phylogenetic analyses of partial or complete SSU rRNA sequences. However, a comprehensive phylogenetic analysis of the genus Chlamydomonas including related genera of the Chlorophyceae is still lacking. In the present communication, we initiate such an analysis by addressing the phylogenetic history of the oogamous species of Chlamydomonas (C. gigantea Dill and C. zimbabwiensis Heimke et Starr) and related isogamous species within the framework of related taxa. Nuclear-encoded SSU rRNA sequences of 32 strains of Chlamydomonas, Chloromonas and Chlorogonium Ehrenberg were determined and incorporated into global phylogenetic analyses of 132 strains of Chlorophyceae, which revealed a well-supported molecular phylogeny of the class. Based on these results and in combination with detailed investigations of morphology and reproduction we emend the genera Chlamydomonas (with C. reinhardtii Dangeard as conserved type) and Chloromonas, and propose the new genera Oogamochlamys, and Lobochlamys. Results Taxon Sampling and Alignment The initial goal of this study was to present a comprehensive analysis of the phylogeny and taxonomy of the oogamous Chlamydomonas species (C. gigantea Dill and C. zimbabwiensis Heimke et Starr). These species have been placed in the section Pleiochloris based on the presence of more than two pyrenoids in the chloroplast (Ettl 1976). This section comprises 36 species (Ettl 1983a; Heimke and Starr 1979), but only four of the species described are available in culture (C. carrizoensis Deason et Bold, C. rubrifilum Korshikov and the two oogamous species; strains of C. carrizoensis and C. rubrifilum have also been designated as C. chlorococcoides Ettl et Schwarz and Chloromonas clathrata Korshikov, respectively; Schlösser 1994). Therefore, we examined all available strains of C. chlorococcoides (3) and Chloromonas clathrata (1). Since both C. carrizoensis and C. rubrifilum have elongated chloroplast perforations/incisions, we also added other Chlamydomonas strains, with chloroplast incisions (including the authentic strains of C. actinochloris Deason et Bold and two varieties of C. augustae Skuja). Furthermore, initial molecular phylogenetic studies by Buchheim et al. (1996) suggested that C. gigantea could be related to C. culleus Ettl (VLE-group 9) and C. segnis Ettl (VLE- Molecular Phylogeny and Revision of Chlamydomonas group 10; see Schlösser 1976). We therefore included several strains of these two v-lysin groups in the analysis (9 of 17 strains available in culture). C. gigantea is characterized by multiple (> 4) contractile vacuoles, a unique situation within Chlamydomonas; however, multiple contractile vacuoles are also known in Haematococcus and in seven species of Chlorogonium (Ettl 1983a; Nozaki et al. 1998). To test the phylogenetic significance of this character, we determinated SSU rRNA sequences of three species of Chlorogonium with multiple contractile vacuoles (Chlorogonium euchlorum Ehrenberg, Chlorogonium capillatum Nozaki, M.M. Watanabe et Aizawa and Chlorogonium elongatum (Dangeard) Dangeard). Nuclear-encoded SSU rRNA sequences of 32 strains of the genera Chlamydomonas, Chloromonas and Chlorogonium (see Table 5 for a list of strains) determined in this study were integrated into two different alignments: (1) a large data set comprising a total of 152 taxa of the Chlorophyceae, Trebouxiophyceae, and Ulvophyceae with four prasinophyte taxa as outgroups; and (2) a smaller data set (34 taxa) including all strains of Chlamydomonas and Chloromonas for which sequences were obtained during this study. Molecular Phylogenetic Analyses Phylogenies inferred from these data sets are summarized in Figures 1 and 2; tree topologies shown were obtained by distance (Fig. 1) or maximum likelihood analyses (Fig. 2); bootstrap percentage values > 50% are indicated for distance (for two different models of DNA substitution: HKY85 and TrN+I+G; see Methods) and maximum parsimony (Fig. 1), and for maximum likelihood, distance and parsimony (Fig. 2) analyses. In the figures, the designation of species of Chlamydomonas and Chloromonas follows the taxonomic rearrangements (including the new genera Oogamochlamys and Lobochlamys) described at the end of the results. The global analysis revealed the Ulvophyceae, Trebouxiophyceae, and Chlorophyceae as separate lineages within the Chlorophyta in accordance with earlier studies (reviews by Chapman et al. 1998; Friedl 1997; McCourt 1995; Melkonian and Surek 1995), but only the Ulvophyceae and Chlorophyceae were supported by moderate bootstrap values in all analysis (Fig. 1). Within the 132 strains of Chlorophyceae examined, 18 independent monophyletic clades comprising 122 strains were recognized (these are indicated in Figure 1 and provisionally named after a representative taxon). Ten strains (marked with an asterisk in Figure 1) could not be al- 267 located to one of the 18 clades, mostly due to their individual long branches and incomplete taxon sampling. Eleven clades belong to the so-called CW group of Chlorophyceae (basal bodies clockw ise displaced; Deason et al. 1991; Nakayama et al. 1996), a well-supported monophyletic lineage (arrow with encircled CW at the node uniting this group in Fig. 1). All Chlamydomonas species (including Chloromonas) belong to this lineage, but are distributed over seven different clades (“ Stephanosphaera” -, “ Polytoma”-, “Monadina”-, “Moewusii”-, “ Oogamochlamys”-, “ Chloromonas”-, and “Reinhardtii”-clade) together with other genera. The polyphyly of the genera Chlamydomonas and Chloromonas was thus clearly revealed (see also previous studies by Buchheim et al. 1996, 1997b; Hepperle et al. 1998; Nakayama et al. 1996). The phylogenetic relationships (i.e. the branching order) between the 11 clades of the CW-group could not be resolved in this study. However, phylogenetic relationships of strains within the clades were often resolved (Fig. 1). Here, we focus on two clades, which contain most sequences determined in this study including the two oogamous species, namely the “ Oogamochlamys”-clade and the “ Chloromonas”-clade. The two oogamous species of Chlamydomonas (C. gigantea and C. zimbabwiensis) are sisters in a monophyletic lineage (described below as Oogamochlamys gen. nov.; Fig 1). They form a larger clade (termed “ Oogamochlamys”-clade) together with two other Chlamydomonas species, i.e. C. culleus and C. segnis (previously placed in the sections Chlorogoniella and Euchlamydomonas respectively), which also form a well-supported lineage (described below as Lobochlamys gen. nov.; Fig. 1). 17 of the 32 strains investigated belong to the “ Oogamochlamys”-clade, which was well supported in all types of analysis (Fig. 1). The other two species of the Pleiochloris section of Chlamydomonas investigated in this study, C. carrizoensis (strain SAG 46.72) and C. rubrifilum (strain SAG 3.85), do not belong to the “ Oogamochlamys”-clade, but are part of another clade (the “ Chloromonas”-clade; described below as Chloromonas in emended form) together with other strains of Chloromonas and Chlamydomonas (the latter again previously placed in different sections). The “ Chloromonas”-clade contains 11 of the 32 strains investigated and was also well supported in all types of analyses (Fig. 1). The global analysis thus not only revealed the polyphyly of the genera Chlamydomonas and Chloromonas, but also the inadequacy of the concept of the subdivision of the genus Chlamydomonas into sections, as both the “ Oogamochlamys”- and the “ Chloromonas”clade contain taxa from different sections. The Molecular Phylogeny and Revision of Chlamydomonas global analysis also showed that the character state “multiple contractile vacuoles” is present in three different clades (“ Stephanosphaera”-, “ Chlorogonium”-, and “ Oogamochlamys”-clade), two of which (“ Oogamochlamys”-clade and “ Stephanosphaera”clade) also contain taxa in which motile cells have only two contractile vacuoles, strongly suggesting that this character has evolved independently in the respective clades. To further evaluate the significance of the phylogenetic inferences presented in the global analysis (Fig. 1), user-defined trees were generated and compared with the topology shown in Figure 1 (tree 1; “best tree”; Table 1) using Kishino-Hasegawa-tests by calculating differences in the log-likelihood values [maximum likelihood (TrN+I+G model; see Methods)] or in the tree lengths (maximum parsimony). In all but two analyses the user-defined trees were significantly “worse” at the P< 0.05 level than the “best tree” (Table 1). Specifying a tree length of zero for the common branch (i.e., collapsing this branch) of the Chlorophyceae (tree 2) or the common branch uniting the CW-group (tree 3) resulted in significantly “worse” phylogenies (Table 1), supporting the robustness of these lineages. However, when the common branch uniting the “ Neochloris”-, “ Hydrodictyon”-, “ Scenedesmus”- and “ Bracteacoccus”-clades [these taxa are sometimes designated as the DO-group (orientation of basal bodies in motile cells directly opposite; Watanabe and Floyd 1989)] was collapsed, the resulting tree topology (tree 4) using the maximum likelihood model was not significantly worse than the best tree (Table 1). This suggests that the phylogenetic relationships between these four clades and the other clades (“ Chaetophora”-, “ Oedogonium” and “ Chaetopeltis”-clade), which are not positioned in the CW-lineage, are not resolved and require further study. User-defined trees were also generated to test the hypotheses that the section Pleiochloris of the genus Chlamydomonas is monophyletic (tree 269 5), or that taxa with multiple contractile vacuoles are monophyletic (trees 6–11); both hypotheses were rejected (Table 1). Finally, when the common branch of a clade (“ Dunaliella”-clade sensu Buchheim et al. 1997a and Nakayama et al. 1996; in Fig. 1, this clade consists of Dysmorphococcus globosus, Chlamydomonas tetragama, and four clades, i.e. the “ Dunaliella”-,“ Polytoma”-,“ Stephanosphaera”- and “ Chlorogonium”-clades), which in our analysis received moderate bootstrap support only in the maximum parsimony analysis (Fig. 1), was collapsed (tree 12), the resulting tree using the maximum likelihood model was not significantly worse than the best tree (Table 1). This results suggests that the phylogenetic relationships between the “Dunaliella” -, “Polytoma” -, “Stephanosphaera” -, and “Chlorogonium” -clades remain currently unresolved requiring further study. In addition to the global phylogenetic analyses (Fig. 1), we also performed refined phylogenetic analyses with a reduced taxon sampling comprising only the strains of the “ Oogamochlamys”- and “ Chloromonas”-clades (34 strains; Fig. 2) but using an increased number of positions (1702 compared to 1642 positions) in the analyses. Because the phylogenetic relationships between the two clades remain unknown and no outgroup could be identified to either of them (see Fig. 1), unrooted analyses were performed. Figure 2 shows the maximum likelihood tree (using the TrNef+I+G model of evolution; see Methods) with bootstrap values referring to maximum likelihood, distance (using the TrNef+I+G model) and maximum parsimony analyses. Within the “ Oogamochlamys”-clade a monophyletic lineage consisting of the oogamous species (C. gigantea and C. zimbabwiensis) was strongly supported in all analyses (Fig. 2). Whereas the SSU rRNAs of the three strains of C. zimbabwiensis are almost identical (all three originated from Southern Africa, see Table 5), there is more genetic diversity among the other five strains. An early diverging Figure 1. Molecular phylogeny of the Chlorophyta based on nuclear-encoded SSU rRNA sequence comparisons. Four prasinophyte flagellates were used as an outgroup for 152 taxa of the Ulvophyceae, Trebouxiophyceae and Chlorophyceae. The strains for which SSU rRNA sequences were determined in this study are indicated in bold (for strain designations see Tables 4 and 5). The phylogenetic tree shown was inferred by the neighbor-joining method based on distances of 1,642 aligned positions calculated by the HKY85 model (Hasegawa et al. 1985). Bootstrap percentage values (> 50% ) are given for neighbor-joining [using the HKY85 model (bold, above branch) and the model of Tamura and Nei (TrN+I+G; not bold, above branch) with estimated shape parameter of the gamma distribution (G = 0.62) and proportion of invariable sites (I = 0.46), which was determined as the best model for the data set by Modeltest 3.04 (Posada and Crandall 1998)] and unweighted maximum parsimony (italics, below branch). The monophyletic clades (18) are provisionally named after a representative taxon. Ten strains (with an asterisk) cannot yet be placed within a clade of the Chlorophyceae. CW (encircled): clade of Chlorophyceae, in which motile cells have clockwise-displaced basal bodies. 270 T. Pröschold et al. Figure 2. Molecular phylogeny of the “ Oogamochlamys”- and “ Chloromonas”-clades inferred by SSU rRNA sequence comparisons using 1,702 aligned positions. The unrooted tree shown resulted from a maximum likelihood analysis [using the model of Tamura and Nei (1993) with estimated gamma shape (G = 0.63), proportions of invariable sites (I = 0.72) and equal base frequencies, TrNef+I+G, calculated as the best model by Modeltest 3.04, (Posada and Crandall 1998)] of 34 taxa; bootstrap percentage values (> 50% ) were determined for maximum likelihood (using TrNef+I+G; bold) neighbor-joining (using TrNef+I+G; bold italics) and unweighted maximum parsimony (not bold) methods. SSU rRNA sequences determined in this study are indicated in bold, authentic strains are marked by an asterisk. The strain numbers and the original designations of the strains are indicated as well as the newly proposed combinations (see also Tables 4 and 5; Cd. = Chlamydomonas, Cm. = Chloromonas). 271 Molecular Phylogeny and Revision of Chlamydomonas Table 1. Comparisons of the neighbor-joining tree in Figure 1 with user-defined trees by Kishino-Hasegawa-tests using maximum likelihood (ML) and maximum parsimony (MP) methods. ML (TrN+I+G)a Tree topologyb Diff-lnLc 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. (22322.4)f 27.4 ± 11.3 18.3 ± 8.7 9.5 ± 6.3 98.8 ± 21.8 74.8 ± 15.1 92.3 ± 17.2 52.6 ± 13.9 80.4 ± 16.5 165.6 ± 27.2 130.3 ± 25.1 4.5 ± 3.9 (best tree; Fig.1) (collap.: Chloroph.) (collap.: “CW”) (collap.: “DO”) (Ccar/Crub; Oochlam) (Cg/Hp; Ogig) (Hz/St; Ogig) (Cg/Hp; Oochl/Lobo) (Hz/St; Oochl/Lobo) (Cg/Hp/Hz/St;Ogig) (Cg/Hp/Hz/St; (collap.: Dun/Buchheim) MP Pd – 0.0153* 0.0357* 0.1350 <0.0001* <0.0001* <0.0001* 0.0002* <0.0001* <0.0001* <0.0001* 0.2491 Diff. lengthe (3789)g 7 7 5 35 27 29 16 21 57 37 6 P – 0.0081* 0.0081* 0.0253* <0.0001* <0.0001* <0.0001* 0.0025* 0.0010* <0.0001* <0.0001* 0.0143* a Maximum likelihood (ML) using the model after Tamura and Nei (1993) with estimated gamma shape and proportions of invariable sites (TrN+I+G). b Tree no.1 (best tree) is identical with Fig. 1; modifications in user-defined trees no. 2–12 are indicated by following abbreviations: Tree 2: collapsed branch of the Chlorophyceae (Chloroph.) Tree 3: collapsed branch of the “CW”-group (“CW”) Tree 4: collapsed branch of the “DO”-group (“DO”) Tree 5: sections Pleiochloris monophyletic (SAG 46.72 Chloromonas carrizoensis [Ccar] and SAG 3.85 C. rubrifilum [Crub] as sister to Oogamochlamys [Oochlam]) Tree 6: “ Chlorogonium”-clade (Cg/Hp) as sister to Oogamochlamys gigantea (Ogig) Tree 7: Haematococcus zimbabwiensis and Stephanosphaera (Hz/St) as sister to Oogamochlamys gigantea (Ogig) Tree 8: “ Chlorogonium”-clade (Cg/Hp) to the basis of the “ Oogamochlamys”-clade (Oochl/Lobo) Tree 9: Haematococcus zimbabwiensis and Stephanosphaera (Hz/St) to the basis of the “ Oogamochlamys”-clade (Oochl/Lobo) Tree 10: “ Chlorogonium”-clade (Cg/Hp) and Haematococcus zimbabwiensis and Stephanosphaera (Hz/St) as sister to Oogamochlamys gigantea (Ogig) Tree 11: “ Chlorogonium”-clade (Cg/Hp) and Haematococcus zimbabwiensis and Stephanosphaera (Hz/St) to the basis of the “ Oogamochlamys”-clade (Oochl/Lobo) Tree 12: collapsed common branch of the “ Stephanosphaera”-, “ Chlorogonium”-, “ Polytoma” and “ Dunaliella”clade (including the single taxa: Chlamydomonas tetragama and Dysmorphococcus globosus; Dun/Buchheim). c Difference in -log-likelihood between the best tree (Fig. 1) and the user-defined tree. d Probability of obtaining a more extreme T-value under the null hypothesis of no difference between the two trees (two-tailed test). e Difference in tree length between the best tree (Fig. 1) and the user-defined tree. f -log-likelihood of the best tree (Fig. 1). g Length of the optimal tree in the maximum parsimony analysis. *User defined tree significantly worse than the best tree at P < 0.05. strain (C. gigantea UTEX 2218; described below as a new species) isolated from Zimbabwe was placed in an intermediate position between the C. zimbabwiensis strains and the other C. gigantea strains. The later diverging four C. gigantea strains formed a well-supported monophyletic lineage (Fig. 2). Among those, strain SAG 21.72, originating from California, emerged first (Fig. 2). This strain was a sister to the other three strains of C. gigantea, which have almost identical SSU rRNA sequences (although they originated from two continents; Table 5). A sister lineage to the oogamous Chlamydomonas species contained nine strains (here assigned to the species C. culleus and C. segnis; Fig. 2). This lineage was also well supported in all analyses (Fig. 2) but in contrast to the C. gigantea/ C. zimbabwiensis lineage did not split up into well-sup- 272 T. Pröschold et al. ported monophyletic groups (Fig. 2). We tentatively identify two groups of strains in this lineage: three early diverging (in a paraphyletic topology) strains designated C. culleus [including the authentic strain (SAG 17.73; neotype)], and a later diverging, moderately-supported lineage of six strains, five of which are authentic strains of previously described Chlamydomonas species (C. segnis, C. fimbriata Ettl, C. gymnogama Deason, C. pallidostigmatica King, C. sajao Lewin). All six strains are here (and in Schlösser 1994) designated as C. segnis (and described within the new genus Lobochlamys as L. segnis, see below). Four of the six strains have almost identical SSU rRNA sequences (Fig. 2), although they originated from Europe, North America and China (Table 5). The “ Chloromonas”-clade consists of three wellsupported lineages (14 strains) plus three strains, which are characterized by long branches and therefore could not be placed with confidence in one of the three lineages (Fig. 2). One lineage consists of seven closely related strains, which we now designate as Chloromonas reticulata (Goroschankin) Wille (type species of Chloromonas; see below). It contains two strains without pyrenoids previously listed under the genus Chloromonas [C. rosae (Ettl H. et O.) Ettl (SAG 51.72; authentic strain) and C. clathrata (SAG 29.83)], three strains previously designated as Chlamydomonas chlorococcoides including the authentic strain (SAG 15.82), and two strains previously placed in other species of Chlamydomonas [C. augustae (SAG 26.86) and C. macrostellata Lund (SAG 72.81)]. Another lineage of four strains is genetically more diverse: two strains referred to Chlamydomonas actinochloris (including the authentic strain, SAG 1.72) are monophyletic and sister to a strain designated as Chlamydomonas bipapillata Ettl (SAG 11-47). These three strains in turn are a sister to the fourth strain designated as Chlamydomonas radiata Deason et Bold (UTEX 966: the authentic strain of this species), which is the earliest-branching taxon in this lineage (Fig. 2). The third lineage within the “ Chloromonas”-clade is represented by three strains, which we designate as Chlamydomonas augustae (it contains the authentic strains of two varieties of the species: SAG 5.73 and SAG 9.87). The phylogenetic relationships between the three lineages within the “ Chloromonas”-clade could not be resolved in this study (Fig. 2). The three remaining strains of the “ Chloromonas”-clade, which could not be significantly positioned, have long individual branches: two species of the section Pleiochloris, Chlamydomonas carrizoensis (SAG 46.72) and Chlamydomonas rubrifilum (SAG 3.85; designated as Chloromonas clathrata in Schlösser 1994) as well as Chloromonas serbinowii Wille (UTEX 492). Bootstrap support for a sister group relationship between C. rubrifilum and C. serbinowii is moderate in all analyses (Fig. 2). To further evaluate the phylogenetic position of several strains in the two clades, user-defined trees were generated and compared with the topology shown in Figure 2 (tree 1; “best tree”; Table 2). Kishino-Hasegawa-tests were performed by calculating differences in the log-likelihood values [maximum likelihood (TrNef+I+G model; for details see Methods)] or in the tree lengths (maximum parsimony). The intermediate position of strain UTEX 2218 between C. zimbabwiensis and C. gigantea (see Fig. 2) was again evident: when UTEX 2218 was placed as sister to C. zimbabwiensis (tree 3; Table 2) this topology, in both maximum likelihood and maximum parsimony methods, was not significantly “worse” at the P < 0.05 level than the “best tree”. However, when UTEX 2218 was placed as sister to the three closely related C. gigantea strains SAG 44.91, SAG 9.84 and UTEX 1753, disrupting the C. gigantea lineage (Fig. 2), the resulting topology was significantly “worse” in both analyses (tree 2; Table 2). This suggests that UTEX 2218 is really intermediate and cannot be placed with confidence in either C. gigantea or C. zimbabwiensis (for morphological observations supporting these conclusions, see below and Table 3). Both, C. gigantea and C. zimbabwiensis are robust lineages in accordance with the high bootstrap values obtained for these lineages in all phylogenetic analyses (Fig. 2): for example, when a tree length of zero was specified for the common branch uniting the four C. gigantea strains, the resulting topology was significantly “worse” than that of the “best tree” (tree 4; Table 2). The stability of the C. segnis lineage within the C. segnis/ C. culleus subclade (= Lobochlamys, see below) was probed by placing the earliest branching strain SAG 17.72 (authentic strain of “ C. fimbriata”) of this lineage (Fig. 2) either as sister to the four closely related strains SAG 1.79, SAG 2.75, SAG 9.83 and SAG 50.84 of C. segnis (tree 5; Table 2) or as sister to the two C. culleus strains SAG 17.73 and SAG 18.72 (tree 6; Table 2). Although this strain is more likely positioned with C. segnis than with C. culleus, none of the specified topologies is significantly “worse” than that of the “best tree” depicted in Figure 2, indicating that the C. segnis lineage is not very robust (in accordance with the only moderate bootstrap support for this lineage in all analyses; Fig. 2). The branching order of the strains within C. segnis remains unresolved as well. The three strains assigned to C. culleus (SAG 18.72, SAG 64.72, SAG 273 Molecular Phylogeny and Revision of Chlamydomonas Table 2. Comparisons of the maximum likelihood tree in Figure 2 with user-defined trees by Kishino-Hasegawatests using maximum likelihood (ML) and maximum parsimony (MP) methods. ML (TrN+I+G)a b c Tree topology Diff-lnL 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. (4840.7)f 28.6 ± 11.6 3.2 ± 3.3 28.6 ± 11.6 2.1 ± 5.9 12.8 ± 8.3 0.9 ± 2.3 1.2 ± 2.0 71.0 ± 13.8 3.8 ± 3.4 10.3 ± 7.6 14.9 ± 7.3 7.3 ± 7.6 2.1 ± 2.6 (Best tree; Fig.2) (UTEX 2218; Ogig minus SAG 21.72) (UTEX 2218; Ozim) (collap.: Ogig) (SAG 17.72; Lseg minus SAG 52.72) (SAG 17.72; Lcul) (SAG 64.72; Lcul) (SAG 64.72; Lseg) (Cmrub; Ogig) (Cmrub; Cmaug) (Cmrub; Cmret) (Cmserb; Cmret) (Cmaug; Cmact) (Cmcar; Cmret) MP P d – 0.0137* 0.3291 0.0137* 0.7215 0.1253 0.6797 0.5410 <0.0001* 0.2611 0.1778 0.0208* 0.1706 0.2096 Diff. lengthe (420)g 8 1 8 0 3 0 1 38 4 9 12 7 1 P – 0.0046* 0.6549 0.0046* 1.0000 0.0833 1.0000 0.3175 <0.0001* 0.2060 0.0389* 0.0073* 0.0896 0.6549 a Maximum likelihood (ML) using the model after Tamura and Nei (1993) with estimated gamma shape, proportions of invariable sites and equal base frequencies (TrNef+I+G). b Tree no.1 (best tree) is identical with Fig. 2; modifications in user-defined trees no. 2–14 are indicated by following abbreviations: Tree 2: Oogamochlamys ettlii (strain UTEX 2218) as sister to the three Oogamochlamys gigantea strains SAG 44.91, SAG 9.84 and UTEX 1753 (Ogig minus SAG 21.72), Tree 3: Oogamochlamys ettlii (strain UTEX 2218) as sister to Oogamochlamys zimbabwiensis (Ozim), Tree 4: collapsed branch of Oogamochlamys gigantea (strains SAG 44.91, SAG 9.84, UTEX 1753 and SAG 21.72; Ogig), Tree 5: Lobochlamys segnis (strain SAG 17.72) as sister to the four Lobochlamys segnis strains SAG 1.79, SAG 2.75, SAG 9.83 and SAG 50.84 (Lseg minus SAG 52.72), Tree 6: Lobochlamys segnis (strain SAG 17.72) as sister to the two Lobochlamys culleus strains SAG 17.73 and SAG 18.72 (Lcul), Tree 7: Lobochlamys culleus (strain SAG 64.72) as sister to the two Lobochlamys culleus strains SAG 17.73 and SAG 18.72 (Lcul), Tree 8: Lobochlamys culleus (strain SAG 64.72) as sister to the five Lobochlamys segnis strains SAG 52.72, SAG 1.79, SAG 2.75, SAG 9.83 and SAG 50.84 (Lseg), Tree 9: Chloromonas rubrifilum (strain SAG 3.85; Cmrub) as sister to Oogamochlamys gigantea (strains SAG 44.91, SAG 9.84, UTEX 1753 and SAG 21.72; Ogig), Tree 10: Chloromonas rubrifilum (strain SAG 3.85; Cmrub) as sister to Chloromonas augustae (strains SAG 5.73, SAG 13.89 and SAG 9.87; Cmaug), Tree 11: Chloromonas rubrifilum (strain SAG 3.85; Cmrub) as sister to Chloromonas reticulata (Cmret), Tree 12: Chloromonas serbinowii (Cmserb) as sister to Chloromonas reticulata (Cmret), Tree 13: Chloromonas augustae (Cmaug) as sister to the group Chloromonas actinochloris (strains UTEX 578, SAG 1.72), C. asteroidea (strain SAG 11-47), and C.radiata (strain UTEX 966) (Cmact), Tree 14: Chloromonas carrizoensis (Cmcar) as sister to Chloromonas reticulata (Cmret). c Difference in -log-likelihood between the best tree (Fig. 2) and the user-defined tree. d Probability of obtaining a more extreme T-value under the null hypothesis of no difference between the two trees (one-tailed test). e Difference in tree length between the best tree (Fig. 2) and the user-defined tree. f -log-likelihood of the best tree (Fig. 2). g Length of the optimal tree in the maximum parsimony analysis. *User defined tree significantly worse than the best tree at P < 0.05. 17.73) formed a paraphyletic assemblage in the topology shown in Figure 2. To test for monophyly of the three strains, strain SAG 64.72 was placed as a sister to the other two strains (tree 7; Table 2). In the maximum parsimony method, this topology was as good (zero difference in tree length) as that of the “best tree” and in the maximum likelihood method it was also not significantly “worse” suggesting that monophyly of the three strains cannot be rejected. However, when strain SAG 64.72 was placed as sis- 274 T. Pröschold et al. ter to five strains of C. segnis excluding “C. fimbriata” (tree 8; Table 2) the resulting topology was again not significantly “worse” than that of the “best tree”. In conclusion, phylogenetic analyses based on SSU rRNA sequence comparisons cannot be used to distinguish C. culleus from C. segnis unequivocally (but see morphological observations below). Within the „ Chloromonas“-clade, Chlamydomonas rubrifilum formed a moderately-supported clade with Chloromonas serbinowii (Fig. 2); however, both sequences are characterized by long branches. To evaluate whether a long branch attraction (LBA) artifact is responsible for their apparent relatedness, we tested the significance of alternative topologies for C. rubrifilum: placing C. rubrifilum as a sister of the C. gigantea lineage (disrupting the “ Oogamochlamys”-clade and probing the monophyly of the section Pleiochloris; tree 9; Table 2) is significantly “worse” than the topology of the “best tree” (a difference in tree lengths of 38 steps in the maximum parsimony method!). In contrast, when C. rubrifilum was placed as sister to C. augustae (tree 10; Table 2) or as sister to C. reticulata (tree 11; Table 2), both resulting topologies were not signifcantly “worse” than that of the “best tree” (with exception of the C. rubrifilum/ C. reticulata lineage in the maximum parsimony method; Table 2). Placing C. serbinowii as sister to the C. reticulata lineage (thus probing the traditional Chloromonas concept, i.e., absence of pyrenoids) resulted in a significantly “worse” topology than that of the “best tree” (tree 12; Table 2). In conclusion, the C. serbinowii/ C. rubrifilum lineage is not robust (despite its moderate bootstrap support). The longer individual branch of C. rubrifilum (compared to C. serbinowii) makes this strain apparently more vulnerable to alternative topologies. To test for the monophyly of the Chloromonas strains with asteroid chloroplasts (tree 13; see below) and the position of Chlamydomonas carrizoensis within the „ Chloromonas“-clade, we placed Chlamydomonas augustae as sister to the lineage consisting of C. actinochloris, C. asteroidea and C. radiata (tree 13) and C. carrizoensis as sister to C. reticulata (tree 14). Both topologies were not significantly “worse” than that of the “best tree”. Light Microscopy of Oogamous Chlamydomonas Species and Comparison with Species Diagnoses The morphology and life history of C. gigantea (five strains; Table 5) and C. zimbabwiensis (three strains; Table 5) have been investigated under standard culture conditions (see Methods) by light microscopy. The results of the morphological observations on vegetative cells (young zoospores and mature cells at the end of the light period; see Methods) are summarized in Table 3. In general, strains fall into three groups represented by C. gigantea (4 strains), C. zimbabwiensis (three strains), and a single strain (UTEX 2218) also previously designated as “ C. gigantea”. Mature cells of C. gigantea and C. zimbabwiensis differ from each other mainly in the number of contractile vacuoles (two in C. zimbabwiensis and up to 200 in C. gigantea; Figs 5a,m, 6m–p), cell size, number of pyrenoids, and in the surface structure of the parietal, massive chloroplast (Table 3). In all strains, the chloroplast surface displays ridges/ grooves (shallow incisions), which run mostly parallel to the cell axis (Figs 5, 6a–d). In C. gigantea the chloroplast surface is coarsely ribbed (Figs 5b, 6a–b), whereas in C. zimbabwiensis the ridges/ grooves are much finer (Fig. 5n). Interestingly, strain UTEX 2218 occupies an intermediate position between C. gigantea and C. zimbabwiensis as the cell size and chloroplast surface structure (Figs 5g–h, 6d) are similar to those of C. zimbabwiensis, whereas the numbers of pyrenoids (3–8) and contractile vacuoles (> 20; Fig. 6 q–t) are more similar to those of C. gigantea (Table 3). Asexual reproduction in all strains investigated starts with the resorption of the flagella. In all strains, mostly four zoospores (rarely two or eight) were formed (Fig. 3c, 4c). In C. gigantea (including strain UTEX 2218) the first division step is a true transverse division (perpendicular to the cell axis) without prior rotation of the protoplast (Fig. 3b), whereas in C. zimbabwiensis the protoplast first rotates by 90º within the parental cell wall before division commences (“false transverse division” according to Ettl 1988; Fig. 4b); the second division plane in both species is perpendicular to the first one. Pyrenoids and eyespots did not disappear during the cell divisions. One zoospore received the parental eyespot whereas the others formed new eyespots. Liberation of zoospores occurred after swelling of the parental cell wall and partial dissolution at the flagellar pole (not shown). Sexual reproduction in all strains is a homothallic proterandric oogamy (not observed in strains SAG 21.72, and SAG 9.84). Cells of C. gigantea (except strain UTEX 2218) differentiating into male gametangia became yellowish-green, accumulated starch grains, dissolved the pyrenoids and divided simultaneously into four protoplasts, followed by successive divisions into 32 or 64 (rarely 128) spermatozoids (Fig. 3d,e). Spermatozoids were released by partial autolysis of the parental cell wall at the flagellar pole (Fig. 3f). Cells of C. zimbabwiensis and strain Molecular Phylogeny and Revision of Chlamydomonas 275 Table 3. Comparison of structural characteristics in the genus Oogamochlamys (= oogamous Chlamydomonas species). O. giganteaa SAG 44.91 Figs 3, 5 a-f, 6 a, b, m-p O. ettlii UTEX 2218 Figs 5g-l, 6 c, d, q-t O. zimbabwiensisb SAG 45.91 Figs 4, 5 m-r nearly spherical broadly roundedcylindrical-oviform nearly spherical broadly roundedcylindrical-oviform nearly spherical broadly roundedcylindrical-oviform Cell size in zoospores in mature vegetative cells 10–15 µm 30–50 × 25–35 µm 10–12 µm 16–27 × 18–22 µm 10–12 µm 15–22 × 15–20 µm Cell wall thin thin thin Cell wall papilla small, rounded small, rounded broad, rounded, two humped or absent Flagellar length about as long as the cell about as long as the cell about as long as the cell Chloroplast shape parietal, massive, parietal, massive, surface with coarse ridges surface with fine ridges parietal, massive, surface with very fine ridges 4–8 10–16(–20) discontinuous, matrix multipartite 3–6 3–8 discontinuous, matrix multipartite 2–4 2–6 discontinuous, matrix multipartite pale red, elliptic to narrowly elongate anterior pale red, elliptic to narrowly elongate anterior pale red, elliptic to narrowly elongate or punctiform anterior Contractile vacuoles 2 apical + many (>20), distributed over the whole cell surface 2 apical + many (>20), distributed over the whole cell surface 2 apical Nucleus position central or slightly anterior central or slightly anterior central or slightly anterior Character Cell shape in zoospores in mature vegetative cells Pyrenoids number in zoospores in mature vegetative cells shape Eyespot shape position in the chloroplast a The Oogamochlamys gigantea strains SAG 21.72 and the authentic strains of Chlamydomonas megalis (SAG 9.84) and Chlamydomonas capensis (UTEX 1753) reveal the same morphological characters. b The Oogamochlamys zimbabwiensis strains UTEX 2213 and UTEX 2214 reveal the same morphological characters. UTEX 2218 differentiated into male gametangia in the same manner as in C. gigantea except that cell divisions were successive from the beginning (as in asexual reproduction) and not initially simultaneous as in C. gigantea (Fig. 4d–f). 16 spermatozoids were formed, which were released by partial autolysis of the parental cell wall at the flagellar pole as a mucilaginous package of cells (Fig. 4g). The dimensions of the teardrop-shaped, naked spermatozoids were 6–10 × 4–6 µm (C. gigantea) and 4–6 × 3–5 µm (C. zimbabwiensis and UTEX 2218). Spermatozoids of all oogamous strains were biflagellate (flagellar length 1.5 times the cell length) with two contractile vacuoles, a pale-green small chloroplast with an eyespot but without a pyrenoid and nucleolus (Fig. 3f). Oogenesis was triggered by the presence of spermatozoids (therefore proterandric); a spermatozoid attached near the cell wall papilla of a vegetative cell and induced its protoplast to round up and, after partial lysis of the wall at the flagellar pole, escape from the cell wall (Figs 3g,h, 4h). The free protoplasts, in most cases without flagella and enclosed by mucilage, represent functional egg cells. Figure 4. Stages of development in Oogamochlamys zimbabwiensis, strain SAG 45.91. a vegetative cell, b divided cell after a 90° rotation of the protoplast, c sporangium with four zoospores, d immature male gametangium, e mature male gametangium, f male gametangium shortly before the liberation of spermatozoids, g liberation of spermatozoids, h hatching of the egg, i zygote. Bar: 20 µm. Figure 3. Light microscopy of developmental stages in Oogamochlamys gigantea (strain SAG 44.91). a vegetative cell, b first transverse division during asexual reproduction, c sporangium with four zoospores, d immature male gametangium, e mature male gametangium, f liberation of spermatozoids, g spermatozoid attached to a vegetative cell, h hatching of the egg, i-n maturing zygotes, i with developing secondary cell wall, j optical transverse section, k surface view, l primary and developing secondary cell wall, m mature zygote with thickened secondary cell wall and accumulation of starch grains and carotenoids, n nontransparent secondary zygote cell wall. Bar: 20 µm. 278 T. Pröschold et al. Figure 5. Schematic drawings of vegetative cells of Oogamochlamys spp . a–f Oogamochlamys gigantea, strain SAG 44.91, g–l Oogamochlamys ettlii spec. nov., strain UTEX 2218, m–r Oogamochlamys zimbabwiensis, strain SAG 45.91, a, g, m optical longitudinal section, b, h, n surface view of the chloroplast with ridges and many contractile vacuoles scattered between the ridges (only in b and h), c, i, o cell in apical view, d, j, p bipartite pyrenoid from young cells, e, k, q multipartite pyrenoids from mature cells, f, l, r multipartite pyrenoid broken apart in mechanically pressed cells. Bar: 20 µm. In C. gigantea (except strain UTEX 2218), the zygote formed a smooth primary wall after fertilization (Fig. 3i), which was replaced by a thick secondary zygote wall ornamented with regular projections (Fig. 3j,k). Zygotes contained many starch grains (Fig. 3i,j) and gradually changed their color from green to brownish-red (Fig. 3l,m). Zygote size and cell wall ornamentation varied within each strain (Fig. 3l-n). In C. zimbabwiensis and in strain UTEX 2218, a smooth primary zygote wall is formed after fertilization (not shown), which was replaced by a thin secondary zygote wall without ornamentations (Fig. 4i). Figure 6. a–l Light micrographs (Nomarski interference contrast) of chloroplast structures and contractile vacuole function in vegetative cells of Oogamochlamys, Lobochlamys, and Chloromonas spp. c, e, g, i, k optical section, a, b, d, f, h, j, l surface view of the chloroplasts, a, b chloroplast with coarse ridges/grooves in Oogamochlamys gigantea (in b, a zoospore), strain SAG 44.91 (fixed cells), c, d chloroplast with fine ridges/grooves in Oogamochlamys ettlii spec. nov., strain UTEX 2218 (live cells), e, f chloroplast with incisions in Lobochlamys segnis, strain SAG 52.72 (live cells), g, h chloroplast with incisions leading to formation of lobes in Lobochlamys culleus, strain SAG 17.73 (live cells), i, j asteroid chloroplast in Chloromonas actinochloris, strain SAG 1.72 (fixed cells), k, l parietal chloroplast with perforations in Chloromonas reticulata, strain SAG 29.83 (fixed cells), m–p Four contractile vacuoles in diastole (black arrows) and systole (empty arrows) in Oogamochlamys gigantea, strain SAG 44.91 (live cell), q-t Two contractile vacuoles in diastole (black arrows) and systole (empty arrows) in O. ettlii spec. nov., strain UTEX 2218 (live cell). Bar: 20 µm. 280 T. Pröschold et al. Zygotes stored starch grains but, in contrast to C. gigantea, remained green. Zygote germination was not observed and could not be induced in any oogamous strain. Several strains, which are collectively assigned here to C. gigantea, are authentic strains of other Chlamydomonas species (SAG 9.84: C. megalis Bischoff et Bold 1963; UTEX 1753: C. capensis Pocock 1962) or have previously been determined as C. pseudogigantea Korshikov (SAG 21.72; Schlösser 1982) or C. gigantea (SAG 44.91; Schlösser 1994). However, the light microscopical examination of these strains under defined culture conditions did not reveal any significant differences (see above); therefore, we follow a proposal by Ettl and Gärtner (1995) and recognize only one species, i.e. C. gigantea (for an emended diagnosis of C. gigantea, see “Taxonomic Revisions and Diagnoses”). For three strains of C. zimbabwiensis no nomenclatural changes at the species level are required; nevertheless, an emended diagnosis is provided below since the original diagnosis (Heimke and Starr 1979) did not include chloroplast characters. Light Microscopy of Isogamous Chlamydomonas and Chloromonas Strains Studied and Comparison with Species Diagnoses 26 isogamous strains of Chlamydomonas and Chloromonas related to the oogamous species (see Fig. 2) were studied by light microscopy under standard culture conditions (see Methods), their desig- Table 4. Nomenclatural changes of Chlamydomonas and Chloromonas strains studied after taxonomic revisions or new determinations. Strain SAG 1.72 SAG 34.72 SAG 5.73 SAG 13.89 SAG 9.87 SAG 11-47 SAG 46.72 SAG 47.72 SAG 26.86 SAG 29.83 SAG 72.81 SAG 15.82 SAG 16.82 SAG 12.96 SAG 51.72 SAG 3.85 SAG 17.73 SAG 18.72 SAG 64.72 SAG 52.72 SAG 17.72 SAG 2.75 SAG 9.83 SAG 50.84 SAG 1.79 SAG 44.91 SAG 9.84 UTEX 1753 SAG 21.72 UTEX 2218 SAG 45.91 UTEX 2213 UTEX 2214 (= UTEX 965) (= UTEX 578) (= UTEX 968) (= UTEX 966) (= UTEX 1969) (= UTEX 1970) (= UTEX 1337) (= UTEX 1057) (= UTEX 1059) (= UTEX 1343) (= UTEX 1349) (= UTEX 1638) (= UTEX 1905) (= UTEX 2277) (= UTEX 1919) (= UTEX 1492) (= SAG 22.98) (= UTEX 848) (= SAG 24.98) Original designation in Schlösser (1982, 1994), and Starr and Zeikus (1993) New designation Chlamydomonas actinochloris Chlamydomonas mutabilis Chlamydomonas augustae Chlamydomonas augustae Chlamydomonas augustae Chlamydomonas bipapillata Chlamydomonas chlorococcoides Chlamydomonas radiata Chlamydomonas augustae Chloromonas clathrata Chlamydomonas macrostellata Chlamydomonas chlorococcoides Chlamydomonas chlorococcoides Chlamydomonas chlorococcoides Chloromonas rosae Chloromonas clathrata Chlamydomonas culleus Chlamydomonas culleus Chlamydomonas culleus Chlamydomonas segnis Chlamydomonas fimbriata Chlamydomonas gymnogama Chlamydomonas pallidostigmatica Chlamydomonas sajao Chlamydomonas segnis Chlamydomonas gigantea Chlamydomonas gigantea Chlamydomonas gigantea Chlamydomonas gigantea Chlamydomonas gigantea Chlamydomonas zimbabwiensis Chlamydomonas zimbabwiensis Chlamydomonas zimbabwiensis Chloromonas actinochloris Chloromonas actinochloris Chloromonas augustae Chloromonas augustae Chloromonas augustae Chloromonas asteroidea Chloromonas carrizoensis Chloromonas radiata Chloromonas reticulata Chloromonas reticulata Chloromonas reticulata Chloromonas reticulata Chloromonas reticulata Chloromonas reticulata Chloromonas reticulata Chloromonas rubrifilum Lobochlamys culleus Lobochlamys culleus Lobochlamys culleus Lobochlamys segnis Lobochlamys segnis Lobochlamys segnis Lobochlamys segnis Lobochlamys segnis Lobochlamys segnis Oogamochlamys gigantea Oogamochlamys gigantea Oogamochlamys gigantea Oogamochlamys gigantea Oogamochlamys ettlii Oogamochlamys zimbabwiensis Oogamochlamys zimbabwiensis Oogamochlamys zimbabwiensis Molecular Phylogeny and Revision of Chlamydomonas 281 nation was reevaluated (Ettl 1983a), and some nomenclatural changes are proposed (Table 4). Authentic strains were compared with the original diagnoses and descriptions. Isogamous Strains Assigned to the “Oogamochlamys”-clade All six strains (including the authentic strain of C. segnis, SAG 52.72) assigned here to C. segnis contain a cup-shaped chloroplast with a thickened basal part extending anteriorly (Figs 6e, 7a). The chloroplast surface exhibits narrow and deep incisions running mostly parallel to the cell axis (Figs 6f, 7b). Incisions in the posterior part of the chloroplast do not perforate the plastid but can extend toward the single pyrenoid which is located in a basal position (Figs 6e, 7a). These characteristics correspond well with the original description of C. segnis (Ettl 1965a), especially with the additional description (Ettl 1965b). Two strains previously designated as C. gymnogama (SAG 2.75) and C. pallidostigmatica (SAG 9.83) within the section Chlamydella (with lateral position of the single pyrenoid; Ettl 1983a) revealed laterally positioned pyrenoids only in freshly liberated zoospores, whereas mature vegetative cells contained a centrally placed, basal pyrenoid. The chloroplast shape (tubular with a cross-bridge containing the pyrenoid; characteristic for the section Pseudagloë) described for C. sajao (Lewin 1984) and the distinctly asteroid chloroplast described for C. fimbriata (Ettl 1965a) were not observed in the two authentic strains SAG 50.84 (C. sajao) and SAG 17.72 (C. fimbriata), instead both strains conformed with the diagnosis of C. segnis. Isogamous sexual reproduction was induced in three strains (SAG 52.72, SAG 2.75 and SAG 17.72) and corresponds with the previous description given by Deason (1967) for C. gymnogama (authentic strain SAG 2.75). Based on these observations the six strains investigated here were assigned to C. segnis requiring nomenclatural changes for C. gymnogama, C. pallidostigmatica, C. sajao, and C. fimbriata (Table 4), and an emended diagnosis, which contains a description of the isogamous sexual reproduction. The three strains assigned to C. culleus (including strain SAG 17.73 designated as “authentic strain (neotype)” in Schlösser 1994) are characterized by a cylindrical/parietal chloroplast with prominent incisions, running predominantly parallel to the cell axis but often also branching transversely. The incisions traverse the chloroplast, leading to the formation of irregular chloroplast lobes (Figs 6g,h, 7c,d). The pyrenoid is located laterally in median/submedian position. For the three strains no nomenclatural changes are required. Figure 7. Schematic drawings of vegetative cells of Lobochlamys and Chloromonas spp. with emphasis on chloroplast structures. a, c, e, g, i, k, m, o optical longitudinal section, b, d, f, h, j, l, n, p surface view of the chloroplast, a, b Lobochlamys segnis, strain SAG 52.72, c, d Lobochlamys culleus, strain SAG 17.73, e, f Chloromonas actinochloris, strain SAG 1.72, g, h Chloromonas augustae, strain SAG 5.73, i, j Chloromonas reticulata, strain SAG 29.83, k, l Chloromonas reticulata, strain SAG 15.82, m, n Chloromonas carrizoensis, strain SAG 46.72, o, p Chloromonas rubrifilum, strain SAG 3.85. Bar: 20 µm. 282 T. Pröschold et al. Table 5. List of taxa examined in this investigation. The strains are available in the SAG (Sammlung von Algenkulturen Göttingen; Schlösser 1994) and UTEX (Culture Collection at the University of Texas at Austin; Starr and Zeikus 1993) algal culture collections. The SSU rRNA sequences are available in the EMBL, GenBank and DDBJ sequence databases under the accession numbers given. Strain species SAG 12-2a Chlorogonium capillatum SAG 12-2e Chlorogonium capillatum SAG 12-2d Chlorogonium euchlorum UTEX 2571 Chlorogonium elongatum SAG 1.72 Chloromonas actinochloris SAG 5.73 Chloromonas augustae SAG 9.87 Chloromonas augustae SAG 13.89 Chloromonas augustae SAG 46.72 Chloromonas carrizoensis SAG 15.82 Chloromonas reticulata SAG 16.82 Chloromonas reticulata SAG 29.83 Chloromonas reticulata SAG 26.86 Chloromonas reticulata SAG 12.96 Chloromonas reticulata SAG 3.85 Chloromonas rubrifilum Original designation and remarks authentic strain of Chlamydomonas actinochloris authentic strain of Chlamydomonas augustae var. eupapillata authentic strain of Chlamydomonas augustae var. ellipsoidea isolated as Chlamydomonas bullosa; designated as Chlamydomonas augustae authentic strain of Chlamydomonas carrizoensis (= Chlamydomonas pyrenoidosa); designated as Chlamydomonas chlorococcoides authentic strain of Chlamydomonas chlorococcoides designated as Chlamydomonas chlorococcoides isolated as Chlamydomonas yellowstonensis; designated as Chlamydomonas augustae isolated as Chlamydomonas nivalis; designated as Chlamydomonas augustae designated as Chlamydomonas chlorococcoides designated as Chloromonas clathrata Isolator and origin Accession number E.G. Pringsheim, 1916, Germany, Berlin-Dahlem; basin with ducks AJ410441 E.G. Pringsheim, 1936, Czechoslovakia, pool near Prague AJ410442 E.G. Pringsheim, 1951, South Africa, Cape Flats; dry mud from De Klip Vlei AJ410443 M. Wood, 1916, USA, TX, Austin; Biology pond AJ410444 T.R. Deason, 1958, TX, USA; soil from Caldwell Co. AJ410445 H. Ettl, before 1973, Czechoslovakia; mucous covers in swamp AJ410452 S. Watanabe, 1979, Indonesia; soil from paddyfield near Lampong/ Sumatra R.W. Butcher, England; salt marsh pool AJ410453 AJ410454 T.R. Deason, 1958, USA, TX; soil from Caldwell Co. AJ410446 K. Schwarz, 1975, Yugoslavia, Dalmatia; soil from Isle Lavsa AJ410449 K. Schwarz, 1975, Yugoslavia, Dalmatia; soil from Isle Lavsa AJ410450 E. Sutton, 1968, USA, OR, Cascade Range; snow from Todd Lake AJ410448 E. Sutton, 1968, USA, OR, Cascade Mts.; on snow AJ410447 T. Reiner, 1993, Germany; soil from Göttinger Wald near Mackenröder Spitze M. Melkonian, 1976, Germany, Pevestorf/Dannenberg; pond under ice AJ410451 AJ410455 Molecular Phylogeny and Revision of Chlamydomonas 283 Table 5. (Continued). Strain species Original designation and remarks Isolator and origin Accession number SAG 17.73 Lobochlamys culleus SAG 18.72 Lobochlamys culleus SAG 64.72 Lobochlamys culleus SAG 52.72 Lobochlamys segnis SAG 17.72 Lobochlamys segnis SAG 2.75 Lobochlamys segnis SAG 1.79 Lobochlamys segnis SAG 9.83 Lobochlamys segnis SAG 50.84 Lobochlamys segnis UTEX 2218 Oogamochlamys ettlii SAG 44.91 Oogamochlamys gigantea SAG 9.84 Oogamochlamys gigantea authentic strain (neotype) of Chlamydomonas culleus F. Hindák, 1969, Czechoslovakia; pond in Nordmähren AJ410461 isolated as Chlamydomonas frankii; designated as Chlamydomonas culleus isolated as Chlamydomonas elliptica var. britannica; designated as Chlamydomonas culleus authentic strain of Chlamydomonas segnis G.M. Smith, 1946, USA, FL; field near Maxville AJ410462 G.M. Smith, 1940, Nicaragua; soil from hog wallow, Bluefields AJ410463 AJ410456 authentic strain of Chlamydomonas fimbriata H. Ettl, 1960, Czechoslovakia; soil from beech forest near Winkelsdorf, Altvater-Gebirge F. Hindák, 1962, Czechoslovakia; Doksy pond Máchovo jezero authentic strain of Chlamydomonas gymnogama T.R. Deason, 1965, USA, AL; soil from Dauphin Island AJ410458 designated as Chlamydomonas segnis S.S. Badour, 1969, Canada, Manitoba, Delta Marsh AJ410457 authentic strain of Chlamydomonas pallidostigmatica authentic strain of Chlamydomonas sajao J.M. King, USA, TX, McDadeCaldwell; roadside pool AJ410459 R.A. Lewin, 1980, China; duckwood growing near Guangzhou, Canton AJ410460 designated as Chlamydomonas gigantea J.W. Heimke and R.C.Starr, 1978, Zimbabwe, Rhodesia; soil AJ410469 designated as Chlamydomonas gigantea U.G. Schlösser, 1969, South Africa, Cape Town; soil from Cape Flats “De Klip” H.W. Bischoff and H.C. Bold, 1960, USA, TX, Llano Co.; soil from Enchanted Rock AJ410465 UTEX 1753 Oogamochlamys gigantea SAG 21.72 Oogamochlamys gigantea SAG 45.91 Oogamochlamys zimbabwiensis UTEX 2213 Oogamochlamys zimbabwiensis UTEX 2214 Oogamochlamys zimbabwiensis authentic strain of Chlamydomonas megalis; designated as Chlamydomonas gigantea authentic strain of Chlamydomonas capensis; designated as Chlamydomonas gigantea designated as Chlamydomonas gigantea AJ410464 AJ410466 R.C. Starr, 1968, Zimbabwe, Rhodesia; soil AJ410467 J. Stein, 1951, USA, CA, Lemoncove; pond soil AJ410468 AJ410470 authentic strain of Chlamydomonas zimbabwiensis U.G. Schlösser, 1969, South Africa, Cape Town; soil from Cape Flats “De Klip” J.W. Heimke and R.C.Starr, 1979, Zimbabwe, Rhodesia; soil designated as Chlamydomonas zimbabwiensis J.W. Heimke and R.C.Starr, 1979, Zimbabwe, Rhodesia; soil AJ410472 designated as Chlamydomonas zimbabwiensis AJ410471 284 T. Pröschold et al. Strains Assigned to the “Chloromonas”-clade All seven strains assigned here to Chloromonas reticulata contain cup-shaped, parietal and massive chloroplasts extending almost to the flagellar poles with distinct, irregularly branched perforations (these traverse the chloroplast, but do not result in the formation of chloroplast lobes; Figs 6k,l, 7i,j). In the five strains with a pyrenoid, the perforations radiate from the laterally positioned pyrenoid mostly parallel to the cell axis; in the two strains without a pyrenoid (SAG 29.83 and SAG 51.72), perforations also run parallel to the cell axis, but do not radiate from a distinct center (Figs 6k–l, 7i–j). The morphological characteristics (except for the presence/absence of a pyrenoid) of all strains correspond well with the description given by Ettl (1983a) for Chloromonas reticulata. Determination of the strains with pyrenoids (using the key provided by Ettl 1983a) was not possible, because strains could not be assigned to one of the two sections Chlamydella or Chlorogoniella. However, all strains could be identified (as C. reticulata) using the Chloromonas key in Ettl (1983a) when the character “presence of pyrenoid” was ignored. The authentic strain of Chloromonas rosae (SAG 51.72) was also identified as C. reticulata because the subdivision of the chloroplast into separate lobes was not observed in culture. Based on these observations, it is proposed to assign all seven strains to the species Chloromonas reticulata and to emend its diagnosis (Table 4 and “Taxonomic Revisions and Diagnoses”, see below). Chlamydomonas actinochloris (two strains: authentic culture SAG 1.72 and strain SAG 34.72, which was misidentified as C. mutabilis; Schlösser 1982), C. bipapillata, C. radiata, and C. augustae exhibit an asteroid chloroplast with a central pyrenoid (incisions to the central pyrenoid; Figs 6i,j, 7e–h). Chlamydomonas carrizoensis (Fig. 7m,n), Chlamydomonas rubrifilum (Fig. 7o,p; both with several pyrenoids) and Chloromonas serbinowii (without pyrenoid) have a cup-shaped, parietal chloroplast with distinct perforations. In all strains except for Chlamydomonas carrizoensis and Chlamydomonas rubrifilum, the observed characteristics correspond to the original diagnoses. The authentic strain of C. carrizoensis (SAG 46.72; mislabeled as C. chlorococcoides in Schlösser 1994 and as C. deasonii in Ettl 1976, 1983a) contains two-four pyrenoids, which are, however, only visible in mature vegetative cells (C. chlorococcoides has only one pyrenoid). In the original diagnosis (Deason and Bold 1960 – under the illegitime designation C. pyrenoidosa; see correction in Chantanachat and Bold 1962), al- though referring to the same strain, the presence of two-five pyrenoids is considered as a constant character and chloroplast perforations are not mentioned. These discrepancies require an emended diagnosis of C. carrizoensis based on observations made during this study (see below). Strain SAG 3.85 (mislabeled as Chloromonas clathrata in Schlösser 1994) could be unambiguously identified as C. rubrifilum (Ettl 1983a) and is characterized by a parietal chloroplast with prominent perforations and 7–11 pyrenoids (Table 4). However, pyrenoids are only visible after staining with azocarmine (see Methods). Emendation of the Genera Chlamydomonas and Chloromonas, and Proposal of the New Genera Oogamochlamys and Lobochlamys Molecular phylogenetic analyses of strains of Chlamydomonas and Chloromonas clearly revealed that both genera are not monophyletic and instead are distributed over several independent lineages (Figs 1, 2). Therefore, the classification of these genera cannot be sufficiently clarified by rearrangements at the species level alone (for taxonomy of species see above). Instead, a comprehensive taxonomic revision at the generic level is required, including new definitions (emendations) of both genera based on monophyletic lineages. Emendation of Chlamydomonas Since species of Chlamydomonas form several independent lineages and are found in six major clades of the Chlorophyceae (+ C. tetragama; see Fig. 1), it is first necessary to define a single lineage that includes the type of the genus, Chlamydomonas pulvisculus (O. F. Müller) Ehrenberg (Monas pulvisculus O. F. Müller). Unfortunately, the identity of the type is not determinable because of the lack of detail in Müller’s drawing. Ehrenberg’s description of the species (Ehrenberg 1838) is equally ambiguous while his illustrations can be tentatively referred to C. reinhardtii, C. noctigama or species of Chlamydomonas in the section Amphichloris. This situation led Ettl (1976) to propose C. reinhardtii as neotype of Chlamydomonas; but a neotype may be chosen only when all material on which the name of the taxon was based is missing. Uncertainly of identity of a holotype species (in this case C. pulvisculus) may be overcome only by designating a species of known identity as conserved type. This procedure requires a formal proposal (Pröschold and Silva, in preparation). The genus is emended to include at most only few species in the “Reinhardtii” -clade (see “Taxonomic Revisions and Diagnoses”). Molecular Phylogeny and Revision of Chlamydomonas Oogamochlamys and Lobochlamys As a consequence of the emendation of Chlamydomonas, those species, which are not monophyletic with C. reinhardtii, must be assigned to other genera most of which will have to be newly described – a process, which is initiated in this contribution for the oogamous species and their isogamous relatives. For oogamous species, the new genus Oogamochlamys is proposed (containing at present three species: O. gigantea, O. zimbabwiensis and O. ettlii spec. nov.). The sister clade of Oogamochlamys which contains isogamous species with incised chloroplasts is here designated as Lobochlamys (with two species: L. culleus, L. segnis; see “Taxonomic Revisions and Diagnoses”). Emendation of Chloromonas Although the defining taxonomic character of the traditional genus Chloromonas (absence of a pyrenoid) must be rejected (strains without pyrenoids are not monophyletic to the exclusion of pyrenoid-bearing strains), Chloromonas can nevertheless be assigned to a single clade (i.e., “ Chloromonas”-clade), which contains the type species of the genus, i.e. Chloromonas reticulata (Ettl 1970; Wille 1903). Because this clade contains strains with one to several pyrenoids as well as strains without pyrenoids, Chloromonas as recognized here is redefined on the basis of chloroplast characters (asteroid chloroplasts or cup-shaped, parietal chloroplasts with irregularly branched perforations) and an emended diagnosis is provided (see “Taxonomic Revisions and Diagnoses”). Chloromonas currently contains eight species: C. actinochloris, C. augustae, C. asteroidea, C. carrizoensis, C. radiata, C. reticulata, C. rubrifilum, and C. serbinowii. Taxonomic Revisions and Diagnoses Chlamydomonas Ehrenberg 1833 emend. Pröschold, Marin, Schlösser et Melkonian Emended Diagnosis: Unicellular Chlorophyta sensu Bremer (1985) with two flagella in clockwise basal body orientation; cell wall thin, with or without papilla, the single chloroplast cup-shaped with thickened basal part, and without incisions, lobes or perforations; one discontinuous pyrenoid in slightly basal position of the chloroplast; eyespot in anterior position; with two apical contractile vacuoles; nucleus in slightly anterior position. Asexual reproduction by zoospores, rotation of the protoplast by 90º before first cell division (“false transverse division”); total lysis of the sporangial wall be- 285 fore release of zoospores through by means of a vegetative lysin enzyme (VLE-group 1, Schlösser 1976). Sexual reproduction by isogamy, homo- or heterothallic; zygotes not ornamented, red. The genus is supported by molecular phylogenetic analysis using SSU rRNA sequence comparisons. Proposed conserved type species: Chlamydomonas reinhardtii Dangeard 1888, Ann. Sci. Nat. Bot. Sér. 7e, 7; pp. 130–136, pl. XII, fig. 29-39 (descr. et ic. prima, iconotypus). Epitype: The strain UTEX 90 permanently preserved in a metabolically inactive state (cryopreservation in liquid nitrogen) in the Culture Collection of Algae at the University of Texas at Austin (UTEX), Department of Botany, The University of Texas at Austin, Austin, Texas 78713-7640, USA. Chloromonas Gobi 1899/1900 emend. Pröschold, Marin, Schlösser et Melkonian Emended Diagnosis: Unicellular Chlorophyta sensu Bremer (1985) with two flagella in clockwise basal body orientation; cell wall thin with or without papilla, the single chloroplast asteroid (with a central pyrenoid) or cup-shaped, parietal and massive with unbranched or branched perforations; one or more discontinuous pyrenoids present or pyrenoid absent; eyespot in anterior position; with two apical contractile vacuoles; nucleus in central or slightly anterior position. Asexual reproduction by zoospores, rotation of the protoplast by 90º before first cell division (“false transverse division”). Sexual reproduction by isogamy, homo- or heterothallic; zygotes ornamented or not ornamented, green. The genus is supported by molecular phylogenetic analysis using SSU rRNA sequence comparisons. Type species: Chloromonas reticulata (Goroschankin 1891) Wille 1903 emend. Pröschold, Marin, Schlösser et Melkonian. Chloromonas reticulata (Goroschankin 1891) Wille 1903 emend. Pröschold, Marin, Schlösser et Melkonian. Wille 1903, Nyt Mag. Naturvidensk. 41, p. 150, pl. IV, fig. 27 (descr. et ic. prima, iconotypus). Basionym: Chlamydomonas reticulata Goroschankin 1891, Bull. Soc. Imp. Nat. Moscou, N. S. 5, p. 124–128, pl. I, fig. 1–9 (descr. et ic. prima, iconotypus). Synonyms: Chlamydomonas clathrata (Korshikov) Pascher 1927, Die Süßwasserflora Deutschlands, Österreichs und der Schweiz Vol. 4, p. 305, fig. 274b (descr. et ic. prima, iconotypus). Chloromonas 286 T. Pröschold et al. clathrata Korshikov in Pascher 1927, Die Süßwasserflora Deutschlands, Österreichs und der Schweiz Vol. 4, p. 305, fig. 274b. Chloromonas rosae (Ettl H. et O.) Ettl 1963, Nova Hedwigia 6, p. 397, Chlamydomonas rosae Ettl H. et O. 1959, Arch. Protistenkd. 104, p. 99, fig. 28–30 (descr. et ic. prima, iconotypus), authentic culture: SAG 51.72 = UTEX 1337. Chloromonas rosae (Ettl H. et O.) Ettl 1963 var. polychloris Ettl 1963, Nova Hedwigia 6, p. 396, fig. 1 (descr. et ic. prima, iconotypus). Chlamydomonas chlorococcoides Ettl et Schwarz 1981, Plant Syst. Evol. 138, p. 119, fig. 1, 2 (descr. et ic. prima, iconotypus), authentic culture: SAG 15.82. Emended Diagnosis: Cells 11–22 × 5–15 µm, ellipsoid, ellipsoid-cylindrical, cell wall thin, papilla broad, rounded or absent, with two flagella about as long as the cell. Chloroplast cup-shaped, parietal and massive with irregular, branched perforations, one discontinuous pyrenoid in lateral position or pyrenoid absent; eyespot pale red, elliptic to narrowly elongate in anterior position; two apical contractile vacuoles, nucleus in central or slightly basal position. Asexual reproduction by four or eight zoospores, rotation of the protoplast by 90º before first cell division (“false transverse division”). Sexual reproduction by isogamy; zygotes ornamented, green. Epitype: The strain UTEX 1970 permanently preserved in a metabolically inactive state (cryopreservation in liquid nitrogen) in the Culture Collection of Algae at the University of Texas at Austin (UTEX), Department of Botany, The University of Texas at Austin, Austin, Texas 78713-7640, USA. culture: SAG 5.73. Chlamydomonas augustae Skuja 1943 var. planctonica Ettl 1976, Beih. Nova Hedwigia 49, p. 404, pl. 64, fig. 1 (descr. et ic. prima, iconotypus). Chlamydomonas augustae Skuja 1949 var. ellipsoidea S. Watanabe 1983, Arch. Protistenkd. 127, p. 231, fig. 5 (descr. et ic. prima, iconotypus), authentic culture: SAG 9.87. Epitype: The strain SAG 5.73 permanently preserved in a metabolically inactive state (cryopreservation in liquid nitrogen) in the Culture Collection of Algae at the University of Texas at Austin (UTEX), Department of Botany, The University of Texas at Austin, Austin, Texas 78713-7640, USA. Chloromonas actinochloris (Deason et Bold 1960) Pröschold, Marin, Schlösser et Melkonian comb. nov. Chloromonas carrizoensis (Deason et Bold in Chantanachat et Bold 1962) Pröschold, Marin, Schlösser et Melkonian comb. nov. Basionym: Chlamydomonas actinochloris Deason et Bold 1960, Phycol. Stud. I, p. 13, fig. 1–3 (descr. et ic. prima, iconotypus). Epitype: The strain UTEX 965 permanently preserved in a metabolically inactive state (cryopreservation in liquid nitrogen) in the Culture Collection of Algae at the University of Texas at Austin (UTEX), Department of Botany, The University of Texas at Austin, Austin, Texas 78713-7640, USA. Chloromonas augustae (Skuja 1943) Pröschold, Marin, Schlösser et Melkonian comb. nov. Basionym: Chlamydomonas augustae Skuja 1943, Arch. Protistenkd. 96, p. 370–371, pl. 10, fig. 2–24 (descr. et ic. prima, iconotypus). Chlamydomonas augustae Skuja 1943 var. eupapillata Ettl 1976, Beih. Nova Hedwigia 49, p. 404–405, pl. 64, fig. 2 (descr. et ic. prima, iconotypus), authentic Chloromonas asteroidea Pröschold, Marin, Schlösser et Melkonian nom. nov. replaced synonym Chlamydomonas bipapillata (Bourrelly 1951) Ettl 1976 non Chloromonas bipapillata Ettl 1979, Beih. Nova Hedwigia 60, p. 118, pl. 25, fig. 1 (descr. et ic. prima, iconotypus) Synonyms: Chlamydomonas bipapillata (Bourrelly 1951) Ettl 1976, Beih. Nova Hedwigia 49, p. 407–408, pl. 57, fig. 5 (descr. et ic. prima, iconotypus). Chlamydomons macropyrenoidosa Skuja 1927 var. bipapillata Bourrelly 1951, Hydrobiologia 3, p. 256, fig. 2: 35–37 (descr. et ic. prima iconotypus). Epitype: The strain SAG 11-47 permanently preserved in a metabolically inactive state (cryopreservation in liquid nitrogen) in the Culture Collection of Algae at the University of Texas at Austin (UTEX), Department of Botany, The University of Texas at Austin, Austin, Texas 78713-7640, USA. Basionym: Chlamydomonas carrizoensis Deason et Bold in Chantanachat et Bold 1962; Deason et Bold 1960, Phycol. Stud. I, p. 16, fig. 8–9 (descr. et ic. prima, iconotypus). Synonyms: Chlamydomonas pyrenoidosa Deason et Bold 1960, Phycol. Stud. I, p. 16, fig. 8–9 (descr. et ic. prima, iconotypus) non Chlamydomonas pyrenoidosa Schiller 1952, Österr. Bot. Z. 99, p. 109–110, fig. 6 (descr. et ic. prima, iconotypus). Chlamydomonas deasonii Ettl 1976, Deason et Bold 1960, Phycol. Stud. I, p. 16, fig. 8–9 (descr. et ic. prima, iconotypus). Emended Diagnosis: Cells 15–28 × 8–25 µm, ellipsoid, ellipsoid-cylindrical, cell wall thin, papilla broad, rounded or absent, with two flagella about as long as the cell. Chloroplast cup-shaped, parietal and massive with irregular, branched perforations; Molecular Phylogeny and Revision of Chlamydomonas three to five discontinuous pyrenoids in mature cells, irregularly distributed, one pyrenoid in zoospores and young vegetative cells; eyespot pale red, elliptic to narrowly elongate in anterior position; two apical contractile vacuoles, nucleus in central or slightly basal position. Asexual reproduction by four or eight zoospores, rotation of the protoplast by 90º before first cell division (“false transverse division”). Sexual reproduction not observed. Epitype: The strain UTEX 968 permanently preserved in a metabolically inactive state (cryopreservation in liquid nitrogen) in the Culture Collection of Algae at the University of Texas at Austin (UTEX), Department of Botany, The University of Texas at Austin, Austin, Texas 78713-7640, USA. 287 Oogamochlamys Pröschold, Marin, Schlösser et Melkonian gen. nov. Basionym: Chlamydomonas rubrifilum Korshikov in Pascher 1927, Die Süßwasserflora Deutschlands, Österreichs und der Schweiz Vol. 4, p. 288–289, fig. 251 (descr. et ic. prima, iconotypus) non Chlamydomonas rubrifilum Korshikov var. africana Bourrelly 1961. Epitype: The strain SAG 3.85 permanently preserved in a metabolically inactive state (cryopreservation in liquid nitrogen) in the Culture Collection of Algae at the University of Texas at Austin (UTEX), Department of Botany, The University of Texas at Austin, Austin, Texas 78713-7640, USA. Chloromonas serbinowii Wille 1903, Nyt Mag. Naturvidensk. 41, p. 151, pl IV, fig. 28 (descr. et ic. prima, iconotypus). Diagnosis: Unicellular Chlorophyta sensu Bremer (1985) with two flagella in clockwise basal body orientation; cell wall thin with or without papilla, the single chloroplast cup-shaped, parietal and massive with ridges/grooves running parallel to the cell axis; several discontinuous pyrenoids, irregular distributed; eyespot in an anterior position; with only two apical contractile vacuoles or with many (>10) additional contractile vacuoles, distributed over the cell surface; nucleus in central or slightly anterior position. Asexual reproduction by zoospores, rotation of the protoplast by 90º before first cell division (“false transverse division”) by species with only two contractile vacuoles or transverse division by species with many contractile vacuoles; partial lysis of the sporangial wall before release of zoospores. Sexual reproduction by oogamy, homothallic, proterandric; zygotes ornamented or not ornamented, green or brownish-red. The genus is supported by molecular phylogenetic analyses using SSU rRNA sequence comparisons. Latin Diagnosis: Cellulae unae viridiplantae (Chlorophyta sensu Bremer 1985), flagellis duobus brevioribus quam corpus cellulae provisae (positio corporis flagellorum in sensu modi horae); papilla praesenta vel absenta; cellula membrana tenuis; chloroplastus unicus, depressionibus fissurisque multis praeditus, duae vacuolae pulsantes vel multae vacuolae pulsantes inter superficies plastidis atque membranam plasmaticum; multae pyrenoides laterales discontinae, per chloroplastum solidum fortuite distributae, stigmate filiformi, paululum anterior. Nucleus magnus cum nucleolum unum, paululum anterior media cellulae. Reproductio asexualis per bipartitionem repetitum endogenam plerumque in cellulas filias. Reproductio sexualis ovogama, spermae, ante ova evolutae, guttiformes sine membrana, flagellae duae corpore unus dimidiatusque longae, paulum sub apice insertae, chloroplastus reducta sine pyrenoide, stigmate distincta. Ova ex cellulae membrana cellulae elapsa spermis inductis. Zygota matura levis, viridis vel spinulis ornata, porphyra. Genus novus confirmatus per SSU rRNA sequentiae geneticae. Type species: Oogamochlamys gigantea (Dill 1895) Pröschold, Marin, Schlösser et Melkonian Epitype: The strain UTEX 492 permanently preserved in a metabolically inactive state (cryopreservation in liquid nitrogen) in the Culture Collection of Algae at the University of Texas at Austin (UTEX), Department of Botany, The University of Texas at Austin, Austin, Texas 78713-7640, USA. Oogamochlamys gigantea (Dill 1895) Pröschold, Marin, Schlösser et Melkonian comb. nov. Basionym: Chlamydomonas gigantea Dill, 1895, Jahrb. wiss. Bot. 28, p. 354, fig. 26–30, 34 (descr. et ic. prima, iconotypus). Chloromonas radiata (Deason et Bold 1960) Pröschold, Marin, Schlösser et Melkonian comb. nov. Basionym: Chlamydomonas radiata Deason et Bold 1960, Phycol. Stud. I, p. 14, fig. 4–5 (descr. et ic. prima, iconotypus). Epitype: The strain UTEX 966 permanently preserved in a metabolically inactive state (cryopreservation in liquid nitrogen) in the Culture Collection of Algae at the University of Texas at Austin (UTEX), Department of Botany, The University of Texas at Austin, Austin, Texas 78713-7640, USA. Chloromonas rubrifilum (Korshikov in Pascher 1927) Pröschold, Marin, Schlösser et Melkonian comb. nov. 288 T. Pröschold et al. Synonyms: Chlamydomonas pseudogigantea Korshikov in Pascher 1927, Die Süßwasserflora Deutschlands, Österreichs und der Schweiz Vol. 4, p. 287, fig. 249 (descr. et ic. prima, iconotypus), Chlorogonium pseudogiganteum (Korshikov) Ettl 1958, Zur Kenntnis der Klasse Volvophyceae. I. in Komárek et Ettl: Algologische Studien, p. 240. Chlamydomonas capensis Pocock 1962, Arch. Mikrobiol. 42, p. 59, fig. 1 (descr. et ic. prima, iconotypus), authentic culture: UTEX 1753 = SAG 22.98. Chlamydomonas megalis Bischoff et Bold 1963, Phycol. Stud. IV, p. 19, fig. 8–13, 88–91 (descr. et ic. prima, iconotypus), authentic culture: SAG 9.84 = UTEX 1492. Emended Diagnosis: Cells 30–50 × 25–35 µm, broad rounded to cylindrical to oviform, cell wall thin with a small rounded papilla, with two flagella about as long as the cell. Chloroplast cup-shaped, parietal and massive, the chloroplast surface with coarse ridges mostly parallel to the cell axis, 10–16(–20) discontinuous pyrenoids, irregularly distributed; eyespot pale red, elliptic to narrowly elongate in an anterior position; many (up to 200) contractile vacuoles, distributed over the cell surface; nucleus in central or slightly anterior position. Asexual reproduction by four (rarely two or eight) zoospores, first division true transverse without prior rotation of the protoplast. Sexual reproduction by oogamy, homothallic, proterandric; first division in gametangium simultaneously resulting in four protoplasts, later divisions successive, 32 or 64 (rarely 128) spermatozoids formed in a gametangium; spermatozoids 6–10 × 4–6 µm, teardrop-shaped, without a cell wall, with two flagella nearly 1.5 times as long as the cell, with two apical contractile vacuoles, chloroplast reduced, pale green, with a distinct eyespot, without pyrenoid. Zygotes ornamented with regular, flat-stopped projections, green to brownish-red. Epitype: The strain SAG 44.91 permanently preserved in a metabolically inactive state (cryopreservation in liquid nitrogen) in the Culture Collection of Algae at the University of Texas at Austin (UTEX), Department of Botany, The University of Texas at Austin, Austin, Texas 78713-7640, USA. Oogamochlamys zimbabwiensis (Heimke et Starr 1979) Pröschold, Marin, Schlösser et Melkonian comb. nov. Basionym: Chlamydomonas zimbabwiensis Heimke et Starr, 1979, Arch. Protistenkd. 122, p. 21, fig. 1 (descr. et ic. prima, iconotypus). Emended Diagnosis: Cells 15–22 × 15–20 µm, broad rounded to cylindrical to oviform, cell wall thin, papilla broad, rounded, two-humped or absent, with two flagella about as long as the cell. Chloroplast cup-shaped, parietal and massive, the chloroplast surface with very fine ridges mostly parallel to the cell axis, 2–6 discontinuous pyrenoids, irregularly distributed; eyespot pale red, elliptic to narrowly elongate or punctiform in anterior position; two apical contractile vacuoles, nucleus in central or slightly anterior position. Asexual reproduction into four (rarely two or eight) zoospores, rotation of the protoplast by 90º before first cell division (“false transverse division”). Sexual reproduction by oogamy, homothallic, proterandric; all cell divisions in gametangium successive, 16 spermatozoids formed in gametangium; spermatozoids 4–6 × 3–5 µm, teardrop-shaped, without cell wall, with two flagella nearly 1.5 times as long as the cell, with two apical contractile vacuoles, chloroplast reduced, pale green, with a distinct eyespot, without pyrenoid. Zygotes not ornamented, green. Epitype: The strain UTEX 2213 permanently preserved in a metabolically inactive state (cryopreservation in liquid nitrogen) in the Culture Collection of Algae at the University of Texas at Austin (UTEX), Department of Botany, The University of Texas at Austin, Austin, Texas 78713-7640, USA. Oogamochlamys ettlii Pröschold, Marin, Schlösser et Melkonian spec. nov. Diagnosis: Cells 16–27 × 18–22 µm, broad rounded to cylindrical to oviform, cell wall thin with a small rounded papilla, with two flagella about as long as the cell. Chloroplast cup-shaped, parietal and massive, the chloroplast surface with fine ridges mostly parallel to the cell axis, 3–8 discontinuous pyrenoids, irregularly distributed; eyespot pale red, elliptic to narrowly elongate in an anterior position; many (up to 20) contractile vacuoles, distributed over the cell surface, nucleus in central or slightly anterior position. Asexual reproduction by four (rarely two or eight) zoospores, first division true transverse without prior rotation of the protoplast. Sexual reproduction by oogamy, homothallic, proterandric; all cell divisions in gametangium successive, 16 spermatozoids formed in gametangium; spermatozoids 4–6 × 3–5 µm, teardrop-shaped, without cell wall, with two flagella nearly 1.5 times as long as the cell, with two apical contractile vacuoles, chloroplast reduced, pale green, with a distinct eyespot, without pyrenoid. Zygotes not ornamented, green. Molecular Phylogeny and Revision of Chlamydomonas Latin diagnosis: Cellulae 16–27 × 18–22 µm, cylindricae usque trunctatae ovataeque; flagellis duobus brevioribus quam corpus cellulae provisae; papilla parva truncata; cellula membrana tenuis; chloroplastus unicus, depressionibus fissurisque multis praeditus, multae vacuolae pulsantes inter superficies plastidis atque membranam plasmaticum; 3–8 pyrenoides laterales discontinae, per chloroplastum solidum fortuite distributae, stigmate filiformi, paululum anterior. Nucleus magnus cum nucleolum unum, paululum anterior media cellulae. Reproductio asexualis per bipartitionem repetitum endogenam plerumque in 4 (2–8) cellulas. Reproductio sexualis ovogama, spermae 4–6 × 3–5 µm, ante ova evolutae, guttiformes sine membrana, flagellae duae corpore unus dimidiatusque longae, paulum sub apice insertae, chloroplastus reducta sine pyrenoide, stigmate distincta. Ova ex cellulae membrana cellulae elapsa spermis inductis. Zygota matura levis, viridis. Holotypus: The strain UTEX 2218 permanently preserved in a metabolically inactive state (cryopreservation in liquid nitrogen) is the nomenclatural type of the name Oogamochlamys ettlii. The type material (holotype) has been deposited in the Culture Collection of Algae at the University of Texas at Austin (UTEX), Department of Botany, The University of Texas at Austin, Austin, Texas 78713-7640, USA. Iconotypus: Fig. 5g–l in this study. Type locality: col. M. A. Pocock as soil, Zimbabwe. Etymology: This taxon is named after Dr. Hanuš Ettl (1931–1997) to honor his contribution to the systematics of the genus Chlamydomonas Authentic culture: UTEX 2218 = SAG 24.98 Lobochlamys Pröschold, Marin, Schlösser et Melkonian gen. nov. Diagnosis: Unicellular Chlorophyta sensu Bremer (1985) with two flagella in clockwise basal body orientation; cell wall thin with mucilage layer around the flagellated cell, with or without papilla, the single chloroplast cup-shaped with thickened basal part cylindrical/ parietal, the chloroplast surface with incisions, partly lobate; one discontinuous pyrenoid, in basal or lateral position; eyespot pale red, elliptic to narrowly elongate in an anterior position; with two apical contractile vacuoles; nucleus in central or slightly anterior position. Asexual reproduction by zoospores, rotation of the protoplast by 90º before first cell division (“false transverse division”); sporangial wall with mucilage layer; total lysis of the sporangial wall before release 289 of zoospores by means of a vegetative lysin enzyme. Sexual reproduction by isogamy, homo- or heterothallic; zygotes not ornamented, green. The genus is supported by molecular phylogenetic analysis using SSU rRNA sequence comparisons. Latin Diagnosis: Cellulae unae viridiplantae (Chlorophyta sensu Bremer 1985), flagellis duobus brevioribus quam corpus cellulae provisae (positio corporis flagellorum in sensu modi horae); papilla praesenta vel absenta; cellula membrana tenuis cum velamentum gelatum; duae vacuolae pulsantes anterior; chloroplastus unicus, fissuris multis praeditus; unum pyrenoide lato discontinum, per chloroplastum solidum lateralis vel media distributa, stigmate filiformi, paululum anterior. Nucleus magnus cum nucleolum unum, paululum anterior media cellulae. Reproductio asexualis per bipartitionem repetitum endogenam plerumque in cellulas filias; sporangium membranum tenuis cum velamentum gelatum. Reproductio sexualis homothallica, per conjunctionem isogametarum nudarum ad zygotas formandas effecta. Zygota matura levis, viridis. Genus novus confirmatus per SSU rRNA sequentiae geneticae. Type species: Lobochlamys segnis (Ettl 1965) Pröschold, Marin, Schlösser et Melkonian Lobochlamys segnis (Ettl 1965) Pröschold, Marin, Schlösser et Melkonian comb. nov. Basionym: Chlamydomonas segnis Ettl 1965, Arch. Protistenkd. 108, p. 336, fig. 42–45, (descr. et ic. prima, iconotypus). Synonyms: Chlamydomonas fimbriata Ettl 1965, Arch. Protistenkd. 108, p. 418, fig. 102, (descr. et ic. prima, iconotypus), authentic culture: SAG 17.72 = UTEX 1349. Chlamydomonas gymno-gama Deason 1967, J. Phycol. 3, p. 109, fig. 8–12, (descr. et ic. prima, iconotypus), authentic culture: SAG 2.75 = UTEX 1638. Chlamydomonas pallidostigmatica King 1972, J. Phycol. 8, p. 120, fig. 1–3, 16 (descr. et ic. prima, iconotypus), authentic culture: SAG 9.83 = UTEX 1905. Chlamydomonas sajao Lewin 1984, Chin. J. Oceanol. Limnol. 2, p. 93, fig. 1 (descr. et ic. prima, iconotypus), authentic culture: SAG 50.84 = UTEX 2277. Emended Diagnosis: Cells 10–18 × 5–15 µm, ellipsoid, ellipsoid-cylindrical to oviform or fusiform, cell wall thin with mucilage layer around the flagellated cell, papilla broad, rounded or absent, with two flagella about as long as the cell. Chloroplast cupshaped with thickened basal part extending anteri- 290 T. Pröschold et al. orly, the chloroplast with distinct incisions, mostly parallel to the cell axis, one discontinuous pyrenoid in central or slightly basal position of the chloroplast; eyespot pale red, elliptic to narrowly elongate in anterior position; two apical contractile vacuoles, nucleus in central or slightly anterior position. Asexual reproduction by four or eight zoospores, rotation of the protoplast by 90º before first cell division (“false transverse division”); sporangial wall with mucilage layer; total lysis of the sporangial wall before release of zoospores by means of a vegetative lysin enzyme (VLE-group 10, Schlösser 1976). Sexual reproduction by isogamy; homothallic; gametes morphologically similar to vegetative cells; zygotes not ornamented, green. Epitype: The strain UTEX 1343 permanently preserved in a metabolically inactive state (cryopreservation in liquid nitrogen) in the Culture Collection of Algae at the University of Texas at Austin (UTEX), Department of Botany, The University of Texas at Austin, Austin, Texas 78713-7640, USA. Lobochlamys culleus (Ettl 1965) Pröschold, Marin, Schlösser et Melkonian comb. nov. Basionym: Chlamydomonas culleus Ettl 1965, Arch. Protistenkd. 108, p. 358, fig. 59, (descr. et ic. prima, iconotypus). Emended Diagnosis: Cells 10–18 × 5–15 µm, ellipsoid, ellipsoid-cylindrical to oviform, cell wall thin with mucilage layer around the flagellated cell, papilla broad, rounded or absent, with two flagella about as long as the cell. Chloroplast cylindrical/ parietal, chloroplast surface with distinct incisions in longitudinal and transverse orientation, partly lobate, one discontinuous pyrenoid in parietal position of the chloroplast; eyespot pale red, elliptic to narrowly elongate in anterior position; two apical contractile vacuoles, nucleus in central or slightly anterior position. Asexual reproduction by four or eight zoospores, rotation of the protoplast by 90º before first cell division (“false transverse division”); sporangial wall with mucilage layer; total lysis of the sporangial wall before release of zoospores by means of a vegetative lysin enzyme (VLE-group 9, Schlösser 1976). Sexual reproduction by isogamy; homo- or heterothallic; gametes morphologically similar to vegetative cells; zygotes not ornamented, green. Epitype: The strain SAG 17.73 permanently preserved in a metabolically inactive state (cryopreservation in liquid nitrogen) in the Culture Collection of Algae at the University of Texas at Austin (UTEX), Department of Botany, The University of Texas at Austin, Austin, Texas 78713-7640, USA. Discussion The Traditional Genera Chlamydomonas and Chloromonas In this contribution on the phylogeny and systematics of Chlamydomonas/ Chloromonas we intended to combine analyses of new and published data (both molecular and structural) with a critical examination of “historical“ descriptions, diagnoses, figures and taxonomic concepts in order to initiate a comprehensive taxonomic revision of the genera Chlamydomonas and Chloromonas. In the SSU rRNA sequence analyses of this study, species of the traditional genus Chlamydomonas (including Chloromonas) occurred in seven clades of the CW-group within the Chlorophyceae (“ Stephanosphaera”-, “ Polytoma”-, “Monadina”-, “Moewusii”-, “ Oogamochlamys”-, “ Chloromonas”-, and “Reinhardtii”-clade). In three clades (“ Stephanosphaera”-, “Moewusii”-, and “Reinhardtii”-clade), they were monophyletic with other genera of colonial, capsal, coccoid or filamenteous green algae and within these clades, species of Chlamydomonas/ Chloromonas were not monophyletic to the exclusion of the other genera. The polyphyly of the genus Chlamydomonas/ Chloromonas was indeed already anticipated by light microscopists (e.g. Ettl 1976, 1983a), but only demonstrated convincingly by molecular phylogenetic analyses (Buchheim et al. 1990, 1996, 1997b; Hepperle et al. 1998; Nakayama et al. 1996). Molecular phylogenetic studies by Nakayama et al. (1996) identified three principal clades within the CWgroup of Chlorophyceae which contained species of Chlamydomonas/ Chloromonas together with other genera and designated them “ Volvox” , “ Tetracystis” and “ Dunaliella” clades, Buchheim et al. (1997a, 1997b) identified five lineages (clades) containing species of Chlamydomonas/ Chloromonas (termed A–E in Buchheim et al. 1997a), and Hepperle et al. (1998) in their molecular phylogenetic analysis of the Phacotaceae found a total of six clades containing species of Chlamydomonas/ Chloromonas. How do these lineages relate to the clades identified in this study? Two of the clades containing Chlamydomonas/ Chloromonas species identified by Nakayama et al. (1996) correspond to the “Reinhardtii”- (“ Volvox”) and “Moewusii”- (“ Tetracystis”) clades in our analysis. The third clade (“ Dunaliella”), for which Nakayama et al. (1996) found support only in a neighbor-joining analysis using a simple model of evolution (Kimura’s two parameter model), was not supported in our study using a complex evolutionary model proposed for our data set by Model- Molecular Phylogeny and Revision of Chlamydomonas test (Posada and Crandall 1998) and consisted of three separate clades, the “ Stephanosphaera”-, the “ Polytoma”-, and the “ Dunaliella”-clade, only the first two contained Chlamydomonas/ Chloromonas species. Four of the five clades of Buchheim et al. (1997a) were recovered in the present analysis, they correspond to the “Moewusii”-clade (clade A), the “ Chloromonas”-clade (clade B), the “Reinhardtii”clade (clade D), and the “ Oogamochlamys”-clade (clade E). Only their clade C (similar to the Dunaliella-clade of Nakayama et al. 1996) was not supported in our analysis (see above). It is likely that the larger taxon sampling used in this study (27 taxa compared to ten taxa or six taxa in the analyses of Nakayama et al. 1996 and Buchheim et al. 1997a, respectively) helped us to resolve the uncertainties regarding the composition of clade C. The six clades containing Chlamydomonas/ Chloromonas species found by Hepperle et al. (1998) were also recovered in our analysis. They correspond to: (1) the “ Stephanosphaera”-clade (their Haematococcaceae), (2) the “ Polytoma”-clade (Dunaliellales II), (3) the “Monadina”-clade (Chlamydomonas I), (4) the “Moewusii”-clade (Chlamydomonas IV), (5) the “ Chloromonas”-clade (Chlamydomonas II), and (6) the “Reinhardtii”-clade (Chlamydomonas III). Overall, the previous molecular phylogenetic investigations using SSU rRNA sequence comparisons already identified ten of the 11 clades within the CW-group described in this study [among the clades without Chlamydomonas/ Chloromonas species the “ Phacotus”- and the “ Dunaliella”-clades were identified by Hepperle et al. (1998; their Phacotaceae and Dunaliellales I), and the “Radiosa”-clade only by Buchheim and Chapman (1992)]. Only the “ Chlorogonium”-clade had previously not been recognized. We sequenced four strains of Chlorogonium belonging to three species and found them to be a sister group to Haematococcus pluvialis; in previous analyses H. pluvialis could not be positioned in phylogenetic trees of CW-group green algae (Buchheim et al. 1996; Hepperle et al. 1998). It is noteworthy that in our analysis (as in Hepperle et al. 1998) no sister group to the “ Dunaliella”-clade [in contrast to previous analyses by Buchheim et al. (1997a,b) and Nakayama et al. (1996)] could be identified, corroborating the conclusion expressed by Melkonian and Preisig (1984), based on ultrastructure, that “ Dunaliella cannot be regarded simply as a “naked” Chlamydomonas”. It is evident, however, that the “ Dunaliella”-clade should not be separated at a larger taxonomic level (e.g. order; Ettl 1983b) from the other clades of the CW-group (as suggested by Melkonian 1990). 291 What is Chlamydomonas and Chloromonas? From this study and earlier molecular phylogenetic analyses (see above), it is clear that the traditional genus Chlamydomonas/ Chloromonas is polyphyletic within the CW-group of the Chlorophyceae and that the genus needs to be revised. On the species level, such a revision will require much more experimental work on the numerous strains of Chlamydomonas in culture as well as careful evaluation of hundreds of descriptions and diagnoses of Chlamydomonas species. On the genus level, a revision of Chlamydomonas must first clarify the identity of its type species (C. pulvisculus, see below) and should take into account that: (1) other morphologically well-circumscribed genera of green algae, e.g., species of Volvox, form monophyletic lineages with species of Chlamydomonas to the exclusion of other Chlamydomonas species; (2) colorless taxa, e.g., Polytoma, form clades with species of Chlamydomonas to the exclusion of other Chlamydomonas species; and finally (3) taxa (strains) without pyrenoids cannot be separated from taxa (strains) with pyrenoids calling into question the justification for the genus Chloromonas in the traditional sense (i.e. “ Chlamydomonas without pyrenoid”). What are the formal type species of Chlamydomonas and Chloromonas? An answer to this question is crucial because these type species will define the taxonomic identity of their respective genera in a revised classification. Only those Chlamydomonas species, which are monophyletic with the type species (to the exclusion of other taxa), will retain their generic identity whereas other species will have to be transferred to other genera. The formal type species of Chlamydomonas Ehrenberg 1833 (conserved against the original spelling Chlamidomonas and two earlier synonyms, Protococcus C. Agardh 1824 and Sphaerella Sommerfelt 1824; see Silva 1980) is Chlamydomonas pulvisculus (O.F. Müller) Ehrenberg 1838, which was described by O.F. Müller (1786) as Monas pulvisculus. Unfortunately, this species is ambiguous as the descriptions and figures by O.F. Müller and Ehrenberg cannot be matched with modern species (see Results). It has therefore not been dealt with in recent taxonomic treatments of Chlamydomonas (Ettl 1976, 1983a), and in fact Ettl (1976) considered this species as invalid („zu streichende Art“). To solve the problem, Ettl (1976) proposed to select C. reinhardtii as the type (neotype) of the generic name instead of C. pulvisculus. We propose to designate C. reinhardtii as conserved type and provide an emended diagnosis of the genus Chlamydomonas (see Results). 292 T. Pröschold et al. The position of C. reinhardtii in molecular phylogenies (Buchheim et al. 1996, 1997b; Hepperle et al. 1998; Nozaki et al. 2000; Fig. 1 in this study) demonstrates that this species is intimately associated with colony-forming Volvocales within the “Reinhardtii”clade (e.g., Tetrabaena, Gonium, Eudorina, Volvox), but not closely related to most other species of Chlamydomonas. In consequence, Chlamydomonas as emended here (sensu stricto) will be confined to at most a few (probably only one) species. In Chloromonas Gobi 1899/1900 emend. Wille 1903, the formal type species, C. reticulata, is a welldescribed taxon (Ettl 1970; Goroschankin 1891) and in this study is represented by seven strains with almost identical SSU rRNA sequences and, with the notable exception of the presence (five strains) or absence (two strains) of a pyrenoid, identical structural characters. C. reticulata is part of one of the CW-group clades, i.e. the “ Chloromonas”-clade, currently with eight species (see Results). Six of these species were previously recognized as Chlamydomonas because of the presence of one or several pyrenoids. Obviously, in Chlamydomonas/ Chloromonas the character state “pyrenoid” has no taxonomic significance (even at the species level; i.e. C. reticulata) and certainly cannot be used as generic character to define Chloromonas as in traditional classifications. It should also be noted that the presence or absence of pyrenoids sometimes depends on environmental conditions such as photoheterotrophy (see Nozaki et al. 1998 for species of Chlorogonium). The two “ Chloromonas” species C. perforata and C. oogama, which were studied by Buchheim et al. (1997b), and which cluster with the “ Stephanosphaera”- and the “Reinhardtii”-clade in our analysis (Fig. 1), reveal pyrenoids, which led Buchheim et al. (1997b) to conclude that they “have been misidentified as Chloromonas taxa or they represent contaminated cultures”. Similarly, a Chlamydomonas strain (SAG 3.85) isolated by one of us (MM) was originally placed in the Pleiochloris section because of the presence of many pyrenoids. Later, it was designated Chloromonas clathrata (Schlösser 1994) as it apparently now lacked pyrenoids. When this strain was reinvestigated during this study, we identified pyrenoids (but only after staining with azocarmine), corroborating the initial observations when the alga (now designated Chloromonas rubrifilum) was isolated. Clearly there is much to be learned about the significance of the presence or absence of pyrenoids in the genus Chlamydomonas. We consider it most likely, and find no conflicting evidence in molecular phylogenetic analyses (this study; Buchheim et al. 1997b; Morita et al. 1999), that strains without pyrenoids evolved from pyrenoid-containing strains by multiple independent losses. It may be possible, however, that such pyrenoid losses occurred more often in some clades than in others. In the two clades, which we investigated in more detail in this study, strains without pyrenoids occurred only in the “ Chloromonas”-clade, not in the “ Oogamochlamys”-clade. One possible consequence of the rejection of the defining character of Chloromonas (“no pyrenoid”) would be the invalidation of this genus. Instead, we prefer to retain the name and validate the taxon Chloromonas (equated with the “ Chloromonas”-clade), which needs to be defined by other characters. Fortunately, the investigation of the chloroplast structure revealed the presence of either asteroid chloroplasts or parietal chloroplasts with perforations in all strains of the “ Chloromonas”clade investigated, characters which are presently unknown in other flagellates of the CW-group of the Chlorophyceae (see Results). Among approximately 130 described species of Chloromonas (Ettl 1983a), about 30 species also display asteroid chloroplasts or parietal chloroplasts with perforations and, though neither cultures nor sequences are available, can be expected to belong to the “ Chloromonas”-clade. As an example, Chloromonas insignis and Chloromonas palmelloides (no pyrenoid, parietal chloroplasts with perforations) were positioned within the „ Chloromonas“-clade using rbc L-sequence comparisons (Morita et al. 1999). Most of the more than 600 species in the traditional genus Chlamydomonas presumably have no phylogenetic affinity to C. reinhardtii and consequently, will have to be transferred to other described genera, such as Chloromonas (see above), or described as new genera. To initiate the description of successor genera of Chlamydomonas (sensu lato), we propose to raise two independent lineages to genus level and designate them as Oogamochlamys and Lobochlamys (see Results and below). Some of our results bear on the question of the species concept in the genus Chlamydomonas. All of the Chlamydomonas (including Chloromonas) species described until now have been established using a morphological species concept (see Ettl 1976, 1983a). Unfortunately, the variability of the diagnostic features used, was not sufficiently appreciated: most descriptions were based on observations of material from natural samples, and often only a few cells were examined. In addition, rigorous discrimination against known species was seldom made when new taxa were described. Another possible reason for inflation in the number of species, ironically, relates to a diagnostic character introduced to subdivide the genus into more managable groups (sections, sometimes termed subgenera): Molecular Phylogeny and Revision of Chlamydomonas the number and position of pyrenoids in relation to the shape and position of the chloroplast (see Ettl 1976, 1983a; Pascher 1927). We and others (Buchheim et al. 1996) have shown that the sections of Chlamydomonas have little meaning in phylogenetic terms. As an example, we note that in the authentic strain of Lobochlamys segnis (= C. segnis), the position of the pyrenoid successively changes during the cell cycle (from zoospore to mature vegetative cell) to reflect the characteristics of three sections, namely Chlorogoniella, Chlamydella and Euchlamydomonas (unpublished observations). It is thus not surprising that authentic strains of Chlamydomonas species, which have been assigned to different sections, may be genetically (and morphologically) so similar that they should be placed in a single species: e.g. C. gymnogama and C. pallidostigmatica (section Chlamydella), C. sajao (section Pseudagloë), C. segnis and C. fimbriata (section Euchlamydomonas) in the species Lobochlamys segnis (this study). Furthermore, the number of pyrenoids per cell is a homoplasy among the strains studied (as in the section Pleiochloris) and can also vary during development within a single strain (as in Chloromonas carrizoensis, see Results). Interestingly, in the majority of the sections of Chlamydomonas the same diversity of chloroplast characters such as different shapes (cup-shaped, asteroid) or different chloroplast surfaces (with or without incisions, perforations, or ridges/grooves) occur in each section, suggesting to us that the concept on which the sections are based, i.e. the number and position of pyrenoids, obscures other taxonomically more informative morphological characters. This conclusion is corroborated by our observations that some aspects of the shape and surface structure of the chloroplast correlate well with results obtained from the molecular phylogenetic analyses: all strains of the “ Oogamochlamys”-clade (Oogamochlamys and Lobochlamys) have cup-shaped/parietal chloroplasts (except for L. culleus, see below) with either surface ridges (Oogamochlamys) or incisions (Lobochlamys) running mostly parallel to the cell axis, and strains of the “ Chloromonas”-clade have either asteroid or cup-shaped/parietal chloroplasts with perforations (the latter not resulting in the formation of chloroplast lobes) but no surface ridges/grooves or incisions (see Results). We can also extend these observations to groups of strains within clades. In strains assigned to Lobochlamys segnis, the chloroplast is cup-shaped with thickened basal part, the chloroplast surface reveals incisions which (at least in the chloroplast posterior) do not traverse the chloroplast. In Lobochlamys culleus, the chloroplast is cylindrical/parietal, the chloroplast surface ex- 293 hibits incisions which lead to the formation of distinct chloroplast lobes. In strains assigned to the oogamous species Oogamochlamys gigantea, Oogamochlamys zimbabwiensis and Oogamochlamys ettlii, the chloroplasts are cup-shaped and parietal with ridges/grooves running parallel to the cell axis, which in O. zimbabwiensis and O. ettlii (perhaps due to the smaller cell size) are less coarse than in O. gigantea. In the “ Chloromonas”-clade, strains assigned to C. reticulata have cup-shaped and parietal chloroplasts with perforations, whereas strains assigned to C. actinochloris (two strains), C. augustae (three strains), C. asteroidea, and C. radiata have asteroid chloroplasts. How the two types of chloroplasts in the “ Chloromonas”-clade relate to each other, cannot yet be decided, because the branching order within this clade could not be resolved using SSU rRNA sequence comparisons (see Fig. 2, Results). The functional significance of the various chloroplast shapes and surface structures also remains to be elucidated. In summary, structural features of the chloroplast seem to be of taxonomic value in the Chlamydomonas/ Chloromonas complex and in general lend support to the results obtained by molecular phylogenetic analysis. However, additional critical observations of a variety of morphological characters over the whole range of strains in this complex are necessary before any firm conclusions can be drawn. Surprisingly, Pascher (1927) avoided the use of structural features of the chloroplast for his definition of subgenera. Instead, he used the homoplasious chloroplast characters (i.e. presence, number and position of pyrenoids in relation to chloroplast shape), which apparently led to the description of artificial subgenera. Another result of this approach was that isolates, which differed only in pyrenoid characters but were otherwise identical, were placed in different subgenera and consequently also separated at the species level. However, these isolates as shown in this study represent only one (morpho)species (e.g., Lobochlamys segnis; Chloromonas reticulata). In conclusion, a questionable subgenus/section-concept in Chlamydomonas led to an inflation in the number of (morpho)species described, many of which will have to be abandoned in the future. Ettl and Schlösser (1992) made a start in this direction by reducing three species of Chlamydomonas to synonyms of a fourth species, C. applanata. In addition, they noted that six further species described only from natural samples could not be distinguished morphologically from C. applanata and suggested that they should no longer be recognized as such. Their conclusions were based on careful light microscope observations of 294 T. Pröschold et al. strains of the various species, and comparison with published diagnoses. Their study was facilitated because the strains selected for light microscopy belonged to a common vegetative lysin enzyme group (VLE-group 7, Schlösser 1976). V-lysins are proteases, which dissolve the sporangial cell walls to allow liberation of zoospores (Jaenicke et al. 1987; Matsuda et al. 1995). Schlösser (1976, 1984) described methods for obtaining v-lysins from strains for cross-specificity testing. From more than 100 strains of Chlamydomonas tested, v-lysins were recovered from 65 strains belonging to 42 different species and were assigned to 15 VLE-groups based on cross-reactivity within groups combined with non-reactivity between different groups (three exceptions were noted revealing non-reciprocal cross-reactivity between different VLE-groups; Schlösser 1984). Based on their analysis of VLEgroup 7 (C. applanata) Ettl and Schlösser (1992) concluded that strains of a VLE-group should be regarded as clones of one (morpho)species. In two strains of this VLE-group (included the authentic strain of “ C. humicola”), SSU rRNA sequences have been analysed and shown to be identical (Gordon et al. 1995). In their molecular phylogenetic analyses, Buchheim et al. (1997a) have shown that all members of VLE-group 14, which include the authentic strains of C. geitleri, C. hindakii, C. monoica and C. pinicola, also reveal almost identical SSU rRNA sequences and can be regarded either as distinct isolates of a single species (C. noctigama) or as a group of sibling species [because of the nonviability of most offspring following “interstrain” crosses (Burrascano and Van Winkle-Swift; Genetics 107:s15, 1984)]. How do our results relate to the VLE-groups of Schlösser (1976, 1984)? V-lysins are not known for the oogamous species (perhaps their concentration is too low for the cross-reactivity tests as only a small hole develops in the sporangial wall) or for strains of the “ Chloromonas”-clade; however, the three strains of Lobochlamys culleus belong to VLEgroup 9 and the six strains assigned to Lobochlamys segnis (including the authentic strains of Chlamydomonas segnis, C. fimbriata, C. gymnogama, C. pallidostigmatica, and C. sajao) belong to VLE-group 10. Although our molecular phylogenetic analyses based on SSU rRNA sequence comparisons could not with confidence separate L. culleus from L. segnis (see Results; less conserved molecular markers such as the internal transcribed spacers of the rRNA operon could be useful), in general, VLE-groups known to be closely related (one-sided cross-reactivity) were also the closest relatives in the SSU rRNA phylogeny [e.g. VLE-groups 1 (C. reinhardtii), 2 (C. debaryana) and 15 (C. zebra) in the “Reinhardtii”-clade, and VLE-groups 12 (C. moewusii) and 13 (C. pitschmannii) in the “Moewusii”-clade; Fig. 1]. It appears that the phylogenetic signal of the SSU rRNA (at least in Chlamydomonas) is informative almost to the level of the (morpho)species (VLE-group). This is exemplified by strain UTEX 2218, which we could neither place in C. zimbabwiensis nor in C. gigantea based on SSU rRNA sequence comparisons (see Results). Morphologically, this strain is intermediate between the two oogamous species. Our unpublished observations show that spermatangia from strain UTEX 2218 cannot induce egg release from vegetative cells of the two other species (strains SAG 44.91 and SAG 45.91 respectively), and conversely, that spermatangia from the other two species cannot induce egg release from vegetative cells of UTEX 2218, suggesting that indeed strain UTEX 2218 represents a new (morpho)species, which is here described as Oogamochlamys ettlii. It has been argued that a biological species concept cannot be applied to the genus Chlamydomonas, because sexual reproduction is unknown for more than 80% of the species described (Ettl and Schlösser 1992). However, in the (few) cases where sexual reproduction, molecular phylogeny (based on SSU rRNA sequence comparisons), morphology (based on light microscopy) and cross-reactivity of v-lysins has been studied in the same Chlamydomonas strains almost complete congruence between these characters has been observed: strains of one VLE-group are morphologically indistinguishable, and they usually represent a group of biological sibling species (the latter separated from each other mainly by postzygotic barriers). Strains of a VLE-group also reveal the most closely related sequences in the SSU rRNA phylogeny and cannot be split into different, significantly supported monophyletic lineages (this study). However, it appears that in the SSU rRNA phylogeny a single VLE-group is not always monophyletic to the exclusion of another (presumably closely related) VLE-group (e.g. VLE-groups 9 and 10 in Lobochlamys; Figs 1, 2), indicating that the phylogenetic signal of the SSU rRNA in such cases is not informative at the (morpho)species level. Nevertheless, the SSU rRNA sequences between strains in the same VLEgroup/(morpho)species often differ from each other (as in L. segnis or in L. culleus). This situation sheds some light on the “phylogenetic species concept” (Baum 1992): if the identification of the smallest diagnosable units is the primary criterion for species delimitation, then any base difference between two sequences could lead to the description of a new Molecular Phylogeny and Revision of Chlamydomonas species. However, when no sister group relationship can be identified between such strains (see above), the sequence difference is obviously not phylogenetically informative. Internal transcribed spacers (in particular ITS-2) of the ribosomal RNA operon have been used successfully in recent years to address the question of the relationship between sequence divergence and species identity in Chlamydomonas and in the colonial Volvocales (Coleman and Mai 1997; Coleman et al. 1998; Fabry et al. 1999). One outcome of these studies has been the recognition of a correlation between the absence of compensatory base changes (CBC) in conserved stem regions of the ITS-2 and the ability of gametes of different strains to form zygotes (Coleman 2000). Groups of strains without CBCs readily mate and form zygotes, i.e., a CBC-clade conforms to a Z (Z for zygote formation) clade. It is likely, but remains to be investigated in more detail, that Z-clades and/or CBC-clades correspond to VLE-groups. If one accepts the notion that a VLE-group corresponds to a (morpho)species of Chlamydomonas/ Chloromonas (in the traditional sense), then the number of (morpho)species can be roughly estimated: the recently emended Chlamydomonas species include about three to five synonymous species based on comparison with authentic strains (Ettl and Schlösser 1992; and this study). If this remains the trend for the future, then the number of species is only about 25% of the known. In addition, there appear to be numerous incompletely described species. Ettl and Schlösser (1992) in their revision of C. applanata noted six incompletely described species (compared to C. applanata and its three synonymous species). If this can be generalized, then of the 600 known species of traditional Chlamydomonas/ Chloromonas only 50–100 species will remain (distributed over several genera, see above). A similar figure is obtained, when we calculate the number of VLE-groups: Schlösser (1976, 1984) found 15 VLE-groups in 42 species. When the incompletely described species are taken into consideration, a maximum of about 70 VLE-groups/ (morpho)species are to be expected. The present study of the “ Chloromonas”- and the “ Oogamochlamys”-clades identified 13 (morpho)species (five in the “ Oogamochlamys”- and eight in the “ Chloromonas”-clade). The Oogamous Species (Oogamochlamys gen. nov.) The four strains designated here Oogamochlamys gigantea, including the authentic strains of Chlamydomonas capensis and C. megalis (Table 5, Results), 295 are very similar to each other in morphology, reproduction and in their SSU rRNA sequences (Fig. 2). Many contractile vacuoles scattered over the cell surface is a feature, which has been previously observed in C. pseudogigantea (Geitler 1954), in C. megalis (Bischoff and Bold 1963), C. capensis (Heimke and Starr 1979; not mentioned by Pocock 1962), and in all strains investigated by us. In contrast, only two apical contractile vacuoles were described for C. gigantea (Dill 1895). The two apical contractile vacuoles occur in all strains investigated in this study, but the numerous others, which are smaller and hidden in the grooves of the chloroplast surface can be overlooked easily. It is thus most likely that Dill (1895) either did not notice them or, alternatively, that the multiple empty circles scattered over the chloroplast surface, depicted in his type figure, represent contractile vacuoles and not starch grains. The first division step of Oogamochlamys gigantea is always, according to our observations, a „true transverse division“ without rotation of the protoplast, as also described for C. pseudogigantea (Ettl 1988; Geitler 1954). The first division step in most Chlamydomonas species is either longitudinal or transverse after a 90° rotation of the protoplast within the cell wall (“false transverse division”). Heimke and Starr (1979) indicated for the type material of C. capensis that the initiation of zoosporogenesis begins with a 90°rotation of the protoplast. F or the same strains, however, we found only true transverse divisions. Dill (1895) described, but did not illustrate, the first division step in C. gigantea as longitudinal. In the same publication, however, he described a true transverse division for C. reinhardtii although for this species, only a longitudinal division of the protoplast after 90° rotation is known (Schlösser 1972). We suppose that Dill had no possibility for long-term observations of dividing cells, and do not put undue emphasis on his statements concerning cell division. Sexual reproduction in Oogamochlamys gigantea is a true oogamy characterized by the differentiation of small, flagellate male gametes with reduced chloroplasts and large immobile female gametes, which escape from their cell walls, becoming naked protoplasts. This was not recognized in some of the earlier descriptions: Pocock (1962) described different stages of this mode of sexual reproduction in C. capensis, but did not observe the release of the egg cell protoplast; she therefore labeled the mode of sexual reproduction in this species heterogamous. Dill (1895) depicted a protoplast of C. gigantea released from its cell wall and also cells with the typical zygote walls; however, he did not recognize 296 T. Pröschold et al. them as stages of the sexual cycle. A complete description of true oogamy was first presented by Geitler (1954) for C. pseudogigantea and later confirmed by Heimke and Starr (1979). Sexual reproduction is triggered by special inducing conditions (Heimke and Starr 1979), which were presumably not met by the authors of C. megalis (Bischoff and Bold 1963). As a result of this comparison, we integrated the three species discussed above (C. pseudogigantea, C. megalis, and C. capensis) into the species first described: Oogamochlamys gigantea Dill (an emended diagnosis is given in the Results section). In O. zimbabwiensis, the first cell division in asexual reproduction is a “false transverse division” unlike the situation in O. gigantea (Heimke and Starr 1979; and this study). Why two closely related taxa such as O. gigantea and O. zimbabwiensis should differ in the mode of cell division during asexual reproduction is not known, however a correlation appears to exist between multiple contractile vacuoles and a true transverse division. We note that this correlation holds for the new species Oogamochlamys ettlii as well as for unrelated taxa with multiple contractile vacuoles (e.g. in the Chlorogonium-clade; Ettl 1988; Hoops and Witman 1985). Conclusions Based on a molecular phylogentic analysis using SSU rRNA sequence comparison and a reevalution of light microscope characters, the traditional concepts of the genera Chlamydomonas and Chloromonas have been revised. Two monophyletic lineages (clades), which contain species previously assigned to Chlamydomonas, are now recognized as new genera (Oogamochlamys and Lobochlamys). This is a first contribution towards the taxonomic revision of Chlamydomonas/ Chloromonas using a combination of molecular phylogenetic analyses and investigations into the morphology and life history of this important and ubiquitous assemblage of “green protists”. Methods Cultures and light microscopy: Strains were obtained from the Sammlung von Algenkulturen, Universität Göttingen, Germany (SAG; Schlösser 1994) and the Culture Collection of Algae, University of Texas at Austin, USA (UTEX; Starr and Zeikus 1993) and are listed in Table 5. The algae were grown in modified “Bold Basal Medium” (3N-BBM+V, medium 26a in Schlösser 1997). The cultures of the ooga- mous Chlamydomonas species (C. gigantea and C. zimbabwiensis) were grown in 100 ml Erlenmeyer flasks at 20 ºC at a photon fluence rate of 50 µEm–2s–1, and a light/dark-cycle 14/10 h. Sexual reproduction was induced by the following method: 10 ml of a dense cell suspension from a 14 day-old culture in Erlenmeyer flasks were sedimented by low speed centrifugation and the pellets were resuspended in 5 ml 3N-BBM+V or in the same medium without sodium nitrate (BBM-N+V) to induce gametogenesis. These suspensions were incubated in shallow watch glasses, which rested on a glass triangle in a petri dish (Starr 1955). The bottom of the petri dish contained 20 ml distilled water in order to reduce evaporation. Samples were controlled microscopically each day. For the continuous observation of single cells (up to two weeks) we used a microchamber (Heunert 1973) containing 3N-BBM+V agar medium (1.5% w/v). For species determination, strains were grown under the same culture conditions as above. After 14 days of culture, mature vegetative cells (near the end of the light period) were observed by light microscopy. Micrographs were made using oil immersion lenses and Kodak Ektachrome (160 ASA) or Fuji RD 135 (100 ASA). The cells were observed in the living state, either free swimming or immobilized by removal of nutrient solution with filter paper or transfer of cells into a gel (Protoslo, No. 88-5141, Carolina Biological Supply Company, USA, or ultralow gelling agarose according to Reize and Melkonian 1989). For staining of pyrenoids, azocarmine was used (5% w/v in 45% v/v acetic acid; see Ettl 1976). For photographic documentation of the chloroplast structures, cells (500 µl algal suspension) were either fixed (50 µl 2% OsO4, 20 µl 25% glutaraldehyde, 430 µl 3N-BBM+V-culture medium, for 30 minutes on ice and washed once in culture medium) or studied alive after they had settled on the slide. Observations were made using Nomarski interference contrast with a Zeiss inverted microscopy (IM35). For documentation of contractile vacuole number and activity, live cells were photographed using an electronic flash attached to a Zeiss inverted microscope (IM). DNA isolation, PCR and sequencing: DNA isolation was performed using a CTAB protocol (Rogers and Bendich 1985; modified by Friedl 1996) followed by the amplification of nuclear-encoded SSU rRNA regions by polymerase chain reactions (PCR; Saiki et al. 1988) using thermocycling protocols and 5-biotinylated PCR-primers (biot.) as described previously (Marin et al. 1998). Amplications of SSU rRNA regions were performed using oligonucleotide primers A (biotiny- Molecular Phylogeny and Revision of Chlamydomonas lated) (Medlin et al. 1988) and ITS055 (Marin et al. 1998). PCR products were sequenced with sequencing methods and primers (82F, 528F, 920F, BR, 920R, 536R) as described by Marin et al. (1998) and Marin and Melkonian (1999). The nucleotide sequence data are avaiable in the EMBL, GenBank and DDBJ sequence databases under the accession numbers listed in Table 5. SSU rRNA sequences were manually aligned with homologous sequences from other Chlorophyta using the Olsen Multiple Sequence Alignment Editing program with respect to primary and secondary structural conservation. Strains designations and EMBL/GenBank accession numbers of published sequences are given in Figures 1 and 2. The alignment is available from the authors upon request. Phylogenetic analyses: Phylogenetic trees were inferred using distance, parsimony, and maximum likelihood criteria using PAUP* versions 4.0b2, 4.0b6, and 4.0b8 (Swofford 1998). Two data sets were used: a large alignment of 156 taxa of Chlorophyta with 1,642 unambiguously aligned positions, and a smaller data set consisting of 34 taxa of Chlamydomonas/ Chloromonas and 1,702 unambiguously aligned positions. To decide on the evolutionary model, which best fitted our data, we used the program Modeltest 3.04 (Posada and Crandall 1998), which employs two statistics, the likelihood ratio test (LRT) and the Akaike information criterion (AIC; Akaike 1974). Based on the results of these tests, the model selected by the hierarchical LRT was the Tamura-Nei model (TrN; Tamura and Nei 1993), with the proportion of invariable sites (I) and the gamma shape parameter (G) for among site variation calculated from the data set (TrN+I+G for the global data set and TrNef+I+G for the smaller data set; base frequencies and substitution parameters were estimated by Modeltest). Maximum likelihood with the Tamura-Nei model with invarint sites and gamma parameter was used as the distance measure in inferring trees using the distance optimality criterion. The same model was invoked in maximum likelihood analyses. For comparison with the TrN+I+G model in the global analysis we also employed the simpler HKY85 model of evolution (without I and G; Hasegawa et al. 1985). Maximum parsimony analyses used a heuristic search with a branch-swapping algorithm (tree bisection-reconnection algorithm; 10 replicates with random order of sequence addition). The confidence of branching in all methods was assessed using 1000 bootstrap resamplings of the data set (except for ML analysis [Fig. 2] was assessed using 100 bootstrap resamplings; Felsenstein 1985). 297 User-defined trees were generated by modifying the treefile of the “best tree” using TreeView (version 1.6.2; Page 1996). To compare user-defined topologies with the “best tree“ the sequence data file and the treefiles were loaded into PAUP* and used for Kishino-Hasegawa-tests (Kishino and Hasegawa 1989); comparisons were based on maximum likelihood (ML; model of evolution selected by Modeltest 3.04, see above) and maximum parsimony criteria (see Tables 1 and 2). Acknowledgements We like to thank Barbara Surek (Cologne) for help with photomicrography, Paul C. Silva (Berkeley), and Vincent Demoulin (Liège) for advice in nomenclatural problems, Robert A. Andersen (Bigelow) for valuable suggestions to improve the manuscript and Siegfried Werth (Cologne) for help with image analysis. This research was supported by grants from the Deutsche Forschungsgemeinschaft to B. Marin, U. G. Schlösser and M. Melkonian. References Adair WS, Snell WJ (1990) The Chlamydomonas reinhardtii Cell Wall: Structure, Biochemistry and Molecular Biology. In Adair WS, Mecham RP (eds) Organization and Assembly of Plant and Animal Extracellular Matrix. Academic Press, San Diego, pp 15–84 Akaike H (1974) A new look at the statistical model identification. IEEE Trans Contr 19: 716–723 Baum D (1992) Phylogenetic species concepts. Trends Ecol Evol 7: 1–2 Bischoff TR, Bold HC (1963) Phycological studies IV. Some soil algae from Enchanted Rock and related algal species. Univ Texas Publ 6318: 1–95 Bremer K (1985) Summary of green plant phylogeny and classification. Cladistics 1: 369–385 Buchheim MA, Chapman RL (1992) Phylogeny of Carteria (Chlorophyceae) inferred from molecular and organismal data. J Phycol 28: 362–374 Buchheim MA, Buchheim JA, Chapman RL (1997a) Phylogeny of the VLE-14 Chlamydomonas (Chlorophyceae) group: a study of 18S rRNA gene sequences. J Phycol 33: 1024–1030 Buchheim MA, Buchheim JA, Chapman RL (1997b) Phylogeny of Chloromonas (Chlorophyceae): a study of 18S ribosomal RNA gene sequences. J Phycol 33: 286–293 Buchheim MA, Turmel M, Zimmer EA, Chapman RL (1990) Phylogeny of Chlamydomonas (Chlorophyta) 298 T. Pröschold et al. based on cladistic analysis of nuclear 18S rRNA sequence data. J Phycol 26: 689–699 Buchheim MA, Lemieux C, Otis C, Gutell RR, Chapman RL, Turmel M (1996) Phylogeny of Chlamydomonadales (Chlorophyceae): A comparison of ribosomal RNA gene sequences from the nucleus and the chloroplast. Mol Phylogenet Evol 5: 391–402 Chantanachat S, Bold HC (1962) Phycological studies II. Some algae from arid soils. Univ Texas Publ 6218: 1–74 Chapman RL, Buchheim MA, Delwiche CF, Friedl T, Huss VAR, Karol KG, Lewis LA, Manhart J, McCourt RM, Olsen JL, Waters DA (1998) Molecular Systematics of the Green Algae. In Soltis DE, Soltis PS, Doyle JJ (eds) Molecular Systematics of Plants II. Kluver Academic Publishers, Boston, pp 508–540 Coleman AW (2000) The significance of a coincidence between evolutionary landmarks found in mating affinity and a DNA sequence. Protist 151: 1–9 Coleman AW, Mai JC (1997) Ribosomal RNA ITS-1 and ITS-2 sequence comparisons as a tool for predicting genetic relatedness. J Mol Evol 45: 168–177 Coleman AW, Preparata RM, Mehrotra B, Mai JC (1998) Derivation of the secondary structure of the ITS1 transcript in Volvocales and its taxonomic correlations. Protist 149: 135–146 Deason TR (1967) Chlamydomonas gymnogama, a new homothallic species with naked gametes. J. Phycol 3: 109–112 Deason TR, Bold HC (1960) Phycological studies I. Exploratory studies of Texas soil algae. Univ Texas Publ 6022: 1–72 Deason TR, Silva PC, Watanabe S, Floyd GL (1991) Taxonomic status of the species of the green algal genus Neochloris. Plant Syst Evol 177: 213–219 Dill EO (1895) Die Gattung Chlamydomonas und ihre nächsten Verwandten. Jahrb Wiss Bot 28: 323–358 Ehrenberg CG (1838) Die Infusionsthierchen als vollkommene Organismen. Voss, Leipzig Ettl H (1965a) Beitrag zur Kenntnis der Morphologie der Gattung Chlamydomonas Ehrenberg. Arch Protistenkd 108: 271–430 Ettl H (1965b) Untersuchungen an Flagellaten. Österr Bot Z 112:701–745 Ettl H (1970) Die Gattung Chloromonas Gobi emend. Wille (Chlamydomonas und die nächstverwandten Gattungen I). Beih Nova Hedwigia 34: 1–284 Ettl H (1976) Die Gattung Chlamydomonas Ehrenberg (Chlamydomonas und die nächstverwandten Gattungen II). Beih Nova Hedwigia 49: 1–1122 Ettl H (1979) Die Gattungen Carteria Diesing emend. Francé und Provasoliella A. R. Loeblich (Chlamydomonas und die nächstverwandten Gattungen III). Beih Nova Hedwigia 60: 1–226 Ettl H (1983a) Chlorophyta I- Phytomonadina. In Ettl H, Gerloff J, Heynig H, Mollenhauer D (eds) Süßwasserflora von Mitteleuropa, Bd. 9. Gustav Fischer, Stuttgart, New York Ettl H (1983b) Taxonomische Namensänderungen und Neubeschreibungen unter den Phytomonadina (Chlorophyta). Nova Hedwigia 35: 731–736 Ettl H (1988) Unterschiedliche Teilungsverläufe bei den Phytomonaden (Chlorophyta). Arch Protistenkd 135: 81–101 Ettl H, Gärtner G (1995) Syllabus der Boden-, Luftund Flechtenalgen. Gustav Fischer, Stuttgart-JenaNew York Ettl H, Schlösser UG (1992) Towards a revision of the systematics of the genus Chlamydomonas (Chlorophyta). 1. Chlamydomonas applanata Pringsheim. Bot Acta 105: 323–330 Fabry S, Kohler A, Coleman AW (1999) Intraspecies analysis: comparison of ITS sequence data and gene intron sequence data with breeding data for a worldwide collection of Gonium pectorale. J Mol Evol 48: 94–101 Felsenstein J (1985) Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783–791 Friedl T (1996) Evolution of the polyphyletic genus Pleurastrum (Chlorophyta): inferences from nuclearencoded ribosomal DNA sequences and motile cell ultrastructure. Phycologia 35: 456–469 Friedl T (1997) The Evolution of the Green Algae. In Bhattacharya D (ed) The Origin of Algae and Plastids. Springer-Verlag, Wien, pp 87–101 Geitler L (1954) Echte Oogamie bei Chlamydomonas. Österr Bot Z 101: 570-578 Gerloff J (1940) Beiträge zur Kenntnis der Variabilität und Systematik der Gattung Chlamydomonas. Arch Protistenkd 94: 311–502 Gobi C (1899/1900) Über einen neuen parasitischen Pilz, Rhizidiomyces ichneumon nov. sp., und seinen Nährungsorganismus Chloromonas globulosa (Perty). Scr Bot Horti Univ Imp Petropol 15: 251–272 Gordon J, Rumpf R, Shank SL, Vernon D, Birky Jr CW (1995) Sequences of the RRN18 genes of Chlamydomonas humicola and C. dysosmos are identical, in agreement with their combination in the species C. applanata (Chlorophyta). J Phycol 31: 312–313 Goroschankin JN (1891) Beiträge zur Kenntnis der Morphologie und Systematik der Chlamydomonaden. II. Chlamydomonas reinhardi Dang. und seine Verwandten. Bull Soc Imp Nat Moscou NS 5: 101–142 Greuter W, Brummitt RK, Farr E, Kilian N, Kirk PM, Silva PC (1993) NCU-3 Names in Current Use for Extant Plant Genera. Koeltz Scientific Books, Königstein Molecular Phylogeny and Revision of Chlamydomonas Hasagawa M, Kishino H, Yano T (1985) Dating the human-ape splitting by a molecular clock of mitochondrial DNA. J Mol Evol 22: 160-174 Heimke JW, Starr RC (1979) The sexual process in several heterogamous Chlamydomonas strains in the subgenus Pleiochloris. Arch Protistenkd 122: 20-42 Hepperle D, Nozaki H, Hohenberger S, Huss VAR, Morita E, Krienitz L (1998) Phylogenetic position of the Phacotaceae within the Chlamydophyceae as revealed by analysis of 18S rDNA and rbcL sequences. J Mol Evol 47: 420-430 Heunert H-H (1973) Microtechnique for the observation of living microorganisms. ZEISS Information 79: 74–85 Hoops HJ, Witman GB (1985) Basal bodies and associated structures are not required for normal flagellar motion or phototaxis in the green alga Chlorogonium elongatum. J Cell Biol 100: 297–309 Jaenicke L, Kuhne W, Spessert R, Wahle U, Waffenschmidt S (1987) Cell-wall lytic enzymes (autolysins) of Chlamydomonas reinhardii are (hydroxy)proline-specific proteases. Eur J Biochem 170: 485–491 Kishino H, Hasegawa M (1989) Evaluation of the maximum likelihood estimate of the evolutionary tree topologies from DNA sequence data, and the branching order of the Hominoidea. J Mol Evol 29: 170-179 Lewin RA (1984) Chlamydomonas sajao nov. sp. (Chlorophyta, Volvocales). Chin J Oceanol Limnol 2: 92–94 Marin B, Melkonian M (1999) Mesostigmatophyceae, a new class of streptophyte green algae revealed by SSU rRNA sequence comparisons. Protist 150: 399–417 Marin B, Klingberg M, Melkonian M (1998) Phylogenetic relationships among the Cryptophyta: Analyses of nuclear-encoded SSU rRNA sequences support the monophyly of extant plastid-containing lineages. Protist 149: 265–276 Matsuda Y, Koseki M, Shimada T, Saito T (1995) Purification and characterization of a vegetative lytic enzyme responsible for liberation of daughter cells during the proliferation of Chlamydomonas reinhardtii. Plant Cell Physiol 36: 681–689 McCourt RM (1995) Green algal phylogeny. Trends Ecol Evol 10: 159–163 Medlin L, Elwood HJ, Stickel S, Sogin M (1988) The characterization of enzymatically amplified eukaryotic 16S-like rRNA-coding regions. Gene 71: 491–499 Melkonian M (1990) Phylum Chlorophyta. Class Chlorophyceae. In Margulis L, Corliss JO, Melkonian M, Chapman DJ (eds) Handbook of Protoctista. Jones and Bartlett Publ, Boston, pp 608–616 Melkonian N, Preisig HR (1984) An ultrastructural comparison between Spermatozopsis and Dunaliella (Chlorophyceae). Plant Syst Evol 146: 31–46 299 Melkonian M, Surek B (1995) Phylogeny of the Chlorophyta: congruence between ultrastructural and molecular evidence. Bull Soc Zool Fr 120: 191–208 Morita E, Abe T, Tsuzuki M, Fujiwara S, Sato N, Hirata A, Sonoike K, Nozaki H (1999) Role of pyrenoids in the CO2-concentrating mechanism: comparative morphology, physiology and molecular phylogenetic analysis of closely related strains of Chlamydomonas and Chloromonas (Volvocales). Planta 208: 365–372 Müller OF (1786) Animalcula infusoria fluviatilia et marina. Hauniae Nakayama T, Watanabe S, Mitsui K, Uchida H, Inouye I (1996) The phylogenetic relationship between the Chlamydomonadales and Chlorococcales inferred from 18S rDNA sequence data. Phycol Res 44: 47–56 Nozaki H, Ohta N, Morita E, Watanabe MM (1998) Toward a natural system of species in Chlorogonium (Volvocales, Chlorophyta): A combined analysis of morphological and rbcL gene sequence data. J Phycol 34: 1024–1037 Nozaki H, Misawa K, Kajita T, Kato M, Nohara S, Watanabe MM (2000) Origin and evolution of the colonial Volvocales (Chlorophyceae) as inferred from multiple chloroplast gene sequences. Mol Phylogen Evol 17: 256–268 Page RDM (1996) TREEVIEW: an application to display phylogenetic trees on personal computers. Comp Appl Biosci 12: 357–358 Pascher A (1927) Volvocales. In Pascher A (ed) Süßwasser-Flora Deutschlands, Österreichs und der Schweiz, Heft 4 Gustav Fischer, Jena, pp 1–506 Pocock MA (1962) Algae from De Klip soil cultures. Arch Mikrobiol 42: 56–63 Posada D, Crandall KA (1998) Modeltest: testing the model of DNA substitution. Bioinformatics 14: 817–818 Reize IB, Melkonian M (1989) A new way to investigate living flagellated/ciliated cells in the light microscope: immobilization of cells in agarose. Bot Acta 102: 145–151 Rogers SO, Bendich AJ (1985) Extraction of DNA from milligram amounts of fresh, herbarium and mummified plant tissues. Plant Mol Biol 5: 69–76 Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, Horn GT, Mullis KB, Erlich HA (1988) Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239: 487–491 Schlösser UG (1972) Chlamydomonas reinhardtii (Volvocales) Asexuelle Fortpflanzung. Film E1318, Institut für den wissenschaftlichen Film, Göttingen. Schlösser UG (1976) Entwicklungsstadien- und sippenspezifische Zellwand-Autolysine bei der Freisetzung von Fortpflanzungszellen in der Gattung Chlamydomonas. Ber Dt Bot Ges 89: 1–56 300 T. Pröschold et al. Schlösser UG (1982) Sammlung von Algenkulturen, Pflanzenphysiologisches Institut der Universität Göttingen. Ber Dt Bot Ges 95: 181–276 Schlösser UG (1984) Species-Specific Sporangium Autolysins (Cell-Wall-Dissolving Enzymes) in the Genus Chlamydomonas. In Irvine DEG, John DM (eds) Systematics of the Green Algae. The Systematics Association Special Volume 27, Academic Press, London, pp 409–418 Schlösser UG (1994) SAG- Sammlung von Algenkulturen at the University of Göttingen. Catalogue of strains 1994. Botanica Acta 107: 113–186 Schlösser UG (1997) Additions to the Culture Collection of Algae since 1994. Bot Acta 110: 424–429 Silva PC (1980) Remarks on algal nomenclature VI. Taxon 29: 121–145 Starr RC (1955) Isolation of sexual strains of placoderm desmids. Bull Torrey Bot Club 82: 261–265 Starr RC, Zeikus JA (1993) UTEX- the culture collection of algae at the University of Texas at Austin. J Phycol Suppl 29: 1–106 Swofford DL (1998) PAUP*, phylogenetic analysis using parsimony (* and other methods), version 4.0b2. Sinauer Associates, Sunderland, MA Tamura K, Nei M (1993) Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Mol Biol Evol 10: 512–526 Watanabe S, Floyd GL (1989) Comparative ultrastructure of the zoospores of nine species of Neochloris (Chlorophyta). Plant Syst Evol 168: 195–219 Wille N (1903) Algologische Notizen IX.-XIV. Nyt Mag Naturvidensk 41: 89–185