Spongomorpha was one of several genera segregated from Conferva by Kützing (1843b: 273). It was characterized by intricate branching that produces ropy thalli and was assigned two species, S. uncialis (O.F. Müller) Kützing (Conferva uncialis O.F. Müller, 1778: 7, pl. 771(1); type locality: unspecified, but somewhere in Denmark of that period, including Holstein, Norway, Iceland, Greenland, and the Faeroes) and S. congregata (C. Agardh) Kützing (Conferva congregata C. Agardh, 1824: 111; syntype localities: North Sea and the Kattegat). Later, Kützing (1845: 237) moved the genus to the Ectocarpaceae and added seven species, three of which are ectocarpoid rather than chlorophycean.
Apparently unaware of Kützing's work, J. Agardh (1846a: 104) also divided Conferva into several genera, including Acrosiphonia, which was characterized as having elongate upper segments that continue to divide and short lower segments that do not divide. Three species were assigned to Acrosiphonia: Conferva lanosa Mertens ex Roth (1806: 291--292, pl. IX; type locality: Norderney, East Frisian Islands, Germany) with C. congregata cited as a synonym, C. arcta Dillwyn (1809 [1802--1809]: 67, pl. E; type locality: Bantry Bay, County Cork, Ireland), with C. centralis Lyngbye (1819: 161, pl. 56C; type locality: ``in sinu Othiniensi'' [Odense Fjord, Denmark]) cited as a synonym, and C. membranacea Hofman-Bang ex C. Agardh (1824: 120--121; type locality: St. Croix, Virgin Islands).
The only species that the original circumscriptions of Spongomorpha and Acrosiphonia had in common was Conferva congregata. This species was based on Lyngbye's concept of C. uncialis (Lyngbye, 1819: 160--161, pl. 56B), which in C. Agardh's opinion differed from true C. uncialis O.F. Müller. If C. congregata had been designated lectotype of both Spongomorpha and Acrosiphonia, the two generic names would be nomenclatural synonyms. As it happened, however, C. congregata was designated lectotype of Spongomorpha by H.K. Phinney (Index Nominum Genericorum card 26/06342, distributed 22 July 1958) while C. arcta was designated lectotype of Acrosiphonia by Jónsson (1991: 176). Phinney's lectotypification was inadvertent since he erroneously attributed it to Setchell & Gardner (1920b: 221). Those authors, however, did not choose between the two original species of Spongomorpha, merely stating that since they ``have generally been regarded as varieties of one'' and since C. `aggregata' (a lapse for congregata) was ``considered to be a synonym of Conferva lanosa Roth'', the type of the generic name may be said to be S. lanosa (Roth) Kützing. Phinney's lectotypification was accepted by Jónsson (1991: 173) but apparently was not known to Burrows (1991: 70), who designated S. uncialis (O.F. Müller) Kützing as lectotype. Even if Burrows had provided the first lectotypification, it would not be tenable according to van den Hoek (1963: 225), who expressed the opinion that Müller's illustration suggests a species of Cladophora, but certainly not Spongomorpha.
Jónsson (1991) made a special effort to find suitable type specimens of the type species of Spongomorpha and Acrosiphonia. He selected a specimen (no. 7089) in the Agardh Herbarium (in LD) as the lectotype of Conferva congregata (hence of Spongomorpha) and assigned it to S. aeruginosa (Linnaeus) van den Hoek (1963: 225), a synonym of S. lanosa (Roth) Kützing. As lectotype of C. arcta (hence of Acrosiphonia), he chose a specimen in Dillwyn's herbarium in the National Museum of Wales (NMW). Because there are two sheets of the holotype collection in Dillwyn's herbarium, the one chosen by Jónsson as ``holotype'' should properly be considered a lectotype. Since Spongomorpha and Acrosiphonia have different type species, both names are usable.
Kützing (1849: 417) had second thoughts regarding the generic distinctness of Spongomorpha and reduced it to the rank of a subgenus or section of Cladophora, but without mentioning Acrosiphonia, thus suggesting that he was unaware of J. Agardh's work. He added the character of slender rhizoidal branches and assigned 21 species to the section. Later, Kützing (1854: 16, pls. 74--84; 1855: 29, pl. 100) again treated Spongomorpha as an independent genus. Farlow (1881: 50) and De Toni (1889: 335), however, chose to treat it as a subgenus and section, respectively, of Cladophora, and neither author mentioned Acrosiphonia. Wille (1890b: 118) followed De Toni, and in addition listed Acrosiphonia as a synonym of Cladophora.
A detailed study of the species assigned to Spongomorpha and Acrosiphonia was made by Kjellman (1893b), who concluded that they were congeneric but distinct from Cladophora. While admitting that Spongomorpha had priority over Acrosiphonia, he chose the latter name for the genus, in which he recognized 20 Scandinavian species. Most if not all of Kjellman's new species have been reduced to synonymy.
The first person to accept both generic names was Wille (1899: 281; 1900: 238), who found that species of this complex fell into two groups with regard to the number of nuclei in each cell. Those with uninucleate cells were assigned to Spongomorpha while those with multinucleate cells were assigned to Acrosiphonia. Most subsequent authors have adopted this treatment, with which the generic lectotypifications are also in agreement. Jónsson (1991) confirmed that the lectotype of Spongomorpha has uninucleate cells while that of Acrosiphonia has multinucleate cells.
A second character of possible diagnostic value was introduced when Fan (1957; 1959) and Hollenberg (1957; 1958b) discovered that Spongomorpha coalita (Ruprecht) Collins [Acrosiphonia coalita (Ruprecht) Scagel et al.], a multinucleate species from northwestern America, is the gametophyte in a life history of which the sporophyte is an endophytic vesicle that had been considered an independent species, Codiolum petrocelidis Kuckuck (1894: 259, fig. 27; type locality: Helgoland, Germany). According to Jónsson (1958; 1959a; 1962; 1970), the gametophyte alternating with C. petrocelidis at Roscoff, Atlantic France, is Acrosiphonia spinescens (Kützing) Kjellman (1893b: 51), a multinucleate species originally described from Morbihan, Atlantic France (as Cladophora spinescens Kützing, 1849: 418). Since A. coalita and A. spinescens are distinct species, the vesicular sporophyte of the former must be called a codiolum stage and not assigned to C. petrocelidis. The host of the codiolum stage of A. coalita is Petrocelis franciscana Setchell & Gardner, the sporophyte of Mastocarpus papillatus (C. Agardh) Kützing, while the host of C. petrocelidis is Petrocelis cruenta J. Agardh, the sporophyte of Mastocarpus stellatus (Stackhouse) Guiry, or Haemescharia hennedyi (Harvey) Vinogradova & Yakovleva (1989: 751), a red crust vegetatively similar to Petrocelis but with a life history involving isomorphic gametophytes and sporophytes. It may be useful to point out that the combination H. hennedyi was also proposed by Wilce & Maggs (1989: 1477) in a publication issued less than a month after the publication of the paper by Vinogradova & Yakovleva (7 June vs. 10 May).
Jónsson (1959b; 1962; 1966) further showed that Spongomorpha aeruginosa (as S. lanosa) alternates with Chlorochytrium inclusum Kjellman (1883: 392--393, pl. 31: figs. 8--17; syntype localities: various in the Arctic Ocean), a vesicular endophyte in Dilsea, Polyides, and other fleshy Rhodophyceae. Kornmann (1961) observed the development of S. aeruginosa (as S. lanosa) from zoospores produced by vesicular endophytes in both Polyides and Petrocelis, and suggested that Chlorochytrium inclusum and Codiolum petrocelidis are phenotypes of the sporophyte of a single species. Jónsson (1963) was skeptical of Kornmann's conclusions, emphasizing that at Roscoff zoospores of Codiolum petrocelidis always gave rise to A. spinescens rather than to Spongomorpha lanosa. Kornmann (1964b) confirmed his previous findings and suggested that what Jónsson was calling C. petrocelidis at Roscoff was not conspecific with C. petrocelidis from Helgoland (the type locality), thus accounting for the difference in the gametophytes.
The discrepancies between Jónsson's findings at Roscoff and Kornmann's at Helgoland underscore inherent difficulties in treating life histories involving an alternation of somatic phases, whether the considerations be morphological, taxonomic, genetic, or phylogenetic. The two somatic phases are subjected to different selective pressures, resulting in evolutionary pathways that are separate yet held together by a common genome. When one phase is a crust or a unicellular endophyte while the other phase is an erect thallus, the difference in selective pressures is readily perceived. It seems reasonable to assume that differential rates and directions of evolution result in pairs of species in which one phase is indistinguishable from one another while the second phase embodies the attributes that define the species. The high degree of morphological variability of the filamentous thalli of the Acrosiphonia-Spongomorpha complex has been recognized taxonomically (Kjellman, 1893b). A commensurate diversity among the endophytic sporophytic vesicles corresponding to Kjellman's microspecies has not been demonstrated, but it seems reasonable to assume that some of these species have indistinguishable sporophytes. Jónsson (1963), in fact, suggested that sporophytes from different species might cohabit the same host. Conversely, Kornmann's finding that zoospores from both Codiolum petrocelidis and Chlorochytrium inclusum give rise to filaments of Spongomorpha lanosa has a possible alternative interpretation, namely, that two species are involved, with distinguishable sporophytes alternating with indistinguishable gametophytes.
Kornmann (1962) proposed that Spongomorpha be circumscribed to include species with a life history involving an alternation of heteromorphic phases, leaving Acrosiphonia to encompass species with other life histories. Using life history as a generic character in this complex cuts across the lines drawn by nuclear number since A. coalita and S. aeruginosa, both of which have an alternation of heteromorphic phases, have multinucleate and uninucleate cells, respectively.
Several life histories have been reported for members of this complex in addition to alternation of heteromorphic phases, but whether these are carried out in nature as well as in culture remains to be demonstrated. In Acrosiphonia grandis Kjellman (1893b: 78--79, pl. III: fig. 17; type locality: Finnmark, Norway) the zygote develops into a codiolum stage, which grows directly into a filamentous plant without formation of zoospores (Kornmann, 1970a). Meiosis has not been observed, but Kornmann (1970b: 301) suggested that it might occur within the codiolum vesicle. In A. sonderi (Kützing) Kornmann (1962: 236), based on Cladophora sonderi Kützing (1845: 208; type locality: Helgoland), filamentous plants of unknown ploidy recycle by means of biflagellate zoospores, sexual reproduction not having been observed (Kornmann l.c.). In A. arcta (Dillwyn) Gain (1912: 31), Kornmann (1962) found that filamentous plants of unknown ploidy recycle by means of isogametes. Meiosis has not been observed, but Kornmann (1970b: 301) suggested that it might occur during germination of the zygote. Later, Kornmann (1965: 40) referred to this species as a ``monoecious diplont'', implying that meiosis occurs in gametogenesis. Kornmann (1962: 232) placed A. spinescens in the synonymy of A. arcta and thus was faced with the problem of addressing the marked discrepancy between Jónsson's demonstration of an alternation of heteromorphic phases and his own results. He suggested (Kornmann, 1964a: 115) that Jónsson had mistaken enlarged zygotes of a species of Ulothrix for the codiolum stage of Acrosiphonia.
Jónsson (1963), who considered A. spinescens and A. arcta to be closely related, noted that uncopulated gametes of A. spinescens could develop directly into filamentous plants and suggested that parthenogenesis might have displaced syngamy in A. arcta at Helgoland. Later, Jónsson (1964) showed that in A. spinescens syngamy was not always followed by karyogamy. The resulting false zygotes gave rise to micto-haploid filamentous plants. In cultures of A. arcta from Helgoland supplied by Kornmann, Jónsson (1968) found widespread suppression of karyogamy. In view of Jónsson's results, Kornmann (1970b) recognized A. arcta and A. spinescens as two distinct species.
While not always in agreement as to the species being investigated, Jónsson (1968) and Kornmann (1970b) shared the view that the basic life history in the Acrosiphonia-Spongomorpha complex is an alternation of heteromorphic phases, from which all other life histories that have been demonstrated in culture have arisen ontogenetically and/or phylogenetically. Thus, life history details would seem to be diagnostic only at the level of species.
We are thus left with two alternatives in treating the Acrosiphonia-Spongomorpha complex: either distinguish two genera on the basis of nuclear number or recognize only one genus (as by Burrows, 1991). We prefer the first alternative.
A potentially complicating factor developed during the preparation of this catalogue when we found that the type of Anadema orientalis J. Agardh (1846a: 103) is conspecific with Spongomorpha indica Thivy & Visalakshmi (1963c: 495), a species that has multinucleate cells and is thus referable to Acrosiphonia. Because Anadema comprised only the one species, that generic name is a taxonomic synonym of Acrosiphonia. The two names were published by J. Agardh in the same paper and thus have equal priority. Considering that Anadema has not been treated by any author subsequent to J. Agardh, we have retained Acrosiphonia.