Compilation
Marsilea coromandelina

The great variability of this species has vexed many students of the past. A. Braun in his first publication on the classification of Marsilea (1864: 416) indicated that coromandelina and trichopoda were mere “Formen” of one species. Despite this, in his more comprehensive study (1871) he maintains the two taxa as species but states in his on trichopoda that it is an intermediate between coromandelina and muscoides. J.G. Baker (Handb. Fern-Allies, 1887), who did not provide a key to the individual species, followed A. Braun in maintaining them as separate species. Later workers studied the genus only on a regional basis which accounts for the increase in the number of synonyms. C. Christensen (1932) defines M. microphylla as follows “its nearest relative is perhaps M. muscoides Leprieur from Senegambia, a species unknown to me …”. The myth that everything occurring in Madagascar must be endemic may also have been a contributory factor towards his decision to describe a new species. In fact M. microphylla is virtually identical with the specimen on which M. muscoides is based. Bremekamp’s M. trichocarpa was founded on a very immature specimen. In more recent material from South Africa (e.g. Codd 4096) the sporocarps do not differ at all from those of tropical African or even Indian material. I have not yet seen any aquatic forms of coromandelina from Africa. All the specimens from southern Angola, Namibia and South Africa are terrestrial forms with very narrowly cuneate leaflets. The interstitial pellucid streaks in the leaflets of M. coromandelina are usually shorter, less prominent, and fewer than in M. distorta. They are much more developed in terrestrial forms than in aquatic ones, and in the latter case one often has to examine the leaflets in strong transmitted light in order to recognise them. The only species with which M. coromandelina can perhaps be confused is M. gibba, which occasionally has slightly bordered sporocarps. On herbarium labels as well as inliterature, e.g. Baker (1887), one finds the specific epithet spelled “ coromandelica ”, based on Burman’s Flora Indica (1767). Since Burman not only published his name in synonymy (under quadrifolia Linnaeus) but was also not sure of its taxonomic status, his name is invalid. Linnaeus (Mant. Pl. Alt.: 308 (1771)) mentions the name as minuta ß coromandelina (as ‘cormandeliana’) without giving it the rank of a variety. He just refers to Burman’s name. Thus the first valid publication was effected by Willdenow, and the type is certainly the specimen in Willdenow’s herbarium.