122 C.E. Moss. 
420 (1796). There are several forms in England of this highly 
variable species in addition to the one noticed (p. 309 (1911) ) by 
the party in Norfolk. Last year, in a fen in Huntingdonshire, I 
noticed that the following characters occurred in all their theoretical 
combinations :—(1) glaucousness (G); (2) non-glaucousness (g) or 
greenness; (3) large petals (P), about once-and-a-half to twice as 
long as the sepals; (4) small petals (p), about as long as the sepals; 
(5) many-flowered cymes (C); and (6) few-flowered cymes (c). All 
the forms grew in watery places in the fen; but, as regards their 
morphological characters, the “g p” forms were with difficulty 
distinguishable from the allied, and also variable S. graminea. In 
fact, by botanists of Bentham’s school, such forms might be held to 
justify a return to the Linnzan view that S. graminea and S. 
dilleniana (=S. palustris=S. glauca) are simply forms of a single 
polymorphic species. 
Now, many of the combinations of the above six characters 
have received names from systematists. Adopting, for the time 
being, the names in Rouy’s Fl. de France, iii, 233 (1896), the following 
plants occurred in the fen in question:—G PS. palustris var. 
communis; ¢ P=S. palustris var. communis subvar. viridis; G p= 
S. palustris var. parviflora; g p=S. palustris var. parviflora subvar. 
viridis. Thus, if only the four characters, G, g, P, p, are considered, 
all the possible combinations are named. If, however, other 
combinations of two characters are considered, there are the 
following combinations unnamed :—G C, Ge, g C, gc; and PC, 
pC, Pc, pe. 
There are also the following eight theoretical combinations to 
consider:—G PC,GPc,GpC,GpaggPC,gPc,gpC gpa 
With regard to these, three of them have been named by Magnier 
in Bull. Bot. Soc. France, xxviii, 82 (1881); and G p C=S. litigiosa 
Magnier; g PC is S.manchit Magnier; andG Pe is S. heterophylla 
Magnier. It will be seen that Magnier uses three characters, and 
names three combinations; but the following combinations he leaves 
without names:—G PC,Gpc,gpC,gPec,andgpc. 
The first form of the species to receive a binominal was g pc 
which was named S. dilleniana by Moench; and, as this name is 
valid, it must, by the rules of nomenclature, be used to designate 
the whole species. The later but more familiar names must 
unfortunately be discarded. Of these, S. palustris, like S. dilleniana, 
referred toa “g” form as neither Linnzeus nor Retzius mentions 
the glaucous character; and Withering’s name S. glauca refers to 
