Critical Plants noticed on the Excursion. 401 
a “ G P ” form. The form figured in Eng. Bot. t. 825 is the “GP c” 
type. Mr. Druce recently (in Brit. Bot. Rep. for 1910, 546-7 (1911) ) 
pointed out that the name S. dilleniana must replace the others. 
It is obvious, I think, that the forms which have been named 
according to the methods of systematists have no stronger theo¬ 
retical claim to be thus brought into prominence than those which 
have not been named. Ought we therefore to name the latter 
combinations, or reject the names of the former? To be logical, 
we should do either the one or the other. 
I wish to ask, however, if it is not the case that the kind of 
variation which occurs in 5. dilleniana is not amenable to the parti¬ 
cular form of naming which systematists in all countries adopt ? Is 
it not more desirable, in cases like this, where (so far as can be 
judged without experiments) characters hybridise and rehybridise 
and continue to be transmitted whole, to name the resulting forms 
by some symbolic method, such as is adopted by Mendelian workers? 
If this were done in systematic works, the combining characters 
would be indicated in the specific description, and the segregating 
forms would not be there named at all, except perhaps symbolically; 
and systematists need no longer, in these cases, encumber their 
works with an almost interminable number of sub-specific divisions 
and names. 
To show how systematic works may be, and indeed often are, 
encumbered by names and synonyms of the forms here being dis¬ 
cussed, let us consider the case of the characters G, g, P, p. Rouy 
(loc. cit.) divides these primarily by the characters of large petals 
(P) and small petals (p), thus having two varieties, P and p; 
then each variety is sub-divided by the character of glaucousness (G) 
and non-glaucousness (g), giving the sub-varieties G p and g p. But 
other botanists might first divide the species by some other corres¬ 
ponding pair of characters, e.g., G and g, and sub-divide these by 
the remaining pairs; and this plan would result in sets of different 
names. 
Again, who would say there are no additional characters to G, g, 
P, p, C, and c, say, for example, with regard to the width of the 
leaves, the width of the petals, the degree of splitting of the petals, 
and so on ? Is each additional combination, as it is discovered, to 
receive a name ? 
If, in these and similar cases, the symbolical method of naming 
were adhered to, systematists would have fewer opportunities of 
becoming controversial in a vain attempt to settle which particular 
