Floras of Great Britain and Central Europe. 237 
Species Limits and Local Races. 
This last task is both difficult and important. It constantly 
leads us into fresh paths, and in more recent times appears to have 
become more confused rather than clearer, since the views held in 
different countries and by different authors as to the extent of the 
limits of single species differ from one another more widely than 
was formerly the case. The confusion has been greatly added to 
by preposterous changes in nomenclature, which pursue their course 
in the vain hope of attaining a goal that will satisfy everyone. At 
the same time the complexity of the subject has naturally been 
increased by new phylogenetic and ecological studies. Thus it is 
evident that a hook like H. C. Watson’s Topographical Botany , at 
one time authoritative, is no longer a safe guide, even in fundamentals, 
and that Druce’s “ List ” with its additions of sub-species and 
varieties is indispensable. In the variability of species we are con¬ 
fronted by old questions in a new form, especially such questions as 
whether variations of the same character over wide areas really 
belong together, or whether they are not rather made up of similar 
ecological variants of independent local origin, analogues to one 
another in various separated parts of the whole area, and thus have 
not the status of a single monophyletic race or sub-species. 
Let me cite a single instance. The question arose as to whether 
a form of Salsola Kali occurring on the Southport dunes was to be 
considered as S. Tragus. Druce’s list does not consider this form 
as indigenous. It is interesting to recall the doubt which such 
critical observers as Mertens and Koch 1 felt a hundred years ago 
on this “ species ” established by Marschall von Bieberstein, and 
how they tried to express its characters in a diagnosis. But it 
appears much more probable that the Tauric-Siberian and West 
European forms are local and analogous, not forming directly 
connected series, and that the circle of variation of each is thus to 
be drawn more narrowly. For this reason it was very valuable to 
hear an expression of the views of the English specialists on the 
limitation of the range of variation of their species. Very many 
common species constantly polymorphic in their German stations 
appeared to me to look very different in England, to represent in 
fact slightly differentiated examples of local endemism. How 
otherwise can we interpret the Birches for instance ? How difficult 
it appears to be satisfactorily (it is not permissible to say “ correctly ”) 
to limit the range of variation of Betula odorata, pubescens and 
1 Deutschlands Flora, 2, pp. 321-2. 
