Floral Evolution. 
397 
tendency in the ancestry to respond to long-established climatic 
conditions, e.g., the environment of the dense tropical forest. 
Within the space at our disposal we cannot aspire to complete¬ 
ness even in the most general examination of the principles of 
classification ; but the foregoing hints may serve to indicate a few 
of the modes of attack upon the all-absorbing questions of phylogeny, 
and to portray the relative value of the various kinds of characters. 
In the search after affinities it is all-important that no consideration 
should be left unnoticed before any conclusion is arrived at; and 
too much stress cannot be laid upon that most obvious, most general, 
and yet so often-neglected principle that Presumptions of affinity 
should not he based upon a single character; the characters employed 
as criteria should be as numerous and as distinctive as possible; and 
biological evidence should be carefully examined in the light of the 
history in descent of the plant-forms under investigation. 
No more fit conclusion to these studies can, perhaps, be found 
than a special notice of the vast field for research which they have 
revealed at every turn. Surely, no branch of botany has suffered 
more through the lack of practical research than has the study of 
the Systematics of Flowering Plants; and this by reason of the 
dearth of known facts of floral development in the great majority of 
the families of Angiosperms. And for the little knowledge that we 
do possess, we must needs turn for the most part to the work of 
more than half a century ago. Indeed, if we shall prove to have 
succeeded in stimulating ever so little the current of research in the 
direction in question, we shall feel that our labour has not been vain. 
