286 BOTANICAL GAZETTE [APRIL 
But even conceding the importance claimed for this habit of early 
fruiting, it is still only one point of six left open for consideration, 
and we may sum up all the characters of /wnartoides that we have been 
reviewing as follows: y 
1. Zhe bud: Commonly pubescent, of varying degrees, only rarely 
smooth ; character not of specific value. 
The lamina: \ts cutting and shape of the segments; character 
wholly unreliable, or of varietal importance only. 
3. Relative position of the sterile and fertile divisions: Sterile division 
nearly sessile, or short-stalked ; scarcely of more than varietal signifi- 
cance. 
4. Haéitat: Character variable and unimportant. 
5. Spores: Shapes and markings as in the other forms (including 
type) ; no specific differences. 
6. Fruiting time: Evidence incomplete and inconclusive, and 
doubtfully of more than varietal significance at the most. 
My conclusion, therefore, is that Lamarck’s Osmunda biternata had 
best remain where Milde placed it, under B. ternatum as a good 
variety. 
But here arises another question which it may be well to consider 
briefly. There is a tendency on the part of some of our later authori- 
ties to give to well marked varieties specific recognition, on the ground 
that it facilitates scientific investigation ; and since, at the best, species : 
are merely the arbitrary definitions devised by man for the convenience 
_of study, there would be no serious objection to this if it were not for 
the numerous intermediate forms that constantly confront us and 
demand recognition. It is all very well to say that such forms may be 
disregarded for purposes of classification, but we cannot dispose of 
them in that way. They are an essential part of Nature’s great scheme 
of evolution, and just as much entitled to recognition as more definite 
forms. I have known collectors who were in the habit of throwing tO 
4 
q 
mee 
one side all puzzling forms that could not be placed readily, so as not 
to disturb the arrangement of species in their collections, but no close 
student of nature would be content with such practices. Nature shows 
| groups, orders, races, and the nearer we approach to her methods the — 
more accurate will our knowledge of her great works become. ae : 
for this reason, therefore, that I prefer that broader recognition of aie 
