282 ON THE FORMS OF POLLEN-GRAINS. 
stock of truly natural characters to assist us in the definition of the 
differences between several kindred species or orders of the vegetable 
ingdom. 
In this point of view, I have already shown the importance of the 
structure and function—form, size, and contents—of the elementary or 
other cells. And such is often the excellence of this kind of character 
that by it alone, without any other whatever, a mere shapeless and 
minute fragment of a plant, at any period of its growth, may be most 
easily and certainly distinguished, sometimes from any other species of 
its genus and frequently from every other nearly allied Order. Nay, 
by this very character simply, a plant may be tried and found wanting 
in a close or true affinity with the Order under which it has been 
placed by systematic botanists. On these points, descriptions and 
references will be found in my contributions to the ‘ Popular Science 
Review,’ October, 1865, and ‘Quarterly Journal of Microscopical 
Science,’ January, 1866 
As to the real sih of the pollen for ordinal or generic characters, 
it will probably rise higher when the facts have been more completely 
ascertained and Gone At present, they have not been sufficiently 
studied ; and so manifold are they, and so vast is this single field of 
observation, that a long time may pass before they can be fully realized 
and reduced to a comprehensive and useful met 
Meantime, I hope to excite more attention to the subject by show- 
ing that even closely allied plants may present a sufficient difference 
in their pollen for specific diagnosis. Since my notice of the pollen 
of Ranunculus arvensis (Ann. Nat. Hist., July, 1865), I have made 
numberless comparative examinations of the pollen of the yellow- 
flowered divided-leaved British Crowfoots; and, as the results have _ 
been always constant and certain, a woodcut is now given of the out- 
lines of the pollen-grains of two species which stand close together 
in Professor Babington’s * Manual of British Botany. It will thus be 
seen how the pollen-grains of Ranunculus arvensis differ, in their 
roughness and much larger size, from those of Ranunculus hirsutus ; 
and the pollen of R. arvensis differs similarly from that of the other 
species of the section just mentioned. 
But it is surprising to find that there is a regular différence of size 
between the pollen-grains of Lotus corniculatus and L. major, plants 
so very closely related that the latter is considered by some eminent 
