180 BERKELEY, ON CRYPIOGAMIC BOTANY. 
we can arrive at anything like accurate views of geographic botany, 
or the distribution of plants over the globe, is by a correct estima- 
tion of species. If two Floras be formed on different principles—while 
in the one the species are accurately limited, and forms which vary only in 
some subordinate point, and not in essential characters, are grouped under 
one common name; in the other, not only every marked variety, but even 
accidental variation, is elevated to the rank of a species,—it is impossible 
to form any correct comparison, and this is the more necessary in Crypto- 
gains than elsewhere, because the species have notoriously such a wide 
diffusion, and because their technical, though not their essential characters, 
are so very variable. The great point in all these cases is never to describe 
from single or imperfect specimens, where there is some form evidently very 
closely allied. It may not be possible, perhaps, always to avoid error, but 
a little caution will be most advantageous, both as to one’s own individual 
character as a botanist, and to science in general. And if species are once 
accurately characterised, there will be far less difficulty than may be imagined 
as to genera. Nothing is more vain than to run down botanists as mere 
makers of species, as though it did not take as much knowledge and tact to 
limit species well, as to ascertain a few detached microscopical facts without 
deriving any general views from such study, or ever seeing the relative 
bearing of such observations. ‘The physiologists of the present day, at 
least too many who have some name in science, are absolutely doing the 
very thing which they profess to despise in species-makers. A proposer of 
bad, ill-defined species is no promoter of science; still less is the so-called 
physiologist who draws from isolated half-observed facts, conclusions which 
the very next observation may entirely destroy. We may regret, indeed, 
sometimes the over-caution of the prince of physiologists, but such over- 
caution is ten thousand fold more praiseworthy, and tends more to the 
advancement of science, than crude, hasty, and ill-considered theories founded 
on imperfect observations, because what it does bring forth is essentially a 
KTnpa €¢ ae, and, even when incomplete, is a sure stepping-stone for the 
acquirement of some further eminence.” 
As we at first stated, it is in the structure of the repro- 
ductive organs in Cryptogamic plants that most has been 
done by the aid of the microscope. On this subject we 
extract the followimg passage from Mr. Berkeley’s luminous 
introductory chapter. 
“ Tt is desirable, again, before entering further on this argument, to say 
a few words on the reproductive organs of Cryptogams, at least on the 
female organs, for there is little or no similarity between the male organs of 
Cryptogams and Phenogams. There are no proper pollen globules, no 
germinating of a cell to bring the walls in contact with the embryo-sac; nor 
is there any agreement between the mode of generation of the grumous 
matter or fovilla and the spermatozoids. 
“In the more simple cases there is nothing at all analogous to flower, but 
certain privileged cells are separated from the threads or compact tissue of 
the matrix, whether naked, or produced within a special tube or sac, and 
constitute the fruit. ‘These germinate almost exactly like pollen grains, and 
reproduce the species. ‘There are, sometimes, several kinds of spores upon 
a plant, all capable of reproduction, though differing in appearance. These 
spores, then, are homologues of the individual cells of Phaenogams, which, 
at times, are equally capable of reproduction in the shape of buds. 
“The spores, or what have the appearance of spores, do not always 
reproduce the plant immediately, even in plants of such a low grade as 
