ce ON THE USE OF POLLEN 
greater comprehension are subordinate to, or contained within groups of 
greater extension.’ It follows, therefore, that we can sub-divide the larger 
groups by characters of minor value; and again, different groups possess 
affinities towards each other in the ratio of the less extension and greater 
comprehension of the group to which they-Zare] subordinate ; so that by 
sub-dividing the larger groups, by characters:the next in importance, and 
repeating this process, as expressed in the solution of the problem, we 
rm groups composed of elements having a greater number of resem- 
blances amongst themselves than exist between them and the members of 
other groups. We judge by the considerable increase of mutual affinities, 
of the proper application of a character to segregation of a small group, 
and therefore we continue sub-division only so long;as characters can be 
found;whose application co-exists with an increasing {multiplication of 
affinities. And lastly, as groups founded on; similar characters (Proposi- 
tion Ist), agree in a number of affinities, we can, by approximating the 
larger groups at those minor ones that are founded on similar characters, 
place the several groups in relations more or less approached in the ratio of 
their mutual affinities. 
Having in the preceding pages endeavoured to determine 
the principles of classification in general, I will now direct 
myself to the consideration of the taxonomic value of the 
pollen. The value of a character being the compound value 
of the organ and of its abstract mode, it is necessary that we 
should consider these things separately ; and first, as regards 
the physiological importance of the organ, the different me- 
thods proposed for determining this, reduce themselves to 
one, viz. its extension; the organ most generally present, 
must be considered as the most essential. Now, trie 
by this standard, the pollen must, at the least, be of equal 
value with the embryo; for it is notorious, that all embryo- 
nate plants are phenogamic. But if the observations of 
M. Corda be received, the pollen is of still greater import- 
ance: for this inquirer states it to be the origin of the em- 
bryo. He says that in Conifere he has observed the pollen- 
tubes penetrating the nucleus of the ovule to the very 
bottom, where they discharge a matter that becomes trans- 
formed into the embryo. And again, Fritzche, the latest and 
most accurate observer, inclines to consider the globule in 
Chara as an anther. And the general tendency among 
