The Classification of Conifers. 245 
3. Extent of sporogenous (“spongy”) tissue in ovule. 
4. Sporogenesis. 
5. Place of deposition of pollen. (Not logically included in 
this class of characters, but probably connected phylogenetically 
with such, and therefore best considered here.) 
6 . Development and structure of male gametophyte. 
7. Early development of female gametophyte. 
8 . Method of cell formation in female gametophyte. 
9. Position of archegonium initials, and of archegonia. 
10. Fertilization. (Relative size of male and female nuclei, etc.) 
11. Development of pro-embryo, and formation and first divisions 
of embryo. 
12. Later development of embryo. (Number of cotyledons, etc.) 
13. Structure and thickness of megaspore membrane. 
Of these, numbers 1 and 4, while they are doubtless important 
characters, show no marked differences, as far as is known, within 
the Coniferales, and may therefore be left out of account. Numbers 
2 and 3 also show only minor differences, except in one or two 
special cases, where such differences merely serve to emphasize 
more important divergences in other respects. 
Place of deposition of pollen. In the very large majority of Conifers 
the pollen is deposited on the apex of the nucellus, though 
occasionally it may be caught before reaching the base of the 
micropyle, as is said to be normally the case in Pseudotsuga, 
where a constriction half way down the micropyle partially divides 
it into two chambers, the pollen being retained in the upper. In this 
genus also the pollen appears to be carried into the upper chamber 
by an inward folding of the apex of the integument, which develops 
a special stigmatic surface. Such specialization, while of great 
interest as being unique among gymnosperms, is probably of much 
less phylogenetic importance than the “ protosiphonogamic ” polli¬ 
nation described by Thomson in Araucaria , and lately found to be 
also characteristic of Agathis. Here the pollen is deposited on the 
cone scale or ligule or in the axil of the cone scale, and not, as a rule, 
on the ovule at all; according to Thomson it may be as much as 
3 cm. from the nucellus in Araucaria , though it is somewhat 
surprising to learn that the whole length of the scale is as much as 
3 cm. at the time of pollination. At that time the ovule is scarcely 
differentiated, and there is never a micropylar tube, as there is in 
most Conifers, the nucellus projecting beyond the end of the 
integument. In Saxegothcea also, the nucellus projects in a similar 
