Spore Formation in Torreya californica. 139 
V. —The Phylogenetic Significance of the Male Gametophyte 
in Gymnosperms. 
In Torreya californica there are no signs of any sterile pro- 
thallial cells, either in the ripe pollen-grain or during its development. 
Coulter and Chamberlain 1 point out that these structures are usually 
very ephemeral, and may easily escape observation, and they con¬ 
clude, “ We venture the opinion, therefore, that one or two vege¬ 
tative cells, more or less evanescent, will be found to be of common 
occurrence among Gymnosperms.” Since this was written, however, 
Coker has carefully looked for these cells in Taxodiuni 2 , Cupressus 
(4 spp), Taxus baccata and four vars., J uni perns (2 spp), Chamcc- 
cyparissas (5 spp), Callitris (1 sp.) Cryptomeria japonica, and Thuja 
orieutalis, and reports their absence in all these cases 3 . Lawson 4 
also states that the most searching examination failed to reveal a 
trace of prothallial cells in Sequoia seiupervirens. I think therefore 
that we are justified in assuming that male prothallial cells are 
entirely absent in a certain number of Gymnosperms. Their pre¬ 
sence or absence is a character to which we should be naturally 
inclined to attach some phylogenetic importance. In the case of 
ferns Heim 5 has brought forward evidence to shew that indications 
as to the affinities of the different groups may reasonably be looked 
for in the gametophyte as well as the sporophyte generation. Fern 
prothalli are notoriously susceptible to the influence of external 
conditions, and Heim carried out cultural experiments to determine 
what characters were dependent on such conditions, and hence 
were of no phylogenetic value. After eliminating these he found 
that growth by an apical cell or a marginal meristem, the presence 
or absence of hairs, and the mode of dehiscence of the antheridia,. 
were characters which were constant throughout groups which 
would be regarded as related on the evidence of the sporophyte 
generation alone. It is clear that the characters of the gametophyte 
can only be used with great caution as indications of affinity. But 
there is nothing peculiar about this; the question of how to distin¬ 
guish ancestral from adaptive characters is one that confronts us in 
every branch of the study of structure. It is the crux of the whole 
situation, and if it could be eliminated morphology would lose at 
once both its difficulty and its charm. In the present state of our 
1 Morphology of Spermatophytes, Part I., New York, 1901. 
2 loc. cit. 
3 Note in Bot. Gaz., May, 1904, ref. to Science, N.S., 3: 424, 1904. 
loc. cit. 
6 Carl Heim. Untersuchungen iiber Farnprothallien. Flora, 1896. 
