424 MORPHOLOGY OF GYMNOSPERMS 



is proceeding in the direction of the ehmination of one of the male 

 cells. Among the Abietineae and Araucarineae the sperms are 

 not distinctly organized cells, but are male nuclei lying free in 

 the cytO]jlasm of the body cell, and this cytoplasm is not always 

 sharply delimited from that of the pollen tube. That male cells and 

 male nuclei have been derived from swimming sperms is a reasonable 

 inference, and if our proposed phylogeny is true, this series of changes 

 has taken place in the evolution of Coniferales. Such changes do 

 not indicate necessarily the relative ages of the groups in which they 

 are observed, but rather their relative rapidity in different groups. 



Oogenesis among gymnosperms does not cover so great a range 

 of reduction, but it is none the less evident. What must be regarded 

 as the most primitive condition among gymnosperms is the association 

 of a definite ventral canal cell with the egg, as in Ginkgo and Pinus. 

 x^mong cycads and many conifers, however, the wall between nuclei 

 of the ventral canal cell and of the egg has been eliminated, so that 

 finally the former cell is represented only by its nucleus; and in Torreya 

 it is probable that even this nucleus does not appear. A further 

 simplification of oogenesis is shown by Welwitschia and Gnetum, in 

 which no definite eggs are formed, but the egg is represented only by 

 a nucleus in a more or less definite sheath of cytoplasm. 



There are certain combinations of characters among the Conifer- 

 ales that may prove to be correlations, but the range of observation is 

 too limited as yet to establish such a claim. In any event, they deserve 

 consideration. From those forms (Thuja, for example) with definitely 

 organized male cells (the most primitive condition of the sperm 

 among conifers) both the male prothallial cells and the wall separat- 

 ing ventral and egg nuclei have been eliminated (probably the most 

 advanced condition of these structures among conifers). On the 

 other hand, those forms (Pinus, for example) which do not organize 

 definite male cells have retained the male prothallial cells. Three 

 structures are concerned in these combinations, and in all of them 

 there is evident a tendency toward elimination, or at least simplifi- 

 cation. It is further evident that this tendency is not shown by all 

 three structures simultaneously. The inferences are that the several 

 lines have proceeded independently, and that the condition of no one 

 of these structures indicates the relative age of the plant to which it 



A 



