EVOLUTIONARY TKXDKXCIKS AMONG GYMN'OSl'KRMS 421 



have disappeared; and of course there is no trace of it when the 

 archegonia are eliminated. 



The distribution of archegonia may be considered in this connec- 

 tion, although the tendencies do not appear to be general. It is becom- 

 ing evident that the ])osition of archegonia is related to the position 

 of the i)ollen tube, which sometimes reaches the embryo sac before 

 the archegonium initials are selected. In cases where the pollen tube 

 assumes a lateral position in reference to the gametophyte (as in 

 Sequoia and Widdringtonia), it has been demonstrated that the latter 

 responds by the selection of numerous deep-seated and laterally 

 placed archegonium initials. It may be inferred, therefore, that the 

 usual micropylar position of archegonia is due to the usual micropylar 

 position of the ti]) of the pollen tube. It may be that numerous 

 scattered and rather indefinitely placed archegonia were a feature of 

 ancient gymnosperms, but there is no evidence for it; on the contrary, 

 the few sections of paleozoic ovules that reveal archegonia, and also 

 the archegonia of heterosporous pteridophytes, suggest the o])posite 

 conclusion. In any event, they tend to become definite in number 

 and are then organized in two ways: either as individual archegonia, 

 each with its own jacket and chamber; or as an archegonial complex, 

 with a common jacket and chamber. The latter may seem to be a 

 specialized condition, exhibited chiefly by Cupressineae, but it also 

 seems to be the natural condition from which to derive the free eggs 

 of W elwitschia and Gnetuni when archegonia are eliminated from the 

 ontogeny. 



THE MALE GAMETOPHYTE 



It is perhaps impossible as yet to determine the character of the 

 male gametophyte of the paleozoic gymnosperms. The evidence is 

 accumulating that it comprised many more cells than do the game- 

 tophytes of most living gymnosperms; but it is not demonstrable 

 whether these supernumerary cells were vegetative or spermatogenous. 

 There are instances of supernumerary cells of both kinds among 

 living gymnosperms, so that they furnish no clue; and the same is 

 true of heterosporous j)teridophytes. The balance of probability, 

 however, is in favor of the view that they were in the main sperma- 

 togenous. 



