452 BOTANICAL GAZETTE [December 



megasporangmm 



microsporocarp 



step in this reduction. 



What causes the abortion in one case of thirty, in another of thirty- 



me 



determined. It would seem largely a matter of chance, since there 

 is no external factor or group of factors responsible. Very often 

 megasporocarp and microsporocarp lie side by side, having begun 

 their development from two apical cells which were formed by the 

 division of a single cell, and having continued growth with their sur- 

 roundings as nearly alike as it is possible to make them. On the 

 other hand, there is no internal condition that we can say, definitely, 

 must be responsible for this. There is of course the possibility that 

 the whole thing is a question of food supply, as Strasburger sug- 

 gested. If there should be some change in the cells of the stalk 

 below the megasporangium making them impermeable to any of the 

 materials which are necessary for growth in the sporangium, the inhi- 

 bition of growth there would be a natural result. That the food supply 

 thus kept from the megasporangium should then be used for the further 

 development of the microsporangium at the base, seems a plausible 

 supposition. In cases where the cells of the stalk later grow up into 

 the megasporangial cavity, one w^ould have to suppose a regaining of 

 permeability of the cells to food materials. Such changes in the cells 

 from permeability to a given substance to impermeability and back to 

 permeability are quite possible physiologically. 



SUMMARY 



I. 



megasporocarp and microsporocarp 



are exactly similar. Each begins development with a single me 

 sporangium and a sporocarp wall. When the megasporangium " 



mother cells are in synapsis 



some 



microsporangia. Growth in the m 



megasporangium now goes on simultaneously for some time. The 

 tapetal walls of the megasporangium disappear and the mother cells 

 separate from each other and round off, while the microsporangia 

 at the base continue develonmont. diffprincr m nnnparance from the 



