II32 



A TEXTBOOK OF THEORETICAL BOTANY 





; ; 

 7 



Fig. 1 1 05. — Cycadeoidea ingens. Bisporangiate flower of one of the Jurassic Ben- 

 nettitales. Stamens large and pinnate, ovules on a central, conical receptacle. 

 The ovules are gymnospermic but otherwise there is close similarity to an 

 angiospermic flower. (After Wieland.) 



to find flowers of diflterent sexual types produced not only on different plants 

 of the same species, but even on the same individual plant. It is unneces- 

 sary therefore to labour the point that a reduction in sexual expression is not 

 only possible but frequently occurs. The opposite view has received a 

 stimulus of late years from the discovery of the Caytoniales, a group of 

 Mesozoic fossil plants, possibly derived from the Pteridospermae, which 

 have arrived at a condition of angiospermy by the inclusion of the ovules 

 within a closed, and apparently stigmatiferous cupule. Their reproductive 

 shoots were unisexual and the possibility that they might have been in the 

 evolutionary line towards the living Angiosperms turned attention to the 

 idea of the primitive nature of unisexual flowers. Acceptance of this view 

 implies that organs of both sexes can arise where only one w^as formerly 

 present. 



Comparative studies show that the distinction of the two sets of sporan- 

 gial parts is by no means fixed. In Conifers, where the cones are normally 

 unisporangiate, it is not uncommon to find abnormal cones which are 

 bisporangiate, that is to say producing both micro- and megasporangia. In 

 Angiosperms also many anomalous cases occur of ovules borne on stamens, 

 pollen grains formed in ovules and organs of mixed character generally, 

 w'hich serve to show^ that it is quite possible for sporangia of both types to 

 appear in a zone w-hich has previously been unisexual. There is however a 

 considerable gap between such abnormalities and the beautiful regularity of 



