95 



Cactaceae, partly on account of its ordinary leaves. May it be 

 that its basal placentation is likewise primitive in the family? 



From the point of view of fertilization it is an advantage to have 

 the ovules near the stigma, that is. at the top of the' ovary, as in 

 apical placentation. In basal and parietal placentation the pollen 

 tube has to go around the longer distance by the walls of the ovary. 

 But, from the point of view 7 of moisture and nourishment of the 

 developing ovule, basal placentation is better, being nearer to the 

 source of moisture and nourishment. Axile placentation combines 

 both advantages, connection for direct fertilization from above and 

 facility for direct nourishment from the mother plant beneath. 



Basal placentation with few ovules or a single ovule occurs in 

 many wind-pollinated plants. The uncertainty of wind pollination 

 evidently gives a chance for only a few or for only one ovule per 

 flower to be fertilized. Basal placentation is naturally associated 

 with few ovules, parietal placentation usually with many ovules. 

 In the transition from gymnospenny to angiospermv there must 

 have been a change from wind-pollination to one of incipient and 

 presumably not very effective insect pollination. " No insect pol- 

 lination, no closed ovary " - that was the view of Arber and 

 Parkin. Angiospermv to them meant the protection of ovules 

 from insect injury. Where there is a single ovule with two or 

 more carpels it is reasonable to assume that the ancestral form had 

 at least one ovule to each carpel. In Gymnosperms the ratio of 

 pollen-grains to ovules is high, and numerous ovules probably were 

 not a characteristic of earl}' Angiosperms. It may be a question 

 whether it is desirable to use the term placentation for other than 

 fused carpels. However, in the Gymnosperms we have in Cyras 

 circinalis something like parietal, and in Juniperus something like 

 basal attachment of ovules. The possibility of basal placentation 

 being primitive corresponds to some extent with the views of 

 Thompson and Hagerup, who say that central placentation, as in 

 Primula, is primitive. However, Dianthus cannot be considered 

 as having primitive placentation. 



The family Aristolochiaceae has wandered into many positions 

 in the various systems. In the Engler system it is admitted to 

 occupy " an isolated position." Long ago Robert Brown suggested 

 that Rafflesia and Aristolochiaceae belong near 1 'assifloraceae. 



