SHORTER ARTICLES AND CORRESPONDENCE 



IS THERE A SELECTIVE ELIMINATION OF OVARIES 

 IN THE FRUITING OF THE LEGUMINOSiE ? 



During the past few years I have had occasion to have counted 

 the number of ovules formed per pod in series of many thou- 

 sands of pods of various species of Leguminosae. While engaged 

 at this work a side problem of considerable interest presented 

 itself. Since time to follow this up has not been forthcoming 

 for a couple of years, I will publish my data and suggestions for 

 the benefit of any unoccupied botanist. 



It is evident that the distribution of ovules, as seen in series of 

 mature pods, does not necessarily represent that of the ovules in 

 the newly formed ovaries. It is possible that the distribution of 

 ovules per ovary in the flower buds may have been modified as 

 the fruits matured by a selective elimination of certain classes of 

 pods. This selective elimination might modify mean or varia- 

 bility or both. One such case is very probably to be found in 

 ovaries with but a single ovule in species normally producing 

 several. If the single ovule be fertilized it may develop into a 

 seed or not, according to various circumstances ; but if it fail to 

 develop, the pod will, in most cases at least, fall from the plant. 

 Whether the same factor is at work in the case of ovule classes of 

 a higher order can only be determined by special investigation, 

 but it seems quite possible that such an influence might affect 

 very materially the constants of mature as compared with newly 

 formed ovaries. 



In the spring of 1907 I attempted to get some light on this 

 question by the collection of material at different times from 

 individual red bud trees. Cercis Canadensis, growing in the 

 North American Tract of the Missouri Botanical Garden. 



The flowers of Cercis are produced in an umbel-like cluster 

 from the old wood. These clusters are composed of from two to 

 seventeen flowers. Many of the inflorescences produce only one 

 to four fruits. The per cent, of flowers which mature fruits 

 varies widely from year to year and from tree to tree, but it is 

 evident that there is a very heavy elimination of ovaries. The 

 question to be answered is: Have the eliminated ovaries a dif- 

 556 



