DEPARTMENT OF EXPERIMENTAL EVOLUTION. 8l 



Quantitative Investigations of Fertility and Fecundity in Plants. — These 

 studies have been carried forward by Dr. Harris, as time permitted, along 

 the lines indicated in Year Book No. 8. Some of the data have been pub- 

 lished and others are in press, while a very large mass of material awaits the 

 final processes of reduction. 



Investigations of Variation, Correlation, and Inheritance of Quantitative 

 Characters in Garden Beans. — As stated in Year Book No. 8, where an indi- 

 cation of the chief problems may be found, Dr. Harris's attention since com- 

 ing to the station has been chiefly devoted to these investigations. The cul- 

 tures of the present season, involving the planting of about 20,000 seeds, 

 were devoted to a comparison under like conditions of the offspring of an- 

 cestors grown for two generations under very dissimilar environmental con- 

 ditions. Owing to the large amount of routine counting, weighing, tabula- 

 tion, and calculation the results from these experiments can not be ready for 

 a considerable length of time. 



Studies in Vegetable Teratology. — The results of about 125,000 dissec- 

 tions of normal and proliferous fruits of Passiflora, grown in 1908 and 1909, 

 are being studied statistically by Dr. Harris to determine various points con- 

 cerning the nature of teratological variation and the frequency of the several 

 types of anomaly to which a fruit may be subject. Until this routine can be 

 carried through, further cultures of Passiflora have been suspended. Breed- 

 ing experiments with teratological beans have been continued. Approxi- 

 mately 50,000 seedlings have been studied. 



Quantitative Studies of Selective Elimination. — Natural selection is one of 

 the factors of organic evolution least investigated by the quantitative biolo- 

 gist. Dr. Harris has attempted to gain some light on the structural characters 

 which are least fitted for development by studying the characters of ovaries 

 which fail to develop to maturity as compared with those which develop. 

 The analysis of data for Staphylea, collected in the spring of 1906 and 1908, 

 justifies the following conclusions: 



The failure of ovaries to develop to maturity is not random, but is de- 

 pendent upon structural peculiarities which would admirably be classed as 

 "fluctuating variations." In short, the elimination is selective, whether we 

 work within the range of variation of the ovaries of an individual or extend 

 the studies to ovaries from a number of individuals. The changes brought 

 about by selective elimination are : 



1. An increase in the mean number of ovules. 



2. A decrease in variability of number of ovules. 



3. A decrease in radial asymmetry as measured by the standard deviation of number 



of ovules per locule for an individual ovary. 



4. A decrease in the relative numbers of ovaries with one or more locules with an 



odd number of ovules. 



5. Possibly an increase in the mean number of locules per fruit. 



Preparations were made for carrying this work forward in 1910, but a 

 severe freeze killed all the material. 



6 — YB 



