142 



CARNEGIE INSTITUTION OF WASHINGTON. 



fourth generation selected the bean {Phaseolus vulgaris) with respect 

 to seedUng characters. For the most part he has not succeeded in 

 isolating strains of definite type, but in a few cases this has proved 

 possible. 



"A strain in which the number of cotyledons and of primordial leaves is far 

 larger than normal has this year been grown in many thousands of individuals. 

 Normally the number both of cotyledons and of primordial leaves is two. The 

 great increase and the high variability in the number of both is shown by the 

 accompanying table, based on a sample of seedlings of the 1915 culture. It is to 

 be noted that the highest frequency for both number of cotyledons and number 

 of leaves is double that which normally occurs. A peculiarity of this line of 

 plants is the abundance of ascidia, already studied in another form, the ash, 

 at this Station by Dr. Shull. At least two other highly but not completely 

 abnormal strains are also available. These differ greatly in their morphologi- 

 cal characteristics, but require further study before they can be tersely charac- 

 terized. 



"Material progress in the isolation of strains in chlorophyll production has 

 been made. Several types of plants have been secured from those producing 

 no chlorophyll to those normal in this regard, but the propagation of such 

 strains, except through normal plants, is very difficult. Enough has been done, 

 however, to indicate the heritability of a number of types." 



SELECTION OF FLUCTUATING VARIATIONS. 



It is still extremely profitable to study the limits of effective selection. 

 The work of Castle might lead us to conclude that through selection 

 practically any character might be gradually modified and — if the 

 process was prolonged sufficiently — might be modified to almost any 

 extent. Two extensive experiments that we are carrying on bear on 

 this point. 



(o) Dr. Banta has continued the selection of Daphnia for greatest 

 and for least sensitiveness to Ught through more than 150 generations 

 (compare Year Book No. 13, p. 131). The results are being prepared 

 for publication. 



(b) Dr. MacDowell has made the following three experiments upon 

 vinegar flies with extra bristles, outlined in Year Book No. 13, pp. 

 132-133: 



To continue selection till there remains no question as to its effect. — A 

 year ago the report covered selections for 21 generations. At present 



