452 EDWIN CARLETON MacDOWELL 



This result indicated that three crosses with a third race had sufficed 

 practically to eliminate whatever differences that had been produced 

 in the minus and plus races respectively by long-continued selection 

 in opposite directions. Those differences were based on residual he- 

 redity, not in changes in the gene proper. 



These results favor the widely accepted view that the single gene is 

 not subject to fluctuating variability, but is stable like a chemical 

 compound of definite constitution and changes only similarly, by definite 

 steps (mutations in the sense of Morgan, not of De Vries). 



The five years since the review included in the first number 

 of this series have yielded large amounts of evidence upon the 

 selection problem. Zeleny and Matoon ('15) selected for in- 

 crease and decrease in the number of facets in the bar-eyed 

 mutant of Drosophila; three generations separated the flies into 

 two groups with different means. The increase in the regression 

 of the means to the parental populations is taken to indicate 

 that selection cannot be expected to carry the bar eye all the 

 way back to normal. Further experiments with this same 

 character were carried on by May ('17). Using flies that had 

 vestigial wings besides bar eyes, selections were made for five 

 generations (2000 facet counts), with no results after the first 

 generation; this first selection brought about a separation of the 

 means of the two lines. Sterility in this stock led to the recom- 

 mencement of the experiment using normal long-winged flies. 

 Six generations of selection in this stock (7000 counts) effected a 

 continuous divergence in the means of the two lines, due to the 

 rise of the means in the high line. Return selections from the 

 sixth generation still were successful. Unfortunately, no con- 

 stants are given for the results of crossing the two lines nor are 

 data available for their calculation. Although the author in- 

 cludes in the summary the statement that 'Hhe results of these 

 experiments indicate that the hereditary differences in this race 

 of Drosophila are due to a large number of small factors," no 

 evidence discriminating between the theories of. selection depend- 

 ing on separate units, and on one varying unit, is afforded. 



Roberts ('18) has selected vestigial winged flies for thirty-four 

 generations with no general advance in the means. Crossing 

 selected inbred flies to normals changed the means, but did not 



