414 



THE AMERICAN NATURALIST [Vol. XLIV 



viduals, because in the vast majority of eases such individuals will be 

 extreme fluctuating variants rather than mutants. 



Jennings 7 states in his paper that ' 1 Systematic and 

 continued selection is without effect in a pure race, and 

 in a mixture of races its effect consists in isolating the 

 existing races, not in producing anything new." And 

 concludes 5 that "Until some one can show that selection 

 is effective within pure lines it is only a statement of 

 fact to say that all the experimental evidence we have 

 is against this." 



Recently Pearl has brought forth a very noteworthy 

 contribution in this line. His evidence is based upon 

 the work which is being done at the Maine Agricultural 

 Experiment Station to determine the effect of selection 

 on fecundity and the inheritance of fecundity in poultry. 

 His conclusions may be summed up as follows: 



Selection fur hiub egg production carried on for nine consecutive 

 years did not lead to any increase in the average production of the 

 flocks. 



The present data give no evidence that there is a sensible correlation 

 between mother and daughter in respect to egg production, or that 

 egg-producing ability (fecundity) is sensibly inherited. 



In this experiment the daughters of " 200-egg" hens did not exhibit, 

 when kept under the same environmental conditions, such high average 

 egg production as did pullets of the same age which were the daughters 

 of birds whose production was less than 2(10 eggs per year. 



Professor Waugh, 8 of the Massachusetts Agricultural 

 Experiment Station, has been making some studies of 

 the variation of peas and the inheritance of the differ- 

 ent fluctuating characters. He has found that the coeffi- 

 cients of heredity for the different characters are very 

 low and are not very significant in the case of any one 

 character. The two characters, length of vine and num- 

 ber of pods per vine, show a coefficient of heredity of 

 .170 and .158, respectively. At the same time he found 

 that there were certain lines that did reproduce their 

 characters to an appreciable extent, which is along the 



