246 



WILLIAM LAWRENCE TOWER. 



All tests of the heritability of this modification have been 

 made with the sixth or later generations, for the reason that the 

 alteration was apparently not at its full development until 

 then, and in that the only means of distinguishing between the 

 two conditions has been the survival test under the winter 

 conditions at Chicago, earlier generations in which there were 

 survivals of this test, would have given only results that had an 

 error in them not possible of estimation. After the sixth genera- 

 tion at Tucson, all the cultures have thus far given the uniform 

 result in crossings with the normal at Chicago, regardless of the 

 direction of the cross, and these summated results are given in 



Table IX. 



TABLE IX. 



In making these tests of the heritability of the alteration in 

 the Tucson cultures, the FI series were all from matings of 

 the first summer generation, whose progeny would hibernate, 

 and for the Fo tests, the over-wintering generations at Tucson 

 and Chicago were crossed at Chicago, and the second summer 

 generation tested. The results show in all that the FI population 

 are in no instance able to survive the test, that the condition of 

 the Tucson cultures is dominant, but that there is a segregation 

 in F 2 and a survival from the test that is for the total series 

 24.4 per cent., about the expected proportion of the extracted 

 recessives. These in breeding in F 3 show no indication of the 

 Tucson condition, either in pairs or mass cultures, all surviving 

 the hibernation test in F 4 in all respects like the normal Chicago 

 materials. 



The behavior in inheritance shows, therefore, no points of 

 interest aside from the fact of its heritability, and the fact 

 that a condition that has arisen rapidly, but apparently progres- 

 sively, behaves in crossing when fully established as a Mendelian 



