HAROLD HEATH AND BLAKE C. WILBUR. 



nal segments varied from nineteen to twenty-one. In specimens 

 of T. angiisticollis, in the same stage of development, the trans- 

 verse head diameter was 2.56 mm., and the number of antennal 

 segments was twenty-two in the greater number of instances, al- 

 though twenty or twenty-one was not an uncommon count. 



Fully developed soldiers in recently established colonies are 

 subject to a wider range of variation than is the rule regarding any 

 other instar. In some instances the head width is but slightly 

 greater than that of individuals in the penultimate instar ; and the 

 number of antennal joints may be that of the foregoing stage or 

 even less, occasional insects having but nineteen segments. The 

 average head width of 26 soldiers of T. nevadensis was 2.64 mm., 

 while 1 8 soldiers of T. angiisticollis measured 2.96 mm. in trans- 

 verse diameter. The number of antennal joints varied from 

 twenty to twenty-three or even twenty-four. 



From the foregoing data it becomes evident that the two species 

 under consideration are structurally closely related, and in each 

 instance their development follows essentially the same path. 

 Yet they have never been found to interbreed in a state of nature, 

 and everywhere each colony of either species invariably appears to 

 maintain its integrity. The number of antennal joints in the same 

 instar is, with rare exceptions, one more in the case of T. angus- 

 ticollis, and its greater head width is also a distinctive feature, so 

 much so in fact that even without the royal pairs the two species 

 can be distinguished without much difficulty. It may be added 

 that the arrangement of the spines over the body has not been 

 found to be of much diagnostic value even where the various 

 instars are involved. Each additional molt adds to the number of 

 bristles, but their location is much the same in both species. 



As noted previously, this study deals only with the develop- 

 ment of the soldier caste in relatively young colonies. In older 

 communities, where the reproductive caste has put in an appear- 

 ance, the history of the soldier is somewhat more complicated. 

 To what extent this is true can only be determined on the basis 

 of much more extended observations, including experimental 

 work, measurements of head, brain, gonad and perhaps other 

 anatomical characters, a study that is well under way with a 

 strong probability that some of the more important problems can 

 be solved within the next few months. 



