PHYLOGENETIC ORIGIN OF TERMITE CASTES. Iiy 



bulk of the brain, the relative size of brain and head, the com- 

 pound eyes, and the sex organs. When the young reproductive 

 nymphs have attained a length of 1.3-1.4 mm. other structural 

 differences are observable that further differentiate them into 

 two kinds of individuals which later develop into two of the 

 three adult reproductive castes, namely: adults of the first form, 

 with long wings, and adults of the second form, with short, scale- 

 like wing pads. The ontogeny of the third adult reproductive 

 caste, without wing stubs or wing pads, is yet to be worked out 

 in R. flavipes. At a later period in the ontogeny body length 

 about 3.75 mm. the worker-soldier nymphs differentiate into 

 two kinds of nymphs which develop into the two sterile adult 

 castes, the workers and the soldiers. 



As a result of the discovery that the ontogenetic cause of 

 termite castes is intrinsic, two questions arise. First, can this 

 intrinsic ontogenetic cause be determined by cytological means, 

 i. e., by examination of the phases of ovogenesis and spermato- 

 genesis? Second, what is the phylogenetic origin of the termite 

 castes? It is with this second question that the present paper 

 deals. 



The phylogenetic mode of origin of the castes of termites, 

 could it be determined, might have an important bearing upon 

 the general question of the evolution of species. The condition 

 of polymorphism, or the existence of several structural forms 

 within a species, indicates that the parent or ancestral form had a 

 tendency to vary. Two categories of variations are recognized 

 today: (i) the variations which have arisen abruptly and are 

 qualitatively unlike the parent condition, and which breed true 

 to type, "sports" of Darwin, "mutations" of DeVries, "dis- 

 continuous variations" of Bateson; (2) the "fluctuating" varia- 

 tions which have arisen gradually and are merely quantitative 

 variations of the parent condition "Darwinian variations," 

 "fluctuations" of DeVries, "continuous variations" of Bateson. 

 It is well known that the selectionists claim that the latter type of 

 variation is heritable 1 while on the other hand the mutationists 

 and the pure line school either doubt or deny that fluctuating 

 variations are inherited. 



1 Castle (1917); Jennings (1917). 



