4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 'jd 



termites (family Kalotermitidae) have no worker caste. The male 

 continues to live with and fertilize the female, and both have a post- 

 adult growth — ^remarkable in the case of the queens of the more 

 specialized family Termitidae. Both of these reproductive forms are 

 long lived. 



In case of the absence of parent reproductive forms of the winged 

 caste, brachypterous (pi. 2, figs. 3 and 4) or apterous (pi. 2, figs. 

 6-8*) reproductive forms of both sexes are present; these forms 

 are distinct castes. There are no fertile workers or soldiers among 

 termites. Among the higher termites the neuters are polymorphic; 

 there may be from one to three different types of soldiers and one 

 to two types of workers of the same species. 



The castes of termites, however, vary with the family and genus 

 as does the progeny of a single pair of termites. The character of the 

 progeny and the definite ratio between the castes also vary with 

 the genus and the age of the colony. Hence, a genetic formula for 

 termites would vary with the genus and age of the colony; the first 

 brood of the incipient colony in some genera consists entirely of 

 neuters. It would take years of study and experimentation to work 

 out such a formula. 



In no case at the present day do termites exist which do not produce 

 as their progeny other castes in addition to the winged, sexual forms 

 which must have been their presocial ancestral condition, i. e., a single 

 male and female sexually associated. 



All the castes of termites are determined in the germ plasm of 

 the species and are due to its inherent properties. As Thompson 

 states, they are " segregants, in the sense of the offspring of Oeno- 

 thera lamarkiana, arising generation after generation by the splitting 

 and recombination of the genes of a heterozygous parent form." 

 Hence the varied modifications among termites really are not due to 

 "adaptations" to environment but let us rather say that favorable 

 modifications have persisted. 



The social life among termites must have begun some time before 

 the Tertiary period, since wingless, sterile castes have been found in 

 gum copal of the late Tertiary period, i. e.., polymorphism existed 

 during the late Tertiary period ; winged termites occur in the shales 

 of the Tertiary period. 



^ This illustration of a remarkable apterous form from Java may be compared 

 with the nymph of the first form of the same species (pi. 2, figs. 9 and. 10) and 

 with the normal apterous form of Reticulitcrmcs Havipcs Kol. illustrated in 

 Bull. 108, U. S. Nat. Mus., figure 63, p. 108 and plate 30, figure 2. 



