THE FEEDING HABIT OF TERMITE CASTES. 299 



definitely. Thus one worker can at least support itself and two 

 soldiers or two second forms. How much more a worker can do, 

 was not determined. 



Why do the second and third form young adults not take 

 food when placed by themselves? A very large number of 

 individuals of the second and third form castes were examined 

 < in -fully for protozoa and in no instance were protozoa found 

 after the final molt. In this connection it is interesting to note 

 that every time any individual of any cast molt-. it- intestinal 

 proto/oa are lost, but in all molts, except the last one of the second 

 and third form-, the protozoa are regained very quickly. Some- 

 times though very rarely indeed as the inte-tine slips off dur- 

 ing the molting pn>cess, a portion of it is eat en before the pn>to/oa 

 die; but, .1- a rule, if an individual at the time of molting is 

 placed to it-ell, it does not regain its intestinal proto/oa and. 

 afi .1 CMiiM-quence, dies within three weeks or thereabout-. But 

 the lir-t form regains its intestinal protozoa alter the Iinal molt. 

 Why is it able to do this, while the second and third forms 

 c.innot ' < >b\ iou-ly, one reason why the second and third forms 

 .iimot live by themselves after the final molt is becaii-e they 

 have lost their protozoa. Hut this does not explain uhy the\ 

 do not take food, for workers continue to feed alter tin- pmto/na 

 ha\e l>een removed from them experimentally. 



Why and how are the protozoa lost. 1 * In an effort to throw 

 light on tin- que-tion. the second form was studied during tin- 

 short period iu-t before the final molt in \\hich it can be distin- 

 guished externalU from the first form. It \\a- found that the 

 pr>ti./oa gradually disappear during this pt-ri"d and that leu, 

 it aii\ , are |>re-ent at the time of the final molt. It must be 

 noticed that then- is also a dimunition in the pn>tn/oa in the 

 lirst form during the period preceding the final molt, but not BO 

 great a- in the second form, and the prot<>/<>a ne\er di-appear 

 entirely. This pr >v;re--ive disappearance of the proto/oa of the 

 reproducti\e castes at this time is perhaps brought about by 

 one or both of two things; namely, the salivary -e. reiion which 

 i- taken during this period of rapid change and development 

 de-troys the pn>to/,.a and, as a consequence, wood-feeding must 

 be gi\en up; or, so much salivary secretion is taken that wood- 

 feeding is thus made unnecessary and is, therefore, given up, 



