970 PROTOZOA AND OTHER ANIMALS 



incubation, starvation, and oxygenation was studied by Yamasaki 

 (1937b). Kirby (1931) stained the axostyle of Trichomonas termop- 

 sidis brown in Lugol's solution; and stated that this may be taken, as 

 Alexeieff pointed out in the case of Tritrichomonas augusta, to be indic- 

 ative of the possible presence of glycogen in the axostyle. In the light 

 of these results, it seems likely that carbohydrate is stored as glycogen 

 in these Protozoa. (Cleveland et al. [1934], however, considered it 

 possible that the substance colored by iodine may not be glycogen, but 

 a breakdown product of cellulose which gives the same reaction as glyco- 

 gen. See also page 981.) 



Since only in the bodies of the flagellates can cellulose be digested, 

 and termites live and develop normally when only cellulose is eaten, 

 the role of the symbionts is evident. According to Hungate (1938), 

 about one-third of the total material removed from wood, adding that 

 acted on in the fore- and mid-gut to the soluble materials present, can 

 be obtained without the aid of the Protozoa. It is possible that materials 

 adequate for nutrition of the insect may be obtained in the diet with- 

 out the Protozoa, as Cleveland (1924) found by feeding humus and 

 fungus-decomposed cellulose. Presumably sufficiently rotted wood would 

 also be adequate; Cleveland (1930) stated that defaunated Cryptocer- 

 cus, which dies in two or three weeks on partially decayed wood or 

 cellulose, will live two or three months on completely decayed wood. 

 Cleveland has conclusively demonstrated that continued survival of de- 

 faunated termites and Cryptocercus is impossible on a natural diet of 

 wood. Hungate's third, therefore, could not provide all necessary sub- 

 stances, it appears; as, if it did, the amount could be multiplied merely 

 by the ingestion of more wood, or further use of that which ordinarily 

 passes to the hind-gut for use of the Protozoa in faunated individuals. 



Experiments in feeding various cellulose-free carbohydrates to 

 termites have been made by Montalenti (1927) and Lund (1930); and 

 to Cryptocercus by Cleveland (1930, 1934). Montalenti kept Kalo- 

 termes flavkolUs alive for several months on soluble starch, alone or 

 mixed with glucose, though the hypermastigotes soon disappeared and 

 the polymastigotes greatly diminished in number. He concluded that the 

 termite could live a long time, if not indefinitely, on soluble carbo- 

 hydrates without Protozoa, but no other worker has confirmed this. 

 Lund's studies were made to determine the effect of various diets on the 



