THE PROTOZOA OF TERMITES 47 



taining the termites at a temperature of 36° C. for twenty-four 

 hours, which killed all of the protozoa in Reticulitcnncs Havipcs, 

 two species of Kalotcruics and one of ProrJujwtcnncs. The method 

 of starvation has been used chiefly for Tcnnopsis, from which the 

 large wood-feeding forms disappear in about a week when no 

 food is provided. The smaller flagellates live for a much longer 

 time, some of them as long as the termites. In a nymph starved 

 for fifty-eight days, there were still a few active Trichomonas and 

 Strchlouiastix and many Triccrconiifus. The most successful 

 method of defaunation is the use of oxygen under pressure, by 

 which all protozoa can be removed without any apparent injury 

 to the host. Light and Sanford (1928) have given a description 

 of the defaunation apparatus which they used in following up 

 Cleveland's (1925^) suggestions. 



Some species of flagellates are more susceptible to oxygenation 

 and deprivation of food than are others. By control of oxygenation 

 and by starvation, or combination of the two, the removal of some 

 species only from the fauna is possible. By various methods of 

 treatment including feeding on different sugars, it may be possible 

 to establish faunas of only one or a few species, and thus study 

 those species more thoroughly than would be possible in the usual 

 mixed infection. 



The work of Light and Sanford (1927, 1928) has opened the 

 way to an investigation of the interesting cjuestion of transfauna- 

 tion of termites. There is a marked specificity in the faunas, all 

 normally faunate individuals containing the same species of flag- 

 ellates in about the same relative numbers. A consideration of 

 the habits of termites renders it unlikely that cross-infection has 

 taken place. Nevertheless, the question arises whether an obliga- 

 tory host-parasitic specificity makes cross-infection impossible. 

 Will the flagellates of one species of termite live in another? 

 Light and Sanford (1927) succeeded in infecting a defaunate 

 nymph Porotenncs with the flagellates of Termopsis, and these 

 lived in their abnormal environment for twelve days. Later (1928) 

 they found that the flagellates of Tcnnopsis could live and multiply 

 for at least a hundred days in the intestines of nymphs of Kalo- 

 termcs Jmbbardi which had been defaunated by oxygenation. 



Very little is known of the physiology of cellulose digestion in 

 termites. The work of Cleveland appears to have demonstrated 

 that the xylophagous flagellates are essential in the process. But 



