2QO L. R. CLEVELAND. 



some of which are present in almost any locality in the United 

 States and many other countries. Material for this study was 

 collected in Man-land. Ternwpsis belongs to the family Kalo- 

 termitidae and Reticitlitermes to the family Rhinotermitidae. 

 So, the two genera selected for experimentation are quite different 

 in structure, habits, location, etc. 



Cellulose from two sources, cotton and wood, was used. That 

 from cotton was the well-known Whatman filter paper No. 43, 

 which is extracted in hydrochloric and hydrofluoric acids and in 

 ether and contains .00006 per cent. ash. But cellulose prepared 

 in this way is subjected to drastic treatment, which might render 

 it more open to attack. In order to test this, fresh western 

 yellow pine wood was made into sawdust, which was dried and 

 ground so as to pass through a 5O-mesh screen. This was 

 extracted with alcohol and benzene and boiled under a reflux- 

 condenser Dore in a personal communication ('20) suggested 

 the use and preparation of lignocellulose to me for six hours, 

 some in 5 per cent. HC1 and some in 5 per cent. NaOH. In the 

 material prepared in this manner, lignin and cellulose, with 

 perhaps a trace of mannose, remain. But lignin has no food 

 value, for when it alone is fed termites they do not live any 

 longer than when they are starved. Hence we are dealing only 

 with the cellulose content of this material, and results obtained 

 by feeding it may be compared in every way with those obtained 

 from feeding the cotton cellulose. 



EXPERIMENTAL. 



a. The Genus Termopsis. 



The experiments were each carried out in (wo parts, which 

 were duplicates except for food. In part one the food was 

 lignocellulose and in part two it was cotton cellulose in the form 

 of Whatman filter paper No. 43. 



In each part of the first experiment, approximately 500 indi- 

 viduals, some soldiers but mostly nymphs, were placed in a 

 large glass vessel which just fitted in a moist chamber. From 

 time to time more food as needed was placed in the glass vessel 

 and occasionally the fecal debris was removed. In the moist 

 chamber around the outer edge of the glass vessel which con- 



