2 76 Phylum Arthropoda 



in the laboratory. More nearly similar genetic constitution within the 

 group, a condition often desirable in experimental work, is thereby as- 

 sured. Furthermore, the excessive cannibalism which is likely to follow 

 the mixing of colonies is avoided. If colonies must be mixed, as is some- 

 times necessary in order to get the numbers of termites desired for a 

 given experiment, about a week is allowed for cannibalism to subside 

 before the mixed group is used in any experiment. 



Individual termites are most safely handled by lifting them on a 

 camel's hair brush or a tapering piece of stiff paper. Large numbers of 

 termites may be transferred from one container to another by pouring 

 them through a trough or funnel made of cellophane. 



Petri dishes or shallow stender dishes are suitable containers for 

 colonies of ioo or fewer termites. For larger colonies refrigerator dishes 

 or moist chamber dishes have been found satisfactory. The lids of the 

 containers, while not air tight, should fit closely enough and be heavy 

 enough to prevent the escape of the termites. Anyone who has observed 

 the crowded condition in termite galleries in nature will realize that 

 there is little danger of getting too many termites into one container so 

 long as there is opportunity for each termite to get at the food provided. 

 In fact, large colonies thrive much better than small cononies. 



The dishes containing the termites are kept in a dark room or cabinet. 

 Large, open dishes of water placed nearby aid in maintaining a desirable 

 humidity. Damp-wood termites live and reproduce at ordinary room 

 temperatures (20 to 23 ° C). They will survive throughout a con- 

 siderably greater temperature range, provided the change is gradual. 

 They should be protected from sudden changes in temperature. 



Rotten, fungus-containing wood constitutes a satisfactory diet for 

 damp-wood termites (Hendee, 1934)- The pieces of wood should be 

 sufficiently heavy and so placed that they may not be moved by the 

 weight of the termites. 



Cook and Scott (1933) describe an artificial diet for termites which 

 consists of sucrose, casein, "Crisco", salts, cod liver oil, and rice polish- 

 ings, all incorporated in an agar gel. Filter paper, while it fails to com- 

 prise a complete diet for termites (Cook and Scott, 1933 ; Hendee, 1934) , 

 affords a convenient source of the carbohydrate portion of the diet. 



The food should be kept slightly damp. If the containers are left 

 undisturbed, the termites will partially seal the lids from the inside with 

 fecal matter. In this way they partially control the humidity within the 

 container and prevent evaporation of water from their food. If the 

 containers are opened frequently, however, a few drops of water should 

 be added to the food every two or three days. 



Fungi have been shown to supply an essential part of the natural diet 

 of termites (Hendee, 1934). At the same time they constitute a poten- 



