ACTIVITY AND DISTRIBUTION 



301 



taken when the species was breeding (see p. 2 1 2) . The soil was kept very 

 moist up to the time the first ovipositor holes were made, because this 

 species lays only in moist soil. After this the wetting of the soil was 

 done very cautiously, so as not to wash the eggs from the ground in 

 steep parts. Accordingly, the holes were not obliterated from day to 

 day. The counts, however, are not accurate for the soil in which a large 

 number were made, because eggs are sometimes laid very close together 

 and adjoining holes destroyed. Some eggs are deposited in irregular 

 cracks and crevices where they are likely to be overlooked. The greatest 

 care was taken to discover every hole made in the soils in which larvae 

 do not occur in nature. Soils in the different lots were arranged in 

 different orders. 



b) Results. Table LXVIII shows the approximate number of holes 

 made in the clay and probably the actual number made in the other 

 soils, together with the number of larvae which appeared: 80 per cent 

 on the steep slope, 98 per cent in clay. 



The count of holes includes some in the first stages of digging, mere 

 scratches on the ground, and others which had been excavated to the 

 usual depth with or without eggs being laid. 



! 

 TABLE LXVIII (55) 



DISTRIBUTION OF OVIPOSITOR HOLES AND LARVAE OF C. purpurea limbalis UNDER 



EXPERIMENTAL CONDITIONS 

 S = steep; L = level. 



c) Factors controlling habitat selection (55). Pairs taken in coitus 

 were placed in cages containing sand only and level clay only. No 

 larvae appeared in either case. The experiment with the level clay 

 has not been repeated. Females placed in cages containing rough, 

 steep clay, deposited eggs. Eggs are also absent from dry soils, whether 

 steep or level. 



