OBSERVATIONS ON SOME SOUTH AFRICAN TERMITES. 353 



appear more abundant during drought, first because they are 

 then more apt to pillage cultivated crops, and secondly 

 because the veld is barer and so they are more readily seen. 

 The thunderstorm fiction seems wholly due to the fact that, 

 in the stillness preceding the breaking of a storm, the noise 

 they make in biting through dry stems of grass is distinctly 

 heard and draws attention to them. The writer cannot say 

 positively that they are not more active at such a time, as it 

 is quite possible that they are more excited, responding, as do 

 many insects, to storm influences ; they seem to be making 

 hurried efforts to get in the last piece of hay before the stomn 

 breaks. 



The foraging is done by the workers, both large and small, 

 callows also assisting in it. The lengths are carried upright, 

 as described and figured for H. havilandi by Sharp, or are 

 merely di-agged along to the mouth of the burrow. Here the 

 bulk is deposited, but some is always being carried in. When 

 a sufficiently large quantity is collected, or towards nightfall, 

 harvesting ceases, and such supplies as have not been broad- 

 casted by the wind are removed below the ground. The 

 soldiers do not accompany the workers, but as a general rule 

 one or two are to be found on guard just within the mouth of 

 the gallery. 



Sometimes, when a veiy large collection of hay has been 

 made, a loose mound of earth has been thrown over it. Such 

 an accumulation may remain so weighted down for several 

 days (PI. XXVI, fig. 3). 



As no mound of any sort overlies the site of the hive, and as 

 no galleries are driven to the surface in its proximity, it is 

 interesting to note the disposal of the considerable quantity of 

 earth excavated in the making of cavities and galleries. As 

 far as possible this soil is utilised for filling up disused galleries 

 and deserted hives ; cavities made by other termites are also 

 filled. Curiously enough these fillings remain dry where they 

 are placed, the soil particles not being easily wetted. 



The following are a few specific instances coming under 

 observation : (1) a deserted hive completely filled with earth 



