354 CLAUDE FULLER. 



particles, the interioi* partitions being mostly intact; (2) a 

 deserted hive, in which the interior had collapsed and fallen 



the bottom, was being filled through openings in the dome 

 of the cavity, the earth particles and fragments of dead 

 Hodotermes being dropped in; (3) the filling in of the 

 small fungus-gardens of T. incertus from holes in the dome 

 of the cavity. To these may be added many cases where 

 cavities of all soi-ts and sizes were found completely filled up. 



Besides acting in this manner, Hodotermes transvaal- 

 ensis brings a great deal of earth to the surface, arranging 

 the loose particles in mounds which, when of any size, strongly 

 resemble small mole-hills (PI. XXVI, fig. 4). These often 

 occur over a fairly wide area, but they do not coincide with the 

 nest-site nor are they necessarily adjacent to the openings from 

 which the termites emerge when bent on gathering provender. 



A hole is driven to the surface which is about a half or one- 

 third of an inch in diameter. The first particles of moist earth 

 are built together as a collar around this, and further particles 

 are pushed up so as to fall outside it. As the mound rises in 

 height so the collar grows into a chimney, the encircling pile 

 of earth supporting it upon its weak foundations (PL XXVI, 

 figs. 5 and 5a) . Often these progress no further, the holes at 

 the top are closed up, and the wind wearing on the hill leaves 

 the apex of the chimney exposed. 



Where large quantities of earth have been thrown out, the 

 inner chimney is found enlarged and branched and of 

 various grotesque forms. It is obvious from the arrangement 

 of the branches that they are constructed so that the soil will 

 fall away from the point on one side of the moundlet where it is 

 discharged (PI. XXVI, figs. 4 and 4a, 4^, 4c). 



Curiously enough the insects do not expose themselves in 

 doing this work. When the pellets are brought to the mouth of 

 the chimney they are moist and cling together so as to cloak 

 it ; and all that one sees is grain after grain taking its place 

 at the apex of the mound and then, on drying, rolling down 

 the incline. If the mound of loose earth is gently removed, 

 the chimney — however bizarre — stands upon its base ; but if 



