OBSERVATIONS ON SOME SOUTH AFRICAN TERMITES. 381 



•gardens in them fill them completely, and if removed resemble 

 irregular and more or less flattened loaves. 



The queen-cell (PI. XXX, figs. 2-5) is situated in the 

 great cavity and, in those nests explored, in a peculiar posi- 

 tion and always on small pedestals. The cell itself is within 

 an irregular lump of clay, forming a i"ough column at or 

 towards one side of the nest. Its site is indicated in both 

 nests, of which diagrams and illustrations are given. 



In one of the two nests illustrated the pedestals of the 

 queen-cell rested upon the stones which formed the bottom 



Text-fig. 11. 



T. badius. A. Yertical section of nest lying over boulders, 

 showing lateral fungus-garden excavations. B. Ground plan 

 of same. ( X -^^.) 



of the cavity, in another it was attached to a shelf which ran 

 around the chamber at about half its height. In the thiee 

 nests the joining of the upper part of the clay column to the 

 cupola was perhaps more indefinite than as figured in the 

 diagram, and the cell itself not quite so mncli a part of the 

 wall. 



Between the two diagrams a considerable difference appears, 

 and the diiference in the nest arrangement displayed is 

 thought to be due to the difi^erent nature of the soil. In the 

 case of the shallow nest it was found that underlying the 

 site there was a layer of stone boulders. To some extent the 

 soil between these had been removed and the spaces so made 

 filled up with fungus-garden ; it was assumed that as the 

 termites had not been able to excavate deeper they had made 

 the larger horizontal extensions shown in the diagrams. 



