The Termites of South Africa 83 



Holmgren sub-divided Odontotermes s.lat., (e.g. Termes sensu 

 Linneus,) into three sub-genera using both imago and soldier 

 characters for this purpose. With one of these sub-genera, 

 Xenotermes, I am not concerned and amongst our species, I quite 

 fail to separate another, C})clotermes from his Odontotermes s.str. 

 either upon imago or soldier characters. 



My proposal therefore is to base the sub-genus Odontotermes 

 mainly and for the present upon soldier characters. 



1. Mandibles not massive, soldiers of one caste; smaller 

 kinds. Headwidth, normally between 1 and 1 * 5 mm. 



Termes. 



2. Mandibles massive, soldiers of two castes or intergrading 

 from small to large; larger kinds with relatively thick 

 heads. Headwidth normally between 1 • 7 and 3 • 5 mm. 



Odontotermes 



This sub-division seems somewhat artificial but is none the less 

 obvious and useful in practice. 



Negligible References and Species: There are a number of 

 references to the presence of species or the distribution of species 

 of this genus in South Africa that must be treated as negligible 

 or subject to correction., The following discussion is submitted 

 with a view to the matter being more thoroughly considered in the 

 light of my remarks. 



1. Termes destructor Smeathman (1781) This is an im- 

 perfectly described and ill defined species with Sierra Leone as 

 its type locality. It needs to be re-discovered there in all its 

 castes and again described from fresh or alcoholic material before 

 any termite from a far-off locality can with assurance be referred 

 to it. Hagen (1858) identified T. cordofanus Kollar as 

 destructor and these and other Nile imagos have come to be 

 regarded as typical of destructor. As the type locality is 3,000 

 miles away to the east, cordofanus may easily be a distinct species. 

 Again, Hagen gave " Port Natal " as a locality for destructor 

 but he makes it clear that he was by no means certain about the 

 identity of the South African insect and commented upon its larger 

 size. Sierra Leone is north of the equator on the Atlantic, Port 

 Natal is south thereof and on the Indian Ocean; moreover, as 



