108 Claude Fuller 



lows Haviland and the Natal insects he reports upon are alsa 

 havilandi. In 1908, Silvestri determined specimens from Nama- 

 qualand and the Kalahari as trincrvius Ramb. (sensu Sjostedt^ 

 and reduced dispar Sjostedt to a synonym thereof. Later, 

 (1914), Holmgren set aside this determination and redescribed 

 Silvestri's insects as a new species under the name kalaharicus and. 

 declared it to be different from both trinervius (sensu Sjostedt) 

 and dispar. From the two descriptions, and with insects from 

 Namaqualand before me, ! can only regard k^^laharicus Holmgren 

 as a form of gemellus Sjostedt. 



T. dispar Sjostedt (1904) is a Zomba insect and only the 

 characters of the soldiers and workers are known. Nevertheless, 

 Holmgren (1914) has described as " forms" of dispar certain 

 material from Zululand giving to some extent the characters of 

 the imago of one of these two forms. Without doubt the two^ 

 forms are one and the same species. As it is altogether impossible 

 to identify an imperfectly defined species of this genus, as is diopar^ 

 unless obtained in the type locality, it is most undesirable that 

 this record of dispar should stand, especially as there is every 

 likelihood that the Zululand insect is not dispar. It should there- 

 fore be referred to as T. zuluensis Holmg. and disparoides be re- 

 legated to the nom. negl. 



T. gemellus Sjostedt (1904) is an imperfectly defined species 

 erected on nasuti and workers in the absence of imagos. Part of 

 the type series is in the South African Museum and these nasutx 

 seem to be undersized (nanitic) representatives of the common 

 nasute found about Capeto^^^l. It so happens that there is no 

 precise information as to the exact type locality of gemellus but as 

 there is no reason to suspect that the material came to the Museimi 

 from far afield, I have treated the predominant S.W. species as 

 gemellus. 



T. irinervoides Sjostedt (1913a) was erected upon four castes 

 and constitutes the most reliable record of any South African 

 species of the genus. Assuming that the printed description 

 abounds in typographical errors, and there is a fair amount of 

 evidence pointing to this, I have redescribed the species from a nest 

 series found in the lower Karroo, the terrain of irinervoides. 



