A Monograph of the Formicidae of South Africa. 611 



tinct forms, i. e. Qj. or ^ major, ^ media and § minor. No ocelli, 

 but sometimes vestiges of same occur as indistinct pits. Frontal 

 carinae more or* less sinuate, usually well defined. Thorax usually 

 widest in front, the pro-mesonotal and meso-metanotal or meso- 

 epinotal sutures usually well defined. Epinotum unarmed, in some 

 subgenera more or less subtruncate. Petiole of varying form, (see 

 subgenera). In the Qj. the head is very large, subtriangular and 

 with the angles rounded, or subcordate, wider than the thorax. In 

 the ^ minor the head is usually elongate, and more often longer 

 than wide. The <$ media has the head intermediate in form between 

 that of the Qj. and of the ^. 



9- Antennae as in the §. Rarely larger than the Qj.; the head 

 also not so wide as in that form. Ocelli not very large. Thorax of 

 the form usual to the sex, the pronotum rarely exposed above. 

 "Wings with a closed radial and I cubital cell. Abdomen usually 

 oblongo-ovate. Otherwise like the Qj.. 



cf, Antennae 13-jointed, scape as long or at least half as long as 

 the flagellum. Ocelli and eyes large. Petiole thicker than in the 

 j"3 or 9- Abdomen small, considerably narrowed towards the apex 

 or lanceolate, the genital armature small. Legs slender and feeble. 

 Mandibles small and usually edentate except one apical tooth. 

 Wings as in the 9- Distribution. Both hemispheres. 



This is an enormous genus comprising 1000 or more different 

 forms. With the exception of C. maculatus and its subspecies and 

 varieties, the identification of our species, with reference to the Qj., 

 is not attended with much difficulty, as they fall into fairly easily 

 recognised subgenera. This however is not the case with C. macu- 

 latus, a species which, with numerous and closely related races and 

 varieties, is distributed over a large part of the globe. In the ^ 

 media and minor castes the specific characters are but slightly 

 differentiated, and it is therefore essential that the Qj. Qj. should be 

 obtained with the smaller ^§ whenever possible. 



The degree of convexity of the thorax is often a valuable dia- 

 gnostic character, although it is difficult to measure without the aid 

 of a drawing-apparatus on the microscope The standard which I 

 have adopted is the ratio of the height to the length, the latter 

 being measured in a straight line from the anterior point of the 

 pronotum to the apex of the epinotum at its junction with the 

 petiole, and the former by the length of the perpendicular from 

 the highest point. These measurements are expressed by the sym- 

 bols H.L. in the following pages, e. g. L. = 3 1 l 2 H. 



39 



