^16 Annals of the South African Museuni. 



For the present, therefore, I prefer to consider H. pcrarmaius as 

 not more than a variety of H. hersclicli. 



Affinities. — The nearest ally of this species seems to l)e H. aniiatus, 

 Burm., from the Lower Coblenz Beds. 



Localities. — Ezelfontein, Ceres (28, 29, 30, 106) ; Gamka Poort 

 (21) ; unknown (24). 



nOMALONOTUS QUERNUS, sp. nov. 

 Plate XXVII. , tig. 1. 



The glabella is imperfect and perhaps does not show the frontal 

 portion, but so far as it is visible it is trajjezoidal in shape, and 

 narrows forwards ; it is distinctly separated both from the cheeks 

 and from the neck segment by deep furrows, and it bears a number 

 of large and coarse tubercles ; the glabellar furrows are very indis- 

 tinct. Cheeks very tumid. 



The thoracic segments bear numerous tubercles. 



Affinities. — The head of this specimen is somewhat similar to that 

 figured by Freeh '■'■'■ as the head of H. subarniatus, C. Koch, from the 

 Upper Coblenz Beds of Lahneck ; but it is considerably larger, and 

 as neither specimen is complete it is impossible to make a satisfac- 

 tory comparison. 



Locality. — Unknown (22). 



. IIOMALONOTUS COLOSSUS, sp. nov. 

 Plate XXYIIL, figs. 1-3. 



By far the most remarkable of all the specimens sent to me arc 

 the remains of a gigantic Homalonotus from Uitkomst. Unfortu- 

 nately the remains are fragmentary, but, judging from a comparison 

 with H. herscheli, the complete trilobite must have had a length of 

 about twenty inches. The fragments were all found in a very large 

 nodule, and are evidently portions of a single individual. 



The largest of the pieces is a portion of the left cheek. The facial 

 suture is visible and runs as in H. herscheli. The posterior margin 

 is much thickened and bears a row of tubercles. On the cheek 

 there is an oval disk showing somewhat vague concentric and radial 



* Lethisa geognostica, Th. i., Bd. 2, Lief, i., p. 218. 



