1934] Hatt, African Manatees 553 



tending forward from the most anterior part of the symphysis, and though 

 this spine does occasionally occur in inunguis, and possibly in senegalensis, 

 it is almost a constant feature of the mandible of the postnatal manatus. 



While not constant either for a species or, probably, even during the 

 life of an individual, the mental foramina of the manatees follow different 

 tendencies in inunguis than they do in manatus and possibly show a 

 slightly different average in senegalensis than in manatus. In inunguis 

 these foramina range in number from 11 to 15 (average 15 — ). The range 

 in manatus is from 4 to 7 (average 6) . In the Congo specimen there are 4 

 mental foramina, and in Blainville's figure 3 are shown on one side. 



As Hartlaub noted, the ventral border of the mandibular ramus, 

 presents a greater curve to the horizontal plane in manatus than it does 

 in the other species; in the others there is no great difference. This 

 curve results from the deepening of the mandibular symphysis and is 

 so excessive in manatus that it is one of the most pronounced features of 

 the species. 



The coronoid process of the manatees presents the usual high varia- 

 tion in shape and cannot be relied upon in diagnosis. Hartlaub concludes 

 that the posterior superior angle in inunguis is constantly hooked and 

 that these hooks are only occasional in manatus. In my series I find that 

 even in old animals the hooks are not always developed in inunguis, 

 though they are usual in the Caribbean manatees. The extent of broad- 

 ening of the coronoid, which Hartlaub regards as a distinguishing charac- 

 ter of the species, appears to me to be very weak and unreliable for separa- 

 tion of senegalensis and manatus material, since in this feature Florida 

 specimens are indistinguishable from the Congo animal. 



The transverse breadth of the condyle of senegalensis is slight com- 

 pared with that of manatus and inunguis, but the character is so relative 

 that, lacking a large series of specimens, the difference is not discernible. 



The mandibular foramen is subject to certain specific modifications. 

 Thus, in the Congo specimen it is separated from the bony sheath by a 

 septum of bone, whereas in inunguis and manatus no such septa occur, 

 with the exception that there are, rarely, converging processes that 

 may nearly touch to form a partial septum. This bony ridge is well 

 developed in the Congo specimen, and is shown in a figure (Plate in, 

 figure 27) accompanying Hartlaub's paper. However, it does not appear 

 in the figure reproduced in Flower's 'Osteology' (1870, Fig. 64, p. 197) 

 and may not be very constant. 



A single specimen supposed to have come from Texas is different 

 from other manatees examined in that the lamina lying medial to the 



