40 EINAR LONNBERG, MAMMALS COLLECTED BY THE SWEDISH ZOOLOGICAL EXPEDITION ETC. 
size better with the specimens described from Baringo and Nandi by Ettior under 
the name of P. furax. But strange to say there are to be seen among the skulls 
from Guaso Nyiri not only such differences which can be explained as due to indi- 
vidual variation but also others which appear to be of more importance. The diffe- 
rences between the skulls of different ages are fully shown by the dimensions of the 
skulls a and 6 as recorded in the accompanying table of measurements. These two 
specimens (Pl. I fig. 4 & Pl. II figs. 1 & 2) were shot in the same place and out of 
the same herd on the north bank of Guaso Nyiri some distance below Chanler Falls. 
Specimen a is a very old male (Pl. II fig. 1) with the canines worn off to the same 
level as the other teeth which al] of them are strongly worn, and the lower incisors 
have even disappeared. Specimen 0 is a fully adult male (PI. I fig. 4, Pl. II fig. 2) 
in its prime of life. Its formidable canines are about 40 mm., and the molars are 
very little worn but the incisors somewhat more. It is evident that the differences 
between skulls a and 6 are due to age, and a comparison of their dimensions reveals 
the differences which are connected with increasing age. 
Specimen c is a rather old male (Pl. I figs 2 & 3) with incisors as well as mo- 
lars strongly worn. The canines are also worn down to a length of about 33 mm. 
Specimen c is intermediate in age between a and 6, and thus if a dimension in c is 
intermediate between the corresponding ones of a and 6, this is evidently a dimen- 
sion which alters with increasing age. A fine example of this is the distance between 
the last molar and the posterior border of the paiate. But there are other dimen- 
sions of this skull ¢ which are not intermediate between those of a and b, nor similar 
to either of them, and the differences thus exhibited by skull c appear to be of more 
importance than the others. 
These differences in dimensions indicate that specimen c has a shorter muzzle 
so that the distance from the mesial point of the superciliary ridge to the lip of 
the premaxillary is considerably shorter than in specimens a and b. The nasals (if 
measured in a straight line) are also shorter in the former than in the two latter (conf. 
the table). The shortness of these measurements is, however, partly due to the fact 
that the orbits are more abruptly raised in c (Pl. I fig. 3) than in a and b (Pl. I 
fig. 4) and the facial surface of the orbital walls forms almost a right angle to 
the flat upper surface of the muzzle in c, while the same angle is somewhat more 
obtuse, and the profile on a level with the orbits more slanting backwards in a 
and b. But the muzzle is really shorter in ¢ than in a and b as is proved by com- 
paring the respective distances from the middie of the lower border of the orbit to 
the tip of the nasals (conf. the table). Skull ¢ is also considerably broader across 
the orbits than skulls @ and 6. This is due to the great breadth of the facial sur- 
face of the outer walls of the orbits in c, because the inside distance between the 
lateral walls of the orbits is even a little shorter than in a and 6. 
Although these cranial differences are not very great, they appear to be of 
importance, especially as they are supported by differences in the colouration of the 
fur as well. Specimens ¢ and d which were shot on the north bank of Guaso Nyiri 
above Chanler Falls and only about one, or one and a half days march to the east 
