654 Bulletin American Museum of Natural History [Vol. LXVI 



approximately the same age, the short, wide skull having come from an 

 individual 890 mm. long, the other from one 795 mm. The skulls them- 

 selves are within one millimeter of being the same in greatest length, but 

 in one (53861) the width is but 44 per cent of the length, while in the 

 other (53864) the width is 52 per cent of the length. That the disharmony 

 exhibited is not due to age differences is shown by the fact that the oldest 

 individuals in the complete series are nearer the shape of the skull from 

 the smaller animal than that from the larger, and that in this series there 

 is a perfect gradation from one type to the other among animals of very 

 nearly the same size. The difference is not one of sex, since these two 

 are females, and the other skulls available show no sex-form relation- 

 ship. That two species, or races, are not involved is attested by the 

 uniformity of other characters and the random distribution of the skull 

 types within the localities represented by the Congo collection and other 

 material coming from points ranging between Liberia and the middle 

 Congo. Neither do the specimens examined lend support to any assump- 

 tion that one type or other is the result of a pathologic condition or an 

 endocrine disturbance, unless disruption of normal development is 

 chronic in the species. Other matters of random variation in the skulls, 

 such as the presence of Wormian bones, the variability of position in 

 sutures, etc., appear to be of about the same magnitude as variation 

 found in the better series of tricuspis. It must, therefore, be concluded 

 that in development the skull of longicaudatus is normally subject to a 

 wide variability. 



The greatest age is not represented in the Congo series. The 

 United States National Museum specimen (No. 220402) from Fernand 

 Vaz, described above for old age characters of the scales, is accom- 

 panied by a skull 6 mm. longer in greatest, basal and palatal lengths, 

 2 mm. longer in nasal length, and 4 mm. longer in mandibular length 

 than the largest skull of the Congo series. 



The fact that the two male skulls are longer than the seven female 

 skulls in the same collection is obviously due in part to the accident that 

 the young individuals were all females, but it is probable that in this 

 species, as in the two others considered here, the males do attain greater 

 size than the complementary sex. 



