604 ON ARCTOCEBUS CALABARENSIS 
and forwards. The anterior external cusp is continued by a curved 
ridge on to the margin of the large anterior basal process of the tooth ; 
and the posterior external cusp is connected by an oblique curved 
ridge with the antero-internal. Thus the tooth acquires a doubly 
crescentic, Rhinocerotic pattern. 
In the next grinder, which is also quadricuspidate, the anterior 
basal prolongation is reduced, and the ridges connecting the pairs of 
cusps are transverse to the axis of the jaw. There is no oblique 
curved ridge. 
The last molar is like the preceding ; but there is a fifth cusp, and 
the ridges which connect the pairs of cusps are oblique, sloping from 
within backwards and outwards, or in the opposite direction to that 
observed in the antepenultimate tooth. 
There is, in both jaws, a much greater difference between the second 
grinder and the third, than between the third and the fourth, so that, 
except for the caution suggested by the similarity of the last premolar 
to the first molar in Galago sennaarensis, &c., one might suspect 
the third tooth to be a true molar, and consider that, in these 
animals, it is a premolar, and not a molar, which is suppressed. 
The roof of the palate of the Angwantibo exhibits altogether nine 
transverse ridges, each of which is convex forwards and concave 
backwards. The eight anterior ones extend from a given tooth to 
the corresponding tooth of the opposite side. The first lies between 
the outer incisors, and the others between the following six grinding- 
teeth. There is a faint ridge behind the last molars, beyond which 
the soft palate continues the roof of the mouth for half an inch. In 
the interval between the first two ridges, in the middle line, there 
is a smal! round aperture which ends cecally ; but in front of this and 
of the first ridge are two small crescentic apertures, the concavities 
of which are turned towards one another. Into either of these a 
style can be passed upwards, outwards, and backwards for a con- 
siderable distance. These passages are the so-called ducts of Stenson. 
Mr. Murray has noticed a similar structure in the palate of Galago 
murinus as “two small orifices (as large, however, as the root of the 
superior incisors) situated in the middle space between the two 
incisors on each side, but a little behind their line. Their position 
suggests an analogy to Jacobson’s vesicles in the Horse; and on 
tracing their origin, we find that they lead to the nasal orifice, 
expanding before they reach it into a sort of sac, which appears to 
communicate, by a narrow and short canai, with the nasal orifice, in 
this respect differing from Jacobson’s sac, which does not communicate 
directly with the exterior” (“Supplementary Remarks on the genus 
