DECEMBER 22, 1899. ] 
young ZT. dyspelor might exhibit exactly the 
measurements of 7. proriger. 
According to Williston the tail termi- 
nates very abruptly in Tylosaurus proriger, 
in contrast with its gradual and slender 
termination in Platecarpus. If this was 
the case in this specimen of 7. dyspelor, we 
should not allow more than 15 inches or 88 
‘centimetres additional, giving us a total 
length of about 29 feet or 8.83 metres. The 
proportions of different regions of the body 
are very characteristic of different genera 
of Mosasaurs. In this individual the total 
of 29 feet or 9 metres is roughly distributed 
as follows: 
Feet. Metres. 
Head and jaw..........-..-. 4 1.22 
INGA cabGe as otAc oa o oes 2 61 
IB AC Kamteysree iter etieealiorac ctl etne 8 2.44 
IMs Goecnecand Beet artrerey stare 15 4.56 
Wo balestacyacvepsutceasicyoucetsiat 29 8.83 
Thus the back is four times the length of 
the neck, twice the length of the head, and 
about one-half the length of the tail. In 
other words, the tail is longer than the other 
regions of the body combined. These pro- 
portions are carefully observed in Mr. 
Knight’s restoration. 
There are positively seven cervicals, the 
number assigned to all the American Mosa- 
saurs by Williston, and this point is of 
considerable importance as bearing against 
the supposed Dolichosaurian affinities of 
the Mosasaurs. In this specimen there are 
certainly twenty two dorsals, while Williston 
assigns twenty-three dorsals to Tylosaurus 
proriger. Merriam assigns twenty-three 
dorsals to Tylosaurus (op. cit., p. 15). 
Williston is undoubtedly correct in placing 
the pelvis upon the first non rib-bearing 
vertebra, which thus represents the sacral. 
In this specimen, as in the living Monitor 
lizards, the 30th vertebra behind the head 
is distinguished by the absence of a rib, and 
by the sudden expansion of the diapophysis. 
This first expanded vertebra, as determined 
SCIENCE, 
SPAll 
by Williston, must be considered the sacral, 
analogous with the most anterior of the two 
sacralsin Varanus. This vertebra is not per- 
ceptibly different in size from the pygals be- 
hindit. Unfortunately the tips of the diapo- 
physes are not preserved, and there is no 
means of demonstrating positively that the 
ilium was attached by joint or ligament. 
There are no lumbars. The number of 
pygals, or non chevron-pearing caudals, can- 
not be determined, because many chevrons 
are not exposed. 
The vertebral formula is therefore as 
follows : 
Cervicals\isaqaseccerecetes ze 
Dorsals, with sternal ribs.... 10 
Dorsals, with floating ribs .. 12 
Sacralsticy-raeicierarspvereeese 1 
Caudals and pygals........ 72+ (=86). 
A most interesting feature is the adaptive 
modification of the mid-caudal centra and 
spines, apparently for the support of a dorsal 
caudal fin. Dr. W. D. Matthew first di- 
rected the writer’s attention to this struc- 
ture. 
Williston has figured the caudals of 7. 
proriger as having spines of a nearly uniform 
height, while in Clidastes velox (op. cit., p. 
152) he describes an extension of the spines 
as probably designed tosupportafin. This 
specimen of 7. dyspelor shows as evidence of 
a fin: 
1. A slight upward elongation of the 
spines in the mid-caudal region, beginning 
at C. 24 (in which the spine measures 19 
centimetres) to C. 39-40 (in which the 
spine rises to 11 centimetres) and subsiding 
to 10 centimetres in C. 58, At the same 
time the spines change from a pointed and 
backwardly directed to a more square, up- 
right, and truncated form. The vertical 
spine is upon C. 39; in front of this the 
spines of C. 1-38 lean backwards ; while 
behind this the spines of C. 40-70 lean 
forwards, or are nearly upright. 2. There 
is some further evidence that the upward 
