1IO ON THE STAGONOLEPIS ROBERTSONI AND THE 
characters, for which I can find no exact parallel in either recent or 
fossil Crocodilta. 
The first of these is exhibited by each of two imperfect natural 
casts of anterior dorsal vertebrae, the strong and broad transverse 
processes of these vertebrae being bent upwards and backwards at an 
angle of 45° to a horizontal or vertical plane. The second peculiarity 
is exhibited by the caudal vertebrae, whose transverse processes come 
off altogether above the neuro-central suture, whereas ordinarily they 
are, as it were, wedged into this suture, and separate the centrum 
more or less completely from its neural arch. 
With regard to the first of the special characters here noted, it 
may be observed that the anterior dorsal vertebree of different species 
of modern Crocodtlia vary a good deal in the extent to which they 
incline upwards and backwards, and those of some Lxaliosaurta 
suffer a still more marked deflection in the same direction ; but it is 
among the Dinosaurian reptiles that the transverse processes of the 
dorsal vertebre take a direction most nearly corresponding to that 
which obtains in Stagonolepis. Without attaching too much weight 
to this circumstance, it will be seen by and by that it is worth while 
to bear it in mind. 
The second peculiarity to which I have directed attention may 
perhaps be the result of the early anchylosis of the caudal trans- 
verse processes with the neural arches. However this may be, the 
character in question is a very exceptional one, and long led me to 
hesitate in regarding the vertebra in question as really caudal. 
The gradual divergence from the strictly Crocodilian type of 
organization which is manifest in the remains to which I have just 
adverted reaches its climax in the next part I have to mention.! 
\ fragment of bone protruding from the surface of one of the blocks 
of sandstone from Lossiemouth was the last to attract my attention 
of all the fossils which have been sent from Elgin. Certain indica- 
tions convinced me that, notwithstanding the extreme fragility of 
the bony substance and the depth to which it seemed to penetrate 
into the sandstone, it was worth some trouble to work out this bone 
completely, and having succeeded, by dint of careful chiselling, in 
removing a considerable quantity of superincumbent matrix without 
damage to the fossil, I was rewarded by the view of the nearly entire 
ventral face of a coracoid bone, of a form very unlike what might 
have been anticipated. For this bone, far from having the trans- 
versely elongated form more or less constricted in the middle, which 
is exhibited by the corresponding part in all the true Crocodilza with 
1 See the final Note, p. 117. 
