SPECIMEN FROM FRANCE 455 



is the tibial condyle. This condition is the reverse of that found in the 

 Megalosauridae. The distal portion of the shaft is less curved than is 

 the proximal half. From the curvature it is concluded that the whole 

 hind limb was much straighter than in the Megalosauridae. This is 

 shown also by the position of the articular face of the caput femoris. The 

 dorsal centra are relatively high. The tibia has a greatly projecting 

 cnemial crest and the crista lateralis is unusually close to the proximal 

 end of the bone. Metatarsal II is nearly half the length of the femur, 

 which is exceptionally long. A portion of the manus is preserved, show- 

 ing very slender phalanges. The first phalange of the thumb is short 

 and the claws are small. An element found isolated is much like meta- 

 carpal V of the Plateosauridse and of Megalosaurus poikilopleuron; it 

 would be surprising, however, if a fifth digit were still present. All of 

 these remains (and more) have been found so closely associated that they 

 probably belong to a single individual. 



All of the former genera belong to the Carnosauria. Of the Coeluro- 

 sauria there are but few forms. Of this group Sarcosaurus woodi, from 

 the English Low^er Lias, was recently described by Andrews. Ilium, 

 pubis, and vertebrse exhibit the characteristic form, which is also seen in 

 Procompsognatlius, Ornitholestes, and Ornithomimus. 



Compsognathus longipes (AVagner), from the Portland of Solenhofen, 

 is well known. I have made a new restoration of the skull and of the 

 whole skeleton. There are ten cervical vertebrae (beginning with the 

 atlas), thirteen dorsals, and five sacrals. The humerus, although not 

 distinctly recognizable at its j^roximal extremity, shows the processus 

 lateralis extending two-thirds of the whole length of the bone. The 

 reconstruction of the manus exhibits a form reminiscent of Ornitholestes 

 and also of Ornithomimus in some respects. The skull, as in all Coeluro- 

 sauria, is much more primitive than in the Megalosaurid^, as is seen 

 from the base of the skull, the palate, the temporal openings, and the 

 large orbits. 



The vertebra from the English Wealden C alamos pondylus foxi, by 

 Lydekker, is a coelurid, and the sacral remains from the same horizon, 

 which Seeley described as Thecospondylus daviesi, seem to me rather 

 different from the other species of that (so-called) genus. I therefore 

 propose the new generic name Tliecoccelurus. 



More satisfactory are the remains of Aristosuchus pusillus (Seeley), 

 also from the Wealden of the Isle of Wight. The vertebrae, pubis, and 

 claw show this specimen to be a true coelurosaurian. 



The claw described by Dollo as Megalosaurus lonzeensis, from the 



