540 Mr. T. II. Withers on a 



the prominent outer orl)ital teeth. The abdominal segments 

 and the iscliium of the third inuxilliped show no special 

 features. 



Many of the Pre-Cretaceous crabs are allied to the 

 Dromiacea, and it miglit be thought tiiat this early Cretaceous 

 type would show features in common with that group. 

 Nearly all the Jurassic crabs probably belong to the extinct 

 family Prosoponidi^, and, except in the case of the unique 

 example of Protocarcinus longipes, H. Woodward, are known 

 only by their carapace. Tiie carapace is generally longi- 

 tudinally oval in outline, and has the transverse furrows — the 

 cervical and lateral — protninently maiked; the shape of the 

 carapace and the disposition of the furrows show that the 

 Prosoponidai are allied to the family Homalodromiidas of the 

 Dromiacea. In our crab the carapace is square rather than 

 longitudinally oblong, although it is not transversely oval as 

 in many modern crabs ; but while the transverse furrows are 

 •well marked, there is barely a hint of affinity with the 

 Dromiacea. The Dromiacea include the least specialized 

 forms of Brachyura, and retain many primitive characters. 

 One of these is the frequent presence of vestiges of the 

 uropods (the sixth pair of abdominal appendages) in the form 

 of small plates intercalated between the last two segments of 

 the abdomen. These intercalated plates are certainly not 

 developed in our crab, and the legs of the last pair, instead of 

 being reduced or elevated on the back as in many Dromia- 

 ceans, are, on the contrary, well developed and modified into 

 swimming-paddles. We cannot, therefore, refer our crab to 

 the Dromiacea, and, in fact, there is nothing at all primitive 

 in any of its characters. 



The square flat carapace and, more especially, the strongly 

 deflexed front suggest Catonetopa, but the form of the cara- 

 pace does not resemble in detail any member of that varied 

 group, and the presence of three lobes on the line of deflection 

 of the front is very unlike the arrangement in any Cato- 

 metopan, where the lobes, often four, are always separated by 

 a median groove. There is, however, one character sugges- 

 tive of the Portunida?, and that is the presence of a transverse 

 ridge — the epigastric line — extending across the gastric 

 region, similar to that seen in the genera Scylla and Neptunus. 

 Moreover, the chela? are not unlike those in the Portunid 

 genus Scylla and in the Miocene genus Gatunia of the family 

 Gatuniidse ; in all three forms, next the articulation of the 

 larger palm with the dactylus, there is a large lobe or tooth 

 directed towards the end of the dactylus, and the basal tooth 



