THE BATRACHIA OF NORTH AMERICA. 21 



ion of true Salamanders was not i>roperly distiiiQaished before the pub- 

 lication of the system of De Blainville, in 1810, who called tlieni the 

 Pscndosauria. He, however, omitted from tliem the Cieciliidie. In this 

 omission he was followed by sU subsequent authors, except Cope, who 

 called the entire order, including' the Ooeciliidie, theUrodela, adopting a 

 name already proposed by Dnmeril, in 1804, for a division of wider 

 scope. On this account the name Urodela is adopted in the present 

 work. TLe name Gradientia was first used to include only Batrachia, 

 without lizards, by Merrem, in 1820. The two branchiate or.lers, Pro- 

 teidaandTrachystomata, were included in theUrodela or Gradientia by 

 all anthors except the following: De Blainville combined them in one 

 order, the Amphibia; Gray, Harlan, and Fitzinger followed, but com 

 bined them with unrelated forms; Dumeril and Bibron kept them to 

 gether with the Cryptobranchidie in a division, Trematoderes, follow- 

 ing Fitzinger (182G); Haeckel follows De Blainville, bat renames the 

 Pseudosauria of that author Sozobranchia, and includes in his Am- 

 phibia the non-related Axolotls. In 18GG* Co[)e first distinguished the 

 Trachystomata and Proteida as orders, and purged them of the Axolotls, 

 which he placed in the Urodela. 



PROTEIDA. 



Os intercalare present; no supra or basioccipitals; 0. maxUlare and 

 yrefrontalia wanting; vomero-palatinum and pterygoideum present, 

 continuous; orbitosphenoid elongate, not forming part of palate; cera- 

 tohyals connate. 



This order agrees generally with the Urodela, but presents one most 

 important feature of difference, in the presence of the Os intercalare. It 

 is this point that gives the Proteida its position between the Stego- 

 cephali and the Urodela, and which indicates the line of connection 

 between the extinct forms of the Carboniferous period and the mod- 

 ern types. 



The hyoid apparatus differs from that of adult Urodela, and resembles 

 their larva3 in having three epibranchials, instead of one only. The 

 second basibranchial is also connected with the first, which is not the 

 case with the Urodela. Tlie centrale is present in both carpus and 

 tarsus, 



No extinct genera are certainly known to belong to this order, but 

 there is one that resembles it nearly, and may belong here. This is the 

 Cocytinust (Cope), which has been found in the coal measures of eastern 

 Ohio. If it be not a larval Stegocephal, it belongs to this order. It 

 would not enter the same family as the recent forms, as it has a small 

 maxillary bone. Another extinct type from the Wealden of Belgium 

 has been suspected to belong to the Proteida, but the cranium is not 



*0n the Arciferous Annra, Journal Academy Philadelphia, 1866, p. 102. 

 t Geolo<i;ical Survey of Ohio, ii, Paleontology. 



