TESTUDINID^. 397 



expected. No part of the nuchal appears. The first neural was wider than the second, and in 

 contact with the second costal. Its greatest width was near 64 mm., while the second has 

 measured 55 mm. in width. As in some specimens of S. nebrascensis and in probably all 

 individuals of the John Day species of the genus, the third neural was tetragonal. Correspond- 

 ing to this condition, the third costal was in contact with the second, third, and fourth neurals. 



The costals are only slightly modified as respects the widths 

 of the opposite ends of each. The measurements shown in the 

 table taken from Sinclair's description and illustration show the 

 extent of the modification. 



The carapacial scutes resembled closely those of S. nebra- 

 scensis. An estimate makes the first vertebral scute to have 

 been about 184 mm. wide in front. The second vertebral was 



Costal. 



Proximal Distal 

 end. end. 



3 68 



55 



52 37 



3 8 5i 



about 108 mm. wide. 



The plastron (text-fig. 503) was, according to Sinclair, broken 

 away at the extremity of the epiplastral lip. It must therefore have been rounded. The 

 anterior lobe had a length of about 100 mm., the base a width of 200 mm., resembling thus 

 the anterior lobe of S. nebrascensis. The entoplastron is rhomboidal in form, thus resem- 

 bling that of some species of Testudo. Its length is 8 1 mm., its width is 88 mm. The 

 portion of the plastron present evidently extended backward to near the hinder border of 

 the hypoplastron; but did not reach the inguinal notches. The bridge must therefore have 

 had a width of at least 172 mm. 



The sulci separating the scutes of the plastron are stated to be very obscure. The gulars 

 and humerals resemble those of 5. nebrascensis. 



It appears to be quite certain that this species belongs to the genus Stylemys. It appears 

 to differ from all the others in having a very broad first neural, which comes into contact with 

 the second costal bone, and in having the entoplastron more pointed behind. From the known 

 John Day species it differs in having the costal bones less wedge-shaped. 



Genus TESTUDO Linnaeus. 



Carapace extensively joined to plastron by suture. Inguinal buttresses rising little, if any, 

 above the costo-peripheral sutures. Second and fourth neurals usually octagonal; the third 

 and fifth usually quadrangular. Epiplastral lip usually projecting abruptly from the general 

 contour of the anterior lobe. Plastron notcht behind. Crushing-surfaces of |aws with longi- 

 tudinal ridges and grooves. No median longitudinal ridge in front of the choanae. Digits 

 much shortened and foot club-shaped. 



Type: Testudo grxca Linnaeus. 



From both Hadrianus and Stylemys this genus is distinguish! by the much higher grade of 

 differentiation which has been attained by the elements of the shell. From Stylemys it is 

 further distinguish!, as it is also from Gopherus, in having no longitudinal ridge in the midline 

 of the upper jaw in front of the choanae. Some living species have the neurals and costals 

 little more differentiated than in Stylemys, and in T. inusitata, described below, none of the 

 neurals is octagonal or quadrangular. 



ANALYSIS OF THE KNOWN NORTH AMERICAN FOSSIL SPECIES OF TESTUDO. 



A 1 . Oligocene species: 



a. Titanotherium beds. 



1. Peripheral sulci deeply imprest; epiplastral lip more than twice as wide as long .. brontnps 



2. Costals very thick; deeply imprest by lines of growth of the horny scutes cxornata 



aa. Oreodon beds. 



3. Lip nearly as wide as long; its sides parallel thomsoni 



4. Shell flattened; no octagonal neural; epiplastral lip abruptly projecting, tootht 



in front laticunea 



5. Lip truncated, three times as wide as long ligotna 



6. Lip truncated, short; pectoral scutes about one-fourth as wide as abdominals. . amphithorax 



7. Lip square; the gulars not reaching the entoplastron quadrata 



8. Lip narrow and wedge-shaped cultrata 



