6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I3I 



species to Rectithyris but this appears to be the only genus at present 

 to which it can be assigned. Points of agreement with Rectithyris as 

 defined by Sahni (1929) appear: the mesothyrid foramen, the easily 

 visible deltidial plate, rudimentary cardinal process, the inner hinge 

 plates, even though they are not strongly developed, and the constric- 

 tion of the broad transverse ribbon to form a median, flattened crest. 

 Deviations from Rectithyris are the curvature of the beak, less strongly 

 triangular loop, and smaller foramen. 



Some points of similarity exist between the Arizona species and 

 Neoliothyrina. The loop of the latter is like that of R. vcspertina and 

 according to Sahni's figures (1929, pi, 9, figs. 19, 20) shows the same 

 type of variation. The loop shown in figure 20 has nearly parallel sides 

 and the crural processes are well anterior to the crural bases, whereas 

 the loop shown in figure 19 has the crural processes located almost at 

 the junction of the crura and the crural bases. The beak characters and 

 other details of Neoliothyrina, however, are too different to permit 

 use of that name. 



No known species of Rectithyris is like this American species ; con- 

 sequently, no direct comparison is possible. 



Variation of the loop. — Variations in parts of the loop are evident 

 in many of the specimens, but these variations could not be correlated 

 with shape or shell differences. In some specimens the brachial valve 

 is distinctly flatter than in others, but this difference did not accord 

 with loop differences. In the young, loop variation is evident in the 

 length of the descending branch anterior to the crural base. One 

 specimen, paratype U.S.N.M. No. 124196a, has the crural process 

 given off almost at the junction of the descending branch with the 

 crural base. In another, plate iB, figure 33, the crural process is lo- 

 cated a short distance anterior to the crural base. In this specimen the 

 crural base appears as a ridge bounding the inner socket plate. This 

 is true of a somewhat larger and more-elongate specimen shown in 

 figures 29-31 on the same plate. This is not true, however, of the 

 largest and oldest loop figured, same plate, figures 34 and 35, in which 

 the descending process is short and the posterior extension of the 

 crural base is buried in the formation of modest inner hinge plates. 

 The presence of inner hinge plates appears to be an age character, at 

 least in this case. 



Abnormal specimen. — This species shows considerable variation in 

 exterior as well as interior features. Such variability is to be expected, 

 but the occurrence of a freak specimen having the crural processes 

 united is unusual. This specimen is paratype U.S.N.M. No. 124195a. 

 The beak is broken and most of the brachial valve broke from the 



