scHucHEUTj DEVELOPMENT OF THE PROTREMATA. 83 



is the Pentameracea, and has almost without exception a straight cardi- 

 nal area. The Pentameracea has 127 species, and its earliest forms also 

 have straight hinge-lines in the IG species of the families Clitambon- 

 iti(la» and Syntrophiidie; but the rostrate family Peutameridit, which 

 attained maxiiiuim development in the Silurian, h;is 87 species. The 

 Strophomenacea has living species, while the Pentameracea disappeared 

 with the Permian. The cause for the rapid extinction of the latter is 

 probably due to the high degree of specialization expressed by the 

 spondylium. 



Two well-marked types of shell form are developed in this order. By 

 far the most prominent is the group which includes the long-hinge fami- 

 lies Kutorginidie, Olitanibonitidte, Billingsellid;e, Strophomenidic, Pro- 

 ductida', Thecidiidje, and Orthidse. The other group, represented by 

 Pentamerida', is largely rostrate in form, but occasionally also develops 

 a straight hinge line. This, however, is never so prominent as in the 

 former group. In the Telotremata the general form is rostrate, but 

 very notable exceptions are present in the families Spiriferidte and 

 Terebratellida', and occasionally in the Ehynchouellid;Te and Athyrid?e. 

 The form of the sliell, however, has no great taxonomic value, and can 

 not be accorded more than generic rank. The predominating type of 

 shell form within an order probably has phyletic value, since the oldest 

 protrematous shells are long-hinged, while the telotrematous shells are 

 usually rostrate. Nevertheless, as indicated above, in the derived forins 

 of both orders there^ are notable exceptions, and these changes are 

 probablj^ always induced by shortening or lengthening of the peduncles. 

 Since Orthorhynchula has a well-developed cardinal area, it is not in 

 itself "evidence of the first significance as indicating the source from 

 which the extensive group of the Ehynchonellas originated." ^ The 

 oldest rhynchonelloids are rostrate shells [Protorliyncha^i minor and 

 P. ? amhigua of the Lower Cambrian), and the ontogeny of several spe- 

 cies of Rhynchonella and of Zygospira has not revealed a long-hinged 

 stage with cardinal areas. There is, therefore, no conclusive i)roof for the 

 deduction of Hall and Clarke, "tliat some of the Ehynchonellida?, early 

 in their [geologic] history, occasionally retain a well-defined cardinal 

 area, and that, in default of other evidence, the presence of this.char- 

 a(;ter may be regarded as indicative of the common origin of Orthis, 

 the Strophomenidje, and the Ehynchonellas."^ 



In this order far more than in any other is found the closure of the 

 pedicle passage and atrophy of the pedicle, together with peculiar 

 special adaptations which entirely or partially replace the functions 

 of the pedicle. In the family Productidse the ventral shell develops 

 more or less abundant tubular spines, either along the cardinal line or 

 over the entire valve. These are always most abundant in, or are 



' Palteontology of New York, Vo]. VIII, Part IT, 1895, p. 336. 



'Ibid., p. ^.42. For further remarks beariug on this subject, see pages 93-9bon the significance of the 

 prodeltidiiun. 



