474 Systematic Paleontology 



and more persistent axials, and a more quadrate wing than that suggested 

 by figure 8. The broadly rounded tapering whorls and the relatively few, 

 approximately twelve, axials of figure 8 eorerspond more closely with the 

 characters implied by Morton's wretched figure than do the straight whorls 

 and numerous axials of figure 7. It is perfectly possible, to be sure, that 

 neither one of the figures represents Morton's species, and it is highly 

 improbable that they both do. Until further light is brought to bear upon 

 the subject in the way of further collections from Morton's type locality at 

 Prairie Bluff, Alabama, it seems permissible to tentatively unite the 

 southern cast which served as Morton's type and the New Jersey shell 

 figured by Whitfield. A new name should be given to the second figured 

 shell under the same name, but as this does not occur in the area under 

 discussion, that may be left for the present in the hope that further data 

 may be available when it is necessary that the form should bear a name. 

 It is probable that both the more rounded and the more angular shells are 

 represented in the species of casts, Rostellaria spirala being allied with 

 the more rounded type, Anchura compressa with the more angular, and 

 Rostellaria compacta probably distinct in part (figs. 18, 19), and in part 

 (figs. 20, 21) referable to the shell illustrated in figure 7. 



There is something' peculiarly unsatisfactory and disheartening in 

 trying to straighten such a tangle of shells and casts, for the range of vari- 

 ation of a species should be established, wherever possible, from well pre- 

 served shells — from the radulse, the biologist would maintain — and until 

 this has been done, there is no way of knowing with any degree of certainty 

 whether or not the casts and the shells are all distinct or all identical, and 

 the only result of such an investigation may be a further confusion of the 

 nomenclature which must be perpetuated in the future synonymies. It 

 may, however, function as a warning to recognize the limitations of poorly 

 preserved specimens and serve as an argument against any attempt to 

 characterize forms which have no characters preserved, unless indeed the 

 stratigraphic significance of the form demands its recognition. 



Occurrence. — Monmouth Formation. Bohemia Mills, mouth of 

 Turner's Creek, on the Sassafras Biver, Cecil County ; Brightseat, Prince 

 George's County. 



