6 DRY.M.EUS, FLORIDA. AND WEST INDIES. 



third or third and fourth bands dotted, .or they have all the bands 

 subcontinuous as in most Floridian shells. 



Floridian specimens (pi. 5, figs. 18, 21, 22, 23, and pi. 20, fig. 32) 

 are the most variable of all. In one lot from the Caloosahatchie 

 river there are two specimens with the typical band-formula 12345, 

 upper three bands dotted ; four with the formula 10345, the upper 

 band very faint,- dotted, the peripheral also dotted; two without 

 bands, and one (pi. 20, fig, 3-2) with faint traces of the basal bands. 

 Mr. Hemphill found the same form in the vicinity of Charlotte Har- 

 bor, Fla., -one- of his specimens'.figured .by Binney (Man. Amer. Land 

 Shells,, p. 408, fig. 449), having the formula 12345, the upper three 

 bands broken into dots. , The shell, in this form, is often. more ven- 

 tricose with shorter spire than in most Cuban marielinus, but this 

 varies a good deal. Mr. Wright's B. hemphilli applies to these forms. 

 It is the ordinary South Florida form of the species, stated to be a 

 "thinner shell than J5. marielinus Poey,. and more corpulent, while 

 the revolving bands are redder, finer and continuous in. the last-named 

 species ; the substance of the shell of B. marielinus is white, while 

 that of B. hemphilli is light amber colored." These supposed differ- 

 ences are largely due to erroneous ideas regarding marielinus, which 

 is really not " white " nor has it " continuous " bands ; and Cuban 

 specimens are .fully as, thin as "J9. hemphilli." 



Specimens from Upper Matacumba Key, near Miami river, and 

 Micco, vary from 12345 to 02345 or Q0345, all bands generally con- 

 tinuous though irregular (pi. 5, (igs. 22, 23). These often exceed the 

 dimensions of .the largest Haitian shells I have seen, reaching as great 

 a size as alt. 21^, diam. Ill, length of aperture 10.J mill. t The shell 

 varies from as obese as typical dominicus to as narrow as Binney's 

 figure of " marielinus." This is what was figured by Binney as B. 

 marielinus. A Cuban specimen before me has the same characters, 

 alt. 17, diam. 8^, length of aperture, 10^ mill. 



Still another Floridian form remains to be noticed : the bandless, 

 oblong shells, light brown, very translucent, with indistinct corneous 

 streaks, alt. 19^, diam. 10, length of aperture 10 mill. (pi. 5, figs. 18, 

 21). The apical whorl is a little more depressed than in the typical 

 dominicus. It occurs at Lake Helen, Volusia Co., and between 

 Mosquito lagoon and the Atlantic. This local variety, which has no 

 name, has been referred by Mr. Wright to his B. hemphilli, but is 

 not mentioned in the original account of that form. Similarly col- 



