PROCEEDINGS OF THE NATIONAL MUSEUM. 93 



ties of the shell. Dr. Fischer also raised the group to family rank, 

 which, if it be compared solely with Cyrenickv, is reasonable, but, if the 

 comparison is with the Mactridfv, and sufiBciently full material is con- 

 sulted, it will be seen that there are really no characters which remain 

 after the characters common to Mactras and Muliuias are excluded, 

 ui)on which even a subfamily (;an be based. The distinctive characters 

 of the genera of the Mactridw merge so gradually from one form into 

 another that we are foi'ced to the opinion that Gray and Desmoulins 

 were riglit, and that the group can only be ranked as a genus, next to 

 Mulinia, in the Mactroid series. 



In the endeavor to come to a well-founded conclusion in regard to 

 the affinities of Gnathodoii, a careful examination was made of the soft 

 parts of G. cuneatus from 3Iobile and Texas; Mactra {SpisulaJ similis, 

 Say, Florida; M. ( S.) polynnma Stm , Alaska; M. ( Mulinia) lateralis 

 Say, Massachusetts; (■yreiia carol incnsis and Cyrcna JlorUJana Conrad, 

 from Florida. I received half a dozen Gnathodons from Mobile Bay 

 alive, by mail, through the kind intervention of Mr. (i. D. Harris ; and 

 others, in alcohol, from Port Lava<'a, Texas, from Mr. J. D. Mitchell. 

 Several errors were found in liang's account of the macroscopic anatomy, 

 leading to the suspicion that he dealt with specimens which had already 

 been removed from the shell when he received them. The following 

 notes were made from the specimens: 



The foot of Gnathodon cuneatus is like that of Mactra^ but shorter and 

 more compressed. There is no external indication of a byssal gland. 

 The retractor muscle of the foot on each side is attached to the under- 

 side of the cardinal border above and near the adductor. 



The siphons of Gnatuujon are short, but united to the ends, as in 

 Mactra. The incurreut siphon is papillose at its orifice, the excurrent 

 siphon smooth-edged, or very finely pai)iliose, differing in different 

 specimens. The external surface of the siphons is of a dark olive 

 color, nearly black where most intense, with a lighter line conforming to 

 the intersection of the vertical i)lane between the valves with the sii)ho- 

 nal commissure. The mantle margin is wide and smooth, the distal edge 

 thin, blending with the papery ei)idermis, the inner edge thick, smooth, 

 and somewhat elevated. The anterior commissure is in front of the 

 adductor, thence backward the lobes are separated three-fourths of the 

 way to the siphons, much as in Mactra. A short distance within the 

 margin, beginning in the posterior half of the shell and extending back- 

 wards to a point under the shade of the valve of the incurreut siphon 

 is an elevated raphe of tissue which divides the incurreut channel. A 

 similar arrangement is found in Mactra, but not in Cyrena. This ridge 

 is probably the seat of sensory tissue analogous to the osphradium of 

 Gastropods. 



Tlie palps are triangular, slender, rather long, the lower ones extend- 

 ing to the posterior fourth of the foot on each side, continuous below 

 the mouth, where they are soldered to the visceral mass, and joined 



