94 MONOGRAPH OF THE GENUS GNATHODON—DALL. 



above with the inner iinterior edges of the somewhat shorter upper pair. 

 The inner surfaces of both are striated, the outer surfaces smooth. 

 Kang indicates the month beh)w the h)wer palps, which is obviously 

 erroneous. 



The gills are of moderate size, two on each side, the inner pair larger, 

 and hung by their upper edges from the visceral mass. The attach- 

 ment of the outer pair is about a millimeter higher, separated from the 

 suture of the inner gills by a fine very tender membrane; the line of 

 attachment divides the outer gill at its upper third, the upper portion 

 is more or less fixed upon the mantle by slender adhesions toward its 

 middle third, and bent downward, but is more free before and behind. 

 This reflected portion of the outer gill is what is often referred to as 

 the " appendix." Both gills are joined by a delicate membrane behind 

 the retractor pedis (where the width of the two gills is approximately 

 equal) to each other, to the j)air of the opposite side and to the 

 siphonal septum, forming a complete partition between the aual and 

 peripedal chambers. This is also found in the various forms of Jfacfra 

 examined and in Cijrena Jloridana^ though in the latter the attachments 

 are extremely delicate. Below the septum in Gnathodon a thin arched 

 membrane forms an imperfect valve at the baseof theincurrent siphon, 

 as in Mactra, but in Cyrena this was hardly perceptible. The whole 

 surface of the gills is finely striated, of a dull cream color, vertically 

 barred with about twenty dark brown transverse lines. In all the 

 other species examined the gills were colorless. In Ranges figure the 

 aual and peripedal chambers are wrongly represented as communi- 

 cating behind the gills, which error was doubtless due to rupture of the 

 membranes. ^ ip 



So far as the soft parts are concerned, it will be seen from the preced- 

 ing notes that Gnathodon^ Mactra, and Gi/rena agree essentially in the 

 general structure and attachments of the gills, in having a separate 

 anal chamber, in the general form of the foot (shortest in Cyrena and 

 longest in Mactra), in the separation of the mantle lobes (somewhat 

 greater in Cyrena), and in the absence of a byssus. 



Gnathodon agrees with Mactra in having the siphons united to their 

 ends and the incurrent one furnished with an imperfect basal valve 

 and with an elevated raphe behind it. It agrees Avith the Mactridw in 

 having an internal cartilage and with Mulinia in having both the 

 (ordinarily external) ligament and cartilage internal and contained in 

 the same socket. In all these features Mactra and Gnathodon differ 

 from Cyrena and its allies, all of which liave only an external ligament 

 set in a groove and separated by an elongated fulcrum, or nympha, 

 from the cardinal border. 



In considering the evidence of the harder parts all the species of 

 Gnathodon must be examined, the typical species being more extreme 

 in its characters than any of the others. It is also necessary to exam- 

 ine very young specimens, which are extremely difficult to get hold of, 

 notwithstanding the abundance of the species on the Gulf coast. 



