452 CARBONIFEROUS LAMELLIBRANCHIATA. 



mucli larger and farther apart in the anterior part of the shell, and closest and 

 narrowed on the posterior surface. 



Observations. — The systematic position of Conocardium has been a matter of 

 much difference of opinion. Agassiz thought the genus belonged to the Brachio- 

 poda. Other observers thought it had a greater affinity to Gardium. Others 

 again referred it to Liinulicard'mm, a genus of Palaeozoic shells whose internal 

 characters have not yet been observed. There is no one character, however, 

 which permits Conocardium to be referred to Gardium ; and, indeed, the pecu- 

 liarities of the genus are so well marked, and they differ from all known genera 

 so tlioroughly that I do not hesitate to adopt the family Gonocardiidds to receive 

 the genus. At the same time it is very difficult, if not impossible, to assign this 

 family its proper systematic position. 



Of recent shells, Tridaciia has the greatest superficial resemblance to Gono- 

 cardium ; but it cannot be said that any of the peculiar characters of the latter 

 genus are to be noted in Tridacna. 



There has always existed considerable doubt as to the proper orientation of the 

 shell of Gonocardium, several authors — Woodward, Barrande, Hall, and Halfar — 

 considering the tubular process to be posterior, and to have contained the siphonal 

 tubes ; others — de Koniuck, Fischer, Neumayr, and Beushausen — maintaining the 

 reverse. After long consideration, and the examination of very perfect specimens 

 of the interior, I consider that the anatomical evidence is in favour of the view 

 that the truncated end with the prolonged rostrum is posterior. The facts Avhich 

 have caused me to arrive at this opinion are the following : 



PI. LI, fig. 10, a very perfect example of Gonocardium Herculeum from 

 Tournai, shows remains of the external ligament, which occupies a depression 

 between the umbones and the long rostrum ; and this surely cannot be placed 

 anterior to the umbones. The anterior adductor scar is in ordinary bivalves the 

 deeper and better marked, and in many genera is bounded behind by a ridge of 

 shell; in Gonocardium (PI. LI, fig. 11 (') the muscle-scar, h, at the compressed and 

 produced end of the valve is very well-marked, and there is a strong ridge which 

 bounds its posterior margin, which in itself would, I think, make the extrusion 

 and withdrawal of siphon-tubes impossible, for this ridge is continued to within 

 a short distance from the margin of the valve. The posterior adductor is, as 

 generally obtains in bivalves, shallow and indistinct, and is remote from the 

 margin of the shell, and in the normal position, if the truncated, expanded surface 

 of the shell be regarded as a very accentuated dorsal slope. 



Between the wing and the body of the shell is a well-marked constriction, 

 corresponding, I think, to the byssal sulcus in other genera ; and where this fold 

 meets the margin the peculiar ridges on the inner edge of the valve are markedly 

 worn. The shell, it is true, gapes very considerably at the winged end, but I 



