86 CARBONICOLA, ANTHRACOMYA, AND NATADITES. 
in Anthracomya being well developed compared with those in Anodon, and as far 
as my observations go, are never eroded. The accessory adductor muscle, too, 
has a different position. I have never been able to make out that Anthracomya 
possessed a byssus, though, in all species, the valves have a well-marked oblique 
sulcus and a sinuated inferior border, indicating the connection with a byssiferous 
ancestor, or that they themselves were byssiferous in the fry. D’Orbigny quotes 
a byssiferous form of Anodon, being found in the river Paranda, South America. 
In my paper on the affinities of this genus in the ‘Quarterly Journal of 
the Geological Society’ (vol. xlix, 1893, p. 258), on similar grounds to those 
advanced here, I contended for the replacement of this genus in the family 
Unionide ; but I was not aware at that time of the extent of the literature of the 
subject, as may be noted by the number of synonyms given. 
De Ryckholt (‘Mélanges Paléontologiques,’ 1848) appears to have referred 
some forms of Anthracomya to Cardinia (Agassiz); but great uncertainty 
prevails about de Ryckholt’s type-forms, which I am unable to trace; Professor 
Dewalque, of Litge, kindly showed me his collection and pointed out forms 
which he had named after de Ryckholt from the same localities whence the 
originals were obtained. In looking through the collection in the Brussels 
Museum I found typical Anthracomya labelled as Cardinia and Mytilus after 
de Ryckholt. From Mytilus, Anthracomya can easily be distinguished by the 
absence of the triangular, wedge-shaped form, the non-terminal position of 
the umbones, the absence of the scars for the byssal muscles, and the erect 
external ligament. Indeed, Anthracomya approaches far more closely to Modiola, 
to which genus Brown (‘Fossil Conch.,’ 1849, p. 172, pl. Ixxi, figs. 24, 25; 
pl. Ixxu, figs. 19, 20, 24, 25, 27, 29) and Romer, 1870 (‘ Geologie und Palionto- 
logie von Oberschlesien,’ p. 76, foot-note), refer typical shells; but Anthracomya 
has not the inflated anterior portion, internal ligament, byssus, or muscle-scars of 
Modiola, and the form and position of the umbones in either genus is markedly 
different. Anthracomya possesses a well-marked lunule, a character absent in 
the family Mytilidae. From the sub-genus Modiolopsis, to which shells having 
the typical appearance of Anthracomya are referred by Hichwald, the same 
characters will serve as a means of separation. 
Sir W. J. Dawson proposed the name Naiadites for certain Coal-measure 
Lamellibranchs from Nova Scotia (‘ Acadian Geology,’ p. 43, 1860), but Mr. Salter 
pointed out that the group contained members of more than one genus (‘ Quart. 
Journ. Geol. Soc.,’ vol. xix, 1863, p. 80), some of which he correctly referred 
to his new genus Anthracomya (see also ‘Quart. Journ. Geol. Soe.,’ 1894, 
vol. 1, p. 437, et seg.; and antea, pp. 25, 26, and 27). 
Ludwig proposed at least three names for members of the genus under notice, 
two of which—Unio and Anodon—need not be more than mentioned here; but it 
