CARBONICOLA ROBUSTA. 45 



erosions (see PI. IV, figs. 8 — 12). It is curious, however, that I have been able to 

 make out this peculiar feature in no other species of Carbonicola, but erosion 

 is present in similar forms from other localities ; and this eroded condition of 

 the umbones is by no means universal, even in the bed mentioned. I conclude 

 that erosion would not take place provided that the periostracum remained intact, 

 and also that the presence of sufficient CaC0 3 to neutralise the free carbonic acid 

 in the waters in which these shells lived would largely prevent erosion. It is 

 still, however, a matter of great uncertainty as to what is the real cause of erosion 

 in Unios ; and, indeed, in some localities they are not thus affected. 



The nearly universal occurrence of truncate and somewhat ventricose varieties 

 of each species suggests the question as to the possibility of the dioecious 

 character of these Coal-measure Lamellibranchs ; this, of necessity, is a question 

 which cannot be settled in the absence of all data, though Ludwig and Achepol 

 have ventured so far as to describe certain tumidities in the upper and posterior 

 part of specimens as due to the situation of the ovary — a somewhat unwarrantable 

 assumption, judging from what is known of the sexual organs of the JJnionidse. 1 



It is interesting to uote that similar variation of forms occurs in the species of 

 Anthracomya and Naiadites 



I have considered it unnecessary to raise Anthracosia to the dignity of a Family, 

 as I think that of the Unionidse to be sufficiently wide to embrace a genus which 

 seems to present only one important difference from the other genera included in 

 it, viz. the abnormal (qua TJnio) situation of the anterior-pedal muscle-scars. 



1. Carbonicola robusta (Sowerby). Plate I, figs. 1 — 6; Plate II, figs. 1 — 6, 9 — 11. 



Unto eobustus, Sow. Trans. Geol. Soc, ser. 2, vol. v, pi. xxxix, fig. 14, 1840. 

 Cakdinia eobusta, de Koninck. Anim. Foss. Belg., 1842-4, p. 71, pi. ii, fig. 1. 

 Unio eobustus, Brown. Foss. Conch., 1849, p. 181, pi. lxxxviii, fig. 16. 

 Caebonicola subconsteicta, var. eobusta, M'Coy. Brit. Palaeozoic Foss., 1855, 



p. 515. 

 Non Antheacosia eobusta, Hull. Coal-fields of Gt. Britain, 1861, pi. i, fig. 1. 



— — Salter. Mem. Geol. Survey, Country round Wigan, 



Sheet 89, p. 37, fig. 2, 1862. 

 Non — — Wardle. Geol. of Leek, 1863, pi. iv, fig. 13. 



Antheacosia cbassa, Achepol. Niederrheinisch.-Westfalische Steinkohlen-Gebirge, 



Atlas, 1880—1883, Supp., pi. i, figs. 19, 20, 

 1880-83. 



— caudata, Achepol. Ibid., Supp., pi. iii, fig. 49, 1880-83. 



— eobusta, Ward. Trans. North-Staffs. Inst. Min. Mech. Engineers, 



vol. x, p. 127, pi. i, fi^. 1, 1890. 



1 Though generally said to have the sexes distinct, I have never by dissection been able to satisfy 

 myself that such was the case in the Unionidae. 



