61 



the left. The shell, therefore, is clearly neither an Anodonta nor a 

 Margdritaiid. 



Estimated length of a fairly typical specimen, two inches and nine 

 lines; actual height of do., fi'oni beaks to base, one inch and nine lines ; 

 maximum thickness, one inch and three lines. 



Probable length of a much more elongated individual, three inches 

 and one line; ac-tual height trom beaks to base, (the beaks being much 

 eroded) one inch and eight lines; thickness, scarcely fifteen lines. 



Tlie shaded pai-t of the figure on Plate IX. is intended as a represeuta^ 

 tion of the elongaled and attenuated variety of this species. In this 

 instance the dotted lines are not added b}- way of restoration, but to 

 show the shape of another individual. The majority of specimens are 

 much shorter, and tjie downward slope of the anterior extremity is 

 usually more decided. Mr. Gabb's partly restored drawing of the 

 original type is slightly inaccurate ; the hinge border behind is too 

 straight and its downward inclination is not sufficiently expressed. The 

 posterior end is too wide and its ujjper margin not convex enough. 



The locality from which Unto Huhbardi was first obtained is thus 

 described by the author of the species : " A single specimen, from the 

 Nanaimo Coal Mine, Vancouver Island, Chico Group, kindly loaned me 

 by Mr. Hubbard, of the Pacific Mail Steamship Company of San Francisco, 

 and to whom I dedicate this species, in recognition of the unostentatious 

 but valuable services he has been rendering to science for a series of years 

 past." The statement that this fossil was found in the Cretaceous Coal 

 fields of one of the Islands of the Vancouver group, is probably a 

 mistake. At any rate it has not been recorded hy any subsequent 

 observer as occurring in that region, nor can any ti-ace of it be found 

 in the large and imjjortant collections made by Mr. Eichardson at 

 these islands during the past five years. On the other hand, it is 

 not only one of the commonest species in the Carbonaceous Shales 

 near Cowgitz on Graham Island, but it is the only mollusc yet detected 

 in them. 



The elongated and attenuated variety of U. Subbardi described above, 

 is barely distinguishable from the Unio Aduncus of Sowerby, a fossil from 

 the Wealden Formation of Tilgate forest, in England. U. Aduncus seems 

 to be rather straighter at the anterior margin, and its base is ajDpa- 

 rently rather more gibbous under the beaks than arc the correspond- 

 ing parts of the Queen Charlotte Island shell, but these differences do 

 not appear to be constant, and in any case are scarcely of specific 

 importance. The figures in the '' Mineral Conchology " are taken from 



