GILLS OF LAMELLIBRANCH. 6^ 



I know, it has never yet been described. That Dreissena, 

 being a fresh-water form, shouhl agree in its gill-structure 

 with Anodon, another fresh-water form, rather than with 

 Mytilus, a far more closely related though marine form, is in 

 itself a most noteworthy fact. A reference to Plate VII and 

 its description will place the main facts before the reader. 

 The specimens which I have studied were procured from the 

 Oxfordshire Canal, where D. polijmorpha is very abundant. 



As in Anodon, we have in Dreissena gill-plates in which 

 concrescence has fully asserted itself. There are no " ciliated 

 junctions " but fibrous inter-filamentar junctions (fig. 28, 

 page 9, Plate VII) as in Anodon, forming a rectangular 

 basket-work with the filaments. Further, the sub-filamentar 

 outgrowths form an extensive layer of tissue beneath the 

 filaments proper, united into one continuous mass by con- 

 crescence, excepting for the " water-passages," which are 

 arranged much as in Anodon, The descending and ascending 

 lamellae of each gill-plate are united by vertically continuous 

 interlaraellar junctions, which resemble those of the outer 

 (brood-bearing) gill-plates of Anodon. There are large 

 vertical vascular canals present in the sub-filamentar tissue, 

 similar to those of Anodon. These are disposed similarly to 

 the large vertical vessels in the outer gill-plate of Anodon, 

 but with less regularity, sometimes occurring hetweeti the 

 points of origin of inter-lamellar junctions instead of at those 

 points. It is remarkable that this should be the structure of 

 the Dreissena's gill-plate, since we do not know that it has a 

 brood-bearing function as has the similarly constituted plate 

 of Anodon. 1 could ascertain no difference of structure 

 between the outer and the inner gill-plates of Dreissena. 



The most remarkable diiferences between the gill-plates 

 of Anodon and Dreissena are the smaller size of the fila- 

 ments in Dreissena ; the presence of two rows of long ciliated 

 latero-frontal epithelium as in Mytilus, in place of one row 

 as in Anodon ; the larger relative size and smaller number 

 of epithelium cells on the filaments of Dreissena; the smaller 

 relative development of the sub-filamentar tissue ; and the 

 different structure of tJie chitinous deposit in the filaments. 

 This chitinous deposit is a thick horse-shoe shaped mass in 

 transverse section resembling that of Anodon, and totally 

 different from the closed delicate ring of chitin Avhich we 

 find in a transverse section of the Area or Mytilus filament. 

 It is even more exuberant (see Plate VII, fig. 26) in 

 development than the similar structure in Anodon, but 

 presents no trace of the denser "rods" which are differentiated 

 in the latter genus. A careful examination of the chitinous 



VOL. XVII,-«*^NBW SER. E 



