64 R. HOLMAN PECK. 



the trabeculse, uniting one such cell-centre to another. In 

 fact, we have a slightly modified series of branched corpuscles 

 joined by their branches, the branches being band-like rather 

 than filamentous. In the spaces between the trabeculee float- 

 ing blood-corpuscles are seen; a granular coagulum some- 

 times occupies these spaces. The fluid in the spaces is the 

 blood-lymph. This kind of lacunar tissue occurs most abun- 

 dantly as the representative of " mesoblast " in the Anodon's 

 gill and other parts. 



In fig. 30 a dense modification of the " mesoblastic ^' 

 tissue is shown, which cannot be distinctly separated from 

 lacunar tissue because it passes into it. The cells here are 

 spindle-shaped and closely packed, but leave occasional nar- 

 row passages among them. You have only to separate them 

 from one another and pull them out into connecting bauds 

 and processes, and you get ordinary lacunar tissue. 



This dense lacunar tissue with fusiform cells occurs in 

 the interlamellar junctions of both Anodon (PI. V, fig. 9) 

 and Dreissena (PI. VII, fig. 25). It probably very readily 

 passes into the wide-set variety with trabeculse, and vice 

 versa. It occurs in many other parts of Lamellibranchs and 

 other Molluscs, and has been wrongly looked upon as the 

 chief or only kind of "connective tissue" in these animals. 



The lacunar tissue of fig. IS is the typical variety. It resem- 

 bles the earliest formed mesoblast in embryonic Gasteropods 

 and other animals, where, however, in parts it is contractile, 

 a property of which we have no evidence in the gill. 



In (fig. 19) a drawing is given to show how the lacunar 

 tissue behaves itself in condensing to form a blood-vessel. 

 The floor of the little horizontal vessel v' is formed by such 

 fusiform cells as those seen in fig. oO; in both cases mere 

 modifications of the trabecular cells. 



Measurements of Anodon Gill- structure. — The breadth of 

 the gill-filament in a well-grown Anodon taken from the 

 outer edge of one row of latero-frontal cells to the outer 

 edge of the opposite row is ^ ',7^th of an inch. 



The length of the large cells of the latero-frontal epithelium 

 is _-^-th of an inch, and the breadth ^j^-^Virth of an inch. 



The lumen of" the water-passages as seen in sections par- 

 allel with the surface of the lamella varies in diameter 

 from the to'ot,^'^ to the ^i-oth of an inch. 



Dreissena. — The structure of the gill in this genus 

 presents highly interesting points for comparison with Mytilus 

 (in which genus it was formerly placed), and with Anodon. 



On the whole the gill-structure of Dreissena is decidedly 

 nearer to that of Anodon than to that of Mytilus. So fur as 



