NOTES ON ANODON AND UNIO. 51 



and left thickened borders together in the middle line, while also 

 increasing their thickness and offering a more solid resistance to the 

 water. Furthermore, when once the thickened borders of the mantle 

 are in apposition and the shell commences to gape, the pressure on the 

 right and left free borders will tend to drive them even more closely 

 together; for the line of the mantle which is attached to the shell 

 must of necessity follow the outward movement of the valves when 

 gaping commences, and the free borders unite to form a bluntly 

 pointed longitudinal ridge with divergent sides ; the pressure of 

 water falls on these divergent sides and drives them together — the 

 whole structure thus acting in the manner of the mitral valve of the 

 human heart. It is probable that the flexible margins of the valves 

 are also driven together by the pressure of water. The diagram 

 exhibited (Plate I. fig. 7) may make this clearer. 



I am inclined to think, then, that a suction of this kind is used 

 to swiftly draw the ova forward into the external gill-plate. Direct 

 observation on this point is well nigh impossible owing to the necessity 

 of disturbing the animal or even partly opening the shell in order to 

 ascertain whether or no ova are in transit. The fact that violent 

 suction does take place in the case of the Glochidia is beyond doubt ; 

 the exact mode of causing the suction is, for our present purpose, of 

 less importance. 



The question naturally occurs, why do not the ova find their way 

 into the internal as well as the external gill? The reason is, I 

 think, twofold. In the first place, the space between the lamella? 

 of the external gill is considerably greater than that between the 

 lamellae of the internal gill. In the second place, as I have ascer- 

 tained by careful dissection of many individuals, the inner lamella 

 of each external gill-plate extends further towards the dorsal surface 

 than the outer lamella of each internal gill-plate, and stretches over 

 towards the middle line so as to greatly diminish or even totally 

 close the aperture leading into the space within the internal gill. 

 In some cases the inner lamellae of the external gill-plates of the 

 right and left sides actually come in contact with one another in the 

 median line posteriorly 1 . 



1 This, of course, applies only to the post-pedal portion of the gill-plates. 

 In the region of the foot the "labour contractions" close the space between 

 the lamellae of the internal gill, as stated by von Baer. 



