MOLLUSCA OF PLYMOUTH. 407 



further behind. This increased height is due to additional 

 chitin formation at the posterior margins of the teeth, and it 

 thus comes about that the " posterior bay " may be entirely 

 obliterated (PI. XXVIII, fig. 7). This has not before been 

 observed for A. punctata, and breaks down one of the pi'e- 

 viously maintained distinctions between the central teeth of 

 A. punctata and A. depilans. 

 (3) That with age the central cusp of the central tooth becomes 

 less prominent and much more obtuse (Nos. 7, 9, and 10) ; 

 that the external lateral cusp becomes reduced in size, and 

 may even disappear (Nos. 5, 9, and 10) ; and that the in- 

 ternal lateral cusp also becomes reduced in size, and, while 

 quite smooth in young individuals, may become in older 

 specimens irregularly serrated on its external edge (Nos. Q, 

 9, and 10). 



Turning now to the results in the case of the few radulae of A. 

 depilans which I have been able to examine, it is seen that, as in 

 A. punctata, the radula varies in. size and number of rows very 

 largely according to age. My smallest radula (No. 11) is a frag- 

 ment only, and possesses thirty-one lateral teeth. It is obviously, 

 however, an older specimen than my largest punctata, for it is 1'6 mm. 

 wider, and probably possessed fifty rows of teeth originally. The 

 central teeth differ remarkably from those of my largest punctata 

 in being of much smaller size in every way. 



The next (No. 12) is a little larger (0'5 mm. wider), and pos- 

 sesses fifty-six rows of teeth, while the number of lateral teeth 

 increases remarkably in passing from the front to the back of the 

 radula. There are thirty-two lateral teeth behind and twenty- 

 three in front. This radula in this respect, therefore, approaches 

 A. punctata very closely, for in No. 10 the lateral teeth were seventeen 

 in front and twenty behind. As regards number of teeth, therefore, 

 there is no ground for specifically separating these two forms. 

 The width of the central teeth is still much smaller than in Nos. 9 

 and 10, but, as in them, the width shows a crescendo followed by a 

 remai'kable diminuendo in size, passing from the front to the back 

 of the radula. It is impossible to avoid the inference that in the 

 growth of this individual the width of the central tooth has never 

 exceeded that which it attains at the climax of its crescendo (0*3 mm.) , 

 and therefore Nos. 9 and 10 (in which the tooth attains a width of 

 0"4 mm.) cannot easily be regarded as stages in its growth. This 

 inference is also strengthened by the other measurements of the 

 central teeth of the radula, and by comparison with Nos. 13 and 14, 

 which exhibit an increase in the size of their teeth with ag-e. This 



