189.°).] NATURAL. SCIENCES OF PHILADELPHIA. 399 



14. The rate of increase in number of the teeth will furnish the 

 means for calculating the rate of the new formation of the radula as 

 a whole. This is more rapid than is generally supposed. 10 



1"). It is evident that the morphology of a radula can be 

 thoroughly understood only by examining and comparing the differ- 

 ent stages of its formation. 



In the text nothing has been said about the methods of preparation 

 and manipulation of the radulse, — not much is necessary. As 

 far as possible, the radula has been removed from the fresh 

 animal unchanged. Caustic alkali has been used where necessary. 

 By careful desiccation, and also by the action of chromic acid solu- 

 tion, for staining, valuable results may he obtained, as the teeth 

 and especially the plates of attachment shrink somewhat and thus are 

 separated from each other, and seen more exactly. But it is almost 

 unnecessary to add that this means can only be used in connection 

 with observations on the intact radula, for without this check artificial 

 might he mistaken for natural features. 



Explanation of Plates 10 and 11. 



Plate 10, Fig. I — Radula of an embryo of Limax campestris Say, 



not quite mature, showing first part of radula 

 formed. 

 II — Radula of a specimen a few days old (after hatch- 

 ing), and measuring 3*5 mm. long. 



A. 30th transverse row (about 42nd of total 

 number). 



B. 45th transverse row (about 57th of total 

 number). 



Ill— Radula of a specimen 7 mm. long, one- half of 



a transverse row, formula c. 10T8. 

 IV — Radula of a specimen 15 mm. long, one- half of a 

 transverse row, formula, c. 14-27. 

 V — Radula of a specimen 15 mm. long, one-half of 

 a transverse row, formula, c. 14-32. 



"'Without repeating the calculations here, it may be said that I found the 

 radula of Umax campestris to form about 800 transverse rows in all — probably 

 considerably more : this would mean that the radula is about eight times entirely 

 changed during life. For Polygyra (Mesodon) thyroides about 2000 transverse 

 rows were found, corresponding to about 16 to 18 total renewals. 



