lO I Tickling, Variation of Planorbis inultiforniis, Broun. 



uous variations. It is shown by the figures just mentioned 

 that the development of the dorsal carina is quite 

 independent of the height of the shell. The ventral carina, 

 however, shows a certain correlation with that character. 

 It is never strongly developed in " high " forms. In "low" 

 forms it shows the same variations as the dorsal carina. 



As regards its ontogeny, the carination exhibits three 

 conditions, as follows : 



{a) The carina may be strongly developed at an early 

 stage and may remain in that condition throughout the 

 growth stages. 



(/;) The carina may be strongly developed at an early 

 stage and then be gradually lost. 



(<r) The carina may be feebly developed at an early 

 stage and ma}' remain so throughout life. 



The significant fact, therefore, is that feebly carinate 

 adults may develop from strongly carinate young, while 

 the reverse appears never to occur. It may be added that 

 the last whorl of an adult shell generally shows decreasing 

 carination. These facts, taken together, make it fairly 

 clear that this character is now in a catagenetic condition. 



The first whorl of the shell is, nevertheless, invariably 

 quite smooth, showing that they have developed from a 

 non-carinate ancestor. Many individuals thus show the 

 complete cycle of the growth and decay of the carination. 



(2) Involution. The extent of overlap of the succes- 

 sive whorls shows well-marked, though not very extensive 

 variation. In a "normal" shell the suture-line coincides 

 exactly with the dorsal carina, but in other cases it may 

 lie a little above or a little below this line. The amount 

 of variation in this respect is well shown by PI. /., Fig. 3 

 A-E. This character shows no marked relation to the 

 height of the spire, the "normal" position of the suture 

 being on the dorsal carina in all cases, while similar varia- 



