MOLLUSCAN GENUS COCIILITOMA. 263 



condition seen in var. obesa shows that this is not a character to 

 be relied upon. It enters the penis sheath in f, half-way down 

 its length. In f a its point of insertion is doubtful. With regard 

 to the actual junction of the vas deferens and penis a general 

 approximation is made in f„ to the condition seen in z„. The 

 length of the vagina and the form and proportions of the sperma- 

 theca differentiate both examples dissected from var. obesa. 



III. Genetic Affinities and Variation of Cochlitoma 



ZEBRA VAR. OBESA AND FULGURATA. 



The author does not feel justified in altering the position 

 assigned by Connolly (1912) and others to G.fulgurata Pfr. as a 

 variety of G. zebra. Although it differs from var. obesa in certain 

 conchological and anatomical features, conchological intermediates 

 (British Museum, Zool. Dept.) have been found which cannot be 

 disregarded. 



It is a commonplace among systematists that between two 

 varieties of a " species" or two " closely allied species'' there may 

 be complete agreement in certain respects and difference in 

 others. It is plain, however, that in the case of the two varieties 

 of C. zebra, the individuals of a variety do not differ from those 

 of a closely allied variety in respect of the same characters. 

 Though they may all differ to the same extent from the typical 

 form in respect of, for example, two characters, they do not all 

 show the same degree of difference in their other characters. In 

 the present instance var. obesa differs from var. fulgurata in 

 having a truncheon-shaped spermatheca with a short duct 

 as opposed to a spatulate spermatheca with a long duct, and a 

 shell with narrow brown and white stripes as opposed to broad 

 and jagged stripes of brown and yellow. But one individual of 

 fulgurata agreed with obesa in having a narrow and pointed 

 cervical lobe, Avhile the other obesa had a broad flat one. Of the 

 three forms that agreed in this respect one (z x ) differed from the 

 two fulgurata and z„ in having a straight columella as opposed to 

 a sinuous one. Again, one example of fulgurata and one of obesa 

 agreed approximately in the length of the vas deferens, while one 

 fulgurata agreed with the two obesa in the relation of kidney to 

 pericardium and differed from its fellow. Again, the same obesa 

 agreed with fulgurata (cf above) in respect of the columella, and 

 yet the relationships of the two obesa are reversed in respect of 

 the form of the rhachidian tooth of the radula. Two conclusions 

 may be drawn from the above data : — 



(1) That the amount of correlated variation in this case is 



very slight. 



(2) That the degree of difference or agreement is independent 



of the type of organ or structure involved ; though gland- 

 ular tissues appear to be more unstable than the others. 



The size and possibly the shape of glands, however, are 



