194 JOURNAL OF CONCHOLOGY, VOL. II, NO. 7, JULY, I905. 



wire net stretched across an iron frame. The mesh of this net must 

 be so small that the young, which are only three mm. in diameter 

 when they emerge, cannot escape through the interstices. These 

 boxes may be kept either outside or indoors ; in the latter case, the 

 earth should be sprinkled lightly with water every five or six days. 

 Lang found ti)at families which he kept in some large cellars, where 

 the windows were open winter and summer, flourished better than 

 those out in his garden. The food, which was given every eight or 

 ten days, consisted of carrots, dried hop leaves, or macaroni ; and 

 great care was taken to remove from the boxes the remains of the 

 previous meal. 



Before starting a breeding experiment with a hermaphrodite animal 

 it is necessary, in order to be certain of the parentage of each indi- 

 vidual, to determine whether self-fertization occurs or not ; it was 

 established that in the case of ff. hortensis and H. nemoralis that this 

 did not take place. ^ It was also discovered that the sperm acquired 

 by snail a from copulation with snail b is capable of living a very long 

 time in the vesicula seminalis of a ; for example, one of the snails iii 

 experiment 49' was fertilized in 1900, and without further copulation 

 produced four families in the years 1900, 1901, 1902, and 1903. 



Professor Lang's experiments fall under two headings : (i), matings 

 between varieties of Helix hortensis; and (2), crosses between Helix 

 hortensis and H. ntmoralis. 



Matings between Varieties of H. hortensis. 

 The two varieties of this species that were experimented with were 

 the five-banded variety whose formula is 12345, and the bandless 

 variety with formula 00000. Three kinds of pairs can be, and were, 

 made : r/, 12345 x 12345 ; b, 00000 x 00000 ; and c, 00000 x 12345. 



A. 12345 =< 12345.3 



The author divides the result of this type of mating as well as that of 

 the next into two classes, one in which, while it is certain that the mother 

 of the brood was five-banded, it is not quite certain that the father 

 bore this character ; and another, in which it is quite certain that both 

 parents were five-banded. The chief value of the records of broods 

 which fall into the first class lies in the information which they supply 

 as to the order in which the bands appear during ontogeny.'* 



The record of broods which fall into the second class shews that 

 five-bandedness breeds true; for of the 142 snails which reached a 

 diameter of seven mm. all were five-banded with a single exception, 



1 p. 443. This and the following references to pages refer to Prof. Lang's memoir. 



2 p. 486. 



3 PP- 455. 454, and 474. 



4 For brevity's sake, 1 shall denote 12345 by 5 and 00000 by o. ' • 



