DARBISHIRE : PROFESSOR LANg's BREEDING EXPERIMENTS. I9S 



in which band 3 was missing, a state of affairs which is described by 

 the formula r 2045 j this occurred in experiment 35, and Lang states that 

 it is the sole instance of what may be regarded as a mutation which 

 has occurred in his experiments. 



B. 00000 X 00000. 



Here, again, T pass over the description of the cases in which it is 

 not quite certain that the father was unhanded, and proceed to those 

 in which it is certain that both parents were unhanded. The families 

 produced by o x o fall fairly sharply into two groups : one in which 

 the broods are composed entirely of o ; and another in which both 

 o and 5 appear. 



To the first group belong experiments 43, 44, and 48 ; though it 

 must be noted that in experiment 43, out of thirty-five young with 

 formula 00000, one shewed a faint indication of banding^; that in 

 experiment 44, out of the fifty-seven young, one example shewed an 

 "uncertain indication of bands i, 2, 3";- and that in experiment 48 

 none of the young had arrived at anything like maturity.^ 



To the second group belong, for example, crosses 41 and 42 : in 

 the former, of the 78 young, 77 were o, and one was 00300 — this 

 probably was a young stage of an adult whose formula would be 5, 

 because no banded individuals which exceeded seven mm. in dia- 

 meter had any other formula than ^'^ ; another example of the second 

 class on which Lang lays much stress is cross 42, in which of forty-one 

 examples (none of which were larger than 5*5 mm.) thirty-one are 

 without discernible banding, while ten are 00300, i.e., snails which, for 

 the reasons just mentioned, will probably become adult as 5. 



C. 00000 X 12345. 



The results of this type of mating fall into two classes : one in 

 which all the offspring are o ; and another in which the broods 

 consist of o and 5. 



To the former class belong experiments 49 — 54. It must, however, 

 be noted that of the fifty-nine offspring in experiment 49, two shew 

 the faintest suggestion of banding ; and that there are ill-defined 

 indications of band 3 in isolated examples of the sixty offspring in 

 experiment 53. Owing to the fact that the young were born before 

 the parents were isolated in experiment 49, it could not be deter- 

 mined whether the brood was produced by the o or by the 5 or 

 by both ; in experiments 50, 51, and 52, the o (of experiment 49) was 



1 p. 482. 



2 p. 482. 



3 p- 484- 



4 Moreover, the condition 00300 though perhaps even commoner than 12345 '" Helix 

 neinoralis, is very rare indeed in H. hortensis, in which the variability of the Ijands is much less 

 4efinite than in H. nemoralis. 



