342 On Shell-banding as a Means of Protection. 
winter ; a collector seeking living shells in their natural 
haunts, since he would be working under conditions some- 
what similar to the birds themselves, would probably secure 
an excessive proportion of unbanded shells. During the 
winter, however, the relative absence of vegetation makes it 
easy to see dead shells of all types, and so the standard 
collection gives a fairly correct representation of their 
proportions. 
Striking confirmation of Mr. Woodruffe- Peacock’s 
observations on the localization of the different varieties was 
obtained while making this collection; thus, in the lane 
near Broxtowe for a distance of 50 yards quite four-fifths of 
the shells found had one band only. 
The complete details of the collections were as follows :-— 
Standard Collection, ‘ Anvil” Collection. 
Unbanded ) icjeccp 25 per cent. 38 per cent. 
band aoe Ste eas HGS 08 aes 
2 Be mae aes wie ak Ada De oe 
Si 5s pate ne tere Ga Mae 
OPA eke eee Q: tes Sigs 
D dos Geass cree eet LV pre OB crcigs 
6 lessthanl  ,, ie 
Thus, although fully two-fifths of the standard collection 
had the normal five bands, little more than half this propor- 
tion of the broken shells were somarked. Further, although 
unbanded shells constituted only a quarter of the standard 
collection, they occurred in greater numbers among the 
broken shells. The chances of an unbanded shell being 
observed are, according to these figures, about three times as 
great as of a normal shell. Stated more concisely, in the 
standard collection there was an average of 2°9 bands per 
ae ; among the broken shells the average was much lower, 
Oper shell. 
teas figures appear to confirm previous work, although 
the present investigation has been carried out in a different 
district and in quite a different manner. Certainly it seems 
clear that banding alone is not a guarantee of safety—as a 
matter of fact, a reasonably large percentage of banded 
shells are detected by the birds. Yet the figures appear to 
show that banded shells are less liable to be seen—or, at 
least, to be eaten—than unbanded. 
In concluding, I should like to express my indebtedness to 
Prof. J. W. Carr, Prof. H. H. Swinnerton, and the Rev. E. A. 
Woodruffe-Peacock for the assistance they have given me in 
carrying out this work. 
