NATURAL SELECTION 203 



The results, if we allow for the rather low numbers, demon- 

 strate the value of the harmonising colours. As Robson 

 {I.e. p. 213) suggested, the selective value of the colour would 

 only be established for animals living freely if it could be 

 shown that it was accompanied by the habit of choosing an 

 appropriate background. Further, the contrast provided in 

 the experiment would be sharper than that usually found in 

 nature. Beljajeff (1927) repeated these experiments, using 

 brown, yellow and green forms of Mantis. On a brown back- 

 ground, out of 20 of each form, 11 green, 12 yellow and 

 4 brown were eaten in a fortnight. In a second experiment 

 some crows in 24 hours ate n green, 12 yellow and 12 brown 

 from the same background. 



(8) Poulton and Saunders {i8gg) : differential elimination of the 

 pupa of Vanessa urticae in different situations. 



The authors exposed the pupae on backgrounds of various 

 kinds (tree-trunks, fences, etc.) at four stations : two in Switzer- 

 land, one at Oxford and one in the Isle of Wight. The 

 mortality was very low in the Swiss loci, which the authors 

 attribute to the lack of insectivorous birds. At the other loci, 

 where the pups were suspended to a background which con- 

 cealed them (from the human observer's eye), there was a lower 

 mortality and more of the pupae emerged. Thus at St. Helens 

 90 were taken by birds (?) and only 8 emerged among those 

 suspended on fences, whereas on backgrounds which served to 

 conceal better, destruction and emergence were more balanced. 

 The numbers in the Oxford experiment were low and of little 

 value. 



The experiments tend to show elimination of pupae if 

 they are placed in conspicuous situations. Experiments in- 

 volving the concealing value of colour led to very ambiguous 

 results and the authors ' cannot make any statement ' as to 

 their value. Moss (1933) came to a similar conclusion after 

 experiments with pupae of Pieris brassicae. 



(9) Boettger (1931) : observations on the selection of Gepea by 

 captive birds. 



The author made experiments on the selection and rejection 

 of various colour- and band-types of C. nemoralis and Arianta 

 arbustorum by captive pheasants in the Berlin Zoological 



