^rAKNING COLOKATIOK. 187 



theory of warning colour in tte fish or prawn ; and, though 

 the matter clearly requires further investigation, it is at least 

 probable. 



It is likely that other cases of brilliant coloration may 

 have a similar signification. 



The remarkable Mantis, first described by Mr. Wood Mason 

 and figured by Mr. Poulton, may be an instance to the point. 

 Both Mr. Wallace and Mr. Poulton * dwell upon the form re- 

 semblances of this insect to an orchid, as well as its colour 

 resemblances. These are, it must be admitted, not a little 

 striking. The proximal joints of the limbs are flattened and 

 radiate out from the body like the petals of a flower, which 

 they also simulate in their pink coloration. It has, however, 

 yet to be proved that insects have any definite jJerceptions of 

 form. A white butterfly will, as Messrs. Geddes and Thomson 

 remark,t " fly naively to a piece of white paper on the ground," 

 evidently influenced by its colour alone. 



In the vegetable world bright colours appear to me to have 

 an attractive rather than a repellent function. The red berries 

 of the service tree and other shrubs are much sought after, 

 as any one can prove, by birds. The advantage of this to 

 the plant is supposed to be the dissemination of the seed ; it 

 has been proved experimentally that the seeds of many plants 

 will pass uninjured through the alimentary tract of a bird, 

 being still perfectly capable of germination. It was Mr. 

 Grant Allen, I think, who made the ingenious suggestion 

 that poisonous berries secured a twofold advantage by their 

 bright colours : the seeds are disseminated, and the death 

 of the bird provides abundant manure for the plant when it 

 pushes its way out of the seed coats. 



* " The Colour of Animals," p. 74. 

 t "The Evohition of Sex," p. 28. 



