330 BRITISH LEPIDOPTERA. 



captivity, even when they were left with their food and disturbed 

 as little as possible ; further that the colour of the cocoons did 

 not appear to depend upon foreign substances. He suggested 

 that a meconial fluid, viscous • and dark brown in colour and 

 containing some faecal matter, voided by the larva, was the material 

 with which the cocoon was coloured. During 1892, further 

 experiments were undertaken and the results detailed {lac. cit., pp. 

 205 et seq.) at considerable length. He had two batches of eggs 

 that gave about 140 larvae of S. pavonia, these were sleeved on 

 hawthorn in the open air, and on July 2nd divided into two 

 sections : 



(1) 66 larvse, covered with a large sleeve of white muslin, containing much 

 crumpled white paper; these gave — 7 cocoons on the white sleeve, iS in white 

 paper, 19 partially attached to white paper and partly to twigs, 9 on leaves or 

 twigs. With one exception, all the cocoons are of the full dark colour, the excep- 

 tion being also a brown cocoon, which is thin and deficient in substance (it is one of 

 the 19). (2) 44 larvse, covered with a large sleeve of black muslin, placed on same 

 bush, containing much crumpled brown paper, darkest obtainable; these gave — 2 

 cocoons in brown paper, 1 between paper and leaves, 4 on black sleeve, 31 in leaves 

 or massed against each other. All these were of full dark colour. 



These experiments were held to show that there was no 

 relation between the colour of the cocoons and that of the substances 

 to which they are attached. Further evidence is given to show 

 that the colouring substance came from the alimentary canal, and 

 after weighing the evidence Bateson concludes that : 



(1) The brown colour of the cocoons is derived from the alimentary canal. (2) 

 It is produced in the digestion of the food. (3) It is probably a chlorophyll 

 derivative. (4) It is imparted to the silk from the mouth of the larva, and perhaps 

 ■ by evacuation from the intestine also. 



May records (Ent. Record, vii., p. 238) an experiment which 

 goes to show that after pale cocoons are spun in confinement, 

 exposure to damp surroundings tends to darken them. He states 

 that from dark cocoons received from Yorkshire, small imagines 

 were bred, that, from a pairing of these, eggs were obtained and 

 the larvae divided into two lots: (1) Fed on abundant supply of 

 fresh whitethorn, in roomy cage, dry, light and well ventilated ; 

 19 spun up producing 18 more or less pale-coloured and only one 

 dark cocoon. (2) Fed on whitethorn, often short of food, under a 

 glass shade without ventilation, in a continuously damp atmosphere ; 

 16 cocoons resulted, every one dark. On November 24th three of the 

 pale cocoons were placed on damp sand in the forcing cage, kept in a 

 kitchen, and three days later the cocoons were noticed to be dark 

 brown ; wishing to make sure that this change was due to moisture 

 three more were placed with them, and on taking them out two 

 hours later the change had already taken place. May suggests 

 that the colour of the cement that is used in the construction of 

 the cocoon and that makes it waterproof, is perhaps influenced 

 by dampness and gives the cocoons their colour. He notes, however, 

 that the cocoons spun by his larvae were thin in texture, and not 

 so hard as those he had had before. Reference to our account of 

 the variation of the cocoons of LacJincis lanestris {aiitco), vol. ii., p. 

 512) will give our view of the cause of the variation in colour. 



Comparison ok cocoons of Saturnia pyri and S. pavonia. — 

 The cocoon of .V. pyri (about z\ inches long) is of a deep brown 

 colour, very similar in shape and construction to that of Saturnia 



