SCIENTIFIC NOTES. 11 



only at the active period of making the cocoon, or the time immediately 

 preceding that period, and that, when the cocoon was begun, its colour 

 was fixed. 



I now took a number of comparatively immature larvae which I knew 

 were not full-fed, although they appeared to be nearly so, and practically 

 starved them into making cocoons. With scarcely an exception the 

 cocoons were pale from whichever of the three bags they were 

 taken. 



It now became clear that the colouring of the cocoon was a 

 physiological process, and within certain limits, the cocoon was coloured 

 independently, and probably not at the exact time each strand of silk 

 was woven into the cocoon, and it was equally clear that the larvae had 

 practically no power whatever to form dark or normal cocoons, if 

 starved for some three to six days before spinning. Therefore the 

 colouring matter was directly connected with nutrition. But how? 



I now took a tin and placed larvae in it, allowing the food to get moist 

 and even to begin to decay, owing to the moisture collecting in the tin. 

 Most of the larvae died, filled with a black fluid, the few who passed 

 through this process, voided an almost perfectly black excrement and 

 spun an almost black cocoon. 



What then was the connection between the colour of the cocoon and 

 the colour of the excrement? They were identical without doubt, and 

 it suddenly dawned on me that it might be that the colouring matter of 

 these cocoons was an intestinal or urinary waste product, and the more I 

 considered the matter, the more certain I became. If I starved my 

 larvcC so that there was no excretory waste, I got white or pale cocoons 

 under almost any condition of surroundings and environment. If they 

 were well fed up to the moment of pupation, and healthy, the cocoons 

 were fairly normally coloured, whilst if there was a tendency induced by 

 external circumstances (especially moisture) to promote excessive 

 secretion, the cocoon became exceptionally dark.^ 



I had now arrived at any rate at a satisfactory explanation of the cause 

 of the coloration, but I was not altogether satisfied how far the larvae 

 could, in a state of nature, respond to their environment, but I should 

 say it was to a moderately considerable extent, and I found that larvre 

 kept healthily, in a roomy glass breeding cage, spun cocoons, which were 

 modified considerably in colour, according as they spun up on white 

 paper, open chip boxes, dead leaves etc. in the cage, but this modifica- 

 tion did not include a complete response to environment, it was only 

 an approach to it in a most general way. 



Yet another ex[)eriment. I opened a dozen dark -coloured cocoons, 

 after the larvae had been at work in them for some 12 hours and 

 were quite invisible, and took out the larvae. These I put into a dark- 

 coloured box, and each and all spun a second cocoon, thin and fragile, 

 and with one exception, white, pure white, showing that the dark 

 colouring matter had been absolutely expended in forming the first 

 cocoon, with the one exception mentioned, and that the larva, although 

 still capable of spinning the white silk, had no further power or capacity 



' I found afterwards that the food in my dark bag (No. i) had a much greater 

 tendency to decay than in the others (probably because I had put it away wet), and I 

 surmise now, that this had something to do with the fact that the first lot of cocoons 

 formed on the sides of this l)ag were so much darker tiian the other sets. 



