1887.] between certain Pupa? and their Surroundings. 101 



bristles were shorn from several mature larvae, and they were placed 

 under exactly the same conditions of light and surroundings (white or 

 gilt being used in three different experiments) as about an equal 

 number of normal larvae, but the pupae of the two sets were almost 

 exactly similar. 



(7.) The whole Skin Surface as Tested by Conflicting Colour Experi- 

 ments. — It has been shown in paragraph 6 that the larvae are to 

 some extent sensitive during Stage (iii), and I had long thought that 

 this stage in which the larvae are suspended motionless, and cannot 

 be greatly affected by disturbance, might be investigated by the appli- 

 cation of strongly conflicting colours to different parts of the larval 

 surface. Black and gilt were obviously shown to be the best colours 

 to select for the purpose, and the experiments were conducted in two 

 ways. In the first the larvae were induced to suspend themselves from 

 sheets of clear glass, by placing them in wide shallow glass boxes, so 

 that the ascent to the glass roof was easily accomplished. As soon 

 as suspension had taken place each of the larvae was covered with a 

 compartmented cardboard tube, of which the septum was perforated 

 by a hole just large enough to admit the larval body. The tube was 

 fixed to the glass sheet with glue, and the upper chamber and upper 

 surface of the septum were lined with one colour, e.g., gold, 

 while the lower chamber and lower surface of the septum were 

 lined with the opposite colour, e.g., black, which also covered 

 the outside of the cylinder, in case the larva should stretch beyond 

 its lower edge. The septum was placed at such a height in 

 the tube that the larval head and rather less than half of the 

 total skin surface (anterior) were contained in the lower chamber, 

 while rather more than half of the skin surface (posterior) was con- 

 tained in the upper chamber. It was thought that the upper chamber 

 would be illuminated too strongly as compared with the lower, 

 because its opening was directed upwards towards the light descending 

 from the window, and therefore compensation was provided by fixing 

 another perforated septum on the upper end of the cylinder, so that 

 its opening was reduced to the same diameter as the perforation in 

 the septum between the two chambers. The results show that I 

 overcompensated for the difference in illumination, for I did not take 

 into account the fact that the larva spins its boss on a comparatively 

 wide layer of silk which it has previously spun over the glass, and 

 which greatly diminishes the transparency of the latter over an area 

 at least equal to the diameter of the tube, which is comparatively 

 thick, and includes the boss itself over the smaller area, corresponding 

 to the perforation in the disk. Hence the resulting pupae were rather 

 lighter when the gilt chamber was below, although the difference was 

 not great. At the close of the experiment I altered the conditions of 

 illumination by removing the upper septum, and then the single pupa 



