102 Mr. E. B. Poulton. On a special Colour-Relation [Feb. 10, 



produced in a tube with the gilt chamber above proved to be the 

 only (5) obtained in the whole of this set of experiments, in which 

 83 pupee had been compared. Such results show that the sensitive 

 surface is not represented by a sense-organ in the head, or with an 

 anterior portion only, but that the whole skin area possesses suscepti- 

 bility. 



The second method of conducting conflicting colour experiments 

 was superior in the more equal illumination of the gilt surface when 

 above or below. Flat wooden trays were covered in each case with 

 black and gilt paper in alternating areas, the two colours meeting 

 along lines which ran across each tray, and along which shelves were 

 fixed covered with gilt paper towards the gilt surface, and black paper 

 towards the black surface. The shelves were perforated close to the 

 tray bottom with holes separated by equal distances, and of such a 

 size as to easily admit the body of a larva with its spines, while the 

 latter as in the compartmented tubes tend to obscure the interval 

 between the larval body and the edge of the aperture. The trays 

 were placed vertically in a strong east light, so that the shelves pro- 

 jected horizontally, the black surface being uppermost in some cases, 

 the gilt surface in others. Suspended larvse were pinned (by the 

 boss of silk) on to the uppermost colour in such a position over the 

 holes that the head and first five segments of each larva passed 

 through a hole into the colour beneath, which tended to produce oppo- 

 site results. The curvature of the larval body brought the head close 

 up to the underside of the shelf, and thus there was no chance of its 

 being influenced by the colour above the shelf. Other larvaa were 

 similarly fixed between the shelves upon one colour only, so as to 

 afford a comparison with the results of the conflicting colours. The 

 pupee obtained were on the whole rather lighter when the gilt surface 

 was above, and hence the gilt surroundings influenced the rather 

 larger posterior part of the skin to a greater extent than in the converse 

 arrangement, when the effective colour was below. Hence on the 

 whole the influence of conflicting colours has ended in as complete a 

 confirmation of the numerous blinding experiments as the necessarily 

 limited conditions of experiment could be expected to produce. 



8. The Nature of the Effects Produced. — The gilded appearance is 

 one of the most metal-like appearances in any non-metallic substance. 

 The optical explanation has never been understood. It has, however, 

 been long known that it depends upon the cuticle, and needs the 

 presence of moisture, and that it can be renewed, when the dry 

 cuticle is moistened. Hence it can be preserved for any time in 

 spirit. If a piece of dry cuticle be moistened on its upper surface 

 the colour is not renewed, but almost instantly follows the application 

 of spirit to the lower surface. Sections of the cuticle resemble those 

 of Papilio machaon described in a previous paper (' Roy. Soc. Proc.,' 



