430 



Mr. VonMonsfurtJier experiments upon 



Very pale blue tissue-paper. 

 Experiment 82 2 



Light blue. 



Experiment 83 



2 1 



= 3 



Deep blue. 



Experiment 84 1 2 



= 3 



Darkest blue. 



Experiment 85 



1 3 



Blue gelatine. 

 Experiment 40 3 



= 8 



The blue backgrounds absorb- 

 ing more and more of all rays 

 except the blue tend to produce 

 dark pupiv, even the faint 

 blue shade of the tissue-paper 

 being accompanied by some 

 slight effect (compare with 

 white paper). The compara- 

 tively slight effect of the darkest 

 blue is almost certainly due to 

 the larvfE having been intro- 

 duced too late (see account of 

 experiment, p. 410). The effect 

 of a blue screen placed in 

 front of a surface of light wood 

 is not very different from that 

 of the blue background. 



It is necessary to say a few words about this com- 

 parison of the effects of different parts of the spectrum, 

 and the frequent antithesis between the results of 

 screens and backgrounds of the same colour. 



In the conclusions at the end of this memoir the 

 colours of both will be given with greater precision, and 

 their effects on all larvae and pupae subjected to them 

 will be compared. 



The results of the coloured backgrounds in strong 

 light are perfectly regular : it is clear that the rays 

 which check the formation of dark sui^erficial pigment, 

 and so allow the underlying green derived pigments to 

 be seen, lie in the orange and yellow. The other parts 

 of the spectrum do not seem to interfere with this power 

 except by diluting the effective part of the reflected light. 

 Thus red alone produced dark pupje, but red with orange 

 and yellow produced green ones : and approaching from 

 the opposite side of the spectrum we see the same thing ; 

 for blue alone, green and blue, and green alone, pro- 

 duced dark pupae, while green, yellow, orange, and red, 

 produced green ones. Similarly white light reflecting 

 all colours produced green pupae. 



These results are perfectly uniform and consistent : 

 they are precisely similar to the behaviour of the PieridcB 

 (Phil. Trans., 1887, B., pp. 427-432), when exposed to 



