tlie colours of certain Lepidoptera. 



303 



Experiment I. 

 Dark Surroundings. 



June 30. — The difference between 

 the large larvre of I. & II. continued 

 to be very marked. The small larva 

 was not seen, and was apparently 

 lost. 



July 11. — Between this and the 

 last comparison the difference be- 

 tween the 2 large larvre had greatly 

 increased, the dark one being almost 

 black. The latter was apparently 

 mature, being larger than that in 

 II. They were therefore painted 

 (July 11), and afterwards preserved 

 (July 13). A few days earlier they 

 were photographed. 



Experiment II. 

 Green Surroundins 



July 11. — The large larva had 

 been very light brown for some 

 weeks. The small larva was equally 

 light. Between this date and June 

 30 the two large larv;v_' were seen by 

 many physiologists and others (Dr. 

 Ijurdon Sanderson, Sir William 

 Turner, Prof. C. Stewart, Prof. 

 Gotch, Dr. Page, and Dr. Bradford). 

 Everyone was much impressed with 

 the extraordinary difference between 

 them. 



The difference between these larvpe is indicated in an 

 uncoloured iUustration to ' Colours of Animals ' (p. 151). 

 The larvte were photographed for me by my friends 

 Mr. F. J. Smith and Mr. G. J. Burch. In both cases 

 they were induced to rest upon white paper spills, and 

 were arranged so that the light fell on the same part of 

 both from the same direction. The photographs did not 

 show nearly so marked a difference as was seen in the 

 larvae themselves. A collotype reproduction from one of 

 Mr. Burch's negatives is shown on Plate XV., figs. 1 (the 

 dark) and 2 (the light larva). Apart from the colour 

 difference the representation of the larvse is extremely 

 good. 



The number of individuals experimented upon in the 

 case of the Catocalidce, and especially in C. elocata, would 

 have been utterly inadequate for the purpose of proving 

 the existence of colour susceptibility. But this principle 

 having been firmly based on the results of experiments 

 with other genera in which very large numbers of indi- 

 viduals have been employed, the evidence now obtained 

 is sufficient to show that the principle applies to the 

 Catocalidcs. The mere extension of a principle to fresh 

 cases of the same class does not of course require any- 

 thing like the same amount of evidence as that which 

 was necessary in the first place to establish the principle 

 itself. 



