244 ANIMAL COLOEATION. 



The study of the development of animals always shows, 

 with more or less clearness, an epitome of the history of 

 the development of the race : every animal, as it has been 

 put, "is compelled to climb up its own genealogical tree." 

 The facts that have been mentioned about the CHearwings can 

 hardly be explained, except on the hypothesis that they have 

 taken their origin from moths with fully scaled wings ; and the 

 farther inference is probable that the Bee Glearwings have 

 more recently lost their scales than some of the others. This 

 probability is perhaps increased by the resemblance which the 

 larva of 8. hombyliformis bears to a Sjihinx larva, while the 

 larva of 8. tipuliformis is different ; but, after all, we do not 

 know positively whether the internal feediog-larva might not 

 be the older. 



The advocates for explaiuiug the facts of mimicry by a need 

 for protection experienced by the mimicking insects could 

 point to Mr. Poulton's experiment with ,S'. hombyliformis and 

 S. hembeciforniis as being more conclusive than they at first 

 appear. The fact that S. bombyliform.is was eaten without 

 hesitation, and that S. bembeciforntis created a suspicion in the 

 lizard's mind, seem to indicate a considerable imperfection in 

 the resemblance. In the detailed record of his experiments,* 

 Mr. Poulton suggested that the resemblance to a bee or other 

 stinging Hymenopteron shown by 8. fuciformis is possibly to 

 be regarded as " a remnant of a former more perfect mimicry, 

 reliance being now jjlaced on powerful flight and concealment 

 during rest." He admits that the instance is a difficulty in 

 the way of the theory. However, remembering the facts of de- 

 velopment, this particular difficulty almost disappears. Seeing 

 that the loss of the scales in the Bee Hawk moth is more 

 recent than the loss of scales in the Hornet Clearwing, it is 

 * Prnc. Zool. Soc, 1887. 



