of Bupiestidse a>wf Elaterldae, 181 



vies, and beds for the pupa excavated in the solid timber ? At 

 first sight an attack seems hopeless on so strong a scientific 

 fortress ; but a closer inspection will, I trust, discover more than 

 one weak point. 



The structure of the larvae of Buprestidse is easily understood, 

 for in every point it exhibits with great consistency the peculiar 

 characteristics of an animal burrowing in wood. Such are the 

 large head deeply seated in the prothorax, the short thick man- 

 dibles formed like a hollow chisel, the large protruding labrum, 

 the short, freely developed lower organs of the mouth, the 

 powerful muscles of the neck, the broad prothorax beset with 

 grains of chitine, the slender body adapted for creeping through 

 galleries, the large anal segment, which is turned straight back- 

 wards. If now the habits of the larva of Melasis are really the 

 same as those of the larva of Buprestidse, its structure must be 

 identical with theirs at least in these principal points. But if 

 the similarity only applies to the external structure, it is clearly 

 insufficient, however striking, to prove a true affinity between 

 these creatures, or even a similar way of feeding; it merely justi- 

 fies us in supposing that they live in similar localities and use a 

 similar mode of locomotion. One would think that such a con- 

 clusion, of which the truth is evident to common sense at every 

 step in the study of living creatures, could not but be sufficiently 

 appreciated and generally adopted. Nevertheless, as I have often 

 urged on other occasions, no principle is oftener sinned against. 

 No source of errors in natural history flows more copiously than 

 that which rises from the confusion of relationship with simi- 

 larity, of affinity with analogy, of typical characteristics with 

 biological modification ; so that it is but too clear that we have 

 as yet advanced but very little towards the great aim of compre- 

 hending the rational consistency of nature. It is this ancient, 

 ever recurring mistake which also in this case has obscured the 

 truth. 



For in spite of the positive and, decisive appearance of the 

 investigations just alluded to, they nevertheless embody a simple 

 impossibility. Instead of a rasp-like chitinous armour on the 

 prothorax, the larva of Melasis has merely a couple of narrow, 

 partly transversely grooved bands of chitine, and its skin is 

 generally thin and weak ; the lower organs of the mouth are 

 rudimentary, and coalesce into a small plate; the labrum is 

 absent; the mandibles are pointed and bent outwards, they 

 have a tooth on their back, and their inner margin presents 

 merely a narrow unarmed edge ; the head is free, rather soft, 

 with only a few firmer bands of chitine forming a frame- 

 work round the epistoma and hypostoma; the anal or tenth 

 abdominal segment, which in the larva of Buprestidse forms a 

 direct continuation of the body, appears here merely as a small 



