COMPARISON OF hXRVJE WITH UNGULATA. 67 



are but so many representations of rasorial or thysa- 

 nuriform larvse, the chief characteristic of which, as 

 every experienced entomologist well knows, is the horn- 

 like spines which crest the head, and give to them the 

 aspect of ruminant caterpillars. But this is not the 

 only mode by which Nature has clearly pointed out to 

 us the harmony and simplicity of her fundamental laws. 

 Thysanuriforra larvae are those only which have the 

 body either terminating in two long filaments resem- 

 bling tails*, or in two lieshy points, as in the caterpillars 

 of our common meadow brown butterflies, f Now, if 

 it be inquired, what are the birds which have the greatest 

 developement of tail? — the merest tyro wiU name the 

 peacocks, the pheasants, and all the typical gallinaceous 

 birds, as possessing this member in its highest state of 

 developement. The analogy of this order of birds, 

 with that composed of the domesticated quadrupeds, 

 long ago pointed out by Linnaeus, is unquestionable ; 

 and we accordingly find that the horse, which stands at 

 their head, has the most beautiful tail among quadru- 

 peds. But an arrangement, it has well been said, it 

 really natural, will stand any test : the most trivial, as 

 well as what appears to be the most important circum- 

 stances, must be taken into consideration. Now, it is 

 notorious, that the most bulky of true quadrupeds are 

 found in the class Ungulata ; there we have the ele- 

 phant, the rhinoceros, the hippopotamus, and the whole 

 family of antelopes and oxen : the largest birds, on the 

 same principle, occur in the rasorial order ; and in like 

 manner the giants of the diurnal butterflies all proceed 

 from the thysanuriform caterpUlars. To pursue these 

 beautiful and astonishing representations further, — for 

 many others might be pointed out, — is surely needless. 

 Enough has been said to establish our proposition, that 

 the types of lepidopterons larvae are only so many re- 

 presentations of the primary types among the most 



* Merian's Ins. pi. 19. fig. 1. ; StoU, pi. 3. figs. 3, 4. 

 f See Lewin's Brit. Ins. pi. 17. to 21. 



F 2 



