106 
SYSTEM OF NATURE. 
bulate; and nothing could coincide more completely with 
the proposed new arrangement. Thirdly, the metamor- 
photic system of Swammerdam (p. 93), Ray, Willughby and 
Lamarck, (reduced to somewhat more precision by the 
observations at pages 102 and 103), shows that the Lepi- 
doptera, Diptera, and a portion of the Neuroptera, are 
amorphous ; the Hymenoptera, Coleoptera, and a portion 
of the Neuroptera, necromorphous; and finally, Orthop- 
tera, Hemiptera, and a portion of the Neuroptera, isomor- 
phous ; so that, in this instance also, the correspondence 
with the proposed arrangement is perfect. 
In my observations on the classes of Vertebrata, I have 
called attention to the parallels pointed out by Linneus as 
existing between tribes of distinct classes, and after sug¬ 
gesting for consideration parallels between placentals and 
marsupials, and between placentals and birds, I have ven¬ 
tured to express an opinion that groups which bear this 
test are natural groups. Now supposing the Pterota and 
Aptera to be provinces equalling the vertebrates in im¬ 
portance, I think it is probable they will submit to similar 
subdivision; but I must freely confess that I have hitherto 
failed in discovering any parallels but such as appear ra¬ 
ther too hypothetical to be of any real value, still I find no 
reason to retract or even modify what I have said on this 
subject: the evident similarity between the reptiles and 
myriapods indicates the existence of a parallel in this 
instance at least, and the similarity between fishes and 
Crustacea suggests the idea of a second parallel; but it is 
most undesirable to follow out the subject, seeing that in 
the end we must have recourse to conjecture. 
I think it will be admitted by those who have given the 
subject the consideration it deserves, that the Pterota and 
