291 [rackard. 



Insects, but with Leuckart, Agassiz and Dana, we would prefer to 

 rank them as an order of the class Insects.* 



In a former communieation.f we proposed a classification of insects 

 into two series of Suborders, (not however agreeing with the Ilaustel- 

 lata and Mandibulata of Clairville.) of which the lower begin with the 

 Neuroptera, and by the Orthoptera and Ilemiptera culminate in the 

 Coleoptera, while the second series rank higher as a whole, beginning 

 with the Dlptera and ending with the Hymenoptera, which thus stand 

 at the head of the Articulata. The hymenoptera differ from all other 

 insects in having the basal ring of the abdomen thrown forward upon 

 the thorax ; in having the three regions of the body more distinctly 

 marked, and more equally developed than in other insects. The 

 mouth-parts are more equally developed, and at the same time more 

 differentiated in structure and function ; there are no abdominal 

 jointed appendages present in the adult form, while the external gen- 

 erative organs are more symmetrically developed, and more com- 

 pletely enclosed within the abdomen in the highest fiimilies, than in 

 any other suborder of insects. They afford the highest types of 

 articulates, being more compact, less loosely put together, and thus 

 presenting less degradational features than any of the other subor- 

 ders ; but the most valuable shujle character is the transfer of the first 

 abdominal ring forwards to the adjoining region, which involves an 

 entire remodelling of the body, throwing forwards the prime ele- 

 ments of the organism, by which it becomes more cephalized, and thus 

 the nervous power rendered more centralized than in all other articu- 

 lates. 



Selecting the Honey bee as the type, being in our view the most per- 

 fectly organized of all insects, we find the head larger and the abdo- 

 men smaller in proportion than in other insects, accompanied with the 

 most equable and compact development of the parts composing these 

 regions. The brain-ganglia are largest and most developed according 

 to the studies of entomotomists. The larvee, in their general form, 

 are more unlike the adult insects than in any other suborder of 

 insects, while the pupae most closely approximate to the imago. 

 They are short cylindrical, footless, worm-like grubs which are help- 



*The Embryology of Arachnids as worked out by Claparede, shows that the 

 larva is strikingly worm-like, distinct rings ("protozoonites") appearing before 

 the biregional arachnid form is assumed. The embryos of two genera of 

 mites, Demodex and Acarus, are at first hexapodous, as Newport has shown that 

 of Julus, a myriapod, to be. The close homologies of the Arachnids and 3Iyri- 

 apods with the Insects (llexapoda) convince us that the three groups, whether 

 we call them orders or classes, are as a whole equivalent to the Crustacea or 

 "Worms. 



t Synthetic Types of Insects. Boston Journ. Nat. Hist. VII. 18G3. How to ob- 

 serve and collect Insects. 2d Annual lieport of Maine State Survey. 18G3. 



