368 NATURAL ARRANGEMENT OP INSKCTS. 



however, they have never less than six joints, and 

 usually more than twelve. The eyes of the Dipteru 

 exliibit some remarkable peculiarities, and frequently 

 sexual discrepancies : these organs, in the male, often 

 occupying nearly the whole of the head ; and they are 

 often densely clothed with hair, — a structure that would 

 almost seem to impede vision ; and another singular 

 circumstance is, that, although the hexagonal structure 

 of the facets of which the composite eyes are composed 

 is universal, yet the several families, genera, s])ecies, 

 and sexes exhibit peculiar arrangements of different 

 sizes of these facets in the same eyes ; a fact, which, 

 although exhibited by Hook in his enlarged eye of 

 Tabanus, was left unnoticed by himself, and never, 

 subsequently, observed or heeded by systematists or 

 anatomists, although it displays such striking pecu- 

 liarities. 



(328.) The general integument of these insects is of 

 as fragile a texture as in many of the Neuroptera, and 

 the rest of their structure is perfectly analogous to the 

 parts of other insects. All these structural particulars, 

 however, undergo an infinity of modifications which are 

 nowhere stationary; and the forms of their bodies are 

 as variously diflferent. These insects are usually ovi- 

 parous, but they present two remarkable exceptions ; 

 for in Sarcophaga they are viviparous, or produce their 

 larvae hatched ; and in the terminal division, the Pupi- 

 para, they are pupiparous, and bring forth their young 

 advanced to the stage of pupae, — and thus form a re- 

 mote analogy to the marsupial animals. Under the 

 several famihes, we shal^. observe other peculiarities 

 incidental to the prelimmary stages of their existence. 

 The transformation of the majority into the pupa state 

 is also very peculiar ; for, instead of spinning a co- 

 coon, or forming a cell, the skin of the larva hardens, 

 and thus makes a case, within which the changes 

 take place. In their perfect state, they are possibly 

 not inferior to any order in the multiplicity of forms 

 which tliey exhibit ; whereas, in the abundance of 



