Larva of Privet Haivh Moth. 83 



posed and prepared, twelve main g-anglionic masses are readily 

 detectible with the naked eye. The first of these is placed above 

 the oesophagus, is divisible into two lobes by a shallow antero- 

 posterior depression, and from its relation to the eyes and antennae^ 

 may be called the 'cerebroid' ganglion. The first sub-oesophageal 

 ganglion, by the commissural junction of which to the cerebroid, 

 the nerve collar is formed, is in closer proximity to the second than 

 this is to the third, or the third to the fourth, or the fourth to the 

 fifth. These five ganglia resemble each other in being more or less 

 heart-shaped, the nerves they give off being directed forwards; 

 the first of them supplies the organs of the mouth, and the other 

 four those of the thorax. The first of these ganglia is represented 

 in the developing Crustacean Astacus fluviatilis by three pairs of 

 ganglia corresponding severally to the mandibles, the anterior, and 

 the posterior maxillae, but, so far as is known, it is connate in the 

 developing insect from the earliest periods. The second, third^ 

 fourth and fifth pairs of ganglia are distinct in development in 

 both these animals ; they are fused into a single ganglionic mass 

 in the adult forms of both ; the single mass thus formed may 

 retain more or less distinct indications of its originally composite 

 character, but in the insect it does not, as in the Crustacean spe- 

 cified (see Prep. 34), fuse with the similarly fused mass supplying 

 the three true jaws. Thus the distinctness of the head of the 

 insect from its thorax is preserved and reproduced in its nervous 

 system, whilst the fusion of head and thorax in the Crustacean is 

 similarly reproduced also. The fact so far as relates to the class 

 Insecta is sometimes expressed by speaking of their brain as con- 

 sisting of a sub-oesophageal as well as of a supra-oesophageal 

 portion, or of a ' cerebellum' as well as a ' cerebrum,' the former of 

 which is in connexion with a chain of ventral ganglia. As however 

 the serial homology of the various ventrally-placed appendages of 

 the articulate Neuropods is universally recognised, this nomen- 

 clature is less to be recommended than one which by numbering 

 the segments of the nerve cord as is done here, enables us at once 

 to see where correspondence has or has nat existed, or been re- 

 tained between the external and the internal organs. Though the 

 line of division between the head and thorax of the insect, both in 

 its larval and its perfect state, has thus a diastema in the ganglionic 

 chain corresponding to it internally, which is not preserved in the 



G 2 



