1 1 Descriptiois of Preparations. 



first post-oral gang-lionic mass consists of the manducatory gang- 

 lion fused with one thoracic ganglion; whilst in the Dccapoda 

 all three thoracic centres have coalesced with the manducatory. 

 In insects, on the other hand, we should say that even when the 

 nerve ganglia attain the very extreme of concentration, the man- 

 ducatory ganglion is always distinct from the rest of the ventral 

 ganglia ; and that the thoracic ganglia, as might be expected from 

 the importance of the parts they supply, have a much greater 

 relative size and independence than the homologous ganglia in 

 Crustacea. The first abdominal ganglion in insects very frequently 

 coalesces with the third thoracic, which is rarely the case in the 

 Crustacea, in which the nerve centres of what Fritz Mliller has 

 called the ^middle body,' very ordinarily are the best developed 

 in their entire ganglionic series. The Arachnida resemble, on 

 the one side, Crustacea, in having their manducatory fused with 

 their thoracic ganglia, and on the other, Insecta, in having more 

 or fewer of the abdominal ganglia coalesced with the thoracic. 

 The Arthrogastrous Arachnida, as the scorpion, may have as many 

 as four post-abdominal ganglia ; Insecta appear never, even in the 

 larval state, to have more than three ; whilst Crustacea may have 

 as many as six ; or may, in the air-breathing Isopoda, as Oniscus 

 Murariiis, have a single ganglionic mass partially coalesced with 

 the last abdominal to represent the entire caudal series, which, 

 together with the post-abdominal segments, may be wholly lost 

 in Laemodipoda. 



On the other hand, though fusion and connation even may be carried 

 out to the extreme degree witnessed in Diptera, as also in Carcinus 

 Maenas amongst Crustacea, and the Araueae amongst Arachnida, it 

 would be wrong to hold that the nerve system of Arthropoda is in no 

 case a guide towards the determination of the homologies of their seg- 

 ments and appendages, or that the necessity existing in particular in- 

 stances for the establishment of consentaneous muscular action in seg- 

 ments and appendages, entirely obliterates all traces of a common typical 

 arrangement of the elements of this system. The innervation of the 

 so-called 'mandibles' or ' cheliceres' in the Arachnida from the supra- 

 oesophageal mass, must be held to prove that these organs are essentially 

 antennae ; and a study of the appended Tables will show, that if we make 

 allowance for the (possible) connation of the three ganglia, which should 

 typically be developed in correspondence with the three manducatory 



