MR. NEWPORT ON THE NERVOUS SYSTEM OF THE SPHINX LIGUSTRI. 403 



converge, as in the Sphinx ; and, in passing along the cords, gather a few filaments 

 from the motor tract of each, and after uniting in the middle line pass backwards to 

 the next ganglion, to be distributed as before. In this species it is curious that the 

 superadded nerves do not seem to unite with the great moto-sensitive nerve, but only 

 with the small nerve behind it, which is given to muscles that, acting diagonally, 

 seem to be much concerned in the function of respiration. But there are facts 

 which might at first incline us to believe that these transverse nerves constitute the 

 visceral or true sympathetic system in insects. Thus, their union with most of the 

 nerves of the body ; their connexion with the anterior lateral ganglia ; the manner 

 in which they receive additional filaments from the cords ; and the existence of a 

 ganglion upon the terminal filament in the Gr-yllus vividissimus, Linn. [Plate XVI. 

 fig. 39. (c, b),] ; and, above all, the existence of clearly defined ganglia at each distri- 

 bution in the Carahi [fig. 38. (c),], or ground Beetles, and in the Mole Cricket, Gryl- 

 lotalpa, in which the ganglia are very distinct, and situated above the great ganglions 

 of the cords. On the other hand, it is only in a few genera of insects that these gan- 

 glia exist ; and it has not yet been proved that respiratory nerves must necessarily be 

 without ganglia. Indeed, it is not improbable that we may hereafter find a much 

 closer connexion between the respiratory and sympathetic systems in the higher ani- 

 mals than has hitherto been imagined. Now the existence of ganglia upon these 

 nerves, although in but a few genera, seems very decidedly to prove that they are 

 not analogous to the simple motor nerves of the body, while their distribution being 

 almost entirely to muscle, and but sparingly to the viscera, seems as clearly to show 

 that they are not analogous to the sympathetic or visceral nerves of F^ertehrata. 



To prove more directly that these nerves are not simply those of motion, but are 

 for the involuntary function of respiration, we must examine the means by which respi- 

 ration in insects is performed. Nine pairs of spiracles, or breathing orifices, are 

 placed in the larva along the sides of the body. Eight of these are in regular succes- 

 sion. They all communicate with longitudinal tracheae, from which several large 

 ramifying branches pass off transversely, nearly opposite to each spiracle. The lon- 

 gitudinal tracheae extend from one end of the body to the other, and communicate 

 freely with the spiracles. A similar arrangement of the tracheae and spiracles exists 

 in the perfect insect, but with this difference, — in the anterior part of the abdominal 

 region, those parts which in the larva are ramifying tracheal tubes, are now altered in 

 structure, some of them being developed into pulmonary sacs or bags, while in the 

 thorax the tracheae themselves are larger than those in the abdomen, and the spiracles 

 are larger, and of a different form. In the thorax, and first pair in the abdomen, they 

 are either elongated, semioval, or straight. The remainder of the abdominal ones are 

 oval. The spiracles are acted upon by two sets of muscles, the one diagonal, and 

 the other oval. The latter act the part of sphincters, the former are connected with 

 the muscles of the segment. Reaumur, Bonnet, and others have clearly proved that 

 the anterior pairs of spiracles in the larva (those of the collar and thorax in the per- 



