474 DISCOVERY REPORTS 



The fibres of the anterior valve nerve, for the main part, pass into the tritocerebral 

 commissure. A part, however, curve forwards towards the antennal region and join 

 the tract of fibres from the posterior valve nerve. 



The fibres of the adductor muscle nerve pass directly into the tritocerebral com- 

 missure, where they terminate in an ovoid mass which, since it has the same appearance 

 as the central body of the brain, I am calling the tritocerebral central body. Close 

 behind this body, but distinct from it, is another neuropilem mass, and behind the 

 latter still a third but not so distinct a mass. Since this region consists of the extremely 

 condensed ventral chain, these centres must correspond to the " sensorisches Ventral- 

 neuropilem" which Zawarzin has figured for the insect Aeschna (1924). 



GIANT FIBRE SYSTEM 



In addition to the commissural tracts and centres of co-ordination, there exists in 

 the nen'ous system of Dolor ia a complicated system of giant fibres. These are un- 

 doubtedly typical neurochord cells, firstly because of the immense diameter of their 

 fibres, and secondly because they are provided with a non-staining medullary sheath. 

 The latter makes the fibres particularly noticeable when they happen to be cut trans- 

 versely, but it also makes it easy to trace them in frontal sections. 



In Fig. 9 I have drawn the most conspicuous of these fibres but, actually, there are 

 a great many more which are not so thick. In fact, there seem to be all sizes from the 

 enormous fibre emerging from the lower antennal nerve down to the fibres of normal 

 size — no longer giant. Thus the dorsal region between the antennulary and the indistinct 

 antennal nuclei is a complete tangle of medium-sized giant fibres. 



The largest of the giant fibres are connected with the antennae. The dorsal antennal 

 nerve appears to consist of a bvmdle of five (Figs. 9, 11). These are labelled a. They 

 can be traced outwards to the antennal basal ganglion, through which they pass, to 

 the immense muscles on the median face of the limb. The individual fibres can be traced 

 into the cytoplasm which covers the outer surface of the muscles (Fig. 11). They appear 

 to end blindly, there being no special end organ. Internally they can be traced to 

 a dorsal region on a level with the anterior valve nerve, where, after giving off a few 

 branches, they end. I could not find their cell bodies, but just in front and above the 

 antennal nucleus there is a group of very large cells, and these may represent their 

 cell stations. 



The very large giant fibre b which runs in the lower antennal nerve innervates muscles 

 on the outer side of the limb, and internally terminates in the same region as the 

 fibres from the upper antennal nerve. In addition, it gives off a branch to the antennal 

 nucleus. 



Although, as I have just stated, I could not find the cell bodies of these fibres, I 

 feel certain, from a careful study of the antennal basal ganglion, that they are not to be 

 found there. They must be somewhere within the nerve ring, and their fibres must 

 represent the motor elements of the big swimming muscles. I am hence calling the 

 region where they terminate internally the antennal motor nucleus. 



