NO. 8 MORPHOLOGY OF INSECT SENSE ORGANS SNODGRASS I 5 



the embryo. The number and arrangement of the gangha differ 

 much in dift'erent insects, but a typical scheme of the gangha and 

 the principal nerves of this system is shown diagrammatically in 

 figure 6. The most anterior ganglion is the median frontal gan- 

 glion (FrGng), situated on the dorsal wall of the pharynx before 

 the base of the brain. It is united laterally by the frontal coui- 

 missures (FrCom) with the tritocerebral regions of the brain, as 

 already described (fig. 4). Anteriorly it gives oft' a small median 

 frontal nerve (FrNv) with branches to the pharynx and the cly- 

 peal region of the wall of the head. Posteriorly a median recurrent 

 nerve (RNv) goes back from the ganglion beneath the brain to 

 a second median ganglion on the pharynx below or behind the 

 brain. This ganglion is known as the asophagcal ganglion or hypo- 

 cephalic ganglion (CEGng). From it is given off on each side a 

 short nerve to a lateral ganglion (LGng) located on the dorso- 

 lateral part of the pharynx or oesophagus (the " oesophageal gan- 

 glion " of Berlese). Each of these lateral ganglia is connected with 

 the posterior part of the brain by a small nerve (a). The median 

 oesophageal ganglion may also give off posteriorly a long median 

 nerve which ends in a gastric ganglion (GGng) on the wall of the 

 stomach. Besides the main nerve trunks, many smaller nerves arise 

 from all the ganglia of the stomatogastric system, which are mostly 

 distributed on the stomodeal and ventricular parts of the alimentary 

 canal, but some go to the walls of the head, the surface of the 

 brain, to the aorta, and to other neighboring organs. 



Janet (1899) has proposed the interesting theory that the three 

 median ganglia of the stomatogastric chain are the primitive ven- 

 tral ganglia of three pre-cerebral segments that have been invagi- 

 nated to form the stomodeum. This* theory, however, must assume 

 first that, prior to the stomodeal invagination, the first six pairs of 

 ventral ganglia migrated upward in the sides of the body and then 

 came together dorsally, the two ganglia of each segmental pair 

 uniting above the alimentary canal. Then, with the retraction of 

 the supposed stomodeal segments, the three pre-cerebral ganglia 

 were drawn inward with their segments, the first one becoming 

 the gastric ganglion, the third becoming the frontal ganglion. Next, 

 the theory must assume that the posterior roots of the frontal- 

 protocerebral connectives became shifted from the protocerebrum 

 to the tritocerebrum ; while, finally it assumes that the transverse 

 commissures of the stomatogastric ganglia have been lost, and that 

 those of the brain ganglia are all united in the suboesophageal com- 

 missure. None of these assumptions is supported by the known 



