338 ERNEST WARREN. 



tion of some of the cells of the cortex is associated with the 

 degenerate character of the compound eyes. 



In fig. 50 a portion of the antennary lobe (a. I.) is shown 

 springing from the ventral surface of the optic ganglion 

 {d. c). The antennary lobe gives off a stout nerve to the 

 antenna (fig. 46, a. n.). 



The canal for the oesophagus, which passes through the 

 cephalic nervous system, is very narrow in front (fig. 53), but 

 it widens out considerably behind (fig. 55). 



The procerebral lobes are not richly provided with cortex 

 in comparison with many other insects. The great bulk of the 

 lobe consists of medulla, as is seen in fig. 53. The nerve- 

 tissue is bounded on the outside by a thin, fiat epithelium 

 (fig. 54, ej). c), consisting of cells which exhibit a consider- 

 able density, and are quite unlike the ordinary cells of the 

 cortex. The epithelium can be traced with more or less 

 clearness over the whole of the nerve-cord. A similar 

 epithelium, but not so sharply defined, was observed in both 

 Termitomimus and Termitodiscus. It would appear 

 that the epithelium is not nervous in nature, and very 

 possibly it is to be regarded as a layer of less differentiated 

 ectoderm; but for any certainty in the matter a study of the 

 development is necessary. The cortex is exceedingly thin and 

 sparse on the lobes (fig. 53) until we reach quite close to the 

 posterior vertical periphery (i.e. at a plane just, in front of the 

 tritocerebrum, tri., fig. 46) ; here the cortex thickens to some 

 extent, and we find a weakly developed mushroom-body consist- 

 ing of a small calyx (fig. 54, cy.) with a stalk or cauHculus (ca.). 

 On the calj^xrestsa small ti*abecula (trh.). The nuclei of the 

 trabecula are smaller than those of the rest of the cortex, and 

 no cell-outlines are discernible. The tissue of the trabecula 

 stains more deeply than the rest of the cortex. At fig. 55 

 the section passes very close to the vertical face and the 

 calyx {cy.) aiid calyx-cells {cy. c.) are disappearing. The rela- 

 tively small amount of cortex which is present is clearly shown 

 in the vertical, longitudinal section given in PI. XVJI, fig. 43. 



There occui-s a brido^e of transverse fibres connecting- the 



