378 . ME. H. M. BERNARD ON THE 



Phrynus, and Myyale (Lankcster), which are large flat cells, each presenting a considerable area on the 

 external surface of the duct. This difference, however, may be simply a matter of support. In Galeodes 

 the duct is not bound together by connective tissue ; hence the compacter arrangement of the cells 

 composing the wall. The nature of the striation found in the coxal glands — viz., that it is due to the 

 arrangement of the protoplasm in the oiiter layers of the cells themselves — is also seen in the long cells 

 of Galeodes, but is not so clear in the case of Scorpio. 



A close examination of these cells, and of the contents of the Inmen of the duct, 

 confirms a previoiis ohservation (12) that the nuclei absorb the excretory matter, swell 

 up, and are finally discharged as large clear vesicles. 



I thought at first that the clear matter in the inner ends of the cells might be 

 discharged or perhaps even the distal ends of the cells themselves mignt break off. 

 The latter hypothesis would accormt better than the former for the finely granular and 

 often vesicular appearance of the discharge. But the variations in the size and character 

 of the nuclei, the fact that groups of small deeply-staining nuclei are found in the act 

 of dividing (PI. XXXIV. fig. 11 a), and further that the discharge can be found at times to 

 consist of vesicles closely agreeing in size with the large vesicular nuclei, suggest that it 

 is the nu.clei which absorb the excretion and are then discharged. If this is the case, there 

 shoxxld be a small resting nucleus in each cell ; I have only rarely succeeded in finding- 

 one. Fig. 11 shows a cell with two nuclei, one of which, the larger vesicular nucleus, 

 seems to be on the point of escaping. The measurements given in the figures refer to 

 Galeodes ; in my sections of Rhax, I found the small staining nuclei averaging 3 /t, while 

 the large vesicular nuclei were 6/«, and the discharged vesicles Q^ and larger. The 

 different sizes of nuclei appear in patches ; where only small staining nuclei are found, the 

 inner ends of the cells do not contain so much clear matter. 



I have already described a similar phenomenon in Scorpio, where the discharge of the nuclei seemed quite 

 as apparent. I have since reinvestigated the subject with other specimens, and have only been confirmed 

 in my opinion. The point is, however, one of such importance that I hesitate to make any dogmatic 

 assertions ; there are, perhaps, other ways of interpreting the phenomena. The whole subject requires 

 investigation with material specially preserved. 



Eor a long time I thought that the duct in Galeodes was actually in communication 

 with the hollow sponge-work of the chitinous plate (PI. XXVIIl. fig. 16,^), as the 

 tubules of the terminal j)ortion appear to be developing out of it (the plate). But I have 

 now convinced myself that this is not the case. The chitinous plate merely forms a 

 suj)port for this and other tissues. 



Histologically, the end tubules forming the great sj)ongy mass in Galeodes (PL XXXI. 

 fig. 5) differ considerably from the main duct. They are bound together liy connective 

 tissue which contains a tangled mass of blood-lacunre and tracheae. The blood seems to 

 run largely in folds of the limiting membrane (PL XXXIV. figs. 13, 14). The tracheae 

 often appear to have thick protoplasmic walls ; at times it seems as if four or five tracheae 

 of various sizes ran through the same protoplasmic mass. There is no striation of the 

 protoplasm of the epithelial cells, which are hyaline, Avith large, deeply staining nuclei, 

 and crowded together. The nuclei here also seem to be fragmenting in groups, and 



