LEWIS, — POLYCH^TE ANNELIDS. 243 



the central corpuscle stains much more intensely than the surrounding 

 zone. Occasionally it is slightly irregular in shape, and, as before stated, 

 sometimes gives the appearance of a single body (Plate 1, Fig. 2), but 

 more often there seemed to be present several — two, three, or four — 

 exceedingly small corpuscles (Plate 4, Fig. 24). 



Vacuoles resembling those described by Rohde and other writers some- 

 times occurred in the protoplasm close to the sheath surrounding the cell 

 (Plate 1, Fig. 3 ; Plate 4, Fig. 25 ; Plate 5, Fig. 29). 



It is not difficult to trace a single giant cell through all the sections 

 in which it is embraced, because this peculiar type of cell occurs so rarely 

 in the nervous system that two such cells would seldom, perhaps never, 

 occur so close together as to permit their being mistaken for each other. 



d. Centrosome. 

 (1) Historical Review. 



It is only recently that centrosomes were first recognized in nerve cells, 

 and for that reason, if for no other, they have been mentioned by only 

 a very few writers. So late as 1894, when Heidenhain wrote his paper, 

 " Neue Untersuchungen liber die Centralkorper und ihre Beziehungen 

 zuni Kern- und Zellenprotoplasma," such a structure was not known to 

 exist in nerve cells. Heidenhain ('94, p. 656) merely suggests the pos- 

 sibility of its presence, for he closes his chapter regarding the universality 

 of the centrosome with these words : " Kiimen wir so zu positiven Re- 

 sultaten, so wiirde es am Ende der van Beneden-Boverischen Theorie 

 von der Ubiquitiit der Centralkorper am Ende nur einen geriugen Ein- 

 trag thun, weun sich schliesslich herausstellen sollte, dass es einige wenige 

 Zellenformen giebt, welche, da sie im erwachseuen Thierkorper nicht 

 meiir theilungsfahig sind, die Centralkorper vollig einblissen. Es wiirde 

 sich, soweit ich das zu beurtheilen vermag, wohl wesentlich nur um 

 die Ganglienzellen handeln. Allein auch hier ist fiir eine Special- 

 untersuchung Aussicht auf Erfolg vorhanden, da doch von einigen Gan- 

 glienzellenformen her sowohl radiiire als auch concentrische Anordnungen 

 bekannt sind." 



Von Lenhossek ('95*) was the first author positively to demonstrate the 

 presence of a centrosome in ganglion cells. He describes very fully for 

 certain of the spinal ganglion cells of the frog a well defined centrosome 

 and sphere, and gives a detailed account of all the parts of the structure. 

 As von Lenhossek was the first author to discover and describe this 

 structure in nerve cells, I have preferred to make use of his terms " cen- 

 trosome " and " sphere," although in some respects the conditions in the 



