No. 5.] HABENULAR TRACTS OF ACIPENSER. 225 



cylinders are thick, strongly varicose fibers which run toward 

 the central cavity of the lobe, and then along it close over or 

 among the ependyma cells toward the fore-brain. From their 

 horizontal portion the axis cylinders give off collaterals which 

 branch once or twice in the outer part of the granular zone. 

 They may enter the glomerular zone. 



(2) The spindle cells (Fig. i, a) measure 8 to 1 6 by 24 to 

 40 /i. The spindle-shaped cell body stands in a radial position 

 in the internal one-third of the granular zone. From its 

 peripheral end arises a dendrite whose branches end in olfac- 

 tory glomeruli in the glomerular zone. From the central end 

 of the cell a thick process runs toward the cavity and usually 

 ends with an enlargement close to or among the bodies of the 

 ependyma cells. From this enlargement an axis cylinder simi- 

 lar to those of the stellate cells runs backward to the fore-brain. 

 Collaterals rise from it toward the glomerular zone. From the 

 central process, and especially from its enlarged end, arise col- 

 laterals which branch and enter the glomerular zone. Some- 

 times the central process is without an enlargement, and it is 

 perhaps to be regarded as part of the axis cylinder. 



(3) The granule cells (Fig. I, c) are numerous rounded or 

 pyramidal cells measuring 8 to 32 ^ in their greatest diameter. 

 They occur at regular intervals throughout this zone. The 

 character of these cells in other vertebrates has been much in 

 dispute. Cajal in his earlier work ('90) considered them as 

 certainly nervous, the peripheral process being the axis cylin- 

 der. Van Gehuchten and Martin ('91) are in doubt as to their 

 character, having found no axis cylinder. Cajal later ('95) 

 compares them with the spongioblasts of the retina, and con- 

 siders them as nerve cells without axis cylinders. Koelliker 

 ('96, pp. 709-713), on the other hand, regards them as certainly 

 neuroglia. In Acipenser I have found these cells presenting 

 characters showing them to be nerve cells conveying impulses 

 from the olfactory fibers to the fore-brain. A description of 

 them follows. Their peripheral processes rise to the glomeru- 

 lar zone and end in olfactory glomeruli ; either forming, with 

 olfactory fibers, glomeruli into which no other central elements 

 enter, or entering glomeruli formed chiefly by other cells 



