34 
continuously through the canalis centralis to where the canal widened 
to form the fourth ventricle. Further search in other species of Tele- 
osts revealed its presence in all examined. 
“ The occurrence of a continuous rod or fibre in such a position 
was so startling and the sharpness and definiteness of its appearance 
so convincing, that IT immediately devoted my time to its investigation, 
and have succeeded in proving, I believe, that this fibre is an organic 
structure, occurring in all classes of Vertebrates, and intimately con- 
nected with the central nervous system which surrounds it, but within 
the lumen of which it lies for the most part free. 
I take this opportunity of acknowledging my indebtedness to Dr. 
G. H. Parker for advice and criticism, to Dr. E. L. Marx who has 
kindly read the manuscript and to Dr. H. C. Bumpus, Scientific Di- 
rector of the Wood’s Holl Laboratory of the U. S. Fish Commission, 
who has extended to me while an occupant of a Harvard University 
- Table opportunities for collecting a large amount of the material on 
which these studies are based. 
It is remarkable that so peculiar a structure of such wide 
occurrence, and of so great importance in the nervous anatomy 
as to persist throughout the vertebrate series should have remained 
practically unknown to this time. This is no doubt largely due to the 
fact that in the study of the spinal cord transverse sections, in which 
this rod is very inconspicuous, have been so exclusively relied on. 
However, recent investigators who have studied the central nervous 
system in longitudinal sections have failed to note this structure prob- 
ably for the reasons that 1) many staining methods fail to bring it 
out clearly, and 2), being a structure of great delicacy and one of 
the first to break up, it shows clearly only in perfectly fresh material 
carefully fixed. 
About forty years ago Rerssner (’60) described a highly refractive 
cylindrical rod 1,5 u in cross section lying within the lumen of the canalis 
centralis of Petromyzon. This rod showed none of the variations in 
form which sometimes occurred in the axis cylinders in the chromic 
acid preparations studied by him. He recognized this rod as being 
wholly unlike the amorphous mass which usually fills the lumen of the 
canalis centralis wholly or in part, and which had been previously 
observed by many investigators. He denied that it could have been 
formed from thrown off epithelial cells or blood corpuscles, as similar 
contents of the canal had been accounted for by Srixuine (59). If, 
he said, it was formed from the coagulation of the albumen of the 
cerebro-spinal fluid, as BIDDER und KUPFFER (’57) had supposed, then 
