STRUCTURE AND DEVELOPMENT OF REISSNEr’s FIBRE. 5 
there is no risk of overhardening, and the material may be 
left in it safely for several days, or even weeks, and, if 
thoroughly washed, is always found to be in good condition 
for staining. It has, too, the further advantage that small 
entire specimens left in it for several days usually do not 
required subsequent decalcification. 
Specimens of small size (small ammococtes, embryo dog- 
fish, Teleosts up to 30 mm., or even 40 mm. in length, 
amphibian larvm, etc.) were, as a rule, plunged whole into 
the fixing fluid ; in the case of larger specimens some dissec- 
tion was necessary. 
Wherever practicable freshly killed material was taken, 
and the brain partially exposed as rapidly as possible, the 
entire animal then being plunged into a large quantity of the 
reagent. Further dissection to complete the exposure of 
the brain was usually carried on while the specimen was 
immersed in the fixing fluid. 
Such specimens as were sufficiently small were sectioned 
entire, all risk of damage to the central nervous system being 
thus avoided. In most cases, however, the length of the 
central nervous system was too great to admit of the cutting 
of longitudinal sections through its entirety. In these, the 
brain (or head) with a considerable portion of the spinal 
cord behind it, and also the terminal jjiece of the spinal 
cord, were removed, but, before the spinal cord was severed, 
the exposed portions of the central nervous system were 
allowed to become thoroughly penetrated and fixed. The 
severance of the S 2 )inal cord might generally be safely per- 
formed, where aceto-bichromate was the i-eagent onployed, 
after the lapse of from half an hour to one hour, according to 
the size of the S])ecimen. 
By the adoption of this ]>recaution the characteristic recoil 
of the fibre, with the resulting tnngle of its free ends, which 
Inis been remarked by several observers (Sargent, ’04, 
Nicholls, ’12), was generally prevented, and the fibre was 
jireserved in its natural relations. In one or two instances, 
however, it was found that Beissuer’s fibi’e had, notwith- 
