800 
ROSS GRANVILLE HARRISON 
tions. These experiments likewise resulted negatively. In no 
case were nerve fibers found extending from the transplanted piece 
free into the cavity, although the pieces themselves often showed 
differentiation of fibers, and in cases where the graft had grown 
fast to the walls of the medullar}^ tube, fibers passed from the 
former to the latter. The only conclusions which could be 
drawn from these results were either that the nerve fibers were 
built up by the differentiation of formed protoplasmic structures, 
according to the view of Hensen and Held, or else that the growing 
fibers were positively stereotropic and hence remained within the 
solid tissue instead of passing out into the surrounding fluid. 
Acting upon the latter assumption, the next step was to try a 
fixed medium. Two such were employed, one of which, gelatine, 
gave no results at all, the transplanted embryonic tissue remain- 
ing entirely unchanged after imbedding. The other, clotted 
frog's lymph, gave the results that are here recorded. It was 
rather to be expected that this medium would yield positive re- 
sults, if indeed such were to be obtained at all, for it would be 
chemically the most natural medium, and the fine net-work of 
fibrin threads, bathed by the fluid serum, would in a measure 
simulate mechanicall}^ the protoplasmic net-work, which, ac- 
cording to Hensen, Held and others, seems to exist in the tissue 
spaces in which the peripheral nerves undergo their early devel- 
opment. 
In the first experiments made with the lymph, the technique 
employed was comparatively simple. The tissue to be studied 
was dissected out of the embryo under the binocular microscope 
in 0.4 per cent sodium chloride or in Locke's solution without 
sugar. It was then transferred to a cover-slip by means of a cap- 
illary pipette, and a drop of lymph draw^n from one of the lymph 
sacs of an anaesthetized frog was quickly dropped upon it. The 
cover-slip was then inverted over a depression slide and the prep- 
aration kept in a moist chamber. In order to avoid evaporation 
while the specimens were under examination it was necessary to 
seal the preparations, which was done most satisfactorily by apply- 
ing'melted paraffine around the cover-slip with the edge of a warm 
plate. 
