806 ROSS GRANVILLE HARRISON 
TABLE 1 
Summarizing the results of the experiments 
TISSUE ISOLATED 
NUMBER OF CASES RECORDED 
NUMBER OF CASES REJECTED 
AS BELOW STANDARD 
NUMBER OF VALID CASES 
NUMBER OF CASES IN WHICH 
PROTOPLASMIC NERVE FILA- 
MENTS WERE FORMED 
PROPORTION OF CASES SHOW- 
ING NERVE FILAMENTS 
NUMBER OF CASES IN WHICH 
MUSCLE FIBRILLA WERE OB- 
SERVED 
NUMBER OF CASES IN WHICH 
MUSCLE TWITCHING WAS 
SEEN 
NUMBER OF CASES IN WHICH 
CILIARY MOVEMENT WAS 
SEEN. 
90 
19 
71 
35 
.49 
4 
12 
2 
Branchial ectoderm 
15 
4 
11 
6 
.55 
7 
Abdominal ectoderm 
18 
5 
13 
5 
Axial mesoderm alone 
30 
8 
22 
P 
3-54 
Notochord 
' 3 
0 
3 
Endoderm (yolk) 
2 
0 
2 
■■ 
Total 
1582 
36 
122 
423 
7-9 
12 
14 
^ In many of these cases no attempt was made to exclude all of the axial meso- 
derm. This accounts for the presence of muscle fibers in some. 
2 The excess of this number over the total number of recorded experiments is due 
to the fact that in some preparations several kinds of tissue were included. 
3 This isolated case is one of a series in which the attempt was made to separate 
the myotomes from the medullary cord along their natural boundary. This is 
very difficult to do with absolute accuracy and it is supposed that in this case some 
cells from the medullary cord were left attached to the mesodermic tissue. In 
subsequent experiments cutting in close proximity to the nervous system was 
avoided and only the lateral portion of the mesoderm was taken. 
* The small number of cases recorded as showing striations is due to the fact that 
the preparations were examined only in toto. Had sections been cut it is believed 
that the number of positive observations would have been considerably larger. 
only in certain cases that the oil immersion could be used, and then 
it was found to have no great advantage over the water immersion. 
A large number of sketches were made, nearly all with the camera 
lucida. In making these especial care was used to show the length 
of the fibers, and the form of the end organ correctly. Owing 
to the extreme fineness of the terminal filaments and the con- 
stant changes which they undergo, it is not, however, possible 
always to draw them with absolute accuracy in every detail. Still 
