Amount of Tail Regenerated to Removed in Tadpoles 441 
duced a fatal injury. In Experiment 4 it was noted that Series 
T (20 mm.) had a mortality of 80 per cent; and that the four 
tadpoles which did survive the loss of 20 mm. of tail, regenerated 
abnormally. All of the tadpoles in the series with 20 mm. 
removed, died, in Experiment 5. Since tadpoles of the same size 
and subjected to the same conditions as these animals just men- 
tioned, regenerated rapidly from the 15 mm. level, the 20 mm. 
level, in the light of the above statements, was considered the 
lowest level (that is the level nearest the tip of the tail) from which 
the removal of the tail produced fatal injury. Injury at the 20 
mm. level, removed about 80 per cent of the tail, the average 
tail length ‘of these tadpoles being 26 mm. The entire tail was 
removed from several tadpoles, independent of the experiments 
mentioned, and all of the animals so operated upon died within two 
or three days after the injury. 
Consider now the levels at which regeneration took place. 
Beginning near the tip of the tail, injuries at all levels were fol- 
lowed by regeneration until the fatal level, 20 mm., was reached. 
The per cent of the part removed, regenerated from the vari- 
ous levels, was the same regardless of the level; that is, the amount 
regenerated was proportional to the amount removed. The 
time, however, consumed in regenerating grew longer, as has 
already been mentioned, as the level of injury approached the 
head. Although the length of the period of regeneration was 
greater for the regeneration from the higher (the more cephalad) 
levels; this period was relatively longer for the series injured at 
the lower (the more caudad) levels, making the rate of regenera- 
tion higher, the more cephalad the injury. Take for example the 
final amounts of Experiment 4 which are typical. Together with 
the time required by each for this regeneration, and the average rate 
per day, in millimeters, they may be grouped in the following 
table: 
