REGENERATION OF ANTENN2Z—SCHMIT-JENSEN. 531 
unfortunate and somewhat lessens the value of the experiments, but 
as they nevertheless brought some new results, I shall here make 
record of them. ; 
The experiments began in November, 1911, and lasted until 
March, 1912. As material I used 50 specimens of newly hatched 
larvee of the Carausius and about 60 halfgrown larve, all issued from 
unfertilized eggs. The amputations were made by the unaided eye 
with a pair of fine scissors; the newly hatched larvee were anesthetized 
by ether, as it was difficult otherwise to handle these small, delicate 
insects. In most cases the amputation was made as exactly as possible 
in the suture between the basal joint and the second joint, in others 
both these joints were preserved. In one lot of the specimens the 
right antenna was removed, in another lot the left, and in a third lot 
both antennz were removed. Fresh ivy leaves were used as food and 
these were daily sprinkled with water. Most of the specimens of 
this hardy insect thrived well under this treatment, though it must be 
admitted that the mortality was considerably above the normal—all 
cases of cannibalism excepted. It was particularly the individuals 
with both antennz removed which had difficulty in surviving. 
The results of these amputations could be surveyed after a few 
months. In some cases the insects died, in others no regeneration 
took place, or at most a small bud appeared on the place of amputa- 
tion, but a third lot showed regeneration with the formation of not 
only a single claw joint, but also a series of connected tarsal joints, 
and in a few of the specimens an additional tibia-like joint. 
Figure 3 illustrates such a regeneration in an imago. It will be 
seen that there is a large claw joint and three other tarsal joints. 
Comparison with figure 4, which illustrates a newly regenerated tar- 
sus, formed after the amputation of a leg, proved that it is truly a 
tarsal formation. 
Figures 6 and 7 illustrate the most perfect regenerations in which, 
in addition to the tarsus, a tibia-like part has been formed. These 
are described below more in detail. 
A striking difference between the regenerations I produced in this 
experiment and the spontaneous case, described above, is that the 
former are not placed on a stalk of antennal joints, but are emitted 
directly from the basal joint, or from the second joint. The regenera- 
tions developed more and more with each succeeding molt, especially 
in the young larve, which had most changes of skin. The simi- 
larity between the tarsus-like antennal joints and the true tarsus 
became gradually greater as the formations grew larger. 
The steps in the development of these regenerations are about as 
follows: The first molt after the amputation rarely produces any real 
new growth; the wound is grown together and may show a short 
