Mr Brindley , Regeneration in Samia ailanthus. 461 
(G) The last larval instar immediately before pupation, and all 
cases had spun by next morning. (Amputation during spinning 
seemed to disturb the larvae very slightly, they almost immediately 
resumed the process of enclosure in the leaf selected, and in no 
case did mutilation seem to delay the act of pupation.) The injury 
was amputation of the basal joint of a posterior leg. Twenty- 
seven imagos were obtained, after a pupation averaging about 
36 days. The injured leg was usually a flattish stump, and in 
the majority of cases there was a projection from this, so that the 
new growth may possibly be regarded as a femur plus an out- 
growth representing tibia and tarsus. In only three cases was 
there a clear suggestion of a tarsus, and in one of these it seemed 
to consist of four joints. There was in no case any indication of 
the claw apparatus. 
(D) The instar as in ((7), but the injury was amputation of 
a posterior leg in the second (or middle) joint. Eighteen imagos 
were obtained, and the injured limb was better formed than in 
(0). The femur was usually very short and broad, but in nearly 
all cases it seemed possible to distinguish tibia and tarsus. In 
two or three cases the apparent tarsus was not divided into joints, 
but in most one or two articulations were more or less indicated. 
In one case only were both the terminal claws well formed. 
It must be understood that the above interpretations of the 
regenerated limb claim only an approximate accuracy, as though 
the larva from its large size is an easy one to injure in exactly 
the desired manner, the imaginal legs are so covered with hairs 
that there is often much difficulty in making out the articulations, 
especially as the regenerated limb is always dwarf and liable to be 
broken in attempts to remove the hairs by brushing. But the 
observations made may be taken as at least emphasizing the great 
variability of result arising from apparently the same degree of 
injury, and as confirming the statements of Newport and Chapman 
in this respect for other Lepidoptera. The want of uniformity of 
result stands out in great contrast from what is known in the case 
of an insect with “ direct” development; e.g. in Stylopyga (Peri- 
planeta) orientalis out of several thousand legs bearing the charac- 
teristic 4-jointed tarsus of regeneration, only 8 were found with 
the tarsus at all malformed. There is much suggestion that the 
mechanism of regeneration is less highly specialised in an insect 
whose post- embryonic development includes metamorphosis than 
in -one which proceeds by simple ecdyses to maturity. In this 
connection it is natural to remember the greater liability of the 
long legs of an orthopterous form to accidental injury compared 
with the short legs of the lepidopterous larva, and the possibility 
that a more specialised mechanism of reproduction has arisen in 
the case of greater need for such. 
