REGENERATION OF ANTENNE—SCHMIT-JENSEN. 535 
Group A.—In this group the amputations were made in the suture 
between the first and second antennal joints, or possibly encroach- 
ing somewhat on the first joint. 
All the regenerations in this group, which includes 14 out of 20 
cases, are small, only 0.5 to 2 millimeters in length. The number of 
joints varies from one to four. The regenerated part is often badly 
crippled, curved, crooked, or spiral. Most frequently one dispropor- 
tionally large claw joint is found together with two or three small, 
plantula-bearing tarsal joints. A claw joint may often be found in 
this group without the presence of plantula on the other joints and 
vice versa. 
Nearly all the cases in this group are in imagines, hence the possi- 
bility is not entirely excluded, that the defective regeneration is due 
to the fact that the amputation was made in a later stage and that 
therefore fewer molts have intervened. Considering the dwarfed pro- 
portions of most of these formations I regard it however, as improb- 
able that further developments could have taken place through inter- 
mediate forms to the very perfect regenerations found in group B. 
Group B.—In this group the amputations were made in the suture 
between the second and the third antennal joint or possibly os 
ing somewhat on the second joint. 
While all the cases within the first group are of nearly similar 
structure and degree of development, this second group, which con- 
tains a minority consisting of six cases, must be divided into two 
subgroups, according to their different composition. 
In the first of these subgroups, which includes four cases, fall the 
largest and highest developed regenerations within the material under 
observation. ‘These consist each of four well-developed tarsal joints 
and one large unjointed tibia-like segment, inserted between the 
place of amputation and the first tarsal joint. The most striking 
of these cases is shown at different stages in figures 6 and 7. A 
description of this interesting regeneration will explain the structure 
of these specimens better than figures. 
With the second antennal joint, which presumably has been partly 
amputated, as a base, is found a tibia-like joint, 2 centimeters long, 
which is constricted at its base and gradually becomes thicker out- 
wardly. It possesses longitudinal ridges, covered with very small, 
stiff, dark brown hairs, exactly as found in the normal tibia. This 
character together with the position of the joint, next to the tarsal 
joints, presumably justifies the characterization “tibia-like.” It 
should be mentioned that the antennal joints have only scattered hairs. 
Connected by a joint with this short tibia-like structure is the long, 
slender first tarsal jomt, which is followed by two short tarsal joints; 
these three joints possess on their plantal surfaces each a well devel- 
oped pair of plantula. On the fourth well-developed tarsal joint is 
