148 
BULLETIN OF THE UNITED STATES FISH COMMISSION. 
Nothing is definitely known about the causes of repetition. It looks at first sight 
as if the deformities in the appendages of the lobster and other arthropods could be 
explained as phenomena of regeneration, though in this case there is addition rather 
than replacement. 
We have seen that where the cheliped is thrown off by reflex muscular contrac- 
tion the bud of a new leg forthwith appears, and under favorable conditions grows into 
a perfect limb. The tissues of this limb are developed out of the formed histological 
elements of the stump. The process of regeneration is begun in this case as a direct 
result of a loss or injury. The plane of fracture lies between the second and third 
joints, and all parts peripheral to the second joint are reproduced. Repetition of parts, 
however, occurs in many insects and vertebrates where no such regeneration of lost 
appendages is known. However, the power of regeneration, which is present in all 
organisms, differs rather in degree than in kind, and Weismanu has shown that it has 
probably been developed in many cases as a means of defense and protection to the 
individual. (See p. 107.) 
In the specimen of Palinurus (No. 808, Bateson, originally described by Leger 
in 1880), where the left penultimate leg bears two supernumerary legs, both of which 
spring from the basipodite, it certainly looks as if what would have taken place in 
the case of loss of the original limb — namely, the growth of a new one from this joint 
— had happened repeatedly, so that instead of the regeneration of one limb at a 
time there is the superaddition of two; but where the limb is not cast off the first 
superadded one is in secondary symmetry and belongs to the opposite side. If such 
an interpretation will apply to this case, the various other repetitions and abuormal 
growths which arise in more peripheral joints, as upon the sixth and seventh, fall into 
the same general category. In the case of the growth of a new limb in consequence 
of loss, however, all parts which are external to the plane of fracture are reproduced. 
In the cases of repetition this is not usually the case, as Bateson has shown. The 
bud which arises on the propodus (as in figs. 187, 190) may by fission give rise to a 
second propodus, but not usually, if ever, to a dactyl. 
It seems as impossible to suppose that such a deformity as that seen in fig. 187 or 
fig. 189 is congenital as that it is the result of injury. The monstrosities which occur 
in the embryo, which are considered in another place, are, however, in some cases at 
least, the result of injury or unfavorable conditions. 
Autotomy, or the casting of the claw at the second joint, is probably directly 
accountable for the rarity of abnormal growths in the limbs of the higher Crustacea. 
It is extremely improbable that any deformity at the extremity of a limb could sur- 
vive autotomy, but the experiments to settle this interesting point have yet to be 
made. While it would appear that the various deformities which have been described 
can not be explained as the results of injuries and the attempted regeneration ot 
injured parts, since the limb is usually thrown off in such cases to be completely 
renewed or it is retained to be completely restored, yet I can not escape the convic- 
tion that the problem is in some way directly concerned with that of regeneration. 
The mechanism by which so complete a structure as a limb is regenerated can not be 
regarded as simple. As Weismann says, the machinery of a cotton factory can not be 
made out of a few simple levers. It is probably exceedingly complex, and it is no 
wonder that the parts do not always work harmoniously, that the thread is sometimes 
knotted or the product useless. 
