Biology of Pachygrapsus crassipes —HlATT 
191 
autotilly, and autopasy all occur in crabs in the 
A 1 stage. It was soon discovered that the limbs 
of newly molted individuals could not be 
grasped or held in any manner because auto¬ 
pasy is effected rapidly at this time. Despite 
utmost care, autopasy has occurred on several 
occasions while newly molted animals were 
being measured. Moreover, appendages are 
frequently found in aquaria containing recently 
molted crabs, which indicates that autotilly is 
common prior to complete sclerotization. 
Although cannibalism is prevalent when 
newly molted crabs are located in an aquarium 
with hard-shelled individuals, none of the latter 
has been observed to consume portions of a 
severed appendage, nor have any severed ap¬ 
pendages been found from which portions had 
been eaten. It seems conclusive, therefore, that 
autophagy is lacking in this species. However, 
crushed appendages tossed into a tide pool 
are readily consumed. The lack of autophagy 
may perhaps be partially explained by the fact 
that the sclerotized integument precludes con¬ 
sumption of the potential food; however, it is 
strange that the soft appendages of a recently 
molted crab are not utilized. 
Regeneration 
Although sufficient time to pursue an exhaus¬ 
tive investigation into the subject of regenera¬ 
tion in P. crassipes was lacking, a cursory survey 
of this important activity of these crabs was 
made. The present study embraces the phenom¬ 
enon of regeneration through three rather dis¬ 
tinct avenues of approach: (1) Meticulous 
periodic examinations of several regenerating 
appendages were undertaken to disclose the 
events which occur during this process; (2) 
data were secured on regeneration and inter- 
molt stages of a select group of crabs, not to 
provide a statistically accurate account of the 
regeneration by intermolt stages, but to find a 
few salient facts concerning regeneration, which, 
when correlated with the information now avail¬ 
able on the intermolt cycle, might provide a 
framework for further investigation of this oft- 
occurring phenomenon in this species; and (3 ) 
to establish the frequency with which this 
species undertakes regeneration in its normal 
environment, the data for regenerating append¬ 
ages on hundreds of collected specimens were 
analyzed. 
Historically it is found that regeneration in 
decapod crustaceans has drawn the attention of 
biologists for more than two centuries. Reaumur 
(1712) first described the regeneration of legs 
of crabs, lobsters, and crayfish. Later, informa¬ 
tion on morphological development (Brooks, 
1882; Herrick, 1896; Perkins, 1927; Yu, 1932; 
and a host of others) was combined with con¬ 
siderable experimental data chiefly presented by 
Zeleny (1905) and Emmel (1907) to formu¬ 
late our present extensive knowledge of this 
phenomenon. 
Before presenting the regeneration data for 
P. crassipes , a brief synopsis of the internal 
morphological transformations which occur 
from the instant of pereiopodal mutilation to 
the onset of regeneration of the new limb will 
be helpful toward an understanding of the 
external limb transformations which occur in 
this species. Our present histological and cytolo- 
gical knowledge of regeneration in decapod 
Crustacea stems primarily from macruran and 
anomuran types (Reed, 1904; Emmel, 1910; 
Paul, 1915 a), with the latter worker contribut¬ 
ing, in addition, some information on the brach- 
yuran group. In general, all three groups follow 
the same fundamental pattern and differ only in 
detail. The following account is fundamentally 
based on regeneration in the hermit crab as 
described by Paul (op. cit.). 
At the fracture plane the sclerotized strata 
of the integument are discontinuous for a part 
of the circumference of the leg. The columnar 
epithelial cells are greatly enlarged at this point 
and possess processes which extend inward, 
meeting those of the opposite side. These fibers 
mat together to form the diaphragm. The appen¬ 
dicular artery passes through a foramen, which 
it completely fills; but the foramen for the 
nerve is a funnel-like prolongation of the dia¬ 
phragm which fits loosely about the nerve, 
thereby affording the venous blood a return 
