Behaviour of the Soldier Crab — Cameron 
The loser is characterized by: (i) lowering 
of the chelipeds, (ii) flexion of the free walk- 
ing legs, ( iii ) flexion of the legs on which he is 
balanced and subsequent dropping to the sub- 
stratum, (iv) lowering of the eyestalks, (v) 
submissive posture in front of the still threat- 
ening winner, and ( vi ) retreat from the vicinity 
of the hole. 
Although the winner may remain in the vi- 
cinity of the hole for a few seconds, he generally 
moves away shortly after the loser has departed. 
A stopwatch was used to time 84 threat dis- 
plays. The longest encounter lasted for 30 sec 
and the shortest for one sec. From the data col- 
lected, the significance of occupancy of a hole, 
and of disparity in the sizes of the participants 
were estimated. Occupancy was indicated by a 
crab’s position either within a hole or in its 
immediate vicinity. Size was subjectively deter- 
mined with categories of "bigger,” "smaller,” 
and "apparently equal.” The error in allotting a 
pair of individuals to the last category may be 
considerable, but not so great as to obscure the 
general picture revealed by the data. That this 
error in categorisation was consistent is indi- 
cated by the coefficients of variability calculated 
for data of threat displays between apparently 
equal-sized crabs on the one hand, and between 
unequal-sized crabs on the other. These are 
57% and 54%, respectively. 
Of 42 threat displays between unequal-sized 
crabs, 38 were won by the larger of the pair. 
Less than 10% of these encounters were won 
by individuals perceptibly smaller than their op- 
ponents. Of 40 threat displays between appar- 
ently equal-sized crabs, 35 were won by the 
occupant of the hole. Clearly, occupancy is an 
advantage. 
The mean time of threat displays between 
apparently equal-sized crabs (9.8 sec) was com- 
pared with that of threat displays between 
unequal-sized crabs (7.8 sec) by use of a "t” 
test. The means are probably significantly dif- 
ferent (p between 0.05 and 0.1), indicating 
that disparity in size of the participants shortens 
the time of threat display, compared with that 
elapsing during threat display between appar- 
ently equal-sized crabs. 
Only four instances of threat display were 
observed in which neither participant was in 
231 
the vicinity of a hole. In each case the larger 
individual won. 
A considerable degree of social facilitation 
was evident in dense populations of soldier 
crabs during this aggressive wandering phase. 
On many occasions threat displays between pairs 
of crabs were joined by a third, fourth, and 
even a fifth individual. Each individual partici- 
pating in these multiple displays assumed the 
typical threatening posture as far as the crowded 
space occupied by the several crabs allowed. 
Threat display occurred throughout the year. 
Breeding also occurred throughout the year. In 
no month from February 1961 to October 1962 
were ovigerous females absent in the popula- 
tions studied. 
The aggressive wandering phase may con- 
tinue for as long as W 2 hr but generally oc- 
cupies about half that time. Threat displays are 
frequent throughout the phase. As many as 12 
such displays were observed simultaneously in 
an area 20 ft square. Aggressive wandering 
diminishes as the individuals remain in holes 
instead of re-emerging, and the whole popula- 
tion gradually disappears beneath the surface. 
For each of these phases of activity during 
the intertidal period, rhe approximate duration 
has been given. It is emphasised that these times 
are highly variable, and that the entire sequence 
is not always completed. Although the initiation 
of each phase is generally synchronised for all 
the individuals of the emerged population, oc- 
casions were noted when feeding, army wan- 
dering, and returning were being performed 
simultaneously in different areas of a large 
beach. 
REACTIONS TO OTHER ORGANISMS 
During this study seven species were ob- 
served to prey on M. longicarpus. They are 
Threskiornis spinicollis (Jameson), the straw- 
necked ibis; Halcyon sordidus (Gould), the 
mangrove kingfisher; Egretta alba (Linne), the 
white crane; Sphaeroides hamiltoni (Richard- 
son), the toad fish; Ocypode ceratophthalmus 
(Pallas), the ghost crab; Metopograpsus mes- 
sor (Forskal), a grapsid crab; and Conuber 
sordida Swainson, a sand snail. The ghost crab 
and the sand snail preyed on juvenile soldier 
