The
ects of Temperature and Desiccation on
Pagu
samuelis an! Pagurus granosimanus at
Pacific urove, alifornia
Richard Forward
Hopkins Marine Station
Stanford University
Pacific Grove, California
Introduction
A crude survey of the hermit crab population indicated a
marked lifference in the intertilal areas occupied by Pagurus
samuelis (Stimpson, 1857) and Pagurus granosimanus (Stimpson, 1859).
It has since been determine by concurrent work of R. Belknap and
John Markham that P.samuelis ranges from O.3 to 1.3 feet intertidally
and P. granosimanus from -.6 to .55 feet. Melody Bollay (1964)
demonstrated that P. samuelis coul withstand gross exposure longer
than P. granosimanus. This suggests that desiccation and temperature
tolerance may be limiting species distribution, and it is these
physical factors which I investigated.
Metod
The hermit crabs were obtained intertidally at Hopkins Marine
Station and kept alive in aquariums supplied with running sea water.
Only adult crabs as jescribed by Schmitt (1921) were used. The two
species were first tested to determine water temperature tolerance
and their lethal water temperature (taken as that temperature at
which 508 of the crabs can live indefinitely). Twenty five crabs of
each species were placed in plastic dishes, to prevent random mixing.
in an aquarium. Using a Precision Thermo-Regulator the 14°0. sea
water was heated to the jesired temperature in order to simulated
the slow heating occuring in exposed tidepools. At intervals of 1.
4, and 8 hours after reaching this temperature the number of crabs
alive was recorled. Air was pumped into the aquarium causing water
circulation, thereby preventing stratification of heated water.
Shell temperatures were investigated by exposing crabs out of
water to bright sun light between 11 A.M. and 3 P.M. Since most
P. samuelis and P,granosimanus occupy Tegula funebralis (A. Adams
1854) shells (Bollay, 1964), only these were used. A 3/16 inch hole
was drilled in the upper most portion of the shell,and a small (1.5mm.
temperature sensitive probe inserted. Temperatures were measured on
a YSl Model 43 Single Channel Tele-thermometer. The crabs were placed
in a plastic dish, whose bottom was covered by sand. Readings were
taken every minute until the animal either vacated the shell or
became completely inactive.
To study normal body temperatures of each species I inserted
the probe, used in the shell temperature experiment, into the thorax
of animals(ut in half)and read temperatures on the Tele-Thermometer.
Using essentially the same procedure, I determined body temperatures
at death from exposure. Equal numbers of each species both in and
out of the shell were again exposed in plastic dishes whose bottom
was covered by sand. The time till death for each animal was recorded,
and species behavior observed. At death the animals were cut in half.
and the probe inserted into their thorax.
Desiccation experiments were begun by comparing rates of water
loss when exposed to bright sunlight in open plastic dishes. I
used five crabs of each species both in the shell and out; their
total weights were taken every 15 minutes. Weight loss was attributed
to water loss and computed as weight loss per unit of original body
weight.
Since different rates of water loss were observed, I next tested
the amount of desiccation tolerated by each species before death.
Each crab was removed from its shell, weighed and placed in
72
individual 50 ml. glass beakers within a desiccator, whose atmosphere
was kept dry by caleium chloride. The time till death and weight
at death were recorded. The weight loss was attributed to water
loss and calculated as percent of original body weight. Tabulations
of results within species were divided into three groups, females,
gravid females and males.
Since similar hermit crabs are isotonic to normal and slightly
concentrated sea water (Prosser, 1950, p.24) possibly they might
die in sea water from which was removed that peroent of water they
lost at death. Therefore I weighed and placed four orabs of each
species in 1.5X sea water which I prepared by removing 38% of its
original water. The number alive and total crab weights were recorded
after 24 and 41 hours.
Their failure to die within the concentrated sea water indicates
their bodies can withstand higher salt concentrations and suggests
that possibly drying of the gills could cause jeath in the desiccator.
Therefore five orabs of each speices were placed in the desiccater,
and at 2 hour intervala 0.02 cc. sea water was injected onto their
gills by a 1/4 co. Tuberculin syringe. The time till death was recorded
and compared with that for normal death by desiccation.
Another possible cause of death during desiccation may be
waste product accumulation, which cannot be removed in the absence
of excess water. This was tested by placing three crabs of each
species (in their shells) within individual air tight jars containing
wet toweling (thereby keeping the atmosphere saturated and preventing
desiccation). The toweling was remoistened with sea water every
10 hours, the weight of the crabs taken originally, after 24 and
41 hours, and number dead recorded at these times.
Résults
The death rates plotted in figure 1 and 2 show P. samuelis can
tolerate high water temperatures better than P. granosimanus.
More exadt temperature levels indicated the lethal temperature
for P.granosimanus to be 29.7°0. and P.samuelis, 31.5°0.. Although
during an inconclusive survey of exposed tidepools, the highest
temperature I observed was 29.0°0., I feel higher values could be
reached for short periods during a low tide on a very hot day. Thus
if P.granosimanus were exposed one hour to 31°0. they would begin
dying, and one hour exposure to 32.9°0. would begin killing P.samuelis.
Since the two species live in an intertidal region, some ability
to withstand exposure out of water is necessary. The shell temperature
experiments (Table I and figure 3) were conducted under the most
extreme field conditions, for crabs are rarely found exposed on dry
sand, but rather are under rocks or nestled in the algae; however,
the results jemonstrate ability to live under these conditions.
Although not indicated by my data, shell temperatures appear to
depend on the degree of sun exposure, for a passing cloud rapidly
lowered the temperature two to three degrees, while its disappearance
increased the temperature.
The representative graph (Figure 3) shows P, samuelis more active
and better able to control shell temperature for a longer time, almost
double that of P.granosimanus. A successful attempt at control was
considered any act by the crab which lowered shell temperature. The
surprising observation was the apparent behavioral response of
P samuelis to control its increased shell temperature. The crab
would come partially out of the shell, begin walking,as if airing
the shell's interior, and then withdraw deep into the shell, thereby
130
lowering the interior temperature. As indicated by the data (Table I)
the average number of successful attempts was 6,and the average
temperature drop found to be 0.5 0. P.samuelis vacated the shell
either due to high temperature, or because the probe was inhibiting
complete withdrawal into the shell.
P.granosimanus appears to lack a behavioral control of shell
temperature, for only 5 out of 10 crabs made one successful attempt
at lowering the temperature. Osually its "walking-withdrawal" response
increased the temperature. The shell vacating tendency also seemed
absent, as 9/10 of the crabs hung half out of the shell in an apparent
narcotized state at about 29.0°0., which interestingly is olose to
their lethal water temperature.
Body temperature experiments revealed some interesting trends.
As expected of poikilotherms both species' normal body temperatures
(Table II)were only slightly higher than the surrounding water
temperature. When exposed without their shells, the time till death
was the same for both species, 45-65 minutes (Table III). The body
temperatures at death seemed dependent on exposure conditions, as
the second day's values were higher than the first. But generally
the mean body temperaturés of P.samuelis are higher than those of
P.granosimanus.
Body temperatures at death within the shell (Table IV) again depend
on exposure conditions with the second days' results being higher than
the first. Contrary to results without the shell, P.granosimanus
generally has a mean temperature higher than P.samuelis. The times
till death for P.granosimanus ranged from 65 to 90 minutes, while
those for P.samuelis were from 65 to 220 minutes. This indicated o
the shell acts as a protective agent against exposure.
O
The behavioral responses of P. granosimanus seam to correlate
with those observed during the shell temperature experiment, as 2
of the 16 crabs vacated their shells, and the rest seemed to die
hanging  to of their body out of the shell. However only one
third of the P.samuelis individuals vacated the shell while the rest
withdrew tightly into its interior. This suggests the 80% exit in
the shell temperature experiment was higher than normal,and probably
due to the probe obstructing the animals complete withdrawal into
the shell.
It is surprising that P.samuelis which occupies a higher inter-
tidal zone than P, granosimanus should lose more water and faster,
(Figure 4amp;5) especially since other crabs living higher in the inter-
tidal have acquired the ability to tolerate exposure. (Pearse 1929) This
problem appears resolved when the walues for percent of desiccation
(Table VæpVI) before death were determined. Clearly P.samuelis can
withstand a greater water loss than P.granosimanus. Within each
species the trend in amount lost goes from male, withstanding the
least, to female, and then gravid females, withsta ding the greatest
loss. The experiment of removing toe females' eggs demonstrated they
still lost the greatest amount of water before death, which suggests
a physiological change within the gravid hermit crab. The relation
between weight an' water lost before death within each species generally
suggests that the larger animals can tolerate a lower percentage
of weight lost as water than the smaller. This correlates with the
casual observation that the larger crabs are lower in the intertidal
zone.

If desiccation leath results from loss of body water theg might
dan
die when placed in 1.5X sea water, Presumably their internal body
water should decrease osmotically, and likewise their weight. Table VII
disproves these assumptions, for the animals did not die and did
gain weight. This suggests that the maintenance of a constant amount
of internal water balances the increased salt concentration. (Also
the denser sea water makes the animals weigh more.)
The results in Table VIII show that neither specles remains alive
if only its gills are moistened. Normally P.granosimanus died from
6-11 hours after being placed in the desiccator and P.samuelis 7-134
hours. The hours till death when I moistened their gills show that
both species die within the normal death time. Thus the cause of
death during desiccation is questionable. Gill drying appears not to
be the cause, but since the crabs seem to adjust to more concentrated
sea water, body water loss may be the primary mortality factor.
The data in Table IX rule out the possibility of waste product
accumulation causing death, for only one P.granosimanus died after
A mata,
41 hours, Tais time is much longer than the normal time interval
observed for mortality within the desiccator. P.samuelis il and /3
and P.granosimanus 1 and 2 seem to lose weight after 24 hours
probably because of initial weighing inaccuracies due to water held
within the shell.
Discussion
I feel the foregoing data can be correlated by a theoretical
anemals

explanation for the causes of death when exposed out of the shell
and within it. When exposed out of the shell both species died
between 45 and 65 minutes with P.samuelis having a higher body
temperature. As shown in Figure 5 the amount of body water lost at
the time of death of P.granosimanus would be 14-19.5% and that of
P.samuelis 16-21.54. As Table VI shows,the percent water lost before
death for P.granosimanus is about the same as the amount lost in
this time interval. This suggests P.granosimanus is dying from water
loss, and its low temperature could be due to water evaporation.
P.samuelis, although losing water faster (according to Table 7) has
not lost enough water to cause death. This sugrests the crabs are
dying from limited water loss combined with high temperature (as
indicated by their higher body temperatures).
However hermit crabs are almost never observed exposed outside
their shells within the intertidal zone. Table IV shows that P.samuelis
within the shell lives longer and has a lower body temperature at
death than P.granosimanus. Interpolating from Figure 4 in the time
interval till death P.ganosimanus should hawe lost 21.5-22.0% of
its water. These high values woul' suggest lesiccation as the killer,
but initially each crab stores within the shell free water which it
extrudes when exposed. Therefore much of this weight loss results
from extruded water, and I seriously doubt the animal has lost enough
water to cause death. Thus I would attribute le th mostly to
higher temperatures, (as indicated by its body temperature) combined
with limite? water loss.
131
Figure 4 demonstrates P.samuelis over the time interval for death
when exposed (Table III) should have lost between 31.5 and 45.0
of its body weight as water. Although they did extrude free water
when exposed,these losses are still much higher than the observed
desiccation death values. Since P.samuelis can maintain lower shell
temperatures, probably by water evaporation, and its body temperature
at death is far below the lethal water temperature, I feel P.samuelis
died from desiccation while maintaining lower body'and shell
temperatures by means of the behavioral response previously described.
Therefore by testing the effects of physical factors, I can
suggest how P.samuelis is better adapted to withstand more extreme
conditions encountered higher within the intertiial zone, but not
why it inhabits this area. P.samuelis as compared to P.granosimanus
can tolerate higher water temperatures possibly encountered in high
exposed tidepools. When exposed to the sun within the shell,
P.granosimanus appears to die mainly from high tempevatures,and it
is this harmful factor against which P.samuelis seems adapted.
Temperature control supersedes water conservation, necessitates the
ability to withstand greater desiccation and elicits the evolution
of a behavioral temperature-controlling response.
Summary
1. fests were performed on P.samuelis and P.granosimanus to
deterwine the effects of temperature and desiccation.
2. F.sanwelis can tolerate higher water temperature than P.Kranosinanus
with P. sammelis lethal temperature at 31.500. ani E.granosimanus 29.700
3. Shell temperature measurements on exposure to sunlight showed t at
P.Samuelis maiatained a lower shell temperature longer an' possessed
an a parent behavioral response for controlling temperature. (In
response to increase temperature the animal woul' ap arently air
the shell by walking and then withiraw deep into its interior, thereby
lowering the shell temperature.)
4. Both species ile at 45-65 minutes out of the shell when exposed to
the sun with P.samuelis body temperature at death higher than P.
granosiranus.
5. When exposed within the shell P.samuelis livei longer and ha
a lower body temperature at jeath than P.granosimanus.
6. Pisamuelis loses more water and faster than P.granosimanus both
in and out of the shell, but t e former can lose about 27.06 (nean
value) of its body weiwht as water before death, the latter oily
18.36 (mean value). In both species rales can tolerate the least
loss, then females; females with egges tolerate the greatest loss.
7. Generally P.samuelis is a'apted to a higher position in the
intertidal zone due to its ability to withstan' an' control temperature.
Table I Shell Temperature Experiment
of Attemps to control
Temp. at which First attempted
Temp.
to control temp.
P. granosimanus
P.samuelis
P.Granosimanus
P.samuelis
none
26.3°.
24.90
none
25.20
24.9
27.
nne
24.1
50
2.80
23.
25.8°0.
noe
26.8
ne
22. 3
mean 25.00c.
mean.5
mean
mean 24.3
Temp. vacate
Temp. hang
out of shell
shell
P.samuelis
P.granosinanus
28.0
00c.
26.00
27.00
30.4 c
380
28.
26.90
no exit
29.(
28
26.
33
30.0
30.80
left shel!
no eit
mean 29.270.
mean 20.
Temp. "ise
time
time
P.granosimanus
P.samuelis
00
8 min.
19.3°0.-28.
min.
.-2
19.
22.
2°0.-27.7
21.8°0.-26
10 min.
9 min.
23.5°0.-28.80
20.2°0.-30.00
min.
min.
nin
23.00c.-2
min.
20.580.-28.200
80
C.-24
o0.-27.080
in.
13 min.
20.9°0.-29.0
19.0°c.-29.
min.
25 mir.
min.
24.5
20.0,C.-32.0
nin.
'c.-29.
23.0°0.-27.2
3.-35.590
min.
min
22.0
21.5°C.-32.5
19.29
C.-29.00
11 min.
min.
23.0 C.-27.
48 min.
6 min.
23.5°0.-24.
mean
mean values
21.2°0.-29.1°0. 11 min.
21.4°c.-28.2°0.
19 min.
C
Body Temperature of fagurus Under Different Conditions
Table II
Normal Body Temp.°.
P. samuelis
P. granosimanus
14.5
14.2
14.5
14.5
14.6
14.8
14.4
14.6
14.5
14.5
mean 14.5°0..
mean 14.500.
Table III
Boiy Temp. "0. at Death out of the Shell When Exposed to Sun
Trial 41
Trial #2 (Next Day)
Time till death 50-65 min.
Time till death 45-65 min.
P. samuelis
samuelis
P.granosimanus
Paganosimanus
7.5
29.5
22.
25.8
28.8
3.2
25.5
24.3
26.8
25.5
27.0
26.5
26.5
27.0
25.0
24..
mean 28.2%0.
Rean 27.10.
25.0
23.8
s -1
S -
24.7
25.0
25.
mean 25.6°0.
mean 24.200.
s = 1
s 2.8
Table IV
Body Temp. 90. at Death When Exposed to Sun in the Shell
Trial 1
P. samuelis
P.granosimanus
Time till death
time till death
65-220 min.
65 min.
temp.
temp.
24.0
26.5
24.5
25.8
26.0
vacated shell
vacated shell
27.0
26.0
25.9
26.5
mean 25.0°.
mean
26.3°.
s =.4
Trial #2 (Next Day)
P.samuelis
P.granosimanus
time till death
time till death
65-160 min.
65-90 min.
temp.
temp.
28.2
28.2
18.0
vacated shell
26.5
29.0
27.9
30.0
vacated shell
29.1
29.
27.9
29.0
28.1
31.0
mean 28.4°0.
nean 28.800.
s =1
s =.9
Desiccation Experiment Table VI
P. granosimanus
Female
Male
lost before
lost before
Weight
weight
death
death
gms.
gms.
1.344
1.068
16.8
13.2
1.394
.462
17.1
13.3
18.4
.960
1.490
13.8
19.3
.63.
14.0
1.140
19.6
.462
14.7
.680
335
23.0
1.140
14.9
Mean percent 19.02
.910
15.8
1.349
16.1
-1.1
1.244
16.7
16.8
1.130
.470
17.0
18.4
.998
18.6
1.007
.336
19.1
957
19.2
19.8
.972
19.9
1.829
20.7
.630
643
20.8
2.
608
Mean percent 16.83
s=2.8
Femaleswith eggs removed
Weight Zlost before
death
18.6
.910
.889
20.9
24.6
420
4ean percent 21.47
s=1.8
Gravid Female
lost before
Weight
death
gms.
.709
14.7
.850
15.3
1.1?
15.5
1.024
15.7
584
16.9
380
17.1
17.3
.896
.863
17.4
.986
17.
1.150
18.2
18.4
.760
541
18.8
.670
19.4
1.170
19.6
20.3
1.130
.780
20.5
0.7
.990
20.8
.847
21.2
.848
399
21.3
23.0
.907
.93
24.0
26.
Mean percent 19.23
s= 2.9
Table V
Desiccation Experiment
P. Samuelis
male
Female
weight
weight
Blost before
lost before
gms.
death
death
gms.
1.134
321
21.3
23.4
.726
21.4
440
25.0
22.4
.850
385
25.1
1.014
25.4
22.5
.562
.480
25.5
349
22.9
560
22.9
26.8
1.103
.763
301
27.2
395
27.4
23.4
1.162
28.6
.490
23.6
1.176
.524
.670
24.1
28.6
236
24.2
29.0
1.312
24.4
29.7
1.310
.478
25.2
413
.588
30.2
25.4
.606
Mean percent 27.38
25.5
1.312
28.2
s = 2.1
537
28.4
660
.965
28.6
28.6
29.0
.972
236
29.0
29.4
.510
29.8
.525
30.0
1.167
359
31.4
31.9
Mean percent 25.87
s =3.2
Females with eggs removed
weight 3 lost before
death
gms.
.824
28.5
.632
30.1
521
Mean percent 30.52
s =1.3
Gravid Female
weight % lost before
death
gms.
.814
24.3
.692
24.0
.958
25.3
25.4
.860
.806
1.440
31.2
629
.586
32.
440
32.5
.946
32.9
36 4
660
Mean percent 29.53
s =4.1
Placement in 1.5X Sea Water Table VII
P. granosimanus
P.samuelis
total weight alive
total weight alive
time
4.075 gms.
originally
3.028 gms.
4.113 gms
3.161 gms
24 hr.
4.259 gms
41 hr.
3.197 gms.
Table VIII
Time till death of first ten crabs under normal conditions in
the desato
P. samuelis
P.granosimanus
Hr.
Hr.
30
10
121
10
134"
11
desiccator when gillare moistened
Time till death
P.Samuelis
P.granosimanus
10t hr.
hr.
104"
10
102
Table IX
Moist Air Experiment
P. granosimanus
P. samuelis
weights in the shell
weights in the shell
originally after 24 hr. after 41 hr
after 24hr. after 41 hr.
originally
5.582 gms.
5.415 gms. 5.387 gms.
5.209 gms.
5.053 gms.
5.081gms.
6.957
6.7
6.796
5.304
5.260
5.217
4.767 "
4.757 dea
4.787
7.333
7.316
7.362
C


O
No. Alive
L
Fig.
N
Ali
O
O

Temp°
10


Fig. 3
C
(D
A
Fig. 4
B
u6
O
O
1
8
Fig. 5
Figure 1. Water Temperature tolerance by P.samuelis.
4) Begin 14°C., Final Temperature 30.100., time to reach temperature
2 hr. 5min. (Same result for all temperatures below 30.100.
B) Begin 14°C., Final Temperature 31.0°0., time to reach temperature
2 1/3 hr. C) Begin 14°0., Final Temperature 31.5 (Lethal Temperature),
time to reach temperature 2 1/3 hr. D) Begin 13°0., Final Temperature
31.9°0. time to reach temperature 2 hr 34 min.. E) Begin 14'C.,
Final Temperature 32.8°0., time to reach temperature 2 hr. 20 min..
Figure 2. Water Temperature Tolerance by P.granosimanus
4) Begin 14°0., Final Temperature 28.9°0., time to reach temperature
14 hr. (Same result for all temperatures below 28.700.)
B) Begin 14°0., Final Temperature 29.700. (Lethal Temperature), time
to reach temperature 1 3/4 hr..
G) Begin 14°0., Final Temperature 30.1°0., time to reach Temperature
2 hr. 5 min.
D) Begin 14°0., Final Temperature 31.000., time to reach temperature
2 1/3 hr.
E) Begin 13°0., Final Temperature 31.900., time to reach temperature
2 hr. 34 min., (Same result for 32.9°0.)
Figure 3. Shell Temperatures on Exposure to Sun
A. Temperature rise for P.granosimanus. Plot ends as animal becomes
narcotized at 29.200.
B. Temperature Rise for P.samuelis. Plot ends as animal vacates
its shell at 28.0°0.
O
Figure 4. Water Lose of Animals In Shells on Exposure to Sun
4) P. samuelis B) P.granosimanus Arrows indicate interval of
time during which animals died.
Figure 5. Water Lose of Animals Out of Shell on Exposure to Sun
A) P.samuelis B) P.granosimanus Arrows indicate interval of
time during which animals died.
Bibliography
Bollay, Mellody, (1964), Distribution and Utilization of Grastropod
Shells by the Hermit Grabs Pagurus samuelis, Pagurus granosimanus,
and Pagurus hirsutiusculus at Pacifle Grove, Callfornia, The veliger
6; Supplement: 71-76, o text figures, 1 table.
Pearse, A.S., (1926), Observation on Certain Littoral and Terrestrial
Animals at Tortugas Florida, with special Reference to Migrations
from Marine to Terrestrial Habitats, Papers from the
Tortugas Lab.
of Carnegie Institute of Wash., Volume XXVI: 205-225.
Prosser, O. Ladd, ed., (1950), Comparative Animal Physiology, 888pp.,
ed.1, W.B. Sanders Company, Philadelphia, London, p 24.
Schmitt, Waldo L., (1921), Marine Decapods of California.
Univ.
Calif. Publ. 200l., 23: 1-470; plts; 1-50; 165 text figs.