SOME E CTORS OF MORTALITY IN LABORATORY POPUIA TIONS OF PAGURUS SAMUELIS STIMPSON (ARTHROPODA, DECAPODA) by Norman Richardson INTRODUCTION Factors influencing the mortality of individual isolated organisms are often relatively simple to illucidate compared to those determining the mortality of a population of interacting individuals. Calhoun (1957, 1962a, 1962b) has amply pointed out the fact that the elements of interaction are not simple but highly complex. Physical as well as psyckological room is often demanded. The complexities of interactions, behaviors, and dominance hierarchies compound the factors at interplay in social animals. Through investigationof some of the latter mentioned factors, the study of the factors influencing mortality in laboratory populations became attractive. Preliminary observations made on populations kept for behavioral studies suggested that the factors of space, availability of shells or food, time, and behavioral interactions might be involved in influencing the rate of mortal- ity in these populations. Behavioral factors, considered in another paper completed during the same period of observation, will not be here considered. A. AVAILABILITY OF SPACE Methods.and results. Aquaria measuring Lx 2t cm. in total floor area were util ized. Movable partitions of wood covered with polyethylene to prevent climbing of the wood by the crabs allowed variance in the floor area. Both running sea water and constant aeration were employed. Crabs collected from the intertidal zones off Hopkins Marine Station, Pacific Grove, California, and Point Pinos, Pacific Grove, were sorted into size classes by measurement laterally across the cephalothorax. Preliminary experiments involving 100 crabs of a size class 2.0 cm plus or minus 0.1 cm. in aquaria L4 x 2to cm. were set up. Considerable aggressive behavior is shown by the animals during the first 1-6 hours of interaction. Subsequently the crabs tend to pile up in the corners of the aquaria, and in no way do thegutilize the whole of the space available to them. A partition iin the aquarium was employed to vary the area. These preliminary experiments indicated that an area 20 x 20 cm. in total floor area could support a maximum of 25 crabs for a minimum of 5 days with no greater than 204 mortality (see experiment D.). This space, arbitrarily designated as 1004 area, was taken as a base for further experiments. In experiments involving 25 crabs this 1006 area was increased, and decreased in a reciprocal experiment, while the ratio of crabs to shells, total number of individuals, and the absence of food were constant. Tank adapted individuals of five days were added to restore the tot al number when mortalities occurred. Fluctuations of 20% of the 100% area were made in the two reciprocal experimemts at 24 hour intervals. The maximum area available to population A was 180% of the 100% 20 x 20 cm. area. The minimum available to population B was 20% of the 100% area. Results are shown in Fig. 1 and table 1. Conclusions. Limitations of space appear to induce high mortality when available area falls below 50% of the 100% pre-established area. When the ratio of crabs to area approaches this 50% value, inter- actions increase. Death is usually by murder of one individual by another of larger size: an individual is approached by another, extracted from its shell, and severed in two between cephalothorax and abdomin. To buffer interactions imposed by this high ratio of crabs to space, double this minimum 50%, or the originally established "minimum" of 100% area (20 x 20 cm/ 25 crabs) will be used as the area for further experiments. B. AVAILABILITY OF SHELLS Methods and results. Aquaria measuring l x 26 cm. were divided into two areas as described above with polyethylene covered partitions. Crabs were sorted into a size class of 1.5 plus or minus 0.1 cm. as before, and Tagula funebralis shells which had been occupied by crabs in the field were cleared of their owners and sorted into a size class of 2.0 plus or minus O.1 cm. measuring the shell at its greatest diameter. Two experiments were completed. In the 220 first experiment the initial ratio of crabs to shells was set at 1:2 . Twelve shelled individuals were placed with 12 additional empty shells in the 100% area. At 21 hour intervals mortalities were countedand new tank-adapted individuals were added for those lost. Area was kept constant, as was the ratio of crabs to shells. At each of these intervals 50% of the unoccupied shells were removed. The results are shown in Figure 2. In the second experiment 25 housed animals were allowed to remain undisturbed in a 100% area for five days. At the end of this time interval no mortalities were recorded for this group. At succeeding 24 hour intervals 10% of the crabs were removed from their shells and returned to the tank. Mortalities were measured over the ten day period of the experiment. Area and the number of individuals remained constant as above. Results are shown in Fig. 3. Conclusions. Within this defined population mortalities do not greatly increase until the ratio of crabs to shells falls below 2:1. It is difficult to infer that this ratio might become an effective cause of death in natural populations in that Belknap (personal communication, Hopkins spring group) found a ratio of crabs to habtah Ashells of 1:l among intertidal and subtidal Pagurus. It is possible, however, that such field conditions might arise to make this ratio a factor in mortality. Such a feld ratio does sggest consdevable cempettion amon crabs for walable shels: each te a evab obaes shells, the pvolaloitty is hich that he wil 22 C. AVAILABILITY OF FOOD Methods and results. Welsh reports in Watermann (1961) that some decapod crus- tacea lave been starved for periods exceeding two months. Sixty five isolated individuals were kept in 50ml. beakers placed in aquaria with aeriation and running sea water for 30 days. See Fig. 1. Results and conclusions. During the 30 day period no more than 5 individuals represent- ing a mortality of 7.6% died. In the initial 15 day period of starvation no individuals were lost. As the experiments involving area, shells and food ran for no longer than two weeks, starvation appears to be a negligible factor influencing mortality rates in these studies. D. TIME AS A FA CTOR Methods and results. Two experiments were performed: in the first the individuals were isolated as in the above experiment. Twenty - five individuals were used. Within this group there was a loss of one individual in fourteen days for a mortality of 1%. See Fig. 5. In the second experiment twenty - five individuals were placed in the 100% area and allowed to remain for fourteen days. After five days five individuals had been lost for a mortality of 20%, but after this time there were no further losses for the remainder of the fourteen days. 222 Conclusions. This experiment, as does experiment C, serves mainly as a control on some of the variables present in A and B. It is here shown that the defined times here used are not factors of excessive mortality in these h boratory populations. E. GENERAL CONCLUSTONS Singular behavioral patterns and those of interacting individ- uals were not considered in this study. Some of the aspects of dominance are considered in papers by Reese (1962) and in ar companion paper by the author. Behavioral factors are undoubtedly one of the principle factors of mortality in lab as well as in natural populations. Pagurus samuelis is highly aggressive, and one is tempted to suggest from casual observations that murder is one of the chief operative mechanisms of population control. Even this is only the manifestation of behavior preceeding it. To attempt to illucidate some of the factors which may elicit such behavior has been the object of this study. SUMNA RY Factors influencing the mortality rates in laboratory populations of Pagurus samuelis Stimpson were examined. Among factors shown to be important are those involving space, avail- ability of shells, and behavioral interactions. Factors negligible to the mortality rates of these studies were starvation and time. ACKNOWLEDGEMENTS The author wishes to recognize the kind aid of Drs. L. R. Blinks . Abbott D. P. Abboco BIBLIOGRAPHY Calhoun, John B., "Social welfare as a variable in population dynamics", Cold Springs Harbor Symposia on Quantitative Biology 22: 339 - 359 (1957). Calhoun, John B., "Population density and social pathology", Scientific American 206: (1962). Calhoun, John B., "The Ecology and sociology of the norway rat", U.S. Department of Health, Education, and Welfare, Public Health Service, U.S. Government Printing Office, Washington, D.C., (1962). Reese, Ernst S., "Submissive posture as an adaptation to aggressive behavig in hermit crabs", Zeitschrift für Tierpsychologie 19 (6): 645 - 651 (1962). Reese, Ernst S., "Behavior mechanisms underlying shell selection by hermit crabs", Behavior 21 (1 - 2): (1963). Welsh, J. H., "Neurohumors and neurosecretions" in Physiology of Crustacea Vol. II. T. H. Waterman (ed.), Academic Press, New York (1961). C C + + 225 C C 2 .. 12-280 227 ++ H T IH +++ +++ + t — + + ++ +++ + + +t t + + — ++ 1 ++ +++ ++++ +++ + + + + —++ ++ ++ + + t +++ + ++ 1 ++ + ++++ +l t t + + + +++++ + t lt + — ++ lt ++ + ++ + — ++ +++ +++ +++ +++++ ttt + ++ ++ l t + ++++ Ppt t tt + + ++ + ttt + tt + + ++++ t +++ + t + + — + + p ++ + ++++ + P t +t ++ ++ + ttt t t pp ++++ + +++ ++ ++ ++++ t l + t l ++ + + m + + —+ + ++ + +++++ tt ++ ++ + + + + + ++ — — ++ + + + + ++ + ++ — + ++ +++ + + 10 Squares to the Inch C O - 223 12-280 C I + t +++ t ++ t ++ +++ ++++ + + ++ + ++ ++++ ++ ++++++ + ++++ ++ + + ++ ++++ + + + + + ++ + ++ HI + + +++ + t + + t ++ + ++ ++ + +++ ++++ + +++ + ++ + + t — + + + + +++ + p tt + t l + + — + +++ ++++ t +++ ++ — + + + ttt + l + l ++ ++ + ++ — — ttttt + + + + + ++ ++ ——— t + + + tt ltt + t l + ++ ++ + + + + + 1+ +++++ + + + +++ + ++ ++ + + ++ ++ tt + ++ tttttttt ++ + 1 l ++ + ++ tt — + ++ +++ + 10 Squares to the Inch 229 2 C 220