252 (2000) 151–157
www.elsevier.nl / locate / jembe
Decision of the asteroid Leptasterias polaris to abandon its
prey when confronted with its predator, the asteroid Asterias
vulgaris
* ´
Stephane Morissette, John H. Himmelman
´ ´
Departement de Biologie and GIROQ, Universite Laval, Quebec, Quebec, Canada G1K 7P4
Received 15 January 2000; received in revised form 26 February 2000; accepted 25 May 2000
Abstract
We examined the decision of the asteroid Leptasterias polaris to abandon its bivalve prey Spisula polynyma when confronted with the predatory asteroid Asterias vulgaris. The frequency of abandonment decreased as the proportion of prey ingested increased from 10 to 40%, and increased with further increases in proportion of prey ingested. The escape response of L. polaris appears to be a compromise between risk of predation and benefits of feeding. This increase in the frequency of abandonment as the prey became further consumed beyond 30–40% likely represented decreasing benefits in the face of continued risk of predation. We propose that the initial decrease in fleeing as the proportion of prey eaten increased from 10 to 40% was because of a decreasing costs in subduing the prey, as there was a coincident decrease in the proportion of prey which were still capable of holding their valves closed. 2000 Elsevier Science B.V. All rights reserved.
Keywords: Leptasterias polaris; Decision making; Escape response; Predation risk; Kleptoparasite; Asterias
vulgaris; Subtidal; Gulf of St. Lawrence
1. Introduction
Predation is a major evolutionary force as it directly limits population growth and indirectly affects prey morphology, physiology, chemistry and behaviour (Sih, 1987). Although most of these modifications occur over evolutionary time, many animals also modify their behaviour in response to changes of predation risk during their lifetime
*Corresponding author. Present address: Department of Biological Sciences, University of Wollongong, Wollongong, NSW 2522, Australia. Tel.: 161-2-422-5871; fax: 161-2-4221-4135.
E-mail address: john.himmelman@bio.ulaval.ca (J.H. Himmelman).
(Lima and Dill, 1990). Such decisions reflect compromises between predation risk and benefits from activities such as feeding, social behaviour and reproduction (Lima and Dill, 1990).
The asteroid Leptasterias polaris is the most important predator of subtidal com-munities of the northern Gulf of St. Lawrence, eastern Canada (Himmelman and Dutil, 1991), and is one of the few carnivores capable of extracting abundant bivalve resources from soft bottom communities. Numerous other carnivores have developed a klep-toparasitic relationship with L. polaris (stealing its food), and their activities reduce the mass consumed by L. polaris by at least 10.4% (Morissette and Himmelman, 2000).
Asterias vulgaris is the major kleptoparasite and often causes L. polaris to abandon its prey (Morissette and Himmelman, 2000). The behaviour of A. vulgaris is aggressive and its attacks on L. polaris can be sublethal (arm autotomy) or lethal (Dutil, 1988; Himmelman, 1991). The response of L. polaris to attacks from A. vulgaris is usually rapid and often involves combat, however, the outcome of this interaction is variable (Morissette and Himmelman, 2000). Sometimes L. polaris succeeds in keeping its prey and A. vulgaris leaves and in other instances, L. polaris tolerates contact with A.
vulgaris and the two feed together. The most frequent situation is that L. polaris, within hours (and sometimes minutes), abandons its prey to A. vulgaris. The varying behavioural responses of L. polaris suggest that it likely takes into account potential energetic gains and predation risk in its decision making processes.
The present study examines decision making by feeding L. polaris when confronted with attacks by A. vulgaris. We quantify the escape response of L. polaris in relation to how much of the prey has been ingested. We predicted that the decision of L. polaris to abandon its prey to A. vulgaris will increase as potential gains, the amount of prey remaining, decrease. In a previous study of kleptoparasitism, we observed that the intervention of kleptoparasites in feeding bouts of L. polaris on the large bivalve Spisula
polynyma caused a decrease in predation time by 60% (Morissette and Himmelman,
2000). This far exceeded the relative decrease in mass of prey tissues lost due to the intervention of parasites (20%) and suggests that the rate of ingestion of L. polaris is more rapid in the early part of feeding bouts and decreases later on. This supports our premise that the costs of abandoning the prey decrease in the latter part of feeding bouts.
2. Methods
ˆ We conducted our study using SCUBA in the subtidal zone at Cap du Corbeau, Ile du Havre in Mingan Islands, northern Gulf of St. Lawrence (508149N; 638359W). We started each trial by providing a large L. polaris (30–50 cm in diameter) in a 50340320 cm cage at 12 m in depth, with a previously weighed 12–14 cm long
Spisula polynyma (Bivalvia). The cages were made of galvanized screening that
prevented entry of other carnivores. After a period of 48, 72 or 96 h, we placed one A.
vulgaris (20–30 cm in diameter) on top of the feeding L. polaris. Then we dived at
starved in a cage for a period of 14 days prior to the trials to increase feeding motivation. We conducted a total of 75 trials during July and August 1994, 25 for each delay period (48, 72 and 96 h) prior to adding the A. vulgaris. The different delay periods ensured that we obtained a broad range of values for the proportion of prey ingested as an initial condition prior to adding A. vulgaris. We determined the total mass (tissues plus shell) of the S. polynyma immersed in seawater in the laboratory just before and at the end of each trial. Also, we determined the underwater mass of the shell at the end of each trial. We then subtracted shell mass from the initial and final total mass to obtain initial and final tissue mass. These values allowed us calculate the proportion of prey tissue ingested. Measuring mass underwater provided a non-intrusive measure of mass and further eliminated error due to to variations in water content (Yamaguchi, 1974). To control for responses of L. polaris which might have been caused by our manipulations, we also ran 19 trials using A. vulgaris mimics made from PVC.
We grouped the trials into five classes according to the proportion of prey ingested. Since the expected values in two of the classes were ,1, it would be inappropriate to
2
compare classes using a x test (Zar, 1984). Thus, we used a randomization procedure (Manly, 1991) to test whether the frequency of abandonment varied among the different
2
classes. We calculated a x value for the observed distribution and compared it to the
2
frequency distribution of x values obtained by randomly assigning observed values to
2
the five classes 10 000 times. Thus, we obtained the exact probability of obtaining ax
equivalent to or greater than that observed.
3. Results
No escape responses were observed in the control trials, where mimic asteroids were introduced into cages in which L. polaris were feeding. In contrast, A. vulgaris always provoked a rapid reaction from L. polaris in the experimental trials (Fig. 1). The frequency at which L. polaris abandoned its prey varied depending on the proportion of
2
the prey which had been digested (x 511.32, P50.022). However, the pattern was not an increase with the increase in the proportion of prey ingested, as we postulated. Rather, the frequency of abandonment decreased as the proportion of prey ingested increased from 10 to 40%, and increased with further increases in proportion of prey ingested (Fig. 2). Observations of prey as they were abandoned by L. polaris showed that in 61% of the trials where ,40% of the prey was ingested, the adductor muscles of the bivalve were still maintaining the valves together (Fig. 2).
4. Discussion
Fig. 1. A typical sequence of events when a feeding asteroid, Leptasterias polaris, is confronted with the predatory asteroid, Asterias vulgaris. (A) A few minutes after A. vulgaris has been placed on the feeding L.
polaris, (B) 45 min later when A. vulgaris has extruded its stomach and is trying to gain access to the prey (at
this point L. polaris almost always has one arm extended, likely to facilitate fleeing), (C) at 90 min, when A.
vulgaris has gained access to the prey, and (D) at 135 min, when L. polaris is abandoning its prey.
1976; Barnard, 1984). In our study, the cost of extracting the prey from the sediment was not considered, as we provided prey to L. polaris. Nevertheless, L. polaris had to expend time and energy to force the valves of S. polynyma open so that the tissues could be digested. This handling cost could be important. For example, in one instance in our study, this step took 96 h. As L. polaris appeared to first digest the mantle, the adductor muscles could remain functional until up to 20–40% of tissues mass was digested.
Fig. 2. Frequency of prey abandonment by the asteroid Leptasterias polaris when confronted with the predatory asteroid Asterias vulgaris (above), and proportion of abandoned prey which still held their valves together at the end of the 3-h experiment (below), in relation to proportion of tissue mass of the prey (the bivalve Spisula polynyma) that has been ingested. Values above bars indicate the number of trials.
the potential food gains for L. polaris are high. Thus, the immediate availability of food should decrease its tendency to flee. Our observations of the condition of the bivalves, indicated that at the point where 30–40% of the prey was ingested, the valves were usually gaping. At this point, the manipulation costs are small, yet there is a large amount of tissue to be ingested. This was where L. polaris most frequently decided not to flee (Fig. 2). The increased frequency in the decision to flee with decreasing remaining prey mass, likely represents decreasing benefits in the face of continued risk of predation as we predicted. We propose that the initial decrease in the frequency of fleeing going from 10–20 to 30–40% of prey ingested was because of the decreasing energetic costs required to subdue the prey. In the 10–20% class, 75% of the S.
polynyma still held their valves closed. Thus, the escape response of L. polaris seems to be a compromise between risk of predation and benefits of feeding.
autotomize arms (Mauzey et al., 1968; Birkeland et al., 1982), however, this response is costly because of the resources needed to regenerate lost structures and the decreased ability to obtain resources (Lawrence and Larrain, 1994). Attacks of A. vulgaris on L.
polaris generally result in fleeing and a loss of food resources, rather than sublethal (arm
autotomy) or lethal predation, because the response of A. vulgaris once L. polaris abandons its prey is usually to feed on the prey remains (Morissette and Himmelman, 2000). The ability of L. polaris to continue feeding if attacks by A. vulgaris do not represent an immediate risk of injury or death likely decreases the potential impact of A.
vulgaris.
Our study explored interactions between large L. polaris and A. vulgaris that had be starved for a 14-day period and provided with a large bivalve as a prey item. Further investigations are needed to explore the influence of such factors as relative size of the two asteroids, their state of hunger, and the species and size of the prey item on the decision of L. polaris to abandon its prey when confronted with predation risk.
Acknowledgements
´
We would like to thank M.C. Giasson, P. Girard and F. Tetreault for their assistance in the field and D.J. Arsenault, J. Bovet and J.N. McNeil for critical comments on the manuscript. We are further indebted D.J. Arsenault for aiding us in applying the randomization procedure. The senior author received support from a FCAR scholarship and the research was funded by NSERC and FCAR grants to J.H.H. [SS]
References
Barnard, C.J., 1984. The evolution of food-scrounging strategies within and between species. In: Barnard, C.J. (Ed.), Producers and Scroungers: Strategies of Exploitation and Parasitism. Chapman & Hall, New York, pp. 95–126.
Birkeland, C., Dayton, P.K., Engstrom, N.A., 1982. A stable system of predation on a holothurian by four asteroids and their top predator. Aust. Mus. Mem. 16, 175–189.
Curio, E., 1976. The Ethology of Predation. Springer, Berlin.
´ ´
Dutil, C., 1988. Partage des ressources alimentaires et comportement de predation des etoiles de mer de la
´ ´
communaute infralittorale du golfe du Saint-Laurent. MSc thesis, Universite Laval.
Himmelman, J.H., 1991. Diving observations of subtidal communities in the northern Gulf of St. Lawrence. In: Therriault, J.-C. (Ed.), The Gulf of St. Lawrence: Small Ocean or Big Estuary. Can. Spec. Publ. Fish. Aquat. Sci, Vol. 113, pp. 319–332.
Himmelman, J.H., Dutil, C., 1991. Distribution, population structure and feeding of subtidal seastars in the northern Gulf of St. Lawrence. Mar. Ecol. Prog. Ser. 76, 61–72.
Lawrence, J.M., Larrain, A., 1994. The cost of arm autotomy in the starfish Stichaster striatus. Mar. Ecol. Prog. Ser. 109, 311–313.
Lima, S.L., Dill, L.M., 1990. Behavioral decision made under the risk of predation: a review and prospectus. Can. J. Zool. 68, 619–640.
Manly, B.F.J., 1991. Randomization and Monte Carlo Methods in Biology. Chapman and Hall, London. Mauzey, K.P., Birkeland, C., Dayton, P.K., 1968. Feeding behavior of asteroids and escape responses of their
Morissette, S., Himmelman, J.H., 2000. Subtidal food thieves: interactions of four invertebrate kleptoparasites with the seastar Leptasterias polaris. Anim. Behav. (in press).
Sih, A., 1987. Predators and prey lifestyles: an evolutionary and ecological overview. In: Kerfoot, W.C., Sih, A. (Eds.), Predation: Direct and Indirect Impacts on Aquatic Communities. University Press of New England, London, pp. 203–224.
