Foraging in Conus Californicus ABSTRACT Conus californicus is the only temperate species of a large genus of predatory prosobranch gastropods. We studied their foraging behavior and diurnal patterns of activity in relation to fullness of the gut. We found 29% of 59 C. californicus obtained from the field had prey in their gut. Of these, the most common prey items were the tube-dwelling polychaetes Sabella crassicornis and Diopatra ornata. In the laboratory, C. californicus digested a full gut of S. crassicornis in 44 to 60 hours. We also found that C. californicus are more active at night than during daylight. In a laboratory experiment, C. californicus with full guts were significantly less active than unfed individuals for two consecutive nights after feeding (P S.01). During the third and fourth nights, activity increased and there were no significant differences between fed and unfed individuals (P2.44). Throughout the experiment, there were never any differences between fed and unfed individuals during daylight hours (p = 0.51). These results suggest that activity of C. californicus is influenced strongly by prey consumption: immediately after feeding, nocturnal activity decreases significantly. After an interval adequate for complete digestion, nocturnal activity resumes rapidly. Foraging in Conus Californicus Introduction The predatory cone snail is a marine neogastropod that occurs widely in tropical waters. Conus is considered to be a successful genus (Olivera 1997) with over 500 species. Conus californicus is the only species found in temperate waters. Cone snails produce potent venoms for purposes such as prey capture and defense (Olivera 1997). Few papers describe behavior and feeding habits of C. californicus, (Saunders and Wolfson 1961) but these studies were in small laboratory aquaria at 21°C which do not represent the colder environment of the snails found in central California at Hopkins Marine Life Refuge. The range of C. californicus extends from the north in the Farallon Islands (San Francisco) to as far south as Bahia Magdalena (Baja California) (Morris et al. 1980). The cone snails used for this study were found in D. ornata beds at Hopkins Marine Life Refuge in the subtidal at -10m depth. D. ornata is a tube-dwelling polychaete that secretes a parchment-like tube, often decorated with red algae and shell debris. D. ornata beds create a refuge habitat (Woodin 1978) for a wide variety of animals such as other polychaete worms, gastropods, and echinoderms. The feeding mechanism of C. californicus is similar to its tropical relatives. Cone snails possess a highly modified radula whose teeth are poison harpoons used to subdue prey. This radula is connected to a venom apparatus that consists of a venom duct which produces the venom, and a venom bulb (Bingham et al. 1996). The bulb ejects the radula tooth and venom through the tip of the proboscis into the prey item (Saunders and Wolfson 1961). The exact firing mechanism is not known. Foraging in Conus Californicus C. californicus waves its siphon slowly from side to side as it moves, and it can sense prey by smell using the osphradium. C. californicus begins to extend its proboscis when it senses prey in the vicinity. The proboscis can extend the length of the snail. After encountering prey, the snail gently touches the edge with the proboscis until the cone decides to harpoon the prey with a radula tooth. A small release of venom can be seen as the live prey is drawn into the buccal cavity. Many studies have examined the toxicology and ecology of tropical species (Olivera 1999, Kohn 1959). C. californicus is the only temperate species of Conus despite the large diversity in the tropics. Our study provides a source of information for a species whose diet and behavior in Monterey is not well-documented. The main focus of this study was to examine foraging behavior, digestion rates, and prey items that occur in the snail's natural environment. We have studied the gut contents of specimens collected from the field and examined cone snail foraging behavior under laboratory conditions. MATERIALS AND METHODS Diet and Sediment Study We collected 59 cone snails for gut analysis in the Hopkins Marine Life Refuge over a 6 week period. We took a core sample of the sediment below each collected snail. Approximate dimensions of the sample are 1Ocm deep with a 157cm' surface area. Snails were dissected within 5 hours of collection. Individuals were relaxed in MgCl, for 15 -30 minutes. The snail shell was gently cracked using pliers and separated from the soft parts. Dissection of the soft parts involved an incision at the mouth. We separated the proboscis and the gut from the remainder of the body in order to examine any prey parts present Foraging in Conus Californicus in it. Using a compound microscope we looked at scrapings from fore, mid and hind gut, and identified larger prey fragments (Hartman 1969, Morris et. al 1980, Smith and Carlton 1975). Core sediment samples were analyzed by counting numbers of each organism found after sieving through a 3x3 mm screen. Digestion Time C. californicus individuals were left in a tank at ambient temperature with no sand for 48 hours. S. crassicornis were removed from their tubes and were fed to C. californicus. The snails ate known masses of worms until they were satiated. Preliminary dissections showed that after 30 hours negligible digestion had occurred. In light of this, snails were dissected starting at 32 hours until 52 hours, at four hour intervals. Undigested portions of the worm remaining in the gut were removed and wet weight was taken. The undigested portions were dried and weighed. A regression line of amount digested over time was used to estimate total digestion time. Activity Patterns and Foraging Behavior We examined the relationship between activity patterns and feeding behavior in three laboratory experiments in two outdoor tanks: an intensive 24-hour study of activity patterns, and two comparisons of fed and unfed snails (over 48-hours and 96-hours). The tanks were 2.24 mx 0.96 m with water depths between 0.34 m and 0.38 m. A layer 3-5 cm thick of fine- grained beach sand was placed at the bottom of each tank. Seawater circulated through the tanks providing ambient temperature (10-12°C). String grid lines attached over the tank created a 11x5 matrix of squares 20 cm x 20 cm. Foraging in Conus Californicus Because the snails bury themselves in the sand during the day, we tagged the snails using individually numbered flags attached with epoxy to the upper portion of their shell. The snails were then left for 48 hours in the tanks to settle into their normal, unfed activity pattern. Effect of tags on snail behavior was not tested but appeared to be negligible. To determine patterns of diurnal and nocturnal activity, we made preliminary hourly observations for 24 hours of nine unfed snails in one tank. We recorded the snail's position in the 1 1x5 matrix and if the snail was buried, above the sand, moving, or on the tank walls. For the 48-hour experiment, we placed 18 snails in each of two tanks and assigned them randomly to experimental, procedural control, and control groups. The snails were left in the tanks for 48 hours in order to settle into normal activity patterns. These snails were then monitored every hour for a 24-hour pre-feeding period and a 24-hour post-feeding period. An experimental group of 10 snails were fed the polychaete Eunice valens until they no longer accepted further food. A procedural control group of 10 snails was stimulated with E. valens until they extended their proboscides, but they were not allowed to eat. A third control group of 14 snails, left untouched and unstimulated, never extended their proboscides or were roused above the sand. Distance that each snail moved was estimated as the number of grid squares crossed by that snail per hour, averaged over each time category (day, night, dusk, dawn). To examine activity over 96 hours, we tagged and placed 18 newly captured snails in each outdoor tank. Procedures described above were repeated except snails were fed and stimulated with S. crassicornis. There were 11 experimentals, 10 procedural controls, and 15 controls, and we did not record a pre-feeding observation. Data from the 48-hour experiment were statistically analyzed using data as though we had only taken readings every three hours during the day and night. This analysis showed the same significant statistical results. Foraging in Conus Californicus Therefore, we recorded snail position and activity on the grid at hourly intervals during dawn (05:00-07:00) and dusk (19:00-21:00) and at 3 hour intervals during night and day. Snail activity was observed for four days after the feeding. Statistical Analysis Movement data were analyzed using a 3-way analysis of variance (ANOVA) with day (4 levels), time (4 levels), and treatment (3 levels) as fixed factors. All tests were conducted at the 0.05 level. Before analysis, homogeneity of variances was determined using Cochran’s test. Data were transformed using square roots, which produced a non-significant Cochran's Test. Two sets of planned comparisons were of interest. The procedural controls (stimulated but not fed) were first compared to the controls (undisturbed). If this comparison was not significant, the procedural control and control were pooled together and tested against the experimentals (fed). These comparisons were conducted on the daytime and nighttime data separately. Sample sizes were unbalanced with 15 controls, 10 procedural controls, and I1 experimentals. RESULTS General Characteristics Of the snails dissected for the gut content analysis, average shell length was 2.86 cm and average width was 1.57 cm. The largest individual dissected was 3.7 cm and the smallest was 1.6 cm long. Sex ratio of dissected individuals was 19 males to 26 females, and 10 were unknown due to lack of gonad development. Foraging in Conus Californicus General Observations Symbionts of C. californicus The peacrab Opisthopus transversus is a commensal of C. californicus (Wolfson 1974). Surprisingly, we also found a peacrab that had been eaten by Conus in the field. Because the crab was still alive, we fed it to another Conus which readily ate it. Three of the 58 Conus that were dissected had a commensal O. transversus in their mantle cavities. The size of these commensal crabs was -2mm and size of the crab that had been eaten was -Icm. Field observations The apparent abundance of C. californicus varied from day to day. We saw the snails both individually and in groups of 3 or more. On different days, C. californicus could be very abundant, less abundant, and scarce in the same D. ornata beds. When they were abundant, there were groups of at least four Conus touching each other, and other individuals were nearby within a small area of 1-2m’. We observed the seastar Pisaster brevispinus eating C. californicus. We also observed a single C. californicus sitting on the aperture of a D. ornata tube with its proboscis extended. This is the only case that a Conus was observed in the field with its proboscis extended. Diet and Digestion Rate The most commonly consumed organisms were the tube-dwelling polychaetes S. crassicornis and D. ornata (Fig 1). These are also the two most abundant organisms in the habitat. Other organisms found in the gut were an unidentified terebellid, Opisthopus transversus (pea crab), and unidentifiable polychaete parts. Of the 59 cone snails sampled, 17 were found with a full gut. Partially digested D. ornata were covered with a white milky Foraging in Conus Californicus coating. Some less digested sabellids were still undergoing muscle contractions upon being removed from the esophagus of C. californicus. In the laboratory, Conus ate Olivella biplicata only after several weeks of starvation. It showed no interest in Nassarius mendicus. We also made laboratory observations of Conus eating the following polychaetes: Pherusa capulata (from the anterior end), Pareurythoe californica, Eunice valens, Glycera sp. The snails ate these prey items in a manner similar to previous observations. D. ornata and Glycera sp. used their jaws and extensible proboscis, respectively, to defend themselves from the Conus snails. Foraging Behavior and Digestion Rates The regression line of percent digested over time (Fig 2) suggests that 60 hours is the maximum time for snails to digest a full gut of S. crassicornis. Despite low correlation (r - 0.228), we can be assured that after 60 hours a known amount of sabellid will be fully digested by the snail. Initial 24-hour study in the laboratory showed that C. californicus are above the surface of the sediment and moving at night, but are sedentary and often completely buried during the day. During the 24 hour pre-feeding period of the 48-hour experiment there were no significant differences among experimental, control and procedural control groups before they were fed or stimulated (Fig. 3). Average night activity was high (0.87 sq-h"). Snails were inactive during the day (0.16 sq-h). During daylight hours we have observed that the cone snail will usually bury itself under the sand. If a prey item is in close proximity the cone snail will be aroused and will rise from the sand. In other cases, the snail will also extend its proboscis from beneath Foraging in Conus Californicus the sand. If the snail does rise from the sand it will begin to probe the prey item. The snail will then fire its radula into the prey, and subsequently ingest the prey whole. Once the experimental snails fed, they moved less on the following night (0.24 sq-h") than unfed snails (0.905 sq-h"). Daytime movements were not significantly different between any of the treated groups of snails throughout the 48-hour experiment. The results of the 48-hour experiment led us to hypothesize that C. californicus are active at night because they are foraging. To test this, we conducted the 96-hour experiment to determine whether the fed snails would resume nocturnal activity after a period sufficient to complete digestion. On the first and second nights after feeding, the fed snails moved significantly less than the pooled control and stimulated animals (P £ 0.01, Table 1). Stimulated and control groups were not different on any of the four nights (P 0.21). By the third night, the fed snails became more active, and they were no longer significantly different from the control or stimulated snails (p = 0.44, Table 1, Fig. 4). The start of the third night corresponds to 53 hours after feeding. Our digestion time course study suggests that by this time, most fed snails should have completed digestion. This increased level of activity continued on night 4 (P-0.84, Table 1) and confirmed the pattern that Conus are actively foraging when their guts are empty. Day activity among stimulated, fed and control groups were never statistically significantly different (p +0.43) (Table 1). Foraging in Conus Californicus DISCUSSION Diet and Digestion The finding that C. californicus eats mostly D. ornata and S. crassicornis contrasts with studies on C. californicus in southern California. Our study shows a more narrow range of prey choice. Other studies have shown C. californicus to eat not only polychaetes and crustaceans but also gastropods, bivalves, cephalopods and fish (Kohn 1966). These previous studies were carried out at southern California ambient temperatures which are up to 10°C warmer than in Monterey Bay. This could lead to more active snails, which may explain more active hunting. We did not encounter most of the prey species studied by Saunders and Wolfson (1961). However in the case of O. biplicata, Saunders and Wolfson (1961) and Kohn (1966) report that C. californicus readily eat the gastropod, whereas central California C. californicus may eat O. biplicata, but not as readily. Since 29% of the snails collected had full guts, this indicates that at a given time, 29% of snails have recently eaten. Snails collected for this study were mainly on top of the sand and were collected during the day. An area of further study would be to examine snails collected at night. They may show different levels of full guts as well as different prey items. Also there might be different densities of the animals above the surface. It is still unclear how C. californicus extracts the worms from their tubes. In the field, an individual C. californicus was sometimes found half buried with its posterior pointing vertically from the sand. In the laboratory, after placing S. crassicornis worms near a buried snail, it would capture and consume the worm with its proboscis without emerging from the sand. We also observed C. californicus moving under the sand without rising. These snails Foraging in Conus Californicus may be foraging for tube worms beneath the sand. Snails may maintain their preferred daytime buried position, but feed at every opportunity. Fragmented portions of D. ornata found in the gut were identifiable by distinctive parapodia and setae. One of the fragments had a developing anterior end with tentacles. Development was similar to a collected D. ornata tube containing a fully developed worm and a second regenerating worm. An area of further investigation is whether Conus has a food preference for immature or regenerating D. ornata. This may also reveal how Conus capture D. ornata. The digestion rate study gave a regression line with low r and high variability. Possible reasons for this poor fit is our use of highly variable wet weights to estimate percent digestion. We consistently found that after 44 hours a high percent of the meal had been digested. One outlier shows that an animal may not have been coping with laboratory conditions. Activity Patterns Because the activity of the snails drops significantly when they have eaten a full meal and rises once digestion has occurred, it is clear that their night activity is associated with foraging. During dawn there is variable but high level of activity, which may be associated with the night to day transition of snail activity. Further field investigations would include how abundance of cone snails varies with different densities of D. ornata. Dense D. ornata may be preferred habitat. Another extension of the foraging behavior study could examine how foraging behavior is affected by mating aggregations that occur during the spring and summer. Foraging in Conus Californicus CONCLUSION Study of foraging behavior and diurnal patterns of C. californicus revealed that of snails captured during the day 29% were feeding at any given time. The most common prey items both in the sediment near by the C. californicus found, and in the dissected gut are the tube- dwelling polychaetes S. crassicornis, and D. ornata. Digestion of a full gut of S. crassicornis takes between 44 and 60 hours. C. californicus forage during the night when they have high levels of activity. Day activity is consistently low for C. californicus that have fed, and those that have not (p=.51). After ingesting a full gut of S. crassicornis, C. californicus become less active until they have finished digesting (pS.01), at which time nocturnal activity returns to levels of snails that have not eaten (p».44). ACKNOWLEDGEMENTS We would like to give our deepest thanks to James Watanabe for his advising, teaching, diving, and enthusiasm at all times over the course of our project. Thanks also to Freya Sommer for diving and identification of invertebrates, and to Professor William Gilly, Joe Wible, Chris Patton and the Spring 2000 class of Hopkins Marine Station. Foraging in Conus Californicus LITERATURE CITED Bingham, J.P., A. Jones, R.R. ., P.R. Andrews., and P.F. Alewood. 1996. Conus venom peptides (conopeptides): inter-species, intra-species and within individual variation revealed by ionspray mass spectrometry. pp. 13-27 in P. Lazarovici, M.E. Spira, E. Zlotkin, eds. Biochemical Aspects of Marine Pharmacology. The Interuniversity Institute for Marine Sciences, Eilat (IUI). Carlton, J T, R.I. Smith. 1975. Light's manual: Intertidal invertebrates of the central California Coast. Third ed. University of California Press, Berkeley, California. Kohn, A. J. 1959. Ecology of Conus in Hawaii.. Ecological Monographs 29: 47-90. 1966. Food specialization in Conus in Hawaii and California. Ecology 47: 1041- 1043. Hartman, O. 1969. Atlas of polychaetes. Allan Hancock Foundation, University of Southern California, Los Angeles. Morris, RH., D.P. Abbot, E.C. Haderlie, 1980. Intertidal invertebrates of California, Stanford University Press, Stanford, California. Olivera, B. 1997. Conus venom peptides, receptor and ion channel targets, and drug design: 50 million years of neuropharmacology. Molecular Biology of the Cell 8:2101-2109. Saunders, P R, Wolfson, F. 1961. Food and feeding behavior in Conus californicus. The Veliger 3: 73-76. Wolfson, F H. 1974. Two symbioses of Conus with brachyuran crabs. Veliger 16: 427-429. Woodin, Sarah Ann. 1978. Refuges, disturbance, and community structure: A marine soft- bottom example. Ecology, 59: 274-284. Foraging in Conus Californicus Table 1. Three-way ANOVA of distance moved data over 96 hours. (TREATMENT fixed, TIME fixed, DAY fixed). Analysis of Variance Sum-of-Squares DF Mean-Square F-Ratio P Source 5.797 2.899 12.259 0.000 TREATMENT 0.000 11.400 3.800 16.071 TIME 0.069 0.833 0.290 0.206 DAY 0.254 1.075 0.376 1.525 TREATMENTXTIME TREATMENTXCYCLE 2.193 0.161 0.366 1.546 4.159 MEXCYCLE 1.954 0.043 0.462 2.624 0.146 0.617 0.888 TREATMENTXTIMEXCYCLE 124.844 0.236 528 Error Cochran's test = 0.06 Night Procedural Control vs Night Control DAY DAY DAY2 DAY 3 2.428 0.169 1.163 0.017 F-value 0.897 0.681 0.120 0.281 Night Experimental vs Night (Control+Procedural Control) DAY4 DAY: DAYI DAY2 0.039 0.598 6.243 F-value: 6.841 0.009 0.843 0.440 0.013 Day Control vs Day Procedural control DAY4 DAY 3 DAY2 DAYI 0.592 0.511 0.544 F-value: 0.032 0.858 0.442 0.475 0.461 Day Experimental vs Day (Control + Procedural Control) DAY4 DAY 3 DAY DAY2 0.432 F-value: 0.070 0.286 0.622 0.593 0.512 0.791 0.431 Foraging in Conus Californicus FIGURE LEGENDS Fig. 1. Frequency of each prey item in the gut of C. californicus and frequency of organisms in surrounding sediments. Organisms included in the graph had frequency n2 3. Other groups of organisms found but not shown in the graph included polychaetes in the families Polynoidae, Nephytydae, Phyllodocidae, Arabellidae, Flabelligeridae, also the Echinoderms: sea urchins, sand dollars, the Molluscs: Rictaxis punctocaelatus, eolid nudibranch, and Crustacean: caprellid. Fig. 2. C. californicus fed a known amount of S. crassicornis. Dissections at four hour intervals and wet weights of gut contents reveal the percent digested. Fig. 3. Unfed C. californicus allocated as experimental, control and procedural control groups. Monitored at hourly intervals. Dusk is from 19:00 - 21:00, and dawn is from 05:00- 07:00. Error bars are the standard errors of the means. Fig. 4. Activity of fed C. californicus during 96-hours after feeding. Monitored at three hour intervals during the day (07:00 - 18:00) and during night (21:00 - 05:00). Error bars are the standard errors of the means. Movement rates at dawn and dusk are not shown. E L 2 + 8 4 2 50 100 150 S 0 — L In Gut In Surrounding Sediment 2 S o Prey Species Fig. 1. Fig. 2. O Digestion Rates 100+ 30 y = 1.584x + 7.732 R2=0.228 60 40 20 40 30 50 60 Time since feeding (hrs) Fig. 3. 2 1.5 1.0 0.5 OO 1.5 1.0 0.5 OO 2.0 1.5 1.0 05 O0 Experlmental — Control — Pocedral Control dawn day dusk night TIME of DAV Fig. 4. 2 1 1.0- 0.8 06 0.4 02 DO 1.0 0.8 O6 0.4 0.2 OO 1.0 08: O6 0.4 0.2 OO H NICHT Fed □ DAV Control L 1 — Stimuated DAV