-2- ABSTRACT Prey selection and foraging patterns are two very im¬ portant behaviors for Leptasterias hexactis. In order to try to understand these behaviors more fully studies would have to be done in the lab where Leptasterias could be closely mon¬ itored and their environment manipulated. This should be done in the most natural circumstances and with the least amount of disturbance to the asteroids to preserve their usual patterns, Clay plates with different prey items to select from and different distributions of prey were suspended in a 80 1 tank. The plates were large enough, it was hoped, to represent an area of rock that would normally be occupied by the Leptasterias. The starfish were never handled but could be observed while foraging and selecting prey. A flat sea table was also used and again the animals were observed but not disturbed. Data obtained from these and smaller finger bowl exper¬ iments has revealed something about the frequency of eating, movement associated with the density of food in the environ¬ ment, and that Leptasterias might possibly recognize different prey densities and choose prey accordingly. It is strongly suspected that prey selection has a component of learning involved in the behavior. The Leptasterias also seem to op¬ timize food choices. Comparisons of calories per individual prey per hour to consume show that the prey with the highest caloric yield are chosen. This is dependant however on the environment, what prey are present and the relative sizes and densities of the prey. 7. Rad INTRODUCTION There are two ways a starfish could feed in its en¬ vironment. It could just feed randomly on the prey it happens to be near, or it could select prey according to certain cri¬ teria. Prey preferences have been documented for many starfish (Menge, 1972a; Feder, 1959). But obviously more than prefer¬ ence must be involved in the selection of a meal. The densities of different preferred prey, the spaceing of the preferred animals and times between contacts with the prey might all affect what is eventually chosen as the final prey, And looking at these parameters in detail might enlighten the question of how the animal selects its food. The more intriguing question, however, is can the Lep¬ tasterias learn to assimilate the data about densities, prey items and contact time in order to make decissions in each different environment about what will be the best prey to select for consumption? The behavior of selecting food is very complex and might have a component of learning about the terrain rather than just an instinctive or biochemical basis. These questions could be studied quite readily in Leptasterias. Leptasterias are small six armed starfish found readily in the lower intertidal zone around Monterey Bay. Because of their small size, they were the best choice for in lab experiments. They could be put into a controlled habitat that approximated in size their own natural habitat. Leptasterias are also food generalists and are found in many habitats with their diet varying according to surroundings. This was also an attribute needed for the experiments since an attempt to determine what prey would be chosen in different habitats was a key component of the preject. Finally, Leptasterias have no distant chemoreception (Menge, 1972) so their knowledge of their surroundings must come from contact. This makes it easier to determine what prey is actually recognized since contact is necessary, rather than trying to decide if the starfish sensed something. T Prag -4- Menge (1972, 1974) studying Leptasterias in the Puget Sound region has found that they eat most in late summer and least in the winter. This might be due to winter storms as well as immobilization of the females due to brooding. Lep tasterias also respond to light, feeding more at night. He has also analysed their diets, including prey eaten and calories of prey consumed. He states that prey is chosen selectively and varies with their habitat. The food eaten by Leptasterias around the Monterey Bay area was not studied, so their regular diet was not necessarily used for the following experiments. In fact the prey items were rather randomly chosen for their abundance and ease of attainment. Since the object of the experiments was only to try to find the criteria for different foods being selected any combination of foods that would be eaten was fine, as long as there were sessile and mo¬ bile foods and there was a definite preference for certain foods. The lack of knowledge of their regular prey proved no problem The question that originally sparked enthusiasm for a detailed look at foraging patterns and prey selection in Leptasterias was that of why a starfish, an animal with limited perceptual abilities, would often pass over numerous prey items while foraging when it seemingly had no assurance that it would find something better with further wandering. Is it possible that starfish are able to optimize prey selection? Does their behavior suggest that they can recognize different environments and prey densities and forage accordingly? If they can forage in a selective and optimal manner, what are the components of this behavior - learning, instinct or a combination of factors? 7 Peock 2. 3. METHODS The determination of what Leptasterias chooses to eat, when to eat and where to eat required a number of different experiments. Five different experimental procedures are described below. They were used to answer specific questions as well as yielding data that was eventually combined to show differences in patterns due to different environments. 1. Six small finger bowls, each 10 cm in diameter, were filled with between 15 and 25 potential prey items and one Leptasterias. The prey items included L. planaxis, C. digitalis, C. scabra, T. funebralis and M. californianus. The bowls were kept in running sea water and checked once a day. Any empty shell was noted for that day and re¬ moved. This set up was used to test food preferences. As the animals slowly ate their favorite foods, they were left with less and less desirable prey to chose from. The order, in which they ate the foods, should serve as an indicator of the preference for that food. Since all food was in such a small space, it was assumed to be equally detectable and thus food choices were by preference rather than chance. One small bowl, with four Leptasterias in it, was filled only with M. californianus. The question addressed by this experiment was to see if providing Leptasterias with only one food source for a certain period would make them ingestively conditioned to the M. californianus and come to prefer it over other food. After conditioning the animals for 19 days, they were transfered to four seperate bowis each containing M. californianus, L. planaxis and C. digitalis. All bowls had running sea water and were checked daily. Three Leptasterias were released into a 801 tank (1m X 2.2m X .4m). Water was constantly running except at a low tide once a day from approximately 8 AM to noon. The lighting was constant and the temperature, taken at intervals over two weeks, was constant at 15'C f .25°. These Leptasterias served as a set of "hungry" starfish and were used as 1 feal 5. -6 controls for eating experiments as well as subjects for chemosensitivity experiments. By "hungry" I mean that no food was in the tank and the starfish received food only when presented with it. This was control of when and how much they ate could be maintained. Six Leptasterias and five Pisaster were established in a sea table (10 cm deep, .75m X .6m). Rocks with prey items were placed in the sea table. The prey was very abundant. One set of experiments was run using B. glandula as the most abundant food item. Rocks covered with B. glandula covered approximately 75% of the bottom of the sea table. Another experiment was run with C. digitalis and C. scabre approximately 70 limpets were placed on six flat rocks in the sea table. There were no changes in the tide. The animals had a natural light/dark cycle, although direct sunlight was kept off them by a shade. The positions and feeding of these starfish were checked periodically. Nine Leptasterias and two Pisaster ochraceus were placed on a series of clay plates that were suspended by string vertically in a 801 tank. The tiles were originally 900cm but were epoxied together to form plates that varied in size from 30 X 60 cm to 30 X 180cm. A variety of limpets, T. funebralis, L. planaxis, B. glandula and M. californianus were introduced onto the plates. Different concentrations of different species were used on the plates, some sparsely populated and others abundantly. The plates were under constant lighting and running sea water (temp - 15°C +.25). The animals received a low tide from approximately 8 AM to noon each day of the experiment. These animals were closely monitored from three to ten hours a day. Obser¬ vations, on a map of the plates, were made as to what potential prey was touched, at what time and at what place on the plate. Feeding choices were observed and handling of prey noted. The shells from the finished food items would fall to the bottom of the tank under the plate and could later be secured to confirm species and make measure¬ ments. The plates and the number of each prey item on them are 7 Ral -7 presented in figure 3. Below is a description of the distri¬ bution of the prey, pertinent facts about the size of the prey and the size of the plates with the number of star'on each. Plate 1- Abundant B. glandula and large M. californianus. This food is easy to capture but takes a long time for handling and eating. In fact the M. californianus proved to be too large (over 3cm) for the Leptasterias to handle efficiently. Two Leptasterias were on the plate which measured 30cm X 90cm. This plate was meant to represent an abundance of sessile organisms in patches. 2- Again abundant B. glandula and large M. californianus. Plate Only one Leptasterias was on this plate which measured 30cm X 60cm. This plate was meant to represent an abun¬ dance of sessile organisms well dispersed. 3- One small patch of B. glandula and small M. califor Plate nianus. This plate was meant to be sparsely populated. Two Leptasterias were on this 30cm X 90cm plate. 4- One small patch of B. glandula and small M. cal¬ Plate ifornianus. Two Leptasterias were on this 30cm X 90cm plate also. This plate was suppose to be abundant rela¬ tive to plate 3. Plate 5- Small to medium M. californianus (small less than 1 cm, medium 1 to 2cm)oon this plate. It measured 30cm X 180cm and contained two Leptasterias and two Pisaster. The plate was meant to represent very abundant conditions. The Leptasterias used were collected from two places, Yankee point and Pescadero point. The sizes ranged from three centimeters to seven centimeters with the average at 4.8 centi¬ meters. The Pisasters used came from many different areas around Hopkins Marine Station. They ranged in size from four to seven centimeters with an average of 5.3 centimeters. T. Peccl RESULTS AND DISCUSSION Food Preference Leptasterias has a food preference hierarchy that is shown in figure 1 and 2. The most obvious evidence for this scheme comes from the isolated Leptasterias in small bowls and their different prey choices. Two different bowl experiments were run. In these experiments, Leptasterias had available, in a small area, 15 to 25 individuals of 4 species of prey in one experiment and 7 individuals of 5 species in another. It is assumed that all prey were encountered and the small size of the container would limit the effectiveness of any escape responses. The most frequently taken species were the limpets and L. planaxis. Rarely were the M. californianus of T. fune¬ bralis taken and never before the more preferred species were gone. B. glandula was not included in the food bowls, but can be placed into this preference hierarchy by observations made on the clay plates and in the sea table. Here B. glandula was the fourth choice after the limpets, L. planaxis and M. califor¬ nianus. One can see this by looking at figure 3 for plates 3, 4 and 5. The preference for B. glandula however is variable. The choice of M. californianus over B. glandula is dependent on the size of the Mytilus and the density of B. glandula. Large M. californianus and plentiful B. glandula will lead to more meals on B. glandula. This is seen again in figure 3 for plates 1 and 2. B. glandula were abundant on the first two plates and M. californianus were to large to handle efficiently, so more B. glandula were eaten. Plates 3 and 4 had only one small patch of B. glandula and smaller M. californianus, so M. californianus were seen to be more preferred. Of course these choices always follow the more preferred limpets and L. planaxis. Even though L. planaxis is known to be preferred from bowl experiments, figure 3 shows that is was not often taken outside of the finger bowls. This is because the animal often crawls out of the searching range of the Leptasterias. In the T teo -9- sea table it would crawl up the sides above the water line. On the plates it would crawl out of the water and up the strings that suspended the plate. Also note that T. funebralis was never taken. This is because this potential prey species has an effective escape response and can avoid Leptasterias. They seem to be able to crawl away faster, or if they feel really pursued, will let go of the plate and fall to the bottom of the tank. C. pelta and limatula also have effective escape responses and were never caught on the plates. C. digitalis, C. paradigitalis and C. scabra are what is meant by all use of the term "limpets" since these are the species easily caught by the starfish. It is possible that the above mentioned prey species, with escape responses, are caught in the wild, where they might possibly be trapped in a crevice or cornered. (Menge, 1972a). But on the clay plates their behavior was effective in preventing capture. Ingestive Conditioning To see if these Leptasterias could be ingestively con¬ ditioned four were placed in a bowl with M. californianus. After 19 days and observations of 22 feedings on Mytilus, they were transferred to individual bowls with a variety of foods. The next day they were observed eating either limpets or L. planaxis. It does not appear that after 19 days of exposure to only M. californianus that they learned to prefer it. This does not mean that they could not be injestively conditioned. They might have a stronger conditioning to the other more pre¬ ferred foods. Or there might be a longer time period needed to condition them. Also they might have to be conditioned to food during a certain time in their life cycle. Ingestive conditioning experiments done on Pisaster ochraceus by Landenberger (1968) showed that it took three months of constant feeding to get the starfish to show a preference of T. funebralis. And this quickly degenerated within a week when presented with a choice of Mytilus or T. funebralis. Later, it will be shown that limpets are not only the prefered choice but also provide the most calories per unit time 1fec -10- for the animal. It would be adaptive then that the Leptasterias not lose the preference for the more desired prey after only a short exposure to another prey. Sensing of Food Menge (1972a) tested the long range chemosensitivity of Leptasterias with a Y maze and obtained negative results. Experiments for ranges of less than a centimeter had not been done however. The "hungry" Leptasterias in the 801 tank were model subjects for chemosensitivity tests since they hadn't eaten for frome2 to 11 days. When individuals were actively searching for food, a potential prey item would be placed on the side of the tank just out of reach of the podia on the tips of their arms. This could be above, below or behind. Often the starfish would pass within millimeters of the prey items and never change course. However, if the potential prey was placed in front of or where one of the arms of the Leptasterias would touch it, the response was always positive and on each occasion the prey was consumed after the first touch. Table 1 shows the number of times that food was closely missed and also shows the positive result when food was encount¬ ered. It is assumed that if the starfish had known food was near, through chemosensory means, it would have moved towards the food, since feeding after contact occurred in 100% of the cases. Contact chemoreception and tactile sensitivity seem to be the only methods that Leptasterias has to detect potential prey items. Both would limit the perceptual field of the Leptasterias to its diameter. It has been stated that in the well mixed intertidal zone, chemosensitivity might be of dubious value since the direction signals could be mixed up (Menge 1972a). However, an ability to sense something close would not have been dis¬ advantageous. Also since asteroids seem to have only tactile or contact shemosensory means to explore their environment, no vision or distant chemosensitivity, a question arises as to why they pass over potential prey items when they have no assurance that they will encounter more desirable ones? 7 Petel -11- Feeding Patterns Frequency of Eating Landenberger (1968) reported that Pisaster ochraceus are insatiable feeders. Data from experiments on the clay plates and from the feeding bowls shows that Leptasterias and small Pisaster ochraceus tend to have eating bouts followed by spaces that range from two to four or more days. Figures 4 and 5 show these results. Feder (1970) also noted "periods when individual sea stars do not feed." The animals will often eat more than one item a day, in fact it may eat two or three in quick succession. Dropping the shell of one prey item and quickly securing another has been observed. The animals often move while eating and could be scanning the area for the next meal or possibly making a determination of prey density which might influence consumption. Animals that are presented with an abundance of food in a small area, like the bowl experiments, did seem to eat a somewhat larger amount than those on the plates. So the frequency of contact and abundance of food might have a posi¬ tive effect on eating response. Movement Related to Searching and Eating Starfish are capable of moving quite rapidly on occasion. The "hungry" starfish in the 80 1 tank were in general active and have been measured moving at over 30cm per minute. On the other hand, when prey is abundant, in the area that the starfish occupies, it will usually remain in one spot and eat an area bare of easily caught prey before moving on, if not disturbed. The sea table experiment provided evidence that Leptasterias and Pisaster, in an area with abundant prey, will tend to stay in that area. Figure 6 shows this. Most of these asteroids spent near or over 50% of the days they were monitored in one place eating the available food. One experiment was run with B. glandula as the predominant food and another experiment with limpets as the most abundant. Either food seemed to be attractive enough, when in such large amounts, to hold the animals in one spot. Observations were made of two Pisasters remaining on a rock with 16 limpets and T fer -12- eating every one of them before they left the rock. The starfish also seem to return to the same rocks where they once found food. But whether this is due to a retention of a geographical location linked with food or just chance wanderings was not determined. It was noted though that the starfish in the sea table knew to stay near the bottom and on the rocks to find their food, while the starfish, on the clay plates, seemed to know that most of their food items migrate to the tops of the plates so that is where they spend most of their search time. Landenberger (1966) was able to train Pisaster ochraceus to move to the bottom of the tank, where food was placed, on signal. He also observed Pisaster re¬ maining near pillars from which most of the Mytilus, that made up their food supply, dropped. Although the plates were arranged to have varying amounts of food in an attempt to see different searching patterns, this was not totally successful. Plates 1 and 2, which had an extreme abundance of sessile prey led to a reduction in foraging and a choice of mainly B. glandula for food. This was expected from the sea table experiments. Plates 3, 4 and 5, which were similar except that 4 and 5 had many more limpets, L. planaxis, and T. funebralis yielded similar foraging acti¬ vities when compared (see Table 2). Either the difference in number of prey between the plates was not significant to yield a large difference in foraging activity or just the act of chosing moving prey requires that the Leptasterias move a certain amount. The Leptasterias on plates 4 and 5 had the oppor¬ tunity to reject more prey items before the final choice. Whereas the Leptasterias on plate 3 didn't have the same op¬ portunity. Also the food on plates 4 and 5 was often bunched so that the starfish on these plates encountered food as fre¬ quently as those on plate 3 though the density encountered was higher. Although movement of the starfish on the clay plates was closely monitored, no clear search patterns were discernable. The starfish seemed to wander along almost randomly as noted by Feder (1970), although they do stay closer to the top where most of the food is. Also, they often pass over many prey items / Peat -13 before selecting one. The criteria for the final choice also could not be discerned. Selection of Prey Items Each potential prey item that is passed near and touched is usually felt out by the arm that touches it. Occasionally the starfish will move over to the prey item as though to eat but then pass it over. As stated before, the actual criteria for selection is not certain but it seems clear that the time between the last meal, the kind of prey item that is contacted as well as how often prey is contacted are important. Figure 7 shows that "hungry" starfish will rarely pass over the first meal they touch, while it is quite common for the well fed starfish on the plates to walk over five to ten highly preferred prey items (including species they will eventu¬ ally choose) before eating. Perhaps because of past experiences plate asteroids assume they will find more prey because food is abundant, whereas the "hungry" starfish in the 80 1 tank only encounter food when it is presented to them so might take the first available since it is all they are likely to find from their knowledge of prey density in their environment. A couple of the "hungry" starfish were fed within periods that the starfish on the plates would normally eat. So as far as consumption of food, they should not have been any hungrier than those on the plates. But they still took the first item given, whereas those on the plates, even after not eating for a few days, would still pass potential prey over. Caloric Basis of Prey Selection Observed prey preferences may stem from a biochemical basis that makes it more advantages that the starfish select a certain prey over another. Figure 8 shows that the calories per food item per hour that it takes to eat the prey item compare favorably to the preference list. C. digitalis, C. paradigitalis, C. scabra and L. planaxis were seen to be prefared. Of these, the limpets yield the most calories per unit time. Another consideration, besides just the time specific yield of calories for a prey item, is how easy it is to catch. This is also shown on figure 8. If a food is high in calories 7 Pea 7. Reck -14- but takes a long time to catch or eat the calories per hour to consume the prey may make it not worth catching. C. pelta is very high on the scale caloricly but it has an effective escape response from starfish that makes it hard to catch if not trapped, which is virtually impossible on the flat vertical clay plates used in the experiment. So C. pelta were never seen eaten. L. planaxis although not as high in calories as limpets are easy to catch if they are in the search range of the asteroid and not too hard to consume so they seem to be preferred after the more catchable limpets. Many starfish will subsist on B. glandula as a food source if it is abundant and there is a lack of more preferred prey that is easy to catch. The B. glandula won't escape and the starfish can stay and eat its fill. T. funebralis not only has a low caloric value but also has an effective escape response. They were never seen eaten in the sea table or on the plates by Leptasterias. The Pisaster did eat T, funebralis in the sea table when B. glandula was predominant, but the Pisasters move faster than Leptasterias and also can handle larger prey items, making T. funebralis a more caloricly feasible food. -15 GENERAL DISCUSSION From the experiments just cited, a few generalizations can be made about some aspects of the feeding behavior of Leptasterias. Although the experiments were performed with an arbitrary set of prey items this would not necessarily be different from a starfish moving into a new environment. The presentation of different prey is still valid and the even¬ tual choices still have signigicance when looked at as a model of feeding strategy. The abundance of the prey items affects gross average movement of the feeding asteroids. It would be expected though that starfish in havitats of abundant food would have less reason to move in search of food or to continue to explore the environment and this is what was observed. Both starfish in the sea table and those on the plates 1 and 2 which were abun¬ dant with sessile organisms are evidence of this. The frequency of eating proved surprising. Landenberger (1968) expressed the opinion that cold blooded animals ex¬ perience neither hunger nor satiation as understood by humans. So why don't the starfish feed at a more constant rate? One thought is that they use the time between meals to explore their habitat. Another is that they consume food at a rapid rate for a few days and then must give their bodies time to digest and distribute the stomachs contents. The fact that starfish are generally more selective when there is an abundance of food and less selective when food is scarce has been observed (Menge, 1972; Sioan, 1980). This investigation indicates they will take a prey item on the first contact when prey densities are low but will pass over prey when densities are high. Both of these would suggest some sort of ability to remember and recall facts as to the content and density of prey in the environment while foraging. It was seen that the starfish in the food bowls had definite prey preferences and these choices remained fairly constant in the experiments with simulated environments. Leptasterias lacks vision and long range chemoreception so unless they have some means of remembering what they have 7 ferd -16- previously encountered, while exploring their environment, it would be hard to understand how they could take a chance in passing up one potential prey item in the expectation of finding another. The fact that 19 days of exposure to another known food item doesn't change the selection of more preferred items, when seen again in small but recognizable amounts, shows that pre¬ ferences are based on more than just short term prior exposure and abundance. Also since preferred items are usually nutri¬ tionally superior to other items of prey, having a preference does serve a purpose and this preference should be exploited as much as possible. To optimize caloric intake per unit of feeding time, including exploration, capture and ingestion, it would not be in the best interest of the starfish to just consume the most plentiful prey or even whatever prey it comes across, but to know what is available in its environment and how it fits into the preference scheme. Actual selection probably depends on the prey species (related to prey preference, ease of capture and nutritional value), the number of contacts with this species in the en¬ vironment and the time between encounters. Depending on the environment a less prefered species might be eaten because the chances of finding the other more preferred is not high. Diets are seen to vary with locality (Mauzey et al., 1968; Menge, 1972a; Menge and Menge, 1974) and since starfish often wander into areas of idfferent prey densities, some component of learning might be associated with feeding behavior in each new situation. Despite their limited means of perception, through searching behavior and tactile assessment of the en¬ vironment, the starfish might learn enough about the surround¬ ings to apply this knowledge to a set of internal criteria to select the best prey item. Of course much more work needs to be done on this problem of selective feeding. Putting more Leptasterias in different environments and increasing the monitoring of contacts with prey and time between meals will be necessary. It seems fairly obvious however that feeding in starfish is not random or chance but is a complex behavior with many components, some environ¬ mental and some intrinsic to the animal. 7 Peat ACKNOWLEDGEMENTS I would like to thank my advisor, Chuck Baxter, for his constant assistance, without which I would have been stumped many times. And also for his enlightening bits of information that often opened up new doors in how to approach and understand a problem or result. Those I have worked near, exchanged ideas with and received help from in the lab also have my warmest thanks. 7. Fead 7 Pecck -18 Feder, H. M. 1959. The food of the starfish, Pisaster ochraceus, along the California coast. Ecol. 40: 721-724. Feder, H. M. 1970. Growth and predation by the Ochre seastar, Pisaster ochraceus, in Monterey Bay, California. Ophelia 8: 161-185. Harrold, C. 1981. Feeding ecology of the asteroid Pisaster giganteus in a kelp forest system. Ph. D. thesis. Ucsc. Landenberger, D. E. 1966. Learning in the Pacific starfish Pisaster giganteus. Anim. Behav. 14: 414-418. Landenberger, D. E. 1968. Studies on selective feeding in the Pacific starfish Pisaster in Southern California. Ecol. 49: 1062-1075. Mauzey, K.P. 1966. Feeding behavior and reproductive cycles in Pisaster ochraceus. Biol. Bull. 131: 127-144. Mauzey, K. P.; C. Birkeland; P. K. Dayton. 1968. Feeding behavior of asteroids and escape responses of their prey in the Puget Sound region. Ecol. 49: 603-619, Menge, B. A. 1972a. Foraging strategy of a starfish in relation to actual prey availability and environmental predictability. Ecol. Monogr. 42: 25-50. Menge, J. L. and B. A. Menge. 1974. Role of resource allocation, aggression and spatial heterogeneity in coexistence of two competing intertidal starfish. Ecol. Monogr. 44: 189-209. Pyke, G. H.; H. R. Pullian; E. L. Charnov. 1974. Optimal foraging: a selective review of theory and tests. O. Rev. Biol. 52: 137-154. Schoener, T. W. 1971. Theory of feeding strategies. A. Rev. Ecol. & Syst. 2: 369-404. Sloan, N. A. 1980. Aspects of the feeding biology of asteroids. Oceanogr. Mar. Bid. Ann. Rev. 18: 57-124. -19- FIGURE LEGENDS Fig. 1 & 2) Number of prey remaining in food bowls over a 12 day period. The foods preferred will be consumed the fastest. So order of preference is limpets, L. planaxis, T. funebralis and M. californianus. Two seperate experiments were run. Fig. 3) The first block shows the number of each prey item that was maintained on the plates. The second block shows the number eaten out of all the prey items that were ever on the plate. The replaced prey as well as the original are included. L - limpets (C. digitalis, C. scabra, C. paradigitalis) PL - Littorina planaxis B - Balanus glandula M - Mytilus californianus T - Tegula funebralis Fig. 4) Shows what food is eaten and how often by the starfish on the plates. Number of meals, total calories for all meals and calories per day are shown for each starfish. B - Balanus glandula D - C. digitalis PD - C. paradigitalis S - C. scabra M - Mytilus californianus PL - L. planaxis Fig. 5) See fig. 4 for key. Shows what food was eaten and how often by the starfish in the feeding bowls. Number of meals, total calories for all meals and calories per day are shown for each starfish. Fig. 6) Bars represent the largest percentage of time, out of the complete time the experiment was run, that the starfish remained in one area. Data was from the sea table experiments, one with abundant limpets and the other with B. glandula. P - Pisaster ochraceus L - Leptasterias hexactis F Peal 0 -20- Fig, 7) Comparison, of the percent of meals taken on the first touch, between hungry starfish which usually have no food, and starfish on clay plates which do have access to food. Fig. 8) Caloric content of individual prey per time to consume as well as ease of capture on the plates. D - C. digitalis PD - C. paradigitalis S - C. scabra P - C. pelta PL - L. planaxis T - T. funebralis B - B. glandula M - M. californianus TABLE LEGENDS Table 1) Chemosensitivity tests. When the starfish does not contact the prey item it seems to not realize it is present. On contact the starfish will take the food. Starfish used were the hungry starfish. Table 2) Average daily movements for individual starfish and description of the different plate environments they occupy. C 7. Peccl -21- -0 0 - — S ONINIVWHA XAAd 4O AHAWON 0 — — - — oa- ONINIVWAA XAAd HO AHAMON 7. Pead + 36 L O oc OS OS oe N OS O PIO — ope — Po OS e 0 oc 0 8 -22- 1 Peag Kep soo soo o su o oun la -23 1Pecte O 9 5 0 - OT- Kep s sreou o seo o steoy JO 'ON — -24 1see a 2 O- -25 Sa 8 a a taatatakakatakavoa- VHAV ANO NI LNHAS AWIL 40 ADVLNADAAd LSADAVI rea -26 1 Petet 2 - 20 : 0 — Z0 2 0 PREY T. funebralis B. glandula L. planaxis californianus C. digitalis paradigitalis C. scabra C. pelta H KCAL, (dry 4.50 3.34 4.90 4.78 4.96 4.99 4.95 5.08 60 54 48 42 30 24 SIZE EATEN .5-1 .5-.8 .5-1 1-1.5 7-1.5 7-1.5 7-1.5 7-1.5 SIZE FOR WT .6 .8 1.4 1.5 1.4 1.3 1.5 D PD -27 WET WT. (Q .020 .001 .046 .054 .14 .14 .12 .15 CAL.7 TIME CALC. TO EAT %H,0 INDV. MPrTa hrsl 5-7 0054 24.30 73.2 8-10 0003 1.002 73.6 5-7 0173 84.77 62.3 83.0 43.97 12-15 0092 3-5 77.1 0321 159.22 .0349 174.15 3-5 75.1 142.56 3-5 76.0 0288 3-5 179.83 76.4 .0354 Easy to cap¬ ture and handle. Easy to capture, harder to handle. Very hard to capture on plates. 1 E S PPLTBM tpeate 0 —1 -28 7 Pecle — 25 — . — OE 2 58. a 9. 5 95 2. tecd