ABSTRACT A study of the activity patterns of Collisella scabra showed that the most prevalent movement occurred during periods of submergence with no difference between daylight and darkness. Furthermore, studies showed that cephalic tentacles are not essential to the homing ability of this species, but they do play a role in the ability to detect the directionality of mucous trails. INTRODUCTION Collisella scabra is a homing species of limpet. This particular species grows its shell margin to fit the substrate at its home site (Davis, 1895; Underwood, 1979; et al). This provides protection against removal by predators, as well as providing resistance to desiccation while the limpet is emerged at low tide (Branch, 1981, 1986). Because these factors depend on the ability of C. scabra to return to its home site and orient properly after a foraging run, a rigid homing instinct is of paramount importance to its survival. The mechanism which C. scabra uses to return home is a topic of some dispute. Several mechanisms have been proposed to explain this homing ability; these mechanisms fall into three categories (Cook 1969,1971; Thomas 1973; et al): 1) Use of external cues: these mechanisms rely on factors not associated with the limpet or the limpet's home rock. These include the position of the sun and/or moon, identifying coastal landmarks, polarization of light, magnetism, and gravity. 2) Dead-Reckoning: this relies on the ability of the limpet to recall either distances travelled and angles rotated or linear and angular accelerations and "compute" an acceptable route home. 3) Use of endogenous clues: these mechanisms rely on factors associated with or incorporated onto the surface of the rock. These include following a mucous trail home, and possessing a topographic memory of the rock surface. The ability of limpets to home after the home rock was displaced or rotated have demonstrated that C. scabra do not rely exclusively on external clues for their homing mechanism (Underwood, 1979). Likewise, the ability of limpets to home in displacement experiments has eliminated dead-reckoning as the sole homing mechanism in C. scabra; limpets cannot rely on remembering their outward trail and be able to return home after being displaced from their original path (Branch, 1981). The third category of mechanisms has proven to include the most likely strategies employed by a limpet to return home following a foraging excursion. Ability to follow mucous trails has been demonstrated for many species of gastropods, including C. scabra (Cook and Cook, 1975; Peters, 1964; Connor, 1984a,b; et al.) Before a detailed study of the homing mechanism used by C. scabra could be undertaken, it was necessary to perform a set of field experiments which would place the homing ability of this species in its proper context. Populations of C. scabra were observed over a diel cycle and its general activity pattern The size of limpets was measured at different recorded. intertidal heights, as well as distance travelled on a foraging run, and time spent foraging. Furthermore, the paths taken by C. scabra in its foraging excursions were recorded. Completion of these observations set the stage for a study of the mechanism(s) used in the homing ability. Several researchers have examined the role that cephalic tentacles play in the homing of gastropods, both in field and laboratory situations. The cephalic tentacles of gastropods are a tactile and an olfactory organ (Peters, 1964), and may be of great importance in the trail-following ability of C. scabra. Field experiments involving the removal of cephalic tentacles have yielded conflicting results. Davis (1895) removed the tentacles from 2 individuals and displaced them 5-6 cm from their home site -- these animals returned to their home site and Davis concluded that tentacles are not indispensable to the homing ability of C. scabra. However, Jessee (1968) removed limpets from their home sites, anesthetized them with magnesium chloride and left them in an aquarium for 3 days to recover before similar displacement. Most of the limpets in this study did not return home and those that did may have relied on an alternate mechanism since directionality in mucous trails is only present for 48 hours (Connor, 1984b). Using a mucous trail to return home requires that the limpet not only be able to detect the mucous trail, but also to distinguish its trail from others and to determine the directionality of the trail. Experimental manipulations of gastropods have been the key factor in demonstrating the ability of these organisms to follow mucous trails, and are the only studies which have demonstrated a Unfortunately, only a directional component to the trail. handful of studies have examined the role of the cephalic tentacles in trail-following experiments. Peters (1964) reports that Littorina planaxis requires its cephalic tentacles to detect mucous trails at all. Behavioral observations by Connor (1984b) suggest that Collisella scabra uses its cephalic tentacles to detect the directional component of the trail. She describes C. scabra touching alternate sides of a trail with its tentacles as it progresses along it. Two studies were undertaken using C. scabra in order to examine the role of the cephalic tentacles in the homing ability of this species. First, a field study was performed to resolve some of the conflict as to whether C. scabra is able to home after excision of its cephalic tentacles, and second, a trail¬ following experiment was conducted to determine the role of cephalic tentacles in this type of homing ability. MATERIALS AND METHODS ACTIVITY PATTERNS Observation of the movements of Collisella scabra were made at 15 minute intervals over a 24-hour period at Cabrillo Point, Pacific Grove on May 30, 1990. Three relatively flat study sites were chosen along a transect at .83 m, .98 m, and 1.29 m above MLLW. Snorkel equipment was used to observe movement during periods of submergence. The shell length of individual limpets was first measured along the line of greatest distance and then their home sites were marked using Wet N' Wild nail color with one mark placed on the anterior portion of the shell and a corresponding mark placed adjacent to the home site. Clear acetate sheets were then placed over the 1 m study sites and a number was written on the transparency corresponding to The home site marks and the individual home sites. transparencies denoting the position of the home site facilitated observations which required distinguishing those limpets which were on their home site from those that were away from it. The positions of 100 individuals in each of the three study sites On a subsequent period of were recorded in this manner. activity, 25 individuals in each study site were chosen at random from the group of motile animals; the time spent foraging and the maximum distance travelled by each of these limpets were recorded. These transparencies were also used to record the paths taken on foraging runs. At 15 minute intervals, the position of a given limpet was marked on the transparency using a lead pencil. TAGGING Individuals which were used in the tentacle removal needed an identifying mark so that they could be replaced adjacent to the correct home site. Limpets were identified using a system of different colored dots to represent Roman numerals: Green - 1; blue - V; Orange - X. These markings remained up to 6 weeks. Home sites were identified using transparencies in the manner described above. TENTACLE REMOVAL IN THE FIELD In removing the cephalic tentacles, I strayed from the methods of previous researchers and found that an ounce of patience was worth a liter of solution of Mgcl, isotonic to seawater. Previous studies involving the removal of cephalic tentacles employed the use of just such a solution to anesthetize animals and facilitate the tentacle excision (Peters, 1964; Jessee 1968). However, C. scabra will readily extend their cephalic tentacles if submerged and placed on their shell. To facilitate tentacle removal and reduce the size range of limpets used, C. scabra ranging from 12-15 mm were used in this study. Although Hewatt (1940) reports limpets smaller than 14 mm do not home, Jessee (1968, and personal observations) indicates that only limpets smaller than 6 mm fail to home. A subsequent report by Sommer (1982) indicates that limpets as small as 2 mm display homing ability although a higher degree of home site changing is observed. In this study, 25 limpets were placed on their shells in a large petri dish and the tentacles, once extended, were removed using a pair of iridectomy scissors. 25 other limpets were removed from the rock and placed in a similar petri dish on their shells to serve as controls. In these field experiments, limpets were then immediately replaced to within 6 cm of their home site on a rising tide, and their ability to return home was noted on the subsequent low tide. MOVEMENT DURING SIMULATED TIDE A tide tank was set up to simulate the rising and falling of the tide. A series of 7/16" holes were drilled in 1" PVC pipe and used as spraybars to simulate wave-action. This was attached to a filtered seawater system through a Rainmatic 2000. This device has the capacity to execute 8 on/off cycles and hence could be programmed for up to 4 days (two high and two low tides per day). A series of 7/16" holes was also drilled vertically in a PVC standpipe to allow for a gradual rising and receding of the tide. TRAIL FOLLOWING EXPERIMENTS Using techniques described by Connor (1984b, modified from Cook & Cook, 1975), the ability of limpets to follow mucous trails was examined. Glass plates were scrubbed with Alconox, dipped in a suspension of carbon particles, and rinsed with Carbon particles adhere to mucous, and distilled water. therefore, exposing the plates to this suspension ensured that mucous was removed from previous experimental trials. The glass plates were then placed in the tide tank described above and over a period of 24 hours collected a layer of fine particles. Trails laid down on these plates appeared as clear trails through the layer of sediment. Once a trail was laid down, the limpet was removed and the plate rotated 90° to account for any external cues. Depending on the experiment, the same limpet or a conspecific was placed .5 cm from the middle of the trail. Equal numbers of limpets were placed on each side of the trail to account for any innate turning preference. Responses to the trail were: 1) no movement (not used in analysis); 2) crossing the trail; 3) following the trail toward the origin; or 4) following the trail away from the origin. Individual limpets were used only once. A trail lasted 30 minutes, and a limpet had to move at least 5 cm along a trail in order for that trial to be recorded as a positive following experience. This trail-following experiment described by Connor (1984b) was repeated using C. scabra collected at Cabrillo Point, PG. In addition, limpets with excised tentacles were tested for their After trail following ability using a similar technique. excision of the tentacles, but before placement on the glass plates, limpets were placed in an aquarium for a period of approximately one hour or until notable movement was observed. From this point, the procedure described above was repeated. RESULTS ACTIVITY PATTERNS Figures la,b,c display the activity of C. scabra over a 24 hour period at three distinct intertidal heights. No activity was observed while the limpets were emerged at low tide (15 daylight observations and 2 during low tide at night). The period of greatest movement was found to correspond to daytime submerged periods at each site, although substantial movement was also observed during awash periods, both day and night, and during nighttime submerged periods. In 8 weeks of observations, C. scabra was found to strictly adhere to this pattern. Figure 2 represents the relationship between the size of the limpet and the study site in which it was found. This figure shows a significant relationship between the size of a limpet and its position in the intertidal region (r - .528, p « .0005). On the average, smaller animals are located lower in the intertidal zone, with the size of the animal increasing with intertidal height. Figure 3 displays the relationship between the size of a limpet and the maximum distance travelled in a given foraging excursion. A signifacant relationship was also found to exist in this case (r - .448, p « .0005). In general, larger animals travelled greater distances while foraging. These relationships do not represent a random sample of the population since study sites were chosen for the presence of C. scabra; however, there is a significant relationship between the size of an animal, its position in the intertidal, and the maximum distance travelled in a given foraging trip. Furthermore, subsequent observations have shown these patterns to be representative of this population of C. scabra. Figures 4a,b,c display some of the common foraging patterns of C. scabra. Platzfrass (fig. 4a) behavior is evidenced by a limpet raising its shell and turning on its home site. Platzfrass or "eating in place" behavior was so termed by Funke (1968 in Connor 1984b) based on behavioral observations which suggested that limpets were feeding on a ring of algal material which grows around the home site in between the perimeter of the shell and the area occupied by the foot of the limpet. The other foraging patterns are similar to patterns observed in other species of homing gastropods. Figure 4b represents foraging runs in which limpets return to their home site by retracing their outward paths; figure 4c represents another common foraging pattern employed by limpets where the outward trail is not retraced in the journey home. TENTACLE REMOVAL IN THE FIELD After allowing sufficient time for the limpets to return to their home sites (one period of submergence), limpets were classified as: 1) home -- on its home site and oriented properly, 2) not home -- away from the home site or not oriented properly, or 3) missing -- not found within reasonable distance of the study area. Results from this field experiment are presented in table I. It is evident from this table that no significant differences in homing ability were found to exist between control limpets and those with excised tentacles. Observations of the movement patterns of these two groups were made on four subsequent daytime high tides and no substantial differences were seen in numbers of limpets moving, average distance travelled, and average foraging time of the two groups. TRAIL FOLLOWING IN THE TIDE-TANK Activity patterns of limpets placed in the tide-tank described above were consistent with those displayed in figure 1 (a-d). The results of the trail following experiments were analyzed using a Chi square test comparing observed results to 503 expectation for trail following vs. trail crossing (i.e., if trails had no effect, half of the limpets would follow the trails, and half would cross it). Of those limpets which followed the trail, frequency of trail following toward the origin vs. trail following away from the origin were again compared to 508 expectation values (Table II). Results of the experiment on intact individuals were consistent with those found by Connor (1984b). Intact individuals followed trails more often than they crossed them. Individuals placed next to their own trails followed them preferentially toward the origin, while individuals placed next to trails of conspecifics followed trails preferentially away from the origin. Limpets with excised tentacles followed trails more often than they crossed them. However, the directional preference found in intact individuals was lost in the limpets lacking cephalic tentacles. DISCUSSION A study of the activity patterns of Collisella scabra at Big Fisherman's Cove on Santa Catalina Island, CA, reports movement only during daytime awash periods (Wells, 1980). Wells defines an awash period as "the time interval during which the zone of turbulence accompanying the tide is present at a particular intertidal height." Apparently, the lack of severe wave action, and accompanying turbulence, at Big Fisherman's Cove caused the awash period,as Wells defines it, to be restricted to hour long periods when the limpets were just covered by the ebbing tide. This pattern of movement is similar to that found by Hewatt (1940) and Wells (1917). However, Brant (1950) reports that feeding excursions occur at high tide with no apparent differences in activity between daylight and darkness. Furthermore, Villee and Groody (1940) report only small individuals of C. scabra were seen to move at all. However, these observations, as well as those of Hewatt (1940) and Wells (1917) were largely restricted to low tide, with high tide observations being made without the use of snorkel or SCUBA equipment. With these extreme differences in activity patterns between different populations of C. scabra, I found it necessary to define the times of greatest movement for the population of limpets used in this study. The sites employed in this study were characterized by moderate wave action. The zone of turbulence was present throughout the entire submerged period, even at the study site lowest in the intertidal region. The difference in the amount of wave action in different studies, and hence the amount of time spent in the zone of turbulence, could account for the differences in the amount of limpet movement at varying degrees of submergence found in these studies. This suggests a detailed study of the correlation of limpet movement with wave action in both the same and geographically isolated sites to determine the relationship between turbulence and activity as well as its potential adaptiveness. This study shows that limpets higher in the intertidal move more during an awash period than animals lower in the intertidal, suggesting that time spent submerged is a factor in the amount of movement seen at different degrees of submergence. In addition, with larger limpets being located in the higher intertidal region (see fig. 2), the size of a limpets could also be playing a role in the different amounts of movement observed during submerged periods. Tidal location, wave exposure of the site, and limpet size may all play a role in a limpet's activity patterns. The differences in nighttime movement seen in these studies It would seem here that the is more difficult to explain. limpets are already more active at Cabrillo Point than at Catalina and thus experience less pressure to be actively feeding at night. However, the amount of movement a limpet is required to undertake for its survival at a given location depends on the type of food, its abundance and the population density, as well as the growth rates of individuals and their reproductive potential. Another explanation is the differences in the types of predators found at different sites. Wells (1980) examined the effects of octopus, a nocturnal predator of limpets, on activity patterns of C. scabra. The lack of nighttime movement of C. scabra recorded by Wells could be due to effects of predation by At Cabrillo Point, octopus or other nocturnal predators. predation by octopus is a rarity at best, and is certainly not common enough to affect activity patterns of C. scabra found at this site. A site or geographic difference in the basic foraging behavior of a limpet with a predator's presence brings up the interesting question of genetic versus learned determination. Studies on this could give ecologists new perspective on species interactions important in population dynamics. The paths a limpet follow in a given foraging run can be grouped into the three categories shown in figure 4 (a,b,c), These are: 1) Platzfrass, 2) taking different outbound and return routes, and 3) retracing the outward bound path. Each of these foraging patterns represents a unique aspect of the limpets homing ability. Limpets partaking in Platzfrass behavior need only reorient themselves properly on their home site when finished. As simple as this process sounds, it is actually quite complex. C. scabra do not orient themselves on their home site by a trial-and-error process of attempting to obtain a proper shell fit with the substrate, but instead will leave their shell raised and turn on their home site until oriented properly and then lower their shell (Sommer, personal communication and personal observation). Mucous laid down previously by the limpet in the Platzfrass behavior, in addition to trapping microscopic algae and diatomaceous material for later feeding (Connor, 1984c), could serve as a directional clue for proper orientation. Limpets employ the use of their cephalic tentacles in this orienting behavior, feeling the area around the home site. Three of the four limpets which failed to return home in the field tentacle removal experiment were classified as "not home" due to improper orientation on the home site. However, the high degree of proper orientation in limpets lacking tentacles, and the inability of these limpets to detect trail directionality suggest that an alternate method may be employed for proper orientation. Lindberg and Dwyer (1983) reported on the mechanism employed by C. scabra in excavating their home site depression. They found that C. scabra secrete acidic mucopolysaccharides from the foot and acidic mucopolysaccharides and carbonic anhydrase from the mantle edge. These substances were shown to soften the rocks to allow for excavation by the radula. To date, the only reports of Platzfrass have been behavioral observations and it could be that the limpets are excavating their home site depression in addition to, feeding on the algal material around the home site which is believed to occur during Platzfrass behavior. This is further evidenced by the fact that limpets placed on glass plates in the tide-tank showed a profundity of Platzfrass behavior, possibly attempting to form the depression of its new home site, as limpets placed away from their home site are apt to do. The limpet could use textural clues of the excavated home site in orienting properly. It may be that C. scabra possess either or both of these mechanisms for proper orientation. Davis (1895) suggested that the mantle tentacles could be used in detecting textural cues associated with the rock; this suggests that cephalic tentacles aid in proper orientation, but are not indispensable to it. The mechanism used in the second type of foraging pattern where different outbound and return routes are taken by C. scabra is not well understood. It is most likely that the animal makes use of clues associated with the rock surface, although those mechanisms which have been ruled out as primary mechanisms (e.g., dead-reckoning or visual clues) could play a role in the ability of a limpet to home along a different return route. Some researchers report that mucous trail-following is the only homing mechanism employed by C. scabra (Hewatt, 1940; Underwood 1979; et However, using the trail-following theory of homing al). exclusively makes it difficult to explain why C. scabra do not always return home along the same path that was laid down at the beginning of the foraging run. It is possible that limpets not returning along the same path are encountering mucous trails previously laid down, but this would be difficult to detect without prolonged monitoring of behavior at a site. Directionality is only present in trails for a period of 48 hours, and it is also unlikely that in displacement experiments limpets encounter mucous trails with such regularity as to be able to find their way home more than 808 of the time, Furthermore, Jessee (1968) showed that scrubbing the surface of a rock with a 328 solution of NaoH to remove mucous trails did not inhibit limpets from homing. Similar mucous trail removing experiments have yielded similar results (Sommer, 1982; Branch 1981). Taken together, these results suggest that one or more mechanisms exist for limpets to home other than trail-following. The last type of foraging pattern taken by C. scabra is typified by retracing an outbound path in the return home and can be easily explained by the mucous trail-following ability in this species (Table II). An individual following its own trail to the place of its origin has obvious consequences on the ability of the limpet to return home. Upon encountering a current trail, the limpet need merely retrace it to its home site. It is not known whether limpets are able to detect directionality of retraced trails, and limpets encountering a previously retraced trail may not have the ability to detect its directionality and follow it home. However, if the limpet encounters one of its trail which hasn't been retraced, such as that in figure 4b, it would be able to follow this trail to its home site. A limpet encountering the current trail of a conspecific would turn and follow the limpet laying down the trail, or meet that same limpet head-on if the latter was on its return journey home. A limpet encountering a conspecific's previously laid, unretraced trail would follow this trail to the other limpet's home site. The tendency of a limpet to follow a conspecific's trail in the opposite direction is less well understood. Several explanations exist, all of which are based only on observations and theory. The first explanation of this phenomenon is a territorial behavior; C. scabra could use this mechanism to avoid territories of other limpets as well as guard their own territory Alternatively, this mechanism could be against intruders. employed by limpets who abandon their home sites to search for the home site of a conspecific. Since mucous trails ultimately lead to a home site, it would be beneficial to any limpet away from its home site (e.g., in the tide tank) to follow the trail of another limpet in order to find a ready-made home site. Finally, it could be a form of clustering behavior similar to that demonstrated in C. digitalis (Millard, 1968). LITERATURE CITED Brant, D.H., 1950, A quantitative study of the homing behavior of the limpet Acmaea scabra. Unpubl. Spec. Prob. Reprt., Dept. Zoology, Univ. Calif. Berkeley. Branch, G.M., 1981, The biology of limpets: physical factors, energy flow, and ecological interactions. Oceanogr. Mar. Biol. Ann. Rev. 19: 235-380. Limpets: their role in littoral and Branch, G.M., 1986, sublittoral community dynamics. The Ecology of Rocky Coasts ch. 7: 97-116. Connor, V.M., 1984a, Foraging energetics in two intertidal limpet species. Unpub. Reprt., Dept. Zoology, Univ. Calif. Davis 94-115. the mechanisms underlying Connor, V.M., Analysis of 1984b, directional trail following in the limpets Collisella scabra and C. digitalis. Unpub. Reprt., Dept. Zoology, Univ. Calif. Davis 117-150. Connor, V.M., 1984c, Stimulation of food species growth by limpet mucus. Science 225: 843-844. S.B., 1969, Experiments on homing in the limpet Siphonaria Cook, normalis. Anim. Behav. 17: 679-682. S.B., 1971, A study of homing behavior in the limpet Cook, Siphonaria alternata. Biol. Bull. 141: 449-457. S.B., O.S. Bannford, J.D.B. Freeman, & D.J. Teideman, 1969, Cook, A study of the homing habit of the limpet. Anim. Behav. 17: 330-339. S.B. and C.B. Cook, 1975, Directionality in the trail¬ Cook, following response of the pulmonate limpet Siphonaria alternata. Mar. Behav. Physiol. 3: 147-155. Davis, J.R.A., 1895, The habits of limpets. Nature 51:511-512 Hewatt, W.G., 1940, Observations on the homing limpet Acmaea scabra (Gould). Amer. Midland Natural. 24 (1): 205-208. Jessee, W.F., 1968, Studies of homing behavior in the limpet Acmaea scabra. The Veliger 11: 52-55 (Supp.) Lindberg, D.R. and K.R. Dwyer, 1983, The topography, formation and role of the home depression of C. scabra. Veliger 25 (3): 229-233. Millard, C.S., 1968, The clustering behavior of Acmaea digitalis. Veliger 11, (Supp.) 45-51. Peters, R.S., 1964, Function of the cephalic tentacles in Littorina planaxis (Philippi). The Veliger 7 (2): 143-148. Sommer, F., 1982, Homing behavior of the limpet Collisella scabra. Unpub. MS on file at HMS Lib. Thomas, R.F., 1973, Homing behavior and movement rhythms in the Proc. pulmonate limpet Siphonaria pectinata (Linnaeus). Malac. Soc. Lond. 40: 303-311. Underwood, A.J., 1979, The Ecology of intertidal gastropods. Adv. Mar. Biol. 16: 111-210. Villee, C.A. and T.C. Groody, 1940, The behavior of limpets with reference to their homing instinct. Amer. Midland Natural. 24 (1): 190-204. The behavior of limpets with particular M., 1917, Wells, reference to the homing instinct. Journ. Anim. Behav. 7 (6): 387-395. Wells, R.A., 1980, Activity pattern as a mechanism of predator avoidance in two species of Acmaeid limpet. J. Exp. Mar. Biol. Ecol. 48: 151-168. FIGURE1 ACTIVITY PATTERN 4.83 m l E S I T D 4.98 m AE C 11.29 m l- —l — 2000 1200 0400 HOURS Fig. I -- The activity pattern of Collisella scabra over a 24-hour period (May 30, 1990); bars represent the percentage of the population active; light conditions indicated by light and dark bars; S, submergence; E, emergence; A, awash period. 2- 20 - 18 - 18 - 14 - 12 - 10 4- 2- 2.8 19 — 18 - 17 - 18 - 15 - 14 - 13- 12 - 11 10 - 9 - 5 - FIGURE 2: SHELL LENGTH VS. INTERTIDAL HEIGHT R = .528 n - 75 P (.0005 2.8 3 32 34 38 38 4 42 Intertidal Height (tt) FIGURE 3: SHELL LENGTH VS. DISTANCE TRAVELLED R =.448 n = 75 P6.0005 + 1 + tt 12 16 20 SHLLET (mm) 3 . . 1 0 E. 5 3 0 0 0 — — 0 0 0 a 5 8 0 —O1O o + 310 0 oo 22 ON 9 o o O — O — 96 9