ABSTRACT: Daily movement patterns of and pool fidelity in the intertidal fish anzis was determined. Individuals showed a high degree of fidelity to a specific tidepool and the majority appeared in their pools on an every day or every other da¬ basis. However, almost 25% of the time the cottids sheltered at some unknown location at low tide. I was also found that individuals of this species do not stay in their tidepools at high tide and are shown to be able to return to their pool of first capture over considerable distances. INTRODUCTION: It has been shown of many different organisms found in the intertidal zone that Individuals within a species can locate a particular area of the intertidal and repeatedly come back to it during the time of the tidal cycle that is most adaptive for its suvivability. The chiton M masa is known to occupy a particular homesite at low tide but moves away from it and back along the same pathway at times when it is submerged or awash. Similarly, the limpet Collisella scabra returns regularly at low tide to a specific area of a rock which closely matches the contour of its shell. However, homing ability is not necessarily consistent among all invertebrates, for Collisella digitalis shows no homing response in some areas, and in other areas it returns to a particular homesite only up to 54% of the time (Morris Abbott and Haderlie, Hvo). Several species of intertidal fish have exhibited the ability to appear in the same tidepools over wo different low tides when they have been seen at considerable distances from the pools in which they had last been observed I was suggested that navigational cues were used to bring the fish back to their "home" pools and assured them safe refuge at low tide (Williams, 1954). However, a recentstudy on an indicated that an absence from the pool in which an individual was first caught did not necessarily mean the fish had stayed from the area (Richkus 1978). Instead, by sampling a range of pools once every bo weeks for a period of 4 months, Richkus presented the ldea that s anas are not particular to one certaln pool but reside in a "home-range" made up of pools in a glen area. This report of cottid movement altered the concept of homing as a return to one specific area of the intertidal that the fish knows will be a tidepool at low tide. It suggested, instead, that these animals may be residential to an area of tidepools and regularly refuge in a number of these with a high fidellyy for an intertidal area rather than to a specific pool. Richkus did not, however, follow the fish on a day to day basis or monitor their activities at high tide. As first postulated by Green (1971), the appearance of the same individual in the same tidepool does not necessarily indicate an Instance of homing, for the fish in question could have stayed in the pool at high tide and never employed a homing behavior at all. Green (1971) noted that us 'regularty" leaves ts tidepool at high tide In sheltered areas of the intertidal, but he also found that in more exposed areas the amount of vacancy is markedly reduced. As of yet, there have been no studles addressing the high tide movements of wnxvtus ana and although the abilly to home reguires first that Individuals leave their tidepools and then return, residency in its tidepool at high tide has not yet been assessed. This study examines the pool occupancy and behavior patterns of anais on a day to day basis over a limited period of time. Over the course of 18 consecutive days lobserved 31 marked fish in 5 tidepools at low tide to determine the fidelity of Individuals to specific tidepools In a given area. Furthermore, to be able to atribute the successive appearances of individuals on different low tide to the existence of a homing behavior, Ireport here on high tide observations made on 11 days within the study period. MATERIALS AND METHODS: STUDY STTE: Field work was conducted in an area of protected rocky intertidal at Hopkins Marine Station, Callfornia (Figure 1). The experimental study site consisted of 5 tidepools of varying dimensions In covering an approximately 4 X 4 m square area (Figure 2). Pool number 6 indicated on the map was not sampled but was only observed for the presence of tagged indlviduals. The pools varled in physlcal dimension and topographical characteristics, and height ranged from +0.62 m to 41.08 m relative to mean lower low water. These pools proved to be particularh suitable for my study because of the abundance of n relative to other cottid specles. In addition, all of the pools that were sampled were free of large cracks and crevices in which the experimental fish could possibly hide, so that on each experimental date my low tide counts of individuals were accurate and complete. Dimensions and characteristics of each of the pools are presented in Appendix 1. CAPTURE, TAGGING, AND RELEASE: All pools were cleared of fish on the low tide before dally observations began, and on each lowtide observation Icaught and tagged all new fish present. Because each tidepool was small enough to capture the fish using onty a dip net and my hand, I did not need to use anesthetic in my capturing technique, thus minimizing disturbance to both the fish and the pool itself. Individuals were brought back to the lab, and tagging occurred within several hours after capture. The fish were briefly anesthetized with 2 drops of 202 Oulnaldine diluted in isopropyl alcohol in approximately 250 ml of sea water to determine length and specles. Loose following a similar method suggested by Willlams (1957) and Green (1971), tagged Individuals of each pool using 1 or 2 colored embroidery beads of unique combinations sewed midway into the caudal musculature just posterior to the second dorsal fin. Becovery from the anesthetic occurred within minutes after the fish was returned to fresh sea water. Individuals were kept in aquaria until the next low tide te allowtime for acclimation to the beads before releasing them back into their pool of capture. When released, all fish showed normal swimming behavior and appropriate color change to their surroundings, so l assumed the effects of Quinaldine and tagging to be minimal. To test my tagging method, Ibeaded 10 fish and kept them in aquaria for 22 days. During this time perlod, onty 1 fish lost its beads due to inadequate knotting, and this happened on the 15th day of observation. Feeding behavior and swimming motions did not at all seem to be altered by the tagging technique. During the course of my tidepool observations, however, recovered many individuals that had a large tear in their caudal muscle, indicating that these were previously tagged individuals that had apparently lost their beads. In these instances 1 noted the length of the fish and retagged the individual anterior to the previous tag. In addition, Iperformed fin clips on the one or both of the pectoral fins in unique combinations to aid in the identification of individuals when beads were lost. On several occaslons found a fish that had already been tagged twice, and retagged the fish a third time in the dorsal musculature. In previous studies that followed fish over a substantial period of time, data on individuals was severel limited because tag loss was substantial. However, my sample size was a suitable number to use the length of the fish in combination with the day in which it was caught to identify those fish that had lost their tags and retag them. A total of 31 fish were tagged over 15 days of observation from May 18 to June 4. FIELDÖBSERVATINS: Pools were examined dally when exposed by low tide from May 18 to June 4. On each occasion lobserved each pool for approximately 10 minutes without pool disturbance, recording each fish that! saw.I then made a thorough search of my pool using the handle of the fish net to observe all those fish that were previousy out of view. All new fish were caught and tagged or retagged as was needed. For a detalled account of daily observations and releases, see Appendix 2. fh Fldelly to a glven pool was determined by the number of times a fish was seen in its original pool of capture divided by the number of low tides observed from the first sighting of the fish to the last and reported as a percentage. Fish that were observed for 3 days or less were not used in the fidelity calculations, for a fish that was seen 3times in its original pool of capture but was seen only over a total of 3 days would show 100 % fide lit when, In fact, the amount of time It was observed was not enough to say the individual was 100% partial to Es pool. DEPR Several tests were done to determine fideliyy of fish to an area by tagging and displacing fish. Icaught and tagged 5 fish from 2 pools, Pool A and Pool B, 50 cm away from each other and 16m away from my area and, released them into Pool 2. On another occasion Iperfonned the displacement experiment on 3 more individuals from a pool 18 m away, Poo C. Individuals varied in length from 2.7 to 5.2 cm. MT-TARe On two occasions lobserved the tide coming in and the tide going out over a 40X 40 cm area, which lcalled Area A, located between pools 3 and 5 (Figure 2). At this time lobserved and recorded behavior of the fish in pool 3 as the water both filled and drained the area. t e On 11 days, from May 22 to June 4,1 snorkeled at a high tide height of 1.2 to 1.3 mrelatwe to mean lower low water so that ! could reach my hand into the pools and accurately assess If Individuals were present or not. After pools were thoroughly checked, lobserved the area between the tidepools and the outside area up to 1 meter In permeter and recorded presence of fish and behavlor. Distance from original pool of capture was estimated using a waterproof slate which was 15 cm in length (See Appendix 3). RESULTS: Low Tide Observations: The population of tagged individuals generally declined over the course of the study perlod as is reflected in Figure 3, with a maximum number of 17 on the first day that all pools were sampled to 8 on the last day of observations. Individual pools tended to vary in their decline, with pools i and 4 maintaining the same population as the frst day observed. Fidely Fidelity to the tidepool of firstrelease is shown in Table 1. Out of the 31 tota number of fish tagged, 21 individuals were observed for 4 days or more and were Included in the fidelity calculation. Ofthe 21 fish, only 3 were found to have a 50% or lower fidelly to a particular tidepool, and 18, or 86% of the population, had a greater than 60% fidelity to the tidepool in which they were first captured. Length varied with fidelly, with the fish In the 4.0 to 4.9 cm range showing a 9125 fidellt while those In the 5.0 to 5.9 range showed onty 56% fidelit (Figure 4). One must note, however, the small sample size of those in the 5.0 cm range, for 3 out of the 6 tagged fish in this length range were never seen more than 4 days and were thus excluded from the calculations. Men Paterns Because watched the pools on a dally basts, could ascertain distinct movement patterns among individuals (See Figure 5). Over the course of the study, only 3 fish were seen In any of the other 4 experimental pools. On each occasion, however, never again saw that individual in the pool of its first capture, suggesting that it had perhaps relocated to a different pool entire ly. 4 individuals each missed 2 consecutive lowtides at the beginning of my observations, but then returned to ther original pools and appeared there on an every day, every other day basis. Only one fish missed 3 low tides in a row, and this individual also returned to a regular schedule of every day appearance at low tide. The great majority of the fish, or 67%6, however, were never gone from their pool for more than a day, with some individuals appearing consistently every day such that lalmost expected to see them on each observation. One of these indlviduals showed an exceptional 100% ridellty by Inhabiting the pool every day for each of my 12 low tide counts (See Appendix 4). Dipkeen On the next low tide after the displacement experiments were performed, all tagged individuals were gone from Pool 2. Upon examination of Pools A, B and C, observed 100% return of beaded fish in Pools A and B, and 2 of the 3 returned to Pool C. The fish that was missing from Pool C was the smallest individual captured, measuring 2.7 cm in length (See Appendix 5). Mid Tide Observations: RRTTAY Ontwo occasions lwatched Area A next to tidepool 3 as the tide level moved lower than this point. Ofthe 7 fish that were subsequently observed to be in the pool when it was completely isolated, 1 saw 4 individuals of Pool 3 in this area approximately 20 cm away from the lip of the pool. The surge was high as the waves moved Into this area, and at times the fish were no more than 2 cm under the water. At one time the surge moyed out so quickly as to leave one individual standed without water for several seconds until the next wave came in. At this time the fish moved towards the lip and eventually down into the pool. Actlviy seemed to be directed to moving slowly down into the pool with each lower surge, and occasionally saw some of the fish exhibit a feeding behavior on some neighboring algae. On one instance saw a tagged fish from Pool 3 Inhabiting Pool 4. Before the pools were isolated, however, the fish moved from Pool 4 into its original pool and remained there till the next high tide. lalso observed the tide as Itcame In over Pool 3 and flooded Area A. As the water moved higher, individuals tended to orient themsewes vertically with thel heads up on the sides of the pool, and with every surge they moved higher on the rock and closer to the surace. As was observed during the tansition as the tide wert out, the fish tended to remain only 2 to 3 cm underneath the surface of the water unti the pool was flooded enough so that they could move into Area A. 5 of the 7 individuals observed at the previous low tide, all within 3.1 to 3.5 cm, were the first fish out of the pool and they were seen in Area Aas soon as t had been covered with water. The same fish that ! had seen move from Pool 4 to Pool 3 as the tide went out swam over the bordering rock and again exited Pool 3 into Pool 4. Activity was conslderable, with the fish actkely moving up with the water as well as feeding at areas not accessible to them at low tide. Upon looking into Pool 3 at this time, also noticed à large, newfish present that had not been there before at the low tide. High Tide Observations: RRTIKY At high tide, the fish within one meter from the edge of the pools were all sitting very stationary on horizontal rock faces or sometimes positioned obviousyy in shallow cracks In the granite.I saw minimal movement at this time, and lwas even able to snorkel directly over the fish without apparent disturbance. did not see fish hidden in the algae or beneath rocks or debris, but saw them instead siting in exposed areas of the study site. On 10 of the 11 high tide observations made, discovered one or more large sculpin » 10 cm in length in the bottoms of the pools, and once lobserved an eel blenny in Pool 3. Surfperch were also seen swimming in the area over my study site. Cunsand Dan hon B The number of fish inside the pools at high tide was exceptionally small. On any given day, the percentage of fish out of their pools was consistently over 80%, with 3 occasions of 100% vacancy (See Table 2). Most of the fish, ranging from 4224 missing on one occasion to 91% on another, disappeared from the entire study area, and a maximum of 7 and a minimum of 1 individual seen on any one day. Of the one third that remained, 95% were seen within 50 cm of their pool of first capture. Fish inside the pools ranged in length from 3.5 to 5.5 cm and 4 individuals were seen twice each in their respective pools. DISCUSSION: Fromthe above resutts it ls apparent that over a course of 18 days, a majorit of Chtus anais individuals have a high fidelity to specific tidepools and regularly return to them at low tide. From the resutts of the displacement studies observed fidelly to a tidepool was not due simply to the fact that those fish were replaced there at low tide after capture; hence, the appearance of an individual tice at à certaln tidepool Indicated that other factors affecting fidellty were Involed. Upon comparing these findings with those of Richkus (1978),was surprised to discover the remarkably small percentage of fish that missed more than 1 consecutive day away from the original pool of capture, for he reported that samplings in bo week intervals showed a low percent of recapture of tagged individuals in their original pools, and he further demonstated a rise in this percent recapture as he sampled al pools In the area. It is interesting to note that ! found only 3 individuals out of 31 that appeared in any one of the other 5 pools observed, comprising only 14 % of the population, as opposed to the 67% of tagged fish that appeared only in their original pools and occaslonally moved elsewhere out of the study site for no more than one low tide. These findings suggest that the daily movements of individuals over an approximately two-week perlod follovs a markedly different pattern than those reported for the same specles in a different habitat, sampled biweekly. An alteration of the earlier schematic due to these new findings suggests that, on the time scale observed, the majority of individuals have either one or more pools to which they are faithful outside ofthe study area, or they have a series of pools in which they reside only bansienty in addition to their primary pool of residence (Figure 6). As it has been reported that an undistured population of an also declinec over a similar period of time (K. Jensen, 1989), the population decline observed ma not be due to the tagging disturbance but can possibly be attributed to the longer- term Intertidepool migration suggested earlier. To determine this, however, periods of dally observations would need to be continued for more than bo weeks and a greater area of tidepools assessed for appearance of individuals. High tide data shovs that members of the Catis anats species cleart do not inhabit their tidepools at high tide and the majority of them do not reside in the area one meter adjacent to these areas. Ir the fish are not in their pools nor are in the immediate vicinity, they must move to areas greater than one meter away, suggesting that perhaps mechanisms of homing are used to direct individuals back to their primary pool of residence. Results from the displacement experiments clearty support the presence of navigational ability sufficient to carry out homing, for within one tidal cycle, individuals as small as 2.8 cm were able to successfully return to the ir respective pools over 16 meters even though the pools themselves were separated only by 50 cm in distance. It would seem plausible that an accurate homing mechanism serves a key evolutionary purpose in these intertidal fish, for if they are accidentally swept away from their pool of residence or bavel conslderable distances In search of food or other attractants, it would be adaptive to have a mechanism to bring them back to a familiar area of the intertidal so that they are not caught above water or arbirarlly rind a tidepool that may not be physically advantageous (le presence of predators, minimal algae cover, insufficient oxygen levels, etc.). In addition, homing permits the exploitation of a larger area of the intertidal, anc movement patterns associated with the tidal cycles, weather, and seasons may necessitate homing as the fish search over a greater area for food resources. The surprising Invasion of cottids larger than 10 cm in length in places that were inhabited by identified fish raises questions as to the behavior and whereabouts of the tagged individuals if they were not within immediate distance of their pools. Observations made at times of tidal moyement in and out of the area elucldates aspects of the behavior of the cottids as their small world is suddenty expanded to a vast subtidal terrain. At these tansitory times, l observed the fish closely following the surface of the water, as if they were waiting until the very last moment when they could either exit the pool or enter it, and they were prepared to cope with both the forces of the waves and the possibility that they would get stranded without water for a short period of time In return for the possible benefits they recelved from reduced predation risk, feeding andlor engagement other unknown activities. I larger cottids Inhabit the tidepools and perhaps take advantage of their benefits at high tide because their own size prevents them from surviving there at low tide, it could be that the smaller cottids equally participate in a downward vertical shit to explolt the resources that can be galned there without competition or predation from larger fish. Although littie is known about the behavior of Chxxxtus ana at high tide and the reasons for frequent movement of individuals over long distances, t ls clear that these fish do not stay in their original tidepools at high tide and are particularly good enough homers to find a specific tidepool again at low tide and reside in it on a farrly regular schedule. These results, therefore, support an alternate definition of homing in htus anai for its behavior does not seem to fit elther a pattern of faithful homing or exploitation of any of a set of pools in a home range. Instead, Clixttus anats may alternate between onty one or to pools in an area that it considers its "home", and may reside there until the need for resources causes it to temporarily leave the area and return at a later time. Fidelity to these pools may alse be important in partitioning the resources of small refuge pools without overloading them. Much additional work needs to be done in the area of possible navigational ability and mechanisms of homing, however, and future observations of individuals during the tidal change may prove to be useful in elucidating much of the behavior of chxus anis that remains unknown. C LITERATURE CITED Green, J.M. 1971. High Tide Movements and Homing Behavior of the Tidepool Sculpin Oligocottus maculosus. J. Fish. Res. Board Can. 28, 383- 389 Jensen, K. 1989. Oxygen consumption of cottids as it re lates to the tidepool as an interactive organism. Stanford University Hopkins Marine Station Final Papers, Biology 175H. (unpublished) Morris, Abbott and Haderlie. 1280 Intertidal Invertebrates of 5tanford Unversiy Press, Stanford,CA california Richkus W. A. 1978. A Quantitative Study of Intertidepool Movement of the Wooly Sculpin Clinocottus analis. Marine Biology 49, 277-284 Smith, S.Y. 1974. Activit Patterns of M sa Stanford University Hopkins Marine Station Final Papers, Biology 175H. (unpublished) Williams, G. C. 1957. Homing Behavior of California Rocky Shore Fishes. Univ. Calif. Publ. Zool. 59, 249-284. 14 C 15 FIGURE LEGEND Study area on a section of the protected rocky intertidal at Hopkins Marine Station, Pacific Grove, California. Relative locations and dimension of experimental tidepoos l.-V. Population decline of originally tagged individuals over the course of the study period. % Fidelity of fish in three length ranges. Number of fish that missed U- 3 consecutive days in a row from ther original pool of capture. Schematic showing the dally movement patterns of Cllnocottus analls to and from the original pool of capture. Numbered arrows indicate the small number of fish that relocated within the study area. TABLE LEGEND Fidelity forthe 21 fish observed for 4 days or more. Number of fish observed in the tidepools at both low tide and the following high tide. Relative percentages are indicated. 8 8 . a- 8 8 L Fg V 2 N a K. -.. .. S +0.96m 10.73m Experimental Study Site (Moderately Sheltered) +1.08m Area A III. +0.64m IV. +0.62m Figure 2 10.78 C 4L Original Population Over Time 8 4 aaataaaaa. Date Figure 3 a 20 3.0 LENGTH OF FISH (in cm) Figure 4 5.0 C relocated CONSECUTIVE DAYS HISSED + Consecutive Days Hissed 1m Figure 6 O TABLE 1 Fish Fabserved loy tides * Times in pool 2 fidelity 16.16 62.50 66.67 64.29 100.00 100.00 100.00 72.73 100.00 81.82 81.82 100.00 90.9 77.78 25.00 75.00 66.67 100.00 50.00 66.6 He 8.62 76.465 6.6 =2 O DATE 5/22 Low High 5/23 Low High 5/24 Low High 5/25 Low High 5/26 Low High 5/29 Low High 5/30 Low High 5/31 Low High 6/1 Low High 6/3 Low High 6/4 Low High TABLE 2 Total * Fish 14 12 13 2 Out of Pool 92.86 88.24 94.12 100.00 81.82 100.00 93.33 81.82 84.62 100.00 85.7 Page D aai - 7 -1 -- — 0 - 3-5 S50 GUNNNNN - 2 - NN ————— So- 5 e 11 - + 8 NNNSN N Sn———---- Noeodun-o B eo 5 — . 2 — 4 O — e Moo — — — e eeee 9 4 oeoe 04 4 e 7 4 — — — — — — — r e er r eesles es enses 0j0n — — 0 04 en 4 5 4 4 C 0 APPENDIX 4 E1S44 MOVEMENT PATTERNS relocated every other day 8 days every da/ erradio then 7 ed/160 3 ed/eo 55 5 ed/e0 missed 2 days/3 ed/eo Odsys every da missed2/2 ed/eo 5 ed/60 Pundist ed 4 ed/eo 2 ed/eo 3 ed/eo relocated 3 ed/eo missed 2 /5 ed/eo 27 6 ed every other day relocsted =21 ed : every day o -eer other day r..- 11. F C O POOL A POOL B POOI C APPEMDIX 5 LENGTH (CA DISTANC 4.3 32 2.8 42 2.7 FROHPOOL 2 2 RETURN 16m 1002 16 m 1002 18 m 673