DISTRIEUTION (Embistecidae) IM THEF OF FOINT CAERIL ic Marine Stanferd University AESTRACT Io examine what factors influence surfperch distribution and behavier, I made cbservations of a protected nearshore area off Foint Cabrille, California at different times of the day and tidal cycle. Within the study area, there was substantial niche diversity in the members of the family The Embictecidae ameng different habitats of similar depths. black surfperch were found in the most pretected area of the study and in highest densities at low tide. The striped and pile surfperch were equal in abundance in all habitats. The rainbow surfperch were seen in small schoels on sandy bottoms. The reef surfperch were the mest abundant species of Embictocid: they inhabited a shallew nearshere region and concentrated in large schools at the entrances to channels during low tides. Niche diversification seems to be a relevant way of viewing dynamic fish distributions and an important factor to be considered in choosing methods for conducting pepulation studies. INTRODUCTIOM Embictocidae (surfperch) is a viviparcus fish family found only in the North Facific and fairly abundantly in Menterey Bay. A diverse population occurs in the protected nearshore area of Feint Cabrille. As with other viviparcus animals, the young are nurtured internally by their mather. These fish have a gestation peried of appeximately five te six months (Eurgess and Axelrod 1984), ending in the springtime. Live-bearing is an evolutionary adaptation to reduce early direct mertalitys thausands less eggs are fertilized yearly. Behavioral and dietary studies have been conducted in on tuo congeneric species, Ex jacksomi and E. lateralis. Hixen (1980) ebserved competitive interactions between the twe species in Santa Barbara and found the striped surfperch to be the dominant competiter aver the shallew algas-covered reef, leaving the black surfperch te forage aver deeper turf. Similar studies at Hepkins Marine Station hy Tom (1988) reported that the black and striped surfperch there share their available prey rescurces and de not differ significantly in distributien. Observations made by Nichels (1979) of the surfperch in a 20 square meter transect in the same area as my study reported sharpnese and shiner surfperch, beth found in schools, to be the most abundant Embictocids, follewed by pile, black striped, and calico surfperch. A personal communigation by C. Baster reported large schoels of shiner surfperch from 1974 to 1988 and an abundance of dvarf surfperch from 1985 tc 1987. MOTERTALS AMD METHODS From May 15 te June 3, 1989, I made cbservations of the distribution and behavier of the Embictocidae in a 3800 square meter section of the protected nearshore zone at Faint Cabrille, California (see figure 1). I used sporkeling and cccasicnally SCUBe diving gear and recorded my counts and observations on underwater paper. Observations were conducted at different times during the day and tidal cycie. The study area was divided into five separate habitats: the time spent in each was recorded and vemained fairly constant for each respective area. A strüctured pattern was follewed on each snorkel survey in order to aveid as much repetitive counting oft individuals as possible. Habitat A (370 square meters), the area under and arcund the railroad tracks, is very well-protected from surge and breaking waves by Seal Rocks. Ehyilospadix covers the shallow intertidal area of the tracksi algae grows all over the submerged tracks and ties which are approximately one foot above the sandy bottom. The area slepes down to a depth of 10-15 feet, where the tracks end and a small grove of Macregystis begins. I usually spent 15 minutes here. Habitat B (1600 square meters), the area protected by Seal Rocks, is marked on its southern boundary by a sand channel. The bottom is characterized by medium-sized and large rocks, thick Ehyllospadix, and sandy patches. Other than the sandy, shallow area near the shore, the approximate depth of this habitat ranges from 4 to 8 feet. Surge is greater in this area than at the railread tracks, but rarely strong. This is a popular swimming area for harber seals. My cbservation time here was abcut 30 minutes. C Habitat C (530 square meters) is the relatively unprotected area between Seal and Bird Rocks. Surge and suell increase fairly dramatically as one moves inte this habitat; waves are often breaking just seuth of Eird Rocks. The depth here is 10 to is feets it contains the sparse edges of the kelp forest and some Ehyllespadix covered rocks. I spent 10 minutes observing this area. Habitat D4(730 square meters), the ares protected by Eird Rocks, is marked en its southern beundary by Fete's Rock. It is a shallow area of approwimately 5 feet in depth with parts being too shallou to sporkel at low tide. The bottom of the southern side is light-colored and sandy, while Ehyllespadig grous thickly in the northern section. Although this habitat is protected from breaking waves, the water currents can be very strong as the water moves through the channel in one direction to the seuth. Observation time here ranged from 10 to 15 minutes. Habitat E (570 square meters) is the channel beyend Bird Rocks that leads into a more exposed section af Monterey Bay. Ehyllespadik'is heavy, but as the area sicpes down to its maximum depth of 15 feet, the surfgrass divides to line the edges of the sandy channel. The channel cpens up onte sone scattered Macrscystis plants which mark the end of my stud, area. This is the mest vielent water of my study: water rushes between the rocks at high speeds, the surge and swell are very large, and waves are often breaking at the mouth of the channel. There is a small, more shallow and protected area on the west side of section E near the shore. I cbserved in this channel for approwimately 10 minutes. RESULTS Threugheut the entire study, I saw a total of li species of Embictocids (see table 1), five of which were abundant encugh for me to hypothesize about their distribution. These five species (the black, striped, pile, rainbow, and reef surfperches) and their average densities in the entire study area are presented in figure 2. The average densities of each species in each habitat are shoun in figure 3. The large variations display that the five diverse habitats contain very different distributions af surfperch, indicating considerable niche diversity among the Embictocids of this region. The graph of percent abundance (figure 4) shows the habitat(s) in which each species seems mest successful er brefers to live. The results are presented separately for each of the major species. Tables a through 7 list the complete data from each habitat. Rlack Surfperch (Embigtsga jagksoni) Altheuch the black surfperch were the fourth mest dense in the entire study area, they had by far the greatest density of any fish in any habitat when found beneath the railread tracks. Likewise, the protected area under the tracks was the enly habitat in which their density was substantial. As the habitats became rougher and less protected, their respective densities seemed to decrease. An interesting and statistically significant correlation was discovered between the number of black surfperch under the railroad tracks and the height of the tide. (See figure 5) Os the tide rises, the number of Ex iacksani beneath the tracks decreases. I usually found these fish concentrated in tight groups of abeut 15 under each of the algae covered railread ties and maintaining a pacing behavior back and forth to either side of the ties. Alang with the relatively calm water in this habitat, the tracks seemed to afford these perch a protected area and a large, open, horizontal plame of water to inhabit. They were all adults (20 to 30 cm in length): I once saw a large black chase a smaller black from beneath a railroad tie. I recognized four pregnant females and one juvenile.1 witnessed black surfperch feeding both by picking and winnewing. On seven sightings threughout the study I cbserved Ex iacksoni moving in odd manners: quivering and flipping to one side seemingly scratching their dorsal fins. Interestingly, I sau higher numbers of black surfperch than usual in the ares protected by seal recks during my two nicht snorkels. They seamed less active nocturnally than any other species. Striped Sumfperch (Embiotoca late 5) The second most dense species in the entire study area, the striped surfperch,, were consistently dense in each habitat and correlated closely in their density and abundance te that of the pile surfperch. They were mest dense in the channel beyand Bird Rocks follewed by the ares between Bird and Seal Rocks: the twe most expesed habitats in the study. Interestingly, 507 and 457 of the Ex lateralis found in those respective regions were juveriles. This species' nest mest dense habitat was beneath the railroad tracks with their congeneric species, the black perch. The bighest density, of pregnant fish were the striped surfperch cbserved ih this areat 457 of them were visibly pregnant. Because of an injury which marked her, one pregnant striped surfperch was seen swimming under the same railroad tie on four separate shorkels which spanned three consecutive days. A påttern that seemed to emerge but did not vield statistically significant evidence was that as my study period passed, the number of pregnant striped decreased and the number of juvenile striped increased. I saw very feu pregnant or young-of-the-year fish of other species in comparison to the numbers I saw of E. lateralis. I seem to have witnessed the end of the striped sürfperch's five- to six-menth gestation pericd. I cheerved on at least seven occassions a pregnant striped surfperch with an injury such as a bloody gash in its musculature. The juveniles often traveled in schools of approximately 10. I witnessed striped surfperch feeding both by picking and winnewing. Oddly, near the end ofmy study period I began to natice many striped perch that seemed very light in coler or bleached out.. I once noted a harbor seal chasing an E. lateralis. Eile Surfperch (Damalichthys vacca) The pile and the striped surfperch were consistently similar in density and abundance throughout the study area. Like the striped, pile surfperch were found to be most dense in the twe mest exposed habitats: between Bird and Seal Rocks and the channel beyond Bird Rocks. Next, they were observed in almost identical density as E. lateralis in the area protected by Seal Rocks. I observed less pile than striped in the railroad tracks habitat and a different behavior: the pile were seen mainly as individuals swimming around the railroad track area as opposed to beneath the tracks. The pile surfperch were the most dense fish seen in the habitat protected by Bird Rocks, the area which contained by far the lowest numbers of surfperch. D. Vacca, as well as E. lateralis, seemed to thrive primarily in the rough, exposed areas but.able te live successfully throughout the study area. The pile surfperch often swam in small groupsiof five to fifteen. I cbserved four juvenile and tuc pregnant pile surfperch. Haimbow Surfperch (Hypsurus carvil The rainbew surfperch were the lesst dense of the five primary Embictocids in the study area. They wereichserved in greatest density in the exposed habitat between Eird and Seal Recks in scheols of 10 to 20 on the sandy botton. The same behavior was observed in the area arcund the railroad tracks and en the scuth edge of the Seal Rocks habitat gnithe border of the sand channel. I often observed rainbow surfperch feeding by picking at the algal turf of the sandy ocean floor. H. Carvi were rarely seen cutside of their schools as individuals. I chserved two noticably pregnant rainbew surfperch. Reef Sunfperch (Micrometrus aurera) The most dense Embictocid in the entire study ares was seen primarily in the area protected by Seal Rocks. A statistically significant correlation was found between the number of reef surfperch in this habitat and the height of the tide: the lower the tide, the greater the number of fish inhabiting this area. (See figure 6) During leu tides, the reef surfperch were observed in large scheols of 50 to 100 in the shallew, sandy, nearshore section. Saveral times I noticed the presence of these schools while standing on the rocks nest to the water at lew tide. The ather fairly abundant appearance the reef surfperch made was in the channel beyend Bird Rocks. Here, they appeared in the more protected section of the channel. In this habitat as well in frent of Seal Rocks they seemed to be more abundant at low tides. (See figure 7) Altheugh the correlation is not statistically significant in the channel, the data suggests a trend that I believe would be supported by more ebservations. DISCUSSION The great variatign in densities of different species in varicus habitsts displays the existence of niche diversity among the five co-occurring Embictocidae species in the stüdy. The protected nearshore zone contains a dynamic population of surfperch, the patterns and distributions of which are not always ebvious from casual observatign. The black surfperch showed a clear abundance in the prttected area beneath the railroad tracks during low tide. They seemite be less successful in rough, exposed areas. Eventheugh I have been SCUEO diving and srerkeling in all areas of immediate prowimity to the railroad tracks, Iam baffled by the question of these black surfperch's habitat during high tide. I think that more distribution studies of the greater surrounding area would be revealing. Eoth the striped and the pile surfperch appeared to be mest successful in the mest esposed habitats altheugh these tue species, more than any ather, were relatively everly distributed threugheut the study area. The strong correlation of these fish seems interesting not only because that trend has not been reported in other areas or at Foint Cabrille before, but alse because these two species have significantly different diets. While the striped surfperch feed from the algal turf, the pile surfperch appear to be selective visual feeders on larger spails and hermit crabs. (Nichols 1979) This separation apparently permits them to expleit their respective rescurces witheut competition. The rainbou surfperch, though nat as abundant as the ether four species, ware consistently found in small schools on sandy bottoms in depths of ten or more feet. Although they did not tecur in the exposed channeli past Eird Rocks which dees have a sandy bottom, they appeared to be successful in the rough but less-exposed area between Bird and Seal Rocks. Ferhaps the rainbow surfperch can live successfully in schools en any sandy bettom se leng as the surge or reugh water is under some critical point. The mest dynamic pepulation af Embistecids was the reaf surfperch: depending on the height of the tide, their numbers ranged from zero to 250. Figure é suggests that there could exist same critical tidal level such as 1.5 feet under which the large schools of reef are always in the Seal Rocks habitat. As with the black perch, I have anly speculative insight inte the high-tide habitat of the Migrometrus auroras they could be broadly spread through the complen rocky shallow water habitat. Accerding te the personal. observations of Chuck. Baxter, large schools of dwarf surfperch (Micrometrus mininus, congener of the reef surfperch) inhabited the nearshore zone off of Feint Cabrille from 1985 to 1987. He also reported similar large schools of shiner surfperch, related in size and appearance to the reef surfperch, from 1974 to 1983. This spring, I cbserved very few individuals of those two species: estreme annual variaticn is gecurring among the populatiens of smaller surfperch.' Future cbservations of the What present schoels of reef surfperch would be interesting. are the seasonal and yearly patterns of their population structures to there de they relccate; scheols of which, if any, species will replace them?, De emmigration and immigration play a large rele in structuring these pepulatiens or are they dependent on the dynamies from reproductive recruitment? The unique life cycle of the Embietocids, which includes the viviparous production of advanced juveniles, makes their population dynamics an interesting comparison to that of other resident species. Observations or habitat manipulations could be conducted in order to determine what factors dictate these yearly variations of high magnitudes. The results concerning the reef and the black surfperch display that the tidal height is a significant variable affecting at least two of the surfperch populations in my study area. I found ne conclusive distributicnal effect af the time of the day, but because of my low numbers of night¬ time cbservations, still believe that the ammeunt of light cculd have and effect on surfperch distribution. Figure 5 presents the density change of Embictecids aver one decade by comparing the average densities that I observed with these recorded in 1979 in another study of a transect in 0 the same regich (Nichols 1979). Whereas the density af black surfperch seems to have remained constant, the densities of striped and pile surfperch have increased substantially while remaining generally consistent with each other. Neither the rainbow nor the schools of reef surfperch were observed in the transect in 1979: it seems that large schools of shiner and sharpnose surfperch were'seen instead. This data cerrelates with the Baxter observations of schools of shiner perch in the study area over the last fifteen years, while this year it seems that few individuals remain. Nichels alse reported scheols of sharpnose surfperch and some appearance of calice surfperch, neither of which species seem to exist aff of Feint Cabrillo this spring. It is interesting that the shiner and sharpnose perch bear size, shape, and some dietary similarities to the reef surfperch and Correspondingly, the rainbew and calico perch share size and pattern characteristics. Nichols' data showed ancther interesting decadal variation that at the same time of year in 1979, larger densities of black and pile juveniles were observed than striped juveniles. (Nichols 1979) This comparisch is contradictory to my observations of large numbers of striped juveniles in 1987. (See figure 9) Eecause the number of pregnant striped decreased and the number of ycung-of-the-year striped increased as my study progressed, I believe that I witnessed a critical peint: the end of E. lateralis' five- to six-menth gestation period. The lack af literature reporting any conclusive dates of the c beginnings and ends of the different species' gestation perieds made my theery difficult te cenfirm. Nichols observations as they relate to my data suggest that gestation does not occur at precisely regular intervals each year, but rather may be dependent on weather, temperature, er ether seasonal variables. Consecutive annual spring ebservations could yield a clearer picture of variability of gestation periods. Intuitively, it seems that in order to bensfit from their avelutjomal adaptation of viviparity and to pretect their small yearly batch of fertilized eggs, the pregnant female surfperch weuld be required te change their behavicr during the conclusion of their gestaticn period. The ventral sides of the pregnant surfperch I chserved were ehvieusly distended and lighter in coler than the rest of their bedy: at this hlcated period they were inevitably less hydrodynamig than usdal, Therefore, it seems appropriate that the pregnant striped surfperch eccured mest densely in the most protected habitat in my study: the railroad tracks. The comparitively high numbers of injured pregnant Exlateralis suggests either that these pregnant perch cannot swim as well as their lithe geunterparts, that the bleated fish are more attractive to their predators, er both of these hypotheses. One suprising result was the different distributions of the twe congeners, E. iacksoni and E. lateralis, in different habitats with approkimately the same depth. According to Hixen (1980), competition is the cause of these c tuo species' segregation by depth in Santa Earbara, California. There, Ex lateralis dominates the shallow, more desirable reef, pushing Ex iscksoni inte deeper water. From recent studies at Hepkins by Tom (1788), the black and striped surfperch off of Foint Cabrille shared an identical fesding niche and did not differ significantly in distribution. My cbservations revealed that the two species were significantly differently distributed aver a lateral area of similar depths. Combining the gut analysis from Tom (1988) and my results, these twe fish are eating similarly in different places: factors other than food preference are determining their habitat cheice. The differing results that Tom (1988) and I found could indicate the differences in cur methods of cbservation and perspectives on cur respective transects. Her observatiomal esperiment was done by cbserving a quadrat of one continuous habitat. This comparison demonstrates clearly that the cheice of transect and method of analyzing it dees affect the results of a transect experiment. The niche diversity thåt I cbserved, specifically among the black and striped surfperch, displays that different potential transects which, from initial observations, lock very similar, in fact have very different surfperch populations. One inevitable concern of observational studies is the cloudiness that surrounds the effect of the observer on the behavier of the subjects. I can only hope that I scared all species away in equal proportions so that my ratios have remained consistent. The fish did not seem daunted by my presences when I was SCUBe diving and cculd lie en the bettem relatively meticnless, surfperch came within centimeters of me, Gnother concern is the effect of the harbor seals on the number of fish that I saw. In the area in front of Seal Rocks, seals regularly follewed me, surrounded ne, and generally demanded my attention as I attempted to observe and count Embictocids. Mest fish did not seem too distürbed and I continually found high pumbers of fish in this habitat. Tuc other hinderances of my Vision were the often poor visibility in the water and the excessive presence'af phyllespadis, meither of which varied significantly. C BIELIOGRAFHY Baltz, Donald M. and Moyle, Feter B. 1982. Life history characteristics of tule perch (Hysterecarpus traski pepulations in centrasting envirenments. Env. Biel. Fish. Vol. 7, No. 3, pp. 229-242. Behrens, David W. 1977. Fecundity and reproduction of the viviparous perches Hypsurus caryi (Ogassiz) and Embiotoca jacksoni Agassiz. Calif. Fish and Game 63(4): 234-252. Eurgess, Dr. Warren C. and Axelred, Dr. Herbert R., 1784, ishes of California and Western Mexice, Facific Marine Fishes, Book 8, T. F. H. Fublications, Inc. Ltd., pp. 2oa9-2039. DeMartini, Edward E., Moore, Thomas O., and Plummer, Kenneth M. 1783. Repreductive and growth dynamics of Hymærpreschen ardentem (Embictocidae) near San Diege, California. Env. Biel. Fish. Vol. 8. No. 1, pp.29-38. Mark A. 1980. Competitive interactions between Hixon, California reef fishes of the Genus Embictoca. Ecology 61(4): 918-931. Mark A. 1981. An experimental analysis of Hixon, territoriality in the California reef fish Embigtega jacksani (Embictocidae). Copeia 1981(3): 453- 665. Laur David R. and Ebeling, Alfred W. 1783. Fredator-prey brelationships in surfperches. Env. Biel. Fish. Vol. 8. No. 3/4, pp. 217-229. Nichols, Andrew W. Spring 1979. Fredation by tidepcol,and nearshore fishes and its impact on the rocky intertidal zone community. Tiffany W. C. May 1988. A regional absense of Tom, competition between congeneric surfperch sharing the same dietary preferences and spatial distribution. M. S. Thesis. Common 81 ac Striped Pile Hainbow Reef Duarf Shiner Barred Silzer Kelp Rubberlip SPECIES OF EMBIOTOCIDAE OBSERVED IN STUDY AREA Scientific Embiotoca jacksen. Embiotoca lateralis Damalichths vacca a Hicrometrus aurora Nierometrus minimus Cmataste a Aphistichus argente Hyperpresopon elliptiu Brachyistius frenat Hhaochilus touotes Table! HABITAT A: UNDER THE RAILROAD TRACKS tidal 2 8 date time level 15-1 1515 15-2 1545 1400 29 1.1 1430 2.6 00 1100 1.7 10 .5 1800 20 1600 3. 1030 0.5 17 26-1 1115 0.4 26-2 1515 J.7 24-3 1900 27-1 1215 0.7 29-2 1545 29- 1730 4.8 750 0-1 0-2 12 11 10-3 1425 I.5 -4 1750 31-1 00 T.6 1-2 1410 1.8 18 1 1-1 -1 1 —— QVERAGE IS.S 0.20 7.75 3.45 0.62 4.83 1.91 0.75 0.04 0.75 DENSITV O.O o.00 0.O2 0.00 o.00 0.01 o.00 o.00 o.00 o.0o Table 2 5 HABITAT E: GREA FFOTECTED BY SEAL ROCKS 2. tidal date time level 1500 2.e 11 0 38 1120 1.8 1400 . .4 4 O 16 8 11 1130 o. 20 26 1715 5-2 29-1 120 e 5. - 10 S8 24 - 1730 4.3 0-1 o 0 42 . 30-2 1245 1 4 1.4 0-3 1645 . 2 0-4 2000 8 5. S1-1 710 31-2 1425 8 117 54 4 13 40 2. 18 O1-1 - 71 1 2 10 QVEERGE 7.61 oaa- LEMSITY o.o o.o . .00 .1 .0 . .0.0 Table 3 HABITAT C: BETWEEN BIRD AND SEAL ROCKS tidal S8 date time level 1145 2.1 0 1 .8 68 14 7. 1700 50 0 O. 1 1145 15 29-1 100 S1 1.3 O +2 160 T.4 -00 4.4 0-2 1 1. 30-3 1710 26 -36 3.8 51-1 240 . - 00 10 01-1 1.8 o1-2 10 5.2 1515 S 14 24 T.2 QVERAGE 2.86 15.1 1 7.8 15.1 5.0 DENSITY 5.00 o.0S 0.00 0.01 0.02 0.01 0.00 HAFITAT D: AREA PROTECTED BY BIED ROCKS tidal P date time level 200 15 40 1430 12 1700 1600 5.6 9-1 1315 1.4 29-2 1840 3.5 -00 4.4 70-1 845 2.9 30-2 1230 1.4 0-3 1 0-4 20 4.3 1-1 950 O 7.2 18 1-2 1510 O 01-1 25 1. 5. 01-2 1225 0 14 15 C QVERAGE 7.06 0.Oe 1.33 8.2 6 0.06 0.08 0.08 0.08 DENSITY 0.00 0.00 0.0 5.01 0.00 o.00 .0 o.o .o0 Tables HABITAT E: CHONMEL FAST EIFD ROCKS tidal date time level 29-1 1. 4 15 104 1.5 29-2 1650 O 3.6 29-3 2015 4.3 10 30-1 00 O 2.8 45 38 30-2 1340 2 1.5 3O-3 1730 4.1 3 17 - 00 3.2 31-21570 39 34 2.7 18 17 45 01 1.5 O 2 45 0 O o 00 18 AVERAGE 1 20.3 1.1 10.2 14.1 16.4 0.5 DENSITY o.oo o.0 0.00 0.01 0.02 0.02 0.00 Table e COMPOSITE DATA FOR THE ENTIRE STUDY PERIOD DENSITY (fish per ten sq m): striped pile rainowre black o.09 average O.19 O.17 O. 04 O.21 tracks O.21 O.13 o.05 0.02 O.60 seal O.OS O.14 O.14 O.36 O.04 O.O5 O.30 o.29 O.11 O.02 between bird O.00 0.10 O. 11 o.o0 O.08 O.2 O.02 O. O.2 channel O.O1 ABUNDANCE (by fish): striped black rainowreef pile 65. 7.45 O. tracks 10.60 12.80 2.65 seal 43.60 .00 34.2 70.42 79.75 8.57 22.00 J.45 1.20 between 12.89 O. 45 O.2 7.70 bird 7.36 27. 2.99 21.89 T.40 channel 20.10 total 0.00 00.00 00.00 100.00 .o0 Table 7 SSION STATISTICS RE Correlation between black surfperch under the tracks and the height of theie Regression Output: 42.090 Constant Std Er ofYEst 12.41612 R Squared O.389240 No. of Observations Degrees of Freedom X Coefficient(s) -7.25476 Std Err of Coef. 1.937537 Correlation between the number of reef surfperch in the area protected by Seal Roc and the height f the tide Regression Output: Constant 159.3666 Std Er of Y Est SS.03032 R Squared O.405103 18 No. of Observations Degrees af Freedom 15 33.7242 X Caefficient(s) Std Err of Coef. 10.21691 Jable 8 c FIGURE LEGEND Figure i - Hap of the protected nearshore zone off of Foint Cabrille which constituted the study area. Habitat boundaries are shewn. Figure 2 - Graph of the average density of the five primary Embictecid species found in the entire study area. Density measured in fish per 10 square meters. Figure 3 - Graph of the average density of each of the five surfperch in each respective habitat. Density measured in fish per 10 square meters. Figure 4 - Fercent abundance measured solely by number of fish observed in each habitat. Neither the size of the habitats nor the differing total pepulations of the species is taken inte acceunt. S Correlation between the number of black zurfperch Figure cbserved under the railroad tracks and the height of the tide. The line of regression was calculated and drawn. Regression statistics presented in Table 8. Figure 4 - Corrolation between the number of reef surfperch observed in the area protected by Seal Rocks and the height of the tide. The line of regression was calculated and drawn. Regression statistics presented in Table 8. figure 7 - Statistically unsignificant correlation between the number of reef surfperch observed in the channel beyond Eird Rocks and the height of the tide. Figure 8 - The density change of the surfperch or Hepkins Marine Life Reserve from 1979 to 1789. The 1979 densities are from Nichols (1979). Figure 9 - The density change of the juvenile surfperch of three speciez of zurfperch on Hepkins Marine Life The 1977 Refuge from spring 1979 to spring 1989. densities are from Nichals (1979). u g D ULE X— PAD OUu Lo5 — ODO at 1 ou 88 990. 1 . 2 L I 1 prolod- ododcoo 388 ooOO6 L L L H . E 5 2 L C Z E L a- I L 1 e; 3 E 41 8 8 9 8 HSU 30 S3GWON 9 L a O — L L I atavaa- 8882908 9 — HSIA 30 SBONON L a L L L E 0 kakakatatakatooaa- 0 HSIS 30 SBOMON N L L - oonn-- ooooo ooooodoo b 01 1d L o 2 D oronto- ooooo oood lbs o1 d su+