Abstract: Traditional ecological models developed in rocky intertidal systems state that species ranges are determined primarily by physical stress in upper tidal zones and biotic interactions in lower tidal zones. Studies of the barnacles Balanus and Chthamalus show that smaller chthamaloids are able to persist in a larger range of habitats than balanoids because of their greater tolerance to physical stress and lower susceptibility to predation. Ttested this model on the barnacles Balanus glandula and Chthamalus spp. at two sites in Monterey Bay which differed in wave exposure and abundance of the common predatory snail, Nucella emarginata. I ran lab experiments to determine prey preferences of N. emarginata, and measured densities of both species of barnacles and of the whelk along transects in the field. B. glandula and Chthamalus spp. distributions were positively correlated, indicating that competition between the two species did not play an important role in determining distribution. Nested ANOVA's of barnacle densities revealed significant differences between sites and among replicate transects within each site for both species. Although whelks were abundant in the exposed site, they did not play a major role in determining barnacle abundance. Lab experiments showed that Nucella emarginata had a strong preference for B. glandula over Chthamalus spp. As this difference did not manifest itself in differential distributions in the field, I conclude that factors other than whelk predation alone regulate the distribution of these species. A strong recruitment event in spring 2001 of B. glandula enabled me to compare distributions of B. glandula adults and recruits. No strong difference was detected. The very high variance in barnacle density in both sites may indicate that barnacle distributions in Monterey Bay are regulated by multiple complex factors. Further research is needed to understand these interactions. Introduction: The concept of distribution is fundamental to ecology. One of the first questions we ask when we are attempting to understand a new species is, where can we find it? Although exact ranges are difficult to determine, it is fundamental to our understanding of écology that we can consistently expect to see certain species in certain habitats in certain regions. The factors that determine species' ranges are complex. However, certain systems lend themselves to the study of ranges. The rocky intertidal is an excellent model system for studying the factors that determine a species' range because biotic and abiotic conditions vary greatly over a very small spatial scale. Species' vertical distributions are determined over a narrow band of coastline where conditions vary from the permanently submerged subtidal to the almost permanently emerged spray zone. Many classic ecological studies have used the rocky intertidal as a model community for studies of the determination of species' range. The work of Connell (1961a, 1961b, 1970) on rocky shores in Scotland and Puget Sound has been hugely influential in intertidal ecology and community ecology in general (i.e. Menge & Branch, 2001). Successive long term studies of the dynamics of barnacle populations established a model for the determination of species range. Barnacles of the genus Balanus were observed to be competitively dominant over barnacles of the genus Chthamalus. In Scotland, where both barnacles were present. Chthamalus was overgrown by Balanus , which is a faster growing barnacle genus (Sanford, 2001). Chthamalus populations persisted due to their increased tolerance to physical stress - which enabled them to live higher on the shore than Balanus. (Connell 1961b) Predation by whelks of the genus Nucella (formerly Thais) had a strong effect on populations of Balanus in Puget Sound. Whereas most mortality of Balanus in Scotland were due to overcrowding, most mortality of Balanus in Puget Sound was due to predation. As a result of this predation, Balanus was limited to areas above the zone in which Nucella was able to persist, thus much like the Scottish chthamaloids, Pacific Northwest balanoids persisted largely in a refuge zone (Connell 1970.) The effects observed on barnacles in these and other studies were largely responsible for the creation of a central paradigm in intertidal community ecology, which has been widely applied: upper limits of species' ranges are determined by physical stresses (i.e. heat, dessication). while lower limits are determined by biotic factors (i.e. competition, predation). Preliminary observations made in April 2001 at Hopkins Marine Station in Monterey Bay caused me to question the applicability of Connell’s observations about Balanus, Chthamalus, and Nucella to the local ecosystem. I set out to test whether predation by Nucella emarginata might be limiting the distribution of Balanus glandula, and in turn, whether competition with Balanus glandula was limiting the distribution of Chthamalus spp. Methods & Materials: Data were gathered at two sites in Monterey Bay, California. The first site was on China Point, a part of the Hopkins Marine Life Reserve at Hopkins Marine Station of Stanford University. China Point is the location of long term research on intertidal ecology and ecophysiology, but is off limits to non-researchers, and is therefore relatively undisturbed. The part of the point sampled is an area of protected outer coast (Ricketts et al. 1985). The second site was in Monterey harbor, just south of the Coast Guard Breakwater. This area is very protected from waves and storms, and is in an area of high human activity. However, the section of rocky intertidal used was relatively inaccesible to foot traffic, and appeared to have little direct human disturbance. Sites were chosen to provide a contrast of whelk abundances. Preliminary observations indicated that whelks (mostly Nucella emarginata, but also limited number of other species) were abundant at China Point, but were completely absent in the harbor. In order to determine relative abundances of different species, I sampled transects perpendicular to the shoreline that ran from the lowest barnacle I could find to the highest barnacle. At China Point, it was not always possible to reach the lowest barnacle due to strong waves action and treacherous rocks. I never found a barnacle growing below the highest feather boa kelp (Egregia menziesii), and thus used this species as a proxy for lowest barnacle in these highly exposed areas. I also ran a horizontal transect through an area of high whelk abundance in order to expand sample sizes. On China Point, I divided sampling equally between high barnacles (growing above the highest clumps of the alga Endocladia), mid zone (growing in the Endocladia zone), and low zone (below the highest barnacle of species Tetraclita). Within zones, I sampled randomly. In the harbor, the distance between highest and lowest barnacles was less, and no clear community zonation was evident, so 1 simply sampled randomly along the transect line, from the lowest point where barnacles were found to the highest. I sampled whelks in a .25 m’ quadrat. I chose one .01 m’ quadrat randomly from within the .25 m’ quadrat to count barnacles. All barnacles of species Balanus glandula and Chthamalus spp. were counted. No attempt was made to differentiate between C. dalli and C. fissus, as they are ecologically similar and impossible to differentiate in the field (Sagarin 1999). Barnacles greater than 4mm in diameter were classified as large. while barnacles less than Imm in diameter were considered to be recent recruits. This size classification was based on preliminary observations of whelk feeding preferences. After sampling, the height of each .01 m’ quadrat was measured relative to bolts placed at known heights above Mean Lower Low Water in the intertidal, 1 conducted lab experiments to elucidate whelk prey preferences. Barnacle- covered rocks were taken from the harbor area. I found rocks in the harbor with a mixture of Balanus and Chthamalus of a variety of sizes. I then mapped all of the rocks. and placed them in flow-through tanks. Each tank contained two rocks, with the goal of giving each tank approximately the same surface area of rock and cover of barnacles. In all, the six rocks used had 1869 barnacles attached to them. Each tank contained five whelks collected from China Point. Whelks were all similar in size, and sizes ranges were similar in all tanks. The tanks were covered, and the whelks were left to feed for a week on the barnacles. I checked the tanks twice a day to monitor the feeding progress and to make sure that whelks were not escaping. When whelks were found on the outer edges of the flow-through tanks, they were returned to the bottom, to encourage them to climb around on the rocks. The experiment was terminated when approximately half of the large Balanus glandula were eaten. After the termination of the experiment, I tallied the number of barnacles on each rock that were dead. I used a small dentist's tool to poke the opercular plates. If opercular plates were missing or fell out when poked, the barnacle was counted as having been eaten. In order to estimate surface area covered by barnacles, I measured basal diameters of a subsample of barnacles, and multiplied by the number of barnacles present in that tank. Data from field and lab experiments were analyzed in Microsoft Excel 98 and in Systat 8.0. In most tests, all barnacles larger than Imm in diameter were grouped together as adults. Except where explicitly examined (i.e. when looking at recruitment). barnacles less than Imm in diameter were excluded from analysis, as they were assumed to be recent recruits. Results: In the lab experiment the whelk Nucella emarginata had a strong preference for Balanus glandula over Chthamalus spp. Percent cover was used as a measure of availability, assuming that cover would be proportional to the frequency of encounter by à randomly-foraging whelk. The availability of both species of barnacle was close to equal in all tanks (Fig. 1). The mean diameter of Chthamalus spp. was 2.23 mm, while the mean diameter of B. glandula was 5.20. There were six times as many individuals of Chthamalus spp. as B. glandula, but many of the Chthamalus were very small. Forty two percent of the individual B. glandula were consumed, compared to only 2.3% of the Chthamalus spp. Pooling all 3 tanks, the whelk's diet was 75.5% B. glandula (Fig. 1). Nucella ate significantly more B. glandula than expected, based on percent cover as a measure of availability (Chi-square = 14.1, P«.001). Analysis of field data revealed a positive correlation between densities of N. emarginata and B. glandula (r = .377, P«.001) (Fig. 2). A stronger positive correlation existed between densities of N. emarginata andChthamalus spp. (r= .511, P5,001) (Fig. 3) N. emarginata was entirely absent at Breakwater Cove, so correlations could not be calculated for that site. The vertical ranges of B. glandula and Chthamalus spp. show little difference. (Figs. 4, 5) Plots of densities vs. intertidal height indicate that both species are found at the same maximum and minimum heights in both regions (although these maximum and minumum values are highly variable between the two sites.) A strong positive correlation was found between the densities of B. glandula and Chthamalus spp. (r - 630, P«.001, Fig. 6) Barnacle distributions were highly patchy in all sites. Abundances per .01 m2 quadrat varied from 0 to over 250. Mean abundance within different transects at each site were highly variable (Figs. 7, 8). Nested ANOVA showed significant differences in mean abundances between sites and between transects within sites (Table 1). Overall numbers were higher in Breakwater Cove than on China Pt., and there was significantly higher variance in numbers of Chthamalus in Breakwater Cove than on China Pt. (F test. P2.05). (Fig. 9) Recruitment of Balanus glandula was unusually high during the spring of 2001 (J. Watanabe, pers comm). Recruitment occurred throughout the vertical range of B. glandula (Figs. 10 & 1 1), but was highly patchy throughought, with recruits uncommon in most quadrats, and highly abundant in a small number of quadrats (Fig. 12). Although there was a significant positive correlation between numbers of adults and recruits (r==.207, P «.05), the tightness of this trend was low because of the high patchiness. Discussion: Results trom the laboratory experiment clearly show that Nucella emarginata préfers to prey on Balanus glandula rather Chthamalus spp. This confirms work by West (1986) on China Pt., which showed that whelks in the field had a diet consisting primarily of B. glandula. Based on this result, and the previous work by Connell (1961a, 1961b. 1970), 1 would expect that areas with high whelk abundance would correlate with areas of decreased B. glandula abundance. With less competition, and a relatively small threat from predation, I would expect Chthamalus spp. to increase in abundance. As discussed below, this differs from what I found in the field. All possible combinations of the three species showed significant positive correlations. While this does not prove that there is no effect of competition or predation. it does show that this effect is fairly weak. If predation interactions were strong, B. glandula densities might be expected to correlate negatively with N. emarginata densities (Connell, 1970), as whelks would eat all barnacles below a certain tidal height. Furthermore, if predation by N. emarginata was the major factor determining abundance of B. glandula, I would expect to see recruitment in areas of high whelk abundance where B. glandula adults were absent. B. glandula recruits were not seen recruiting to any areas where B. glandula adults were absent. On the other hand, as B. glandula is a sessile species, and it is to be expected that its predators will be found in areas where it is present. Although 1 did not see the very strong interactions reported by Connell in Puget Sound (1970), it is probable that whelks play some role in regulating the abundance of B. glandula. Space competition between B. glandula and Chthamalus spp. did not appear to be strong. The tightest correlation between pairs of species was that between B. glandula and Chthamalus spp., indicating that where one species was abundant, the other was also abundant. This is most likely not due to positive interactions between the two species. but rather is due to the fact that where suitable habitat was present for one species, it was also present for the other. Areas occupied by macroalgae, cleared by limpets, covered in mussels, or in sandy gaps in the granite outcrops are all examples of types of habitat frequently encountered in sites where barnacles were mostly or completely absent. Those areas where barnacles were present were rarely completely covered. Free space was nearly always present (although it was not specifically studied in this experiment.) Only one out of the 155 quadrats I examined appeared to fit the description of Connell (1961b of Balanus growing so densely as to cover all available space and outcompete Chthamalus. Even in this one exceptional quadrat, numerous Chthamalus were found growing among the Balanus. Thus, I conclude that interspecific competition between these two species is not an important factor governing their distributions in southern Monterey Bay. The clear preference shown by Nucella emarginata for Balanus glandula is not surprising. Paine (1981) proposed that the chthamaloid barnacles have taken an overall evolutionary strategy of remaining too small to be eaten. I did not examine the size preference of N. emarginata, but I did measure all barnacles in the field that were being actively drilled by whelks. I was unable to find any that were less than 4 mm in basal diameter. This size is larger than one standard deviation from the mean diameter of Chthamalus spp., but is nearly one standard deviation smaller than the mean for B. glandula. Thus, it seems probable that by staying small, Chthamalus avoids becoming prey of Nucella. It is also possible that Chthamalus that grow to a size approaching 4 mm in basal diameter are rapidly eaten by whelks. If this were the case, I would expect Chthamalus from Breakwater Cove, where whelks are absent, to be larger than Chthamalus from Hopkins. I did not test this hypothesis, but the rocks on which a mean basal diameter of 2.23 mm was measured came from Breakwater Cove. This appears to lend support to Paine’s hypothesis. The high degree of variability in population densities seen between the two sites. and between transects within each site indicates that neither of these species are competitevly dominant, but rather are part of a larger and more complex system. It is likely that population densities are regulated by a variety of factors, rather than simple competition and predation between the three species I studied. Recruitment patterns, predation by other species, parasitsm (Lopez, 2001), physical stresses, and competition with a variety of other space occupiers (macroalgae, limpets, and the mussel Mytilus californianus) are all major space occupiers in Monterey Bay) probably play important roles in determining space occupancy in the rocky intertidal. While I cannot completely rule out predation effects by N. emarginata and competition between B. glandula and Chthamalus spp. as determinators of distribution, it appears that neither plays an overwhelming role compared to factors that have not yet been elucidated. The rocky intertidal has proven a fertile ground for understanding the distribution of species. Fürther investigation of the factors that determine the distribution of these species in Monterey Bay may contribute to our understanding of distribution in many habitats. 12 Acknowledgements: This paper would not exist without the patient support of Jim Watanabe. I cannot thank him enough for his advice and the large amount of time he devoted to this project. Many other professors and grad students were helpful along the way. Fiorenza Micheli, Eric Sanford and Rafe Sagarin contributed both ideas and papers from their libraries. Joanna Nelson and Luke Hunt aided in surveying tidal heights. The other students in the spring class helped create a positive learning environment. Daniel Serres and Galen Weston kept me entertained and excited through dawn tides, dark nights of labor on computers, and bright nights of friendship and the cheapest, tastiest vegan food known to Monterey Bay. Literature Cited: Connell, J. H. 1961a. Effects of competition, predation by Thais lapillus, and other factors on natural populations of the barnacle Balanus balanoides. Ecol. Monogr. 31: 61- 104. Connell, J.H. 1961b. The influence of interspecific competition and other factors on the distribution of the barnacle Chthamalus stellatus. Ecology 42: 710-723. Connell, J.H. 1970. A predator-prey system in the marine intertidal region. Balanus glandula and several predatory species of Thais. Ecol. Monogr. 40: 49-76. Lopez, J. 2001. Spatial Distribution, physiological tolerances, and respiration rates of larvae of the intertidal fly, Oedoparena glauca. Final Papers Biology 175H, Stanford University, Miller Marine Biology Library. Menge, B.A. and G.M. Branch. 2001. Rocky intertidal communties, in Marine community ecology, M.D. Bertness, S.D. Gaines, M.E. Hay eds. Sinauer Associates, Sunderland Ma. Paine, R.T. 1980. The forgotten roles of disturbance and predation. Paleobiology. 7: 553-560. Ricketts, E.F., J. Calvin, J.W. Hedgepeth, and D.W. Phillips, 1985. Between Pacific Tidesw, 5“ ed. Stanford University Press, Stanford, Ca. Sagarin, R.D., J.P. Barry, S.E. Gilman, C.H. Baxter, 1999. Climate-Related Change in an intertidal community over short and long time scales. Ecol. Monogr. 69: 465-490. Sanford, E. and B.A. Menge, 2001. Spatial and temporal variation in barnacle growth in à coastal upwelling system. Mar. Ecol. Prog. Ser. 209: 143-157. West, L. 1986. Interindividual variation in prey selection by the snail Nucella emarginata. Ecology. 67: 798-809. Table 1. ANÖVA table for Nested Anova comparison of means between sites and between transects within sites. Data were square root transformed prior to analysis ((x+0.5)) Source Sum-of-Squares df Mean-Square F-ratio F Between Sites 607.992 1 607.992 13.540 0.006 Between Transects 359.220 8 44.903 2.847 0.006 Error 1892.626 120 15.772 Figure Legends Fig. 1. Predation on Balanus glandula and Chthamalus spp. by Nucella emarginata in laboratory experiment. Bars on left are the percent of the rock surface covered by each barnacle species. Bars on right are measurements of the percent of the whelk's diet belonging to each barnacle species. (n = 3 tubs, 1869 barnacles). Fig. 2. Relationship between abundance of Balanus glandula and Nucella emarginata on China Pt. N. emarginata were sampled in 25 m’ quadrats, and a .01 m’ quadrat was chosen randomly within the .25 m’ quadrat to sample barnacle abundance. (n = 95 quadrats) Fig. 3. Relationship between abundance of Chthamalus spp. and Nucella emarginata on China Pt. N. emarginata were sampled in 25 m’ quadrats, and a .01 m2 quadrat was chosen randomly within the .25 m’ quadrat to sample barnacle abundance. (n= 95 quadrats) Fig. 4. Vertical distributions of Balanus glandula and Chthamalus spp. in the intertidal zone at China Pt. Densities based on counts in .01 m2 quadrats. (n= 95 quadrats) Fig. 5. Vertical distributions of Balanus glandula and Chthamalus spp. in the intertidal zone at Breakwater Cove. Densities based on counts in .01 m’ quadrats. (n= 55 quadrats) Fig. 6. Relationship between abundance of Balanus glandula and of Chthamalus spp. in quadrats. Data from Breakwater Cove and China Pt. are combined. (n= 150 quadrats) g. 7. Variation in mean abundances of barnacles between vertical transects on China 2t. Transects were placed perpendicular to shore, from the highest barnacle to the lowest barnacle, spaced unevenly over an area with a total spread of 20 meters. Means are based of 15 quadrats per transect. Fig. 8. Variation in mean abundances of barnacles between vertical transects in Breakwater Cove. Transects were placed perpendicular to shore, from the highest barnacle to the lowest barnacle, spaced unevenly over an area with a total spread of 20 meters. Means are based on 10 quadrats per transect. Transect four (with much lower abundances of both species) ran through an area of smaller cobbles which had lower abundances of barnacles. g. 9. Variation in mean abundance of barnacles between Breakwater Cove and China Pt. These two sites are approximately 2 km apart. China Pt. is considered protected open coast, while Breakwater Cove is within Monterey Harbor, and thus highly protected. (n- 75 at Hopkins and 55 at Breakwater Cove) Fig. 10. Vertical distribution of Balanus glandula recruits and adults on China Pt. Densities based on counts in .01 m’ quadrats. (n = 95 quadrats) Fig. 11. Vertical distribution of Balanus glandula recruits and adults on China Pt. Densities based on counts in .01 m’ quadrats. (n = 55 quadrats) Fig. 12. Relationship between numbers of recruits and adults of Balanus glandula in quadrats. (n = 150 quadrats) Fig. 1 0.9 0.8 5 04 Fig. 2 Predation by Nucella in Lab » Balanus a Chthamalus % of barnacle % of barnacles surface area that were eaten Number of Balanus vs. Number of Nucella 200 150 100 50 10 Number of Nucella per .25 m2 quadrat Fig. 3 Number of Chthamalus vs. Number of Nucella 150 2 100 111. Number of Nucella per .25 m2 quadrat Fig. 4 Balanus and Chthamalus distributions in Breakwater Cove 2.5 2.0 1.5 S 1.0 S S - SPECIES Balanus Chthamalu 0.0 100 200 300 Number of Bamacles per .O1 m2 quadrat Fig 5 Fig 6 Balanus and Chthamalus distributions on China Pt. 3 32 15.12.2 14 1.2 :: : - -.. .... ....... ....... ....... . ..... . . ..- — 100 200 300 Humber of Baracles perO1 me quadrat Numbers of Balanus vs. Numbers of Chthamalus 300 5 200 £ 1004- b. 200 300 100 Numbers of Balanus per .01 m2 quadrat SPECIES Balanus Chthamalu Fig. 7 05 Fig. 8 Variation Among Transects on China Point Variation Among Transects at Breakwater Cove Balanus Chthamalus E Balanus EChthamalus Fig. 9 16 Fig. 10 Variation between Breakwater and China Pt. EBalanus HChthamalus Breakwate Hopkins Balanus Recruitment on China Pt. SIZE Adults Recruits 300 Number of barnacles per .O1 m2 quadrat