THE NATURAL DISAPPEARANCE OF A TOP CARNIVORE AND ITS IMPACT ON AN INTERTIDAL INVERTEBRATE COMMUNITY: THE INTERPLAY OF TEMPERATURE AND PREDATION ON COMMUNITY STRUCTURE (GULF OF CALIFORNIA).
Item Type text; Dissertation-Reproduction (electronic)
Authors BOYER, EDWARD HENRI.
Publisher The University of Arizona.
Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
Download date 24/09/2021 05:37:14
Link to Item http://hdl.handle.net/10150/184109 INFORMATION TO USERS While the most advanced technology has been used to photograph and reproduce this manuscript, the quality of the reproduction is heavily dependent upon the quality of the material submitted. For example:
• Manuscript pages may have indistinct print. In such cases, the best available copy has been filmed.
• Manuscripts may not always be complete. In such cases, a note will indicate that it is not possible to obtain missing pages.
• Copyrighted material may have been removed from the manuscript. In such cases, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, and charts) are photographed by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each oversize page is also filmed as one exposure and is available, for an additional charge, as a standard 35mm slide or as a 17"x 23" black and white photographic print.
Most photographs reproduce acceptably on positive microfilm or microfiche but lack the clarity on xerographic copies made from the microfilm. For an additional charge, 35mm slides of 6"x 9" black and white photographic prints are available for any photographs or illustrations that cannot be reproduced satisfactorily by xerography.
8715710
Boyer, Edward Henri
THE NATURAL DISAPPEARANCE OF A TOP CARNIVORE AND ITS IMPACT ON AN INTERTIDAL INVERTEBRATE COMMUNITY: THE INTERPLAY OF TEMPERATURE AND PREDATION ON COMMUNITY STRUCTURE
The University of Arizona PH.D. 1987
University Microfilms International 300 N. Zeeb Road, Ann Arbor, MI48106
PLEASE NOTE:
In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark_..f_.
1. Glossy photographs or pages L 2. Colored illustrations, paper or print if 3. Photographs with dark background ~
4. Illustrations are poor copy ___
5. Pages with black marks, not original copy __
6. Print shows through as thero is t~,d on both sides of page ___
7. Indistinct, broken or small print on several pages J
8. Print exceeds margin requirements __
9. Tightly bound copy with print lost in spine ___
10. Computer printout pages with indistinct print ___
11. Page(s) lacking when material received, and not available from school or author.
12. Page(s) seem to be missing in numbering only as text follows.
13. Two pages numbered . Text follows.
14. Curling and wrinkled pages __
15. Dissel1ation contains pages with print at a slant, filmed as received ____ 16. Other______
University Microfilms International
THE NATURAL DISAPPEARANCE OF A TOP CARNIVORE AND ITS IMPACT ON AN INTERTIDAL INVERTEBRATE COMMUNITY: THE INTERPLAY OF TEMPERATURE AND PREDATION ON COMMUNITY STRUCTURE
by
Edward Henri Boyer
A Dissertation Submitted to the Faculty of the DEPARTMENT OF ECOLOGY AND EVOLUTIONARY BIOLOGY
In Partial Fulfillment of the Requirements For the Degree of DOCTOR OF PHILOSOPHY
In the Graduate College THE UNIVERSITY OF ARIZONA
198 7 THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE
As members of the Final Examination Committee, we certify that we have read
the dissertation prepared by __E_d_w __ a_r_d ____ H_e_n __ r_i ___ B__ o~y_e_r ______
entitled The natural disappearance of a top carnivore and its
impact on an intertidal invertebrate community: the
interplay of temperature and predation on community
structure.
and recommend that it be accepted as fulfilling the dissertation requirement
for the Degr~Of~jlQSQPbJT
~ Da t'0'he ~/S,2 ~~;b Date ; l 4/~z ·ff7 Date' / /3 ftlf 1= Date c4ri: I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement. Dissertation~ Director Date STATEMENT BY AUTHOR This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library. Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author. This dissertation is dedicated to: My Father Larry Larson Steve Mackie Nick Yensen Donald Thomson mentors iii PREFACE The work presented in this dissertation would not have been possible if it weren't for a series of fortuitous events and the generosity and cooperation of a number of fellow naturalists. I just happened to be in the right place at the right time, equipped with baseline data, when an important predator suffered a sudden and catastrophic population crash. It is because of the vision and wisdom of my major professor, Dr. Donald Thomson, who saw the possibility and importance of this project, and the generosity of Nick Yensen who gave willingly of his field notes, thoughts, and time that this project was possible. Dr. Thomson opened my eyes to the value and excitement of field instruction in the Gulf of California. Without the inspiration and friendship of my friend, Steve Mackie, I never would have learned the special joy of looking under rocks and into tidepools in the intertidal zone, nor would I have suspected how fascinating invertebrates could be, nor would there have been any baseline data upon which this research is based. Throughout the long period of field work (which began while I was an undergraduate at A.S.U.) many people have assisted in collecting data and turning over boulders. For iv v this help I thank the following people: G:(a Fasanella Helias, Lucinda Davis Maseman, Steve Mackie, L.Y.Maluf, Peggy Turk Boyer, Cynthia Howser, Jim Mackie, Becky Simmons, Rick Boyer, Jon Schumaker, Lisa Doner, and the numerous students in field courses. My understanding of the community at Puerto Penasco and of ecology in general has benefitted from discussions with Mike Dungan, Robert Paine, Steve Schuster, Rick McCourt, Yvonne Maluf, Fernando Zapata, Curt Lively, Pete Raimondi, Katrina Mangin, Jim Malusa, Guy Hoelzer, Julio Betancourt, Ray.Turner, and all the members of my committee. I thank the graduate students of the Dept. of Ecology and Evolutionary Biology who together form a supportive and creative environment in which to work. I thank Bob Fry for his patient guidance in the use and understanding of statistical procedures but I take rsponsibility for any statistical inpropriety. Thanks also to Margaret Kurzius for doing most of the figure illustration. I wish to acknowledge Mike Magatagan who taught me computer programming and basic statistics. Mike was once a student of mine but he absorbed everything I had to offer and then became my teacher. I thank him for his friendship and inspiration. Janet VOight, Meriel Brooks, Sarah Mesnick, Ronnie Sidner, Judy Eastridge, Andrea Sorey, Judy Stahl, my family and many other friends have been invaluable as sources of support during the final stages of th~s project. I vi acknowledge the members of and programs offered by Werner Erhard and Associates for support and contributions to my effectiveness and ability to finish this project. Much of the field work was carried out while using facilities at the Centro de Estudios de Desiertrs y Oceanos (CEDO) in Puerto Penasco, Sonora, Mexico. To the director, Peggy Turk, and the others that have made CEDO a reality I am grateful. CEDO is an important scientific and cultural link between our countries as well as a peaceful and beautiful place to do field work. Andrea Addison-Sorey, Nick Yensen, D.A. Thomson, Rick Boyer, J.R. Hendrickson and Astrid and Jim Brown offered important editorial comments and no doubt-~mproved the readability and value of the dissertation. I wish to thank my fiancee, Judy Stahl, for putting up with my obsession and anxiety over this project. She has been a great support and often gave me back my perspective. Finally I acknowledge my parents who never failed to make me feel important, -intelligent and loved. They have always told me to do my best and that would be good enough. They have had unswerving confidence in me and have assisted me financially without hesitation during my years in college and graduate school. They set an example of excellence that I hope to follow throughout my entire life. TABLE OF CONTENTS Page LIST OF TABLES •.•...•.•...... •...... •...... viii LIST OF ILLUSTRATIONS ...... •....•...... •.....••... ix ABSTRACT •...•.•..•.•.....••••...... xi 1. INTRODUCTION •...... •...... 1 2. STUDY SITE AND ORGANISMS ...... •...... 7 Physical environment ...... •...... 7 Study organisms .•..•...... •...... 9 3. METHODS •..••...... 15 Trophic relationships .....•.•...... 15 Barnacle density •...... •...... 15 Density of Mobile invertebrates ...•...... 16 Community data analysis ••...•...... •...... 17 4. RESULTS AND DISCUSSION .•.•...... •...... •. 20 Trophic relationships ...... 20 Individual species densities & distributions.24 Prey species .....••...... •.•...... •.... 25 Barnacle density ...••...... 25 Mobile invertebrate densities ...... 27 Non-prey species .•..•...... •...... 36 Guild s truc ture •...... 44 Prey vs. non-prey species ...... 44 Trophic guilds ...... ••...... 49 Community patterns •...... 58 Overall abundance ...... 58 Diversity ...... 58 General discussion ...... •...... 62 Heliaster as "keystone species" ...... 63 5. SUMMARY AND CONCLUSIONS ...... •...... 70 APPENDIX A:CAUSES OF THE Heliaster DIE-OFF ...•..... 74 APPENDIX B:BEFORE AND AFTER LIST OF INVERTEBRATES AT STATION BEACH ...... 80 LITERATURE CITED •...... •...... 83 vii LIST OF TABLES Table Page 1. Study organisms (Heliaster prey) at Station Reef ...... 11 2. Study organisms (non-prey of Heliaster) at Station Beach reef ...... 12 3. Helias ter ku bini j i stomach con ten ts ...... 21 4. Species density summary table ...... 43 5. Prey and non-prey ranks ...... 47 6. Spearman rank correlation coefficients for prey and non-prey in the boulder and reef flat zones ..... 48 7. Carnivore, herbivore, and detritivore ranks ...... 52 8. Spearman rank correlation coefficients for carnivores, herbivores, and detritivores in the boulder and reef flat zones ...... 53 9. Number of species, H', and J diversity measures for the 20 species sub-community at Station Beach Reef. '...... 59 10. Density of Heliaster kubiniji between 1974 and 1984 at Station Beach Reef ...... 78 viii LIST OF ILLUSTRATIONS Figure Page 1. Aerial photograph of Station Beach Reef ...... 8 2. Deviations from monthly mean sea surface temperatures for 1966-1984 ...... 10 3. Photograph of Heliaster kubiniji ...... •...... 14 4. Heliaster dominated food webs from the intertidal communities of the Northern Gulf of California ..... 23 5. Mean density (+/- 95% confidence intervals) for Chthamalus anisopoma for 1974,1975 and 1979-1983 ... 26 6. Mean density (+/- 95% confidence intervals) for Tegula ~ and Nerita funiculata •...... 28 7. Mean density (+/- 95% confidence intervals) for Mitrella gutata and Morula ferruginosa ...... 29 8. Mean density (+/- 95% confidence intervals) for Columbella aureomexicana and Turbo fluctuosos ...... 30 9. Mean density (+/- 95% confidence intervals) for Liocerithium judithae and Chiton ~ ...... 34 10. Mean density (+/- 95% confidence intervals) for Neorapana tuberculata and Acanthina angelica ...... 35 11. Mean density (+/- 95% confidence intervals) for Aplysia californica, Othilia tenuispina, and Echinomen tra van brun ti ...... 37 12. Mean density (+/- 95% confidence intervals) for Ophioderma ~, Ophioneries annulata, and Selenkothuria lubrica ...... 39 13. Mean density (+/- 95% confidence intervals) for Palythoa ignota, Eurythoe complanata, and Clibanarius digueti...... •...... 41 ix x LIST OF ILLUSTRATIONS -- Continued Figure Page 14. Mean density C+/- 95% confidence intervals) for prey and non-prey guilds ...... 45 15. Mean density C+/- 95% confidence intervals) for carnivore, herbivore, and detritivore guilds ...... 50 16. Rank vs. abundance curves for the 20 species sUb-community at Station Beach Reef in the boulder and reef flat zones ...... 57 ABSTRACT The predatory sea star, Heliaster kubiniji, has been hypothesized to be a "keystone species" that is instrumental in maintaining diversity in the intertidal zones of the northern Gulf of California (Paine 1966). Four hundred and one Heliaster stomach samples collected in 1974-1976 from Station Beach, Puerto Penasco, Sonora, Mexico indicated that Heliaster consumed a variety of prey but preferred barnacles, Chthamalus anisopoma (93% of all prey items). The catastrophic decline of Heliaster in the Gulf of California in 1978 resulted in mortalities approaching 100% (Dungan et al 1982). Pre-die-off community data from the intertidal reef at Station Beach (Mackie and Boyer 1977) were compared with post-die-off (1981-1984) data to determine effects of the Heliaster disappearance. Community structure data included abundance and intertidal distributions of 20 species of macro-invertebrates including 11 Heliaster prey and 9 non-prey species and consisted of 6 carnivores, 7 herbivores,S detritivores and 2 filter feeders. Comparisons were made in"two distinct habitats: a basalt boulder habitat (Heliaster's preferred habitat) and a reef flat habitat. In the reef flat zone three prey species increased density xi while seven were unchanged; two non-prey species decreased while seven were unchanged. In the boulder zone six prey species increased, three decreased and two were unchanged; one non-prey increased, three decreased and five were unchanged. An examination of prey vs. non-prey guild structure indicated an increase in prey guild density in the boulder zone from 1976 to 1981 and a reshuffling of rank order of prey density. Analysis of trophic guild structure revealed an increase in carnivore guild density (but no change in ranks) in the boulder zone only and an increase in herbivore guild boulder density (with change in ranks) between 1976 and 1981. No significant changes in species diversity were observed during the study. Heliaster did not act as a keystone species in this community, but it influenced the abundance and structure of prey species and trophic guilds. A continuum is proposed for the potential effects of predation on community structure. xii CHAPTER 1 INTRODUCTION Understanding the structure of biological communities and the forces that create or maintain them is of great interest to ecologists. A community is essentially a distinct natural division of the biosphere that we can identify and hopefully comprehend, and it forms much of the context in which evolution occurrs. In a practical sense, it is at the community level that many impacts of human disturban~e become apparent. In order to better predict the consequences of ecological disturbances (such as the greenhouse effect, species extinctions, habitat degradation, or nuclear" warfare), ecologists must understand the processes that determine the structure and dynamics of communities. Many community ecologists have demonstrated the value of experimental manipulations in revealing underlying processes affecting community structure. It is rare that we have the opportunity to study the response of a complex community to a natural perturbation of large magnitude and duration. The word "community" is defined here as an association of biological organsims that reside in a relati~ely clearly defined area, that interact with one 1 2 another either directly or indirectly, and that experience the same basic set of physical conditions. Numerous factors may affect the structure and character of a marine communi ty, such as the frequency and severity of disturbances (Connell 1978, Sousa 1979), competitive interactions (Menge and Sutherland 1976, Brown et al 1979), age of the community (Pianka 1966), larval recruitment (Underwood and Denley 1984; Connell 1975) and predation (Paine 1966, 1969, 1971, Dayton 1971, Connell 1971, 1975). Predation may affect community structure in several important ways: 1) It may influence the resource utilization patterns and conditions tolerated by the prey, thereby affecting their distribution and abundance (Connell 1961, 1975; Duggings 1983; Kotler 1984). 2) Predation may alter patterns of herbivory (Dayton 1985; Duggins 1983; Garrity and Levings 1981). 3) Predation may mediate competitive coexistence (Connell 1971; Dayton 1971; Dayton and Hessler 1972; Dayton et al 1974; Harper 1969; Lubchenco 1978; Morin 1983; Caswell 1978; Day 1977; Inouye et al 1980). 4) Predators may reduce species number if selective for non-dominants (Wiltse 1980; Addicott 1974). 5) Predation may increase diversity by creating a mosaic of patches which are in different successional states (Menge and Sutherland 1976; Duggins 1983; Lewis 1986). The keystone species hypothesis represents one extreme of the potential role of predation in a community. A 3 keystone species is a top carnivore that maintains high species diversity by preventing one dominant prey species from monopolizing resources and competitively excluding other species. Paine's work with Pisaster (a carnivorous sea star) in the Pacific Northwest clearly defines a situation in which a predator can be critical to the structure of the community (Paine 1966, 1969). Species diversity significantly decreased when the top carnivore, Pisaster, was removed from this rocky intertidal community. The decrease in species number was a result of the monopolization of space by the mussel Mytilus. Following Paine's work with Pisaster, a number of other cases of apparent keystone species have been reported. Paine (1971) demonstrated that the removal of the predatory sea star Stichaster from a New Zealand rocky intertidal zone lead to a decrease in species diversity and the monopolization of space by mussels. Estes and Palmisano (1974) suggested that sea otters may act as keystone species in some Aleutian islands by preying on urchins which, when unchecked, prevent kelp forest development. Dayton and Hessler (1972) and Dayton et al (1974) have shown that in deep sea benthic communities predatory echinoderms allow competitive coexistence of sponges. Other examples include work by Glasser (1978, 1979) with plankton communities, and Summarco (1980, 1982), who suggested that the urchin Diadema may be a keysone species in tropical coral reef-algae communities. 4 It was the suggestion by Paine (1966) that the predatory sea star Heliaster kubiniji ma¥ be a keystone species in Northern Gulf of California intertidal communities that motivated the study reported here. This study is based on the natural and catastrophic decline of Heliaster kubiniji. Heliaster was the most common and conspicuous intertidal asteroid throughout the Gulf of California (Brusca 1980, Steinbeck and Ricketts 1941) until the time of its decline. Dungan et al (1982) report the rapid die-off of Heliaster in the intertidal zone at Puerto Penasco, Sonora, Mexico in the Summer of 1978: "Scores of disintegrating sun stars were first observed during 18 to 20 June 1978 in the intertidal zone around Puerto Penasco. Animals initially exhibited whitish lesions on the aboral surface. The lesions rapidly enlarged until the entire animal fragmented, and the decaying parts of the same individual were left in close proximity. High concentrations of bacteria were found in the lesions but it is not known whether bacterial infection was the primary cause of death. Mortality of R. kubiniji approached 100 percent; within 2 weeks the sun star had 'virtually disappeared from this site." The authors further state that the symptoms of the dying individuals closely resembled those of sun stars kept in aquaria which die from temperature stress or pollution. They suggest that a period of warm sea-surface waters in the winter of 1977-1978 that continued through the spring of 1978 was either directly or indirectly responsible for the 5 die-off (see Appendix A). Since the initial die-off, Heliaster population density has shown no sign of recovery at Puerto Penasco, and on summer censuses of the Baja California coast in 1978-1985 Heliaster was extremely rare, even in locations where pre-die~off censuses indicated it was commonly found. The questions raised by the disappearance of this hypothesized keystone species are: (1) What happened in the community since the Heliaster die-off? , (2) Was Heliaster a keys~one species that prevented monopolization of intertidal space by a competitive dominant? I have been able to address these questions only because of the existance of community data collected prior to the disappearance of Heliaster. These data come from several sources, one of which is a collection of Heliaster stomach contents samples collected by N. Yensen in 1974-1976 at Station Beach. These samples wer.e preserved but not analysed until I did so in 1979. The main data on community structure are from Mackie and Boyer (1977) who collected data on intertidal distribution and abundance for 27 species of invertebrates from Station Beach, Puerto Penasco. This study compares the pre-die-off and post-die-off abundance and intertidal distributions of 20 species of invertebrates of the community in which Heliaster was a member. This group of species included Heliaster prey (11 species) and non-prey (9 species), 6 carnivores (6 spp.), herbivores (7 spp.), detritivores (5 spp.) and filter feeders (2 spp.). This study has three objectives: 1) to document the events in the community over a long time span as it experienced a significant physical (temperature) and ecological (Heliaster decline) perturbation; 2) to determine if Heliaster was a k~ystone species, and to elucidate more clearly its role in the community; and 3) to assess what this natural perturbation has revealed about community stability and the forces maintaining intertidal community structure. CHAPTER 2 STUDY SITE AND ORGANISMS Physical environment This study was conducted on a strip of intertidal shell-hash beach rock (coquina) known as "Station Beach Reef" that lies 3 km. east of Puerto Penasco, Sonora, Mexico on the Northern Gulf of California (31 08'30" N., 113 33'00" W.). The intertidal zone in this area is extensive due to the extreme tides of the Northern Gulf (mixed semi-diurnal tides with a maximum spring tide range of 7 meters). Station Beach Reef is characterized by numerous tidal pools, areas of exposed flat coquina beach-rock and a large field of small basalt boulders (Fig. 1). This boulder field was the preferred habitat of Heliaster kubiniji and represents the most physically heterogeneous and species rich microhabitat on the reef. The northern Gulf is a shallow basin bordered on the east by the arid Sonoran desert and on the west by the high arid mountains of Baja California that block it from the Pacific climate. As a result, the semi-isolated northern Gulf recieves little rainfall (less than 3 inches/year) and experiences extreme yearly temperature variations; saa 7 Fig. 1. Aerial photograph of Station Beach Reef. The shoreline runs due east-west; the photo was taken from the east (by P.Turk) 00 9 surface temperatures at Puerto Penasco may exceed 32° C in the summer and may drop below 12°C in the winter. Onshore sea surface temperatures (SST) at ·Puerto Penasco have been measured daily since 1964 (at the Environmental Research Lab and the Centro de Estudios de Desiertos y Oceanos) and deviations from mean monthly temperatures for these data are shown graphically in Fig. 2. Three distinct temperature events stand out in this record. The first is the cold period in 1970-1971 that has been linked to a winterkill for fish (Thomson and Lehner, 1976). The second is the 1977-1978 warm period that is hypothesized to have caused the catastrophic mortalities of Heliaster in the summer of 1978 (Dungan et aI, 1982). The third event is the 1981-1984 warm period that may have prevented Heliaster populations from recovering. Two tailed t-tests of post-1980 temperature deviations indicate a warm period; 1981, 1983, and 1984 have mean deviations significantly greater than zero (p<.02; p<.05; p<.OI respectively). Study organisms During the course of this study observations were made on populations of 21 of the most conspicuous and abundant macro-invertebrates inhabiting Station Beach Reef. A summary of the abundance and feeding habits of these organisms is presented in Tables 1 and 2. These 21 species include 11 molluscs (10 of which are gastropods), 6 10 67 68 69 70 7.1 72 73 14 75 1. .... (J c as ~ E .g 78 77 78 79 88 8.1 82 83 a4 c :a;0 .; 2 . Q) C '----+---31 Year Fig. 2. Deviations from monthly mean sea surface tempera tures for 1966-1984. Each point is the deviation from the mean temp. for that month. Peiod 1 is the fish kill (Thomson and Lehner 197&); period 2 is the Heliaster die-off; period 3 is the warm period of 1'781-1984. 11 Table 1. Study organisms (Heliaster prey) at Station Beach Reef. Relative abundance (A=abundant; C=common; U=uncommon; R=rare) and diet are indicated. Sources are footnoted at bottom .. Species Name Abundance Diet (Source) MOLLUSCA:POLYPLACOPHORA Chiton virgulatus U Algal film & turf (1) MOLLUSCA:GASTROPODA Acanthina angelica C Barnacles (1,2) Columbella aureomexicana C Macroalgae (1,3) Liocerithium judithae R Detritus (3,4) Mitrella guttata A Polychaete, detritus (5) Morula ferruginosa A Barnacles, Gastropods, Bivalves (1,6) Neorapana tuberculata R Bivalves & gastrpods (1) Nerita funiculata A Algal film, diatoms (1) Tegula ~ A Macroalgae (1) Turbo fluctuosus A Macroalgae (1) CRUSTACEA:CIRRIPEDIA Chthamalus anisopoma A Plankton feeder (1,7) SOURCES: (1) Brusca 1980; (2) Yensen 1979; (3) R.Houston pers. comm.; (4) R. Houbrick pers. comm.; (5) D. Lindberg pers. comm.; (6) Lively C. and P. Raimondi in prep.; (7) Barnes 1974 12 Table 2. Study organisms (non-prey of Heliaster) at Station Beach Reef. Relative abundance (A=abundant; C=common; R=rare) and diet are indicated. Sources are footnoted at bottom. Species Name Abundance Diet (source) CNIDARIA:ANTH020A Palythoa ignota A Plankton (1) ANNELIDA:POLYCHAETA Eurythoe complanata C Detritus (1) MOLLUSCA:GASTROPODA Aplysia californica R Macroalgae (1,2) CRUSTACEA:DECAPODA Clibanarius digueti C Detritus (1,3) ECHINODERMATA:ECHINOIDEA Echinometra vanbrunti C Encrusting algae & invertebrates (4) ECHINODERMATA:OPHIUROIDEA Ophioderma ~ C Detritus & small inverte brates (1,4) Ophioneries annulata C Detritus (1,4) ECHINODERMATA:ASTEROIDEA Heliaster kubiniji R Barnacles, Gastropods, Bi valves (1,4) Othilia tenuispina R Detritus, small anemones, mucus (1,4) ECHINODERMATA:HOLOTHUROIDEA Selenkothuria lubrica C Suspended matter (1,4) SOURCES: (1) Brusca 1980; (2) R. Houston pers. comm.; (3) A. Harvey pers. comm.; (4) L.Y. Maluf pers. comm. 13 echinoderms (including Heliaster), 2 crustaceans, 1 polychaete and one anthozoan. Eleven of these are prey of Heliaster whereas the other 9 are non-prey; there are 6 carnivores, 7 herbivores, 5 . detritivores and 2 filter feeders. Brief descriptions of the presumed keystone predator and its preferred prey follow. The sunstar, Heliaster kubiniji (family Heliasteridae), is a medium sized (25 cm diameter; 19-25 arms) and highly mobile predator which ranges from the Gulf of California to Nicaragua (Maluf 1987; see Fig. 3). Until 1978 it was the most common asteroid of the Gulf of California intertidal zones. Heliaster lives under basalt boulders or in reef crevices and forages during night low tides or incoming high tides. Information on the life history and ecology of Heliaster can be found in the following sources; Xantus (1860), Madsen (1956), Caso (1961), Paine (1966), Brusca (1980). The barnacle, Chthamalus anisopoma, is a favored prey of Heliaster and of the gastropods Acanthina and Morula. Chthamalus is a small barnacle (4-5 mm in diameter) and at Station Beach it is almost entirely restricted to the tops of the basalt boulders in the mid-intertidal zone. Chthamalus grows rapidly, becomes mature 6 weeks after settlement and breeds throughout the year. Most Chthamalus mortality results from predation and intraspecific crowding (Malusa 1986; see also Dungan 1984 and Lively 1984). 14 Fig. 3. Photograph of Heliaster kubiniji. Shown here on barnacle covered basalt boulder. Size is 20 cm. diameter. Photo by D.A. Thomson. CHAPTER 3 METHODS Trophic Relationships In order to understand the trophic relationships of He1iaster, contents of 401 He1iaster stomachs were collected (at Station Beach) and preserved by N. Yensen between April, 1974 and June 1976. Stomach contents were collected by placing the sea stars in plastic coritainers (subsequent to a feeding period) until non-digestible contents were evacuated. The samples were examined by the author in 1979 at which time all items were identified to species and counted (a total of 5,213 prey items). Prey importance was determined by numerical abundance in the samp18s. Biomass of Chthama1us and the gastropod prey were estimated as total live wet weight to the nearest O.lg. These samples are now in the care of N. Yensen. Barnacle density In order to determine barnacle density, decimeter square (10 cm. x 10 cm.) areas were marked on basalt boulders in the intertidal zone at Station Beach' and photographed repeatedly over time. Slides of the decimeter square areas were then projected onto a grid of 100 randomly 15 16 located dots. The number of dots which fell on barnacles (live vs. dead) or free space were then counted and mean percent ~over of boulder surface was calculated for each category. I analyzed 105 slides taken by Yensen in 1974 and 1975 and 503 slides that I took since 1979 using this method. In most cases 20 boulders were sampled per month. All slides are of boulders from the the same study sites on the intertidal reef at Station Beach and in some cases the same boulders studied by Yensen. Densities of Mobife Invertebrates The pre-die-off community structure data are from Mackie and B~yer (1977). In this study seven transects were laid out on the inte~tidal reef at Station Beach running from the high to the low intertidal zone. Two distinct habitats were sampled: the basalt boulder field and the flat coquina platform reef. One-square-meter quadrats were placed every 5 meters along those transects and examined. For each quadrat the following data were collected: date, quadrat number, distance from shore reference, tidal height, habitat (boulder vs. flat reef), and number of individuals of 27 species of macro-invertebrates and several species of algae. A total of 114 quadrats were sampled in the spring (Jan., Feb.,and Mar.) .of 1976; 63 quadrats were on the reef flat and 51 were in the boulder field. These are the best data on 17 community structure before the die-off. My post-die-off community sampling methodology was designed to provide comparable data. For four years, beginning in the spring of 1981 I repeated the sampling done by Mackie and myself in 1976. Three transects were placed running from the high to low intertidal zone on the reef at Station Beach in the same location that we had placed our earlier transects. Both boulder field habitat and reef flat habitat were sampled. One-square-meter quadrats were placed every 3 m. along the transects. One transect was sampled in each of three months: Jan., Feb., and March. A total of 310 quadrats were recorded this way. In 1981 75 quadrats were sampled (37 boulder field, 38 reef flat); in 1982 79 samples (34 boulder field, 45 reef flat); in 1983 81 samples (35 boulder field, 47 reef flat); and in 1984, 76 samples (37 boulder field, 39 reef flat). Community Data Analysis The structure of the subcommunity represented by the 21 invertebrate species was analysed by first examining the mean density and intertidal distribution of each species over the experimental period in che two habitats (boulder zone and reef flat zone). The data cover 5 years; 1976, 1981, 1982, 1983, and 1984 (pre-die-off refers to 1976 and post-die-off refers to 1981-1984 combined). In most cases 18 mean densities were considered significantly different (at the .05 level) if the 95% confidence intervals did not overlap. In all cases a 2-way-ANOVA (on log transformed data; time vs. habitat) and the Tukey multiple mean comparison test was used to further evaluate or verify changes in mean density. Broader community patterns were evaluated by use of diversity indices: nu~ber of species, H' (Shannon-Wiener index), and J (eveness). For each year and habitat I counted the number of species (excluding Heliaster), and calculated H' and J using data from all species except Heliaster. Chthamalus was excluded because its density was expressed in % cover, and Palythoa was excluded because of the huge number of individuals in the samples. These data alone would have completely determined the calculated values. I also compared lists of common invertebrate species found at Station Beach by previous workers with what could be found since 1978. A "master species list" of common intertidal invertebrates at Station Beach was compiled from the following sources: Pickens, 1965; Hershey, 1969, Mohr, 1970, Brusca, 1980; and Mackie and Boyer, 1977. Species that were found at Station Beach between 1979 and 1986 (after the Heliaster die-off) were checked off against the "master species list". Combinations of species were compared to determine how different guilds behaved over time. Two classifications 19 of guilds were used: prey 6f Heliaste~ or non-prey, and trophic guilcis: carnivore, herbivore, or detritivore. Carnivores included Acanthina, Morula, Mitrella, Neorapana, Ophioderma, and Othil~a. Herbivores included Turbo, Tegula, Nerita, Columbella, Aplysia, Chiton and Echinometra. Detritivores include Liocerithium, Clibanarius, Ophioneries, Selenkothuria, and Eurythoe, Finally, a series of rank correlation tests, based on abundance ranking of the members of these guilds, was performed in order to determine if guild structure was conserved over time (e.g. was 1976 vs. 1981 rank order highly correlated?). Rank order was based on density within guilds with the most'abundant gpecies given the rank of and tied ranks were calculated as the average rank. Spearman rank correlation coefficients were calculated using the procedure for tied ranks and compared to one another using the z statistic procedure (Zar 1984). CHAPTER 4 RESULTS AND DISCUSSION Trophic Relationships The analysis of 401 Heliaster stomach contents samples yielded over 5,000 individual prey items representing approximately 40 species of marine invertebrates. A list of prey items (ranked by numerical abundance), apd their percent composition of the diet by numbers is prese~ted in Table 3. Barnacles, Chthamalus anisopoma, were clearly the preferred prey comprising 93% of the items found in the samples. The next most abundant prey item, the small mussel Brachiodontes semilaevis, comprised only 2 % of the diet. The other important prey in order of numerical abundance were the snails Liocerithium judithae, Turbo fluctuosus, Tegula ~, Littorina aspera, Anachis ~. and Mitrella guttata. A total of 37 species were identified in the samples but since some of these organisms could only be identified to the genus level there were possibly anywhere from 40 to 50 species represented in the samples. Heliaster clearly prefers barnacles when measured by numerical abundance but the relative contribution of the other species to the diet is different when expressed in 20 21 Table 3. Heliaster kubiniji stomach contents. Prey species are ranked by numerical abundance of individuals found in 401 Heliaster stomach contents samples. Proportion of total number of prey items (total = 5,230) is also given for each species (* indicates less than 0.5% of total). SPECIES # INDIVIDUALS % TOTAL Chthamalus anisoQoma 4,854 93.0 Brachiodontes semilaevis 100 2.0 Liocerithium judithae 45 1.0 Littorina .2..I!...:.. 28 .5 Turbo fluctuosus 28 .5 Tegula ~ 24 .5 Anachis ~ 23 .5 Mitrella guttata 21 .5 ... Cardita affinis 11 "- ,~ Terebra ~ 10 Turitella .2..I!...:.. 6 ,~ Chiton .2..I!...:.. 5 ~::: Nerita funiculata 4 * ... Acanthina angelica 3 '0 Morula ferruginosa 2 ~, ,,- Petrolisthes ~ 7 Polychaete worms 5 '0 Bryozoans 5 "--'- Limpets 4· ~::: 18 other species 45 1 22 terms of biomass: Herbivorous gastropods (Littorina, Turbo, Tegula) = 40%, other gastropods = 10%; Brachiodontes 2% and barnacles = 37% of the biomass. Heliaster with its many arms and dense concentration of tube feet around the mouth seems to be well adapted to tearing off chunks of barnacles; in fact most of the barnacles found in the stomach samples were in clumps of 20 to 25 individual tests still attached to each other. It could be that many of the other prey items are taken inadvertently in the process of consuming barnacles or that Heliaster consumes everything it comes across while it is on its way to the barnacle patch. Because Heliaster lives in crevices or under boulders it is likely to encounter various mobile invertebrates as it moves to the top of the boulder where the barnacles are found. Heliaster may be capable of switching to or surviving on different sources of nutrition (e.g. Brachiodontes or herbivorous gastropods) when its preferred prey is rare or unavailable. In fact there is considerable variation in the diet; percent barnacles in the stomach samples varied from as low as 5% to as high as 99% of the prey items. A food web (actually a subweb of this community) has been drawn.based on these stomach contents, information from the literature, and personal observations of prey species at Statio~ Beach (see Fig. 4a). Heliaster acts as a secondary, tertiary and even quaternary consumer in this web but gains 23 a) HeUaster Bivalves H61blvorous Gastropods Barnacles Chltons 1 sp. 8 spp. 14 spp. 2 spp. b) Hellaster Bivalves Herbivorous Gastropods B~rnacl6s Ch;ton~ Brachiopod spp. spp. 13 spp. 14 spp. 3 2 1 sp. Fig. 4. Heliaster dominated food webs from the intertidal communities of the Northern Gulf of California. Web Ca) is based on stomach sanples collected by N. Yensen in 1974-1976; (b) is ~,lodified from Paine (1966) • 24 the majority of its nutrition from lower trophic levels (i.e. barnacles and herbivorous gastropods). A similar structure constructed by Paine (1966) is shown for comparison (Fig. 4b) and although the two foodwehs differ in some details the overall structure is the same; Heliaster is a top carnivore in a complex and diverse food Iveb consuming a wide variety of prey but with a strong preference for barnacles. It is important to note that Paine's food web is based on observations from spring 1962-1964 from 3 different locations in the No~thern Gulf of California; Puerto Penasco (Pelican Point), Sonora and San Felipe and Puertocitos, Baja California Norte. Paine found barnacles comprised 51% of the diet with bivalves accounting for 31% compared to 93% and 2% respectively in my analyses. This suggests considerable variation in Heliaster diet over long or short term time spans as well as over its geographic range. Individual Species Densities and Distributions Mean density (individuals/square meter) and 95% confidence intervals around the mean for every year of sampling in each of the two habitats are presented· for 20 study species in figures 5-13. These figures also show mean density and 95% confidence intervals for the period before the Heliaster die-off (1976) and the combined post-die-off data (1981-1984). 25 Because of the very low density of Heliaster in the reef flat prior to the die-oft (mean .09 ind/m, s.e.= +/-.11) compared to the boulder zone (.45 ind/m, s.e. .23) the reef flat is treated as if it were a control for the Heliaster removal; Heliaster were never abundant there. If some factors other than the removal of Heliaster (e.g. temperature) were affecting population densities then changes, if any, should be of equal magnitude in both habitats. I will demonstrate that many of the changes in population density were restricted to the boulder habitat and restricted to prey species. Prey species Barnacle Density." Chthamalus anisopoma density (mean percent cover) varied significantly over time but with no clear pattern (Fig. 5) • Four out of five pairs of consecutive years of data had significantly different mean percent cover. A comparison of the combined 1974 and 1975 data with the combined 1979-1983 data indicates that over the long run there has been no significant change in percent cover (mean % cover = 64.5 and 59.3 respectively, p>.05 There is no evidence (even on boulder tops) of any consistent response of Chthamalus to release from Heliaster predation. Chthamalus experiences great variation in abundance over space and time (Dungan 1984, Lively 1984). During the year, peak "settlement periods have great 26 a) Chtham alus b) 100 - 90 - a::: IJJ 80 - o> u 70 - I Z IJJ 60 - ,.....,..... U a:: - LLIa. 50 - 40 - I i 74 75 798081 82 83 BEFORE AFTER YEAR Fig. 5. Mean density (+/- 95% confidence intervals) for Chthamalus anisopoma for 1974, 1975 and 1979-1983, Before=1974 and 1975 combined; After=1979-1983 combined. 27 influence on the percent cover of Chthamalus (Dungan 1984, Malusa 1986). Other factors that influence Chthamalus density, such as predation by Morula, Acanthina, crabs, fish and others, and competition for space frbm algae and mussels (Brachiodontes), vary in importance both seasonally and from year to year (Yensen 1979, Lively and Raimondi in prep.). Chthama1us also experiences intense intraspecific competition for space and forms hummocks which eventually falloff and provide space on boulder tops. Disturbances in the form of boulder rolling or boulder smashing also cause barnacle mortality. It is possible that any small response by Chthamalus to the Heliaster die-off was masked by these other factors. However, if Heliaster predation regulated Chthamalus density a measurable increase in barnacle population density would be expected, but no such event occurred. Mobile Invertebtate Densities. Six species of Heliaster prey exhibited dramatic increases in boulder zone densities between 1976 and 1981 (FLg.s 6, 7, 8). Tegula spp. (3 species, I.rugosa, I. mariana, and I. corteziana were not distinguished in the 1976 data) and Nerita funiculata, both of which are herbivorous gastropods and Heliaster prey, show patterns of dramatic populatiori increase in the boulder zone and virtually no change at all in the flat reef area (F~g. 6). Tegula boulder zone density rose from 1.57 in 1976 to 19.22 for 1981-1984 combined (p<.OOI) and ~erita boulder 28 a) regula b) Tegula 40 32 a:: l.I.I I- 24 l.I.I ::E l.I.I 16 a:: o \\ 1 , 0 ...... &.0=..----"""- ...... -- 76 81 82 83 84 BEFORE AFTER YEAR Fig. 6. ~1ean density (+/- 95% confidence intervals) for Tegula spp. and ~erita funiculata. Open circles or bars~dicate reef flat densities; filled hars or circles indicate boulder zone densities. 3efore =1976; After=1981-19D4 combined. 29 a) Milrella b) Milrel/a 40 32 a::: IJJ ..... 24 IJJ ~ UJa::: 16 I 1 I 76 81 82 83 84 BEFORE AFTER YEAR Fig. 7. Mean density (+/- 95% confidence intervals) for Mitrella gutata and Morula ferruginosa. Open circles or hars indicate reef flat densities; filled circles or bars indicate boulder zone densities. 3efore = 1976; After = 1981-1984 combined. 30 a) CO/iJmbe//o b) CO/iJmbel/o 10 a:: LaJ I- 6 LaJ :E LaJ 4 a:: « :::::> 0 2 --- ~ en f -- \ 0 , en" \ I -'« :::::> C) TiJrbo d} TiJrbo 0 > 20 O· Z 16 IJ. 0 12 a:: lJ.J III :E 8 :::::>z 4 ~---f---f-_ '-2 12 \ 0 \ I I I 76 81 82 83 84 BEFORE AFTER YEAR Fig. 8. ~I e and ens it Y (+ /- 95% con f ide n c e in t e r val s) for Columbella aureomexicana and Turbo fluctuosos. Open circles or bars indicate reef flat densities; filled circles or bars indicate boulder zone dens ities. Before=197fi; After=1981-1984 combined. 31 zone density increased from 0.94 ind/m in 1976 to an average density of 24.81 ind/m for the years 1981-1984 combined (p<.001). None of 1981-1984 densities differed from one another (at the .05 level) in either case. Mitrella gutatta and Morula ferruginosa, both carnivourous gastropods and Heliaster prey, also exhibited dramatic increases in boulder zone densities (Fig. 7). Between 1976 and 1981 Mitrellaincreased over ten fold (2.69 ind/m to 29.05 ind/m; p<.001) while Morula increased nearly as much (1.92 ind/m to 18.16 ind/m; p<.OOl). Mean boulder zone densities for 1981-1984 did not differ from one another in both cases. Reef flat densities were low initially and have stayed low throughout the study period in the case of Mitrella and no change in intertidal distribution is indicated. Morula on the other hand did show some reef flat density fluctuation but the 1976 reef flat density (5.05) is not different than the 1984 density (5.64). Morula is clearly more abundant in the boulder field at this time while in 1976 it was more abundant on the reef flat. Morula and Heliaster have an extensive dietary overlap; both species. consuming Chthamalus, Brachiodontes (which are more abundant in ~he bouler zone; Morula's preferred prey is probably Brachiodontes) and other gastropods. This may be a case of Morula utilizing more of its fundamental niche in the absence of Heliaster whereas in the presence of Heliaster it may have been restricted to the reef flat (its 32 realized niche) by predation, predator avoidance and competition with Heliaster. Columbella auromexicana, an herbivorous gastropod; was originally more abundant in the reef flat than the boulder zone (Fig. 8a). Density in both habitats increased initially and then decreased until it was no higher in 1984 than in 1976 (1.90 vs. 1.33 reef flat and 1.49 vs. .43 boulders; n.s in both cases). The intertidal distribution of Columbella has not shifted throughout the study period (Fig. 8b). The large herbivorous snail, Turbo fluctuosus, also follows the pattern of sustained increase in the boulder zone following the Heliaster decline; 1981 - 1984 densities do not differ from one another but all are significantly higher than the 1976 density (p(.OOl;Fig. 8c). There is also a significant increase in the reef flat in 1981 and 1982 and then a decline back to pre-die-off densities (1976 mean 1.19, 1984 mean = 1.26 n.s.). Turbo shows no change in intertidal distribution pattern (Fig. Sd). Both Turbo and Columbella were prey of Heliaster .and may have shown signs of predation release in the boulder zone. The amount of gastropod prey taken by Heliaster may have been disproportionately higher in the reef flat than in the boulder zone, enough to have had an impact on these species (which had relatively high reef flat densities in 1976). 33 Liocerithium ]udithae and Chiton ~. are two Heli~~t~ prey that seem to have truly decreased in abundance (Fig. 9). Both species exhibit the same pattern of dramatic decline in the boulder zone (1981-1984 Liocerithium densities were all lower than the 1976 density; p<.005; in the case of Chiton 1981, 1982 and 1984 are all significantly lower than the 1976 density; p<.005; 1983 was n.s.). In the reef flat, where both of these species were very rare, there was no detectable change in density nor was there any shift in intertidal distribution. Both of these species are more temperate rather than troptcal in distribution (ranges for both species are Bahia Magdalena on the Pacific side of the Gulf and throughout the Gulf) and may have been affected by the warm temperature periods in 1978 and 1981-1984. These species may have expereinced increased predation from Morula and Mitrella (this could be true for other species as well). The chiton may have had increased competition for algae from the other herbivorous species which have increased in abundance (e.g. Turbo and Tegula) Neorapana tuberculata, a large predatory snail and occasional Heliaster prey, is rare at Station Beach. Neorapana reef flat densities showed no change and densities in the boulder zone were very low (0.13 to 0.33 ind./m) with high variances (Fi~. lOa). Comparison of before and after data (1976 vs. 1981-1984) suggests that densities did not change significantly in either habitat (Fig. lOb). 34 a) Liocerithium b) L iocerithium 20 16 0: IJJ I- 12 IJJ ::E w 8 «0: :::::> 0 4 en ...... en 0 \1 ...J« :::::> c) Chiton d) Chiton Q > 5.0 Q Z 4.0 u.. 0 3.0 0:: W CD ::E 2.0 :::::> z 1.0 0 76 81 82 83 84 BEFORE AFTER YEAR Fig. 9. Mean density (+/- 95% confidence intervals) for Liocerithium iudithae and Chiton ~. Open circles indicate reef flat densities; filled circles or bars indicate boulder zone densities. Before=lo76; After=1981-1984. 35 a) Neorapana b) Neorapana 0.5 0.4 n:: IJ.J ~ 0.3 ::E 0.2 0.1 I c) Acanfhina d) Acanthina 20 16 IJ. o 12 0: IJ.J ~ 8 :::> Z 4 76 81 82 83 84 BEFORE AFTER YEAR Pig. 10. Xean density (=/- 95% confidence intervals) for ~eorapana ~uberculata and Acanthina angelica. Open circles or bars indicate reef flat densities; filled circles or bars indicate boulder zone densities. Before = 1976; After = 1981-1984 combined. 36 Acanthina angelica population density appeared to decrease but also showed large variance (Fig. 10c,d). This snail, a barnacle specialist and Heliaster prey, is a Gulf of California endemic. In the reef flat, where densities were initially low, there has been no significant change in the population. In the boulder field on the other hand densities have signifcantly decreased (2-way ANOVA and Tukey tests show 1976 density significaatly higher from each succeed~ng ye~r's data, p<.OOI, and that 1981-1984 do not differ from one another). This apparent decrease may be due to the fact that Acanthina has a patchy distribution (Yensen 1979; Turk 1981). Additional data on Acanthina population density (also collected by the author) within Acanthina patches in 1980, 1981, and 1982 yielded densities of 5.31, 3.73, and 12.77 respectively. Non-prey species Since 1978 both Othilia tenuispina, a small five-armed sea star and Gulf of California endemic, and Aplysia californica, the sea hare, have become quite rare at Station Beach (Fig.ll). In the reef flat, the preferred habitat of Othilia, the once common sea star decreased to zero density and did not increase during the study. Boulder zone densities show no real trend as the 1976 density was not significantly different from zero. Othilia were found in 1978 with the same white lesions as associated with the 37 0) Aplysia b) Aplysia 0.5 0.4 0.3 a::: w 0.2 I IJJ 0./ ~ IJJ o a::: "' 0.4 CJ) -l lL. o o e) Echinometru f) Echinometra a::: IJJ 0.5 CD ~ 0.4 ::> z 0.3 0.2 0./ o 76 8/ 82 83 84 BEFORE AFTER YEAR Fig. 11. Mean density (+/- 95% confidence intervals) for Aplysia' californica, Othilia., tenuispina, and E~hinometra ~anb~tiriti. Open circles or bars ind icate reef flat density; filled circles or bars indicate boulder zone density. Before=1976; After = 1981-1984. 38 Heliaster die-off and numerous individuals were found dead or dying at Station Beach at that time. In the case of Aplysia, populations in both habitats decreased to zero density and have remained so throughout the study. Aplysia is known to prefer colder waters (the range extends from Northern California to the Gulf of California) and is known to be seasonal at Station Beach found more frequently in fall and winter than in spring and summer. The sea urchin, Echinometra vanbrunti was rare at Station Beach but is now commonly found in the boulder zone (Fig. 11e,f). Statistically this trend is not clear; Tukey tests indicate that 1976 and 1981 are different (p<.OS) but 1976 does not differ from 1982, 1983, or 1984. Echinometra is one of the most common urchins in the Gulf and ranges throughout the Gulf and south to Peru and the Galapagos Islands (Brusca 1980); in the Gulf it is most abundant in warmer waters of the southern Gulf. It seems likely that Aplysia, Othilia and Echinometra were affected by the same temperature increase associated with the Heliaster decline. The two ophiuroids, Ophioderma panamense and Ophioneries annulata, and the sea cucumber, Selenkothuria lubica, had stable population densities during the study in both habitats (Fig 12). The densities of these three non-prey species seem to have been unaffected by either the 39 0) Ophioderma b) Ophioderma 2.0 1.6 1.2 a: w 0.8 1- W :E 0.4 W 0 a: « ::> c) Ophioneries d) Ophioneries 0 4.0 CJ) ...... 3.2 CJ) «-' 2.4 ::> c 1.6 > Ci Z '0.8 \ 9=--_.&._--9--- l1.. 0 \j iii 0 e) Selenkothuria f) Selenko fhuria a: w 8.0 CD :E 6.4 ::> Z 4.8 3.2 1.6 o \, iii 76 81 82 83 84 BEFORE AFTER YEAR Fig. 12. Mean density (+/- 95% confidence intervals) for Ophioderma ~., Ophioneries annulata, and S'elenkothuria luhrica. Open circles or bars indi cate reef flat density; filled circles or bars indicate boulder zone density. Before = 1976; After = 1981-1984 combined. 40 Heliaster disappearance or the elevated temperature phenomenon. Palythoa ignota, a colonial anemone, is found in great abundance on the lower reef flat at Station Beach but there are few small colonies in the boulder zone. There has been no change in density in either habitat during this study although there is great variation in density due to a clumped distribution (Fig. 13a,b). The fireworm, Eurythoe complanata, did not change in the reef flat but in the boulder zone the 1984 density is significantly higher than 1976 (p<.OOI) but 1981, 1982, and 1983 were not (Fig. 13c). Comparing before and after densities (Fig. 13d) suggests that Eurythoe has been relatively stable over the study period. The last species presented here is the endemic hermit crab, Clibanarius digueti. Clibanarius appears to have had a decrease in the boulder zone (Fig. 13e,f). Densities in 1976, 1981, and 1982 do not differ from one another although 1983 and 1984 are significanly lower than any of the previous years (p<.OOI). The comparison of combined before and after data do show a significant decrease (p<.05) in the boulder zone. There was no significant change in the reef flat zone. Clibanarius .is found primarily in clumps of individuals (Mackie and Boyer, 1977 found approximately 87% of all Clibanarius in clumps of 50 or more) and often in tight bunches of hundreds of 41 0) Polythoo b) Polythoa 200 160 120 c:: uJ 80 I- W ::E 40 w 0 'If=:==- 1 c:: « 0) d) :::::> £urythoe £urythoe 0 CJ) 5.0 ...... 4.0 CJ) «-' 3.0 :::::> 0 2.0 > 0 Z 1.0 0 ( .( lL.a e) Cli/Janariu5 f) Clibanarius 0:: W In ::2E :::::> z 20 10 0 \~i 76 81 82 83 84 8EFORE AFTER YEAR Fig. 13. Mean density (+/-.95% confidence intervals) for Palythoa ignota, Eurythoe complanata, and Clibari~riu~ digti~ti. Open circles or bars indicate reef flat density; filled circles or bars indicate boulder zone density. Before = 1976; After = 1981- '1984 combined. ~~~------==-----~~~~-=====~~------. 42 individuals. Clibanarius is not a prey of Heliaster (Yensen pers. comm.) but because it is shell limited it could have depended on Heliaster predation on gastropods as a source of free shells. In summary, with respect to habitat the 20 species examined follow different patterns in the boulder and reef flat zones (Table 4). In the boulder zone 7 species increased in density, 6 decreased and 7 were unchanged. In the reef flat 3 species increased, 2 decreased, and 14 species were unchanged. The prey species and non-prey species also followed different patterns. In the boulder zone 6 prey species increased, 3 decreased and Z were unchanged. In the reef flat 3 prey species increased and the other 7 species were unchanged. Non-prey sp~cies showed the following pattern: in the boulder zone 1 increased, 3 decreased and 5 were unchanged. In the reef flat 2 non-prey decreased and 7 species were unchanged. The only non-prey species that did change were Aplysia, Othilia, Echinometra and Clibanarius three of which are presumed to have been affected by elevated sea surface temperatures. It is clear from this synopsis that changes in abundance were distinctly more profound among Heliaster prey species and in the boulder zone habitat. 43 Table 4. Species density summary table. Number of species that increased (+), decreased (-), or showed no change (0) in density over the study period in two habitats. BOULDER ZONE REEF FLAT 7 + 3 6 2 7 o 14 Prey Non-Prey Prey Non-prey 6 1 + 3 o 3 3 o 2 2 5 o 7 7 44 . Guild Structure Prey vs. Non-prey species The data from ten species of Heliaster prey have been combined and the mean density of individuals in the prey guild and the 95% confidence interval around the mean has been calculated for each year in both habitats (Fig. 14a; see Table 1 for a list of these prey). Chthamalus anisopoma was not included in this analysis because the percent cover data were not comparable. There was clearly a pronounced increase in combined prey species only in the boulder zone where Heliaster was formerly abundant. Prey density increased between 1976 and 1981 (3.18ind/m to 10.09ind/ffi; p<.001) followed by a stable period. In the reef flat there was a small increase in prey density followed by a return to the pre-die off levels by 1984; 1981 (2.06ind/m) and 1982 (1.94ind/m) both are higher than 1976 (.85ind/m, p<.05); 1984 (.99ind/m) is not significantly higher than 1976. Non-prey species were also grouped together and analyzed (Fig. 14b). Palythoa ignota, the colonial anemone, and Clibanarius digueti, the hermit crab, were not incuded in this part of the analysis because both had extremely high variances (as explained earlier) and densities. This would have masked any pattern that the non-prey guild may have exhibited (see Table 2 for non-prey species list). The combined density of non-prey in the reef 45 a) Prey 15 12 a: IJJ I IJJ 9 ::E IJJ 6 a: oCt => o 3 en ! ...... Q en 0----,.-\ \~\ 1 I I I ...J oCt b) Non - prey o=> > 2i z 1.00 IJ.. 0 .75 a: lJ.J CD ::E .50 => z .25 a Q a '9\ ? i i 76 81 82 83 84 YEAR Fig. 14. Mean density C+/- 96% confidence intervals) for Prey and Non-prey guilds. Open circles indicate reef flat density; filled circles or bars indi cate boulder zone density. 46 flat was quite low but was stable over the entire study period (Fig. 14b). Densities were higher in the boulder zone, but there was no change over the study period. Distribution of abundance among prey and non-prey species was examined by assigning ranks based on density within the guild (Table 5). Spearman rank correlation coefficients for 1976 vs. 1981 and for 1981 vs. 1984 were calculated and compared to one another within both habitats (Table 6). There was strong evidence for a major restructuring of the community following the Heliaster disappearance, but only among prey species and only in the boulder zone where He1iaster predation had been intense. In the boulder zone the prey rank order was greatly disturbed over the period of the He1iaster crash (R=-.10 n.s.) but was quite stable over the 1981-1984 period (R=.92; p<.0005). These two rank correlation coefficients are significantly different (z=3.02;p<.05). In the reef flat both 1976 vs. 1981 and 1981 vs. 1984 prey ranks were very highly correlated (R=.92, p<.0005 and R=.85, p<.0025 respectively) and these two correlation coefficients do not differ from one another (z=.58, n.s). Thus the ranks of prey species in the reef flat were stable over the period in which the H~liaster crash occurred and equally stable over a period in which no catastrophe occurred. In the boulder zone both 1976 vs. 1981 and 1981 vs. 1984 non-prey ranks showed high correlations (R=.86; p<.005 47 Table S. Prey and non-prey ranks. Density ranks of prey and non prey species over the five years of the study period in the two habitats; Boulder zone and Reef flat. PREY Boulder Zone Reef Flat 76 81 82 83 84 76 81 82 83 84 Species Acanthina 2 6 6 8 6 6 8.5 5 5 8.5 Morula 5 3 1 3 5 1 1 1 1 1 Turbo 6 5 5 5 4 3 3 2 2 3 Tegula 7 4 4 2 2 5 4 4 4 4 Nerita 8 2 3 4 1 8 7 8.5 7 8.5 Columbella 9 7 7 9 7 2 2 3 3 2 Liocrerithium 1 9 8 6 8 4 5 8.5 6 6 Mitrella 4 1 2 1 3 9.5 8.5 6 9 5 Chiton 3 8 9 7 9 7 6 8.5 9 8.5 Neorapana 10 10 10 10 10 9.5 10 8.5 9 8.5 NON-PREY Boulder zone Reef flat 76 81 82 83 84 76 81 82 83 84 Ophioderma 2 4 2 2 3 6 3 3 2 6 01!hioD~ri!i1iZ 5 5 6 5 7 6 6.5 7.5 6.5 6 S!il~DkQthllLi il 4 3 4 3 1 8.5 6.5 4.5 6.5 6 Clibanarius 1 2 1 6 5 2 2 2 3 6 ~ill:£tbQa 3 1 5 1 4 1 1 1 1 1 Eurythoe 6 6 3 4 2 6 6.5 4.5 6.5 6 Echinometra 9 7 7 7 6 8.5 6.5 7.5 6.5 2 Othilia 8 8.5 7 8.5 8.5 3 6.5 7.5 6.5 6 Al!lysi g 7 8.5 7 8.5 8.5 4 6.5 7.5 6.5 6 48 Table 6. Spearman rank correlation coefficients for prey and non-prey in the boulder and reef flat zones. Spearman rank correlation coefficients for 1976 vs. 1981 and 1981 vs. 1984 are shown for prey (10 spp.) and non-prey (9 spp.) in the two habitats. Z statistic compares R values (significant Z indicates R values are different). PREY Boulder Zone Reef Flat 1976 vs. 1981 Rs=-.10 n.s. Rs= .92 p<'0005 1981 vs. 1984 Rs= .92 p<.0005 Rs= .85 p<.0025 z=3.02 p<' 01 z= .58 n. s. NON-PREY Boulder zone Reef flat 1976 vs. 1981 Rs= .86 p<'005 Rs= .64 p<.05 1981 vs. 1984 Rs= .63 p<.05 Rs= .40 n. s. z= .68 n. s. z= .57 n. s. 49 and R=.63;p(.05 respectively) and these correlations do not differ from one another (z=.68, n.s.;Table 6). In the reef flat there was no difference between 1976-1981 correlation (R=.64; p(.05) and the 1981-1984 correlation (R=.40 n.s.;z=.57, n.s.). There was no "shake up" in the ranks of the non-prey species during the study in either habitat. In summary every test of the prey and non-prey guild densities and rank correlations fits with the outcome expected if Heliaster predation had a large effect on the community. Large changes in density and rank order of abundance occurred but only in the prey guild, only in the boulder zone and only over the period 1976-1981. Trophic Guilds Mean density of individuals in three trophic guilds (carnivores, herbivores and detritivores) and 95% confidence intervals were calculated for each year in each habitat and are presented in Fig. 15. The density of carnivores (6 species, 4 of which are Heliaster prey increased substantially in the boulder zone between 1976 (2.46ind/m) and 1981 (8.13 ind/rn;p(.OOI) and then stabilized at a level slightly higher than the 1976 level (Fig. 15a). In the reef flat there was no significant change in overall density of carnivores over the study period. The herbivore guild includes five prey species and two non-prey species. In the boulder zone there has been a 50 a) Carnivores 10 8 ..,.. 6 a: w 4 1 I W :E 2 f H---f--t 1 l \ O ~ ~\rl--~I--~I--~I-- b) Herbivores T 10 T 8 (/) ..J c::( :::> ....~ Cl 4 > Cl 2 ~ !\~ IJ.. o Ij I I I o c) Detritivores 10 8 r 6 4 2 l~ o T 1" 1 76 81 82 83 84 YEAR Fig. 15. Mean density (+/- 95% confidence intervals) for carnivore, herbivore, and detritivore guilds. Open circles indicate reef flat densities; filled circles indicate boulder zone densities. 51 dramatic increase over the 1976 density (1976=1.17ind/m and 1981=7.52ind/m: p (Fig.15b). Herbivores show a slight peak in density in 1981 in the reef flat (1981 > 1976; pe001) but 1976, 1982, 1983, and 1984 are not significantly different from one another. The detritivore guild includes one Heliaster prey and four species that are non-prey (see Tables and 2). There was a sharp decrease in detritivore density in the boulder zone (Fig. 15c.). None of the 1981-1984 densities differ from one another but are all significantly lower than the 1976 density (p<.OOl in all cases). This decline is almost due entirely to Clibanarius and Liocerithium both of which show sharp declines in density (Fig.s 9 and 13). If both Clibanarius and Liocerithium were excluded from this guild the pattern would be one of no change in density in both habitats. Rank correlation a~alysis was ~erformed for each of the trophic guilds following the same procedure used for the prey and nonprey guilds. Rank orders for each of the three trophic guilds are presented in Table 7 and rank corre1atios are presented in Table 8. No great change occurred in the carnivore rank order in either habitat during the study. Rank correlations between 1976 and 1981 loJere not significantly different from the 1981 vs. 1984' correlation 52 Table 7. Carnivore, herbivore and detritivore ranks. Density ranks of carnivore, herbivore, and detritivore species over the five years of the study in the two habitats; Boulder zone and Reef flat. CARNIVORES Boulder Zone Reef Flat 76 81 82 83 84 76 81 82 83 84 Species As;;aDtbiDa 1 3 3 4 3 3 2.5 4.5 2 4.5 Morllla. 3 2 1 2 2 1 1 1 1 1 Mi tI:el l a 2 1 2 1 1 5.5 2.5 2 4.5 2 Neor:apana 5 5.5 5.5 5 4 5.5 5 4.5 4.5 4.5 Opbioderwa 4 4 4 3 5 4 5 4.5 4.5 4.5 Othjlja 6 5.5 5.5 6 6 3 5 4.5 4.5 4.5 HERBIVORES Boulder zone Reef flat 76 81 82 83 84 76 81 82 83 84 Turbo 2 3 3 3 3 2 2 1 1 2 Iegllla 3 2 2 1 2 4 3 3 3 3 llIerita 4 1 1 2 1 6 5 5.5 4 6 Colllwhella 5 4 4 4 4 1 1 2 2 1 CbjtOD 1 5 5 4 5 5 4 5.5 6 6 EcbiDowetra 7 6 6.5 6 6 7 6.5 5.5 6 4 Aplys;a 6 7 6.5 7 7 3 6.5 5.5 6 6 DETRITIVORES Boulder zone Reef flat 76 81 82 83 84 76 81 82 83 84 Opb;oDer;es 3 3 2 2 3 3.5 3 2 1 3.5 Sekenkotbllr:ia 4 2 4 3 1 5 4.5 3.5 4.5 3.5 Ellq~tblJe 5 5 3 4 2 3.5 4.5 3.5 4.5 3.5 C]ibanar:ills 1 1 1 5 5 1 1 1 2.5 3.5 Ijccex:jtbjllm 2 4 5 1 4 2 2 4 2.5 1 53 Table 8. Spearman rank correlation coefficients for carnivores, herbivores and detritivores in the boulder and reef flat zones. Spearman rank correlation coefficients for 1976 vs. 1981 and 1981 vs. 1984 are shown for carnivores (6 spp.), herbivores (7 spp.) and detritivores (5 spp.) in the two habitats. Z statistic compares R values (significant Z value indicates R values are different). CARNIVORES Boulder Zone Reef Flat 1976 vs. 1981 Rs= .81 n. s. Rs= .39 n. s. 1981 vs. 1984 Rs= .87 p<.05 Rs= .82 n. s z= .04 n.s z= .88 n. s. HERBIVORES Boulder zone Reef flat 1976 vs. 1981 Rs= .46 n.s Rs= .72 p<.05 1981 vs. 1984 Rs=1.00 p<.0025 Rs= .79 p<.05 z= 6.27 p<.0005 z= .20 n. s. DETRITIVORES Boulder zone Reef flat 1976 vs. 1981 Rs= .87 n.s Rs= .60 n. s. 1981 vs. 1984 Rs= .34 n. s. Rs=-.30 n. s . z= . 96 n. s. z= .97 n. s. 54 in either habitat (z=.04 and .88 both n.s.). So rank order was aD stable over a period that included the Heliaster die-off (1976-1981) as it was over a period that did not include such a disturbance (1981-1984). The low correlation coefficient for 1976 vs. 1981 reef flat carnivores is due to a large number of tied ranks in the reef flat. Major changes in the rank order of herbivores occurred over the period of the Heliaster decline but only in the boulder zone (Tables 6 and 7). Herbivore rank correlation was disturbed between 1976 and 1981 (R=.464) in the boulder zone but was significantly higher for 1981 vs. 1984 (R=1.00; 2=6.27, p<.0025) while in the reef flat the correlation between 1976 and 1981 (R=.721, p<.05) is no different than that for 1981 and 1984 (R=.785, p<:05;Z=.20 n. s. ) . Detritivore rank correlation coefficients for 1976 vs 1981 are not significantly different from those for 1981 vs. 1984 in either habitat (reef flat 2=.974; boulder zone 2=.955; both n.s). So there has been no real change in the rank order of abundance in this guild over the study period in either habitat. The low correlation coefficients in the 1981 vs. 1984 cases were due to the tied ranks and did not really reflect large changes in the rank order. Results of the trophic guild density and rank analysis can be summarized as follows. Herbivores showed a large increase in boulder zone density (and possibly some 55 reef flat increase) and there was a large shake up in the boulder zone rank order between 1976 and 1981 with a period of stable ranks following. Carnivores were unchanged in the reef flat but increased in abundance and stayed abundant in the boulder zone and there was no change in rank order. This guild in effect increased density as a unit in the boulder habitat. Detritivores seem to have gone through some change but most of this was due to one species (Clibanarius); the remainder of the guild was relatively stable in density and rank order. Herbivores experienced intense predation pressure from Heliaster and carnivorous gastropods, and this was not equally distributed throughout the guild. I suspect that the dramatic increase in density along with the shake up in the ranks (restricted to the boulder zone) is most likely due to differential release from predation for different guild members and possibly the results of competitive interactions within the guild. It is this level of the food web that responded most to the Heliaster crash because of direct effects in terms of relaxed Heliaster predation and indirectly through the carnivore guild's increased predation on these species (possibly altering competitive interactions). The disappearance of Heliaster may have affected each of the members of the carnivore guild equally. ~any of these species overlapped Heliaster's diet in such a way that 56 the resources freed up by Heliaster's demise may have benefitted each species more or less equally. Also each of these species was probably similarly affected by Heliaster predation. For these reasons the guild behaved somewhat as a unit; increasing density in response to newly available resources (such as Chthatmalus, Brachiodontes or other gastropods) and slightly relaxed predation pressure. Densities of all 20 species from the study have been ranked within each habitat and the rank vs. abundance curves for reef flat and boulder habitats are presented in Fig. 16a and b. The curves for the boulder zone seem to show that the years following the Heliaster die-off are all similar but they differ from the pre-die-off curve. These curves (which are necessarily constrained to certain shapes) suggest that a new state has been attained in the community, more so in the boulder zone then tbe reef flat. The reef flat curves indicate that there was a temporary increase in dominance followed by an apparent trend back toward the original distribution of abundance. The type of shift in abundance distribution one would expect from the removal of a keystone species (~ steep curve intersecting the ahscissa at a lower rank number) is not seen here. 57 a) 1976 1 0 0 196 1 9 •6 1982 e 0 1 983 7 • 1984 6 R e e f flat zone ..... >- (/) 2 c Q) 1 "c co 2 4 6 8 1 0 1 2 CI) E b) Rank -- 40} CI) , () c 0 1976 co • 1 98 1 "c 3°1 IJ 1 982 ::I .Q 25 0 1983 < III 1984 20 Boulder zone 1 5 10 5 2 4 6 8 1 0 1 2 Rank Fig. 16. Rank vs. abundance curves for the 20 species sub- community at Station Beach Reef in the boulder and reef flat zones. 58 Community Patterns Overall abundance The data from all of the study species, with the exception of Heliaster, Chthamalus, and Palythoa were combined and an overall density (no. ind./m) was calculated for each year and each habitat. This approach was used as a crude measure of the standing crop. The mean density in this case represents the average number of individuals (of 18 species) per square meter for a particular year and habitat. A two-way ANOVA indicated that there was no effect of time on overall abundance (F=.579, n.s.) nor was there any interaction between time and habitat (F=.255, n. s. ) . The other factor, habitat, was significant reflecting the large difference in abundance of invertebrates between boulder zone and reef flat. Diversity Number of species, HI, and J for each habitat and year is presented in Table 9. Heliaster, Palythoa, and Chthamalus were not included in these calculations so the maximum possible number of species found would have been 18. In the boulder zone there was no clear pattern of change in number of species over time, but in the reef flat there is a clear indication of a steady decline in number of species from 14 in 1976 to 7 in 1984. This result is deceiving, 59 Table 9. Number of species, HI, and Jdiversity measures for the 20 species sub-community at Station Beach Reef. Heliaster, Chthamalus and Palytoa were not included in these measurements. BOULDER ZONE REEF FLAT 76 81 82 83 84 76 81 82 83 84 # Spp. 17 15 14 16 16 14 11 9 9 7 HI 2.06 2.04 2.12 1.98 1.87 1.69 1.48 1.23 1.24 1.25 J .73 .75 .80 .71 .68 .64 .62 .56 .56 .64 60 however, because most of the species that dropped out in the reef flat (Ophioneries, Ophioderma, Selenkothuria, Eurythoe, Acanthina, Nerita, and Chiton) were so rare to begin with that 95% confidence intervals for their 1976 densities included zero. The only two reef flat species that truly became locally extinct along the transects were Othilia and Aplysia (most likely at the same time and from the same cause as Heliaster). H' values indicate a slight decrease in diversity in the reef flat and this is due to the rare species going to zero density. The measure of evenness on the other hand was quite stable throughout the study in the reef flat. In the boulder zone a similar pattern holds although the decrease in diversity is much less; evenness was again stable over time. A species list of common intertidal invertebrates at Station Beach before and after the Heliaster die-off is shown in Appendix B. This list shows that most common species of invertebrates from station beach are still present and that the only species that have been lost were either rare or were heavily collecte by tourists. The observations of overall abundance and diversity measures can be summarized as follows: There has been no detectable change in the "standing crop" (as estimat'ed by numbers of individuals). The members of the community have stayed basically the same but the distribution of abundance among those community members has been shifted around a bit. 61 The prey and herbivore guild data best illustrate this last point. There have been slight changes in diversity of the sub-community represented by the 21 study organisms with the rare species most affected. No clear competitive dominant species has appeared in the community in the absence of Heliaster as predicted by the keystone species hypothesis. Most, if Got all, of the most common species found in early studies are still found at Station Beach and furthermore many of the missing species are those that may have died out with Heliaster, fell victim to heavy collecting by tourists, or were rare to begin with. In many ways this community shows a great deal of stability over a period in which a major preturbation occurred. General Discussion It is necessary to address two problems with this study before proceeding with any discussion of these results. All of the data concerning abundance over time suffer from two disadvantages: the gap in data between 1976 and 1981, ana the fact that the pre-die-off community composition is based primarily on one year of data (1976). It is rare that nature provides us with the type of natural experiment that a catastrophic population crash creates. It is even rarer that we are prepared (with baseline data) to take advantage of such an event, but this is exactly what occurred here. The 1976 samples (Mackie and Boyer 1977) provided the only reliable community-structure data available from a time close to (but before) the Heliaster die-off. Other sources of data on some of my study species, while not statistically comparable to the data from Mackie and Boyer (1977), indicate that in 1976 the community was not noticably different from other pre-die-off years. The large gap in the data is due to the fRet that the ecological potential of the Heliaster die-off was not even recognized until late spring, 1979 at which time I began work on stomach sample analysis and Heliaster censuses. Transect studies did not begin until spring 1981. It is likely that some species responded to the Heliaster 62 63 die-off within the period of this 1978-1981 gap, and that the 1981-1984 data show the tail ends of those responses. However, Heliaster did not reach low densities until the summer of 1978, and there is no reason to believe that important changes would have completely reversed or disappeared by spring 1981 (less than three years after the Heliaster crash). Some species may have taken one or two years to show a response. Although an important phase of the community response was missed, substantial and persistent effects have been detected. This response is confined to the prey species and perhaps a few other species in the boulder zone habitat. The community may still be adjusting to the Heliaster removal or it may have reached a new stable state. The results of this study are reliable enough to demonstrate several important ecological consequences of the Heliaster die-off. Heliaster as "keystone species" Heliaster kubiniji was not acting as a keystone species in this community at the time when it was abundant on Station Beach reef. This conclusion is based on the fact that the removal of Heliaster failed to produce the two most important effects of keystone species removal as defined by Paine (1966, 1969): 1) a substantial drop in local species diversity (number of species) and 2) competitive exclusions 64 of species by the monopolization of some primary resource by a competitive dominant that was formerly held in check by the keystone predator. If Heliasterwas not a keystone species then two important questions arise; 1) Why not? and 2) What exactly was Heliaster's role in this community? The answers to these questions may be of more value to understanding community structuring processes than the discovery that Heliaster was not a keystone species. The only potential resource monopolizer at Station Beach would be the barnacle, Chthamalus anisopoma, but it cannot monopolize space because of the heterogeneity of the substrate in this ~abitat. Also, because Chthamalus is apparently not limited by predation at Station Beach, there is no major predator, not even Heliaster, preventing it from monopolizing space (even on boulder tops). Therefore the initial conditions in this community were such that basic assumptions of the keystone species hypothesis did not apply. I must point out that Paine did not actually say that Heliaster was acting as a keystone at Station Beach but that this may be true in the Northern Gulf of California in general. In Puerto Penasco he suggested that Heliaster may be important to the community at Pelican Point, a steeply sloping granitic rocky intertidal zone (Dungan, 1984, 1986, did find some increase in barnacle abun~ance after 1978). 65 Another example of a proposed. keystone species whose natural decline resuted in limited community impact is the case of Diadema antillarum in the Caribbean. Jackson and Kaufman (1987) state that Diadema is not a keystone predator in the cryptic reef community. Diadema was important as a grazer on reef algae but apparently had no real impact on the encrusting invertebrate community under coral heads or in crevices because it did not prey on dominant competitors. The diversity of the carnivore guild and the fact that there i9 much dietary overlap among the guild members is a phenomenon that may be important to the maintainance or stability of the system. Morula consumes Brachiodontes and Chthamalus predominantly and also preys on other gastropods and molluscs. Acanthina is a barnacle specialist and Neorapana consumes a variety of sessile invertebrates (although probably not barnacles) and other gastropods. In addition to these species there are a number of other carnivores in this community, more mobile crabs and fish among others, that were not included in this study. The real carnivore guild is much larger than indicated in my food web. If added together these species' diets would cover nearly the total of the Heliaster feeding niche. The resources freed by Heliaster's decline were taken up by this guild as a whole which compensated for the Heliaster removal. Though compensation was not perfect, as can be seen by the shifts in herbivore abundance and rank order, it may 66 be the major reason for the relative stability of this community as it passed through a significant ecological catastrophe. The large number of carnivores in this system is analogous to a table with many legs. If one of the legs were removed or cut short the table might tip somewhat but the objects on the table (the analogue of the diversity of the community) would not all falloff. A keystone species situation would be more like a table with three legs all together at one end and a fourth leg, which represents the keystone, alone supporting the other end of the table. If this one leg is removed, the table tips over and most of the objects on top slide off. This "stable-table model" symbolizes what I perceive to be the process holding the diverse invertebrate community together at Station Beach. The complex and heterogeneous physical environment supports a diverse assemblage of invertebrates and algae. The large number of species in each guild with high diet overlap (herbivores as well as carnivores) make for a system that can withstand the removal of important members with only a modest reshuffling of their abundance and distribution. Although not a keystone species in this community, Heliaster is nevertheless an important predator. I propose a continuum for predator importance as a factor influencing community structure. At one end of the spectrum would be a predator that has virtually no measurable impact on the 67 overall structure of the community. At the other end of the spectrum would be the keystone species whose activity is all important to community structure. Many predators probably fall toward the no-impact end of the spectrum, and only under certain conditions can a single carnivore species be considered to be in the keystone end of the range. Heliaster would be toward this end of the spectrum but not a keystone, since its removal did not produce the large-scale diversity reduction that occurred with the removal of Pisaster (Paine 1966), Stichaster (Paine 1971) or Sea Otters (Estes and Palmisano 1974). Factors determining a predator's position on the predator-impact continuum include the following: 1. Predation intensity (density and feeding rate) 2. Diversity of the carnivore gUild. 3. Dietary overlap among the guild members. 4. Presence of a competitively dominant species capable of resource monopolization. 5. Degree to which that competitive dominant is held in check by the predator In a situation where there is intense predation, low carniviore diversity and low carnivore diet overlap, and a resource mbnopolizer held in check by a predator then the predator may act as a keystone species. In the case of Heliaster there was .no monopolizer but there was intense predation as well as high guild diversity and diet overlap, 68 and there was a substantial impact of the Heliaster removal. I predict that there will be a shift back to a community more or less identical to the pre-die-off status in the event of a Heliaster recovery. The present structure appears to have been stable over the 1981-1984 period and there is reason to believe that the pre-die-off community structure was stable for as long as we have records from the area. This could be a case where there was a shift from one stable state (pre-die-off) to another (post-die-off), and that Heliaster's presence or absence pushed the system from the first state to the second. Perhaps there are mUltiple stable states in this community (see Sutherland 1974) or an infinite number of stable states that depend on the magnitudes of biotic and abiotic variables that affect abundance and distribution of species. Morula should be heavily preyed upon and effectively excluded from the boulder zone if Heliaster returns. Prey species should be cropped back as Heliaster modifies its diet to take advantage of the most abundant and nutritional food sources. Once prey species are cropped back to pre-die-off densities, the Heliaster diet should shift back to chiefly barnacles. Guild densities and rank orders should probably approach their pre-die-off status. Also, other species that were affected by temperature such as Aplvsia, Othilia, and Echinometra may return to previous densities when the warmer water period ends. I predict no significant 69 change in diversity if Heliaster does return, and I predict that the recovery will occur only if there are several consecutive years of cooler water temperatures. In the event that Heliaster does not return to its previous abundance at Station Beach the community will most likely maintain the new stable but dynamic state that is visible now. This state will last until the next major perturbation such as a temperature change, the return of Heliaster, or some other event. CHAPTER 5 SUMMARY AND CONCLUSIONS 1. Heliaster kubiniji was an abundant top carnivore in the Northern Gulf of California intertidal invertebrate community at Station Beach, Puerto Penasco, Sonora, Mexico. Heliaster's diet consisted of a wide variety of invertebrates with barnacles (Chthtmalus anisopoma) making up 93% of the prey. Bivalves and gastropod prey are the next most often taken prey (5%). These observations are similar to those of Paine (1966) except that Paine found barnacles to be less important in the diet (51%) and gastropods to be more important (40%). Paine proposed that Heliaster may act as a "keystone species" in this community by preying on a species that would otherwise monopolize resources and competitively exclude species. 2. A catastrophic decline of Hp.liaster populations at Station Beach and throughout the Gulf of California occurred in the. summer of 1978 and mortality approached 100% (Dungan et al 1982). Population density remained extremely low through the study period (1981-1984), and to the present time (1987). Dungan et'al (1982) suggest that the population crash was the result of elevated sea surface temperature which were related to a 1977-78 climatic event. The 70 71 subsequent years, 1981, 1983, and 1984 have all been \ 3. Pre-die-off and post-die-off densities of 20 species (11 prey and 9 non-prey) from the intertidal community at Station Beach· have been compared. All measurements were made in two distinct habitats: a boulder zone where Heliaster was formerly abundant, and a reef flat zone where Heliaster was formerly rare. Results indicate more frequent and severe density changes in the boulder zone rather than the reef flat and in Heliaster prey rather than non-prey species. In the boulder zone six prey species increased, three decreased and two were unchanged while one non-prey increased, three decreased and five were unchanged. In the reef flat zone three prey species increased while seven were unchanged and two non-prey decreased while seven were unchanged. Several species became extremely rare during the study. At least two of these, Othilia and Aplysia, are thought to have succumbed, like Heliaster, to some temperature-related cause. Chthamalus and Acanthina; prey species that presumably would respond most strongly to Heliaster removal, were virtually unaffected. Morula, a ..... _ ..- ..• ------_. 72 predator with a diet similar to Heliaster, has expanded from reef flat habitat use to utilize the boulder zone as well. 4. Species were grouped together in the following guilds: Prey and Non-prey, Carnivores, Herbivores, and Detritivores. Following the die-off there was a large increase in prey guild density as well as a shift in the order of numerical abundance within the guild, but these changes occurred only in the boulder zone. Carnivores and herbivores both showed substantial increases in density but only in the bOlllder zone. Carnivore rank order was stable throughout the study while herbivores exhibited a shake up in the abundance ranking between 1976 and 1981 in the boulder zone. Density and/or rank order changes occurred during the period (1976 1981) that included and immediately followed the disappearance of Heliaster. There were no significant changes from 1981 to ]984, suggesting the establishment of a new, stable arrangement in the absence of Heliaster. 5. There was no change in total density (all species combined) during the study period which suggests no dramatic change in the standing crop. Slight decreases in diveristy (no. species, H' and J) were due to loss of a few, originally rare species in the reef flat. 6. I conclude that Heliaster did not act as a keystone species in this community, because: a) the initial 73 conditions necessary for the existance of a keystone predator (the presence of a potential resource monopolizer held in check by the predation of a single top carnivore) did not exist at Station Beach, and b) a diverse carnivore guild with a high degree of dietary overlap with Heliaster compensated for the absence of Heliaster predation in controlling prey densities. 7. This community was in a stable dynamiC state before the Heliaster crash, it has reached a new dynamic state which appears to be stable, and I predict that it will return to the pre-die-off status with the return of Heliaster. APPENDIX A CAUSES OF THE Heliaster DIE-OFF The disappearance of Heliaster from the Gulf of California is not unprecedented. A number of recorded cases of echinoderm mass mortalities from various areas around the world have been recorded during the last ten years (Pearce and Hines 1979; Lessios et al. 1984; Sheibling 1984; Jackson and Kaufmann 1987). Often echinoderms are the only organisms affected in the community, and when other taxa are affected, echinoderms are usually the most severly affected group. On the Pacific coast of San Diego, California, 7 species of asteroids exhibited mass mortalities in 1978 and during the period from 1980 through 19$3, with a peak in 1983 possbly due to El Nino Southern Oscilation (Shroeter and Dixon 1984). In the Mediterranean Sea, three species of sea urchins showed disease symptoms in 1978 (Fenaux 1984). Accumulations of tests of these species were found along the European coasts of the western Mediterranea. Diseased urchins exhibited lesions on the tests with loss of epidermis and the disease spread to other individuals by contact. However, the most extensive known mass mortality of marine invertebrates occurred in Panama in 1983 where there was a die-off of the urchin, Diadema antillarum throughout 74 75 the Caribbean. Mortalities were reported to be 95-100% in certain localities (Lessios et al. 1984). Sheibling (1984) reports on the mortality of the sea urchin, Stongylocentrotus, in Novia Scotia from 1980 to 1983 where mortalities were 99% along 400 km of coastline. Most of these reports point out that such catastrophes have a potentially important impact on communities because of the ecological importance of echinoderms as predators or as important grazers (Jangoux 1984) but there have been few such studies of these impacts. Pearce and Hines (1979) is one such study on the expansion of kelp forests after sea urchin die off. Furthermore, many researchers have suggested that temperature was involved either directly or indirectly in these events. Maes and Jangoux (1984) have looked at diseased urchins from the Mediterranean, North sea, Atlantic and Pacific (California) oceans and found similar symptoms: loss of epidermis; formation of discolored lesions on the surface; spine loss; and pentration of the infection to the coelom. They have named this the "bald sea urchin" disease. In some cases amoeba have been found in the diseased tissue (Sheibling 1984). Although pollution is the only known stress to cause disease in inverte~rates (Jangoux 1984), stress may convert an infection to a full-blown disease (Johnson 1984). Also, because many viruses are not host-specific, what may be a low-level infection in one species may spread to another 76 species where it is disastrous (Johnson 1984). I propose the following explanation for the Heliaster die-off. Heliaster probably suffered from the bald sea urchin disease, based on the similarity of the symptoms reported by Dungan et al (1982) with those given by Maes and Jangoux (1984). Heliaster may have experienced a yearly cycle of parasite load with a host of microorganisms as normal flora within the tissues of the organism. In the summer when sea temperatures were high, the infection became more of a burden but never erupted into disease because the cold temperatures of winter slowed parasite growth rates. During the winter of 1977-1978 temperatures did not reach normal lows. Consequently as spring temperatures warmed up, parasite populations were already fairly abundant. By summer the infectious agents reached some threshold level and blossomed into the disease that we observed at Station Beach. High temperatures as such do not bring on the disease (summer 1978 was no warmer than other summers), but an extended period of elevated sea temperatures does. If this scenario is true, it may explain why Heliaster has not yet come back; temperatures have been predominantly warm since 1978. Both 1981 and 1983 were El Nino years and the 1981-1984 period is warmer than any similar length of time during the last 25 years at Puerto Penasco. There is a possiblilty that Heliaster will never return to its previous abundance at Station Beach. Since 77 1978 there has been no increase in abundance of Heliaster at Station Beach (see Table 10) or throughout the Gulf;Heliaster censuses on summer field courses in the Gulf of California indicate that Heliaster is still extremely rare. Heliaster's reproduction method,· spawning by separate sexes with fertilization taking place in the water column, make achieving a successful spawn difficult at such low density. Young Heliaster may have a difficult time getting reestablished in the community if their competitors have increased in their absence. Heliaster can spawn after one or two years, so if successful reproduction had occurred within three or even four years after the crash, increases in abundance should be evident by this time. ~!o such increases have been recorded. 78 Table 10. Density of Heliaster kubiniji between 1974 and 1984 at Station Beach Reef. For each date,the area sampled, number of He1iaster found, density (individuals/square meter) and sources for the data are 1isted.(Y=N.Yensen field notes; M=Maluf field notes; MB=Mackie and Boyer 1977; B= Boyer pers. obs.) DATE AREA SAMPLED # He1iaster DENSITY (square meters) found (ind/m2) SOURCE 2/8/74 100 17 0.17 Y 2/9/74 314 42 0.13 " 2/10/74 314 35 0.11 " 3/8/74 314 40 0.13 " 7/19/74 10 7 0.70 :vi 7/19/74 18 12 0.67 " 7/21/74 5 2 0.40 " 7/21/74 15 4 0.27 " 7/21/74 6 4 0.67 " 7/21/74 5 0 0.00 " 11/2/74 100 16 0.16 Y 11/17/74 100 28 0.28 " 12/7/74 25 12 0.48 II 12/14/74 100 49 0.49 " 2/11/75 100 39 0.39 II 2/11/75 100 26 0.26 II 6/14/75 10 13 1. 30 ~I 9/20/75 100 25 0.25 Y 10/14/75 100 29 0.29 II 1/24/76 9 3 0.33 NB 2/15/76 17 3 0.18 " 3/20/76 74 18 0.24 II 4/18/76 6 5 0.83 " ------Widespread mortalities in the summer of 1978. No data. ------7/25/79 150 0 0.00 B 9/8/79 150 a 0.00 " 9/21/79 150 0 0.00 " 10/5/79 150 2 0.01 " 10/19/79 150 3 0.02 " 11/3/79 150 0 0.00 " 11/18/79 150 3 0.02 " 1/19/80 150 1 . 0.01 " 2/15/80 150 6 0.04 " 3/15/80 150 5 0.03 " 10/25/80 150 1 0.01 " 2/6/81 27 0 0.00 " 79 Table 10.--Continued 3/6/81 24 a 0.00 B 4/3/81 28 a 0.00 " 2/26/82 28 a 0.00 " 3/26/82 24 a 0.00 " 4/24/82 28 a 0.00 " 1/27/83 27 2 0.07 " 2/25/83 27 2 0.07 " 3/24/83 28 1 0.04 " 1/22/84 25 0 0.00 " 2/19/84 24 0 0.00 " 3/15/84 27 1 0.04 " APPENDIX B BEFORE AND AFTER LIST OF INVERTEBRATES AT STATION BEACH The table presented here is a list of species of invertebrates at Station Beach Reef. Species commonly found before the Heliaster die-off are listed with the source for the citing. Species from the list that were found at Station Beach after the die-off are indicated by an asterisk. Total # species after/before is shown next to each taxon name. The sources given are: 1 = Brusca 1980; 2 Pickens 1965; MB Mackie and Boyer 1977; M = Mohr 1970; H = Hershey, 1969. SPECIES NAME Source MOLLUSCA:POLYPLACOPHORA (2/2) * Chiton virgulatus 1,2,MB,H * Stenoplax conspicua 1,2 MOLLUSCA:GASTROPODA (24/28) * Acanthina angelica 1,2,MB,M Acmaea discors 2,H * Aplysia californica 1,2,MB * Berthellina ilisima 1 * Cerithium maculosum 1,2 * Collisella atrata 1 * Collisella standfordiana 2,H * Columbella fuscata 1,2,MB,H * Crepidula striolata 2,M Crepidula uncata 2 * Cypraea annettae 1,2, M * Diodora inaegualis 2,MB,H Hexaplex erythrostomas 1,2 * Hipponix pilosus 2,MB,M 80 81 APPENDIX B--Continued. SPECIES NAME SOURCE * Liocerithium judithae MB * Littorina aspera 1,2,MB * Mitrella ocellata 2,MB,M,H * Morula ferruginosa 2,MB,H * Muricanthus nigritus 1,2,M * Neorapana tuberculata 2,MB,H * Nerita funiculata MB * Nerita scabricosta 1,2 Oliva ~ * Onchidella binneyi 1,2, H ,~ 1...:.. rugosa 1,2,MB * Tridachiella diomedia 2 * Turbo fluctuosus 1,2,MB,H * Vermicularia pellucida 2 MOLLUSCA:BIVALVIA (11/12) * Arca 2 * ~eeveana 1,2,H * B;achiodontes semilaevis 2 * Cardita affinis 1,2 * Isognomon janus 2 * Lima pacifica 1,2 * Lithophaga aristata 2,MB,M,H L. attenuata MB,M ,~ L. spatica 2,H * Modioiolus capax 2 * Ostrae conchaphila 2 * Pinna rugosa 2 MOLLUSCA:CEPHALOPODA (2/2) * Octopus digueti 1 ~'< ~ fitchi MB CRUSTACEA (25/26) :::C Ala cornuta 1,2 ~:c Alpheus ~ 2,MB ::Jc Balanus tintinnabulum 1,2 'Jc Chthamalus anisopoma 2,MB * Clibanarius digueti 1,2,MB Cyclograpsus escondidensis 1 * Epialtus minimus 1, MB * Eriphia sguamata 1,2,MB ,~ Geotice americana 1,2 * Leptodius occidental is 1 i.' Ligia occidentalis 1,2 ,~ Lysmata californica 1 ,~ Pachycheles setimanus 1,2 82 APPENDIX B--Continued. SPECIES NAME SOURCE * Paguristes anahuaces MB * Pagurus lepidus MB * Palemon ritteri 2,MB * Pilumnus gonzalensis 1 * P. limosus 1 ,.~ p:- townsendii 1,2,MB ,~ Petrolisthes armatus 1,2 * .t:.. gracilis 1,2 ,~ .t:.. hirtipes 1,2 * Pylopagurus roseus 2 * Tetraclita sguamosa 1,2,MB * Tetragrapsus iouyi 1 * Tylos punctatus 1 ECHINODERMATA:ECHINOIDEA (2/3) Arbacia incisa 2 * Echinometra vanbrunti 1,2 * Eucidaris thouarsii 1,2 ECHINODERMATA:OPHIUROIDEA (7/7) * Ophiactis savignyi 1,2 * Ophiocoma aethiops 1,2,MB ,~ O. alexandri 1,2,MB ~c OJ)hioderma panamense 1,MB * O. teres 1,2 * ophioneries annulata 1,2,MB * Ophiothrix spiculata 1,2 ECHINODERMATA:ASTEROIDEA (2/2) * Heliaster kubiniji 1,2,MB * Othilia tenuispina 1,2,MB ECHINODERMATA:HOLOTHUROIDEA (6/7) * Brandtothuria arenicola 1,2 * 1L.. impatiens 1,2 ,~ Chiridota .2..P...:.. MB * Isostichopus fuscus 1 * Neothyone gibbosa 1 ,2 Pentamera chierchia 1,2 * Selenkothuria lubrica 1,2,MB TOTAL ...... 81/89 (.91) 83 LITERATURE CITED Addicott,J.F. 1974. Predator and prey community structure: an experimental study of the effects of mosquito larvae on the protozoan communities of Pitcher Plants. Ecol. 55:475-492. Barnes,R.D. 1974. Invertebrate Zoology. 3rd ed. W.B. Saunders Co., Philadelphia. 807 pp. Brown, J.H., D.H. Davidson, and O.J. Reichman. 1979. An experimental study of competition between seed-eating desert rodents and ants. Amer. Zool. 19:1129-1143. Brusca, R.C. 1980. Common Intertidal Invertebrates of the Gulf of California. University of Arizona Press. Tucson, Arizona. 513 pp. Caso, M.E. 1961. Los Equinodermos de Mexico. Ph.D. Dissertation. Dept. BioI., Univ. Nac. Aut. Mexico. 388 pp. Caswell,H. 1978. Predator mediated coexistence: a non equilibrium model. Amer. Natur. 112:127-154. Connell,J.H. 1961b. Effects of competition, predation by Thais lapillus, and other factors on the populations of the barnacle Balanus balanoides. Ecol. Monogr. 31 (1): 61-104. Connell, J.H. 1971. On the role of natural enemies preventing competitive exclusion in some marine animals and in rain forest trees. In: Proc. Advan. Study Inst. "Dynamics of numbers in populations". Oosterbeck 1970. P.J. Boer and G. Gradwell eds. Center for agricultural Publishing and Documentation. Wageningen, The Netherlands. Connell,J.H. 1975. Some mechanisms producing structure in natural communities.In Ecology and evolution of communities. M.L. Cody and J.M. Diamond eds. Harvard Univ. Press. Cambridge, Mass. Connell,J.H. 1978. Diversity in tropical rain forests and coral reefs. Science. 199:1302-1310. 84 Day,R.W. 1977. Two contrasting effects of predation on species richness in coral reef habitats. Marine Biology, 44:1-5. Dayton,P.K. 1971. Competition, disturbance and community organization: the provision and subsequent utilization of space in a rocky intertidal community. Ecol. Monogr. 41:351-389. Dayton,P.K. 1985. The structure and regulation of some South American kelp communities. Ecol. Monogr. 55(4):447-468. Dayton,P.K. and R.K. Hessler. 1972. Role of biological disturbance in maintaining diversity in the deep sea. Deep-Sea Res. 19:199-208. Dayton,P.K., G.A. Robilliard, R.T.Paine, and Linnea B. Dayton. 1974. Biological accomodation in the benthic communtiy at McMurdo Sound, Antarctica. Ecol. Monogr. 44(1):105-128. Duggins,David O. 1983. Starfish predation and the creation of mosaic patterns in a kelp-dominated community. Ecol. 64(6):1610-1619. Dungan, M.L. 1984. Experimental analysis of processes underlying the structure of rocky intertidal community in the northern Gulf of California. Ph.D. diss. University of Arizona, Tucson, Arizona. Dungan, M.L. 1986. Three-way interactions:barnacles, limpets, and algae in a Sonoran desert rocky intertidal zone. Amer. Natur. ]27(3):292-316. Dungan, M.L., T.E. Miller, and D.A. Thomson. 1982. Catastrophic decline of a top carnivore in the Gulf of California rocky intertidal zone. Science. 216:989-991. Estes,J.A., and J.F. Palmisano. 1974. Sea otters: their role in structuring near-shore communities. Science. 185:1058-1060. Fenaux, L. 1984. Mass mortality of echinoids in the Mediterranean Sea. The Echinoderm Newsletter. 1984. Garrity,Stephen C. and Sally C. Levings. 1981. A predator-prey interaction between two physically and biologically constrained tropical rocky shore gastropods: direct and indirect and community effects. Ecol. Monogr. 51(3):267-286. 85 Glasser,J.W. 1978. The effect of predation on prey resource utilization. Ecol. 59(4):724-732. Glasser,J.W. 1979. The role of predation in shaping and maintaining the structure of communities. Amer. Natur. 113:631-641. Harper, J.L. 1969. The role of predation in vegetational diversity. In: Brookhaven Symp. BioI. 22:48-62 Hershey, J.R. 1969. Molluscs as indicators of Rocky Intertidal Zonation at Puerto Penasco, Gulf of California. Biological Studies in the Gulf of California. Vol. 7(1). University of Arizona. Inouye,R.S., G.S.Byers, and J.H.Brown. 1980. Effects of predation and competition on survivorship, fecundity, and community structure of desert annuals. Ecol. 61(6):1344-1351. Jackson, J.B.C. and K.W. Kaufmann. 1987. Diadema antillarum was not a keystone predator in cryptic reef environments. Science 235:687-689 Jangoux, M. 1984. Diseases of Echinoderms. Helgolander meersuntersuchungen. 37:207-216. Johnson, P.T. 1984. Viral diseases of marine invertebrates. Helgolander meersuntersuchungen. 37:65-98 Kotler, B.P. 1984. Risk of predation and the structure of desert rodent communities. Ecol. 65(3):689-701. Lessios,H.A. D.R. Robertson, and J.D.Cubit. 1984. Spread of Diadema mass mortality through the Caribbean. Science 226:336-337. Lewis,Sarah M. 1986. The role of herbivorous fishes in the organization of a Caribbean reef community. Ecological Monographs 56(3):183-200. Lively, C.M. 1984. Competition, predation, and the maintainance of dimorphism in an acorn barnacle (Chthamalus anisompoma) population. Ph.D. diss. University of Arizona, Tucson. Lively, Curtis M. and Peter Raimondi. In prep. Seasonality, predation, and mussel/barnacle interactions in the northern Gulf of California. 86 Lubchenco, J. 1978. Plant species diversity in a marine rocky intertidal community: importance of herbivore food preference and algal competitive abilities. Amer. Natur. 112:23-39. Mackie,S.A., and E.H.Boyer. 1977. Intertidal zonation of macroscopic invertebrates on the coquina reef at Playa de Estacion, Puerto Penasco, Sonora, Mexico. Bios. 148:120-128. Madsen, F.J. 1956. Reports of the Lund University Chile Expedition 1948-1949. 24. Asteroidea, with a survey of the Asteroidea of the Chilean Shelf. Acta/Univ. Lund Andra. Afd. N.S. 52(2):1-53, 6pls. Maes, P. and M. Jangoux. 1984. The bald sea urchin disease: a biopathological approach. Helgolander meersuntersuchungen. 37:217-224. Maluf, L.Y. 1987. Classification and distribution of Eastern Pacific Echinoderms. Ph. D. dissertation. University of Arizona, Tucson. 350 pp. Malusa, J.R. 1986. Life history and environment in two species of intertidal Larnacles. BioI. Bull. 170:409-423. Menge,B.A., and J.P.Sutherland. 1976. Species diversity gradients: synthesis of the roles of predation, competition, and temporal heterogeneity. Amer. Natur. 110:351-369. Mohr, P.B. 1970. Intertidal comparison of Station Beach and La Jolla Cove. Unpublished manuscript. Morin,Peter J. 1983. Predation, competition and the composition of larval anuran guilds. Ecological Monographs 53(2):119-138. Paine,R.T. 1966. Food web complexity and species diversity. Amer. Natur. 100:65-75. Paine,R.T. 1969a. A note on trophic complexity and community stability. Amer Natur. 103:91-93. Paine,R.T. 1971. A short-term experimental investigation of resource partitioning in a New Zealand rocky intertidal habitat. Ecol. 52(6):1096-1106. 87 Pearce, J.S. and A.H. Hines. 1979. Expansion of central California kelp forest following the mass mortality of sea urchins. Mar. BioI. 51:83-91. Pianka, E. R. 1966. La ti tudinal gradients in species diversity: a review of concepts. Amer. Natur. 100(910):33-46. Pickens, P.E. 1965. The Rocky Intertidal Zones at Puerto Penasco. Biological Studies in the Gulf of California. Vol 3. University of Arizona. Schroeter, S. and J. Dixon. 1984. Echinoderm disease in north San Diego County, California. The Echinoderm Newsletter. 1984. Scheibling, R. 1984. Mass mortalities of echinoids (Strongylocentrotus droebachiensis) off Nova Scotia, Canada. The Echinoderm Newsletter. 1984. Sousa, W.P. 1979. Disturbance in marine intertidal boulder fields: The nonequilibrium maintainance of species diversity. Ecol. 60(6):1225-1239. Steinbeck, J.E. and E.F. Ricketts. 1941. Sea of Cortez. A leisurely journal of travel and research. Viking Press, New York. 596 pp. Summarco, P.W. ]980. Diadema and its relationship to coral spat mortality: grazing, competition, and biological disturbance. J. Exp. Mar. BioI. Ecol. 45:245-258. Summarco, P.W. 1982. Echinoid grazing as a structuring force in coral communities: whole reef manipulations. J. Exp. Mar. BioI. Ecol. 61:31-42. Sutherland, J. 1974. Multiple statble points in natural communities. Amer. Natur.108:859-873. Thomson, D.A. and C.E. Lehner. 1976. Resilience of a rocky intertidal fish community in a physically stable environment. J. Exp. Mar. BioI. Ecol. 22:1-29. Turk, M.J. 1981. Intertidal migration and formation of breeding clusters of labial-spine morphs of the Thaid gastropod, Acanthina angelica. M.S. thesis. University of Arizona, Tucson. 66 pp. 88 Underwood, A.J. and E.J. Denley. 1984. Paradigms, explanations and generalizations in models for the structure of intertidal communities on rocky shores. In: Ecological Communities; Conceptual issues and the evidence. Strong, D.R., D.S. Simberloff, L.G. Abele and A.B. Thistle editors. 1984. Princeton University Press, Princeton, N.J. 614 pp. Wiltse,W.I. 1980. Effects of Polinices duplicatus (Gastropoda:Naticidae) on infaunal community structure at Barnstable HRrbor, Mass., USA. Marine Biology. 56:301-310. Xantus, John. 1860. Description of 3 new speciesof starfishes from Cape San Lucas. Proc. Acad. Nat. Sci. Philad.1860:568. Yensen, N.P. 1979. Function of the labial spine and effect of prey size on swithcing "polymorphism" of Acanthina angelica. Ph.D. Dissertation. University of Arizona, Tucson. 62 pp. Zar,J.H. 1974. Biostatistical Analysis. Prentice-Hall Inc. Englewood Cliffs, N.J. 620 pp.