THE FUNCTION OF THE LABIAL SPINE AND THE EFFECT OF PREY SIZE ON "SWITCHING" POLYMORPHS OF ACANTHINA ANGELICA (GASTROPODA: THAIDIDAE)
Item Type text; Dissertation-Reproduction (electronic)
Authors Yensen, Nicholas Patrick
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 02/10/2021 13:50:02
Link to Item http://hdl.handle.net/10150/298486 INFORMATION TO USERS
This was produced from a copy of a document sent to us for microfilming. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the material submitted.
The following explanation of techniques is provided to help you understand markings or notations which may appear on this reproduction.
1. The sign or "target" for pages apparently lacking from the document photographed is "Missing Page(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure you of complete continuity.
2. When an image on the film is obliterated with a round black mark it is an indication that the Him inspector noticed either blurred copy because of movement during exposure, or duplicate copy. Unless we meant to delete copyrighted materials that should not have been filmed, you will find a good image of the page in the adjacent frame.
3. When a map, drawing or chart, etc., is part of the material being photo graphed the photographer has followed a definite method in "sectioning" the material. It is customary to begin filming at the upper left hand corner of a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again—beginning below the first row and continuing on until complete.
4. For any illustrations that cannot be reproduced satisfactorily by xerography, photographic prints can be purchased at additional cost and tipped into your xerographic copy. Requests can be made to our Dissertations Customer Services Department.
5. Some pages in any document may have indistinct print. In all cases we have filmed the best available copy.
University Microfilms International 300 N. ZEEB ROAD. ANN ARBOR. Ml 48106 18 BEDFORD ROW, LONDON WC1R 4EJ, ENGLAND 7912542
YENSEN* NICHOLAS PATRICK THE FUNCTION OF THE LABIAL SPINE AND THE EFFECT OF PREY SIZE ON "SWITCHING" POLYMORPHS OF ACANTHINA ANGELICA (GASTROPODA: THAIDIDAE).
THE UNIVERSITY OF ARIZONA* PH.D.* 1979
University Microfilms International 300n zeebroad, ann arbor.mi woe
© 1373
NICHOLAS PATRICK YENSEN
ALL RIGHTS RESERVED PLEASE NOTE: In all cases this material has been filmed 1n the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark .
1. Glossy photographs
2. Colored illustrations
3. Photographs with dark background
4. Illustrations are poor copy
5. Print shows through as there 1s text on both sides of page
6. Indistinp^,stinp^, broken or small print on several pages throughout ^ 7. Tightly bound copy with print lost in spine
8. Computer printout pages with indistinct print
9. Page(s) lacking when material received, and not available from school or author
10. Page(s) seem to be missing in numbering only as text follows
11. Poor carbon copy
12. Not original copy, several pages with blurred type
13. Appendix pages are poor copy
14. Original copy with light type
15. Curling and wrinkled pages
16. Other
University Microfilms International 300 N ZEEB RD.. ANN ARBOR. Ml 48106 >313) 761-4700 THE FUNCTION OF THE LABIAL SPINE AND THE EFFECT OF PREY SIZE
ON "SWITCHING" POLYMORPHS OF ACANTHINA ANGELICA (GASTROPODA:
THAIDIDAE)
by
Nicholas Patrick Yensen
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
19 7 9
Copyright 1979 Nicholas Patrick Yensen THE UNIVERSITY OF ARIZONA
GRADUATE COLLEGE
I hereby recommend that this dissertation prepared under my direction by Nicholas Patrick Yensen entitled The Function of the Labial Spine and the Effect of Prey Size on "Switching" Polymorphs of flcanthina angelica (Gastropoda: Thaididae) be accepted as fulfilling the dissertation requirement for the degree of Doctor of Philosophy
AJ3i/. 7 As members of the Final Examination Committee, we certify that we have read this dissertation and agree that it may be presented for final defense. X r Wu'. l4*dk. 3- 0 AJc~o~'. ///20 7g > VWv Q-C zO Final approval and acceptance of this dissertation is contingent on the candidate's adequate performance and defense thereof at the final oral examination. 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 allow able without special permission, provided that accurate acknowledgement of source is made. Requests for permission for extended quotation from or reproduction of this manu script in whole or in part may be granted by the copyright holder. To Mathew Oliver Xantus A friend, poet, and artist. iii PREFACE The following dissertation conceptually represents more than I had originally intended. It does, however, represent a synthesis of an information sequence provided by my esteemed colleagues and environment. And, as such, the mathematical conception appeared quite fortuitously, yet required two humbling days to realize on paper. After which its simplicity suggested it to be well known and/or trivial. The former abated leaving my thoughts to the latter. Two months elapsed and I included it here. In doing so, I have clearly marked my path. I gratefully acknowledge the members of my commit tee, Drs. D. A. Thomson, R. W. Hoshaw, J. H. Brown, E. A. Stull, S. M. Russell and C. T. Mason for their patience, courtesy and helpful advice. Dr. D. A. Thomson, my disser tation director, deserves special acknowledgment for his continual backing and support in my endeavors. Dr. R. W. Hoshaw has generously given me access to his laboratory facilities and has helped in numerous ways. R. M. McCourt and J. Hoffman have helped with field work and are greatefully appreciated for this. R. Abugov, G. Byers, E. H. Boyer, C. Brand, R. C. Brusca, M. L. Dungan, L. T. Findley, K. G. Gage, M. R. Gilligan, R. S. Houston, iv V S. Kessler, S. A. Mackie, J. N. Norris, L. Y. Maluf, P. Pepe, E. K. Snyder, J. Eads, P. J. Turk, J. and N. Wilt, and A. E. Yensen have provided assistance without which this study would not have been completed. C. N. Hodges, D. Moore, J. Ure, and 0. Villavicencio have been most helpful in allowing me use of the facilities at Unidad Experimental, Puerto Penasco, Sonora, Mexico. Dr. G. Cubas, Curator of Malacology, has graciously allowed me use of the Malacology Collection at the Universidad Nacional Autonomia de Mexico, Mexico City, D. F., Mexico. Dr. R. S. Felger, C. A. Stigers, R. C. Wilkinson and B. L. Tapper of the Arizona-Sonora Desert Museum have been very generous with their facilities and computer terminal. I have particularly benefited from discussions with Drs. J. H. Connell, J. R. Hendrickson, E. C. Pielou, W. B. Miller, P. E. Pickens, and A. E. Yensen and from graduate students R. Abugov, M. L. Dungan, C. A. Flanagan, C. E. Lehner, P. McKie, J. Short, and R. C. Wilkinson. I would like to thank D. A. Thomson, Ed Gage, and R. Abugov, who have read earlier drafts of the manuscript in its entirety. Any errors, however, are mine. TABLE OF CONTENTS Page LIST OF ILLUSTRATIONS vii LIST OF TABLES viii ABSTRACT ix INTRODUCTION 1 MATERIALS AND METHODS 4 The Study Predator 4 The Field Study Sites 7 Laboratory Studies 13 RESULTS 16 Observations on the Function of the Labial Spine . . 16 The Effect of Diet on Spine Size: The Diet- regulator Hypothesis 23 An Optimal Foraging Model with a Barrier Term ... 41 DISCUSSION 46 The Labial Spine 46 Foraging Strategy 47 The Bioenergetic Barrier, the Model and Disruptive Switching 50 CONCLUSIONS 54 APPENDIX A: DICE-LERAAS GRAPH OF 30 ACANTHINA ANGELICA SHELL LENGTH AND SPINE LENGTH MEASUREMENT DIFFERENCES (mm) BETWEEN REPEATED CALIPER-RULER, CALIPER-CALIPER, AND RULER-RULER MEASUREMENTS ... 56 APPENDIX B: LIST OF ACANTHINA ANGELICA SPINE AND SHELL LENGTHS AND DIMENTIONS OF PREY BARNACLES, TETRACLITA STACTALIFEM, IN FORAGING RATE EXPERIMENT #1 58 LITERATURE CITED 59 vi LIST OF ILLUSTRATIONS Figure Page 1. Map of Acanthina distributions in the Gulf of Cal if or nia 5 2. Map of study sites, northern Gulf of California . . 8 3. Map of study sites near Puerto Penasco, Mexico . . 10 4. Map of primary study site, Playa Estacio'n (Station Beach) 12 5. Scatter graph of 143 Acanthina angelica spine lengths (mm) on shell lengths (mm) ... 25 6. Scatter graph of 332 Acanthina angelica spine lengths (mm) on shell length (mm) ... 26 7. Dice-Leraas graph of Acanthina angelica spine lengths (mm) according to substrates . . 27 8. Graph of mean labial spine lengths (mm) of Acanthina angelica on mean barnacle heights (mm) 29 9. Regrowth of broken long spines 30 10. Spine reduction in snails fed small barnacles . . 32 11. Spine elongation in snails fed large barnacles . . 33 12. Effects of diet on short-spined snails 35 13. Effects of diet on long-spined snails 36 14. Weight gains in short-spined snails 39 15. Weight gains in long-spined snails 40 vii LIST OF TABLES Table Page 1. Consumption in Foraging Rate Experiment #1 . . . . 19 2. Consumption in Foraging Rate Experiment #2 . . . . 21 3. Apertural lip growth and change in spine length of 40 Acanthina angelica 38 viii ABSTRACT The predatory snail Acanthina angelica uses its labial spine to open barnacles by a mechanism called spin- ing. This conclusion is based upon observations that feeding A. angelica: 1) lunge, 2) leave scratch marks on the barnacle's opercula, 3) orient themselves with the spine nearest to a natural separation of the barnacle's opercula, 4) may consume barnacles without drilling, 5) consume barnacles significantly faster when they don't drill, and 6) drill when the spine is too short to reach the opercula. Field observations demonstrated that A_. angelica is poly morphic for spine size with short-spined individuals occurring among small barnacle prey (Chthamalus) and long- spined individuals occurring among large barnacle prey (Tetraclita). The spine size of these two morphs was experimental ly shown to be regulated by barnacle prey size. It was further shown that the snails could "switch" to the other morph. That is, large-spined Acanthina fed small barnacles developed significantly shorter spines than controls and short-spined Acanthina fed large barnacles developed sig nificantly longer spines than controls. Such switching ix X experimental groups gained significantly less weight than controls maintained on their normal prey. A field test of this diet-regulatory mechanism showed a high correlation of spine size to barnacle size. Data from diet-regulator and foraging experiments were used to derive a theoretical "barrier term" based on the cost to switch from one spine morph to the other. It is mathematically demonstrated that this barrier to switching may remain even though there may be a greater frequency in consumption of a "suboptimal" resource, i.e., S > f (I ba i b I ) - f (J - J ) when f > f but that (I - I ) << (J - ajab ij ba a J ) where f and f are the frequency of consumption of b i j resources I and J, respectively, I and J are the costs to a b consume I and J, respectively, using inefficient mechanisms (improper spine length); while I and J are the costs to a b consume I and J, respectively, using efficient mechanisms (proper spine length). This mathematical model may be used to describe the prerequisites for progressive, temporal and disruptive switching. Within the context of this model it is suggested that A. angelica meets the requirements for disruptive switching and that it may be used to explain the occurrence of the spine polymorphism. And finally, it is suggested that the barrier term and the associated inequality are general phenomena basic to evolutionary theory and describe the process of increasing niche speci ficity in the reduction of competition in evolving systems INTRODUCTION Gastropod predatory mechanisms have received some attention in the literature and are reasonably well known (e.g., drilling — Carriker 1955, 1961; harpooning — Kohn 1959; wedging — Wells 1959, Paine 1962). However, the function of the conspicuous labial spine, characteristic of the ubiquitous neogastropod families, Thaididae, Muricidae, and Fasciolariidae, is not well understood. The literature concerning the function of the labial spine is contradictory. MacGinitie and MacGinitie (1949) were the first to suggest that the labial spine was used directly in food gathering. They reiterated their claim in 1968 (p. 371): "On the outer lip of certain snails there is a tooth, or spine, that has a very definite use. Snails possessing such a tooth live on barnacles, small mussels, etc. as exemplified by Acanthina spirata. This snail crawls over a barnacle, inserts its tooth between the two parts of the operculum and while the two piece door is thus wedged apart, inserts its proboscis into the barnacle and eats it. Such snails as Acanthina and Ceratostoma can use the tooth as a wedge on small mussels and nestling clams." Paine (1966a) rejected this observation. Based on some careful observations of spine positions in feeding Acanthina angelica, he concluded that the spine's function is to 1 2 anchor the snail to the substrate while drilling. Jane Menge (1974, p. 313), studying the relatively short-spined A. punctulata, agreed with Paine and further suggested that the spine, "...serves as a point of contact with the substratum and thus stabilizes the predator while it is drilling and eating prey that lift the anterior portion of the shell off the substratum." I have observed the use of the labial spine in sev eral populations of A. angelica in the vicinity of Puerto Penasco, Sonora, Mexico, where the major prey items are the barnacles Tetraclita stactalifem and Chthamalus fissus. My observations show that although the inferences of previous investigators may be contradictory, their observations are compatible. The first aspect of my study attempts to clarify and quantify the labial spine's function in A. angelica. The second aspect considers the effect of switch ing to prey of different sizes on the spine length of the predator. Numerous studies have been done of the effect of predator switching on the prey and are primarily concerned with population regulation and stability (Holling 1965, Murdoch and Oaten 1975, Reed 1969, Landenberger 1968, - Murdoch 1969, Manly, Miller and Cook 1972, Lawton, Beddington and Bonser 1974, to name a few). The effect of switching on the predator has not been so well studied and those studies that have been done all examine predator 3 behavior patterns and how they reflect on prey population regulation and stability (Tinbergen I960, Royama 1970, Fox 1970). I know of no studies of switching predators asso ciated with freely switching morphologies and their effects on the foraging behavior of the predator. The present study documents a case in which prey specialization limits the capabilities of the predator by causing associated morpho- loqical changes and making switching difficult and costly. The study specifically considers a limitation on optimal foraging theory, as described by Schoener (1971), and how the theory may be altered to include a barrier term that would consider interference caused by a morphological bar rier to feeding. The morphological barrier in A. angelica is its spine. I hypothesize that the labial spine size is influ enced by prey size and that if A. angelica switches to preying on large barnacles it is inefficient to switch back to small barnacles. Thus, spine size would act as a barrier to switching. And, in contrast to optimal foraging theory, a model is proposed which suggests that it can be ineffi cient for predators to switch, regardless of prey density. An inequality is derived based on actual switching data. MATERIALS AND METHODS The Study Predator The spine bearing carnivorous snail, Acanthina angelica I. Oldroyd 1918 (Thaididae) is endemic to the Gulf of California (Fig. 1). It ranges south on the mainland side of the Gulf at least to Guaymas, Sonora. At Mazatlan, near the mouth of the Gulf, it is "replaced" by the smaller A. brevidentata (Wood 1828) which extends south to Paita Peru (Keen 1971). On the peninsular side of the Gulf of California A. angelica seems to disappear south of the midriff islands. At the tip of the Baja California peninsu la a species similar in appearance to A. angelica, A. tyrianthina Berry 1957 occurs. Acanthina tyrianthina ranqes northward on the outer coast of Baja California to Bahia Magdalena. Here it is replaced by A. lugubris (Sowerby 1822) which extends north into southern California. At Playa Estacion (Station Beach), Puerto Penasco, Sonora, Mexico, A. angelica is abundant and I have meas ured densities of 1500 per square meter during "mating" aggregations. At other times it may disperse and occur at much lower densities, averaging 4.25 per square meter. The greatest densities were observed intertidally at a height of + 1.0 meters but it typically ranges from 0.0 to + 2.3 m. 4 5 m »« 1fO v •> "30, 110 Fig. 1. Map of Acanthina distributions in the Gulf of California. — The map shows the allopatric distributions of four eas.tern Pacific species of Acanthina: A_. lugubris (Sowerby 1822), A. tvrianthina Berry 1957, A^. angelica T. Oldroyd 1^18, and A. brevidentata (Wood 1828). 6 This corresponds well with the lower and upper limits of the barnacles Chthamalus fissus and Tetraclita stactalifem. Although the smaller Chthamalus may extend up into the zone of the larger Tetraclita, Tetraclita are rarely found below + 1.0 m. High densities of Acanthina are primarily a result of breeding clusters that occur from December to March (Wolfson 1970). Prom anatomical studies (Houston 1976) it may be inferred that A_. angelica does not have external fertilization. It may also be inferred that copulation occurs during the egg capsule laying aggregations since both males and females occur in such clusters and that once a cluster has begun to form, all the individuals in the area will move to the cluster. Egg capsules, approximately 5-mm hiqh and 2-mm wide, are attached in dense mats to under sides of rocks. Up to 3000 capsules have been observed attached to one boulder measuring 40 cm in length. Each capsule may contain from 200 to 500 eggs which may reach a diameter of 250/'m (Wolfson 1970, Houston 1976). I observed that in two to three weeks 10-15 fully developed juvenile snails 1-2 mm in length emerged from a capsule and did not pass through a planktonic phase. It is inferred by the size of these juveniles and the absence of the other 185 to 490 eggs that emerging juveniles consume their potential siblings in a bout of prenatal cannibalism Hemingway (1976) has observed juvenile siblings of A. 7 spirata drill and eat each other. My observations show that in the first few weeks the juveniles, each with a rudi mentary labial spine, rapidly grow to about 10 mm in length. Growth is inferred to become progressively slower such that a shell length of 20 mm is not reached until four to five months. Adult size (approximately 30 mm) is attained in about eight to ten months (about September to November), since according to Houston (1976) individuals are capable of reproducing at this size. Acanthina angelica is believed to live at least one year'although it may live much longer. Lifespans, survivorships, and aging techniques are still poorly known for gastropods in general. In the areas studied A. angelica was observed to eat only barnacles thus supporting Paine's (1966a) observa tion that A. angelica is a barnacle specialist. I have, however, kept 14 individuals in an aquarium for four months with no food other than algae. The Field Study Sites The majority of the studies were conducted in the vicinity of Puerto Penasco, Sonora, Mexico along the eastern coast of the Gulf of California and approximately 62 miles south of the Arizona-Sonora border and Sonoita, Sonora (Fig. 2). Snails were collected and/or studied from seven sites in the vicinity of Puerto Penasco (Rocky Point), Playa Estacion (Station Beach), Playa de Las Conchas, and Estero 8 CALIFORNIA 100 km roads y Puerto u ; PtiiaseaV Co bo I Lobos GOL FO Puerto Liberiod} Dosomboqurs. , do los SorislA . DE f\\ Kinov Guaymai B AJA CALIFORNIA C A LIFORNIA SUR Fig. 2. Map of study sites, northern Gulf of California 9 Morua (Joe's Estuary)(Fig. 3). Additional collections were studied from Cabo Lobos, Puerto Lobos, Puerto Libertad, and Desemboque de los Seris (Desemboque de San Ignacio), Sonora; Bahia de San Luis Gonzaga, Baja California del Norte; and Salina Cruz, Oaxaca. The Bahia Cholla site is located on the south side of the westward facing Bahia Cholla and northeast of Punta Pelicano. The substrate is composed primarily of granitic boulders although some beachrock is present. Both Tetraclita and Chthamalus are present although Tetraclita is found higher here. Balanus sp. is common in sandy areas. The area is protected from the open Gulf but receives silty and potentially polluted water. Punta Pelicano is an exposed granitic headland showing good zonation along its fairly steep shore. Again Tetraclita occurs higher in the intertidal region than Chthamalus. The zonation, however, is clearly more dis tinct. The water is reasonably free of sediment except during ebb tide when water from adjacent Bahia Cholla flows past the point. The Playa Norse site lies east of Punta Pelicano and is also composed primarily of granite. The slope is gentler and the zonation is not quite as evident. The granite outcrops are strewn with craggy boulders and are exposed to the open Gulf. They lie adjacent to a beachrock-formed lagoon and a sandy beach. Both Tetraclita and Chthamlus are reasonably abundant. 10 Bohio Chollo Playa Mors* Puerto Penasco Punta Penasco Playa^=2^> Moriia E,,aci6n Playaae' Las Conchas G O LF O DE CALIFORNIA 5 mi 10 k roads Fig. 3. Map of study sites near Puerto Penasco, Mexico. 11 The Playa Estacion intertidal zone (Fig. 4) con sists of a shell fragment beach running due east from Punta Penasco. The beach consists of coarse shell sand from the high tide level (approximately + 5.5 m) to about the + 2.3 m tide level (near mean high water — MHW) where Tetraclita covered shell-hash beachrock terraces jut out from under the sand. Below these terraces Chthamalus- covered basalt boulders clutter a beachrock platform which gradually slopes for 30-40 m into large tidepools and drainage channels located at about + 0.3 to + 0.6 m (near mean low water — MLW). Beyond the tidepools the beachrock forms a second terrace which then drops to a sand floor (at - 0.6 to - 1.2 m, low water — LW) that extends southward out into the Gulf. During and after storms the barnacle areas may be completely inundated by sand. The Playa de Las Conchas site is the littoral east ward extension of Playa Estacion. It is considerably flatter and has fewer and smaller basalt boulders. It does, however, have numerous loose chunks of beachrock and like the basalt they may be covered with Chthamalus. Tetraclita is rare and Balanus is fairly common. The area is dominated by shifting sand which is constantly covering and uncovering the beachrock platform. And again like Playa Estacion it is fully exposed to the open Gulf. The El Desemboque de los Seris site is located approximately 3 km north of the Seri indian village. 12 100 coastal vcgilgtion r^'j «*«;v • • a . • a • ,# «and" 'o> ' • ° IFORNIA Fig. 4. Map of primary study site, Playa Estacion (Station Beach). — The area illustrated is the marine preserve and is located 3 km east of Puerto Penasco, Sonora, Mexico. A) Casa Garcia. B) Unidad Experimental and marine station complex. C) Upper beachrock terrace and primary habitat for Tetraclita. D) Beachrock ridges formed by north- south fractured beachrock. E) Station Platform, an elevated beachrock platform. F) Station Pool Terrace, a lower beachrock terrace covered with an algal turf. G) Station Pool. H) Long Pool. I) Lower beachrock terrace. J) Basalt boulder covered beachrock platform. 13 Again there is an upper sand beach which gives way to a very gently sloping cobble composed primarily of granite. The cobbles were only moderately covered with Chthamalus and Tetraclita was rare. The site is quite sandy and may re ceive some sediment from the very small estuary 200 meters to the north. It is exposed to the open Gulf but may receive some protection from Isla Tiburon to the south. The Puerto Libertad site is north of a granite head land making it a semi-protected, steep intertidal region. The zonation is similar to the Punta Pelicano site although it is slightly compressed due to the smaller tidal amplitude. Tetraclita were abundant yet few Acanthina were seen in this higher zone. The Bahia de San Luis Gonzaga and the Salina Cruz, Oaxaca sites were not visited. Specimens were studied in the Malacological Collection of the Universidad Nacional Autonomia de Mexico, Mexico City, D. F. Laboratory Studies Laboratory studies and experiments were conducted at the Unidad Experimental (a marine station facility operated cooperatively by The University of Sonora and The University of Arizona and is located on Playa Estacion (Station Beach) approximately three km southeast of Puerto Penasco, Sonora, Mexico. 14 The feeding efficiency experiments were conducted in plastic tubs with a nylon mesh cover secured with nylon cord for easy inspection. Diet-regulator experiment #1 was con ducted in two 40-liter plywood aquaria painted with epoxy. Diet-regulator experiment #2 was conducted in six 16-liter covered nalgene buckets. All experiments had running sea water (21 to 27 degrees centigrade) from a sand filtered well. Additional observations were made in a 40-liter glass aquarium at The University of Arizona. Barnacles for the laboratory experiments were col lected primarily from Playa Estacion although a few were also collected from Punta Pelicano. Pieces of Chthamalus covered beachrock were taken from the Las Conchas site, comparable pieces of Tetraclita covered beachrock substrate were broken loose from the Playa Estacion reef using a claw hammer. All extraneous organisms (mussels, limpets, rock oysters, etc.) were removed form the beachrock substrate. Snails were marked by filing a flat place on the shell and numbered with india ink using a Rapidograph pen. The number was then coated with clear finger nail polish. Each snail was numbered twice as a safety precaution against loss of an identifying number. After two to four months some shells were renumbered to maintain two intact numbers. Labial spines were originally measured using a red plastic ruler (Sterling tradename) calibrated in millimeters. Measurements were taken by placing the ruler posterior to the spine along the labial margin. Measurements were estimated to the nearest 0.1 mm. Shell lengths were taken by placing the shell above the ruler and observing from a distance of 30 to 40 cm in order to maintain a constant parallactic er ror (+ 1.11 mm, see Appendix A). Measurements were made from the tip of the spire to the end of the siphonal canal. Actual lengths of eroded shells are given. No attempt was made to estimate the pre-eroded lengths since bad eroded shells were rare and not used. Measurements made after November 1977 were made using a hard plastic dial caliper calibrated in 0.1 mm. An analysis of the error of the ruler-made measurements, in particularly the spine measurements estimated to the nearest 0.1 mm, showed sur prising repeatability in comparison to the dial caliper measurements (Appendix A). The measurements by the differ ent methods are significantly different, however, necessitating a correction factor to be used when comparing measurements by ruler to those by caliper (shell length — 1.11 mm, spine length — 0.31 mm; see Appendix A). All snail weights were taken damp-dry, using top- loading Mettler balances (P165 and P1000). The weights were recorded to the "nearest 100 mg. RESULTS Observations on the Function of the Labial Spine Paine (1966a, p. 18) stated, "...although Acanthina was observed during all stages of the feeding process, no function was noted for the apertural spine." To support this he presented a table of 52 observations of long-spined Acanthina angelica that were feeding on Tetraclita stactalifem. He found that 34 (67%) had their spines out side of the barnacle's opercular opening and that 30 of these were also drilled on the outside. Of the remaining 18 that had their spines inside the opercular opening only 10 of these were also drilled on the inside. In some similar field observations on 19 A. angelica that were in a "feeding" position over the large barnacle Tetraclita, in which I also noted the condition of prey, I found that none had bored into any of the barnacles, yet 63% of the barnacles were lax and unresistant to prodding, part ly consumed or consumed. Three of the barnacles did have "scratch" marks on the opercula. I found only 37% of the Acanthina with their spines outside the barnacle's opercular opening in contrast to Paine's 67%. I also noted that of those with their spines outside, 71% were partly or com pletely consumed, and that of those with their spines 16 inside, only 17% were partly or completely consumed. This suggests that penetrated barnacles may be easier to feed upon if the spine is outside; and further that having the spine inside may be advantageous if the barnacle has not been penetrated. Although the data only suggest a spine function, it is clear that the barnacles were being consumed without being bored. It would be difficult to observe A^ angelica using its spine as a wedge as suggested by MacGinitie and MacGinitie (1968, p. 37) since the foot, man tle and shell obscure the spine when it is inside the opercular opening." However, snails with their spines inside the opercular opening tend to orient themselves so that the spine would be at the posterior end of the operculum where the plates seal imperfectly. Snails so oriented thrust their shells forward and downward, further suggesting that the spine may be used as a wedge. Such observations and those of Paine (1966a) began to shed light on the complex mechanisms involved in Acanthina foraging. Paine perceptively suggested that A. angelica at Puerto Penasco might be polymorphic for spine length. In addition he correctly suggested that the long- spined Acanthina feeds on the large Tetraclita while the short-spined morph feeds on the small Chthamalus. He then took some short-spined snails back to his laboratory where he fed them the Pacific coast Balanus glandula, a barnacle unfamiliar to _A. angelica. All of these short-spined snails bored their prey. Paine's observations suggested a short-term foraging rate experiment (Foraging Rate Experiment #1) in which I fed 15 long-spined and 15 short-spined, starved Acanthina 14 large barnacles (Tetraclita) to each group of 15 snails to see if there was any advantage in having a long spine since spines of many short-spined individuals were not long enough to reach the barnacles' opercular plates. The shell lengths of the long- and short-spined groups were not significantly different (Appendix B). The long- and short-spined groups were examined after 22.5 hours for barnacles that were: 1) consumed, 2) successfully being attacked, 3) unsuccess fully being attacked, and 4) unconsumed and not being attacked. In the long-spined group significantly more of the barnacles had been consumed than in the short-spined group where only one barnacle was consumed (Table 1). The mor phology of this single consumed barnacle was such that the spine of a short-spined Acanthina could reach the opercular plates. The experiment was allowed to continue for two weeks at which time the remaining barnacles were examined. The tests of consumed barnacles from the long-spined Acanthina tank had no bore marks, although one barnacle test had "scratch" marks on the operculum. The barnacles from the 19 Table 1. Consumption in Foraging Rate Experiment #1. Numbers of Tetraclita stactalifem consumed, successfully being attacked, unsuccessfully being attacked or unconsumed and not being attacked by starved Acanthina angelica after 22.5 and 336 hours. The Chi square of attacked and unattacked at t = 22.5 hours yields P < 0.01. HOUR Number Successfully Unsuccessfully Unconsumed Spine consumed being being and not being type attacked attacked attacked 22.5 Short 1 1 2 10 Long 10 2 1 1 336.0 Short 13 0 1 0 Long 14 0 0 0 20 short-spined Acanthina tank, however, had 10 cases of bor ings. Only one barnacle survived the two-week trial. It was a large individual from the short-spined snail tank and had been unsuccessfully bored twice from the outside. Using the same two groups of long- and short-spined Acanthina the experiment was immediately repeated so as to compare foraging rates of recently fed snails (Foraging Rate Experiment #2). The barnacles were examined at 12- hour intervals for 60 hours. After 60 hours when barnacles were no longer being consumed by either group of snails the experiment was terminated. The results (Table 2) were simi lar to the first experiment although the feeding rate was considerably reduced. The long-spined snails again fed sig nificantly faster (P < 0.03) than the short-spined group. In the long-spined group 11 barnacles had been consumed, compared to five by the short-spined group. In the long- spined group one barnacle was not attacked, whereas in the short-spined group nine were not attacked. Paine (1966a) maintained some short spined Acanthina angelica in his laboratory in Seattle, Washington. Fifteen of these were starved and then fed Balanus glandula. Paine noted (p. 18): "The snails invariably spent some time orienting themselves on the barnacles. Then the spine lo cated between the eyes of a crawling snail, was hooked over the outside rim of the barnacle or in some convenient cranny, after which the barnacle was drilled." 21 Table 2. Consumption in Foraging Rate Experiment #2. Numbers of Tetraclita stactalifem consumed, successfully being attacked, unsuccessfully being attacked or unconsumed and not being attacked by recently fed Acanthina angelica after 24, 36, 48 and 60 hours. The Chi square of attacked and unattacked yields P < 0.03. HOUR Number Successfully Unsuccessfully Unconsumed Spine consumed being being and not being type attacked attacked attacked 24 Short 12 1 11 Long 13 6 5 36 Short 3 1 2 9 Long 6 3 2 4 48 Short 4 0 1 10 Long 9 1 3 2 60 Short 5 0 1 9 Long 11 0 3 1 Further, "...little variation in this procedure was noted, and never was the spine forcibly thrust between the prey's opercular plates." Paine's observations are consistent with mine in that we both observed that short-spined Acanthina will drill their prey, Tetraclita. My observations, how ever, also show that long-spined Acanthina may consume Tetraclita without boring them. Thus demonstrating that the feeding behavior is more complex than it was originally . inferred to be. That Acanthina has an alternate foraging mechanism and that that mechanism may be spining barnacles open is verified by the following observations: 1) long-spined Acanthina orient themselves on unpenetrated barnacles with their spines inside the barnacle's opening and toward the posterior where there is a natural separation of the barna cle covers, 2) such positioned snails have been observed to "lunge" such that the spine must be thrust in the direction of the natural separation, 3) scratch marks may be found on the opercula of barnacles consumed by long-spined snails, 4) such barnacles were consumed much faster than drilled ones, and 5) barnacles were drilled only when the spine was too short to reach the opercula. In an experiment similar to Foraging Rate Experiment #1 and # 2, small barnacles, Chthamalus, were fed to starved (no food except for algae for four months) short- spined Acanthina. The snails immediately attacked the barnacles (who, interestingly, began mating). The spines were observed to be placed directly down on the barnacles and although the spine was then hidden from view by the mantle, the shells of the snails were observed to be thrust forward and down in what might be interpreted as a "lunge." Since it was not determined which particular barnacle of a clump was attacked, it was not possible to correlate lunges with successes. It was noted, however, that within a few days the snails were clearly bulldozing "empty" barnacle shells from the rock with their spines suggesting that con siderable force could be exerted with the spine (or possibly that the barnacle shells were loosened by chemical means). The pressure produced by the spine could conceivably be great if Acanthina used the spine as a lever. The calcu lation of this pressure and associated mechanical advantage is currently being studied by a graduate student at The University of Arizona (J. Short, per. comm.). The Effect of Diet on Spine Size: The Diet-regulator Hypothesis Paine (1966a) after observing the variability of Acanthina angelica spine lengths at Puerto Penasco com mented (p. 19) "It is tempting to suggest that the shorter spined individuals were probably eating small barnacles while the longer spined ones were consuming mature Tetraclita. Further field data are needed to resolve the reality of such a polymorphism." He then graphed the relationship of spine-shell lengths of a few snails from Puerto Penasco and from near San Felipe. The San Felipe data showed linear relationships whereas the Puerto Penasco data (N = 14) was distinctly curvilinear. I tested Paine's suggestion that Acanthina is poly morphic for spine length by graphing the relationship of 143 spine-shell lengths of snails from Playa Estacion (Station Beach)(Fig. 5). Three hundred and thirty two addi tional spine-shell lengths were also graphed (Fig. 6) with data recorded as to whether the snails were collected from basalt boulders (denoted by dots) encrusted only with Chthamalus or from beach rock (denoted by +) encrusted pri marily with the large barnacle Tetraclita. All snails were collected from the same general area and from the same tide level. From these data it is clear that adult Acanthina from Playa Estacion (Station Beach) are polymorphic for spine size. The significantly different regressions of the two groups support Paine's suggestion that the short-spined individuals are feeding on small barnacles (Chthamalus) and the long-spined individuals are feeding on the large barna cles (Tetraclita). Four hundred and fifty three spine lengths were then graphed according to substrate type (boulder Chthamalus, beachrock Chthamalus, and beachrock Tetraclita) to see if spine lengths are significantly greater in the large barnacle (Tetraclita) area (Fig. 7). 25 8- BY EYE -4- a. «/» 2- •• 10 20 30 35 40 SHELL LENGTH (mm) Fig. 5. Scatter graph of 143 Acanth.ina angelica spine lengths (mm) on shell lengths (mm). — Regression lines are fitted by eye and are suggestive of a spine polymorphism inferred to be the result of foraging in different barnacle areas. Specimens were collected from basalt boulders covered with Chthamalus and a beachrock terrace" covered with both Tetraclita and Chthamalus on 6 December 1975 from Playa Estacion (Station Beach) study site at Puerto Penasco. 26 • IN CHTHAMALUS + IN TETRACLITA ++ O" wi 20 30 35 SHELL LENGTH (mm) Fig. 6. Scatter graph of 332 Acanthina angelica spine lengths (mm) on shell lengths (mm). — The sig nificantly different regression lines clearly show A. angelica to be polymorphic for spine size in different barnacle areas. One hundred and ninty eight specimens were collected from basalt boulders covered with Chthamalus (dots) and 134 were from a beachrock terrace covered with Tetraclita (+) between 4 and 20 February 1977 from playa Estacion (Station Beach), Puerto Penasco. 27 o z z & BOULDER BEACHROCK BEACHROCK C HTHAMAIUS CHTHAMALUS TETRACLI TA Fig. 7. Dice-Leraas graph of Acanthina angelica spine lengths (nun) according to substrates. — The graph shows that the spine polymorphism is associated with barnacle substrate types. The beachrock Chthamalus area was located between the other two areas. Open boxes represent 95% confidence inter vals and the hashed boxes represent standard deviations. The specimens were collected from the + 1.2 meter to the + 1.8 meter tide level between 4 and 20 February 1977 at the Playa Estaciofi (Station Beach) study site, Puerto Penasco. 28 Clearly, the spine lengths were significantly different in the three areas. The group with the intermediate spine lengths were collected from an area in between the other two areas. The diet-regulator hypothesis was then tested by correlating spine size with barnacle size at various locali ties in the northern Gulf of California (Fig. 8). These observations reinforce the hypothesis that diet can regulate spine lengths but does not elucidate the mechanisms involved in changing the spine length. An increase in spine length could be caused by a short-spined snail feeding on a large barnacle by having its spine-secreting mantle adjacent to the spine, and possibly extending below it, as the feeding snail reached into the barnacle's cavity. Casual observa tions, however, indicate that the spine growth may occur at other times. Quiescent snails typically wrap the mantle around the spine so that they could be depositing calcium carbonate on the spine, causing it to grow. I have measured the rate at which broken spines will grow by breaking the spines of some long-spined Acanthina. The spines would tend to grow rapidly back to their original size even though the snails were not fed (Fig. 9). To test if diet is causally related to spine size (diet-regulator hypothesis), 10 short-spined and 10 long- spined snails of various sizes were fed small barnacles (Chthamalus). A similar set of 20 snails were fed large 29 •G E •H X •— 19 UiZ UI 0.49 R = 0.94 Z ui • A 20 MEAN BARNACLE HEIGHT (mm) Fig. 8. Graph of mean labial spine lengths (mm) of Acanthina angelica on mean barnacle heights (mm). — The data from study sites in the northern Gulf of California (Sonora) show a high correlation (r = 0.94) between spine lengths and barnacle size. The study sites and sample sizes (snails, barnacles) were: A) Desemboque de los Seris (30, 32); B) Playa de las Conchas (7, 15); C) Punta Pelicano (Pelican Point)(30, 31); D) Playa Norse (Norse Beach)(31, 30); E) F) and G) Puerto Libertad (124, 156)(109, 148) and (15, 10), respectively; and H) Punta Pelicano (32, 30). 30 + 2.0 E3- -eH SHQRT LONG SHOUT CONG SHOUT ALONG CONTROL LONG SPINEO BROKEN Fig. 9. Regrowth of broken long spines. — Dice-Leraas graph of spine growth (nun) of long-spined Acanthina angelica individuals that had their spines broken and of those that had their spines left intact showing the increased growth rate of broken spines. All snails that had their spines broken were originally long-spined. The spine growth experiment was over a 9-day period beginning on 11 October 1977. Open bars represent the 95% confidence interval. barnacles (Tetraclita). After three months the spine and shell lengths were re-measured and graphed (Fig. 10). All long-spined Acanthina fed small barnacles developed smaller spines as predicted by a diet-regulator hypothesis. The short-spined Acanthina fed on small barnacles did not significantly change their spine lengths, nor did the long- spined Acanthina fed on the larger barnacles (Fig. 11). The short-spined Acanthina fed on larger barnacles, however, produced results that at first appeared confusing. All snails over 25 mm long increased their spine lengths as pre dicted by the diet-regulator hypothesis. Four snails, 20 mm or less in length, however, developed shorter spines. These seemingly contradictory data became understandable when it was observed that these small snails could crawl inside the barnacles where they were able to sit directly on the oper cular plates thus making a long spine unnecessary. The experiment was repeated because in the first experiment the short-spined snails had significantly shorter shell lengths than the long-spined snails. In this second diet-regulator experiment (Diet-regulator Experiment No. 2) the variables were as before except that there was no sig nificant difference between the shell lengths of the experimental groups and that the sample size was increased from 10 to 30 individuals per group. An additional two groups of 15 long- and 15 short-spined snails were fed cracked or shelled Chthamalus and another set of 30 snails 32 Fig. 10. Spine reduction in snails fed small barnacles. Graph of spine length (mm) changes on shell length (mm) changes of 10 short-spined and 10 long-spined Acanthina angelica showing the reduction of long spines when fed the small barnacle Chthamalus. Arrows represent spine/shell length vectors of individual snails with the origin at the initial spine/shell measurement and the head at the terminal spine/shell measurement. The data are taken from the Diet-regulator Experiment #1 which was run from 21 February to 1 June 1977. 33 / I x i i 2 0 2 5 SHELL LENGTH (mm) Fig, 11. Spine elongation in snails fed large barnacles. — Graph of the spine length (mm) changes on shell length (mm) changes of 10 short-spined and 10 long-spined Acanthina angelica showing the elongation of short spines when fed the large barnacle Tetraclita. Arrows represent spine/ shell length vectors of individual snails with the origin at the initial spine/shell length measure ment and the head at the terminal spine/shell length measurement. The data are taken from the Diet-regulator Experiment #1 which was run from 21 February to 1 June 1977. * 34 were fed cracked or shelled Tetraclita. This manipulation was done to determine if spine-length change was influenced by barnacle flesh rather than barnacle size or shape. Again, short-spined snails did not significantly change their spine lengths when fed Chthamalus but grew significantly longer spines when fed Tetraclita (Fig. 12). The long-spined snails did not significantly change their spine lengths when fed Tetraclita but "grew" significantly shorter spines when fed Chthamalus (Fig. 13). The groups fed cracked Chthamalus tended to have shorter spines than those fed cracked Tetraclita, but the difference was not statistically significant. Changing the spine length may be a more difficult barrier to a snail switching to smaller barnacles than to snails switching to large barnacles. There are at least two possible mechanisms: 1) abrasion of the spines, or 2) ex tending the adjacent apertural lip and thus "enveloping" the spine. Examination of spines under a dissecting scope showed little or no sign of abrasion. It is not unrealis tic, however, to suggest that the eroded spines are quickly smoothed by the calcium carbonate secreting mantle. Counter evidence comes from the diet-regulator experiments. The labial margin of the snail's aperture was marked with India ink at the beginning of an experiment such that the growth of the lip could be measured. Spines and aperture growth were measured every few weeks. A decrease in spine length 35 9- 8- 7- E E X d" >- (9 Z / ui 3- Z 7*' / 4- 7 / v 7" 7' =E / •• r /' Z -/ 3- N: 29 30 14 13 aT I i I CHTHAMALUS TETRACLITA CRACKED CRACKED CHTHAMALUS TETRACLITA Fig. 12. Effects of diet on short-spined snails. — Dice- Leraas graph of spine lengths (mm) of short- spined Acanthina angelica at the termination of Diet-regulator Experiment #2 (22-24 February 1978) showing the increased spine length of large barna— (Tetradita) fed snails. Open boxes represent standard errors of the mean and hashed boxes represent the 95% confidence intervals. Dotted line indicates the mean spine lengths at beginning of the experiment. 36 CHTHAMALUS TETRACLITA CRACKED CRACKED CHTHAMALUS TETRACLITA Fig. 13. Effects of diet on long-spined snails. — Dice- Leraas graph of spine lengths (mm) of long-spined Acanthina angelica at the termination of Diet- regulator Experiment #2 (22-24 February 1978) showing the decreased spine length of small barna cle (Chthamalus) fed snails. Also, note the tendency for a reduction in spine length of snails fed cracked Chthamalus. Open boxes represent the standard errors of the mean and hashed boxes rep resent 95% confidence intervals. Dotted line indicates the mean spine lengths at beginning of the experiment. was usually accompanied by at least an equivalent amount of aperture lip growth (Table 3). It was not uncommon to find that no apertural growth occurred on some of the snails during some of the intervals, yet no significant decrease in spine length was observed for these individuals. Thus we may infer that decreased spine length occurs primarily by envelopment rather than by erosion. Having demonstrated that the snails can switch spine lengths we may pose the question, "Can the snails switch back and forth without suboptimal caloric output?" To an swer this question the weight gains of snails with "proper" spine lengths were compared to the weight gains of snails with "improper" spine lengths. That is, I compared the weight gains of the long-spined snails with the weight gains of short-spined snails when fed large and small barnacles. Examination of Diet-regulator Experiment #1 re vealed that indeed the short-spined snails that were fed the small Chthamalus did significantly better (as determined by weight gain) than the short-spined snails that were fed the large Tetraclita (Fig. 14). And the long-spined snails that were fed the large Tetraclita did significantly better than the small Chthamalus that were fed Tetraclita (Fig. 15). There was no significant difference between the shell lengths of the respective pairs of compared snails. Diet- regulator Experiment #2 could not be used in this sort of a comparison, since members of two experimental groups 38 Table 3. Apertural lip growth and change in spine length of 40 Acanthina angelica. — Data are from Diet- regulator Experiment #1 (20 Feb. to 1 June 1977). Spine type/diet Indiv.# Lip growth (mm) Spine change (mm) Long-spined 1 1.2 -1.2 fed 2 4.0 -1.7 Chthamalus 3 0.4 -1.0 4 1.1 -1.2 5 3.2 -1.7 6 7.1 -2.1 7 0.2 -0.2 8 0.2 -0.6 9 1.5 -1.3 10 0.5 -1.1 Long-spined 11 5.1 +0.5 fed 12 0.8 -0.7 Tetraclita 13 11.6 +0.3 14 3.3 -1.4 15 4.0 +0.3 16 2.0 -0.3 17 3.0 -0.7 18 1.5 +0.3 19 2.0 +0.3 20 8.6 -0.1 Short-spined 21 0.9 +0.4 fed 22 5.3 +0.6 Chthamalus 23 1.7 -0.5 24 8.5 -0.4 25 7.8 -0.4 26 9.8 +0.4 27 12.5 +0.1 28 9.5 +0.3 29 9.3 +0.9 30 29.5 +0.6 Short-spined 31 1.0 +1.7 fed 32 0.8 +0.9 Tetraclita 33 0.1 +1.0 34 3.9 +1.1 35 3.6 0.0 36 4.9 +0.2 37 4.0 -1.0 38 6.3 -0.6 39 3.5 -0.8 40 3.2 -0.1 39 CHTHAMALUS TETR ACLITA Fig. 14. Weight gains in short-spined snails. — Dice- Leraas graph of weight gain (g) in 20 short-spined Acanthina angelica in Diet-regulator Experiment #1 showing a significantly greater weight gain for those fed the small barnacle (Chthamalus). Open boxes represent standard errors of the mean and hashed boxes represent 95% confidence intervals. 40 CHTHAMALUS TETRACLITA Fig. 15. Weight gains in long-spined snails. — Dice- Leraas graph of weight gain (g) in 20 long-spined Acanthina angelica in Diet-regulator Experiment #1 showing a significantly greater weight gain for those fed the large barnacle (Tetraclita). Open boxes represent the standard errors of the mean and hashed boxes represent 95% confidence intervals. 41 escaped and lost an unknown amount of weight. Also, feeding in Diet-regulator Experiment #2 was variable. The results of Diet-regulator Experiment #1, however, sufficiently dem onstrates that the switching snails did poorly in comparison with non-switching snails (P < 0.05). The conclusion that snails with an improper spine length do poorly leads to an obvious optimal foraging strat egy. Clearly, a snail that spent much time switching would do poorly in comparison with a non-switching snail, given that all other factors are equal. To induce switching the benefits from switching would have to be proportionately greater. We might then predict that an optimally foraging snail would "consider" the cost to change an improper spine length to a proper one. This cost to switch or bioenergetic barrier might be considered a morphological interference to optimal foraging. An Optimal Foraging Model with a Barrier Term First, assume that the goal of a predator is to max imize its net caloric intake while feeding. Secondly, assume that the predator does not make mistakes either a priori or ex post facto. Note, that these assumptions do not represent the real world, but rather the limits that the suboptimal, mistake-ridden real world approaches in the evolutionary process. 42 Allow that the predator, Acanthina angelica, be placed where there is only one food resource, "i" (Chthamalus) or "j" (Tetraclita) ; and that the predator may have either of two morphological adaptations, "a" (short spine) or "b" (long spine), which allows the predator to maximize its net caloric gain (C) in the respective food resources, i.e., "a" in "i" and "b" in "j". Thus we know that in resource "i": C > C (1) ai bi and that in resource "j": C > C (2) bj aj from experimental data. Hence, we may suggest that the predator will expend the energy to switch spine lengths (S or S , bioenergetic ab ba barrier terms) if and only if the cost of switching is less than the potential gain due to increased predation efficien cy, that is: S < C - C (3) ba a b and, S < C - C (4) ab b a Conversely, we may predict that the predator will not switch if and only if: S > C - C (5) ba a b and, S > C - C (6), ab b a all of which is supported by experimental data. We can now make some interesting predictions based on these inequalities. For instance, given (2), the right hand side of (4) must be positive. And therefore, we must allow that given sufficient time, (4) will be true and the snail switches to a long spine. If S is very large, re- ab quiring time greater than ab the time to mean reproductive age, then (6) will be true. Likewise, a similar argument mav be made for the relationship between (3) and (5). Consider then a short-spined snail in an aquarium with both Chthamalus (i) and Tetraclita (j) available. Following Emlen (1966) we find that: C = f (I - I ) + f (J - J ) (7) a i a j a where f is the frequency (or proportion) of i and j in the diet, I and J are the caloric values of i and j, respec tively. and I and J are the respective caloric costs to a a eat i and j with a short spine. Likewise, we may derive the similar equation: C = f (I - I ) + f (J - J ) (8) b i b j b where I and J are the respective caloric costs to eat i b b and j with a long spine. 44 By substituting (7) and (8) into (5) we see that the bioenergetic barrier term inequality will hold and the pred ator will not switch spine lengths when: S >f(I-I)-f(J-J) (9). ba i b a j a b Since we know from experimental data that (I - I ) and b a (J - J ) are positive, we may predict certain conditions a b for which the switching barrier will exist. The inequality (9) will be satisfied and the predator will not switch when: f < f 0 (I - I ) < (J - J ) (10), i j b a a b or f > f 0 (I - I ) << (J - J ) (11). i j b a a b Since the right hand side of the inequality (9) is < 0 the preceeding assumes that the time to mean reproductive age is not important. The second statement (11) makes the inter esting prediction that the predator will not switch spine lengths even though the improper prey is more frequent in the predator's diet! In summary, this simple model describes the bioenergetic conditions that will produce and maintain polymorphism. Although temporal and non-discrete effects have not been considered in the present writing for the sake of simplicity, note that (9) may hold for short-term periods even though the right-hand side of the inequality is 45 positive. Furthermore, if this time period is greater than the time to mean reproductive age, polymorphism may still occur. DISCUSSION The Labial Spine To my knowledge no one has actually observed the labial spine penetrate between the opercular plates of a barnacle since the snail's foot and mantle cover the spine when it is inside the barnacle's opening. So let us brief ly consider mechanisms other than spining that may be used to attack and consume barnacles without leaving bore marks. It does not seem likely that smothering is the mechanism involved since I have subjected these same spe cies of barnacles to anaerobic conditions for up to two days without fatalities, while a barnacle may be completely consumed within a few hours of a "lunging attack." A more reasonable mechanism might involve a chemical relaxant since this could speed up the handling time. Still, it does not clearly explain the shell thrusting and the asso ciated scratch marks on the operculum. Thus, if a chemical relaxant is being used, it is quite possibly being used concurrently with the spining mechanism. All present evidence indicates that Acanthina angelica may either bore or spine its prey. A re examination of Paine's observations might clarify why he 46 47 concluded that Acanthina does not spine its prey. First, it may be noted that when Paine did his observations he was also doing field work on the food web of the sunstar, Heliaster kubiniji (see Paine 1966b), which occurs below the Tetraclita zone. This infers that he would be observ ing A. angelica at only the beginning and end of a low tide period to study Heliaster. At the beginning of the tide period the snails would most likely be "in pursuit" of prey items, while at the end of the tide period only those snails that could not readily spine and consume a prey item might be expected to be using the time-consuming drilling mechanism. Paine"s low correlation of spine posi tion with respe'ct to drill site is also reasonable since, as Paine has already suggested, the spine might be used by drilling snails simply as support. It would be interesting to know if these boring animals had either poorly fitting spines or spines too short to reach the barnacle's opercu lum as was the case in my feeding efficiency experiments. It may also be noted that Paine took short-spined snails to his laboratory in Washington where he fed them an unfa miliar Pacific coast barnacle. It is reasonable to expect these snails to drill rather than spine. Foraging Strategy It was clearly shown that in an aquarium situation snails will alter their spine size in response to a change in diet and that the response was a result of barnacle size and not diet. A close correlation was also found between spine length and barnacle size at various field locations in the northern Gulf of California. Weight gain studies further showed that snails with the proper spine length foraged more efficiently than snails with improper spine lengths. Knowing this we can attempt to make some predic tions about Acanthina's foraging strategy. Consider a snail with a spine too short for an encountered barnacle. The snail then has the option of spending more energy to find a barnacle that its spine will work on or drilling the animal at hand. These kinds of considerations have been reasonably well worked out at the theoretical level (Emlen 1966, 1968; Rapport 1971; Murdoch 1972) and will not be elaborated upon here. A third option, however, is availa ble to Acanthina angelica. It may expend the energy needed to change its improper spine length to a proper length by depositing additional shell material. In most cases this would undoubtedly involve a greater expenditure of energy than drilling the barnacle or crawling to anoth er. Thus we might predict that spine size changes would be more likely to occur as the snail encountered more and more barnacles in which a spine length change would facilitate foraging efficiency. 49 A second consideration would be the cost to back- switch. Consider a snail that develops a longer spine but does not then encounter large barnacles such that it is op timal to backswitch. If, however, we define optimal switching as being a resource switch that results in a greater long term efficiency or caloric gain for the pred ator, we must conclude that the snail made a "mistake" if it would have done better without switching the first time. Excluding genetic predisposition, a snail can respond only on the basis of its past encounter rate and thusly might be led into a suboptimal switch, say from short to long spine. And then, because of the high cost to backswitch, the snail does not change back to the short spine length. It is tempting to suggest that this is what is occurring at Playa Estacion (Station Beach) since there are clearly two morphs for spine size. Also it may be noted that snails between 25 and 30 mm in shell length appear to either switch to a long spine or maintain the short spine as they pass this "decision" point. Perhaps 25 mm is the minimum optimal size for using a large spine on Tetraclita and that above 30 mm it becomes suboptimal to backswitch since the reduc tion of spine length comes primarily from envelopment. And since the time for reduction of spine length is slow, par ticularly in comparison to the rate at which it can be increased, a long-spined snail might be more apt to wait for the proper food, drill its prey or attempt to use its 50 suboptimal spine. Thus the cost of switching, and espe cially from long to short, may act as a bioenergetic barrier enhancing the occurrence of two spine morphs. The Bioenergetic Barrier, the Model, and Disruptive Switching Switching may not occur when the cost to switch (the bioenergetic barrier) is greater than the loss to in efficient foraging. This keeps a predator from switching even though the less preferred prey is more frequent in the diet, as was clearly demonstrated in the mathematical model. The situation becomes more complex when one consid ers that a short-spined snail has the option to drill its prey and that in such a case, drilling may be more frequent as a result of conditioning, although Murdoch (1969) has shown that where preference is strong (as in the present case) the predator is not easily conditioned. The model does not attempt to consider the drilling option, condi tioning, nor that the predator must make its "decisions" based only on past prey encounters, again excluding genetic predisposition. The model, however, is intended to predict switch ing in the evolutionary most fit and/or genetically predispositioned individuals. The model then is an induc tive product derived from the particular foraging pattern of one species of marine snail. In this context let us 51 consider the general importance of the model and its ap plicability to other systems. For instance, humans "preying" on a non-lactose diet may encounter a bioenergetic barrier in switching to a lactose diet. This barrier is clearly the time and caloric conversion cost to switch metabolic pathways and produce lactase. This bioenergetic barrier may be so large that a sudden switch is fatal. The concept may also apply to the foraging strate gies of developing organisms that progress from one resource to another in a progressive switch. For example, the birth of a mammal is timed such that it occurs after the developing organism has put a considerable proportion of its energy budget into converting from a placental to a mammary resource capability. A second example of progres sive switching is the lepidopteran passing from a foliage predator to a nectar predator. Here the bioenergetic barrier may be represented by the caloric cost of metamor phosis. It is assumed that these progressive switches result (at least originally) in an optimal switch while backswitching is not typically within the range of natural variability. Further, when the progressive switch is not optimal one could expect neotany to evolve. Temporal switching is well known and in the present context would be termed optimal backswitching. The American Kestral may serve as an example since it switches from insects to rodents seasonally (S. Mills, Univ. of Ariz. grad. student, per. comm.). In such cases where the switch is frequent and/or predictable one might expect the bioenergetic barrier to be as small as possible, perhaps at the expense of foraging efficiency in each habitat. The bioenergetic barrier also describes a form of switching that I call disruptive switching. Disruptive switching may be defined"as a partial switching of a pop ulation or deme in which backswitching is suboptimal. This appears to be the present case since A. angelica with its two spine morphs occurs in a coarse-grained habitat (the barnacle areas) such that a snail in the Tetraclita area would find it bioenergetically suboptimal to shorten its spine. And a snail in the Chthamalus area would find it bioenergetically suboptimal to crawl 30 to 40 meters to a patch of Tetraclita. It is interesting to speculate that if, following Maynard-Smith (1966), there was a 30% selec tive advantage between the two morphs, the disruptive switching would allow disruptive selection to occur with the tenable result being disruptive speciation. This as pect, however, is well beyond the scope of this paper, although I have observed that short- and long-spined indi viduals mate and reproduce in separate areas. Still, it has been shown that the labial spine is clearly a morph ological barrier such that the two polymorphs are quite possibly the result of disruptive switching. 53 In conclusion, the model derived from this study of snail spines describes a mathematical framework for barrier formation. It might also inductively provide a theoretical basis for evolutionary barrier formation in a world in which individuals react molecularly and/or macroscopically according to their own unique histories and thus produce the variability necessary to the evolutionary process. In this process the barrier term may describe a mechanism for increasing niche specificity and subsequent reduction of competition in an evolving system. CONCLUSIONS 1) The predatory snail Acanthina angelica may ei ther drill or spine barnacles open for consumption, thus verifying the observations of MacGinitie and MacGinitie (1949; 1968, p. 371) and Paine (1966a). 2) Snails with spines either too short to reach the barnacle's opercular plates and/or inferred to be poor ly fitting into a natural separation of the opercular plates will drill the barnacles when "more suitable" prey are not available. 3) Spining may be done faster and more efficiently (as measured by weight gain) than drilling. 4) It was shown that spine size may be regulated by size and/or shape of the barnacles fed to the snails in an aquarium situation. 5) Field observations showed a high correlation (r = 0.94) between barnacle size and spine size of popula tions at various sites in the northern Gulf of California. 6) A mathematical relationship, S > f (I - I ) ba i b a - f (J - J ), was derived, based on the data of the diet j a b requlator and foraging efficiency experiments, that demon strates that a snail may continue to prey on an "improper" 54 55 sized barnacle when the cost to switch (a bioenergetic bar rier) was too great for an optimal switch. 7) It is suggested that the barrier term inequali ty is a general phenomenon and that it might be used to describe progressive, temporal and disruptive switching in an evolutionary context. APPENDIX A DICE-LERAAS GRAPH OF 30 ACANTHINA ANGELICA SHELL LENGTH AND SPINE LENGTH MEASUREMENT DIFFERENCES (mm) BETWEEN REPEATED CALIPER-RULER, CALIPER-CALIPER, AND RULER-RULER MEASUREMENTS Specimens were 15 short- and 15 long-spined individ uals from diet regulator experiment #2 and were numbered 121 to 150. Measurements were taken on 10 December 1977. Open blocks represent 95% confidence intervals and the hashed blocks represent the standard deviations. The conversion values for ruler to caliper measurements are: - 1.11 mm (length) and - 0.31 mm (spine). Conversion values have been used on all measurements prior to December 1977. 56 57 + 1.5- I/ + 1.0- €I/ r-4. E E + 0.5- ui U z / "T 0- / / —E Z ui / 2 ui / oc •0.5- 3 z i/i < -1.0- -1.5- 1 CAlS PER CALIPER RULER CALIPER CALIPER RULER RULER CALIPER RULER RULER CALIPER RULER SHELL LENGTH SPINE LENGTH APPENDIX B LIST OF ACANTHINA ANGELICA SPINE AND SHELL LENGTHS AND DIMENTIONS OF PREY BARNACLES, TETRACLITA STACTALIFEM, IN FORAGING RATE EXPERIMENT NO. 1 SNAILS ______—BARNACLES Spine Shell Distance Opening length length (mm) to diameter Height Basis (mm) (mm) operculum (mm) (mm) (mm) (short-spined Acanthina tank) 2.9 33.4 *1.9 6.9 7.4 14.4 3.3 32.3 3.0 3.9 11.3 20.4 3.3 36.2 3.3 2.8 10.8 16.9 3.3 37.0 3.7 4.9 11.7 22.4 3.6 41.5 3.7 4.8 11.1 21.9 4.0 40.5 4.1 4.2 8.7 21.7 4.0 41.3 4.3 4.6 7.7 22.7 4.1 33.3 4.6 7.3 13.8 24.1 4.1 37.1 4.6 7.3 16.3 26.5 4.1 39.0 4.8 5.6 12.3 23.0 4.3 41.3 5.1 7.4 17.6 26.8 4.4 40.1 5.7 6.5 12.6 27.0 4.6 36.7 5.9 6.4 13.3 31.5 4.9 41.0 6.4 8.7 12.6 26.7 5.2 38.4 (long-spined Acanthina tank) 7.0 37.5 3.0 6.4 2.5 24.4 7.2 31.7 4.1 7.2 9.4 27.9 7.7 38.7 4.3 5.7 14.3 28.9 8.5 36.2 4.5 5.3 14.1 27.5 8.5 39.3 4.5 4.1 10.0 22.9 8.5 40.7 4.5 3.0 8.4 19.4 8.6 38.3 4.8 5.6 14.0 29.0 8.7 37.9 5.0 6.0 14.3 23.9 8.8 38.4 5.1 7.2 15.1 26.4 8.9 37.4 5.3 4 < 7 9.9 22.0 8.9 38.0 5.4 5.7 13.4 29.7 9.3 36.1 5.7 5.7 11.2 23.1 9.4 37.9 6.0 6.7 12.5 22.4 9.4 39.6 6.3 7.4 15.7 24.5 10.0 36.5 •Consumed after 22.5 hours. 58 LITERATURE CITED CARRIKER, MELBOURNE ROMAINE 1955 Critical review of biology and control of oyster drills Urosalpinx and Eupleura. Spec. Sci. Rep.: Fish. no. 148: 1-150. 1961 Comparative functional morphology of boring mech anisms in gastropods. Am. Zool. 1: 263-266. EMLEN, JOHN MERRITT 1966 The role of time and energy in food preference. Am. Natur. 100(916): 611-617. 1968 Optimal choice in animals. Am. Natur. 102: 385-390. FOX, LAUREL RUTH 1970 Effects of past history and experimental condi tions on the functional response curve of the predator Notonecta fioffmani (Hemiptera: Notonectidae). M.A. thesis, Univ. of California, Santa Barbara. HEMINGWAY, GEORGE T. 1976 A comment on Thaisid "Cannibalism." Veliger 18 (3): 341. HOLLING, CRAWFORD STANLEY 1965 The functional response of predators to prey den sity and its role in mimicry and population regulation. Mem. Entomol. Soc. Can. 45: 1-60. HOUSTON, ROY SEAMANDS 1976 The structure and function of Neogastropod repro ductive systems: with special reference to Columbella fuscata Sowerby, 1832. Veliger 19 (1): 27-46. 59 60 KEEN, (ANGELINE) MYRA 1971 Sea shells of tropical west America. 2nd ed. Stanford Univ. Press, California, 1064 p. KOHN, ALAN JACOBS 1959 The ecology of Conus in Hawaii. Ecol. Monogr. 29: 47-90. LANDENBERGER, DONALD EDWIN 1968 Studies on selective feeding in the Pacific star fish Pisaster in southern California. Ecology 49: 1062-1075. LAWTON, J. H., J. BEDDINGTON, AND R. BONSER 1974 Switching in invertebrate predators. Iri Ecol- logical Stability, M. B. Usher and M. H. Williamson (eds.), Chapman and Hall. MacGINITIE, GEORGE, AND NETTIE MacGINITIE 1949 Natural history of marine animals. McGraw-Hill, New York, 473 p.; illus. 1968 Natural history of marine animals. 2nd ed. McGraw-Hill, New York, 523 p.; illus. MANLY, B. F. J., P. MILLER, AND L. M. COOK 1972 Analysis of selective predation experiment. Am. Natur. 106(952): 719-736. MAYNARD-SMITH, JOHN 1966 Sympatric speciation. Am. Natur. 100(916): 637-650. MENGE, JANE LUBCHENCO 1974 Prey selection and foraging period of the preda- ceous rocky intertidal snail Acanthina punctulata. Oecologia (Berlin) 17: 293-316. MURDOCH, WILLIAM W. 1969 Switching in general predators: experiments on predator specificity and stability of prey popu lations. Ecol. Monogr. 39(4): 335-352. 61 1972 The functional response of predators. Quantita tive Evaluation of Natural Enemy Effectiveness, Int. Congr. Entomol. (Canberra): 335-341. MURDOCH, WILLIAM W., AND ALLEN OATEN 1975 Predation and population stability. Adv. Ecol. Res. #9. OLDROYD, IDA S. 1918 List of shells from Angel and Tiburon Islands, Gulf of California, with description of a new species. Nautilus 32(1): 26-27. PAINE, ROBERT T. 1962 Ecological diversification in sympatric gastro pods of the genus Busycon. Evol. 16: 515-523. 1966a Function of labial spines, composition of diet, and size of certain marine gastropods. Veliger 9(1): 17-24. 1966b Food web complexity and species diversity. Am. Natur. 100(910): 65-75. RAPPORT, DAVID JOSEPH 1971 An optimization model of food selection. Am. Natur. 105(946): 575-587. REED, R. C. 1969 An experimental study of prey selection and regu latory capacity of bluegill sunfish (Lepomis macrochirus). M.A. thesis, Univ. of California, Santa Barbara. ROYAMA, T. 1970 Factors governing the hunting behavior and selec tion of food by the Great Tit (Parus major L.). J. Anim. Ecol. 39: 619-668. SCHOENER, THOMAS WILLIAM 1971 Theory of feeding strateqies. Annu. Rev. Ecol. Syst. (2): 369-404. TINBERGEN, L. 1960 The natural control of insects in pine woods. I. Factors influencing the intensity of predation by songbirds. Arch. Neerl. Zool. 13: 265-343. WELLS, HARRY W. 1958 Feeding habits of Murex fulvescens. Ecology 39(3) : 556-558. WOLFSON, FAY HENRY 1970 Spawning notes V. Acanthina angelica I. Oldroyd, 1918 and Acanthina lugugris (Sowerby, 1821). Veliger 12(3): 375-377.