ERROR LOADING HTML FROM SOURCE (http://ncf.sobek.ufl.edu//design/skins/UFDC/html/header_item.html)

Predation by Octopus vulgaris at Cayos Cochinos, Honduras

Permanent Link: http://ncf.sobek.ufl.edu/NCFE004433/00001

Material Information

Title: Predation by Octopus vulgaris at Cayos Cochinos, Honduras An Analysis of Prey Selection Using the Framework of Optimal Foraging Theory
Physical Description: Book
Language: English
Creator: Petro, Caitlin
Publisher: New College of Florida
Place of Publication: Sarasota, Fla.
Creation Date: 2011
Publication Date: 2011

Subjects

Subjects / Keywords: Optimal Foraging Theory
Octopus
Cayos Cochinos
Genre: bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Octopus vulgaris is a major predator of benthic organisms in the intertidal reef at Cayos Cochinos, Honduras. Because they are capable of capturing a variety of species using different foraging techniques, their predatory behaviors can have a significant impact on community trophic structures and population dynamics. Field work was conducted in the intertidal reef at Cayos Mejor as part of a pilot study aimed at characterizing the prey items that were returned to the dens of O. vulgaris. Major prey items were determined based on midden contents and the relative availabilities of prey species were approximated using randomized quadrats. Subsequent lab work was performed to assess the degree to which energetics influence prey selection in the field. Specifically, handling times and energy contents were determined for species of both low and high electivity. Factors such as patch-utilization and predator avoidance were found to play a significant role in the foraging behavior of O. vulgaris, suggesting that O. vulgaris is a time-minimizing forager.
Statement of Responsibility: by Caitlin Petro
Thesis: Thesis (B.A.) -- New College of Florida, 2011
Electronic Access: RESTRICTED TO NCF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE
Bibliography: Includes bibliographical references.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The New College of Florida, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Local: Faculty Sponsor: Gilchrist, Sandra

Record Information

Source Institution: New College of Florida
Holding Location: New College of Florida
Rights Management: Applicable rights reserved.
Classification: local - S.T. 2011 P46
System ID: NCFE004433:00001

Permanent Link: http://ncf.sobek.ufl.edu/NCFE004433/00001

Material Information

Title: Predation by Octopus vulgaris at Cayos Cochinos, Honduras An Analysis of Prey Selection Using the Framework of Optimal Foraging Theory
Physical Description: Book
Language: English
Creator: Petro, Caitlin
Publisher: New College of Florida
Place of Publication: Sarasota, Fla.
Creation Date: 2011
Publication Date: 2011

Subjects

Subjects / Keywords: Optimal Foraging Theory
Octopus
Cayos Cochinos
Genre: bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Octopus vulgaris is a major predator of benthic organisms in the intertidal reef at Cayos Cochinos, Honduras. Because they are capable of capturing a variety of species using different foraging techniques, their predatory behaviors can have a significant impact on community trophic structures and population dynamics. Field work was conducted in the intertidal reef at Cayos Mejor as part of a pilot study aimed at characterizing the prey items that were returned to the dens of O. vulgaris. Major prey items were determined based on midden contents and the relative availabilities of prey species were approximated using randomized quadrats. Subsequent lab work was performed to assess the degree to which energetics influence prey selection in the field. Specifically, handling times and energy contents were determined for species of both low and high electivity. Factors such as patch-utilization and predator avoidance were found to play a significant role in the foraging behavior of O. vulgaris, suggesting that O. vulgaris is a time-minimizing forager.
Statement of Responsibility: by Caitlin Petro
Thesis: Thesis (B.A.) -- New College of Florida, 2011
Electronic Access: RESTRICTED TO NCF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE
Bibliography: Includes bibliographical references.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The New College of Florida, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Local: Faculty Sponsor: Gilchrist, Sandra

Record Information

Source Institution: New College of Florida
Holding Location: New College of Florida
Rights Management: Applicable rights reserved.
Classification: local - S.T. 2011 P46
System ID: NCFE004433:00001


This item is only available as the following downloads:


Full Text

PAGE 1

PREDA TION BY OCT OPUS VULGARIS A T CA YOS COCHINOS, HONDURAS: AN ANAL YSIS OF PREY SELECTION USING THE FRAMEWORK OF OPTIMAL FORAGING THEOR Y BY CAITLIN CLAIRE PETRO A Thesis Submitted to the Division of Natural Sciences New College of Florida In partial fulfillment of the requirements for the degree Bachelor of Arts Under the sponsorship of Dr Sandra Gilchrist Sarasota, Florida May 201 1

PAGE 2

Acknowledgments The research for this thesis was funded by the Explorer s Club, the New College Alumni Association, the New College Council of Academic Af fairs, the Sarasota Shell Club, and Dr Sandra Gilchrist. Thank you to my committee members for their help and advice. Special thanks to my advisor Dr Sandra Gilchrist for all of her support and encouragement during my time at New College and for her commitment to providing students with the opportunity to study in Honduras. Thank you to Dr Peter Sandusky and Thomas V allieres for their help with the bomb calorimeter Thank you to my friends and family for their moral support and confidence in my ef forts. Special thanks to Eliza Cameron, who helped me tremendously with my field work and my data analysis. ii

PAGE 3

T able of Contents Acknowledgments ..............................................................................................................ii T able of Contents ...............................................................................................................iii List of T ables and Figures ..................................................................................................iv Abstract ..............................................................................................................................vi 1. Introduction ....................................................................................................................1 1.1 Optimal Foraging Theory ...................................................................................1 1.2 Octopus Foraging ...............................................................................................5 1.3 Feeding Behavior ...............................................................................................7 1.4 Prey Preference ................................................................................................10 1.5 Habitat and Dens ..............................................................................................13 1.6 Middens ............................................................................................................16 2. Methods ........................................................................................................................18 2.1 Field Research .................................................................................................18 2.1.1 Midden Sampling ...................................................................................19 2.1.2 Prey Abundance and Distribution ...........................................................23 2.2 Laboratory Research .......................................................................................24 2.2.1 Prey Handling .........................................................................................24 2.2.2 Bomb Calorimetry ..................................................................................26 3. Results ......................................................................................................................... 28 3.1 Octopus Dens ................................................................................................ 28 3.2 Midden Contents ........................................................................................... 34 3.3 Prey A vailability .............................................................................................42 3.4 Handling T imes ..............................................................................................49 3.5 Ener gy Content ...............................................................................................52 4. Discussion ....................................................................................................................54 4.1 Patch Utilization ..............................................................................................54 4.2 Prey Preference ................................................................................................56 4.3 Ener gy Content and Handling T ime ................................................................57 4.4 Octopus Dens ..................................................................................................59 5. References ....................................................................................................................61 Appendices ........................................................................................................................65 Appendix A: Field Data ......................................................................................65 Appendix B: Handling T imes .............................................................................70 Appendix C: Bomb Calorimetry Data ................................................................71 iii

PAGE 4

List of Figures and T ables .......... ............ ....... ............ ............ .......... Figure 1 Overview of External Octopus Anatomy ! 8 .......... ............ Figure 2 Features Inside the Buccal Cavity of ...... ............ .......... O. vulgaris 9 ............ ............ ........ ........ T able 1 Summary of Diet Composition Across V arious Studies 12 .......... ............ ...... Figure 3 The Cayos Cochinos Archipelago Natural Marine Monument 18 .......... ............ ............ ............ ........ Figure 4 A Section of the Study Site on Cayos Mejor ! 19 ......... ............ ............ ............ ........ Figure 5 Flagged Midden with Octopus Inside of Den ! 20 .......... ............ ..... ............ ............ ........ Figure 6 Map of Den Locations and Microhabitats ! 21 .......... ............ Figure 7 Measurement Standards for Bivalves, Gastropods .... ............ ............ ............ ............ ............ ........ and Crab Carapaces ! 22 .......... ............ ....... ............ ............ ............ ........ Figure 8 Handling T ime T ank and Subject ! 25 ............ ............ T able 1 Features of 5 Dens of ...... ............ ............ ............ ........ O. vulgaris ! 28 .......... ............ .............. ............ ............ ............ ............ ............ ............ ........ Figure 9 Den 1 ! ! 30 ........ ............ .............. ............ ............ ............ ............ ............ ............ ........ Figure 10 Den 2 ! ! 31 ........ ............ .............. ............ ............ ............ ............ ............ ............ ........ Figure 1 1 Den 3 ! ! 32 ........ ............ .............. ............ ............ ............ ............ ............ ............ ........ Figure 12 Den 4 ! ! 33 ........ ............ .............. ............ ............ ............ ............ ............ ............ ........ Figure 13 Den 5 ! ! 34 ............ ............ T able 2 Prey of Octopus vulgaris ............ ........ Found in Midden Remains 35 ........ ............ Figure 14 Overall Contents of 5 Middens of O. vulgaris ........ ............ ............ ............ ........ Collected Over a 14-Day Period ! 36 ............ ............ T able 3 Size V ariations Between Five Most Common ............ ............ ............ ............ ........ Prey Items Found in all Dens !! ! 36 ........ ............ Figure 15 T otal Amount of Prey Remains Found in Each Midden ..... ............ ............ ............ ........ Over the 14-Day Sampling Period ! 37 ............ ............ ......... ........ T able 4 Quantities of Prey Items Found in Individual Middens 37 ............ ............ T able 5 Number of Prey Remains in Each Midden During .......... ............ ............ ............ ............ ........ T otal Collection Period ! 40 ........ ............ ..... ............ ............ ........ Figure 16 Overview of Individual Midden Contents ! 41 ............ ............ T able 6 Potential Prey Items for O. vulgaris Found in 120 0.5 m 2 Quadrats W ithin Approximately 6,075 m 2 .... ............ ............ ........ of Backreef at Cayos Cochinos, Honduras ! 44 ............ ............ .......... ............ ........ T able 7 Abundances of Prey Items in Cayos Cochinos 45 ........ ............ .......... ............ ............ ........ Figure 17 Electivity (E) of Prey Species by T ype ! 46 ............ ............ T able 8 Electivity of Prey Species at Cayos Mejor as Represented ... ............ ............ ............ ............ ............ ........ by Midden Remains ! 47 ........ ............ .......... ............ ........ Figure 18 Electivity (E) of Prey Species by Microhabitat 48 ............ ............ T able 9 Mean Handling T imes (minutes) and Sizes (cm) for Three Prey Species of ... ............ ............ ............ ............ ........ O. vulgaris ! 49 ........ ............ ...... ........ Figure 19 Correlation Between Mussel Height and Handling T ime 50 ........ ............ Figure 20 Correlation Between Flame Scallop Height and ............ ............ ............ ............ ............ ............ ........ Handling T ime ! 51 iv

PAGE 5

........ ............ Figure 21 Correlation Between Crab Carapace W idth and ........... ............ ............ ............ ............ ............ ........ Handling T ime ! 52 .......... ............ T able 10 A verage Ener gy Contents of M. edulis and L. lima ... ........ Determined From Four Bomb Calorimetry Measurements 52 ........ ............ Figure 22 Comparison of Ener gy Contents and Handling T imes for M. edulis and L. lima .. ............ ............ ............ ............ ............ ........ ! 53 v

PAGE 6

Predation by Octopus vulgaris at Cayos Cochinos, Honduras: An Analysis of Prey Selection Using the Framework of Optimal Foraging Theory Caitlin Petro New College of Florida, 201 1 ABSTRACT Octopus vulgaris is a major predator of benthic or ganisms in the intertidal reef at Cayos Cochinos, Honduras. Because they are capable of capturing a variety of species using dif ferent foraging techniques, their predatory behaviors can have a significant impact on community trophic structures and population dynamics. Field work was conducted in the intertidal reef at Cayos Mejor as part of a pilot study aimed at characterizing the prey items that were returned to the dens of O. vulgaris. Major prey items were determined based on midden contents and the relative availabilities of prey species were approximated using randomized quadrats. Subsequent lab work was performed to assess the degree to which ener getics influence prey selection in the field. Specifically handling times and ener gy contents were determined for species of both low and high electivity Factors such as patch-utilization and predator avoidance were found to play a significant role in the foraging behavior of O. vulgaris, suggesting that O. vulgaris is a time-minimizing forager Dr Sandra Gilchrist Division of Natural Sciences vi

PAGE 7

Chapter 1: Intr oduction 1.1 Optimal Foraging Theory For motile or ganisms, feeding involves actively seeking out food items, or foraging. The act of foraging consists of a variety of behavioral decisions regarding issues such as location, searching strategy and prey selection. Optimal foraging theory attempts to predict these decisions by comparing them to the behaviors exhibited by a theoretical optimal forager (MacArthur and Pianka, 1966). Due to the critical importance of successful foraging to an or ganism s survival, this theory suggests that natural selection will favor foraging behaviors that maximize fitness. W ithin this context, fitness is measured as a function of a number of foraging parameters; common parameters include but are not limited to ener gy gain, rates of encounter and handling times (Pyke, 1984; Burrows and Hughes, 1990). By adopting foraging behaviors that maximize fitness as defined by relevant parameters, or ganisms are said to be exhibiting optimal foraging. The fundamental premise of optimal foraging theory is that foraging behavior has a heritable component that can be passed on to of fspring. Natural selection will favor the traits of individuals that contribute the most to subsequent generations, thereby resulting in a gradual shift in the population from the average foraging behavior towards the foraging behavior which gives maximum fitness (Pyke et al., 1977). Studies on various or ganisms have shown that foraging parameters do in fact have genetic components, and that these components may be directly related to fitness (Gosselin and Chia, 1996; Gibbons et al., 2005). Foraging behaviors are the products of complex interactions 1

PAGE 8

between these genetic components and learning. Although certain aspects of behavior may be af fected by heritable mechanisms, experience and learning are ultimately required to achieve optimal foraging ef ficiency (Gibbons et al., 2005). Optimal foraging theory can be divided into four categories: ( a ) optimal diet, ( b ) patch choice, ( c ) allocation of time to patches, and ( d ) movement between patches (Pyke, 1984). According to optimal diet theory or ganisms should select food items in such as way as to maximize their net rate of ener gy intake. Food items are added to the diet in rank order which is determined by the ratio of food value to handling time. Food value is commonly measured by ener gy content, although nutrient composition can also have an impact (Snellen et al., 2007). According to this theory the decision to include a food item in the diet is independent of that item s abundance and, instead, depends only on the abundances of food types of higher rank. As a result, an increase in the abundance of higher ranking food types should lead to greater specialization on these food types and the removal of less preferred prey types from the diet (Pyke et al., 1977). This ef fect can be modeled using Ivlev s index of electivity which measures the preference of a food item based on its inclusion in the diet relative to its proportional availability (Ivlev 1961): (1) where p i is the proportion of the i th species in the diet, and q i is the proportional availability of the i th species in the environment. Electivity ranges between -1 (complete avoidance) to +1 (greatest preference). An electivity of 0 indicates that a prey type is consumed in proportion to its field abundance. 2

PAGE 9

Or ganisms that learn from experience and modify foraging behavior based upon this knowledge have dynamic foraging parameters. The degree of prey specialization an or ganism exhibits determines how these parameters are altered. Foraging experience allows individuals to evaluate novel prey as well as estimate prey abundance. Foraging behavioral changes occur in response to quality and abundance of available food items, and are limited by the ability to analyze these conditions. In habitats where high-ranking food items are not very abundant, individuals can learn to use a exible search image to forage efciently increasing their ability to generalize. In habitats where high-ranking food items are abundant, individuals can learn to become more efcient at locating and handling preferred prey increasing their degree of specialization (Pyke, 1984). Selectivity does not always increase in proportion to foraging experience. Nucella emar ginata hatchlings with little foraging experience are more selective than late juveniles and adults, demonstrating that that species and size preference can exist in individuals who have no prior foraging experiences. The ener gy content of the preferred sizes of Mytilus sp. was not higher than the ener gy content of other prey species that were available. Nucella emar ginata hatchlings do not select Mytilus sp. to maximize ener gy gain. Instead, the prey preference exhibited here is designed to help the hatchlings locate the protective microhabitats small Mytilus sp. typically inhabit, thereby increasing their chance of survival during a period of heightened vulnerability (Gosselin and Chia, 1996). The problem of patch selection (as predicted by optimal patch choice theory) is similar to that of food selection in that both involve selection or rejection based on rankings. Patches are ranked based on factors such as resource abundance and travel distance (MacArthur and Pianka, 1966). Foraging experience can also af fect patch 3

PAGE 10

selection because it allows individuals to learn the locations and types of available patches. This allows individuals to adjust encounter rates with various patch types, thereby increasing foraging ef ficiency as well as the ability to specialize. The amount of time that is allocated to a certain patch is dependent on various factors. Generally more time should be allocated to patches with more abundant resources. However in situations where animals can forage in an area with some degree of exclusivity the abundance of food in a patch at a later time becomes a function of its behavior in the present. In this case, it should manage its resources and forage in additional patches, rather than maximizing the initial yield at the cost of poorer yields later (Baum et al., 1999). Optimal foraging behaviors are generally described as behaviors that maximize an individual s net rate of ener gy intake. In reality the relationship between foraging tactics and fitness is more complex; it is lar gely dependent on the animal in question as well as its surrounding habitat (Pyke, 1984). If an animal has a limited amount of foraging time and its fitness increases in proportion to its ener gy gain, its maximum fitness will occur when it maximizes its net rate of ener gy intake in the given amount of time. Such animals have been referred to as ener gy maximizers. However if an animal has a fixed ener gy requirement and requires time to perform other activities, it should obtain maximum fitness by minimizing the time required to acquire the necessary amount of ener gy Such animals have been referred to as time-minimizing foragers (Pyke et al., 1977; Leite et al., 2009). Predation pressures can alter optimal foraging tactics if they produce a conflict with increasing the net rate of ener gy intake. Under the threat of 4

PAGE 11

predation, foragers are predicted to engage in a trade-of f between maximizing foraging ef ficiency and minimizing predation danger They have been found to avoid patches with higher predation risk, regardless of resource density (Koy 2008). A better understanding of foraging behavior is necessary to increase our knowledge of community trophic structures and population dynamics. In order to understand food web structures, it is necessary to understand the nature of competitive interactions and predator -prey relationships that exist within a community (Drossel et al., 2001). The components of optimal foraging theory can provide insight into the nature of many coevolutionary systems, including competition, predator -prey dynamics, and symbiotic relationships (Snellen et al., 2007). Optimal foraging theory can also be used to assess population regulation and diversity within a community Generalist predators stabilize populations and increase the ability for or ganisms to coexist within a particular habitat (Y amamoto, 2004). Specialist predators feed disproportionately on preferred prey thereby producing or exacerbating species population variations (Symondson et al. 2002). 1.2 Octopus Foraging Octopuses are intelligent and opportunistic predators. They lack stereotyped foraging behaviors and, instead, modify their hunting tactics in response to the environment (Mather 1991; Leite et al., 2009). As a result, their foraging distances and patterns vary depending on the location and availability of prey Mather and ODor (1991) found that O. vulgaris in Bermuda hunted within a circular area of 15 m diameter 5

PAGE 12

in the intertidal zone, while Forsythe and Hanlon (1997) reported foraging distances up to 120 m for O. cyanea located on a coral atoll in French Polynesia. Observations of octopus foraging suggest that they are time-minimizing foragers, rather than ratemaximizing foragers (Leite et al., 2009). T o minimize foraging time and exposure to predators, octopuses hunted intensely and captured prey unselectively They have been found to forage within microhabitats containing patches of preferred prey resulting in the consumption of many species with specialization on a few (Leite et al., 2009). In the field, octopuses have been found to forage in a highly speculative manner Octopus cyanea, O. vulgaris, and O. insularis have all been observed using chemotactile exploration to assess surrounding habitats and locate hidden prey items during foraging (Y arnall, 1969; Mather 1991; Leite et al., 2009). Although octopuses do not rely primarily on their keen vision for prey capture, it is still used in a variety of ways during foraging. V isual landmarks are thought to be used for orientation, allowing octopuses to construct a visuospatial map so that they can navigate between likely food patches and find their way back to their dens (Leite et al., 2009). V ision is also used to recognize predators and to determine the cryptic body patterns that are appropriate to conceal the octopus from predation (Forsythe and Hanlon, 1997). Finally visual cues allow octopuses to modify their foraging behaviors in response to the environment. Mather (1991) observed O. vulgaris switching from chemotactile exploration to a jet-propelled attack when a prey item attempted to escape. Octopuses have been shown to alter their foraging tactics depending on the location and availability of prey While many foraging behaviors have been identified, 6

PAGE 13

the majority of these behaviors can be classified either as pouncing or groping. Pouncing occurs when an octopus envelops a small coral head or rock with its arms and web. When the coral or rock is too lar ge to be surrounded by the web, the arms are instead inserted into crevices to remove hidden prey items. This behavior is referred to as groping. Forsythe and Hanlon (1997) found that O. cyanea spent approximately 60% of its time away from its den actively searching for food by either pouncing or groping; the remainder of it s time was spent either moving (21%) or sitting (19%). Movements that occur while foraging can be classified either as crawling or swimming. Octopuses are capable of swimming in two orientations: forward swimming and fast-backward swimming is commonly used to make rapid, extended trips back to dens at the conclusion of a foraging trip (Forsythe and Hanlon, 1997). Prey can be consumed while out foraging or brought back to den. Y arnall (1969) observed O. cyanea collecting and paralyzing up to four crabs before returning back to the den to eat. 1.3 Feeding Behavior Octopuses possess a wide variety of morphological and behavioral variations that allow them to consume a diverse array of prey Prey-handling behaviors are dependent upon factors such as type of prey prey size, and octopus size. When attempting to penetrate bivalve shells, octopuses have been found to use two main methods: pulling and drilling (McQuaid, 1994). Before pulling occurs, the bivalve is positioned underneath the octopus interbranchial web with the valve mar gin facing its buccal mass (shown in Figure 1) (Steer and Semmens, 2003). The octopus can 7

PAGE 14

then attempt to separate the two valves using force exerted by its arms and suckers. Each pull, which lasts for approximately five seconds, is visibly marked by tension in the arms (McQuaid, 1994). The strength of the pulling force is lar gely dependent upon the area of suckers in contact with the shell and their degree of adhesion to the surface. The contact area can be increased by using the proximal portion of the arms, which contain lar ger suckers. The ef ficiency of sucker adhesion is determined by the outer surface of the shell and varies among bivalve species (Fiorito and Gherardi, 1999). If an octopus is unable to pull apart a bivalve due to size, strength, or poor surface adhesion it will resort to drilling (McQuaid, 1994; Steer and Semmens, 2003). Figure 1. Overview of external octopus anatomy Image modified from: http://www .bumblebee.or g/ invertebrates/Cephaopoda.htm Steer and Semmens (2003) observed that O. dierythraeus rotated their prey 90 o when shifting from pulling to drilling. Following this rearrangement, the octopus drilled 8 Buccal mass Mantle W eb Arm

PAGE 15

for 60 90 minutes before returning the bivalve to its original position and pulling the two valves apart. The hole is drilled by a combination of mechanical and chemical forces. The posterior salivary glands secrete corrosive saliva that degrades the shell, thereby allowing more ef ficient penetration by the radula and salivary papilla (Figure 2) (Fiorito and Gherardi, 1999). Nixon and Maconnachie (1988) found these secretions to be essential in the drilling process. After the hole is completed, toxic salivary secretions are injected into the prey by a duct in the salivary papilla. Ghiretti (1959) identified the main toxic component of these secretions as the glycoprotein cephalotoxin (as reviewed in Altman and Nixon, 1970). These secretions paralyze the bivalve, allowing it to be easily opened by pulling. The drilling penetration site has been found to vary depending on the species of octopus. While O. vulgaris has been found to drill predominantly into the adductor muscle (Nixon and Macconachie, 1988), O. dierythraeus was observed to tar get the valve periphery (Steer and Semmens, 2003). These variations may either be the result of varying prey physiology or preferred grip. Figure 2. Features inside the buccal cavity of O. vulgaris Image from: Altman and Nixon, 1970. 9

PAGE 16

In contrast to bivalves and other armored prey items, gastropods are drilled more frequently than pulled due to morphological characteristics that make it dif ficult to pull them apart. Anderson and colleagues (2008) found that approximately 60% of gastropod prey remains found at middens belonging to O. vulgaris had been drilled. When O. vulgaris first captures a gastropod, it has been observed to insert the tip of one arm into the aperture to check for the presence of an or ganism. If the shell is inhabited, it is brought under the web and positioned against the buccal mass with the aperture pointing outward. Completion of the hole was found to take between one and two hours. Gastropod drilling holes are generally located in the spire (Arnold and Arnold, 1969). Once the soft tissue of the prey is exposed, the two chitinous beaks located within the buccal mass (Figure 2) are used to remove the flesh and break it up into smaller pieces. The breakdown of food prior to entrance into the buccal cavity helps to limit distension of the esophagus, which is in close proximity to the brain (Altman and Nixon, 1970). The tissue is broken up by the raising and lowering of the upper beak onto the lower one, which functions primarily as a support for the beak muscles. The food particles are then impaled by the radula and carried into the buccal cavity where they are transported to the esophagus by the lateral buccal palps (Altman and Nixon, 1970). Once the majority of the tissue is consumed, the radula is used to clean any remaining flesh of f of the shell or exoskeleton. 10

PAGE 17

1.4 Prey Preference Shallow-water octopuses have been found to act as major predators of both sessile and motile species within benthic communities. They are equipped with a variety of capture and handling techniques, thereby enabling them to consume a diverse array of prey (Steer and Semmens, 2003). Middens of O. bimaculatus were found to contain the remains of five dif ferent prey groups: snails, crustaceans, bivalves, hermit crabs, and sedentary grazers (Ambrose, 1984). These groups are consistent with the literature on prey selection in other octopus species, with the potential addition of fish and other octopuses to the diet (Altman and Nixon, 1970; Smith, 2003). Despite extensive research on octopus diets, the factors influencing prey preference are still lar gely unknown. Data on the subject are gathered mostly from indirect methods, owing to the dif ficulty of observing unaltered behaviors in the field. After field observations, stomach content analysis is regarded as the most accurate method, as midden counts can be misrepresentative of actual diet (see section 1.6) (Smith, 2003). Numerous studies have used laboratory choice experiments to determine prey preference (Ambrose, 1984). However these models involve limited prey choice and lack important factors, such as predation risk and resource distribution, which can influence foraging behavior (Pyke et al., 1977). Diet composition has been found to vary significantly between entire populations as well as for individual octopuses. Because they are common predators on a variety of prey octopuses have been frequently classified as generalists (V incent et al., 1998; Leite et al., 2009). However numerous studies have shown that individuals within a generalist 11

PAGE 18

12 Species Location Sampling T echnique Crustaceans Bivalves Gastr opods Other Author Y ear O. vulgaris Honduras Midden contents None 82.4% 17.6% None Hamman 2009 O. vulgaris Honduras Midden contents >90% < 5% None None Rosebrock 2000 O. vulgaris Caribbean Midden contents 30% 51% 19% None Anderson et al. 2008 O. insularis Brazil Midden contents 70% 17.5% 12.5% None Leite et al. 2009 O. vulgaris Bermuda Midden contents 2 1 rare None Mather 1991 O. bimaculatus California Midden contents 5% 6% >70% >10% Ambrose 1984 O. bimaculatus California Laboratory Observations 1 2 3 None Ambrose 1984 O. cyanea French Polynesia Field Observations Xanthid crabs Common Common None Forsythe & Hanlon 1997 O. vulgaris Mediterranean Midden contents <20% 37.6% 42.4% minimal Ambrose & Nelson 1983 O. vulgaris South Africa Stomach Contents None Mussel None None Smale & Buchan 1981 O. vulgaris South Africa Midden contents 12.8% 12.8% 71.5% 3.6% Smith 2003 O. vulgaris South Africa Stomach Contents 51.6% 0% 28.6% 19.7% Smith 2003 O. vulgaris South Africa Field Observations 1 1.1% N/A 77.8% 1 1.1% Smith 2003 T able 1. Summary of diet composition across various studies.

PAGE 19

population can exhibit varying degrees of specialization. Midden contents collected from a population of O. vulgaris in the Caribbean were of 75 species and were widely dispersed between bivalves, gastropods, and crustaceans. Individual midden contents, however showed that some octopuses were specializing on certain taxa or individual species (Anderson et al., 2008). Field observations of O. insularis revealed similar patterns. While the overall diet included over 55 species, half of the total prey occurrences were of four species (Leite et al., 2009). T able 1 provides a summary of the results from numerous studies on octopus prey preference and diet composition. Prey selection in the field is the product of complex interactions between factors such as experience, resource distribution and abundance, and predation risk. Numerous studies have shown that these interactions are consistent with the predictions made by optimal foraging theory Ambrose (1984) found that O. bimaculatus added prey types to its diet in rank order depending on their relative availabilities. Highly preferred species that were relatively rare made up only a small portion of the diet, which was mostly composed of abundant species with moderate preference rankings. Preferences involving gastropods demonstrated selection between dif ferent species, as low-ranking species that were abundant were not selected. Preference rankings were determined using combined data from laboratory choice tests and midden contents. Ener getic return also appeared to influence selection, as was demonstrated by McQuaid (1994). Using laboratory observations, McQuaid demonstrated that O. vulgaris selected sizes of mussels that maximized their rate of ener getic return, as measured by E/H. The E/H value is 13

PAGE 20

determined by dividing the ener getic value of the prey by the ener gy required to open and consume the prey 1.5 Habitat & Dens Octopuses are widely distributed across the world. The most common octopus, Octopus vulgaris, has been found to occur in many areas, including the Caribbean, of f the British Isles, of f South Africa, and in the Mediterranean (Mather 1991; Smale and Buchan, 1981; Ambrose and Nelson, 1983). They are a coastal species, commonly found in the intertidal or shallow subtidal zones, within tropical waters at depths between 0.2 m and 2.0 m (Mather and ODor 1991). Octopuses are capable of adjusting to many dif ferent ecosystems; they have been found to inhabit coral reefs, sandy bottoms, bedrocks or boulders, and grass flats. Benthic octopuses are sedentary or ganisms. They select and modify shelters known as dens, where they spend up to 70% of their day (Mather 1991). The existence of suitable shelters has been shown to impose significant constraint on octopus distributions (Katsanevakis and V erriopoulos, 2004). Octopuses are common in more structured areas, where den availability is high. In French Polynesia, Octopus cyanea was found to select only subtidal dens in the backreef in approximately 1 m of water They occupied naturally occurring holes in the hard substratum, which were found to be in excess. (Forsythe and Hanlon, 1997). In southern California, Octopus bimaculatus shelters were found more frequently in bedrock and boulders than any other substrate type (Ambrose, 1982). In Cayos Mejor O. vulgaris was found to inhabit primarily either 14

PAGE 21

crevices in the major reef structure, areas of patch reef, or holes in a submer ged rock pile (Hamman, 2010). Den availability is often more limited in open areas, especially on soft sediment. In these areas, shelter construction is dependent on the octopus ability to locate suf ficient solid materials such as rocks, shells, and human debris (Katsanevakis and V erriopoulos, 2004). The construction of individual dens has been found to vary depending on the availability of suitable shelter locations and building materials. On hard substratum, octopuses have been found to dig or blow out crevices filled with sand and other debris (Ambrose, 1982; Forsythe and Hanlon, 1997). Dens can be excavated in a similar manner on soft sediment, leading to the formation of wells or holes underneath rocks. A well consists of a hole dug into the sediment that is lined around the inner wall with solid objects, such as rocks and shells. In the latter construction, an inclined cavity is dug out from underneath a rock (Katsanevakis and V erriopoulos, 2004). Smaller octopuses, such as Octopus joubini and juvenile O. vulgaris, have been found to inhabit empty mollusk shells when available (Mather 1991). Solid materials of human origin have also been seen to function as den construction materials. Katsanevakis and V erriopoulos (2004) found that the densities of juvenile O. vulgaris increased in areas of soft sediment that were enriched with artificial dens constructed of human litter Among the most frequently occupied objects were plastic water bottles, tires, aluminum cans, and cement blocks. Individual dens do not function as permanent residences. Octopuses have been shown to move between shelters, although the length of time spent occupying one shelter 15

PAGE 22

can vary Octopus cyanea were found to continuously reside in one den for a maximum of 23 days (Y arnall, 1969), while juvenile O. vulgaris have been found to have an average occupancy length ranging between 4.5 and 12.4 days (Mather and ODor 1991). Octopus bimaculatus of f the coast of southern California remained in the same den for upwards of a month, with some octopuses occupying a shelter for over five months (Ambrose, 1982). Dens are frequently modified, both at the beginning of occupancy and throughout an octopus s residency In French Polynesia, O. cyanea pulled rocks or rubble over the opening of their shelter during the day (Forsythe and Hanlon, 1997). Octopus bimaculatus were observed moving stones and other loose debris around inside their dens (Ambrose, 1982). Juvenile O. vulgaris have been found to block of f lar ge den entrances with items such as mollusk shells, rocks, and crab carapaces (Mather 1991). Rocks and stones have also been found around the perimeter of the den, away from its opening. 1.6 Middens Prey that are caught during foraging are frequently brought back to the den for consumption. As a result, dens are commonly surrounded by piles of discarded prey items. These piles are referred to as middens. Although midden contents vary between individual octopuses, they often contain bivalve and gastropod shells as well as crab carapaces. Middens are common for many octopus species, although dens belonging to O. cyanea (Forsythe and Hanlon, 1997) and O. bimaculatus (Ambrose, 1983) have been 16

PAGE 23

observed without them. They are used by researchers both to locate octopus dens and as sources of information regarding prey choice and consumption (Mather 1991). Although middens are valuable tools for studying octopus prey they provide an incomplete picture of an individual s actual diet. Prey items that are consumed away from the den are not represented in the midden contents, which commonly results in underestimation of daily food intake. Mather (1991) found that juvenile O. vulgaris consumed approximately 70% of its total prey at its den and 30% while out foraging. The decision to bring items back to the den was significantly correlated with the distance from the site of prey capture to the den. Size and dif ficulty opening prey had no ef fect on the decision by O. vulgaris to return items to the den. Therefore, the items in the middens were considered to be representative of the range of prey that were consumed. In a study by Smith (2003), midden contents of O. vulgaris in South Africa did not appear to be representative of actual diet. When compared to stomach contents, midden items had a relatively low concentration of crustaceans and soft-bodied or ganisms, but a relatively high concentration of shelled mollusks. Discrepancies such as these often result from the dif ferential loss of individual prey items from middens. Remains from small crustaceans can be removed by currents or gravity while heavier remnants from lar ge bivalves and chitons are frequently buried (Mather 1991). Gastropod shells can be removed by hermit crabs (Gilchrist, 2003), as well as by waves and currents (Ambrose, 1983). Ambrose (1983) hypothesized that O. bimaculatus may have been removing shells surrounding the den to decrease their visibility to potential predators. T o increase the accuracy of midden estimates, prey items must be collected 17

PAGE 24

shortly after they are discarded. Mather (1991) estimated that only 50% of prey remains were visible near the den five days after consumption. 18

PAGE 25

Chapter 2: Methods 2.1 Field Research The field research was conducted from July 13 to July 30, 2010. The study site was located at Cayos Cochinos Mejor approximately 30 km northeast of La Ceiba, Honduras. This location is part of the Cayos Cochinos Archipelago Natural Marine Monument, an area that is protected under Legislative decree 1 14-2003. Data were collected from the back reef of the south western bay of Cayos Mejor (Figure 3). Figure 3. The Cayos Cochinos Archipelago Natural Marine Monument. The study site is indicated by the black arrow 19

PAGE 26

The area where the study was performed is shown in Figure 4. Data were collected from the site two times per day once for midden sampling and once for prey abundance quadrats. Data regarding daily weather and water conditions were also recorded. Figure 4. A section of the study site on Cayos Mejor The arrow points north. Data were collected approximately 60 m north of the dock and 100 m south of the dock. 2.1.1 Midden Sampling Octopus dens were located by snorkeling along approximately 200 m of shoreline at a water depth of 0.5 to 3 m. The area was surveyed for middens, piles of clean shells and other debris that would indicate the presence of an octopus. When a midden was located, it was further inspected to determine whether an octopus was present. Octopuses 20

PAGE 27

were identified as O. vulgaris by the presence of dark rings around the circumference of their suckers as well as the lack of a dark ring surrounding their eyes. Dens occupied by O. vulgaris were marked with a red flag as shown in Figure 5 and the midden contents were removed. Figure 5. Flagged midden with octopus inside of den. The octopus is indicated by the black arrow Five dens were marked in total, although the octopus in Den 1 abandoned its den on the sixth day of collection. For each den, the surrounding environment was noted and depth measurements were taken. The dens were mapped relative to each other and to surrounding landmarks using transect tape. A map of the den locations is shown in Figure 6. 21

PAGE 28

Figure 6. Map of den locations and microhabitats. Red represents the reef areas, green represents the grass flats, and yellow represents sandy regions. The areas in blue were unexplored. Octopus dens are indicated by black circles. The arrow points to the north. Measurements are in meters. 22

PAGE 29

The midden contents were collected from each den once a day for a total of 14 days. Each item was labeled and identified to either genus or species using the procedure described in Emerson & Jacobson (1976). Items that could not be identified on the island were photographed and identified at a later time. Intact prey items were measured using a caliper The measurement standards for dif ferent prey types are summarized in Figure 7. All items were inspected for damage and mollusk shells were examined for drill holes. When all observations were recorded, the items were returned to the water away from the den. The prey items were analyzed and individuals diet breadths were calculated as: 1 / i 2 (2) where i is the proportion of the i th species in the diet. The average sizes of each of the most prominent prey items were calculated. These data were used to determine the samples used for the handling time and calorimetry tests. Figure 7. Measurement standards for bivalves, gastropods, and crab carapaces. (A) Bivalves were measured with the umbo in the dorsal position and (B) gastropods were measured with the apex in the dorsal position. (C) Crab carapace widths were measured parallel to their fronts. 23

PAGE 30

2.1.2 Prey Abundance and Distribution A 0.5 square meter quadrat was used to sample the most abundant or ganisms present in the intertidal area of the reef. Only populations of mollusk bivalves, gastropods, and crabs were sampled. The quadrat was placed in a total of 120 locations within a total area of roughly 6,075 m 2 at depths between 0.5 and 3 m. The quadrats were split between three dif ferent microhabitats. Forty were performed in a sandy area, forty in a region of submer ged aquatic vegetation (SA V), and forty in the rocky area of the reef. Each of these environments were measured using transect tape and a map of the entire area was created (Figure 4). The total areas of the dif ferent microhabitats were roughly 1,586 m 2 (sand); 4,015 m 2 (SA V); 474 m 2 (coral). The quadrats were assigned to a specific location within a microhabitat using a random number generator ( www .random.or g ). The first number indicated the distance travelled in meters along the perimeter of the microhabitat from the corner nearest the dock. The second number was used to determine the distance travelled in meters towards the shoreline. Half of the quadrats were collected north of the dock and half were taken south of the dock. In the SA V and sandy areas, the area within the quadrat was inspected by manipulating the grass and sand by hand to uncover any items buried in a thin layer of sediment. In the rocky habitats, the quadrats were inspected carefully to prevent injury or damage to the corals. Items found within the quadrats were photographed and recorded. The prey items were analyzed and the mean number per square meter were calculated for each species. These data were combined with the midden content results to calculate food preference using Ivlev s Index of Electivity (Eq. 1). 24

PAGE 31

25 Figure 8. Handling time tank (top) and subject (bottom). The PVC pipe shelter is indicated by the black arrow and the octopus in the tank is indicated by the white arrow (top picture). The test subject was a juvenile weighing 142.2 g.

PAGE 32

2.2 Laboratory Research 2.2.1 Prey Handling The prey handling tests were conducted from February 18 to March 22, 201 1 at Pritzker Marine Laboratory at New College of Florida in Sarasota, Florida. The test subject was a 142.2g juvenile O. vulgaris obtained from T om s Caribbean T ropicals, Inc. The octopus was isolated in a 60 x 180 x 65 cm tank with a 40-cm section of PVC pipe for shelter (Figure 8). The octopus was allowed time to habituate to its new environment before the experimental trials began, during which it fed primarily on crustaceans from Sarasota Bay T wo weeks prior to the handling time tests, the octopus was fed bivalves ( Mytilus edulis ) to increase its handling familiarity Formal observations of prey handling commenced approximately 40 days after the octopus arrived at the lab. The experimental procedure was adapted from Fiorito and Gherardi (1999). The octopus was starved for 24 h before the handling tests, which were performed at night. Experimental trials began when a prey item was dropped into the tank approximately 30cm from the octopus s shelter The trials ended when the octopus finished consuming the prey item or after 1 h if the prey item was never consumed. The times of events were recorded with a stopwatch to the nearest second by continuous direct observation. The total handling time (T h ) was divided into two components: the time required to open the prey (T o ) and the time required to eat it (T e ). T o started when the octopus began to attempt to open a prey item and ended when feeding began. The beginning of feeding was marked by a pulsing of the octopus head, which occurred as the flesh was torn away 26

PAGE 33

from the shell or carapace. T e ended when the prey remnants were pushed out from under the web. Handling time tests were performed for Mytilus edulis (mean height = 6.1 cm; SD = 0.26 cm), Lima lima (mean height = 6.4 cm; SD = 0.66 cm) and Mithraculus sculptus (mean width = 1.8 cm; SD = 0.32 cm). Lima lima and M. sculptus were both found in the natural diet of O. vulgaris. Although M. edulis was not found in the natural diet of O. vulgaris mussels of similar morphology ( Modiolus modiolus ) were. The octopus was fed approximately 5-7 g (shell free wet weight) of food per day Prey were added to the tank one at a time. Behavioral observations, such as the use of pulling vs. drilling, were recorded in addition to handling times. The discarded shells of M. edulis and L. lima were removed and examined macroscopically for drill holes. 2.2.2 Bomb Calorimetry The bomb calorimetry was performed at Eckerd College in St. Petersbur g, Florida from February 25 to March 25, 201 1. The tests were performed using a 1 15 V Parr Oxygen Bomb Calorimeter (Model No. 13031). The bomb was filled to a pressure of 25 atm with oxygen and submer ged in 2,000 mL of distilled H 2 O. The calorimeter was calibrated using 1.0 g benzoic acid tablets. The bomb was calculated as having a heat capacity equal to 10,620 43 J / o C. Mytilus edulis and L. lima were used for the calorimetry tests. The bivalves were removed from the water and left to sit for approximately 15 minutes before the samples were prepared. The tissue was separated from the shell, rinsed with distilled water 27

PAGE 34

blotted, and weighed. The samples were dried for 36 hrs at 60 o C in a dehydrator The dried tissue was removed from the dehydrator and finely ground using a mortar and pestle. The ground samples were stored with a desiccant in an airtight container until needed. Six tests were performed for each species. 28

PAGE 35

Chapter 3: Results The following data were gathered as part of a pilot study with the goal of characterizing the prey items that were returned to the dens of O. vulgaris The species that are described are not representative of the entire diet of O. vulgaris as data associated with prey consumed away from the den are not included. 3.1 Octopus Dens A total of five dens inhabited by O. vulgaris were found (Figures 9 -13). Each den was located in one of three types of microhabitats: the major reef structure (reef), sandy substratum (sand), or a bed of submer ged aquatic vegetation (SA V). The depth and surrounding microhabitat of each den is shown in T able 1. Four out of five of the dens were found on soft sediment, either in sea grass beds or sand flats. The sea grass beds were composed primarily of Thalassia testudinum. The dens were all found at similar depths except for Den 2, which was located in a crevice in the reef approximately 1 m from the shoreline. All of the dens appeared to remain inhabited by O. vulgaris for the duration of the study except for Den 1, which was abandoned on the sixth day of collections. All five of the octopuses were juveniles and were similar in size (approximately 200 300 g). T able 1. Features of 5 dens of O. vulgaris. 29 Den Depth Microhabitat Location 1 2.5 m SA V Hole inside of patch reef structure 2 1.0 m Reef Crevice within major reef structure 3 2.0 m SA V Crevice in submer ged rock pile 4 3.0 m SA V Hole inside of rocky structure 5 2.5 m Sand Hole in the ground

PAGE 36

Figure 9. Den 1. Octopus location is marked with a white arrow The discarded prey remnants are visible near the entrance to the den. This den was located in a patch reef structure in a sea grass bed. The octopus appeared to abandon its den on the sixth day of collection after an eel was found in its place. 30

PAGE 37

Figure 10. Den 2. This den was located in a crevice in a major reef structure, approximately 1 m from the shoreline. Midden contents were frequently absent from this den, possibly due to removal by waves. 31

PAGE 38

Figure 1 1. Den 3. This den was located in a crevice in a submer ged rock pile in a sea grass bed at a depth of 2.0 m. The octopus location is indicated by the white arrow 32

PAGE 39

Figure 12. Den 4. This den was located in a hole inside of a rocky structure in a sea grass bed, approximately 3.0 m in depth. The octopus (indicated by the white arrow) is shown eating a gastropod. A mango pit (indicated by the black arrow) was used to seal of f the den opening when the octopus was not active. 33

PAGE 40

Figure 13. Den 5. This den was located in a hole in the sandy substrate, at a depth of approximately 2.5 m. The entrance to the den is indicated by the white arrow Discarded prey remnants are visible to the right of the den. 34

PAGE 41

3.2 Midden Contents A total of 194 prey items were collected from 5 middens. They were classified into 30 taxa of bivalve (63.3%), gastropod (26.7%), and crustacean (10.0%) species (T able 2). While the number of species was lar ge, 70.1% of the prey remains were of five species: Anadara notabilis (18.6%), Atrina rigida (18.0%), Modiolus modiolus (17.5%), Codakia sp. (8.23%), and Glycymeris decussata (7.73%). T en other species each contributed to over 1% of the total prey occurrences. In total, the top 15 species accounted for 92.2% of all the midden contents. The remaining 15 species were each found only once during the collection period. In total, these 15 species contributed to 7.73% of the total midden contents over the course of the three week study The relative abundances of individual prey types are shown in Figure 14. Bivalves were the most common prey items found at the middens, accounting for 88.7% of the total midden contents. Gastropods and crustaceans were found much less frequently contributing to 7.75% and 3.55% of the total midden contents, respectively All of the gastropod shells that were found in the middens had drill holes. Drill holes on bivalve shells were more dif ficult to locate due to the absence of one of the valves and/or shell damage. The thin shells of A. rigida were almost always broken, indicating that they were opened by force. Drill holes were identified on 7 of the 34 M. modiolus remnants. However this value is not considered to be representative of the actual proportion that were drilled. The sizes of the five most common prey items were found to vary according to species (T able 3). 35

PAGE 42

Coded Letter T axon Species T otal number Percent of T otal Items Den occurrence C Bivalvia Anadara notabilis 36 18.6 4 H Bivalvia Atrina rigida 35 18.0 3 F Bivalvia Modiolus modiolus 34 17.5 4 D Bivalvia Codakia sp. 16 8.23 2 E Bivalvia Glycymeris decussata 15 7.73 4 K Bivalvia Saxidomus giganteus 8 4.12 2 O Gastropoda Oliva cir cinata 8 4.12 2 B Bivalvia Lima lima 5 2.58 1 N Crustacea Calappa gallus 5 2.58 2 S Bivalvia Ar ca sp. 4 2.06 1 M Bivalvia Glycymeris sp. 3 1.55 1 Y Bivalvia Ctenocar dia sp. 3 1.55 1 Z Bivalvia Laevicar dium laevigatum 3 1.55 1 R Bivalvia Pinctada sp. 2 1.03 1 U Gastropoda Cerithium atratum 2 1.03 1 A Bivalvia Chione paphia 1 0.52 1 G Bivalvia T ellina listeri 1 0.52 1 I Bivalvia Macr ocallista maculata 1 0.52 1 J Bivalvia T ellina laevigata 1 0.52 1 L Gastropoda Cyphoma gibbosum 1 0.52 1 P Bivalvia Lima pellucida 1 0.52 1 Q Gastropoda Cypraea sp. 1 0.52 1 T Bivalvia Cyrtopleura costata 1 0.52 1 V Bivalvia Lithopoma tectum 1 0.52 1 W Gastropoda Semicassis sp. 1 0.52 1 X Bivalvia T ellina lineata 1 0.52 1 AA Gastropoda Thais deltoidea 1 0.52 1 BB Crustacea Mithraculus sculptus 1 0.52 1 CC Crustacea Hepatus epheliticus 1 0.52 1 DD Gastropoda Str ombus gigas 1 0.52 1 T able 2. Prey of Octopus vulgaris ( n = 194) found in midden remains. The designated letters code for the indicated species in subsequent graphs. 36

PAGE 43

Figure 14. Overall contents of 5 middens of O. vulgaris collected over a 14-day period. Letters correspond to prey types indicated in T able 2. Species Range of W idth (mm) A vg. W idth (SD) (mm) Range of Height (mm) A vg. Height (SD) (mm) Anadara notabilis ( n = 36) 9 43 29.82 (8.8) 7 29 20.18 (4.9) Atrina rigida ( n = 35) 40 66 55.53 (7.2) 105 139 122.7 (9.5) Modiolus modiolus ( n = 34) 20 29 23.31 (2.6) 31 51 43.08 (6.9) Codakia sp. ( n = 16) 13 53 29.43 (1 1.6) 16 63 35.57 (14.7) Glycymeris decussata ( n = 15) 16 27 20.88 (3.5) 18 30 24.19 (3.9) T able 3. Size variations between five most common prey items found in all dens. Bivalves were measured with umbo in the dorsal position. Each midden accumulated remains of a dif ferent amount of prey during the 14day collection period (Figure 15; T able 4). The average midden produced remains of 39.4 4.6 prey items during the entire collection period, with an average of 3.2 0.16 0 5 10 15 20 25 30 35 40 A B C D E F G H I J K L M N O P Q R S T U V W X Y Z AA BB CC DD T otal Number of Items Collected Prey T ype 37

PAGE 44

remains per day Midden 1 had the lar gest mean number of items per day However this value may be less accurate than the others due to the reduced sampling period. Midden 2 had the smallest total amount of prey items, despite having a full collection period. Prey remnants may be have been removed from Midden 2 by waves, as it was located near the shoreline in a more turbulent area. Figure 15. T otal amount of prey remains found in each midden over the 14-day sampling period. Midden Number of Collection Days T otal Items Mean Number of Items Per Day (SE) 1 6 31 5.17 (0.37) 2 14 24 1.71 (0.21) 3 14 49 3.50 (0.13) 4 14 43 3.07 (0.19) 5 14 50 3.57 (0.28) T able 4. Quantities of prey items found in individual middens. 0 12.5 25.0 37.5 50.0 Midden 1 Midden 2 Midden 3 Midden 4 Midden 5 Number of Prey Items Midden 38

PAGE 45

The midden contents varied between each den, with most octopuses showing signs of generalization. An overview of the composition of each midden is shown in T able 5 and Figure 16. The relative abundances of dif ferent taxa in each midden were consistent with the overall midden contents. Bivalves were the most common prey group, comprising between 79.2% (Midden 2) and 93.0% (Midden 4) of the contents of each midden. Gastropod remains were less common, although they were found to occur in each midden at least once. The percentage of gastropod remains in each midden varied between 2.0% (Midden 5) and 20.8% (Midden 2). These values could be skewed by the possible removal of gastropod shells from the middens by hermit crabs. Crustacean remains were only found in Middens 1 and 5, comprising 8.0% and 9.6% of the total prey items, respectively Analysis of the midden contents revealed dietary variations between individuals. During the 14-day collection period, individual octopuses had represented remains of 8 to 12 prey species in the middens, with a mean value of 9.0 1.5. Individuals exhibited varying degrees of specialization on dif ferent prey species. Diet breadth was generally high for all of the middens. The lowest relative diet breadth belonged to the octopus from Midden 4, due to the frequent remains of A. rigida at the midden. Midden 2 had the highest relative diet breadth, however this value may be misrepresentative due to the potential loss of prey remains from the midden by waves. The most common species to occur in the overall diet were consumed by at least two of the five octopuses. Anadara notabilis was consumed by four A. rigida by three, M. modiolus by four Codakia sp. by two, and G. decussata by four The relative rates of consumption of these species varied 39

PAGE 46

between individual octopuses. Although A. notabilis remains were found in four of the five middens, 40.5% and 43.2% of its overall occurrences were from Middens 1 and 2, respectively Atrina rigida remains were found in three of the five middens and accounted for 18.0% of the total midden contents. However 73.0% of its occurrences were from Midden 4. 40

PAGE 47

Midden number Midden number Midden number Midden number Midden number 1 2 3 4 5 Bivalves Anadara notabilis 15 3 16 3 Atrina rigida 2 8 27 Modiolus modiolus 4 13 3 14 Codakia sp. 2 13 Glycymeris decussata 2 4 3 7 Saxidomus giganteus 3 5 Lima lima 5 Ar ca sp. 4 Glycymeris sp. 3 Ctenocar dia sp. 3 Laevicar dium laevigatum 3 Pinctada sp. 2 Chione paphia 1 T ellina listeri 1 Macr ocallista maculata 1 T ellina laevigata 1 Lima pellucida 1 Cyrtopleura costata 1 Lithopoma tectum 1 T ellina lineata 1 Gastr opods Oliva cir cinata 5 3 Cerithium atratum 2 Cyphoma gibbosum 1 Cypraea sp. 1 Semicassis sp. 1 Thais deltoidea 1 Str ombus pugilis 1 Crustaceans Calappa gallus 3 2 Mithraculus sculptus 1 Hepatus epheliticus 1 Number of prey items 31 24 49 43 50 Number of prey species 9 8 8 8 12 Diet breadth (B) 3.6 6.4 4.6 2.4 5.5 41 T able 5. Number of prey remains in each midden during total collection period. Diet breadth was calculated using Eq. 2.

PAGE 48

Figure 16. Overview of individual midden contents. Letters correspond to prey types indicated in T able 2. The 15 remains of prey types that contributed to >1% of the total midden contents are shown. Other codes for the 15 prey types that contributed <1% of the total midden contents. Midden 1 Midden 2 Midden 3 Midden 4 Midden 5 42

PAGE 49

3.3 Prey A vailability A total of 75 individual prey items were sampled from the 120 x 0.5 m 2 quadrats. They were classified in 25 taxa of gastropod (48.0%), bivalve (44.0%), and crustacean (8.0%) species (T able 6). Bivalves were the most abundant prey items, accounting for 45.3% of total items found. Gastropods were less abundant than bivalves, comprising 37.3% of the total items collected. Crustaceans were relatively rare and contributed only 17.3% to the total items found. The five most abundant species were: A. rigida (14.7%), L. lima (14.7%), M. scultpus (13.3%), C. gibbosum (9.3%), and O. cir cinata (8.0%). These five species accounted for >60% of the total prey items that were found. Out of the 25 taxa, 14 species were found only once. The dif ferent microhabitats varied in prey composition. Only one species, O. cir cinata, was found to occur in all three microhabitats. Species that were located in two of the three microhabitats were generally more abundant in one, as was seen with A. rigida and M. sculptus. The coral area contained the lar gest number of individual prey (40.0%), followed by the sea grass beds (34.7%), and the sand (25.3%). Prey type diversity was greatest in the sea grass beds, which contained 14 of the 25 dif ferent species. The sea grass beds also contained 8 of the 14 (57.1%) species that were sampled only once between all of the quadrats. These results may be reflective of the total area of the SA V microhabitat (4,015 m 2 ), which was the lar gest of the three. Sea grass beds can also function as a protective habitat for small, benthic or ganisms. Both factors could result in increased species diversity within SA V The most abundant species in SA V were A. rigida; 10 of the 1 1 A. rigida were collected from SA V The sandy area contained 43

PAGE 50

mostly crustaceans (58.0%), which accounted for 1 1 of the 13 total crustaceans found in all the quadrats. The coral area contained a variety of gastropods and bivalves. All of the L. lima were found in crevices between rocks in the coral microhabitat. Densities of the prey species were calculated for microhabitats as well as the overall area (T able 7). The overall densities of the prey types were lar gely dependent on the microhabitats they were located in and their corresponding areas. As a result, these values deviated from the percent occurrences given in T able 8. Atrina rigida was predicted as having the highest mean density (0.35 per m 2 ) because it was found primarily in sea grass beds, which covered >60% of the sampling area. Lima lima was predicted as having one of the lowest mean densities (0.04 per m 2 ), although its percent occurrence was equal to that of A. rigida. Overall, the most abundant prey were predicted to be: A rigida M. sculptus, C. gibbosum, H. epheliticus, and L. laevigatum. All of these species were found to some extent in the sea grass beds. 44

PAGE 51

Coded Letter T axon Species T otal number Percent of T otal Items Microhabitat Occurrence Microhabitat Occurrence Microhabitat Occurrence Coded Letter T axon Species T otal number Percent of T otal Items Sand SA V Coral C Bivalvia Atrina rigida 1 1 14.7 1 10 -T Bivalvia Lima lima 1 1 14.7 --1 1 B Crustacea Mithraculus sculptus 10 13.3 9 1 -K Gastropoda Cyphoma gibbosum 7 9.3 -3 4 H Gastropoda Oliva cir cinata 6 8.0 2 1 3 W Gastropoda Cittarium pica 5 6.7 --5 D Crustacea Hepatus epheliticus 3 4.0 2 1 -E Gastropoda Cerithium atratum 2 2.7 1 -1 I Bivalvia Laevicar dium laevigatum 2 2.7 -2 -F Bivalvia Macr ocallista maculata 2 2.7 2 --S Bivalvia Pinctada sp. 2 2.7 --2 A Bivalvia T ellina lineata 1 1.3 1 --G Gastropoda Leucozonia nassa 1 1.3 1 --J Gastropoda Natica livida 1 1.3 -1 -L Bivalvia Petricola pholadiformis 1 1.3 -1 -M Bivalvia Glycymeris sp. 1 1.3 -1 -N Bivalvia Ar chidae sp. 1 1.3 -1 -O Gastropoda Conus mindanus 1 1.3 -1 -P Gastropoda Str ombus gigas 1 1.3 -1 -Q Bivalvia Ar ca decussata 1 1.3 -1 -R Bivalvia Saxidomus giganteus 1 1.3 -1 -U Gastropoda Fissur ellidae sp. 1 1.3 --1 V Gastropoda Fasciolaria lilium 1 1.3 --1 X Gastropoda Lithopoma tectum 1 1.3 --1 Y Gastropoda Thais deltoidea 1 1.3 --1 T able 6. Potential prey items ( n = 75) for O. vulgaris found in 120 0.5 m 2 quadrats within approximately 6,075 m 2 of backreef at Cayos Cochinos, Honduras. The designated letters code for subsequent graphs. 45

PAGE 52

Species Mean (SD) No. per 0.5 m 2 Mean (SD) No. per 0.5 m 2 Mean (SD) No. per 0.5 m 2 Mean No. per m 2 in total area Species Sand SA V Coral Mean No. per m 2 in total area Atrina rigida 0.03 (0.03) 0.25 (0.08) 0.00 ( ) 0.35 Lima lima 0.00 ( ) 0.00 ( ) 0.28 (0.09) 0.04 Mithraculus sculptus 0.23 (0.07) 0.03 (0.03) 0.00 ( ) 0.16 Cyphoma gibbosum 0.00 ( ) 0.08 (0.04) 0.10 (0.05) 0.12 Oliva cir cinata 0.05 (0.03) 0.03 (0.03) 0.08 (0.04) 0.05 Cittarium pica 0.00 ( ) 0.00 ( ) 0.13 (0.06) 0.02 Hepatus epheliticus 0.05 (0.03) 0.03 (0.03) 0.00 ( ) 0.07 Cerithium atratum 0.03 (0.03) 0.00 ( ) 0.03 (0.03) 0.02 Laevicar dium laevigatum 0.00 ( ) 0.05 (0.03) 0.00 ( ) 0.07 Macr ocallista maculata 0.05 (0.03) 0.00 ( ) 0.00 ( ) 0.03 Pinctada sp. 0.00 ( ) 0.00 ( ) 0.05 (0.03) 0.01 T ellina lineata 0.03 (0.03) 0.00 ( ) 0.00 ( ) 0.02 Leucozonia nassa 0.03 (0.03) 0.00 ( ) 0.00 ( ) 0.04 Natica livida 0.00 ( ) 0.03 (0.03) 0.00 ( ) 0.04 Petricola pholadiformis 0.00 ( ) 0.03 (0.03) 0.00 ( ) 0.04 Glycymeris sp. 0.00 ( ) 0.03 (0.03) 0.00 ( ) 0.04 Ar chidae sp. 0.00 ( ) 0.03 (0.03) 0.00 ( ) 0.04 Conus mindanus 0.00 ( ) 0.03 (0.03) 0.00 ( ) 0.04 Str ombus gigas 0.00 ( ) 0.03 (0.03) 0.00 ( ) 0.04 Glycymeris decussata 0.00 ( ) 0.03 (0.03) 0.00 ( ) 0.04 Saxidomus giganteus 0.00 ( ) 0.03 (0.03) 0.00 ( ) 0.04 Fissur ellidae sp. 0.00 ( ) 0.00 ( ) 0.03 (0.03) 0.00 Fasciolaria lilium 0.00 ( ) 0.00 ( ) 0.03 (0.03) 0.00 Lithopoma tectum 0.00 ( ) 0.00 ( ) 0.03 (0.03) 0.00 Thais deltoidea 0.00 ( ) 0.00 ( ) 0.03 (0.03) 0.00 T able 7. Abundances of prey items in Cayos Cochinos. 46

PAGE 53

The prey abundance results were analyzed with midden occurrences to estimate food preferences using Ivlev s Index of Electivity (T able 8, Figure 17). These data represent only the items that were returned to the dens, not all of the prey consumed. Electivity ranges between -1 (complete avoidance) to +1 (greatest preference). An electivity of 0 indicates that a prey type is consumed in proportion to its field abundance. Data were excluded for prey types having either midden proportions or field proportions equal to 0.01 while the other proportion is equal to 0.00. The estimated electivities for these species were inaccurate, as they predicted either complete avoidance or greatest preference based on only one data value. The estimated electivity for bivalves indicated that they were preferred, while gastropods and crustaceans were avoided. Anadara notabilis, M. modiolus, Codakia sp., and G. decussata all appeared to be selectively consumed. Atrina rigida was selected in proportion to its abundance. Lima lima, C. pica, C. gibbosum, and M. sculptus appeared to be avoided. Figure 17. Electivity (E) of prey species by type. B represents bivalves, G represents gastropods, C represents crustaceans. Proportions correspond to values indicated in T able 8. 47 0 0.225 0.450 0.675 0.900 B (E=0.33) G (E=-0.64) C (E=-0.62) Proportion of T otal Items Proportion in middens Proportion in eld

PAGE 54

Species No. in middens Proportion in middens Field abundance Proportion in field Electivity Bivalves 171 0.88 33 0.44 0.33 Anadara notabilis* 36 0.19 0 0.00 1.00 Ar ca sp.* 4 0.02 0 0.00 1.00 Atrina rigida* 35 0.18 1 1 0.15 0.09 Codakia sp.* 16 0.08 0 0.00 1.00 Ctenocar dia sp.* 3 0.02 0 0.00 1.00 Glycymeris sp.* 3 0.02 1 0.01 0.33 Glycymeris decussata* 15 0.08 1 0.01 0.78 Laevicar dium laevigatum* 3 0.02 2 0.03 -0.20 Lima lima 5 0.03 1 1 0.15 -0.67 Macr ocallista maculata 1 0.01 2 0.03 -0.50 Modiolus modiolus* 34 0.18 0 0.00 1.00 Pinctada sp. 2 0.01 2 0.03 -0.50 Saxidomus giganteus* 8 0.04 1 0.01 0.60 T ellina lineata 1 0.01 1 0.01 0.00 Gastr opods 16 0.08 28 0.30 -0.64 Cerithium atratum 2 0.01 2 0.03 -0.50 Cittarium pica 0 0.00 5 0.07 -1.00 Cyphoma gibbosum* 1 0.01 7 0.09 -0.80 Lithopoma tectum 1 0.01 1 0.01 0.00 Oliva cir cinata* 8 0.04 6 0.08 -0.33 Str ombus gigas* 1 0.01 1 0.01 0.00 Thais deltoidea 1 0.01 1 0.01 0.00 Crustaceans 7 0.04 13 0.17 -0.62 Calappa gallus 5 0.03 0 0.00 1.00 Hepatus epheliticus 1 0.01 3 0.04 -0.60 Mithraculus sculptus 1 0.01 10 0.13 -0.86 T able 8. Electivity of prey species at Cayos Mejor as represented by midden remains. Data are excluded for species for which either proportion in diet or proportion in field is equal to 0.01 while the other proportion category is equal to 0.00. Data for entire taxa include all prey items sampled. Symbols indicate a species microhabitat: SA V *, sand, coral 48

PAGE 55

Electivity was also calculated for the total prey types in each of the three microhabitats (Figure 18). The results indicated that prey items located in SA V (E = 0.51) were selected and brought back to the dens with greater frequency than items in sand or coral. Prey items from coral areas (E = -0.72) appeared to be avoided almost entirely These results suggest that patch utilization plays a major role in the foraging behavior of the octopuses that were examined. The selection of patches may be influenced by factors such as prey abundance, travel distance, and predation pressures. Some of the most common prey types in the middens ( A. notabilis, Codakia sp, M. modiolus ) are located in SA V microhabitats. Although they were abundant in the middens, these items were not found in any of the SA V quadrats. These prey items may have been located in dense aggregations within the SA V that the octopuses utilized while foraging. Figure 18. Electivity (E) of prey species by microhabitat. The values above the bars indicate the total number of prey items in each category 49 0 0.2 0.4 0.5 0.7 0.9 SA V (E = 0.51) Sand (E = -0.53) Coral (E = -0.72) Proportion of total items Proportion in middens Proportion in eld

PAGE 56

3.3 Handling T imes Handling times were recorded for 28 individual prey items, belonging to three species (T able 9). The total handling time (T h ) was divided into the time required to open the prey item (T o ) and the time required to eat it (T e ). The octopus used dif ferent handling methods for each of the species. Mytilus edulis were drilled, L. lima were pulled apart using force, and M. sculptus were penetrated by the beak and consumed immediately The octopus attempted to open M. edulis using force before drilling, although it usually pulled only once or twice before switching to drilling. Species Height (SD) W idth (SD) T o (SD) T e (SD) T h (SD) Mytilus edulis* (n = 13 ) 6.06 (0.26) 2.95 (0.24) 90.01 (5.55) 18.93 (1.51) 108.94 (5.91) Lima lima (n = 5 ) 6.40 (0.29) 5.12 (0.61) 5.20 (0.70) 16.12 (2.44) 21.31 (2.83) Mithraculus sculptus (n = 10) 1.60 (0.34) 1.81 (0.32) -13.62 (1.27) 13.62 (1.27) T able 9. Mean handling times (minutes) and sizes (cm) for three prey species of O. vulgaris. Drilled items are indicated by *, all other items were opened by force. The handling time components were lar gely dependent on the method of consumption that was used. Drilling (of M. edulis ) took the longest of the three methods, with T h ranging between 99.6 min and 108.9 min. When drilling was used, T h was not correlated to mussel height (r = 0.34; p > 0.25, Figure 19) due to the variations in T o T e was correlated to mussel height when drilling was used (r = 0.98; p < 0.01, Figure 19). 50

PAGE 57

Figure 19. Correlation between mussel height and handling time. T e (top) was correlated to mussel height (r = 0.98; p < 0.01). T h (bottom) was not correlated to mussel height, due to variations in T o Pulling (of L. lima ) took less time than drilling, with T h ranging between 17.5 min and 24.3 min. In these cases, T h was correlated to scallop height (r = 0.93; p < 0.02, Figure 20), as was T e (r = 0.98; p < 0.01, Figure 20). "#$" %%#"! %$#&" &'#!! "#! "#" (#! (#" $#! "#$%&'( )*++",#-"&./ 0#$1%( (! $" )! %!" %&! "#! "#" (#! (#" $#! 20 3,#/3'4,&'.#0&%"#$%&'( )*++",#-"&./ 0#$1%( 51

PAGE 58

Figure 20. Correlation between flame scallop height and handling time. T e (top) was correlated to scallop height (r = 0.98; p < 0.01). T h (bottom) was also correlated to scallop height (r = 0.93; p < 0.02). Handling times for M. sculptus were the shortest of the three methods, with T h ranging from 1 1.2 min to 15.2 min. T o was not recorded for these trials, as it was too dif ficult to dif ferentiate between the breaking of the carapace and the consumption of the flesh. T h *#$" )#"! %*#&" %)#!! ( $ "#$%&'( 5,3%"#61 3,,27#-"&./ 0#$1%( (#&" %&#"! %+#$" &"#!! *#!! *#$" "#"! (#&" $#!! 20 3,#/3'4,&'.#0&%"#$%&'( 5,3%"#61 3,,27#-"&./ 0#$1%( 52

PAGE 59

was correlated with the width of the carapace for these trails (r = 0.93, p < 0.01, Figure 21). Figure 21. Correlation between crab carapace width and handling time. T h was correlated to carapace width (r = 0.93; p < 0.01). 3.4 Ener gy Content The average ener gy content for M. edulis and L. lima are shown in T able 10. Ashfree dry weight (AFDW) was calculated using conversion factors that were determined by Ricciardi and Bour get (1998). Four trials were performed for each species and the results were averaged. Species J g1 AFDW (CI) Cal mg1 AFDW (CI) A verage # T ( o C) Mytilus edulis 21, 239 (216) 5.06 (0.05) 1.08 Lima lima 21,647 (224) 5.15 (0.05) 1.01 T able 10. A verage ener gy contents of M. edulis and L. lima determined from four bomb calorimetry measurements. 95% confidence intervals are indicated in parentheses. Shell-free dry weights were converted to ash-free dry weights (AFDW) using conversion factors outlined in Ricciardi and Bour get (1998). 53 + %& %( !#$" %#"! &#&" '#!! 20 3,#/3'4,&'.#0&%"#$%&'( 89 3:#839 3731"#;&40/#$1%(

PAGE 60

Figure 22 provides a rough comparison of the handling times and ener gy contents for dif ferent sizes of M. edulis and L. lima The handling times correspond to the same data from section 3.3. The wet weights (shell included) of the bivalves used for the handling time tests were converted to ash-free dry weight (AFDW) using conversion factors from Ricciardi and Bour get (1998). The ash-free dry weights were then used to calculate approximate caloric contents for each of the items. The position of L. lima on the graph demonstrates its high ener gy content and low handling time relative to M. edulis. These data points are only approximations and are not intended to represent absolute values. They are used solely to demonstrate overall variations between M. edulis and L. lima. 54 0 2500 5000 7500 10000 12500 15000 0 30 60 90 120 150 Energy Content (cal) T otal Handling T ime (min) Mytilus edulis Lima lima Figure 22. Comparison of ener gy contents and handling times for M. edulis and L. lima The values correspond to the bivalves used during the handling time tests. Caloric content was calculated from approximate AFDW which was estimated from wet weight, using conversion factors discussed in Ricciardi and Bour get (1998).

PAGE 61

Chapter 4: Discussion The goal of this study was to examine how this generalist predator forages and selects its prey considering the ef fects of prey availability and distribution. The results provided insight into prey selection by O. vulgaris using the framework of optimal foraging theory According to optimal foraging theory or ganisms are expected to forage such a way as to maximize their net ener gy gain. This behavior is commonly associated with the consumption the most profitable prey items, leading to an increased level of specialization and the exclusion of less profitable items from the diet (Pyke et al., 1977; Pyke, 1984; Snellen et al., 2007). In the present study the midden contents left by O. vulgaris were indicative of a generalist diet, although individuals exhibited varying degrees of specialization. They showed greater selectivity for microhabitats, rather than individual species. These results suggested that patch-utilization and predator avoidance played a significant role in the foraging behavior of O. vulgaris This is supported in the literature as octopuses have been found to exhibit behaviors associated with timeminimizing foraging (Leite et al., 2009). Conclusions are limited, however by the nature of the data collected and the limited number of dens that were available for analysis. 4.1 Patch Utilization Octopuses feed opportunistically and are capable of consuming a wide variety of prey They are intelligent predators and are capable of learning the distributions of dif ferent prey and dense food patches (Mather 1991). As soft-bodied or ganisms, their 55

PAGE 62

fitness is more dependent on predation avoidance than choosing individual prey items with maximum ener gy content. Therefore, O. vulgaris would be expected to select prey that are closer to their dens, closely spaced (in dense patches), and in habitats with minimal predation risk. The results from the present study were consistent with these findings. Prey remnants left at the middens suggested that O. vulgaris lacked a narrow search image and, instead, foraged in patches containing preferred prey A comparison of the electivities of the three microhabitats indicated that prey located in grass beds were preferentially returned to the dens, while prey from coral and sand were avoided. Sea grass beds have been known to function as a protective habitat for many benthic or ganisms (Gosselin et al., 1996), as was demonstrated by the diversity of prey found in the SA V quadrats. Sea grass beds also pose a minimal predation risk for O. vulgaris. Eels are frequently described as the major predators of octopuses, particularly in reef environments. At Cayos Cochinos, eels are commonly found in areas with with rock and coral substrate (Hamman, 2010). The apparent avoidance of coral areas may have occurred so as to minimize exposure to predators. By selectively foraging in sea grass beds, O. vulgaris would be minimizing exposure to predators, while also increasing encounter frequency with a diverse array of benthic or ganisms, particularly bivalves. The sandy substrate contained the lowest concentration of potential prey items. Foraging in these microhabitats would therefore reduce the frequency of prey encounters, increase foraging duration, and increase possible exposure to predation. A voidance of these areas is therefore consistent with the time-minimization hypothesis. 56

PAGE 63

4.2 Prey Preference Generalist predators foraging on a diverse diet encounter a wide range of defenses in intertidal areas. Sessile or ganisms often possess morphological defenses, while motile species are capable of predator avoidance or counterattack behaviors (Y amamoto, 2004). T ime-minimization and optimal foraging require the selection of prey that are easiest to capture and have the shortest possible handling time. Analysis of the midden remnants indicated that bivalves were, by far the most preferred prey types. These findings were consistent with a similar study performed in the same location the previous year (Hamman, 2010). Bivalves were the most abundant of the three prey types in the area, although they were not much more abundant than gastropods. A comparison of the electivities of the the prey types indicated that bivalves were preferred, while gastropods and crustaceans were avoided. Bivalves may have been preferred because they are sessile and require shorter handling time, in contrast to gastropods which frequently require drilling (Steer and Semmens, 2003). In addition, many of the more abundant gastropods (such as C. pica and C. gibbosum) were very small and may not have been a viable ener gy source for the octopuses. However the prevalence of bivalve remnants at the middens may have been the result of selective foraging within the SA V rather than an indication of preference. Conclusions regarding prey preference are limited in this study due to the methods used. Midden contents represent only a fraction of an individual s total diet, as they fail to account for prey that are consumed while foraging as well as the dif ferential loss of certain remnants from the den (Ambrose, 1983). Midden contents have been 57

PAGE 64

found to retain bivalve and gastropod shells with greater frequency leading to a possible underestimation of the consumption of crustaceans and soft-bodied or ganisms (Smith, 2003). The removal of gastropod shells from the middens by hermit crabs (Gilchrist, 2003) may have also played a role in the abundance of bivalve remains found. The accuracy of the data could be improved in the future through the combination of multiple sampling methods. Foraging observations with limited researcher interaction would be the most useful and least invasive of the possible sampling methods and could account for any inaccurate representations of prey selection associated with reliance on midden contents. Remote video could also be used to determine when the octopuses leave and return to their dens. V ideo could also be used to calculate the amount of time spent eating prey after returning from foraging. 4.3 Ener gy Content and Handling T ime Remnants of Modiolus modilous were common in the middens. They were determined to have a high electivity as none were found in the quadrats. Lima lima were less common in the middens. They were determined to have a low electivity as they were found frequently in crevices in the major reef structure. Handling observations for the two species (with M. edulis used in place of M. modiolus ) revealed dif ferent strategies and handling times. Mytilus edulis took longer to open than L. lima because they required drilling, while L. lima were simply pulled open. McQuaid (1994) found that pulling mussels cost approximately 1.29 times the ener gy of drilling. However when handling times were accounted for pulling was still more ener getically ef ficient. In the 58

PAGE 65

present study Lima lima were also found to be a better source of ener gy than M. edulis. T ogether these results suggest that prey distribution played a lar ger role in preference than handling time or ener gy content. Lima lima were found exclusively in coral microhabitats, while M. modiolus have been found to occur in mussel beds in sea grass beds. Low electivity for L. lima, despite a relatively high ener getic return and short handling time, further suggests that predator avoidance was more important in shaping the foraging behavior of O. vulgaris than prey preference or ener getic return. The conclusions from this portion of the study are limited. The handling time tests were performed using only one subject, which was approximately half the size of the octopuses that were observed in Cayos Mejor The size of the test subject may have severely af fected the observed handling times. A lar ger subject may have been able to open M. edulis using force, as McQuaid (1994) found that lar ger octopuses were able to open mussels more easily by force. A lar ger subject may also have been able to open L. lima and M. sculptus more easily These ef fects could be better understood in a future study by using more test subjects that were more similar in size to those observed in the field. The ener gy content determinations may also have been inaccurate, due to the preparation method. The samples were dried at 60 o C, while the literature states that the preferred preparation method for bomb calorimetry is freeze-drying (Beukema, 1997). This method of drying may have resulted in the retention of water in the sample, as well as the loss of volatile component. This could lead to significant underestimates, not only of dry weight but also of calorific content. However all of the samples were prepared 59

PAGE 66

using the same method, so the errors in the study should be consistent. T o attain more accurate estimates of ener gy content, a lyophilizer should be used for sample preparation. 4.4 Octopus Dens The dens in the current study were all located in at least one of the three microhabitats. Three of the five dens were in sea grass beds, one was in the sandy substrate, and one was located in a crevice in the major reef structure. The den locations that were examined for this study may have altered the results in a number of ways. Most importantly it is unknown whether the octopus chose dens located near preferred prey or simply brought prey back to the dens that were closest. Dens in SA V may have been selected because they contain patches of preferred prey and minimal predation risk. This explanation is likely as the entire intertidal area was surveyed very carefully and dens appeared to be more common in the SA V However it is also possible that the prevalence of SA V dens had an ef fect on the prey preference results. Mather (1991) found that distance played a major role in the decision to bring prey back to the den. Therefore, the apparent preference for SA V prey (as indicated by the SA V electivity) could have resulted from the higher sampling of SA V middens. 60

PAGE 67

References Aguado Gimnez, F and B. Garca Garca. 2002. Growth and food intake models in Octopus vulgaris Cuvier (1797): influence of body weight, temperature, sex and diet. Aquacultur e International 10, (1): 361-77. Altman, J. and Nixon., M. 1970. Use of the beaks and raduala by Octopus vulgaris in feeding. Journal of Zoology 161, (1): 25-38. Ambrose, R. and Nelson, B. 1983. Predation by Octopus vulgaris in the Mediterranean. Marine Ecology 4, (3): 251-61. Ambrose, R. F 1982. Shelter Utilization by the molluscan cephalopod Octopus bimaculatus Marine Ecology Pr ogr ess Series 7 (1): 67-73. Ambrose, R. F 1983. Midden formation by octopuses: the role of biotic and abiotic factors. M arine Behaviour and Physiology 10, (1): 134-44. Ambrose, R. F 1984. Food preferences, prey availability and the diet of Octopus bimaculatus V errill. Journal of Experimental Marine Biology and Ecology 77, (1): 29-44. Anderson, R., W ood, J. and Mather J. 2008. Octopus vulgaris in the Caribbean is a specializing generalist. Marine Ecology Pr ogr ess Series, 371, (1): 199-202. Arnold, J. and Arnold, K. 1969. Some aspects of hole-boring predation by Octopus vulgaris American Zoologist, 9, (1): 991-6. Baum, W ., Schwendiman, J. and Bell, K. 1999. Choice, contingency discrimination, and foraging theory Journal of the Experimental Analysis of Behavior 71, (1): 355-73. Beukema, J. J. 2008. Caloric values of marine invertebrates with an emphasis on the soft parts of marine bivalves. Oceanography and Marine Biology: an Annual Review 35, (1): 387-414. Burrows, M. T and Hughes, R. N. 1990. V ariation in growth and consumption among individuals and populations of dogwhelks, Nucella lapillus : a link between foraging behaviour and fitness. Journal of Animal Ecology 59, (1): 723-42. 61

PAGE 68

Drossel, B., Higgs, P and McKane, A. 2001. The influence of predator -prey population dynamics on the long-term evolution of food web structure. Journal of Theor etical Biology 208: 91-107. Fiorito, G. and Gherardi, F 1999. Prey-handling behaviour of Octopus vulgaris (Mollusca, Cephalopoda) on Bivalve preys. Behavioural Pr ocesses 46, (1): 75-88. Forsythe, J. and Hanlon, R. 1997. Foraging and associated behavior by Octopus cyanea Gray 1849 on a coral atoll, French Polynesia. Journal of Experimental Marine Biology and Ecology 209, (1): 15-31. Gibbons, M. E., Fer guson, A. M. and Lee, D. R. 2005. Both learning and heritability af fect foraging behaviour of red-backed salamanders, Plethodon ciner eus. Animal Behaviour 69: 721-32. Gilchrist, S. L. 2003. Hermit crab population ecology on a shallow coral reef (Bailey s Cay Roatan, Honduras): octopus predation and hermit crab shell use. Memoirs of Museum V ictoria 60: 35-44. Gosselin, L. A. and Chia, F S. 1996. Prey selection by inexperienced predators: do early juvenile snails maximize net ener gy gains on their first attack? Journal of Experimental Marine Biology and Ecology 199: 45-58. Hamman, E. 2010. Spatial analysis of octopus dens and predation. Ivlev V S. 1961. Experimental ecology of the feeding of fishes. New Haven: Y ale University Press. Katsanevakis, S. and V erriopoulos, G. 2004. Den ecology of Octopus vulgaris Cuvier 1797, on soft sediment: availability and types of shelter Scientia Marina 68: 147-57. Kayes, R. J. 1974. The daily activity pattern of Octopus vulgaris in a natural habitat. Marine Behaviour and Physiology 2: 1337-3439. Leite, T S., Haimovici, M. and Mather J. 2009. Octopus insularis (Octopodidae), evidences of a specialized predator and a time-minimizing hunter Marine Biology 156: 2355-67. MacArthur R. H. and Pianka, E. R. 1966. The optimal use of a patchy environment. The American Naturalist 100: 603-09. Mather J. A. 1991. Foraging, feeding and prey remains in middens of juvenile Octopus vulgaris (Mollusca: Cephalopoda). Journal of Zoology 224: 27-39. 62

PAGE 69

Mather J. A. and O'Dor R. K. 1991. Foraging strategies and predation risk shape the natural history of juvenile Octopus vulgaris Bulletin of Marine Science 49, (1-2): 256-69. McQuaid, C. D. 2002. Feeding behaviour and selection of bivalve prey by Octopus vulgaris Cuvier Journal of Experimental Marine Biology and Ecology 177, (1): 187-202. Nixon, M. and Boyle, P 1982. Hole-drilling in crustaceans by Eledone cirr hosa (Mollusca: Cephalopoda). Journal of Zoology 196: 439-44. Nixon, M. and Maconnachie, E. 1988. Drilling by Octopus vulgaris (Mollusca: Cephalopoda) in the Mediterranean. Journal of Zoology 216 (1): 687-716. Paine, R. T 1971. The Measurement and application of the calorie to ecological problems. Annual Review of Ecology and Systematics, 2: 145-64. Pyke, G. H. 1984. Optimal foraging theory: a critical review Annual Review of Ecology and Systematics, 15: 523-75. Pyke, G. H., Pulliam, H. R. and Charnov E. L. 1977. Optimal foraging: a selective review of theory and tests. The Quarterly Review of Biology 52, (2): 137-54. Ricciardi, A. and Bour get, E. 1998. W eight-to-weight conversion factors for marine benthic macroinvertebrates. Marine Ecology Pr ogr ess Series 163: 245-251. Rosebrock, T 2000. An Analysis of the Co-habitation of Octopus vulgaris Cuvier and Octopus briar eus Robson at Bailey s Cay Roatan, Honduras. Smale, M. and Buchan, P 1981. Biology of Octopus vulgaris of f the east coast of South Africa. Marine Biology 65, (1): 1-12. Smith, C. D. 2003. Diet of Octopus vulgaris in False Bay South Africa. Marine Biology 143: 1 127-33. Snellen, C. L., Hodum, P J. and Fernandez-Juricic, E. 2007. Assessing western gull predation on purple sea urchins in the rocky intertidal using optimal foraging theory Can. J. Zoo l. 85: 221-31. Steer M. A. and Semmens, J. 2003. Pulling or drilling, does size or species matter? An experimental study of prey handling in Octopus dierythraeus Journal of Experimental Marine Biology and Ecology 290: 165-78. 63

PAGE 70

Symondson, W O. C., Sunderland, K. D. and Greenstone, M. H. 2002. Can generalist predators be ef fective biocontrol agents? Annual Review of Entomology 47: 561-94. Thayer G. W ., Schaf f, W E., Angelovic, J. W and LaCroix, M. W 1973. Caloric measurements of some estuarine or ganisms. Fishery Bulletin 71, (1): 289-96. V incent, T L. S., Scheel, D. and Hough, K. R. 1998. Some aspects of diet and foraging behavior of Octopus dofleini (W ulker 1910) in its northernmost range. Marine Ecology Pr ogr ess Series 19: 13-29. Y amamoto, T 2004. Prey composition and prey selectivity of an intertidal generalist predator Muricodrupa fusca (Kuster) (Muricidae). Marine Ecology Pr ogr ess Series 25: 35-49. Y arnall, J. 1969. Aspects of the behaviour of Octopus cyanea Gray Animal Behaviour 17, (4): 747-54. 64

PAGE 71

Appendix A: Field Data Midden Contents Den 1 1 2 3 4 5 6 C D E F G H L N Q T otal 3 4 1 3 3 1 1 1 2 2 1 1 1 1 1 1 2 1 1 6 5 6 6 4 4 Den 2 1 2 3 4 5 6 7 8 9 10 11 12 14 B 2 1 2 C 1 1 1 K 1 2 O 1 2 1 1 P 1 R 1 1 S 1 1 2 T 1 T otal 2 2 2 2 2 2 1 2 3 0 1 3 2 Den 3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 C 2 1 1 1 3 3 2 1 1 1 E 1 1 1 1 F 1 1 2 1 2 1 1 2 2 H 3 1 1 1 2 O 2 1 W 1 X 1 Y 1 2 T otal 3 4 4 3 3 3 4 3 4 4 4 3 4 3 65

PAGE 72

Den 4 1 2 3 4 5 6 7 8 9 10 11 12 13 14 C 1 1 1 E 1 1 1 F 1 1 1 H 3 1 1 3 1 3 2 2 3 4 1 1 2 J 1 M 2 1 U 1 1 AA 1 T otal 3 3 3 4 3 2 4 3 4 3 4 3 2 2 Den 5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 A 1 D 2 1 1 1 2 1 1 1 2 1 E 1 1 1 1 1 1 1 F 1 2 2 3 2 2 1 1 I 1 K 1 1 2 1 N 1 1 V 1 Z 1 2 BB 1 1 CC 1 DD T otal 4 4 4 3 2 3 2 5 4 4 3 5 5 2 66

PAGE 73

Quadrat Coordinates and Contents SA V Side of Dock Coor dinates Contents 1 North 21, 3 A. rigida 2 North 25, 11 empty 3 North 27, 18 A. rigida, L. laevigatum A. rigida, L. laevigatum 4 North 16, 19 A. rigida 5 North 9, 15 Natica livida 6 North 37, 12 empty 7 North 6, 12 A. rigida 8 North 45, 4 empty 9 North 22, 17 L. laevigatum 10 North 13, 13 empty 11 North 8, 2 empty 12 North 4, 8 C. gibbosum 13 North 9, 7 A. rigida 14 North 28, 12 empty 15 North 53, 15 (2) A. rigida 16 North 16, 7 P pholadiformis P pholadiformis 17 North 42, 12 empty 18 North 54, 14 empty 19 North 47, 10 Glycymeris sp. 20 South 10, 13 Ar chidae sp. 21 South 23, 3 empty 22 South 71, 1 empty 23 South 2, 23 Conus midanus Conus midanus 24 South 11, 24 S. gigas 25 South 46, 7 empty 26 South 52, 11 C. gibbosum 27 South 86, 2 H. epheliticus 28 South 31, 21 A. decussata 29 South 8, 15 empty 30 South 34, 2 empty 31 South 55, 6 A. rigida 32 South 99, 1 empty 33 South 27, 5 A. rigida 34 South 6, 2 S. giganteus 35 South 34, 16 M. sculptus 36 South 50, 18 empty 37 South 14, 9 Oliva sp. 38 South 39, 10 C. gibbosum 39 South 56, 18 empty 40 South 5, 7 A. rigida 67

PAGE 74

Sand Side of dock Coor dinates Contents 1 North 29, 2 empty 2 North 12, 10 T ellina lineata 3 North 11, 34 (2) M. sculptus 4 North 7, 1 empty 5 North 18, 10 empty 6 North 19, 3 empty 7 North 19, 7 M. sculptus 8 North 18, 3 A. rigida 9 North 2, 17 H. epheliticus 10 North 8, 13 empty 11 North 5, 35 empty 12 North 11, 13 M. sculptus 13 North 14, 9 empty 14 North 11, 21 empty 15 North 30, 2 M. sculptus 16 South 10, 22 Cerithium atratum Cerithium atratum 17 South 63, 2 empty 18 South 3, 18 empty 19 South 59, 7 empty 20 South 32, 8 empty 21 South 7, 20 Macr ocallista maculata Macr ocallista maculata 22 South 41, 15 empty 23 South 51, 15 M. sculptus 24 South 7, 13 empty 25 South 9, 12 Leucozonia nassa Leucozonia nassa 26 South 12, 16 empty 27 South 40, 4 empty 28 South 4, 21 M. sculptus 29 South 42, 2 M. sculptus 30 South 18, 11 Oliva sp. 31 South 30, 15 empty 32 South 48, 15 M. sculptus 33 South 65, 1 Macr ocallista maculata Macr ocallista maculata 34 South 5, 9 empty 35 South 15, 13 empty 36 South 36, 1 H. epheliticus 37 South 17, 17 empty 38 South 30, 10 empty 39 South 33, 13 empty 40 South 58, 4 Oliva sp. 68

PAGE 75

Coral/rocks Side of Dock Coor dinates Contents 1 North 27, 4 Pinctada sp. 2 North 22, 4 (2) L. lima 3 North 19, 1 Fissur ellidae sp. Fissur ellidae sp. 4 North 8, 4 F lilium 5 North 10, 3 empty 6 North 13, 4 L. lima 7 North 10, 1 L. lima 8 North 9, 5 empty 9 North 26, 4 Pinctada sp. 10 North 17, 3 empty 11 North 30, 2 empty 12 North 14, 3 C. pica 13 North 15, 5 L. lima 14 North 26, 1 C. gibbosum 15 SW 6, 3 (3) Oliva sp 16 SW 8, 5 L. lima 17 SW 6, 5 empty 18 SW 13, 3 empty 19 SW 3, 1 C. pica, C. gibbosum C. pica, C. gibbosum 20 SW 30, 1 C. gibbosum 21 SW 31, 5 empty 22 SW 16, 4 (2) L. lima 23 SW 29, 1 empty 24 SW 20, 2 L. tectum 25 SW 16, 3 empty 26 SW 23, 2 empty 27 SW 33, 5 empty 28 SE 2, 5 T deltoidea 29 SE 2, 3 empty 30 SE 3, 5 Cerithium atratum Cerithium atratum 31 SE 21, 1 (2) C. pica 32 SE 14, 2 empty 33 SE 19, 2 empty 34 SE 15, 3 C. gibbosum 35 SE 7, 1 L. lima 36 SE 10, 1 L. lima 37 SE 27, 2 empty 38 SE 12, 3 C. pica 39 SE 9, 1 empty 40 SE 4, 2 L. lima 69

PAGE 76

Appendix B: Handling T ime Data Mytilus edulis Muss el 1 Muss el 2 Muss el 3 Muss el 4 Muss el 5 Muss el 6 Muss el 7 Muss el 8 Muss el 9 Muss el 10 Muss el 11 Muss el 12 Muss el 13 Heigh t (cm) 5.7 6 6.1 6.2 6.7 6.1 5.8 5.8 6.1 6.3 6 5.9 6.1 Width (cm) 3.2 2.9 2.7 3 3.5 2.8 2.8 3 2.7 3.2 3 2.7 2.8 W eig ht (g) 15.74 16.25 19.35 19.56 24.38 21.57 14.93 16.48 18.36 19.37 22.86 14.88 22.76 T o 97.32 81.19 83.12 87.56 97.47 94.22 94.09 91.18 88.56 90.32 95.28 82.48 87.31 T e 17.09 18.42 19.07 19.54 22.35 19.33 17.11 17.34 19.03 21.02 18.5 18 19.26 Th 114.4 99.61 102.2 107.1 119.8 113.6 111.2 108.5 107.6 111.3 113.8 100.5 106.6 Lima lima FS 1 FS2 FS3 FS4 F5 Height (cm) 6.9 6.6 6.3 6.9 5.3 Width (cm) 5.5 5.4 5 5.6 4.1 W eight (g) 45.82 40.88 37.53 43.29 32.49 T o (min) 5.49 5.16 4.11 6.03 5.19 T e (min) 17.54 17.32 15.17 18.27 12.28 Th (min) 23.03 22.48 19.28 24.3 17.47 Drilled? no no no no no Mithraculus sculptus 1 2 3 4 5 6 7 8 9 10 Height (cm) 1 1 1.6 1.6 1.8 1.7 1.7 2 1.8 1.8 Width (cm) 1.3 1.2 1.7 1.9 2 2 1.9 2.2 1.9 2 T o (min) T e (min) 12.07 11.18 13.43 13.58 15.04 14.22 14.37 15.24 13.07 14.01 Th (min) 12.07 11.18 13.43 13.58 15.04 14.22 14.37 15.24 13.07 14.01 70

PAGE 77

Appendix C: Bomb Calorimetry Data Benzoic Acid T rial 1 ,& ,% -./0 1, 213345 67 89: .45 67 &+#&*' &"#$+% &#*(& !#))+* !#!!&" 71 ,92.453.;7 ,45<7 &"#$+ '! &"#$+ (! &"#$$ )! &"#$$ %&! &"#$$ %"! &"#$( %+! &"#$( &%! &"#+& &*! &(#$ &$! &$#* '!! &$#(+ ''! &+ '(! &+#!) ')! &+#%+ *&! &+#&& *"! &+#&( *+! &+#&$ "%! &+#&$ "*! &+#&+ "$! &+#&+ (!! &+#&+ ('! &+#&+ ((! &+#&+ &" &( &$ &+ &) %$" '"! "&" $!! y = 5.953E-5x + 28.243 y = -0.0001x + 25.781 "%7"9 3 0*9 "#&'#<". ""+#8",1&*+# !&%"#&'#+"1 2'4+

PAGE 78

Benzoic Acid T rial 2 ,& ,% -./0 1, 213345 67 89: .45 67 &)#$'+ &$#&)$ &#**% !#))+ !#!!%) 72 ,92.4537 ,45<7 &$#' '! &$#&) (! &$#&+ )! &$#&+ %&! &$#&( %"! &$#&( %+! &$#&( &%! &$#'* &*! &+#%& &$! &+#++ '!! &)#&& ''! &)#*" '(! &)#"( ')! &)#(* *&! &)#$ *"! &)#$% *+! &)#$* "%! &)#$* "*! &)#$* "$! &)#$" (!! &)#$" ('! &)#$" ((! &)#$" ()! &)#$* $&! &)#$* &$#!! &$#$" &+#"! &)#&" '!#!! &!! *!! (!! +!! y = 1.111E-5x + 29.738 y = -0.0002x + 27.297 "%7"9 3 0*9 "#&'#<".9 ""+#8",1&*+ !&%"#&'#6"1 2'4+

PAGE 79

Benzoic Acid T rial 3 ,& ,% -./0 1, 213345 67 89: .45 67 &(#)%& &*#*&* &#*++ !#)''" !#!!'% 73 ,92.4537 ,45<7 &*#&+ '! &*#*& (! &*#*' )! &*#** %&! &*#** %"! &*#** %+! &*#** &%! &*#** &*! &*#*( &$! &*#++ '!! &"#+* ''! &(#'% '(! &(#"( ')! &(#$% *&! &(#+% *"! &(#++ *+! &(#) "%! &(#)& "*! &(#)* "$! &(#)( (!! &(#)( ('! &(#)( ((! &(#)( ()! &(#)( $&! &(#)( &*#!! &*#$" &"#"! &(#&" &$#!! &!! *!! (!! +!! y = 7.142E-5x + 26.912 y = 9.524E-5x + 24.424

PAGE 80

Benzoic Acid T rial 4 74 &*#!! &*#$" &"#"! &(#&" &$#!! &!! *!! (!! +!! y = 9.524E-5x + 24.167 ,92.4537 ,45<7 &*#%& '! &*#%( (! &*#%+ )! &*#%+ %&! &*#%+ %"! &*#%+ %+! &*#%+ &%! &*#&* &*! &"#%* &$! &"#+ '!! &(#%+ ''! &(#') '(! &(#"% ')! &(#"+ *&! &(#(* *"! &(#(( *+! &(#(+ "%! &(#() "*! &(#$ "$! &(#$% (!! &(#$% ('! &(#$% ((! &(#$% ()! &(#$% $&! &(#$%

PAGE 81

Calibration V alues m QBA BA err or wir e Qwir e wir e err or T2 T1 delta T 0.9984 26430 6.99 0.0025 5858 0.34 28.243 25.781 2.462 0.998 26430 6.99 0.0019 5858 0.34 29.738 27.297 2.441 0.9935 26430 6.99 0.0031 5858 0.34 26.912 24.424 2.488 0.989 26430 6.99 0.0037 5858 0.34 26.686 24.167 2.519 delta T err or E QBA E Qwir e E T E C^2 E C` C 0.01 1.153190 0.05609224 1897.28775 1898.49704 43.5717459 10723.9468 0.01 1.17311778 0.05706152 1961.33025 1962.56043 44.3007949 10810.4343 0.01 1.12921444 0.05492602 1801.89326 1803.0774 42.4626589 10561.2399 0.01 1.10159217 0.05358245 1699.7857 1700.94088 41.2424645 10385.4484 Mean C 10620 J/C Mean C Err or 42.89 J/C Mytilus edulis Calorimetry Data :91/ =>? @4 !"#$%&'() @9: .4A:9B:40 B4; B2CD3 09BE4 ? @4 *"#'(+, @9: .4A:9B:40 B4; B2CD3 09BE4
PAGE 82

M. edulis T rial 1 ,92.49E4H.; BE-3 .2A.: 1 0D: .49E4<./;9D3 !#! &'#)$ '!#! &*#!& (!#! &*#!& )!#! &*#!& %&!#! &*#!' %"!#! &*#!' %+!#! &*#!' &%!#! &*#%% &*!#! &*#*+ &$!#! &*#$% '!!#! &*#+( ''!#! &*#)* '(!#! &*#)) ')!#! &"#!& *&!#! &"#!* *"!#! &"#!( *+!#! &"#!( "%!#! &"#!( "*!#! &"#!(" "$!#! &"#!$ (!!#! &"#!$ ('!#! &"#!+ ((!#! &"#!+ ()!#! &"#!+ $&!#! &"#!+ $"!#! &"#!+ $+!#! &"#!+ 76 &'#!! &'#$" &*#"! &"#&" &(#!! &!!#! *!!#! (!!#! +!!#! y = 25.08 y = 0.0003x + 23.995 "%7"9 3 0*"#&'#8",1&*+ !&%"#&'#+"1 2'4+

PAGE 83

M. edulis T rial 3 ,92.4IE4H.; BE-3 .2A.: 1 0D: .49E4<./;9D3 !#! &'#+" '!#! &'#+" (!#! &'#+( )!#! &'#+( %&!#! &'#+( %"!#! &'#+( %+!#! &'#+( &%!#! &'#)* &*!#! &*#%* &$!#! &*#'+ '!!#! &*#"( ''!#! &*#(' '(!#! &*#$& ')!#! &*#$" *&!#! &*#+! *"!#! &*#+& *+!#! &*#+& "%!#! &*#+' "*!#! &*#+' "$!#! &*#+* (!!#! &*#+* ('!#! &*#+* ((!#! &*#+* ()!#! &*#+* $&!#! &*#+* &'#$ &*#! &*#' &*#( &*#) &!!#! *!!#! (!!#! +!!#! y = 24.84 y = 5.953E-5x + 23.852 "%7"9 3 0*9 "#&'#8",1&*+ !&%"#='#6"1 2'4+ 77

PAGE 84

M. edulis T rial 4 ,92.49E4H.; BE-3 .2A.: 1 0D: .49E4<./;9D3 !#! &&#*$ '!#! &&#"! (!#! &&#"% )!#! &&#"% %&!#! &&#"% %"!#! &&#"& %+!#! &&#"& &%!#! &&#"* &*!#! &&#)* &$!#! &'#&! '!!#! &'#'$ ''!#! &'#*+ '(!#! &'#"* ')!#! &'#(! *&!#! &'#(& *"!#! &'#(& *+!#! &'#(* "%!#! &'#(*" "*!#! &'#(" "$!#! &'#(" (!!#! &'#(" ('!#! &'#(( ((!#! &'#(( ()!#! &'#(( $&!#! &'#(( &&#! &&#" &'#! &'#" &*#! &!!#! *!!#! (!!#! +!!#! y = 7.143E-5x + 23.611 y = 0.0002x + 22.485 "%7"9 3 0*9 "#&'#8",1&*+ !&%"#&'#6"1 2'4+ 78

PAGE 85

M. edulis T rial 5 ,92.49E4H.; BE-3 .2A.: 1 0D: .4IE4<./;9D3 !#! &'#"+ '!#! &'#") (!#! &'#") )!#! &'#(! %&!#! &'#(! %"!#! &'#(! %+!#! &'#(! &%!#! &'#(& &*!#! &*#!! &$!#! &*#&+ '!!#! &*#*& ''!#! &*#"* '(!#! &*#(! ')!#! &*#(* *&!#! &*#(( *"!#! &*#($ *+!#! &*#(+ "%!#! &*#() "*!#! &*#$! "$!#! &*#$! (!!#! &*#$! ('!#! &*#$! ((!#! &*#$!! ()!#! &*#$! $&!#! &*#$! &'#! &'#" &*#! &*#" &"#! &!!#! *!!#! (!!#! +!!#! y = 5.951E-6x + 24.697 y = 0.0001x + 23.585 "%7"9 3 0*9 "#&'#8",1&*+ !&%"#&'#6"1 2'4+ 79

PAGE 86

Lima lima T rial 1 ,92.49E4H.; BE-3 .2A.: 1 0D: .49E4<./39D3 !#! &'#(" '!#! &'#("" (!#! &'#("" )!#! &'#(( %&!#! &'#($ %"!#! &'#($ %+!#! &'#($ &%!#! &'#($ &*!#! &'#($ &$!#! &'#)* '!!#! &*#%+ ''!#! &*#'$ '(!#! &*#*+ ')!#! &*#"( *&!#! &*#(! *"!#! &*#(* *+!#! &*#(( "%!#! &*#($ "*!#! &*#(+ "$!#! &*#() (!!#! &*#$! ('!#! &*#$! ((!#! &*#$! ()!#! &*#$! $&!#! &*#$! $"!#! &*#$! &'#! &'#" &*#! &*#" &"#! &!!#! *!!#! (!!#! +!!#! y = 24.7 y = 0.0001x + 23.65 "%7"9 3 0*9 "#&'#8",+&*+ !&%"#='#6"1 2'4+ 80

PAGE 87

L. lima T rial 2 !#! &'#"+ '!#! &'#(& (!#! &'#(& )!#! &'#(& %&!#! &'#(& %"!#! &'#(& %+!#! &'#(& &%!#! &'#$& &*!#! &*#!* &$!#! &*#'& '!!#! &*#*+ ''!#! &*#"( '(!#! &*#(! ')!#! &*#(' *&!#! &*#(* *"!#! &*#(( *+!#! &*#(+ "%!#! &*#() "*!#! &*#$! "$!#! &*#$! (!!#! &*#$! ('!#! &*#$! ((!#! &*#$! ()!#! &*#$! &'#! &'#" &*#! &*#" &"#! %$"#! '"!#! "&"#! $!!#! y = 24.7 y = 0.0001x + 23.601 "%7"9 3 0*9 "#&'#8",+&*+ !&%"#&'#6"1 2'4+ 81

PAGE 88

L. lima T rial 3 ,92.49E4H.; BE-3 .2A.: 1 0D: .4IE4<./39D3 !#! &'#* '!#! &'#"& (!#! &'#"& )!#! &'#"& %&!#! &'#"& %"!#! &'#"& %+!#! &'#"& &%!#! &'#"+ &*!#! &'#)( &$!#! &*#& '!!#! &*#'* ''!#! &*#*% '(!#! &*#*( ')!#! &*#" *&!#! &*#"& *"!#! &*#"* *+!#! &*#"* "%!#! &*#"* "*!#! &*#"* "$!#! &*#"* (!!#! &*#"* &'#! &'#" &*#! &*#" &"#! %"!#! '!!#! *"!#! (!!#! y = 24.54 y = 23.52 "%7"9 3 0*9 "#&'#8",+&*+ !&%"#&'#6"1 2'4+ 82

PAGE 89

L. lima T rial 4 ,92.49E4H.; BE-3 .2A.: 1 0D: .49E4<./39D3 !#! &'#*& '!#! &'#*( (!#! &'#*$ )!#! &'#*+ %&!#! &'#*+ %"!#! &'#*+ %+!#! &'#"( &%!#! &'#)" &*!#! &*#%+ &$!#! &*#'% '!!#! &*#'$ ''!#! &*#*& '(!#! &*#*" ')!#! &*#*$ *&!#! &*#*+ *"!#! &*#*+ *+!#! &*#*+ "%!#! &*#*+ "*!#! &*#*+ "$!#! &*#*+ &'#! &'#" &*#! &*#" &"#! %"!#! '!!#! *"!#! (!!#! y = 24.48 y = 0.0006x + 23.428 "%7"9 3 0*9 "#&'#8",+&*+ !&%"#&'#6"1 2'4+ 83


ERROR LOADING HTML FROM SOURCE (http://ncf.sobek.ufl.edu//design/skins/UFDC/html/footer_item.html)