This dissertation investigated the ‘disco’ clam Ctenoides ales (Limidae), which is the only bivalve to have a behaviorally-mediated flashing display. Topics covered include (i) mechanisms, ultrastructure and movement that produce the flashing, (ii) the fitness value (function) of the flashing, (iii) the clams’ sensory abilities and vision, and (iv) the clams’ ecology, distribution and habitat. The flashing occurs on the clams’ mantle lip. Electron microscopy revealed two distinct tissue sides; one highly scattering side that contains dense aggregations of spheres composed of silica (white), and one highly absorbing side that does not (red). High-speed video confirmed that the two sides alternate rapidly, creating the appearance of flashing. Optical modeling suggested that the sphere’s diameter is nearly optimal for scattering light, especially at shorter wavelengths, which predominate in the ocean. This simple mechanism produces a striking optical effect. Three potential hypotheses for the fitness value of the flashing were investigated; conspecific attraction, prey luring, and/or predator deterrence. The lack of movement toward other C. ales when given visual cues in behavioral trials, as well as the clams’ inability to resolve flashing in other C. ales suggested conspecific attraction was not the function of the flashing. The lack of significant differences in prey abundance in experiments testing flashing versus non-flashing clams suggested prey luring was also not the function of the flashing. Predator deterrence is considered a possible function of the flashing due to (i) sulfur presence in the clam’s tissues, (ii) behavioral responses by predators during feeding trials as well as tissue preferences that suggest potential distastefulness, and (iii) the clams’ increase in flash rate when exposed to predators in the lab and the field. The presence of photosensitive pigments (rhodopsin, tubulin and retinochrome) was suggested from immunohistochemistry results in the ~40 eyes of C. ales in collaboration with A. Nahm-Kingston (University of Maryland Baltimore County). Transmission electron microscopy done in collaboration with R. Dubielzig, L. Teixeira, and C. Schobert (University of Iowa Veterinary School) confirmed that the morphology of the eye was inconsistent with image formation or the ability to resolve flashing in conspecifics. Therefore, the visual capability of C. ales is most likely used for predator detection. The clams’ distribution, depth, main habitats, projected sex ratios, movement based on size (sex) and sphere properties with depth were identified through SCUBA research at four field sites. The clams are found throughout the Indo-Pacific, from depths 3m to ≥50m. Their movement doesn’t vary with sex, and they exhibit a clumped distribution seemingly skewed towards males, which compounds their vulnerability in aquaria collections and highlights the need to determine their conservation status.

Autotomy is the shedding of a body part as a means of secondary defense against a predator that has already made contact with the organism. This defense mechanism has been widely studied in a few model taxa, specifically lizards, a few groups of arthropods, and some echinoderms. All of these model organisms have a hard endo- or exo-skeleton surrounding the autotomized body part. There are several animals that are capable of autotomizing a limb but do not exhibit the same biological trends that these model organisms have in common. As a result, the mechanisms that underlie autotomy in the hard-bodied animals may not apply for soft bodied organisms. A behavioral ecology approach was used to study arm autotomy in the octopus, Abdopus aculeatus. Investigations concentrated on understanding the mechanistic underpinnings and adaptive value of autotomy in this soft-bodied animal.

A. aculeatus was observed in the field on Mactan Island, Philippines in the dry and wet seasons, and compared with populations previously studied in Indonesia. Frequency of encountered arm loss and habitat characteristics were recorded and compared between sites, seasons, sex, and arm position. The incidence of autotomy was highest in Indonesia; nearly half of the sampled octopuses were found missing or regenerating arms from the base of the body. Most of the octopuses were male. There was no significant difference in the incidence of autotomy between Indonesia and the Philippines or between the wet and dry seasons in the Philippines. Individuals lost between one and five arms, one arm being the most common. The eight arms from each individual were not equally lost. The anterior pairs of arms were more susceptible to loss due to their function as exploratory arms. The male third right arm used for reproduction was lost less frequently.

Attempts to fully describe the behaviors associated with autotomy besides limb detachment have not been made for any organism. Multiple arm autotomy events were observed in A. aculeatus to generate an inventory of behaviors that are characteristic of autotomy. Sequential analysis of these behaviors was conducted and found to follow a particular order during autotomy. Some of the classified behaviors include other secondary defense mechanisms, such as jetting and inking.

To understand the mechanism for arm loss at the tissue level, histological sections and force measurements were collected from autotomized arms to characterize the autotomy zone. Individuals were also decerebrated to establish the relative importance of the central nervous system in the occurrence of autotomy. Histology revealed a zone of weakness with no specialized adaptations for loss between proximal suckers four and eight. An average tensile force of 0.645 ± 0.145 MPa was required for autotomy to occur. Eight out of ten decerebrated individuals did not autotomize an arm post-decerebration suggesting that the central nervous system plays a role in arm autotomy.

Lastly, the costs and benefits of arm autotomy to A. aculeatus were assessed by studying the effects of arm loss in whole body and autotomized arm locomotion. Individuals with fewer arms were not found to exhibit any negative impact to their crawling or jetting behaviors. The autotomized arms post detachment performed complex behaviors and suctioned to various surfaces for prolonged periods of time suggesting potential adaptations in the arm to distract a predator while the octopus escapes.

Overall, the octopus provided a good model system for studying the mechanisms and adaptive value of autotomy in a soft-bodied animal. There multitude of arms were easy to manipulate and induce autotomy in the lab. The relatively high frequency of individuals missing arms in the wild suggest that autotomy plays an important defense role for A. aculeatus and needs further characterization to identify the underlying cause. Efforts to study autotomy in other non-model organisms will broaden and enhance our knowledge on what seems like an extreme, but common defense mechanism.

The deep sea is the largest habitat on Earth and holds the majority of its' animal biomass. Due to the limitations of observing, capturing and studying these diverse and numerous organisms, little is known about them. The majority of deep-sea species are known only from net-caught specimens, therefore behavioral ecology and functional morphology were assumed. The advent of human operated vehicles (HOVs) and remotely operated vehicles (ROVs) have allowed scientists to make one-of-a-kind observations and test hypotheses about deep-sea organismal biology.

Cephalopods are large, soft-bodied molluscs whose defenses center on crypsis. Individuals can rapidly change coloration (for background matching, mimicry, and disruptive coloration), skin texture, body postures, locomotion, and release ink to avoid recognition as prey or escape when camouflage fails. Squids, octopuses, and cuttlefishes rely on these visual defenses in shallow-water environments, but deep-sea cephalopods were thought to perform only a limited number of these behaviors because of their extremely low light surroundings.

The Monterey Bay Aquarium Research Institutes' ROVs were used to determine whether or not deep-sea squids are limited in their behaviors compared to their shallow-water relatives. First, eighteen species of deep-sea squids were observed to release ink when encountered by an ROV (Chapter 2). Ink release was observed from the surface to below 1800 m and included six types of ink release, some of which have not been observed from shallow-water cephalopods.

Ink release could serve as a visual defense by blocking the cephalopods' silhouette or bioluminescence stimulated by the squids' movement, or causing predators to confuse ink pseudomorphs with potential prey. It could also, or instead, have a chemical function by having distasteful or noxious qualities, blocking predator olfactory senses, stimulating feeding behaviors, or functioning as an attractant so that predators attack the ink. To investigate potential chemical functions, ink collected from deep-sea and shallow-water cephalopods was analyzed using liquid chromatography - mass spectroscopy (Chapter 3). The goal was to compare ink chemical composition between species to see if deep-sea species or a few particular species had vastly different chemical compositions that may warrant further investigations.

An ethogram - a catalogue of every behavior a species performs - of the deep-sea squid Octopoteuthis deletron Young 1972 was produced from over eight hours of ROV video footage of 76 individuals (Chapter 4). Octopoteuthis deletron is capable of numerous color, posture, and locomotor changes. Individuals can also change patterns of bioluminescence produced by their arm-tip photophores, leading to the first description of bioluminescence as a cephalopod behavioral component. Additional work was performed with O. deletron to determine if this species can autotomize arms (Chapter 5). ROV observations, laboratory and in situ experimentation, and histological sectioning were conducted. Octopoteuthis deletron can autotomize any of the eight arms at numerous points along the length. This species is the first squid described to autotomize arms, has the uncommon capability of economy of autotomy, and is one of very few animals that can perform attack autotomy.

Our understanding of cephalopod behavioral ecology has been greatly advanced by the research presented here. The hypotheses that deep-sea squids do not release ink and that their defensive behaviors are very limited were both falsified. Clearly there is much additional work to be done on these and other deep-sea animals to understand the ecology of the deep sea.

Understanding the reciprocal interactions between animal behavior and other inter-related systems such as physiology, morphology, ecology, and evolution has been called a “Grand Challenge” for organismal biology. Behavior offers two unique contributions to our understanding of organism response to environmental change: 1) behavior is a rapid and reversible response, and 2) organisms can directly influence their surrounding environment, and thus the stimuli their physiological and morphological systems are exposed to, by “choosing” their environment.

My dissertation utilizes the temperate octopus Octopus bimaculatus to understand the reciprocal interactions between behavior and ecology in a human-altered landscape. I use a combination of animal population surveys, dietary analysis, mathematical models, and acoustic telemetry to understand the feedback loops been behavior and ecology. Octopuses play a key predatory role in shaping communities that is unmatched by any other invertebrate. Additionally, their “live fast, die young” life history strategy means their populations respond quickly and dramatically to changes in the environment, which suggests they could be an indicator of ecosystem change. Specifically, I addressed three fundamental questions about octopus behavior and ecology on Santa Catalina Island, CA: 1) What are key environmental variables influence octopus environmental choice?; 2) Where and when do octopuses move?; and 3) Does octopus diet reflect differences in predatory behavior within a human-altered habitat?

The first question addressed a key gap in current knowledge of the environmental variables that influence octopus abundance and distribution in a rocky reef ecosystem by combining intensive population surveys with imperfect detection modeling. Binomial mixture modeling is used when the likelihood of false non-detection is high, and is ideal for studying highly cryptic species like octopuses. Abundance and detection probability was modeled with site and survey-level covariates such as abiotic variables, octopus predator and prey abundances, and habitat structural characteristics. No single abundance covariate explained the data for either year, but detection in 2013 was best modeled with Julian date. Model-averaged estimates of abundance had high ranges of possibility, and all correlations with estimated octopus abundance and environmental variables had weak support. These results highlight the high variability in patterns of octopus abundance and the necessity for integration of multiple environmental factors to elucidate drivers of octopus abundance and small-scale distribution.

The second question quantified movement and habitat use of individual octopuses. This study pioneered acoustic telemetry research on small octopus species and was the first of its kind in the continental United States. I collected and tagged O. bimaculatus and recorded their position over the course of approximate 2 weeks and measured movement continuously over one day for each animal. I found that O. bimaculatus is highly mobile and, combined with my survey data, concluded that they do not stay in a single den for more than a few days. Additionally, movement distances and diurnal movement patterns are highly variable, suggesting a behavioral response to avoid predation.

The third question aimed to identify the diet of O. bimaculatus using Stable Isotope Analysis (SIA) and assess if octopus diet differed between Marine Protected Area (MPA) sites and non-protected sites. SIA allows for the indirect assessment of diet because the isotopic composition of prey is incorporated with reliable fidelity into the predators’ tissues. I analyzed the carbon and nitrogen stable isotope ratios in octopuses and their prey collected inside and outside the Blue Cavern MPA. I found that octopus diet differs between MPA and non-MPA sites and octopuses located within the MPA have a more varied diet than those located outside. This difference indicates a change in the predator-prey relationships with the establishment of the MPA and suggests larger changes in the community structure.

Together, these findings yield novel insights into what influences the populations and behavior of octopuses in a rocky reef environment, and lay the groundwork for directly testing how octopuses will respond to and influence the changes in their surrounding community.

ABSTRACT

The Evolution of Social Monogamy and Biparental Care in Stomatopod Crustaceans

By Mary Louisa Wright

Doctor of Philosophy in Integrative Biology

University of California, Berkeley

Professor Roy L. Caldwell, Chair

Although social monogamy and biparental care have been extensively studied in birds, mammals, and fish, the evolutionary origins and maintenance of these phenomena are not well-understood, particularly in invertebrate taxa. The evolution of social monogamy is of interest because current theory predicts that both males and females will usually gain more fitness from mating with multiple partners. Furthermore, biparental care should only occur when males and females both gain more fitness benefits from providing parental care than from investing time and energy into mate searching. Given these expectations, under what environmental and social conditions will social monogamy and biparental care arise and do the same conditions maintain monogamy and biparental care on an evolutionary time scale? Long-term social monogamy, which occurs when a male and female pair for longer than a single breeding cycle, has been reported in eight genera of Lysiosquilloid stomatopods. Furthermore, the Lysiosquilloidea also contains the only marine crustacean genus (Pullosquilla) in which biparental care has been systematically studied. This dissertation examines the evolutionary maintenance and origins of both biparental care and long-term social monogamy in the Lysiosquilloidea, using experimental manipulations, ecological surveys, and comparative, phylogenetically-based methods.

Chapter 1: The maintenance of biparental care

I examined the fitness costs and benefits of biparental care in the stomatopod Pullosquilla thomassini using an experimental manipulation of the number and sex of care providers. In the absence of any care, egg clutch survival and growth decreased. However, neither the number, nor the sex of the care providers had a significant effect on changes in egg clutch mass. Parental care treatment did not affect ovary size, the total number of eggs in a clutch, or egg size. Thus, while parental care increases production of offspring, uniparental care by either sex is sufficient to achieve this goal. Males providing uniparental care lost more weight than those providing biparental care or no care. This may lead to sexual conflict over female desertion. These results suggest that biparental care is not evolutionarily maintained in P. thomassini by increasing the number of offspring hatching in an egg clutch. Instead, I hypothesize that biparental care may be evolutionarily maintained in P. thomassini by increasing the rate of egg clutch production and facilitating double-clutching.

Chapter 2: The effects of environmental and demographic variation on pairing behaviors

I examined the effects of environmental and demographic variation on pairing behaviors, egg clutch production, and burrow distribution in two sympatric stomatopod crustaceans, P. litoralis and P. thomassini . These small (<16mm) stomatopods are found as heterosexual pairs in U-shaped burrows in coral patch reef ecosystems. Coral patch reef ecosystems consist of coral heads of varying sizes separated by sand flats; in this heterogeneous environment, the coral heads host high levels of invertebrate and fish diversity and abundance in comparison to the sand flats. I hypothesized that environmental heterogeneity in the coral back reef environment would affect pairing behaviors, egg clutch production, and burrow distribution of Pullosquilla species due to gradients in food abundance or predation by fish. I tested this hypothesis on demographic and environmental data collected from a survey of Pullosquilla species collected in a patch reef in Moorea, French Polynesia. My findings indicate that proximity to coral heads is an important factor in structuring the demography and pairing behaviors of P. litoralis, but not its congener, P. thomassini. The directionality of the relationships between proximity to coral heads and several demographic traits suggests that gradients in fish predation are responsible for these patterns in P. litoralis. This suggests that selective pressures from fish predation may play an important role in the maintenance of pairing behaviors in P. litoralis. Determining the causes of differences in demographic patterns and pairing behaviors of P. litoralis and P. thomassini may yield a better understanding of the evolutionary maintenance of social monogamy in stomatopod crustaceans.

Chapter 3: The evolutionary origins of long-term social monogamy in stomatopods

I examined two hypotheses for the evolutionary origins of long-term social monogamy in stomatopod crustaceans using comparative, phylogenetically-based methods. One of the most commonly posited explanations for the evolution of social monogamy is that biparental care is required to successfully raise offspring. A prediction of this hypothesis is that biparental care should evolve in a clade before or at the same time as social monogamy. I tested this prediction by reconstructing ancestral states of social monogamy and biparental care on a Maximum Likelihood tree of 66 stomatopod species and found that long-term social monogamy evolved before biparental care in the Lysiosquilloid stomatopods. This indicates that a need for biparental care did not lead to the origin of social monogamy in this clade. Based on my finding that predation influences pairing behaviors in P. litoralis (Chapter 2) and the observation that all known socially monogamous stomatopods are sit-and-wait predators, I propose an alternative hypothesis for the origin of social monogamy in stomatopods. Sit-and-wait predation evolves as a strategy to maximize energy intake while minimizing predation risk when a lineage lives in an environment where both prey items and potential predators are abundant. I therefore hypothesized that a suite of behaviors, including burrowing, sit-and-wait predation, and social monogamy, that allowed stomatopods to escape high levels of predation evolved in the Lysiosquilloidea. I tested two predictions of this hypothesis: 1) social monogamy should evolve more often in burrow-dwellers living in soft-bottom substrates and 2) the evolution of long-term social monogamy should be correlated with the evolution of sit-and-wait predation. I tested this hypothesis on a Maximum Likelihood phylogeny of 66 stomatopod species using ancestral state reconstructions and Pagel's (1994) test of correlated trait evolution and found that burrowing, sit-and-wait predation, and social monogamy evolved sequentially in the Stomatopoda. Long-term social monogamy may have evolved as a way of further maximizing the fitness benefits of the sedentary lifestyle associated with sit-and-wait predation. This novel evolutionary route to long-term social monogamy may be associated with the shallow benthic marine environments that most Lysiosquilloids inhabit.

Conclusions

The findings of my thesis emphasize the importance of studying a diversity of taxa and environments when trying to understand the evolution of important behavioral traits. For example, it is often assumed that when biparental care is widespread in a species, it increases the viability of the current brood of offspring. However, in P. thomassini there is no evidence that biparental care increases either the survival or development of embryos (Chapter 1). Additionally, the evolution of social monogamy in many animals is attributed to a need for biparental care. The Lysiosquilloid stomatopods appear to provide a counter-example in which social monogamy likely facilitated the evolution of biparental care and other form of paternal effort (Chapter 3). Instead, my findings support the hypothesis that long-term social monogamy and sit-and-wait predation may have evolved to decrease mortality from predation during foraging and mate searching. The role of the risk of predation during mate searching in the evolution of social monogamy has received relatively little attention in the large body of literature on mating system evolution, but it appears that predation plays an important role in determining pairing behaviors and burrowing distributions in P. litoralis (Chapter 2). Taken as a whole, these findings provide compelling justification for studying the evolution of behaviors in a wide diversity of taxa.

It was long thought that dark, nutrient depleted environments, such as the deep sea and subterranean caves, were largely devoid of life and supported low-density assemblages of endemic fauna. The discovery of hydrothermal vents in the 1970s and their subsequent study have revolutionized ecological thinking about lightless, low oxygen ecosystems. Symbiosis between chemosynthetic microbes and their eukaryote hosts has since been demonstrated to fuel a variety of marine foodwebs in extreme environments. We also now know these systems to be highly productive, exhibiting greater macrofaunal biomass than areas devoid of chemosynthetic influx. This dissertation research has revealed chemosynthetic bacteria and crustaceans symbionts that drive another extreme ecosystem - underwater anchialine caves - in which a landlocked, discrete marine layer rests beneath one or more isolated layers of brackish or freshwater.

Most anchialine caves contain low invertebrate abundances, yet some have inexplicably large biomasses of shrimp and remipedes, a rare crustacean class endemic to caves. Since only anecdotal evidence of remipede feeding in the field has been published and little ecological information about anchialine organisms is known, in situ studies are crucial to understanding whether there are direct or indirect links between chemosynthetic input into anchialine foodwebs and remipede abundance. This dissertation research integrates studies of geochemistry, microbiology, invertebrate food web dynamics, and behavior in these extreme ecosystems. Findings include the following: 1) highly stratified anchialine caves support communities of freeliving chemosynthetic bacteria, 2) Typhlatya pearsei, an anchialine shrimp, harbors chemosynthetic endosymbiotic bacteria, 3) the remipede, S. tulumensis, may also harbor chemosynthetic symbionts ectobiotically 4) food webs in the cave are complex and vary spatially, driving community structure changes 5) remipedes likely use chemical gradients and specialized chemosensory receptors to navigate in their lightless environment.

This combined body of research provides the first systematic study into anchialine ecology and has found them to be rich in taxa for comparative study with other chemosynthetic systems. For example, the finding of the first chemosynthetic endosymbiosis in a crustacean, T. pearsei, and the first chemosynthetic epibiosis in a remipede, S. tulumensis, allows for a variety of comparative coevolutionary studies. Typhlatya, a genus of troglobytic shrimp, are found in anchialine systems across the Caribbean and Bermuda and likely all contain chemoendosymbiotic bacteria. As such this discovery provides a springboard for comparative research into the transmission and evolution of chemosynthetic symbiotic associations. The most interesting outcomes, however, may result from comparing ecological and ecosystem function patterns described in caves to those known for deepwater as well as molluscan and annelid systems. Furthermore, anchialine systems, a primary source for agricultural, domestic, and industrial water use, are especially susceptible to climate change. Therefore, not only are studies of ecosystem function such as this imperative to conserving endemic cave fauna and their suitable habitat, but they may also help in the preservation of clean water caches for future generations.

The stomatopod crustaceans, or mantis shrimps, are marine predators that stalk or ambush prey and that have complex intraspecific communication behavior. Their active lifestyles, means of predation, and intricate displays all require unusual flexibility in interacting with the world around them, implying a well-developed ability to learn. Stomatopods have highly evolved sensory systems, including some of the most specialized visual systems known for any animal group. Some species have been demonstrated to learn how to recognize and use novel, artificial burrows, while others are known to learn how to identify novel prey species and handle them for effective predation. Stomatopods learn the identities of individual competitors and mates, using both chemical and visual cues. Furthermore, stomatopods can be trained for psychophysical examination of their sensory abilities, including demonstration of color and polarization vision. These flexible and intelligent invertebrates continue to be attractive subjects for basic research on learning in animals with relatively simple nervous systems.