Ectotherms rely on external heat to attain target body temperatures which can vary based on the animal’s current physiological activity. Many ectotherms become thermophilic (“heat-loving”) during crucial physiological processes like digestion and reproduction, behaviorally thermoregulating to increase body temperature higher than what they otherwise prefer. However, there is a positive relationship between body temperature and water loss that dictates increasing body temperature typically elicits an increase in water loss. Animals that inhabit areas where water is at least seasonally limited (e.g., deserts, wet-dry forests) may face a tradeoff between prioritizing behavioral thermophily to optimize physiological processes versus prioritizing water balance and potentially sacrificing some aspect of total performance capability.It is thus far unknown how reduced water availability and subsequent dehydration may influence thermophily in ectotherms. I hypothesized that behaviorally thermoregulating ectotherms exhibit thermophily during critical physiological events, and the extent to which thermophily is expressed is influenced by the animal’s hydric state. Using Children’s pythons (Antaresia childreni), I investigated the effects of dehydration on behavioral thermophily during digestion and reproduction. I found that dehydration caused a suppression in digestion-associated thermophily, where dehydrated snakes returned to pre-feeding body temperature sooner than they did when they were hydrated. In contrast, water deprivation at different reproductive stages had no effect on thermophily despite leading to a significant increase in the female’s plasma osmolality. ii Additionally, the timing of water deprivation during reproduction had differing effects on plasma osmolality and circulating triglyceride, total protein, and corticosterone concentrations. My research provides evidence of the sensitive and complex dynamic between body temperature, water balance, and physiological processes. At a time when many dry ecosystems are becoming hotter and drier, my investigation of dehydration and its influence on thermal dynamics and physiological metrics provides insight into cryptic effects on the vital processes of digestion and reproduction.

Pollinator populations globally have declined at concerning rates in recent years, which is problematic given that roughly a third of all food production depends on them. Managed honey bee colony losses in particular have alarmed beekeepers and scientists, especially in the United States. Widespread agrochemical use has been implicated as one of the major causes of these colony losses. While the lethal effects of agrochemicals often receive the most attention, sublethal effects can occur at lower doses and can substantially weaken colonies over time. Impaired associative learning ability is a sublethal effect of a number of agrochemicals, and is particularly concerning, as it may hinder the abilities of bees to forage for food or find their way back to the colony. Here, I focus on the fungicide Pristine® (active ingredients: 25.2% boscalid, 12.8% pyraclostrobin), which is sprayed on honey bee-pollinated crops during bloom and is known to poison bee mitochondria at ppm levels. First, I show that Pristine® impairs performance on an associative learning assay in the laboratory. Next, I show that Pristine® alters carbohydrate absorption in honey bees, providing a possible mechanism underlying this impaired learning performance. Finally, I demonstrate that Pristine® interacts with high temperatures to induce homing failure in exposed bees. My results raise concerns that this common fungicide may not be safe for pollinators and will be relevant to policymakers as they make decisions surrounding the regulation of fungicide use in agriculture.

Biogeography places the geographical distribution of biodiversity in an evolutionary context. Ants (Hymenoptera: Formicidae), being a group of ubiquitous, ecologically dominant, and diverse insects, are useful model systems to understand the evolutionary origins and mechanisms of biogeographical patterns across spatial scales. On a global scale, ants have been used to test hypotheses on the origin and maintenance of the remarkably consistent latitudinal diversity gradient where biodiversity peaks in the equatorial tropics and decreases towards the poles. Additionally, ants have been used to posit and test theories of island biogeography such as the mechanisms of the species-area relationship, being the increase of biodiversity with cumulative land area. However, there are still unanswered questions about ant biogeography such as how specialized life histories contribute to their global biogeographical patterns. Furthermore, there remain island systems in the world’s biodiversity hotspots that harbor much less ant species than predicted by the species-area relationship, which potentially suggests a place ripe for discovery. In this dissertation, I use natural history, taxonomic, geographic, and phylogenetic data to study ant biodiversity and biogeography across spatial scales. First, I study the global biodiversity and biogeography of a specialized set of symbiotic interactions between ant species, here referred to as myrmecosymbioses, with an emphasis on social parasitism where one species exploits the parental care behavior and social colony environment of another species. In addition to characterizing a new myrmecosymbiosis, I use a global biogeographic and phylogenetic dataset to show that ant social parasitism is distributed along an inverse latitudinal diversity gradient where species richness and independent evolutionary origins of social parasitism peak within the northern hemisphere where the least free-living ant diversity exists. Second, I study the unexplored ant fauna of the Vanuatuan archipelago in the South Pacific. Using approximately 10,000 Vanuatuan ant specimens coupled with phylogenomics, I fill in a historical knowledge gap of South Pacific ant biogeography and demonstrate that the Vanuatuan ant fauna is a novel biodiversity hotspot. With these studies, I provide insights into how specialized life histories and unique island biotas shape the global distribution of biodiversity in different ways, especially in the ants.

Aggregation is a fundamental principle of animal behavior; it is especially significant tohighly social species, like ants. Ants typically aggregate their workers and brood in a central nest, potentially due to advantages in colony defense and regulation of the environment. In many ant species, when a colony must abandon its nest, it can effectively reach consensus on a new home. Ants of the genus Temnothorax have become a model for this collective decision-making process, and for decentralized cognition more broadly. Previous studies examine emigration by well-aggregated colonies, but can these ants also reach consensus when the colony has been scattered? Such scattering may readily occur in nature if the nest is disturbed by natural or man- made disasters. In this exploratory study, Temnothorax rugatulus colonies were randomly scattered in an arena and presented with a binary equal choice of nest sites. Findings concluded that the colonies were able to re-coalesce, however consensus is more difficult than for aggregated colonies and involved an additional primary phase of multiple temporary aggregations eventually yielding to reunification. The maximum percent of colony utilization for these aggregates was reached within the first hour, after which point, consensus tended to rise as aggregation decreased. Small, but frequent, aggregates formed within the first twenty minutes and remained and dissolved to the nest by varying processes. Each colony included a clump containing the queen, with the majority of aggregates containing at least one brood item. These findings provide additional insight to house-hunting experiments in more naturally challenging circumstances, as well as aggregation within Temnothorax colonies.

Dominance behavior can regulate a division of labor in a group, such as that between reproductive and non-reproductive individuals. Manipulations of insect societies in a controlled environment can reveal how dominance behavior is regulated. Here, I examined how morphological caste, fecundity, group size, and age influence the expression of dominance behavior using the ponerine ant Harpegnathos saltator. All H. saltator females have the ability to reproduce. Only those with a queen morphology that enables dispersal, however, show putative sex pheromones. In contrast, those with a worker morphology normally express dominance behavior. To evaluate how worker-like dominance behavior and associated traits could be expressed in queens, I removed the wings from alate gynes, those with a queen morphology who had not yet mated or left the nest, making them dealate. Compared to gynes with attached wings, dealates frequently performed dominance behavior. In addition, only the dealates demonstrated worker-like ovarian activity in the presence of reproductive individuals, whereas gynes with wings produced sex pheromones exclusively. Therefore, the attachment of wings determines a gyne’s expression of worker-like dominance behavior and physiology. When the queen dies, workers establish a reproductive hierarchy among themselves by performing a combination of dominance behaviors. To understand how reproductive status depends on these interactions as well as a worker’s age, I measured the frequency of dominance behaviors in groups of different size composed of young and old workers. The number of workers who expressed dominance scaled with the size of the group, but younger ones were more likely to express dominance behavior and eventually become reproductive. Therefore, the predisposition of age integrates with a self-organized process to form this reproductive hierarchy. A social insect’s fecundity and fertility signal depends on social context because fecundity increases with colony size. To evaluate how a socially dependent signal regulates dominance behavior, I manipulated a reproductive worker’s social context. Reproductive workers with reduced fecundity and a less prominent fertility signal expressed more dominance behavior than those with a stronger fertility signal and higher fecundity. Therefore, dominance behavior reinforces rank to compensate for a weak signal, indicating how social context can feed back to influence the maintenance of dominance. Mechanisms that regulate H. saltator’s reproductive hierarchy can inform how the reproductive division of labor is regulated in other groups of animals.

Understanding why animals form social groups is a fundamental aim of sociobiology. To date, the field has been dominated by studies of kin groups, which have emphasized indirect fitness benefits as key drivers of grouping among relatives. Nevertheless, many animal groups are comprised of unrelated individuals. These cases provide unique opportunities to illuminate drivers of social evolution beyond indirect fitness, especially ecological factors. This dissertation combines behavioral, physiological, and ecological approaches to explore the conditions that favor group formation among non-kin, using as a model the facultatively social carpenter bee, Xylocopa sonorina. Using behavioral and genetic techniques, I found that nestmates in this species are often unrelated, and that non-kin groups form following extensive inter-nest migration.Group living may arise as a strategy to mitigate constraints on available breeding space. To test the hypothesis that nest construction is prohibitively costly for carpenter bees, I measured metabolic rates of excavating bees and used imaging techniques to quantify nest volumes. From these measurements, I found that nest construction is highly energetically costly, and that bees who inherit nests through social queuing experience substantial energetic savings. These costs are exacerbated by limitations on the reuse of existing nests. Using repeated CT scans of nesting logs, I examined changes in nest architecture over time and found that repeatedly inherited tunnels become indefensible to intruders, and are subsequently abandoned. Together, these factors underlie intense competition over available breeding space. The imaging analysis of nesting logs additionally revealed strong seasonal effects on social strategy, with social nesting dominating during winter. To test the hypothesis that winter social nesting arises from intrinsic physiological advantages of grouping, I experimentally manipulated social strategy in overwintering bees. I found that social bees conserve heat and body mass better than solitary bees, suggesting fitness benefits to grouping in cold, resource-scarce conditions. Together, these results suggest that grouping in X. sonorina arises from dynamic strategies to maximize direct fitness in response to harsh and/or competitive conditions. These studies provide empirical insights into the ecological conditions that favor non-kin grouping, and emphasize the importance of ecology in shaping sociality at its evolutionary origins.

The living world is replete with easily observed structural adaptations (e.g. teeth, claws, and stingers), but behavioral adaptations are no less impressive. Conspecific aggression can be defined as any harmful action directed by one animal at another of the same species. Because it is a potentially risky and costly behavior, aggression should be elicited only under optimal conditions. In honeybees, nestmate recognition is considered the driving factor determining whether colony guards will aggress against other honeybees attempting to gain entry to the colony. Models and empirical research support the conclusion that nestmate recognition should be favored over direct kin recognition. Thus, bees tend to use environmentally mediated cues associated with their colonies (e.g. colony odors) to recognize nestmates. The framework of nestmate recognition suggests that non-nestmates should always be aggressed against while nestmates should always be accepted. However, aggression towards nestmates and acceptance of non-nestmates are seen in a wide variety of eusocial insects, including honeybees. These are typically classified as rejection errors and acceptance errors, respectively. As such, they can be explained using signal detection theory and optimal acceptance threshold models, which postulate that recognition errors are inevitable if there is overlap in the cues used to distinguish “desirables” (fitness-enhancing) from “undesirables” (fitness-decrementing) conspecifics. In the context of social insects desirables are presumed to be nestmates and undesirables are presumed to be non-nestmates. I propose that honeybees may make more refined decisions concerning what conspecifics are desirable and undesirable, accounting for at least some of the phenomena previously reported as recognition errors. Some “errors” may be the result of guard bees responding to cues associated with threats and benefits beyond nestmate identity. I show that less threatening neighbors receive less aggression than highly threatening strangers. I show that well-fed colonies exhibit less aggression and that bees from well-fed colonies receive less aggression. I provide evidence that honeybees may decrease aggression towards nestmates and non-nestmate not involved in robbing while increasing aggression towards non-nestmate from a robber colony. Lastly, I show that pollen bearing foragers, regardless of nestmate identity, receive little to no aggression compared to non-pollen bearing foragers.

As technological advancements in silicon, sensors, and actuation continue, the development of robotic swarms is shifting from the domain of science fiction to reality. Many swarm applications, such as environmental monitoring, precision agriculture, disaster response, and lunar prospecting, will require controlling numerous robots with limited capabilities and information to redistribute among multiple states, such as spatial locations or tasks. A scalable control approach is to program the robots with stochastic control policies such that the robot population in each state evolves according to a mean-field model, which is independent of the number and identities of the robots. Using this model, the control policies can be designed to stabilize the swarm to the target distribution. To avoid the need to reprogram the robots for different target distributions, the robot control policies can be defined to depend only on the presence of a “leader” agent, whose control policy is designed to guide the swarm to a particular distribution. This dissertation presents a novel deep reinforcement learning (deep RL) approach to designing control policies that redistribute a swarm as quickly as possible over a strongly connected graph, according to a mean-field model in the form of the discrete-time Kolmogorov forward equation. In the leader-based strategies, the leader determines its next action based on its observations of robot populations and shepherds the swarm over the graph by probabilistically repelling nearby robots. The scalability of this approach with the swarm size is demonstrated with leader control policies that are designed using two tabular Temporal-Difference learning algorithms, trained on a discretization of the swarm distribution. To improve the scalability of the approach with robot population and graph size, control policies for both leader-based and leaderless strategies are designed using an actor-critic deep RL method that is trained on the swarm distribution predicted by the mean-field model. In the leaderless strategy, the robots’ control policies depend only on their local measurements of nearby robot populations. The control approaches are validated for different graph and swarm sizes in numerical simulations, 3D robot simulations, and experiments on a multi-robot testbed.

Despite theoretical models predicting that signals should only evolve if they convey honest information, dishonest signals may persist. Interestingly, crustaceans have been crucial in furthering biologists understanding of how and why dishonest signals persist; because many crustaceans wield claws that function as dishonest signals. For example, male fiddler crabs have claws that grow to large sizes but are incapable of inflicting severe damage to opponents, thus acting as a dishonest signal of their strength. Although initial work suggested that dishonest signaling was common throughout Crustacea, biologists understanding of the generality of dishonest communication is lacking. To resolve these issues, I combined morphological, behavioral, and comparative studies to investigate whether crayfish engage in dishonest communication. First, I found that regenerated claws in virile crayfish (Faxonius virilis) produce 40% weaker pinching forces compared to original claws. These results suggest that claw regeneration in crayfish may be the functional mechanism that produces dishonest signals. Second, I conducted two studies that investigated what traits determine dominance in staged contests; one on intrasexual contests in both male and female F. virilis, and a second between intra- and interspecific contests between male F. virilis and male red swamp crayfish (Procambarus clarkii). In both studies, I did not find support the hypothesis that large but weak claws function as dishonest signals; because claw size did not predict the outcome of signaling interactions and claw strength did not predict the outcome of physical fights. Lastly, I conducted a comparative study between six species of crayfish — three stream-dwelling species that use their claws as weapons and signals, and three burrowing species that use their claws for excavating burrows. Despite all six species possessing claws that unreliably predicted claw strength, I found no support for the hypothesis that their claws function as dishonest signals in any of these species. Thus, my dissertation results suggest that despite having claws that unreliably predict their strength, such unreliable signals do not equate to dishonest signals. Altogether, my work highlights the importance of collecting behavioral data in studies of dishonest communication and stresses the importance of separating unreliable signals from dishonest signals.

One of the most pronounced issues affecting the management of fisheries today is bycatch, or the unintentional capture of non-target species of marine life. Bycatch has proven to be detrimental for many species, including marine megafauna and pelagic fishes. One method of reducing bycatch is illuminated gillnets, which involves utilizing the differences in biological visual capabilities and behaviors between species of bycatch and target fish catch. To date, all studies conducted on the effects of net illumination on bycatch and target fish catch have been conducted at night. In this study, the effects of net illumination on bycatch, target fish catch, and market value during both night and day periods at Baja California Sur, Mexico were compared. It was found that i) net illumination is effective (p < 0.05) at reducing bycatch of finfish during the day and at night, ii) net illumination at night is more effective (p < 0.05) at reducing bycatch for elasmobranchs, Humboldt squid, and aggregate bycatch than during the day, iii) time of day did not have an effect (p > 0.05) on sea turtle bycatch, and iv) net illumination did not significantly (p > 0.05)affect target catch or market value at night or during the day. These results suggest that net illumination may be an effective strategy for reducing finfish bycatch in fisheries that operate during the day or across 24 h periods, and is especially effective for reducing elasmobranch, Humboldt squid, and total bycatch biomass at night.