Journal of Experimental Biology

Sensory processing of environmental stimuli during locomotion is critical for the successful execution of goal-directed behaviors and navigating around obstacles. The outcome of these sensorimotor processes can be challenged by head movements that perturb the sensory coordinate frames directing behaviors. In the case of visually-guided behaviors, visual gaze stabilization results from the integrated activity of the vestibuloocular reflex and motor efference copy originating within circuits driving locomotor behavior. A recent videographic study showed that echolocating bats exhibit inflight head stabilization during a target identification and landing task, though compensatory timing of the bats sonar signals was not reported. In the present investigation we tested hypotheses that head stabilization is more broadly implemented during epochs of exploratory flight, and is temporally associated with emitted sonar signals, which would optimize acoustic gaze. This was achieved by measuring head and body kinematics with motion sensors secured to the head and body of free-flying Egyptian fruit bats. These devices were integrated with ultrasonic microphones to record the bats sonar emissions and elucidate their temporal association with periods of head stabilization. Head accelerations in the Earth-vertical axis were asymmetric with respect to wing downbeat and upbeat relative to body accelerations. This indicated that inflight head and body accelerations were uncoupled, outcomes consistent with the implementation of head movements that limit vertical acceleration during wing downbeat. Furthermore, sonar emissions during stable flight occurred most often during wing downbeat and head stabilization, supporting the conclusion that head stabilization behavior optimized sonar gaze and environmental interrogation via echolocation. Summary statementDirect measurements of head and body kinematics from affixed motion sensors revealed head stabilization behaviors during exploratory flights in bats. Most sonar emissions were temporally correlated with this behavior, thereby contributing to the optimization of acoustic gaze.

Force-length (F-L) and force-velocity (F-V) properties characterize skeletal muscles intrinsic properties under controlled conditions, and it is thought that these properties can inform and predict in vivo muscle function. Here, we map dynamic in vivo operating range and mechanical function during walking and running, to the measured in situ F-L and F-V characteristics of guinea fowl (Numida meleagris) lateral gastrocnemius (LG), a primary ankle extensor. We use in vivo patterns of muscle tendon force, fascicle length, and activation to test the hypothesis that muscle fascicles operate at optimal lengths and velocities to maximize force or power production during walking and running. Our findings only partly support our hypothesis: in vivo LG velocities are consistent with optimizing power during work production, and economy of force at higher loads. However, LG does not operate at lengths on the force plateau ({+/-}5% Fmax) during force production. LG length was near L0 at the time of EMG onset but shortened rapidly such that force development during stance occurred almost entirely on the ascending limb of the F-L curve, at shorter than optimal lengths. These data suggest that muscle fascicles shorten across optimal lengths in late swing, to optimize the potential for rapid force development near the swing-stance transition. This may provide resistance against unexpected perturbations that require rapid force development at foot contact. We also found evidence of passive force rise (in absence of EMG activity) in late swing, at lengths where passive force is zero in situ, suggesting that dynamic history dependent and viscoelastic effects may contribute to in vivo force development. Direct comparison of in vivo work loops and physiological operating ranges to traditional measures of F-L and F-V properties suggests the need for new approaches to characterize dynamic muscle properties in controlled conditions that more closely resemble in vivo dynamics.

In nearly all animals, light sensing mediated by opsin visual pigments is important for survival and reproduction. Eyeless light-sensing systems, though vital for many animals, have received relatively less attention than forms with charismatic or complex eyes. Despite no single light sensing organ, the sea anemone Nematostella vectensis, has 29 opsin genes and multiple light-mediated behaviors throughout development and reproduction, suggesting a deceptively complex light-sensing system. To characterize one aspect of this light-sensing system, we analyzed larval swimming behavior at high wavelength resolution across the ultraviolet and visual spectrum. N. vectensis larvae respond to light at least from 315 to 650 nm, which is a broad sensitivity range even compared to many animals with complex eyes. Swimming in the water column is induced by ultraviolet (UV) and violet wavelengths until 420 nm. Between 420 and 430 nm a behavioral switch occurs where at wavelengths longer than 430 nm, larvae respond to light by swimming down. Swimming down toward the substrate is distinct from light avoidance, as animals do not exhibit positive or negative phototaxis at any wavelength tested. At wavelengths longer than 575 nm, animals in the water column take increasingly longer to respond and this behavior is more variable until 650 nm where larval response is no different from the dark, suggesting these longer wavelengths lie outside of their sensitivity range. Larval swimming is the only motile stage in the life history of N. vectensis, and increased swimming activity in the water column could lead to greater dispersal range in potentially damaging shallow environments with short-wavelength light exposure. Longer wavelength environments may indicate more suitable substrates for metamorphosis into the polyp stage, where the individual will remain for the rest of its life. Future work will test whether this robust behavior is mediated by multiple opsins.

The critical oxygen partial pressure (Pcrit) is most commonly defined as the oxygen partial pressure below which an animals standard metabolic rate can no longer be maintained. It is widely interpreted as measure of hypoxia tolerance, which influences a species aerobic scope and, thus, constrains biogeography. However, both the physiology underlying that interpretation and the methodology used to determine Pcrit remain topics of active debate. The debate remains unresolved in part because Pcrit, as defined above, is a purely descriptive metric that lacks a clear mechanistic basis. Here we redefine Pcrit as the PO2 at which physiological oxygen supply is maximized and refer to these values, thus determined, as Pcrit-. The oxygen supply capacity () is a species- and temperature-specific coefficient that describes the slope of the relationship between the maximum achievable metabolic rate and PO2. This is easily determined using respirometry and provides a precise and robust estimate of the minimum oxygen pressure required to sustain any metabolic rate. To determine , it is not necessary for an individual animal to maintain a consistent metabolic rate throughout a trial (i.e. regulation) nor for the metabolic rate to show a clear break-point at low PO2. We show that Pcrit- can be determined at any metabolic rate as long as the organisms oxygen supply machinery reaches its maximum capacity at some point during the trial. We reanalyze published representative Pcrit trials for 40 species across five phyla, as well as complete datasets from six additional species, five of which have not previously been published. Values determined using the Pcrit- method are strongly correlated with Pcrit values reported in the literature. Advantages of Pcrit- include: 1) Pcrit- is directly measured without the need for complex statistics that hinder measurement and interpretation; 2) it makes clear that Pcrit is a measure of oxygen supply, which does not necessarily reflect hypoxia tolerance; 3) it alleviates many of the methodological constraints inherent in existing methods; 4) it provides a means of predicting the maximum metabolic rate achievable at any PO2, 5) Pcrit- sheds light on the temperature- and size-dependence of oxygen supply and metabolic rate and 6) Pcrit- can be determined with greater precision than traditional Pcrit.

Although mitochondrial respiration is believed to explain a substantial part of the variation in whole-animal basal (BMR) or resting metabolic rate (RMR), few studies have addressed the relationship between organismal and cellular metabolism and how this may vary in environments where individual demands for energy differ. We investigated the relationship between whole-individual metabolic rate, measured in temperatures ranging thermoneutrality to far below thermoneutrality, and mitochondrial respiration of intact or permeabilized blood cells in two separate studies on wild great tits (Parus major L.). Our results show that, in permeabilized cells, there are significant positive relationships between BMR or RMR and several mitochondrial traits, including phosphorylating respiration rate through both complexes I and II (i.e., OXPHOS respiration). However, surprisingly, the LEAK respiration (i.e., basal respiration that mainly counteract for proton leakage) was not related to BMR or RMR. When measurements were performed using intact blood cells, BMR was positively related to ROUTINE respiration (i.e., mitochondrial respiration on endogenous substrates) in one of the two studies, but no other mitochondrial traits could explain variation in BMR or RMR in any thermal environment. These studies seem to show that the level of activation of mitochondrial metabolism as well as the permeabilization status of blood cells play a primary role on the extent to which blood metabolism might explain variations in the whole-individual metabolic rate.

Both male and female ticks have a strong innate drive to find and blood-feed on hosts. Carbon dioxide (CO2) is considered a critical behavioral activator and attractant for ticks and an essential sensory cue to find hosts. Yet, how CO2 activates and promotes host-seeking in ticks is poorly understood. We studied CO2 responses in the black-legged tick Ixodes scapularis, the primary vector for Lyme disease in North America. Adult males and females were exposed to 1, 2, 4, or 8% CO2, and changes in walking behavior and foreleg movement were analyzed. We find that CO2 is a potent stimulant for adult Ixodes scapularis, even at lower concentrations (1%). Behavioral reactions depend on the animals state: Walking ticks increase their walking speed, while stationary ticks start to wave their forelegs and begin to quest - both behaviors resembling aspects of host-seeking. Furthermore, Ixodes scapularis has no clear concentration preference and is not tuned more robust to breath-like CO2 concentrations ([~]4%) than to the other concentrations tested. As soon as the CO2 level is above a certain threshold, Ixodes scapularis react, indicating that CO2 acts as a behavioral activator and can be used as a long-distance cue to detect approaching hosts. Moreover, we provide convincing evidence that the foreleg Hallers organ is not necessary for CO2 detection. Even with disabled or amputated Hallers organ, Ixodes scapularis respond robustly to CO2, signifying that there must be CO2-sensitive structures important for tick host-seeking that have not yet been identified.

Withdrawal StatementThe authors have withdrawn this manuscript because of the need for further data analysis. Therefore, the authors do not wish this work to be cited as reference for the project. If you have any questions, please contact the corresponding author.

Despite breathing water using their tidally ventilated rectal gills, dragonfly nymphs show a surprising ability to maintain oxygen (O2) extraction from the water during hypoxia. However, an increase in convective O2 transfer is insufficient to sustain aerobic demands by itself, which suggests that diffusive mechanisms must also be involved. This study examines the contributions of changing the O2 partial pressure gradient (PO2) and/or O2 conductance across the rectal gill in maintaining O2 extraction efficiency (OEE) of dragonfly nymphs during hypoxia. Data were collected using the same custom-designed respiro-spirometer described in a previous study with the addition of an implanted O2 sensor to measure hemolymph PO2. Results show that the implantation of the O2 sensor does not affect the respiratory and ventilatory response of nymphs to hypoxia. Hemolymph PO2 fell from 6.3 {+/-} 1.6 kPa at normoxia to 2.5 {+/-} 0.6 kPa at 16.0 kPa, which resulted in the PO2 diffusion gradient remaining statistically constant at these two water PO2s (17.5 {+/-} 1.7 and 15.4 {+/-} 0.7 kPa during normoxia and 16.0 kPa respectively). Beyond 16.0 kPa, a progressive reduction in hemolymph PO2 was unable to sustain the diffusion gradient. Mathematical modeling revealed that while the addition of hemolymph PO2 in tandem with ventilation frequency was able to elevate OEE during 16.0 kPa to that of normoxia, both were still insufficient during severe hypoxia and required an increase in O2 conductance. Estimating the change in whole-gill conductance showed that nymphs are indeed increasing their conductance as the water becomes hypoxic, demonstrating a reliance on both diffusion gradient and O2 conductance to enhance diffusive O2 transfer in conjunction with convective mechanisms to maintain O2 extraction during hypoxia.

Research on birds suggests that extreme weather events during development may have long-lasting consequences on form and function. The underlying cellular mechanisms mediating such phenotypic effects are poorly studied. We raised Japanese quail in warm (30{degrees}C) or cold (10{degrees}C) temperatures from hatching until adulthood, and then measured mitochondrial metabolism in intact blood cells at representative normothermic body temperature (41{degrees}C) and a hyperthermic temperature (45{degrees}C) that quail commonly attain when heat stressed. To investigate whether any developmental effects were reversible, half of the cold- and warm-acclimated birds were assigned to a common garden (20{degrees}C) 3 weeks before the measurements. Across groups, hyperthermia was associated with increased proton leak, but decreases in both phosphorylating respiration (where ATP is produced) and working capacity of the mitochondria. Cold-acclimated birds were more strongly affected by heat stress: the increase in proton leak was 1.6-fold higher, and the decrease in phosphorylating capacity during endogenous respiration was 1.7-fold greater, compared to warm-acclimated birds. These differences did not remain in the common-garden birds. Our study suggests that developmental cold acclimation is traded off against heat tolerance at the level of cellular metabolism, with implications for our understanding of avian responses to climate change.

The ability of ectothermic animals to live in different thermal environments is closely associated with their capacity to maintain physiological homeostasis across diurnal and seasonal temperature fluctuations. For chill-susceptible insects, such as Drosophila, cold tolerance is tightly linked to ion and water homeostasis obtained through a regulated balance of active and passive transport. Active transport at low temperature requires a constant delivery of ATP and we therefore hypothesize that cold-adapted Drosophila are characterized by superior mitochondrial capacity at low temperature relative cold-sensitive species. To address this, we investigated how experimental temperatures 19-1 {degrees}C affected mitochondrial substrate oxidation in flight muscle of seven tropical and temperate Drosophila species that represent a broad spectrum of cold tolerance. Mitochondrial oxygen consumption rates measured using a substrate-uncoupler-inhibitor-titration protocol showed that cooling generally reduced oxygen consumption of all steps of the electron transport system across species. Complex I is the primary consumer of oxygen at benign temperatures, but low temperature decreases complex I respiration to a much greater extent in cold-sensitive species than in cold-adapted species. Accordingly, cold-induced reduction of complex I correlates strongly with CTmin (the temperature inducing cold coma). The relative contribution of alternative substrates, proline, succinate and glycerol-3-phosphate increased as temperature decreased, particularly in the cold-sensitive species. At present it is unclear whether the oxidation of alternative substrates can be used to offset the effects of the temperature-sensitive complex I, and the potential functional consequences of such a substrate switch are discussed. Summary statementMitochondrial oxygen consumption decreases at low temperature, particularly in cold-sensitive Drosophila species, which turn to oxidation of alternative substrates as complex I-supported respiration is impaired.

Thermal variation has complex effects on organisms and they deal with it by combining behavioral and physiological thermal tolerance. However, we still do not understand well how these two types of traits relate to body condition (e.g. size, hydration) and environmental variables (e.g. relative humidity), some of which are typical aspects of thermal tolerance experiments (warming rates, start temperature). We explored these interactions using a set of experiments that sequentially measure behavioral (Voluntary Thermal Maxima) and physiological thermal tolerance (Critical Thermal Maxima) for individuals of Atta sexdens rubropilosa (Forel, 1908). We found non-linear effects of body size on behavioral thermal tolerance and refuted the traditional hypothesis that body size increases ants physiological thermal tolerance. Hydration state and humidity had complex effects on behavioral and physiological tolerance. However, both tolerance measures increased with heating rates and start temperature. Our work helps understanding how an ectotherm integrates stimuli affecting its thermal tolerance to decide which temperatures to avoid. We discuss implications for the ecology of ants, their labor division, and for their susceptibility to climate warming and drought. Summary StatementHere we show how internal (body size, hydration level) and external factors (heating rate, relative humidity) affect leaf-cutting ants behavioral and physiological responses to temperature rises.

Visual animal communication, whether to the same species or to other species, is largely conducted through dynamic and colourful signals. For a signal to be effective, the signaller must capture and retain the attention of the receiver. Signal efficacy is also dependent on the sensory limitations of the receiver. However, most signalling studies consider movement and colour separately, resulting in a partial understanding of the signal in question. We explored the structure and function of predator-prey signalling in the jumping spider-tephritid fly system, where the prey performs a wing waving display that deters an attack from the predator. Using a custom-built spider retinal tracker combined with visual modelling, and behavioural assays, we studied the effect of fly wing movement and colour on the jumping spiders visual system. We show that jumping spiders track their prey less effectively during wing display and this can be attributed to a series of fluctuations in chromatic and achromatic contrasts arising from the wing movements. These results suggest that displaying flies deter spider attacks by manipulating the movement biases of the spiders visual system. Our results emphasise the importance of receiver attention on the evolution of interspecific communication.

Understanding the cellular and physiological mechanisms underlying muscle remodeling requires model systems that allow rapid, reliable, and quantitative assessment of muscle state. The cricket Gryllus lineaticeps naturally undergoes non-pathological striated muscle breakdown (histolysis), making it a promising system for studying this process. However, current assessments of muscle state are largely qualitative, subjective, and poorly standardized across experiments. Here, we developed and validated a continuous, quantitative muscle color metric to objectively capture histolysis progression and functional changes in muscle. We show that this metric robustly tracks variation in muscle color across remodeling stages, including the challenging fully transparent stage, and strongly predicts protein content, mitochondrial abundance, and iron content in a muscle- and trait-specific manner. The reproducibility of these relationships across independent datasets demonstrates the generality and robustness of this approach. By providing a rapid, objective, and biologically informative proxy of muscle state, this framework not only advances the utility of G. lineaticeps as a model for muscle remodeling but also offers a strategy for exploring the cellular dynamics underlying age-related muscle diseases and disorders, addressing an increasing public health concern in aging populations.

Photosynthetic animals produce oxygen internally, providing an ideal lens for studying how oxygen dynamics influence thermal sensitivity. The sea slug, Elysia viridis, can retain functional chloroplasts from its food alga Bryopsis plumosa for months, but retention is limited when fed Chaetomorpha sp., limiting potential oxygenic benefits. We fed slugs each alga and exposed them to 17{degrees}C (their current yearly maximum temperature) and 22{degrees}C (the increase predicted for 2100), to examine plasticity in thermal tolerance and changes in oxygen uptake when fed and starving. We also examined slugs under increased illumination to examine a potential tradeoff between increased oxygen production, and a faster rate of chloroplast degradation. Following exposure to these conditions, we performed ramping trials, subjecting them to acute thermal stress to determine their thermal tolerance. We also measured oxygen uptake before and after ramping. We observed increases in thermal tolerance for specimens exposed to 22{degrees}C, indicating they acclimated to temperatures higher than they naturally experience. Fed slugs exhibited higher rates of oxygen consumption before exposure to acute thermal stress, and suppressed their oxygen uptake more after it, than starved slugs. Under higher light, slugs exhibited improved thermal tolerance, possibly because increased oxygen production alleviated host oxygen limitation. Accordingly, this advantage disappeared later in the starvation period when photosynthesis ceased due to chloroplast digestion. In conclusion, E. viridis can suppress metabolism to cope with heat waves, however, starvation influences a slugs thermal tolerance and oxygen uptake, so continuous access to algal food for chloroplast retention is critical when facing thermal stress. Summary StatementOxygen has been implicated in determining an ectotherms thermal sensitivity. Examining photosynthetic (and therefore oxygen-producing) sea slugs under various conditions helps elucidate how oxygen and other factors impact thermal tolerance.

Variation in tissue-specific metabolism between species and among individuals is thought to be adaptively important; however, understanding this evolutionary relationship requires reliably measuring this trait in many individuals. In most higher organisms, tissue specificity is important because different organs (heart, brain, liver, muscle) have unique ecologically adaptive roles. Current technology and methodology for measuring tissue-specific metabolism is costly and limited by throughput capacity and efficiency. Presented here is the design for a flexible and cost-effective high-throughput micro-respirometer (HTMR) optimized to measure small biological samples. To verify precision and accuracy, substrate specific metabolism was measured in heart ventricles isolated from a small teleost, Fundulus heteroclitus, and in yeast (Saccharomyces cerevisiae). Within the system, results were reproducible between chambers and over time with both teleost hearts and yeast. Additionally, metabolic rates and allometric scaling relationships in Fundulus agree with previously published data measured with lower-throughput equipment. This design reduces cost, but still provides an accurate measure of metabolism in small biological samples. This will allow for high-throughput measurement of tissue metabolism that can enhance understanding of the adaptive importance of complex metabolic traits.

Bumblebees navigate complex environments where collisions with obstacles can impair flight performance. While bees possess innate collision avoidance reflexes, they may benefit from learning to identify and avoid high-risk collision areas using environmental cues. However, the temporal dynamics of how bees form associations between visual cues and collision experiences remain unclear. We investigated whether bumblebees associate visual cues perceived before or after collision events with a movement direction. Individual foragers were trained to navigate through a flight tunnel containing a transparent barrier and an LED panel that switched colours immediately after the bees first collision. Following training, we tested bees responses to each colour cue without the barrier. Bees demonstrated significant preferences for avoiding the previously blocked side when presented with either the pre-collision colour (81% correct responses) or post-collision colour (71% correct responses), while showing no preference when no colour cue was presented. Individual analysis revealed that 61% of bees responded to both cue types, while others showed selective responses to specific temporal windows. These results demonstrate that bees can form associations with visual cues encountered in different time windows relative to negative experiences, revealing temporal flexibility in associative learning that contributes to successful navigation in cluttered environments.

One of the most stressful factors for insects is increasing temperature because of the risk of potentially fatal dehydration linked to their small size. We used respirometry to study the effect of both temperature and body mass on water loss and metabolic rate in individual workers of the polymorphic ant species Messor barbarus. As expected, we found that large ants exposed to increasing temperatures have a lower rate of water loss than small ants and that their mass-specific metabolic rate increases more slowly. However, counterintuitively, the measure of worker sensitivity to changes in temperature, as assessed by the instantaneous Q10 value (i.e., the rate of change across 10{degrees}C temperature intervals), shows that large ants are more sensitive than small ants to changes in temperature in terms of both water loss and metabolic rate. Such differential thermal sensitivity allows to make testable predictions on the temporal distribution of foraging activity among workers of different sizes in polymorphic ant species, as well as how these species may alter their colony demographics in response to rising temperatures.

Feedforward models are mechanisms enabling an agent to predict the sensory outcomes of its actions. It can be implemented in the nervous system in the form of efference copies, which are copies of motor signals that are subtracted from the sensory stimulation actually detected, literally cancelling the perceptual outcome of the predicted action. In insects, efference copies are known to modulate optic flow detection for flight control in fruit flies. Much less is known, however, about possible feedforward control in other insects. Here we investigated whether feedforward control occurs in the detection of horizontal optic flow in walking ants, and how the latter is integrated to modulate their locomotion. We mounted Cataglyphis velox ants within a virtual reality set-up, allowing us to manipulate the relationship between the ants movements and the optic flow it perceives. Results show that ants do compute a prediction error by making the difference between the expected optic flow according to their own movements and the one it perceived. Interestingly, this prediction does not control locomotion directly, but modulates the ants intrinsic oscillator, which produces continuous alternations between right and left turns. Whats more, we show that the prediction also involves proprioceptive feedback, and is additionally modulated by the visual structure of the surrounding panorama in a functional way. Finally, prediction errors stemming from both eyes are integrated before modulating the oscillator, providing redundancy and robustness to the system. Overall, our study reveals that ants compute robust predictions of the optic flow they should receive using a distributed mechanism integrating feedforwards, feedbacks as well as innate information about the structure of the world, that control their locomotion through oscillations.

The majority of insects can acclimate to changes in their thermal environment and counteract temperature effects on neuromuscular function. At the critical thermal minimum a spreading depolarization (SD) event silences central neurons, but the temperature at which this event occurs can be altered through acclimation. SD is triggered by an inability to maintain ion homeostasis in the extracellular space in the brain and is characterized by a rapid surge in extracellular K+ concentration, implicating ion pump and channel function. Here, we focused on the role of the Na+/K+-ATPase specifically in lowering the SD temperature in cold-acclimated D. melanogaster. After first confirming cold acclimation altered SD onset, we investigated the dependency of the SD event on Na+/K+-ATPase activity by injecting an inhibitor, ouabain, into the head of the flies to induce SD over a range of temperatures. Latency to SD followed the pattern of a thermal performance curve, but cold acclimation resulted in a left-shift of the curve to an extent similar to its effect on the SD temperature. With Na+/K+-ATPase activity assays and immunoblots, we found that cold-acclimated flies have ion pumps that are less sensitive to temperature, but do not differ in their overall abundance in the brain. Combined, these findings suggest a key role for plasticity in Na+/K+-ATPase thermal sensitivity in maintaining central nervous system function in the cold, and more broadly highlight that a single ion pump can be an important determinant of whether insects can respond to their environment to remain active at low temperatures.

The gills are hypothesized to play a key role in early vertebrate evolution by replacing the skin as the primary site of gas exchange. In this scenario, water flow across the gills used for suspension feeding in ancestral deuterostomes was coopted for breathing in stem vertebrates to facilitate the evolution of larger, active fishes. This hypothesis is supported by a stem-vertebrate origin for structures that increase gill capacity for breathing. However, these structures might have instead enhanced an already dominant capacity at the gills of invertebrate deuterostomes rather than mark a shift from the skin. To test this, we characterized gill function for gas exchange in the suspension-feeding hemichordate acorn worm Protoglossus graveolens. We measured oxygen uptake and ammonia excretion in whole worms and worm halves with or without gills at 10{degrees}C and during an acute thermal challenge at 20{degrees}C to maximize gill recruitment. Gills did not enhance oxygen uptake or ammonia excretion, suggesting they are not a primary site of gas exchange. This is the first test of gill function for gas exchange in a suspension-feeding invertebrate deuterostome, and it provides essential support for the long-hypothesized vertebrate origin of breathing at gills and its role in early vertebrate evolution.