HORMONAL AND PHEROMONAL EFFECTS OF 20-HYDROXYECDYSONE IN THE AMERICAN LOBSTER, HOMARUS AMERICANUS

20-hydroxyecdysone (20E), the active principle of the molting hormone in the American lobster has significant effects on the animals’ agonistic behavior and has been shown to influence the outcome of agonistic interactions. Animals injected with 20E are significantly more aggressive than saline-injected control animals, and premolt animals, which have high circulating levels of 20E in their hemolymph, are more successful than intermolt animals in agonistic interactions. 20E has been shown to act as an internal modulator of neuromuscular physiology: there is an increase in the amplitude of excitatory post-synaptic potentials in the claw opener muscle and a decrease of them in the abdomen when 20E is perfused across the neurons. In addition to its humoral action, 20E appears to be an important signaling molecule sensed by the animal’s antennules, since the behavior of animals change when they are exposed to 20E. The purpose of this study was twofold: to reassess the internal hormonal effects of 20E on agonistic behavior in lobsters, and to provide biochemical evidence for the presence of 20E receptors on the antennules. Fights were conducted between small lobsters injected with 20E and large lobsters injected with saline. The nephropores of each lobster were blocked to eliminate urine signals between the combatants. Using an ethogram, the frequency and intensity of aggressive, defensive and avoidance behaviors of animals in experimental fights were compared to those in control fights (large and small lobsters injected with saline). A significant difference was found in the aggressive content in the behavior of animals engaged in experimental fights and that of the animals engaged in control fights, such that the difference in aggressive content of defensive behaviors between 20E injected animals and their opponents was less than its difference between saline-injected animals and their opponents. These results suggest the aggressiveness of the defensive behavior of smaller treated animals was closer to that of their larger opponents than the behavior of smaller control animals was to their opponents. A post-hoc analysis comparing the control animals in this study to control animals in a similar experiment in which lobsters were injected with 20E and allowed to urinate freely showed that blocking urine release changes the dynamics of an agonistic interaction between lobsters. Since 20E was previously shown to affect the neuromuscular properties of the claw opener muscle, force experiments were performed to test the effect of ecdysteroids on the claw closer muscle. A customized force transducer was constructed to measure the force and duration generated by the closer muscle of male and female lobsters after injection with alpha-ecdysone or 20E. The differences in force and duration before and after injection of 20E or alpha-ecdysone was compared to their differences after injection of saline. Alpha-ecdysone significantly increased the force generated by female crusher and cutter claws, and 20E potentially increased the force in female crusher claws. The results suggest that circulating ecdysteroids influence the claw closer muscle of females and could be a factor influencing agonistic interactions. Because previous behavioral experiments indicated that 20E could be perceived by lobsters and could alter their behavior, experiments were performed to determine whether a 20E receptor (EcR) existed on the antennules of lobsters. In order to visualize the presence of an EcR, various tissues from lobsters were dissected, soluble and insoluble fractions extracted, and spot blots and Western blots performed. Spot blots indicate the presence of a 20E receptor in both the soluble (cytoplasmic/nuclear) and insoluble (membrane-associated) fractions of walking legs and eyestalks, but only in the membrane-associated fraction of the guard setae and aesthetasc sensilla. Western blots and Mass Spectrometry returned several different molecular weights for the EcR (75 kDa, 50 kDa, 40 kDa). The presence of an EcR in the membrane-associated fraction confirms that 20E can be perceived by the antennules of lobsters, while the various molecular weights suggest different isoforms may exist, which is consistent with various insect and crustacean species.

behaviors between 20E injected animals and their opponents was less than its difference between saline-injected animals and their opponents. These results suggest the aggressiveness of the defensive behavior of smaller treated animals was closer to that of their larger opponents than the behavior of smaller control animals was to their opponents. A post-hoc analysis comparing the control animals in this study to control animals in a similar experiment in which lobsters were injected with 20E and allowed to urinate freely showed that blocking urine release changes the dynamics of an agonistic interaction between lobsters.
Since 20E was previously shown to affect the neuromuscular properties of the claw opener muscle, force experiments were performed to test the effect of ecdysteroids on the claw closer muscle. A customized force transducer was constructed to measure the force and duration generated by the closer muscle of male and female lobsters after injection with alpha-ecdysone or 20E. The differences in force and duration before and after injection of 20E or alpha-ecdysone was compared to their differences after injection of saline. Alpha-ecdysone significantly increased the force generated by female crusher and cutter claws, and 20E potentially increased the force in female crusher claws. The results suggest that circulating ecdysteroids influence the claw closer muscle of females and could be a factor influencing agonistic interactions.
Because previous behavioral experiments indicated that 20E could be perceived by lobsters and could alter their behavior, experiments were performed to determine whether a 20E receptor (EcR) existed on the antennules of lobsters. In order to visualize the presence of an EcR, various tissues from lobsters were dissected, soluble and insoluble fractions extracted, and spot blots and Western blots performed.
Spot blots indicate the presence of a 20E receptor in both the soluble (cytoplasmic/nuclear) and insoluble (membrane-associated) fractions of walking legs and eyestalks, but only in the membrane-associated fraction of the guard setae and aesthetasc sensilla. Western blots and Mass Spectrometry returned several different molecular weights for the EcR (75 kDa, 50 kDa, 40 kDa). The presence of an EcR in the membrane-associated fraction confirms that 20E can be perceived by the antennules of lobsters, while the various molecular weights suggest different isoforms may exist, which is consistent with various insect and crustacean species.
v ACKNOWLEDGEMENTS I would first like to thank Dr. Gabriele Kass-Simon for her guidance over the past six plus years. I am grateful for the opportunity she gave me, and her patience in the completion of this dissertation. I would also like to thank all faculty members who Most importantly I need to thank my family for all of their moral (and sometimes financial) support over the course of my graduate school career. My parents have always been very supportive and I wouldn't have been able to complete my degree without their support. Finally, I need to thank my wife for her understanding of the time commitment needed to obtain a PhD. I will finally be able to answer her favorite question: 'When will you be done? ' vi PREFACE This dissertation is prepared in manuscript format. Chapter one contains general background information and the rationale for each of the three major experiments conducted. Each experiment is separated into its own chapter and has been prepared for publication in the format of the relevant journal.
Chapter Two addresses the effects of 20-hydroxyecdysone (20E) on the agonistic interactions of lobsters, and has been prepared for Biological Bulletin.
Appendix A containing all figures, tables and raw data not included in the paper prepared for publication is included at the end of the dissertation.
Chapter Three describes the location and molecular weights of a receptor for ecdysone (EcR) in lobsters, and has been prepared for Chemical Senses.
Chapter Four addresses the effects of 20E and alpha-ecdysone on the claw closer muscle in lobsters, and has been prepared for Biological Bulletin. Appendix B containing all figures, tables, and raw data not included in the paper prepared for publication is included at the end of the dissertation.
In summary, this dissertation addresses the pheromonal and hormonal actions of the molting hormone, 20E, on agonistic interactions in the American lobster, Homarus americanus. It is the first study to show biochemical evidence that a membrane bound EcR exists in lobsters, which could contribute to the immediate changes in agonistic behavior of lobsters exposed to 20E. It also describes the effects of alpha-ecdysone on the force produced by the claw closer muscle.
vii  Table 3. Raw data for male cutter claws treated with 20E………………….258 Table 4. Raw data for female cutter claws treated with alpha-ecdysone…...260 Table 5. Raw data for male cutter claws treated with alpha-ecdysone……..262 Table 6. Raw data for female cutter claws treated with saline……………...264 Table 7. Raw data for male cutter claws treated with saline………………...266 Table 8. Raw data for female crusher claws treated with 20E……………...268 Table 9. Raw data for male crusher claws treated with 20E………………..270 Table 10. Raw data for female crusher claws treated with alpha-ecdysone..272 Table 11. Raw data for male crusher claws treated with alpha-ecdysone….274  Agonistic encounters play an important role in the life of lobsters; they are involved in procuring shelters, defending and maintaining those shelters, mating, and foraging success . Some of the factors affecting the outcomes of agonistic encounters include the physical characteristics of the animals, such as weight, carapace size, and chelae size. Large animals who weigh more and have greater carapace and chelae size  will often take on a dominant position over smaller animals (Cobb and Tamm, 1975). In staged encounters, larger lobsters will win significantly more encounters than smaller lobsters of the same sex . Size also plays a role in the formation of dominance hierarchies in lobsters; larger lobsters are dominant over smaller lobsters in social settings . Similar hierarchies exist in crayfish, where larger animals are dominant over and have more access to food resources than smaller, subdominant males . Shelter competition between lobsters is influenced by size and prior residence; larger animals and animals who maintain prior residence are more aggressive and successful in obtaining/maintaining shelter (O'Neill and Cobb, 1979;Cromarty et al., 1999).

Effects of hormones on agonistic behavior:
Along with size, internal hormones and external chemical signals (pheromones) have been shown to influence the outcome of agonistic interactions between lobsters. The effects of hormones and pheromones are complex and can lead to a wide spectrum of effects on animals engaged in an agonistic interaction.
The hemolymph titers of the molting hormone, 20-hydroxyecdysone (20E), varies over the molt cycle of lobsters. Premolt animals (D 1 and D 2 stage animals about to molt) with increased levels of ecdysone in their hemolymph and urine , are dominant over and more aggressive than their intermolt C-stage opponents in a confrontation . Evidence has been presented that injected 20E makes lobsters more aggressive, as in staged combats between large and small lobsters, smaller lobsters injected with the hormone are significantly more aggressive than small control animals injected with saline . The injection of 20E may mimic the increased 20E titers that occur in D-stage animals about to molt. 20-hydroxyecdysone isn't the only hormone shown to affect agonistic behavior; amines, such as serotonin and octopamine, have been shown to affect behavior (Kravitz, 1990;Kravitz, 2000). In lobsters, injection of serotonin causes sustained flexion of the limbs and abdomen: claws are opened and held forward, walking legs are flexed directly under the body, and tails are loosely flexed and tucked under the body . Octopamine has the opposite effect on the posture of lobsters; injection causes sustained extension of the limbs and abdomen, legs and tail are held rigidly straight just above the substrate, and the tail and claws are fully extended . Evidence has also been presented that injection of serotonin increases the aggression of previously subordinate lobsters (Huber et al., 1997). A dominance hierarchy was established between two lobsters, and subordinate animals were removed and injected with serotonin. The serotonininjected subordinate animals were then re-introduced to the same dominant individual from the first encounter. The fight duration and intensities of the serotonin-injected animals were three times as great as those of saline-injected control animals, which suggests that serotonin made the animals more aggressive (Huber et al., 1997).

Effects of hormones on neuromuscular physiology:
The effects of 20E and serotonin on behavior include their effects on neuromuscular physiology and synaptic transmission; 20E has been shown to affect the neuromuscular electrophysiology of the claw and abdomen. Over the molt cycle, animals in premolt stage D produce significantly larger excitatory junctional potentials (EJP's) and significantly fewer inhibitory junctional potentials (IJP's) in the claw opener muscle . In the presence of 20E (which is present in greater quantities in the hemolymph of premolt animals), there is an increase in EJP amplitudes and miniature excitatory junctional potential (MEJP) frequency in the opener muscle (used in threat displays). There is also, a decrease in EJP amplitudes in the abdominal muscles used in the escape response, which corresponds to the effect of pre-molt and post-molt hemolymph on EJPs . These findings are consistent with the agonistic behavior of premolt animals.
In crayfish, 20E has also been shown to act as a hormone that alters the internal physiology of neurotransmitter release . Cooper and his colleagues have shown that 20E decreases the probability of neurotransmitter vesicle release in the walking legs, and that a mixture of 20E and serotonin increases neuron firing frequency in the slow-adapting muscle receptor organ (MRO) of the abdomen. In early experiments (Fischer and Florey, 1983) both octopamine and serotonin were shown to increase nerve-evoked tension, amplitudes of EPSPs, amplitudes of synaptic currents and the effectiveness of excitation-contraction coupling through an increase in neurotransmitter release in the claw opener muscle of crayfish.
In concordance with their effects on postures and behavior mentioned above, in the neuro-muscular junctions of lobsters, octopamine and serotonin induce contractures and the appearance of Ca 2+ action potentials. Serotonin facilitates transmitter release and dopamine relaxes muscle baseline tension, decreasing contraction . Serotonin and octopamine were also found to act in the central ganglion and affect only slow (postural) flexor and extensor muscles, each with an opposite pattern of activation on the excitatory and inhibitory neurons that innervate the muscles. Octopamine acts on the excitatory extensor neuron and the inhibitory flexor neuron, and serotonin acts on the excitatory flexor neuron and inhibitory extensor neuron . The results show that amines may act as neurohormones in the lobster nervous system, affecting behavior. High circulating levels of octopamine or serotonin could cause sustained extension or flexion of limbs, which could affect the mobility of a lobster. If a lobster is unable to flex or extend claws or walking legs, then they may be at a disadvantage in an agonistic interaction.
In the Pacific spiny lobster, serotonin significantly reduced the strength of graded synaptic transmission, and octopamine significantly increased the strength of graded synaptic transmission at all pyloric dilator synapses (Johnson and Harris-Warwick, 1990). Serotonin was shown to reduce the pre-and post-synaptic input resistance, while octopamine did not change the input resistance. The results suggest that the different amines work through different mechanisms and affect the same neurons differently (Johnson and Harris-Warwick, 1990).
Ecdysteroids may also have differential effects on the neuromuscular properties of the claw closer muscle of crusher and cutter claws, particularly as the two claws differ in their muscle fiber types, motoneuron innervation and contractile properties . Fast twitch fibers quickly reach a maximal tension, which rapidly decays, while slow twitch fibers gradually increase in tension with a slow decay phase . Crusher claw closer muscles contain only slow twitch fibers with long sarcomeres, which allows the crusher to maintain force for a long period of time . Cutter claws contain mostly fast twitch fibers with short sarcomeres and a small ventral band of slow twitch fibers, which leads to a quicker fatigue than crusher muscle . The closer muscle in both crusher and cutter claws are innervated by the fast closer excitor neuron (FCE) and the slow closer excitor neuron (SCE), however, cutter closer muscles have mostly FCE while crushers have a mixture of both (Wiersma, 1955;. Generally, SCE synapses are more fatigue resistant and recover more quickly than FCE synapses, but the synapses at both the FCE and SCE in crusher claws are more resistant to fatigue than those in the cutter . During an agonistic interaction, lobsters perform many behaviors with their claws, including grabbing, hitting, pinching and claw locks Reinhart et al., submitted;Sipala et al., unpublished). These claw behaviors are important, as the strength or duration of a squeeze/pinch may affect the outcome of the agonistic interaction. Also, a larger, thicker claw might deliver more force during a hit than a smaller claw, which could cause more harm. Based on the known neurophysiological effects of 20E, serotonin, and octopamine described above, the different neuromuscular properties of the crusher and cutter claw could lead to different responses to ecdysteroids. Any differential response by the claw closer muscle to ecdysteroids may affect the action of the claw during an agonistic encounter and therefore the outcome the interaction between animals. Such differences would be reflected in differences in patterns of behavior in pre-and intermolt animals and could help explain why premolt animals are more successful in agonistic encounters than intermolt animals.

Effects of pheromones on agonistic behavior:
Along with internal hormonal effects, external pheromonal signals have also been shown to affect the outcome of an agonistic interaction. Urine signaling appears to play a large role in determining the outcome of a fight: the ability of a lobster to 'smell' urine is very important in both establishing dominance hierarchies and in individual recognition in lobsters engaged in agonistic encounters. If urine release is blocked, then dominance hierarchies are not established (Karavanich and Atema 1998;Karavanich and Atema, 1991) and lobsters are not able to recognize each other in subsequent encounters, which results in increased fighting before dominance is reestablished during the secondary encounter (Kaplan, 1993;Karavanich and Atema, 1991). If the aesthetasc sensilla are removed or made anosmic, lobsters are not able to recognize previous opponents or established dominance relationships and spend more time fighting than lobsters who could smell normally (Johnson and Atema, 2005;. The same is true for crayfish, where ablation of aesthetasc sensilla results in fights of longer duration between previous combatants than unablated control pairs (Horner et al., 2008). Further, there is evidence that both the timing of urine release and the contents of the urine affect agonistic encounters between lobsters. It has been shown that the winners of fights release significantly more urine than losers, and the removal or prevention of urine release in subsequent encounters between the same pair of animals abolishes a previously established dominance relationship (Breithaupt and Atema, 1993;Karavanich and Atema, 1998;Breithaupt et al., 1999;Breithaupt and Atema, 2000;Breithaupt and Eger, 2002). These results highlight the importance of the urine as a means of assessing opponents in an agonistic encounter. If the urine signal or the ability to smell are removed, then the behavior of the lobster changes.
One component of urine that has been found to affect the outcome of agonistic encounters is the active principle of the molting hormone, 20E Reinhart et al., submitted). Recent evidence presented by , Reinhart et al. (submitted) and Cromarty et al. (unpublished) indicates that 20E acts as a pheromone that changes the behavior of other lobsters.  found that the behavior of female animals exposed to a plume of 20E was different than control animals exposed to a plume of artificial sea-water (ASW). In these experiments, the nephropores of each animal were blocked, thereby eliminating urine release into the tank, and 20E was puffed across the antennules of large lobsters while their small opponents were made anosmic. Large female lobsters who had a plume of 20E puffed across their antennules performed significantly more aggressive, defensive and avoidance behaviors than large control animals in staged confrontations.
The small non-exposed animals became significantly more aggressive, presumably in response to the larger animal's overall arousal. The change in behavior of the exposed individual can be attributed to the "smelling" of the 20E in the odor plume, suggesting it acts as a pheromone. Reinhart et al. (submitted) performed the same experiment with male lobsters, and found that the behavior of males exposed to a plume of 20E was different than control animals exposed to a plume of artificial sea-water (ASW).
Reinhart found that male lobsters exposed to 20E performed more defensive behaviors than ASW exposed control animals. Also, the opponents of the 20E exposed animals performed significantly more aggressive behaviors than the opponents of ASW exposed animals. These results also led to the conclusion that the change in behavior of the exposed individual could be attributed to the "smelling" of the 20E in the odor plume. One important distinction between the results of Coglianese and Reinhart is that males and females responded differently to 20E: Females responded to 20E exposure by becoming more aroused, increasing aggressive, defensive and avoidance behaviors, whereas males simply increased the frequency of defensive behaviors. In electrophysiological experiments, Cromarty et al. (unpublished) found that the olfactory receptor neurons (ORN bundles) of female lobsters exhibit a dose dependent response to 20E, supporting that idea that 20E may be perceived by the antennules of lobsters during an agonistic encounter.
Molting is a slow process that takes place over several days or weeks, and activation of receptors for 20E most likely act via genomic mechanisms. Steroid hormones have been traditionally considered to work through genomic mechanisms, where steroids enter a cell, bind to a specific receptor in the cytosol or nucleus, and activate transcription that leads to changes in gene expression and results in the production of proteins that have a biological function (Losel et al., 2003). This mechanism is generally slow-acting. The pheromonal effects of 20E described above, however, result in immediate changes in behavior, which cannot be explained by slow-acting genomic mechanisms.
A model for the fast-actions of steroid hormones involves a non-genomic mechanism, wherein steroids have an immediate effect on physiological function (Losel et al., 2003). The physiological effect of steroids that act through non-genomic mechanisms can be seen within seconds of exposure to the hormone, ruling out any models that involve changes in the transcription levels of genes. The activation of an outer membrane-bound receptor, or signaling via a second messenger pathway are likely causes of the immediate physiological changes observed to occur in response to the hormone (Losel et al., 2003).
In recent studies in numerous insects and crustaceans, ecdysteroids have been shown to have fast-acting effects, suggesting the presence of an additional nonclassical steroid hormone signaling pathway (Spencer and Case, 1984;Tomaschko, 1999;Thummel and Chory, 2002;Schlattner et al., 2006). Rapid non-genomic effects have been found to act in exocrine glands, the central nervous system, motor neurons, neuromuscular junctions and sensory cells of numerous organisms (Schlattner et al., 2006). Compared to the prolonged and slow process of ecdysone induced molting, the non-genomic effects of ecdysone exposure are immediate; changes are sometimes observed within a matter of seconds or milliseconds (Tomaschko, 1999).
In the California spiny lobster, Spencer and Case (1984) found an increased action potential firing-frequency in the lateral antennule one second after exposure to both 20E and alpha ecdysone. In American lobsters, 20E has been shown to have immediate effects on neuro-muscular properties. In the presence of 20E, there is an increase in EJP amplitudes and miniature excitatory junctional potential (MEJP) frequency in the opener muscle, as well as a decrease in EJP amplitudes in the abdomen . In Drosophila, Ruffner et al. (1999) found reduced transmitter release in the ventral abdominal muscles within one minute after exposure to 20E. In crayfish, there is a decreased amount of neurotransmitter release in the opener muscle of the first walking leg within 20 minutes after exposure to 20E .  found increased action potential firing frequency in the muscle receptor organ of crayfish 10 seconds after exposure to 20E.
As described above, 20-hydroxyecdysone also appears to be a pheromone that immediately alters the agonistic behavior of lobsters Reinhart et al., submitted) and is sensed by ORN in the antennules (Cromarty et al., unpublished). The external perception of 20E in lobsters and the immediate change in behavior of lobsters exposed to 20E suggest that a membrane bound receptor must be present on the aesthetasc sensilla. In spiny lobsters, Panulirus argus, the aesthetasc sensilla are the sensory cilia of the olfactory receptor cells whose nuclei are located within the antennules themselves (Ache and Derby, 1985;Grunert and Ache, 1988).
Histological studies show that the aesthetasc sensilla are innervated by the dendritic extensions of multiple bipolar receptors, with the soma gathered in a cluster at the base of the sensillum inside the antennule itself (Ache and Derby, 1985;Grunert and Ache, 1988). The anatomy of the antennules in the Spiny lobster suggests that any receptor for 20E must be a membrane bound receptor, as no nucleus or cytoplasm exists in the sensory hair itself. A similar morphology is presumed to exist in the American lobster, although the histology of the antennules in the American lobster has not been examined. The activation of membrane bound receptors is fast-acting, and could explain the immediate change in behavior observed by  and Reinhart et al. (submitted), and the immediate neural response to 20E observed by Cromarty et al. (unpublished).
The presence of a membrane bound ecdysone receptor has been isolated from the anterior silk gland of the silkworm, Bombyx mori (Elmogy et al., 2004). 20E aids in the initiation of apoptosis of the anterior silk gland, and Elmogy et al. (2004) found a putative membrane receptor located in the plasma membrane. This membrane receptor exhibited saturable binding to 20E and the authors suggest that the receptor is likely to be an integrated membrane protein. The presence of a membrane bound receptor for 20E in the silkworm supports the idea that a membrane bound receptor for 20E exists in other insect and crustacean tissues. Srivastava et al. (2005) discovered a G-protein coupled ecdysone receptor in Drosophila, so it is possible that an EcR in lobsters might act through a second messenger pathway.

Rationale of Dissertation:
This dissertation investigates both the internal and external effects of 20E on agonistic interactions in lobsters.
One of the questions raised by the Bolingbroke, Coglianese and Reinhart experiments (cited above) is what role do the internal hormonal effects and the external pheromonal effects of 20E play in the agonistic interactions of lobsters. In  lobsters were injected with 20E and allowed to urinate, and both the hormonal and pheromonal effects of 20E were present. In  and Reinhart et al. (submitted), lobsters had their nephropores blocked and were not injected, but 20E was puffed from one lobster onto another allowing for only the pheromonal signal of 20E. The purpose of the experiments in Chapter Two was to determine the effects of the 20E hormonal signal alone on the aggressive behaviors of female American lobsters, Homarus americanus. To do this, lobsters were injected with 20E and the nephropores were blocked, effectively eliminating urine released into the water.
The apparent pheromonal effects of 20E shown by , Reinhart et al. (submitted) and corroborated by the response from ORN by Cromarty et al. (unpublished), lead to experiments in Chapter Three. The purpose of these experiments was to find biochemical evidence of a membrane bound EcR on the antennules of lobsters and to provide a molecular weight for the receptor. The presence of a membrane bound receptor would support the idea that lobsters are able to perceive 20E, and that the actions are too quick to be explained by genomic mechanisms.
The importance of claws in agonistic interactions and the effects of 20E on the neuromuscular physiology of lobsters and crayfish lead to the experiments in Chapter Four. The purpose of these experiments was to determine the effects of ecdysteroids, 20E and alpha-ecdysone on the claw closer muscle in lobsters. Since 20E affects neuromuscular physiology in the claw opener muscle and abdomen, it is possible that it affects the claw closer muscle. If 20E or alpha-ecdysone change neuromuscular properties, such as increasing the force or duration of a squeeze, a lobster that has high circulating levels of ecdysteroids may have an advantage in an agonistic interaction. In lobsters, 20-hydroxyecdysone (20E) has been shown to act as both an internal modulator of neuromuscular physiology and as an external pheromone that affects behavior. The purpose of this study was to reassess the internal hormonal effects of 20E on agonistic behavior in lobsters. Experimental fights were conducted between small lobsters injected with 20E and large lobsters injected with saline.
Control fights consisted of small and large lobsters injected with saline. The nephropores of each lobster were blocked to eliminate urine signals between the combatants. Using an ethogram, the frequency and intensity of aggressive, defensive and avoidance behaviors of animals in experimental fights were compared to those in control fights. Significance was found between the differences in aggressive content of animals engaged in experimental fights to animals engaged in control fights. These results suggest the aggressiveness of the defensive behavior of smaller treated animals was more similar to that of their larger opponents than the aggressiveness of defensive behaviors of smaller control animals to their larger opponents. A post-hoc analysis comparing our control animals to control animals from a similar experiment in which lobsters were injected with 20E and allowed to urinate freely showed that blocking urine release changes the dynamics of an agonistic interaction between lobsters.

Introduction:
Agonistic encounters play an important role in the life of lobsters; they are involved in procuring shelters, defending and maintaining those shelters, mating, and foraging success . Some of the factors affecting the outcomes of agonistic encounters include weight, carapace size, and chelae size; large animals weigh more and have greater carapace and chelae size . In staged encounters, larger lobsters will win significantly more encounters than smaller lobsters of the same sex . Size also plays a role in the formation of dominance hierarchies in lobsters, as larger lobsters are dominant over smaller lobsters in social settings . This is also true of crayfish, where larger animals are dominant over and have more access to food resources than smaller, subdominant males .
Along with size, internal hormones and external chemical signals (pheromones) have been shown to influence the outcome of agonistic interactions between lobsters. The effects of hormones and pheromones are complex and can lead to a wide spectrum of effects on animals engaged in an agonistic interaction. Recent evidence indicates that 20-hydroxyecdysone (20E), a hormone that modulates molting in American lobsters, also acts as a pheromone Reinhart et al., unpublished).
The hemolymph titers of the molting hormone, 20-hydroxyecdysone (20E), varies over the molt cycle. Premolt animals (D 1 and D 2 stage animals about to molt) have increased levels of ecdysone in their hemolymph and urine , and are dominant over and more aggressive than their intermolt C-stage opponents in a confrontation . Evidence has been presented that injected 20E makes lobsters more aggressive, as in staged combats between large and small lobsters, smaller lobsters injected with the hormone are significantly more aggressive than small control animals injected with saline . The injection of 20E may mimic the increased 20E titers that occur in D-stage animals about to molt (D1 and D2), which are correlated with increased aggression in D-stage animals .
20E has been shown to affect the neuromuscular electrophysiology of the claw and abdomen. In the claw opener muscle, the amount of opening depends on the patterned interaction between excitatory and inhibitory junctional potentials Kass-Simon and Govind, 1989). Over the molt cycle, animals in premolt stage D produce significantly larger excitatory junctional potentials (EJP's) and significantly fewer inhibitory junctional potentials (IJP's) in the claw opener muscle . In the presence of 20E (which is present in greater quantities in the hemolymph of premolt animals), there is an increase in EJP amplitudes and miniature excitatory junctional potential (MEJP) frequency in the opener muscle (used in threat displays). Also, there is a decrease in EJP amplitudes in the abdomen (used in the escape response), which corresponds to the effect of premolt and post-molt hemolymph on EJPs . This correlates with the finding that lobsters are more aggressive just before molting.
In crayfish, 20E has been shown to act as a hormone that alters the internal physiology of neurotransmitter release . Cooper and his colleagues have shown that 20E decreases the probability of neurotransmitter vesicle release in the walking legs, and that a mixture of 20E and serotonin was effective in increasing neuron firing frequency in the slow-adapting muscle receptor organ (MRO) of the abdomen. In lobsters, injection of serotonin causes sustained flexion of the limbs and abdomen, where claws are opened and held forward, walking legs are flexed directly under the body, and tails are loosely flexed and tucked under the body . Furthermore, evidence has been presented that injection of serotonin increases the aggression of previously subordinate lobsters (Huber et al., 1997). Dominance hierarchies were established between two lobsters, subordinate animals were removed and injected with serotonin, and were then re-introduced to the same dominant individual from the first encounter. These serotonin injected animals had a fight duration and intensity level three times that of a saline injected control animal, which suggested that serotonin made them more aggressive (Huber et al., 1997). However, recent studies indicate that the removal of serotonin also increases the duration of fighting behavior in lobsters . This suggest that the concentration of serotonin, per se, is unlikely to be the determining factor in the level of aggression. This is consistent with the earlier biochemical studies indicating that serotonin does not vary significantly over the molt cycle .
Internal hormonal effects are not the only factors that affect the outcome of a confrontation, as urine signaling appears to play a large role in determining the outcome of a fight. There is evidence that both the timing of urine release and the contents of the urine affect agonistic encounters. It has been shown that the winners of fights release significantly more urine than the losers, and the removal or prevention of urine release in subsequent encounters between the same pair of animals abolishes a previously established dominance relationship (Breithaupt and Atema, 1993;Karavanich and Atema, 1998;Breithaupt et al., 1999;Breithaupt and Atema, 2000;Breithaupt and Eger, 2002). One component of urine that has been found to affect the outcome of agonistic encounters is the active principle of the molting hormone, 20E Reinhart et al., submitted).
Recent evidence presented by , Reinhart et al. (submitted) and Cromarty et al. (unpublished) indicates that 20E may not only act as a hormone, but also as a pheromone that changes the behavior of other lobsters.  found that the behavior of female animals exposed to a plume of 20E was different than control animals exposed to a plume of artificial sea-water (ASW). In these experiments, the nephropores of each animal were blocked, thereby eliminating urine release into the tank, and 20E was puffed across the antennules of large lobsters while their small opponents were made anosmic. Large female lobsters who had a plume of 20E puffed across their antennules performed significantly more aggressive, defensive and avoidance behaviors than large control animals in staged confrontations. The small non-exposed animals became significantly more aggressive, presumably in response to the larger animal's overall arousal. The change in behavior of the exposed individual can be attributed to the "smelling" of the 20E in the odor plume, suggesting it acts as a pheromone. Reinhart et al. (submitted) performed the same experiment as Coglianese with male lobsters, and found that the behavior of males exposed to a plume of 20E was different than control animals exposed to a plume of artificial sea-water (ASW). Reinhart found that male lobsters exposed to 20E performed more defensive behaviors than ASW exposed control animals. Also, the opponents of the 20E exposed animals performed significantly more aggressive behaviors than the opponents of ASW exposed animals. These results also led to the conclusion that the change in behavior of the exposed individual could be attributed to the "smelling" of the 20E in the odor plume. In electrophysiological experiments, Cromarty et al. (unpublished) found that the olfactory receptor neurons (ORN bundles) exhibit a dose dependent response to 20E, supporting that idea that 20E may be perceived by the antennules of lobsters during an agonistic encounter.

One of the questions raised by the Bolingbroke, Coglianese and Reinhart
experiments is what role do the internal hormonal effects and the external pheromonal effects of 20E play in the agonistic interactions of lobsters. In  lobsters were injected with 20E and allowed to urinate, and both the hormonal and pheromonal effects of 20E were present. In  and Reinhart et al. (submitted), lobsters had their nephropores blocked and were not injected, but 20E was puffed from one lobster onto another allowing for only the pheromonal signal of 20E. The purpose of the present experiment was to determine the effects of the 20E hormonal signal alone on the aggressive behaviors of female American lobsters, Homarus americanus. To do this, lobsters were injected with 20E and the nephropores were blocked, effectively eliminating urine released into the water. Given that urine appears to affect the outcome of aggressive encounters and that the injection of 20E appears to increase the aggressiveness of lobsters, blocking the pheromone signal leaves only the hormonal effect. The experimental set-up was identical to , with the exception that the nephropores were blocked on all animals.

Methods:
Animal Procurement and maintenance: Female American lobsters, Homarus americanus, were obtained from local fisherman and the Rhode Island Department of Environmental Management from inshore waters off the coast of Narragansett Bay, RI. Animals were maintained in natural circulating unfiltered seawater tanks at the Narragansett Bay Campus, on a 12hr light/dark cycle. Water temperature and salinity were ambient, ranging from 10-20°C and 28-33ppt, respectively. Animals were fed fish scraps, supplied by a local fish market, twice weekly. No animal was fed 48 hours prior to a fight. Animals were banded and kept in separate tanks with compartmentalized large gauge wire cages to prevent physical interactions between any lobsters prior to a fight. All animals were returned to Narragansett Bay after 2 weeks.

Experimental set-up:
A total of 20 fights (10 experimental and 10 control) were carried out between July-August 2005 and July-August 2006. Experimental fights consisted of a small 20hydroxyecdysone-injected lobster pitted against a large saline-injected lobster; control fights consisted of saline-injected small lobsters versus saline-injected large animals.
The large animal in all fights was identified by a rubber band placed on the endopodite of the crusher claw between the cheliped joint and the insertion of the dactyl. The band was placed in such a way that it did not interfere with the normal movement of the joint or the claw as a whole. In order to prevent urine release into the tank during the fights, the nephropores of each lobster were covered with aquarium tubing sealed at one end with sealing wax. Aquarium tubing was first cut to a size of approximately 2 cm, one end was blocked with sealing wax, allowed to cool and tested for leaks.
The nephropore blockers were glued over the nephropores with Super-Glue on the morning of the fight. Although the nephropores were blocked, the actual release of urine from the lobsters was not blocked during the fight, as the nephropore blockers collected the urine that was released by the lobsters during the fight.
The pre-fight injection protocol consisted of 4 injections of 20E or lobster saline 12 hours apart, and with 12 hours between the fourth injection and the fight. glucose 11; Tris-maleate 10;pH 7.4 (Meiss and Govind, 1979). Experimental animals were injected with enough 20-hydroxyecdysone to result in a final hemolymph concentration inside the body of 600 ng/ml. Stock aliquots of 20E (1mg/mL saline) were frozen at -80°C. The volume of stock solutions injected was that which was estimated on the basis of the animals weight to result in a final hemolymph concentration of 600ng/ml. This weight/volume estimate was generated by , by measuring the hemolymph volume bled from lobsters of known weights and fitted to a linear curve, having the values: y = 0.26x -54.33, where y is the hemolymph volume and x is the weight of the animal. The volume estimated from the equation was then used in a ratio to determine the amount of 20E stock solution needed to be injected in order to obtain a final concentration of 600ng/mL. The ratio used was: 0.0006mg/1mL = X mg/hemolymph volume of interest (y from the equation). For example, an animal with a total hemolymph volume of 50mL would receive an injection of 0.03 mL stock 20E solution, whereas one with a hemolymph volume of 100mL would receive an injection of 0.06 mL stock solution. The amount of saline injected in the control animals was calculated the same way. A Dremel electric drill was used to drill a small hole through the outer portion of the carapace above the presumed level of the cardium through which a 20-gauge needle could be inserted into the remaining carapace layer. After injection, the hole was plugged with dental wax to prevent bleeding.

Analysis:
Each fight was analyzed for aggressive, defensive and avoidance behaviors using the behavioral ethogram developed by  and modified by  and   (Table 1). The ethogram ranks each behavior on a numerical scale, called the Rank of Aggression scale, where the most aggressive behaviors receive the highest number, and the least aggressive behaviors (avoidance behaviors) receive the lowest numbers. For clarity, a summary of the considerations used in the ethogram is repeated here: Aggressive behaviors are defined as any behavior in which the animal attempts to cause damage to their opponent or signal a threat of such a behavior. Defensive behaviors are defined as behaviors that attempt to ward off aggressive behaviors by an opponent.
Avoidance behaviors are defined as any behaviors in which the animal attempts to get away from, or avoid its opponent. Along with behaviors directed towards opponents, behaviors designated as wall behaviors were also recorded. Wall behaviors are defined as any aggressive or defensive behavior that is directed towards the walls of the fighting tank, rather than towards the opponent. Definitions of all behaviors in the ethogram are listed in Table 2.
The fights were analyzed by two people, one of whom (MS) had initially staged the confrontations. The number of times each behavior was performed by each animal was noted (Frequency) and recorded into a computer program that kept a running total of the number of behaviors and also the rank of each behavior. After the frequency of behaviors were tabulated, two more parameters were calculated in order to assess the relative aggressiveness of each animal, the Rank Frequency and the Average Rank. The Rank Frequency (RF) for each animal was calculated by multiplying the Frequency of each behavior by its Rank of Aggression value, in order to reflect relative aggressive intensity. The RF value accounts for animals that may perform a low total number of aggressive behaviors, but perform many highly Single factor Analyses of Variance (ANOVA) were used to determine significance between control and experimental animals. In our analysis, small experimental animals (20E injected) (Treated, T) were compared to small control animals (saline injected) (Control, C), and large experimental animals (opponents of 20E injected animals (OT)) were compared to large control animals (opponents of saline injected small animals (OC)). ANOVA's were performed on all three parameters measured, Frequency, Rank Frequency and Average Rank for all behaviors recorded (aggressive, defensive, avoidance) both with and without wall behaviors.
ANOVA's were also performed on each pair within a fight; large experimental plus small experimental vs. large control plus small control (OT + T vs. OC + C), large experimental minus small experimental vs. large control minus small control (OT -T vs. OC-C). Values were considered significant at p < 0.05 and potentially significant (strong trend) at 0.05< p <0.08.

Post-Hoc Analysis:
During the course of analyzing the data of the present study, it became apparent the results were different from those of .
Since no significant differences were found among the aggressive or avoidance behaviors and only one significant difference was found in defensive behaviors, the question arose as to whether blocking the nephropores changes the dynamics of Values were considered significant if the p-value was < 0.05.

All Behaviors (Aggressive, Defensive and Avoidance):
No significant differences were found for all behaviors between treated animals (T) and their counterpart controls (C) in any of the parameters measured with or without wall behaviors (Frequency, Rank Frequency and Average Rank). Nor were there any significant differences found between the opponents of treated animals (OT) and their counterpart controls (OC) for any parameters measured with or without wall behaviors. Similarly, there were no significant differences found between OT + T vs. OC + C, or between OT -T vs. OC -C for any parameters measured with or without wall behavior.

Aggressive Behaviors:
No significant differences were found for any aggressive behaviors between treated animals (T) and their counterpart controls (C) in any of the parameters measured with or without wall behaviors (Frequency, Rank Frequency and Average Rank). Nor were there any significant differences found between the opponents of treated animals (OT) and their counterpart controls (OC) for any parameters measured with or without wall behaviors. Similarly, there were no significant differences found between OT + T vs. OC + C, or between OT -T vs. OC -C for any parameters measured with or without wall behavior.

Defensive Behaviors:
No significant differences were found for any defensive behaviors between treated animals (T) and their counterpart controls (C) in any of the parameters measured with or without wall behaviors (Frequency, Rank Frequency and Average Rank). Nor were there any significant differences found between the opponents of treated animals (OT) and their counterpart controls (OC) for any parameters measured with or without wall behaviors. Similarly, there were no significant differences found between OT + T vs. OC + C for any parameters with or without wall behaviors.
Significance was found in the differences in Average Rank between the OT -T vs. OC -C with wall behaviors (41.8 and 60.7, respectively; F 1,18 =4.9, P=0.04) ( Figure 1) indicating that the disparity between the two combatants in control fights might have been greater than in treated fights. Since the Average Rank is larger for the control fights, these results indicate that the aggressiveness of the defensive behavior of smaller treated animals was more similar to that of their larger opponents than the aggressiveness of the defensive behaviors of smaller control animals and their larger opponents. No other significant differences were found between OT -T and OC -C animals for any parameters measured with or without wall behaviors.

Avoidance Behaviors:
No significant differences were found for any avoidance behaviors between treated animals (T) and their counterpart controls (C) in any of the parameters measured (Frequency, Rank Frequency and Average Rank). Nor were there any significant differences found between the opponents of treated animals (OT) and their counterpart controls (OC) for any parameters measured.

Post-Hoc Results:
The post-hoc analysis results suggests that the removal of the urine/chemical signal from an agonistic interaction changes the behaviors of the combatants in the following ways.

Aggressive Behaviors:
Significant differences in the Frequency of aggressive behaviors were found between the control experiments in which the nephropores were blocked (Sipala) and those in which they were not blocked (Bolingbroke). C S animals performed significantly more aggressive behaviors than C B animals (183.4 and 93.2, respectively, p=0.004), while OC B animals performed significantly more aggressive behaviors than OC S animals (359.9 and 242.2, respectively, p=0.009) ( Figure 2). When the aggressive behaviors of the control animal were subtracted from the behaviors of the opponent control animal in a single fight, the difference between OC B -C B was larger than the difference between OC S -C S (266.7 and 59, respectively, p=0.001) ( Figure 2).

Defensive Behaviors:
Significant differences in the Frequency of defensive behaviors were found between the experiments such that C B animals performed significantly more defensive behaviors than C S animals (110.8 and 51.3, respectively, p=0.0000004) ( Figure 3).
When the defensive behaviors of the two animals in a single fight were added together, OC B +C B performed significantly more defensive behaviors than OC S +C S (152.2 and 99.1, respectively, p=0.001) ( Figure 3). When the defensive behaviors of the control animal were subtracted from the behaviors of the opponent control animal in a single fight, OC B -C B was larger than OC S -C S (-69.4 and -3.5, respectively, p=0.0002) ( Figure 3).

Avoidance Behaviors:
Significant differences in the Frequency of avoidance behaviors were found between the experiments such that C B animals performed significantly more avoidance behaviors than C S animals (81.5 and 59.6, respectively, p=0.03) (Figure 4). When the avoidance behaviors of the two animals in a single fight were added together, OC B +C B performed significantly more avoidance behaviors than OC S +C S (135.2 and 107.5, respectively, p=0.007) ( Figure 4).

Discussion:
The purpose of this study was to determine whether increased blood titers of 20-hydroxyecdysone affected the aggressive behavior of female American lobsters, Homarus americanus. The only significant difference found in the present study was the difference in Average Rank of Defensive Behaviors between the OT -T and OC -C with wall behaviors (41.8 and 60.7, respectively; F 1,18 =4.9, P=0.04), indicating that the difference in the aggressiveness of defensive behaviors of treated animals was more similar to that of their opponents than the aggressiveness of the defensive behaviors of control animals and their opponents. Therefore, the disparity between the two combatants is greater in the control fights than in the fights with treated animals. This could be due to the hormonal effect of 20E making treated animals more aggressive than saline-injected control animals. The increased aggression of the treated animals could cause the OT animals to increase their level of aggression in defensive response, which could lead to a decreased difference in the total amount of defensive behaviors between the T and OT animals. It was expected that the injection of the hormone alone (with no urine release) would increase the aggressiveness of lobsters due to its effects on physiological processes: increase in EJP amplitudes and miniature excitatory junctional potential (MEJP) frequency in the opener muscle, as well as a decrease in EJP amplitudes in the abdomen ). However, our experiments failed to find any significant differences in aggressive behaviors between hormone-injected treated animals and saline-injected control animals (or their opponents) for any category of behaviors tested. The lack of significance in other parameters raised the possibility that an experimental artifact had been created by removing the urine signal from the interaction, since in comparable studies in which an olfactory signal was present, a number of significant differences were found Reinhart et al., unpublished).
This led to a post-hoc analysis to compare the behaviors of the control animals of the present study with the control animals in the Bolingbroke and Kass-Simon study, since the only treatment differences between these two sets of animals was that the control animals in the Bolingbroke and Kass-Simon study were capable of receiving a urine signal, while those in the present study were not. The differences in behaviors found between the present experiment and  indicates that removing the urine signal changes the dynamics of agonistic interactions in lobsters. Since the only obvious consistent difference between the control animals was the lack of a urine signal in the present study, the difference in behavior may be due to the removal of the urine signal.
In an agonistic interaction between a large and a small lobster, large lobsters win significantly more encounters than smaller lobsters .
Concomitantly, larger animals will often take on a dominant position over smaller animals (Cobb and Tamm, 1975), evicting them from shelters and initiating more confrontations than smaller animals . Furthermore, once dominance is established between two lobsters in a staged confrontation, dominant 'winners' continue to perform aggressive behaviors toward the subordinate 'losers,' who perform submissive or avoidance behaviors (Karavanich and Atema, 1998). The present study and Bolingbroke and Kass-Simon (2001) do not directly assess dominance or 'winners' and 'losers' in a fight, but addresses the aggressiveness of the behaviors of the lobsters engaged in the interaction. Since the fights consist of a large animal versus a small animal, the fight is biased in favor of the larger animal, and the behaviors of each animal in the interaction should follow those described in the previous experiments: larger animals should be dominant over smaller animals and perform more aggressive behaviors, while smaller animals should be submissive and perform more submissive/avoidance behaviors.

The control animals in Bolingbroke and Kass-Simon (2001) (C B and OC B )
appear to follow similar behavioral patterns to the normal dominant/subordinate interaction, while C S and OC S animals do not. The differences found between OC B -C B and OC S -C S suggests that there is a smaller difference in the total number of aggressive behaviors between the lobsters in a single fight when the urine signal is removed. In OC B animals, the total number of aggressive behaviors is greater than that of the C B animals, leading to a larger difference. The difference in the aggressive behaviors between OC S and C S was much smaller than that of Bolingbroke's, which means that the number of aggressive behaviors of the smaller animal was closer to that of the larger animal. Once the urine signal is removed, small animals that would ordinarily become less aggressive in an interaction with a larger opponent did not, which accounts for the smaller difference in aggressive behaviors between OC S -C S than OC B -C B . The removal of urine changes the behavior of a small animal engaged agonistic interaction with a larger animal is that C S animals perform significantly more aggressive behaviors than C B animals, and C B animals perform significantly more defensive and avoidance behaviors. OC B animals performed significantly more aggressive behaviors than OC S animals, which contradicts the finding that non-urine signaled control animals are more aggressive than urine-signaled controls. Since small C S did not become more defensive or subordinate, the large OC S did not become more aggressive in response to a more defensive subordinate smaller animal. This may be why OC B perform more aggressive behaviors than OC S . Another reason OC B may be more aggressive is explained by the fact that OC B + C B perform significantly more defensive and avoidance behaviors than OC S + C S . Since C B immediately perform more defensive and avoidant behaviors in response to a larger dominant animal, the additive value of those behaviors in the individual fight is very high due to the high frequency of behaviors by the smaller animal. This could further explain why OC B are more aggressive than OC S , as the larger OC B would increase its aggressiveness in response to a smaller subordinate animal, while OC S would not increase their aggressiveness because their smaller opponent does not exhibit the same frequency of defensive and avoidance behaviors as a urine-signaled animal.
These results together suggest that blocked animals engaged in an agonistic interaction are not able to assess each other through urine signals, resulting in a change of the dynamics of the encounter. The ability of each lobster to asses each other via urine signals is an important determinant in the outcome of agonistic interactions (Breithaupt and Atema, 1993;Karavanich and Atema, 1998;Breithaupt et al., 1999;Breithaupt and Atema, 2000;Breithaupt and Eger, 2002), and the lack of urine signals in the present study affects the outcome of the encounter. Bolingbroke's OC is able to receive a urine signal from its smaller combatant, alerting the larger animal that the smaller lobster is weaker, and therefore increases its aggressiveness. Sipala's OC does not receive a urine signal, so it is not able to assess the strength of its smaller combatant, and its aggressiveness does not increase as much as Bolingbroke's OC. Bolingbroke's OC aggressiveness has been increased by the perceived weakness of its smaller combatant, and since Sipala's OC can't assess its opponent, it is less aggressive than Bolingbroke's OC. Conversely, Sipala's C did not receive a urine signal from a blocked larger opponent, and therefore increased its aggressiveness due to the lack of a 'strong' signal from a larger animal.
The ability of a lobster to 'smell' urine is very important in both establishing dominance hierarchies and individual recognition in lobsters engaged in agonistic encounters. If urine release is blocked, then dominance hierarchies are not established (Karavanich and Atema 1998;Karavanich and Atema, 1991) and lobsters are not able to recognize each other in subsequent encounters, leading to increased fighting before dominance is re-established (Kaplan, 1993;Karavanich and Atema, 1991).
Furthermore, if the aesthetasc sensilla are removed or made anosmic, lobsters are not able to recognize previous opponents or established dominance relationships and spend more time fighting than lobsters who could smell normally (Johnson and Atema, 2005;. The same is true for crayfish, as the ablation of aesthetasc sensilla results in fights of longer duration between previous combatants than unablated control pairs (Horner et al., 2008). These results highlight the importance of lobsters being able to smell urine and gain some kind of assessment of the animal opposite them in an agonistic encounter. If the urine signal or the ability to smell are removed, then the behavior of the lobster changes. The lobsters in the present study were not able to smell the urine of their opponent lobsters, which may have prevented them from assessing their opponent, thereby affecting their behavior.    (28) An extension of the entire body with chelipeds and tail extended, such that the animal has made its profile as thing and long as possible Back/front truce (26) Opponents aligned with the tail of one in close proximity or touching the opponent Backing away (8) Walking backward with contact while facing an opponent Claws in front of head preparing to confront; the large ready differs from meral spread in that the claws are not raised as high or spread as far apart; the small ready is a lesser large ready

Horner
Retreat (10) Walking backward without contact while facing an opponent Shielding (60)        Ecdysone and its active metabolite, 20-hydroxecdysone (20E), are steroid hormones that regulate molting in insects and crustaceans and coordinate alterations in the transcription of groups of genes required to control this process (Waddy et al., 1995). Molting is a slow process that takes place over several days or weeks, however, in recent studies in numerous insects and crustaceans, Ecdysteroids have been shown to have fast-acting effects, suggesting the presence of an additional nonclassical steroid hormone signaling pathway (Spencer and Case, 1984;Tomaschko, 1999;Thummel and Chory, 2002;Schlattner et al., 2006).
Steroid hormones have been traditionally considered to work through genomic mechanisms, where steroids enter a cell, bind to a specific receptor in the cytosol or nucleus, and activate transcription that leads to changes in gene expression and results in the production of proteins that have a biological function (Losel et al., 2003). This mechanism is generally slow-acting, as it sometimes takes several hours or days to alter patterns of gene transcription after the hormone enters the cell. A second model for the actions of steroid hormones involves a non-genomic mechanism, wherein steroids have an immediate effect on physiological function that cannot be explained by the slower classical mechanisms (Losel et al., 2003). The physiological effect of steroids that act through non-genomic mechanisms can be seen within seconds of exposure to the hormone, ruling out any models that involve changes in the transcription levels of genes. The activation of an outer membrane-bound receptor, or signaling via a second messenger pathway are likely causes of the immediate physiological changes observed to occur in response to the hormone (Losel et al., 2003).
Rapid non-genomic effects have been found to act in exocrine glands, the central nervous system, motor neurons, neuromuscular junctions and sensory cells of numerous organisms (Schlattner et al., 2006). Compared to the prolonged and slow process of ecdysone induced molting, the non-genomic effects of ecdysone exposure are immediate; changes are sometimes observed within a matter of seconds (Tomaschko, 1999). In the California spiny lobster, Spencer and Case (1984) found an increased action potential firing frequency in the lateral antennule 1 second after exposure to both 20E and alpha ecdysone. In American lobsters, 20E has been shown to have immediate effects on neuro-muscular properties. In the presence of 20E, there is an increase in EJP amplitudes and miniature excitatory junctional potential (MEJP) frequency in the opener muscle, as well as a decrease in EJP amplitudes in the abdomen . In Drosophila, Ruffner et al. (1999) found reduced transmitter release in the ventral abdominal muscles within 1 minute after exposure to 20E. In crayfish,  found a decreased amount of neurotransmitter release in the opener muscle of the first walking leg within 20 minutes after exposure to 20E.  found increased action potential firing frequency in the muscle receptor organ of crayfish 10 seconds after exposure to 20E.
Recent evidence presented by  and Reinhart et al. (submitted) indicates that 20E acts as a pheromone that leads to an immediate change in the behavior of lobsters.  found that the behavior of female animals exposed to a plume of 20E was different than control animals exposed to a plume of artificial sea-water (ASW). Large female lobsters who had a plume of 20E puffed across their antennules performed significantly more aggressive, defensive and avoidance behaviors than large control animals in staged confrontations. The change in behavior of the exposed individual was attributed to the perception of the 20E in the odor plume, suggesting it acts as a pheromone. Reinhart et al. (submitted) performed a similar experiment as Coglianese with male lobsters, and found that the behavior of males exposed to a plume of 20E was also different than control animals exposed to a plume of artificial sea-water (ASW). Reinhart found that male lobsters exposed to 20E performed more defensive behaviors than ASW exposed control animals. These results led to the conclusion that the change in behavior of the exposed individual could be attributed to the perception of the 20E in the odor plume.
In earlier electrophysiological experiments, Cromarty et al. (unpublished) found that the olfactory receptor neurons (ORN bundles) in the antennules of American lobsters exhibit a dose-dependent response to 20E, which supports the idea that 20E may be perceived by the antennules of lobsters and this perception is responsible for the alteration of behavior during an agonistic encounter. Because the physiological and behavioral changes occur within milliseconds or seconds of ecdysone exposure, these result together suggest the presence of a non-genomic receptor on the lateral antennules.
Here, we present evidence for a membrane-bound and nuclear ecdysone receptor (EcR) with comparable molecular weights in various tissues of the American lobster, Homarus americanus. The presence of a receptor for 20E in the membrane fraction suggests the possibility that a membrane bound receptor is responsible for the immediate physiological response exhibited by lobsters exposed to 20E.

Methods:
Animal procurement and maintenance:

Homogenization and Membrane Protein Fractionation:
The tissue was placed into a homogenization tube and macerated sequentially with two different buffers: homogenization buffer (20 mM Tris-HCl, 2 mM EDTA) and detergent Buffer (

Protein Characterization by Western Blotting:
The homogenized samples were then either spotted directly onto nitrocellulose paper or loaded into a 10% SDS-PAGE gel. Samples run on SDS-PAGE gels were then transferred to nitrocellulose. A total of three spot blots were made during each experiment: one spot blot for an amido black total protein stain and two blots for staining with two different anti-mouse secondary antibodies (Biotin-linked goat antimouse and LICOR Odyssey Goat anti-mouse IR DYE 680LT). We used two different types of standards as controls for the two different types of blots: BSA was used to quantify the total protein blot and a standard mouse IGG was used for the two antibody blots. The total protein stain spot blot was stained with 0.25 % Amido Black in order to determine if protein was indeed extracted for each sample. This blot was later compared to the blots labeled with EcR-specific antibodies in order to eliminate background artifacts. If a receptor was positively identified on an antibody spot blot, but inadequate protein was extracted from that tissue sample on the amido black spot, then that result could be thrown out as background staining.
The staining protocol for the Biotin linked Goat Anti-Mouse secondary antibody was as follows. The blots were blocked overnight in the fridge with 5%  Transfer blots were run at a constant current of 150 Volts for 30 min in transfer buffer.
The staining procedure for the Western Blots is the same as for the LICOR antibody stained dot blots, described above.
In order to confirm bands found on Western Blots and to specify accurate   Figure 8).

Discussion:
In this study, we have presented the first biochemical evidence for a membrane In spiny lobsters, Panulirus argus, the aesthetasc sensilla are the sensory cilia of the olfactory receptor cells whose nuclei are located within the antennules themselves (Ache and Derby, 1985;Grunert and Ache, 1988). Histological studies shows that the aesthetasc sensilla are innervated by the dendritic extensions of multiple bipolar receptors, with the soma gathered in a cluster at the base of the sensillum inside the antennule itself (Ache and Derby, 1985;Grunert and Ache, 1988).
Although we have not studied the histology of the antennules in the American lobster, we have made the assumption that a similar morphological arrangement exists, and our spot blot data appears to confirm this. Since we shaved the sensilla off of the antennule at the base, we would have separated the cell bodies from the dendritic extensions that extend into the sensillum itself. Therefore, our GS and SE samples did not contain any cell bodies (nucleus or cytoplasm), and positive staining for the EcR must represent a membrane receptor, not a cytoplasmic one.
We found several molecular weights for the EcR, which suggest that different isoforms exist in lobsters. Isoforms of the EcR have been found in several insect and crustacean species, including Drosophila (Talbot et al., 1993), Manduca sexta (Fujiwara et al., 1995;Jindra et al., 1996), Bombyx mori (Kamimura et al., 1997) Crangon crangon (Verhaegen et al., 2011) and Homarus americanus (Tarrant et al., 2011). In Drosophila, three different isoforms exist for the EcR: EcR-A (91 kDa), EcR-B1 (93 kDa) and EcR-B2 (73 kDa) (Talbot et al., 1993). Ann Tarrant  whereas we obtained our results directly from dissected tissue. A possible reason for this could be due to alternative splicing of the EcR, which has been widely reported in crustaceans (Chung et al., 1998;Wu et al., 2004;Kim et al., 2005;Asazuma et al., 2007;Kato et al., 2007). Also, Tarrant et al. (2011)  and we used adult intermolt animals. In crustaceans, it has been shown that there is differential expression of the EcR over the molt cycle (Durica et al., 1999;Asazuma et al., 2007;Kato et al., 2007;Hirano et al., 2008). inside the cell and the phosphoinositide-3-kinase pathway is activated. When 20E or alpha ecdysone binds, the mitogen-activated protein kinase pathway is activated (Srivastava et al., 2005). With respect to the non-genomic functions of the ecdysone receptor, the presence of a putative membrane bound ecdysone receptor has been isolated from the plasma membrane of the anterior silk gland of the silkworm, Bombyx mori (Elmogy et al., 2004). The receptor (57 kDa) exhibited rapid saturable binding to PonasteroneA, and these quick association/dissociation kinetics are characteristic of membrane bound receptors, not nuclear/cytoplasmic receptors (Elmogy et al., 2004).
The presence of a membrane bound receptor for 20E in the silkworm correlates with our finding of a membrane bound receptor for 20E in lobsters. These membrane bound receptors could act through a second messenger pathway, as observed in Drosophila, or through another pathway initiated in the membrane.       4a. 4b.

Figure 5a.
Mass Spectrometry results for Walking leg pellet (membrane) fraction.

7a.
7b.  whether ecdysteroids also affect the closer muscle, a customized force transducer was constructed to measure the force and duration generated by the closer muscle of male and female lobsters after injection with alpha-ecdysone or 20E. The difference in force and duration before and after injection of 20E or alpha-ecdysone was compared to their differences after injection of saline. Alpha-ecdysone significantly increased the force generated by female crusher and cutter claws, and 20E also potentially increased the force in female crusher claws. The results suggest that circulating ecdysteroids influence the claw closer muscle and could be a factor influencing agonistic interactions.

Introduction:
Agonistic encounters play an important role in the life of lobsters; they are involved in procuring shelters, defending and maintaining those shelters, mating, and foraging success . Some of the factors affecting the outcomes of agonistic encounters include physical characteristics of the animals, such as weight, carapace size, and chelae size; larger animals weigh more and have greater carapace and chelae size . In staged encounters, larger lobsters will win significantly more encounters than smaller lobsters of the same sex .
Size also plays a role in the formation of dominance hierarchies in lobsters, as larger lobsters are dominant over smaller lobsters in social settings . This is also true in crayfish, where larger animals are dominant over and have more access to food resources than do smaller, subdominant males .
During an agonistic interaction, lobsters perform many behaviors with their claws, including grabbing, hitting, pinching and claw locks Reinhart et al., submitted;Sipala et al., unpublished). These claw behaviors are important, as the strength or duration of a squeeze/pinch may affect the outcome of the agonistic interaction. On average, male lobsters have larger crusher and cutter claws than female lobsters of the same carapace size . This is also true in crayfish, where for a given body length, males have larger chelae that generate a greater force than do the chelae of females of the same size (Wilson et al., 2009).
Along with body size, hormones have been shown to influence the outcome of agonistic interactions in crustaceans. In lobsters, the hemolymph titers of the molting hormone, 20-hydroxyecdysone (20E), varies over the molt cycle. Premolt animals (D 1 and D 2 stage animals about to molt) have increased levels of ecdysones in their hemolymph and urine , and are dominant over and more aggressive than their intermolt C-stage opponents in a confrontation . Evidence has been presented that injected 20E makes lobsters more aggressive; in staged combats between large and small lobsters, smaller lobsters injected with the hormone are significantly more aggressive than small control animals injected with saline . The injection of 20E may mimic the increased 20E titers that occur in D-stage animals about to molt (D1 and D2), which are correlated with increased aggression in D-stage animals .
One reason circulating 20E may affect the outcome of an agonistic interaction has to do with its effect on neuromuscular physiology. 20-hydroxyecdysone has been shown to affect the neuromuscular electrophysiology of the claw and abdomen in lobsters , as well as alter neurotransmitter release in crayfish . 20E decreases the probability of vesicular neurotransmitter release in the walking legs , and a mixture of 20E and serotonin increased neuron firing frequency in the slow-adapting muscle receptor organ (MRO) of the abdomen . In the claw opener muscle of lobsters, animals in premolt stage D produce significantly larger excitatory junctional potentials (EJP's) and significantly fewer inhibitory junctional potentials (IJP's) than intermolt animals .
In the presence of 20E (which is present in greater quantities in the hemolymph of premolt animals), there is an increase in EJP amplitudes and frequency of miniature excitatory junctional potentials (MEJP) in the opener muscle (used in threat displays) . There is also a decrease in EJP amplitudes in the abdomen (used in the escape response) , which corresponds to the effect of pre-molt and post-molt hemolymph on EJPs .
Ecdysteroids may also have differential effects on the neuromuscular properties of claw closer muscle of crusher and cutter claws, as the two claws differ in their muscle fiber types, motoneuron innervation and contractile properties . Fast twitch fibers quickly reach a maximal tension, which rapidly decays, while slow twitch fibers gradually increase in tension with a slow decay phase . Crusher claw closer muscles contain only slow twitch fibers with long sarcomeres, which allows the crusher to maintain force for a long period of time . Cutter claws contain mostly fast twitch fibers with short sarcomeres and a small ventral band of slow twitch fibers, which leads to a quicker fatigue than crusher muscle . The closer muscle in both crusher and cutter claws are innervated by the fast closer excitor neuron (FCE) and the slow closer excitor neuron (SCE), however, cutter closer muscles have mostly FCE while crushers have a mixture of both (Wiersma, 1955;. Generally, SCE synapses are more fatigue resistant and recover more quickly than FCE synapses, but the synapses at both the FCE and SCE in crusher claws are more resistant to fatigue than those in the cutter .
The purpose of this study was to determine whether alpha-ecdysone and its active principle, 20E, alter the squeezing properties of the crusher and cutter closer muscle in male and female lobsters.

Methods:
Animal procurement and maintenance: Male and female American lobsters, Homarus americanus, were obtained from local fisherman and the Rhode Island Department of Environmental Management from inshore waters off the coast of Narragansett Bay, RI. Animals were maintained in natural circulating unfiltered seawater tanks at the Narragansett Bay Campus, on a 12-hr light/dark cycle. Water temperature and salinity were ambient, ranging from 10-20°C and 28-33ppt, respectively. Animals were fed fish scraps, supplied by a local fish market, twice weekly. All lobsters used were intermolt C-stage animals in perfect condition, i.e., all eight walking legs, claws, antennae and antennules were intact, with no other signs of physical damage or shell disease. The tanks were compartmentalized, so animals could not physically interact with each other prior to use. After animals were weighed and measured, one claw was chosen to be the test claw and left unbanned for the entirety of the experiment. This allowed the claw to have free movement prior to the testing period in order to prevent atrophy of the claw.
The other claw was banded for the entirety of the experiment, which ensured that the unbanned claw was the only claw that could grab the force transducer used to measure force.

Preparation of test substances and injection protocol:
In order to prevent bias during data acquisition and analysis, one of us number coded stock aliquots of 20E, alpha-ecdysone and saline. Each number was then assigned to a given sex and claw type, ensuring that male and female cutters and crushers were allotted equal numbers of 20E, alpha-ecdysone and saline. The following experiments were performed: Male cutter and crusher with 20E, alphaecdysone, and saline and female cutter and crusher with 20E, alpha-ecdysone, and saline.
20E and alpha-ecdysone aliquots were made at a concentration of 1mg/ml, and frozen at -80°C. Saline aliquots had a composition in (mM/L) of: NaCl 472; KCl 10; MgCL 2 *6H 2 O 7; CaCl 2 16; glucose 11; Tris-maleate 10;pH 7.4 (Meiss and Govind, 1979) and were frozen at -80°C. Experimental animals were injected with enough gauge needle could be inserted into the remaining carapace layer. After injection, the hole was plugged with dental wax to prevent bleeding.

Force Measurements:
In order to measure the force generated by the claw, a customized force transducer with strain gauges leading to a customized Wheatstone bridge was constructed similar to that described by Wilson et al. (2009) and plugged into the Wheatstone bridge. In order to waterproof the transducer, shrink tubing was placed over every exposed wire connection point, and epoxy was layered over the strain gauges and shrink tubing to create a tight seal. The Wheatstone bridge was connected to a Power Lab (ADInstruments), which recorded the deflections in each strain gauge. The transducer was calibrated by hanging known weights from each squeezing surface to determine the change in voltage for each weight. These results were graphed and a best-fit line and equation were generated. Since there were two strain gauges attached to each offset mending brace, the total force for one squeeze was obtained by adding the force from each strain gauge. A sample Power Lab recording is provided (Figure 1).

Experimental protocol:
The squeeze injection protocol consisted of an initial pre-squeeze followed by 4 injections of 20E, alpha-ecdysone or saline 12 hours apart. The post-squeeze was taken 12 hours after the fourth injection. All pre-and post-squeezes were taken after 7PM.
A total of three pre-squeezes and three post-squeezes were recorded from each subject. During the acquisition of squeezes, lobsters were placed into 10-gallon opaque Tupperware bins filled with approximately nine gallons of water and allowed to acclimate for one hour. All experiments were performed under red light to mimic nighttime conditions. The squeezing protocol was as follows: the transducer was slowly lowered into the water and the lobster was allowed to grab and release the transducer. Once the lobster released the transducer, the transducer was slowly removed from the bin and re-introduced for the second squeeze one minute after the first squeeze ended. The third squeeze was recorded one minute after the second squeeze ended. In some instances, three squeezes could not be elicited from a lobster, or some squeezes were not forceful enough to be picked up by the force transducer.
The transducer was only presented three times to each lobster for a maximum of one minute each, with one minute in between each presentation. Only measurable squeezes were used for analysis, any instances where no measurable squeezes were elicited were thrown out. Because several trials had to be discarded, the following Values were considered significantly different at P < 0.05 and potentially significant (trend) at 0.07 < P < 0.05.

Results:
In female crusher claws, the average difference between post and pre-squeezes for animals injected with alpha-ecdysone (71.4 N) was greater than the average difference between post and pre-squeezes for those treated with saline (-55.5 N) (Student's t-test, P=0.009) (Figure 2). In female cutter claws, the average difference between post and pre-squeezes for animals treated with alpha-ecdysone (18.1 N) was greater than the average difference between post and pre-squeezes for those treated with saline (-17.7 N) (Student's t-test, P=0.024) (Figure 3). Potential significance was found in the force of female crusher claws treated with 20E compared to saline p=0.07). No other significant differences were found in crusher or cutter claws for any other parameter measured in males or females.
An analysis was made of all the pre-squeezes for all animals (control, before hormone injection) to compare the force and duration generated by crusher and cutter claws for males and females. The average maximum force generated by crusher claws (93.9 Newtons) was greater than that by cutter claws (46.7 Newtons) (Student's t-test, p=0.02), while no differences were found between maximum duration of the squeeze by crusher and cutter claws. Also, there were no sex differences in the average force or duration from cutter or crusher claws when all pre-squeezes were analyzed. The average force generated by male (43.1 N) and female (49.9 N) cutter claws were not significantly different (Student's t-test, P= 0.3), nor were the average force generated by male (93.1 N) and female (93.7 N) crusher claws (Student's t-test, P= 0.5).

Discussion:
Our results suggest alpha ecdysone causes a significant increase in the maximum force produced in female cutter and crusher claws. This increase in force caused by alpha-ecdysone may contribute to the fact that lobsters are more successful in agonistic interactions immediately prior to the molt. Levels of alpha-ecdysone and 20E spike immediately prior to molt  so increased levels of alpha-ecdysone may increase the force generated by claws, which could help the lobster gain an advantage in a fight. If a lobster is able to squeeze with a stronger force than its opponent, then the opponent may be able to sense that strength and withdraw from the confrontation. Intermolt lobsters, who have lower levels of circulating alpha-ecdysone, may produce a lower closing force than premolt lobsters, which could be a reason they lose more encounters with premolt lobsters. Since claws are an important factor in the outcome of an agonistic interaction, anything that changes the mechanisms of the claw could affect the agonistic interaction itself.
No differences were found in male cutter or crusher claws for force or duration with 20E or alpha-ecdysone. The fact that no differences were found in male lobsters with either alpha-ecdysone or 20E suggests that ecdysteroids have differential effects on the closer muscle in males and females. However,  showed that 20E has a significant effect on the claw opener and abdominal flexor muscles in male lobsters. Given that 20E affects the opener and abdominal flexor muscles in males, we expected the closer muscle to respond to 20E in males and females. There were no changes in the closing force among male claws, but there was a potentially significant response in females (see below). It is possible that differential sexual responses occur in the claw closer muscle, as differential responses to ecdysteroid exposure have been shown in previous studies Reinhart et al., submitted).  puffed 20E across the antennules of female lobsters engaged in an agonistic interaction, and Reinhart et al. (submitted) performed the same experiment with male lobsters. Females responded to 20E exposure by becoming more aroused, increasing aggressive, defensive and avoidance behaviors, whereas males simply increased the frequency of defensive behaviors. It is possible that a similar sex-dependent differential response to ecdysteroids may exist in the claw closer muscle.
One reason our results with 20E did not prove to be significant could be due to the high variance in our data. In female crushers injected with 20E, the average difference between the post and pre-squeezes was 109.2 N, which was not significant at the 5% level when compared to saline (-55.5N), although a trend or potential significance at the 7% level was demonstrated (Student's t-test, P=0.07). However, the mean difference in 20E was actually greater than the difference between alphaecdysone (96.9N) and saline (-55.5N), which was significant. The standard error in crusher claws treated with 20E was ±96.9 N, while the standard error in crusher claws treated with alpha was only ±39.8N. One cause of the large variance was the fact that there were instances where lobsters did not squeeze the transducer at all or squeezed with such low force that it did not register on the transducer. Further, the fact that differences in the crusher claw of females were potentially significant but that differences in the cutter claw were not could be attributed to the different muscle fiber types and motoneuron innervation in crusher and cutter claws, describe earlier.
A second reason that the effects of 20E were not significant may be due to metabolic considerations. Since we injected 20E into the abdomen, but tested the effects at the periphery, it is possible the injected 20E became degraded or metabolized before it reached its target area. In contrast, because alpha-ecdysone is converted into 20E in the peripheral tissues by 20-hydroxylase (Mykles, 2010), this conversion within the target tissue would result in an exposure to 20E in the muscle.
It is possible that the alpha-ecdysone we injected made its way into the peripheral tissues and was converted into 20E, and once converted, it may be the 20E that is having the effect on the closer muscle, and not alpha-ecdysone. In this regard, we have reported as unpublished observations in an earlier study that we were unable to show that alpha-ecdysone caused a change in the neuromuscular properties of the claw or abdomen . It is also possible that the injected 20E was excreted so that its concentration would have been too low to have an effect on the claw closer muscle.  perfused 20E directly over the neuromuscular preparation, so there was no metabolism or excretion of 20E before it could have an effect, unlike our injections.
Although our data had a high variance, analysis of all pre-squeeze data revealed similar patterns and trends for force observed in other crustaceans. In crustaceans, in general, larger animals have larger claws that generate greater force than smaller claws on smaller animals Vye et al., 1997;Wilson et al., 2009). In lobsters, Vye et al. (1997) found a general increase in contraction force with an increase in claw dimensions, although results in that study were also highly variable. Another characteristic of crustaceans is that crusher claws, or major chelae, generate a greater overall force than cutter, or minor, chelae Govind and Blundon, 1985;Vye et al., 1997). Our data indicates that crusher claws generate more force than cutter claws, on average; crusher claws have many slow twitch fibers with long sarcomeres, while cutter claws have fast twitch fibers with short sarcomeres . Finally,  did not find any significant sexual differences in the force generated by the claws, a finding that is corroborated by our data.
In summary, alpha-ecdysone causes a significant increase in the force of crusher and cutter claws of female American lobsters, and a potentially significant increase in the force of female crusher claws treated with 20E, suggesting that circulating ecdysteroids could be a factor that leads to the success of premolt animals in agonistic interactions.  Each window represent the force exerted on one of the two strain gauges. The force from each window was added together to obtain the overall force of the squeeze.  Tables  Table 1. Summary of Aggressive behaviors.
Values are means + SEM, N = 10. Significant P-values are in bold.
Since there are two squeezing surfaces and two strain gauges, the total force exerted by the claw was obtained by adding the force of each strain gauge. This accounts for the force that is exerted on each squeezing surface, and not just the top or bottom squeezing surface. In order to accurately asses the force, it is necessary to include the force exerted on both squeezing surfaces.

Analysis of integral of squeeze:
Along with force and duration described in the manuscript, we also analyzed the integral of the squeeze (area under the curve). The integral was calculated using the Power Lab integral function, and the maximum pre-squeeze integral was subtracted from the maximum post-squeeze integral. No significant differences were found for the integral for any parameters measured, however, a potentially significant trend was found in female cutter claws treated with 20E (Student's t-test, p = 0.06) (Table 1). These results suggest that 20E potentially lowers the force duration, which is in contrast the fact that the force itself is potentially greater in the presence of 20E, and that the duration was not significantly affected. Figure 1a.

Figures
Force transducer.
Force transducer is inside the Tupperware bin, connected to the Wheatstone bridges on the table.     Tables  Table 1. Summary of the average difference between post-squeezes and presqueezes for force, duration and integral.

Summary and Raw Data
Values are means + SEM. Force measured in pounds, duration measured in seconds and integral measured in lbs/sec. Significant Pvalues are in bold.  Table 2.
Raw data for female cutter claws treated with 20E.
Force is measured in pounds, duration is measured in seconds, and integral is measured in lbs/sec.  Table 3. Raw data for male cutter claws treated with 20E.
Force is measured in pounds, duration is measured in seconds, and integral is measured in lbs/sec.  Table 4. Raw data for female cutter claws treated with alpha-ecdysone.
Force is measured in pounds, duration is measured in seconds, and integral is measured in lbs/sec.  Table 5. Raw data for male cutter claws treated with alpha-ecdysone.
Force is measured in pounds, duration is measured in seconds, and integral is measured in lbs/sec.  Table 6. Raw data for female cutter claws treated with saline.
Force is measured in pounds, duration is measured in seconds, and integral is measured in lbs/sec. Table 7.
Raw data for male cutter claws treated with saline.
Force is measured in pounds, duration is measured in seconds, and integral is measured in lbs/sec.  Force is measured in pounds, duration is measured in seconds, and integral is measured in lbs/sec.  Table 9. Raw data for male crusher claws treated with 20E.
Force is measured in pounds, duration is measured in seconds, and integral is measured in lbs/sec.  Table 10.
Raw data for female crusher claws treated with alpha-ecdysone.
Force is measured in pounds, duration is measured in seconds, and integral is measured in lbs/sec. Table 11. Raw data for male crusher claws treated with alpha-ecdysone.
Force is measured in pounds, duration is measured in seconds, and integral is measured in lbs/sec. Table 12.
Raw data for female crusher claws treated with saline.
Force is measured in pounds, duration is measured in seconds, and integral is measured in lbs/sec. Table 13. Raw data for male crusher claws treated with saline.
Force is measured in pounds, duration is measured in seconds, and integral is measured in lbs/sec.