Abstract Intertidal organisms can spend a great deal of time out of water and must be physiologically adapted to tolerate terrestrial conditions. I hypothesized that one of the physiological adaptations made by intertidal organisms is reduction of metabolic rate during the low tide period. To test this hypothesis, I investigated the ability of the intertidal porcelain crab Petrolisthes cinctipe, to regulate its heart rates during the transition from water to air. Heart beats were monitored by impedance electrodes at 10°C, 15°C, and 20°C during one hour of immersion followed by one hour of emersion. Heart rates were recorded using a PowerLab data acquisition system and were averaged for each 10 minute period of the experiment. Linear regression analysis was used to determine if heart rate varied with body size using crabs with carapace width ranging from 9mm to 16 mm. A significant decrease in heart rate with increasing body size was found. Heart rates were corrected for body size in comparisons of the effects of emersion. No significant effect of emersion was found (p-value of 0.079.) The temperature dependence (Qio) of heart rate also did not differ between immersion or emersion. In future studies, variables such as size should be more strenuously controlled for, and this may yield clearer results. Introduction An important theme in evolution is the phenomenon of terrestrialization, whereby aquatic species developed abilities to live on land and respire in the air. The intertidal zone is an ideal location to explore mechanisms of terrestrialization because organisms inhabiting this habitat must be adapted to both land and water. Potential challenges that these animals face during transition from water to land include heat stress, desiccation, and difficulties in breathing because properties of air and water are vastly different (Burggren and McMahon, 1981). One possible adaptation to such stressful conditions made by marine organisms when adapting to land would be reduced respiration as a means of reducing loss of water of the respiratory surfaces, especially the gills. Lowering heart rate could be one mechanism for reducing the threat of desiccation. Arthropods have been used as model organisms for many studies of terrestrialization (Burggren, 1992). In this experiment, intertidal porcelain crabs will be explored. The effect of marine organisms adapting to land could be observed in the alternating tidal cycles, where twice daily low tides simulates the process of terrestrialization. The intertidal porcelain crab Petrolisthes cinctipes can be emersed during every low tide, and spend over half its life out of water. They are found on the Northeastern Pacific coastal shores living underneath rocks and in mussel beds at the mid intertidal zone. P. cinctipes are also regularly subjected to large changes in temperature during tidal cycles where, during emersion, temperatures may read 31’C. Their Arrhenius break temperature of heart rate in response to temperature is 31.5°C (Stillman and Somero, 1996). One adaptation to emersion in response to the difficulties of respiration in P. cinctipes is that these crabs possess a membraneous structure on the ventral merus on each walking leg, which functions as a respiratory structure (Stillman and Somero, 1996). What I want to investigate is how P. cinctipes are affected by alternating between air and water in these microscale simulations of terrestrialization. Are their metabolisms lowered? Cardiac output is a good indication of the rate of metabolism, and it has been seen to be generally lower in terrestrial crabs relative to marine species (Burggren, 1992). Thus, I asked if an effect of emersion in P. cinctipes could be a lowered metabolism indicated by a decrease in heart rate. Materials and Methods Collection and Care Specimens were collected during low tide from two intertidal sites, Fisherman’s Wharf in Monterey, CA and Hopkins Marine Station in Pacific Grove, CA. Exposed rocks that had crevices underneath were selected and turned over to collect crabs. The crabs were placed into a small cooler filled with seawater and transported to Hopkins Marine Station, where they were then placed into aquaria with flowing ambient seawater at 13+2’C. They were kept in a rectangular Rubbermaid tupperware container (21x13cm) that had its bottom and sidewalls cut out so that only its frame was being used. The walls were replaced with plastic wired mesh, which allowed flow of seawater through the system and provided crabs with support. The crabs were allowed at least 24 hours to acclimate before use in experimentation. Crabs were fed Cyclop-Eeze (Argent Laboratory, Redmond, WA) cultured micro-crustaceans every Monday, Wednesday, and Friday. Experimental Materials and Design Effects of emersion were observed at 10'C, 15°C, and 20'C. The crabs were placed into an experimental chamber constructed with a rectangular Rubbermaid ice chest (36x25cm). A refrigerated water bath (RC6 CS Lauda, Brinkmann, Germany) was used to keep the water temperature constant inside the chamber. Styrofoam lids were devised to loosely cover the water chamber to control the air temperature and keep the specimens inside moist when suspended in air by metal clamps. The inside of the chamber was lined with pipes and tubes in which anti-freeze circulated to control the temperature of the water bath. An aquarium air stone at the bottom provided oxygen. The chamber also contained two thermocouples calibrated to each other, one to measure the temperature inside the chamber, and one that was inserted into the crab. Before each run, the crabs were left undisturbed in the chamber in water for at least 15 minutes to exclude the influence of preparation activities and stimulation. Each run consisted of six crabs immersed in filtered seawater. Heart rate of immersed crabs was recorded for one hour, at which time the water was siphoned out to a set level to simulate low tide. The crabs were suspended in air for another hour. 12 crabs were used for each temperature, making a total of 36 runs. An experimental run with a fake "crab" crafted with a sponge whose soft side was cut into the average size of a crab was done to test for evaporative water loss. The imitation "crab" was weighed, left suspended in air for six hours, and then weighed again. Preparation of crabs Only male crabs were used. Females were not used because differences in the brooding cycles might have complicated the results. Small numeric paper tags were glued to the left claws of the crabs for identification. The tags did not hinder the crab's natural mobility or ability to pinch. The widths of carapaces were measured with a Vernier Caliper Type 6914 (Scienceware Bel-Art Products, Pequannock, NJ). Some crabs sustained sustained injury to their exoskeleton before the entire sequences of runs were performed, so replacement crabs were used. SuperGlue gel was used to attach a thin band of cork onto the back of each crab to allow metal clasps to grasp and suspend the specimen. Shaped like a long trapezoid, the narrow end of the cork was attached to the crab, whereas the broad and flat end was clasped. Pin-sized holes were made through the carapace on either side of the heart and impedance electrodes made of 0.025mm diameter ceramic-coated copper wire, whose tips had about Imm of the ceramic insulation removed, were inserted through each opening and glued in place (Stillman and Somero, 1996). Of each of the two sets of six crabs, number three received an additional pin-sized hole near the previous right one, into which a thermocouple was inserted to measure temperature inside the crab. The impedance electrode wires were connected to an impedance converter model 2991 (UFI, Morro Bay, CA), and the heart rate was monitored with PowerLab/16sp (AD Instruments, Pty, Ltd, Castle Hills, Australia). Data Analysis Heart rate was monitored continuously for the duration of each experimental run and expressed as average beats min“. Data were then averaged over ten minute intervals and graphs against time were generated. To better compare heart rates between immersion and emersion, the averages for each hour in air and water were calculated. To test for a correlation between crab size and heart rate, linear regression analysis was employed to generate regression lines for each temperature. The average of the three regression lines was then used to correct the heart rates for effects of size. An Analysis of Covariance was performed with the corrected data. The factors temperature and treatment (air or water) were orthogonal to each other, and n = 12 replicates. The heart rates were transformed into square roots before this test was performed to gain homoskedasticity. Qjos were also calculated and compared between 10°C to 15°C and 15°C to 20°C in air versus water. One other possible statistical test that had been considered was a Chi Square. However, this could not be applied to a two by three model, in this case two treatments (air or water) x three temperatures. If performed, the difference in heart rate between air and water would have been compared within each temperature separately. This would have necessitated three separate Chi Square tests, one for each temperature. I understood that this would have a lower statistical power than an Analysis of Covariance. To get an idea of what the results might have been, single factor Anovas were performed with each temperature orthogonal to treatment, and no statistical significance was seen in any of the three temperatures. Therefore, an Analysis of Covariance was chosen. Results Temperature The thermocouple inside the chamber showed no change in temperature throughout each experiment, whether done in water or in air. Similarly, the thermocouple inserted into the crab revealed that temperature remained constant. When evaporative cooling was explored, the initial weight was 4.2340.2g, and after six hours suspension in air, 4.7740.2g. This demonstrates no net water loss by the crabs during the experiments. Effects of emersion Plots of heart rate versus time (figs 1, 2) generally showed a pattern in which the initial heart rate under immersion (i.e. the very first data point) was higher than the final heart rate in air at the end of the experiment. 86% (30 of 36 runs) indicated decreased heart rates in the comparison, and of the six that did not, one came from 10°C, two from 15°C, and three from 20°C. There was also consistently a great of deal of variation in heart rate, sometimes over a 3-fold difference both within each crab and amongst crabs overall. These trends were seen at all temperatures. To more easily compare differences, the heart rates for each hour in air and water were averaged (figs 3, 4), and 86% (30 of 36 runs) decreased during emersion. Of the six crabs’ whose heart rates increased, one was from 10°C, two from 15°C, and three from 20°C. These were not the same crabs that exhibited higher rates at the end of the experiments when the initial and final rates were compared (see above). The average heart rate in water was divided by the average heart rate in air to obtain percentages of overall difference (fig. 5). The percentage differences amongst crabs ranged from 30.69% increase in heart rate to a 38.48% decrease. When all 36 experimental runs were averaged, a net decrease of 9.22% was calculated. Effects of size The carapaces of the crabs ranged in width from 9mm tol6mm and averaged 13mm. Linear regression analysis comparing carapace size to heart rate showed a significant relationship at all temperatures. For example, at 15°C, the P-value was 0.001, and R°= 0.463 (fig. 6). To incorporate the effects of size into the analysis of the effects of emersion on heart rate, the average of the regression lines for all three temperatures was used to correct for size. Heart rate data were square root transformed for this analysis. When size-corrected, the heart rate data did not show a difference between conditions of immersion and emersion (P-0.079). Effects of temperature Looking for differences in temperature dependence of heart rate between immersion and emersion, Qjos showed no differences for either 10°C to 15°C or 15°C to 20°C (fig. 7). Discussion It has been seen in previous studies that heart rate is directly related to temperature in a substantial number of crustacean species (Stillman and Somero, 1996; McGaw, 2002) This was one of the main reasons why the experiment was performed at 10°C, 15°C, and 20°C. Not only was I investigating whether or not heart rates decreased as an effect of emersion, but also, if this were to occur, could the amount of heart rate dependence upon temperature between air and water be different at different temperatures? This relationship was analyzed with Qjos, and it indicated no such difference. Another aspect that arose due to the specimen’s heart rate dependence on temperature was whether or not temperature was absolutely controlled for throughout each entire experimental run. If not, this could have confounded the results. As mentioned previously, a general decreasing, albeit non-significant, trend had been seen in heart rate as the crabs were moved from water to air, and my hypothesis would suggest that this was due to the effects of emersion. However, one alternative hypothesis is that if the crabs had consistently experienced even a slightly lower temperature while suspended in air, then this could very well have been responsible for the perceived decrease in heart rate. Hence, controlling the temperature both in air and water was vital. There were many things that could have affected the temperature that the crabs experienced when suspended in air. A Styrofoam lid was crafted to cover the top of the chamber, and in this way separate to some extent the air inside from out. The cover should also have been effective in shielding most random breezes. However, it was devised to fit loosely over the top because sufficient airflow was needed to facilitate oxygen circulation to prevent hypoxia. To test the efficacy of the lid in controlling temperature, a thermocouple was immersed in water, and then left suspended in air when low tide was simulated during the experimental runs. Because no difference was seen between immersion and emersion, effective control of temperature by the lid was demonstrated. One other concern regarding the temperature that the crabs experienced in the air was whether or not evaporative cooling was occurring. If this were the case, then the temperature that the crabs experienced would have differed from the set air temperature. To test for evaporative cooling, the control experimental run was performed with the fake crab". The lack of water loss verified that evaporative cooling was not a contributing factor. One other affirmation that evaporative cooling was not occurring was that the thermometer inserted inside the crab also showed no signs of change when emersion occurred. Note that although evaporative cooling was controlled for in the experimental runs, this does not preclude that crabs do undergo evaporative cooling in their natural environment. There was a great deal of variation in heart rate: up to 3-fold differences were recorded both within each crab and amongst different crabs. Variation amongst different crabs tended to lower the power of statistical tests such as Analysis of Variance. One way in which this was slightly corrected for was by performing instead an Analysis of Covariance. This was only made possible by the fact that a correlation was found between the sizes of the crabs and their heart rates. Linear regression analysis was performed demonstrating this correlation’s statistical significance, and the average of the regression lines was used to correct for the heart rates before analyzing the data and square root transforming it. Although this new P-value of 0.079 was considerably lower than the one obtained from a regular Anova, it is still not statistically significantly different. Conclusions Although it has been seen that cardiac output in terrestrial crabs is generally lower than in marine crabs, this decrease was not detected in P. cinctipes as an effect of emersion. The natural variation in an organism’s heart rate was unavoidable, and this made comparing heart rates between immersion and emersion a challenge. Additionally, metabolic rates in general can be strongly influenced by other innate properties such as body size, physiological state (e.g. molting stage), and activity level (Burggren, 1991). This makes it difficult to conclude that because a statistically significant decrease in heart rate was not found, the metabolic rate did not decrease either. Instead, it has been determined that more stringent size control for the crabs is needed to weed out some of the natural variation in heart rate, and in future studies, a larger number of replicates should also be employed to increase the statistical power of the experiment. Acknowledgements I thank Prof. George Somero for his guidance and support. He asked just the right thought provoking questions and gave a continuous string of intellectual insights. Special thanks to Prof. Jonathon Stillman for his time and dedication, without which this never would have been accomplished. Likewise, the patient assistance and encouragement from Prof. James Watanabe was invaluable. This work was supported by Hopkins Marine Station of Stanford Uiniversity, and so I would like to recognize all of its wonderful staff, students, and professors. Literature cited Burggren, W. W. (1991). Does comparative respiratory physiology have a role in evolutionary biology (and vice versa)? In A. J. Woakes, M. K. Grieshaber, and C.R. Bridges (eds.), Comparative insights into strategies for gas exchanges and metabolism, pp. 1-13. Cambridge University Press, Cambridge Burggren, W. W. (1992). Respiration and circulation in land crabs: Novel variations on the marine design. Am. Zool. 32, 417-427. Burggren, W. W. and McMahon, B. R. (1981). Oxygen uptake during environmental temperature change in hermit crabs: adaptation to subtidal, intertidal, and supratidal habitats. Physiol. Zool. 54, 325-333. McGaw, I. J. (2003). Behavioral thermoregulation in Hemigrapsus nudus, the amphibious purple shore crab. Biol. Bull. 204: 38-49. Stillman, J. H. and Somero, G. N. (1996). Adaptation to temperature stress and aerial exposure in congeneric species of intertidal porcelain crabs (genus Petrolisthes): correlation of physiology, biochemistry and morphology with vertical distribution. J Exp. Biol. 199:1845-1855. Figure legend Fig. 1. Average heart rates versus time in a given experimental run for specimens 1 to 6 at 15°C. Fig. 2. Average heart rates versus time in a given experimental run for specimens 7 to 12 at 15°C. Fig. 3. Average heart rates during immersion compared to emersion for specimens 1 to 6 at 15°C. Error bars are standard errors of the mean. Fig. 4. Average heart rates during immersion compared to emersion for specimens 7 to 12 at 15°C. Error bars are standard errors of the mean. Fig. 5. Ratios of average heart rates during immersion divided by average heart rates during emersion to display percent differences for specimens 1 to 12 at 15°C. Fig. 6. Q1o values for air and water in 10°C to 15°C, and 15°C to 20°C. Error bars are standard errors of the mean. Fig. 7. Average heart rates in a given experimental run versus crab size for specimens 1 to 12 at 15°C and linear regression line. 5 3 8 8- Heart rate (beats min *) O O A- S 14 8 0. o 8 OGAON 3 8 100 )O Heart rate (beats min *) 9 15 5 e 8aaaa- Heart rate (beats min o 5 wate Heart rate (beats min *) 5 percent below 100 indicates decrease heart rate in air 8 Qjo values 8 8 20 0 Heart rate (beat min") *. * . *.. / . 4