Abstract 1 quantified the spatial distribution, physiological tolerances, and respiration rates of larvae of the intertidal fly, Oedoparena glauca. O. glauca larvae consume the barnacle, Balanus glandula, and use the shell for protection and pupation. The percent of barnacles infested with larvae and pupae was greatest in a high-intertidal crevice (29%), moderate on upper north-facing intertidal walls (-3%), and zero at lower tidal heights and on south-facing walls. Thermal stress, high salinity, and submergence are physiological stresses imposed on the larvae and pupae due to life in the intertidal zone. In laboratory tests of thermal tolerance, all larvae died following 3 hours of exposure to 39°C, but tolerated temperatures 838°C. Pupae appeared more sensitive to thermal stress, with 0% eclosion by pupae treated at 35°C, 38°C, and 41°C for four hours. Intertidal datalogger records and field measurements of internal B. glandula temperatures exceeded 36°C, near the thermal limit of O. glauca larvae. In addition to this thermal stress, during periods of aerial exposure, B. glandula was significantly more saline in vivo than the surrounding seawater, perhaps due to evaporative cooling by the barnacle. In tests of submergence tolerance, larvae survived 8 days of submergence in filtered seawater held at 13°C. but all died after 10 days. The pupae were apparently more sensitive to submergence; 40% of pupae submerged for two hours eclosed within 16 days, whereas there was only 12.5% eclosion of pupae submerged for 24 hours. The larvae were observed not to respire during six hours of submergence. However, the larvae respired at a rate of 0.4317 umol Oy/hr/g, over four days of submergence. This study illustrates that O. glauca larvae are hardy and capable of tolerating thermal stress and submergence. In contrast, the pupae may be more sensitive to these environmental conditions, perhaps restricting their geographical distribution. Introduction The rocky intertidal zone experiences large temperature fluctuations and extended periods of submergence. This environment is extremely stressful for a terrestial insect, and may explain why insects rarely inhabit and reproduce in the intertidal zone. One little studied terrestial insect that occupies this niche is the intertidal fly, Oedoparena glauca. It is the first and only confirmed record of a dipterous predator of barnacles (Burger et. al 1980). O. glauca larvae prey on the acorn barnacle Balanus glandula. O. glauca larval development consists of three instars (Burger et. al 1980). This report focuses on the second and third larval stages and the pupae (Figures 11 and 12). The adult flies are 5-Omm in length (Figure 10), and feed primarily on diatoms. O. glauca inhabits the Endocladia-Balanus zone within the intertidal zone. There are only two other reports of Diptera associated with barnacles, both from Europe. Roubond (1903) found fly larvae among the Semibalanus balanoides. Mercier (1921) observed larvae living within barnacles, and hypothesized that these larvae were also predaceous on barnacles. Neither account reared larvae to adults, so the identification of the species was not certain. There are many unusual stresses that the O. glauca must overcome to survive in intertidal habitats. Glynn (1965) determined that the center of the Endocladia-Balanus zone may spend 27% percent of the time submerged. For a terrestial insect, this is a tremendous amount of time to be submerged in saline water. Furthermore, temperature may fluctuate a great deal in the intertidal zone, and change rapidly depending on the tide. Since the adult cannot tolerate submergence, O. glauca must coordinate behavior such as oviposition with periods of low tide emersion. 1 quantified the physiological tolerances of the second and third instar larvae and pupae. Although the mechanisms for the tolerance of much of the physiological stress is unknown, it is clear that the larvae are capable of withstanding severe stress. The pupae appear much more sensitive to thermal stress and submergence than the larvae. These tolerances may have important effects on both the local and geographical distribution of this species. Materials and methods Organisms Pupae and second and third instar larvae of Oedoparena glauca were collected from high intertidal crevice habitats around a surge channel on the wave exposed point at Hopkins Marine Station in Central California (Figure 13). This channel contains many wave-force meters installed by Mark Denny and his lab group. Collected larvae and pupae were placed in petri dishes with raised lids. A small opening was kept in the Parafilm sealing the petri dishes to prevent anoxia. The larvae and pupae were kept moist and held in air at approximately 15°C above a flowing seawater tank. During the treatments of the larvae, the larvae were not fed. Wet masses of second and third instars were between 0.0020g and 0.0170g. Wet weights of the pupae were approximately 0.0150g. Spatial distribution The abundances of larvae and pupae were surveyed in seven habitats around the wave-exposed surge channel. These seven habitats were located in a high intertidal crevice and on north and south-facing walls at low, mid, and high tidal heights. The tidal heights were obtained by surveying the middle of these zones (Table 1). From these seven habitats, 100-150 acorn barnacles, Balanus glandula, were haphazardly removed from the rock with a screwdriver. To determine percent infestation by Oedoparena glauca larvae and pupae within these habitats, the barnacle body was removed from each B. glandula and the contents of each barnacle shell were scrutinized in the laboratory. Temperature loggers Temperature dataloggers (Optic StowAway, Onset Computer Corp., Pocasset, MA) were installed in the high and low, north and south-facing habitats from 28 May 2001 thru 1 June 2001. A smaller temperature datalogger (Tidbit, Onset Computer Corp.) was placed in the high crevice for the same period. All of the dataloggers were set to record the ambient temperature every ten minutes. Barnacle Temperatures Barnacle temperatures were taken periodically in the field from 30 May 2001 through 1 June 2001. The internal temperature of the barnacle was taken by opening the opercular plates with a scalpel and inserting a thermocouple inside the shell. Ten barnacle temperatures were taken at each site during each sampling. Barnacle temperatures were also taken periodically in a 10 cm radius around one temperature datalogger to test whether barnacle temperatures were correlated with datalogger records Thermal tolerance A 1.8 m long aluminum block was used to establish a thermal gradient to test the tolerances of the larvae and pupae. On one side of the aluminum block, a heater was attached, and on the other side a water bath. The temperature set points were manipulated until a preferred thermal gradient was maintained. The aluminum block was drilled with holes, four across and 31 lengthwise, so that 2.0 mL Polypropylene flat top microcentrifuge tubes could be inserted within the aluminum block. The microcentrifuge tubes were punctured on the cap to prevent anoxia. The upper thirds of the tubes were filled with tightly fitting foam cores that had been soaked in a solution of filtered seawater and antibiotics (15 mg/L penicillin-G and 15 mg/L streptomycin). Three thermal tolerance experiments were performed on the second and third instar larvae. Five different temperatures were chosen per experiment. The first of three experiments had temperatures of 24°C, 28°C, 32°C, 36°C, and 40°C. The second had temperatures of 32°C. 33°C, 34°C, 35°C, and 36°C. In the final experiment, larvae were maintained at temperatures of 36°0, 37°C, 38°C, 39°C, and 40°C. A control tube with moistened foam was placed in the fourth hole at each experimental temperature along the block. A thermocouple was inserted through a hole in the top of these tubes, and the temperatures were monitored to assure stable conditions during the experiment. Larvae were exposed to experimental temperatures for three hours, followed by a three hour recovery period. During the recovery period, the larvae were placed into petri dishes covered with Parafilm through which a small hole had been made and held at 15°C above the seawater tank. Responsiveness to stimuli immediately following the treatment and following the recovery period were scored. In a similar experiment, pupae in microcentrifuge tubes with seawater moistened foam cores (see above) were exposed to 35°C, 38°C, or 41°C for four hours (n = 3 pupae x 3 tubes per temperature). Following the period of exposure, pupae were placed in petri dishes with raised lids and held at 15°C above the flow tank. Percent eclosion was scored after 16 days. Submergence tolerance To test submergence tolerance of the second and third instar larvae, larvae were placed into 2 mL screw top glass vials that were filled with a solution of filtered seawater and antibiotics (15 mg/L penicillin-G and 15 mg/L streptomycin). A single larva was placed within each of three vials per treatment. The vials were submerged in a 13°C water bath and maintained up to (but not submerging) the level of the bottom of the air-permeable screw top. Larvae were submerged in 6, 12, 24, 48, 96, 144, 192, 240, and 264 hour treatments. Following submergence, the larvae were allowed to recover for three hours. During the three hour recovery period, the larvae were placed inside petri dishes and maintained at 15°C, as described above. As before, responsiveness to stimuli was scored immediately following the treatment and following the recovery period. Pupae were subjected to a similar esperiment. Pupae were placed into plastic and nylon mesh tea strainers (PermaBrew Tea Infuser, Upton Tea Imports, Hopkinton, MA). A large beaker was filled with the solution of filtered seawater and antibiotics (see above), and the tea strainers were submerged at 13°C for either 2 hours or 24 hours (n = 8-10 pupae per treatment). Following the submergence treatment, the pupae were placed inside petri dishes at 15°C and percent eclosion was quantified after 16 days. Respirometry These respirometry trials were designed to test if the larvae were respiring during short and long submergence periods. A water-jacketed microrespiration chamber was attached to a water bath and maintained at 13°C. Data were recorded using PowerLab (model 8SP, ADInstruments, Mountain View, CA) and the PowerLab Chart Program. Filtered seawater treated with antibiotics (15 mg/L penicillin-G and 15 mg/L streptomycin) was aerated with an air pump prior to setting 100% 02 saturation. Six gas-tight test tubes (-3 mL volume) were then filled with the treated seawater. In the six hour trial, four of the six tubes had two larvae placed in each and two tubes were used as controls. In the four day trial, three of the tubes had two larvae in each and three tubes were used as controls. All tubes were maintained in the dark at 13°C. At the end of the treatments, samples of water from each tube were injected into the microrespirometry unit and percent O2 saturation was quantified. Barnacle Osmolarity On 31 May 2001, two B. glandula in each habitat were sampled near the time when they were likely experiencing their maximum daily temperatures for that day. Immediately following removal from the rock, internal fluid was extracted from the barnacle using a 1 mL syringe and 20G1 PrecisionGlide needles. As a control, a sample of untreated seawater was also collected in the Hopkins Marine Life Refuge. The barnacle fluid and seawater were analyzed in a Wescor 5500 Vapor Pressure Ösmometer. Differences in osmolarity among the fluid samples and the seawater were analyzed using a t-test. Results Spatial distribution The highest infestation of O. glauca larvae and pupae was in barnacles of the high zone crevice (Figure 1). Infestation was moderate (2.7%) in the north-facing high habitat and in the north- facing mid-habitat (3.8%). There was no infestation in any of the south-facing habitats and zero infestation in the low north-facing habitat. Barnacle field temperatures The dataloggers recorded temperatures above 35°C in the south-facing low and high habitats as well as in the crevice habitat (Figures 2 and 5). The maximum recorded temperatures occurred on 31 May 2001 (Figure 3). Highly stressful temperatures, above 35°C, were recorded by the dataloggers only in the south-facing and the high crevice habitats. A tight correlation was found between datalogger readings and the internal temperatures of nearby barnacles (Figure 4, R2 = 0.7576), indicating that the datalogger records were representative of the temperatures that larvae of O. glauca would experience within these habitats. Thermal tolerances O. glauca's lethal thermal limits were 38-39°C, with zero larvae responding to stimulus immediately following the 38°C treatment, and 67% responding following the three hour recovery period (Figure 6). All larvae subjected to treatments 2 39°C were unresponsive immediately following the treatments, and none recovered. Temperatures of 36°C also seemed stressful to the larvae; although all larvae were able to recover, 66% were unresponsive immediately following exposure to 36°C The pupae of O. glauca seemed more sensitive than the second and third instar larvae. No pupae had eclosed in the 16 days following the three treatments (35°, 38°, 41°C). Although larvae were able to tolerate temperatures as high as 38°C, the pupae appeared unable to withstand the 35°C treatment. Submergence tolerances The second and third instar larvae of O. glauca were all responsive following submergence for as long as four days (Figure 7). After six days submergence, 33% of the larvae were unresponsive following the treatment and were unable to recover. Following eight days of submergence, zero larvae were responsive immediately following submergence. However, following the three hour recovery period, 67% were responsive. All larvae were unresponsive following treatments of 10 days or longer and no larvae recovered from these treatments. Again, pupae of O. glauca appeared more sensitive than the second or third instar larvae. 40% of pupae had eclosed in the 16 days following the 2 hour treatment, compared with 12.5% eclosion following the 24 hour treatment. Respirometry No measurable respiration was found during a six hour period of submergence. The O, saturation did not differ among control tubes and tubes with larvae. In contrast, after the four day trial, a significant percentage of O2 was removed from the tubes containing larvae. The average respiration rate was 0.4317 umol/hr/g. Barnacle Ösmolarity The osmolarity of the barnacle fluid collected from the seven habitats on 31 May 2001 was significantly greater than that of the seawater samples collected on the same day (Figure 9, t-test. p50.001). Discussion O. glauca, a terrestrial insect, is highly unusual in that it has been able to establish a niche in a marine environment by having larvae and pupae tolerant of physiological stresses associated with the intertidal zone. The results of this study suggest that the second and third instar larvae of O. glauca are more tolerant to these physiological stresses than are the pupae. This difference is most evident in the thermal tolerance responses of the larvae and pupae. Although larvae showed significant signs of stress, they were able to tolerate temperatures as high as 38°C. In contrast, pupae did not eclose following exposure to either 35°C or 38°C. Pupae may take 28 days to eclose (Burger et al. 1980) so the results may not have shown all pupae that would have eventually eclosed (age of pupae used in the study was unknown). However, many pupae did eclose (40%) in the 2hr submergence treatment, indicating the lack of eclosion was a real effect, It is possible that the rapid development within the pupae leading up to eclosion makes the pupae more susceptible to thermal stress. Pupae were most abundant in the crevice habitat that also had the longest exposure to temperatures that appear lethal to a pupa in the laboratory. It appears that on these warm days, the larvae are also living in a hypersaline barnacle, perhaps due to evaporative cooling by the barnacle. However, the temperatures recorded by the datalogger may not be indicative of the internal barnacle temperature in the high crevice. The thermocouple on the datalogger may have been too close to the rock and may not have accurately recorded the ambient temperature. In addition, a different datalogger model was required for the narrow crevice, and this smaller instrument may have had different thermal properties than the other four dataloggers. If this is not the case, it is unclear how the pupae deal with these elevated temperatures in the field, because they were unable to do so in the lab. It is also uncertain why these sites with high temperature exposure are chosen by the reproductive female. When submerged, larvae of O. glauca do not appear to respire initially, but do respire with extended submergence. It is unclear whether the onset of respiration is delayed, or whether the initial respiration was too small to be detected above the noise of the system. The mechanism by which respiration occurs during submergence is unknown in these larvae. Further work is needed to elucidate how these larvae respire in both air and water. With larvae and pupae of O. glauca able to tolerate long durations of submergence, one might predict that high numbers of these individuals would be found in the low zone. This seems even more likely based on the sensitivity of larvae and pupae to periods of thermal stress, which are generally more severe at higher tidal heights. However, although there appears to be a preference for northern exposure over southern exposure, both low habitats showed zero infestation by O. glauca. Newly hatched O. glauca flies may require time to harden their wings. to become competent to fly. This phase of their maturation may be difficult in the low zone due to shorter periods of aerial exposure. Furthermore, pupae are more sensitive to submergence, making distribution higher in the intertidal zone favorable. These facts seem consistent with Burger et al.’s (1980) observation that pupae in WA state were distributed higher in the intertidal zone than the larvae. Pupal distribution may also be decreased in the low zone due to decreased barnacle density, resulting in a less attractive target for ovipositing females. The southern geographical limit of O. glauca is reported to be in Central California (Burger et al. 1980). This boundary could arise from the pupae living near their lethal limit, especially on unseasonably warm days. This southern geographical limit may be established by the availability of reproductive sites that are high enough in the intertidal zone to minimize submergence, yet are not exposed to temperatures above the lethal limit of the pupae. Acknowledgements 1 would like to thank Eric Sanford for all of his generous help, time, attitude, and teaching. I would also like to especially thank George Somero and Jim Watanabe for all of their advice, expertise, and criticism. Jason Podrabsky and Peter Fields also added valuable assistance in the lab, and Chris Harley (University of Washington) provided valuable references and insight during the inception of this project. Finally, I would also like to thank all of the 175H instructors for making this a tremendous experience. Literature Cited Burger, JF, JR Anderson, MF Knudsen. 1980. The habits and life history of Oedoparena glauca (Diptera: Dryomyzidae), a predator of barnacles. Proc. Entomol. Soc. Wash. 82(3): 360- 37. Glynn, PW. 1965. Community composition, structure and interrelationships in the marine intertidal Endocladia muricata - Balanus glandula association in Monterey Bay, California. Beaufortia 12: 1-198. Mercier, L. 1921. La larve de Limnophoro aesruum Villen., Diptere marin. Compt. Rend. Acad. Sci. Paris 173: 1410-1413. Rouboud, M. 1903. Sur les larves marines de Dolichapodes attribueés au genre Aphrosylus (Walker). Bul. Mas. Nat. 9: 338-340. Tables Table 1: Tidal heights in feet above Mean Low Low Water for the seven intertidal habitats sampled for fly larvae and pupae. Habitat Tidal Height (ft. above MLLV) 5.49, 4.79 North-facing low wal North-facing mid wall 6.61 North-facing high wall 765 Crevice 7.10 430, 5.03 South-facing low wal South-facing mid wall 5.33 South-facing high wall 5.91, 5.82 14 Figure Legends Figure 1. Infestation of barnacles (B. glandula) in the seven intertidal habitats: a high crevice, low, mid, and high tidal heights on north and south-facing walls. Figure 2. Datalogger temperatures for 30 May 2001 through 1 June 2001 in the high zone crevice, and low and high zones on north and south-facing intertidal walls at the Hopkins Marine Life Refuge. Figure 3. Temperatures for 31 May 2001, the day with the highest peak temperatures for the period 30 May 2001 thru 1 June 2001. Figure 4. Calibration curve for datalogger temperatures versus the measured internal temperatures of barnacles (B. glandula) within 10 cm of the south-facing high zone datalogger. Figure 5. Time of exposure in the high zone crevice, low and high zones of north and south- facing walls to temperatures above 30°C, 33°C, and 35°C for the period 30 May 2001 through 1 June 2001. Figure 6. Thermal tolerance of O. glauca larvae. Percentages of larvae responsive immediately after temperature treatment (dotted line), and final survival after a three hour recovery period (solid line) (n = 3-9 larvae per termperature) Figure 7. Submergence tolerance of O. glauca larvae. Percentages of larvae responsive immediately after submergence treatment (dotted line), and final survival after a three hour recovery period (solid line) (n = 3 larvae per submergence treatment). Figure 8. Percentage of pupae eclosed after 16 days following submergence treatments of 24 hours and 2 hours (n = 8-10 pupae per treatment). Figure 9. Difference in osmolarity between barnacle fluid and seawater samples. Error bars are standard errors of the mean. Figure 10. View of an Oedoparena glauca fly that eclosed in the laboratory Figure 11. View of a third instar Oedoparena glauca larva. Figure 12. View of a pupa of Oedoparena glauca. Figure 13. Map and picture of the study site located within the Hopkins Marine Life Refuge, Pacific Grove, CA. Figures Figure 1 Infestation of Barnacles on North vs. South-facing Wave Exposed Vertical Walls SW high vertical wall %larvae %pupae SW mid vertical wall SW low vertical wall High crevice NW high vertical wall N mid vertical wall Nlow vertical wall 5 20 25 30 35 Percent Barnacles Infested 16 Figure 2 Field Temperatures at Hopkins Marine Life Refuge — Crevice - North-facing High 35.00 North-facing Low -South-facing High 30.00 - South-facing Low 25.00 20.00 15.00 10.00 5/28/01 12:00 5/29/01 12:00 5/30/01 12:00 5/31/01 12:00 6/1/01 12:00 6/2/01 12:00 AM AM AM AM AM AM Time 17 Figure 3 Field Temperatures on 31 May 2001 -Crevice 35.00 North-facing High — North-facing Low 30.00 -South-facing High - South-facing Low 25.00 20.00 15.00 S 10.00 5/31/01 12:00 5/31/01 4:48 5/31/01 9:36 5/31/01 2:24 5/31/01 7:12 6/1/01 12:00 PM AM PM AM AM AM Time 18 Figure 4 34.0 29.0 24.0 19.0 14.0 10.00 Datalogger Temperatures vs. Internal Barnacle Temperatures V=0.9353x + 1.9106 R2 = 0.7576 15.00 25.00 20.00 30.00 Datalogger Temps (°C) 35.00 Figure 5 Crevice Variation in Thermal Exposure Among Intertidal Habitats EAbove 30°C HAbove 33°0 EAbove 35°C North-facing Low North-facing High South-facing Low South-facing High Habitats Figure 6 100 75 25 Thermal Tolerance of Larvae in Laboratory Gradient = - % Responsive after thermal stres % Survival after recovery 36 38 40 32 34 24 26 28 30 Temperature (?C) Figure 7 Submergence Tolerance of Larvae 100-— % Survival after recovery 90 - % Responsive after submergence 80 60 40 30 20 k 50 100 150 250 200 Time (hrs) 22 Figure 8 Percentage of Pupae Eclosed following Submergence Treatments 15 35 40 45 O 5 10 20 30 % Eclosion after 16 days 23 Figure 9 850 900 Barnacle Salinity vs. Sea Water Salinity □Sea Water Osmolarity EBarnacle Fluid Ösmolarity 1050 1000 1100 950 Ösmolarity (mmollkg +SE) 24 1150 1200 igure 10 Figure 11 2 mm Figure 12 2 mm Figure 13 Prg ** ä 0 70 + Sel s 2 28