Abstract The mysid shrimp A. sculpta was observed to associate closely with the kelp canopy of Macrocystis beds during the day. At night but almost never during the day, it was found in plankton tows through open water adjacent to the kelp forest, suggesting nocturnal movement away from kelp. In constant light A.sculpta remained closely associated with kelp in aquaria and virtually never moved into adjacent open water. When subjected to a light-dark 14:10 regime which coincided with day and night, more mysids moved away from kelp at night than during the day. In constant darkness A. sculpta showed an endogenous circadian behavioral pattern of close association with kelp during the subjective day and movement away from kelp during the subjective night. This behavior is discussed in terms of possible selective advantages, such as avoidance of predation and feeding on zooplankters which are present in the water column at night. Introduction A. sculpta is a mysid shrimp, about lem long, which inhabits the kelp canopy of Macrocystis beds in California. Limbaugh (1955) and Clarke (1971) observed this crustacean in the southern part of its range. They found it associates closely with kelp blades in the canopy where its behavior and cryptic coloration render it inconspicuous. Concealment is essential for its survival, since many species of fish prey upon it (Clarke, 1971). Its complex forward appendages enable it to filter feed on phytoplankton and other particulates (Banner, 1948; Tattersall, 1951). It is also known to consume pieces of kelp and to prey on zooplankters such as copepods (Hobson, 1976). There is evidence that A. sculpta does not always maintain close association with kelp since it was found in plankton tows through open water adjacent to kelp forests (Hobson, 1976; Miller, 1976). During preliminary studies I became interested in the suggested horizontal migration of A. sculpa, and I undertook this research to investigate a possible diel pattern of close daytime association with the kelp canopy followed by movement away from kelp at night. As well as field studies, I carried out laboratory experiments in aquaria where the habitat was simulated in miniature and stocked with mysids. Materials and Methods Studies were conducted at Stanford University's Hopkins Marine Station in April and May of 1978. Collections and field studies were conducted near a fixed buoy 50M from the edge of the Station's south kelp forest and in an adjacent sand channel. Surface tows with a 333um mesh zooplankton net, O.5M in diameter, were made through the canopy and in open water 10-20M from the kelp bed. A rowboat was used to tow the net in one study and skin divers towed it in another. The mysids collected were placed in aquaria with running sea water. Half of each aquarium contained suspended blades of Macrocystis pyrifera while half consisted of an open water section. Three light regimes were used: 1) a light-dark cycle of 14:10 (LD 14:10), 2) constant light (LL), and 3) constant dark (DD). Light intensity near the surface of the open water was 10-20 footcandles in all lighted aquaria. On a sunny day light intensity near the surface of open water in the field was slightly greater than that used in the laboratory. Aquaria contained either 100 or 200 mysids and the number of animals in the open water was counted at two-hour intervals. A dim flashlight shaded with red cellophane was used to take readings in darkness as preliminary studies showed its effect on the mysids to be minimal. Results In all field studies and preliminary field work, virtually no mysids were found in open water during the day but some were found in open water at night (Figure 1). The Pairwise Chie Test in conjuction with the Formula for Combining Probabilities for Independent Tests of Significance (X* Combined) was used to analyze field data (Sokal and Rohlf, 1969). In the rowboat study the nighttime peak number of mysids detected in open water is significantly greater than the daytime values (X Combined, p£05). In the diver study the nighttime value for the number of mysids counted in open water samples was slightly greater than the daytime value, but a statistical test failed to show significance (X* Combined, p£.20). Also in the diver study, the number of mysids detected in the kelp canopy at night was slightly less than that during the day but was again insignificant (X Combined, p£.20). In the LD 14:10 experiment, peak numbers of mysids were detected in the open water during hours of darkness (Figure 2). A mean was calculated for three consecutive nighttime readings starting with 2200h and for three consecutive daytime readings starting with 1000h. The Student's t-test for the Comparison of Two Percentages (PTT) was used to collate the two means (Sokal and Rohlf, 1969). The nighttime mean was significantly greater than the daytime mean (PTT, p£.05 ). In the LL experiment, the mysids remained closely associated with kelp during all periods of observation (Figure 2). At no time was more than 2% of the total population detected in open water. In the DD experiments, peak numbers of mysids in the open water were detected during the subjective night (Figures 3 and 4). In the first DD experiment this pattern persisted for about one week, though the peak values diminished. Since 92% of the mysids survived the first DD experiment, the decline was probably not a function of loss of animals. The mean night value was significantly greater than the mean day value for diel periods in all DD experiments (PTT, po.05 ). Discussion Both laboratory and field results suggest that A. sculpta remains closely associated with blades of Macrocystis during the day but does not maintain this close relationship with kelp at night. Daytime observations made while snorkeling in the kelp canopy and the adjacent open water corroborate these results with the added information that the mysids are so cryptic in both behavior and coloration that they are seldom seen. The mysids which were observed, however, were intimately associated with kelp. In the rowboat study, the fact that the nighttime peak value for the number of mysids detected in open water is significantly greater than the daytime values (x* Combined, p.05) suggests movement away from kelp at night. Taken in the context of Hobson's (1976) and Miller's (1976) work together with preliminary studies, all of which showed an increase in the number of mysids detected in open water at night over daytime values, these results suggest a behavioral pattern which lates to laboratory findings. These results do not necessarily show mass movement from the kelp bed across long distances to open water. It is possible A. sculpta comes out of hiding among kelp blades and swims in water immediately adjacent to them. Perhaps only some of these mysids move far enough away from kelp to be taken in plankton tows 10-20M off the canopy. A better method of demonstrating this behavior may be to dip net in open spaces in the canopy during the diel cycle. Possible advantages for a behavioral pattern of nocturnal movement away from kelp were not examined in this research, but previous work gives grounds for speculation. Hobson (1976) shows that A. scupta's consumption of copepods increases at night over the amount consumed during the day. In addition, the number of copepods and other zooplankters present in the water column is known to be greater at night than during the day (Hobson, 1976). This abundant food source and the protection against predation which the darkness provides probably combine to make A. scupta's behavioral pattern highly adaptive. An interesting and unexpected result obtained from the DD experiments was the existence of an endogenous rhythm of association with kelp during the subjective day and movement to the adjacent open water during the subjective night. For a small animal, which may have to move some distance to reach the protection of kelp before daylight subjects it to diurnal or crepuscular predatory fishes, the ability to anticipate dawn provided by an endogenous clock would be essential. If any compass orientation is required to home back to the kelp canopy a clock is again necessary to account for movement of celestial objects used as navigational cues. Many intriguing aspects of this patterned behavior remain to be examined. Characterization of the behavior in the field is only fragmentary and suggestive. Properties of the biological clock such as its free running period and Zeitgeber also provide a fertile area for further investigations. Summary 1) Field observations showed that A. sculpta remains close to kelp blades in the Macrocystis canopy during the day. 2) A. sculpta was detected in open water adjacent to the kelp forest at night but virtually never during the day. 3) In laboratory experiments in constant light, mysids remained closely associated with kelp and almost none moved away from kelp into adjacent open water. 4) In LD 14:10 in aquaria, significantly greater numbers of mysids were observed in open water during darkness than during light (PTT, p.05). 5) In DD A. sculpta exhibited an endogenous circadian behavioral pattern of close association with kelp during the subjective day and movement away from kelp during the subjective night. The nocturnal abundance of zooplankters and the protection against predation which darkness affords probably combine to make A. sculpta's nighttime movement away from kelp highly adaptive behavior. Works Cited Banner, Albert H. 1948. A taxonomic study of Mysidacea and Euphausiacea (crustacea) of the northeastern Pacific. Transactions of the Royal Canadian Institute, 20: 97-102. Clarke, William D. 1971. Mysids of the southern kelp region. pp. 369-380 in Wheeler J. North ed. The biology of giant kelp beds (Macrocystis) in California. Verlag Von Cramer. Germany. Hobson, Edmund S. and Chess, James. 1976. Trophic interactions among fishes and zooplankters near shore at Santa Barbara, California. Fish. Bull. U.S., 74: 567-598. Limbaugh, Conrad. 1955. Fish life in the kelp beds and the effect of kelp harvesting. University of California Institute of Marine Resources, 55-59: 134. Miller, W. Allen. 1976. The effects of giant kelp on the distribution of plankton and nutrients. Unpubilshed Manuscript. Carleton College, McAllistir, Minnisota. Sokal, Robert R. and Rohlf, F. James. 1969. Biometry and practice of statistics in biological research. W. H. Freeman and Company, San Francisco. 608-620. Tattersall, W.M. and Tattersall, Olive. 1951. The british Mysidacea. Ray Society. London. 408. 10 Acknowledgments I wish to thank my advisors Charles Baxter and Lawrence Harding for their guidance, fellow students for helping with plankton tows during all hours of the night, and Carl Johnson and Richard Roux for extricating me from Monterey Bay after one such ill-fated voyage. Figure 1 Figure 2 Figure 3 Figure 4 11 Illustrations Mysids Observed/Zooplankton Sample Versus Time % Total Population in Open Water (LD and LL) Versus Time % Total Population in Open Water (DD EXP. 1) Versus Time % Total Population in Open Water (DD EXPS. 2 and 3) Versus Time Figure 1 Mysids Observed/Zooplankton Sample versus Time The mean number of mysids detected in three replicate surface tows of a 333um mesh zooplankton net, 0.5M in diameter is graphed versus time for two different studies. In the rowboat study each tow lasted three minutes and tows were made only in open water 10- 20M from the kelp forest (boxed points). In the diver study, 50M tows were made through the kelp canopy (points which are not boxed) as well as through the open water. Vertical bars indicate standard error. v Ee DD o r — MYSIDS/ SAMPLE S r 20 20 e 14 Figure 2 % Total Population Observed in Open Water (LD and LL) Versus Time The % of the total mysid population observed in open water in aquaria (as opposed to a section of each aquarium which contained kelp) is graphed versus time for separate studies which used different light regimes. In the light-dark (LD) experiment a 14:10 cycle was used and one tank was lit with natural light which diffused through a laboratory skylight while another tank was lit artificially. In the graph of constant light (LL) experimental results vertical bars denote standard error. All LD and LL aquaria contained 100 mysids and were exposed to a light intensity of 10-20 footcandles. POP. IN OPEN WATER 0 — C 16 Figure 3 % Total Fopulation in Open Water (DD EXP. 1) Versus Time The % of the total mysid population observed in open water in aquaria (as opposed to a section of each aquarium which contained kelp) is graphed versus time for three replicate tanks which each contained 100 mysids and were exposed to constant darkness (DD). Vertical bars denote standard error. 4 + % POP. IN OPEN WATER f 2 18 Figure 4 % Total Population in Open Water (DD EXPS. 2 and 3) Versus Time The % of the total mysid population observed in open water in aquaria (as opposed to a section of each aquarium which contained kelp) is graphed versus time for two constant dark (DD) experiments. In experiment 2 there were 200 mysids/replicate while in experiment 3 there were 100 mysids/replicate. Three replicates were used in both experiments. Vertical bars denote standard error. 2 — r1 POP. IN OPEN WATER —