Abstract Remote operated vehicle (ROV) dives in the Monterey Bay Canyon revealed the ubiquitous presence of an unidentified, orange-pigmented mysid. An analysis of videotape footage taken on numerous ROV dives during a one year time period was done to examine the distribution, habitat, and behavior of the mysid. Samples were also taken using a suction sampler on the ROV and taxonomic and observational studies were performed on these samples. It was found that the mysid may be a new species and a proper identification continues to be pursued. A preference for soft sediment was also found in the mysid, and it is felt that this may be a reflection of available habitat type and feeding requirements. Introduction For the past two years the Monterey Bay Aquarium Research Institute (MBARI) has been conducting investigations of the Monterey Bay Canyon using a submersible remote operated vehicle (ROV). Currently the tether length permits a maximum operating depth of four hundred fifty meters. Scientists have seen small, orange organisms with golden eyes hovering in the hyperbenthic zone on numerous dives. They were thought to be mysids, but as none were in hand to identify them, they were labeled temporarily as "golden-eyes". Limitations in our knowledge of mysid taxonomy, ecology and physiology as noted by Krygier and Murano (1988), coupled with the ubiquitous nature of golden-eyes, make them animals worthy of study. Additionally, the ROV provides a rare opportunity to view these animals in their natural environment, and to get live specimens from the deep ocean. The purpose of my study was to determine an exact taxonomic label for the mysid and to better understand how they fit into the benthic microenvironments they inhabit. An analysis of "golden-eye" distribution by depth and habitat was done by reviewing videotapes taken during ROV dives. In addition, specimens were collected using a suction sampler mounted on the RÖV and observed in holding tanks at the Monterey Bay Aquarium where they continue to live. Those animals which did not survive the transfer to holding tanks in the aquarium were preserved for use in taxonomic studies. Materials and Methods Golden-eye were recorded on betacam videotape footage taken by a DXC 3000 Sony camera that was mounted on a remote operated vehicle (ROV) during dives from March 20, 1989 to March 19,1990. The five primary dive sites studied cover a range of different areas and habitats in the Monterey Bay Canyon (Fig. 1). The names and locations for each site are as follows: Canyon Wall C4-C5 (36 degrees 42'N 122 degrees 00'W), Soquel Canyon (36 degrees 49'N 121 degrees 59'W), Canyon Wall Meander (36 degrees 45' N 121 degrees 58'W), Pt. Joe (36 degrees 37'N 122 degrees 01'W), and North Wall (36 degrees 48' N 122 degrees 05'W). The videotapes were annotated by a technician who recorded the comments of scientists along with a listing of the organisms seen during the dives. The video tapes are time coded and tied to a master clock so that annotated frames are linked to all recorded physical and navigational data. Access to sets of these data can be gained through a relational computer database. An analysis of one hundred thirty single frames that had been annotated with golden-eye and known depth was carried out noting date, site, depth, number of golden-eye in frame, habitat type, other organisms in association, and behavior. In order to accurately count all of the golden-eye in an annotated frame adjacent frames were also examined. The single frames were recorded on optical disk for easier access and review at later times in the study. An estimation of the area captured in a given videotape frame was calculated setting the length of a sharply focused golden-eye in the frame equal to the average length based on specimens in hand. This conversion factor was used in transforming dimensions on the monitor to natural dimensions. The calculated area was then used to compute golden-eye density (Harea) for a standardized 25X25 centimeter quadrat. investigated the relationship of density to habitat type by first classifying notes on habitat into three categories and then comparing them with calculated average densities for each frame. The three types of habitat were classified as follows: 1. Hard substrate- on or within approximately one meter of a hard, vertical rock wall face 2. Rubble- mixture of soft detritus or sandy sediment and low lying rocks 3. Soft sediment- detritus or sandy sediment with no large rocks within one meter Specimens were collected May 7, 1990 and May 23,1990 at the Canyon Wall C4-C5 site using the suction sampler mounted on the ROV. The May 7th dive yielded only one golden-eye from a depth of 370 meters. It was kept in cold sea water for transport to the Monterey Bay Aqurium, where it was put in a holding container with circulated water maintained between five and seven degrees Celsius. The container was then placed in a darkened room, and we observed the specimen for ten days before it died. It was then fixed in sea water and10% Formalin, and stored in 70% alcohol. On the May 23rd dive approximately 30 specimens were collected from a depth of 340 meters and transported in cold sea-water to an aquarium holding tank. Six died in the transit and were fixed and preserved in the same fashion as the one above. These seven organisms were observed under a dissecting scope, and using a calibrated ocular micrometer, total body length was measured from the anterior margin of the carapace to the posterior end of the telson, excluding spines or setae. Illustrations were drawn using a camera lucida. The surviving 25 specimens continue to be kept in a cold-water holding tank. Observations of their behavior have been noted. They are being fed brine shrimp nauplii and rotifers daily and appear to be doing very well in holding. Observations continue to be made by a professional aquarist. Results Taxonomic efforts placed golden-eye in the Suborder Mysida, Family Mysidae, Subfamily Mysinae using Tattersall (1951), Mauchline (1980), and Kathman (1986). A further classification can not confidently be placed on the organism at the present time, but identification is being pursued. Sizes of the seven measured specimens ranged from 6.88 to 8.80 millimeters, with an average of 7.69 millimeters. Camera lucida drawings of the body parts used in identification of mysids are shown in Figures 24-2F. Distinguishing characteristics of the animal include its orange pigmentation and large, globular golden eyes with large retinas. The rostrum is rounded and does not extend far beyond the eyestalks. The antennal scale is setose all around. Both the endopod and exopod of the uropod are setose all around, and there is a statocyst on the endopod. The telson is cleft, and has short spines which begin at about two-thirds of the total length of the distal margins and run posteriorly through the cleft region. The pleopods appear uniramous, but ambiguity in the characterization of uniramity and biramity make it difficult to assess this trait with confidence. The analysis of average density by depth in 10 meter intervals for all of the sites showed the five highest average densities were contained in the 350 to 390 meter range. The highest average density was 15.1 golden-eye per 625 square centimeter quadrat in the 320 to 330 meter range. However, this was based on only one frame (Fig. 3). The C4-C5 site had the highest average density of the five sites with 8.97 golden-eye per 625 square centimeter quadrat. (Fig. 4) Average densities for the three different habitat types revealed the highest density above soft sediment (8.01 golden-eye per 625 square centimeters) and the lowest density above hard substrate (5.89 golden-eye per 625 square centimeters) (Fig.5). The following is a list of some of the organisms seen in areas with golden-eye: Sponges Hagfish Holothuroids Psolus sp. Eptatretus sp. Anemones Parastichopus johnsonii Stomphia sp. Opisthobranchs Gorgonians Asteroids Plerobranchia sp. Sergestids Rathbunaster caifornicus Tritonia diametia Sergestes similis Stylasterias forreri There was one golden-eye sitting in the upper rim of a conical-shaped sponge, and what appeared to be two other golden-eye were further down in the spiral lying motionless. Golden-eye were seen regularly swimming in and out of hagfish burrows. We performed qualitative behavioral observations of the golden-eyes both from the videotape and the aquarium holding tanks. Golden-eye swim with a hovering motion just above the substrate. They were only seen four times, in the 130 frames that were analyzed, to be swimming a significant distance away from the substrate. In only one frame were they seen to be quiescent above detritus material. In viewing other frames, a rapid snapping of the abdominal segments at the juncture to the thorax was seen when a quick escape response action was initiated. The clearest illustration of this behavior was seen when a golden-eye swam into an Anthomastus ritteri tentacle and escaped in a fraction of a second. When examining one specimen in a dish under a dissecting scope it began to do something best described as a headstand. Mauchline described this type of behavior as a modification of the general filter feeding method. The thoracic exopods are used to generate a water current capable of moving surface material to the feeding appendages for filtration from the water (Mauchline 1980). The golden-eye appeared to have no problems maintaining its orientation from the dish sides, and the use of the antennal flagellum for this purpose was evident. It also did an extensive picking process through particles and smaller organisms in the dish and rapidly discarded most of them posteriorly by a sweeping motion of the thoracopod endopods. Discussion The inability to classify the organism beyond the subfamily level not only reflects the possibility that it is an undescribed species, but also points out some of the difficulties encountered in mysid taxonomy. I primarily used Tattersall (1951), Mauchline (1980), and Kathman's et. al (1986) in my efforts to key out the organism, and found that all three of these references lacked too many definitive characteristics at important junctures to confidently carry the golden-eye through the key. Mauchline points out that the incomplete collection of mysids, especially male species, has created problems with classification. There is a real need for a comprehensive and updated taxonomic review of the Mysidacea with a sound key and illustrations of all known species. It is difficult to draw firm conclusions from the data of average density according to depth due to a lack of sampling at certain depths. However, the depth range of the mysid appears to start on the shelf break below two hundred meters. It should also be noted that in a recent dive golden-eye were found at 450 meters in fairly high densities. The extension of the RÖV depth range to one thousand meters in the fall of 1990 should be very helpful in determining the lower depth range of the animal. An understanding of the benthic habitat and the biology of the mysid may add insight into the distribution of golden-eyes in the Monterey Submarine Canyon. The highest density of golden-eye being found at the C4-C5 site may relate to the patchiness and diversity of suitable habitat types at that location. The C4-C5 site is characterized by hard, vertical wall faces, and very rugged, rocky terrain, with patches of rubble, and soft sediment slopes of variable size. There is also an abundance of detrital material settled on the non vertical substrate. Golden-eye were often found swimming at the base of rocky, vertical wall faces, or in and out of rock crevices. The data show that the greatest densities were near soft sediment, where detrital accumulation is favored. The high degree of habitat diversity at the C4-C5 site with many suitable habitat types may legislate against a dominance by predators which may be distributed at a different patch scale. The expanded range of the ROV will again be helpful in investigating the habitat available with increased depth. It will be interesting to see if the trend of decreased available hard, rock substrate and substrate relief with increasing depth affects golden-eye distribution as such conditions have previously been shown to affect hard substrate communities (Lissner 1989). Another possible explanation for the low density of golden-eye on hard, vertical faces is that their methods of feeding favor a softer sediment habitat. They exhibit two types of feeding: filtration of suspended food material, and ingestion of large food masses (Crouau 1989). It seems likely that since filter-feeding is a major means of gathering food that this activity would be favored in a softer material, because it would be easier to generate a hydrodynamic current with lighter and less attached organic material. Additional support for a preference for soft sediment is the relatively fragile structure of the golden-eye exoskeleton. Normally benthic mysids have a more robust exoskeleton which protects them it they go burrowing for food in coarse sediment (personal communication with Dr. Peter Slattery). The lighter exoskeleton of the golden-eye suggests that it is not a burrower, and this is corroborated by video and laboratory observation. Finally, when dives were conducted at the C4-C5 site to collect golden-eye, the golden-eyes were most often found on plateaus where softer detritus matter was present or near crevices where rock and soft bottom met. The great numbers and close proximity of golden-eye to a variety of organisms and substrate types suggest they may be scavengers or involved in commensal relationships. This would make a great deal of sense, since mysid diets are generalized and include a great variety of organic materials (Mauchline, 1980). More detailed studies of exactly what the golden-eye are feeding on and further investigations of their interactions with other organisms are needed and may tell us more about their role in the benthic environment. References Cite Crouau, Y. (1989) Feeding mechanisms of the Mysidacea. Functional Morphoogy of Feeding and Grooming in Crustacea, Issues 6:153-171. Kathman, R.D., W.C. Austin, J.C. Saltman, and J.D. Fulton. (1986). Identification manual to the Mysidacea and Euphausiacea of the northeast Pacific. Can. Spec. Publ. Fish Aquat. Sci. 93: 411p. Krygier, E.E. and M. Murano (1988) Vertical distribution and zoogeography of oceanic mysids from the northeastern Pacific Ocean. Bull. Ocean Res. Inst. Univ. of Tokyo. 26(1):109-122. Lissner, Andrew (SAIC, ed.) (1989) Benthic Reconnaissance of Central and Northern California OCS Areas Vol.1 and II. Prepared by Science Applications International Corp. and MEC Analytical Systems. Mauchline, J. (1980) The biology of mysids and euphausiids. Advances in Marine Biology, 18:1-681. Tattersall, W.M. (1951) A review of the Mysidacea of the United States National Museum. Bulletin U.S. Natl. Museum., No. 201, 1-292. Figure Legend Figure 1: Map of the Monterey Bay Canyon showing the five major dive sites where the remote operated vehicle (ROV) was submersed and operated. Figure 2: Drawings done with camera lucida of some of the distinguishing anatomical features of the golden-eye. A. dorsal view, anterior end; B. antennal scale (41 total setae all around); C. uropod; D. telson; E. 5th pleopod; F. 3rd endopod. Figure 3: Graph illustrating average density (4/625 sq. cm) at 10 meter depth intervals for all of the sites combined. Numbers in parentheses indicate total number of frames viewed at a particular depth. Figure 4: Graph illustrating average density (4/625 sq. cm) for all depth intervals at each site Figure 5: Graph illustrating average density (4/625 sq. cm) for each of the three habitat types. L ... .. . . . Q oo .. . D 6o 2000 S -Z- Fiqure 2: Anatomical Features of the Golden-Eye Hysio A J a -a 63 — 5 S (5) 5 (8) (2)8 (6 (2) (01) (98) 2 (VI) o (0E (9)E (1) 91 (1 (0) e (0) o (1) 9 (2E (1) E (DSE (0) o s e 2 8S 8 5 - 8 Tigure 4: Golden-Ege Average Density By Site 4.77 Meander 5.18 Pt. Joe 631 Soquel 6.18 N. Van 8.97 C4-C Average Densitg (2/625 sq. cm) Figure 5: Golden-Ege Average Density By Habitat Type Soft sedient 8.01 Rubbl 761 Hard substrate 5.89 Average Density (2/625 sq. cm) Acknowledgements would like to thank all those who assisted me in completing this project. The Monterey Bay Aquarium Research Institute (MBARI) was generous in providing access to videotape and in allowing some ROV time to be spent on viewing and collecting golden-eye. Öf particular help at MBARI were Lynn Lewis, Bruce Gritton, and Chris Harrold. Gilbert Van Dykhuizen at the Monterey Bay Aquarium was generous in time and resources with his help in getting holding space and trying to work out the identification of the golden-eye. Cindy Ashy was very helpful in suggesting references and supplying equipment. Ladd Johnson provided assistance in using the camera lucida and offered some insightful comments in interpreting my data. Alan Baldridge and his library staft were very helpful and bent over backwards to get needed references. Finally, I wish to thank my insightful advisor Chuck Baxter. He taught me a great deal about how science truly works and the value of critical thinking skills.