UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The association between the black rockfish Sebastes melanops Girard and beds of the giant kelp Macrocystis.. Leaman, B. M. 1976

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


UBC_1976_A6_7 L42.pdf [ 9.04MB ]
JSON: 1.0093866.json
JSON-LD: 1.0093866+ld.json
RDF/XML (Pretty): 1.0093866.xml
RDF/JSON: 1.0093866+rdf.json
Turtle: 1.0093866+rdf-turtle.txt
N-Triples: 1.0093866+rdf-ntriples.txt
Original Record: 1.0093866 +original-record.json
Full Text

Full Text

The Association between the Black Eockfish (Sebastes melanojas Girard) and Beds of the Giant Kelp (Bacrocystis integrifolia Bory) in Barkley Sound, British Columbia by Bruce Michael Leaman B.Sc., Simon Fraser University, 1972 A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science at the Institute of Animal Resource Ecology He accept this thesis as conforming to the reguired standard The University of British Columbia February,1976 (c) Bruce Michael Leaman, 1976 » In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Depa rtment The University of British Columbia Vancouver 8, Canada Date ii ABSTRACT The association between the black rockfish, Sebastes melanogs Girard and the giant kelp, Macrocy.st.is integrifolia Bory, was examined in terms of habitat utilization. Distributional patterns of juveniles frequenting kelp beds were documented by three-dimensional positioning of observed fish with the kelp bed boundaries used as a frame of reference. Significantly higher numbers of smaller (30 - 100 mm) fish occurred inside the kelp bed than outside, both diurnally and nocturnally. A diel migration of larger (101 - 200 mm) fish occurred, showing significantly higher numbers of fish in the kelp bed nocturnally. Very few of the largest (201 - 250 mm) fish were observed in the bed and none of these exhibited this diel shift. Wider-scale sampling indicated an increasing cline in the size of fish when moving from inshore to the open sea. Diet analysis does not indicate dependence by the fish on kelp bed diet items. Transporting experiments failed to demonstrate any significant homing ability within the size-class of fish studied. Artificial habitats employing plastic * kelp* were constructed to test the assumption that the fish were responding to the amount of physical shelter provided by the kelp. Results show a highly significant numerical response by the fish population, to the changes in the amount of artificial shelter provided. It is concluded that the species utilizes the physical shelter of the kelp bed primarily as a refugium from nocturnal predators. The value of the kelp bed to the species lies in its enhancement of recruitment through increased survival of the 0 to I age-classes of fish. iii Xllil 51 PAGE Abstract . . ...» ii Table of Contents iiList of Tables v List of Figures .vii Acknowledgements x 1. Introduction 1 2. Methods and materials 4 2.1 Description of the study site 4 2.2 Associated fish at the study site .................. 5 2.3 Collecting 7 2.4 Tagging and growth 7 2.5 Diet .. 9 2.6 Distribution and abundance ...............,<>........ 10 2.7 Laboratory experiments 15 2.8 Manipulative experiments ...........................16 3. Results 22 3.1 Physical and biological parameters of the site 22 3.2 Distribution of the species 30 3.3 Population dynamics of Sebastes melanops ........... 35 3.4 Diet 41 3.5 Localized distribution and abundance ..44 3.6 Environmental correlates of distribution 54 3.7 Fidelity and quantification of the association ..... 57 4. Discussion 65 4.1 General distribution 64.2 Barkley Sound distribution 69 iv 4»3 Sebastes mglano^s in Macrocyjstis beds 73 4. 4 Diet 75 4.5 Activities in the kelp bed 79 4.6 The nature of the association ...................... 87 4.7 Significance of the kelp bed to the fish 90 5. Conclusions 92 6. Literature cited 4 7. Appendices ..........................107 V LIST OF TABLES PAGE Table I. Abundances of fishes at the study site, other than Sebastes melanous ................................. 29 Table II. Comparison of mean growth rates of Sebastes melano£s as determined by direct and indirect measurements 40 Table III. Major food items of each size class cf Sebastes melanogs, with a statistical comparison of the frequencies of dominance of these items ............... 44 Table 17. Statistical comparison of the die! positions of the cumulative totals of fish observed ................ 53 Table V. Environmental information for each date supplying data for visibility-distribution analysis ............. 56 Table VI. Initial conditions, manipulations and results of experiments examining the phototaxis of Sebastes melanops and the influence of rheotaxis on this response 57 Table VII. Numbers, sizes and habitats to which Sebastes melanop_s were transported, together with the numbers and sizes of fish exhibiting homing behaviour 61 vi Table VIII. Levels of shelter and fish associated with Frame I, together with an analysis of the variance in these numbers in relation to changes in the shelter provided 63 vii LIST OF FIGURES PAGE Figure 1. Location of the study area 6 Figure 2. Construction details of the fluorescein tag ..... 9 Figure 3. Sampling locations in Barkley Sound ............. 13 Figure 4. Study site and the surrounding area ............. 14 Figure 5. Photograph and dimensions of the habitat frame .. 19 Figure 6. Underwater view of Frame I 20 Figure 7. Plastic used in the shelter experiment 20 Figure 8. Underwater views of Frame II and the kelp bed studied 21 Figure 9. Low-level aerial photograph of the study kelp bed 22 Figure 10. Bathymetric map of the study site .............. 25 Figure 11. Aerial view of the study site illustrating orientation of logs on the beach 26 Figure 12. Regression of tide range on bottom current at viii the kelp bed 26 Figure 13. Seasonal change in the average occlusion of the water column by kelp tissue. 0.8 m above the substrate 28 Figure 14. Abundances and depths of Sefcastes jelanojgs capture by Oregon trawlers. 1963 - 1971 .............. 31 Figure 15. Sampling locations in Barkley Sound for Sebastes melanop.s 3 2 Figure 16. Sebastes spp. larvae obtained in plankton samples in Barkley Sound. 0 - 3 m deep, January -May, 1974 34 Figure 17. Net captures of Sebastes roelanojgs at the kelp bed by size classes 36 Figure 18. Population estimates of Sebastes melanoos based on recaptures and observations of tagged fish ......... 37 Figure 19. Age - length regressions for Sefcastes jglanops by scale and otolith aging ............................ 39 Figure 20. Freguency of diet item occurrence in Sebastes melanops ..............................................42 ix Figure 21. Seasonal change in the occurrence of mysids and clupeids in the diet of Sebastes jelanogs ............. 43 Figure 22. Coincidence of kelp presence and fish capture at the kelp bed 46 Figure 23. Physical nature of the rock slope area ......... 47 Figure 24. Observed abundances of fish at the rock slope ., 48 Figure 25. Cumulative nocturnal and diurnal observations of fish postion at the kelp bed. 30 - 100 mm fish .... 50 Figure 26. Cumulative nocturnal and diurnal observations of fish position at the kelp bed. 101 - 150 mm fish .. 51 Figure 27. Cumulative nocturnal and diurnal observations of fish postion at the kelp bed. 151 - 200 mm fish ... 52 Figure 28. Changes in fish position with water clarity .... 55 Figure 29. Tracking record of fish tagged with fluorescein tag 59 Figure 30. Transporting of fish and returns during fidelity experiments .................................. 60 Figure 31. Fish at Frame I at 14.2 m* and 33.2 m* levels X of shelter 62 Figure 32. Surface canopy of the plants in late Auqust,1974 81 JCKHOJILEDGEMENTS I wish to thank the members of my supervisory committee, Drs. T.H.Carefoot, G.C.Hughes, J.D.McPhail and R.F.Scagel, for their thoughtful comments on the design of the project and a draft of the manuscript. To Dr. $.J.Wilimovsky, my research supervisor, I am grateful for criticisms at various stages of the research, comments regarding a preliminary draft of the manuscript and many other considerations. A field project such as this necessitates considerable assistance while diving. fly colleague, Dr. Jeffrey Marliave, who gave generously of his time and expertise, aided me on numerous and often trying occasions. To the following diving partners, I am also grateful: Glenn Cota, Myriam Haylock, Susan von Melville, Trudie Mullin, Dan Pace, Chris Tanner and Heather Washburn. Joseph and Richard Garcia shared their many observations on the natural history of the fish fauna of the Bamfield region with me. Wendy Craik kindly criticized initial segments of the manuscript. Laboratory experiments were conducted and field studies were based at the Bamfield Marine Station of the Western Canadian Universities Marine Biological Society. Steve Borden of the Biological Data Centre (I&RE) gave advice on statistical procedures. I thank Dr. Louis Druehl of Simon Fraser University for logistical help in the field and other considerations. Alex Thompson, lighthouse keeper at Cape Beale, and Edwin Niska, of the Oregon Fish Commission, graciously allowed me access to unpublished information referred to in the thesis. 1 1. INTRODUCTION The biota of the coastal waters of British Columbia is characterized by a distinct seasonality of the abundance and diversity of species, particularly on those segments of coastline adjacent to the open Pacific Ocean. One of the most prominent floristic features of the summer months is the presence of stands of large phaeophyte algae, the upper portions of which form canopies floating on the surface waters. There are, primarily, two algae which form these conspicuous canopies in British Columbia; the giant kelp, Macrpcystis integrifolia Bory, and the bull kelp, Nereoc^stis luetkeana (Hertens) Postels and Ruprecht. Beds of these large plants provide considerable structural heterogeneity in the otherwise almost unstructured water column. This increased heterogeneity gives rise to a proliferation of the species in the animal and plant communities in the vicinity of the beds. The biology of kelp beds has received considerable attention in California. North (1972) provides an extensive bibliography of research on beds of Macrocystis pyrifera (Linneaus) C. figardh. The works reviewed by North are primarily of a botanical nature, though ecological interactions of various animal species found in the beds are also included. Research into the biology of kelp beds in British Columbia has been limited. Scagel (1959) included B.C. studies in his review of the relationship of plants and animals in the marine environment. There has been little £ublished material about research on B.C. kelp beds since 1960, although a considerable amount of effort has been expended in these habitats, including both floristic and faunistic studies. Many animals (invertebrate and vertebrate) found in kelp beds have a direct relationship with kelp plants. They live either in or on the plants, or the plants provide their trophic resource base. Species with indirect relationships are those which may be associated either with kelp plants themselves, or with the associated flora and fauna. The fish fauna of kelp beds may contain either type of species. The fishes of kelp beds have been examined both as a community (Limbaugh, 1955; Davies,1958; Quast,1968; Ebeling et al.,1972) and as individual kelp bed species (Skogsberg,1939; Wales,1952; 0'Connell,1953; Boedel,1953; Williams and Williams,1955; Phillips, 1957; Follett,1960; Limbaugh,1961,1962; Turner and Ebert,1962; Gotshall et al.,1965). All of the above papers dealt with fishes in California kelp beds and, in general (e'xcept Limbaugh, 1961), detail their interactions with the plants in a qualitative fashion only. The fishes of B.C. kelp beds have not received a great deal of attention; I know of no published material that deals specifically with these forms. Sebastes melangps Girard is the most common neritic fish resident in the kelp beds of Barkley Sound,B.C. Juveniles of this species are found abundantly in beds of Macrocystis iii§3£if2lis and in lesser amounts in beds of Nareocystis 1 ueJUsejuia.» Previously, I had observed that the abundance of fish was, roughly, inversely correlated with their size, and that some fish could be found consistently within the kelp bed. Larger fish, however, were not found in or near kelp beds, but instead, near the open coastline. 3 The present study was designed to examine the hypothesis that S. jelanojas is dependent upon kelp beds in some fashion and that this fish undergoes a transition in habitat association with age; from an inhabitant of kelp beds when young, to an inhabitant cf moderate relief coastal areas when older. In the initial stages of the study I attempted to determine how the fish uses the kelp bed as a habitat. Several aspects of the ecology of the species were examined to categorize the links between the fish and the kelp bed. The later stages of the project were largely occupied with the testing of a subsequent hypothesis as to the primary nature of the association betseeen the two species. 4 2. METHODS AND MATERIALS The study was conducted during the period June,1973 -September,1974 in Barkley Sound, near Bamfield,B.C. The field site was located in a small group of islets (Boss Islets) at position 48°52.45*N, 125°09.65»8; the kelp bed studied was situated in a small cove having a southwestern aspect (Figure 1). 2.1 Description of the study, site The cove containing the kelp bed studied was characterized in terms of several physical and biological parameters: area, topography, current velocities and kelp density. (i) The area of the study site was calculated from a large scale map of the vicinity, as well as aerial and surface photographs taken at the same tidal heights, The total area of the site was taken as the average of these three measurements. (ii) Topography of the bottom was determined by sounding with a metred, hand line. The values obtained were then incorporated into a bathymetric map. (iii) Current magnitudes were extrapolated from daily wind and sea state data taken at Cape Beale lighthouse, approximately 12.8 km from the study site. In situ current measurements were taken at the study site over a short period of time. These data were then related to the Cape Beale data and a correlation plot constructed for the purposes of extrapolation throughout the year. The in situ measurements were obtained with a TSK Flow 5 Meter1 over a fifty-day interval during May - June 1974. (iv) Kelp density was estimated with a visual occlusion method employing SCUBA. A checkered, plywood panel 0.83 m2 (9 ft2) was viewed from 2 m distance, normal tc the panel, while the panel was held vertically, its base in contact with the substrate. Kelp density was taken as the mean percentage of the panel occluded by the kelp, when viewed from the four cardinal compass points. This procedure was repeated approximately once per week, at both random and specific locations in the kelp bed, during the periods of kelp presence. 2.2 Associated fish at the site The fish assemblage of the area was partially known from previous work (Druehl, Green and Leaman, unpublished data) in this area and other similar sites. This work had included both netting and ichthyocide sampling. All species (other than §• ffl§isa2ES) were enumerated during the netting program of the current study to determine the total fish complement of the site. 1 Tsurumi-Seiki Kosakusho Co., Yokohama 6 Location of the study area 7 2.3 Collecting Fish [unless specified otherwise, •fish' refers to Sebastes melanoDS ] were collected for diet, population and movement studies with nylon monofilament gill nets of 1.27 (0.5 in), 2.54 (1.0 in), 3.81 (1.5 in), 5.08 (2.0 in) and 6.35 (2.5 in) cm (wet, stretched mesh) sizes. All nets were not employed simultaneously, rather, two nets, of 2.5** cm difference in mesh size, were set at any given time. Nets were set 1.0 - 1.5 hours before sunset and retrieved <0.5 hours after dawn in order to: (i) maximize the number of living fish in the net, and (ii) improve the probability of identifying the stomach contents of the fish. Nets were set in a T-shaped or L-shaped pattern at various locations, in and out of the kelp bed. Specimens of S. melano£s were also obtained from ether locations to complete an age-length regression. These additional fish were obtained using a sling spear and a Norwegian-type cod jig. 2.4 Tagging and growth Fish recovered from the nets alive were anaesthetized, measured (TL), tagged, allowed to recover from the anaesthetic and released. The anaesthetic employed was MS-222 (tricaine methane sulfonate). Several different models of tags were employed for the population study, all of the nylon 1 anchor' type (Dell,1968). All tags were inserted in the dorsal musculature of the fish, ventral to the base of the dorsal fin. An additional tag (Figure 2) was developed for short term 8 tracking studies. This tag was composed of a variable length of five mm (I.D,) glass tubing which was mated to the basic nylon anchor tag. The tube was then filled with powdered Na~ fluorescein dye and the open end of the tube plugged with epoxy glue or silicone sealant. Fish were tagged and the epoxy or silicone plug removed immediately prior to release. It was possible to track the fish from the surface (if water clarity permitted), or from in the water, by the trail of dye which diffused from the tag as the fish swam. The length of time over which the tag was effective was a simple function of the amount of dye in the tag (and, hence, of tag size), e.g., a 55 mm tag lasted approximately five hours. Data for population estimates were gathered both frcm net sampling and from observational sweeps through the study area, using SCUBA. Population estimates were generated through multiple censusing of the fish in the area for relative numbers of marked and unmarked fish, using the modified formula for a Schnabel capture-recapture estimate of Chapman (1954). Growth of the fish was measured by growth of recaptured fish, and interpolation from an age-length regression. The regression was constructed from data taken from all size classes of fish. Fish were aged by both scale and otolith annuli counts. 9 Figure 2 Construction details of the fluorescein taq 2.5 Diet Fish recovered dead in the gill nets were frozen, upon return from the field and analyzed at a later time. Only the contents of the stomachs were analyzed because intestinal contents were, generally, unrecognizable or impossible to totally recover. Stomachs were removed from the animal, fullness estimated, and the contents emptied into a finger bowl for microscopic examination. The sagittal otoliths and a scale sample were alsc remcved from all size classes of the fish recovered. 10 The analyses of the stomach contents were summarized on forms, an example of which is shown in Appendix I. Contents were scored by percentage occurrence and by percentage dominance, A dominant item was any item which represented > 50% of the total volume of the contents, estimated by eye. The dominant item selected on this basis yielded a general estimation of the major components of the fish's diet. Volume was the basic criterion employed because an analysis based on the freguency of occurrence of items alone confers equal statistical weight to all occurrences of any given item, without regard to the guantity of the item in the stomachs, A volume-based analysis has the additional advantage of incorporating some aspects of the energy contribution of diet items, although results regarding energy input through diet items must be interpreted cautiously if no data are available regarding caloric content and assimilation efficiency (Paine,1971), Specimens whose stomach contents exhibited no clearly dominant item, or whose contents were too digested to determine dominance, were assigned no dominance. 2.6 Distribution and abundance The primary emphasis of the study was centred on the single kelp bed, but the overall regional distribution of Sebastes melanogs »as also investigated through jig and speargun captures, returns from a trolling cruise, reports from local fishermen and extrapolations from published data on the geographic and bathymetric distribution of the species. Jig and speargun samples were made on the southeast side of Trevor 11 Channel from Aguilar Point to Whittlestone Point; the trolling cruise covered the area at the mouth of Trevor Channel and that between Folger Island and Edward King Island (Figure 3). The technique employed in establishing the distribution and abundance of S. melanoj3£ at the study site was to enumerate and size fish during underwater observational sweeps. The observational data encompassed physical parameters at the time of observation and information on the fish sighted. A typical observational summary sheet is illustrated in Appendix II. Sizes of fish were estimated and when necessary or possible, size was measured against background features, to which the fish could be adequately compared, using a pencil marked in centimetres. The accuracy of size estimation was checked by comparing the estimate with data cn fish which were tagged or marked and observed on the same day. The position of the fish in relation to the edge of the kelp bed was determined by comparison of sighting location with position along metred transit lines established on the bottom. There were three such lines of 35 mm metred, polypropylene rope, laid out 15 m normal to the edge of the kelp bed (Figure 4). The lines were anchored to bolts set into pre-packaged units of concrete which had been formed into rock crevices. These lines were only necessary during the initial observational dives until familiarity with the microtopography of the area was achieved, allowing accurate estimate of position. The observational sweeps were varied as to starting point, to minimize behavioural responses which could give rise to patterned positioning of the fish. 12 Observation dives diel activities of the their distribution and attempted to monitor fish, in addition to abundance. all aspects of the seasonal features of Figure 3 Sampling locations in Barkley Sound Imperial Eagle Channel Sanford Island Transit line Frame II Kelp boundary Framel—(-• Boundary rock substrate Trevor Channel Figure 4 Study site and surrounding area 15 2.7 Laboratory £2LBi:£i.EiL2£§ Several laboratory experiments were conducted at the WCUMBS1 Bamfield Marine Station, in an attempt to resolve conflicting field observations on nocturnal behaviour of the fish. The fish exhibited different distributional patterns under, apparently, similar nocturnal conditions. Observations suggested that light and current might be interacting to determine some aspects of the nocturnal distributional pattern. A series of experiments was conducted in which study fish were subjected to varying intensities of current and light, to determine if there was any interaction betweeen their phototactic and rheotactic responses. Experiments were conducted in a 2700 litre aguarium inside a shaded area (illustrated in Appendix III), with fish being held in a similar aguarium prior to experiments, A single 40 watt microscope bulb controlled by a rheostat supplied the light. Actual light intensities were determined through interpolation from a plot of rheostat values versus lux produced, as measured with a photometer (Photovolt Corporation, Model 501-M). Experimental current regimes were produced with three submersible pumps (773 1/hr each) which were positioned at several locations within the aguarium. The position of the fish and the orientation in the water column, assumed under each experimental condition, were noted. Behaviour between changes in levels of the factors was recorded. Universities Marine Biological Society 1 Western Canadian 16 2.8 Mani. £ ulat i ve experiments Two field experiments were conducted to determine the fidelity of the association between the fish and its habitat, and the nature of the relationship between the amount of physical shelter provided and the number of fish using it. After these experiments, an artificial habitat, simulating the shelter aspect of the kelp bed, was constructed to examine the nature of the relationship between the fish and the shelter provided by the kelp. (i) Previous work in Barkley Sound (Druehl, Green and Leaman, unpublished data) had shown that S. melanojgs (and ether fish) were attracted to artificial 'beds' of kelp established in areas at least 70 m from the nearest kelp bed. These 'beds' consisted of elevated, rectangular frames covered with wire mesh. Plants, attached to bricks, were placed on the mesh in various densities. I repeated the essential features of this experiment without the metal frame, employing thirteen plants attached to clay bricks. These plants were taken from a kelp bed approximately 1000 m distant from the study site and were established at a location approximately 30 m from the nearest kelp. The edge of the existing bed at the study site was also extended 8 m, employing more plants from the same location as those used in the previous experiment. This second experiment was designed in an attempt to extend the contiguous distribution of the fish observed at the edge of the study kelp bed during daily observations. A third experiment attempted to determine if there was any fidelity in the relationship of the fish and a given habitat 17 location or a given habitat type. Fish were captured and tagged at the study site and then transferred to several areas at varying distances from the site. In the instances where the fish were transported to another habitat containing Sebastes melanop_s, an equal number of similar sized fish were removed from this habitat, to ensure that the transported fish were not excluded frcm occupying this habitat by density-dependent factors. While the removal of these fish may have avoided a strictly numerical exclusion of the transported fish, it did not necessarily preclude, nor was it possible to control for, behavioural exclusion. The assumption must be made, therefore, that no antagonistic reactions existed between the two groups of conspecifics. The transported fish were tracked by net capture on the line of their transport to determine if they could return (•home*) to the location from which they were taken; or whether they would remain in the new location to which they were transported. In addition, observations were made at the •home' and the 'transport1 sites to determine the fate of the transported fishes. (ii) The capture of fish by size class over time was monitored and compared with changes in kelp density. The shelter provided by the kelp was estimated as kelp density, obtained by the method described in part 1. (iv) (p.7) of this section. The artificial habitat (Figure 5) was constructed of 1.9 cm diameter iron reinforcing rod with corner brackets and rod holders brazed onto the frame. The dimensions of the habitat frame are given in the figure. The frame was partially 18 asssembled, transported to the study site and assembled under water (Figure 6). Shelter at the frame was provided by pieces of green 0.15 mm plastic cut into a shape to simulate kelp plants (Figure 7). The strips were cut into several lengths to enable them to reach the substrate from the supporting rods; the plastic was weighted at the bottom and attached to the supporting rods with clothes pins. The shelter provided by the plastic was taken to be the combined area of the strips. The amount of shelter provided by the plastic was changed periodically, by adding or removing pieces of plastic, and the numbers of fish associated with the frame was recorded during observation dives. A second habitat frame was constructed and modified such that the pieces of plastic floated upward from the substrate (Figure 8), rather than hanging down as at Frame 1. Observations and manipulations at this frame were similar to those conducted at Frame 1. 20 Figure 7 Plastic used in the shelter experiments 21 Figure 8 Underwater views cf Frame II and the study kelp bed 22 3. RESULTS 3.1 Physical and biological characteristics of the site The study cove comprised an area of appproximately 1000 in2 (995 m2) , as calculated by measurements taken from an aerial photograph, obtained through the Department of Lands, Forests and Water Resources of the Province of British Columbia (BC 726 1-052). The kelp bed did not occupy the entire area of the cove, rather, the surface canopy of the kelp was estimated from a low - level aerial photograph (Figure 9) to be approximately 700 a2 , or 10% of the area of the cove. Figure 9 Low-level aerial photograph of the study kelp bed 23 Figure 10 is a bathymetric map of the site constructed employing the data obtained through hand soundings, The substrate is not uniform throughout the cove; there are several anomalies which affect the distributional patterns of the fish. The most prominent of these are the two trenches at the east and west sides of the cove, These depressions below the general level of the substrate have important ramifications on both the diurnal and nocturnal activities of the juvenile Sebastes me la nop_s. a second anomaly at the site is the flat central portion of the cove, delineated by the crosshatched area in Figure 4 (p. 14). The significance of this central area is its substrate; the substrate of this area is sand with scattered, small (<5 cm diameter) rock, rather than the continuous, conglomerate rock of the rest of the cove. Since they reguire rock substrate for attachment, the plants are in very low abundance in this area. an examination of the low -level photograph illustrates this discontinuity. The hydrographic conditions at the site were almost exclusively the result of conditions in Imperial Eagle Channel. Infrequent, large scale (>70 km/hr) winds from the southeast gave rise to waves striking the site from this direction. Predominant winds in the summer months were from the west, southwest and northwest and while southeast and east winds did occur during the summer, they were generally of too weak to generate waves impinging on the site from this direction. The result of the orientation of the site and the seasonal meteorological conditions is that the primary wave force was received onto the mouth of the cove from Imperial Eagle Channel. 24 The position and the orientation of the logs in the supralittoral region (Figure 11) of the cove illustrates the uniformity of this exposure. The second major hydrographic influence in the region of the site was that of tidally generated currents passing through the two small channels, separating Sanford Island and Fleming Island, which link Trevor Channel with Imperial Eagle Channel. While tidal currents did not generate bottom currents in the immediate vicinity the kelp bed studied (Figure 12), their effect was considerable at the rock slope just to the east of the kelp bed. On extreme tides (4.5 m) tidal currents as rapid as 2.5 m/min were recorded at this location. The influence of these tidally generated currents on fish distribution will be presented in part 3.6 of this section. The seasonality of the oceanographic conditions at the site was paralleled by a concomitant seasonality cf kelp presence. Figure 13 details the change, over time, of the occlusion of the water column by kelp tissue. The degeneration of kelp tissue, beginning in August, is a result of several biological and oceanographic processes (e.g., degree of epiphytization, sea temperature, grazing pressure, etc.). While all of these factors influence the amount of tissue in the water column, the dramatic decrease in occlusion in August/September was closely correlated with the magnitude of south, southwest, and west winds or, nfore generally, the amount of stress to which the plants were subjected. 25 Figure 10 Bathymetric nap of the study site Regression of tide range on bottom current at the kelp bed 27 The fish assemblage at the study site, other than Sebastes melanojas, is presented in Table I. The sampling methods employed differ in their effectiveness to sample any given segment of a habitat, rotenone is most effective when sampling those fishes associated with bottom features and gill nets are most effective in obtaining larger water column fish. The overlap between the two methods is, generally, a result of either large specimens of epi-benthic fish being caught in gill nets, or juveniles of water column forms being affected by the rotenone. 28 OQULEION OF WATER DULJfvM BY KELP 1373-74 1DO. 80.1 BOol 70-1 60-1 50-1 40.1 30.1 S0«1 ajo«l -\ h -\ 1 1 \- H 1 h ^ JU AU SE CC ND DE JA FE MA AP MY JM JUAU SE 1373 1974 TIME Figure 1.3 Seasonal change in average water column occlusion by kelp tissue. 0.8 m above the substrate 29 TABLE I Abundances of fishes, other than Sebastes melanojgs, at the study site as determined through gillnet and chemical sampling. Gillnet sampling, this study; rotenone sampling,1971 Species Engraulis mordax Hexagrammos decagrammus Sebastes cau.ri.nus Ehacochilus vacca Sgualus aca.nth.ias Cymatogaster aggregata Hemilepidotus hemilepidotus Embiotoca lateralis Aulorhynchus flavidus Hexagrammos superciliosus fieriuccius productus Blepsias cirrhosus Raja binoculata Clupea harengus Oncorhynchus kisutch Salmo gairdneri Porichthys notatus Sebastes paucispinis Ophiodon elongatus Scorpaenichthys marmcratus Artedius meanyi Oligocottus maculosus Xiphister atropurpureus Artedius lateralis Pholis laeta Xererpes fucorum Anoplarchus purpurescens Oligocottus snyderi Apodichthys flavidus liparis florae Artedius harringtoni Gibbonsia montereyensis Artedius fenestralis Enophrys bison 30 3,2 Distribution of the species The scope of the study was not sufficiently large to investigate the widespread distribution of the various life stages or size classes of S. melanojDS; however such information for adults has been obtained from an unpublished report on commercial trawl fishing along the Pacific coast (kindly supplied by Mr. Edwin Niska of the Oregon Fish Commission). Figure 14 is a map of the central Pacific coast of North America on which are indicated the positions, depths and yields of the adults of the species that were collected by Oregon - based trawlers. The bathymetric mode of abundance, as determined through this type of sampling, is in the 0 - 53 m depth range. The coastal regions, at the positions where these catches were made, are such that the depth range sampled is less than 10 km from the coastline. A limited amount of sampling, other than at the study site, was done in Barkley Sound during the project. Figure 15 is a map of Barkley Sound on which are marked the numbers and average sizes of fish caught at several locations during this sampling program. An increasing cline in the average size of the fish is indicated when sampling from north to south in Trevor Channel. While the abundances of smaller size class juveniles (<150 mm) in Barkley Sound were not extensively investigated, observational samples in Wacrpcystis beds and in beds of the more exposed plant, Nereocystis luetkeana, show significantly higher numbers of S. malanops in beds of the less cumatophytic plant. 31 Figure JU Abundances and depths of Sebastes melanops capture by Oregon trawlers. 1963 - 1971 Figure J5 Sampling in Barkley Sound for Sebastes melano£S 33 Plankton toss in Earkley Sound yielded numerous Sebastes spp. larvae (Figure 16), however, the appearance of 40 - 50 mm juveniles of S. melanojas in mid-May indicates that none of the specimens collected was of this species. These 40 - 50 mm juveniles began to appear in the area of the kelp beds soon after the initiation of Microcystis sporophyte growth. Significantly higher numbers of Sebastes spp. larvae were obtained in sheltered waters, both in surface and in subsurface hauls, than in open channels (e. g., 700 vs. 3, Trevor Channel vs. Ross Islets). The absence of any published description of the larva of S. melanges, as well as the large degree of intraspecific variation and interspecific similarity in larvae of the genus (S.J.Westrheim, personal communication) make the assignment of specific identity to any of the larvae questionable, at best. No fish obtained at the study site were sexually mature and no gravid females were obtained in the course of sampling in the rest of Barkley Sound. All females caught which were sexually mature had ovaries at the stages 2 or 7 of Westrheim et al. (196 8). 35 3.3 Population dynamics of Sebastes jelanop_ s at the site Collections of S. jelanojp_s, obtained by nettings over the period of the study are presented by size classes in Figure 17. The information presented in these figures was collected during the course of the population sampling program, conducted to obtain data cn the dynamics of the species in the area. Population estimates of the numbers of fish residing in the area of the study site were derived separately from net captures of tagged fish and from underwater observation of relative numbers of marked and unmarked fish {Figure 18}. 'Besidence* of fish at the site was assumed on the basis of both direct and indirect evidence; the former, being recaptures of fish several times over the course of the study, including overwintering individuals, and the close similarity in population estimates from season to season. Indirect evidence of residence includes observations of tagged fish in the area of the site over long periods of time, with no sightings in adjacent areas, and in situ tracking of fish movements with fluorescein tags. f€T CAPTURES 1373-1974 30-100 M-M-50-. 40-as... 30... £5... so.. is... SO: s... f£T CAPTURES 1S73-1S74 101-350 M-M-1 ... 1 ~- 1 _ 1 - 1 _ 1 1 1 1 1 1 1 1 1 OIAUSEOCNOCEJAFEMAAPMVJJ 1373 1S74 Tike .i l*T CAPTURES 1S73-1S74 151-300 M-M-JU AU 30. SO-15.1 so.. 40. b = ^ 33'.. IS... I' I . II 1 1 1 i H JJ AU SE 0C NO 0E JA FE UA AP W JMJUAU 1373 1374 TttE fCT CAPTURES 1S73-1S74 E01-E50 M-M-u. ED' IS-l 10.1 5-1 1 1 1 1 1 1 1 1 1 1 1 1 JUAUSEOCNDCEJAFEMAAPMVJH 1373 1S74 TttE TIME Figure 17 Net captures of Sebastes melanops at the kelp bed by size classes 37 CHAPMAN POP- ESTIMATE - OBSERVATIONS 13000. 18000' HOOO. 1C000< sooo< BOOO-7000. eooo? sooo> •4000-300O> aooo* iflOO. 0-f < 1 1 1 1 1 1 1 1 1 1 1-JUAU5E0CNCIDEJAFEMAAPMY Jl JU AU 1373 JS74 CHAPMAN POP- ESTIMATES - NETTUSCS EE CC | •4500-4300*. 3900«. J 3B00-. : 3300-. 3000•. : : 8700'. : : SCO.. E1O0.. : laoo.. • ! : 1500.. 13X1.. 3X>- : : BOO: 300-. < i i 0* i -XX-H 1 1 1 1 1 1 1 1 1 (- H 1 JJAUSEOCrCIDEJAFEMAAPMirj^JUAU 1S73 1374 TIME Figure J8 Population estimates of Sebastes melanpjas based on recaptures and observations of tagged fish 38 In addition to the population estimates generated with recapture data, changes in the lengths of these fish during the interval from initial tagging allowed calculation of their growth rates. Table II provides mean growth rates for two size classes of fish, as determined by the above method. Included in this table are the growth rates for these size classes as determined by regressions of age upon length, with age determination by both scale and otolith annuli estimation (Figure 19). A Chi-sguared test for differences in the growth rates of the two size classes (as determined through recapture data) showed a significant decrease (p<.05) in growth rate for the 151 - 200 mm fish; the null hypothesis is that the rate was constant through the entire size range examined. There was also a significant difference between the growth rates of fish whose growth was interpolated from regression plots and those of 101 -150 mm fish. The results presented in the table illustrate both the advantages of direct growth calculation and the lack of resolution of growth afforded with interpolation of age-length regressions. The age classes, corresponding to these size ranges, are also presented in this table. 39 AGE-LENGTH REGRESSION SCALES S -8 E5> SO- 75iCO425<130^a»SSOe75a30SS5SOa75-«X)-e5-)SO-«75e0O. LENGTH IN VM« AGE-LENGTH REEFEESIDN OTDLXTHS !. Y = -0-4336 * C-3JE7-£:-01«X N = -44 5 0> £5« 50- 75<UX)<125d33<L75e30eS'^e75a003Sa50a75^0-4S-^^5SOO-LENGTH IN MM. Figure 19 Age-length regressions for Sebastes melanops by scale and otolith aging 40 TABLE II Comparison of mean growth rates of Sebastes mel,ano£S as determined by direct (recapture) and indirect (aging) measurements r - — • - T ~T_ T T | Size-class j Age | Recaptures | Growth rate 1 „ i . ., i _. , i i —'-j- — —1 - i • 1 • i | 1. 101-150 mm I I-II | 9 | 104 mm/yr | j 2.151-200 mm III-III ] 10 | 58 mm/yr j i i i' j J Aging Method I No. | Size Range J Growth rate | ] j. , ^,,.,, ,_i ] ! 1 + t ! |3.Otolith aging|44 fish| 84-484 mm | 60 mm/yr | j4.Scale aging |25 fish! 98-515 mm | 68 mm/yr J , | . i. ^ j j Growth rate difference| Chi-sguared value j | 1. vs. 2. | 12.76* | I 1. vs. 3. 1 11.94* | I 1. vs. **. 1 7.73* ! | 3. vs. 4- 1 0.49 j i J. ,. , . J * - significantly different (p<0.05) 41 3.4 Diet Analysis of the stomach contents of net-caught specimens was performed in two ways: - freguency of occurrence of diet items in the stomachs; and - freguency of dominance (as previously defined) of any item in a given stomach. These data are differentiated for all fish combined, and for the three size classes analyzed (Figure 20). The results of Chi-sguared statistical tests for differential frequencies of dominance, between the two most common diet items for each size class are listed in Table III. The major habitat where each of these diet items is found is also included in this table. All of the size classes examined exhibited a relatively narrow total diet spectrum. The primary food input was restricted to two different diet items and was consistent for any specific time cn an annual basis. The seasonal change in the occurrence of the two primary diet items was examined as the number of fish of each size class whose stomach contained the item. Figure 21 represents the collation of this information for the period of the study. *2 DIET ITEM5 - ALL FISH Amphipods J.Empty Polychaetes MZMMMMZM ; ))>n Mysids Clupeids Unident. Fish Cumaceans Eggs •Icopepods lEuphausids Zoea H 1 1 1-0. 10. • 3D. 30- -40. 3D. GO- 70-PERCENT FREOLENCY DIET ITEMS* 30-100 M-M-Empty Amphipods Crust. Remains Clupeids mm Unident. Fish Euphausids _) 1 1 ! 1 |_ 0. ID. 33. 30. O- 50- 60- 70- B0> 9Q.1XX). PERCENT FREOLENCY DIET ITEM5 101-150 M-M- DIET ITEMS 151-200 M-M-W2M2MMMZMM' Empty Mysids Amphipods MCrobs ' jH Crust. Remains . J Mussels W//////////////////M Clupeids Unident. Fish Copepods Zoea 0- ID- SO- 3D- AO- SO- BO- 70-PERCENT FREOJENCY Empty Polychaetes Mysids Amphipods Crabs J Mussels Uniaent. Fish Zoea Clupeids _ 1 1 1 1 • 1— 1 1 1 » 0> 10- SO- 30- 40. 50- BO- 70- BO- 90-PERCENT FREOJENCY frequency ot diet it«» occarr«nc« in Ssbj^jcjj lsl3Q2£3. Open bars « occurrence, kittened bars • do»ln»nc« 43 SEASONALITY CF DIET 1XOJRRENCE TIME Figure 2JT_ Seasonal change in the occurrence of mysids and clupeids in the diet of Sebastes melanops. 44 lASkl III Major food items of each size class of Sebastes melanoDs, as determined through gut analysis, with a statistical comparison of the percentage freguencies of dominance of these items Size Class Stomachs Examined Major Food Items Type Freg. of Dominance 30-100 mm 10 clupeids unid. fish euphausids P P 66.7%* 33. 7% 101-150 mm 79 mysids clupeids P P 51.7S* 36. 2% 151-200 mm 39 clupeids mysids amphipods P P B 73.7%* 10. 5X * - significantly greater (p<.05) P - pelagic/water column form B - benthic form 3.5 Localized distribution and abundance (i) Seasonal: Fish were not present throughout the year at the location of the study kelp bed, rather, fish were present only when the kelp plants were present. Figure 22 is a composite of previously presented figures which illustrates the close correlation between the abundance of net caught fish and 45 the presence of kelp. Nettings during the period from October,1973 to April,1974 did not yield any fish from those locations netted during the rest of the study (Figure 22). Fish in the size range 201 - 250 mm were uncommon at the kelp bed throughout the study. In addition to the study kelp bed, a second area of consistent fish occurrence was monitored for the duration cf the study. This location was a sloping area, immediately to the southeast of the study site, composed of uneven, large (1.0 - 3.0 m diameter) boulders extending 2.4 -9.1 m in depth and meeting the same sand flat which fronted the kelp bed area (Figure 23). Seasonal abundances of fish at this area showed an approximate inverse correlation with the presence of fish at the kelp bed (Figure 24). This correlation was more apparent for 30 - 100 mm fish than for 101 - 100 mm fish. (ii) Daily: Considerable variation was evident in the numbers of fish seen on any given day or series of days. This variation existed within larger scale changes in fish abundance at any location and was present at both high and low levels thereof. Experimental work conducted during the project must be considered in light of this variation. A more detailed treatment of variation in abundance is included in section 3.7 (P. 57). 46 NTT CAPTURES 1373-1374 30-10? M-M« f£J CAPTURES 1S73-1S74 101-150 M-M-Kelp presence BO-TOE NTT CAPTURES 1373-1374 151-200 M-M-9S.J 80-. io-. Ketp pre'sence Kelp presence • t -I 1 ^ 1 1 JUAJ5tO:tOCEAfEWtfUV _N 1373 1374 TOE NET CAPTURES 1373-1374 E01-25O M-M-SO*. 40-. 35-1 fa EE.. 1 SO-15-1 •4. 1 1 1 1 1 1 1 1 1 1 1 1 H 1 JU/USCOCNOCEJAFEKAAPl^ -M JJ AU 1373 1S74 -I H 1 1 1 1 1 H 1 1 1 1 TOE JUAUSE0CN3IX^FEMAAPMY_N 1373 1374 TIME Figure 22 Coincidence of kelp presence and fish capture FISH ABUNDANT AT RCCK SLOPE 30-100 MM FISH ASLN3ANZE AT ROCK 5LCPE 101-150 MM b TIME -*6 1 1 1 1 -+* Hfc^H 1 t-ND CE JA FE MA AP MY JNJUALI 1S73 1S74 TIME FISH ABLMWCE AT ROCK SLOPE 151-500 MM FISH ABLN3AJCE AT ROCK 5LEPE 331-550 M M 53' 40-S 301 M b 30. JJO-1 I -E 1 1 1 1 1 1 NO DE JA FE MA AP MY JNJU AU SE 1S73 1H74 TIME Observed abundances of fish at tha rock slope 49 (iii) Diel: S. jelanop,s juveniles undertake diel migrations throughout the immediate vicinity of the kelp bed. The extent of these migrations varies, often extensively, between different size classes of fish but the initiation, duration and completion remain relatively consistent. The lengths of the migrations are closely related to the size of the fish involved. Size-class distribution graphs for 1973 and 1974 (Figures 25 to 27) establish cumulative totals of in situ nocturnal and diurnal positions of the fishes, using the kelp bed as a frame of reference. Negative and positive numerals on the abscissa correspond to positions (in metres) shoreward and seaward from the outside of the bed, respectively. Statistical analyses (Table IV) of the positional abundances of the size classes of the fish show significantly greater numbers of fish inside the bed than outside. In 1973 this situation existed both diurnally and nocturnally. In 1974 the distribution cf the fish showed an alteration of habitat utilization by the larger size classes; in this year, 30 - 100 mm fish showed a consistent association with the kelp bed, both nocturnally and diurnally. Fishes of sizes 101 - 150 mm and 151 - 200 mm displayed a diel transition such that significantly higher numbers of fish were present outside the bed diurnally. Conversely, the nocturnal distribution patterns were the same as in 1973. Observations indicate that the inward migration of fish takes place before sunset and, in a similar fashion, the outward migration occurs after sunrise. No differential timing of the migration was noted between the two size classes. -id' -UJ' -a* -b- -A. -ti.' o« £• A' B.» B» 10» 12• Inside Outside Figure 25 Cumulative nocturnal and diurnal observations of fish position. 30 - 100 BD fish 51 Inside Outside Cuoulative nocturnal and diurnal observations of fish position. 101 - 150 nn fish 52 50-40.1 30.1 EO-I 10.1 SIZE CLASS DI5TN' 1373 151-200 m + Diurnal x Nocturnal -15»14. -12. -10. -B- . -4» 0« £• 4« 6» B» ID- IH- 14.15. 40-30-1 20-10.1 SIZE CLASS DI5TN. 1SGM ±51-200 MM + Diurnal x Nocturnal •X-X--15-14• -12« -10. -B. -S- -4. Inside 4« 6« B« 40 « 12' 14'1S» Outside Figure 27 Cuaulative nocturnal and diurnal observations of fish postion. 151 - 200 BB fish 53 TABLE IV Statistical comparison of diel positions of the cumulative totals of fish observed 1973 Size 30-100 mm 101-150 mm 151-200 mm Day I/O 1412*/1003 132*/45 118*/2 Night I/O 379/329 (+1m) 400*/308 14*/2 9/3 (+1 m)10*/2 Diel shift NO No No 1974 —+~ Size Day I/O Night I/O Diel shift 30-100 mm 101-150 mm 151-200 mm 4449*/1844 25/381* 13/105* 280*/37 191*/7 14/8 ( + 1 m) 18*/4 No Yes Yes (+1 m) - data includes those fish at +1.0 m as being inside the bed. * - significantly greater (p<.05) I - inside the kelp bed 0 - outside the kelp bed 54 3.6 Environmental correlates of distribution The diurnal distribution of 30 - 100 mm fish was consistent, in that higher numbers of fish were consistently found inside, rather than outside the bed. The observations of fish inside the bed were further analyzed to determine if any smaller scale migrations may have occurred and to determine if the water column was being exploited in any differential fashion by the various size classes of fish. Figure 28 illustrates the distribution of all fish under conditions of 2.0, 2.5, 3.5, 4.0 and 8.0 metres visibility. Data on the times, tidal heights and meteorological conditions on each occasion supplying visibility-distribution data, are presented in Table V. Efforts were made to observe the distribution patterns under approximately eguivalent conditions, There was no significant difference in the average height of the fish in the water column, nor was there any difference among the size classes - either within each sampling occasion or among different ones. 55 Changes in fish position with water clarity 56 TABLE V Environmental conditions on each date supplying data for visibility-distribution analysis j ; , 1 ; ! 1 Date jMeteorology I J, 1 27/VII/74I Clear,sunny I 25/VIII/74J " ! 28/VIII/74| " I 10/71/74 | " I 1/IX/74 } " Visibility Tide 8.0 m I 1.80 m 115:00 4.0 m I 2.25 m | 16:00 3.5 m J2.13 m |13:50 2.5 m 12.04 m I15:00 2.0 m 12.01 m 116:30 Time A second set cf environmental correlates with fish distributional patterns was that of nocturnal illumination and current magnitude. This influence was shown primarily at the rock slope, since tidal range had little effect on currents at the study kelp bed (Figure 12). Fish appeared to exhibit contradictory phototactic responses under apparently similar illumination; on moonless nights, fish could be found, on different nights, to be either down in rock crevices or up in the water column. Further observation suggested that the magnitude of the current in the area may have been influencing the fish's response to light. Table VI presents the results of laboratory experiments examining the influence of rheotaxis upon phototaxis. The presence of both phototaxis and rheotaxis was first established and was subseguently followed by the introduction of a current into the phototactic experiment to determine if any behavioural modification was induced from the observed 'normal' response. Results indicate an overriding 57 effect of current on the 'normal1 positive phototaxis, such that current of sufficient intensity will negate the phototaxis exhibited by the fish. The photonegative response indicated in the table refers to movement by the fish away from the light, which is associated with their movement toward the current. TABLE VI Initial conditions, manipulations and results of experiments examining the phototaxis of Sebastes melanojas and the reversal of this response generated by current changes I Date 21/1 22/1 1/11 2/II 11/II 18/II 1VIII No. | Initial Fish| Light Phototaxis H +-IInitial |Current 2 2 2 2 4 4 4 1441.27 lux|773 1/hr 1441.27 " |773 " 1441.27 | 0.11 1441.27 | 172. 16 J212.94 |773 | 773 |773 | 773 |773 Change in Light -269.00 lux -269.00 -269.00 + 4.09 -269.00 -172.05 -176.35 Besponse +photo •photo +photo +photo +photo + p hoto +photo Influence of Bheotaxis Date No. | Initial |Initial J Change in | FishJ Light |Response| Current | Response ; .|_ H f- H 7/1 1 2 | 0.05 lux| +photo 1+2319 1/hr ! +rheo/-photo | 3/II 1 2 1441.27 " I •photo 1+1546 ti 1 +rheo/-photo J 11/11 | 4 1441.27 " \ +photo | +773 1 +rheo/-photo | 11/11 1 4 I 0.56 " I •photo t +773 n 1 +rheo/~ photo J 18/II | 4 I 172.16 " I +photo | +773 1 +rheo/-photo | 18/II | 4 |172.16 " I •photo | +773 ii 1 +rheo/-photo j J.. ± .... _ _ J, .x +photo - photopcsitive response +rheo - rheopositive response 3.7 Fidelity and suantification of association (i) The experiment involving the establishment of a small group of plants 30 m distant from the kelp bed was successful to 58 the extent that the juvenile S. melanosis were observed among the plants diurnally, where none had been noted in the area previously, however no fish were observed nocturnally. The extension of one segment of the bed margin by 8m served to extend the contiguous distribution of fish normally associated with the bed. No differences in behaviour or abundance of fish were noted, concerning either those fish at the 'new' bed margin, or those at the artificially established bed, in relation to those observed within the natural bed. Tracking of fish with fluorescein tags indicated that the association of the fish and the kelp bed was relatively intimate. A typical tracking record is illustrated in Figure 29. Some cf the tracked fish displayed wide ranging forays (+15 m) but, generally, maintained a close enough association with the bed such that they could return to it when rapid advances toward them were made. Figure 30 illustrates numbers of fish and dates of the transport of tagged fish from the study site. Transplants were performed to test the fidelity of the association between the fish and the kelp bed as a habitat. These experiments examined the fish's ability to return ('home*) to a site of initial capture. A statistical analysis of the results of subsequent sampling for homing (Table VII) does not support the concept of significant homing ability to the site of initial capture, by these size classes of fish. 59 Ficjure 29 Tracking record of fish tagged with fluorescein tag 60 Figure 30 Transporting of fish and returns during fidelity experiments. 61 SiBLE VII Numbers, sizes and habitats to which Sebastes melanojas were transported, together with the numbers and sizes of fish exhibiting homing behaviour i— —r i - X | Date | No. ofj Size | Transport | No. | i i Fish | Range | Habitat | Returned | t _ j ...., . _ ...... i . _ i ., „. ,,„ i 1 X l 1 i I 18/VI/74J 7 | 118-123 mm Jopen water J 0 1 | 25/VI/74J 8 J 123-144 mm jopen water | 1-144 mm | j 4/VII/74| 4 1 134-191 mm | kelp bed | 0 | |18/VII/74| 4 1 120-151 mm |rock bottomJ 0 i |25/VII/74J 7 | 110-160 mm | kelp bed | 1-149 ram \ ! _! [_ I [ . _.. . J (ii) In general, the experimental work involving the habitat frames centred on the upright Frame I, This was because the fish were not attracted to Frame II as well as to Frame I; abundance of fish at the former was uniformly an order of magnitude less than at the latter. The results presented, therefore, refer only to Frame I data. Table VIII details information on the levels of shelter provided at the frame and the numbers of fish observed during each interval of that level (Figure 31). Figure 3J Fish at Frame I at 14.2 m2 and 33.2 m2 levels of shelter 63 TABLE VIII Levels of shelter and fish associated with Frame I, together with an analysis of the variance in these numbers in relation to changes in the shelter provided r T T r i 1 | Date j Level Fish at] Fish at jProportion off i lof Shelter 1 Frame | kelp bed I fish at Frame) j j j. -„ _ T i • -1 11 Vv i 23. 7 m2 J 3 • I 8 1 0.27 I I20/VII j II 125 | 373 ! 0.25 | |20/VIIII II 3 I 13 ! 0.19 | i22/VIIIj II 156 ! 138 ! 0.53 | J 24/VIIIJ 33.2 m.2 | 125 | 213 ! 0.37 | |25/VIII} •i 80 j 336 ! 0.19 I |27/VIIIj ii 175 | 196 ! 0.47 | |28/VIIIJ n 181 373 0.33 | I 1/IX | II 7 ! 523 ! 0.01 J" I 2/IX | ii 55 169 ! 0.25 1 I 3/IX | 14.2 m2 | 10 120 S 0.07 | I 4/IX j II 20 292 I 0.06 | | 5/IX | II 85 283 I 0.23 | 4 , i j. _ _ _ i i . i i i r ; T I|Source of | Degrees of 1 Sum of j Mean t • F' 1 j Variation t Freedom |Sguares j Sguare | Statistic J 4 J 1 J | j 1 J J J J j Among I 2 1 o. 61 | 0. 30 | 16.67** J i Within 1 10 1 o. 22 | 0. 018 1 1 j Total 1 12 1 o. 83 | ! ! L j J. - L. . J i 64 The numbers of fish at the frame are presented as the proportions of the total number of fish seen on the observation day. The transformation of the raw data into this standardized form was necessary tc compensate for the large inherent variability in the numbers of fish observed on any given day. A transformation of this nature bears the justifiable assumption that both the kelp bed and the habitat frame draw upon the same •universe' of fish to populate their environs. The results of a single classification analysis of variance {ANOVA) for the three levels of shelter are also included in the table. The *F* statistic resulting from the ANOVA (16.67) indicates that the variance in the numbers of fish observed under each shelter treatment is significantly accounted for by the changes in the treatment, rather than by factors outside the experimental design. 65 4. DISCUSSION 4.1 General distribution The use of kelp beds by the fish should be examined in relation to its total geographic and bathymetric range. The geographic distribution of Sebastes melanops was not delimited by its describer (Girard 1856), The first major range statement being made by Jordan and Evermann (1898:1783), who described the distribution as extending from "...Monterey to Kadiak £=Kodiakj, most abundant northward; very abundant at Sitka ...at San Francisco, much less common than S. my.stijrus; about Humboldt Bay much more common being the most abundant food fish.", The known distributional limits have been extended only slightly since that time. Miller and Lea (1972:96) describe the range as extending from Paradise Cove [California] to Amchitka, Alaska..,". A similar situation exists regarding the depth range. Early papers infer that the species is a shallow water form; the increasing depth range reported in succeeding years reflects an increased sampling efficiency and effort rather than the the abundance of the species at the depths involved. The current delimitation of depth is probably an accurate estimation of the true bathymetric range of the species, since sufficient effort has been expended in deeper waters but has produced no specimens of S. melanojas. The reported (Miller and Lea, 1972) depth range (3 - 92 m), when correlated with the microgeographic range of the fish, leads to a description of fish's distribution (as an adult) as being closely associated with the coastal waters. A manuscript compilation, of the results of catch surveys (Niska, 1973) on S. melanogs shows a mode of abundance 66 of adults from 0 - 50 m deep in areas from 1 - 10 km from the coastal region. An earlier publication by Alverson, Pruter and fionholt (1964) indicated a much deeper mode of abundance (92 -181 m) than the Oregon report. Several features of the earlier study should be considered when assessing the distributional patterns of the adults: firstly, the sampling effort was considerably higher in the 9 2 - 181 m depth range; secondly, abundances were gauged by the freguency of occurrence in total trawls; and, finally, trawling was done primarily during the winter months. An increased sampling effort might have indicated greater abundances in shallower water. The use of a freguency of occurrence measure masks abundances. With regard -to this point, the highest freguency of occurrence and the greatest yield of the species (by weight) did not occur at the same depth range; the maximum yield occurred at shallower depths. The seasonal nature of the fish's movements may indicate that the reports do not represent distinct differences in the modes of abundance. Alverson (1960) has postulated, on the basis of catch data, that S. alutus Jordan and Gilbert undergoes a seasonal migration, moving into deeper water during the winter. In view of the large numbers of demersal fish which undergo such a seasonal migration, it is probable that S. melanojjs follows suit, this may account for the apparently distinct distributions in the two reports. The fishing areas (Figure 14) are generally low to moderate relief exposed segments of coastline or closely adjacent to them. Commercial catches of the fish reflect relatively low abundance of the adults in most areas (when compared to S. 67 alutus and S. £i;nnic[er Jordan and Gilbert, or the other fish most abundant in the same hauls with S. melanoo_s) accessible to commercial fishing. Dunn and Hitz {196 9) reported a spent female specimen of S. melanop,s 385 km southeast of the Alaskan peninsula, over water 2000 m deep, but in the absence of similar offshore records of the species, this must be regarded as anomalous. The inshore coastal distribution of the species has received more attention than that of the nearshore, because it is a component of the sport fishery catches in California, Oregon and, to some extent, Washington and British Columbia. Hhile S. melanops rarely exceeded 5% of the total inshore catch, it ranked 12th by numbers and 8th by weight for all species of sportfish landed between Oregon and Port Arguello,California from 1957 - 1961 (Frey,1971:75). Sport fish catches lend corroborative evidence to the inshore nature of the species; Frey describes it as being, "...primarily a shallow reef species...". Delacey, Miller and Borton (1972) record the largest collections of S. melanops, taken by the University of Washington, as coming from the inner reaches of Puget Sound (Colvos Passage). Adult distribution, therefore, does not appear to be strongly associated with the occurrence of kelp beds, although populations may be within several kilometres of them. Published reports and comments by commercial fishermen, as well as personal sampling, estatlish the inshore adult populations as being in moderate relief areas, close to the coastline or shore, often with considerable exposure to wave action. 68 The larva of the species has never been described, ahlstrom (1965), LeBrasseur (1970) and Haldron (1972) have described the broad coastal distribution of "Sebastes species' for the California - Eaja, northeast Pacific and Oregon B.C. areas, respectively. Their results definitely show that the larval distribution has a strong coastal or inshore (<914 m in depth) component. The highest abundances of Sebastes species larvae, excepting seamount stations, were found within 10 kilometres of the coastline, Published accounts of the distribution and abundance of juveniles are few, if any; I have been unable to locate any reference which treats the distributional aspects of the younger life stages of the species in anything except gualitative form, Hubbs and Schultz (1933) described a form related to S. ffi«lano£s based upon 5 specimens obtained in a salmon trap, in 1926, at the mouth of the Columbia River, They designated the new species as Sebastodes [-Sebastes] columbianus and differentiated it from Sebastodes melanges on the basis of eye diameter, colour, and body configuration. The meristic counts published overlapped with those of S. melanojas. The specimens employed in the description of S. eglumbianus were all over 400 mm TL, yet the authors stated in their diagnosis of the species that "...young not occurring in tide pools of reefs.", whereas the young of S. melano_ps were described as "...developing in tide pools."(Hubbs and Schultz,1933:21). The point of significance is that, in Barkley Sound waters, the young of S. jelano^s are rarely, if ever, found in tide pools. The form described by Hubbs and Schultz was reduced to subspecific status 69 by Alverson and Welander (1952) for Washington and B.C. waters. Phillips (1957), however, does not designate any subspecific forms of S. melanoDS, nor do any of the California works following his paper, e.g., Bitz (1965) or Miller and Lea (1972). Specimens of British Columbia forms have been uniformly described as Sebastes melanges, and I follow this practice for the Barkley Sound forms. 4.2 Barkley Sound distribution The first record of the species in Barkley Sound was reported by Evermann and Goldsborough (1907) who described a specimen taken by the United States Fisheries Commission steamer Albatross on September 23,1888. This record is guestionable because the station records of the Albatross do not indicate that it was in Barkley Sound on this date. My sampling within Barkley Sound indicates a macro-distributional pattern correlated with the size of the fish. No data are available which allow determination of any similar patterns within the coastal range of the adults susceptible to commercial fishing, Commercial fishermen of the Bamfield region attest that, while the adults are occasionally caught in the inner parts of Barkley Sound (Satellite Passage, Dodger Channel) the largest specimens of the species are caught past the mouth of the Sound. These reports corroborate the results of the sampling of this study; however, they should be interpreted with caution for two reasons. Fishing vessels of the Bamfield region are salmon trollers and are, basically, upper water column samplers; and fishermen describe the fish as 'black bass' and at 70 least two species caught at the. mouth of Barkley Sound are commonly denoted as such (S. melanops and S. sistinus Jordan and Gilbert). The latter species is, generally, found only at the mouth of the Sound during the summer months, and appears to accompany the northward intrusion of warmer water during this period. In summary, the adults of the species appear to be concentrated at the mouth of Barkley Sound, rather than being distributed more uniformly, or in a different biased fashion, throughout the area. Such a distribution, if maintained annually, establishes the centre of reproductive effort as being along the coastline, or slightly farther away, if the fish undergo a seasonal migration. Insufficient data exist to confirm this, since no gravid females were taken during the study and there are no published accounts of either gravid female capture or their occurrence/distribution. Results of sampling in the present study show a definite increasing cline in the size of fish captured, when sampling from the head to the mouth of the Sound (Figure 19). This cline is corroborated by observational data. These data also establish higher abundances of 0, I and II age-classes in the inside of the Sound, than at the mouth. A plausible life history of the species must account for this distribution. The plankton tows taken in Trevor Channel and in the Deer Group islands in 1973-1974 showed significantly higher numbers of Sebastes spp. larvae in sheltered waters of islands and bays than in open channels. Hourston (1958,1959), working in Barkley Sound, found a similar distribution of juvenile herring in 71 regard to these two types of hydrographic situations, and postulated that such a distribution was achieved through dispersal from the spawning grounds (primarily on the northwest side of -the Sound) and gradual accumulation of the juveniles in these sheltered areas. He concluded that such dispersal was neither passive, with the currents, nor was an active searching on the part of the juveniles for such environments. Hoursfon believed, on the basis of several tests of distributional correlates, that the aggregations of juveniles were the result of primarily random movement by the fish; movement which diminished when a suitable environment was found. The attractiveness of the sheltered areas was thought to be related to the decrease in physical stress which the herring experienced while there. He also demonstrated an increased abundance of the fish in the most sheltered areas of the bays and inlets; around, "...docks, log booms, kelp patches and rock piles." (Hourston,1959:303). The concentrations of Sebastes spp. larvae found in my study generally mirror that reported for juvenile herring. The fine-scale distribution of the larvae may or may not have reflected the same pattern, but with regard to the kelp patches, it was impossible to determine this because the new year*s growth of kelp sporophytes was insufficient to establish visible kelp patches. Although the abundance of larvae was greater in many areas which later contained kelp beds, there was at least one large bay (fioguefeuil Bay,Figure 3) which supports both large kelp beds and S. melanop_s populations during the summer, within which very few Sebastes spp. larvae were taken. The 72 distribution of S. mej.anoj)s larvae can only be extrapolated from that of observed specimens of Sebastes spp. The size of the juveniles of S. melanogs appearing In the kelp beds in mid-May indicates that the time of spawning for this species agrees with the February Puget Sound spawning records (Delacy et al.,1964), rather than with the April Vancouver Island reports (iestrheira et al. , 1 968) . The segregation between the observed abundances of adults and those of larvae may be explained by physical oceanographic factors. Doe (1952) found high surface transport out of Barkley Sound on the west side of the Sound during the spring. This surface transport was primarily a result of freshet conditions in the Toguart and Somass rivers, A corresponding inward flow of water took place on the east side of the Sound and in deeper water on the west side, during this period. It is plausible that S. melano^s larvae, if extruded at the coastline, could be passively carried into the Sound with these currents. Observational samples show S. melanops juveniles, especially 30 - 100 mm fish, to be more abundant in beds of J2il££2£istis intec[rifolia than in beds of the more exposed Nejceocystis luetkeana. Higher numbers of fish were found in M. intecjrifolia beds in sheltered waters (Ross Islets, Eoguefeuil Bay) than in beds of the same plant in more exposed locations (2nd Beach, Cape Beale). These observations, when coupled with the cllnal variation in size of sampled fish, lead to the conclusion that the observed distribution of fish is related to the degree of exposure of the habitat or, more generally, the wave stress imposed upon the fish. This 73 conclusion may not reflect the only cause and effect mechanism at work but it is definitely a major component in the determination of the distributional pattern of this size group of juveniles. 4.3 Sebastes melano£s in Microcystis beds The presence of the species in beds of Macrocystis is limited to the juveniles in that none of the fish from the kelp beds were found to be sexually mature. The smallest fish in the kelp bed were 40 - 50 mm young-of-the-year. These fish would be designated as the 0 - I age-class and would include those fish up to 100 mm. The population of the fish at the site included three or, infreguently, four age-classes {Table II). The abundance of the fish at the site was reasonably constant on an annual basis for the ages II to IV fish. Major fluctuations in the population were a result of yearly recruitment by the 0 - I age fish (Figure 18). The recruitment and growth of these fish to a size which was capable of being tagged is reflected in the sudden increase in estimates of population in August,1974. The recruitment of these individuals was obvious in observations beginning in June of the year but it was not possible to tag them until early august, due both to the small fish (<80 mm) being unsusceptible to net capture and unable to support a tag. The close agreement between population estimates computed in late fall, when the kelp bed was degenerating and in the late spring, when the new year's sporophyte growth began, infers an approximately constant population overwintering at the site. These estimates may indicate a probable low value of natural mortality over winter, as well as a minimum residency of tagged individuals of at least one year. The similarity of these estimates may also indicate that mortality, followed by immigration was occurring; however, this would necessitate differential mortality cf untagged fish in order to maintain similar estimates, and I think this situation unlikely. The initial increase in the population estimate (by both methods) in early 1974 is a result of the previous year's juveniles becoming accessible to the tagging program. The large increase in the population estimate in August,1974 is not mirrored in 1973 because very few 0 - I juveniles were tagged in 1973 and, the recruitment of this age class of fish was higher in 1974. A third possible factor responsible for this increase was the addition of the two habitat frames at the site in this year. The degree to which these may have increased the resident population is unknown, however an influence is inferred by the facts that the increase in the population was almost exclusively through this size class and that these were the fish which were attracted to the habitats. A degree of learning by the fish is indicated by the consistently lower estimates generated using observational data over those using recapture data. A higher proportion of tagged vs. untagged fish were observed than were represented in the net captures. This is most probably accounted for by net avoidance on the part of tagged fish and concomitant decrease in recaptures. Observational data collection did not necessitate capture and handling and the sampling technigue was not biased 75 toward untagged individuals. In light of these considerations, I feel that the estimates generated using the observational data are the more accurate of the two. In addition, the availability of the tagged fish to observation, as opposed to recapture, is responsible for the lower degree of fluctuation in the observation estimates. 4,4 Diet An examination of the diet of those fishes present at the site leads to the conclusion that the population in this area directs the majority of its foraging activity to the water column, rather than to the benthos. The clupeiform fish and mysids which form the dominant portion of the diet are water column inhabitants. Neither of these items were found to be in higher abundance inside the kelp bed than outside. Clupeiform fish, of the size found in the stomach contents, were never observed inside the bed, albeit they maintained an intimate spatial relationship with it at times. In the few instances when S. j§lano£s were actually observed feeding on clupeids, the situations were very similar. A school of Clup_ea hargngus oallasi Valenciennes would pass close to the bed and an individual S. melanojjs would dash out and capture one of the small herring. On every occasion, the individual capturing the herring had pieces of the uningested herring torn from its mouth by conspecifics. Some alternation in the use of trophic resources on a seasonal basis, similar to those described by Gotshall et al. (1965) is evidenced in Figure 21. While some overlap 76 exists, there is a segregation between the periods of maximum occurrences of mysids and clupeids in the stomachs of the fish. It is known that the young-of-the-year herring appear near kelp beds in late May (Hourston,1958,1959; Marliave, personal communication). Mysid blooms have been noted in 1974 and 1975 to occur in early May. These times of occurrences correspond generally to the times cf occurrences of these diet items in the fish's stomachs. I do not have guantitative data on the abundances of these two primary prey items in the environment and it is not possible to construct any index of trophic selection such as those similar to Ivlev's (1961) ' E* index, e.g., frame (1974) or Gerald (1966). The anlysis of the diet data differs from previous diet analyses performed under a similar data handicap. I have tried to eliminate the subjectivity of the methods of Tester (1932) and Hynes (1950) as well as avoid a sole reliance on the occurrence of diet items (Quast,1968), The dominance data employed are conservative estimates of the contributions of various diet items rather than optimistic ones such as results from an analysis based on occurrence of items alone. A more detailed analysis, such as that of Carr and Adams (1973) was not possible, while a less detailed one (Fritz,1974) did not yield sufficient information. The analysis in my study attempts to improve on the basic occurrence/volume presentation (Heufcach et al.,1963; Gotshall et al.,1965). No experiments on the feeding behaviour of Sebastes melanogs were conducted. It must remain unknown, on the basis of currently available information, whether the abundances of 77 items in the diet represent preference or specialization (Bryan and Larkin,1972; Landenberger,1968; Hood,1968) or a passive response to differential abundances of prey items in the environment (Gerald,1966; Murdoch,1968). 'Switching1 (after Murdoch, 1968; Boiling, 1965; Tinbergen, 1960) by S. j§lano£j= would result in a similar record of diet item occurrence if such a seasonal alteration of mysids and clupeid fish had occurred; however the studies of Murdoch (1968) and Landenberger (1968) indicate that even sophisticated experimental design will not prove the existence of such switching. Strong preferences by predators tend to negate differential abundances of prey items in the laboratory and preference testing is generally inconclusive when the predator is faced with multiple prey items. In addition, Holling (1959) found that predators would occasionally select ncn-preferred diet items even when the preferred item was abundant. Increasing the levels of the preferred item even further yielded an increasing utilization of the alternate diet items. While the size, appearance, habits and taste of the prey may act to govern their abundance in the diet (Allen,1941), Gerald (1966) found good correlation between forage ratios of diet items and their abundance and/or availability in the field. The latter indicates a purely passive feeding response on the part of the predator, when faced with differential abundances of diet items. I use 'passive' in the sense that the alternate prey is eaten merely because it, "...offers a sufficiently higher rate of reward..." (Croze,1970:79), rather than actually connotating greater preference, in the sense of Murdoch (1968). 78 Although it is impossible to determine the degree of selectivity, if any, which the fish exhibit toward their total diet spectrum, an analysis of the relative contribution of the various diet items can be made. The statistical analysis of the volume - based dominance data infers a general reliance by the species on clupeids and mysids for the vast majority of its energy input during its presence in the kelp bed. The presence of mysids as the dominant item for 101 - 150 mm fish, while the other two size classes exhibit clupeid dominance, warrants comment. Firstly, the sample size for the 30 - 100 mm fish is very small for a positive indication of major diet input. Secondly, those herring eaten by the smallest fish were larval or juvenile forms and thus, probably not attractive to the larger fish as a diet item. For the duration of their presence in the kelp bed, fish of the largest size class examined (151 -200 mm) exhibit overwhelming reliance on clupeid fish. while the fish have been shown to utilize clupeids and mysids to a greater degree than other diet items during their period of residence in the kelp bed, no evidence is available which describes their primary diet input during the winter months, when they adopt an alternate habitat. The seasonality of their summer food sources (Hourston,1959) reguires an alternate winter diet. The metabolic rate of the fish and hence their feeding rate, is necessarily smaller in the winter months; evidence for this is both direct (temperature control of metabolic rate) and indirect (annual checks in scale circulus formation). The identity of the winter diet is unknown but in consideration cf the habitat associations of their summer forage items, I would 79 assume that S. me la nop. § utilizes pelagic species as its primary energy source during the winter as well. Previous work in Barkley Sound (Druehl, Green and Leaman, unpublished data) has shown that S. melanojjs in other locations utilizes sandlance (Ammody.tes hexapterus Pallas) as a food source late in the summer. later work in Nereocystis luetkeana beds (Leaman, in preparation) has shown that shoals of 80 - 100 mm sandlance appear in these beds in late August - early September and present a demonstrably exploitable resource. 4.5 Activities in the kelp bed The utilization cf the kelp bed as a habitat by the fish was strongly seasonal as a direct result of the seasonality of the bed itself (Figure 13). The onset of sporophyte growth in May - April was not immediately accompanied by the appearance of Sebastes melanges in the bed. The time lag between growth initiation and the appearance of fish is a function of the degree of water column occlusion by the growing plants. Fish did not occur until after the percentage-occlusion had exceeded 75%. The apparent disparity between the appearance of the fish at this occlusion value and the maintenance of fish at lower values later in the summer is related to the morphometry of the plant throughout its growing season and the technigue of measurement of occlusion. The plant grows from a basal holdfast and proliferates to the surface through a graduated and continuous series of fronds. The progress of the growth of these fronds through the water column, albeit via intercalary meristematic tissue, is analagous to the growth cf a terrestrial 80 shoot. The water column is not occluded gradually as a whole \ but progressively from the bottom to the surface. The fish were not present in the bed until the plants were approximately one-half of the distance to the surface. Ihe occlusion measurements were made only 0.8 m above the substrate and were not necessarily representative of the entire water column exploited by the fish, which is primarily a mid-water-column form. The maintenance of fish in the kelp bed in August - September, when the occlusion values fell below 60$, again reflects the plant's morphometry and the measurement technigue. Lower laminae of the fronds degenerate in late summer resulting in the decreased occlusion values. Upper and mid-water laminae of the plants remain viable during this period (Figure 32) and continue to provide an occluded water column. The fish appear to be unaffected by the changes in the lower portions of the plants. This situation contrasts with the response of birds to changes in the vertical structure of terrestrial forests (Hagar, 1960; Martin,1960), wherein bird communities change both qualitatively and guantitatively with changes in the successional status of the forest. It also contrasts with the species changes associated with differences in forest foliage profile (HacArthur and MacArthur,1961). 81 Figure 32 Surface canopy of the plants in late August,1974 Daily observations of fishes in the kelp bed were characterized by a large variation in the numbers of fish seen, similar to observations in California kelp beds (Quast,1968; Turner et al.,1969). This variation in sightings related primarily to the smallest size class of fish due to their greater abundance. Several features may have acted singly, or in concert, to mask the actual abundance of fish. The foremost of these features was the clarity of the water; water in which visibility is decreased through phytoplankton blooms, terrestrial run-off or particulate detritus, severely hampers observation. I consistently made cnly one set of cbservational sweeps through the kelp bed to minimize the possibility of duplicate enumeration. Observations made during periods of 82 lowered visibility were more susceptible to duplicate sightings because of the possibility of undetected movement of the fish to unenumerated areas. Observations made during or after storms often yielded few sightings because the fish were sheltering in crevices, both in and around the kelp bed. The additional time reguired to make a more thorough search of these refugia resulted in a decrease in the total time/area observations. Similarly, the illumination in the bed (distinct from water clarity) affected my perception of the fish in the water column. The variation in sightings is thought to be primarily a function of factors affecting observations^ rather than a reflection of actual variation in the numbers of fish present. This does not negate the fact that such variation in numbers undoubtedly occurs on a daily basis. The actual variation in the numbers of fish inside the bed may be an expression of the abundance of the larger size classes of fish, since they are known to undergo more extensive migrations than the smaller fish. The local distribution of the fish and their diel activities have parallels in both other kelp bed studies and in limnological work. Limbaugh (1955) found a similar pattern of greater abundance of fish at the edge of California kelp beds. He did not, apparently, make any observations of the clines in the abundance of fishes with position in the kelp bed, other than noting that the juveniles of some species appeared to be more abundant in the inshore rocky zone than inside the kelp bed. The larger size of the plants in California and the greater depth in which they grow introduces an added habitat complexity which is not mirrored in British Columbia kelp beds. 83 Limbaugh was able to distinguish whether species were primarily bottom, mid-kelp or canopy fishes. The shallowness of my study site precluded any depth distinction of this nature. Sebastes melano£_s appeared to exploit the entire water column, save for the surface layer above the thermocline (at approximately -1.0 m). Quast (1968) also distinguished utilization of segments of the kelp bed by various species, although he did not detail the smaller scale movements of individual species. Neither of these studies quantified the pattern of abundance of kelp bed forms within the bed, nor did they encompass diel observations. Their qualitative comments about various species included an assessment of the •dependency* of the species on the kelp bed biotype but their assessment of 'dependency' was primarily based upon the the degree of crypsis in the colouration of the fish. Fish showing no cryptic colouration were assumed to be non-dependent on the kelp bed. S. melanogs would fall into this category. The lack of diel observations in the aforementioned studies hamper the comparison between S. melanogs and the probable California analogues, S. serranoides Eigenmann and Eigenmann, faralabrax clathratus Girard and S. saxicola (Gilbert). These species appear to occupy the same type of habitat as S. melanops but diet analysis (Quast,1968) indicated that they forage more on benthic invertebrates than S. melanops does. High values of clupeoid fish occurrence in the stomach contents of the California fishes was thought to be a function of the sampling technigue ('chumming'), rather than an expression of the normal feeding behaviour of the fish. The diel activities of S. jelanop_s are thought to be a 8U result of the feeding benefits derived from an extended diurnal range and the survival benefits associated with the utilization of a nocturnal refugium from predators, such as Ophipdon elgngatus Girard, S. jjaucisp,insis (Ayres) , Sgualus acanthias Linnaeus and Salmo aairdneri Richardson. Diel changes in gut content and sampling for diet items within the kelp bed do not indicate that the kelp bed is utilized as a food resource base fry Sebastes melanops. More recent work in California kelp beds (Bray, personal communication) has investigated the diel activities of OxyJulius californica Gunther, Phanerodon furcatus Girard and Brachyistius frenatus Gill in terms of foraging and habitat preferences. Activities of these species were determined from laboratory swimming experiments, limited field observations and the diel fluctuation of gut contents. All species were diurnal feeders; 0. californica sheltered by burrowing in sand at night. No information was available on the v migrations of these fish. Bray has also investigated the diel activities of Chromis punctipinnis (Cooper) and H^£erp.rgsopon ajraentjjum Gibbons. The former species feeds by day in the kelp bed and hides in holes at night. The latter species schools inshore during the day and migrates out to the kelp beds to feed at night. The activities of H. argenteum are the converse of those of Sebastes melang£s. The activities of S. m§lano£s in Barkley Sound appear to be analogous to those of some species in California kelp beds but it is difficult to accurately compare them because the information on the southern species is either of a gualitative nature only, or it is unfinished or unpublished. 85 Work cn coral reef fishes by Hobson (1973) elucidated their diel feeding migrations, characterized by movement away from the reef at night to feed and schooling near the reef diurnally, in response to predation pressure. This type of migration, although temporally opposite, is similar to that of S. melanops. except for the schooling behaviour. The tropical counterparts are certain sciaenids and pomadasyids. Baumann and Kitchell (1974) described diel feeding migrations between diurnal limnetic feeding areas and the nocturnal littoral area in juvenile lepomis macrochirus Raflnesgue, which correspond to the movements of the larger size classes of S. melanops but not to the smallest. In addition, the migration by L. macrochirus was primarily by the smallest juveniles (the larger fish remaining in the littoral area) and was thought to be solely related to feeding; no mention of refuge seeking is made. With regard to the activities of Sebastes melanops, the diel migration of 101 - 200 mm fish, which occurred in 1974, was not noted in 1973. I believe that these migrations are density - associated phenomena. The abundance of the smaller (30 - 100 mm) juveniles was very much higher (approximately five times) in 1974 than in the previous year (Figure 18) and since this size-class was the most intimately associated with the kelp bed, the competition for trophic resources in the immediate vicinity of the kelp bed would have been concomitantly higher in that year. The larger fish, being relatively less vulnerable to predation, were able to avoid this increased competition by exploiting more distant trophic resources which were unavailable to the C - I juveniles. 86 The behaviour of S, melanops has biological correlates in terms of feeding and shelter. There are also environmental correlates of these activities. Increased water clarity results in a movement of the smallest size-class of fish deeper into the kelp bed. This response appears to be a graded response above a certain threshold level (Figure 28). Distribution of 30 - 100 mm fish is centered about the edge of the bed when visibility is less than 3.5 m. Movement into the kelp bed is initiated under conditions of increased water clarity. The juveniles are not cryptically coloured to a great degree and the increased water clarity would increase their vulnerability to predation by visual predators, hence the inward movement with its concomitant increase in shelter and security. The larger size classes of fish do not appear to participate in this retreat, reflecting their size-associated decreased vulnerability. Size-dependent susceptibility to predation has been detailed for a wide variety of organisms, both aguatic and terrestrial (F,E,Smith, 1954; fi.L.Smith,1966; Pianka,1974) and the behaviour of S, melanops juveniles is only one of several mechanisms which organisms employ tc escape predation on juvenile forms (crypsis, shelter-seeking, mimicry, morphological adaptations, etc.). The alteration of the distribution of the fish through its rheotactic response is another environmentally correlated facet of the behaviour of this species. The significance of this behaviour is open to speculation. The species is apparently exchanging a degree of security, which it gains through its association with the kelp bed or rock slope, for a benefit derived from its hovering postion in the tidal currents. 87 Several explanations may be presented for this behaviour. Given the fact that the species is primarily a water column forager, it may simply be optimizing its foraging opportunuties; the response may be the expression of a faculty necessary in the adult stage of its life to maintain its position in exposed habitats; or, the currents generated may exceed the •crtical velocity* (Pavlov et al.,1972) at the substrate level, forcing the animal up into the water column. The existing information is not sufficient to indicate the value to the fish that such behaviour imparts, but the assumption must be made that it does not evidently act to the fish's detriment. 4.6 The nature of the association In the analysis of the association of Sebastes melanops and beds of Hacrocy.stis, the data gathered indicated that the fish was using the bed as a physical refugium, probably from nocturnal predation. The diel distribution patterns clearly indicated the proportionate increase in the nocturnal utilization, especially when considering the larger size classes. Results of experimental plant manipulation showed that the distribution of the plant has a definite effect on the distribution of the juvenile fish, both in terms of extending an existing distribution and attracting fish to a new location. The transporting experiments did not show a significant homing ability on the part of S. melanops juveniles, when displaced from their point of capture in the study kelp bed. A homing ability may be present in the adults, after the fashion of S. flayidus (Carlson and Haight,1972) but this ability has not been 88 demonstrated for the juveniles of any species of Sebastes. The lack of returns of transported fish indicated that the juveniles were either unable to home (or had not yet developed the ability) or, that there was no feature of a kelp bed that was sufficiently important to the species to warrant the development, over evolutionary time-scale, of an ability to return to a given bed. The habitat frame experiments were conducted to elucidate what general feature of any kelp bed supported the obvious association between the two species. They were designed to provide only the shelter component of the kelp bed and to do so in an abiotic fashion, which would test whether this component was of significant importance in attracting fish. The highly significant response of the juveniles to the changes in the shelter provided (Table VIII) clearly indicates that the physical shelter provided by the plastic (and by analogy, the kelp) represents a component of the habitat which may potentially determine the numbers of fish present within the environs of a given bed. The importance of the shelter component of a habitat in attracting fish populations has been indicated previously by the provision of artifical substrates (Fujimura and Kami, 19 58; Kami,1962; Carlisle, Turner and Ebert,1964; Ogawa,1966,1967, 1968; Ogawa and Onoda,1966). None of these studies attempted to quantify the numbers of attracted fish with the amount of shelter provided at a given size of reef, rather, they examined various sizes of reefs. Ogawa amd Aoyama • (1966) did find that the provision of plastic strips inhanced the attractiveness of any artificial reef but they neither quantified this enhancement, nor manipulated the number 89 of plastic strips, Senta (1965) related the results of studies conducted in Japan ' on the association of fishes and floating clumps of seaweed. The fishes found in and around these clumps were primarily pelagic, oceanic forms and Senta regarded the fishes as using them mainly as refugia from predation. He found no relation between the size of the clumps and the number of fish present but he did find that the degree of structural heterogeneity of the clumps was a major factor in attracting fish. In addition, he found that fish exhibited varying degrees of association with the seaweed, Small Sebastes inermis Valenciennes leave these seaweed clumps and move into the nearshore seaweed beds when they reach 50 - 60 mm, The response of S. • melanops•to plastic * kelp1 parallels the response of pelagic fish noted in this study. Quast (1968) stated that the fish utilized the kelp as orientation points in the water column to establish their contact with the bottom. I do not believe this is the case with S. melanops because visibility is not related to the position of the fish in the water column. The fish appears to be using the physical presence in the water column as shelter rather thas an an orientation point with the substrate. Hunter and Mitchell (1968) found a disproportionately large number of fish surrounding their three-dimensional raft when comparing its size to two-dimensional rafts of similar size. They inferred that the size of the three-dimensional raft might not be as important as in a similar two-dimensional one although they employed only one size of three-dimensional raft. My 90 studies with three-dimensinal shelter suggest that, while a three-dimensional object is more attractive, a similar relationship between size and abundance of fish exists with three-dimensional objects, as does with two-dimensional ones. The results of my studies using artificial seaweed dc not illustrate that the amount of shelter provided by the kelp plants is the sole cause and effect relationship governing the abundance of juvenile S. melanops present; however, they do provide proof that this component of the habitat is one to which the fish show a very strong in situ numerical response. 4.7 Significance of the keljD bed to the fish The information presented in this thesis, together with published material, indicate that the association between S. melanops and M. integrifolia in Barkley Sound is almost exclusively concerned with non-sexually mature juveniles. Diet analysis does not imply a reliance by the fish on a food resource base which is strongly associated with kelp beds. A strictly biotic association between these two species is not evidenced by this study. I believe that the significance of the kelp bed to S. melanops is that o;f a physical refugium from predators. While feeding in the kelp bed undoubtedly occurs, observations of feeding and inference from diel migrations indicate that adjacent environments contribute the majority of the food resource base. The association is more constant for smaller size class fish and the dimunition of it with increasing size infers that the smaller fish are more dependent on the kelp bed than larger 91 ones. The distribution of the adults of the species and the increasing cline in the sizes of fish observed in Barkley Sound, lead me to believe that the vulnerability of the young to predation results in a dependence of the species on kelp beds, and probably other sub-littcral refugia, for the initial survival of the yearly recruitment to the population. The appearance of recently transformed juveniles in the early stages of the yearly sporophyte growth illustrates both the coordination of spawning with kelp bed development and the provision to the fish of some advantageous feature about the kelp bed. Experimental results show that the juvenile fish will respond to the provision of physical shelter in a manner similar to their response to natural kelp beds. The results lead tc the conclusion that the shelter of the kelp bed is of primary importance to the juvenile Sebastes melanojjs. 92 5. CONCLUSIONS (i) The life history of Sebastes melanops encompasses a change in habitat association with the size and age of the fish. Adults are associated with the coastal zone while the juveniles associate with inshore features, particularly kelp beds. Young are spawned in late January - early February and appear in the kelp beds in mid - May. Juveniles spend up to four years in these inshore areas. An increasing cline in the size of fish from the inshore to the coastal zone is present in Barkley Sound. (ii) Higher abundances of juvenile S. melanops are found in JSS££2£istis -jntegrif olia beds than in beds of the more cumatophytic Nereocjstis luetkeana. In addition, the abundance of juveniles in Microcystis beds is inversely correlated with the degree of hydrodynamic exposure of the bed. (iii) In the kelp bed studied, three, or occasionally four, age classes of juveniles are found in the beds, none of which are sexually mature. The absolute abundances of II, III and IV age-class fish are, generally, consistent over time. Fish recruited to a kelp bed often overwinter in the same area during the absence of kelp. (iv) Smaller-sized juveniles (30 - 100 mm) are more intimately associated with the kelp bed than either of the other two size classes. The latter fish occasionally make diel feeding migrations from the kelp bed, returning to the bed before nightfall. These migrations are thought to be a response to the greater feeding competition resulting from higher densities of population recruits in some years. 93 (v) The kelp bed represents neither a feeding nor a reproductive sanctuary for the fish. Adjacent environments supply the primary food resource base. Fish may be ".resident* around a particular kelp bed during their inshore phase, but experiments indicate no fidelity to that bed should displacement occur. Distributional pattern of fish around the kelp bed has both biotic and abiotic correlates. Abundance of fish is both qualitatively and quantitatively correlated with kelp abundance. Fish adopt an alternate habitat durinq periods of kelp absence. At some size, in excess of 200 mm, inshore forms undergo a gradual migration to the coastal zone. (vi) The kelp bed presents several ecological benefits to the fish, the main one of these being the physical shelter of the plants and their occlusion of the water column. The provision and manipulation of artificial seaweed shelter induces a numerical response in the fish population, similar to its response to the shelter provided by the kelp plants. (vii) The kelp bed represents an important physical refugium to the fish. Its primary value is in the enhancement of the survival of the C - I age class juveniles. 94 6. LITEEATORE CITED Ah 1st ro ID, E. H. 1965. Kinds and abundances of fishes in the California current region based on egg and larval surveys. Calif. Coop. Ocean. Fish. Invest. 10:31-52. Allen,K.Eadway. 1941. Studies on the biology of the life stages of the salmon (Salmo salar). J. Anim. 10 (1) ;47-76. Alverson,D.L. 1960. A study of annual and seasonal bathymetric catch patterns for commercially important groundfishes of the Pacific Northwest coast of North America. Pac. Mar. Fish. Comm. Bull. 4. 66 pp. Alverson,D.L., A.T.Pruter and L.L.Ronholt. 1964. A Study of Demersal Fish and Fisheries of the Northeast Pacific Ocean. H.R.MacMillan Lecture in Fisheries. Inst. Fish., Univ. British Columbia, Vancouver,B.C. 190 pp. Alverson,D.L. and A.D.Welander. 1952. Notes on the scorpaenid fishes of Washington and adjacent areas, with a key for their identification. Copeia 1952 (3): 138-143. Barsukov,V.V. 1964. Taxonomy of the fishes of the family Scorpaenidae. Trans. Fed. Sci. Res. Inst, of Mar. Fish, and Ocean. (VNIRO). Vol. 52. Pt. 3:263-266. early Ecol. 95 Baumann,P.C. and J.F. Kite hell. 1974. Diel patterns of distribution and feeding of bluegills in Lake Wingra, Wisconsin. Trans. Am. Fish. Soc. 103(2):255-280. Bryan,J.E. and P.A.Larkin. 1972. Food specialization by individual trout. J, Fish. Res. Bd. Can. 29 (11):1615-1624. Carlisle,J.G., C.H.Turner and E.E.Ebert. habitat in tthe marine environment, and Game. Fish Bull. 124. 93 pp. Carlson,H.R. and R.E.Haight 1972. Evidence for a home site and adult homing in §ebastes flavidus. J. Fish. Res. Bd. Can. 29(7):1011-1014. Carr,W.S. and C.A.Adams. 1973. Food habits of juvenile marine fishes occupying seagrass beds in the estuarine zones near Crystal River, Florida. Trans. Am. Fish. Soc. 102 (3):511-540. Chapman,D.G. 1954. The estimation of biological populations. Ann. Math. Stat. 25:1-15. Croze,Harvey. 1970. Searching image in carrion crows. Zeitschrift fur Tierpsychologie, Suppl. 5:86 pp. Davies,David H. 1958. The relationship between sportfishing 1964. Artificial Calif. Dept. Fish 96 in the kelp beds and the harvesting of kelp off the coast of California. University of California Institute of Marine Resources Ref. 58-4. 56 pp. Delacy, A.C, C.R.Hitz and R.L.Dryfoos. 1964. Maturation, gestation and birth cf rockfish (Sebastodes) from Washington and adjacent waters. Wash. Dept. Fish. Fish, Res. Pap. 2 (3):51-67. Delacy,A.C, B.S.Miller and S.F.Borton. 1972. Checklist of Puget Sound fishes. Univ. Wash., Sea Grant Publ. WSG72-3. 43 pp. Dell,M.B. 1968. A new fish tag and rapid, cartridge-fed applicator. Trans. Am. Fish. Soc. 97(1):57-59. Doe,L.A•£. 1952. Current and net transport in Loudon Channel, April 1950. J. Fish. Res. Bd. Can. 9(1);42-64. Dunn,J.B. and C.R.Hitz. 1969. Oceanic occurrence of Sebastodes melanges in the central N. pacific. J. Fish. Res. Bd. Can. 26 (11):3094-3097. Ebeling,A.W., R.Larson, W.Alevizon and F.A.DeWitt,Jr. 1971. Fishes of the Santa Barbara kelp forest. (Abstract) University Press, University of Southern California. 1 p. Follett,W.I., D.Gotshall and J.G.Smith. 1960. Northerly 97 occurrences of the scorpid fish, Medialuna californiensis (Steindachner) with meristic data, life history notes and discussion of the fisheries. Cal. Fish and Game 46(2)1165-175. Frame,D.W. 1974, Feeding habits of young winter flounder: prey availability and diversity. Trans. Am. Fish. Soc. 103 (2):261-269. Frey,H.w- 1971. California's Living marine Resources and Their Utilization. Calif. Dept. Fish and Game. 148 pp. Fritz,E.S. 1974. Total diet comparison in fishes using Spearman rank correlation coefficients. Copeia 74{1):210-214. Fujimura,T. and H.T.Kami, 1958. Studies on the effect of shelter on the standing crop of fishes. July 1,1957 to June 30,1958, U. S. Fish and Wild. Serv. Fish. Abstract 3 pp. Gerald,J.W. 1966. Food habits of the longnose dace, Rhinichthys cataractae. Copeia 1966 (3): 478-485. Girard,CF. 1857. Contributions to the ichthyology of the western coast of the United States from specimens in the museum of the Smithsonian Institution. Proc. Acad. Nat. Sci. Phila, (1856) 8:131-137. 98 Gotshall,D.W., J.G,Smith and A.Holbert. 1965. Food of the blue rockfish, Sebastodes mystinus. Cal. Fish and Game 51 (3) : 147-162. Hager,D.C, 1960. The interrelationships of logging, birds and timber regeneration in the Douglas fir region of northwestern California. Ecology 41(1):116-125. Heubach,William, R.J.Toth and A.M.HcCready. 1963. Food of the young of the year striped bass (Soecus saxatilis) in the Sacramento - San Joaguin river system. Cal, Fish and Game 49 (4) : 224-239. Hitz,C.B. 1965. Field identification of the northeastern Pacific rockfish (Sebastodes). 0. S. Fish and Wild. Serv. Circ. 203. 61 pp. Hobson,E.S. 1973. Diel feeding migrations in tropical reef fishes. Helgolander wiss. Meeresunters. 24:361-370. Holling,C.S. 1959. The components of predation as revealed by a study of small mammal predation on the European pine sawfly. Can. Ent. 91:293-320. Hourston,A.S. 1958. Population studies on juvenile herring in Barkley Sound,B.C. J. Fish. Res. Bd. Can. 15 (5):909-960. 99 1959. Effects of some aspects of environment on the distribution of juvenile herring in Barkley Sound,B.C. J. Fish. Res. Bd. Can. 16 (3) : 283-308. Hubbs,Carl L. and L.P.Schultz. 1933. Description of two new American species referable tc the rockfish genus Sebastodes, with notes on related species. Univ. wash. Publ. Biol. 2 (2): 15-44. Hunter,John R. and C.T.Mitchell. 1668. Field experiments on the attraction of pelagic fishes to floating objects. J. Cons. perm. int. Explor. Mer. 31(3):427-434. Hynes,H.B.N. 1950. The food of freshwater sticklebacks with a review of methods used in the study of the food of fishes. J. Anim. Ecol. 19:36-58. Ivlev,V.S. 1961. Experimental Ecology of the Feeding of Fishes. Translated by Douglas Scott. Yale Univ. Press. 302 pp. Jordan,D.S. and B.W.Evermann. 1898. The Fishes of Middle and North America. Bull. 0. S. Nat. Hus. 47 (2): 1-3313. Kami,H.T. 1962. Sea sled and SCUBA reconnaisance of inshore areas and studies of effects of artificial shelters on standing crops of fishes. U. S. Fish and Wild. Serv. Fish. Abstract. In: Bibliography on Artificial Reefs, 100 Coastal Plains Center for Marine Development Service. Publ. 72-2:40. Landenberger,D.E. 1968. Studies in selective feeding in the Pacific starfish, Pisaster, in S. California. Ecology 49:1062-1075. LeBrasseur,Robin. 1970. Larval fish species collected in zooplankton samples from the NE Pacific ocean. Fish. Res. Bd. Can. Tech. Rpt. 175. 47 pp. Limbaugh,Conrad. 1955. Fish life in the kelp beds and the effect of kelp harvesting. University of California Institute of Marine Resources IMR Ref. 55-9. 158 pp. 1961. Life history and ecological notes on the black croaker. Cal. Fish and Game 47 (2): 163-174. 1962. Life history and ecological notes on the tnbenose, Aulorhynchus flayidus, a hemibranch fish of western North America. Copeia 1962(3):549-555. MacArthur,R.H. and J.W.MacArthur. 1961. On bird species diversity. Ecology 42 (3):594-598. Martin,N.D. 1960. An analysis of bird populations in relation to forest successsion in Algonguin Provincial Park,Ontario. Ecology 41(1):126-140. 101 Miller,D.J. and R.N.Lea. 1972. Guide to the coastal marine fishes of California. Calif. Dept. Fish and Game. Fish Bull. 157:1-235. Murdoch,William W. 1969. Switching in general predators: experiments on predator specificity and stability of prey populations. Ecol. Mono. 39:335-354. Niska,Edwin. 1973. MS. Species composition of rockfishes in the catches by Oregon trawlers 1963-1971. Appendix Tables 1-27. Prepared for the 14th Ann. Meeting of Int. Groundfish Conf. Fish. Comm. of Oregon. 35 pp. North,W.J. 1972. (ed.) The Biology of Giant Kelp Beds (!• integrifolia) in California. J Kramer. Lehre,Germany 1971. 600 pp. 0'Connell,C.P. 1953. Life history of the cabezon. Calif. Dept. Fish and Game. Fish Bull. 93:1-76, Ogawa,Y. 1966. Experiments on the attractiveness of artificial reefs for marine fishes. Iv. Attraction of stone bream and rudder fish to black objects on a white background. Bull. Tokai Reg. Fish. Res. Lab. 45:137-145. (English summary) . 1967. Experiments on the attractiveness of artificial reefs for marine fishes. VII. Attracticn of 102 fishes to the various sizes of model reefs. Bull. Jap, Soc. Sci. Fish. 33:801-811. (English summary). 1968. Experiments on the attractiveness of artificial reefs for marine fishes. VIII. attraction of young yellowtail to the model fish reefs. Bull. Jap. Soc. Sci. Fish. 34:169-176. (English summary). and H.Aoyama 1966. Experiments on the attractiveness of artificial reefs for marine fishes. V. Attraction of fishes to imitative sea-weed. Bull. Jap. Soc. Sci. Fish. 147-144. (English summary). • and Y. Ononda. 1966. Experiments on the attractiveness of artificial reefs to marine fishes. VI. Attraction of common sea bass to model reefs. Bull. Jap. Soc. Sci. Fish. 155-163. (English summary). Paine,R.T. 1971. The measurement and application of the calorie to ecological problems. Ann. Rev. Ecol. System. Vol. 2:145-164. Pavlov,D.S., Yu.N.Sbikin, A.Ye.Vaschinnikov and A.D.Mochek. 1972. The effect of light intensity and temperature on the current velocities critical to fish. Voprosy Ikhtiologii 12(4):703-711. Phillips,J.R. 1957. A review of the rockfishes of California 103 (family Scorpaenidae). Calif. Dept. Fish and Game.Fish Bull. 104:1-158. Pianka, E. B. 1974. Evolutionary. EcojLoajr. Harper and Bow. N.Y. 356 pp. Quast,J.C. 1968a. Some physical aspects of the inshore environment, particularly as it affects kelp bed fishes. In: Utilization of kelp bed resources in southern California. Calif. Dept. Fish and Game. Fish Eull. 139:25-34. 1968b. Fish fauna of the rocky inshore zone. Calif. Dept. Fish and Game. Fish Bull. 139:35-56 1968c. Estimations of the populations of the standing crops of fishes. Calif. Dept. Fish and Game. Fish Bull. 139:57-80. :57-80. 1968d. Observations on the food and biology of the kelp bass Paralabrax clathratus with notes on its sportfishery at San Diego, California. Calif. Dept. Fish and Game. Fish Bull. 139:81-108. 196 8e. fishes., Calif. 139:109-142. Observations on the food of kelp bed Dept. Fish and Game. Fish Bull. 104 1968£. The effects of kelp harvesting on the fishes of the kelp beds. Calif. Dept. Fish and Game. Fish Bull. 139:143-150. Roedel,P.M. 1953. Common ocean fishes of the California coast. Calif. Dept. Fish and Game. Fish Bull. 91:1-84. Scagel,R,F, 1959. The role of plants in relation to animals in the marine environment. In: Marine Biology. Pratt,I. and J.E.McCauley (eds.) Twentieth Annual Biology Collogium. Corvallis,Oregon. 1959:9-29. Schultz,L.P. 1958. Keys to the fishes of Washington, Oregon and closely adjoining regions. Univ. fash. Publ. Eiol. 2 (4): 103-128 (5th printing). Senta,Tetsushi. 1965. Importance of drifting seaweeds in the ecology of fishes. Jap. Fish. Resour, Ccnserv. Assoc. Fish. Res. Ser. No. 13. 55 pp. S.kogsberg,Tage. 1952. The fishes of the family Sciaenidae (Croakers) of California. Calif. Dept. Fish and Game.Fish Bull. 54:1-62. Smith,Fredrick,E. 1954. Quantitative aspects of population growth. In: Dynamics of Growth Processes. E.Boell (ed.) Princeton Univ. Press 304 pp. 105 Smith,Robert,!. 1966. Ecology and Field Biology. Harper and Row,N.Y. 686 pp. Tester,A. L. 1932. Food of the small-mouthed black bass in some Ontario waters. Univ. Toronto Publ. Biol. 46. 38 pp. Tinbergen,Leo. 1960. The natural control of insects in pine woods, i. Factors influencing intensity of predation by songbirds. Arch, neerl. Zool. 13:225-243. Turner,Charles H. and E.E.Ebert. 1962. The nesting of Chromis punctiEinnis (Cooper) and a description of their eggs and larvae. Cal. Fish and Game 48 (4):243-248. Turner, CH., E.E.Ebert and R. R. Given. 1969. Man-made reef ecology. Calif. Dept. Fish and Game. Fish Bull. 146. 221 pp. Baldron,K.D. 1972. Fish larvae collected from the northeastern Pacific Ocean and Puget Sound during April and May,1967. Nat. Ocean. Atmos. Admin., Nat. Mar. Fish. Serv., Spec. Sci. Rpt. Fish. 663:1-16. Hales,J.H. 1952. Life history of the blue rockfish, Sebastes listinus. Calif. Dept. Fish and Game. Fish Bull. 37:1-22. 106 Westrheim,S.J., W.S. Hailing and D.Davenport. 1968. MS. Preliminary report on maturity, spawning season and larval identification of Rockfish (Sebastodes) collected off British Columbia in 1967, Fish. Res. Bd. Can. Manuscript Report 9519 23 pp. Williams,George C. and Doris C.Williams. 1955. Observations on the feeding of the opaleye, Girella nig.rica.ns. California Fish and Game 41 (3):203-208. Wood,Langley. 1968. Physiological and ecological aspects of prey selection by the marine gastropod Urosalpinx cinerea, (Prosobranchia: Muricidae). Malacologia 6(3):267-320. 107 i££§ndix I Fullness of i to pods ana pin pcx^S. Crabs t.ru4)a4Can remain^ ] rl'^Vi (.C^r>£-i for i~ ? is V> ( U r\ i r.{ i ? i cc*.) Mores «^cA Commen Diet analysis sunaary sheet 1 Appendix II 108 Example of an observational record Experimental aquarium 


Citation Scheme:


Usage Statistics

Country Views Downloads
China 14 29
United States 5 2
Japan 4 0
Canada 2 2
France 1 0
City Views Downloads
Shenzhen 10 29
Tokyo 4 0
Beijing 4 0
Ashburn 3 0
Victoria 2 1
Mountain View 1 0
Unknown 1 32
Juneau 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items