UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Some factors affecting rainbow trout Ginetz, R. M. J. 1972

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1972_A6_7 G55.pdf [ 7.87MB ]
Metadata
JSON: 831-1.0101880.json
JSON-LD: 831-1.0101880-ld.json
RDF/XML (Pretty): 831-1.0101880-rdf.xml
RDF/JSON: 831-1.0101880-rdf.json
Turtle: 831-1.0101880-turtle.txt
N-Triples: 831-1.0101880-rdf-ntriples.txt
Original Record: 831-1.0101880-source.json
Full Text
831-1.0101880-fulltext.txt
Citation
831-1.0101880.ris

Full Text

SOME FACTORS A F F E C T I N G RAINBOW TROUT (Salmo gairdneri) PREDATION ON MIGRANT S O C K E Y E SALMON (Oncorhynchus nerka) FRY ,_ by R. M . J . GINETZ B.Sc , University of British Columbia, 1970 A THESIS SUBMITTED IN PARTIAL F U L F I L M E N T OF T H E REQUIREMENTS FOR T H E D E G R E E OF MASTER OF SCIENCE in the Department of Zoology We accept this thesis as conforming to the required standard T H E UNIVERSITY OF BRITISH COLUMBIA Apr i l 1972 In presenting t h i s thesis i n p a r t i a l fulfilment of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t freely available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Ronald M. J . Ginetz Department of Zoology The University of B r i t i s h Columbia Vancouver 8, Canada Date 13 April 1972 i i A B S T R A C T Various aspects of rainbow trout (Salmo gairdneri) predation on migrating sockeye salmon (Oncorhynchus nerka) fry and rainbow trout eggs were studied during 1970-71 in specially-constructed artificial streams, and in the laboratory. Tests involving sockeye fry as prey were conducted near Babine Lake, British Columbia, while those using rainbow trout eggs were done near Abbotsford, British Columbia. Examination of the effects of physical factors such as water velocity, water turbidity, and light intensity on predation on migrating sockeye fry showed mortality to be inversely related to water velocity and water turbidity; inversely related to light intensity at low light levels but directly related at very low levels. Other stream tests showed mortality to be inversely related to the amount of exposure of predators to fry, before the beginning of a nightly fry migration. Exposing predators to abnormal light for varying periods of time, immediately prior to fry migration, reduced over-all mortality during fry migration. Mortality was not proportional to the length of exposure of predators to abnormal light. F r y experience with predators was shown to increase the ability of fry to escape or avoid predation on subsequent predator encounters. Additional experience served to further increase their ability to escape or avoid predators. Conclusions drawn from stream tests and a behavioral study are that experienced fry migrate in a manner rendering them less susceptible to predation and the migration pattern (compact and in mass) is influenced in part, or completely, by encounter and escape from predator-prey interactions experienced earlier. "Handling" or other fright-evoking stimuli appeared to have similar effects. Rainbow trout feeding on colored fish eggs indicated color iii preference patterns which are influenced by background coloration (color contrast between food and background), and light intensity. Pre-ference was for colors showing the most contrast with the background at a particular light intensity. At low light levels, on a pale-blue background, preference was for lighter colors, while it was for darker colors at high light levels. Mortality differences increased proportionately with contrast between colors. Finally, trout displayed what appears to be a behavioral preference for red, and possibly blue, regardless of surrounding environ-mental conditions. In a food deprivation study rainbow trout displayed an S-shaped hunger response curve when fed on eggs. Indications were that rainbow trout will feed to gut capacity when given the opportunity. Finally, beyond an upper limit of food deprivation, the amount of food eaten by an individual remains fairly constant. iv T A B L E OF CONTENTS Page T I T L E P A G E i A B S T R A C T ii T A B L E OF CONTENTS iv LIST OF FIGURES vi LIST OF T A B L E S ix LIST OF APPENDICES x A C K N O W L E D G E M E N T S xii INTRODUCTION 1 SECTION I. PHYSICAL FACTORS AND THEIR E F F E C T S ON RAINBOW TROUT PREDATION ON MIGRANT S O C K E Y E S A L M O N FRY 3 Introduction. 3 Materials and General Methods 3 Field Data 3 Statistics 14 Methods and Results .'. 14 Mortality Associated with Light Intensity, Water Turbidity, Water Velocity, and Fry Development 14 Mortality of Fully-Developed River and Channel Fry at Low Light Intensities 23 Relationship Between Exposure Time of Predators to.Sockeye Fry Before Migration Begins, and F r y Mortality 25 Discussion 26 SECTION II. EXPERIMENTS WITH FOREIGN OBJECTS AND LIGHT ACCLIMATIZATION 29 Introduction. 29 Methods and Results 29 Predator Light Acclimatization Prior to Fry Migration and its Effect on Predation of Fry 29 V Page F r y Mortality from Predation in the Absence and Presence of White Styrofoam Strips 32 •3-2 Discussion  SECTION III. FRY E X P E R I E N C E WITH PREDATORS AND ITS E F F E C T ON RAINBOW TROUT PREDATION ON MIGRANT S O C K E Y E SALMON FRY 34 Introduction. 34 Methods and Results 34 Survival of Naive Fry and Fry with an Experience of Predation 24 Hours Previously 34 Survival of Naive Fry and Fry with an Experience of Predation 48 Hours Previously 39 Survival of Successive Groups of Naive Fry and the Same Group of Experienced F r y 40 Fry Migration Patterns 41 Behavioral Study of Experimental F r y 44 Survival of Enumerated and Non-enumerated Fry 52 Discussion 53 SECTION IV. RAINBOW TROUT COLOR P R E F E R E N C E AND HUNGER RESPONSE TO FOOD DEPRIVATION. 55 Introduction. 55 Materials and Methods 56 Field Data 56 Color Preference 60 Relationship Between Background Coloration and Color Preference &2 Relationship Between Light Intensity and Color Preference 5^ Effects of Time of Food Deprivation on Hunger of Rainbow Trout 6 8 Discussion 70 G E N E R A L DISCUSSION 73 R E F E R E N C E S 7 5 APPENDICES 76 vi LIST OF FIGURES Figure Page 1 Map showing approximate location of studies conducted at Fulton River, near Babine Lake, B . C 4 2 Map showing layout at Fulton River and adjacent spawning channels 5 3 Artificial stream apparatus used in the spring of 1970. Top: looking downstream; Bottom: looking upstream 6 4 Artificial stream apparatus used in the spring of 1971. Top: overall view of Channel No. 2 with apparatus in foreground; Bottom: close-up of apparatus, looking downstream 7 5 Experimental stream. Top: predator-prey interaction area. Bottom: temporary partition separating prey acclimatization area from interaction area 9 6 Artificial stream prey traps. Top: closed sock-like trap. Bottom: open sock-like trap with captured prey 10 7 • Converging throat-traps of Spawning Channel No. 2. Top: trap in fishing position. Bottom: looking directly down into trap where fry are captured 11 8 Upper fan-traps of Channel No. 1, in fishing position 12 9 Floating inclined plane-trap located below Channel No. 2 outlet 12 10 Development stages of sockeye salmon alevins (fry) 16 11 Mortality differences between fully-developed (Stage #5) and under-developed (Stage #3) fry in relation to fry density 17 12 Mortality differences between Fulton River and Spawning Channel No. 1 fry 17 13 Artificial streams equipped with lights and black poly-propylene for experiments requiring controlled lighting. Top: looking upstream. Bottom: looking downstream 19 vii Figure Page 14 F r y migration patterns in streams having clear water, simulated moonlit sky conditions, and water velocity of 12+2 cm/sec 21 15 F r y mortality differences in relation to light intensity, water turbidity, fry development and water velocity. Top: velocity of 12 _+ 2 cm/sec. Bottom: velocity of 21 + 2 cm/ sec 22 16 Mortality differences between Fulton River and Channel No. 1 fry at low levels of light intensity (0.05, 0.10, 0.20, and 0.30 ft-c) 24 17 F r y mortality differences in relation to exposure time of predators to sockeye fry before migration begins. Each light level represents the beginning of an exposure period; the 0.05 ft-c level representing the shortest period and 25.0 ft-c the longest period 27 18 Relationship showing the effect that varying periods of predator light acclimatization prior to fry migration has on fry survival during migration (Data, Appendix IV) 31 19 Mortality differences between fry migrating in streams with and without white styrofoam strips 31 20 Mortality differences between Fulton River and Channel fry in relation to fry density 36 21 Survival differences between: (a) naive, and experienced fry having 24 hours inexperience; (b) naive and experienced fry having 48 hours inexperience; (c) successive groups of naive fry, and the same group of experienced fry; (d) enumerated and non-enumerated fry (Data, Appendix VIII).. 38 22 Migration patterns of naive and experienced fry through streams containing predators, and of naive fry through streams without predators 42 23 Fibreglass tub into which captured fry are washed from converging throat-traps 45 24 F r y on wet-surfaced counting table 45 viii Figure Page 25 Artificial stream apparatus suspended in Marshall Creek near Abbotsford, B . C . Top left: looking downstream. Top right: looking upstream. Bottom: two streams enclosed in black polypropylene for an experiment requiring controlled lighting 57 26 Artificial stream prey traps. Top: prey trap consisting of fine mesh netting attached to coarse screen. Bottom: trap in position for capturing prey 58 27 Relationship showing the extent of rainbow trout color preference by (a) the number of selections made by predators for a particular color when that color formed part of a two-color combination of eggs; (b) the number of eggs eaten when a single color was presented alone (Data, Appendix X) 61 28 Experimental prey. Top: black, yellow, red and blue eggs. Bottom: two-color combinations shown by columns, from left to right — red-black, blue-red, red-yellow, yellow-blue, black-yellow, and blue-black 63 29 Relationship showing the effect of background coloration on rainbow trout color preference by (a) the number of selections by predators for a particular color when that color formed part of a two-color combination of eggs; (b) the number of eggs eaten when a single color was presented alone (Data, Appendix XI) 64 30 Relationship showing the effect of light intensity on rainbow trout color preference when colored eggs were presented in two-color combinations of (a) yellow and blue; (b) red and blue; (c) blue and black; (d) red and black; (e) yellow and black; and when presented as single colors (red, blue, black and yellow) (Data, Appendix XII) 67 31 Relationship showing the effect of food deprivation on hunger of rainbow trout. Data (Appendix XIII) fitted to von Bertalanffy equation 69 ix LIST OF T A B L E S Table Page I Behavior of naive and experienced fry prior to initial stimulus, and after application of stimulus 49 II Behavior of enumerated and non-enumerated fry prior to initial stimulus, and after application of each stimus 50 X LIST OF APPENDICES Appendix Page I Results of three-way analysis of variance (a) on the main effects of treatments, velocity and time on fry mortality; and (b) between the various treatment effects on fry mortality. Analyses are two-tailed tests of means 76 II Results of three-way analysis of variance with fry type, light intensity, and time as main effects for comparing mortality differences between fully-developed Fulton River and Spawning Channel No. 1 fry at the different light intensities. Analysis is two-tailed test of means 77 III Results of two-way analysis of variance with light intensity and time as main effects for comparing fry mortality at the different levels of light intensity. Analysis is two-tailed test of means 78 IV Results of (a) one-way analysis of variance, with unequal replication, on the effect of the period of predator light acclimatization on fry survival, and (b) Duncan's new multiple range test of comparisons between treatment means. Analysisis one-tailed test of means 79 V Experimental results showing fry survival differences in relation to periods of predator light acclimatization. 80 VI Results of two-way analysis of variance on the main effects of the presence and absence of styrofoam strips, and time, for comparing mortality differences between fry in the absence and presence of styrofoam strips. Analysis is one-tailed test of means 81 VII Results of two-way analysis of variances with fry type and time as major effects for comparing differences in survival between (a).naive and experienced fry having 24 hours inexperience; (b) naive and experienced fry having 48 hours inexperience; (c) successive groups of naive fry and a group of experienced fry; (d) enum-erated and non-enumerated fry. A l l analyses are one-tailed test of means 82 Appendix Page VIII Experimental results showing survival differences between (a) naive, and experienced fry having 24 hours inexperience; (b) naive, and experienced fry having 48 hours inexperience; (c) successive groups of naive fry, and the same group of experienced fry; and (d) enumerated and non-enumerated fry IX Results of three-way analysis of variance with (a) fry type, time and days as main effects for comparing migration differences between naive and experienced fry; and (b) stream type, time and days as main effects for comparing migration differences for naive fry in streams with and without predators. A l l analyses are one-tailed tests of means 85 X Experimental results showing mortality differences between red, orange, yellow, green, blue, brown and black eggs 86 XI Experimental results showing mortality differences between red, yellow, blue and black eggs on red, yellow, blue and black backgrounds 87 XII Experimental results showing mortality differences between red, yellow, blue and black eggs at four light intensities (0.1, 0.4, 0.8 and 1.2 ft-c) 88 XIII Experimental results showing the effect of food deprivation on hunger of rainbow trout 89 xii ACKNOWLEDGEMENTS The author would like to express his gratitude to Dr. P. A . Larkin, who supported the presentation of this thesis, guided the analysis, and made many helpful suggestions. Thanks are due to Dr. C. S. Holling and Dr . N. J . Wilimovsky, who reviewed the manuscript. I am indebted to the Fisheries Service Branch of the Department of the Environment, and to the British Columbia Fish and Wildlife Branch, for their assistance, without which the problem could not have been under-taken. Sincere thanks are extended to all those who assisted in carrying out many of the experiments; and to my wife, Sharon, who gave encourage-ment and assistance throughout the study. 1 INTRODUCTION Predation on migrant salmon (genus Oncorhynchus) fry could seriously affect adult salmon population densities, especially if fry popu-lations are small. Predation is probably the most important cause of mortality during the relatively short period of migration. Hunter (1959) estimated mortality from fish predation on pink (O. gorbuscha) and chum (O. keta) fry emigrating from Hooknose Creek (North Central British Columbia) as between 23 per cent and 86 per cent for a ten-year period. Neave (1953) developed the concept of depensatory mortality (greater proportional losses at lower densities) as the reason for failure of an increase of pink salmon populations when they had been reduced to low levels of abundance. Both Ricker (1950) and Larkin (1971) have ex-plored theoretically the effects of depensatory mortality, caused by pre-dation, on salmon population fluctuations. There is certainly adequate circumstantial evidence of the significance of predation, but rather little field evidence of the machineries of predation. In a preliminary study (B.Sc. Honors thesis, 1970) the author confirmed in experimental situations that rainbow trout (Salmo gairdneri) predation on migrant sockeye (O. nerka) fry was inversely related to density and fry size. Additionally, rainbow trout were found to be capable of selective feeding; i.e., they have the ability to develop an association between external appearance of a food item and reward. This study, also dealing with predation on migrant sockeye fry, consisted primarily of artificial stream tests. Various aspects of pre-dation were studied using sockeye fry and dead rainbow trout eggs as prey and rainbow trout as predators. Studies involving fry were done at Babine Lake, British Columbia, while those with eggs were at Abbotsford, British Columbia, in the interim period between annual fry migrations. 2 Aspects of predation examined through the course of the study are briefly outlined below. The effects of physical factors such as light intensity, water turbidity and water velocity, and fry experience, were examined in relation to predation on sockeye fry. Experiments with foreign objects and predator light.acclimatization were conducted to show how predation might be reduced without implementing management schemes involving predator removal or stream arki lake poisoning. A l -though several factors influence a predator's choice of food, time com-mitments permitted examination of only one — color preference. Color preference of rainbow trout was examined in relation to light intensity and background coloration. Finally, a preliminary study was conducted on the hunger response of rainbow trout to obtain some idea of feeding and digestion characteristics. 3 SECTION I PHYSICAL FACTORS AND THEIR E F F E C T S ON RAINBOW TROUT PREDATION ON MIGRANT S O C K E Y E SALMON FRY Introduction Most natural sockeye salmon streams are subject to extreme environmental variation during spring runoff. This environmental variation involves changes in physical factors, some of which include water velocity, water turbidity, and light intensity. Since the stream environment is variable at the time of fry migration, it appeared obvious to examine physical factors in relation to their effects on rainbow trout predation on migrant sockeye fry. This section of the report deals with that examination. Materials and General Methods Field Data The following information applies to most of the studies conducted during the periods Apr i l to mid-June, 1970 and 1971, at Fulton River near Babine Lake, British Columbia (Figures 1 and 2). Additional information, relative to a particular study, is presented in the "Methods and Results" section of that study. A l l studies, except one in the laboratory, were conducted in a series of identical artificial streams (Figures 3 and 4) suspended, in parallel, in the new Fisheries Service (Department of the Environment) artificial spawning channel located near Babine Lake. Streams were trough-like structures, constructed of 1.6 cm plywood, with dimensions 7.3 m x 0.6 m x 0.5 m. To simulate natural stream bed conditions, bottoms were sparsely covered with gravel of size range 2.0 cm - 10.0 cm. Water flowed through the streams at a depth of 30 cm and the velocity for 4 Figure 1. Map showing approximate location of studies conducted at Fulton i iver, near Babine Lake, B. C. Figure 2. Map showing layout of Fulton River and adjacent spawning channels. 6 Figure 3. Artificial stream apparatus used in the spring of 1970. Top: looking downstream; Bottom: looking upstream. I I Figure 4. Artificial stream apparatus used in the spring of 1971. Top; overall view of Channel No. 2 with apparatus in foreground; Bottom: close-up of apparatus, looking downstream. 8 all tests except one was 12 + 2 cm/sec. Velocity was controlled by weir-like valves attached to the stream inlets. Each artificial stream consisted of a predator-prey interaction area (Figure 5) occupying the first 6.1 m from the inlet, and a prey-trapping area occupying the last 1.2 m of a stream. The interaction area was constructed in a manner enabling the upper 1.0 m to be a temporary prey acclimatization area. Tops had 2.5 cm x 3.8 cm Gauge-14 screening; 2 inlets had 1 mm soft metal window screening; the downstream end of the interaction area had 1 cm^ Gauge-l6 screening. Prey traps (Figure 6) consisted of a smaller, open or closed sock-like trap fitted by a steel coupling (3.8 cm dia) to a larger fike-trap of dimensions 0.9 m x 0.6 m x 0.4 m. Predators (Salmo gairdneri) of fork length size range 25 cm-35 cm, were captured from Fulton River by beach-seining and angling. Attempts were made to equalize size distribution in all streams. Prior to preliminary experimentation, predators were acclimatized to artificial stream conditions for 96 hours. Predators contracting fungus infections during experimen-tation were replaced with new fish taken from a reserve stock held in a holding pen located in Fulton River. Depending on the experiment, prey (O. nerka fry) were captured and taken from the following locations — the converging throat-traps of Channel No. 2 (Figure 7), the upper and lower fan-traps of Channel No. 1 (Figure 8), and a floating inclined plane-trap (Figure 9) located approxi-mately 5 m downstream from the Channel No. 2 outlet. Samples of about 5000 fry were taken daily, between 2300 and 0100 hours (peak migration period), and held in the laboratory in 225 litre polyethylene containers of water with an airstone, until the next morning. Between 0900 and 1400 hours, test fry were selected out of the sample of 5000 fry and then held, undisturbed, in 2.5 litre polyethylene containers with an airstone 9 Figure 5. Experimental stream. Top: predator-prey interaction area. Bottom: temporary partition separating prey acclimatization area from interaction area. Figure 6. Artificial stream prey traps. Top: closed sock-like trap. Bottom: open sock-like trap with captured prey. 11 Figure 7. Converging throat-traps of Spawning Channel No. 2. Top: trap in fishing position. Bottom: looking directly down into trap where fry are captured. 13 until about 2130 hours. (For experiments dealing with fry experience, only fry measuring 29 mm in length were used. To allow for easy measurement, fry were anaesthetized in a solution of 200 ml of water and 1 ml of 2-phenoxyethanol.) Test fry were then transported to the stream for testing. Shortly thereafter (about 2225 hours) fry were introduced into the stream acclimatization areas, allowed to acclimatize to stream condi-tions for 15 minutes, and then released to migrate down the streams. (For experiments dealing with fry experience, acclimatization occurred during the period when natural light intensity fell from 1.0 to 0.2 ft.c.) After allowing fry to migrate down the streams, survivors were recovered from the traps at 0800 hours the following morning. On recovery, fry were transported to the laboratory, counted, and then depending on the nature of the experiment, were either released or retained for further tests. The method of obtaining numerical results of fry survival was to count all fry present, both in and upstream from the prey traps, on completion of a daily test (for clarity, daily tests are termed "trials"). The laboratory study was aimed at observing fry behavior in response to different external stimuli. The apparatus consisted of two 10 gallon aquaria with circulating water, each enclosed in separate compartments of a two-compartment chamber constructed of 1.27 cm plywood and black, 4-mil polypropylene. Three 7 w a t t General Electric incandescent light bulbs, connected to a Variac and suspended above each aquarium, were used for lighting. The entire chamber was covered with polypropylene to allow for controlled lighting and to prevent any outside movement by personnel from affecting prey behavior. An observation chamber, consisting of two sheets of polypropylene, with the innermost having a 7.5 cm x 25.0 cm slit in front of each aquarium, was attached to the experimental chamber in a manner enabling the observer to sit un-detected between both sheets and observe fry activity. 14 Light intensity measurements were required for some experiments. Below water surface readings were obtained with a 20 cm Secchi disc, while a Gossen foot candle meter (Trilux Model) fitted with a scotopic (human eye) correction filter was used for surface and above-surface readings. By controlling artificial illumination levels with Variacs, spectral character-istics of the light were altered considerably, reds predominating at the lowest intensities. Readings in foot-candles, and depth of light penetra-tion, were therefore very approximate. Statistics The statistical analysis applied to the experiments conducted at Babine Lake was the analysis of variance. A l l data were transformed to arc sin values prior to statistical analysis. Third level interactions and error discrepancies were pooled together in determining F values. A .05 significance level was applied to all tests. Prior to analyses of data on fry migrations, data for 2400, 0030, 0100, 0800 hours were pooled together to constitute one period. Analysis was then done on four time periods. Methods and Results Mortality Associated with Light Intensity, Water Turbidity, Water Velocity, and F r y Development Two preliminary tests were conducted to obtain data that would assist in the choice of fry density and source of fry for this experiment. Because the variable to be tested is the percentage of fry which die under various circumstances, it is statistically convenient if the number eaten is as large as possible, and if mortality approximates 50 per cent and the resulting distribution of proportions surviving is near normal. Treatment effects are then most readily determined. A randomized design was used to determine the density-mortality 15 relationship. Development stages #3 and #5 (Figure 10) Channel No. 1 fry, over the range of density 50 to 300, were allowed to migrate down eight streams each containing five rainbow trout predators. Four groups of each fry type were tested once a day over a six-day experimental period. Fry density for all streams on a given day was constant, however, it increased each day by increments of 50 with the density on the first day being 50 and on the sixth day, 300. Over the range of density from 50 to 200, approximately 20 per cent of stage #3 fry were eaten. Beyond 200 there was no further increase in number eaten, and the percentage removal is accordingly less (Figure 11). For stage #5 fry, mortalities were slightly less at all densities and the number eaten was asymptotic at about 20 fry. Since percentage mortality was about the same for the 50 to 200 density range, and because of daily time commitments, 45 fry was chosen as a suitable experimental density. A 5 x 5 randomized design was used to determine the effect of fry source and age on mortality. Stage #3 and #5 fry from Fulton River and Spawning Channel No. 1 were treated in the same manner as described above, but the density per trial was equal and constant at 150. Over five replicates channel fry experienced more similar mortalities than river fry (Figure 12). Mortality differences between the two stages of channel fry were about 10 per cent, but about 20 per cent for river fry. F r o m these results it seemed advantageous to use Spawning Channel No. 1 fry in the following experiment. With the information from the preliminary tests, effects of light intensity, water turbidity, water velocity, and fry development on fry mortality were examined under a 2 x 2 x 2 randomized factorial design. Treatment levels were: moonlit and cloudy night light intensities, clear and turbid water, high and low water velocities, and under and fully-developed fry. The design was such that each group of five predators was subjected to one treatment combination once under each of two water velocities, 12+2 cm/sec and 21 + 2 cm/sec. Eight experimental Figure 10. Development stages of sockeye salmon alevins (fry). 17 60 r O E N S I T Y Figure 11. Mortality differences between fully-developed (Stage #5) and under-developed (Stage #3) fry in relation to fry density. Figure 12. Mortality differences between Fulton River and Spawning Channel No. 1 fry. 18 streams, each enclosed in black, 4-mil polypropylene (Figure 13) and equipped with a row of 12 clear 7^-watt G . E . incandescent light bulbs in circuit with a Variac, were used for lighting. Lights were positioned 45 cm apart and 1.25 m above the water surface. Streams were enclosed to allow easy control of artificial light intensities and to prevent influence from natural light. The two levels of light intensity were obtained by duplicating, with artificial light, qualitative measurements of the natural light trans-mittance in the experimental streams, through 30 cm of water, during moonlit and cloudy night intensities. Secchi disc readings were taken with the eye positioned 30 cm above the water surface. At a depth of 30 cm the white portion of the disc was barely visible under cloudy night intensities but clearly visible under moonlit sky conditions. Turbid water was simulated by introducing a pre-mixed solution of organic dye (Bismarck Brown Y) and water into the streams. The original mixture, from which samples were taken, consisted of 50 grams of dye and 408 litres of water. The dye solution flowed into each of four streams, by gravity flow, from 22.5 litre metal containers equipped with airline tubing. The rate of flow through the tubing was about 0.4 litres per minute. Secchi disc readings were taken to measure turbidity, caused by the dye solution, for cloudy and moonlit sky conditions. For a cloudy night the white portions of the disc were visible up to a depth of 20 cm, while for moonlit sky conditions they were barely visible at the 30 cm depth. For a particular trial, development stage #3 and #5 fry were removed from the larger sample of 5000 fry and each placed in separate water-filled containers. Four groups of each fry type, each with a density of 45, were selected from these samples for testing. Attempts were made to equalize size distribution for each development stage by rejecting all fry appearing too large or small in comparison with the normal size range. Figure 13. Artificial streams equipped with lights and black poly-propylene for experiments requiring controlled lighting. Top: looking upstream. Bottom; looking downstream. 20 Prior to the beginning of a trial, test fry were acclimatized to experimental stream conditions for 15 minutes while predators were acclimatized for 60 minutes. Each trial was terminated at 0230 hours the following morning. Previously determined fry migration patterns (Figure 14), and personal observation, showed that under the experimental conditions migration was all but complete by 0230 hours. At that time lights were turned off and the stream valves opened to increase water velocity and force any fry outside the traps inside. Over a 16-day experimental period the effect of each treatment combination on fry mortality was tested eight times under each of the two water velocities. F r y mortality from predation was substantially influenced by the variables in this experiment (Figure 15). Mortalities were higher for stage #3 fry than for stage #5, higher for moonlit than cloudy night sky conditions and higher for clear than turbid conditions. A l l of these effects were statistically significant (Appendix I). There were no significant inter-actions, indicating that the various factors act additively. Contrasting Figures 14a and 14b, there are apparently higher mortalities at the lower velocity. The difference between low and high velocities was significant. Additionally, the interaction between velocities and treatments was signi-ficant. This seems to be related to the greater spread in the treatment differences at the lower velocity. At present no explanation can be given for this spread. Expectations were that treatment differences would be the same at both velocities. Thus, sockeye fry mortality from rainbow trout predation appears to be inversely related to turbidity, velocity and fry development, but directly related to light intensity at levels characteristic of moonlit and cloudy night sky conditions. The results have indicated that a reduction in light intensity had a greater effect on fry mortality than turbidity or fry development, and that turbidity had a greater effect than stage of fry development. Figure 14. F r y migration patterns in streams having clear water, simulated moonlit sky conditions, and water velocity of 12 + 2 cm/sec. 22 z u i t-< U l or LU m Z LEGEND X STAGE 5, MOONLIT, CLEAR o STAGE 3, MOONLIT, TURBID • STAGE 3, MOON LIT, CLEAR a STAGE3, CLOUDY, CLEAR J o STAGE5,MOONLIT, TURBID A STAGE 3, CLOUDY, TURBID » STAGE 5.CLOUDY,TURBID A STAGE 5,CLOUDY, CLEAR Figure 15. Fry mortality differences in relation to light intensity, water turbidity, fry development and water velocity. Top; velocity of 12 +_ 2 cm/sec. Bottom: velocity of 21+2 cm/sec. 23 Mortality of Fully-Developed River and Channel Fry at Low Light Intensities A 2 x 4 x 4 randomized factorial design was used to examine mortality differences between fully-developed Fulton River and Spawning Channel No. 1 fry under simulated dusk light intensities of 0.0 5, 0.10, 0.20 and 0.30 ft-c. The design was such that each group of five predators was subjected to each treatment combination once throughout the experiment. Experimental routine and the stream apparatus was the same as for the previous experiment, except that experimental fry density was increased to 50. Streams were divided into two groups of four. For the first four trials one group was used in testing river fry and the other for channel fry. This procedure was reversed for the fifth and sixth trials. The original intention was to conduct eight trials, but due to time commitments for further experiments, this experiment was terminated after the sixth trial . Thus, a total of six replicates was obtained for each of the four light levels and for each fry type. F r y mortality from predation was substantially influenced by the variables in this experiment (Figure 16). Mortalities were inversely related to light intensity and at a given light intensity mortalities were greater for channel than river fry. A l l of these effects were statistically significant (Appendix II). Additionally, the interaction between fry type and light intensity was significant, meaning that mortality differences between the two fry types were different from each other at the different light levels. Significance could have resulted from the large mortality difference between fry types that occurred at the 0.05 ft-c level. In com-parison with the results obtained from the previous experiment involving mortality under moonlit and cloudy night sky conditions, the present results are opposite to that which might be expected. Apparently mortality 24 Figure 16. Mortality differences between Fulton River and Channel No. 1 fry at low levels of light intensity (0.05, 0.10, 0.20 and 0.30 ft-c). 25 increases down to low light levels, but decreases at very low light levels. Finally, indications are that Fulton River fry are better able to avoid rainbow trout predation than are Spawning Channel No. 1 fry. Since river fry were slightly larger than channel fry at the time, the mortality difference is probably related to the size-mortality relationship mentioned earlier. Relationship Between Exposure Time of Predators to Sockeye F r y Before Migration Begins, and Fry Mortality A 5 x 5 randomized factorial design was used to examine fry mortality in relation to the exposure time of rainbow trout to sockeye fry before nightly fry migration begins. Since fry migration begins at low light intensities, the assumption was made that if experimental fry were introduced into the streams at high light intensities, they would remain at the stream inlet, exposed to the predators, until light intensities declined to a level at which they began migrating. From the assumption, experimental treatment levels (exposure periods) were set to begin at different light levels (<:0.05, 0.10, 1.60, 12.5, and 25.0 ft-c) and to end when fry migra-tion began (<0.0 5 ft-c). This experiment was conducted under natural lighting. The experimental routine was the same as for the previous experiments, except for the period of fry acclimatization to stream conditions and the time of fry release from each of the five streams used in the experiment. In testing the effects of a particular exposure period, a group of 50 Channel No. 1 fry was introduced into a stream acclimatization area approximately five minutes prior to the time when the light intensity fell to the level marking the beginning of that exposure period. When the ambient light intensity reached the appropriate level, fry were released from the acclimatization area and became exposed to the predators. Since five exposure periods were tested daily, and each began at a different time, fry release varied 26 accordingly. Additionally, some day-to-day variation occurred in the time at which a particular exposure period began; the decline in natural light intensity varied from day to day due to changes in cloud cover. After five experimental days, a total of five replicates was obtained for each exposure period. F r y mortality from predation was influenced by the length of time fry were exposed to the predators before downstream migration began (Figure 17). The results (Appendix III) indicated that fry mortality was significantly different among the different exposure periods, and that an inverse relationship existed. A large mortality difference occurred between the shortest exposure period and the four longer ones. A possible reason for this might be that the shorter exposure periods provide fry with a smaller time period to prepare themselves for escaping predators or detecting the predators before fry migration begins. As the length of exposure time increases, more time is available to the fry for detecting predators. In general, expectations are that fry emerging at very low light intensities would have a smaller chance of detecting and avoiding predators than those emerging at higher light intensities. Consequently, the former would experience heavier predation than the latter. Discussion The foregoing results indicate that physical factors can influence the rate of predation of rainbow trout on salmon fry. Accordingly, one would expect that rates of predation could vary from stream to stream, from place to place, and from time to time within a single stream. There is also the expectation that manipulation of physical conditions could be used to reduce predation. For example, one might illuminate a stream, for a number of days, to prevent fry from migrating and then on a cloudy night turn the lights off and allow the fry to migrate in a mass. Another method would be to flush a stream with a large volume of water every 27 30r R E P L I C A T E S Figure 17. Fry mortality differences in relation to exposure time of predators to sockeye fry before migration begins. Each light level represents the beginning of an exposure period; the 0.0 5 ft-c level representing the shortest period and 25.0 ft-c the longest period. 28 evening during the peak period of migration (which can be done on the Fulton River). A third method would be to make a stream more turbid than normal during nightly migrations. Combinations of any two or three methods might also produce the desired effect of reducing predation. The measurement of light intensity in the present experiments was inadequate and further more critical work should probably precede any large scale field trials aimed at reducing predation by changing light intensity. Turbidity in the present experiments was variable and difficult to control. Therefore, more critical work should be done before increasing turbidity is used as a method for reducing the predation effect. Before flushing is implemented, its possible effect on the whole stream environ-ment should be evaluated. Flushing could cause erosion and scouring in spawning beds. Premature fry present could suffer significant mortalities from mechanical injury and suffocation. In general, expectations are that manipulation of physical condi-tions could be used to reduce predation of salmon fry during the migratory stage. However, mortality in later life stages from other causes (parasites, disease, starvation, predation, etc.) could supersede, in terms of over-all salmon production, the predatory effect occurring during the fry migration stage. Nevertheless, information from the present experiments can be used to advantage if predation on migrant salmon fry is contributing signi-ficant losses to a salmon population. 2 9 SECTION II EXPERIMENTS WITH FOREIGN OBJECTS AND LIGHT ACCLIMATIZATION Introduction In the past, schemes such as predator removal and stream or lake poisoning have been thought of as ways of reducing predation on young salmon. It is felt that other less destructive methods could be implemented to achieve the same result. This study involves examining ways of reducing predator efficiency during fry migration, which, if feasible, could be used to reduce predation. One test involves subjecting predators to abnormal light immediately prior to a nightly fry migration, while another involves introducing small foreign objects in a stream with migrating fry. Methods and Results Predator Light Acclimatization Prior to Fry Migration and its Effect on Predation of Fry Expectations were that, if rainbow trout predators were subjected to abnormal light immediately prior to fry migration, a period of predator night blindness would occur when light levels were quickly reduced to a low level. Fry migrating during the period of predator night blindness should escape predation, resulting in a higher over-all survival rate than if pre-dators did not experience temporary night blindness. Also, expectations were that fry survival would be directly related to the length of time pre-dators were acclimatized to the high light levels. The two experimental streams (Figure 4) used in this experiment were equipped for controlled artifical lighting in a manner similar to that described in previous experiments. Differences in the apparatus were 30 that two sets of 12 lights were used for each stream, and the predator -prey interaction area was partitioned off, above the water surface, from the prey acclimatization area. Also, there were 18 lights above the interaction area and six above the acclimatization area. With a completely randomized design, fry survival differences were examined for each of the treatment levels (80, 0, 20, 10 and 40 min of predator light acclimatization, in that order) once in each stream over a five day experimental period. After repeating the procedure, a third attempt was made, but due to time commitments, the experiment termi-nated after testing the 80 and 0 min treatments. This incomplete replica-tion resulted in obtaining six replicates for the 80 and 0 min levels, and four for the 20, 10 and 40 min levels. Prior to the beginning of a trial, predators (seven per stream) were acclimatized to 6.0 ft-c of light for the period designated for testing in that particular trial. Prey (100 fully-developed Channel No. 1 fry per stream) were acclimatized to stream conditions under natural light. Acclimatization occurred during the period when natural light levels fell from 1.0 ft-c to 0.1 ft-c. To allow for natural light acclimatization, the polypropylene covering the fry acclimatization areas was removed. When natural light intensity reached 0.1 ft-c, the covers were replaced, artificial lighting was adjusted to 0.1 ft-c in the interaction and fry acclimatization areas, and fry were released to migrate downstream. For the 0 minute treatment level, predators were acclimatized to stream conditions under the 0.1 ft-c light intensity in which all trials occurred. In this trial acclimatization of predators was of the same duration as for prey. Fry survival was influenced by the variable in this experiment (Figure 18). The 10, 20, 40 and 80 min acclimatization periods had a significant effect (Appendix V) in increasing fry survival in comparison to survival when the predators did not experience abnormal light acclima-1 0 0 PERIOD OF LIGHT ACCLIMATIZATION BY PREDATORS (MINUTES) Figure 18. Relationship showing the effect that varying periods of predator light acclimatization prior to fry migration has on fry survival during migration (Data, Appendix IV). Figure 19. Mortality differences between fry migrating in streams with and without white styrofoam strips. 32 tization (0 minute treatment). When predators experienced the 6.0 ft-c light level, fry survival was about 20 per cent greater than when they did not. Since survival differences among the 10, 20, 40 and 80 min periods were not significant, indications are that beyond the 10 minute period fry survival will not increase regardless of the length of time predators are exposed to the 6.0 ft-c light intensity. Apparently fry survival is related to the period of predator light acclimatization, but only over the range of exposure up to about 20 minutes. Fry Mortality from Predation in the Absence and Presence of White Styrofoam Strips Expectations were that predation on migrating salmon fry could be reduced if predators became confused or frightened during the actual fry migration. This experiment involved examining the effect white styrofoam strips have on reducing predation of fry when the strips are introduced into a stream with migrating fry. Four experimental streams, each with five predators, were used in testing mortality differences in the presence and absence of strips. Strips were 30 cm in length and 0.3 cm in diameter. A total of 300 strips was used in each of two streams for every trial. Fry were from Channel No. 1 and density per stream was 50. With a randomized experimental design, two groups of fry with styrofoam strips, and then two groups without strips, were introduced into the fry acclimatization areas at 2225 hours on each of six days. After acclimatization, fry and strips were released. Fry recovery was in the usual manner. After six trials, a total of 12 replicates was obtained for each treatment. Fry mortality differences in this experiment (Figure 19) were not significant (Appendix VI). Mortalities in the presence and absence of 33 white styrofoam strips approximated 20 per cent. A possible reason for these unexpected results is that the strips did not function in the intended manner. Once released from the acclimatization areas with the fry, they became attached to the stream sides and would not drift downstream as fry migrated downstream. Attraction was probably due to strong surface tension, at the surface along the stream sides, or from electrostatic attraction between the styrofoam and the wooden stream sides. Because the strips did not function as intended, the possibility exists that sockeye fry migrating in the presence of unnatural artifacts, such as styrofoam strips, will experience lower predation than fry migrating without these artifacts. Further tests should be conducted to prove or disprove the idea. Discussion Some of the foregoing results indicate that less destructive methods than predator removal, or stream and lake poisoning, might be implemented to reduce predation on young migrant salmon. For example, if heavy mor-tality is occurring below artificial propagation facilities such as spawning channels or hatcheries, predation might be reduced by illuminating a stream, or portions of it, prior to normal nightly fry migration, and then switching the lights off at the time when normal peak migration occurs. If predators are actually temporarily blinded by the illumination change, pre-dation on migrating fry should be significantly less. Although the experi-ment with foreign objects was unsuccessful, expectations are that predation could be reduced if predators became confused or frightened during fry migration. Expectations are that ways other than those suggested exist in which predation can be reduced. Accordingly, one would expect the rates of predation to vary with the method and that success would not always be achieved. The techniques used in the present experiments were not the best and further work should precede large scale field trials to reduce predation by the methods described above. SECTION III FRY EXPERIENCE WITH PREDATORS AND ITS E F F E C T ON RAINBOW TROUT PREDATION ON MIGRANT SOCKEYE SALMON FRY Introduction Experience of both predators and prey is a relevant component influencing the outcome of predator-prey interactions. Experience of rainbow trout did not appear, from previous studies, to be an important factor to pursue as part of a short term research program. Apparently rainbow trout can switch from one food type to another very quickly, and do not show a progressively more efficient predation rate with time. Consequently, emphasis-was placed on examining fry experience with predators in relation to its effect on rainbow trout predation on migrant sockeye fry. Methods and Results Survival of Naive Fry and Fry with an Experience of Predation 24 Hours Previously "Experienced" fry are those having the experience of migrating down an artificial stream containing predators. "Naive" fry are those lacking that experience. It was assumed that fry taken from a spawning channel would not have previous experience with fish predators since predator populations in channels are minimal compared to natural streams. As was done previously, a preliminary experiment was conducted to assist in the choice of fry density and fry source for this and other experiments dealing with fry experience. One experiment, a randomized 5 x 5 factorial, was done to determine the fry density and fry source. Over the range of density 50 to 250, stage #3 Channel No. 1 and Fulton River fry, 35 and stage #5 from the same source plus Channel No; 2 were tested against predators in artificial stream tests. Prior to this experiment it was decided that only fully-developed fry would be used in the following experiment. Under-developed fry were included in the test for other reasons. Over the range of densities from 50 to 250, fully-developed Channel No. 1 fry experienced greater mortalities than fully-developed Channel No. 2 and Fulton River fry (Figure 20). Average mortality for Channel No. 1, Channel No. 2, and river fry. approximated 10 per cent, 6 per cent and 3 per cent respectively. Since Channel No. 1 fry suffered the greatest percentage mortality, they were chosen as experimental fry for the following experiments. Additionally, the density chosen was 75 because greatest mortality occurred in the 50 to 100 density range. To test for survival differences between experienced and naive fry, four to six experimental streams (Figure 4) were used and the pre-dator density varied from four to six. For all trials, two groups of experienced fry, and then depending on the number of streams used, a minimum of two groups of naive fry were introduced at random into the streams, allowed to acclimatize, and then released. Naive fry were from the channel migration of the previous night, while experienced fry were the survivors of naive fry tested one day previous. Originally, the experimental procedure was to test the two fry types every second day; however, on May 16 of the experimental period (Appendix Vlll-a) this was changed to daily tests to provide the time required to conduct further experiments. "With this procedural change, a total of 16 replicates was obtained over a 24-day experimental period. The variation in predator density and experimental streams used in this experiment resulted from attempts to increase fry mortality per trial. It was hoped that mortality would approximate 50 per cent, therefore 36 LEGEND 100 150 DENSITY 200 STAGE 3 A CHAN. NO. I o RIVER STAGE 5 A CHAN. NO-1 x CHAN. N0.2 • RIVER 250 Figure 20. Mortality differences between Fulton River and Channel fry in relation to fry density. 37 enabling easier determination of treatment effects. On May 6, the number of streams used was reduced from six to five and predator density was increased from four to five. On May 12, streams used were reduced to four and predator density increased to six. With the acquisition of more predators, stream number was increased to five on May 14, where it remained for the duration of the experiment. From May 18 to 23, five streams were used, but only the survivors from four were used in the analysis of data. During that period stream #2 contained only four preda-tors (two were removed because of fungus infection), while the remaining four each contained six. Since mortality did not increase with predator density, it was assumed that four predators would provide naive fry with the same experience as six predators. Survivors of naive fry from stream #2 provided the additional fry (experienced) needed in maintaining the regular 150 (75 fry for each of two streams) required per trial. Additionally, this enabled the experiment to proceed on a regular day-to-day basis. Survival differences existed between naive and experienced fry (Figure 21a). Over the experimental period mean difference in survival was 13.3 per cent, with experienced fry having 94.5 per cent survival and naive fry 81.2 per cent. Indications are that experience gained by a group of naive fry 24 hours previous was statistically significant (Appendix Vila) in increasing their survival relative to fry without experience. Experi-mental results show two reversals occurring, one on the third trial, another on the fifteenth. In general, the results suggest that once migrating sockeye fry encountered and escaped from a rainbow trout predator, the experience gained enhances ability to escape rainbow trout predation at least during the next 24 hours. 3 8 REPLICA TES Figure 21. Survival differences between: (a) naive, and experienced fry having 24 hours inexperience; (b) naive and experienced fry having 48 hours inexperience; (c) successive groups of naive fry, and the same group of experienced fry; (d) enumerated and non-enumerated fry (Data, Appendix VIII). 39 Survival of Naive Fry and Fry with an Experience of Predation 48 Hours Previously Six experimental streams (Figure 4), each with six predators, were used in this experiment. Naive fry were from the previous night's channel migration, while experienced fry were the survivors of naive fry tested two nights (48 hours) previously. During the holding period, experienced fry were left undisturbed in the laboratory, in ten-gallon aquaria which were under a constant light intensity of 5 ft-c. For each trial, which occurred every second day, two groups of experienced and then four groups of naive fry were introduced at random into the stream, acclimatized, and then released. A total of six replicates was obtained over a 14-day experimental period. Experienced fry had greater survival than naive fry in this experiment (Figure 21b). Over the experimental period, mean difference in survival was 11.8 per cent, with experienced fry having 86.8 per cent survival and naive fry 7 5.0 per cent. Indications are that the initial experience, even after 48 hours, was statistically significant (Appendix Vllb) in maintaining survival of experienced fry above that of naive fry. The results show a reversal for the third trial. A possible explanation is that in this particular trial, naive fry were not obtained by the standard procedure. Due to human error, a sample of naive fry was not obtained the previous evening and, to compensate for this, fry were taken, with dip nets, from the lower end of Channel No. 1 during daylight hours. The possibility exists that these fry already had experience with predators in the channel. Personal observation on subsequent days showed that large numbers of fry were always present at the site from which the experimental fry were taken. Past observation has revealed that many fry do not migrate from the channel unless forced to do so. If the fry tested on the third trial were residents with predator experience, chances are the trial was actually a test between groups of experienced fry. 40 Survival of Successive Groups of Naive Fry and the Same Group of Experienced Fry Two experimental streams (Figure 4), each containing six pre-dators, were used in testing for survival differences, with time, between a single group of experienced fry and successive groups of naive fry. For each trial, naive fry were taken from the channel migration of the previous night, while the survivors of a single group of experienced fry were used over the entire experimental period. Fry density, although equal in both streams for any one trial, decreased with every trial. The density for successive trials (85, 82, 68, 64, 55 and 50) was dependent upon the mortality incurred by the experienced group from the beginning of one trial to the next. Mortality occurred from predation, and from handling in recovery. The experimental procedure was identical to that described for the previous experiments. Al l trials were two days apart, the reason being that the last experiment was conducted at the same time and required all six experimental streams. Consequently, the present experiment was conducted on alternate days. Six replicates were obtained over a 12-day experimental period. The two fry types displayed survival differences in this experi-ment (Figure 21c). Mean difference in survival, after six trials, was 15.2 per cent, with experienced fry undergoing 97.7 per cent survival and naive fry 82.5 per cent. The effect of experience was statistically signifi-cant (Appendix Vile). Examination of the survival differences shows them to increase as the experienced fry gain more experience. This is parti-cularly evident when comparing the first with the last two trials. This suggests that repeated experience increases the ability of migrant sockeye fry to escape rainbow trout predation. 41 Fry Migration Patterns Attempts were made, by examining fry migration patterns, to evaluate why experienced fry were better able to survive in a predator situation. Migration patterns for the fry types, and for naive fry migrating streams without predators, were examined using four and two streams respectively. The pattern for experienced and naive fry migrating streams with predators was obtained during the actual experiment on survival differences involving fry with experience 24 hours previously. A separate test, to determine whether migration patterns are actually influenced by the presence of predators, was conducted by allowing naive fry to migrate in streams without predators and comparing the resultant pattern with that for streams with predators. In both experiments, fry introduction, acclimatization, and release were identical to that described for the preceding experiments. Fry recovery was conducted every 30 minutes after release from the acclimatization areas, for six consecutive periods a trial. Any fry not migrating into the traps were retrieved at 0800 hours the following morning. Experiments were conducted over a four-day period. Migration patterns (Figure 22) were almost the same for both naive and experienced fry migrating in the presence of predators. Approxi-mately 60 per cent of the fry migrated down the streams within the first hour. After this the numbers declined with time. Statistical analysis (Appendix IXa) showed no significant difference in the migration patterns between the fry types from day to day. The main effect of "time" was significant indicating that both fry types displayed a significant migration pattern, the number of fry migrating varying with time. There were no significant interactions, indicating that the various factors act additively. Although the "time x fry type" interaction was not significant, it is felt that the observed slight differences may be attributed to experienced fry 42 •LEgENP o NAIVE TIME Figure 22. Migration patterns of naive and experienced fry through streams containing predators, and of naive fry through streams without predators. 43 escaping predation by migrating somewhat more quickly. If one relates these migration differences to predation, schooling and individual differences observed in a behavioral study (discussed later) could explain the survival differences that occurred between the fry types. In the behavioral study experienced fry tended to form more compact schools than naive fry. Accordingly, a compact school could escape predation more effectively than a loose school of naive fry. Experimental results (Figure 22) show that migration patterns displayed by naive fry in the presence and absence of predators were significantly different (Appendix IXb). In streams with predators, the majority of fry migrated as a group within the first hour of each trial, while in streams without predators, migration was spread out over a longer time span. The "time x stream-type" interaction was significant indicating that the difference in proportion of fish migrating was not the same at the different times. These results plus the delayed peak occur-ring in streams with predators, suggest that predators invoke an awareness in fry which in turn influences compactness of schooling and timing of migration. However, results from the previous stream tests suggest that experience of predators is more effective than awareness of predators in increasing the ability of migrant sockeye fry to survive in a predator situation. Earlier, the slight differences in migration patterns between naive and experienced fry were attributed to experienced fry escaping predation and naive fry undergoing predation. Schooling and individual differences (observed in behavioral study) resulting from predator experience appear to be the mechanism by which experienced fry migrate past predators and maintain a higher survival rate than naive fry. Inherent in this is the assumption that experienced fry had actually gained experience through encounter and escape from predators. A question arises as to whether this assumption is valid. What would happen in a survival test 44 between naive fry, and naive fry having had the experience of migrating down a stream in the absence of predators? The factor responsible for experienced fry having higher survival could well have been the experience of migrating down an experimental stream. This experience could in-fluence schooling and individual differences which in turn could assist them in escaping predators. Handling of fry through the experimental procedure also could increase their awareness of danger, rendering them less susceptible to predation than naive fry. Legitimate arguments exist about the factor, or factors, responsible for the survival differences between naive and experienced fry in the stream tests. However, until further experiments are conducted, the belief is that the observed migra-tion pattern and behavior of experienced fry, render them less susceptible to predation, and this over-all behavior is influenced in part or completely by earlier encounter and escape from predators. Behavioral Study of Experimental Fry A behavioral study was conducted, in the laboratory, to determine: (1) the mechanism(s) by which experienced fry maintain a higher survival than naive fry; and (2) whether survival differences are dependent on pre-dator experience, or whether "fright" evoked by other means would bring the same result. The study involved examining group and individual responses by experienced and naive fry, and enumerated and non-enumerated fry, to different types of arbitrarily chosen external stimuli. Enumerated fry are subjected to a capture, count and release program conducted by the Fisheries Service to assess total annual fry production from the various spawning areas. In the enumeration process, migrant fry are captured in converging throat traps (Figure 7), washed into and drained out of a 227 litre fibreglass tub (Figure 23) into a 12 litre polyethylene bucket half filled with water. The bucket of water, with fry, is transferred to an enumeration room where fry are transferred to a 45 Figure 24. Fry on wet-surfaced counting table. 46 wet-surfaced counting table (Figure 24) and counted. As the fry are individually counted they are handswept into a trough of flowing water which extends back to the river. Experimental fry were collected in dipnets as they were swept from the counting table. Fry undergoing enumeration are subjected to a rapid and dramatic change in light inten-sity of from 0.1 ft-c in the channel, or river, to about 80 ft-c on the counting table. In the experimental chamber described earlier, the four fry types were subjected to two forms of light intensity fluctuation, and vibration. Experienced and naive fry were also observed in the presence of a single rainbow trout (fork length 26 cm) predator. To eliminate the possibility of density specific behavioral patterns occurring for enumerated and non-enumerated fry, observations were made at densities approximating 50, 200, 300 and 400. Fry were observed at each density once. Shortage of experienced fry restricted fry densities to 50. New samples of naive and experienced fry were observed each day of a two-day study period. Only two fry types were observed at any one time. Acclimatization to aquarium conditions, which occurred prior to actual observation, varied depending on the fry types undergoing observation. Enumerated fry collected from the counting table between 2300 and 0100 hours were placed directly into one aquarium of the darkened experimental chamber for acclimatization. At about the same time, non-enumerated fry, captured in the inclined plane trap below the Channel No. 2 outlet, were collected and transferred, in total darkness, to the second aquarium in the experimental chamber. Both groups remained in total darkness until 1100 hours the following morning, at which time the chamber lights were turned on, by Variac, to an intensity of 1.0 ft-c. During the next 30 minutes light intensity was raised to 8.0 ft-c in steps. At 1110 hours it 47 was raised to 4.0 ft-c; at 1120 to 7.0 ft-c; and at 1130 to 8.0 ft-c where it remained until application of the various external stimuli. Acclimatization of naive fry was the same as described above, but the collection procedure was different. Approximately 5000 fry were col-lected from Channel No. 1, between 2300 and 0100 hours, and transported to the laboratory in a covered 227 litre polyethylene container filled almost to capacity with water. At the laboratory a sample of about 50 fry was taken from the larger sample and placed into one aquarium of the darkened experimental chamber. Experienced fry, survivors of a stream test conducted during the previous night, were collected from the stream traps at 0800 hours, held in the laboratory until 10 50 hours, and then placed into the second aquarium of the darkened experimental chamber. Between 1100 and 1130 hours, with naive fry in one aquarium and experienced fry in the other, the chamber lights were raised to 8.0 ft-c. Thereafter, acclimatization occurred as described for enumerated and non-enumerated fry. Application of external stimuli, excluding the predators, and ob-servation of the resultant spatial distributions and behavior was conducted from 1830 to 1850 hours. After application of each stimulus, fry behavior was observed for five minutes. Application of various stimuli, in chronolo-gical order, were: (1) six consecutive fluctuations in light intensity from 8.0 ft-c to 0 ft-c to 8.0 ft-c; (2) tapping on each aquarium glass for 10 seconds; and (3) a rapid light intensity change to 0.1 ft-c, where it was maintained for five minutes and then returned to 8.0 ft-c. The effect of a predator stimulus on the behavior of naive and ex-perienced fry was examined in the following manner. At 1830 hours, spatial distributions and behavior were observed and recorded. Immediately thereafter two predators were introduced, one into each aquarium, and the resultant spatial distributions and behavior, observed. 48 Group and individual differences existed between naive and experienced fry in response to the external stimuli (Table I). A l l groups of experienced fry schooled prior to, and in response to, the stimuli while on several occasions naive fry did not. Relative to each other, experienced schools were always compact and almost always contained all fry, while naive schools were loose, except on one occasion when compact schools were formed in response to the predator. Naive schools generally contained only about half of the fry. Individual experienced fry remained motionless within a school in response to stimuli. Naive fry were more active in that they generally maintained some forward swimming movement. At times they appeared bewildered and swam in all directions, particularly in response to the vibration stimulus. The observed behavioral differences between naive and experienced fry might be the reason for survival differences in the stream tests. By migrating in compact schools and remaining motionless when near predators, experienced fry may migrate, undetected, past the predators. Enumerated and non-enumerated fry displayed behavioral differences comparable to those observed for experienced and naive fry (Table II). As for experienced fry, enumerated fry were always in loose schools prior to, and in compact schools after, application of the stimuli. As for naive fry, non-enumerated fry never schooled prior to a stimulus. However, schooling did occur in response to all stimuli, which was not observed for naive fry. No explanation can be given for this difference at this time. Individual behavior of enumerated and non-enumerated fry appeared similar to experienced and naive fry behavior, respectively. A comparison of Tables I and II presents a clearer picture of the simi-larities and differences observed between the fry types in response to the various stimuli. T A B L E I . B e h a v i o r o f n a i v e a n d e x p e r i e n c e d f r y p r i o r t o i n i t i a l s t i m u l u s , a n d a f t e r a p p l i c a t i o n o f s t i m u l u s . D A Y F r y T y p e D e n s i t y p e r A q u a r i u m P r i o r t o a p p l i c a t i o n o f I n i t i a l S t i m u l u s E X T E R N A L S T I M U L I R a p i d L i g h t F l u c t u a t i o n V i b r a t i o n S i n g l e L i g h t F l u c t u a t i o n P r e d a t o r s N a i v e 50 - s l o w s w i m m i n g a b o u t a q u a r i u m - n o s c h o o l - s l o w s w i m m i n g a b o u t a q u a r i u m - no s c h o o l - r a p i d m o v e m e n t i n t o c o r n e r s ; - s h o r t l y t h e r e a f t e r , s l o w s w i m m i n g a b o u t a q u a r i u m - n o s c h o o l - s l o w s w i m m i n g a b o u t a q u a r i u m - n o s c h o o l - r a p i d m o v e m e n t i n t o c o r n e r s o n i n t r o d u c t i o n o f p r e d a t o r ; - s h o r t l y t h e r e a f t e r , t w o m o t i o n l e s s c o m p a c t s c h o o l s , o n e o n e a c h s i d e of p r e d a t o r E x p e r i e r c e d 50 - s l o w s w i m m i n g w i t h i n l o o s e s c h o o l - m o t i o n l e s s w i t h i n c o m p a c t s c h o o l - m o t i o n l e s s w i t h i n c o m p a c t s c h o o l - h a l f m o t i o n l e s s i n c o m p a c t s c h o o - f e w m o t i o n l e s s i n c o r n e r s - r e m a i n d e r j o i n s c h o o l a f t e r l i g h t r a i s e d t o 8 .9 f t - c - r a p i d m o v e m e n t i n -t o c o r n e r s o n i n t r o -d u c t i o n o f p r e d a t o r - s h o r t l y a f t e r o n e m o t i o n l e s s c o m p a c t s c h o o l f o r m e d - f e w m o t i o n l e s s i n c o r n e r s o n s i d e o p p o s i t e s c h o o l a n d p r e d a t o r N a i v e 50 - s l o w s w i m m i n g a b o u t a q u a r i u m - n o s c h o o l - f r y s l o w s w i m -m i n g w i t h i n l o o s e s c h o o l ; a f e w f r y f r e e s w i m m i n g o u t s i d e s c h o o l . - r a p i d m o v e m e n t i n t o c o r n e r s ; - s h o r t l y t h e r e a f t e r a b o u t h a l f s l o w s w i m m i n g a b o u t a q u a r i u m - r e m a i n d e r i n l o o s e s c h o o l - h a l f f r e e s w i m -m i n g a b o u t a q u a r i u m - r e m a i n d e r i n l o o s e s c h o o l - r a p i d m o v e m e n t i n t o c o r n e r s o n i n t r o d u c t i o n o f p r e d a t o r - a b o u t 15 f r y f o r m l o o s e s c h o o l - e v e n t u a l l y a l l s l o w s w i m m i n g E x p e r i e r c e d 50 - s l o w s w i m m i n g w i t h i n l o o s e s c h o o l - m o t i o n l e s s w i t h i n c o m p a c t s c h o o l - m o t i o n l e s s w i t h i n c o m p a c t s c h o o l - h a l f m o t i o n l e s s i n c o m p a c t s c h o o l - r e m a i n d e r j o i n s c h o o l a f t e r l i g h t r a i s e d t o 8 .0 f t - c - r a p i d m o v e m e n t i n -to c o r n e r s o n i n t r o -d u c t i o n o f p r e d a t o r - s h o r t l y t h e r e a f t e r t w o m o t i o n l e s s c o m p a c t s c h o o l s f o r m e d , one o n e i t h e r s i d e o f p r e d a t o r T A B L E I I . B e h a v i o r o f e n u m e r a t e d a n d n o n - e n u m e r a t e d f r y p r i o r t o i n i t i a l s t i m u l u s , a n d a f t e r a p p l i c a t i o n o f e a c h s t i m u l u s . D P r i o r t o E X T E R N A L S T I M U L I A Y F r y T y p e D e n s i t y a p p l i c a t i o n o f i n i t i a l s t i m u l u s R a p i d L i g h t F l u c t u a t i o n V i b r a t i o n S i n g l e L i g h t F l u c t u a t i o n 1 N o n -E n u m 50 - s l o w s w i m m i n g a b o u t a q u a r i u m -no s c h o o l - h a l f s l o w s w i m m i n g i n l o o s e s c h o o l - r e m a i n d e r s l o w s w i m -m i n g a b o u t a q u a r i u m - r a p i d m o v e m e n t i n t o c o r n e r s - h a l f s l o w s w i m m i n g i n l o o s e s c h o o l , - r e m a i n d e r s l o w s w i m m i n g a b o u t a q u a r i u m E r n i m 50 - m a j o r i t y s l o w s w i m m i n g i n l o o s e s c h o o l - r e m a i n d e r s l o w s w i m m i n g a b o u t a q u a r i u m - m o t i o n l e s s w i t h i n c o m -p a c t s c h o o l - r a p i d m o v e m e n t i n t o c o r n e r s - s h o r t l y t h e r e a f t e r a l l f o r m m o t i o n l e s s c o m -p a c t s c h o o l - m o t i o n l e s s w i t h i n c o m p a c t s c h o o l •% N o n -E n u m 2 0 0 - s l o w s w i m m i n g a b o u t a q u a r i u m - m a j o r i t y s l o w s w i m -m i n g a b o u t a q u a r i u m - r e m a i n d e r s l o w s w i m -m i n g w i t h i n l o o s e s c h o o l - r a p i d m o v e m e n t i n t o c o r n e r s ; t h e r e a f t e r h a l f s l o w s w i m m i n g i n l o o s e s c h o o l - r e m a i n d e r s l o w s w i m -m i n g a b o u t a q u a r i u m - m a j o r i t y s l o w s w i m m i n g i n l o o s e s c h o o l - r e m a i n d e r m o t i o n l e s s i n c o r n e r s L E n u m 2 0 0 - m a j o r i t y s l o w s w i m m i n g a b o u t a q u a r i u m - m a j o r i t y m o t i o n l e s s w i t h i n c o m p a c t s c h o o l - f e w m o t i o n l e s s i n c o r n e r s a n d s o m e 6 l o w s w i m m i n g a b o u t a q u a r i u m - r a p i d m o v e m e n t i n t o c o r n e r s - s h o r t l y t h e r e a f t e r m a j o r i t y f o r m m o t i o n l e s s c o m p a c t s c h o o l - r e m a i n d e r m o t i o n l e s s i n c o r n e r s a n d s o m e s l o w s w i m m i n g a b o u t a q u a r i u m - m a j o r i t y w i t h i n c o m p a c t s c h o o l - r e m a i n d e r s l o w s w i m m i n g i n c o r n e r s a n d a b o u t a q u a r i u m T A B L E I I . ( c o n t i n u e d ) D P r i o r t o E X T E R N A L S T I M U L I A Y F r y T y p e D e n s i t y a p p l i c a t i o n o f i n i t i a l s t i m u l u s R a p i d L i g h t F l u c t u a t i o n V i b r a t i o n S i n g l e L i g h t F l u c t u a t i o n N o n -E n u t n 3 0 0 - s l o w s w i m m i n g a b o u t a q u a r i u m - s l o w s w i m m i n g - n o s c h o o l - r a p i d m o v e m e n t i n t o c o r n e r s - s h o r t l y t h e r e a f t e r m a j o r i t y f r e e s w i m -m i n g i n l o o s e s c h o o l - r e m a i n d e r s l o w s w i m -m i n g i n c o r n e r s a n d a b o u t a q u a r i u m - m a j o r i t y s l o w s w i m m i n g i n l o o s e s c h o o l - r e m a i n d e r s l o w s w i m m i n g a b o u t a q u a r i u m 3 E n u m 300 - m a j o r i t y s l o w s w i m m i n g a b o u t a q u a r i u m - r e m a i n d e r s l o w s w i m m i n g w i t h i n l o o s e s c h o o l - m a j o r i t y f o r m t w o m o t i o n l e s s c o m p a c t s c h o o l s - f e w m o t i o n l e s s i n c o r n e r s a n d s o m e f r e e s w i m m i n g a b o u t a q u a r i u m - r a p i d m o v e m e n t i n t o c o r n e r s - s h o r t l y t h e r e a f t e r t w o m o t i o n l e s s c o m p a c t s c h o o l s f o r m e d - f e w s l o w s w i m m i n g i n c o r n e r s - m a j o r i t y m o t i o n l e s s w i t h i n c o m p a c t s c h o o l - r e m a i n d e r s l o w s w i m m i n g i n c o r n e r s 4 N o n -Enum 400 - s l o w s w i m m i n g about a q u a r i u m - no s c h o o l - s l o w s w i m m i n g about a q u a r i u m - n o s c h o o l - r a p i d m o v e m e n t i n t o , c o r n e r s - s h o r t l y t h e r e a f t e r m a j o r i t y s l o w s w i m -m i n g w i t h i n l o o s e s c h o o l - r e m a i n d e r s l o w s w i m -m i n g a b o u t a q u a r i u m - m a j o r i t y f o r m a c t i v e , c o m p a c t s c h o o l - f e w m o t i o n l e s s i n c o r n e r s a n d s o m e s l o w s w i m m i n g a b o u t a q u a r i u m . E n u m 400 - s l o w s w i m m i n g a b o u t a q u a r i u m - n o s c h o o l - m a j o r i t y m o t i o n l e s s i n c o m p a c t s c h o o l - r e m a i n d e r s l o w s w i m -m i n g a b o u t a q u a r i u m - r a p i d m o v e m e n t i n t o c o r n e r s - s h o r t l y t h e r e a f t e r t w o m o t i o n l e s s c o m p a c t s c h o o l s f o r m e d - m a j o r i t y m o t i o n l e s s i n o n e c o m p a c t s c h o o l - r e m a i n d e r s l o w s w i m m i n g a b o u t a q u a r i u m Survival of Enumerated and Non-Enumerated Fry-It was shown that enumerated and non-enumerated fry, and ex-perienced and naive fry, display common behavioral differences. Since stream tests showed experienced fry to have a higher survival than naive fry, enumerated and non-enumerated fry were similarly tested to deter-mine whether the enumeration effect would increase the survival of enumerated fry. Four experimental streams, each with six predators, were used in this experiment. Experimental fry came from the Channel No. 2 migr tion of the previous night. Enumerated fry, collected from the counting table, were treated in the usual manner in preparation for testing. Non-enumerated fry were captured in the inclined-plane trap during normal channel migration, but were not removed until 0900 hours the following morning when test fry were selected and prepared for testing. With a completely randomized design, two groups of enumerated and then two groups of non-enumerated fry were introduced into the streams, acclimatized and released. Trials were conducted every secon day. Six replicates were obtained, over a 12-day experimental period. The results (Figure 21d) indicate that the enumeration process has a significant effect (Appendix Vlld) on fry in increasing survival over non-enumerated fry. Mean difference in survival was 9.7 per cent with enumerated fry experiencing 90.1 per cent survival and non-enumerated fry 80.4 per cent. Apparently migrating sockeye fry undergoing and surviving enumeration without injury have a greater ability to avoid or escape predation than non-enumerated fry. The causative effect could have been "fright" evoked by the extra handling that enumerated fry experienced during preparation for tests. 53 Discussion The foregoing results indicate that rainbow trout predation on migrant sockeye fry is affected by fry experience with predators. Apparently, if an individual sockeye fry encounters and escapes a predator, the experience obtained from the interaction increases the ability of that individual to escape a predator in following predator-prey interactions. Additional experience serves to further increase ability to escape a pre-dator. Indications are that the initial experience appears to invoke a unique behavioral response in the fry which is reflected in a specific schooling and migratory behavior. Experienced fry form compact schools and remain motionless in a predator-prey or other fright-evoking situation, whereas naive fry behavior approaches opposite reactions. The belief, then, is that experienced fry migrate in a manner rendering them less susceptible to predation and the migration pattern (compact and in mass) is influenced in part, or completely, by encounter and escape from predators experienced earlier. There is some indication from the survival differences observed between enumerated and non-enumerated fry that fright evoked by external stimuli may cause the same result as for fry experiencing a predator encounter. Apparently the enumeration process in some way increased the ability of enumerated fry to escape predators. The results suggest ways of reducing trout predation on young salmon fry migrating from artificial propagation facilities. For example, by exposing a large number of fry to a small group of predators prior to migration, fry experiencing the predators and escaping should experience lower mortality during subsequent migration. Predation might be reduced by implementing a scheme comparable to that which enumerated fry experience. Perhaps if fry were subjected to a fright-evoking stimulus prior to migration, predation would be reduced considerably. 54 The present results do not show conclusively the cause(s) of the observed survival differences between the various fry types. For example, it is not known whether the predators, or just the experience of migrating down an artificial stream, is responsible for the survival differences between experienced and naive fry. Survival differences between enumerated and non-enumerated fry suggest that handling of fry causes a fright response in fry which makes them more efficient at escaping predators than those not experiencing handling. More conclusive evidence regarding the requirement of fry experience of a predator as the cause of survival differences could be obtained from tests matching river fry against channel fry. River fry should have experienced predators prior to capture because of large populations of predators present in the river. Consequently, in stream tests between river and channel fry, river fry should experience higher survival than channel fry. If this were to happen, the results would support the theory that predators are responsible for the survival differences observed in the present experiments. Evidence could also be obtained by allowing fry to experience a stream migration in the absence of predators, and then testing them against fry with predator experience. If the latter experienced higher survival, one could agree with the theory of predator experience. Until further tests are done, the belief is that experience with predators increases the ability of sockeye fry to escape predation in its next predator encounter, and this ability is proportional to the amount of experience attained. 55 SECTION IV RAINBOW TROUT COLOR PREFERENCE AND HUNGER RESPONSE TO FOOD DEPRIVATION Introduction Among the many factors influencing a predator's selection for a suitable food item are the physical characteristics of that food item. Some of these factors probably include size, shape, odor, movement and color of prey. In a stream situation, not all of these factors would be as impor-tant as others. For example, odor within a stream would probably be unimportant as an immediate stimulus; movement would probably be insignificant because most organisms remain attached to the substrate or their passive drift with current is by far the greatest component of their movement. Color, size and shape could be considered important to a predator for various reasons, ranging from ease of food detection to the food size-mouth size relationship. As a starting point, color was chosen as the parameter to be tested in some detail. In this study, color preference was examined in relation to the effect of both background coloration and light intensity. Among the many factors affecting a predator's response for food, hunger is most significant. According to Holling (1966) the hunger com-ponent of his predation model could conceivably affect nearly all the sub-components of the model. Apparently most of a predator's activities in response to prey are affected by hunger. In this study, rainbow trout were tested for their hunger response with time in hopes of determining an expected hunger curve for a proposed study on the digestion rates of rainbow trout when feeding on sockeye fry. Eventually, studies of this type will lead to estimation of the total number 56 of sockeye fry that could be consumed by rainbow trout during a downstream fry migration at Fulton River. Materials and Methods Field Data This study, conducted at the site of the British Columbia Fish and Wildlife Commission Trout Hatchery, near Abbotsford, B.C., involved a series of artificial stream tests and a laboratory study. The four streams (Figure 25), suspended in Marshall Creek, were almost identical to those used at Babine Lake. They were screened on top 2 2 with 2.5 cm Gauge-12 screening, and at the inlets and outlets with 1 cm Gauge-18 screening.to prevent predators from escaping and foreign debris from entering. A l l streams were pale blue in color, but in one experiment the bottom coloring was changed. Water flowed through the streams at a depth of 30 cm. Velocity, controlled by adjustable, wooden-valve-like structures located at the stream inlets, was 38+2 cm/sec. Prey traps, 2 bag-like structures (Figure 26) consisting of 2.5 cm Gauge-12 screening and No. 5 mesh, were located in the last 0.4 m of each stream. They were constructed in a manner enabling easy installation, and removal, from the streams. Predators (Salmo gairdneri) of fork length size approximating 2 5 cm and of domestic stock, were obtained from the hatchery two years previously. During this period they were used in tests similar to those described below and were never moved from the artificial streams. Predator density per stream was five. Prey were dead rainbow trout eggs obtained from the Commission's hatchery at Summerland, B.C. For stream tests, eggs were colored with "Rit" fabric dye prior to use. Coloring was done by soaking the eggs for 57 Figure Z5 . A r t i f i c i a l stream apparatus suspended in Marshall Creek near Abbotsford, B.C. Top left: looking downstream. Top right: looking upstream. Bottom; two streams enclosed in black polypropylene for an experiment requiring controlled lighting. 58 Figure 26. Artificial stream prey traps. Top: prey trap consisting of fine mesh netting attached to coarse screen. Bottom: trap in position for capturing prey. 59 48 hours in a solution of 350 ml of water and 0.5 gm of dye. Five minutes prior to the beginning of a trial the eggs were removed from the dye solu-tion and placed in 50 ml of water. This was done to bring out the dye color in the eggs. For stream tests conducted during late morning hours, (intensity of light reflection on the water surface varied from 330 to 380 ft-c), the experimental procedure was as follows. At 0930 hours, streams were cleaned of all silt and organic debris that collected after the previous trial, and the prey traps were installed. At 1000 hours, eggs were introduced into the streams, at the inlets, and allowed to roll with the current down-stream into the traps. After a 10 minute period, the traps were removed and the uneaten eggs in the traps were counted. One stream test using two streams was conducted at night under controlled artificial lighting. The streams (Figure 32) were modified for controlled lighting in exactly the same way as described for experiments on physical factors conducted at Babine Lake. A food deprivation study was conducted, in the laboratory, on ten two-year-old rainbow trout. A l l fish were isolated from each other in individual 7-gallon aquaria arranged so that the fish could not see each other. Each aquarium was covered with black 4-mil polypropylene on all sides except one — the front surface. The top surface had a 2.54 cm^ hole through which fish were fed. Water was supplied to each aquarium by gravity flow from a nearby spring-fed pond. Automatic siphons were used to maintain circulation and a water temperature of 5.0 + 0.5°C. The entire experiment was conducted under laboratory lighting (40-watt fluorescent lights) and the intensity of light reflection on the front surface of each aquarium was 32 +• 2 ft-c. Accumulated digestive wastes were removed from the aquaria, by siphoning, every second day. 60 Predator food was dead rainbow trout eggs taken from the same stock used for the stream tests. In this experiment the eggs were not dyed. Additionally, only eggs of uniform size were fed to the predators. Predators of native origin, and of fork length size range 14.3 cm to 16.4 cm, were obtained from the hatchery. These fish were raised from the egg stage on enriched fish food and had never eaten rainbow trout eggs prior to this experiment. Color Preference Color preference by rainbow trout was examined using four experi-mental streams and a completely randomized experimental design. Eggs were fed in lots of 70, comprising 35 each of all combinations of two colors from the series: red, orange, yellow, blue, green, brown and black. A l l combinations were presented once to each of four groups of predators over a 28-day experimental period. No two groups of predators experienced the same color combination on an experimental day, although all groups even-tually experienced all combinations. Also, eggs of each color were pre-sented with eggs of the same color, once to each group of predators. Rainbow trout displayed a definite color preference pattern (Figure 27) in this experiment. Aside from a few exceptions (Appendix X), both red and blue eggs consistently experienced higher mortalities when in the presence of any of the other colors. For two color combinations (Figure 27a), red and blue appear to rank equally in color preference, but in the red-blue combinations, red eggs experienced a higher mortality three out of four times. Conversely, when eggs of one color (Figure 27b) were pre-sented alone, blue eggs experienced greater mortality than red eggs, with the mean difference being 5.75. From results of the two color combinations, the feeling is that red had the highest preferences. Comparison oi all colors showed preference, 61 Figure 27. Relationship showing the extent of rainbow trout color pre-ference by (a) the number of selections made by predators for a particular color when that color formed part of a two-color combination of eggs; (b) the number of eggs eaten when a single color was presented alone (Data, Appendix X). 62 from highest to lowest, to be red, blue, black, brown, orange, yellow and green. Results suggest that under the experimental conditions (pale blue streams and high light intensities) preference is for darker colors, or that preference declines with color brightness. Additionally, preference tends to decline as contrast between the food and background color declines. There is some indication that red and blue elicit a greater feeding response than other colors regardless of the environmental surroundings. For example, both red and blue are highly preferred and yet color contrast between red, blue and pale blue does not appear, to the human eye, to be as great as between black and pale blue. To provide more information to support or reject the above sug-gestions, experiments were conducted examining the effects of background coloration, and light intensity on rainbow trout color preference. Relationship Between Background Coloration and Color Preference The effect of background coloration on rainbow trout color pre-ference was examined in four streams using a completely randomized experimental design. Two color combinations of red, blue, black and yellow eggs (Figure 28), and eggs of each color (one-color combination) were presented alone to the predators in streams, each having a bottom coloration almost identical to one of the egg colors. Each two-color and one-color combination was presented once in each stream over a 10-day experimental period, and no two groups of predators experienced the same combination on an experimental day. Egg densities were the same as in the previous experiment. Rainbow trout color preference was substantially influenced by background coloration (Figure 29). For two-color combinations (Figure 29a), Figure 28. Experimental prey. Top: black, yellow, red and blue eggs. Bottom: two-color combinations shown by columns, from left to right—red-black, blue-red, red-yellow, yellow-blue, black-yellow, and blue-black. a. PI / / / / LEGEND V7\ BLUE EGGS I I YELLOW EGGS 111 RED EGGS 111 BLACK EGGS RED BACKGROUND i ! 1 7 ! '*•**".'! L ; i YELLOW BACKGROUND i i i i BLUE BACKGROUND 7 BLACK BACKGROUND 9 z o o ° o o c 3) o o ro 3) z o > 2 o z CO 0 = o O <->4 "TI z CO PI i d PI ;IJI5 PI — t .ll*' '* ••*» (,•;•'. - '* '** p] 70 z p o 50 o o 30 w m 10 Figure 29. Relationship showing the effect of background coloration on rainbow trout color preference by (a) the number of selections by predators for a particular color when that color formed part of a two-color combination of eggs; (b) the number of eggs eaten when a single color was presented alone (Data, Appendix X I ) . 65 highest preference on a red background was blue and then yellow; on a yellow background, red and then black; on a blue background, yellow and then red; on a black background, red and then yellow. Similar results were obtained for one-color combinations (Figure 29b). A l l trials except those on a blue background resulted in low mortalities for eggs the same color as the background. On a blue background, blue eggs experienced the third largest mortality. In any case, indications are that food selec-tion by rainbow trout will be influenced by the amount of color contrast between the food item and its surrounding environment. Experimental results also support the suggestion that red, and possibly blue, are highly attractive to predators regardless of environmental surroundings. On a black background the greatest contrast of the colors tested would appear to be yellow, yet red eggs experienced the highest mortality. Blue eggs experienced the third largest mortality on a blue background, yet contrast appears to be the lowest. Also, mortalities were very similar to that of red and yellow eggs. Relationship Between Light Intensity and Color Preference The effects of light intensity on rainbow trout color preference were examined in two streams (Figure 25) using a randomized block experimental design. Two-color and one-color combinations of red, blue, black and yellow eggs (Figure 28) were presented to two groups of predators under artificial light intensities of 0.1, 0.4, 0.8 and 1.2 ft-c. Each color combination was presented once in each stream over the 40-day experi-mental period, and no one combination was presented at the same light intensity on an experimental day. Egg densities were the same as in the previous experiments. Prior to each trial, predators were acclimatized, for one hour, to the designated light intensity of that trial. 66 Experimental results (Figure 30) indicate that color preference is affected by light intensity. Over the range tested, lighter colored eggs experienced greater mortalities than dark colored eggs. The red-yellow combination resulted in near equal mortality but results for single color combination show that yellow eggs experienced greater mortality. The results indicate the order of color preference at low light levels to be yellow, red, blue and black in that order. Contrasting mortality differences between colors in two-color combinations shows them to in-crease as contrast between colors increases. For example, total mortality differences for the red-yellow, red-blue and red-black combina-tions were 9, 77 and 118 eggs respectively. Similar results occurred for the yellow-blue and yellow-black combination. The results also show mortality to be directly related to light intensity between 0.1 and 0.8 ft-c. Above 0.8 ft-c it remained fairly constant except for the yellow-red and yellow-blue combinations which continued to increase. Mortality differences decreased slightly between 0.8 and 1.2 ft-c for the yellow-blue and yellow-red combinations. Comparison between these experimental results with those obtained in the daylight experiment shows yellow to be highly pre-ferred at low light levels and least preferred at high levels. A possible reason for this difference is that with increased light intensity contrast between yellow eggs and a pale blue background decreases, while at low light levels it increases. If contrast between yellow and pale blue varies, mortality would vary accordingly. The opposite should occur for red and blue eggs in that contrast is highest at high light levels and lowest at low lightlevels. Since mortality differences between red and yellow, and blue and yellow decrease between 0.8 and 1.2 ft-c, contrast may begin to change in this range. Expectations are that as light intensity increased, the proportion of red or blue eggs eaten, in comparison to yellow, would be greater. The reversal occurring between red and yellow at the 1.2 ft-c level supports this possibility. 67 50 n Z t u < u CO o IS Ul u. o cc UJ m I D Z Z < UJ .2 70r, u . O <• UJ Z W CD Z o U l z OI 0.4 0 8 1.2 L I G H T I N T E N S I T Y L E G E N D o R E D • B L A C K x Y E L L O W A B L U E 50 r 0.1 0.4 0 8 1.2' L I G H T I N T E N S I T Y Figure 30. Relationship showing the effect of light intensity on rainbow trout color preference when colored eggs were presented in two-color combinations of (a) yellow and blue; (b) red and blue; (c) blue and black; (d) red and black; (e) yellow and black; and when presented as single colors (red, blue, black and yellow)- (Data, Appendix XII). 68 Effects of Time of Food Deprivation on Hunger of Rainbow Trout In the laboratory individual fish were acclimatized to aquarium conditions and food (rainbow trout eggs) for seven days, starved for 72 hours, and then fed to satiation. Each fish was then deprived of food for various time intervals ranging from 12 to 144 hours. At the end of each deprivation interval each fish was fed one egg every 30 seconds until satiation, or until 10 uneaten eggs had accumulated on the aquarium bottom. Observation of feeding behavior during the acclimatization period showed that all individuals displayed an all-or-nothing feeding response when presented with food. For almost every feeding period acceptance or rejection of food was clearly apparent. If a fish rejected an egg after eating several, it would generally reject any additional eggs presented. The reason for presenting ten additional eggs after rejection was merely to check on satiation. The time of satiation or the beginning of a food deprivation interval was arbitrarily chosen at the time five minutes prior to when the last egg was presented to a fish. For example, if the last egg was pre-sented at 110 5 hours, satiation occurred and food deprivation began at 1100 hour s. Experimental results (Figure 31) show how the amount of food required to satiate the experimental fish changed with the time of food deprivation. The S-shaped curve signifies that as time increased, hunger increased at a progressively increasing rate up to the 72-hour deprivation interval, and at a progressively decreasing rate thereafter to a sustained maximum. Attempts to fit the results, by the vom Bertalanffy growth equation, such as was done by Holling (1966) to describe a negatively-accelerated rise to a plateau, were unsuccessful. In fitting the present results to the von Bertalanffy equation, estimates of "K", a constant for the rate of food eaten, diverged rather than approached each other. 69 TIME DEPRIVED OF FOOD (HOURS) Figure 31. Relationship showing the effect of food deprivation on hunger of rainbow trout. Data (Appendix XIII) fitted to von Bertalanffy equation. 70 The results suggest that beyond a point further increase in the length of food deprivation will result in sustained maximum consumption. This is shown at the 96, 120 and 144 hour intervals when the number of eggs consumed was about 43. Holling (1966) states that the plateau of a hunger curve is a measure of gut capacity. Consequently, when given the opportunity to feed on sockeye fry, rainbow trout will do so until gut capacity is reached. In laboratory studies conducted by Ivlev (1945; 1961), carp (Cyprinus carpis) feeding on non-living food, roach (Rutilus rutilus caspicus) feeding on Chironomid larvae, and bleak (Alburnus alburnus) feeding on Daphnia, all displayed negatively accelerated hunger curves (from Holling, 1965). One would expect the rainbow trout hunger curve to be the same or similar when feeding on sockeye fry. The unexpected S-shaped curve obtained in the present study is probably due to the hard outer shell of the eggs. Personal observation on stomach samples from Rocky Mountain Whitefish (Prosopium williamsoni) held in the laboratory, has shown that sockeye salmon eggs may remain unassimilated in the stomach for two days or more. Consequently, in determining digestive characteristics of rainbow trout feeding on sockeye fry, one should use sockeye fry as the experimental food. Consumption, assimilation and evacuation character-istics probably are entirely different than for eggs. The results of the present experiments raise the possibility that predators could be fed with salmon eggs as a means of satiating their hunger for a substantial period of time, during which the predation on migrating fry would be much reduced. Discussion Stringer and Hoar (195 5) demonstrated that some colors elicit greater aggressive behavior by under-yearling Kamloops Trout (Salmo 71 gairdneri) than others. Kwain and McCrimmon (1967) showed that yearling trout can distinguish between black and white bottom colorations. When exposed to a series of white light intensities over either white or black backgrounds, preference was for black. The present experimental results indicate that rainbow trout display color preference patterns which are influenced by background coloration and light intensity. Preference was for darker colors at high light levels and for lighter colors at low levels. As the contrast between background coloration and food color increased, predation increased. Additionally, predation differences increased proportionately with con-trast between food colors. The results also suggest that regardless of surrounding environmental conditions, rainbow trout display innate pre-ference for red, and possibly blue, during feeding. Since the physical characteristics of streams are variable from place to place, and from time to time, expectations are that color pre-ference would be variable within and between streams. For example, during spring runoff some streams are turbid while others are clear; therefore, color preference would vary accordingly. One would also expect spatial and temporal variation in color preference due to the unstable nature of stream bottoms. Added to this is the effect of water-surface and below-surface light intensities on food colors and background colorations. The existence of color preference patterns by rainbow trout could be used as a tool for reducing predation of migrant sockeye fry. For example, if foreign objects were introduced into a stream in an attempt to confuse or detract predators away from migrating fry, one would use objects of a color which is highly preferred by the predators. Obviously the color(s) chosen should have the sharpest contrast with the background coloration at low light levels. Variation in the physical characteristics of 72 a stream would make it difficult to implement such a predation reduction scheme on a large scale. Therefore, the scheme would probably be best confined to areas of a stream wher'e predation is unusually heavy, such as spawning channel outlets, river mouths, narrow sections of a stream, or any other areas where fry are extremely vulnerable to predation. In the present experiments, predators were of domestic origin and of the same age class. No information was obtained on variability of color preference with age or origin. These possible sources of variation should be examined prior to large scale field trials aimed at reducing pre-dation using colored foreign objects. The results for the food deprivation experiment indicate that rainbow trout display an S-shaped hunger response curve when feeding on eggs rather than the expected negatively accelerated curve. Reasons for the deviation are probably associated with the digestibility of experimental foods used in each case. There was some indication that rainbow trout will feed to gut capacity if given the opportunity. Finally, beyond an upper limit of food deprivation, the amount of food eaten by an individual remains fairly constant. A full scale program to estimate annual loss of migrant sockeye from predation should include consumption, assimilation and evacuation studies using sockeye fry as the experimental food. Inferences drawn from studies using a different food type, especially eggs, will probably be invalid. They nevertheless raise the possibility that eggs could be "fed" to predators in a stream, and because of their slow digestibility, would cause reduced predation on fry for as much as two days. 73 GENERAL DISCUSSION The foregoing has demonstrated how physical factors, fry experience with predators, color preference, predator light acclimatiza-tion prior to fry migration, and possibly the introduction of foreign objects into a stream, could influence predation on young salmon fry. It was shown that manipulation of physical factors such as water turbidity, water velocity, and light intensity, can reduce predation on fry. Predator efficiency may be reduced by subjecting predators to abnormal light prior to fry migration or by introducing foreign objects into a stream with migrating fry. Allowing migrant fry experience with predators was shown to improve their ability to escape or avoid predators on subsequent encounters. There was some indication that handling of fry would create the same effect. It was suggested that one might increase fry survival during migration by allowing large numbers of fry to experience predators prior to migration. It was shown that rainbow trout display color preference patterns which are influenced by background coloration and light intensity. The suggestion was that color preference could be applied to the foreign objects scheme of reducing predation in that the objects could be of a color highly preferred by predators. If predators spent more time preying on the colored objects, obviously less time would be spent on migrating fry, resulting in reduced predation on fry. Finally, it is possible that various combinations of the above factors could be used to reduce rainbow trout predation on salmon fry. The information gained from this study could be applied to full scale field trials aimed at reducing predation on salmon fry. However, it would seem more desirable to pursue the study, filling in the gaps sug-gested and then proceeding into full scale trials. Implied in this paper is the fact that a reduction in predation on migrating fry would be reflected in future adult populations; that is, 74 increased population densities would result. In the case of pink (Oncorhynchus gorbuscha) or chum (Oncorhynchus keta) salmon which migrate directly to the sea on emergence, this is not a poor assumption. However, for sockeye a year or more in lake residence prior to going to the sea, it could be a poor assumption. For example, it is quite possible that fry production in excess of a lake's carrying capacity would only create a situation where more severe density-dependent relationships could occur. Large fry populations have a greater chance of experiencing parasitic infections, or starvation from lack of food caused by over-cropping. Additionally, predator populations could increase, which in turn would be reflected by smaller salmon populations in subsequent years. While this may be true in Babine Lake, it certainly is not the case in areas where salmon runs are being initiated or rehabilitated. For example, the Upper Adams and the Quesnel-Horsefly river systems, in the Fraser drainage in North Central British Columbia, have lakes of high carrying capacity. In these areas techniques related to those described in this paper could be used to rehabilitate or initiate salmon runs. They might also be used to establish pink salmon runs in off years. More generally, in the large amount of literature on predation, consideration is given to the experience of predators and its bearing on predator success in capturing prey. Little consideration has been given to prey experience with predators and its effect on the outcome of predator-prey interactions. One reason for the lack of concern is that most predation studies place emphasis on predator activity. Another reason is that the prey used are generally slow moving, or dead, or incapable of learning to appreciate the significance of a predator. In learning more about predator-prey interactions, more emphasis must be given to prey experience. Only in this way will the complexities of predator-prey relationships be better understood. 75 R E F E R E N C E S Ginetz, R. M . 1970. Predator-prey interactions that affect survival of migrant sockeye salmon (Oncorhynchus nerka) fry. B.Sc. thesis, University of British Columbia, Vancouver. 37 pp. Holling, C S . 1965. The functional response of predators to prey density and its role in mimicry and population regulation. Mem. Ent. Soc. Can., 45:1-60. . 1966. The functional response of invertebrate predators to prey density. Mem. Ento. Soc. Can., 48:1-86. Hunter, J . G . 1959. Survival and production of pink and chum salmon in a coastal stream. J . Fish. Res. Bd. Canada, l6(6):835-886. Kwain, W. and H. McCrimmon. 1967. The behavior and bottom color selection of the rainbow trout Salmo gairdneri Richardson, exposed to different light intensities. Animal Behaviour, 15:75-78. Liar kin, P. A . Simulation studies of the Adams River sockeye salmon (Oncorhynchus nerka). J . Fish. Res. Bd. Canada, 28:1493-1502. Neave, N. 1953. Principles affecting the size of pink and chum popula-tions in British Columbia. J . Fish. Res. Bd. Canada, 9(9):450-491. Ricker, W.E. 1950. Cycle dominance among the Fraser sockeye. Ecology, 31(l):6-26. Stringer, G . E . and W. S. Hoar. 1955. Aggressive behavior of under-yearling Kamloops trout. Can. J . Zool., 33:148-160. Swihart, C.A. and S .L. Swihart. 1970. Colour selection and learned feeding preferences in the butterfly, Heliconius chartitonius Linn. Animal Behaviour, 18:60-64. 76 APPENDIX TABLE I. Results of three-way analysis of variance (a) on the main effects of treatments, velocity and time on fry mortality; and (b) between the various treatment effects on fry mortality. Analyses are two-tailed tests of means. (a) Source of Variance d.f. SS MS F Velocity 1 656.90 656.90 226.91 ** Treatments 7 3038.52 434.10 149.93 Days 7 75.79 10.82 3.74 -Vel. x Treat. 7 89.00 12.71 4.39 * Vel. x Days 7 27.29 3.89 1.34 -Treat, x Days 49 166.70 3.40 1.17 -Vel. x Treat. x Days (Error) 49 141.80 2.89 Total 127 4196.00 (b) Source of Variance d.f. SS + MS F Light 1 1810.32 164.72 ** Turbidity 1 326.86 29.74 ** Fry type 1 878.21 79.90 *# Light x Turbidity 1 2.50 0.22 -Fry x Turbidity 1 6.79 0.61 -Fry x Light 1 2.85 0.25 -Fry x Light x Turbidity (Error) 1 10.99 Total 7 3038.52 d.f. = degrees of freedom; SS = sums of squares; MS = mean square = SS/f; F = ratio of MS of effect/residual MS; - =,not significant; * = significant at 5% level; ** = significant at 1% level; Treatments (listed in part (b)). 77 APPENDIX TABLE II. Results of three-way analysis of variance with fry type, light intensity, and time as main effects for comparing mortality differences between fully-developed Fulton River and Spawning Channel No. 1 fry at the different light intensities. Analysis is two-tailed test of means. Source of Variance d.f. SS MS F Fry type 1 72.89 72.89 21.73 ** Light 3 64.09 21.36 6.37 * Days 3 6.68 2.23 0.66 -Fry type x Light 3 50.74 16.91 5.04 * Fry type x Days 3 11.42 3.81 1.13 -Light x Days 9 37.54 4.17 1.24 -Error 9 30.17 3.35 Total 31 273.53 d.f. = degrees of freedom; SS = sums of squares; MS = mean square = SS/f; F = ratio of MS of effect/residual MS: = not significant; * = significant at 5% level; ** = significant at 1% level; Fry type = fully developed Fulton River and Spawning Channel No. 1 fry; Light = light intensities of 0.05, 0.10, 0.20, 0.30 foot candles. 78 APPENDIX TABLE III. Results of two-way analysis of variance with light intensity and time as main effects for comparing fry mortality at the different levels of light intensity. Analysis is two-tailed test of means. Source of Variance d.f. SS MS F Light 4 38.91 9.73 26.38 ** Days 4 1.95 0.49 1.32 -Error 16 5.90 0.37 Total 24 47.76 d.f. = degrees of freedom; SS = sums of squares; MS = mean square = SS/f; F = ratio of MS of effect/residual MS; = not significant; ** = significant at 1% level; Light = light intensities of <.05, 0.1, 1.6, 12.5, 25.0 foot candles. ( 79 APPENDIX TABLE IV. Results of (a) one-way analysis of variance, with unequal replication, on the effect of the period of predator light acclimatization on fry survival, and (b) Duncan's new multiple range test of comparisons between treatment means. Analysis is one-tailed test of means. (a) Source of Variance d.f. SS MS F Treatments 4 1623.66 405.92 9.94 ** Error 19 776.18 40.85 Total 23 2399.84 (b) Treatments (minutes) 0 40 10 20 80 Treatment means 77.83 94.50 95.75 97.25 98.17 Significance d.f. = degrees of freedom; SS = sums of squares; MS = me an s quar e = SS / f; F = ratio of MS of effect/residual MS; ** = Significant at 1% level; Treatment = periods of predator light acclimatization. = significantly different (5%) from mean not underlined. APPENDIX TABLE V. Experimental results showing fry survival differences in relation to periods of predator light acclima-tization. Acclimati- _ _ Stream 1 Stream 2 zation Day Period Mortality Survival Mortality Survival 1 80 0 100 0 100 2 0 14 86 16 84 3 20 4 96 2 98 4 10 3 97 10 90 5 40 5 95 8 92 6 80 2 98 4 96 7 0 13 87 16 84 8 20 5 95 2 98 9 10 4 96 0 100 10 40 4 96 5 95 11 80 5 95 0 100 12 0 39 61 35 65 81 APPENDIX TABLE VI. Results of two-way- analysis of variance on the main effects of the presence and absence of styrofoam strips, and time, for comparing mortality differences between fry in the absence and presence of styrofoam strips. Analysis is one-tailed test of means. Source of Variance d.f. SS MS F Treatments 1 1.38 1.38 0.08 Days -5 9.98 1.99 0.12 Error 5 81.83 16.36 Total 11 90.43 d.f. = degrees of freedom; SS = sums of squares; MS = mean square = SS/f; F = ratio of MS of effect/residual MS; = not significant; Treatments = with and without styrofoam strips. APPENDIX T A B L E VII. Results of two-way analysis of variances with fry type and time as major effects for comparing differences in survival between (a) naive and experienced fry having 24 hours inexperience; (b) naive and experienced fry having 48 hours in-experience; (c) successive groups of naive fry and a group of experienced fry; (d) enumerated and non-enumerated fry. A l l analyses are one-tailed tests of means. (a) Source of Variance d.f. SS MS F F r y type Day s Error Total 1 15 15 31 1501.11 546.93 774.96 2823.00 1501.11 36.46 51.66 29.06 0.71 ** (b) Source of Variance d.f. SS MS F F r y type Days Error Total 1 5 5 11 257.70 208.17 230.96 696.83 257.70 41.63 46.19 5.58 0.90 (c) Source of Variance d.f. SS MS F F r y type Days Error Total 1 5 5 11 757.31 510.18 313.06 1580.55 757.31 102.04 62.61 12.10 1.63 * (d) Source of Variance d.f. SS MS F F r y type Days Error Total 1 5 5 11 182.36 302.65 118.42 603.43 182.36 60.53 23.68 7.70 2.56 * d.f. = degrees of freedom; SS = sums of squares; MS = mean square = SS/f; F = ratio of MS of effect/residual MS; - = not significant; * = significant at the 5% level; ** = significant at the 1% level; Fry type = naive and experienced fry. APPENDIX TABLE VIII. Experimental results showing survival differences between (a) naive, and experienced fry having 2 4 hours inexperience; (b) naive, and experienced fry having 4 8 hours inexperience; (c) successive groups of naive fry, and the same group of experienced fry; and (d) enumerated and non-enumerated fry. (a) Number Pre- Fry survival in streams of dators * = experienced % % Date Streams per Survival Survival 1 9 7 1 Used Stream 1 2 3 4 5 ' 6 Naive Experienced May 3 6 4 6 0 5 8 * 5 3 6 0 * 5 8 4 5 7 2 . 0 7 8 . 6 5 6 4 6 3 6 6 6 8 * 6 1 6 0 6 4 * 8 3 . 3 8 8 . 0 7 5 5 7 5 * 6 6 6 6 7 1 5 9 * 9 0 . 1 8 9 . 3 9 I I I I - - 6 0 7 5 * 7 5 * 6 9 4 4 7 6 . 8 1 0 0 . 0 1 1 I I I I - - 5 0 7 5 * 6 3 7 4 * 4 2 6 8 . 8 9 9 . 3 1 3 4 6 -- 7 1 * 6 7 6 4 6 9 * 8 7 . 3 9 3 . 3 1 5 5 I I - - 6 5 5 2 7 5 * 7 4 * 5 5 7 6 . 4 9 9 . 3 1 6 i t - - 6 0 6 9 * 7 5 * 5 4 5 5 7 5 . 1 9 6 . 0 1*7 i t . - - 5 5 5 9 7 4 * 7 0 * 6 3 7 8 . 7 9 6 . 0 1 9 6, e x c e p t #2 h a d 4 ' - - 7 0 7 1 * 6 0 5 8 7 4 * 7 8 . 7 9 6 . 7 2 0 I I - - 6 5 7 3 7 5 * 7 5 * 6 0 8 8 . 7 1 0 0 . 0 2 1 I I I I - - 6 7 6 4 7 3 * 6 0 7 5 * 8 2 . 7 9 8 . 7 2 2 I I - - 7 0 7 0 * 7 0 7 2 * 6 6 9 0 . 7 9 4 . 7 2 3 I I - - 6 7 7 2 * 7 0 * 5 3 6 0 7 5 . 3 9 4 . 7 2 4 6 - - 6 0 7 3 * 5 8 7 4 * 5 7 8 1 . 0 9 8 . 0 Fry survival in streams % % * = experienced Survival Survival Date 1 2 3 4 5 6 Naive Experiencec May 2 7 5 0 6 7 * 5 6 5 9 7 5 * 6 1 7 4 . 6 9 6 . 0 2 9 6 5 6 2 7 0 * 5 8 6 3 7 3 * 8 2 . 3 9 5 . 3 3 1 6 7 7 0 7 1 * 6 2 4 2 * 5 7 8 5 . 3 7 5 . 3 June 2 6 9 * 5 9 5 7 5 6 5 5 * 4 6 7 2 . 7 8 2 . 7 4 6 1 7 3 * 5 8 6 1 4 1 5 8 * 7 3 . 7 8 7 . 3 6 6 4 * 5 1 6 3 6 2 * 4 0 3 1 6 1 . 3 8 4 . 0 84 APPENDIX TABLE VIII. (continued) S T R E A M 1 S T R E A M 2 F r y H a n d l i n g F r y % H a n d l i n g D a t e D e n s i t y T y p e M o r t a l i t y S u r v i v a l M o r t a l i t y T y p e M o r t a l i t y S u r v i v a l M o r t a l i t y May - 2 6 8 5 N a i v e 6 9 2 . 9 0 E x p . 1 9 8 . 8 2 2 8 8 2 E x p . 3 9 6 . 3 7 N a i v e 3 9 6 . 3 0 30 6 8 N a i v e 11 8 3 . 8 0 E x p . 3 9 5 . 6 1 J u n e 1 64 E x p . 3 9 5 . 3 6 N a i v e 31 5 1 . 5 0 3 5 5 N a i v e 7 8 7 . 2 0 E x p . 0 1 0 0 . 0 5 5 50 E x p . 0 1 0 0 . 0 0 N a i v e 7 8 6 . 0 0 (d) Date Fry survival in streams * - enumerated i 3 4 5 6 % Survival Non-Enumerated % Survival Enumerated May- 26 70 75* 72 62* 94.7 91.3 28 71* 65 71* 66 87.3 94.7 30 66* 62 58 67* 80.0 88.7 June 1 68 75* 70* + 49 78.0 96.7 3 70* 62 51 60* 75.3 86.7 5 56 70* 45 54* 67.3 82.6 85 APPENDIX TABLE IX. Results of three-way analysis of variance with (a) fry type, time and days as main effects for comparing migra-tion differences between naive and experienced fry; and (b) stream type, time and days as main effects for comparing migration differences for naive fry in streams with and without predators. A l l analyses are one-tailed tests of. means. (a) Source of Variance d.f. SS MS F Time 3 1776.67 592.22 21.87 ** Fry type 1 36.51 36.51 1.35 -Days 3 24.32 8.12 0.30 -Time x Fry type 3 64.74 21.58 0.80 -Fry type x Days 3 7.45 2.48 0.09 -Time x Days 9 438.33 48.70 1.80 -Error 9 243.75 27.08 Total 31 2591.77 (b) Source of Variance d.f. SS MS F Time 3 217.50 72.50 2.57 Stream type 1 1120.51 120.51 4.27 * Days 3 60.03 20.01 0.71 -Time x Type 3 1649.04 549.68 19.46 ** Type x Days 3 19.86 6.62 0.23 -Time x Days 9 484.40 53.82 1.91 -Error 9 254.31 28.25 Total 31 2805.65 d.f. = degrees of freedom; SS = sums of squares; MS = mean square = SS/f; F = ratio of MS of effect/residual MS; = not significant; * = significant at 5% level; ** = significant at 1% level; Fry type = naive and experienced; Stream type = streams with and without predators. 86 APPENDIX T A B L E X . Experimental results showing mortality differences between red, orange, yellow, green, blue, brown and black eggs. M O R T A L I T Y DAY S T R E A M 1 S T R E A M 3 S T R E A M 4 S T R E A M 2 1 B=31; Br=22 Br=34; Y=27 R=31; 0=23 Bk=22; Br = 10 2 R = 23; G=5 R = 28; B=22; 0=25 Br = 18 Y = 14 0=0 3 Y=33 G=39 0=33 G=27 Y=13 4 B = 35; Y=34 B = 35; 0=30 Bk=35; Y=34 B = 33 Bk=26 5 Br=46 Bk=32; 0=24 B k =29 ; G=14 R = 2 9 Br=22 6 R = 58 R=31; Bk=14 B=24 G=13 G=28 0=23 7 B=62 Bk=50 R=33 Y=20 Br=30; G=26 8 Br=35; Y=28 Bk=18; Br = l l B=20 Br = 12 R = 25 0=25 9 R = 26; B=24 0=35 Y=21 R = 22 G=13 Br=26; 0=21 10 G=51 Y = 25 G=25 Y=63 0=62 11 B = 33; 0=26 B = 33 Bk=21 B = 35 Y=24 Bk=32; Y = 23 12 0=26; Bk=l8 R = 27 Br = 22 Br = 53 Bk=33; G=22 13 R=25; Bk=25 0 = 20 G=20 R = 53 B=27 G=15 14 Bk=36 Br=35; G=34 B = 57 R=26 Y=22 15 Bk=34; Br=22 R = 31; 0=28 Br = 27; Y=26 B=23 Br = 22 16 0=24; Y=16 0=25 Br=25 B = 34 R=30 R=33 G=16 17 G=28; Y=14 0=58 G=52 Y=43 18 B = 27; Bk=20 Bk=34; Y=21 B = 30; 0=26 B = 31; Y=17 19 R=22; Br = 15 Bk=33; G=25 Bk=29; 0=22 Br=48 20 O=30; G=18 Bk=66 Bk=21; R = 16 R=43 21 Br=33; G=23 R = 35 Y=29 Bk=66 B=48 22 R = 32; 0=27 B = 33 Br=27 Br = 33; Bk=33 Br=20; G=18 23 0=23; Br = l6 R = 23 G=12 0=25 Y = 16 R=26; B=21 24 0=40 Y=47 Y = 35 G=29 G=26 25 Bk=35; Y = 2 9 B = 35; Y=27 B = 35 Bk=26 B=21; 0=19 26 Bk=33; G=17 Br =43 R=26 ; Br=23 Bk=25; 0=12 27 B=34; G=7 R = 57 0=32 ; G=26 R=21; Bk=l6 28 R = 31; Y=4 B=67 Br=22; G=l6 Bk=43 B = blue; Br = brown; R = red; B l = black O = orange G = green 87 APPENDIX T A B L E XI. Experimental results showing mortality difference between red, yellow, blue and black eggs on red, yellow, blue and black backgrounds. M O R T A L I T Y S T R E A M 1 S T R E A M 2 S T R E A M 3 S T R E A M 4 red yellow blue black DAYS background background background background 1 R=22, B=27 R=48 Y=30; B=25 R=24; Bk=ll 2 B = l l ; Bk=4 B = 18; Y=8 Bk=22; Y=16 B = 34 3 R=15 R=26; Y = l l R=24; Bk=10 R=20; B=10 4 R=24; Bk=19 Bk=36 R = 26; B=21 Y = 14; Bk=9 5 Y=42 Bk=21; B=19 Bk=35 R=41 6 Bk=39 R=24; Bk=14 R=68 R=28; Y=25 7 B=30, Y = 15 Y = 24; Bk=20 Y = 22; R=17 B = 16; Bk=13 8 Y=33; R=ll R=19; B = 9 B = 56 Y=32 9 B=49 Y = 51 B = 33; Bk=27 Bk=18 10 Y=35; Bk=17 B = 53 Y=66 Y = 18; B = l l B = blue; R = red; Bk = black; Y = yellow 88 APPENDIX TABLE XII. Experimental results showing mortality differences between red, yellow, blue and black eggs at four light intensities (0.1, 0.4, 0.8 and 1.2 ft-c). Color Light MORTALITY Combination Intensity STREAM 1 STREAM 2 red + blue L l R=14; B=6 R=ll; B=7 L2 R=20; B=6 R = 19; B=7 L3 R=21; B=13 R=29; B=18 L4 R=25; B=15 R=27; B=17 blue + black L l B=2; Bk=2 B=2; Bk=3 L2 B=22; Bk=10 B = l l ; Bk=5 L3 B=30; Bk=12 B = 17; Bk=l6 L4 B=29; Bk=13 B=24; Bk=10 red + red L l R=6 R=7 L2 R=30 R=17 L3 R=46 R=30 L4 R.= 54 R=41 red + black L l R=16; Bk=3 R=ll; Bk=l L2 R=19; Bk=9 R = 12; Bk=4 L3 R=32; Bk=10 R=26; Bk=6 L4 R=32; Bk=15 R=24; Bk=6 yellow + yellow L l Y=20 Y=17 L2 Y=31 Y=36 L3 Y=35 Y=39 L4 Y=68 Y=61 black +black L l Bk=4 Bk=5 L2 Bk=l6 Bk=17 L3 Bk=26 Bk=30 L4 Bk=30 Bk=39 yellow + blue L l Y = 14; B=7 Y = 13; B=2 L2 Y=25; B=22 Y=25; B = 18 L3 Y=28; B=2I Y=22; B=21 L4 Y=29; B=25 Y=26; B=26 red + yellow L l R=12; Y=9 R=2; Y=7 L2 R=14; Y=10 R = 16; Y = 13 L3 R=16; Y=16 R=25; Y=17 L4 R=30, Y=29 R=29; Y=34 blue + blue L l B=8 B=0 L2 B=18 B=21 L3 B=35 B=30 L4 B=49 B=40 yellow + black L l Y=14; Bk=23 Y = 16; Bk=23 L2 Y=25; Bk=29 Y=22; Bk=28 L3 Y=23; Bk=32 Y=25; Bk=31 L4 Y=25; Bk=31 Y=26; Bk=43 B = blue; L l = 0.1 ft -c; Bk = black; L2 = 0.4 ft -c; Y = yellow; L3 = 0:8 ft -c; R = red. L4 = 1.2 ft -c. 89 APPENDIX TABLE XIII. Experimental results showing the effect of food deprivation on hunger of rainbow trout. DEPRIVATION PERIODS Fish 12 hours 24 hours 48 hours 72 hours 96 hours 120 hours 144 hours 1 2 6 16 30 52 46 51 2 4 7 10 29 52 47 54 3 0 6 14 23 33 53 49 4 0 10 21 35 50 34 41 5 0 1 16 45 49 36 37 6 1 , 6 12 32 40 48 48 7 0 2 9 26 34 39 32 8 0 11 16 32 38 52 49 9 2 14 18 21 29 35 33 10 1 6 13 32 45 44 43 Total Eaten 10 69 145 30 5 422 434 437 X 1.0 6.9 14.5 30.5 42.2 43.4 43.7 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0101880/manifest

Comment

Related Items