EFPEQTS GE TEES ESTER-4538 CRGSSBREEDENG 0? C030 SALMON (oncoamcmts msurcu) 0R EARLY fiflGRTALETY ARI GROWTX “tests {so fix: Dawes cf M. 5. WWW“ STAYE UNWERS’ET’Y Robert L. Hite i973 “A‘ ‘ LIBRARY L Michigan Stave University ABSTRACT EFFECTS OF THE INTER-AGE CROSSBREEDING 0F COHO SALMON (oncomnvaws xrsm'cn) 0N EARLY MORTALITY AND GROWTH By Robert L. Hite In fall 1969 inter-age crossbreeding of coho salmon was initiated in an effort to develop a strain suitable for fresh water existance. Precocious male (jack) coho salmon were utilized in fall 1969 and 1970 to fertilize ova of age III females. It was hoped progeny of this cross would display faster growth and earlier maturity and thus through reduction in generation time, an acceleration in the selection process over time. No reduction in egg fertilization was evident in 1970 from the utilization of jack coho milt in contrast with milt from age 111 males. Mortality prior to hatching was greater for jack-fertilized ova in 1969 than for control ova. In 1970, however, control ova exhibited higher mortality up to hatching when compared with jack- fertilized ova. Mortality of control fry-juveniles from normal causes was higher in 1970 and 1971 than that experienced by experi- mental fish. Experimental coho were somewhat larger than control fish at time of release in July 1970 and 1971. No returns of marked experimental and/or control coho salmon released in Thonpson Creek in 1970 and 1971 were reported. EFFECTS OF THE INTER-AGE CROSSBREEDING 0F COHO SALMON (ONCORHYNCHUS KISUTCH) 0N EARLY MORTALITY AND GROWTH By Robert L. Hite A THESIS Stbmitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Fisheries and Wildlife 1973 ACKNOWLEDGEMENTS It is appropriate at this time to express my gratitude to those who provided assistance and direction in the course of this study and in the formulation of professional philosophy. To Dr. Peter Tack, under whom this investigation was conducted, for his guidance during the study and the many stimulating discussions we have had in the past. Dr. Eugene Roelofs provided critical review of this manuscript and numerous suggestions pertinent to the study. His excellent instruction during my formal schooling and genuine concern for the student was greatly appreciated and will long be remenbered. To the "Claw", Dr. David Jude, I owe many thanks for his statistical assistance, many hours of computer time and unique editing style. Without his aid this undertaking would probably still be incomplete. And to my*wife Patricia, whose assumption of the many domestic reSponsibilities and excellent typing made it possible for me to finalize this thesis. This investigation was conducted under a research grant provided by the Sport Fishery Research Foundation. This support made it possible for my continued education and is deeply appreciated. ii TABLE OF CONTENTS Page INTRODUCTION .......................... 1 METHODS AND MATERIALS ..................... 4 Egg Collection and Incubation 1969-197D .......... 4 Cultural Methods 1970 ................... 7 Food and Feeding ................... 8 Sampling and Marking Procedure ............ 8 Cultural Methods 1971 ................... 9 Food and Feeding ................... 12 Sampling and Marking Procedure ............ 12 Stocking Location .................. 12 RESULTS AND DISCUSSION 1969-1970 ................ 16 Incubation Mortality ................... 16 Rearing Mortality ..................... 18 Growth .......................... 18 Planting and Returns ................... 24 RESULTS AND DISCUSSION 1970-1971 ................ 25 Fertilization ....................... 25 Incubation Mortality ................... 27 Rearing Mortality ..................... 27 Growth .......................... 31 Carrying Capacity ..................... 44 Food Conversion ...................... 48 Planting and Returns ................... 50 SUMMARY AND CONCLUSIONS .................... 56 LITERATURE CITED ........................ 60 APPENDIX ............................ 65 Table LIST OF TABLES Mortality of eggs during incubation, December, 1969. Mortality totals include unfertilized eggs ....... Mortality of experimental and control coho salmon from post-hatching period through final weighing, Decenber 16 - June 17, 1970 .............. Mean total length, weight and calculated standard error of experimental and control coho salmon on six sampling dates in 1970 ............... Fertilization and mortality of experimental and control coho salmon e gs taken in Novenber, 1970, by individual female tray). Mortality totals include all fertilized eggs removed up to hatching . . . . Mortality of experimental and control coho salmon fry from hatching through final weighing, June 4, 1971. Coho groups 1 and 2 artificially spawned on Novenber 3, 1970 .................. Mortality of experimental and control coho salmon fry from hatching through final weighing, June 4, 1971. Coho groups 3 and 4 artificially spawned on November 19, 1970 .................... Mean total length in millimeters and calculated standard error of experimental and control coho salmon on nine weighing dates in 1971 ......... Mean weight in grams and calculated standard error of experimental and control coho salmon on nine weighing dates in 1971 ................. Analysis of variance to test regression line linearity of mean total length vs time for group 1 experimental coho salmon on nine weighing dates in 1971 ........................ iv Page 17 19 20 26 28 36 37 LIST OF TABLES - Continued Table 10. 11. 12. 13. Analysis of variance to test regression line linearity of mean total length vs time for group 2 control coho salmon on nine weighing dates in 1971 . . . Analysis of variance to test regression line linearity of mean total length vs time for group 3 experimental coho salmon on nine weighing dates in 1971 ......................... Analysis of variance to test regression line linearity of mean total length vs time for group 4 control coho salmon on nine weighing dates in 1971 . . . Food conversion of experimental and control coho salmon, February 12 - June 4, 1971. All weights expressed in kilograms ................ APPENDIX Chemical analysis of'water supply utilized for cultural operations during 1970-1971 ......... Analyses of Ewos salmon starter and grower feed (locke, 1969) .................... Total number, weight and carrying capacity of experimental and control coho salmon (artificially spawned on Novenber~3, 1970) on nine weighing dates in 1971 . . . ................. Total number, weight and carrying capacity of experimental and control coho salmon (artificially Spawned on November 19, 1970) on nine weighing dates in 1971 .................... Page 41 41 49 66 67 74 75 Figure LIST OF FIGURES Heath vertical-stack incubator and recirculation tank used for egg incubation in 1970 .......... Holding tanks and apparatus used in the rearing of experimental and control coho salmon in 1971 ...... Thonpson Creek showing release site of experinental and control coho salmon in 1970 and 1971 . . . ..... Growth of experimental and control coho salmon on six sampling dates in 1970. Comparison of mean total length is shown in A and group weight in B. Time is expressed as days from fertilization ........ Regression of mean total length fOr experimental and control coho salmon on six sampling days in 1970 . . Mean total length of experimental and control coho salmon on nine sampling dates in 1971. Length of salmon groups 1 and 2 artificially spawned on Nov- ember 3, 1970, is displayed in A. Total length of coho groups 3 and 4, spawned on November 19, 1970. is shown in B ...................... Mean weight of experimental and control coho salmon on nine sampling dates in 1971. Heights of coho groups 1 and 2, spawned on Novenber 3, 1970, are shown in A. Heights of coho groups 3 and 4, spawned on November 19, 1970, are shown in B .......... Regression of mean total length against time for four coho salmon groups reared in 1971. Groups 1 and 2 are experimental and control fish respectively, artificially spawned on Novenber'3, 1970. Groups 3 and 4 are experimental and controls respectively, spawned on November 19, 1970 .............. vi Page 11 14 22 23 33 35 39 LIST OF FIGURES-~Continued Figure 9. 10. 11. Density of experimental and control coho salmon in rectangular rearing tanks used in 1971. Pounds of group 1 experimental salmon per cubic foot of water, theoretical carrying capacity and tank water exchange rate (ER = number of changes per hour) is shown in A. Height of group 2 control salmon, carrying capacity and tank exchange rate is shown in B .......................... Density of experimental and control coho salmon in rectangular rearing tanks used in 1971. Pounds of group 3 experimental coho salmon per cubic foot, theoretical carrying capacity and tank water exchange rate (ER = number of changes per hour) is shown in A. Height of group 4 control salmon, carrying capacity and tank exchange rate is shown in B ..... Trap installed in Thompson Creek in July 1971 to monitor emigration of coho juveniles into Lake Michigan ........................ APPENDIX Mean total length of all experimental and control coho salmon groups on nine sampling dates in 1971. Experimental and control groups 1 and 2 artificially spawned on November 3, 1970. Experimental and control groups 3 and 4 artificially spawned on NoveMber 19, 1970.. . . ..................... Mean weight of all experimental and control coho salmon groups on nine sampling dates in 1971. Experimental and control groups 1 and 2 artificially spawned on November 3, 1970. Experimental and control groups 3 and 4 artificially spawned on Novenber l9 , 1970 ................... Salmonid carrying capacity at 12.8 - 15.0 C using water exchange rates and fish size (Nesters, 1970) . . . vii Page 45 47 52 69 71 73 INTRODUCTION The drastic reduction of major piscivore fish stocks in the Great Lakes following 1940 (Miller, 1958; Smith, 1968) and the extreme abundance of the planktivorous alewife (Alosa pseudoharengus) in the late 1950's (Tody and Tanner, 1966; Smith, 1970) were prevalent factors in the decision of the Michigan Department of Natural Resources to introduce the coho salmon (Oncorhynchus kisutch) into Lake Midiigan in 1966. The major obstacle barring successful introduction appeared to be the acclimation of a true anadromous species to a fresh water environment. Prior introductions of chinook salmon into the Great Lakes in 1873 and 1921 (Borgeson, 1970a) net with failure as natural spawning populations did not emerge. Only one previous introduction of coho salmon into Great Lakes' waters has been recorded (Trautman, 1935) when Ohio planted chinook and coho salmon into Lake Erie in 1933. Of subsequent catches reported, only one chinook salmon was positively identified and successful reproduction did not materialize. Atlantic salmon (Salmo salar), however, existed in great nunbers in Lake Ontario before declining in the late 1800's because of overfishing and unfavor- able ecological changes in spawning streams (Huntsman, 1944). Pink salmon (Oncorhynchus gozbuscha) planted in Lake Superior in 1956 apparently made the necessary physiological adaptation to fresh water, as spawning populations have successfully perpetuated the species for several generations (Schumacher and Eddy, 1960; MacKay, 1963). 2 Steelhead trout (Salmo gairdneri), like coho, have a similar fresh- water phase in their life cycle, but often spend from two to three years in the stream before enfigration occurs (Wagner et a1., 1963; Withler, 1966). In West Coast areas, where juvenile steelhead and coho salmon typically frequent the same streams, competitive interaction is mini- mfized by utilization of slightly different habitats (Hartman, 1965). It would thus appear from the similar ecological requirements exhibited by steelhead and coho elsewhere, that Michigan streams now sustaining spawn- ing runs of steelhead would be equally suitable for coho salmon. Adult coho salmon resemble the steelhead trout in morphology and reproductive requirements, often spawning in small tributary streams from late Septenber through Decenber, depending upon latitude. The fenale usually deposits around 2700 eggs (Neave, 1948) in a depression sc00ped out of gravel. Young alevins hatch in early spring and generally remain in fresh water fer one year before migrating to the sea the following spring. Scale checks have revealed that at least a few fry ndgrate to sea in the spring following hatching (Fraser, 1917), possibly being displaced by density-dependent intra or interspecific competitive interactions (Chapman, 1962). Age at maturity of North American coho stocks is apparently correlated with latitude. Precocious males (known as jacks or grilse) approaching age 11 return in significant nunbers to the parent stream after one summer at sea in California waters (Shapovalov and Taft, 1954). Certain Alaskan stocks consist mainly of four-year-old fish at maturity, with some reaching six years of age (Drucker, 1972). Michigan's first coho plantings were of the Columbia River stock (Tody and Tanner, 1966), where a three-year life cycle is predominant. 3 It has been demonstrated from Montana introductions that coho salmon of sea-run parentage could be artificially spawned with fair success after conpleting their life cycle in fresh water (Beal, 1955). West (1965) recorded similar observations at the Mt. Shasta Hatchery in California, where coho salmon were maintained for three generations in fresh water and indicated that through a selective breeding program a fresh water strain could probably be developed for stocking inland waters. Such a selective breeding program was initiated at Michigan State University in the fall of 1969, with the ultimate goal being the development of a coho salmon bred exclusively to thrive in the fresh water environment characteristic of Michigan's Great Lakes and tributary streams. It was hypothesized that if rate of growth and age at maturity were sex-related characteristics dependent upon the male fish, substantial progress toward that goal could be achieved through utilization of'early-maturing jacks or grilse for fertilization purposes. If successful, such mating would achieve a reduction in generation time from three years to two years and a resultant acceleration of the selection process over a period of time. Specific objectives of this study were the evaluation of fertilization results, incubation mortality, rearing mortality, growth and adult returns of jack-fertilized and normally fertilized coho salmon progeny. METHODS AND MATERIALS The study was conducted over a duration of two years, beginning in fall, 1969. Certain changes in cultural procedure were implemented as necessary and/or by acquisition of new equipment. Eggfgollection and Incubation 1969-1970 The project was initiated on October 31,1969, with collection of eggs at the Platte River State Fish Hatchery near Honor, Michigan. Ova were removed from six mature females by incision of the abdominal cavity. Eggs from three females were each fertilized with nfilt from two jack coho salmon for experimental purposes. Scales were then taken from all grilse utilized for artifical spawning to ascertain correct age. Eggs of the three remaining females were each fertilized with milt from two adult (age 111) males to provide control fish. A11 eggs were placed in separate quart containers, labeled and transported to laboratory facilities in the Natural Resources Building, Michigan State University (MSU). A sinfllar procedure was used for eggs taken on November 3, 1970 for the second phase of cultural operations. Ova from two females were removed and fertilized to serve as controls. Eggs for experimental fish, i.e., jack-fertilized eggs, were obtained from four female coho salmon. Again, two grilse were used to provide milt for each group. Collection of additional eggs on November 19 was necessitated by substantial mortalities encountered in the incubation of eggs taken on November 3, 1970. Incubation of all eggs taken in 1969 and 1970 was accomplished in trays of a Heath vertical-stack incubator (Figure l). Filtered Figure 1. Heath vertical-stack incubator and recirculation tank tank used for egg incubation in 1970. East Lansing city water maintained at approximately 10 C was used throughout incubation in 1969. Chemical analyses of selected water quality parameters are provided in Appendix Table A. Eggs taken in 1970 were incubated in a Heath unit mounted on a 560-liter stock tank. Partial water recirculation was acconplished for maintenance of desired tenperature. Incubation was initiated at 14 C, but lowered to 10.5 C on Decenber 2, 1970 with the addition of a Cooling unit. Adequate aeration was achieved by submersible punps placed in the recirculation tank. Eggs from each female were maintained in separate trays of the incubator. After all eggs had reached the eyed stage, dead eggs and/or fry were removed every buo days by hand picking to determine percent fertilization and conpare mortalities between control and experimental groups. Ten percent glacial acetic acid was used to clear all eggs picked until internal development was evident. Cultural Methods 1970 ‘ Experinental fry hatched from eggs taken in Noverrber, 1969 were placed in a 380-liter stock tank upon reaching the swim-up stage and control fry were held in a 560-1iter oval stock tank. Oxygenation of water was acconplished by injection of freshwater into the tanks under pressure. Water tenperature was held at approximately 12 C. Larger capacity holding facilities became available May 1, 1970, and all salmon were transferred to the Fishery Research Building on the MSU campus. Control and experimental fish were held in two rectangular fiberglass tanks, 2.13 x 0.61 x 0.43 meters. Water depth was maintained at 0.38 meters. Surplus control salmon were held in bro insulated stock tanks, each having a 560-1iter capacity. Filtered tap water was supplied to all tanks by gravity flow through a head tank with surface agitator. 8 Flow to each tank was approximately 1.9 liters (0.5 gallons) per minute. Constant photoperiod was maintained through the study using over— head flourescent light augmented by natural light through existing windows during daylight hours. Growth rate of salmon appears to be enhanced by exposure to constant photoperiod (Eisler, 1957). Food and Feeding Lack of adequate cold storage facilities for the Oregon moist pellet (0MP), used predominantly in West Coast hatcheries, necessitated the use of a dry food that could be stored without cooling. Availability of the Ewos dry diet for Atlantic salmon developed by Aktiebolaget Ewos of Sweden (Locke and Linscott, 1969) prompted its use throughout the duration of the study. An analysis of food composition and particle size is presented in Table B of the Appendix. Feeding of fry was initiated with starter-sized particles five to six times a day for the first two weeks to induce the feeding response. Stbsequent feeding of excess rations were inplemented three times daily. Exoess rations for this study are defined as all food the salmon would consume in a period of 15 to 20 minutes. Progressively larger-sized food grades - through # 3 - were utilized as warranted by growth. Samling and Marking Procedure Sanples of weight and length of 25 experimental (jack-fertilized) and control salmon were recorded at 30-day intervals. Fish from each group were selected by herding all fish to one end of the tank and then netting the approximate nunber from the concentrated group. Sanple fish were anesthesized with MS 222 (tricaine methanesulfonate) and total length measured to the nearest millimeter when mobility was sufficiently inpaired. All fish sanpled were weighed as a group through May 30. 0n the final sampling date (June 17, 1970) individual total lengths and weights of all remaining experimental fish and 25 control salmon were taken. Excess water'was blotted off each fish with tissue paper and weighing accomplished to the nearest hundredth gram on a Mettler 1200 scale. All salmon were marked for planting by removal of the adipose and right pelvic fins (marks were assigned by the Great Lakes Fishery Commission). Experimental coho salmon were distinguished from control fish by a partial clip of the dorsal fin, extending approximately one half the length anteriorly from the posterior base of the fin. Cultural Methods 1971 Coho salmon fry hatched from eggs taken in the fall of 1970 were concentrated in 189-liter oval fiberglass tanks for approximately two weeks for feeding initiation and then transferred to large rectangular tanks identical to those used in 1970 (Figure 2). All fry were maintained separately by experimental and control groups and by egg collection date in four holding tanks. Fry hatched from eggs taken on Novenber 3 were designated as group 1 experimental and group 2 controls. Experi- mental and control fry from the November 19 egg collection were designated as group 3 and 4 respectively. Because of space limitations, no attempt was made to separate fry by individual female. Water temperatures were maintained at approximately 15 C for the duration of the holding period with the aid of three submersible heaters (3500 watts total) mounted in a plexiglass head tank. Aeration of influent water was accomplished by a surface agitator in the head-tank system and by submersible pumps placed in each holding tank to provide constant surface aeration. 10 Figure 2. Holding tanks and apparatus used in the rearing of experimental and control coho salmon in 1971. 12 Food and Feeding Food consisted of the Ewos F-48 salmon starter and the larger particle size (F-49 grower feed) used in 1970. Frequency of feeding was as previously described. To obtain conversion estimates, food utilization was recorded when preweighed (500 to 1000 gram) units of food were exhausted in daily feedings of excess rations. Gross food conversion efficiency was calculated by dividing weight gain for the sampling period (February 12 - June 4, 1971) by weight of food fed (Brown, 1957). No attempt was made to estimate or to incorporate in conversion calculations fead which had settled on the tank bottoms. Samplingjand Marking Procedure Experimental and control salmon were weighed and measured bi-weekly from February 12 through June 4, 1971. Fifty fish were sampled from each tank with individual total length and weight recorded. Juvenile experimental and control salmon were marked for release with finclips identical to those used in 1970. StockingfiLocation The original stocking location was scheduled to be a small tributary stream of Lake Huron. Stream suitability, however; was dependent upon size, location, facilities for capturing returning adults and authoriza- tion from the Michigan Department of Natural Resources. Thomson Creek, below the existing Thompson State Fish Hatchery, six miles west of Manistique, Michigan (T41N, R16W, sec 29), was ultimately designated as the stocking location. The hatchery facility, located at the confluence of Williams and Thompson Creeks, lies about one half-mile from where the stream enters Lake Michigan (Figure 3). 13 Figure 3. Thompson Creek showing release site of experimental and control coho salmon in 1970 and 1971. 15 Downstream from the hatchery Thompson Creek consists of a series of shallow gravel riffles, sandy runs and pools. Streanlwidth ranges from three to five meters with an average depth of approximately 0.3 meters. Stream discharge was measured at 10.3 cubic feet per second (CFS). Plantings of an Alaskan strain of coho salmon were made in Thomson Creek in May 1967 (Borgeson, 1970b) with additional plantings of Oregon coho strains in spring 1968. Annual stocking of salmon smolts in the spring provides the creek with a transient fish population, some of which remain through the summer and emigrate the following spring. Along with some naturally reproduced juvenile coho salmon and rainbow (steelhead) trout, resident populatiOns of brook and brown trout are present. Results and Discussion 1969-1970 Incubation Mortality, A total of 15,216 jack-fertilized (experimental) eggs and 5,826 control eggs fertilized with milt from normal three-year old male coho salmon were taken at the Platte River State Hatchery on October 31, 1969. Mortality of experimental eggs throughout the duration of incuba- tion was 7,551 or a total of 49.6 percent (Table l). The number of control eggs removed from trays during incubation was substantially lower, with 1,370 (23.5 percent) recorded. Mortality totals for both groups in 1969 included unfertilized eggs as well as ova in which inter- nal development was evident. The high mortality of experimental eggs during incubation in 1969 would apparently indicate reduction in jack fertility and/or aberration in enbryonic development. Because fertilized and unfertilized eggs were not differentiated in mortality records maintained, it was not possible to demonstrate any difference in fertilization rates between control and experimental eggs. Thus, high incidence of mortality recorded for experimental eggs may in fact have been simply low percentage of fertilization. Average fecundity reported for some West coast coho stocks (Foerster, 1944; Hunter, 1948; Neave, 1948; Wickett, 1951; Salo and Bayliff, 1958) was less than 3,000 eggs per female. Fecundity of Michigan coho also appears to be somewhat less than 3,000 per female. From these averages it would appear that 15,246 experimental eggs taken from three females to be quite high and 5,826 control eggs from three fish to be low. Possible error in labeling egg containers 16 17 Table 1. Mortality of eggs during incubation, December, 1969. Mortality totals include unfertilized eggs. Tray no. Fish group Total number Mortality Percent of eggsa of total 0 Experimental - 3187 - 1 Experimental - 1777 - 5 Experimental - 2587 - 15.216 7551 49.6 2 Control - 532 - 3 Control - 507 - 4 Control - 331 - 5,826 1370 23.5 aNunber of eggs per female not recorded 18 and/or placement in trays prior to this investigators assumption of the project nay have influenced the disparity in mortality rates exhibited by control and experimental eggs in 1969. Rearing_Mortality Fry began emerging from control and experimental eggs on December 10 with hatching completed by December 16, 1969, 46 days from date of fertilization. High mortality of both groups was experienced from the post hatching period through 105 days (Table 2). On February 4. 1970 a majority of experimental fry were lost when the aeration system failed. At this time 353 experimental fry remained. Total mortality of control fry, expressed as percent of total, was ten percent higher for the first 50 days following hatching. m Samples for weight and length were taken from experimental and control fry on February 4, 1970, subsequent to the mortality occurring on that date. Experimental fry were slightly longer than control salmon at 97 days (Table 3). With the exception of the succeeding sampling date (122 days from fertilization) experimental fry remaining continued to exhibit greater weight and length on all sampling dates (Figure 4). Regression of total length versus time plotted for experimental and control coho salmon reared in 1970 (Figure 5) revealed little difference in slope or rate of growth for the two groups. Comparison of final mean weight and length measurements taken on June 17, 1970 (Student t-test) indicated that no significant difference existed between control and experimental fish at the 0.05 level. 19 Table 2. Mortality of experimental and control coho salmon from post-hatching period through final weighing, Decenber l6 - June 17, 1970. Experimental Coho a Control Coho P Date Days from Mortality Percent Mortality Percent Interval Fertilization of Total of Total Dec. 16-25 46-55 922 12.03 861 19.32 Dec. 26-Jan. 4 56-65 43 0.56 85 1.91 Jan. 5-14 66-75 550 7.16 343 7.70 Jan. 15-24 76-85 222 2.90 126 2.83 Jan. 25-Feb. 3 86-95 136 1.77 160 3.59 Feb. 4-13 96-105 5439 C 70.96 262 5.88 Feb. 14-23 106-115 7 roz m_mp q.¢~ owe m.F~ Noam wmuu —apcmswcmaxm 5 mp L$55262 mmmm e.op «mm m.mm mpmm eumm paucmswcmaxm N ap L0nem>oz «app m.¢m mmmp ¢.Fm comm moon —ocucou o m gunsm>oz mowp m.m~ on“ e.mm comm Nuym Pmacmsvgmaxu m m cunem>oz as 0.2 5. EN «.3 38 32:33 a m 3532 New ~.mm mmm o.P~ was, mem poucmewgmaxm m m L89.532 3mm mama .6 9.232.. 3:33: 33:29. @3253“. c3353“. 3:52 96.5 .3252 332—8 $952 :83“. we 333...: 233; $95: :38. 5: >2.— mmmm 38 .2232. o..— g: c9059. 83 325.93.? :n 832; £33 .3333: .23.: 32.8 $333.: .3 .22 £22.30: 5 :33 8?... 55a... 2.8 3.580 Ea $22233 mo 3:33... new 853233". .e 22:. 27 Inc1bation Mortality Experimental and control ova taken on Noverrber 3 exhibited the highest mortality up to hatching, ranging from 28.3 to 70.0 percent (Table 4). Mortality of ova collected on Novenber 19 was appreciably lower, ranging from 10.4 to 28.9 percent for the two coho egg groups. The high loss of eggs taken on Novenber 3, 1970 is attributed to three factors: failure of the smmersible pump circuit for an unknown nunber of hours on Novenber 4, 1970; the initial incwation tenperature of 14 C maintained through Novenber 30; and mechanical injury througi handling. While dissolved oxygen levels undowtedly dropped to low levels during loss of water circulation on Novenber 4, greatest cbtrimental effect on egg development was probably temperature increase. Alderdice et .31., (1958) demonstrated that chum salmon (o. keta) eggs can withstand oxygen levels of less than 1 mg/l at early incubation (4 degree days). Allen (1958) thougit the high mortality of coho eggs prior to hatching to be due in part to incubation at temperatures higher than optimum for the species (11.4 C). Egg mortality for sockeye salmon was significantly himer at 14.2 C and above, when compared to lower temperatures (Conbs, 1965). And Atlantic salmon eggs incubated at 12.2 C experienced high losses while eggs at 10 C hatched with no difficulty (Markus, 1962). Reaflngiortali ty Mortalities of group 1 experimental and group 2 control coho salmon from hatching through final weigiing on June 4, 1971 are shown by lO-day intervals in Table 5. Expressed as a percentage of total, experimental fry exhibited higher mortality for the first 30 days following hatching. Greatest mortality (percent of total) 28 Table 5. Mortality of experinental and control coho salmon fry from hatching through final weigiing, June 4, 1971. Coho groups 1 and 2 artificially spawned on Novenber 3, 1970. Experinental 13 Control 2b Date Days from Mortality Percent Mortality Percent Interval Fertilization of Total of Total Dec. 7-16 32-41 529 18.4 160 14.0 Dec. 17-26 42-51 115 4.0 28 2.5 Dec. 27-Jan. 52-61 77 2.7 22 1.9 Jan. 6-15 62-71 55 1.9 56 4.9 Jan. 16-25 72-81 123 ‘4.3 27 2.4 Jan. 27-Feb. 82-91 70 2.4 18 1.6 Feb. 5-14 92-101 85 3.0 56 4.9 Feb. 15-24 102-111 10 -< 1.0 128 11.2 Feb. 25-Mar. 112-121 10 107 9.4 Mar. 7-16 122-131 7 15 1.3 Mar. 17-26 132-141 4 10 0.9 Mar. 27-Apr. 142-151 12 15 1.3 Apr. 6-15 152-161 7 7 <1 .0 Apr. 16-25 162-171 3 2 Apr. 26-May 5 172-181 5 1 May 6-15 182-191 7 2 May 16-25 192-201 3 0 May 26-Jun. 4 202—211 4 3 Total Mortality 1126 39.2 657 57.5 Nunber Surviving 1745 60.8 485 42.5 aPost-hatching survival = 2871 bPost-hatdiing survival = 1142 29 between the two coho groups, however, was experienced by control fish. Mortalities of experimental fish after 101 days from fertilization were less than one percent. Control salmon continued to experience mortalities for 30 days beyond this time (131 days from fertilization). Immediately fellowing hatching, group 4 control salmon exhibited higher frequency of mortality (expressed as percentage of total) than group 3 experimental fish by lO-day intervals through final weighing on June 4, 1971 (Table 6). Total mortality for group 3 experimental and group 4 control coho salmon fer the rearing period was 22.7 and 43.5 percent respectively. Rearing mortalities experienced by fry fbllowing hatching in 1970 and 1971 were higher for all control coho salmon groups. It is possible the phenomenon of heterosis normally displayed by progeny of inter and intra-specific breeding is operative in the crossbreeding of two age groups. Clearly thougi, comparison of a much larger sample of experimental and control fish would be necessary befbre any infer- ences could be made from such a cross. 30 Table 6. Mortality of experimental and control coho salmon fry from hatching through final weighing, June 4, 1971. Coho groups 3 and 4 artificially spawned on November 19, 1970. Experimental Coho 3a Control Coho 4b Date Days from Mortality Percent Mortality Percent Interval Fertilization of Total of Total Dec. 24—Jan. 2 34—43 191 4.9 196 10.9 Jan. 3-12 44-53 103 2.6 89 5.0 Jan. 13-22 54—63 73 1.9 69 3.8 Jan. 23-Feb. 1 64-73 239 6.1 42 2.3 Feb. 2-11 74-83 155 3.9 107 6.0 Feb. 12-21 84-93 32 <1.0 83 4.6 Feb. 22-Mar. 3 94-103 18 103 5.7 Mar. 4-13 104-113 12 25 1.4 Mar. 14-23 114-123 14 60 C '<1.0 Mar. 24-Apr. 2 124-133 8 2 Apr. 3-12 134-143 18 2 Apr. 13-22 144-153 12 0 Apr. 23—May 2 154-163 14 1 May 3-12 164-173 16 3 May 13-22 174-183 4 0 May 23-Jun. 4 184-193 8 0 Total Mortality 917 22.7 782 43.5 Nunber Surviving 3121 77.3 1015 56.5 a Post-hatching survival 4038 b Post-hatching survival 1797 c 52 control salmon which perished when maintenance workers overloaded submersible pump circuit were not included in percent of total 31 93.11611. Group 1 experimental coho salmon were larger than group 2 control salmon on all nine sampling dates in 1971 (Figures 6 A and 7 A). On six of the first seven sampling dates group 4 control salmon‘were larger than group 3 befbre being surpassed in size by the experimental fish on the last two sampling dates (Figures 6 B and 7 B). Mean total length and weight for respective experimental and control fish are provided by date in Tables 7 and 8 (Graphical comparison of mean total length and weight of all coho groups together is depicted in Appendix Figures A'and B). Probability of a linear growth phase was indicated by the plotted mean total length of the four coho groups. Subsequently, regression of mean total length with time was examined. Regression of total length over time equations for experimental and control coho groups 1 to 4 are displayed in Figure 8. Analysis of variance tests performed fer line linearity and deviation are given in Tables 9 - 12. To test the basic hypothesis of no difference in total length between fish groups, the pooled Student t-test (p = 0.05) was used to compare the overall mean of respective regression lines for salmon of similar age (i.e., group 1 vs 2 and group 3 vs 4). For salmon spawned on November 3, 1970, experimental coho were fbund to be significantly longer and therefore larger than group 2 control fish. Group 4 coho. spawned on Novenber 19, 1970, however, were significantly longer than group 3 experimental fish over time (nine sampling dates). On April 9, 688 group 1 and 1513 group 3 experimental coho were removed from their respective tanks and placed in separate holding tanks. Total length of both experinental groups increased after density 32 Figure 6. Mean total length of experimental and control coho salmon on nine sampling dates in 1971. Length of salmon groups 1 and 2 artificially spawned on November 3, 1970, is displayed in A. Total length of coho groups 3 and 4, spawned on Novenber 19, 1970, is shown in B. (mm) lENG‘I'H MEAN TOTAL 33 110 r“ ,+ ,/ ,0. A + EXPERIMENTAL l ..-A 9' OwnoL 2 ./jz’ 2/,- 90 / .0 -'7f-//’/A AA” .K/’ 10 ,,/‘fi/ 2t , . "I l// 60 ii! //A I'/ .l/I / ,A/ so ,4 x’ / ‘K/5 40 A/ 30 1 l 1 L L L 1 1 J no 3’ .p anunmnu. 3 100 A (hoax 4 .Al 4‘7 90 no ,.-" ,/K/' 70 / .+ K .ij' so 2””? so %/ 40 921/ so I I I I I I I I I I2 26 I2 26 9 23 7 21 4 nonunv MARCH APRIL MAY JUNE 34 Figure 7. Mean weight of experimental and control coho salmon on nine sampling dates in 1971. Weights of coho groups 1 and 2, spawned on November 3, 1970, are shown in A. Weights of coho groups 3 and 4, spawned on November 19, 1970, are shown in B. (gramS) HEIGHT MEAN c—l-IJ—ul O—IN Nwhmciwoom 11 10 thTOleQ N 35 + EXPERINENTAL l A CONTROL 2 + EXPERIMENTAL 3 A CONTROL 4 12 26 12 26 FEBRUARY MARCH 23 APRIL MAY 21 JUNE a; A mm 8.0 A 8 mm; A a 3.0 A 82 e 2% m: A a £5 A 8 8.. A 3 25 A 3 Z a: :3 A em :5 A em 8.. A em 35 A 8 A a: R; A R 35 A K S; A me 35 A mm 8 7:3 8.3mm EgAS SAA K REA? 8 7:2 3; A mm $5 A 3.. 21A No 8.0 A 8 mm 6:: Rd A mm 85 A em a; A 3 £5 A G 2 5.32 2.0 A 3 :5 A 3 was A e. :3 A 8 8... $28.. 8.0 A mm :5 A R 8.0 A z. 85 A S N. 528“. A¢VFogA=ou Amvfimucmewgmaxm Amy—ogucou APV—mucmewgoaxm mcwwwmmm :2 .2 $9.83: :2 .m 59.83: "38 eavpa~eprutaa .anp c? mmpau mcvcmwmz one: co cospom ogou pocucou ecu Faucmswgmaxo we cogcm usmvcaum umquaupmo can memAmEvppce :_ sumcmp papa» cum: .5 mFaMF 37 00.0H¢N.0 00.0H 05.0 0¢.0H 0V0 _.0.0H0¢._._. v 9.50 «0.0 H 00.x 2.0 H 00.5 00.0 H 0V0 «0.0 H 50.0 _.N >6: 00.0H00.0 S..0H 00.0 00.0H 050 2.0H FOK 5 >6: 2.0 H 0&6 :30 H 0¢.0 0N0 H N56 2.0 H 004 00 _._La< 2.0 H 00.0 N_..0 H 00.0 0_..0 H 00.0 :30 H 00.0 mm _._La< _._..0 H 00... 50.0 H 00... 320 H 00.0 2.0 H 3.0 00 no.3: No.0 H 00... .00 H N0.~ 2.0 H 0¢._. no.0 H _.0.N 0.. :0ng «0.0 H 00.0 #00 H 05.0 00.0 H s50 No.0 H 0_.._. 00 bwagnmn. 00.0 H 3.0 _.0.0 H 00.0 00.0 H N0.0 00.0 H 00.0 N_. ago—Ina“. Aeqfiocucou AMWHmAuuaHuumxm Afiupogucou a_vmunquawuummm1| memo mcvpaeam K2 .2 Lm€w>oz K2. .0 sm€m>oz "mama =o_pa~.2.ecaa .Fump a. macaw acegaea; o=.= co coepam ozou Focucou uca Paucaevcmaxw mo Lasso ugmucmum uaampaupmu ecu maacm cw Agave: cam: . 0 oz: Figure 8. 38 Regression of mean total length against time for four coho salmon groups reared in 1971. Groups 1 and 2 are experimental and control fish respectively, artificially spawned on Novenber 3, 1970. Groups 3 and 4 are experimental and controls respectively, spawned on Novenber l9, l970. N oz< F $0010 u =0Hh<-4~hmmm roam m><0 ("I") HLSNZ-l‘l 5 2m «.2 a: E E m: g m: 2: 2w 2 —«4.««iafluidui4—..4444daiudld—Juwuqnqmu«qu~+u«quad—“uuququuuuud “a .«u—u«.l4..i..— .. 8 L 8 1 8 1 8 1 2 1 8 :mmm.o.mm:.2- a» a .5528 O. x ~_mm.o+§m.2- u> m dhfifiunom D L 8 :5.c+32.2- u> N 4258 < \.... x 83.3...«32- u» H #555 + 1 8. P r P k b r b p _ b b. O: 5 8. a: a: S. 5 m: a 3 ¢ 92 0 anoxwuzgzddhxmm :9: 0:3 W34 'IVlOl 40 Table 9. Analysis of variance to test regression line linearity of mean total length vs time for group 1 experimental coho salmon on nine weighing dates in 1971. Source Sum of Squares D.F. Mean Squares F Groups 248415.38 8 31051.92 584.16 Linear Regression 247508.28 l 247508.28 1909.99 ** Deviation 907.10 7 129.59 2.44 N.S. Error 34073.54 641 53.16 Total 282488.92 649 ** Significantly linear at p = 0.01 N.S. Deviation from regression line not significant at p = 0.01 Table 10. Analysis of variance to test regression line linearity of mean total length vs time fer group 2 control coho salmon on nine weighing dates in 1971. Source Sum of Squares D.F. Mean Squares F Groups 130693.75 8 18670.54 209.07 Linear Regression 130036.01 1 130036.01 1186.21 ** Deviation 657.73 7 109.62 1.23 N.S. Error 33203.00 441 89.30 TOtal 165700.47 449 ** Significantly linear at p = 0.01 N.S. Deviation from regression line not significant at p = 0.01 41 Table 11. Analysis of variance to test regression line linearity of mean total length vs time for group 3 experimental coho salmon on nine weighing dates in 1971. Source Sum of Squares D.F. Mean Squares F Groups 254375.77 8 31796.97 625.86 Linear Regression 251223.01 l 251223.01 557.78 * Deviation 3152.76 7 450.39 8.84 ** Error 32568.48 639 50.97 Total 286944.25 647 * Significantly linear at p = 0.01 ** Deviation from regression line significant at p = 0.01 Table 12. Analysis of variance to test regression line linearity of mean total length vs time fOr group 4 control coho salmon on nine weighing dates in 1971. Source Sum of Squares D.F. Mean Squares F Groups 169200.00 8 21150.00 345.86 Linear Regression 168750.00 1 168750.00 2625.00 ** Deviation 450.00 7 64.29 1.05 N.S. Error 26967.88 441 61.15 Total 196167.88 449 .** Significantly linear at p = 0.01 N.S. Deviation from regression line not significant at p = 0.01 42 reduction when contrasted with the plotted mean total length of control fish. This increase in growth was apparently compensatory for any lag in growth created by less than optimal conditions under greater fish densities. This density-related lag in growth by group 3 experimental salmon explains the deviation from regression line linearity noted in Table 5. “esters (1966) experienced a similiar resunption in growth upon thinning coho juveniles after critical density levels had been reached. Brown (1957) found growth of brown trout (Salmo trutta) to be greatest at intermediate fish densities. Overcrowded trout ate less and used food less efficiently than fish not in crowded conditions. Trout with much space grew erratically, as it seened a certain amount of "soci a1" stimulation was conducive to rapid growth. Egg size has been shown to determine initial size and subsequent growth of juveniles up to 3 months (Bilton, 1970; Fowler. 1972). As egg size was not determined, initial differences in size of fry may have been maternally dependent. with subsequent differences a function of differing specific growth rates. With increasing fish density availability of dissolved oxygen and the accumulation of excretory products become critical factors influencing growth. It has been demonstrated that efficiency of food conversion changes little with reduction in oxygen concentrations from 8.3 to 5.0 mg/l. Rate of food consumtion and growth, however. «heline markedly at the lower oxygen levels (Herrmann at .31., 1962). Dissolved oxygen was generally maintained near 6 mg/l in experinental tanks and at somewhat higher levels in less crowded control tanks. The principal excretory product. anmonia (reported as nitrogen) was 43 found at levels at high as 0.55 mg/l in experinental tanks prior to density reduction on April 9, 1971. Toxicity of anmonia is dependent upon quanity of the un-ionized form (Burrows, 1964) and is a function of tenperature and pH (Downing and Merkens. 1955). At a pH of about 8.1, encomtered during rearing operations in 1971, the calculated quanity of un-ionized ammonia (Spotte, 1970) present would be approxi- mately 0.029 mg/l. While not lethal, amonia concentrations at such levels are undoubtedly detrimental to growth. The natural variability of behavori a1 and morphological diaracter- istics within populations and between individuals of a species is quite evident. It is the desirable genetic variations which are exploited in the selective breeding process. and growth is one sudi parameter subject to natural fluctuations. The primary purpose of growth comparison between experimental and control coho salmon groups was to attribute any differences found to parental influence. The conflicting objectives of achieving rapid growth and production of maximum nunbers of experimental salmon. however, did not facilitate a valid corrparison of growth because of the resulting descrepancy in fish nunbers and density between control and experimental tanks. Thus. from the limited sample of the existing gene pool (Platte River "strain") differences in growth displayed may well have been due to the inherent variability within individuals from which eggs and/or milt were obtained, egg size or density dependent relations of oxygen and accumulation of excretory products. 44 Carrying Capacity The increase in fish density (pounds/cubic foot) on successive sampling dates was markedly higher for experimental coho salmon groups 1 and 2 (Figures 9 - 10). Hypothetical carrying capacity (Appendix Figure C) was exceeded prior to and on February 26 by experimental fish groups 3 and 1 respectively.‘ For control salmon groups 4 and 2, carrying capacity was not exceeded before March 26 and April 23, 1971. Actual number of fish, weight and carrying capacity by sampling date is shown in Appendix Tables C and D. The relationships of temperature, dissolved oxygen levels, water exchange rates, feeding levels, fish size and metabolite production to salmcnid carrying capacity have been thoroughly discussed (Haskell, 1955; Burrows and Combs, 1968; Hilloughby, 1968; Westers, 1970; Liao, 1971). Crowding of fry has been shown to be advantageous in inducing feeding response (Westers, 1966). This was substantiated by observations during the present study. An upper limit or asymptote, however, dependent upon space and/or physiological requirements previously mentioned, is undoubtedly reached at some point in time -- beyond which rate of growth is impaired. 45 1.8 — 1.5 _. + EXPERIMENTAL 1 1 4 _ ——— CARRYING CAPACITY ER=0.30 1.2 - 1.o - .8 - .6 — § .4 — LL 3 .2 - 3 o r- a: 8‘3 I I I I I I I I I g. .. B o. 6 + 601mm 2 ——— CARRYING CAPACITY .4 1.— __..—-—-""".-‘- 2 r- 0 .— I I I I I I I I 12 25 12 25 9 23 7 21 4 FEBRUARY MARCH APRIL MAY Juve Figure 9. Density cf experimental and control coho salmon in rectangular rearing tanks used in 1971. Pounds of group 1 experimental salmon per cubic foot of'water, theoretical carrying capacity and tank water exchange rate (ER = number of changes per hour) is shown in A. Height of group 2 control salmon, carrying capacity and tank exchange rate is shown in B. Figure 10. 46 Density of experimental and control coho salmon in rectangular rearing tanks used in 1971. Pounds of group 3 experimental coho salmon per cubic foot, theoretical carrying capacity and tank water exchange rate (ER = nunber of changes per hour) is shown in A. Height of group 4 control salmon, carrying capacity and tank exchange rate is shown in B. 47 1.2 " * q LO 7 N“ CARRYING CAPACITY .3 — .6 — .4 — .2 — o ._ I I I , I I I J 12 26 12 26 9 23 7 2] 4 FEBRUARY MAR“ APRIL my JUVE 48 Food Conversion Conversion of fOOd by experimental and control salmon from February 12 through June 4, 1971 is displayed in Table 13. Experimental coho groups 1 and 3 had similar food conversions of 1.32 and 1.36 respectively. Conversions of 1.96 and 1.17 emibited by control groups 2 and 4, were the highest and lowest ratios for all coho groups. The efficiency of food conversion into fish flesh is apparently a function of fish size, tenperature, type of food and rate of growth (Brown, 1957). Certainly, other variables such as fish density influence conversion. Under experimental conditions Hesters (1966) obtained conversions of 0.92 to 1.50 for coho salmon at densities ranging from 5.1 to 6.8 pounds/cubic foot. Conversions ranging from 1.3 to 1.5 are considered normal for coho salmon reared in Michigan's salmonid production facilities (Robertson, per. comm.). 49 p.55: 52. 29¢ @323me a 2., mm; mm; mm; cowflgcoo 3.2 8.8 om.m 84m em“. too“. we 23m: «m5 31% «Te 2.3 58 25.83 8.0 3.8 84 no.8 $5.53 22 55 8e me? .3952 q 25a. $6 mp; 8.0 8; $533 camp Rmm 80 mm? 03.52 N; Enigma. 33:8 Ecmfigmaxm 3.55... ”3:27.33 3:95 For: w m N F "gawk K2 12 2.3532 :2 .m 39.262 "38 covumN0Fvugmu .Kmp .e «:3. .233on 5 ummmflnxm £53: :< 2 Paine“. .553 0:8 33:8 EB 322.283 mo c32m>=ou v8... :3 Sam.— 50 Plantingland Returns Experimental salmon and control coho salmon were finclipped on June lO-l4, 1971 and dipped in a solution of l:500 formalin for approximately l0 minutes. All salmon were transported to the Thompson State Fish Hatchery on successive trips. June 22-23, l97l. In transit mortality was less than 10 fish. The juvenile salmon were acclimated in hatchery rearing tanks for two weeks prior to their release into Thompson Creek on July 7, l97l. A total of about 4,800 experimental and about 1,450 control salmon were released. A trap net (Figure ll) was positioned across Thompson Creek, l00 meters above U.S. 2 to monitor rate of emigration into Lake Michigan. Five experimental and seven control salmon were recovered on July 8, 1971. The silvery appearance indicative of parr-smolt transformation was not evident on any of the captured fish. Surveil- ance of downstream migrants was discontinued after local vandals removed the trap lead net in the early morning hours of July 9, 1971. Posters describing finclips of planted fish were again placed by the stream mouth and public parking lot. Jacks from this plant were not reported by fishermen or recovered by hatchery personnel during the normal egg taking operations in fall, l97l. Adult salmon anticipated from the l970 plant were not reported.. Notification received from hatchery personnel in Decenber l97l indicated that many juvenile salmon were observed in Thompson Creek, below the hatchery. Samples taken on Decenber's. l97l verified that the fish were of the July l97l plant. Approximately 500-l000 salmon were estimated to be schooled immediately below the hatchery outfall. 51 Figure ll. Trap installed in Thompson Creek in July l97l to monitor emigration of coho juveniles into Lake Michigan. 53 Measurements of length and weight of 10 juvenile coho salmon indicated little growth occured since the July release. It is apparent from the capture of a few marked salmon downstream from the release site on July 8 that not all fish remained in Thompson Creek overwinter. These fish were in all probability the weaker individuals unable to withstand existing stream velocity. l‘ne degree of downstream displacement of other juveniles, however, was not deter- mined. Chapman (l962) attributed the aggressive territorial behavior of juvenile coho salmon to be an important factor in causing downstream nfigration. Undoubtedly, density-dependent interactions among planted and resident.salmonids contributed to any such movement into Lake Michigan. Failure of coho juveniles planted in Thompson Creek in early July to migrate was attributed to late time of release and general absence of smolting condition. The effect of proper release timing upon survival and eventual returns of artificially reared salmonids is critical. Fish released prior to completion of parr-smolt transformation probably remain in the stream where they are subject to operant mortality factors (Wagner, 1968). Once released, the low survival of hatchery reared fish has been well documented (Needham et a1, 1944; Schuck, l948; Wales, 1954 and Miller, l953, 1958). To minimize duration of tine juveniles spend in tributary streams prior to emigration, it would appear that release should be coordinated with onset of smolting. Juvenile steelhead in Michigan waters normally migrate in May and June (Stauffer. 1968). Peak emfigration periods for coho juveniles from California and Washington waters ranged from April 15 to June 1 (Shapovalov and Taft, 1954; Kabel and German, 1966 and 54 Salo and Bayliff, 1958). Coho salmon stocked in Mighigan tributary streams have generally been released in late March or April, depending upon stream conditions (Robertson, per. comm). And successful rate of returns resulting from releases since 1966 would seemingly indicate these months to be near the optimum release period. Size at tine of release is acknowledged to exert considerable influence on survival, returns and age at maturity. Large chinook salmon (o. tshawytscba) juveniles returned in greater nurrbers in comparison to releases of smaller fish (Warner et at, 1961). Ricker (l962) attributed the high ocean survival of Cultus Lake sockeye smolts to large size at time of seaward migration. And the faster growing hatchery-reared steelhead (i.e., largest fish) have exhibited greatest survival in the Alsea River, Oregon (Wagner et a1, 1963). As of the final sanpling on June 4, 1971 (the mean total length (i) of all four salmon groups was 100.5 nm (4 inches). Mean weight of all fish was approximately 10 graus. In California, Washington and British Colunbi a investigations the average fork length of wild coho smolts at time of emigration ranged from 104 to 127 run or 4.1 to 5.1 inches (Shapavalov and Taft, 1954; Salo and Bayliff, 1958 and Fraser, 1920). By converting these lengths to total length and number of fish per pound size at emigration would range from about 35.8 to 19.7 salmon per pound. For introduction into Michigan waters, the Oregon Fish Commission reconnended that coho juveniles be no less than lS/pound at time of release (Tody and Tanner, 1966). Assuming rate of growth remained constant from the period of June 4 until time of transport to Thompson Hatchery on June 22-23, 1971, the projected size for all salmon groups would be around 107 an (4.2 55 inches) or approximately 33/pound. This is considerably smaller than recommended size at release and/or size at which parr—smolt transformation occurs. As transformation to smolting condition is dependent primarily upon size (Johnson and Eales, 1970) it is assumed that experimental and control fish had not yet achieved this physiological condition at time of release in 1971, SUMMARY AND CONCLUSIONS The following results were obtained through utilization of precocious male (jack) coho salmon for fertilization of ova from age III females in the fall of 1969 and 1970: 1. No reduction in percent of egg fertilization was evident in records maintained in 1970er the use of jack coho milt in contrast with milt from age III males. Mortality of eggs prior to hatching was markedly higher for experimental (jack-spawned) eggs in 1969. In 1970, however, mortality of fertilized control eggs was greater. Post-hatching mortality of control coho was greater than that of experimental fishprior to February 4, 1970 at which time most experimental fish were lost through equipment failure. Rearing mortalities experienced by fry-juveniles in 1971 were smstantially higher for control fish. Experimental coho salmon reared in 1970 were slightly larger at time of release than control salmon reared at greater densities. In 1971, two experimental coho groups were somewhat larger than control salmon reared under less crowded conditions. Theoretical carrying capacities calculated for holding tanks in 1971 were exceeded in early rearing stages by experimental salmon. Gross food conversion of two experimental salmon groups from February 12 througi June 4, 1971 was 1.32 and 1.36. Control fish exhibited conversions of 1.96 and 1.17. 56 57 7. Returns of coho salmon released in Thonpson Creek in 1970 and 1971 have not been reported. The majority of salmon released in 1971 appeared to overwinter in the creek and thus returns could be anticipated through fall of 1973. From this investigation it can be seen that coho salmon can be reared from egg to near smolt size in 8 months througi utilization of heated water. Acceleration to time of smolting has similarly been achieved with Atlantic salmon in heated water (Markus, 1962) and coho salmon reared in brackish water (Garrison, 1965). If it is cbsirable to obtain parr-smolt transformation within one year rather than the normal 18-month incwation-rearing period, use of heated water may acconplish this objective. In large hatchery operations, however, it is dorbtful if such a practice would be economically feasible. Savings from reduction in rearing time would have to outweigi heating costs and tangible benefits to management must exist. To obtain smolting condition by optimum release time early maturing (i.e., early running) fish could be chosen for egg lots. Salo and Bayliff, (1958) demonstrated that taking eggs from early running adults produced corresponding early runs of jacks and adults in succeeding years. Results from the present study are inconclusive, however, in respect to advantages derived through utilization of early maturing males (coho jacks) for fertilization purposes. Slight differences in growth observed may well have resulted from differing environmental factors rather than parental influence. The rapid growth achieved 58 by rainbow trout and chinook salmon through selective breeding (Donaldson and Menasveta, 1961 and Donaldson, 1970) arrply demonstrates the potential of exploitation of genetic variability. Successful too, was the selection for early maturity in rainbow trout (Lewis, 1944 and Millenback, 1950), although in the latter study early maturity resulted in smaller sized fish. Similar to the objectives of the present study, Neave and Prichard (1947) selected for early maturation in coho salmon by using early maturing males (jacks) for fertilization purposes. The resulting adults, however, exhibited only the usual proportion of jacks in the run and the experiment was discontinued after one gener- ation. Donaldson (per. conm.) indicated that two-year male coho are mated with three-year-old females at the Bureau of Sport Fisheries Hatchery at Eagle Creek, Estacada, Oregon. This mating is reported to produce a net increase in the total population with a "fantastic" nunber of jacks returning at larger than average size. Donaldson further indicated that two-year-old chinook males bred to three- year-old precocious females increased rate of growth significantly. The sport fishery for Atlantic salmon in Ireland is reported to be based entirely on a grilse (both males and females) run achieved througi selective breeding (Robertson, per. com.). The successful escapement from the comercial fishery is predicated on smaller sized adults. The acclimation of an introduced or exotic species to a new environment normally occurs through genetic plasticity over many generations of natural selection. Existing data would seemingly indicate that if early maturity was desired in Lake Michigan coho stocks, it could be achieved through a selection for this trait over 59 many generations. Since the introduction of the coho in 1966, some degree of natural selection has obviously taken place and will continue as long as the coho persists in any abundance. A two-year generation time midit well accelerate the rate of the selection process but a resultant reduction in size at maturity would probably not be acceptable to management and the fishing prlic. LITERATURE CITED Alderdice, O. F., W. P. Wickett, and J. R. Brett. 1958. Sorre effects of tenoorary exposure to low dissolved oxygen levels on Pacific salmon eggs. J. Fish. Res. Bd. Canada 15: 229-250. Allen, 0. H. 1958. Survival through hatching of eggs from silver salmon (mcorhyndzus kisutch). Trans. Amer. Fish. Soc. 87: 207-219. Beal, Fred R. 1955. Silver salmon (Oncorhynchus kisutah) reproduction in Montana. Prog. Fish-Cult. 17: 79-81. Bilton, H. T. M. S. 1970. Maternal influences on the age of maturity of Skeena River sockeye salmon (moorhynchus nerka) . Fish. Res. Bd. Canada Tech. Rept. 167: 20 pp. Borgeson, D. P. 1970a. Log of the chinook. Michigan Natural Resources Magazine, Septenber-October Vol. 39(5): p 365. Borgeson, D. P. (ed.) 1970b. Coho salmon status report 1967-1968. Fish Management Rept. No. 3. Fish Division, Michigan Department of Natural Resources. Brown, M. E. 1957. Experinental studies of growth, pp. 361-400. In M. E. Brown (ed.). Physiology of Fishes, Vol. 1, Academic Press, Inc., New York. Burrows, R. E. 1964. Effects of accumulated excretory products on hatchery-reared salmonids. U. S. Dept. of the Interior, Bureau of Sport Fisheries and Wildlife. Res. Rept. 66, 12 pp. Burrovs, R. E., and Bobby D. Conbs. 1968. Controlled environments for salmon propagation. Prog. Fish-Cult. 30: 123-136. Chapman, 0. W. 1962. Aggressive behavior in juvenile salmon as a cause of emigration. J. Fish. Res. Bd. Canada 19: 1047-1080. Conbs, Bobby D. 1965. Effect of temperature on the development of salmon eggs. Prog. Fish-Cult. 27: 134-137. Donaldson, E. M., J. 0. Funk, F. C. Withler, and R. B. Morley. 1972. Fertilization of pink salmon (mcorhynchus gorbuscha) ova by spermatozoa from gonadotropin-injected juveniles. J. Fish. Res. Bd. Canada 29: 13-18 60 61 Donaldson, Lauren R., and Deb Menasveta 1961. Selective breeding of chinook salmon. Trans. Amer. Fish. Soc. 90: 160-164. Donaldson, Lauren R. 1970. Selective breeding of salmonid fishes. In: William J. McNeil (ed.) Marine Aquiculture. Oregon State University Press. Corvallis, Oregon: 65-74. Donaldson, Lauren R. 1972. Personal Communication. University of Washington, Seattle. Downing, K. M., and J. C. Merkens. 1955. The influence of dissolved oxygen concentration on the toxicity of un-ionized ammonia to rainbow trout (Salmo gairdherii Richardson). Ann. Appl. Biol. 43: 243-246. Drucker, Benson. 1972. Some life history characteristics of coho salmon of the Karluk River system, Kodiak Island, Alaska. Fishery Bulletin 70: 79-94. Eisler, R. 1957. Influence 0f light on the early growth of chinook salmon. Growth 21: 197-203. FOerster, R. E. 1944. Appendix IV. Report for 1943 of the Pacific Biological Station, Nanaimo, B. C. Annu. Rept. Fish. Res. Bd. Canada 1943: 22-26. Fowler, Laurie G. 1972. Growth and mortality of fingerling chinook salmon as affected by egg size. Prog. Fish-Cult. 34: 66-69. Fraser, C. McLean. 1917. On the life history of the coho. Contrib. Can. Biol. for 1915-1916: 39-46. Fraser, C. McLean. 1920. Growth rate in Pacific salmon. Trans. Royal Soc., Canada Ser. 3, Vol. 13, Sec. 5. pp. 163-226. Garrison, R. L. 1965. Coho smolts in ninety days. Prog. Fish-Cult. 27: 219-220. Hartman, G. F. 1965. The role of behavior in the ecology and interaction of underyearling coho salmon (ancorhgnchus kisutch) and steelhead trout (Salmo gairdheri). J. Fish. Res. Bd. Canada 22: 1035-1081. Haskell, David C. 1955. Weight of fish per cubic foot of water in hatchery troughs and ponds. Prog. Fish-Cult. 17: 117-118. Herrmann, R. B., C. E. Warren, and P. Doudoroff. 1962. Influence of omgen concentration on the growth of juvenile coho salmon. Trans. Amer. Fish. Soc. 91: 155-167. Hunter, J. G. 1948. Natural propagation of salmon in the central costal area of British Columbia. Fish. Res. Bd. Canada, Prog. Rept. Pacific Coast Station, 77: 105-106. 62 Hmtsman, A. G. 1944. Why did Ontario salmon disappear? Trans. Royal Soc. Canada, Sect. 5: 83-101. Johnston, C. E., and J. G. Eales. 1970. Influence of body size on silvering of Atlantic salmon (Salmo salar) at parr-smolt transformation. J. Fish. Res. Bd. Canada 27: 983-987. Kabel, C. S., and E. R. German. 1966. Some aspects of steelhead and silver salmon hatchery operation. Resources Agency of Calif., Dept. of Fish and Game (Unpublished). Lewis, R. C. 1944. Selective breeding of rainbow trout at Hot Creek Hatchery. Calif. Fish and Game 30: 95-97. Liao, P. B. 1971. Water requirenents of salmonids. Prog. Fish-Cult. 33: 210-224. Locke, D. O., and S. P. Linscott. 1969. A new diet fbr landlocked Atlantic salmon and lake trout. Prog. Fish-Cult. 31: 3-10. MacKay, w. E. 1963. Fishes of Ontario. The Bryant Press Ltd., Toronto, Canada, 300 pp. Markus, H. C. 1962. Hatchery-reared Atlantic salmon smolts in ten months. Prog. Fish-Cult. 24: 127-130. Millenbach, Clifford. 1950. Rainbow brood-stock selection and observations on its application to fishery management. Prog. Fish-Cult. 12: 151-152. Miller, Richard B. 1953. Conparati ve survival of wild and hatchery- reared cutthroat trout in a stream. Trans. Aner. Fish. Soc. 83: 120-130. Miller, Richard B. 1958. The role of competition in the mortality of hatchery trout. J. Fish. Res. Bd. Canada 15: 27-45. Miller, Robert R. 1957. Origin and dispersal of the alewife (Alosa pseudoharengus), and the gizzard Shad (Dorosoma cepedianum) in the Great Lakes. Trans. Amer. Fish. Soc. 86: 97-111. Neave, F. 1948. Fecundity and mortality in Pacific salmon. Trans. Royal Soc. Canada, Ser. 3. Sect. 5, 42: 97-105. Neave, F., and A. L. Prichard. 1947. 00 small salmon tend to produce small salmon? Canadian Fish-Cult. 2: 12-13. Neediam, Paul R., and D. W. Slater. 1944. Survival of hatchery-reared brown and rainbow trout as affected by wild trout populations. J. Wildlife Management 7: 18-23. 63 Ricker, W. E. 1962. Comparison of ocean growth and mortality of sockeye salmon during their last two years. J. Fish. Res. Bd. Canada 19: 531-560. Robertson, John M. 1973. Personal Communication. Fish Division, Michigan Department of Natural Resources. Salo, E. 0., and W. H. Bayliff. 1958. Artificial and natural production of silver salmon (Oncorhgnchus kisutch) at Minter Creek, Washington. Wash. Dept. Fish., Res. Bull. 4, 76pp. Schuck, H. O. 1948. Survival of hatchery trout in streams and possible methods of improving the quality of hatchery trout. Prog. Fish-Cult. 10: 3-14. Schumacher, R. E., and Samuel Eddy. 1960. The appearance of pink salmon (Oncorhgnchus gozbuscha Walbaum) in Lake Superior. Trans. Amer. Fish. Soc. 89: 371-373. Shapovalov, Leo, and Alan C. Taft. 1954. The life histories of the steelhead rainbow trout (Salmo gairdheri), and silver salmon, (Oncorhgnchus kisutch), with special reference to Waddell Creek. CalifOrnia, and recommendations regarding their management. Calif. Dept. of Fish and Game, Fish Bulletin No. 98, 375 pp. Smith, Stanford H. 1968. Species succession and fishery exploitation in the Great Lakes. J. Fish. Res. Bd. Canada 25: 667-693. Smith, Stanford H. 1970. Species interactions of the alewife in the Great Lakes. Trans. Amer. Fish. Soc. 99: 754-765. Spotte, Stephen H. 1970. Fish and Invertebrate Culture. Water Management in Closed Systems. Wiley-Interscience, a Division of John Wiley and Sons Inc. New York, 145 pp. Stauffer, T. M. 1968. Migration and growth of rainbow trout parr in Black River, Michigan. M. S. Thesis, Michigan State University. East Lansing, Michigan, 41 pp. (Unpublished). Tody, W. H., and H. A. Tanner. 1966. Coho salmon for the Great Lakes. Mich. Dept. of Conservation, Fish Management Rept. No. 1. Trautman, M. B. 1935. The attempted introduction of Pacific salmon in Lake Erie waters. Mich. Dept. of Conservation, Inst. Fish. Res. Rept. No. 298. 4 pp. Wagner, H. M., Richard L. Wallace, and Homer J. Campbell. 1963. The seaward migration and return of hatchery-reared steelhead trout (Salmo gairdheri Richardson), in the Alsea River, Oregon. Trans. Amer. Fish. Soc. 92: 202-210. 64 Wagner, H. H. 1968. Effect of stocking time on survival of steelhead trout (Salmo gairdneri), in Oregon. Trans. Amer. Fish. Soc. 97: 374-379. Wales, J. H. 1954. Relative survival of hatchery and wild trout. Prog. Fish-Cult. 16: 125-127. Warner, G. H., and D. H. Fry, Jr., and A. N. Culver. 1961. History of yearling king salmon marked and released at Nimbus Hatchery. Calif. Fish and Game 47: 343-355. West, Delbert A. 1965. Freshwater silver salmon (Oncorhynchus kisutch Walbaum). Calif. Dept. Fish and Game 51: 210-212. Westers, Harry. 1966. A density and feeding study with coho salmon. Fish Division, Michigan Department of Natural Resources. (Unpublished). Hesters, Harry. 1970. Carrying capacity of salmonid hatcheries. Prog. Fish-Cult. 32: 43-46. Wickett, W. P. 1951. The coho salmon population of Nile Creek. Fish. Res. Bd. Canada, Prog. Rept. Pacific Coast Station, 89: 88-89. . Willoughby, Harvey. 1968. A method for calculating carrying capacities of hatchery troughs and ponds. Prog. Fish-Cult. 30: 173-174. Withler, I. L. 1966. Variability in life history characteristics of steelhead trout (Salmo gairdheri) along the Pacific Coast of North America. J. Fish. Res. Bd. Canada 23: 365-393. Withler, F. C., and R. B. Moreley. 1970. Sex-related parental influences on early development of Pacific salmon. J. Fish. Res. Bd. Canada 27: 2197-2214. APPENDIX 65 Table A. Chemical analysis of water supply utilized for cultural operations during 1970 - 1971.* Parameter Reported as Well Water Tap Water Alkalinity mg/l CaCO3 331 292 Ammonia mg/l N 0.12 0.03 Chloride mg/l 01 0.6 5.2 Copper mg/l Cu 0.01 0.01 Hardness mg/l CaCO3 332 328 Iron mg/l Fe 0.1 0.1 Lead mg/l Pb 0.1 0.1 Conductivity Mi cronhos/ cm 650 682 Nitrate mg/l N 0.00 0.01 Ni trite mg/l N 0.00 0.00 Ph05phorus total mg/l P 0.13 0.85 Sulfate mg/l $04 4.0 6.0 Zinc mg/l Zn 0.015 0.015 * Analyzed June 1, 1971, by Institute of Water Research Water Quality Laboratory, Michigan State University. 67 .3553 82» EB mpmgmcE fammmom: :m 223:8 postage we. N .czbm 39833 u 5.53? $5.35 .33 omoncoaocfiéua £352; .93 atom 5.338 I c 3; 6355355 Esvaau £33: .chEmnooocguv N m :Emut, .Aocvxoutav m FEB; 40:32.63: m 553; 42752.3 m 5235 4:33:85 v. 523? .N 5233 5 2:53; .< 5.5:: Nmpagmfiz m.m - m.e PF - om op. . mo. a pactsopw> 3 - mg 8 .. E 8. - 8. m 58 : 322528 m.N - m._ Fmp . mom mo. . mo. N m._ F sway; , 3.28 m8 2 289 o._ - m.p mom - mmP.P mo. - mo. N e.o— m— gm< M., - o.F mmp._ - oNN.N mo. 1 pc. F m.¢ w an; 2333 0.: mm 539:. 83.5 Ammficflw canon. 20:: onto. 3:8..qu 3:823 59.3 ta 5.: to 8.5 Ba 33. 2:: 2:1 2:: cm: .3952 £3»ng 886 38 .8395 you“. 23.33 335 com... mmmapaé tom... A82 .38.: no.8 333 new .823... 28:3 85 mo naming .m 2.3... 68 Figure A. Mean total length of all experimental and control coho salmon on nine sanpling dates in 1971. Experimental and control groups l and 2 artificially spawned on Novenber 3. 1970. Experimental and control groups 3 and 4 artificially spawned on Novenber 19, 1970. 69 mzzo >m<=mmmu ¢ FN m mm a 0N NF 9N Np _dq:d—dd___—.——44j_—d—dqqqfiddqqqq—d-_——A—___————q#——__——dd—qfi«41__d_- +<1UO cm cm on oh om cop o—— ("I") H19N3‘l 1v101 wan Figure B. 70 Mean weight of all experimental and control coho salmon groups on nine weighing dates in 197l. Experimental and control groups 1 and 2 artificially spawned on November 3, 1970. Experimental and control groups 3 and 4 artificially spawned on November 19, l970. mzsn >