M\||\H\\N\\N\l \ m“mumsimmnuumwu 7‘--- w lUlUHllUHIIIMJUIUIIUIIlll'lllll‘ll‘IUHIUUIHW 31293 00785 1581 This is to certify that the thesis entitled EVALUATION OF TRIPLOID INDUCTION IN CHINOOK SALMON (ONCORHYNCHUS TSHAWYTSCHA) USING MICROWAVE RADIATION AND GROWTH COMPARISONS OF DIPLOID AND TRIPLOID CHINOOK SALMON presented by Paul Spruell has been accepted towards fulfillment of the requirements for Master of Science degree in Fisheries and Wildlife DateA 1151181: 4,1989 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution ‘ LIBRARY Michigan State L University W L PLACE ll RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or baton die due. DATE DUE DATE DUE DATE DUE L__JL EW- —_1| l- l H I MSU Ie An Afflrmdive Action/Equal Opportunity Imitation cWMt EVALUATION OF TRIPLOID INDUCTION IN CHINOOK SAEMON (Qngggnxngggs ISfiAflXIfiQflA) USING MICROWAVE RADIATION AND GROWTH COMPARISONS OF DIPLOID AND TRIPLOID CHINOOK SALMON BY Paul Spruell A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTERS OF SCIENCE Department of Fisheries and Wildlife 1989 m m to r:- o 3) ABSTRACT EVALUATION OF TRIPLOID INDUCTION IN CHINOOK SALMON (W W) USING MICROWAVE RADIATION AND GROWTH COMPARISONS or DIPLOID AND TRIPLOID curnoox SALMON BY Paul Spruell As part of a continuing study heat shocks were used to produce 143,607 triploid chinook salmon for stocking in the Great Lakes by the Michigan Department of Natural Resources. Although these heat shocks are effective in producing triploids, they also cause high mortality in treated eggs. Microwave radiation was investigated as a novel approach to induction of triploidy in chinook salmon. Shocks from a 750 watt microwave oven were administered to treatment units of 0.25 L of chinook salmon eggs. Shocks were applied over a range of duration and intensity combinations. Twenty continuous and three intermittent shocks were applied from 10-20 minutes post fertilization. Only three triploids were detected. Microwave radiation did effect mortality, but was not effective in the production of triploids. During an eight week growth study diploid and triploid chinook salmon reared in 110 L aquaria were fed either 2% or 4% wet body weight daily. Growth in tanks containing 100% diploids, 100% triploids, and a 50% mixture of diploids and triploids was evaluated. No difference (p>0.05) in growth was observed with in feeding regimes. ACKNOWLEDGEMENTS I would first like to thank my committee, Dr. Donald Garling, Dr. Howard Tanner, and Dr. James Asher for their suggestions in the completion of this degree. Special thanks go to Dr. Carling, my major professor for his help during my work, as well as to Dr. Tanner for working to maintain funding for this project. I would also like to thank those individuals and companies which have helped fund this project. Specifically, the American Fishing Tackle Manufacturers Association and Bill Cullerton Sr. and his staff at the Chicagoland Sport Fishing, Travel 8 Outdoor Show. In addition, I would like to thank all contributors and buyers from the Chicagoland auction, whose donations combined to yield the majority of the funding for my project. I would also like to thank the Michigan Department of Natural Resources, Fish Division for their continued support of the triploid chinook project, despite a less than optimal run in 1989. This includes all personnel at the Little Manistee, and Boardman River Weirs, and the Wolf Lake Hatchery. Thanks also to Dr. Kathy Brooks of M80 for allowing me to use her lab and flow cytometer for ploidy determination. iii In addition to Dr. Brooks, I would like to thank Vonnie VanderPloeg for patiently running the flow cytometer, even when things weren't running smoothly, and Corey Casey for enduring my continual presence in the lab. I would also like to thank Angelo Napolitano for his help in obtaining chicken blood to use as a control. I would also like to acknowledge Ric Westerhof for his help upon my arrival and all individuals who have voluntarily assisted in carrying out my research, the MDNR production runs, and fund raising activities, Ken Cain, Todd Grischke, Tom Heron, Jay Hesse, Scott Sowa, Chris Starr, and Andy Westmaas. I owe a special debt of gratitude to Roger Glass, Paul Wilbert, and Bill Young. To Roger for his never ending work at our lab, which entailed anything we needed done, to Paul and Bill for providing continual advice, assistance, and support in the design and execution of my project. Finally, I would like to thank my parents and family for their support throughout my academic career. iv LIST OF TABI‘ES...OOOOOOOOOOOOOOOOOIOOOOOOOOOOOOOOOOOO LIST OF TABLE OF CONTENTS vii FIGURES...0.0000......OOOOOOOOOOOOOOO00...... V111 INTRODUflION...0.000000000000000...OOOOOOOOOOOOOOOOOO LITEMTURE REVIEWOOOOOOOOOCOOOOOOOOOOOOOOOOOOOOOOOOO. I. II. III. IV. V. VI. VII. CHAPTER I. II. III. IV. CHAPTER Natural polyploids.......................... Techniques to Induce Polyploidy............. Gynogenesis................................. Androgenesis................................ Triploid Hybrids............................ Gonadal Development and Sterility........... Identification of Triploids................. I. ATTEMPTED INDUCTION OF TRIPLOIDY IN CHINOOK SALMON (QNCORYNCHUS Ififlfiflllfigflb) USING MICROWAVE RADIATION............ Abstract.................................... Introduction................................ Materials and Methods....................... Results..................................... Discussion and Conclusions.................. II. EVALUATION OF GROWTH DIFFERENTIAL IN DIPLOID AND TRIPLOID CHINOOK SALMON (QHQQBHXEQHQE TSEAEXI§QEA)-oooooooo AbstractOOOOOOOOOOOOOO0.00.0000...0.0.000... V 1 3 11 15 15 15 19 28 33 36 36 II. IntrOductionOOOOOOOOOOOOOOOOOOOOOOOO00...... III. Materials and Methods....................... Iv. ResultSOOOOOOOOOOOOOO0.000000000000000000000 v. DiscuSSion.OOOOOOOOOOOOOOOOOO0.0000000000000 CHAPTER III. MICHIGAN DEPARTMENT OF NATURAL RESOURCES PRODUCTION RIJNSOOOOOOOOOOOOOOOOO CHAPTER V. SUMMARY AND CONCLUSIONS.................. Appendix AOOOOOOOOOO0.0000000000000000000.0.00...O... REFERENCESOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO vi 36 39 42 42 47 51 53 56 LIST OF TABLES BEES Examples of Induction of Triploidy by Heat Shock in Fishes................... 6 Summary of chinook salmon egg mortality after microwave treatments andministered in catcher, 1987.00.00.0000000.0.0....O... 29 Average temperature (°C) resulting from microwave shocks applied to 0.25 L of chinook salmon eggs in 1987................ ...... . 20 Number and Location of Triploid Chinook Salmon Stocked by the Michigan Department of Natural Resources in Spring 1989.......... 46 vii LIST OF FIGURES Timing of steps in the production of triploid chinook salmon using heat shocks at the Little Manistee River Weir near Stronach, Michigan ............................ Timing of the steps in microwave shocks applied to chinook salmon eggs in 1987 at the Little Manistee River Weir near Stronach, Michigan............................ Timing of the steps in constant microwave shocks applied to chinook salmon eggs at the Little Manistee River Weir near Stronach, Michigan in 1988.................... Timing of the steps in intermittent shocks applied to chinook salmon eggs at the Little Manistee River Weir near Stronach, Michigan in 1988.................... Effect of various microwave power seittings on chinook salmon mortality holding shock duration constant............................ Effect of various microwave shock durations on chinook salmon mortality maintaining a consztant power setting........................ Growth of diploid and triploid chinook salmon fed at 4% wet body weight daily grown seperately or mixed in a single 110 L aquarium............ Growth of diploid and triploid chinook salmon fed at 2% wet body weight daily grown seperately or mixed in a single 110 L aquarium............ viii 21 22 26 26 3O 31 43 44 INTRODUCTION In 1984, Michigan State University, in conjunction with the Michigan Department of Natural Resources (MDNR), began stocking triploid chinook salmon in the Great Lakes. The goal of this project was the production of a trophy fishery for the Great Lakes. Sterile triploids are hoped to reach a larger ultimate size due to their extended life span. Chapter I is a summary of the relavent literature available on induced polyploidy in fishes. Production runs of 200,000 triploid chinook salmon have been produced using heat shock techniques developed at MSU. Although heat shock reliably produces 95-100 percent triploids, mortality rates are often in excess of 80 percent. This high mortality necessitates shocking two million eggs to obtain the desired 200,000 fish for stocking. The high mortality rate also makes increased care of eggs necessary. If an alternate means to produce triploids could be developed which did not result in high mortality, the cost and effort required to produce triploids could be reduced. Chapter II details experiments attempting to use microwave radiation as an alternative to heat shock. It was hoped that a short duration microwave shock would induce triploidy without causing the high mortality rates associated 1 2 with heat shock. Using both constant and intermittent microwave shocks, a variety of intensities and durations were attempted. There has been much speculation concerning the performance and behavior of triploids. Researchers have reported triploids to grow faster, slower, and at the same rate as diploids of the same species. Anecdotal reports have also indicated that diploids may have a competitive advantage over triploids under some conditions. Chapter III is a report of a statistically valid comparison of the growth of diploids and triploids grown alone and under competitive conditions. In 1988 heat shock was used to produce triploids for stocking by the MDNR. A slight modification was made in the procedure used in past years. Survival rates for heat shocked eggs in 1988 were improved by approximately 8 percent over previous years. Although we cannot be certain that this increase in survival is due to the modification, it is a possibility that deserves further attention. A description of the modification and a record of stocking locations and numbers are given in Chapter IV. CHAPTER I. LITERATURE REVIEW W Polyploidy is usually lethal in animals, but is relatively common in plants (Strickberger, 1985). Only lower vertebrates and more primitive organism appear to survive polyploidy. There are examples of naturally occurring polyploid insects, crustaceans, amphibians, reptiles, and fish (Strickberger, 1985). These organisms produce offspring parthenogenically. Some organisms, such as the amazon molly (ggggilia formgga), maintain a natural state of triploidy by gynogenesis (Schultz and Kallman, 1968). The amazon molly is believed to be of hybrid origin. In present populations, sperm from the male serves only to stimulate egg development, but its genetic material is not incorporated into the embryo. Triploidy is maintained by an endomitotic division that results in a hexaploid cell. After meiosis, the resulting egg is triploid. (Schultz, 1967). c ' u 0 Several techniques have been used to induce polyploidy in fishes which are generally based on earlier experiments with amphibians (for review see Fankhauser, 1945). Triploid fish which have been produced by artificial induction have proven to be sterile. Sterile fish have commercial potential, and thus induced polyploidy has been studied 4 extensively in the past few years. The earliest attempts to produce polyploid fishes utilized cold shocks to disrupt chromosome migration. Cold shocks have proven to be species specific and variable in their degree of success. Swarup (1959a) reported the occurrence of some mosaics in cold shocked stickleback (Casterosleus agglgatgs). Cold shock has been used to produce triploid plaice (Purdom, 1972), Iilapia 331g; (Valenti, 1975), Carp (gyprinus carpig) (Gervai et a1., 1980), and channel catfish (Igtalurgs punctatgs) (Walters et a1., 1981b). Cold shocks did not induce triploidy or tetraploidy when administered to Atlantic salmon (Salmg galgr) (Lincoln et a1., 1974) and to brook trout (Salyelingg fightinalis) (Lemoine and Smith, 1980). Researchers have attempted to produce both triploids and tetraploids using chemicals. In past attempts, eggs have been submersed in a chemical solution during part of their development. Cytochalasin B, which blocks cytokinesis has been used to treat eggs of Atlantic salmon (Allen and Stanley, 1979; Refstie et a1., 1977) and rainbow trout (Qgggxnynghug mykisg) (Refstie et a1., 1977; Refstie, 1981). Polyploid mosaics were produced but no pure triploids or tetraploids resulted from these treatments. Similar results were obtained when brook trout eggs were exposed to colchicine (Smith and Lemoine, 1979). Triploidy was successfully induced (>90%) when rainbow trout eggs were 5 exposed to nitrous oxide coupled with elevated pressure (Shelton et a1., 1986). Neither nitrous oxide nor pressure alone resulted in any triploids at the levels used in this study. Heat shocks have proven to be the most reliable means of producing polyploids in most fishes. Experiments involving heat shock have been summarized by Westerhof, 1988 (Table 1). It has been suggested that the heat shock may denature the spindle fibers during meiosis two (Briggs, 1947; Fankhauser and Godwin, 1948). Rieder and Bajer (1979) have shown heat shocks depolymerize the microtubules in cultures of newt lung cells. The disruption of second polar body extrusion during meiosis is the probable cause of unreduced ova in fishes (Purdom, 1983; Thorgaard, 1986). The success rate of heat shocks has varied from 13 percent in channel catfish (Bidwell et a1., 1985) to 100 percent in rainbow trout (Chourrout, 1982: Lincoln and Scott, 1983; Solar et a1., 1984), Atlantic salmon (Benfey and Sutterlin, 1984; Johnstone, 1985), and chinook salmon (Hill et a1., 1985; Westerhof et a1., 1988). Treatments must be optimized for each species of interest (Thorgaard, 1986). The optimal treatment consists of shock temperature, shock duration, and time after fertilization at which shocks are administered. Johnstone (1985) demonstrated that induction rate and mortality are positively correlated, and suggested that the maximum percentage of triploid induction may not be 6 Table 1. Examples of Induction of Triploidy by Heat Shock in Fishes Heat Time Shock After Primary Temp. rert. Duration Ploidy Surv. Author and Species (C) (min) (min) % % Year (1900) sturgeon 34 1-60 3 52.3 --- Vasetskii 67 tilapia 32-38 14 1h 10 50 Valenti 75 trout 27-30 60 10 50 --- Chourrout BO trout 37-38 10 1 50 10 Thorgaard 81 trout 26 25 20 90-100 63 Chourrout 82 trout 27-28 40 10-15 100 --- Lincoln 83 coho 24-30 10 10 85 47 Utter 83 pink 29 10 10 72 74 Utter 83 chinook 29 10 '10 60 88 Utter 83 Atlantic 32 5 5-15 100 80 Benfey 84b trout 28 35-40 10 90 50 Lou 84 trout 26-28 40 10 90-100 50-57 Solar 84 catfish 40 80-90 1 13 --- Bidwell 85 chinook 28.5 10 10 100 70 Hill 85 Atlantic 30 20 10-12 100 67 Johnstone 85 trout 38 40 10 90 40 Bye 86 trout 29 10 10 91-96 48-79 Thorgaard 86 tilapia 39.5 3 3.5-4 60 60 Don 88 chinook 28.5 10 10 100 45 Westerhof 88 modified from Westerhof (1988) 7 the most efficient use of eggs due to the increasing mortality associated with the more extreme treatments. This may be disputed if 100 % sterile fish are necessary for implementation of a successful management scheme. Hydrostatic pressure is one of the most promising means of blocking the extrusion of the second polar body. It has been used with a great degree of success to produce triploid rainbow trout (Chourrout, 1984; Lou and Purdom, 1984) and Atlantic salmon (Benfey and Sutterlin, 1984a). As well as tetraploid rainbow trout (Chourrout, 1984). It has been suggested that hydrostatic pressure may be a more desirable mechanism for polyploid induction than heat shock since mortality resulting from pressure treatments is usually lower than for heat shocks (Solar et a1., 1984). Pressure treatments are limited in the volume of eggs that can be treated at one time, but this limitation may be overcome in time (Thorgaard and Allen, 1987). Wis The primary effect of triploid induction techniques is the production of a diploid ovum. If the eggs were stimulated to continue development after the inducing shock were applied without adding additional genetic material, as is the case if irradiated sperm are used, a gynogenic diploid would be produced (Purdom, 1983; Thorgaard, 1986). This procedure has been used successfully to produce gynogenic 8 rainbow trout (Chourrout, 1980; Chourrout, 1984) and zebrafish (graghyganig rerig) (Streisinger et a1., 1981). There is increasing interest in gynogenesis in fish species in which the male is the heterogametic sex since all female populations could be produced. These populations could be used for control of reproduction, or the development of inbred lines (Purdom, 1976; 1983: Chourrout, 1982). dr e 's In addition to fish with all maternal inheritance, androgenic fish can be produced in which all genetic material is of paternal origin. These fish result when eggs are irradiated by gamma radiation to inactivate the maternal genome (Purdom, 1969, Parsons and Thorgaard, 1984). After fertilization with normal sperm, first cleavage is blocked (Parsons and Thorgaard, 1985). Androgenic fish could produce all male lines using sex reversed males coupled with a screening procedure to isolate those fish of the appropriate chromosome composition. All male populations produced by androgenesis might be used in a manner similar to gynogenic fish to produce inbred lines for scientific study (Thorgaard, 1986). Triploid_flxbrid§ A final possibility in chromosome manipulation involves the production of triploid hybrids. It has been shown that 9 in some fish crosses the triploid hybrid survives better than the corresponding diploid hybrid (Chevassus et a1., 1983; Scheerer and Thorgaard, 1983). Triploid hybrids have three potential aquaculture and fisheries management applications. The higher survival of triploid hybrid fish might make otherwise impractical diploid hybrids feasible (Refstie et al., 1982). Inducing triploidy in hybrids that already show a skewed sex ratio would further reduce the risk of unwanted reproduction (Thorgaard and Allen, 1987). Finally, desirable traits from two species might be combined in one fish (Utter et al., 1983). MW Both male and female triploids generally display a reduction in size of the gonads. The reduction is greater in females than males (Thorgaard, 1986). A gonadal somatic index (GSI) is regularly used to compare development of gonads between fish. The index is calculated by dividing the weight of the gonad by the total weight of the fish (Johnson et al., 1986; Benfey and Sutterlin, 1984a). Benfey and Sutterlin (1984a) found testicular development was reduced by 7.2% while ovarian development was reduced by 52% in Atlantic salmon. In triploid coho salmon GSI in males was 35.7% of normal diploids, while GSI of females was 11.8% of the normal diploid (Johnson et al. 1986). The unequal effect of triploidy on gonadal development has also been observed in 10 rainbow trout (Thorgaard and Gall, 1979; Lincoln and Scott, 1983:1984; Solar et al. 1984) channel catfish (Walters et a1., 1982), plaice x flounder (Elatightnys £1g§3§)(Purdom, 1972) and turbot (figgphthalmug maximus) x brill (Sggphthalmus :hgmpgg) hybrids (Lincoln, 1981a: b). Although gonadal development is usually impaired in triploids it is not completely eliminated. Organisms that have an odd number of chromosome sets are usually sterile (Cassoni et al., 1984; Strickberger, 1985). Thus despite the development of gonads, abnormal development of gametes render the individuals sterile. The presence of a third homologue may prevent proper pairing and migration during meiosis thus preventing the completion of gametogenesis (Solar et a1., 1984; Strickberger, 1985). Male triploid rainbow trout produced aneuploid spermatozoa that produced no viable offspring when used to fertilize eggs (Benfey et al., 1986). Allen et al. (1986) induced spermiation in a triploid grass carp by injection of carp pituitary. The resulting sperm was found to be aneuploid, and eggs fertilized with this sperm were not viable. Artificially induced triploid fishes are functionally sterile. Much of the interest in using triploidy in aquaculture and management depends on the elimination of the secondary sex characteristics, in addition to sterility. It has been shown that elimination of gonadal development by surgical castration prior to maturity eliminated post- 11 spawning mortality in kokane salmon (Robertson, 1961). The development of secondary sexual characteristics were not exhibited in spontaneous triploid rainbow trout (Thorgaard and Gall, 1979). Based on the sex biased differential reduction in gonadal development, only triploid females may demonstrate all of the traits desired by fisheries scientists (Thorgaard, 1986). n ' 'c o Triploids are usually not identifiable by morphometric observations. In carp, when a female of the scattered scale type was crossed with a male of the normal scale pattern, the resulting triploid offspring displayed a minor disturbance in scale pattern when compared to diploids (Gervai et al., 1980). Swarup (1959b) reported detectable variation in certain body parts in triploid stickleback, but they were identical in most morphometric characteristics. Other researchers have not detected morphometric means to distinguish triploids from diploids based on empirical observation (Walters et al., 1982b; Cassani et al., 1984; Johnson et al., 1986). Triploids can be identified by direct observation of chromosome number by karyotyping. Kligerman and Bloom (1977) developed an inexpensive method to make chromosome preparations from solid tissues. Although this method can be used reliably to produce quality chromosome spreads, it takes 12 considerable practice to become proficient at both the slide preparation, and accurate counting of the spreads (Sweet, 1986). Embryos contain a high percentage of replicating cells. Thus they usually provide quality preparations for karyotyping (Myers, 1986). Tissue cultures can also be used to provide chromosome spreads. Wolters et al. (1981b) found cultured lymphocytes from macerated kidney tissue of channel catfish to provide good chromosome spreads. Most fishes contain a large number of small chromosomes and the exact number may vary in cells from one individual and between fish of the same species (Blaxhall, 1983). It is necessary to do several counts on each specimen to identify variation caused by counting and variation found naturally in the population. The limited number of spreads from each specimen that can be analyzed make detection of mosaics difficult (Thorgaard et al., 1982; Wattendorf, 1986). The number of fish from each treatment which can be tested is also limited, thus the percentage of triploids produced may not be accurately estimated. Other methods have been developed to identify the ploidy of cells by direct observation under a microscope. Microfluorimetry and microdensitometry can be used to measure the DNA content of individual cells (Gervai et al., 1980; Johnstone, 1985). The number of nucleoli present in each cell can also be used to determine ploidy (Phillips et al., 1986). The nucleoli are silver stained and then counted. 13 Diploids were found to have either one or two nucleoli, while triploids had one, two, or three nucleoli. Although no diploids possessed cells with three nucleoli, the overlap in number per cell makes observation of many cells necessary to insure accurate estimates of ploidy and there is no way to detect mosaics. Although other ”direct observation" methods are an improvement over karyotyping, they are still labor intensive and time consuming. The nuclear volume of erythrocytes has been found to differ significantly between diploids and triploids (Swarup, 1959; Purdom, 1972; Cimino, 1973; Valenti, 1975; Allen and Stanley, 1978; Lemoine and Smith, 1980). Nuclear volume can be measured and ploidy determined on this basis. The distribution of nuclear volume of diploids and triploids has been found to overlap in some fish species making the detection of mosaics difficult (Purdom, 1972). Variation in nuclear volume within individual fish has also been detected (Allen and Stanley, 1978; Lemoine and Smith, 1980). Thorgaard and Gall (1979) found extensive variation both within individuals and among individuals of each ploidy type. Thus a comparison between individuals of known ploidy level must be done on a species to insure the validity of ploidy determinations using nuclear volumes (Walters et al., 1982a). An improvement over direct measures of erythrocyte nuclei is the use of a Coulter Counter and channelizer (Wattendorf, 1986). This method allows measurement of l4 thousands of cells, and thus is more sensitive to the presence of mosaics. Despite its advantages, Johnson et al. (1984) found that a Coulter Counter gave inconclusive results in 11% of their coho and chinook salmon blood samples. The use of a flow cytometer allows rapid analysis of ploidy with unequalled precision (Callis and Hoehn, 1976; Thorgaard et al., 1982; Allen, 1983; Johnson et al., 1984). Flow cytometry measures the fluorescence of stained DNA as the cell is passed through a laser (Allen, 1983). The cells are passed through the laser in single file driven by either air pressure or a vacuum (Dowining et al., 1984). As the cells pass through the laser, the amount of fluorescence is converted into an amplified electrical signal, digitized, and displayed. The type of display generated varies depending on the specific information sought by the researcher (Downing et al., 1984). Frequency histograms are usually used to determine ploidy (Thorgaard et al., 1982; Allen, 1983). CHAPTER II. ATTEMPTED INDUCTION OF TRIPLOIDY IN CHINOOK SALMON (QNQQBHXNQHQS.TEHAKXTEQHA) USING MICROWAVE RADIATION ABSTRACT Methods have been developed to produce large numbers of triploid fish using heat shock. However, these techniques cause high levels of mortality. Microwave radiation was explored as a potential source of shock to induce triploidy. In addition to lowering mortality, we hoped microwave radiation would be less temperature dependent and have application over a broad spectrum of species. Chinook salmon eggs were shocked using a 750 watt microwave oven over several time-power combinations. Eggs were shocked from 10-20 minutes after fertilization, in a microwave bundt cake pan to provide uniform exposure. Although mortality rates were increased in some treatments, very few triploids were detected. This would indicate that microwave radiation, while having some effect on the eggs is not specific in depolymerizing microtubules without fatally damaging other components in the egg. INTRODUCTION There are many potential advantages to using genetic manipulations to produce sterile fish in aquaculture and fisheries management. One way to produce sterile fish is genetic manipulation to produce triploid individuals 15 16 (Lincoln and Scott, 1983, 1984: Benfey and Sutterlin, 1984b). Triploids are sterile due to disrupted gametogenesis caused by the presence of an extra set of chromosomes (Purdom, 1983). Sterile triploid fish could be used to stock waters in which control of reproduction is desirable to protect endemic organisms (Purdom, 1983: Cassani et al., 1984) or to prevent stunting (Allen and Stanley, 1979: Wolters et al., 1982b; Lou and Purdom, 1984). Elimination of sexual maturation in salmonids is of particular interest since secondary sex characteristics, such as reduced growth rate and decreased flesh quality, could be avoided by using sterile triploids (Lincoln et al., 1974: Lemoine and Smith, 1980). Precocial maturation is another problem of economic importance that might be eliminated using triploids (Solar et al., 1984). Finally, as a result of their extended life span, triploid pacific salmon might reach a larger ultimate size than diploids (Thorgaard, 1986). If properly managed, these fish could provide a trophy fishery for the Great Lakes (Westerhof et al., 1988). Although triploidy is rare in most fishes, some naturally reproducing triploid fish have been described (Schultz, 1967: Schultz and Kallman, 1968; Menzel and Parnell, 1973). Triploidy also occurs spontaneously at a low level in individuals of some species (Cuellar and Uyeno, 1972; Gold and Avise, 1976; Thorgaard and Gall, 1979) Many fish that are normally diploid have shown little or 17 no adverse effects when artificially induced to triploids (Purdom, 1972; Walters et al. 1981b; Chourrout, 1980; Benfey et al. 1984; Johnson et al. 1986). A wide variety of techniques have been used to prevent the extrusion of the second polar body during meiosis, resulting in a triploid fish with two complements of the maternal chromosomes and one complement of the paternal chromosomes (Purdom, 1983). Chemicals that inhibit chromosome migration have been used to produce polyploid mosaic fish, but have rarely produced pure triploids (Refstie et al., 1977; Allen and Stanley, 1979: Smith and Lemoine, 1979). Hydrostatic pressure has been used to produce both triploid (Benfey and Sutterlin, 1984b: Chourrout, 1984; Lou and Purdom, 1984) and tetraploid fishes successfully (Chourrout, 1984). Since only a limited volume of eggs can be treated at one time using pressure it is not applicable to treating large numbers of large eggs (Thorgaard, 1986). Temperature shock has been the most common method used for triploidy induction in fishes. Cold shocks have been used to induce triploidy in the plaice and its hybrid with the flounder (Purdom, 1972), the channel catfish (Wolters et al., 1981), and tilapia (Don and Avtalion, 1988). Cold shocks have not been used successfully to produce triploid salmonids (Lincoln et al., 1974; Lemoine and Smith, 1980). Triploid salmonids have been successfully produced by heat shocks (Chourrout, 1980: Thorgaard et al., 1981; Benfey and 18 Sutterlin, 1984b; Johnstone, 1985; Johnson et al. 1986: Westerhof et al., 1988). Heat shock has been used to produce large numbers of high percentage triploid chinook salmon for the Michigan Department of Natural Resources (MDNR) Great Lakes stocking program. High mortality rates (85-90 percent) have been associated with these treatments (Westerhof, 1988). Producing the number of fish desired for the MDNR production runs is not difficult since excess eggs are usually available. However, if eggs are in short supply, production of 200,000 triploids may not be possible in some years (see Chapter III). The high mortality rate also dictates increased care by MDNR workers. Moribund eggs must be removed or antifungal treatment administered more frequently to prevent fungal growth. Thus, a means to produce triploids that significantly lowered mortality would decrease the number of eggs necessary and the effort required to produce the triploid fish. In this study, we attempted to induce triploidy in chinook salmon using microwave radiation as the shock. It was hoped that a short duration microwave treatment could be used as the shock, and that mortality would be decreased. The possibility of using microwave shocks was also appealing in that they are easy to administer and would enable triploids to be produced using equipment readily accessible to both fish farmers and management agencies. Microwaves 19 would also utilize a shock other than temperature which might allow more uniform application over a wide variety of species, adjusting for developmental rate and volume of the eggs being treated. MATERIALS AND METHODS Prior to the onset of our experiments with chinook salmon eggs, the uniformity of heating produced by a Sears 750 watt microwave oven was examined. Garden peas were used to simulate salmon eggs and to determine the appropriate procedure to produce consistent heating throughout a treatment group. Salmon eggs were neither available no expendable, thus peas were selected due to their similar size and shape. Our preliminary trials compared heating uniformity in peas heated in 250 ml glass beaker, a 9x9x2 inch glass baking dish, and a microwave safe bundt cake pan. Peas were heated at various settings and temperatures were taken at several points throughout the pan. We concluded that a microwave-safe bundt cake pan produced the most consistent and uniform heating. We also determined that for short treatments (5 1 min.), no benefit was gained by using a rotating platform in the oven. Eggs of salmonids are not easily observed to determine the time of developmental events (Chourrout, 1982). Great Lakes chinook salmon eggs are large, opaque and colored orange making direct observation difficult. Triploidy can be 20 induced in chinook salmon using heat shocks applied 10 minutes post fertilization as outlined in figure 1 (Hill et al., 1985; Westerhof et al., 1988). We relied on results from the successful induction of triploidy by heat shock to determine the appropriate time to apply the microwave shock. Our first attempt to induce triploidy using microwave radiation was conducted in October of 1987. Eggs were randomly sampled from female chinook salmon collected by the MDNR and the shocking experiments were conducted at the Little Manistee River Weir near Stronach, Michigan on October 13, 1987. Chinook salmon were spawned using standard procedures (Piper et al., 1982). Eggs were removed from ripe females by injecting compressed air at 2 psi pressure into the body cavity. Eggs from 6 females were mixed in a 5 gallon pail. Milt was obtained by applying pressure to the abdomen of ripe males. Milt from 5 males was collected and mixed in a plastic wash bottle. Treatment units of 0.25 L of eggs were fertilized without water in plastic quart containers. Two minutes after fertilization, river water was added to the quart containers to increase the percent of eggs fertilized (Piper et al., 1982). Immediately prior to shocking, eggs were drained and placed in a bundt cake pan for shocking. At ten minutes post fertilization eggs were shocked (figure 2). Shocks were administered at 25, 50, 75, and 100% power. Three replicates of each power level was tested at 15, 30, 45, and 60 seconds. 21 .unwa50fiz .nuucouum uses uses uo>wm moumwcmz oauqu enu um mxoonm use: mafia: soaamm xoocwco macadauu mo coauusooua onu cw woman «0 mafiaqa .H muswfim A.=aev arse mm an N o _ _ _ _ 1. M u “.4 M I. u a E a S J 1. 1 .a a 1. o u. Hoou scam ammo macaw 1 W“ a n . .. o. o z a «q o t. .4 u A I. e o «q u xoonm use: M. u oo m.m~ 22 .ccwHEOHz .numcouum new: news omumficmz macaw; may um mama ca wwwm coaamm xoocanu ou voaaeam mxoonm u>m30uuwe ca macaw mo wcHaHH .N ouswfim 755 as: In 0‘ In on ow OH N o _ _ _ __ «4 M 1 e m. a J E «é D E a u a m. .41 a z A mm. o U. Hooo 30am w ammo emcwu .L .4 1 s o z a. q nu u m A I. N. E O 3 Au to O u FL 23 After shocking the temperature of each treatment group was determined by placing a thermometer in the bundt cake pan at several places and averaging the recorded temperature. Eggs were returned to the plastic container, slow cooled to ambient temperature of river water over a 15 minute period by addition of cool water, and water hardened for one hour. Water hardening results from the uptake of water by the outer membrane of the egg. This makes eggs more resistant to mortalities due to handling and transport. The plastic containers of eggs were placed in a cooler to maintain constant temperature and transported to the Fisheries Research Lab at MSU. At the MSU lab, each replicate was randomly placed in a vertical stack incubator (Heath Technica). Incubator tray were divided, so up to four different treatment lots could be reared in one tray. All moribund eggs were removed from the stacks at two to three day intervals to prevent fungal growth. Moribund eggs were counted to calculate mortality rates for each treatment. Swim up fry were placed in 110 L aquaria. Each aquarium was supplied with 13 °C well water at a rate of 0.25 liters per minute and equipped with an external stand pipe to maintain constant water level (Garling and Wilson, 1976). Fish were fed to satiation at least three times daily using biodiet salmon starter. Fish were maintained in these tanks until they reached sufficient size for ploidy analysis. Due 24 to space limitations and the low survival from some treatment groups, some replicates were combined and raised in one tank. Shortly after all of the yolk sac had disappeared, blood was taken for ploidy analysis using flow cytometry as described in Appendix 1. Five fish from each replicate were sampled, in those treatments which had been combined, 15 fish were sampled. To provide a more reliable estimate of triploid induction rate, an additional 15 fish were tested in those replicates in which triploids were detected. In the Fall of 1988 a second attempt to induce triploidy using microwave radiation was made. The 750 watt Sears microwave oven was used as the source of the shock. On October 21, 1988 chinook salmon eggs were collected, fertilized, and water activated at the Little Manistee Weir as described for treatments done in 1987. Based on 1987 results, two variations in shocking were administered. It was hypothesized that a one minute shock did not produce high rates of triploidy because only a low percentage of the eggs were at a susceptible stage of development. Results from heat shock studies indicate that the optimal time for application of shock is from 10-20 minutes post fretilization. A 10 minute microwave treatment to span this entire range of development would have been desirable; however, trials in 1988 proved 10 minute shocks even at 10 percent power caused 100 percent mortality in treatment groups of 0.25 L. If one assumes the percentage of 25 eggs in the critical developmental stage occurs as a normally distributed function of time, the maximum number of eggs in this stage would occur at 15 minutes post fertilization. To maximaize the number of eggs in the critical period during exposure to microwave radiation the following variations of our initial shocks were applied. In the first variation, eggs were shocked for five minutes, beginning 12.5 minutes after fertilization (Figure 3). Power settings of 5, 10, 15 and 20 percent were used for the continuous shocks. In the second variation power settings of 40, 50, and 60% were used. Eggs were shocked for 30 seconds, at two minute intervals starting at 10 minutes post fertilization, and continuing to 18 minutes post fertilization (Figure 4). This pattern yeilded a total of 2 minutes of shock, over a period of 8 minutes of development. Unlike 1987, eggs in 1988 were kept in water during shocking. This was done to prevent oxygen stress during the longer shock period. A rotating platform was also used in the microwave to insure equal treatment to all eggs during the prolonged exposure. All eggs were water hardened for one hour and transported to the MSU lab for incubation as described for treatments made in 1987. At swim up, fry were placed in 110 L aquaria and maintained as described above, until ploidy analysis could be conducted. Ploidy analysis was done using flow cytometry as in 1988. 26 .mmma ca coma50a: noesouum use: use: um>am emumacmz mauuaq on» um move coaamm xoocwco cu ocwaoon mxoonm m>e30uofifi unnumcoo cw woman no mcfiswa .m shaman A.:ssv szcs mm mm cu oH N c _ _ _ i _ _ _ n m m a m .... s 1 1. 1 d a 1. O U. I T. 1 e Hooo 33m name was: I D. 3 2 a 3.093 1. a. u 25395.“... M. "w... P O 1. H To 0 u 27 in C‘ :uodsuezq. __ .OOOH cu cuwfinuwz .nuecouum use: use: uo>qm ueumacmz eduuwg era as ammo cosacm soocwnu cu vmaficcc mxoonm c>uzouufia ucwuuwsuouca cu adobe «0 wcfiefie .e unawfim A.:ssv msz ON OH N O ._ _ _ _ m n I. a e a . a u. u. Hooo 33m name one: 1 I. o .L I E T. D. D 2 . n. n U A T. P O «a u To 0 u L O 1 r mxoosm e>e30uuHE 28 RESULTS In 1988, two triploid fish were detected, one from a 15 second treatment at 100 percent power and one from a 45 second treatment at 50 percent power. Mortality figures were calculated and are presented in Table 2. 'Shocks exceeding one minute in duration at full power, and shocks of 10 minute duration at 10 percent power were lethal. Figures 5 and 6 demonstrate the effects of varying time and power setting. In Figure 5, average percent mortality is plotted versus power setting, holding shock duration constant. Figure 6 is a similar graph maintaining power setting as the constant. Mortality increased with increasing percent power and with treatment duration as did the observed temperature (Table 3). In 1989, one triploid was detected from an intermittent shock, at 50 percent power. Mortality records could not be kept accurately in 1989 because of unexpected problems at the laboratory. On December 10, 1988 a pump failure at the MSU laboratory caused significant mortality in all treatment groups despite our efforts to maintain water flow from an auxiliary source. Due to a ruptured pipe inside of a filter tank we experienced periodic discharges of filter media from a new iron removal system. Although the effect of the media on the eggs is unknown, those eggs in the upper trays of our heath stacks were coated with the media on several occasions. Eggs in the lower trays were not exposed to the media since it was trapped by the screens and eggs in TABLE 2. 29 Summary of chinook salmon egg mortality after microwave treatments andministered in October, 1987. Treatments are expressed as time in seconds/% power setting. Mortality due to treatment refers to mortalities recorded 24 hrs after exposure to microwave radiation. Mortality after treatment is the fraction of individuals remaining after day one which died prior to transfer from incubator trays to aquaria. (** differs from control p<0.05) n MORT DUE AVE. % MORT. TRT H AVE. % MORT. TO TRT AFTER TRT 15/25 . 4o gfl 12 T 32 15/50 40 13 32 15/75 50 h 41 28 15/100 54 H 32 34 30/25 53 30 34 30/50 45 20 32 30/75 63 u 38 41 30/100 72 n so 46 45/25 53 25 39 45/50 61 4o 37 45/75 85 H 71 ** I 55 ** 45/100 87 H 76 ** 56 ** 60/25 69 J 41 49 60/50 82 n 67 4* 45 60/75 97 fl 9o ** 7o 4. 60/100 100 H 100 ** -- IC 45 8 38 EC 45 12 4o 3O .muumufiaauu monsu mo mufiHmDuoa some ecu mucomcuemu ucfioa comm .Amvcooom ca coaueusv xuonm a DO ucnumcoo coaumucv xuonm wcfipaon huwamuuoa cosacm xoocusu co mwcauucm Meson u>msouoaa msoaum> mo uuemmm .m ouswwm mmzfiyuhzmocmm Oop on em O1 ON — _ _ p b P _ Genus a men.“ 0 I 81. + apt. e I I l op cow AJINUAKWNINBOUEd 31 .muunofiamuu noun» «0 some ecu mucmmuumuu ucwom zoom .Auesom unmoumn n av chuuuw um3om accumcoo o mcflc«ou:fioe %ufiamuuoa COEHmm xoocfino :o mcowuouso xoocm m>o30uows moofiuo> no vacuum .O unsawm 20.5.5150 goo-5 Oh cm on O— . _ _ _ _ O I O— ocpna o R .. a o . I 8 our“. + I O0 mule I I Ow 1.3 I 8 I on I OO I Om OO— AJJ'IVIHOW .LNBOBBd 32 Table 3. Average temperature (°C) resulting from microwave shocks applied to 0.25 L of fertilized chinook salmon eggs in 1987. 15 3O 45 60 Power Settings (%) 25 19 20 27 3O 50 22 25 27 37 75 21 27 32 41 100 21 3O 4O 53 Initial temperature = 12.5 °C 33 the upper trays. Survival was sufficient for ploidy analysis, but comparison of mortalities is not possible. DISCUSSION AND CONCLUSIONS Microwave shock treatments were not effective in induction of triploidy in chinook salmon when applied ten minutes after fertilization. Although three triploids were detected from treatment groups exposed to microwave shocks, we can not be certain they are the result of the microwave treatment. Other researchers have reported triploids in normal fish populations. Gold and Avise (1976) found a spontaneous triploid in a sample of nine California roach Hespgzgleuggs symmetrigus. A single triploid fathead minnow Eimgphalgs prgmglgg was observed from a sample of fifteen fish (Gold, 1986). These are the two reports of viable triploids in natural populations of bisexual teleosts. In hatchery stocks, one triploid rainbow trout was detected from a sample of 18 (Cuellar and Uyeno, 1972). Thorgaard and Gall (1979) found 6 of 11 full sibs, and 1 of 20 cousins of a rainbow trout family to be spontaneous triploids. Detection of triploids in relatively small samples would indicate that triploids occur naturally at some low rate. Thus, our observation of three triploids in over 350 fish sampled may correspond to their frequency in the Great Lakes population. However, in four years of ploidy analysis using flow cytometry, we have never detected any triploids from 34 approximately 200 control fish tested. The possibility also exists that the triploids were produced as a result of exposure to microwave radiation. However, their low frequency of occurrence would indicate that microwave radiation is not an effective shock treatment to induce triploidy. Triploid chinook salmon have been produced by our lab for five years using heat shocks yielding 95-100 percent triploids (Westerhof et al, 1988: see chapter IV). Microwave shocks were administered at the same time during development as heat shocks have been. However, it is possible that if administered at an earlier or later stage in development microwaves might produce triploids by a mechanism other than depolymerization of spindle fibers. It is also possible that the triploids were produced by the heat generated in the microwave oven, instead of the microwave radiation itself. The average temperatures for groups treated for 15 seconds at 100 percent power, and for 45 seconds at 50 percent power were 21 and 27 C respectively. It has been shown that one minute duration heat shocks applied ten minutes post fertilization induced triploidy in rainbow trout, with the incidence of triploidy increasing with shock temperature (Thorgaard et al., 1981). If triploids were being produced by the heat or a heat-microwave interaction, incidence of triploids should increase as a function of average temperature resulting from microwave shocks. This was not observed. 35 A final possibility is that microwave radiation does inhibit the extrusion of the second polar body; but, also disrupts normal cell structure causing mortality. Data from both 1987 and 1988 demonstrated that some eggs were killed immediately after microwave shock. Mortality data from 1987 indicated that the survival of eggs that were subjected to microwave radiation was decreased beyond the initial deaths due to the shock itself (Table 2). This would indicate that either the microwave shocks or the heat produced by the microwave had some effect on the developmental processes in the egg. Figures 5 and 6 represent the effect of shock intensity and duration. In Figure 5, mortality increases linearly with a slope of approximately one was expected in our symmetric design. Figure 6 however does not reflect the expected result. Mortality increased at a lower rate than expected. This would indicate that for microwave shocks, the intensity of the shock has a more dramatic effect on egg survival than duration of shock. We had hoped microwave radiation could be used to induce triploidy. However, our data suggests that it is not effective for polyploid induction when administered ten minutes after fertilization. Based on our experience, 28.5 C heat shock applied for 10 minutes, 10 minutes after fertilization is the optimal method for producing large numbers of triploid chinook salmon (Westerhof et al., 1988). CHAPTER III EVALUATION OF GROWTH DIFFERNTIALS IN DIPLOID AND TRI PLOID CHINOOK SALMON ABSTRACT The use of triploid fish in both aquaculture and management applications has increased dramatically in the past years. As this use continues to increase, more information is needed concerningthe differences and similarities between diploid and triploid individuals. Conflicting reports have been submitted regarding the growth of triploid individuals. An experiment was designed to measure the growth of diploids and triploids reared seperately and in competition, at two feed rations. Twenty fish, either 100 percent diploid, 100 percent triploid or a mixture of 50 percent of each ploidy type, were reared in 110 L aquaria. Three replicates of each ploidy combination were fed rations at either two or four percent body wet weight daily, adjusted at two week intervals. The study continued for eight weeks. No significant difference (p<0.05) in growth rate was detected between treatments within a feeding regime. Feed conversions calculated for tanks containing a single ploidy level did not differ significantly (p<0.05) between diploids and triploids or feeding levels. INTRODUCTION There is significant potential in using sterile triploid fishes in aquaculture and management (Donaldson and Hunter, 1982: Purdom, 1983; Thorgaard, 1983: Benfey and 36 37 Sutterlin, 1984). Sterility which results from induced triploidy could eliminate undesirable traits associated with the onset of sexual maturity, including decreased flesh quality, decreased growth rate, and precocial maturation. However, before triploids can be used on a wide spread basis, more information must be compiled concerning potential differences in performance between diploids and triploids (Thorgaard, 1983). Theoretically, triploids and diploids should grow at similar rates until the onset of sexual maturation (Purdom, 1983; Thorgaard, 1983). Studies have indicated that there is little or no difference between the growth of juvenile diploid and triploid stickleback (Swarup, 1959), rainbow trout (Thorgaard and Gall, 1979) carp (Gervai et al., 1980), plaice x flounder hybrids (Lincoln, l981),coho salmon (Johnson et al., 1986) and chinook salmon (Westerhof, 1988). However this may not be the case for all species under all experimental conditions. Triploid Atlantic salmon were found to be significantly longer than diploids but did not differ in weight (Benfey and Sutterlin, 1984). Other studies have indicated that diploids and triploids may grow at different rates. Rainbow trout triploids reared in fiberglass tanks were smaller than diploids (Solar et al.1984). Cassani and Caton (1986) found triploid grass carp (Qtenonharxngedon idella) grew slower than diploids if reared together in ponds, but no difference if they were 38 reared separately in indoor greenhouse tanks. Triploid channel catfish have been shown to be heavier than diploids at 8 months of age (Wolters et al., 1982) and triploid tilapia were larger than diploids at 14 weeks of age (Valenti, 1975). Triploid hybrid grass carp (Qtengpharynggdgn idella x Ernenhthalmucthxs.nebilis) were shown to grow faster than diploids (Cassani et al., 1984) as were triploid fancy carp produced by crossing two strains (Taniguchi et al., 1986). Both authors however, indicate that the increased growth rate may have been the result of genetic differences in parents. It has been suggested that the sterility of triploids, not the triploid condition is responsible for some of the observed differences in growth (Thorgaard and Allen, 1987). Comparisons of growth rates between diploids and triploids are based on fish maintained in similar, but not identical environments. Some comparisons that have been reported are based on observations made as part of other studies. In some studies the feeding rates, fish size and age, water temperature, and descriptions of rearing facilities have been omitted making replication of the experimental design impossible. Variation in these conditions between treatments and studies could lead to significantly different results in the same species. Finally, the comparison of fish held in separate environments gives no information concerning diploid-triploid interactions. Although neither diploids or 39 triploids should have a competitive advantage, anecdotal observations have indicated the possibility of differential performance between diploids and triploids. Our study was designed to compare the growth of triploids and diploids alone or in equal proportions in the same tank. Fish were fed at fixed rates that were maintained throughout the study and adjusted periodically to compensate for growth. This design allows not only the comparison of growth rates, but also provides the first available information on potential competitive interactions between diploids and triploids. MATERIALS AND METHODS In March of 1989, diploid and triploid chinook salmon used in the growth study were randomly sampled from fish reared by the MDNR for the Great Lakes stocking program (see Chapter III) at the Wolf Lake Hatchery. On April 7, 1989 the flow cytometery was used to confirm ploidy (see Appendix 1). The heat-shocked fish tested were 95 percent triploid from a sample of twenty fish. Diploids were marked using a right pectoral clip, while triploids were marked with a left pectoral clip. Prior to fin clipping, fish were anestisized using tricane methane sulfonate (Finquel MS-222, Argent Chemical Laboratories) at 5 grams per gallon. All fish were reared in 110 L aquaria as described in Chapter I. One of the six ploidy-food ration combinations 40 was assigned to each of 18 tanks, resulting in three replicates for each combination. Diploids were anestisized, weighed and assigned to the appropriate tank. Ten triploids were then collected at random and weighed. These ten fish were assigned to the mixed tank containing the ten diploids that were most similar in weight. Triploids were then assigned to those tanks designated as all triploid. This process was followed to reduce the possibility of a competitive advantage for either ploidy type due to initial size differences. Fish were fed at either 2 percent or 4 percent wet body weight daily on a wet feed basis (approximately 1.5 percent and 3 percent dry weight basis). Prior to the onset of each day's feeding, tanks were siphoned to remove fecal material, uneaten food, and any other material in the bottom of the tank. A net was placed over all tanks to prevent fish from jumping out of the tanks. Feeding began the day following the initial weight samples. Fish were fed 14mm biodiet three times a day to total the percent body weight assigned to each tank. A minimum of three hours separated each feeding. Fish were weighed at two week intervals, to adjust the amount of feed being fed. Tanks that included both diploids and triploids were sorted, and weighed separately. To facilitate handling and reduce stress, all fish were anesthesized before weight samples were taken. Fish were not fed on the day they were weighed. After fish had been weighed, they were held in 41 fresh water while their tanks were scrubbed to remove any algae that had accumulated on the side of the tanks. The fish used in this study were known to have bacterial kidney disease (BKD) (John Gnath, MDNR fisheries pathologist, pers. comm.). During the study nine fish died exhibiting the signs of BKD. Since BKD is an inherited disease, and the deaths were not associated with the treatments, these fish were replaced by a fish of equal weight which had been held in separate tanks and fed the same diet. Handling and the siphoning of tanks each resulted in the death of one fish, which were also replaced with appropriately sized fish from the axillary tanks. No fish mortalities were observed that were not attributed to handling, tank siphoning, or BKD. After an eight week period, fish were maintained without feed for two days to insure all feed had been digested and waste expelled. Fish were then separated and weighed as described for biweekly weight samples. Linear regression analysis was used to find the best fit line to describe the growth rate of each tank. The slope of the regression line was used as the growth rate for each tank. A one way analysis of variance was used to test for significant differences of the mean growth rate of each ploidy combination for both feed rations (Gill, 1978). At the conclusion of the study, feed conversion was calculated for those tanks containing only one ploidy level. Feed conversion is defined as weight gained/weight of feed 42 fed. It was impossible to calculate feed conversion in mixed tanks since the amount of feed eaten by each ploidy type could not be measured. A two way analysis of variance was used to test for differences between feeding rations, and difference among treatments. RESULTS No significant difference in growth was found between ploidy combinations at each feeding ration (p<0.05). As expected, fish fed 4 percent body wet weight daily grew significantly faster at all ploidy combinations than fish fed 2 percent body wet weight daily. Feed conversion was calculated for those tanks that contained only one ploidy type. No significant difference (p<0.05) was found among treatments at either feed ration nor was there a significant difference between feed rations (P<0.05). Figures 7 and 8 are graphs of average daily weight versus time. Most of the growth curves for individual tanks were correlated in a linear response. In no case was r2 found to be less than 0.920. This would indicate that the average slope of the regression lines for each treatment should be an accurate estimate of the true growth rates. DISCUSSION We predicted that juvenile diploid and triploid chinook salmon should grow at similar rates either alone or 43 .aswuoovn A OHH mamcam a Ca ouxfis no >amucummom cacao aafloo u20wms xoon um: ucuuumo 03D cum cOEHem xoocwco cacaowuu one cacaoflo no nusouo .h musmfim m>