ma INFLUENCE or SALlNiTYON ? PROTEIN REQUIREMENTS AN0 - UTILIZATION BY RAINBOW TROUT _.- :1}?j;:;f:_;; V 1:; 'SALMO GAIRDNER! AND com SALMON ?.-_._”i ; i-:,;,\j;;1; jg; OHCGRHYNCHUS «130100 .I' j. . mass for the Degree of Ph D 5 ' MECHTGAN STATE UNIVERSITY . 2997355; IBRAHIM H ZEITOUN . 1973 = ............... LIBRARY Michigan State UniverS fv - fl, v.91-.. This is to certify that the thesis entitled THE INFLUENCE OF SALINITY ON PROTEIN REQUIREMENTS AND UTILIZATION BY RAINBOW TROUT, SALMO GAIRDNERI AND COHO SALMON, ONCORHYNCHUS KISUTCH presented by IBRAHIM H. ZEITOUN has been accepted towards fulfillment of the requirements for Ph.D. . . ' . degree in _F__her_les & wildlife L52; gm Major professor Date April 9, 1973 0-7639 ABSTRACT THE INFLUENCE OF SALINITY ON PROTEIN REQUIREMENTS AND UTILIZATION BY RAINBOW TROUT, SALMO GAIRDNERI” AND COHO SALMON, ONCORHYNCHUS KISUTCH BY Ibrahim H. Zeitoun Rainbow trout, Salmo gairdneri andcxflu: salmon, Oncorhynchus kisutch juveniles maintained at 10 p.p.t. or 20 p.p.t. salinity were used in this research. Seven diets were tested from 30 to 60 per cent protein in five per cent increments. Gram weight gain, protein retention, total diet efficiency, protein efficiency ratio (P.E.R.) and net protein utilization (N.P.U.) methods were employed to evaluate the dietary protein levels. All the methods except that of total diet efficiency indicated that the minimum requirement of protein in the diet for rainbow trout at 10 p.p.t. and the coho salmon maintained at either salinity was approximately 40 per cent. Rainbow trout fingerlings at 20 p.p.t. exhibited a higher require- ment (45 per cent). Total diet efficiency values reached a maximum for all tests at 50 per cent dietary protein. Salinity effects in coho salmon were negligible when the Ibrahim H. Zeitoun fish were acclimated to the high salinity and salinity effects were not apparent if the osmoregulatory mechanism(s) were already developed. The techniques that have been established by nutritionists for estimating protein re- quirements of mammals and birds were successfully em— ployed for fish. Analysis of variance and Duncan's multiple range tests were used to determine the signifi- cance of the average values. THE INFLUENCE OF SALINITY ON PROTEIN REQUIREMENTS AND UTILIZATION BY RAINBOW TROUT, SALMO GAIRDNERI AND COHO SALMON, ONCORHYNCHUS KISUTCH BY Ibrahim prZeitoun A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Fisheries and Wildlife 1973 Dedication To my beloved country, Egypt and to my parents. ACKNOWLEDGMENTS I wish to express my gratitude to the following: my major professor, Dr. Peter Tack and Committee Members Drs. Duane Ullrey, Eugene Roelofs and Wayne Porter for their advice, patience and encouragement;Ihz John Halver for hhssupervision and useful suggestions concerning this work; Dr. William Magee for his valuable help regarding the statistical analyses and presentation of the data; Dr. Werner Bergen for reviewing the manuscript and criticism. My thanks are also due to all the personnel at Bowman Bay Field Station of the Western Fish Nutrition Laboratory, particularly Mr. Clarence Johnson, Mr. Duane Gahimer, Mr. Alton Anderson and Mr. Fred Hunger for their sincere assistance and friendship during the course of the work; and the Ford Foundation for supporting me through a scholarship that made this work possible. iii INTRODUCTION . . MATERIALS AND METHODS RESULTS . . . . Weight Gain Methods Gross (or total) Diet Efficiency TABLE OF CONTENTS Protein Efficiency Ratio . Biochemical Analysis Protein Apparent Net Retention Maintenance Mortality . Hematology . DISCUSSION . . . SUMMARY . . . . LITERATURE CITED APPENDIX . . . . of the Carcass Utilization and iv Page 20 20 31 36 38 42 50 51 52 55 78 80 87 Joe. ‘ -..’ Composition Composition Composition Composition Initial weights of coho salmon and rainbow LIST OF TABLES trout fingerlings in grams . . . . Average daily gain and average daily feed of different formulated diets of the mineral mixture (premix of the vitamin mixture (premix of the oil mixture (premix #4) #5) #1) 10 ll 22 for rainbow trout for the ten week period at two week intervals . . . . . . . . . . . Average daily gain and average daily feed for coho salmon for the ten week period at two week intervals . . . . . . . . . . . Initial and final specific growth average rate of the ten weights and the rainbow trout at after 10. ll. 12. 10 p.p.t. Initial and final specific growth average rate of week period weights and the rainbow trout at 20 p.p.t. after the ten week period . . Initial and final average weights and the Specific growth rate of coho salmon at 10 p.p.t. after the ten week period Initial and final average weights and the specific growth rate of coho salmon at 20 p.p.t. after the ten week period Diet efficiency coefficients of rainbow trout and coho salmon fingerlings as influenced by the dietary protein level at the end of 10 weeks 0 O O O 29 3O 3O 35 Table Page 13. Protein efficiency ratios of coho salmon and rainbow trout fingerlings fed different levels of protein . . . . . . . . . 37 14. Body composition of rainbow trout and coho salmon fingerlings after 10 days of adaptation to 10 p.p.t. or 20 p.p.t. sea water at the beginning of the experiment . . . 39 15. Major body constituents of rainbow trout fingerlings maintained at 10 p.p.t. or 20 p.p.t. and fed different percentage of protein . . . . . . . . . . . . . . . . . . 4O 16. Major body constituents of coho salmon fingerlings maintained at 10 p.p.t. or 20 p.p.t. and fed different percentages of protein . . . . . . . . . . . . . . . . . . 41 17. Apparent N.P.U. values of rainbow trout fingerlings fed different levels of protein and maintained at two different salinities [MSE = l] . . . . . . . . . . . . . 47 18. Apparent N.P.U. values of coho salmon fingerlings fed different levels of protein and maintained at two different salinities [MSE = 13] . . . . . . . . . . . . 47 19. Total serum protein and hematocrit as influenced by salinity and dietary protein level for rainbow trout and coho salmon . . . . . . . . . . . . . . . . . . . . 53 20. Protein consumed per day in grams per kilogram of gain by fish maintained at 10 p.p.t. or 20 p.p.t. and fed with varying protein concentration . . . . . . . . 63 vi LIST OF FIGURES Figure Page 1. Standard curve relating salinity and electrical resistance of water at 10 C . . . . 12 2. Illustration of the cones in which the fish were maintained during the test. Water volume was 25,380 m1 and average water flow rate was 1300 ml/minute with the arrows indicating the flow direction . . . 15 3. Nitrogen growth index of the rainbow trout over a period of 10 weeks . . . . . . . . . . 23 4. Nitrogen growth index of the coho salmon over a period of 10 weeks . . . . . . . . . . 23 5. Relationship between gross efficiency and percentage protein in the diet for rainbow trout maintained at 10 p.p.t. and 20 p.p.t. salinity . . . . . . . . . . . . . . . . . . . 33 6. Relationship between gross efficiency and percentage protein in the diet for coho salmon maintained at 10 p.p.t. and 20 p.p.t. salinity . . . . . . . . . . . . . . 33 7. Apparent net protein utilization (N.P.U.) in relation to dietary protein levels for rainbow trout and coho salmon . . . . . . 45 8. Relationships between protein retention and dietary protein levels for rainbow trout and coho salmon maintained at 10 p.p.t. and 20 p.p.t. salinity . . . . . . . . 48 9. Weekly thermal range for the rainbow trout at 10 p.p.t. and 20 p.p.t. salinity . . . . . 6O 10. Weekly thermal range for the coho salmon at 10 p.p.t. and 20 p.p.t. salinity . . . . . . . 6O vii INTRODUCTION Fish nutrition is a developing branch of fisheries science which is concerned with the determination of nutri- tional requirements of fish for growth and other life processes. Included in this science is the interaction of these requirements with the various biotic and abiotic factors of the environment. Studies of the effect of en- vironmental variables on body composition and quantita- tive nutritional requirements of fish are limited. Halver (1957) developed a test diet that could maintain fish without symptoms of nutritional deficiency. Brett (1971) conducted an experiment which established that Halver's diet (50 per cent protein) had the highest efficiency of the diets tested and that the protein-energy ratio (P/E) was more favorable than many other diets used in various laboratories to maintain and rear trout and salmon. This diet was modified and used satisfactorily in a series of projects aimed at determining the protein requirements of sockeye salmon, Oncorhynchus nerka, and rainbow trout, Salmo gairdneri (Halver, et al., 1964) and the Chinook salmon, g. tschawytscha (DeLong, et al., 1958). Recently Cowey, et al., (1972) estimated the protein requirement of the young plaice, Pleuronectes platessa using freeze- dried cod muscle and shrimp meal. The essential amino acid requirementsflgommmmu .onm .mmm who mam>ma camponm mqflmmmuonfl on“ mcflpfl>oum mpmflo onu OD Uaommounoo SOHQB mnmnfidc “map muoumnonmq GOHMHHDDZ nmflm anmgmmz mQBH maa muH mua mna NHH Ana HHH comm Umnmflcflm CH - Hmuw3 "HGDDME who we N» an AA mm NHH Hma N eflmnona Axmx\amox LOHDMH chMOHmIMHHoHMU o.ooa 0.00H 0.00H o.ooa o.ooa o.ooa 0.00H o.¢ o.v o.v o.w o.v o.v o.v muspxflfi Hmnmcflz o.m o.m o.m o.m o.m o.m o.m muspxflfi HHO o.m o.m o.m o.m o.m o.m o.m ouspxfla aHEmpH> o.wH o.mm o.mm o.mm o.mm o.mv o.mw cflupxmo w.wa N.ma o.mH m.oa m.m v.m N.n caveamw m.mv m.Hv o.mm m.wm v.om o.mm m.mm Gammmo om mm om ms ow mm om eoHnHmOQEOO oflmmm aw .Goflumupcmocoo sampoum auspmflo mpmwp UmDMHSEHOM psmHmMMHo mo QOHDHmOQEOUII.H manna percentage of water in the diet is high. In addition, better physical structure of the diets was maintained if the water content of such diets was lowered. The use of purified diets made it possible to change the concentra- tion of a given constituent with a minimum disturbance of other nutrient relationships. Protein was supplied as casein and gelatin, car- bohydrates as dextrin, and fat as a mixture (premix no. 4, Table 2) of corn oil, cod liver oil and alpha-tocopheryl acetate. Minerals were supplied as reagent grade salts (premix no. 5, Table 2), and vitamins as pure crystalline compounds (premix no. 1, Table 3). Two salinities were used in the experiment, 10 p.p.t. and 20 p.p.t. Salinity was determined by using the conductivity method and plotting the curve of known salinity waters and their corresponding electrical re- sistance at known temperatures (Figure l). The electrical resistance of water was checked twice daily to adjust the flow rate of the saline and fresh water supply going to the mixing cones as was necessary to maintain the re- sistance within :_4.0 ohms of the estimated value of the curve. The curves were replotted as needed because of the effect of temperature on water conductivity. Water conductivities were determined using the YSI Model 311 conductivity meter. 1Yellow Springs Instrument Co., Yellow Springs, Ohio. Table 2.--Composition of the mineral mixture (Premix #5) Ingredient Grams * Salt mixture no. 2 USP X111 100.000 AlCl3 . 6H20 0.015 K1 0.015 CuCl 0.010 MnSO4 . H20 0.080 CoCl2 . 6H20 0.100 ZnSO4 . 7H20 0.300 *Salt mixture ingredients (Nutritional Biochemi- cals Corp., 1972 Diets manual, page 6, Cleveland, Ohio). CaH4(PO4)2 . H20 CaC3H503 . 5H20 Fe C6H507 . 3H20 MgSO4 KH2P04 Na2H4(PO4)2 . H20 NaCl Ingredient Calcium biphosphate Calcium lactate Ferric citrate Magnesium sulfate Potassium phosphate (dibasic) Sodium biphosphate Sodium chloride Percentage 13.58 32.70 2.97 13.20 23.98 8.72 4.35 10 Table 3.--Composition of the vitamin mixture (Premix #1) Ingredient Grams Alpha-cellulose 8.000 Choline chloride 0.500 Inositol 0.200 Ascorbic acid 0.100 Niacin 0.075 Calcium pantothenate 0.050 Riboflavin 0.020 Menadione 0.004 Pyridoxine . HCl 0.005 Thiamine chloride . HCl 0.005 Folic acid 0.0015 Biotin 0.0005 Cyanocobalamin* 0.500 ml *Premix # 2 Vitamin B12 10 mg/500 ml H20 11 Table 4.--Composition of the oil mixture (Premix #4) Ingredient Grams Corn oil 7.0 Cod liver oil (3139 USP units vitamin A) 2.0 D,L-Alpha-tocophery1 acetate* 0.04 *Must be warmed in water bath before using. 12 Figure 1.——Standard curve relating salinity and electrical resistance of water at 10 C 13 30 15 2 2 x 10 6 2 P. 0. + 15 x I 6 0. 7 ..m . 7 2 m -5 .- V: F _ _ _ _ o O O 0 O 8 7 6 5 4 w 2:3 .mozfimamm $23 44253.6 SALINITY, p.p.t. 14 Prior to the mixing of fresh and sea water, each was run through a "Crystaleen" fiberglass filter tank Model C-72 to remove Suspended materials. Filtered water was then exposed to ultra violet light in an "Aquafine" liquid sterilizer Model MP 4 PVC3 to inhibit bacterial growth and minimize disease problems. Fish were treated twice with 1 ppm Furanase4 in the first week of the experi- ment in order to control infection and cure the fish of Sporocytophaga s22, Treatments were applied twice for one hour. Water flow was shut off during each application. Duplications of the Gahimer, gt_al., (1971) cones were available for each tested diet at each salinity. The cones were modified in the coho salmon experiment to re- ceive two water inlets instead of the one originally pro- posed (Figure 2). The modification was necessary to eliminate the complete dependence on one inlet which sometimes became blocked, causing the accidental suffoca- tion of fish, particularly on stormy days when filtration efficiency dropped. In addition, an air pump was installed to supply additional air to the mixing cones. Water tem— perature was recorded daily from one cone in each salinity 2Howard Construction Corp., 4547 N. Scottsdale Road, Scottsdale, Arizona. 3Aquafine Corp., 1230 Sunset Boulevard, Los Angeles, California. 4Dainippon Pharmaceutical Co., Ltd., Research Laboratories, Osaka, Japan. 15 Figure 2.--Illustration of the cones in which the fish were maintained during the test. Water volume was 25,380 m1 and average water flow rate was 1300 ml/minute with the arrows indicating the flow direction l6 35.6 cm 45.7 cm WATER INLET ——>T WATER OUTLET RUBBER SEAL WIRE BASKET AL TO DRAIN 17 level. Photoperiods and light intensity were maintained uniformly by placing a lO-watt light bulb over each cone. Fish were subjected to nine hours of illumination daily. A constant and uniform counterclockwise flow rate in all cones was maintained. Depending on fish size, each cone was stocked initially with 50 rainbow trout or 40 coho salmon fingerlings. Fish were acclimated to the new en- vironment for 10 days during which time they were fed twice daily the same diet to be used in the tests. The construction of the cones allowed waste products to be easily eliminated. Diets were prepared biweekly according to the Halver (1957) method and frozen until needed. Each diet was grated and fed twice a day, ad libitum, on a rigid time schedule. Grams of food consumed biweekly by each fish lot were determined. A total of ten weeks was used for each feeding experiment. Each diet was fed slowly to eliminate leaching of water soluble constituents and feeding stopped as soon as any portion reached the cone's basket. Dead fish were removed and mortality recorded daily. Weighing and counting of fish took place biweekly. The weighing technique followed that of Halver (1957). Diets and collective fish weights for each cone were measured to the nearest gram. Average fish weights were then estimated to the nearest tenth of a gram. 18 At the start and termination of each experiment, fish were starved for 48 hours after which representative samples from duplicate cones were taken for chemical analysis of the body. After the last weighing period, fish were fed once and blood samples were collected and hematocrit values were measured twelve hours later. This was done to eliminate changes which might have taken place in the blood due to the influence of starvation, digestion or absorption. Blood was pooled in test tubes by severing the caudal peduncle of several fish to collect a suffi- cient amount. The blood was then centrifuged for 15 minutes and the serum was transferred to screw-capped plastic tubes and stored in a freezer for total serum protein determination. Very little or no hemolysis was observed. Body composition determinations for initial and final samples started by thawing the fish and cal— culating the wet weight for each fish. 'Fish'were'placed into tared large-mouth bottles and dried in ovens at 95 C to a constant weight. The moisture and dry matter were calculated and the dry fish were blended in'a blender to form fine homogenous samples. Blending and homogenizing were facilitated by using isopropyl alcohol. Aliquots were allowed to dry in vacuum ovens at 55 C. Homogenates from the same cone were then placed together and mixed again using a mortar and pestle with isopropyl alcohol as 19 a mixing agent. Then the samples were dried again and kept in screw-capped bottles under refrigeration (2-3 C). Dried samples were analyzed for nitrogen by the micro-Kjeldahl method (A.O.A.C., 1970). Crude protein was estimated by multiplying nitrogen values by 6.25. Lipid contents were established by the Folch, gt_§1., method (1957). Ash values were determined by burning the dry samples in a muffle furnace at 500 C. Total serum protein was determined using the biuret method, applying the procedures of'O'Brien;'eE;al., (1968). Crystallized human plasma albumen5 was used as a standard. Absorption values were read on a Beckman model B spectrophotometer at 545 mu. Statistical analyseS‘were computed using the analysis of variance. 'Mea *differences were compared using Duncan's multiple range test;“Mean square errors (MSE) or standard errors (:SE) accompanied the tables and figures to identify the range of the means. 5Dade Division, American Hospital Supply Corp., Miami, Florida. RESULTS In this experiment fish were maintained at two salinities in order to determine (1) the minimum require— ments of dietary protein necessary to attain maximum weight gain and protein retention and, (2) the efficiency of growth and protein utilization. Each experiment was divided into two phases. The initial one was basically biological, involving the determination of weights of fish and diets and mortalities. The subsequent phase in— volved laboratory techniques. It is well understood that the main role of a successful dietary protein is to supply the body with a mixture of amino acids of appropriate pro- portions for maintenance and the synthesis of tissues (Maynard and Loosli, 1969). The success or failure of either species reared in each water salinity and fed various levels of protein was based on several accepted methods in animal nutrition. Weight Gain Methods The weight gain method depends on the fact that an inadequate supply of protein in the diet will reduce the weight gain or terminate growth completely.' This method is sensitive, particularly in testing individual 20 21 amino acid requirements in animals, but it has also been employed in evaluating the overall effect of protein (McLaughlan and Campbell, 1969). In these experiments percentage gain was employed instead of gram gain because average initial weights of the rainbow trout assigned to different diets were significantly different statistically (Table 5). Also, some fish nutritionists (Halver, 1957; and DeLong, §E_al., 1958 and 1962) have used the same parameter in their studies in determining some of the nutritional requirements of Oncorhynchus spp. Percentage gain was plotted against the percentage protein in the diet to determine the nitrogen growth index of each fish group (Figures 3 and 4). Nitrogen growth index is de- signated as the slope of the straight line of weight gain plotted against percentage protein in the diet (Allison, 1964). The nitrogen growth index for rainbow trout was 7.7 for those tested at 10 p.p.t. sea water and 7.6 for those at 20 p.p.t., while the nitrogen growth in— dex for coho salmon at 10 and 20 p.p.t. were 3.5 and 2.5, respectively. Minimum protein requirements were estimated by using the method of Fisher, gt_§1., (1957). A regres- sion line (Y = aX + b) was fitted to the data over the as- cending portion of the percentage gain response by the method of least squares. A straight line parallel to the x-axis was established by averaging the highest observed means which did not differ from each other as determined 22 ucmummmflo an .pcmHmMMHU Amo.o v mv magaMOHwHamHm mum mBQHHomemsm UmBOHH0m mcflpmmn Momma mEmm may Hoods mafia mEMm mnu co mcwmz n.0 00.0 00.0H 0.0a 00.0 0.0a 0.0a 0.0a 0.0a 0.0a 0.0H coeH0m onoo 00.0 0.0 0.0 0H.0 nN.0 00.0 00.0 00.0 00.0 000.0 00.0 usouu zoncfl0m 00H om . 0H mmH 00 00 00 00 00 mm om mmflommm .u.m.m .myflcflamm w.GOHuMHmeocoo zflmgoum %HMD0HQ mfimnm cfl.mmnflaummcflm psoup BOQCHMH paw Gofiamm onoo mo mpnmflmz HmeHcHII.m GHQMB 23 Figure 3.——Nitrogen growth index of the rainbow trout over a period of 10 weeks* [iSE = 10.10] Figure 4.--Nitrogen growth index of the coho salmon over a period of 10 weeks* [:SE = 2.05] *A regression line ( Y= a)(+ b) was fitted to the data over the ascending portion of the percentage gain re— sponse. A straight line parallel to the x-axis was esta— blished by averaging the highest observed means which did not differ from each other as determined by the analysis of variance. PERCENTAGE GAIN in WEIGHT PERCENTAGE GAIN in WEIGHT 24 220 fi 20 p.p.t. o—-— IO p.p.t. 8 2000 I80— I60- I40- I20— IOO- 80- 60- 4o— 20- 04,11 [11/ L 1 4 1 l 1 1 L 1 0 I0 I5 20 25 30 35 4O 45 50 55 60 60 PERCENTAGE PROTEIN in the DIET F so - F500 --',‘—O : . I0 p.p.I. O ‘f—. , 20 p.p.t. 40- 30- 20- ; 1 ’1 ’ 1 1 1 1 L 1 n 00—” 20 30 35 4o 45 so 55 so PERCENTAGE PROTEIN in the DIET 25 by the analysis of variance. The average value obtained was used as the Y determinant in establishing the position of this horizontal line. The point of intercep— tion of these two lines provided an estimate of the pro- tein requirement. The line interceptions for rainbow trout indicated a minimum requirement of approximately 40 and 45% crude protein at 10 p.p.t. and 20 p.p.t. sea water, respectively. The coho salmon line interceptions suggested that 40% crude protein in the diet is the minimum dietary protein level for fish maintained in either of the two salinities to obtain the maximum weight gain. Rainbow trout fed a 60% crude protein diet showed a decline in growth similar to that reported by DeLong gE_§1., (1958) for chinook salmon that were fed 65% crude protein at 47 F. Also, Cowey, gt_al., (1972) demonstrated that high dietary protein repressed the growth rate of plaice. The average daily gain and average daily feed for rainbow trout and coho salmon as influenced by percentage dietary crude protein and salinity are summarized in Tables 6 and 7. The average daily gain and average daily feed data resulted from calculating fish days for each two week group (Appendix A). Fish days for each group of fish was the sum of days each fish in the group sur— vived. This was necessary to minimize the errors in accounting for the loss of diet used by the deceased fish 26 usmnmwmflc an omzoaaom 000000: .DcmeMMAU Amo.o v.mv hHDCMOAMHcmHm mum mgmfluomummsm Hoflma 080m 050 Horas mafia 080m may no mammz 0.3.m .m can mom 0 omma omha ovma Dvom QmON 00mm cmmm mxom3 ad How m Ham mam m omma Omba owma Qmom amom wawm cmwm mxmmz How m mom mam m onha omha owna avom nhom wowm mahm mxmmB How 0 me «ma 5 oama Oboa omwa acmna Qoom Mmmm wavm mxmms MOM m mma OMH o QNNH nvma nmma gamma 0Coma gamma gamma mxmm3 How 08 .Ummw haflmp mmmno>¢ m mma mma m ooma Obma coma ovwa omna vaa MOHH mxmmz 0H How m mma mma m oama oama oama 0mma onna nmma 00m mxmm3 MOM v mma mma h nmmH nwwa nmma nmma Qvna avoa 00m mxmo3 How m mva mva 0H 0mmH OSSH owna 0mma coma Qwaa 0mm mxmmz Mom m mHH boa ma nmma amma nmva QHNH vaa 000 0mm mmeB “ow m0H 0m 0H 00H 00 00 00 00 00 00 00 m8 .cwmm MHHMG mmmnw>¢ .u.m.m .wuflcflamm w.coflumnucwocoo cfimwoum mumumflo mam>umucw xwm3 030 um poaumm #003 can map How usouu Boncfimu How comm SHAMU mmmnm>m tam mama >Hfl$p mmmum>¢al.w magma .u:0u0wuwp Amo.o v my SaucMOHwflamHm 0H0 mumflnomn0msm uc0n0wwflp wn ©03OHHOM mcflpm0£ HOWME 0800 0:0 H0Uaz 0GNH 0800 0:0 :0 0:002 27 0.3.0 m Hoa MONA m nvm nooa noaa avaa Dooa meH wx003 oH now 0 003 Ho 0 am maoa w awn Dam nNm Dom mam mNNH x m m 003 Ho 0 on 0mm h Dam 2mm new gun 2mm MHHH mx 0 m m m0 mm o 3N0 va 30mm Qmm now 000 mx003 0 How N mN SN m HN NN mN AN AN Hm mx003 N How we .000m maflmp 0mmu0>¢ N up 0mm m 00m oNoa 0mm Ohm Dam .Mmm mx003 ca How N 00 0H0 0 £00 nmm 3mm 0mm Mom mON 0x003 m now m mm 0mv m Qwv £00 Qmm Qmm 00H 00H 0x003 0 How 0 SN ma Ha 0N mN 0m 0m ON H 0x003 0 How 0 0| 0| 0 v: o 0 0 HNI 0H: mx003 N How mu“ om 0H mm“ 00 00 00 00 mm om ma .cflmm Sadat 0mmu0>¢ .0.m.m .mgflcflamm w .soflpmnuc0ocoo 0H0Doum mumu0flo mam>u0ucfl M003 03M um poau0m R003 :00 030 now coEHMm 0:00 How U00“ SHNBU 0mmn0>m paw qflmm haflmp 0mmn0>dll.n 0Hnme 28 as well as weight loss caused by the removal of the dead fish. The weight gain data suggested that the fish were more sensitive to the dietary protein level than to the degree of salinity. Beyond the minimum requirements of protein, the growth of fish was not significantly different in either salinity. Apparently, the rainbow trout showed a clear response to the protein levels in the diets in the first two weeks of the test but the coho did not until after the fourth week. An inverse relationship could be established between the average daily feed and the per- centage protein fed until the protein level for maximum growth was met. To provide a useful way to compare growth of different size fish, the mean specific growth rate for each of the duplicate groups that received the same treatment was used, and since specific growth rate usually takes into account the time factor (0&0, this method was recommended by Brown (1957). Tables 8, 9, 10 and 11 summarized the specific growth rate and the initial and final average weights of coho salmon and rainbow trout maintained at two different salinities. Instantaneous growth rate or the specific growth rate (GR) was derived from the equation: Table 8.--Initial and final average weights and the specific growth 29 rate of rainbow trout at 10 p.p.t. after the ten week period Dietary protein concentration, % Item 30 35 40 45 50 55 60 :SE In?t1a1 average 6.4 6 4 6 2 6.6 6 5 6 5 7 1 0.22 weight, g F1?a1 average 14.6 16.8 18.9 19.1 19.7 19.3 20.1 0.55 weight, g SPeCIflc 9rOWth 1.2 1 4 1 6 1.5 1.6 1.5 1.5 0.55 rate Table 9.--Initial and final average weights and the specific growth rate of rainbow trout at 20 p.p.t. after the ten week period Dietary protein concentration, % Item 30 35 40 45 50 55 60 :SE Initial average 6.5 7 3 6 3 6 0 6.2 6 1 7 3 0,22 weight, g FlPal average 13.8 17.1 18.1 19.4 19.6 19.5 20.9 0.55 weight, g SPe°1flC grOWth 1.1 1 2 1 5 1 7 1.6 1 7 1.5 0.05 rate 30 Table 10.--Initial and final average weights and the specific growth rate of coho salmon at 10 p.p.t. after the ten week period Dietary protein concentration, % Item 30 35 4O 45 50 55 iSE Initial average . 14.3 14.1 14.3 14.4 14.2 14.3 0.10 weight, 9 Final average . 16.6 17.5 21.6 21.8 21.5 21.4 0.35 weight, g Specific growth rate (X10) 2.1 3.0 5.9 5.9 5.9 5.8 0,20 Table ll.--Initia1 and final average weights and the specific growth rate of coho salmon at 20 p.p.t. after the ten week period Dietary protein concentration, % Item 30 35 40 45 50 55 iSE Initial average . 14.8 14.7 14.7 14.6 14.6 14.8 0.10 weight, 9 Final average weight, 9 Specific growth rate (x10) 2'3 3.2 5.0 5.2 5.6 5.1 0.20 31 log W - log W GR = e t e o At where Wt and wO were the average weights of fish at the end and beginning of the test, respectively, and At = 70 days on test. Instantaneous growth rates of both species maintained at the two different salinities gave the same results as the weight gain method. Gross (or Total) Diet Efficiency A parameter of particular interest to many fish physiologists is the gross efficiency of diet or gross efficiency of growth (Winberg, 1956; Brown, 1957; Kinne, 1960; Paloheimo & Dickie, 1966; and Brett, g£_§l., 1969). This measurement expresses the efficiency of the conver- sion of food to fish tissue. Gross (or total) diet ef- ficiency was symbolized by Warren and Davis (1967) as: E = G/I t where E is the diet efficiency, G is the t gain in weight and I is the diet intake. In these tests, the gross diet efficiency calculations were based on the wet weight of the average fish and the average dry weight of the diet as fed. Therefore, the diet efficiency ratios tended to be higher than if both gain in weight and weight of the diet ingested were 32 calculated on a dry basis (Figures 5 and 6). The diet efficiency ratios of the coho salmon and rainbow trout finger— lings established that salinity was of minor consequence while the dietary protein level was a major influence (Table 12). In contrast to the above, Kinne (1960) found that the growth efficiency of the desert pupfish, Cyprinodon macularius, increased with salinity. Over a sixteen week period and at 30 C the efficiency of food conversion reached a maximum at 15 p.p.t. salinity and decreased in the following order: 35 p.p.t. and then fresh water; and the diet efficiency ratios were 14.4, 10.6 and 8.8, respectively. The desert pupfish responded in a different fashion than did coho salmon or rain- bow trout to saline media since the pupfish is basically a brackish water species. The effect of the type of food on the conversion of dietary nitrogen to fish flesh was demonstrated by Pandian (1967a). He reported that the nitrogen conversion rate of Megalops cyprinoides was dependent on the quality of food ingested. The nitrogen conversion was 22.0% for fish fed prawn, Metapenaeus monoceros, whereas those fed mosquito fish, Gambusia affinis, converted 35.5% of nitrogen. He related this difference to the reduction of food intake on the prawn diet due to the bulk of the exoskeleton and not to its low digestibility. In the rainbow trout and coho salmon tests, where casein diets were fed, digesti- bility was assumed to be equal for the diets tested, and diet intake was not restricted. 33 Figure 5.--Relationship between gross efficiency and per- centage protein in the diet for rainbow trout maintained at 10 p.p.t. and 20 p.p.t. salinity [:SE = 0.24] Figure 6.--Relationship between gross efficiency and per- centage protein in the diet for coho salmon maintained at 10 p.p.t. and 20 p.p.t. salinity [:SE = 0. 69] GROSS EFFICIENCY (XIO) GROSS EFFICIENCY (XIO) 34 30 l l l l l 1 J 35 4O 45 5O 55 60 65 PERCENTAGE PROTEIN in the DIET l l l l l l 30 l 35 40 45 50 55 60 65 PERCENTAGE PROTEIN in the DIET .pc0H0MMHp Amo.o v mv waua000m0cmflm 0H0 mpmHHUmH0QSm 0G0H0MMHU an U030HHOM mcflp00£ HohmE 0Emm 050 H0UGD 0CHH 0600 050 Go mc0020.p.o.n.m mo.o om.o Nb.o mo.o emo.H oNo.H oom.o 0mm.o va.o 0VmN.o coEHmm 0:00 ”w Ho.o 0m.o 0m.o No.0 0vo.a 050.H 0mo.H Uom.o 00m.o Qnm.o 0mw.o usouu Boncflmm 00H 00 0H 00H . 00 00 00 00 00 mm 00 m0flo0mm .D.m.m .>UHcHHmm w.cOHp0Huc0ocoo afl0poum NH000HQ mx003 OH mo Uc0 0:0 00 H0>0H :H0uonm humu0flo 030 mn o0ocmdawafl mm mmcflau0mcflm GOEHmm osoo paw 050”“ Boncflmm mo mDC0HOHmm0oo mocmflOHmm0 #0HQII.NH 0HQMB 36 Protein Efficiency Ratio Protein efficiency ratios (P.E.R.), introduced by Osborne, et_§1., (1919), were employed in estimating minimum dietary protein requirements. P.E.R. is defined as the coefficient of grams gain per gram protein consumed. P.E.R. calculations are theoretically more precise than estimates of gross diet efficiency since dietary protein intake is the only item of the diet that was taken into account. Morrison and Campbell (1960) stated that P.E.R. for rats is a function of sex, age, genetics and length of feeding trial. Therefore, this information should be supplied in order to eliminate confusion when comparing data. In the present work, sex was difficult to determine because rainbow trout tested were yearlings while the coho salmon were slightly over a year old. Genetics of the fish would be considered uniform and the feeding trial lasted for a period of 10 weeks. P.E.R. figures were substantially influenced by the dietary protein level and not by the osmotic pressure of the medium. Table 13 presents the P.E.R. values as a function of percentage crude protein fed in the diet. P.E.R. reached the high- est values around 40 and 45% dietary protein levels for both experimental species. Approximately 40 to 45 per cent crude protein in a diet for rainbow trout or coho salmon would be appropriate to support maximum gain. The data reported by Cowey, et al., (1972) on plaice showed 37 .uc0H0MMHp Amo.o v my MHNGMONMHamHm 0H0 mumfiuomn0m50 pc0u0meU SQ UmonHOM mafipm0£ Momma 0300 0:0 H0005 0QHH 0800 030 no 0:002 w~o~gsm Ho.o mm.H N0.H HH.o 00m.a ovoN 0H0.N oONN now.a 0mm.o coaamm 0:00 No.0 mm.a 0m.H 0o.o ©0S.H 0mm.a omoN oHo.N 00H.N QNm.H 0N0.H psouu 3OQQH0m 00H 0m 01.. 00H 00 00 00 00 00 mm 00 00Ho0mm .p.m.m .mpHCNHmm w .coflumnpc0ocoo nH0pOHm mmmu0fim cfl0poum mo ma0>0a pc0H0MMH© 00m meHHH0mch usouu BOQcflmn paw coEHMm osoo mo moflpmu moc0000ww0 000poymII.mH 0Hn09 38 that protein efficiency ratio was highest at a 40 per cent dietary protein level although they estimated a minimum dietary protein requirement of 50 per cent for this species. These differences were not explained. Biochemical Analysis of the Carcass Phillips, et_al., (1957) reported that increases in fish weight were not reliable indicators of true growth. Weight gain might be a result of fat deposition or other substances not directly related to true growth. Maynard and Loosli (1969) recommended the application of a slaughter test for the experimental animal. Proximate analyses, particularly crude protein determinations, were used in order to measure protein retention and net protein utilization. Total protein retention is a better indica— tor of the growth index than weight changes (Hegsted, 1964) although both methods tend to yield the same conclusion. Table 14 summarizes the percentage composition of the major constituents of rainbow trout and coho salmon, re- spectively, after a considerable period of acclimation to the two salinities studied but before differences in dietary protein levels were instituted. The proximate constituents of the carcass at the end of the experiment are listed in Tables 15 and 16 for both species maintained at 10 p.p.t. and 20 p.p.t. sea water. All determinations are expressed on a dry matter basis. 39 Table 14.--Body composition of rainbow trout and coho salmon finger- lings after 10 days of adaptation to 10 p.p.t. or 20 p.p.t. sea water at the beginning of the experiment Salinity Constituent 10 p.p.t. :SE 20 p.p.t. :SE Rainbow Trout Dry matter, % of fresh weight 20.9 1.24 21.3 0.66 Protein, % of dry matter 69.8 0.03 71.1 0.10 Lipid, % of dry matter 22.4 0.15 20.4 0.90 Ash, % of dry matter 8.2 0.04 8.1 0.01 Coho Salmon Dry matter, % of fresh weight 23.9 0.14 25.0 0.16 Protein, % of dry matter 62.7 0.27 62.3 0.17 Lipid, % of dry matter 26.3 0.59 27.5 0.12 Ash, % of dry matter 11.1 0.17 10.4 0.07 40 02 00.0 00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 000008 080 mo 0 .:00 00.0 v m 00.0 00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 000008 080 mo 0 .00000 mz 00.0 00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 000008 080 mo 0 .8000080 02 00.0 00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 000003 00080 00 0 .000008 080 awnwuwamm 00.0 v 0 00.0 00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 000008 000 mo 0 .000 00.0 v m 00.0 00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 000008 080 mo 0 .00008 00.0 v m 00.0 00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 800008 080 00 0 .8000080 00.0 v m 00.0 00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0:0003 00000 00 0 .000008 Sun nmqmqmlmm 00800 00 00 00 00 00 00 00 00 08000000800 IHMHcmHm mm: Hmuoa . w .8000000800800 800ponm wumumflo 800uoum mo m0mmuamou0m 08080mm00 00m 080 .u.m.m ON 00 .u.@.@ OH #0 0080008008 mmcflau0mcflm usouu zoacflmu mo 008050000800 upon “ohszI.mH 0HQ09 41 .60000m m806000 >06 00 0 8o 6000m« 00.0 v m 06.0 00 o.mm m.mm ©.mm N.0m N.NN m.0m 00.0 v m mm.o >0 N.©© 6.00 0.00 m.m® m.m© 0.06 000008 >06 00 w .800 000008 >06 00 w .60m00 000008 >06 00 0 08000000 0£m003 80000 00 0 .000008 >0Q 000008 >06 00 w .804 000008 >06 00 0 060m00 000008 >06 00 0 .800000@ 08 00.0 00 0.00 0.00 0.00 0.00 0.00 0.00 080003 80000 00 0 .000008 000 00800 00 00 -00 00 00 00 00 80800 0800 10008000 008 00000 0 .0 0 .8000000800800 800000m >00000a 800000m mo 0000080000m 080000006 600 680 .0.m.m ON 00 .0.m.m 00 00 6080008008 0680000m800 808000 0800 00 008080000800 >60Q nommzlu.00 00009 42 The reported values of protein, lipids, and ash were the means of duplicate or triplicate determinations of the dry composite of all fish in each dietary treatment. Dry matter was determined on individual fish representing 10 per cent of the population in each dietary treatment. Standard error (:SE) or the mean square error (MSE) were computed to evaluate the overall range of the means. Cowey, gE_§l., (1972) studied the plaice and found that increasing dietary protein levels were associated with increasing tissue protein levels. But in the studies with rainbow trout and coho salmon, while there were significant dif- ferences between the tissue protein levels produced by different diets, there was no linear relationship between them. Apparent Net Protein Utilization and Protein Retention Two major criticisms of the P.E.R. method are that (1) it does not allow for maintenance requirements, which are neglected by assuming that all the nitrogen intake is utilized for growth, and (2) it is based on the assumption that weight gain is constant in composition, which is not necessarily valid (Bender & Doell, 1957; and Allison, 1964). Other approaches have thus been used to evaluate the ef- fect of differing dietary protein levels upon fish growth, namely net protein utilization (N.P.U.) and protein 43 retention. Average fish weights and protein values from the proximate analysis of the carcasses at the beginning and at the termination of the experiment were used to cal- culate these values. The protein retention values were based on the following equation, which is similar to the procedure used by Gerking (1952) and Pandian (1967b). body protein at ' body protein at th Protein retained = end of the test, - beginning of the = protein retained in grams Apparent N.P.U. X 100 protein consumed during the test in grams The apparent N.P.U. values in these tests were not derived from the equation used by Miller and Bender (1955) which was: N P U (bOdy N of test grouP) - (body N of nonprotein group) N consumed by test group X 100 Since the test lacked a group of fish that had been fed a protein-free diet, the values resulting from these studies should not be considered the net values of protein utili- zation but rather the apparent net protein utilization. Although N.P.U. (or apparent N.P.U.) has a theoreti- cal advantage in allowing for maintenance requirements, the data derived from studies of rainbow trout and coho salmon did not alter the conclusions resulting from 44 examinations of P.E.R. and growth index curves. Figure 7 demonstrates the effect of dietary protein on the apparent N.P.U. The values after the 40 per cent level of protein tend to plateau and exhibit only nonsignificant differences. The influence of protein levels in the diet was signifi- cant (P < 0.05) but the salinity effect was negligible. Analysis of the data indicated a significant interaction of salinity and dietary protein level upon apparent N.P.U. for the rainbow trout (Table 17) but not for coho salmon. In the case of the rainbow trout, the mean value of apparent N.P.U. at 60 per cent dietary protein was significantly lower than the preceding values. A similar decline of apparent N.P.U. was observed in coho salmon at high dietary protein levels but this difference was not significantly different from previous values. It should be noted that the highest protein level used for coho salmon was only 55 per cent and higher levels might have further depressed apparent N.P.U. values. These findings are consistent with the depression of percentage gram gain at higher protein levels and are similar to observa- tions of DeLong, gt_al., (1958) and Cowey, et_3l., (1972). In regard to protein retention determinations, Figure 8 represents the pattern of fish reaction to both diets and salinity. Although the curves showed the same response to various diets as indicated by previous methods the data were expressed in terms of gram protein gain 45 Figure 7.—-Apparent net protein utilization (N.P.U.) in relation to dietary protein levels for rainbow trout [iSE = 0.50] and coho salmon [:SE = 1.8] APPARENT N.P.U. 4O 35 30 25 20 IO 46 Rainbow trout ~fl \ \ Coho solmon‘n l I L l l l L 25 3O 35 4O 45 50 55 60 PERCENTAGE PROTEIN in the DIET 47 Talale l7.—-Apparent N.P.U. values of rainbow trout finger- lings fed different levels of protein and main- tained at two different salinities [MSE = l] Dietary protein concentration, % Salinity 30 35 40 45 50 55 60 10 p.p.t. 21.4a 26.4b 34.4C 32.4C 32.5C 33.4c 31.3C 2o p.p.t. 24.6a 25.8ac 34.3b 34.2b 35.5b 32.4b 28.4c a,b,c Means on the same line under the same major heading followed by different superscripts are significantly (P <0.05) different. Table 18.——Apparent N.P.U. values of coho salmon finger— lings fed different levels of protein and main- tained at two different salinities [MSE = 13] Dietary protein concentration, % Salinity 30 35 40 45 50 55 10 p.p.t. 16.0 24.3 35.7 34.3 34.7 33.7 20 p.p.t. 10.4 28.8 36.4 27.5 33.8 29.4 Mean 13.2a 26.5b 36.1c 30.9bc 34.2C 31.5bc a,b,c Means on the same line under the same major heading followed by different superscripts are significantly (P <0.05) different. 48 Figure 8.-—Relationships between protein retention and dietary protein levels for rainbow trout [iSE = 0.08] and coho salmon [:SE = 0.05] maintained at 10 p.p.t. and 20 p.p.t. salinity PROTEIN RETENTION,g 3.0 - 49 Rainbow trout 2.5 - 2.0 - |.5 - I.O - 0.5 — L I l 1 l A l 25 30 35 4O 45 5O 55 60 PERCENTAGE PROTEIN in the DIET 50 and thus could be used to rate the protein levels of the diets by eliminating the interference of the alterations in the body composition of the test animal. The observed- decline of the N.P.U. at high dietary protein levels might be a result of the drop in the efficiency of utilization of the protein. A portion of the dietary protein is ex- pected to be deaminated to supply the body with the necessary energy for growth and maintenance. The deamina- tion process requires a high expenditure of energy. There- fore, 50 or 55 per cent dietary protein would be the thres- hold limit of protein incorporated in a casein'diet after which any increase in the protein level will decrease the efficiency of protein utilization. Maintenance The maintenance requirement is defined as the level at which the animal will neither lose nor gain weight, in other words, when percentage growth equals zero. From this definition, an attempt was made to esti- mate the protein level required for maintenance. The nitro- gfinm growth index curves were manipulated by extending the ascending line downward until they met the x-axis which represented the percentage dietary protein at zero growth. This manipulation was based on the assumption that weight gain and percentage protein intake were directly related until the point of inflection. Hegsted 51 and Chang (1965a) related gain in weight of rats to nitro-- gen intake and found that they were linearly related from zero growth until maximum growth was reached. Although the minimum protein requirements for growth of the two groups of coho salmon at 10 p.p.t. and 20 p.p.t. sea water were not substantially different, the two groups differed slightly in the estimated maintenance protein re- quirement which was around the 25 per cent level, whereas the maintenance protein level of the two groups of rainbow trout fingerlings was close to 15 per cent crude protein. It appeared from this approach that the osmotic pressure of the medium exerted little influence over the maintenance requirement and could be neglected. Thus, estimation of the maintenance protein requirement for fish at either of the two salinities would be valid. Mortality Mortality was determined on a daily basis. Dead fish were removed at the time of their discovery. The amount of diet they used before death was subtracted from the total diet used by the whole group. This was facilitated by determining the average daily feed which is dependent on the number of fish days. Rainbow trout main- tained at 20 p.p.t. sea water showed a significantly higher mortality rate than those maintained at 10 p.p.t. and the values were 9.4 and 2.1 per cent, respectively. The coho 52 salmon mortalities were not significantly different between the two salinities, 2.1 and 2.7 per cent. The protein level in the diet had no significant effect on the percentage mortality at either salinity level. Appendix A summarizes the biweekly number of deceased fish in relation to the treatment and salinity. It was apparent that mortality started in the first week of the experiment in:spite 0f the acclimation period. Hematology In the case of rainbow trout, hematocrit deter— minations were significantly higher for fish maintained at 20 p.p.t. than those at 10 p.p.t. Total serum protein values of rainbow trout were not influenced by either salinity or dietary protein level. Contrary to these results, the coho salmon ex- hibited higher total serum protein values for those fish maintained at the lower salinity level. The mean values of total serum protein and hematocrits for coho salmon and rainbow trout are listed in Table 19. However, salinity comparisons were not complete since the fish maintained at 20 p.p.t. receiving the 30 per cent protein diet were not tested and those receiving the 35 per cent diet were suffering from hypoxia. Oxygen deficiency or the capture of fish may affect the blood composition par— ticularly the relative proportion of plasma protein fractions (Bouck & Ball, 1965 and 1966). 53 .080000006 Amo.o v 80 >008000008800 000 008000000880 080000006 >8 60300000 8806008 0on08 0800 080 00688 0800 0800 080 80 08002 m~c~0~nhm .8008 0080 00 00080000 000880 00000 0 80 60008 ..0.8.8 om 00 8000 000 6088000 003 0.0 00 0m000><0 00.0 0.0 00.0 00.0 00.0 0.0 0.0 0.0 0.0 000.0 08 000\0 .8000008 88000 00009 808000 0800 0.00 0 .0000000808 0.0 08 000\0 .8000008 88000 00009 08000 3088008 800 00 00 000 00 00 00 00 00 00 om .0.8.8 .>008000m w 88000000800800 8000008 >000008 808000 0800 680 08000 3088000 000 00>00 8000008 >000006 680 >0080000 >8 6008080080 00 0000000808 680 8000008 88000 000oa||.m0 0080B 54 Hematocrit was not determined for the coho salmon because the values would not be valid as hypoxia would increase the hematocrit values (Schiffman & Fromm, 1959). DISCUSSION The experiments were designed to provide statisti- cal parameters for determining the difference between re- sponses of fish maintained at two different salinities and also to measure the sensitivity of fish to different dietary protein levels. The results have potential use- fulness when applied to the culture of fingerlings of Salmonidae or other species in brackish water. Compara— tive methods to evaluate the nutritive value of protein were employed in order to determine the best procedure to evaluate future nutritional bioassays. Proteins are the basic constituents of the cells, tissues, and organs. They play a vital role in growth and all the regulatory mechanisms in the animal body. The minimum protein requirements for growth are important to minimize costs of fish rearing since protein is the most expensive item in formulating artificial diets. Minimal requirements of protein, as salinity in the rearing tanks changed, had not been previously determined. DeLong, gt_§l., (1958) found that a rise in temperature increased the minimal protein requirements of chinook salmon. 55 56 The amount of protein required for growth was used to support three important functions: body main~ tenance, formation and growth of tissues, and allowance- for losses in metabolism. The amount of the protein that is required for the tissue formed may be estimated from proximate analysis data (Maynard and Loosli, 1969). The output of the tests established that growth was accomplished with each level of protein in the diets. In addition, an inadequate percentage protein intake did not block growth, terminate it, or exhibit any protein deficiency symptoms in the fish. Each group responded relative to the level of protein tested as long as such level was above the maintenance requirements and the pro- tein biological value appropriate. The percentage weight gain and protein retention methods established that salinity influenced the protein requirements of rainbow trout fingerlings but not those of the coho salmon. In the case of the rainbow trout, the percentage of dietary protein for maximum growth in- creased as the salinity increased. The problem was to determine with accuracy the minimum requirement of pro- tein which produced the maximum weight gain or protein retention. Therefore, the values given for the protein requirement were approximated to the nearest dietary pro- 'tein level tested. 57 The rise of protein requirements with salinity seemed to be a function of the osmotic pressure of the medium. The importance of proteins in regulating the-in- ternal fluids of the animals against a hypertonic medium through the binding of some ions in indiffusible com- plexes and thereby inducing a Donnan equilibrium was dis- cussed by Potts and Parry (1963). However, this mechanism accounted for only a small part of the overall mechanisms of hyperosmoregulation. Concurrently, the rainbow trout at 20 p.p.t. salinity showed a substantially higher percentage of mor- tality than those at 10 p.p.t. Apparently, rainbow trout fingerlings at higher salinities expended more energy to maintain their osmotic equilibrium, resulting in greater stress and frequently death. In the case of the coho salmon, the stress induced by the two salinities studied was minimal and mortalities were low and not statistically- different. Otto (1971) found that the highest growth values of coho salmon juvenile were obtained at salinities of 5 to 10 p.p.t. throughout the presmolt period. Canagaratnam (1959) stated that coho salmon fry grew faster in water of 12 to 18 p.p.t. than in fresh water. Otto and McInerney (1970) concluded that the preferred Salinity for the presmolts of coho salmon increased from 7-8 p.p.t. in June to 13-14 p.p.t. in February. Salinity irreference was determined by introducing a salinity 58 gradient and observing turning frequency as the gradient approached the fish and by observing evidence of an alarm response. The results of Kepshire and McNeil (1972) and" Bullivant (1960) showed that chinook salmon fingerlings grew better at intermediate salinities (17-18 p.p.t.) and in fresh water than in 100% sea water. These authors attributed the poorer performance in 100% sea water to the greater expenditure of energy by the fry to maintain osmotic equilibrium. Conte and Wagner (1965) and Conte, et_al., (1966) reported that seaward migration of steel- head trout and coho salmon is preceded by sea water adapta- tion. Also, euryhalinity mechanisms developed earlier in coho salmon than in steelhead trout. These mechanisms were a function of size and independent of age. Therefore, the increased dietary protein requirements with increased salinity (based on prevention of mortality) may be neces- sary to overcome the stress of maintaining osmolarity of body fluids since weight gain and protein retention of the two groups maintained in the two salinities did not differ substantially. Stress diminished as the fish grew larger and this was demonstrated in the coho salmon which under- went smolt transformation. The initial weights 0f the coho salmon were more than twice those of the rainbow trout and the initial total lengths were considerably greater in the case of the coho salmon (11.2 i 0.3 versus ‘7.5 i 0.2). There was no gain in weight until after the 59 second week of the test for the coho salmon in spite-of the availability of food. This may be attributed to the temperature influence on appetite described by Brett, gt_al., (1969) with sockeye salmon. Reduction of metabolic rate of poikilothermic animals was directly related to temperature (Fry, 1957). However, these results agreed with those of Vanstone and Markert (1968) who noticed that laboratory-reared coho salmon grew in a steadily de- creasing rate in early winter. Therefore, the timing of the test was not pr0per from the standpoint of supporting optimum growth and late winter or spring would be more appropriate seasons because the experiments were dependent on natural water and consequently on natural water temper- atures. The temperature gradients were plotted weekly (Figures 9 and 10). Unfortunately, a precise comparison of the two species is not possible because of the different age groups and because the two experiments were conducted under completely different thermal conditions. .Temperatures of the water for the coho salmon were more or less close to those of the rainbow trout only after the fourth week of the experiment. The weight gain and protein retention data indi- cated that beyond the minimum protein required for maxi— Inum gain in weight and protein retention, the values .leveled off or decreased, without significant differences kbetween dietary protein levels. The biological capability 60 Figure 9.--Weekly thermal range for the rainbow trout at 10 p.p.t. and 20 p.p.t. salinity Figure 10.--Week1y thermal range for the coho salmon at 10 p.p.t. and 20 p.p.t. salinity TEMPERATURE (°c) TEMPERATURE (°C) 61 9 8 7 6 I I l 1 l l l l J I 2 3 4 5 6 7 8 9 IO N0. of WEEKS l2 - II P- 4 9 .— 5 4 I2 - IO p.p.t. 5 3 # N0. of WEEKS L 1 I bi- m)- (DP 62 to achieve the maximum weight gain and protein retention appeared to be species specific and environmentally de- pendent. Gerking (1952) termed the point at which no further growth would take place as the growth potential of the species. Pandian (1967b) related the same term to the protein conversion rate value beyond which the values declined. Protein retained in each fish is con- sidered as the net protein resulting from the difference between protein consumed and protein excreted. Warren and Davis (1967) designated the same caloric difference as the scope for growth. The daily consumption of protein in grams per kilogram of weight gain by rainbow trout and coho salmon fed diets with varying protein concentration is shown in Table 20. The data show that maximum gain with minimum protein intake was achieved by both species on diets con- taining 40 per cent protein. The values were 477 and 465 g protein/day/kg gain for rainbow trout and coho salmon, respectively. At dietary protein concentrations of 35 and 30 per cent, appreciably greater amounts of protein were consumed per day per kg of weight gain. This is a reflection of both a higher feed intake on low dietary protein concentrations and poorer efficiency of dietary protein utilization. For rainbow trout raised in 20 p.p.t. salinity, the minimum protein intake per kilogram gain was achieved on the 50 per cent protein diet. However, 63 Table 20.——Protein consumed per day in grams per kilogram of gain by fish maintained at 10 p.p.t. or 20 p.p.t. and fed with varying protein condentration Dietary protein concentration, % 30 35 40 45 50 55 60 Rainbow Trout 10 p.p.t. 675 597 459 514 497 517 568 20 p.p.t. 738 635 495 483 471 510 581 Mean 704 616 477 499 484 514 575 Coho Salmon 10 p.p.t. 1336 846 507 510 505 534 20 p.p.t. 1117 616 422 489 475 538 Mean 1227 731 465 500 490 536 64 this value was not significantly different from those on 45 or 40 per cent protein. These results agreed with those obtained from the nitrogen index curves which showed that the rainbow trout exhibited a higher protein requirement at 20 p.p.t. than at 10 p.p.t. Total diet efficiency values responded directly to the quantity of protein in the diets. They increased7 as the amount of protein intake increased; These results coincided with the findings of Pandian (1967a) who at- tributed the decline in the conversion efficiency to the reduction in the quantity of food intake. The salinity had a nonsignificant effect on the efficiency of food conversion.‘with a.tendency for higher values for those fish maintained at a higher salinity. Kinne (1960) re- ported that growth rates of the euryhaline teleost Cyprinodon macularius and efficiencies of food conversion were higher at 35 p.p.t. than in fresh water, while Otto (1971) found that the growth rates, food intake, and gross food conversion efficiency had the highest values at salinities of 5 to 10 p.p.t. throughout the presmolt period of the coho salmon. The efficiency of food conver— sion is dependent on many factors and general statements should not be made. Conditions favorable to one fish would not necessarily be favorable to another. The best response is believed to depend on the environment to which the fish has been adapted and in which it has been reared. 65 Higher food conversion efficiency took place in fish hatched and grown in the same salinity than in fish trans- ferred into another salinity after hatching (Kinne, 1962). Maximum gross efficiency of the diet output in the case of rainbow trout and coho salmon fingerlings was identi- cal. The maximum was reached at 50 per cent dietary pro- tein. This implied that diet no. 8 would be the most favorably balanced dietin.spite of the indications that maximum weight gain, specific growth rate and protein re- tention were reached below the 50 per cent protein level. Brett (1971) obtained 48 per cent for the gross conversion of Halver's (1957) 50 per cent protein diet, which was the highest among many other diets that were compared. Although the available data were meaningful, they could not be compared with other data simply because the cal- culations of the coefficients were based on the wet weight gain of the fish and the dry weight of the ration. Since body composition is dependent on the food quality, physio- logical status of the animal, growth rate, and environ- mental factors (Love, 1957 and 1970; Maynard and Loosli, 1969), water content of the gain was variable. Therefore, the dry weights of fish gain were recommended and used by Brown (1957), Kinne (1960 and 1962), Otto (1971), and Brett, gt_al., (1969). The application of wet weight of fish gain was for three reasons: (1) the results could be applied by the common fish farmers and culturists with 66 no laboratories as a quick and practical method to com- pare the efficiency of diets; (2) the evaluations of the overall dietary protein and protein intake were conducted using more precise methods whose values were presented earlier in the text; and (3) the method and the formula were already in use at the Western Fish Nutrition Labora— tory. Dry weights of gain could be calculated by multi- plying the average dry matter of the body composition by the average weight of the fish at the beginning and at the end of the test. It was clear that by increasing the level of pro- tein of the diet above maintenance level, the diet effi- ciency coefficients increased, approaching a maximum level at which the efficiency stabilized or declined. It has been previously stated that the quantity of the diet con- sumed was related to the gross efficiency of growth, reaching a maximum value at the maximum diet intake when the net efficiency was constant (Warren and Davis, 1967). Paloheimo and Dickie (1966), after reviewing some valuable growth articles, stated that the logarithm of gross efficiency dropped as the quantity of food increased. In such a case the decline would be a function of food quality and not food quantity. Gross efficiency of growth of sockeye salmon fingerlings was temperature dependent (Brett, gt_a1., 1969). A decline of the logarithm of gross efficiency took place at the progressively higher 67 dietary intakes associated with increasing temperatures. The highest values occurred between 5 and 10 C. The inflection point of dietary efficiency of both the coho salmon and the rainbow trout took place at and be- yond the 50 per cent dietary protein level. This level of inflection was 5 to 10 per cent higher than the conclusions from percentage weight gain and protein retention. The im- portance of a proper balance between energy and protein was discussed by Ringrose (1971) for brook trout. He found that the calorie-to-protein ratio (C/P) of 75 kcal/kg/% protein was approximately optimum for producing maximum growth and feed efficiency. The C/P ratios of the diets used in the rainbow trout and coho salmon tests ranged be- tween 131.0 in the case of diet no. 269 and 66 for diet no. 273. Based on the apparent optimum dietary protein levels it could be said that C/P of 98.0 and 87.0 were the optimum ratios for producing maximum weight gain for the coho salmon and rainbow trout fingerlings. Biological Value Of protein is a term used to define the per~ centage of protein absorbed and utilized by the body of the animal. It is dependent on the amounts and propor- tions of the essential amino acids available to the body. Tryptophan was included as an indispensable amino acid for the growth of rainbow trout (Shanks, et_al., 1962). In addition, the knowledge that gelatin is deficient in tryptophan would tend to cause the belief that there 68 may have been some variation of the protein biological value of the tested diets. Some protein reserve in the body of fish may be helpful in countering stressful conditions. Excess pro- tein, however, would be wasted through burning or trans- formation to fat and deposition as fat tissues since storage is very limited. Also, protein would be used as energy if the supplied energy was insufficient or the protein fed was low in quality (Phillips, 1969). There- fore, it was desirable to maintain an appropriate calorie- to-protein ratio in the diet which would promote optimum efficiency of energy utilization because carbohydrates have a sparing action on proteins. Phillips and Brockway (1959) stated that high dietary protein level would in- crease the metabolic rate of the fish to get rid of the toxic nitrogenous products which resulted from the deamina- tion processes. Although the protein efficiency ratio (P.E.R.) and net protein utilization (N.P.U.) gave essen— tially the same relative ratings of the various levels of dietary protein as were obtained by the gain in weight and protein retention methods, the need for some informa- tion on the efficiency of the protein utilization by fish in relation to the quantity and quality of protein intake would be valuable. The P.E.R. method was reported to be more repro- ducible than that of N.P.U., which was believed to be a 69 :more complex and difficult procedure (Chapman, gt_§l., 1959; Derse, 1960; and Middleton, et_§l,, 1960). Apparent- ly the determinations of MSE showed that apparent N.P.U. values varied with the level of the dietary protein tested more than did the P.E.R. determinations. However, P.E.R. and N.P.U. outcomes showed similar responses that increased with dietary protein until they reached maximum values at the best balanced diet. Therefore, it was concluded that P.E.R. might be correlated with N.P.U. since this correla- tion has been frequently established in mammals (Bender, 1956; Bender & Doell, 1957; and Henry, 1965). Chapman, et_al., (1959) studied six protein sources of different qualities with rats and reported that P.E.R. determinations were more widely spread between low and high quality proteins than were either the net protein ratio or the net protein utilization. From the results, P.E.R. values tended to level off when the maximum values of 2.1 and 2.2 for rainbow trout and coho salmon, respectively, were reached at ap- proximately the 40 per cent protein level, although a slight decline was noticeable at higher levels of protein intake. Rippon (1959) found that rats which received .87 g protein/day and those which received 1.47 g/day did not show a significant difference in the calculated P.E.R. This suggested that both intakes were higher than the minimum level of protein intake that gave maximum weight 7O gain. The reported values of P.E.R. by Cowey, et_al., (1972) for the plaice were relatively lower than the values calculated for rainbow trout and coho salmon. The high values of P.E.R. might be due to the higher digestibility of casein diets than the freeze-dried cod muscle and shrimp meal diets used by Cowey, et_al., (1972). Obviously the protein efficiency ratio procedure was-subject to a few handicaps such as: (1) it does not allow for maintenance by not including a control group of animals fed on a protein-free diet and, (2) it assumes that the increase in weight of the body is proportional to the protein retained, which is not always a valid as- sumption. In spite of the above mentioned limitations, this method is generally applicable to fish bioassay and the results rated the overall nutritive values of proteins in the diets better than the total diet effi- ciency method. Efficiency of conversion of dietary pro- tein to tissue protein seemed to level off after the maximum growth was obtained or when the species growth potential (Gerking, 1952) was attained. Excessive protein intake will be eliminated since the capacity of the body to store protein is limited. The results of the studies of Morrison and Campbell (1960) established that P.E.R., which is correlated to N.P.U. (Bender & Doell, 1957), varied not only with the quantity and the quality of the dietary protein, but also with the duration of the test 71 and the sex of the experimental animal. They suggested- that the factors studied should be standardized if repro- ducible results were desirable. To compensate for the maintenance allowance which was not considered in the P.E.R. calculations, Bender and Doell (1957) proposed a modified method by including a control group fed on a protein-free diet. This new method was called the net protein ratio (N.P.R.) and could be converted to a percentage scale (termed protein retention efficiency, P.R.E.) by multiplying N.P.R. by sixteen. The factor sixteen was derived by determination of the per- centage protein in new tissue of the growing rat. No serious attempt was made to calculate the N.P.R. and the P.R.E. in this study since the equation requires a non- protein fed group. Net protein utilization (N.P.U.) determinations which were proposed to measure the percentage of food protein that was retained by the fish were subjected in the rainbow trout and coho salmon tests to criticism identical to that mentioned for P.E.R. calculations, which was the lack of data from a control group fed a protein— free diet. Hence, the values of N.P.U. in this article would not be regarded as absolute since maintenance allowance of protein was not taken into account. There- fore, the term apparent net protein utilization was pro- posed as more appropriate. The apparent values of N.P.U. 72 were defined as the percentage protein retained in'a given period of protein intake. Accordingly, these values tended to be numerically lower than would be true if a control group had been included. Chapman, et_al.,(l959) estimated the P.E.R. for weaning rats fed 10 per cent protein as casein was equal to 2.5 in a four week assay. The N.P.U. for the same period was 72.2. The results in these two experiments established that P.E.R. and N.P.U. without any doubt varied with the tested level of protein. The find- ings of Morrison, et al., (1963) and Henry (1965) were similar to this conclusion. Rippon (1959) demonstrated that P.E.R. values were highly related to the Biological Value of the proteins and Bender (1956) reported a good correlation between P.E.R. and N.P.U. Therefore, the Biological Value of any protein would vary with the level of protein tested. The same conclusion was given by Maynard and Loosli (1969). Miller and Payne (1961) claimed that diets in which more than half of the energy was supplied in the form of proteins were unable to support growth in rats. Also, they reported that a decline in the net protein utilization was observed with increases in the protein concentration of the diets if the dietary protein was above the maintenance level. Recently, Cowey, gt_al., (1972) claimed identical findings with the plaice. The results of the rainbow trout and coho salmon tests did not confirm those of Miller and Payne (1961) and 73 Cowey, et_§1., (1972) but correspond to those of Njaa (1962) who implied that N.P.U. leveled off at high pro— tein levels. Morrison, §t_31., (1963) failed to support the claim of Miller and Payne (1961), and they concluded that the decrease in N.P.U. with an increase of protein concentration of the diet found by Miller and Payne (1961) might be a result of diet deficiency in auxiliary growth factors. In addition, Hegsted and Chang (1965b) reported that N.P.U. decreased only after the maximal growth was achieved. The relationship of net protein utilization and Biological Value of a given protein was expressed as.B.V. = N.P.U./TD by Miller and Bender (1955) where TD referred to the true digestibility of the protein ingested which was independent of the percentage protein in the diet in rats (Forbes, et_gl., 1958) and in rainbow trout (Nose, 1967). The digestibility of Halver's (1957) diet that was applied in these tests was proven to be higher than two other natural and two other formulated diets (Brett, 1971). Therefore, the findings of these experiments should be adjusted by any commercial firm in order to meet the natural needs of fish at different water salinities to attain maximum growth and healthy conditions. The data under investigation were subjected to one important handicap which was the significant difference of the initial weights. This problem was minimized by employing 74 the instantaneous growth rate method or applying the per— centage gain values instead of using absolute gram gain in determining the growth curves. Hegsted and Chang (1965a) stated that log-dose response curves with gain were sig- moid in nature and recommended their use instead of the more complex procedures. In fisheries science, infor- mation on the efficiency of protein utilization was limited. Gerking (1952 and 1955) designated the term per- centage utilization of protein for growth instead of the widely used term "biological value" which refers to that percentage of dietary protein retained by fish from the absorbed protein portion. Whatever the term that he may have used, he found that as the body weight of fish in- creased, the ability of fish to utilize the nitrogen com- ponent of the food for growth decreased. Pandian (1967b) applied the term "efficiency of protein conversion" to the definition of biological value. The tested fish were the euryhaline fishes Megalops cyprinoides and Ophiocephalus striatus. He reported a conclusion identical to that of Gerking (1952). In both experiments, natural foods were tested and no formulated diets were employed. Determina- tions of Biological Value of the protein were criti- cized by Hegsted and Chang (1965a) since they over— estimated the nutritive values of low quality proteins. In mammals, Forbes, et_al., (1958) found that the biologi— cal value decreased with the increase of the concentration of protein in the diets. 75 No comparable data were available in fisheries science and few individuals have attempted to apply to fishes the techniques used in mammalian and avian studies. All the methods that were used in these two bioassays rated the various protein levels in the same order and there was no serious difference in the output obtained. One of the most interesting points that emerged was the deviation of the total diet efficiency inflection point from those resulting from the rest of the methods. 'There- fore, the total diet efficiency method would be the least accurate method to be recommended to evaluate the dietary protein level. Generally speaking, the amount of protein in the diet exerts a biological influence on the growth rate. Low protein levels cause lower growth rate, growth efficiency, and protein conversion. Any relative increase in these is related directly to the increase of the protein level until the species specific growth poten- tial is reached and after which growth or conversion will stabilize or decline. The significant increase in the hematocrit values for the rainbow trout at 20 p.p.t. rather than at 10 p.p.t. was related to the loss of some water from the blood in the process of homeostasis in higher salinity media. The in- significant effect of various dietary proteins on the total serum protein and hematocrit values of the rainbow trout implied that the fish, due to the limited ability 76 to store protein, responded similarly to the protein in— take as long as it was above the maintenance level of protein. In spitecflfthe marked significance of total serum protein of the coho salmon as influenced by the concentra- tion of protein intake the data could not be decisively discussed due to the disaster the test experienced on the 65th day with the loss of a few fish by asphyxia.' However, it was deemed advisable to present such data if needed for future comparison. Wedemeyer and Chatterton (1970 and 1971) attempted to establish normal ranges for rainbow trout and coho salmon. They stated that 2 to 6 g/100 ml and 1.4 to 4.3 g/100 ml were the normal total serum protein values for rainbow trout and coho salmon respectively while the hema- tocrit values as determined for the coho salmon juveniles ranged between 32.5 and 52.5 per cent. ‘Consequently, the figures obtained in this study indicated that the tested coho salmon and rainbow trout juveniles were within the normal ranges proposed. The environmental salinity might induce a positive effect on the protein content of the blood. Love (1970) stated that the specific gravity of the blood is dependent on the amount of protein in euryhaline fish. The changes that would occur in the fish when transformed from one sa- linity to another were discussed thoroughly by Winberg 77 (1956) and Love (1970). These changes in the rate of metabolism,blood composition and urine flow revert to the original patterns after a short period of adaptation to the new environment. These findings might account for the insignificant difference of the total serum protein values of the rainbow trout in the two salinities. SUMMARY The study was conducted at Bowman Bay Field Station, Anacortes, Washington of the Western Fish Nutri- tion Laboratory, U.S. Fish and Wildlife Service. Juvenile rainbow trout, Salmo gairdneri and coho salmon, Oncorhynchus kisutch were used to determine pro- tein requirements as influenced by salinity of 10 p.p.t. or 20 p.p.t. The purified-diet method was employed, using seven protein levels that were supplied as casein and gelatin. The range of the dietary protein 1eVels was 30 to 60 per cent in 5 per cent increments. Evaluation of dietary protein levels and the reactions of the fish to environmental salinity was based on several techniques developed by mammal and bird nutritionists as well as methods which have been used by fish physiologists and nutritionists. Percentage weight gain,gram protein gain (protein retention), gross or total efficiency of growth, specific growth rate, protein efficiency ratio (P.E.R.) and net protein utilization (N.P.U.) were criteria employed. All the criteria, except for gross efficiency, in- dicated that 40 per cent protein in the diet was the 78 79 minimum value to support maximum growth of the two groups of coho salmon and that of rainbow trout at 10 p.p.t. salinity, while 45 per cent dietary protein concentration was the minimum level for the rainbow trout raised in 20 p.p.t. salinity. Growth leveled off after reaching a maximum figure followed by a decline if the protein level of the diet ex- ceeded 55 per cent. An inverse relationship was established between gain in weight and feed/gain. The ability of the fish to withstand higher osmotic pressure than that of their bodies depends largely upon the size, age, and physiological status of the fish . Dietary protein levels for maintenance were pre- dicted for both the rainbow trout and coho salmon by ex~ tending the growth index curves downward to meet the zero growth level. Maintenance levels were close to 15 and 25 per cent dietary protein for rainbow trout and coho salmon, respectively. The calorie-to-protein ratio of the diets that pro- duced maximum gain ranged between 98.0 and 87.0 kcal/kg/% protein. Analyses of variance and Duncan multiple range tests were used to determine the significance of the data. Either standard error (:SE) or mean square error (MSE) accompanied each table or figure to give pertinent infor- Ination about the range of the means. LITERATURE CITED LITERATURE CITED Allison, J.B. 1964. The nutritive value of dietary pro- teins, p. 41-86. In Mammalian Protein Metabolism, Vol. 2. Academic Press Inc., New York. Association of Official Analytical Chemists. 1970. Official Methods of Analysis. 11th edition. A.O.A.C., Washington, D.C. Bender, A.E. 1956. Relation between protein efficiency and net protein utilization. Brit. J. Nutr. 10: 135-143. Bender, A.E. and B.H. Doell. 1957. Biological evaluation of proteins: a new aspect. Brit. J. Nutr. 11: 140-148. Bouck, G.R. and R.C. Ball. 1966. Influence of capture methods on blood characteristics and mortality in the rainbow trout (Salmo gairdneri). Trans. Amer. Fish. Soc. 95: 170-176. Bouck, G.R. and R.C. Ball. 1965. Influence of a diurnal oxygen pulse on fish serum proteins. Trans. Amer. Fish. Soc. 94: 363—370. Brett, J.R. 1971. Growth responses of young sockeye sal- mon, Oncorhynchus nerka, to different diets and 'planes of nutrition. J. Fish Res. Bd. Canada 28: 1635-1643. Brett, J.R., J.E. Shelbourn and C.T. Shop. 1969. Growth rate and body composition of fingerling sockeye salmon, Oncorhynchus nerka, in relation to temper- ature and ration size. J. Fish. Res. Bd. Canada 26: 2363-2394. Brown, M.E. 1957. Experimental studies on growth, p. 361- 418. In_The Physiology of Fishes, Vol. 1. Academic Press Inc., New York. Bullivant, J.S. 1960. The influence of salinity on the rate of oxygen consumption of young quinnat salmon (Oncorhynchus tschawytscha). New Zealand Journal of Science 4: 381-391. 80 81 Canagaratnam, P. 1959. Growth of fishes in different salinities. J. Fish. Res. Ed. Canada 16: 121-130. Chapman, D.G., R. Castillo and J.A. Campbell. ‘1959. Evaluation of protein in foods. 1. A method for the determination of protein efficiency ratios. Can. J. Biochem. and Physiol. 37: 679—686. Conte, F.P. and H.H. Wagner. 1965. Development of osmo- tic and ionic regulation in juvenile steelhead trout Salmo gairdneri. Comp. Biochem. Physiol. 14: 603-620. Conte, F.P., H.H. Wagner, J. Fessler and C. Gnose. 1966. ' Development of osmotic and ionic regulation in juvenile coho'salmon Oncorhynchus kisutch. Comp. Biochem. Physiol. 18: 1-15. Cowey, C.B., J.A. Pope, J.W. Adron and A. Blair. 1972. Studies on the nutrition of marine flatfish. The protein requirement of plaice (Pleuronectes platessa). Brit. J. Nutr. 28: 447-456. Davies, P.M.C. 1963. Food input and energy extraction efficiency in Carassius auratus. Nature 198: 707. DeLong, D.C., J.E. Halver and E.T. Mertz. 1962. Quanti— tative threonine requirements of chinook salmon at two water temperatures. J. Nutr. 76: 174-178. DeLong, D.C., J.E. Halver and E.T. Mertz. ‘1958. Nutri— tion of salmonoid fishes: Protein requirements of chinook salmon at two water temperatures. J. Nutr. 65: 589—599. Derse, P.H. 1960. Evaluation of protein quality (Biological method). J. A.O.A.C. 43: 38-41. Dupree, H.K. and J.F. Halver. 1970. Amino acids essen- tial for the growth of channel catfish, Ictalurus punctatus. Trans. Amer. Fish. Soc. 99: '90-92. Fessler, J. L. and H. H. Wagner. 1969. Some morphological and biochemical changes in steelhead trout during the*parrsmolt transformation. J. Fish. Res. Bd. Canada 26: 2823-2841. Fisher, H., D. Johnson, Jr. and G.A. Leveille. 1957. The phenylalanine and tyrosine requirement of the grow- ing chick with special reference to the utiliza- tion of the D-isomer of phenylalanine. J. Nutr. 62: 349—355. 82 Folch, J., M. Lees and G.H. Sloane Stanley. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226: 497-509. Forbes, R.M., L. Vaughan and M. Yohe. 1958. Dependence of biological value on protein concentration in the diet of the growing rat. J. Nutr. 64: 291- 302. Fry, F.E.J. 1957. The aquatic respiration of fish, p. 1-63. In The Physiology of Fishes, Vol. 1. Academic Press Inc., New York. Gahimer, G.D., A.R. Anderson and C.L. Johnson. 1971. In- expensive experimental aquariums. Prog. Fish- Cult. 33: 242. Gerking, S.D. 1955. Influence of rate of feeding on body composition and protein metabolism of bluegill sunfish. 'Physiol. Zoo. 28: 267—282. Gerking, S.D. 1952. The protein metabolism of the sun- fishes of different ages. Physiol. Zoo. 25: 358-372. Groves, T.D.D. 1970. Body composition changes during growth in young sockeye Oncorhynchus nerka in fresh water. J. Fish. Res. Bd. Canada 27: 929— 942. Halver, J.E. 1957. Nutrition of salmonoid fishes: Water- soluble vitamin requirements of chinook salmon. J. Nutr. 62: 225—243. Halver, J.E. and W.E. Shanks. 1960. Nutrition of salmon- oid fishes: Indispensable amino acids for‘sockeye salmon. J. Nutr. 72: 340-345. Halver, J.E., L.S. Bates and E.T. Mertz. 1964.‘ Protein requirements for sockeye salmon and rainbow trout. Federation Proceedings 23: 1778. Hegsted, D.M. 1964. Proteins, p. 115-179.£n Nutrition: A Comprehensive Treatise, Vol. 1. Academic Press, New York. Hegsted, D.M. and Y. Chang. 1965a. Protein utilization in growing rats. I. Relative growth index as a bioassay procedure. J. Nutr. 85: 159—168. 83 Hegsted, D.M. and Y. Chang. 1965b. Protein utilization in growing rats at different levels of intake. J. Nutr. 87: 19-25. Henry, K.M. 1965. A comparison of biological methods with rats for determining the nutritive value of proteins. ‘Brit. J. Nutr. 19: 125-135. Kepshire, B.M.,Jr. and W.J. McNeil. 1972. Growth of premigratory chinook salmon in sea water. Fishery Bulletin 70: 119-123. Kinne, O. 1962. Irreversible nongenetic adaptation. Comp. Biochem. Physiol. 5: 265-282. Kinne, O. 1960. Growth, food intake, and food conversion in a euryplastic fish exposed to different tem- peratures and salinities. Physiol. Zoo. 33: 288- 317. Love, R.M. 1970. The Chemical Biology of Fishes. Academic Press, New York, 547 p. Love, R.M. 1957. The biochemical composition of fish, p. 401-418. In The Physiology of Fishes, Vol. 1. Academic Press Inc., New York. Maynard, L.A. and J.K. Loosli. 1969. Animal Nutrition. McGraw-Hill Book Co., New York 613 p. McCartney, T.H. 1969. The effect of dietary protein level and methionine supplementation on the growth, chemical composition, and total sulfur of fingerling brook trout. Cortland Hatchery Report No. 36 for the year 1967. Fish. Res. Bu11.No. 31: 20-24. McLaughlan, J. M. and J. A. Campbell, 1969. Methodology of protein evaluation, p. 391-422. In Mammalian Protein Metabolism, Vol. 3. Academic Press Inc., New York. Middleton, E.J., A.B. Morrison and J. A. Campbell. 1960. Evaluation of protein in foods. VI. ‘Further factors influencing the protein efficiency ratio of foods. Can. J. Biochem. Physiol. 38: 865-869. Miller, D.S. and P.R. Payne. 1961. Problems in the pre- diction of protein values of diets. Brit. J. Nutr. 5: 11-19. 84 Miller, D.S. and A.E. Bender. 1955.- The determination of the net utilization of proteins by'a shortened' method. Brit. J. Nutr. 9: 382-388. Morrison, A.E., Z.I. Sabry, N.T. Gridgeman and J.A. Campbell. 1963.’ Evaluation of protein in foods. VIII. Influence of quality and quantity of die— tary protein on net protein utilization. Can. J. Biochem. Physiol. 41: 275-281. Morrison, A.B. and J.A. Campbell. 1960.‘ Evaluation of protein in foods. V. Factors influencing the protein efficiency-ratio of foods. J. Nutr. 70: 112-118. Njaa, L.R. 1962. A note on the method of Miller and Payne for prediction of protein value. Brit. J. Nutr. 16: 185-190. Nose, T. 1967. On the metabolic fecal nitrogen in young rainbow trout. Tokyo, Bull. Freshwater Fish. Res. Lab. 17: 97-105. O'Brien, D., F. A. Ibbott and D.O. Rodgerson. ‘1968. Labor- atory Manual of Pediatric Micro-Biochemical'Tech- niques. Harper and Row Publishers, New York 367 p. Osborne, T. B. , L.B. Mendel and E.L. Ferry. 1919. A method of expressing numerically the growth- promoting value of proteins. J. Biol. Chem. 37: 223-229. Otto, R.G. 1971. Effects of salinity on the survival and growth of presmolt coho salmon, Oncorhynchus kisutch. J. Fish. Res. Ed. Canada 28: 343-349. Otto, R.G. and J.E. McInerney. 1970. Development of salinity preference in presmolt coho salmon, Oncorhynchus kisutch. J. Fish. Res. Bd. Canada 27: 793-800. Paloheimo, J.E. and L.M. Dickie. 1966. Food and grthh of fishes. II. Effects of food and temperature on the relation between metabolism and body weight. J. Fish. Res. Bd. Canada 23: 869-908. Pandian, T.J. 1967a. Transformation of food in Megalops cyprinoides. I. Influence of quality of food. Marine Biol. 1: 60-64. 85 Pandian, T.J. 1967b. Intake, digestion, absorption and conversion of food in fishes Megalgps cyprinoides and Ophiocephalus striatus. Marine Biol. 1: 16- 32. Parker, R.R. and W.E. Vanstone. 1966. Changes in chemi- cal composition of control British Columbia pink salmon during early sea life. J. Fish. Res. Bd. Canada 23: 1353-1384. Phillips, A.M.,Jr. 1969. Nutrition, digestion and energy utilization, p. 391-432. In Fish Physiology, Vol. 1. Academic Press Inc., New York. Phillips, A.M., Jr" D.L. Livingston and H.A. Poston. 1965. The effect of protein and calorie levels and sources on the growth of brown trout. Fish. Res. Bull., 28: 11-19. Phillips, A.M., Jr. and D.R. Brockway. 1959. Dietary calories and the production of trout in hatcheries. Prog. Fish-Cult. 21: 3-16. Phillips, A.M., Jr” D.R. Brockway, F.E. Lovelace and H.A. Podoliak. 1957. A chemical comparison of hatchery and wild brook trout. Prog. Fish-Cult. 19: 19-25. Potts, W.J.W. and G. Parry. 1963. Osmotic and Ionic Regulation in Animals. The Macmillan Co., New York. 423 p. Ringrose, R.C. 1971. Calorie—to-protein ratio for brook trout (Salvelinus fontinalis). J. Fish. Res. Bd. Canada 28: 1113-1117. Rippon, W.P. 1959. A comparison of several methods for estimating the nutritive value of proteins. Brit. J. Nutr. 13: 243-260. Schiffman, R.H. and P.O. Fromm. 1959. Chromium-induced changes in the blood of the rainbow trout, Salmo gairdneri. Sewage and Industrial Wastes 31: 205- 211. Shanks, W.E., G.D. Gahimer and J.E. Halver. 1962. The indispensable amino acids for rainbow trout. Prog. Fish-Cult. 24: 68-73. Vanstone, W.E. and J.R. Markert. 1968. Some morphologi- cal and biochemical changes in coho salmon, Oncorhynchus kisutch, during parrsmolt transfor- mation. J. Fish. Res. Bd. Canada 25: 2403-2418. 86 Vanstone, W.E., J.R. Markert, D.B. Lister and M.A. Giles. 1970. Growth and chemical composition of chum (Oncorhynchus keta) and sockeye (Oncorhynchus nerka) salmon fry produced in spawning channel and natural environments. J. Fish. Res. Bd. Canada 27: 371-382. Warren, C.E. and G.E. Davis. 1967. Laboratory studies on the feeding, bioenergetics, and growth of fish, p. 175-214. In The Biological Basis of Freshwater Fish ProductiSH. Blackwell Scientific Publications, Oxford. Wedemeyer, G. and K. Chatterton. 1971. Some blood chemistry for the juvenile coho salmon, Oncorhynchus kisutCh. J. Fish. Res. Bd. Canada 28: 606-608. Wedemeyer, G. and K. Chatterton. 1970. Some blood chemistry values for the rainbow trout, Salmo gairdneri. J. Fish. Res. Bd. Canada 27: 1162-1164. Weisbart, M. 1968. Osmotic and ionic regulation in embryos, alevins, and fry of the five species of Pacific salmon. Can. J. Zool. 46: 385-397. Winberg, G.G. 1956. Rate of Metabolism and Food Require- ments of Fishes. Fish. Res. Bd. Canada, Transl. series no. 194. 202 p. APPENDIX 7 8 I! .momoSHGoumm_nww3uon.©ouooflogm mm imam coop mo Hones: one AHV ooo oo ooom ooo ooom ooo oomm oooo ooom ooo ooo oo ooom ooo ooom ooo oomm ooo oooa ooo ooo oo moon ooo ooom ooo ooom Emoo ooom ooo ooo oo ooom moo omom moo ooom moo ooom Eooo Eooo oo oomm ooo omom ooo ooom ooo mmmm ooo Eooo oo ooom ooo Hoom ooo ooom ooo ooom ooo Emoo oo ooom ooo ooom ooo oomm ooo ooom ooo ooo om ooom moo mmom moo moom moo ooom moo Amoooo oo ooom omo ooom omo momm omo omom ooo omo oo momm moo ooom moo omom moo ooom moo Amoooo om ooom ooo ooom ooo oomm ooo ooom ooo ooo om ooom ooo ooom ooo Hoom ooo ooom ooo Eooo om ooom ooo ooom ooo oomm ooo ooom ooo ooo om Hmuoe mxmmS m Hmuoe mxomB N Houoa mxmwz m Hmuoe mxmmB m mxmwB N .moopoupcmogoo Sum ape ohm 6cm pmH componm hHoumHQ I) .p.@.@ OH um usoup Boogoou mo mmmo zoom mo Hobsozlu.H manna 88 .mommnucmumm :monoQ woumoflocm mm zoom pomp mo Hogans one I! ooom ooo ooom ooo ooom Emoo ooma Eooo Eooo oo momm moo omom moo ooom Eooo ooom ooo- Eooo oo ooom ooo ooom ooo omom Eooo ooom moo. Amvooo oo ooom ooo ooom ooo mmom aooo ooom moo Amoooo oo ooom ooo ooom Eooo ooom Eooo momm ooo Eooo oo ooom moo omom. ASooo ooom ooo ooom ooo. Eooo. oo Homo Amoooo moom Eomo ooom Amoomo ooom ooo Emoo oo ommm ooo moom ooo mmom Emoo momm moo Amoooo oo oomm ooo ooom ooo ooom ooo momm Eooo Eooo oo ooom ooo oomm ooo ooom ooo ooo oo oomm ooo moom ooo ooom EHmo momm Emoo Amvomo om ooom ooo moom ooo ooom ooo ooom Eooo ooo om omom Eooo Hmom Amommo ooom Eooo ooom ooo ooo om oomm Eooo ooom Eomo omom aooo ooom Emoo Amcooo om Hopoe mxoms N Hopoe mxooa N Hmwoa mxmmB N Hmuoe. mxooz N mxooz N w .gompmnusoogoo npm sue bum osN umH compoum wuopomo .n.m.m om no noomn sonommm oo omoo Homo oo monaoz-u.m magma 1:.- 89 .mommnucwHMQLGmmBumn oopoomocm mm Snow comp mo Hones: was ooom ooo oomm ooo ooom ooo ommm ooo ooo oo ooom Amoooo oomm ooo ooom ooo ommm ooo ooo oo ooom ooo oomm ooo ooom ooo ommm ooo ooo oo ooom ooo ooom ooo ooom Eooo ommm ooo ooo oo ooom ooo oomm ooo ooom ooo ommm ooo ooo oo ooom ooo oomm ooo ooom ooo ommm ooo ooo om ooom ooo oomm ooo ooom. ooo ommm ooo ooo oo ooom ooo oomm ooo ooom ooo ommm ooo ooo oo moom mmo oomm moo ooom amoo oomm Amvooo ooo om ooom ooo oomm ooo ooom ooo ommm ooo ooo om omom moo oomm Emoo ooom Amoooo ommm ooo ooo om ooom ooo oomm ooo ooom ooo ommm ooo ooo om Hopoa mxwma N Hobos mxmo3 N Houoa mxmoB N Houoa mxme N mxwoz N w .comumupcmogoo sum sue bum ocN uma Componm xuopomo .u.m.m 0H pm QOEHmm ozoo mo mmmo :mmmmfiuuonaszll.m magma 90 'll! .mmmonpcmnmm,cm03umn-wmpoomoqm mm nmmm pomp mo Honfioa $39 ooom ooo ommm ooo ooom .ooo ommm ooo ooo oo ooom ooo oomm ooo ooom ooo ommm ooo ooo oo omom omo momm Amoomo ooom Eooo ommm ooo ooo oo ooom ooo oomm ooo ooom ooo ommm ooo ooo oo ooom ooo ommm ooo ooom aooo ommm ooo ooo om ooom ooo ommm ooo ooom ooo ommm ooo ooo mo ommm ooo moom ooo moom ooo ooo Aoomoo Amomoo oo moom ooo oomm ooo omom 1m+oooo ommm ooo ooo oo ooom ooo oomm ooo ooom ooo ommm ooo ooo om ooom ooo oomm ooo ooom ooo ommm ooo ooo om ooom ooo oomm ooo moom ooo ommm Eooo ooo om ooom ooo oomm ooo ooom ooo ommm ooo ooo om Hopoe mxomz N Hopoe mxmmB N Houoe mxmmB N HMDOB mxmm3 N mxmmz N w .cowpmuucmocoo sum one cum cam pom compose snowman 1.79m ON Em c953 onoo mo mmmo swam mo Honeszllé manme We EM 1me 3 1293 ‘lHllHHHflIIlfil