UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Sexual maturation in sea- and freshwater phases of Pacific salmonids and investigations of reproductive… Smith, Jack Lloyd 1993

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

Item Metadata


831-ubc_1993_spring_smith_jack.pdf [ 3.74MB ]
JSON: 831-1.0086160.json
JSON-LD: 831-1.0086160-ld.json
RDF/XML (Pretty): 831-1.0086160-rdf.xml
RDF/JSON: 831-1.0086160-rdf.json
Turtle: 831-1.0086160-turtle.txt
N-Triples: 831-1.0086160-rdf-ntriples.txt
Original Record: 831-1.0086160-source.json
Full Text

Full Text

SEXUAL MATURATION IN SEA- AND FRESHWATER PHASES OFPACIFIC SALMONIDS AND INVESTIGATIONS OF REPRODUCTIVEHORMONE INVOLVEMENT IN OSMOREGULATORY ADAPTATIONbyJACK LLOYD SMITHB.Sc., The University of VictoriaA THESIS SUBMITTED IN PARTIAL FULFILMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIESDEPARTMENT OF ZOOLOGYWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAFEBRUARY 23, 1993©  JACK LLOYD SMITH, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature) Department of  Zoc / q The University of British ColumbiaVancouver, CanadaDate Mo—, 1 %^3DE-6 (2/88)ABSTRACTThe changes in reproductive parameters and hormones, and blood sodium during thefinal weeks of sexual maturation in wild coastal salmon, chinook, Oncorhynchustshawytscha, and steelhead, 0. mykiss, were documented for stocks spawning in two similarriver systems on the west coast of Vancouver Island, British Columbia, Canada. Chinookwere blood-sampled and sacrificed at intervals during migration from the ocean into brackishNitinat Lake, through the completion of final maturation in Nitinat River, and at the post-spawning stage. Gonadosomatic index (GSI), hepatosomatic index (HSI), and oocytegerminal vesicle stage and hydration level were measured in sacrificed chinook andsteelhead. Stamp River steelhead were serially sampled for blood during the monthpreceding and the month following final maturation.Blood samples were analysed by specific radioimmunoassay for gonadotropin (GtH),estradio1-1713 (E2), testosterone (T), 11-ketotestosterone (11-KT), and 17a,2013-dihydroxyprogesterone (17,20-P). Blood sodium was measured by flame photometry.GSI and hydration level of oocytes increased in females of both species between theearliest stages of maturation and ovulation. Patterns of hormone concentration were similarin chinook and steelhead, but the peaks in concentration of the steroids at each stage weremuch lower in steelhead. Concentrations of E2 and T declined from initial levels to ovulationin females, while GtH and 17,20-P increased. In chinook males, T concentration peaked anddeclined before spermiation, while 11-KT was highest at spermiation, and declinedthereafter. The concentration of 11-Kr was higher than that of T in steelhead males atspermiation. Little change in GtH and 17,20-P concentration occurred in steelhead males,but these hormones increased in relation to onset of spermiation in chinook.AbstractFertility was compared between chinook and chum salmon, 0. Beta, maturing inbrackish and fresh water. Fertility was low in both brackish and freshwater chinook femalesat the beginning of the chinook migration, when salinity was high in Nitinat Lake. Bloodsodium was higher in brackish water females, but there was no correlation between eggmortality and blood sodium at ovulation. Later in the season, as salinity in net pensdecreased, chinook fertility increased. At the beginning of the chum migration season, whensalinity was lowest, chinook fertility was equal to that of chum maturing at the samelocation.The relationship between plasma reproductive hormone concentration and the abilityto osmoregulate in sea water was compared in post-spawning steelhead and coho salmon, 0.kisutch, transferred from fresh water to brackish holding facilities. All hormones declined tolow levels in steelhead while still in fresh water, within 3 weeks of spawning, but persistedat spawning levels for at least 6 weeks in coho. Steelhead blood sodium was stable in therange indicating osmoregulatory adaptation as salinity was increased to full sea watershortly after transfer. Coho blood sodium was higher in 2/3 sea water than that of steelheadin full sea water. High levels of reproductive hormones may interfere with the ability ofsalmonids to osmoregulate in sea water.iliTABLE OF CONTENTSAbstract^  iiTable of Contents^ ivList of Figures  viAcknowledgements^ ixCHAPTER 1 1GENERAL INTRODUCTION^  1CHAPTER 2^ 9ANALYTICAL METHODS COMMON TO MOST EXPERIMENTS^ 9A. RADIOIMMUNOASSAY PROCEDURES^ 9B. BLOOD SODIUM MEASUREMENT 13C. STATISTICS^ 14CHAPTER 3^ 15NORMAL MATURATION OF CHINOOK AND STEELHEAD^ 15A. INTRODUCTION^ 15B. MATERIALS AND METHODS^ 16A. Sampling of maturing chinook 16B. Sampling of steelhead 21C. RESULTS^ 22A. Chinook 22I. Migration and spawning of chinook^ 22II. Water quality parameters^ 23III. Changes in reproductive parameters 28IV. Changes in reproductive hormones^ .36B. Steelhead^ 50I. Migration and spawning of steelhead 50II. Changes in reproductive parameters^ 51III. Changes in reproductive hormones 56D. DISCUSSION^ 66A. females 66B. males 70CHAPTER 4^ 73MATURATION OF CHINOOK AND CHUM IN BRACKISH WATER ^ 73A. INTRODUCTION^ 73B. MATERIALS AND METHODS^ 74C. RESULTS^ 77I. Reproductive parameters 77II. Reproductive hormones^ 79III. Water quality^ 86ivTable of ContentsD. DISCUSSION^ 88CHAPTER 5^ 92SW TRANSFER OF POST-SPAWNING SALMON AND STEELHEAD^ 92A. INTRODUCTION^ 92B. MATERIALS AND METHODS^ 93C. RESULTS  94I. General observations  94II. Changes in reproductive hormones^  96III. Blood sodium^ 100D. DISCUSSION  101CHAPTER 6^ 105SUMMARY AND CONCLUSIONS^ 105A. NORMAL MATURATION OF CHINOOK AND STEELHEAD^ 1051. Females^  1052. Males 106B. MATURATION OF CHINOOK AND CHUM IN BRACKISH WATER^107C. SEAWATER TRANSFER OF SALMON AND STEELHEAD^ 109REFERENCES^ 111LIST OF FIGURESFigure 1. Nitinat system study area, with inset showing proximity of Nitinat and Stampsystems on the west coast of Vancouver Island, British Columbia.^ 18Figure 2. Stamp/Somass system showing the location of Robertson Creek Hatchery. . . . . 19Figure 3. Temperature profiles of Nitinat Lake, Aug. 19 to Sept. 5, 1989, at chinook samplingsites 1-3. ^ 25Figure 4. Salinity profiles of Nitinat Lake, Aug. 19 to Sept. 5, 1989, at chinook sampling sites1-3. ^ 26Figure 5. Dissolved oxygen profiles of Nitinat Lake, Aug. 19 to Sept. 5, 1989, at chinooksampling sites 1-3. ^ 27Figure 6. Water content of oocytes and eggs, and gonadosomatic index of chinook femalesmigrating from the ocean to spawning grounds and undergoing sexual maturation. . 31Figure 7. Plasma sodium of chinook salmon migrating from the ocean to spawning groundsand undergoing sexual maturation^ 32Figure 8. Gonadosomatic index and hepatosomatic index in male chinook migrating tospawning grounds and undergoing sexual maturation.^ 35Figure 9. Gonadotropin and Estradiol in female chinook migrating to spawning grounds andundergoing sexual maturation. ^ 37Figure 10. Blood sodium, Gonadotropin, Testosterone, and 17a,201I-dihydroxyprogesteronein ovulating female chinook at Robertson Creek Hatchery. ^  38Figure 11. Plasma testosterone and 17a,20l-dihydroxyprogesterone in female chinookmigrating to spawning grounds and undergoing sexual maturation. ^ 41Figure 12. Plasma gonadotropin in male Chinook migrating to spawning grounds andundergoing sexual maturation. ^ 43Figure 13. Blood sodium, Gonadotropin, Testosterone, 11-ketotestosterone, and 17a,2013-dihydroxyprogesterone in sper -miating male chinook at Robertson Creek Hatchery. . . 44Figure 14. Plasma 11-Ketotestosterone and testosterone in male chinook migrating tospawning grounds and undergoing sexual maturation. ^ 47Figure 15. Plasma 17a,203-dihydroxyprogesterone in male chinook migrating to spawninggrounds and undergoing sexual maturation. ^  49viList of FiguresFigure 16. Gonadosomatic index, hepatosomatic index and oocyte water content in maturingStamp River steelhead females at Robertson Creek Hatchery. ^  53Figure 17. Gonadosomatic index and hepatosomatic index in maturing Stamp Riversteelhead males at Robertson Creek Hatchery.^ 55Figure 18. Plasma gonadotropin in serially sampled steelhead females undergoing sexualmaturation at Robertson Creek Hatchery, Feb.-March, 1988. ^ 58Figure 19. Plasma estradio1-1713 in serially sampled steelhead females undergoing sexualmaturation at Robertson Creek Hatchery, Feb.-March, 1988. ^ 59Figure 20. Plasma testosterone in serially sampled steelhead females undergoing sexualmaturation at Robertson Creek Hatchery, Feb.-March, 1988. ^ 60Figure 21. Plasma 17a,200--dihydroxyprogesterone in serially sampled steelhead femalesundergoing sexual maturation at Robertson Creek Hatchery, Feb.-March, 1988.. . .61Figure 22. Plasma testosterone in serially sampled steelhead males undergoing sexualmaturation at Robertson Creek Hatchery, Feb.-March, 1988. ^ 63Figure 23. Plasma 11-ketotestosterone in serially sampled steelhead males undergoingsexual maturation at Robertson Creek Hatchery, Feb.-March, 1988. ^64Figure 24. Plasma 17a,200—dihydroxyprogesterone in serially sampled steelhead malesundergoing sexual maturation at Robertson Creek Hatchery, Feb.-March, 1988..^.65Figure 25. Mortalities to the eyed stage among chinook salmon eggs taken from females inbrackish (BW) or fresh (FW) water, October 8-26, 1988, and fertilized with milt from FWor BW males. ^  81Figure 26. Water content of oocytes and eggs, and gonadosomatic index in chinook salmonfemales at Nitinat Lake and Hatchery, 1988 and 1989. ^  82Figure 27. Blood and ovarian fluid sodium in chinook salmon females at Nitinat Lake andHatchery, 1988 and 1989. ^ 83Figure 28. Plasma hormone levels in chinook salmon females at Nitinat Lake and Hatchery,1988 and 1989. ^  84Figure 29. Plasma hormone and sodium levels, and egg mortality rates in chum salmonfemales at Nitinat Lake and hatchery, Oct. 28, 1988.^  85Figure 30. Water quality parameters at Nitinat Lake net pen site, October 18, 1988. . .^87Figure 31. Plasma hormone levels in spawning (March 10, FW) and spent (March 31, FW;April 9, BW) steelhead^ 97viiList of FiguresFigure 32. Plasma hormone levels in spawning (Dec 13 and 27, FW) and post-spawning (Jan19-Feb 22, BW) coho males, winter 1988. ^  98viiiACKNOWLEDGEMENTSThe Department of Fisheries and Oceans Canada and its representatives haveprovided crucial assistance and support during all phases of this project. Within theEnhancement Services Branch I am particularly indebted to: Nitinat Hatchery and staff,including Rob Brouwer for his indulgence, and Steve Emmonds for his skepticism; RobertsonCreek Hatchery and staff, especially Mike Wolfe and Tom Forrest; Big Qualicum Hatchery,especially Lorne Hepting and Ian Seaton; and Iain MacLean of Conuma Hatchery.Within the Biological Sciences Branch I owe thanks to Dr. E. M. Donaldson of theWest Vancouver Lab for putting the physical resources of the radioimmunoassay lab at mydisposal, and to Helen Dye, chemist and den-mother, for making them work. I also thankDr. Craig Clarke, John Blackburn, and John Jensen of the Pacific Biological Station.Special recognition is due to Sarah Casely, who made room in her life for me and "theproject", for more years than either of us anticipated.This thesis would not have been initiated, nor completed, were it not for the exampleof my mother, Isabelle Smith, inspiring me to dream and persist.ixCHAPTER 1GENERAL INTRODUCTIONJuvenile anadromous Pacific salmonids undergo a smolting process which permitstransition from a hypoosmotic to a hyperosmotic medium. After feeding 1 to 6 years in theocean, where growth and gametogenesis are largely completed, adults return to coastalwaters in response to cues related to daylength and temperature (Lam, 1983). From inshoreareas they migrate up rivers and streams, reverting to obligatory hyperosmoregulation, andupon completing sexual maturation, deposit their stenohaline eggs in freshwater (Potts andRudy, 1969).The degree to which final maturation is completed in seawater (SW) or brackishwater (BW; salinity 5-25 ppt) before re-entry into freshwater (FW) varies between speciesand stocks (Healey, 1991). All salmonid species are capable of completing their entire lifecycles in FW (Hoar, 1976). The anadromous forms may proceed upstream directly on arrivalat the river mouth, several months in advance of final maturation and spawning, or schoolfor varying periods, in nearby bays and estuaries, in some cases entering FW in fully maturecondition.In some cases, fish may be forced to remain in BW beyond the point of finalmaturation. For example, if autumn rains are delayed following a dry summer, water levelsin coastal rivers may be low, resulting in either dry riverbeds or reduced holding volumeaccompanied by high temperature and low oxygen. In combination with seal and bearpredation these factors may drive fish out of the river or prevent initial entry. In addition,salmon hatchery broodstock programs may, out of convenience or necessity, cause fish to be1held in BW through final maturation. For example, capture of broodstock in estuaries with aseine net may be more reliable and economical than pursuing them after they have enteredand dispersed in the river. Estuarine-captured broodstock may be ripened (held to maturity)in SW or BW netpens if resources for holding in FW are not available.Investigators have had mixed success with BW or marine ripening of salmonidbroodstock. Fertilization rate of sperm from SW-ripened coho, Oncorhynchus kisutch, wasequal to that of FW-ripened fish, but fertilization rates of SW-ripened eggs were lower(Clarke et al., 1977). Wertheimer and Martin (1981), held FW-captured coho in BW, andfound no negative effect on adults, fertilization rate or survival to eyed stage in embryos. Ina subsequent study (Wertheimer, 1984) there were significant losses of pink, 0. gorbuscha,and coho adults, as well as lower fertility in survivors, under a similar regime. Reduction infertilization rate was found in chum, 0. Beta, (Stoss and Fagerlund, 1982; Lam et al., 1982)and in coho, by Sower and Schreck (1982), in fish held to maturity in BW or SW. Thesestudies also examined blood osmolality and sodium during maturation. Fertilization ratewas high if eggs were taken from individuals with blood sodium levels no higher at ovulationthan those of SW-adapted salmon before maturation. Fertilization rates in individuals withblood sodium levels during final maturation which were higher than those of SW adaptedfish, were much lower.Both Lam et al., (1982) and Sower and Schreck (1982) concluded that finalmaturation in SW is possible in chum and coho. If contact with FW is not permitted soonafter ovulation/spermiation, however, viability of gametes declines more rapidly in SW heldfish than in FW held fish. Personal observations of fertility rates of chum and chinooksalmon, 0. tshawytscha, from a holding site at brackish Nitinat Lake, near the west coast ofVancouver Island, indicate there may be species differences among salmon in the ability to2produce viable gametes in BW. Fertilization rates for chum eggs consistently averaged 97%,whereas those for chinook averaged 20% with a range between individuals of 0 - 80%. Thereason for this difference in fertility of chinook and chum at the same site is not known.Based on the studies cited above, however, I hypothesize that a difference in maturationphysiology exists such that hypoosmoregulation is compromised in chinook, but not chummaturing at the BW net pen site. Dysfunction in osmoregulation in fish held in BW maylead to elevated blood sodium, which is correlated with infertility of eggs.It has been suggested that ovulated salmon have exhausted their energy reservesand cannot meet the higher energy requirements of hypoosmoregulation. Under theseconditions fish gradually lose ground in the effort to retain water and keep sodium andchloride out of the blood and body cavity (Lam et al., 1982; Sower and Schreck, 1982). Giventhe salmon's ability to convert much of its muscle tissue to energy (Ando et al., 1986), fishwhich have fully utilized their energy reserves should have reduced muscle mass. However,based on observations of hundreds of chinook and chum carcasses during artificial spawningactivities at Nitinat hatchery and the net pen site, there is no gross difference in themusculature of ripe chinook or chum killed in BW and those killed in FW. Therefore theenergy shortage argument does not seem entirely satisfactory. An alternative possibility isthat the hormones controlling reproduction interfere in some way with the ability toosmoregulate in brackish water.Regardless of the duration of any holding period between arrival in inshore watersand SW/FW transition there are a number of reproductive processes, controlled by thehypothalamo-pituitary-gonadal axis, which take place before viable gametes are produced(see Fostier et al. 1983; Wallace and Selman, 1981). Some of these, such as vitellogenesis,are initiated while fish are still feeding in the marine environment, well before FW entry3(Wallace and Selman, 1981). Others, such as ovulation (Nagahama, 1990) and spermiation(Liley et al., 1986; Scott and Sumpter, 1989), are more abrupt, and occur just days or hoursbefore gametes are fully formed and ready for release. All are regulated by a sequence ofinterdependent hormonal actions which defines the precise timing of events.Seasonal changes in reproductive hormones and their correlation with developmentalstate of the gonads have been studied in rainbow trout, 0. mykiss, (Baynes and Scott, 1985;Liley et al., 1986a; b; Scott et al., 1983), brown trout, Salmo trutta, (Crim and Idler, 1978),sockeye salmon, 0. nerka, (Schmidt and Idler, 1962; Truscott et al., 1986), as well as pink, 0.gorbuscha, (Dye et al., 1986), coho, (Sower and Schreck, 1982; Fitzpatrick et al., 1986), chum,(Ueda et al., 1984a; Hirano et al., 1990), masou, 0. masou (Sufi et al., 1982; Yamauchi, 1983 )and amago salmon, 0. rhodurus, (Ueda et al., 1985; Kagawa et al., 1983).The most important hormones have been identified as a pituitary glycoprotein,gonadotropin (GtH), and several steroids including an estrogen, estradiol-1713 (E2), theandrogens testosterone (T) and 11-ketotestosterone (11-KT), and the progestin 17a, 2013-dihydroxyprogesterone (17,20-P). Although the mechanisms of action of these hormones arenot fully understood, they are considered to be the critical links in the chain of maturation,in mediating gametogenesis and final maturation.Gonadotropin is a dimeric pituitary glycoprotein believed to direct or initiate mostprocesses in the course of gonadal development and maturation. Two GtH's with distinctbeta subunit amino acid sequences and differing carbohydrate content are recognized insalmonids (Idler and Ng, 1983; Suzuki et al., 1988a;b; Kawauchi et al., 1989; Swanson et al.,1991). Designated GtH I and II, they are somewhat homologous to mammalian FSH andLH, respectively, in their beta subunit amino acid sequences (Itoh et al., 1988). They are alsoreferred to as vitellogenic and maturational GtH, respectively.4Vitellogenic GtH (vGtH) is thought to be important in early gonadal development,when highest plasma levels are recorded (Suzuki et al., 1988c). In immature females itpromotes thecal cell secretion of T, aromatization of T to E2 by granulosa cells (Dickhoff andSwanson, 1989) and uptake of the yolk precursor molecule, vitellogenin, by oocytes (Tyler etal., 1991). In immature males testicular androgenesis is promoted by endogenous vGtH(Swanson et al., 1989). As fish approach final maturation vGtH decreases in the blood andmaturational GtH (mGtH) begins to rise (Suzuki et al., 1988c). Accordingly, the actions andmeasurement of mGtH are of exclusive concern in this study and all subsequent uses of theterm "GtH" refer to the maturational form.Incubations of GtH in vitro with whole follicles (Scott et al., 1982) and separatedthecal and granulosa layers (Kagawa et al., 1982; Kanamori et al., 1988) reveal stage- anddose-dependent responses in their metabolism of E2, T, 17-hydroxyprogesterone, and 17,20-P. Similar incubations of testicular tissue indicate stimulus of T, 11-KT (Schulz, 1986a) and17,20-P (Ueda et al., 1984b) secretion in rainbow trout, and direct control of sodium iontransport by sperm duct epithelium in brook trout, Salvelinus fontinalis, sperrniation(Marshall et al., 1989).Among the sex steroids, the first indication of the onset of final maturation infemales is a decline in plasma E2, signalling the end of vitellogenesis (Scott et al., 1983).Receptors for E2 are located in the liver, and during the oocyte growth stage E2 is bound atcytosolic and nuclear sites (Maitre et al., 1985; Pottinger, 1986; Pottinger and Pickering,1990) stimulating vitellogenin manufacture and release into blood. A decline in E2 is seen atthe end of oocyte growth, followed rapidly by a drop in circulating vitellogenin (Wallace et al.,1987). The factor which initiates the decline in E2 is not known.5Testosterone is present in the blood of salmonids during most phases ofgametogenesis in either sex (Scott and Sumpter, 1983; Baynes and Scott, 1985; Sumpter etal., 1984). Specific binding has been demonstrated in testis Sertoli cells (Schulz and Blum,1988; Foucher and le Gac, 1989) and pituitary of trout (Peute et al., 1989). The action of T inspermatogenesis has not been determined, though its multiple binding sites in trout testessuggest its importance (Schulz and Blum, 1988). In the ovary, where it may function purelyas a substrate for aromatization (Kanamori et al., 1988), no receptor-like binding has yetbeen demonstrated. Nevertheless, T rises (often sharply, in some reports) at the completionof gametogenesis (Sumpter et al., 1984; Truscott et al., 1986; Fitzpatrick et al., 1986),presaging the next phase of steroidogenesis which brings on fmal maturation.11-KT is the major male-specific androgen produced by the salmonid testis (Idler etal., 1971). Testicular production of 11-KT is responsive to exogenous GtH in vivo, more sothan that of T, especially around the time of spermiation (Ueda et al., 1985). In vitro studiesshow that sensitivity of 11-KT synthetic pathways to GtH is highest, and level of responsegreatest, in testicular tissue at the end of spermatogenesis (Schulz and Blum, 1990). Miuraet al. (1991) demonstrated that 11-KT is the only androgen capable of stimulating all phasesof spermatogenesis in vitro in fragments of eel testis. As is the case for T, plasma 11-KT isnot strongly associated with any particular germ cell stage, but the consistently higher levelsseen around final maturation make it a useful indicator of normal reproductive function(Scott and Sumpter, 1989).First identified in significant amounts in spawning sockeye salmon by Idler et al(1960), 17,20-P was shown by Jalabert (1976) to be the most potent hormonal stimulator ofin vitro maturation (germinal vesicle breakdown or GVBD) in rainbow trout folliculated ordenuded eggs. Similar findings have been reported for oocytes of amago salmon and rainbow6trout (Nagahama et al., 1983). Jalabert et al. (1978) demonstrated the effectiveness of 17,20-P as an in vivo inducer of final oocyte maturation in coho. 17,20-P was also shown to be apotent stimulus of spermiation in testes of these same species (Ueda et al., 1985). In males,the beginning of spermiation has a better correlation with the first appearance and rise of17,20-P in plasma than it does with either T or 11-KT (Schulz and Blum, 1990). Sangalangand Freeman (1988) demonstrated 17,20-P to be the major steroid produced by spermiatedtestes and mature, pre-ovulatory ovaries of Atlantic salmon.Osmoregulation is known to be under hormonal control in salmonids. Growthhormone (Bolton et al., 1986; 1987) and cortisol (Redding et al., 1984; Bjornsson et al., 1987)are involved in SW adaptation of salmon smolts. The action of prolactin is required in FWadaptation of salmon previously adapted to SW (Hirano et al., 1985; Hasegawa et al., 1986).Although this thesis is not concerned directly with the osmoregulatory hormones, there is thepossibility that reproductive hormones may influence the process of osmoregulation. Seawater adaptability is impaired by exogenous androgens in salmon smolts (Lundqvist et al.,1989). In maturing adults held in SW blood sodium is elevated as plasma sex steroidsincrease despite levels of cortisol, growth hormone, and prolactin similar to those of SWadapted fish (Hirano et al., 1990). State of osmoregulatory adaptation will be assessed onthe basis of blood and ovarian fluid sodium concentration.In order to assess the physiological response of fish to an environmental disturbanceor constraint, it is necessary to have a clear picture of the physiological profile under"normal" undisturbed conditions. In Chapter 3 of this study I examined changes inendocrine status, reproductive parameters, and osmoregulation in chinook salmon, fromtheir first appearance in inshore waters to the completion of spawning. This is the firstexamination of reproductive physiology in coastal chinook salmon. Endocrine profiles and7reproductive parameters were also studied in the FW phase of adult steelhead trout, aspecies entering FW considerably in advance of final maturation and capable of survival andreentry of SW after reproduction. Reproduction in resident rainbow trout has been welldocumented (Scott et al., 1983), but how well those findings apply to steelhead is speculativeat present. In terms of their reproductive physiology, chinook and steelhead, arguably themost endangered of the Pacific salmon in British Columbia, are the least studied. Assuccessful spawning of the remaining populations becomes of greater concern our knowledgeof the basis of this function will require greater detail, of the kind provided in this chapter.In Chapter 4, the BW maturation of chinook was examined, with emphasis onviability of eggs matured in Nitinat Lake and Nitinat River. One aspect of osmoregulation,the ability to maintain blood sodium levels within a certain range while maturing in BW wasalso examined. This may help to identify abnormalities in blood sodium or hormoneconcentration, which correlate with reduced fertility. Some critical comparisons were alsomade with chum salmon under the conditions of estuarine capture and artificial spawning,as this species is known to spawn normally under slightly saline conditions.Chapter 5 compares the post-spawning response to SW transfer of adult coho, whichnormally die after spawning, and steelhead, which are capable of recovery and SW re-entryafter spawning. Coho (jacks and adult males), of smaller size than chinook, were used in thisstudy for ease of handling. There are also indications from scale circuli patterns thatprecociously maturing male coho (jacks) may return to the ocean and resume feeding.Reproductive hormones and blood sodium were measured to determine whether SWadaptability returns in the presence of high levels of these hormones after reduction ofgonadal mass through spawning.8CHAPTER 2ANALYTICAL METHODS COMMON TO MOST EXPERIMENTSA. RADIOIMMUNOASSAY PROCEDURESPlasma sex steroid levels were measured using radioimmunoassays described andvalidated by van Der Kraak et al. (1984) for testosterone (T), 176-estradiol (estrogen or E2)and 17a-hydroxy-206-dihydroxyprogesterone (17,20-P), and by Dye et al. (1986) for 11-ketotestosterone (11-KT). In preparation for assay, plasma samples were diluted 20-fold insteroid assay buffer and incubated at 70°C for 1 hour in covered 12 x 75 mm borosilicateculture tubes. After cooling to room temperature for 1/2 hour the tubes were centrifuged at1,335 g for 15 min at 4°C then decanted into plastic tubes and stored at -20°C. Thedenaturation of steroid binding proteins, advocated by Scott et al. (1982), eliminates the needfor plasma steroid extraction prior to assay.The assay buffer was a 0.05 M phosphate buffer comprised of 5.75 g/1 dibasic sodiumphosphate and 1.315 g/l monobasic sodium phosphate plus 1.0 g/1 gelatin, in distilled water.After heating to 37° C for 1 hour to dissolve the gelatin and adjusting pH to 7.6, 65 mg/1sodium azide was added to prevent bacterial growth.Standards of 11-Kr (lot 2697, Syndel Laboratories Ltd., Vancouver, B.C.),testosterone (T-1500, Sigma), 17 beta-estradiol (E-8875, Sigma) and 17,20-P (Q-1850,Steraloids Inc., Wilton, NH), dissolved in ethanol at 1 mg/ml initially, were serially diluted at10-fold intervals to 10 ng/ml in the assay buffer. The working standards were prepared byfurther serial-dilution at 2-fold intervals from 5,000 to 39 pg,/ml. Tritiated steroids were9obtained from Amersham Canada Ltd. (Oakville, Ontario) for 11-KT (TRK.676), testosterone(TRK.402) and 17 - estradiol (TRK.322). Tritiated 17,20-P was made at the West VancouverLaboratory by Helen Dye from labelled 17a-hydroxyprogesterone, using the method of Scottet al (1982).Antibody to 11-KT was a gift from Dr. T. G. Owen (Helix Biotech Ltd., Richmond,B.C.), and was used at a dilution of 1:3500 or 1:4000. Remnants from each batch were frozenand added to the next batch made up from 1:100 stock. Antibodies to testosterone (61-315)and estrogen (61-305, ICN Immunobiologicals) were prepared as stock solutions by adding 5ml assay buffer to 1 vial of lyophilized antibody (1:5). Aliquots were removed and diluted to1:21 (E2) or 1:50 (T) as needed for the assays while unused portions were stored frozen at thestock dilution. Antibody to 17,20-P, a gift from Dr. A. P. Scott (MAFF Fisheries Laboratory,Lowestoft, England), was stored at 1:100 and used at working dilutions of between 1:4,000and 1:16,000.In the radioimmunoassays of 11-KT and 17,20-P duplicate 12 x 75 mm borosilicatetubes contained 100 ul of standard or denatured plasma, 50 ul of antibody and 50 ul oftritiated steroid (2,000 cpm). For testosterone and estrogen the tubes contained 200 ul ofstandard or denatured plasma, 200 ul of radiolabelled steroid (5,000 cpm) and 200 ul ofantibody. Activity of the tritiated hormone (total counts) was monitored in each assay byincluding duplicate tubes wherein standard, antibody and charcoal (see below) were replacedby buffer. Non-specific binding and reference controls had standard and antibody in theformer case, and standard only in the latter case, replaced with assay buffer. Aftervortexing, all tubes were incubated overnight at room temperature.The reaction was stopped the following day by immersion in ice-water for 15 minutes.Ice-cold dextran-coated activated charcoal (0.5 g/1 dextran T-70 [Pharmacia (Canada) Ltd.,10Dorval, Quebec] and 5.0 g/1 activated charcoal [C-5260, Sigma] in assay buffer) at 1 ml (11-KT and 17,20-P) or 200 ul (testosterone and estrogen) was added (except in total countstubes) to bind any unbound steroid. After vortexing and a further 10 min stand in ice-water,tubes were centrifuged at 1,335 g for 10 min at 4° C and decanted into glass scintillationvials. Each vial received 10 ml of Scinti Verse II (So-X-12, Fisher) or Ecolite (ICN) and wasvigorously shaken and counted on a LKB Wallac 1214 RackBeta liquid scintillation counter(Wallac Oy, Turku, Finland) for 3 min each.Plasma GtH was measured by slight modification of the method of Peter et al. (1984).Antibody and hormone for standards and labelling were received from Dr. J. P. Sumpter(Brunel University, Uxbridge, England). Phosphate buffer stock for the iodination [0.5 M,0.84 g potassium dihydrogen phosphate (KH2PO4 , BDH) and 6.3 g dibasic sodium phosphate(Na2HPO4, S-9763, Sigma) in 100 ml double distilled deionized water (DDI), pH 7.5] wasprepared fresh each time. Column buffer was a 0.08 M barbital buffer [5.0 g sodium barbital(B 22-500, Fisher), 3.25 g sodium acetate (BDH), 0.1 g thimerosal (T-5125, Sigma) and 34.2ml 0.1 N hydrogen chloride made up to 11 in DDI, pH 8.6]. Assay buffer was made by addingBovine serum albumin (BSA, Sigma) to 0.08 M barbital buffer at the rate of 5 g/l.Iodination of salmon GtH followed the procedure of Peter et al. (1984) for carp GtH,with modification. The shipping vial containing 10 ul (1 mCi) Na 125 1 (Amersham) was usedas the reaction vessel. The solution was first buffered with 70 ul 0.5 M phosphate bufferbefore the addition of 5 ug of sGtH (41.6 ul of a 120 ug/ml solution in phosphate buffer).Mixing of this volume was achieved by drawing up once and expelling from a pipette tipwithin the reaction vessel. Following this, 40 ul of 0.5 M phosphate buffer and 25 ulchloramine T [10 mg chloramine T (Calbiochem) in 10 ml 0.05 M phosphate buffer] wereadded to the vessel. After gentle shaking of the recapped vessel for 90 s the reaction was11stopped with the addition of 100 ul sodium metabisulphate [24 mg sodium metabisulphate(Fisher, S-244) in 10 ml 0.05 M phosphate buffer] and 200 ul potassium iodide [100 mgpotassium iodide (Fisher P-411C) in 10 ml 0.05 M phosphate buffer].The contents of the vessel were transferred to a 1.1x20 cm Sephadex G-50 columnprimed with 2 nil 5% BSA in barbital buffer (1 g BSA in 20 ml barbital buffer). Two hundredul potassium iodide, used to flush the vessel, were transferred to the column as well. Whenthis volume (approximately 690 ul) had percolated into the bed, the column was flushedthrough with column buffer during the collection of 16 fractions of 1 ml (20 drops) each.Activity was counted in 10 ul aliquots from each fraction in a gamma counter (Picker, Pace-1)to identify the first peak (usually appearing in fractions 6-10), corresponding to labelledsGtH. Low activity fractions and the second peak containing inorganic 1251 were discarded.Degradation of the tracer was slowed by adding 200 ul of 5% BSA in column buffer to thestock solution. Tracer stock was viable for about 1 month when stored at 4° C.The RIA for GtH was conducted at 4° C, encompassing 4 days on a 24-hour cycle, or 2days on a 12-hour cycle. Duplicate 12 x 75 borosilicate glass culture tubes contained either50 ul of standard (0.9-240 ng/m1) or plasma, 200 ul GtH antibody (1:10,000-1:14,000) in 1:100normal rabbit serum [NRS, (Calbiochem, 869019)] in assay buffer, and approximately 5,000cpm/200 ul tracer. Assay buffer replaced standards in maximum binding and non-specificbinding controls, and NRS replaced anti-GtH in non-specific binding controls. The rackholding assay tubes was vigorously shaken and incubated at 4° C. At 24 (or 12) hours, tubeswere again shaken and incubated at 4° C for a further 24 (12) hours, followed by addition of200 ul goat antibody to rabbit gamma-globulin (GARGG) (Calbiochem 539844, Terochem,prepared by the addition of 20 ml assay buffer to 1 vial of lyophilized GARGG). After a finalincubation of similar duration, the tubes were shaken prior to centrifugation at 3000 rpm for1220 min at 4° C. In the 12-hour version of the assay, tubes were held at room temperature 1hour after each shake, except in the last instance. Centrifuged tubes were immediatelyaspirated and counted in a LKB Wallac 1272 Clinigamma counter.In calculating plasma steroid or GtH concentration, non-specific binding counts werefirst subtracted from counts for each standard or plasma tube. Counts were then expressedas percentage of the reference counts (between 0 and 100 %). Steroid or proteinconcentration was determined by reference to the standard curve of percent binding versuslog concentration. Samples which bound less than 20% ( off the linear part of the standardcurve) were diluted in assay buffer and reassayed. Those which bound more than 80 % wereinterpreted as having undetectable levels of hormone.B. BLOOD SODIUM MEASUREMENTPlasma sodium and potassium were determined by the flame photometrictechnique of Blackburn and Clarke (1987). Briefly, 5 ul of plasma was diluted in 1 nillithium diluent and aspirated into a Turner (Case Instrument) clinical flame photometer.Up to 20 point-standards at 160 mMo1/1 sodium : 8 inMo1/1 potassium (160/8) and 140/5 wereinterspersed in an average run of 250 plasmas. Additionally standard curves consisting of 0- 250 in steps of 50 mMo1/1 were run at the beginning, middle and end of each run of plasmas.Use of this method removes the necessity of running duplicate samples for each plasma.C.STATISTICSAll data are expressed as mean ± standard error. Tukey HSD test was used todetermine differences between several means (P < 0.05). Analysis of serial sampling datawas by repeated measures ANOVA and Tukey HSD test. Arcsin and log 10 transformations13were used to obtain homogeneity of variance. Differences between two means weredetermined by t-test. Statistical analyses were performed using Systat version 5.0.CHAPTER 3NORMAL MATURATION OF CHINOOK AND STEELHEADA. INTRODUCTIONThe first aim of this chapter is to describe changes in reproductive condition andendocrines in chinook from the time of their first appearance in inshore marine waters, tocompletion of spawning. Identification of irregularity in physiological parameters in salmonmaturing in BW requires knowledge of changes in those parameters during normal SW/FWtransition and final maturation. Reproductive parameters and hormones were measured inchinook males and females in SW just at the end of the feeding and growth phase and thebeginning of final maturation, through the SW/FW transition period to the post-spawningstage. One aspect of osmoregulatory function, blood sodium, was also analysed. For thispurpose chinook from two similar river systems on the west coast of Vancouver Island wereexamined. The majority of samples were obtained from Nitinat River chinook, while post-spawning (spent) samples were obtained from Stamp River chinook.A similar description of reproductive parameters and endocrinology during finalmaturation was made in (Stamp River) steelhead, a species which often spends severalmonths in FW before spawning. Although steelhead could not be sampled during theSW/FW transition period comparison of final maturation in the two species may beinstructive. Adaptation to FW and final maturation occur more or less concurrently inNitinat River chinook, but are separated by several months in Stamp River late summersteelhead. It was anticipated that any involvement of reproductive hormones in FW15adaptation or conversely, influences that osmoregulatory processes exert on reproductivephysiology, would be manifested as differences in hormone profiles around the time of finalmaturation in the two species. In addition I expected to find differences in reproductivehormone profiles which help to explain the difference in post-spawning survival potentialbetween chinook and steelhead.Reproductive condition was evaluated on the basis of gonadosomatic index (GSI),hepatosomatic index (HSI), external signs of gender, and skin pigmentation. In females,oocyte germinal vesicle position and degree of hydration (or water content) were recorded.Changes in reproductive hormone levels were related to gonadal development and migratorystage.B. MATERIALS AND METHODSA. Sampling of maturing chinooka. Nitinat chinookChinook salmon (4-6 yr, 5-20 kg) in Nitinat Lake were captured by gillnet during theperiod Aug 19 to Sept 10, 1989. First appearance of fish in the lake was determined by sonarscanning and test fishing. Fishing was carried out between dusk and dawn at threelocations (Fig. 1) in the lake : Site 1: outlet (seaward) end, Aug. 19 (4 females, 5 males); Site2: mid way along the length of the 20 km lake, Aug. 29 (4 females, 5 males); and Site 3: inletend, Sept. 5 (5 females, 5 males). A 600 m x 12 m, 12 cm mesh monofilament gillnet was setacross the lake for one hour between midnight and dawn. As fish were landed over the16stern, external signs of maturation were recorded and blood was obtained by caudal puncturein 10 ml heparinized vacutainer tubes fitted with 18 gauge 3.5 cm hypodermic needles.Blood samples were kept on ice 1 hr before spinning in an IEC portable benchtop centrifuge.Plasma samples and fish were held on ice for up to 2 hr until the vessel returned toshore, at which time plasmas were frozen at -20' C, and fish were sampled for body weight,gonads and livers. Gonads and livers were weighed to obtain GSI (gonad weight x 100/ bodyweight) and HSI (liver weight x 100/ body weight). In the case of females the oocytegerminal vesicle position was recorded. A sample of 50 oocytes from each female wasweighed to the nearest 0.1 g. Oocytes were reweighed after 24 h drying at 80' C followed by1 h in a desiccator at room temperature. This was also done with chinook eggs donated by asalmon sportfishing guide from 8 females angled in Barkley Sound 25 km northwest ofNitinat Lake's outlet, Aug. 5-15, 1989.Three methods were employed in the daytime capture of fish in the freshwaterphase. Seine and gill nets were used in the middle reaches of the river to obtain maturingfish in transit (5 females, 5 males), while crowding and dipnetting was the method used formature fish in the hatchery raceways (8 females, 11 males). Blood samples as well as gonad,liver and body weights were obtained in the same manner as in the Nitinat Lake samplingprogram.Temperature, salinity and oxygen were measured at the surface and at depths of 1, 3,6, and 10 meters in Nitinat Lake during the sampling period.17PACIFIC OCEANLokepensSite 3Site 2KILOMETRES1=som===i0^5Figure 1. Nitinat system study area, with inset showing proximity of Nitinat and Stampsystems on the west coast of Vancouver Island, British Columbia. Also shown areNitinat Lake chinook sampling sites 1-3, the "Lakehead" area, and adult chinooknetpen maturation site, in relation to Nitinat Hatchery.18Figure 2. Stamp/Somass system showing the location of Robertson Creek Hatchery.19b. Stamp River chinookAs sampling of Nitinat chinook was carried out during the hours of both daylight anddarkness 24-hour blood sampling program was conducted at Robertson Creek Hatchery(RCH) (Fig. 2) to determine the effect of sampling time on endocrine and blood sodium levels.Stamp River chinook have a migration pattern similar to that of Nitinat River chinook. OnOct. 25, 1990, seven unovulated females (5-13 kg) were selected on the basis of ovarianfirmness as determined by palpation. Seven males were selected on the basis of a generalhealthy appearance, as all males examined at this stage were spermiating and expressedmilt when handled. Fish were removed to covered, flow-through aluminum containers of 2001 capacity, and allowed to acclimatize for 2 hours before being anaesthetized (0.4 ml 2-phenoxyethano1/1) and sampled.Samples of blood were taken from all 14 fish at 1530 h Oct. 24 and 1330 h Oct. 25,1990. For sampling, water flow was interrupted in sequence at each unit containing one fish.Three-fourths of the water was drained away and an emulsified 50:50 2-phenoxyethanol:water mixture was added through a gap in the lid to give a final concentration of 0.6 mIll.After three minutes lids were removed, narcotized fish were rolled on their sides in the waterand 5 ml of blood was removed from the caudal vasculature. Lids were replaced and waterflow was reestablished at 20 litres per minute within 5 min of the initial interruption. At2000, 0040, 0630, and 0920 h 5 fish of each sex were sampled, the remaining two of each sexserved as controls for handling stress. After the final blood sample, fish were killed with ablow to the head and sampled for germinal vesicle stage in oocytes, as well as GSI and HSI.Spent chinook were captured by gillnet in Glover Creek adjacent to the RCH site inNov. 1989. Specimens sacrificed on Nov. 5 and 13 (4 females, 6 males) were sampled20following the procedure described above for Nitinat chinook. In addition 5 males and 5females were bled and transferred to individual holding tanks on Nov. 8. On Nov. 11 deadfish were removed from the tanks and sampled for GSI and HSI while live fish wereanaesthetized and bled. On Nov. 13 the same procedure was followed and the remaining 4live fish were sacrificed and sampled for GSI and HSI.B. Sampling of steelheadStamp River steelhead (2.63 ± 0.65 kg) were collected from surplus broodstock inholding ponds at RCH in December/January 1987-88. These fish were maintained in a 2 x 30x 2.5 m deep covered concrete raceway. To control fungus the raceway (and fish) wereflushed for 1 h each week with malachite green at 1 ppm until the approach of finalmaturation in the hatchery brood fish, in the spring of 1988. On sampling days fish (6females, 5 males) were crowded to one end of the raceway, dipnetted individually from theraceway into a 4001 aluminum tub with 2-phenoxyethanol at 0.5 m1/1, bled, palpated andswabbed with malachite (5 ppt). Seven ml blood were collected weekly or biweekly betweenFebruary 5 and March 10, 1988, and the approximate dates of ovulation or spermiation wererecorded. These fish were not sampled for GSI and HSI on the final sample date as theywere held beyond final maturation and had lost many eggs on the floor of the raceway.Other fish of the same stock and original holding pond were sacrificed on several occasionsduring the serial-sampling program, to provide GSI and HSI data. Steelhead maturing in1988, were sacrificed on Feb. 1, 12 and 19 and on March 3. For fish maturing in 1989, fromwhich examples of earlier stages of maturation were sought, samples were obtained on Dec.8 and 27/88 and Jan. 15/89.21Blood plasma samples from all chinook and steelhead were analyzed for GtH, T, and17, 20-P by specific RIA described in Chapter 2. Female plasma samples were analyzed forE2 and male plasma samples for 11-KT. Representative plasma samples from females wereanalyzed for 11-KT to confirm undetectable levels, and likewise for E2 in males.C. RESULTSA. ChinookI. Migration and spawning of chinooka. Nitinat chinookIn 1989 chinook first appeared in Nitinat Lake on Aug. 19 and were captured at Site1 at the outlet end of the lake on that date. These first fish were silvery, resemblingmidsummer or feeding fish in coloration, and the sexes could not be externally differentiated.Fish were less silvery when captured in the middle reach of the lake (Site 2) on Aug. 29.Chinook appeared at the inlet end of the lake (Site 3) on Sept. 5. Skin colour was coppery,and females were distinguishable from males on the basis of abdominal width. Theirpresence at the mouth of Nitinat River, 2 km away, was confirmed 2 days later. Catch sizeon the 3 sampling dates in the lake were 9, 13 and 89, respectively, with only 4 females ineach of the first 2 catches.The number of fish at the mouth of the river and adjacent bay increased to Sept. 26.During this period darker fish with brown, red, and yellow flanks were seen in the lowertidal pools of the river, but few ventured above the tidal influence before Oct. 2. After Oct. 222fish in full spawning colours, devoid of silvery scales, were seen from Nitinat Hatchery to themouth of the river. The large schools at the head of the lake began to diminish after thatdate. Spawning began on Oct. 8 and continued until Nov. 4. Most spawning occurred in theriver 2 km on either side of the hatchery.b. Stamp River chinookChinook adults entered the Somass River enroute to the Stamp River around Sept.7 in 1989. Peak migration through the fish ladder at Stamp Falls occurred between Oct. 5and 15. Peak spawning activity occurred between Oct. 18 and 30 in Glover Creek. Spentchinook were observed in Glover Creek between Oct. 23 and Nov. 8.II. Water quality parametersa. Temperature (Fig. 3)Nitinat Lake was stratified with respect to temperature at all sampling sites.Surface temperature ranged from 1W C at Site 1 to 22' C at the head of the lake anddecreased, most rapidly between 1 and 3 m, to a uniform 13' C at 10 m depth over the lengthof the lake. Surface temperature remained relatively constant at the head of the lake frommid-summer until Sept. 27 when a slow decline began in response to the change of season.Summer temperatures in the river (12*-1T C) persisted until this time, then fell to W C atthe beginning and T C at the end of the peak spawning period.Water temperatures in Stamp River fall from September highs of around 14.5^ C to7° in November, and 2° in December. Temperature rises to 4° in February and T in March.23b. Salinity (Fig. 4)A salinity gradient was recorded at all lake sampling sites. The highest surfacesalinity was found at Site 1 (28 ppt) where an increase to 30 ppt between 1 and 3 m occurred.At Sites 2 and 3 surface salinity was slightly lower (25 and 21 ppt, respectively), but likewiseincreased to 30 ppt in the first 3 m of depth.c. Oxygen (Fig. 5)Dissolved oxygen in the surface layer at Site 1 ranged from 10 to 12 ppm during thesampling period. A slow decrease to 8 ppm occurred between 1 and 3 m, below which nochange was seen. From Site 2 to the head of the lake surface oxygen levels were lower thanthose at Site 1, and were depleted to lower levels at depth. Oxygen level fell to 5 ppm atabout the 4 m mark. Chinook cannot survive at oxygen levels less than 5 ppm, andthroughout most of the length of the lake the movements of adult salmon are restricted tothe top 4 m of water.d. hydrogen sulphideUnder normal conditions there is a stable layer of highly toxic hydrogen sulphide inthe anoxic zone from 4 m depth to the bottom of the lake. While the concentration of 112Swas not measured directly during the course of this study its presence was confirmed in both1988 and 1989 when inversions occurred at various times of the year, bringing thecharacteristic smell to the surface. Mortalities of chum adults were observed in the fall of1988 and 1989, following the peak chinook migration period.24Figure 3. Temperature profiles of Nitinat Lake, Aug. 19 to Sept. 5, 1989, at chinook samplingsites 1-3.25Figure 4. Salinity profiles of Nitinat Lake, Aug. 19 to Sept. 5, 1989, at chinook sampling sites1-3.26Figure 5. Dissolved oxygen profiles of Nitinat Lake, Aug. 19 to Sept. 5, 1989, at chinooksampling sites 1-3.27III. Changes in reproductive parametersA. Femalesa. Gonadosomatic indexNitinat chinookA steady rise in GSI (Fig. 6) took place in females from the first sample within theconfines of Nitinat Lake (12%) up to the penultimate sample in mid-river, prespawning fish(20%). No change was in evidence between the mid-river, unovulated fish and ovulated fishat the hatchery. Gonad weight averaged over all stages of maturation was 2.00 ± 0.53 kg.Stamp chinookGonadosomatic index in Stamp River serially sampled females was similar (22 ±1.7%) to that of ripe Nitinat females. Mean gonad weight was 2.12 ± 0.52 kg. In spentfemales GSI fell to 1.79 ± 0.76%.b. Germinal vesicle stageNitinat chinookOocytes from females in the first 3 samples (Aug. 19 and 29, Sept. 5) taken in NitinatLake were in the central (premigrating) germinal vesicle (CGV) stage. In 75% of mid-riverfish (Oct. 10) oocytes were in migrating germinal vesicle (MGV) stage, and 25% had oocytesin the peripheral germinal vesicle (PGV) stage. In ripe fish at the hatchery (Oct. 14) 900were at PGV or germinal vesicle breakdown (GVBD).28Stamp chinookAmong the serially sampled fish 3 of 7 females were ovulated, or partially so, andtheir oocytes were in the GVBD stage. Of the unovulated. fish 2 were in GVBD and 2 were inPGV stage.c. Water content of oocytesOocytes of females angled in the ocean west of Barkley Sound in August had thelowest hydration level at 50.2% (Fig. 6). Water content of oocytes from BW females wasobtained from the Sept. 5 Nitinat Lake sample (52%). Upon ovulation at Nitinat Hatchery57% of the egg mass was comprised of water.d. Hepatosomatic indexNitinat chinookThere was a slight though not significant increase in HSI between Aug. 19 and Sept.21 in Nitinat Lake females. Thereafter a slight decrease in the mid-river sample occurred,followed by a further decrease to 1.53% at ovulation.Stamp chinookHepatosomatic index of Stamp females at ovulation was significantly lower (1.00%, P< 0.003) than in Nitinat females at ovulation. There was a significant increase to 1.66% inspent fish after oviposition.29e. Blood sodium (Fig. 7)Nitinat chinookThe highest blood sodium concentrations (205 mM/1) in females were encountered insamples taken at the lower end of Nitinat Lake (SW). Significant decreases occurred in the 2subsequent lake fish samples. Blood sodium in FW female groups was lower than in mostBW groups except the Sept. 5 lake sample, and followed a decreasing trend from mid-riverthrough pre-ovulatory and ovulated fish. The lowest blood sodium (128 mM/1) levels of anyFW fish were recorded from ovulated females at Nitinat hatchery.Stamp chinook (Fig. 10)Plasma sodium was slightly higher in ovulated Stamp females than in Nitinatfemales. In spent females blood sodium was higher than in ovulated females at Nitinat orRCH.30Figure 6. Water content (Water) of oocytes and eggs, and gonadosomatic index (GSI) ofChinook females migrating from the ocean to spawning grounds and undergoingsexual maturation. SW = seawater (west coast of Vancouver Island); S1-S3 = Sites 1-3 (Nitinat Lake); MR = mid-river; Mature = ovulated at Nitinat Hatchery; Spent =spent fish at Robertson Creek Hatchery. Each value represents the mean + SEM.indicates significant difference from previous sample.31Figure 7. Plasma sodium of chinook salmon migrating from the ocean to spawning groundsand undergoing sexual maturation. Sites 1-3: Nitinat Lake; Mid R = mid-river;Mature = ovulated or spermiated at Nitinat Hatchery; Spent = spent fish atRobertson Creek Hatchery. Each value represents the mean + SEM. ''' indicatessignificant difference from previous sample.32B. Malesa. Gonadosomatic index (Fig. 8)Nitinat chinookGonadosomatic index in male chinook increased between Aug. 19 (Site 1) and Sept. 5(Site 3). In FW mid-river males before spermiation GSI was higher than in ripe males atNitinat Hatchery.Stamp chinookIn males that were spent or holding with spent females in Glover Creek GSI variedwidely (3.5-10.0%) averaging 6.00%, as compared with 6.34 ± 0.88% in Stamp River malesused in the 24-hour serial sampling study on newly mature fish.b. Hepatosomatic index (Fig. 8)Nitinat chinookNo changes were seen in HSI amoung males captured in Nitinat Lake through to themid-river sample. A highly significant increase (to 1.87%, P < 0.01) in HSI took placebetween mid-river fish and ripe fish at the hatchery.Stamp chinookIn recently spermiated males, sacrificed after serial sampling, HSI was significantlylower (1.06%) than in spermiated Nitinat males. Hepatosomatic index in spent males washigher than in the recently spermiated males.33c. Blood sodiumNitinat chinookChanges in blood sodium (Fig. 7) of male chinook showed the same trend as infemales from Nitinat Lake through to the mid-river stage. Wider variation in male bloodsodium level was seen in all lake samples however. A slight rise was seen in mid-rivermales. No changes took place between the mid-river stage and spermiated males at thehatchery.Stamp chinook (Fig. 13)Mean blood sodium in Stamp males was similar to that in Nitinat males. Bloodsodium was significantly lower in spent males.34Figure 8. Gonadosomatic index (GSI) and hepatosomatic index (HSI) in male chinookmigrating to spawning grounds and undergoing sexual maturation. Sites 1-3:Nitinat Lake; Mid R = mid-river; Mature = spermiated fish at Nitinat Hatchery;Spent = spent fish at Robertson Creek Hatchery. Each value represents the mean +SEM.^* indicates significant difference from previous HSI sample. GSI valueswhich are similar as determined by Tukey HSD (P > 0.06) are identified by the samesuperscript.35IV. Changes in reproductive hormonesA. Femalesa. Gonadotropin (GtH)Nitinat chinookPlasma levels of GtH varied only slightly around 7.5 ng/ml in fish captured in NitinatLake (Fig. 9), with a trend toward increasing values from the mouth to the head of the lake.The rise in GtH values between lake females and those in mid-river fish was not significant.There was a significant increase to a mean of 15.6 ng/ml in ovulating females sampled at thehatchery.Stamp chinookThe distribution of oocytes among germinal vesicle and ovulatory stages betweenfemales was mirrored in the profiles of some plasma hormones. While there are indicationsof significant differences in some reproductive hormone levels between females, in light ofthe small sample sizes within groups analysis of these data was not conducted. As well,there was scant evidence of diurnal cycling of plasma hormone and blood sodiumconcentrations in individuals, and the 24-hour profiles have not been reported here.However, hormone and sodium values for 3 ovulated and 7 spermiated Stamp River fish atthe beginning of serial sampling were compared to those for mature Nitinat fish to assess thevalidity of taking samples of spent fish from Glover Creek at RCH.36Figure 9. Gonadotropin (GtH) and Estradiol (E2) in female chinook migrating to spawninggrounds and undergoing sexual maturation. Sites 1-3: Nitinat Lake; Mid R = mid-river; Mature = ovulated fish at Nitinat Hatchery; Spent = spent fish at RobertsonCreek Hatchery. Each value represents the mean + SEM. * indicates significantdifference from previous sample.37Figure 10. Blood sodium (Na), Gonadotropin (GtH), Testosterone (T), and 17a.208-dihydroxyprogesterone (17,20-P) in ovulating female chinook at Robertson CreekHatchery.38Plasma GtH (Fig. 10) level in ovulated Stamp chinook was similar to that in ovulatedNitinat females. Gonadotropin was significantly elevated, relative to recently ovulatedfemales in sacrificed spent females and in serially sampled females, and remained at thislevel or increased in serial samples.b. Estradio1-1713 (E2)Nitinat chinookHighest plasma E2 (79 ± 16.2 ng/ml) was measured in females caught on Aug. 19 atthe outlet of Nitinat Lake (Site 1). In the two subsequent lake captures (Sites 2 and 3) E2was at or below the limit of detection (4 ng/ml) (Fig. 9). E2 was higher in the mid-riverfemales than in Site 2 and 3 females, and was depleted from the plasma of ovulated femalesat the hatchery.Stamp chinookNo E2 was detectable in the plasmas of peH-ovulatory and spent chinook at RCH.c. Testosterone (T)Nitinat chinookPlasma T (Fig. 11) was detected in all females sampled in the lake (31 to 242 ng/ml),and variation was extreme at all sites. Coefficient of variation was lowest in the Site 1sample (13%) and highest in the Site 2 and 3 samples (41 and 55%, respectively). Mid-riverfemales had the highest T levels while at ovulation, levels had returned to those seen in lakefish.39Stamp chinook (Fig. 10)Extreme variability in plasma T level was seen in ovulating fish at RCH as well (214± 98 ng/ml). Plasma T was at its lowest in spent fish, and did not change in the serialsamples.d. 17a,2013-dihydroxyprogesterone (17,20-P)Nitinat chinookLake female 17,20-P levels (Fig. 11) were a few ng/ml above the detection limit of 4ng/ml and no trend toward an increase in lake fish was in evidence. No change was seen inmid-river fish, but the concentration was greatly elevated in ovulated fish at the hatchery(1018 ± 373 ng/ml).40Figure 11. Plasma testosterone (T) and 17a,2013-dihydroxyprogesterone (17,20-P) in femalechinook migrating to spawning grounds and undergoing sexual maturation. Sites 1-3: Nitinat Lake; Mid R = mid-river, Mature = ovulated fish at Nitinat Hatchery;Spent = spent fish at Robertson Creek Hatchery. Each value represents the mean +SEM. * indicates significant difference from previous sample.41Stamp chinookPlasma 17,20-P level was as high (Fig. 10) in ovulating Stamp females as in matureNitinat females. There was a significant drop in 17,20-P in both the sacrificed and seriallysampled groups of spent fish. There was a slight decline indicated in the two spent femaleswhich survived through subsequent samples.B. Malesa. GonadotropinNitinat chinook (Fig. 12)GtH in males at Site 1 in the lake was 6.9 ng/ml, while the hormone was not detectedin the mid-river males. In spermiated males GtH increased (14.6 ng/ml) significantly overthe highest level recorded in the lake.Stamp chinook (Fig. 13)Recently mature males at RCH showed somewhat lower levels of GtH overall thanNitinat males. Gonadotropin reached its highest plasma level in spent males sacrificed atRCH 29.1 ng/ml), but was lower than this in spent males which were serially sampled.42Figure 12. Plasma GtH in male Chinook migrating to spawning grounds and undergoingsexual maturation. Sites 1-3: lower, mid and upper Nitinat Lake; Mid R = mid-river;Mature = spermiated fish at Nitinat Hatchery; Spent = spent fish at Robertson CreekHatchery. Each value represents the mean + SEM. * indicates significant differencefrom previous sample.43Figure 13. Blood sodium (Na), Gonadotropin (GtH), Testosterone (T), 11-ketotestosterone(11-KT), and 17a.2013-dihydroxyprogesterone (17,20-P) in spermiating male chinookat Robertson Creek Hatchery.44b. 11-Ketotestosterone (11-KT)Nitinat chinookThe level of plasma 11-KT (Fig. 14) was significantly lower in the first two lakesamples than in the last lake sample at Site 3. Concentrations in mid-river males weresimilar to those of Site 3 males, but increased significantly in spermiated fish sampled at thehatchery.Stamp chinook11-KT level was slightly lower in spermiating Stamp males (Fig. 13) than in Nitinatmales at the same stage, but was significantly lower than either of these groups in spentfish.c. TestosteroneNitinat chinookTestosterone (Fig. 14) level in males was high in the first sample, at Site 1 (130ng/ml) and declined significantly in the subsequent lake samples. There was a significantincrease to a peak in plasma T in the mid-river sample followed by a marked decline inspermiated males sampled at the hatchery.Stamp chinookSpermiating males at RCH had plasma T levels (Fig. 13) similar to those in males atNitinat hatchery. Plasma T in spent sacrificed males at RCH was significantly lower than inspermiated males. Starting values in the serially sampled group of spent males were similarto this and did not change in subsequent samples.45d. 17a,2013-dihydroxyprogesteroneNitinat chinook (Fig. 15)Plasma 17,20-P in males was at or slightly above the detection limit of the assay (4ng/ml) in lake samples from Sites 1-3, showing little change over the 3-week period. Similarlevels were seen in mid-river fish, but a steep increase occurred in spermiated fish at thehatchery.46Figure 14. Plasma 11-KT and T in male chinook migrating to spawning grounds andundergoing sexual maturation. Sites 1-3: lower, mid and upper Nitinat Lake; Mid R= mid-river; Mature = spermiated at Nitinat Hatchery; Spent = spent fish atRobertson Creek Hatchery. Each value represents the mean + SEM. * indicatessignificant difference from previous sample.47Stamp chinookIn recently spermiated males at RCH (Fig. 13), plasma 17,20-P was slightly lowerthan in Nitinat males of uncertain spermiation date. Plasma 17,20-P concentrations inspent males remained high in both groups of spent fish at RCH.48Figure 15. Plasma 17,20-P in male chinook migrating to spawning grounds and undergoingsexual maturation. Sites 1-3: lower, mid and upper Nitinat Lake; Mid R = mid-river,Mature = spermiated fish at Nitinat Hatchery; Spent = spent fish at Robertson CreekHatchery. Each value represents the mean + SEM. '' indicates significant differencefrom previous sample.49B. SteelheadI. Migration and spawning of steelheadThe migrating population of Stamp River steelhead in 1988 and 1989 was estimatedto be approximately 2500 in both years. In the lake-fed Stamp/Somass system completedrying of the riverbed in late summer and fall does not occur in the lower reaches as it doesin Nitinat River, and determination of the exact timing of the first migrants is not possible.Given the propensity of this late-summer stock to enter FW several months before finalmaturation it is also difficult to identify a pronounced peak in spawning. In 1989 steelheadwere recorded at the Stamp Falls ladder as early as Sept. 20, and numbers increased untilcounting was discontinued on Nov. 10, but a significant proportion of the run, comprised ofboth hatchery and wild fish, may remain below the ladder for months.The fish examined in this study were hatchery stock and the timing of their arrivaland maturation (and spawning in Glover Creek) in the 1988 brood year will be taken asrepresentative. Adult steelhead first appeared at RCH on Oct. 4 and continued to arriveuntil Nov. 5. Fish had lost all silvering in the scales and were brownish dorsally and whitishventrally. The sexes were distinguishable on the basis of snout shape and the red lateralstripe in males. These fish were held and blood-sampled when adults entered Glover Creekto pair and begin final maturation on Feb. 1, 1988. First spawning of steelhead began inearly March in Glover Creek and continued until March 29. Secondary sexualcharacteristics and spawning coloration were similar to those of adults first entering thehatchery in October. Synchronization of spawning activity was not as pronounced as inchinook; during most observation periods immature, ripe and spent fish could be seen50holding over the same gravels. The last spent steelhead were observed dropping to the lowerreaches of Glover Creek on April 5.II. Changes in reproductive parametersA. Femalesa. Germinal vesicle stageGerminal vesicle stage was variable between females in each sample except forMarch 3 (all GVBD) in 1988 broodstock, and Dec. 8 (all MGV) in 1989 broodstock. Oocytes inthe least mature, CGV stage, were not present in these fish. Oocyte stage was a mixture ofPGV and GVBD in all samples between Dec. 8 and ovulation in spring. GVBD stage wasseen in 25, 33, 40, 80, and 80% of females on Dec. 27, Jan. 15, Feb. 1, Feb. 12, and Feb. 19,respectively. Two germinal vesicle stages in the same ovary (PGV and GVBD) were seen intwo females.b. Gonadosomatic indexLowest GSI (13.1 ± 1.7%) was seen in females (Weight: 2.43 ± 0.72 kg) of most recentarrival at the hatchery, on Dec. 8, 1988 (Fig. 16). In ovulated fish, representing 20, 10, and100% of the last three samples respectively, GSI was marginally higher (18.1 ± 1.9%, P =0.053). Highest GSI occurred in Feb. 1, 1988 females however, where no difference in GSIwas seen between PGV and GVBD females.51c. Water content of oocytesWater content rose slightly in the samples collected from adults maturing in 1988,from 51.3% in maturing oocytes to 55.6% in ovulated eggs, although the difference was notsignificant (Fig. 16). There was no difference in hydration level between oocytes examined in1989 broodstock, representing the earliest obtainable, and the earliest oocytes collected from1988 broodstock.d. Hepatosomatic index (Fig. 16)The highest HSI (1.44%) occurred in females in the earlier stages of final maturation,and the lowest (1.07%) occurred at ovulation. Although the difference between these twodates was significant, HSI fluctuated in the intervening samples.52Figure 16. Gonadosomatic index (GSI), hepatosomatic index (HSI) and oocyte water content(Water) in maturing Stamp River steelhead females at Robertson Creek Hatchery.The data are presented in the correct order with respect to time elapsed since arrivalat the hatchery, with values for fish sampled early from the 1989 spawnerspresented ahead of values for fish sampled late from the 1988 spawners. Numbersinside bars represent sample sizes. Each value represents the mean + SEM. Valueswhich are similar as determined by Tukey HSD (P > 0.05) are identified by the samesuperscript.53B. Malesa. Gonadosornatic indexIn male steelhead (Weight: 2.88 ± 0.62) sacrificed at RCH there was a slight increasein GSI (Fig. 17) on Feb. 1, followed by a significant decline in the next sample. Values werein the 3.5% range in early samples of 1989 spawners. GSI fluctuated (4.1-2.7%) in fishsampled after more prolonged holding in 1988 spawners. Milt production by mature malesteelhead was not large when compared to the volumes produced by chinook, but fluid waspresent in the sperm ducts of most steelhead at the time of gonad removal. Milt volume wasnot measured in the sacrificed males, but there did not appear to be an increase in the lastsample from 1988 spawners.b. Hepatosomatic indexThere were no significant increases in HSI between consecutive samples (Fig. 17).There was a significant increase between the first two samples and the last sample (March3).54Figure 17. Gonadosomatic index (GSI) and hepatosomatic index (HSI) in maturing StampRiver steelhead males at Robertson Creek Hatchery. The data are presented in thecorrect order with respect to time elapsed since arrival at the hatchery, with valuesfor fish sampled early from the 1989 spawners presented ahead of values for fishsampled late from the 1988 spawners. Numbers inside bars represent sample sizes.Each value represents the mean + SEM. HSI values which are similar asdetermined by Tukey HSD (P > 0.05) are identified by the same superscript.indicates significant difference from previous GSI value.65III. Changes in reproductive hormonesA. Femalesa. Gonadotropin (Fig. 18)Six female steelhead were serially sampled in this phase of the study, which relatesplasma hormone profiles to ovulation date. The study was begun on Feb. 5 1988, andterminated on March 10. Two fish were found to have ovulated on Feb. 10, three more onFeb. 28, and the last one on March 9. Plasma GtH was low or undetectable until a few daysbefore ovulation. During ovulation GtH rose from about 8.5 to 25 ng/ml, and continued torise in ovulated fish for 2-3 weeks after ovulation.b. Estradiol-17f3 (Fig. 19)A consistent decline in E2 occurred in all females. In pre-ovulatory fish E2 rangedbetween 1.9 and 11.6 ng/ml, reaching a fairly uniform level of 2 ng/ml at ovulation, droppingto below the 1 ng/ml range or to below the limit of detection afterwards.c. Testosterone (Fig. 20)The pattern of plasma T concentration was similar to that of E2 in relation toovulation, with the difference that starting and final T values were higher. Levels declinedbetween the first sample and ovulation in all fish, and generally declined further afterovulation.56d. 17a,2013-dihydroxyprogesterone (Fig. 21)Changes in 17,20-P titre were variable between females. In most fish 17,20-P titrewas the highest of the steroid hormones studied around the time of ovulation. However, invarious fish, levels were high before and declining at ovulation while in others, highest levelswere attained after ovulation.B. Malesa. GonadotropinGonadotropin was not detected in the plasma of any male steelhead.Five males were sampled in the study, and an attempt was made to determine thetime of spermiation among them. Milt was first expressed by 4 males on Feb. 10, and by thefifth on March 4, but quantities varied greatly between fish. One male died between Feb. 26and March 4.67Figure 18. Plasma gonadotropin (GtH) in serially sampled steelhead females undergoingsexual maturation at Robertson Creek Hatchery, Feb.-March, 1988. Sample datesare standardized to the day of ovulation for each fish. F1-6 = females 1-6.58Figure 19. Plasma estradio1-175 (E2) in serially sampled steelhead females undergoingsexual maturation at Robertson Creek Hatchery, Feb.-March, 1988. Sample datesare standardized to the day of ovulation for each fish. F1-6 = females 1-6.59Figure 20. Plasma testosterone (T) in serially sampled steelhead females undergoing sexualmaturation at Robertson Creek Hatchery, Feb.-March, 1988. Sample dates arestandardized to the day of ovulation for each fish. F1-6 = females 1-6.60Figure 21. Plasma 17a,2013-dihydroxyprogesterone (17,20-P) in serially sampled steelheadfemales undergoing sexual maturation at Robertson Creek Hatchery, Feb.-March,1988. Sample dates are standardized to the day of ovulation for each fish. F1-6 =females 1-6.61b. Testosterone (Fig. 22)Plasma T was lower in males than in females at the outset of sampling and declinedto the end of the study by an average of 75%.c. 11-hetotestosterone (Fig. 23)Throughout the sampling period 11-KT levels in males were equal to, or greaterthan, T levels. Most fish showed a decline from highest levels of 11-KT approximately 10days before spermiation to lower levels at spermiation. A further decline followedspermiation in all fish, but the percent change within fish was not as large as for T.d. 17a,2013-dihydroxyprogesterone (Fig. 24)The level of 17,20-P in males did not reach the levels seen in females, remainingrelatively low (15 ng/ml) at spermiation. A gradual rise beginning 1 week after spermiationwas clear in two males which survived to the end of sampling, but in the other two there wasno change.62Figure 22. Plasma testosterone (T) in serially sampled steelhead males undergoing sexualmaturation at Robertson Creek Hatchery, Feb.-March, 1988. Sample dates arestandardized to the day of first milt expression for each fish. M1-5 = males 1-5.63Figure 23. Plasma 11-ketotestosterone (11-KT) in serially sampled steelhead malesundergoing sexual maturation at Robertson Creek Hatchery, Feb.-March, 1988.Sample dates are standardized to the day of first milt expression for each fish. M1-5= males 1-5.64Figure 24. Plasma 17a.206-dihydroxyprogesterone (17,20-P) in serially sampled steelheadmales undergoing sexual maturation at Robertson Creek Hatchery, Feb.-March,1988. Sample dates are standardized to the day of first milt expression for each fish.M1-5 = males 1-5.65D. DISCUSSIONA. femalesReproductive factors in female chinook salmon at the end of their SW residence andtransition into the FW environment have been documented in this portion of the study.Additionally, reproductive hormone concentrations in the peri-ovulatory stage of chinook,maturing shortly after FW entry, and steelhead, maturing over a longer period in FW, havebeen examined in detail.Both species were found to gain gonadal mass (GSI) between earlier and later stagesof oogenesis. Other studies have found that GSI may increase (Ueda et al., 1984a) orundergo little change (McBride et al., 1986) in maturing salmon. Hydration level of oocytesrose consistently as females of both species matured, indicating that water is absorbedduring the last stages of maturation. In chinook this constitutes an increase in waterdemand between first entry of brackish Nitinat Lake and ovulation in the river (approx. 0.07kg/ovary).Oocyte germinal vesicle stage was variable between individuals in most samplesfrom both chinook and steelhead. Asynchronous maturation in a spawning population is notunusual in salmonid species which can maintain fertility of gametes for several days afterfinal maturation (see Wallace and Selman, 1981). The two most advanced stages (PGV andGVBD, respectively) were not seen in chinook in the brackish water environment of NitinatLake, indicating a degree of synchronization between migratory stage and oocytematuration. Only in steelhead were oocytes of different maturation stage seen in the sameovary (PGV and GVBD). The fact that these were the two most advanced stages suggests66that oocytes progress more rapidly from PGV to GVBD than they do from one stage to thenext in earlier stages.Blackburn and Clarke (1987) tested osmoregulatory ability in juvenile salmonimmersed in SW for 24 h, and found that fish which had completed smoltification maintainedblood sodium level below 170 mM/1. Levels above 200 m.M/1 were observed in one sample oflower Nitinat Lake adult chinook newly arrived from the ocean. Similar high levels recordedby Sower and Schreck (1982) in maturing coho were associated with high adult mortality andlow egg survival. The high blood sodium level in Nitinat chinook may have been a normaltransient effect related to cessation of feeding and beginning of final maturation, but thepossibility of stress or netting injuries contributing to high sodium influx in these fish cannotbe discounted. Higher catches in subsequent lake samples, further into the brackish lake,allowed the rejection of fish suffocated by the net. Blood sodium levels in these fish fallwithin the normal range for adult salmon in SW (Sower and Schreck, 1982; Lam et al., 1982).Blood sodium decreased after transition from Nitinat Lake to FW in female chinook anddeclined further at ovulation.Similarities were seen in the patterns of change of reproductive hormones in chinookand steelhead females. Prior to ovulation gonadotropin was at basal levels, rose around thetime of ovulation, and reached highest levels in post-ovulation or spent fish of both species.This is a common pattern in salmonids (Scott et al., 1989; Dye et al., 1986) where an increaseof mGtH at the end of gametogenesis is believed to initiate production/release of 17,20-P,GVBD, and ovulation (Kawauchi et al., 1989). The continuous rise of GtH after ovulationand spawning may be a consequence of reduced steroid feed-back inhibition of pituitarygonadotrops (Jalabert, 1980; Scott et al., 1983) with the sequential decline of E2, T, and17,20-P, respectively.67A marked drop in plasma E2 took place in Nitinat Lake females after first entry offish from the ocean. Estradiol was low or declining after the first sample from steelhead aswell, dropping to basal levels before ovulation. Hepatic production of vitellogenin would beexpected to substantially decline within days of E2 disappearance from the blood (Wallace etal., 1987), and circulating vitellogenin would be rapidly depleted by oocyte uptake (Tyler etal., 1991). A large part of any increase in GSI after the decline in E2 in both chinook andsteelhead may therefor be attributable to water of hydration. Plasma E2, associated morewith vitellogenesis than final maturation, was higher in mid-river Nitinat chinook than inthe last fish sampled in the lake. Between these two sampling dates the commencement ofautumn rains caused a rise in river level and an increase in the depth of the FW layer overthe length of the lake. The occurrence in the mid-river sample of fish at a stage ofmaturation, assessed on the basis of plasma E2 level, which would be expected in fishmidway along the lake suggests fish may be induced to move more rapidly into FW bydecreasing temperature or increasing river flow.Between the decline of E2 and the rise of 17,20-P at ovulation plasma T rose to apeak in mid-river chinook, then declined thereafter. Similar preovulatory peaks of androgenwere reported in coho (Sower and Schreck, 1982), chum (Ueda et al., 1984a), and rainbowtrout (Sumpter et al., 1984). In serially sampled steelhead T declined from the firstsampling date in every female irrespective of ovulation date, and was depleted to nearlyundetectable levels in the last sample. The sampling procedure itself, or someenvironmental influence, may have been acting to affect T, but not other hormone levels.These observations nevertheless suggest that a reduction or cessation of T production occursin steelhead, as well as in chinook, sometime before ovulation.68A surge in plasma 17,20-P occurred at ovulation in chinook. The plasmaconcentration was much higher in chinook than in steelhead at ovulation in this study, andabout two-fold higher than any previously reported values (Dye et al., 1986; Ueda et al.,1984a; Van Der Kraak et al., 1984; Scott et al., 1983) for this hormone in a salmonid species.The reason for the exceptionally high level in chinook is not clear. Samples from mature fishin this study may have been taken nearer to the event of egg expulsion from the follicle, andrelease of 17,20-P into the body cavity, than in other studies. If rapid absorption takes placeblood levels would accurately reflect the levels in the body cavity. Rapid response in blood17,20-P level to changes in volume of eggs and ovarian fluid in the body cavity seems to occurin rainbow trout either manually stripped (Springate et al., 1984) or allowed to spawnnaturally (Liley and Rouger, 1990). This and the fact that the ovary gradually losessteroidogenic capacity after spawning (Van Der Kraak and Donaldson, 1986) explains thesignificant decrease in plasma 17,20-P in spent chinook.Plasma 17,20-P concentrations at ovulation were near the highest recorded for eachindividual or increased for a short time thereafter, in most steelhead females. Levels in thefinal samples from 5 of the 6 females were lower than at ovulation. Dead eggs in significantnumbers were present on the raceway floor after ovulation. Stripping eggs generally acts tolower plasma 17,20-P levels (Liley and Rouger, 1990) and partial loss of eggs in some of thesesteelhead may be involved in the variability of patterns between females. As fertility checkswere not performed the significance of an uninterrupted fall in 17,20-P from the first samplein one female is not known. Lowest levels of 17,20-P at ovulation occurred in steelhead,possibly as a result of lower GtH stimulation in the period preceding ovulation.69B. malesAs in females, blood sodium in male chinook captured at the lower end of NitinatLake was quite high. The trend in subsequent lake samples was toward lower levels as fishgained access to lower salinity in the lake. Blood sodium declined further as maturationapproached and the fish entered FW. A significant decline in blood sodium occurred betweennewly spermiated and spent males. This may take place during removal of sodium (Marshallet al., 1989) from the sperm duct lumen in the formation of seminal fluid.Gonadosomatic and hepatosomatic indices were not very useful in the evaluation ofmaturation in male chinook. As a relatively small percentage of total weight, small changesin gonad or liver weight may be obscured by increase in body weight due to water absorptionby somatic tissues (Kirschener, 1991) on the entry of BW and FW. Conversely, a significantdecline in body weight may occur as a result of starvation. Analysis of gonad and liverweights in relation to some measure of condition coefficient (body weight in relation tolength) may be a more useful way of monitoring change in those parameters.Gonadotropin is usually found to be lower in males than in females at a given stageof maturation (Dye et al., 1986; Fitzpatrick et al., 1986; Sumpter and Scott, 1989). In thisstudy GtH was about equal between the sexes in Nitinat chinook. Lowest levels of plasmaGtH were seen in mid-river Nitinat males, a finding which suggests, as do the E2 values formid-river females, that chinook in earlier stages of maturation will enter the river as soon asaccess is available.In Stamp steelhead males, GtH was undetectable throughout the maturationprocess, even after spermiation. This does not exclude the possibility of a role for GtH inspermiation. A peak in plasma GtH below the detection limit of the assay used in this studywas found in spermiating resident rainbow trout males Sumpter and Scott (1989). The70striking difference in GtH levels between chinook and steelhead males may help to explainthe difference in post-spawning life histories between these species. In chinook, gonadalandrogenesis is driven to a higher level than in steelhead by high levels of GtH. Highandrogen level in salmonids induces hyperadrenocorticism and elevated cortisol, whichmobilizes energy reserves (Robertson and Wexler, 1959; Donaldson and Fagerlund, 1969)and eventually leads to tissue degeneration. Androgens decline in mature and post-spawning chinook, but not to the low levels seen in mature steelhead. The magnitude of theprematuration peak, or perhaps the persistence of appreciable post-spawning androgenlevels may not allow the cessation of cortisol production. Without elimination of cortisol,tissue catabolism in chinook would continue unabated.Levels of 11-KT increased steadily in males migrating out of Nitinat Lake and up theriver, to peak in mature fish at the hatchery. Testosterone level peaked before that of 11-KTand was much lower than 11-KT in mature males. In steelhead males 11-KT was higherthan T at spermiation, and both androgens were declining. In sockeye (Truscott et al., 1986)and pink salmon (Dye et al., 1986) plasma T was lower than 11-KT and fell steadily towardfinal maturation. In coho (Fitzpatrick et al., 1986) T level fell while 11-KT rose with theapproach of spawning. Taken together these observations suggest that 11-KT has somefunction in late spermatogenesis not shared by T (Schulz and Blum, 1990), but this functionhas not been determined (Scott and Sumpter, 1989).In Nitinat chinook males a 20-fold surge of 17,20-P at the time of final maturationwas similar, albeit of somewhat smaller amplitude, to that in females. Unlike femaleshowever, the rise in 17,20-P in males continued after final maturation. Salmonids continueto produce milt for days or weeks after initial spermiation. The persistent high levels of17,20-P during this period was interpreted by Ueda et al. (1985) as evidence for control of71spermiation by this hormone. However, its major effect may be solely on milt volume. OnlyGtH is able to stimulate Na+ transport from the sperm duct lumen (Marshall et al. 1989),which creates the ionic environment required by maturing sperm.In general 17,20-P in male steelhead did not approach the levels in chinook, and didnot trend toward a notable peak at spermiation. Differences in milt volumes betweenspermiated steelhead and chinook seem to be correlated with the differences in 17,20-Plevels between the two species. Both the low levels of GtH and 17,20-P and low miltvolumes in steelhead may be an effect of prolonged confinement in concrete raceways. Theabsence of redd building activities by females may have deprived males of the stimulus Lileyet al. (1986) identified as complementary to 17,20-P and milt production in this species.72CHAPTER 4MATURATION OF CHINOOK AND CHUM IN BRACKISH WATERA. INTRODUCTIONThe experience at Nitinat hatchery has been that fertility of chinook femalescompleting final maturation in net pens in brackish Nitinat Lake is lower than in femalesmaturing in FW. The major objective of this phase of the study was to determine whetherfertility of fish maturing in BW differs from that of fish maturing in FW. If such a differenceexists can it be related to differences in plasma hormones or other physiological features?For example, Sower and Schreck (1982) found that fertility of SW-held coho was adverselyaffected in individuals with high blood sodium shortly before, but not necessarily duringovulation. Blood sodium level in female coho maturing normally in FW was between 135 and150 mM/1, while in SW-maturing fish levels ranged between 175 and 200 mM/1.Peaks in chum salmon migration and spawning occur 3-4 weeks after those ofchinook in the Nitinat system. A substantial proportion (40-70%) of chum eggs incubated atNitinat hatchery come from females captured and held to final maturation in the lake.Fertilization rate is high in egg lots from these fish maturing at the same net pen locationwhere attempts to obtain comparable fertility in chinook have been largely unsuccessful.Similar high fertility was seen in chum captured and held at another BW location on thewest coast of Vancouver Island (Lam et al., 1982). Because fertility of females maturing inBW and FW is equal in chum but not in chinook, the question is raised as to whetherphysiological differences can be detected in chum and chinook that may relate to the abilityto produce viable eggs in BW.73A secondary aim of this chapter was to examine hormone profiles and blood sodiumlevels in preovulatory chinook delayed at the Nitinat lakehead by low water levels in theriver. The question in that situation is whether reproductive hormone profiles in the blood ofBW females are different from those normally seen in FW females at a similar stage ofmaturation. It is also of interest to know if blood sodium level in these fish indicates osmoticimbalance in BW and whether this correlates with hormone profiles.B. MATERIALS AND METHODSIn October of 1988 maturing chinook females were captured by seine on severaloccasions in Nitinat Lake and held to final maturation and ovulation in BW net pens nearthe mouth of Nitinat River. In the first group, females were held in the net pen until severalhad ovulated, a practice followed with females maturing at the hatchery, and eggs weretaken from 8 fish on Oct. 8. Eggs were obtained on the same date from a group of femalesallowed to mature in FW (n=8). The post-ovulation age of eggs at the time of fertilizationvaried by as much as 5 days within these groups. Blood and ovarian fluid was not collectedfrom these fish as sampling equipment was not available at the net pen site on this occasion.Both groups of eggs were fertilized with sperm from males maturing in FW.Eggs were transported with several volumes of air in sealed plastic buckets andfertilized, after temperature acclimatization, within 3 h of collection. In clean, dry 10 1plastic buckets 10 ml sperm was mixed into each batch of eggs (approx. volume 21) andallowed to stand for 5 min. Approximately 2 volumes of water were poured into each bucket,gently stirred and allowed to stand for 1 min. The eggs were then rinsed with severalvolumes of water and poured into upwelling Heath tray incubators (Heath Techna Corp.)74supplied with 8' C well water at 111/min. The same fertilization procedures were followedwith subsequent groups of eggs, with modifications noted.A second group of eggs from fish of relatively recent ovulation, was taken on Oct. 14.From fish captured and/or determined to be unripe on Oct. 10, 6 net pen and 6 hatcheryfemales were randomly selected. Shortage of space at the net pen site did not allow for theholding of male broodstock in great numbers, but 3 were maintained for comparison with FWmales. Half the eggs of three Oct. 14 females from each location were fertilized with spermfrom net pen males, and the other half with sperm from FW males. Each of these 6 malesfertilized eggs from one hatchery and one net pen female. Eggs of the 3 remaining femalesfrom each location were fertilized with sperm from FW males, such that each of 6 differentmales fertilized all the eggs of 1 female. Eggs that had been divided and fertilized with thetwo types of sperm were incubated in trays divided with plexiglass partitions.After rains commencing Oct. 12 had brought the river level up, and the numbers offish holding at the lakehead had decreased, a third group of net pen females, evaluated asnew arrivals, was established, on Oct. 15. These fish (n=14) were checked every 2 days andtheir eggs, ovarian fluid and blood taken when ripe between Oct. 17-26. A third group ofeggs ovulated in FW was obtained from females (n=16) arriving ripe at the hatchery duringthe same period, but blood and ovarian fluid were not obtained from these fish. Eggs fromboth types of females were fertilized with sperm from FW males. Mortality was recorded foreach of the above tests at the eyed egg stage.Hormone profiles, blood sodium, and other reproductive parameters of the last groupof BW females (Oct. 17-26) in 1988 were compared with those from females arriving newlyripe at the hatchery on Oct. 12, 1989. The first rains came earlier in 1989 than in 1988, butit was possible in both years to differentiate, on the basis of coloration and general condition,75between ripe fish which had been delayed in the lake or lower river and those which came infresh. Fish which had been delayed in the lake were very dark-skinned and those delayed inthe lower river were infected with patches of fungus. Varying proportions of the eggs fromdelayed fish were in the so-called "water-hardened" state; enlarged, hard to the touch, andinfertile. Lighter coloured, nonfungused fish were judged to be most similar to the Oct. 17-26, 1988 fish among the FW types encountered in 1989. Eggs from these females (n=8) werefertilized with sperm from FW males.A comparison of hormone levels, blood sodium, and egg hydration was also madebetween the 2 above-mentioned groups and females captured at the lakehead on Sept. 21,1989. Chinook had been holding in this area (see Fig. 1) for several weeks, and 200 wereseined and transferred to the net pen site for holding to maturity. Five females weresacrifice-sampled 2 hours after capture. Mortalities of 50% in the main group forced thedecision to release the survivors on Sept. 23. Although no ovulated eggs were obtained forfertility tests, it was important to document the physiological condition of fish that had beendelayed in BW.On Nov. 4 of 1988 chum salmon females maturing at the net pen site (n=7) andhatchery (n=7) were sampled. Reproductive hormone and sodium levels were analyzed inthe plasmas of these fish. After fertilization of eggs (approx. volume 1 1) from both locationswith milt (5 ml) from FW males, mortality was recorded at the eyed stage.In the interest of simplifying the language of description, eggs from femalesmaturing in BW are referred to as "BW eggs". Similar abbreviations yield the terms "FWeggs", "BW milt" and "FW milt", respectively for gametes matured at a given location. Theeggs produced by one female are referred to as a "batch".76C. RESULTSI. Reproductive parametersa. Egg fertilityChinookHighest mortalities occurred in chinook eggs taken at the lake pens on Oct. 8, 1988(Fig. 25). Mortality was above 90% in the eggs from 3 females in this group. Losses in eggstaken at the hatchery were highest on Oct. 8 as well, and 48% mortality was the highest atthat location.Mean mortality was lower on Oct. 14 than on Oct. 8 in eggs taken at both locations.Fertility of milt obtained from BW and FW males was compared in this test. High mortality(58-98%) occurred in 2 of the 3 batches of BW eggs divided and fertilized with BW and FWmilt, and in the third batch, mortalities were below 8%. Within these batches, mortality wassimilar whether fertilization was with FW or BW milt. In the 3 undivided batches of BWeggs, fertilized with FW milt from 3 previously unused males, mortality did not exceed 20%.After fertilizations by the same males used on divided batches of BW eggs thehighest mortality in divided batches of FW eggs was 6.2%. In the 3 undivided batches of FWeggs, fertilized with the last 3 FW males, mortality ranged between 15 and 76%.Lowest mortalities among BW eggs occurred in those obtained between Oct. 17-26and fertilized with FW milt. Mortality was high in three batches of these eggs (100, 23, and22%, respectively), but relatively low in the other 11 batches. In the FW females fertilizedover the same period mortality ranged between 1.5 and 15.2%. Mortality among FW eggswas lowest in those fertilized on Oct. 12, 1989.77In many groups of eggs from BW and FW and on most dates, mortality varied widely.In the Oct. 8 BW eggs for example, mortality ranged from 2.5 to 100%, but a bimodaldistribution was not indicated, rather the range of possible mortality rates between "high"and "low" were represented. Neither Tukey's HSD test nor the Mann-Whitney test identifieddifferences between any groups. There was only a low correlation between blood sodium andmortality.Churn (Fig. 29)There was no difference in mortality to the eyed stage between chum eggs from lakepen and hatchery ripened females (4 and 5.3%, respectively) fertilized with sperm from FWmales.b. Gonadosomatic index (Fig. 26)ChinookFemales sampled at the lakehead in 1989 had significantly lower GSI than both theBW and FW ovulated groups.c. Germinal vesicle stageOnly the fish sampled at the lakehead in 1989 held oocytes in PGV stage; eggs fromall ovulated fish were in GVBD.78d. Water content of oocytesChinookOocytes from unovulated females at the lakehead had the lowest hydration level (Fig.26). Ovulated eggs from the lake pen fish were significantly higher in water content, andeggs from hatchery females contained more water than lake pen eggs.e.Blood and ovarian fluid sodiumChinookHigh blood sodium (191 mM/1) was seen in lakehead females (Fig. 27). In lake penfemales at ovulation, blood sodium was significantly higher than in hatchery females, andsignificantly lower than in lakehead females. Ovarian fluid (Fig. 27) sodium was also higherin lake pen fish (158 mM/1) than in hatchery fish (144 mM/1).Chum (Fig. 29)Blood and ovarian fluid sodium did not differ between net pen and hatchery chumfemales (blood: 148 and 147 mM/1, ovarian fluid: 150 and 151 mM/1, respectively).II. Reproductive hormonesa. GonadotropinChinookPlasma GtH (Fig. 28) was lowest in the unovulated females caught at the lakeheadon Sept. 21, 1989. Between the ovulated fish at the lake pens or hatchery, highest levelswere encountered in the lake pen group, though the difference was not significant.79ChumPlasma GtH (Fig.29) in chum females ovulating at the lake pens was 15.4 ± 6.3ng/ml, compared to 23.6 ± 12.9 ng/ml at the hatchery.b. TestosteroneChinookThere was no difference in T level (Fig. 28) in the plasmas obtained from females atthe 3 locations (range 128-156 ng/ml).ChumTestosterone (Fig. 29) in lake pen females was significantly elevated compared tohatchery females.c. 17,20-PChinookIn lakehead females 17,20-P was only a few ng/ml above the detection limit (Fig. 28).In ovulated fish at the lake pens and hatchery the levels were much higher, 450 and 1020ng/ml respectively, and significantly different (P < .03).ChumIn lake pen females 17,20-P levels (Fig. 29) averaged 233 ± 192 ng/ml. Highvariability was seen in hatchery females as well (178 ± 103 ng/ml), and the differencebetween the 2 groups was not significant.80Figure 25. Mortalities to the eyed stage among chinook salmon eggs taken from females inbrackish (BW) or fresh (FW) water, October 8-26, 1988, and fertilized with milt fromFW or BW males, eg., BW/FW = BW female fertilized with FW milt. Valuesrepresent mean + SEM.81Figure 26. Water content (Water) of oocytes and eggs, and gonadosomatic index (GSI) inchinook salmon females at Nitinat Lake and Hatchery, 1988 and 1989. Solid bars:unovulated fish holding at lakehead, Sept. 21, 1989; Diagonal hatching: recentlyovulated at net pens, Oct.17-26, 1988; Cross hatching: recently ovulated at hatchery,Oct. 12, 1989. Values represent mean + SEM. * indicates significant difference frompreceding value.82Figure 27. Blood and ovarian fluid sodium in chinook salmon females at Nitinat Lake andHatchery, 1988 and 1989. Solid bars: unovulated fish holding at lakehead, Sept. 21,1989; Diagonal hatching: recently ovulated at net pens, Oct.17-26, 1988; Crosshatching: recently ovulated at hatchery, Oct. 12, 1989. Values represent mean +SEM. * indicates significant difference from preceding value.83Figure 28. Plasma hormone levels in chinook salmon females at Nitinat Lake and Hatchery,1988 and 1989. GtH=gonadotropin, T=testosterone, 17,20-P=17a,205-dihydroxyprogesterone. Solid bars: unovulated fish holding at lakehead, Sept. 21,1989; Diagonal hatching: recently ovulated at net pens, Oct.17-26, 1988; Crosshatching: recently ovulated at hatchery, Oct. 12, 1989. Values represent mean +SEM. Hormone levels which were similar as determined by Tukey HSD Test (P >0.05) are identified by the same superscript.84Figure 29. Plasma hormone (see Fig. 28 for abbreviations) and sodium levels, and eggmortality rates in chum salmon females at Nitinat Lake and hatchery, Oct. 28, 1988.Solid bars: recently ovulated at net pens; Hatched bars: recently ovulated athatchery. Values represent mean + SEM. * indicates significant difference frompreceding value.85III. Water qualitya. TemperatureSurface temperature had decreased to 10 . C following the onset of seasonal rains(Oct. 18) at the lake pen site in 1988. The thermocline (13.5° C) was encountered at 1 m (Fig.30), inverted due to the density difference between FW and BW. Temperature in the first 2m at the net pen site had fallen to T C by the Nov. 4 chum sampling date. On Sept. 21, 1989the lakehead surface temperature was 16' C, with a thermocline at 1 m.b. SalinityThe surface water was nearly fresh (2 ppt) at the net pen site in 1988, on Oct. 18, butsalinity increased sharply (Fig. 30) within the first metre to 22 ppt and reached 24 ppt at 2.5m, the depth of the net pen. The FW layer extended to 2 m on Nov. 4. The salinity profile atthe lakehead was similar to the profile at Site 3 in 1989.c. OxygenBetween 2 and 2.5 m, oxygen decreased from 6.2 to 4.3 ppm at the net pen site (Fig.30), on Oct. 18. Oxygen level remained above 6 ppm to a depth of 3 m on Nov. 4. At thelakehead in 1989, the anoxic layer was located 3 m below the surface.86Figure 30. Water quality parameters at Nitinat Lake net pen site, October 18, 1988.87D. DISCUSSIONMortalities were higher, though not significantly so, in chinook eggs taken fromfemales at the net pen site during a period of high salinity and temperature, and low oxygen(Oct. 8 and 14) than in eggs taken after Oct. 17. The incidence of mortality was also higherin eggs from FW females taken earlier in the season than in those taken later. These werefish arriving at the hatchery soon after the first rise in river level permitted access to FW.Their exact ovulation dates are unknown, but it can be assumed with some confidence thatthey had been in BW only a few days prior to arrival at the hatchery. Plasma hormone andsodium levels seen in the Sept. 21, 1989 lakehead chinook may be representative of theirphysiological state prior to FW entry. If so, effects of high prematuration blood sodium,similar to those postulated by Sower and Schreck (1982), may have contributed to the highmortality rate in the first eggs taken at the hatchery.Mortalities from the last group of chinook eggs taken at the net pens were almost aslow as in FW chinook eggs taken at the same time, and in chum eggs from the net pens andhatchery one week later. By the time the third sample of chinook eggs was taken, higherflow rate from the river increased the depth of the FW layer, resulting in higher dissolvedoxygen and lower temperature and salinity in the top 2 m of water at the net pen site. Thesesame conditions existed when the chum were sampled for fertility. Normally theseconditions prevail during the majority of the hatchery's chum spawnings at the net pen site.Although mean egg mortalities were slightly higher for late October net pen chinookthan for hatchery chinook and chum, the higher average for net pen chinook was caused by 388of the 11 fish examined. One of these, suffering 100% egg mortality, had higher blood andovarian sodium than the average for the group. The other 2, each suffering 20% mortality,had higher than average blood sodium but average ovarian fluid sodium. A strongcorrelation between high blood and ovarian fluid sodium and egg mortality was reported bySower and Schreck (1982) in coho held to maturity in SW, and similarly for chum (Lam et al.,1982; Morisawa et al., 1979) and pink salmon Wertheimer (1984) in SW and BW.In the current study there was no significant correlation between blood sodium andegg mortality. However, high sodium was seen in the blood of preovulatory chinook at thelakehead before the river level rose. Similar high levels were seen by Sower and Schreck(1982) in preovulatory coho held in SW. Blood sodium was somewhat lower at ovulation, butin some of these fish, low fertility nevertheless developed. A similar process may be at workin chinook in Nitinat Lake. Brackish water and FW eggs used in the Oct. 8 fertility testscame from fish which in the preovulatory stage may have been exposed to the same waterconditions as those experienced by the Sept. 21 lakehead fish. High blood sodium developedin these fish, and it may be in this period that the future viability of unovulated oocytes iscompromised. It would appear that the process in question does not affect the entire ovarysimultaneously, or that it is reversible by the availability of FW, as no batches of eggs takenin FW on Oct. 8 sustained mortalities as high as those seen in BW eggs taken on the samedate.Although the influence of concurrent high temperature and low oxygen may bedirectly or indirectly involved, it is clear that some chinook females produce a higherproportion of infertile eggs during late summer conditions of high salinity existing at theNitinat Lake net pen site. Fertility of males does not seem to be as strongly affected.89During the periods of lowest river flow, unacceptably high mortality of chinook adultsoccurs during net pen holding in BW, as seen in the group of fish from which the Sept. 21sample was obtained. High mortality occurred in this group in the 2 days following capture.I suggest that this is due to high temperature and low oxygen whose combined effects are toincrease the ventilation rate of gills. The increasing volume of water passing the gills mightincrease the dehydrating and hypernatremic effects of high salinity on the fish, leading toosmotic imbalance and death. Fertility of eggs in the survivors of these conditions wouldalmost certainly be diminished.In years when the change of season coincides with arrival of chinook at the lakehead,and cooler, well oxygenated water of low salinity becomes available at the net pen site, thepractice of estuarine capture and maturation may be a viable option for chinook broodstock.Chum salmon normally arrive at the lakehead later in the season than chinook, when watersalinity, temperature, and oxygen levels have moderated somewhat. High survival of adultsand eggs is usually obtained when chum broodstock are held in the net pens. It appears thisis not due necessarily to a superior hypoosmoregulatory ability of chum as compared tochinook, but rather to an improvement in water conditions between the peaks in chinook andchum migration.Highest blood sodium was seen in chinook delayed at the lakehead by low riverwater. Plasma T was as high in lakehead females as in ovulated females, but 17,20-P andGtH were low. In mature females at the net pens, T and GtH were at levels similar to thosein females maturing at the hatchery, but 17,20-P levels, although much higher than inlakehead fish, were half those in hatchery females.The reduction of 17,20-P in fish maturing in BW suggests a degree of suppression ofmaturation processes under saline conditions, as seen by Sower and Schreck (1982) in coho.90Plasma 17,20-P levels in mature BW fish were many times higher than in fish holding at thelakehead, not surprisingly as lakehead fish were not mature. In order to determine whetherdelayed entry of FW may have curtailed maturation in lakehead fish, a comparison ofmaturation timing between groups from the same school held in BW and FW would berequired. The question of which reproductive hormones might correlate with reduced abilityto osmoregulate in BW is of greater concern in this thesis however, and the observations inlakehead fish do not strongly implicate 17,20-P. Lakehead fish had blood sodium levelshigher than the critical level of 170 mM/1 defined by Blackburn and Clarke (1987) for smoltsadapting to SW. At the same time T levels were as high in the lakehead fish as those inmature fish. Based on the observation that blood sodium rose while prolactin, thefreshwater-adapting hormone, remained low in SW-maturing chum Hirano et al. (1990)suggested that: "high levels of maturational (and other) steroids such as 17,20-P are involvedin the development of the impaired hypoosmoregulatory ability in Pacific salmon at the timeof spawning". The possibility that 17,20-P may interfere with the ability to osmoregulate inSW, cannot be ruled out. The observation that T was high in Nitinat lakehead chinookpoints to this hormone as having a likelihood of involvement in the failure tohypoosmoregulate as well.91CHAPTER 5SW TRANSFER OF POST-SPAWNING SALMON AND STEELHEADA. INTRODUCTIONAs demonstrated in Chapter 4, delayed FW entry of maturing chinook, in whichreproductive hormones are high, results in high blood sodium. Fertility of chinook maturingin BW or FW in the early part of the season, following protracted BW residency, is lowerthan that of chinook maturing later in the season, when access to FW increases. In thesesituations high levels of reproductive hormones may be interfering with hypoosmoregulation.After completion of spawning steelhead do not inevitably undergo further decline ofcondition to the point of death, as is the case in chinook and the other species of Pacificsalmon (Healey, 1991). A proportion of post-spawning steelhead populations travelsdownstream after spawning and readapts to SW (Hart, 1980). This leads to the question ofwhether there are post-spawning differences in reproductive hormones between steelheadand salmon which affect the ability to adapt to SW. In particular, these contrasting life-histories provide an opportunity to examine responses in sex hormone levels andosmoregulatory adaptation to the challenge of SW transfer. For example, although there aresimilarities in the profiles of reproductive hormones in the last few weeks of maturation inchinook and steelhead the peaks of concentration for most hormones are considerably lowerin steelhead. I suggest this is evidence for the existence of a greater degree of proximatecontrol over reproductive hormones in steelhead than in other salmon. This allows a widerrange of responses to environmental conditions, including salinity.92Mayer et al. (1990) could not find differences in the levels of 5 endogenous androgensto correlate with the difference in SW adaptability in mature and immature Atlantic salmonparr. On the other hand, high plasma levels of exogenous sex steroids impaired SWadaptability during smoltification in amago salmon (Miwa and Inui, 1986). Reproductivehormone levels in post-spawning in chinook and mature steelhead are considerably higherthan in the Atlantic salmon parr of Mayer et al. (1990). Will the higher reproductivehormone levels of mature salmonids permit the delineation of a more clear-cut relationshipbetween impairment of SW adaptability and reproductive endocrinology? Does SW transferprecipitate a reduction of circulating reproductive hormones in the SW readaptation processof steelhead?These questions are addressed in the current section which focuses on post-maturesteelhead and coho. Spent steelhead and spermiating male coho were transferred to SWfacilities and blood-sampled over a period of weeks as salinity was gradually increased. Bothmature jack and adult coho were examined in this study to determine whether a difference inendocrine levels and post-spawning survival and SW adaptability exists. Blood was analysedfor hormone and sodium levels, and observations were made on behaviour and condition.B. MATERIALS AND METHODSOn March 31 1988, Glover Creek steelhead were captured by electroshock holdingover the spawning gravel in the only available spawning area in the creek, blood-sampled,tagged, and transported by tank truck to holding in 1/3 SW (10 ppt) at Pacific BiologicalStation (PBS) in Nanaimo, B.C. Salinity was increased to 2/3 SW on April 10 and to full SW(29 ppt) on April 21. Fish were anesthetized and force-fed once (whole or cut herring), andsampled on a weekly or biweekly basis.93Sampling was conducted on mature male coho during the winter of 1988 and springof 1989 at Big Qualicum and Robertson Creek Hatcheries, and subsequently at PBS. AtRCH, spermiating coho jacks were blood-sampled once and discarded on December 27, forcomparison with mature adult males. Blood sampling was kept to a minimum in the seriallysampled adults and jacks to conserve their health through to SW transfer. At Big QualicumHatchery adult male coho were blood-sampled as captured between Dec.11 and 13 to providea representative sample for spermiating adults. A group of unsampled spermiating adultmales was collected at the same time.Coho jacks from RCH (n=8) and adults (n=8) from Big Qualicum were transported to1/3 SW at PBS on Dec. 17 and 18. Salinity was increased to 1/2 SW on Jan. 11, 1989 and to2/3 SW on Feb. 4. Fish at PBS were anaesthetized (0.4 m1/1 2-phenoxyethanol) and force-fedoxytetracycline treated fish food (Oregon Moist Pellets: Moore-Clarke, LaConner, Wash.) andstripped of milt once weekly. Blood sampling was conducted on a biweekly basis until Feb.22.C. RESULTSI. General observationsA. Activity, feeding, growth, and survivalSteelheadSteelhead were alert and energetic from the beginning of the study. Force-feeding ofsteelhead with herring pieces was conducted on April 9, thereafter active feeding took place94when whole herring was offered on non-sampling days. Growth was evident in some fish.Survival over the 2 months of sampling was 60%.CohoJacks were more active than adults, but not as active as steelhead, and relativelyeasy to catch. Force-feeding was necessary in all feeding sessions. No growth took place inadults or jacks. Over the 2 month sampling period mortality was 100% in jacks and 70% inadults. In dead fish and those killed on Feb. 22, stomachs were full to capacity withundigested food.B. Milt production and skin colourSteelheadMilt was expressed by all males when collected at RCH. Only one fish was producingmilt after the first week in 1/3 SW, and this ceased after the third week.Dorsal spots began to lighten and a silvery sheen characteristic of SW fish returnedto the flanks, with fading of the red lateral stripe, around the third week after capture in allbut the milt-producing male, which remained darker than the other steelhead.CohoBoth adults and jacks expressed milt through to the end of the study. All cohoremained dark and kept their spawning colours.95II. Changes in reproductive hormonesA. GonadotropinSteelheadAll samples, including the first from fish electroshocked in FW, showed nearlyundetectable levels of GtH (1.5 ng/ml). Average values from ovulated and spermiatedsteelhead sampled 3 weeks earlier (March 10) are included here (Fig. 31) to illustrate themagnitude of the decline in hormone levels. GtH was higher in the 2 males which died afterSW transfer, and in the spermiating male, but was undetectable after the first 2 SWsamples.CohoPlasma GtH in FW was similar in coho adult males and jacks (Fig. 32). By the timethe first sample in BW was taken, one month after transfer from FW, jack mortalities hadreduced their number by 50% to 4 fish. GtH was not detectable in two of these fish (below 3ng/ml), but was relatively high in the other two. On Feb. 4, plasma GtH could be detected inonly one of the three remaining fish, and on Feb. 22 no jacks were left alive.In adult males, 5 fish survived to the first BW sample, with similar GtH levels withinthe group (15.4 ng/ml). GtH was detectable at around 6 ng/ml in two of three survivors onFeb. 4, and reached 6.5 and 19 ng/ml in the two surviving to Feb. 22.96Figure 31. Plasma hormone levels in spawning (March 10, FW) and post-spawning (March31, FW; April 9, BW) steelhead, spring 1988. Solid bars: gonadotropin, diagonalhatching: testosterone, cross hatching: 17,20-P, plain: il-ketotestosterone. Numbersabove groups of bars give sample size. Values represent mean + SEM.97Figure 32. Plasma hormone levels in spawning (Dec 13 and 27, FW) and post-spawning (Jan19-Feb 22, BW) coho males, winter 1988. Solid bars: gonadotropin, diagonalhatching: testosterone, cross hatching: 17,20-P, plain: 11-ketotestosterone. Numbersabove groups of bars give sample size. Values represent mean + SEM.98B. TestosteroneSteelheadHighest testosterone (20 ng/ml) was measured in one of the males that died beforethe third week in BW, and in the spermiating male. In all other steelhead, includingfemales, T was undetectable at all sampling times in SW.CohoPlasma T in jack coho before transfer to BW averaged 49 t 12 ng/ml, decreasingsignificantly in the next sample.Similar T levels were seen in adult male coho before transfer (64 t 33 ng/ml), halvingin the first month in BW. In the second month, T dropped to below 1/4 of the FW value.C. 11.-KetotestosteroneSteelheadNo spent fish had measurable levels of plasma 11-KT.CohoHigh levels of 11-KT were recorded in FW jacks. These levels remained unchangedone month later in BW, and were halved in the last survivors on Feb. 4.Plasma 11-KT level in adult males in FW was similar to that in jacks, and followed asimilar course to the Feb. 4 sample. No further decline was seen in the last survivors of thisgroup on Feb. 22.99D. 17a, 200—DihydroxyprogesteroneSteelhead17,20 -P was not detectable in the plasmas of male or female steelhead.CohoPlasma 17,20-P was high in FW jacks and did not change during the first month inBW. In the Feb. 4 sample levels had declined significantly to 1/2 the FW value.High levels of 17.20-P were also present in FW adult males. Levels declined after 1month in BW and rose again in the last sample.III. Blood sodiumSteelheadSpent steelhead electroshocked from Glover Creek before transport to PBS hadplasma sodium levels of 155 mM/1. Seventeen days later, after 1 week in 10 ppt BW, plasmasodium was still at 165 mM/1. After 1 month (May 28) in full SW plasma sodium wasunchanged. There was no difference between males and females.CohoIn jacks and adult coho in FW, blood sodium was similar to that in steelhead (153and 154mM/1, respectively). One month later, after 1 week in 20 ppt BW, jacks had levels ofblood sodium similar to those in adults (164 and 163, respectively). Two weeks later, still in20 ppt, blood sodium level in jacks was 160 mM/1, and in the 3 adults 172 mM/1. Two adultssurviving another 18 days in 2/3 SW to the final sample had blood sodium levels of 173 mM/1.100D. DISCUSSIONAll reproductive hormones dropped below the detection limit in spent steelhead.This occurred within 1 week of spawning in Glover Creek. Estimation of spawning time ofthese fish is based on the observation that floods had pushed all gravid and spent fish out ofthe creek 10 days before electroshocking began. When the flood subsided, only gravid fishwere seen in the first week; in the following week spent fish were in the majority. A declinein GtH starting about 20 days after ovulation in spent female rainbow trout has beenreported (Scott et al., 1983). The GtH level remained elevated for a longer period in males,which produce milt for many weeks after the onset of spermiation (Sumpter and Scott 1989).A post-spawning decline in T and 17,20-P to undetectable levels in resident rainbow troutwas also more immediate in females than in males, but still required 30 days. The morerapid decline of all sex hormones in male and female steelhead after spawning may be animportant difference between resident and anadromous strains of this species. Rapidelimination of reproductive hormones from the body may facilitate the changes inosmoregulatory hormones required before reentry of SW by steelhead.In mature, hand-stripped male coho reproductive hormones slowly declined over twomonths in BW. Spermiation was also maintained during this period. It is not clear to whatdegree continued high testicular production of steroids is required to maintain plasma levelsin this range. Idler and Truscott (1963) suggested that persistence of gonadal steroids in theblood of spent sockeye may be due in part to impaired clearance. I suggest that thepersistence of GtH in the blood of coho provides sufficient stimulus to the gonads forcontinued steroid and milt production. Production eventually declines, perhaps as energyreserves are depleted after many weeks.101The continued responsiveness of the gonads to GtH may shed light on otherquestions regarding general (including osmoregulatory) systems failure as spawningcondition is maintained. Changes in the histology of kidneys, liver, interrenals, gills, andstomach of migrating and maturing pink salmon were described as incremental degenerationby McBride et al., (1986). Loss of function and responsiveness to hormonal signalspresumably follows a similar slow course in those organs, and eventually reaches the gonadsas well. No selective shutting down of systems is in evidence, in fact the potential forcomplete return of normal function of all organs seems to persist at least until the time ofspawning. In sockeye salmon gonadectomized a few weeks before final maturation generalhealth and proper function of all other organs returned to normal (McBride et al., 1963).Similarly in chinook, gonadectomy a few days before final maturation allowed extension ofnormal life for several months (McLean, 1993 in press).In contrast, coho jacks and adult males gonadectomized after the onset ofspermiation (in a study conducted as part of the current chapter, but which yielded littleadditional information) did not regain vigour. For example, as in the intact fish there was noresumption of digestive and peristaltic activity in the gastrointestinal tract. Cumulativechanges in many systems eventually do define a point in maturation after which recovery isnot possible. In the natural situation continued production of GtH in Pacific salmon maysuffice to explain the persistence of high levels of sex steroids. The sex steroids may initiate,directly or indirectly a process of degeneration in all tissues, and also interfere with theability to osmoregulate in SW.Concurrent with the decline in sex steroid levels, hypoosmoregulatory abilitydeveloped almost immediately in steelhead after spawning. Blood sodium did not vary assalinity was increased in the holding tank and was normal in 29 ppt for SW adapted salmon,102within 1 month after transfer from FW. Morisawa et al. (1979) recorded similar levelsduring the prespawning SW migrations of chum salmon. Salinity in the coho tank was notincreased to full SW in the light of evidence that the continuing high levels of sex steroidswould not permit survival in high salinity. In 20 ppt BW blood sodium in coho was higherthan that in steelhead in full SW. I suggest that the disappearance of reproductivehormones from the plasma of spent steelhead, prior to departure from the spawning area inthe creek, constitutes a preadaptation for a return to SW after spawning. This reduction ofreproductive hormones may be the event which initiates abandonment of the redd anddownstream movement of spent fish. A rapid decline in sex steroids immediately afterspawning would also remove any degenerative/catabolic activity which may be stimulated bythe levels circulating in spawning fish.Any correlation between clearance of reproductive hormones and return ofhypoosmoregulatory ability does not amount to proof of direct osmoregulatory action,however. Decline of sex steroids may create the permissive environment for the resumptionof production of the osmoregulatory hormones. Cortisol (Bjornson et al., 1987; Redding etal., 1984), growth hormone (Clarke et al., 1977; Richman and Zaugg, 1987), and prolactin(Bolton et al., 1987; Hirano et al., 1987), are known to directly control function of the organsof osmotic/ionic regulation through their effects on membrane permeability and Na+/K÷-ATPase activity. Concurrent measurement of reproductive hormones and prolactin duringthe SW maturation of farmed Atlantic salmon was performed by Anderson et al. (1991).Prolactin peaked and declined before androgen and E2 levels rose and no change in osmoticability occurred. Similar studies of the relationship between reproductive hormones, cortisol,and growth hormone have not been reported. There is also a need for studies of the effects ofexogenous sex steroids on hypoosmoregulation, cortisol, and growth hormone in non-103maturing fish in order to elucidate links which may exist between the controllingmechanisms of reproduction and osmoregulation.CHAPTER 6SUMMARY AND CONCLUSIONSA. NORMAL MATURATION OF CHINOOK AND STEELHEAD1. FemalesThe changes in reproductive parameters and hormones during the final weeks ofsexual maturation in wild coastal salmon; chinook, Oncorhynchus tshawytscha, andsteelhead, 0. mykiss, were documented for stocks spawning in two similar river systems onthe west coast of Vancouver Island, British Columbia, Canada. The two species are similarin that the bulk of their growth and oogenesis take place in the marine environment whilespawning occurs in freshwater after the growth phase. Both species normally make thetransition from seawater to freshwater before fmal maturation. The major difference in life-history is that chinook spawn in the fall and die soon after; steelhead although enteringfreshwater in fall, spawn in spring and often survive to return to the ocean.Oocytes of both species sequester vitellogenin from the blood until 2 or 3 weeksbefore fmal maturation, and continue to hydrate through to ovulation. There are nodifferences in GSI, HSI or oocyte water content between the species at ovulation.Patterns of hormone concentration are similar in chinook and steelhead, but thepeaks in concentration of the steroids at each stage are lower in steelhead. A decline in E2preceded a rise in maturational GtH, which is believed to initiate final maturation in salmon(Dickhoff and Swanson, 1989), by about two weeks. Presumably vitellogenesis continues insteelhead for several months after FW entry. Plasma T was relatively high in both species at105the beginning of the decline in E2, and declined towards ovulation. GtH was lower atovulation in steelhead than in chinook, not reaching the levels seen in ovulating chinookuntil several days after ovulation.The somewhat lower level of the maturation inducing hormone, 17,20-P, in steelheadmay be due to the lower level of GtH stimulation leading up to ovulation in that species. Inchinook females an abrupt rise in plasma 17,20-P, from basal levels to the highest levels,occurred at ovulation.Blood sodium was elevated in chinook captured in high salinity water near the outletof Nitinat Lake. At all other times and locations during unimpeded migration blood sodiumwas in the range of normal values for unstressed salmon (Lam et al., 1982; Sower andSchreck, 1982; Blackburn and Clarke, 1987).2. MalesGonadotropin was detectable in the first samples of Nitinat male chinook before FWentry and rose as maturation commenced, peaking in spent fish. In steelhead males GtHwas not detectable at any time during final maturation. The difference in GtH productionmay explain the considerably lower levels of maturational steroids produced in steelhead ascompared to chinook.Testosterone and 11-KT rose with approaching maturation in chinook. Between themid-river sample of sub-mature fish and the sample of mature fish at Nitinat hatchery Tdeclined and 11-KT rose, such that 11-KT was higher than T at spermiation. Both steroidsdeclined after spermiation in chinook. 11-KT concentration was higher than that of T insteelhead at spermiation as well. These patterns, seen in other salmonid males during106maturation (Dye et al., 1986; Ueda et al., 1984a), indicate that 11-KT is more involved withthe later stages of spermatogenesis than is T.A surge of 17,20-P accompanies spermiation in chinook, while in steelhead 17,20-Plevel was maintained at quite low levels. It is not clear whether the low levels of 17,20-P inspermiating steelhead males is normal for the species or a consequence of suppressionresulting from the holding conditions for these fish. In chinook 17,20-P continued to riseafter the onset of initial spermiation, and large volumes of milt were obtained from spentfish; evidence which supports the concept of 17,20-P as the maturational hormone in male aswell as female salmon (Ueda et al., 1985; Schulz and Blum, 1990), but is also consistent withthe view of Marshall et al. (1989) that 17,20-P is involved with the production of seminalfluid.B. MATURATION OF CHINOOK AND CHUM IN BRACKISH WATERIn 1988, mortality was higher in eggs taken earlier in the season from both BW andFW chinook than in eggs taken later in the season at the same locations. The first femalesmaturing in FW had recently left BW after extended residence in an environment of highsalinity and temperature. The first females maturing at the BW net pen site were exposed tothe same conditions for a longer period. Fertility was highly variable in the first femalesfrom either location, and the high average mortalities were due to exceptionally highmortality in a few batches of eggs.Subsequent samples taken from the net pens and hatchery after the beginning ofautumn rains had lower mortality rates, but the incidence of high mortality remained107greater in BW eggs. Mortality rates after fertilizations with sperm from males maturing ateither location suggest that maturation site does not affect fertility of sperm. A few batchesof eggs having exceptionally high mortality again brought the average up for eggs fromeither location. The occurrence of batches of eggs with near 100% mortality was reported forcoho by Sower and Schreck (1982) and for chum by Lam et a/.(1982) who linked thephenomenon with high blood sodium in females before or during ovulation.Fall weather had brought about a marked lowering of temperature and salinity andan elevation of dissolved oxygen in the lake by the time eggs were taken from the third groupof chinook in 1988. Mortality was slightly higher in these eggs than in eggs taken well intothe rainy season at the hatchery in 1989, and considerably lower than in the first 2 groups of1988. One batch of eggs in the last BW group suffered 100% mortality and the majority ofthe others had mortalities comparable to those sampled in the hatchery over the sameperiod. Blood sodium in the BW females was significantly higher than in hatchery females.In particular sodium was highest in the female contributing 100% mortality. However, aclear correlation between blood sodium and mortality of eggs, as found by Sower and Schreck(1982), could not be demonstrated in this study. The occurrence of high mortality in the firstgroup of FW eggs, from females exposed to high salinity in the days immediately precedingovulation, suggests this is a sensitive period of fmal maturation.An important finding of this study was the fact that mortality in the majority of thelast batches of chinook eggs from the net pens was no higher than in chum eggs taken at thenet pens and hatchery one week later in the 1988 season. Toward the end of the chinookmigration period in Nitinat Lake improvement of environmental conditions, stemming fromlower salinity and temperature, and higher oxygen, are more favourable to net penmaturation. The perceived difference between chum and chinook does not reflect a basic108physiological difference leading to differences in ability to withstand holding in BW. Thepast differences in egg mortality rate between BW chum and chinook instead reflect thedifference in time of arrival at the lakehead. Chum arrive later when conditions are alreadymore favourable, and do not normally experience osmoregulatory difficulties. In years whensubstantial rains precede or coincide with the arrival of chinook at the lakehead estuarinecapture and holding of chinook should be considered as a viable option.C. SEAWATER TRANSFER OF SALMON AND STEELHEADThe hypothesis that transfer of steelhead into SW may bring about a rapid decline inreproductive hormones that facilitates SW readaptation could not be tested in this study.Reproductive hormones declined to basal levels immediately after spawning, before fish hadabandoned the spawning area in the creek. The rapid post-spawning decline in reproductivehormones sets steelhead apart from coho (and other Pacific salmon) which continue toproduce and secrete GtH, 17,20-P and androgens for some weeks after spermiation.Hand-stripping of milt does not have the effect of lowering sex hormones or GtH, orimproving hypoosmoregulatory ability in mature coho. Steelhead resumed feeding, lost theirspawning colours, and adapted to SW soon after disappearance of reproductive hormonesfrom the blood and transfer from FW. Coho are not capable of any of these physiologicalchanges, and appear to be "locked" into a program of high steroid production. That thegonads of mature coho continued to respond to GtH stimulation is reason to believe otherorgans, including those involved in hypoosmoregulation, were operative and capable ofresponding to hormonal stimulation. Coho were unable to regulate blood sodium however,109and the presence of high levels of sex steroids in the blood at this time suggests interferenceby the sex steroids in the control mechanisms of osmoregulation.Other methods of examining the hypothesis of steroid interference in (hypo)osmoregulatory ability may be available in the use of gonadectomized or otherwise sexuallydysfunctional fish, where steroid metabolism by the gonads can be eliminated.110REFERENCESAnderson, 0., Skibeli, V., Haug, E. and K. M. Gautvik, 1991. Serum prolactin and sexsteroids in Atlantic salmon, Salmo salar, during sexual maturation. Aquaculture,96:169-18.Ando, S., Yamazaki, F., Hatano, M. and K. Zama, 1986. Deterioration of chum salmon(Oncorhynchus keta) muscle during spawning migration-M. changes in proteincomposition and protease activity of juvenile chum salmon muscle upon treatmentwith sex steroids. Comp. Biochem. Physiol. 83B:325-330.Baynes, S. M. and A. P. Scott, 1985. Seasonal variations in parameters of milt productionand in plasma concentration of sex steroids of male rainbow trout, Salmo gairdneri.Gen. Comp. Endocrinol. 57: 150-160.Bjornsson, B. Th., Yamauchi, K., Nishioka, R. S., Deftos, L. J. and H. A. Bern, 1987. Effectsof hypophysectomy and subsequent hormonal replacement therapy on hormonal andosmoregulatory status of coho salmon, Oncorhynchus kisutch. Gen. Comp.Endocrinol., 68:421-430.Blackburn, J. and W. C. Clarke, 1987. Revised procedure for the 24 hour seawater challengetest to measure seawater adaptability of juvenile salmonids. Can. Tech. Rep. Fish.Aqua. Sci. 1515.Bolton, J. P., Collie, N. L., Kawauchi, H. and T. Hirano, 1987. Osmoregulatory actions ofgrowth hormone in rainbow trout, Salmo gairdneri. J. Endocrinol., 112:63-68.Bolton, J. P., Takahashi, A., Kawauchi, H., Kubota, J. and T. Hirano, 1986. Development andvalidation of a salmon growth hormone radioimmunoassay. Gen. Comp. Endocrinol.62:230-238.Clarke, W. C., Griffioen, W. and A. Solmie. 1977. A factorial mating study comparing thefertility of pen-reared and wild coho salmon, Oncorhynchus kisutch. Fish. Mar. Serv.Can. Tech. Rep. 748 9pp.Crim, L. W. and D. R. Idler, 1978. Plasma gonadotropin, etradiol, and vitellogenin andgonad phosvitin levels in relation to the seasonal reproductive cycle of female browntrout. Ann. Biol. Anim. Biochem. Biophys., 18:1001-1005.Dickhoff, W. W. and P. Swanson, 1989. Functions of salmon pituitary glycoproteinhormones: "The maturational surge hypothesis." In Progress in ComparativeEndocrinology. Proceedings of the Eleventh International Symmposium onComparative Endocrinology. A. Epple, C. G. Scanes and M. H. Stetson (eds.), Wiley-Liss, Toronto.111Donaldson, E. M. and U. H. M. Fagerlund, 1969. Cortisol secretion rate in gonadectomizedfemale sockeye salmon, Oncorhynchus nerka: effects of estrogen and cortisoltreaatment. J. Fish. Res. Bd. Canada 26:1789-1799.Dye, H. M., Sumpter, J. P., Fagerlund, U. H. M. and E. M. Donaldson, 1986. Changes inreproductive parameters during the spawning migration of pink salmon,Oncorhynchus gorbuscha (Walbaum) J. Fish Biol. 29:167-176.Eddy, F. B. (1981). Effects of stress on osmotic and ionic regulation in fish. In "Stress andFish" (A. D. Pickering, Ed.), pp. 77-102. Acedemic Press, New York.Fitzpatrick, M. S., Van Der Kraak, G. and K. Schreck, 1986. Profiles of plasma sex steroidsand gonadotropin in coho salmon, Oncorhynchus kisutch, during final maturation.Gen. Comp. Endocrinol. 62:437-451.Fostier, A., Jalabert, B., Billard, R., Breton, B. and Y. Zohar, 1983. The gonadal steroids. InW. S. Hoar, D. J. Randall and E. M. Donaldson (eds.). Fish physiology Vol. 9A.Academic Press, New York. pp. 277-372.Foucher, J. L. and F. Le Gac, 1989. Evidence for an androgen binding protein in the testis ofa teleost fish, Salmo gairdneri R.: A potential marker of Sertoli cell function. J.Steroid Biochem., 32:545-552.Hart, J. L., 1980. Pacific Fishes of Canada. Fish. Res. Bd. Can. Bulletin 180. John DeyellCompany, Ottawa.Hasegawa, S., Hirano, T. and H. Kawauchi, 1986. Sodium retaining activity of chum salmonprolactin in some euryhaline teleosts. Gen. Comp. Endocrinol., 63:309-317.Healey, M. C., 1991. Life history of chinook salmon. In C. Groot and L. Margolis (eds.).Pacific Salmon Life Histories. UBC Press, Vancouver. pp.313-393.Hirano, T., Ogasawara, T., Hasegawa, S., Iwata, M. and Y. Nagahama, 1990. Changes inplasma hormone levels during loss of hypoosmoregulatory capacity in mature chumsalmon, Oncorhynchus )seta, kept in seawater. Gen. Comp. Endocrinol. 78(2):254-262.Hoar, W. S., 1976. Smolt transformation: evolution, behaviour, and physiology. J. Fish. Res.Board Can., 33:1233-1252.Idler and Ng 1983. Teleost gonadotropins: isolation, biochemistry, and function. In W. S.Hoar, D. J. Randall and E. M. Donaldson (eds.). Fish Physiology Vol.9A. AcademicPress, New York. pp. 187-212.Idler, D. R. and B. Truscott, 1963. In vivo metabolism of steroid hormones by sockeyesalmon. (A) Impaired hormone clearance in mature and spawned Pacific salmon (0.nerka). (B) Precursors of 11-ketotestosterone. Can. J. Biochem. Physiol. 41:875-887.112Idler, D. R., Fagerlund, U. H. M., and A. P. Ronald, 1960. Isolation of pregnen-4-ene-17a,20b-diol-3-one from the plasma of Pacific salmon, Oncorhynchus nerka.Biochem. Biophys. Res. Commun., 2:133-137.Idler, D. R., Horne, D. A. and G. B. Sangalang, 1971. Identification and quantification of themajor androgens on testicular and peripheral plasma of Atlantic salmon, Salmosalar, during sexual maturation. Gen. Comp. Endocrinol., 16:257-267.Itoh, H., Suzuki, K. and H. Kawauchi, 1988. The complete amino acid sequence of 0-subunitsof two distinct chum salmon GTHs. Gen. Comp. Endocrinol., 71:438-451.Jalabert, B. and B. Breton, 1980. Evolution de la gonadotropin plasmatique t-GtH apresl'ovulation chez la Truite arc-en-ciel, Salmo gairdneri R., et influence de la retntiondes ovules. C. R. Acad. Sci. Paris, 290:799-801.Jalabert, B., 1976. In vitro oocyte maturation and ovulation in rainbow trout, Salmogairdneri, northern pike, Esox lucius, and goldfish, Carassius auratus. J. Fish. Res.Board Can. 33:974-988.Jalabert, B., Goetz, F. W., Breton, B., Fostier, A. and E. M. Donaldson, 1978. Precociousinduction of oocyte maturation and ovulation in coho salmon, Oncorhynchus kisutch.J. Fish. Res. Board Can., 35:1425-1420.Kagawa, H., Yong, G., and Nagahama, Y. 1983. Relationship between seasonal plasmaestradio1-1713 and testosterone levels and in vitro production by ovarian follicles ofamago salmon, Oncorhynchus rhodurus. Biol. Reprod. 29:301-309.Kanamori, A., Adachi, S. and Y. Nagahama, 1988. Developmental changes in steroidogenicresposes of ovarian follicles of amago salmon, Oncorhychus rhodurus, to chumsalmon gonadotropin during oogenesis. Gen. Comp. Endocrinol., 72:13-24.Kawauchi, H., Suzuki, K., Itoh, H., Swanson, P., Naito, N., Nagahama, Y., Nozaki, M.,Nakai, Y. and S. Itoh, 1989. The duality of teleost gonadotropins. Fish Phys.Biochem. 7:29-38.Kirschner, L. B., 1991. Water and ions. In C. L. Prosser (ed.) Environmental and MetabolicAnimal Physiology. Wiley-Liss, Toronto.Lam, C. N. H., Jensen, J. 0. T. and D. F. Alderdice, 1982. Preliminary study of low gameteviability in adult chum salmon, Oncorhynchus keta, held in sea pens at DesertedCreek, Hisnit Inlet„ B.C. Can. Tech. Rep. Fish. Aq. Sci. NO.1133.Lam, T. J. 1983. Environmental influences on gonadal activity in fish. In. W. S. Hoar, D. J.Randall and E. M. Donaldson (eds.). Fish Physiology Vol. 9B. Academic Press, NewYork. pp. 65-116.113Liley, N. R. and Y. Rouger, 1990. Plasma levels of gonadotropin and 17a,2013-dihydroxy-4-pregnen-3-one in relation to spawning behaviour of rainbow trout, Oncorhynchusnzykiss, Waulbam. J. Fish Biol., 37:699-711.Liley, N. R., Breton, B., Fostier, A. and E. S. P. Tan, 1986a. Endocrine changes associatedwith spawning behaviour and social stimuli in a wild population of rainbow trout,Salmo gairdneri I. Males. Gen. Comp. Endocrinol. 62:145-156.Liley, N. R., Fostier, A., Breton, B. and E. S. P. Tan, 1986b. Endocrine changes associatedwith spawning behaviour and social stimuli in a wild population of rainbow trout,Salmo gairdneri II. Females. Gen. Comp. Endocrinol. 62:157-167.Lundqvist, H., Borg, B. and I. Berglund, 1989. Androgens impair seawater adaptibility insmolting Baltic salmon, Salmo salar. Can. J. Zool., 67:1733-1736.Maitre, J. L., Mercier, L., Dolo, L. and Y. Valotaire, 1985. Characterization of specificestradiol receptors, induction of vitellogenin and vitellogenin mRNA in the liver ofrainbow trout, Salmo gairdneri. Biochimie, 67:215-225.Marshall, W. S., Bryson, S. E. and D. R. Idler, 1989. Control of ion transport by the spermduct epithelium of brook trout, Salvelinus fontinalis. Fish Phys. Biochem. 7:331-336.Mayer, I., Lundqvist, H., Berglund, I., Schmitz, M., Schulz, R. and B. Borg, 1990. Seasonalendocrine changes in Baltic salmon, Salmo salar, immature parr and mature maleparr. I. Plasma levels of five androgens, 17a,20[3-dihydroxyprogesterone, and 170-estradiol. Can. J. Zool. 68:1360-1365.McBride, J. R. and A. P. van Overbeeke, 1969. Hypertrophy of the interrenal tissue insexually maturing sockeye salmon, Oncorhynchus nerka, and the effects ofgonadectomy. J. Fish. Res. Bd. Canada 26:2975-2985.McBride, J. R., Fagerlund, U. H. M., Dye, H. M. and J. Bagshaw, 1986. Changes in structureof tissues and in plasma cortisol during the spawning migration of pink salmon,Oncorhynchus gorbuscha (Walbaum). J. Fish. Biol. 29:153-165.McBride, J. R., Fagerlund, U. H. M., Smith, M. and N. Tomlinson, 1963. Resumption offeeding by and survival of adult sockeye salmon, Oncorhynchus nerka, followingadvanced gonad development. J. Fish. Res. Bd. Canada 20:95-100.McLean, E., 1993. Microsurgery: gonadectomy. In P. Hochachka and T. Mommsen (eds.)Biochemistry and Molecular Biology of Fishes. vol. IV. Elsevier, Amsterdam (inpress).Miura, T., Yamauchi, K., Takahashi, H. and Y. Nagahama, 1991. Hormonal induction of allstages of spermatogenesis in vitro in the male Japanese eel, Anguilla japonica. Proc.Nat. Acad. Sci. USA, 88:5774-5778.114Miwa, S. and Y. Inui, 1986. Inhibitory effects of 17 methyltestosterone and estrdiol-17 onsmoltification of sterilized amago salmon, Oncorhynchus rhodurus. Aquaculture53:21-39.Morisawa, M., Hirano, T. and K Suzuki, 1979. Changes in blood and seminal plasmacomposition of the mature salmon, Oncorhynchus keta, during adaptation tofreshwater. Comp. Biochem. Physiol. 64A:325-329.Nagahama, Y., 1990. Endocrine control of oocyte maturation in teleosts. In Progress inComparative Endocrinology. Proceedings of the Eleventh InternationalSymmposium on Comparative Endocrinology. A. Epple, C. G. Scanes and M. H.Stetson (eds.), pp. 385-392. Wiley-Liss, Toronto.Nagahama, Y., Hirose, K., Young, G., Adachi, S., Suzuki, K. and B. Tamaoki, 1983. Relativein vitro effectiveness of 17a,2013-dihydroxy-4-pregnen-3-one and other pregnenederivatives on germinal vesicle breakdown in oocytes of ayu, Plecoglossus altivelis,amago salmon, Oncorhychus rhodurus, rainbow trout, Salmo gairdneri, and goldfish,Carassius auratus. Gen. Comp. Endocrinol., 51:15-23.Peter, R. E., Nahorniak, C. S., Chang, J. P. and L. W. Crim, 1984. Gonadotropin releasefrom the pars distalis of goldfish, Carassius auratus, transplanted beside the brain ofinto the brain ventricles: Additional evidence for gonadotropin release-inhibitoryfactor. Gen. Comp. Endocrinol., 55:337-346.Peute, J., Schulz, R., Glazenburg, K., Lambert, J. G. D. and V. Blum, 1989. Pituitary steroidsin two teleost species: Immunohistological and biochemical studies. Gen. Comp.Endocrinol., 76:63-71.Pottinger, T. G. and A. D. Pickering, 1990. The effect of cortisol administration on hepaticand plasma estradiol-binding capacity in immature famale rainbow trout,Oncorhynchus mykiss. Gen. Comp. Endocrinol., 60:264-277.Pottinger, T. G., 1986. Estrogen binding sites in the liver of sexually mature male andfemale brown trout, Salmo trutta L. Gen. Comp. Endocrinol., 61:120-126.Potts, W. T. W. and P. P. Rudy, Jr., 1969. Water balance in the eggs of the Atlantic salmon,Salmo solar. J. Exp. Biol. 50:223-237.Redding, J. M., Schreck, C. B., Birks, E. K. and R.D. Ewing, 1984. Cortisol and its effects onplasma thyroid hormone and electrolyte concentrations in fresh water and duringseawater acclimation in yearling coho salmon, Oncorhynchus kisutch. Gen. Comp.Endocrinol. 56:146-155.Robertson, 0. H. and B. C. Wexler, 1959. Hyperplasia of the adrenal cortical tissue in Pacificsalmon (Genus Oncorhynch us) and rainbow trout, Salmo gairdneri, accompanyingsexual maturation and spawning. Endocrinology, 65:225-238.115Rudy, P. P. Jr. and W. T. W. Potts, 1969. Sodium balance in the eggs of the Atlantic salmon,Salmo salar. J. Exp. Biol. 50:239-246.Sangalang, G. B. and H. C. Freeman, 1988. In vitro biosynthesis of 17a,2013-dihyclroxy-4-pregnen-3-one by the ovaries, testes and head kidneys of the Atlantic salmon, Salmosalar. Gen. Comp. Endocrinol., 69:406-415.Schmidt, P. J. and D. R. Idler, 1962. Steroid hormones in the plasma of salmon at variousstages of maturation. Gen. Comp. Endocrinol., 2:204-214.Schulz, R. and V. Blum, 1988. Testosterone immunoreactivity in rainbow trout, Salmogairdneri Richardson, testis. Gen. Comp. Endocrinol., 73:80-89.Schulz, R. and V. Blum, 1990. Steroid secretion of rainbow trout testis in vitro: variationduring the reproductive cycle. Gen. Comp. Endocrinol., 80:189-198.Scott, A. P. and J. P. Sumpter, 1989. Seasonal variation in testicular germ cell stages and inplasma concentrations of sex steroids in male trout, Salmo gairdneri, maturing at 2years old. Gen. Comp. Endocrinol. 73:46-58.Scott, A. P., Sheldrick, E. L., and A. P. F. Flint, 1982. Measurement of 17a,2013-dihydroxy-4-pregnen-3-one in plasma of trout, Salmo gairdneri Richardson: Seasonal changes andresponse to salmon pituitary extract. Gen. Comp. Endocrinol., 46:444-451.Scott, A. P., Sumpter, J. P. and P. A. Hardiman, 1983. Hormone changes during ovulation inthe rainbow trout, Salmo gairdneri Richardson. Gen. Comp. Endocrinol. 49:128-134.Sower, S. A. and C. B. Schreck, 1982. Steroid and thyroid hormones during sexualmaturation of coho salmon, Oncorhynchus kisutch, in seawater or fresh water. Gen.Comp. Endocrinol. 47:42-53.Springate, J. R. C., Bromage, N. R., Elliot, J. A. and D. L. Hudson, 1984. The timing ofovulation and stripping and their effects on the rates of fertilization and survival toeying, hatch, and swim-up in the rainbow trout, Salmo gairdneri R.. Aquaculture,43:313-322.Stoss, J. and U. H. M. Fagerlund, 1982. Influence of saltwater spawning and stress onquality of sex products of chum salmon, Oncorhynchus keta. In C. J. J. Richter and H.J. Th. Goos (eds.). Proceedings of the first international symposium on reprductivephysiology of fish. Pudoc, Wegeningen, Netherlands.Sumpter, J. P. and A. P. Scott, 1989. Seasonal variations in plasma and pituitary levels ofgonadotropin in males and females of two strains of rainbow trout, Salmo gairdneri.Gen. Comp. Endocrinol., 75:376-388.Sumpter, J. P., Scott, A. P., Baynes, S. M., and P. R. Witthames. 1984. Early stages of thereproductive cycle in virgin female rainbow trout (Salmo gairdneri Richardson).Aquaculture 43:235-242.116Suzuki, K., Kawauchi, H. and Y. Nagahama, 1988a. Isolation and characterization of twodistinct gonadotropins from chum salmon pituitary glands. Gen. Comp. Endocrinol.71:292-301.Suzuki, K., Kawauchi, H. and Y. Nagahama, 1988b. Isolation and characterization ofsubunits of two distinct gonado-tropins from chum salmon pituitary glands. Gen.Comp. Endocrinol. 71:302-306.Suzuki, K., Nagahama, Y. and H. Kawauchi, 1988c. Steriodo-genic activities of two distinctsalmon gonadotropins. Gen. Comp. Endocrinol. 71:452-458.Swanson, P., Bernard, M., Nozaki, M., Suzuki, K., Kawauchi, H., and W. W. Dickhoff, 1989.Gonadotropins I and II in juvenile coho salmon. Fish Phys. Biochem. 7:169-176.Swanson, P., Suzuki, K., Kawauchi, H. and W. W. Dickhoff, 1991. Isolation andcharacterization of two coho salmon gonadotropins, GtH I and GtH II. Biol. Reprod.44:29-38.Truscott, B., Idler, D. R., So, Y. P. and J. M. Walsh, 1986. Maturational steroids andgonadotropin in upstream migratory sockeye salmon. Gen. Comp. Endocrinol. 62:99-110.Tyler, C. R., Sumpter, J. P., Kawauchi, H. and P. Swanson, 1991. Involvement ofgonadotropin in the uptake of vitellogenin into vitellogenic oocytes of the rainbowtrout, Oncorhynchus mykiss. Gen. Comp. Endocrinol. 84:291-299.Ueda, H., Hiroi, 0., Hara, A., Yamauchi, K. and Y. Nagahama, 1984a. Changes in serumconcentration of steroid hormones, thyroxine, and vitellogenin during spawningmigration of the chum salmon, Oncorhynchus keta. Gen. Comp. Endocrinol. 53:203-211.Ueda, H., Kagawa, H. and S. Fujimoto, 1985. Immunoelectron microscopic localization ofgrowth hormone in the pituitary glands of two teleosts, tilapia, Sarotherodonmossambicus, and amago salmon, Oncorhynchus rhodurus. Gen. Comp. Endocrinol.59:149-154.Ueda, H., Kambegawa, A. and Nagahama, Y. 1984b. In vitro 11-ketotestosterone and 17a,20(3-dihydroxy-4-pregnen-3-one production by testicular fragments and isolatedsperm of rainbow trout, Salmo gairdneri. J. Exp. Zool. 231:436-439.Van Der Kraak, G. and E. M. Donaldson, 1986. Steroidogenic capacity of coho salmonovarian follicles throughout the periovulatory period. Fish Physiology andBiochemistry 1:179-186.Van Der Kraak, G., Donaldson, E. M., Dye, H. M., Hunter, G. A., Rivier, J. E. and W. W.Vale, 1987. Effects of mammalian and salmon gonadotropin-releasing hormones andanalogues on plasma gonadotropin levels and ovulation in coho salmon,Oncorhynchus kisutch. Can. J. of Fish. Aquat. Sci. 44:1930-1935.117Van Der Kraak, G., Dye, H. M. and E. M. Donaldson, 1984. Effects of LH-RH and De-Gly10[D-A1a6] LH-RH-Ethylamide on plasma sex steroid profiles in adult femalecoho salmon. Gen. Comp. Endocrinol. 55:36-45.Wallace, R. A. and K. Selman, 1981. Cellular and dynamic aspects of oocyte growth inteleosts. Am. Zool. 21:325-343.Wallace, R. A., Selman, K., Greeley, M. S. J., Begovac, P. C. and Y. W. P. Lin, 1987. Currentstatus of oocyte growth. In D. R. Idler, L. W. Crim and J. M. Walsh (eds.).Proceedings of the third international symposium on reproductive physiology of fish.Memorial University, St. John's, Newfoundland, Canada. pp.167-177.Wertheimer, A. C. and R. M. Martin, 1981. Viability of gametes from adult anadromous cohoripened in an estuarine net pen. Prog. Fish Cult., 43:40-42.Wertheimer, A. C., 1984. Maturation success of pink salmon (Oncorhynchus gorbuscha) andcoho salmon (0. kisutch) held under three salinity regimes. Aquaculture, 43:195-212.118


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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


Related Items