Sexual Differences in Homing Profiles and Shortening of Homing Duration by Gonadotropin-Releasing Hormone Analog Implantation in Lacustrine Sockeye Salmon (Oncorhynchus nerka) in Lake Shikotsu

Relationships between maturational status and migration behavior of homing chum salmon Oncorhynchus keta in inner bays of the Sanriku coast

The correlations among gonad maturity and various homing behaviors of chum salmon, Oncorhynchus keta, were evaluated using acoustic tracking of tagged fish in Otsuchi Bay, Japan. There was a negative correlation between the time duration from release of tagged fish until river entry and the plasma 17α, 20β-dihydroxy-4-pregnen-3-one (DHP) levels, an indicator of final maturation. Females with high DHP entered the rivers soon after the release, whereas females with low DHP (<10 ng/ml) took a few days to more than one week until river entry. Similar correlation was also found in males. A pattern of river entry correlated with maturational conditions was also observed in fish entering the rivers of neighboring bays. DHP concentrations of fish caught in the rivers were consistently higher. On the other hand, more than half of released salmon departed from the bay regardless of their plasma DHP level, suggesting that maturational status does not force homing adults to enter the most available nearest rivers. Fish entering the rivers experienced ambient temperatures less than 8 °C, which is approximately 5 °C lower than that of the bay. These results indicate that homing salmon hold their position in the bay until just before spawning, which may be attributable to low temperature avoidance. This characteristic type of river entry may be suitable to geographical features and thermal regimes of this region.

Life-history strategies in salmonids: the role of physiology and its consequences

Salmonids are some of the most widely studied species of fish worldwide. They span freshwater rivers and lakes to fjords and oceans; they include short- and long-distance anadromous migrants, as well as partially migratory and non-migratory populations; and exhibit both semelparous and iteroparous reproduction. Salmonid life-history strategies represent some of the most diverse on the planet. For this reason, salmonids provide an especially interesting model to study the drivers of these different life-history pathways. Over the past few decades, numerous studies and reviews have been published, although most have focused on ultimate considerations where expected reproductive success of different developmental or life-history strategies are compared. Those that considered proximate causes generally focused on genetics or the environment, with less consideration of physiology. Our objective was therefore to review the existing literature on the role of physiology as a proximate driver for life-history strategies in salmonids. This link is necessary to explore since physiology is at the core of biological processes influencing energy acquisition and allocation. Energy acquisition and allocation processes, in turn, can affect life histories. We find that life-history strategies are driven by a range of physiological processes, ranging from metabolism and nutritional status to endocrinology. Our review revealed that the role of these physiological processes can vary across species and individuals depending on the life-history decision(s) to be made. In addition, while findings sometimes vary by species, results appear to be consistent in species with similar life cycles. We conclude that despite much work having been conducted on the topic, the study of physiology and its role in determining life-history strategies in salmonids remains somewhat unexplored, particularly for char and trout (excluding brown trout) species. Understanding these mechanistic links is necessary if we are to understand adequately how changing environments will impact salmonid populations.

Expression dynamics of the genes for the hypothalamo-pituitary-gonadal axis in tiger puffer (Takifugu rubripes) at different reproductive stages

Tiger puffer, Takifugu rubripes, a commercially important long-distance migratory fish, return to specific spawning grounds for reproduction. To clarify reproductive neuroendocrine system of the tiger puffer, the changes in the expression levels of the genes encoding three gonadotropin-releasing hormones (GnRHs), gonadotropin-inhibitory hormone (GnIH), GnIH receptor (GnIH-R), kisspeptin and kisspeptin receptor in the brain and gonadotropin (GTH) subunits, growth hormone (GH) and prolactin (PRL) in the pituitary were examined in the tiger puffer captured in the wild at different reproductive stages, namely immature and mature fish of both sexes, and post-ovulatory females that were obtained by hormonal treatment. The amounts of three gnrh mRNAs, gnih, gnih-r, fshb and lhb were substantially increased in the mature fish compared to the immature fish, especially in the females, and these augmented expressions were drastically decreased in the post-ovulatory females. gh expression showed a slight increase in the mature males. In contrast, kiss2, kiss2r and prl did not show significant changes in the males but significantly decreased in the post-ovulatory females. The present results demonstrate the expression dynamics of the hypothalamo-pituitary-gonadal axis genes associated with the reproductive conditions and the possible involvement of the GnRH/GnIH/GTH system in the regulation of the sexual maturation and spawning in the wild tiger puffer.

Reduced lactate dehydrogenase activity in the heart and suppressed sex hormone levels are associated with female-biased mortality during thermal stress in Pacific salmon

Female-biased mortality has been repeatedly reported in Pacific salmon during their upriver migration in both field studies and laboratory holding experiments, especially in the presence of multiple environmental stressors, including thermal stress. Here, we used coho salmon (Oncorhynchus kisutch) to test whether females exposed to elevated water temperatures (18°C) (i) suppress circulating sex hormones (testosterone, 11-ketotestosterone and estradiol), owing to elevated cortisol levels, (ii) have higher activities of enzymes supporting anaerobic metabolism (e.g. lactate dehydrogenase, LDH), (iii) have lower activities of enzymes driving oxidative metabolism (e.g. citrate synthase, CS) in skeletal and cardiac muscle, and (iv) have more oxidative stress damage and reduced capacity for antioxidant defense [lower catalase (CAT) activity]. We found no evidence that a higher susceptibility to oxidative stress contributes to female-biased mortality at warm temperatures. We did, however, find that females had significantly lower cardiac LDH and that 18°C significantly reduced plasma levels of testosterone and estradiol, especially in females. We also found that relative gonad size was significantly lower in the 18°C treatment regardless of sex, whereas relative liver size was significantly lower in females held at 18°C. Further, relative spleen size was significantly elevated in the 18°C treatments across both sexes, with larger warm-induced increases in females. Our results suggest that males may better tolerate bouts of cardiac hypoxia at high temperature, and that thermal stress may also disrupt testosterone- and estradiol-mediated protein catabolism, and the immune response (larger spleens), in migratory female salmon.

Sensory mechanisms of natal stream imprinting and homing in Oncorhynchus spp

Juvenile Oncorhynchus spp. can memorise their natal stream during downstream migration; juveniles migrate to feed during their growth phase and then they migrate long distances from their feeding habitat to their natal stream to reproduce as adults. Two different sensory mechanisms, olfaction and navigation, are involved in the imprinting and homing processes during short-distance migration within the natal stream and long-distance migration in open water, respectively. Here, olfactory functions are reviewed from both neurophysiological studies on the olfactory discrimination ability of natal stream odours and neuroendocrinological studies on the hormonal controlling mechanisms of olfactory memory formation and retrieval in the brain. These studies revealed that the long-term stability of dissolved free amino-acid composition in the natal stream is crucial for olfactory imprinting and homing. Additionally, the brain-pituitary-thyroid and brain-pituitary-gonadal hormones play important roles in olfactory memory formation and retrieval, respectively. Navigation functions were reviewed from physiological biotelemetry techniques with sensory interference experiments during the homing migration of anadromous and lacustrine Oncorhynchus spp. The experiments demonstrated that Oncorhynchus spp. used compass navigation mechanisms in the open water. These findings are discussed in relation to the sensory mechanisms involved in natal stream imprinting and homing in Oncorhynchus spp.

Involvement of hormones in olfactory imprinting and homing in chum salmon

The olfactory hypothesis for salmon imprinting and homing to their natal stream is well known, but the endocrine hormonal control mechanisms of olfactory memory formation in juveniles and retrieval in adults remain unclear. In brains of hatchery-reared underyearling juvenile chum salmon (Oncorhynchus keta), thyrotropin-releasing hormone gene expression increased immediately after release from a hatchery into the natal stream, and the expression of the essential NR1 subunit of the N-methyl-D-aspartate receptor increased during downstream migration. Gene expression of salmon gonadotropin-releasing hormone (sGnRH) and NR1 increased in the adult chum salmon brain during homing from the Bering Sea to the natal hatchery. Thyroid hormone treatment in juveniles enhanced NR1 gene activation, and GnRHa treatment in adults improved stream odour discrimination. Olfactory memory formation during juvenile downstream migration and retrieval during adult homing migration of chum salmon might be controlled by endocrine hormones and could be clarified using NR1 as a molecular marker.

The Olfactory Transcriptome and Progression of Sexual Maturation in Homing Chum Salmon Oncorhynchus keta

Reproductive homing migration of salmonids requires accurate interaction between the reception of external olfactory cues for navigation to the spawning grounds and the regulation of sexual maturation processes. This study aimed at providing insights into the hypothesized functional link between olfactory sensing of the spawning ground and final sexual maturation. We have therefore assessed the presence and expression levels of olfactory genes by RNA sequencing (RNAseq) of the olfactory rosettes in homing chum salmon Oncorhynchus keta Walbaum from the coastal sea to 75 km upstream the rivers at the pre-spawning ground. The progression of sexual maturation along the brain-pituitary-gonadal axis was assessed through determination of plasma steroid levels by time-resolved fluoroimmunoassays (TR-FIA), pituitary gonadotropin subunit expression and salmon gonadotropin-releasing hormone (sgnrh) expression in the brain by quantitative real-time PCR. RNAseq revealed the expression of 75 known and 27 unknown salmonid olfactory genes of which 13 genes were differentially expressed between fish from the pre-spawning area and from the coastal area, suggesting an important role of these genes in homing. A clear progression towards final maturation was characterised by higher plasma 17α,20β-dihydroxy-4-pregnen-3-one (DHP) levels, increased pituitary luteinizing hormone β subunit (lhβ) expression and sgnrh expression in the post brain, and lower plasma testosterone (T) and 17β-estradiol (E2) levels. Olfactomedins and ependymin are candidates among the differentially expressed genes that may connect olfactory reception to the expression of sgnrh to regulate final maturation.

Physiological mechanisms of imprinting and homing migration in Pacific salmon Oncorhynchus spp

After several years of feeding at sea, salmonids have an amazing ability to migrate long distances from the open ocean to their natal stream to spawn. Three different research approaches from behavioural to molecular biological studies have been used to elucidate the physiological mechanisms underpinning salmonid imprinting and homing migration. The study was based on four anadromous Pacific salmon Oncorhynchus spp., pink salmon Oncorhynchus gorbuscha, chum salmon Oncorhynchus keta, sockeye salmon Oncorhynchus nerka and masu salmon Oncorhynchus masou, migrating from the North Pacific Ocean to the coast of Hokkaido, Japan, as well as lacustrine O. nerka and O. masou in Lake Toya, Hokkaido, where the lake serves as the model oceanic system. Behavioural studies using biotelemetry techniques showed swimming profiles from the Bering Sea to the coast of Hokkaido in O. keta as well as homing behaviours of lacustrine O. nerka and O. masou in Lake Toya. Endocrinological studies on hormone profiles in the brain-pituitary-gonad axis of O. keta, and lacustrine O. nerka identified the hormonal changes during homing migration. Neurophysiological studies revealed crucial roles of olfactory functions on imprinting and homing during downstream and upstream migration, respectively. These findings are discussed in relation to the physiological mechanisms of imprinting and homing migration in anadromous and lacustrine salmonids.

Physiological mechanism of homing migration in Pacific salmon from behavioral to molecular biological approaches

The amazing abilities of Pacific salmon to migrate long distances from the ocean to their natal streams for spawning have been investigated intensively since 1950's, but there are still many mysteries because of difficulties to follow their whole life cycle and to wait their sole reproductive timing for several years. In my laboratory, we have tried to clarify physiological mechanisms of homing migration in Pacific salmon, using four anadromous Pacific salmon (pink, Oncorhynchus gorbuscha; chum, Oncorhynchus keta; sockeye, Oncorhynchus nerka; masu, Oncorhynchus masou) in the north Pacific Ocean as well as two lacustrine salmon (sockeye and masu) in Lake Toya and Lake Shikotsu, Hokkaido, Japan, where the lakes serve as a model "ocean". Three different approaches from behavioral to molecular biological researches have been conducted using these model fish. First, the homing behaviors of adult chum salmon from the Bering Sea to Hokkaido as well as lacustrine sockeye and masu salmon in Lake Toya were examined by means of physiological biotelemetry techniques, and revealed that salmon can navigate in open water using different sensory systems. Second, the hormone profiles in the brain-pituitary-gonadal (BPG) axis were investigated in chum salmon and lacustrine sockeye salmon during their homing migration by means of hormone specific time-resolved fluoroimmunoassay (TR-FIA) systems, and clarified that salmon gonadotropin-releasing hormone (sGnRH) plays leading roles on homing migration. Third, the olfactory functions of salmon were studied by means of electrophysiological, behavioral, and molecular biological techniques, and made clear that olfactory discriminating ability of natal stream odors. These results have discussed with the evolutional aspects of four Pacific salmon, sexual differences in homing profiles, and the possibility of dissolved free amino acids (DFAA) as natal stream odors for salmon.

Endocrine control of sexual behavior in teleost fish

Sexual behavior is one of the most profound events during the life cycle of animals that reproduce sexually. After completion of gonadal development that is mediated by various hormones, oviparous teleosts perform a suite of behaviors, often termed as spawning behavior. This is particularly important for teleosts that have their gametes fertilized externally as the behavior patterns ensures the close proximity of both sexes for gamete release, fusion and ultimately the production of offspring. As in other vertebrates, sexual behavior of fish is also under the control of hormones. Testicular androgen is a requirement for male sexual behavior to occur in most fish species that have been studied. Unlike tetrapods, however, ovarian estrogen does not appear to be essential for the occurrence of female sexual behavior for fish that have their gametes fertilized externally. Prostaglandins produced in the ovary after ovulation act as a trigger in some teleosts to induce female sexual behavior. Potentiating effects of gonadotropin-releasing hormone in the brain on sexual behavior are reported in some species. Under endocrine regulation, male and female fish exhibit gender-typical behavior during spawning, but in some fish species there is also some plasticity in their sexual behavior. Sex changing fish can perform both male-typical and female-typical sexual behaviors during their lifetime and this sexual plasticity can also be observed in non-sex changing fish when undergoing hormonal treatment. Although the neuroanatomical basis is not clear in fish, results of field and laboratory observations suggest that some teleosts possess a sexually bipotential brain which can regulate two types of behaviors unlike most other vertebrates which have a discrete sex differentiation of their brain and can only perform gender-typical sexual behavior.

Testing the synergistic effects of GnRH and testosterone on the reproductive physiology of pre-adult pink salmon Oncorhynchus gorbuscha

To test the hypothesis that the hypothalmic gonadotropin-releasing hormone (GnRH) and testosterone (T) co-treatment stimulates both the hypothalmo-pituitary-gonadal (HPG) and hypothalmo-pituitary-interrenal axes, the reproductive and osmoregulatory responses of pre-adult pink salmon Oncorhynchus gorbuscha were compared after GnRH and T administration either alone or in combination. Relative to controls, neither GnRH nor T treatment resulted in significantly greater ovarian or testicular growth, but co-treatment significantly increased ovarian growth after 5 months. Interestingly, the stimulation was undetectable after 3 months. However, once daily photoperiod began shortening after the summer solstice, c. 2 months before the natural spawning date, GnRH+T-treated females were stimulated to produce larger ovaries. Final fish body length and the size of individual eggs did not differ among treatment groups. GnRH+T eggs, however, showed signs of advanced vitellogenesis relative to GnRH-treated and control eggs, whereas T-treated eggs became atretic. Testis size increased significantly from initial values and most males were spermiating, but this growth and development were independent of hormone treatments. Final plasma ion, metabolite and cortisol concentrations did not differ among treatment groups. It is concluded that GnRH+T co-treatment was effective in stimulating female but not male maturation. GnRH and T treatment, however, presumably had little effect on the hypothalmo-pituitary-interrenal axis as observed by ionoregulatory status.

Neuromodulatory effects of gonadotropin-releasing hormone on retinotectal synaptic transmission in the optic tectum of rainbow trout

Gonadotropin-releasing hormone (GnRH) is a hypophysiotropic decapeptide that stimulates the release of gonadotropins from the pituitary. In addition, there are extra-hypothalamic GnRH neurons that project to all regions of the brain and whose function remains unknown. Here, we investigated the effects of GnRH on retinotectal synaptic transmission, the synapses of which are formed between retinal fibers and tectal periventricular neurons that express GnRH receptor mRNA. We used rainbow trout as our study model. The excitatory postsynaptic currents (EPSCs), which were evoked by electrical stimulation of the retinal fibers and recorded in periventricular neurons, were suppressed by antagonists of ionotropic glutamate receptors. EPSCs were increased by application of each of two types of GnRH (GnRH2 and GnRH3) in the trout tectum. Such facilitation lasted for at least 10 min after application of the GnRH. To our knowledge, this is the first report of GnRH modulating conventional synaptic transmission in the brain, suggesting that tectal GnRH enhances tectal sensitivity for retinal inputs. Furthermore, such long-lasting facilitation might occur across all the brain regions innervated by GnRH neurons, and GnRH might simultaneously switch neuronal activities in the brain regions relevant to reproductive behaviors.

Roles of periventricular neurons in retinotectal transmission in the optic tectum

The midbrain roof is a retinorecipient region referred to as the optic tectum in lower vertebrates, and the superior colliculus in mammals. The retinal fibers projecting to the tectum transmit visual information to tectal retinorecipient neurons. Periventricular neurons are a subtype of these neurons that have their somata in the deepest layer of the teleostean tectum and apical dendrites ramifying at more superficial layers consisting of retinal fibers. The retinotectal synapses between the retinal fibers and periventricular neurons are glutamatergic, and ionotropic glutamate receptors mediate the transmission in these synapses. This transmission involves long-term potentiation, and is modulated by hormone action. Visual information processed in the periventricular neurons is transmitted to adjacent tectal cells and target nuclei of periventricular neuron axonal branches, some of which relay the visual information to other brain areas controlling behavior. We demonstrated that periventricular neurons play a principal role in visual information processing in the teleostean optic tectum; the effects of tectal output on behavior is discussed also in the present review.

Molecular regulation of gonadotropin secretion by gonadotropin-releasing hormone in salmonid fishes

Gonadotropin-releasing hormone (GnRH) plays a central role in the control of reproductive function in vertebrates. In salmonids, salmon GnRH (sGnRH) secreted by preoptic GnRH neurons regulates gonadal maturation through stimulation of synthesis and release of pituitary gonadotropins (GTHs). In addition, several lines of our evidence indicate that sGnRH is involved in spawning behavior, and serves to integrate the gonadal maturation with the reproductive behavior. A growing number of studies show that the effects of GnRH are mediated by multiple subtypes of GnRH receptors, successive multiple signaling pathways, and finally multiple transcription factors which act cooperatively to stimulate transcription of GTH subunit genes. This complex regulatory system of the action of GnRH may serve as a molecular basis of divergent physiological strategies of reproductive success in various vertebrate species. In this article, recent data on the molecular mechanisms of action of GnRH are reviewed with special reference to the regulation of synthesis and release of GTHs in the pituitary of salmonids to elucidate the multifunctional action of GnRH.

Year-to-year differences in plasma levels of steroid hormones in pre-spawning chum salmon

Changes in plasma levels of steroid hormones in pre-spawning chum salmon (Oncorhynchus keta) were examined for 6 years in association with sexual maturation. Fish were sampled along their homing pathway from the coastal sea to the spawning ground from 1995 to 2000. Plasma levels of testosterone (T), 11-ketotestosterone (11KT), estradiol-17beta (E2), 17alpha,20beta-dihydroxy-4-pregnen-3-one (DHP), and cortisol were determined by enzyme immunoassays. Sexual maturity was comprehensively estimated by gonadosomatic indices, histology of gonads, nuptial color, spermiation or ovulation ratio. Since the plasma levels of steroid hormones and sexual maturation differed from year to year, they were compared with year-to-year variation of sea surface temperature (SST) of coastal sea to study influence of oceanographic environment on these physiological data. The SST of the migratory route varied among the years, so that we classified the 6 years into cool, intermediate, and warm years. Concerning maturity, the males that returned to the natal hatchery in the warm years were sexually more advanced than those in the cool years. Furthermore, histological data suggested that final oocyte maturation occurred before arrival at the hatchery in one of the warm years, i.e., 1999, while it occurred at the hatchery in one of the intermediate years, i.e., 2000. In the males, T and 11KT levels increased significantly on midway of the homing route in the warm years, whereas they did not show any noticeable changes in the cool years. Furthermore, the levels of T and 11KT on midway of the homing route in the warm years, i.e., 1998 and 1999, were significantly higher than those in one of the cool years, i.e., 1995, in both sexes. In the females, the levels of E2 decreased during upstream migration. Conversely, those of DHP considerably elevated at spawning ground in all years examined. The levels of cortisol were different from year to year regardless of the SST. The present results showed that there were year-to-year differences in plasma levels of steroid hormones and maturity, and some of them may be influenced by the year-to-year variation of SST.

Seasonal changes of responses to gonadotropin-releasing hormone analog in expression of growth hormone/prolactin/somatolactin genes in the pituitary of masu salmon

Gonadotropin-releasing hormone (GnRH) is considered to stimulate secretion of growth hormone (GH), prolactin (PRL), and somatolactin (SL) at particular stages of growth and sexual maturation in teleost fishes. We therefore examined seasonal variation in the pituitary levels of GH/PRL/SL mRNAs, and tried to clarify seasonal changes of responses to GnRH in expression of GH/PRL/SL genes, in the pituitaries of growing and maturing masu salmon (Oncorhynchus masou). Pituitary samples were monthly collected one week after implantation with GnRH analog (GnRHa). The levels of mRNAs encoding GH, PRL, and SL precursors in single pituitaries were determined by a real-time polymerase chain reaction method. The fork lengths and body weights of control and GnRHa-implanted fish of both sexes gradually increased and peaked out in September of 2-year-old (2+) when fish spawned. GnRHa implantation did not stimulate somatic growth, nor elevate gonadosomatic index (GSI) of 1+ and 2+ males, whereas it significantly increased GSI of 2+ females in late August to early September. The GnRHa-implanted 1+ males had higher levels of GH and PRL mRNAs in July, and SL mRNA from June to August than the control males. The levels of GH, PRL, and SL mRNAs in the control and GnRHa-implanted 1+ females, however, did not show any significant changes. Afterward, the PRL mRNA levels elevated in the control 2+ fish of both sexes in spring. GnRHa elevated the GH mRNA levels in both males and females in 2+ winter, and the PRL mRNA levels in females in early spring. Regardless of sex and GnRHa-implantation, the SL mRNA levels increased during sexual maturation. In growing and maturing masu salmon, expression of genes encoding GH, PRL, and SL in the pituitary is thus sensitive to GnRH in particular seasons probably in relation to physiological roles of the hormones.

The effects of thyroxine or a GnRH analogue on thyroid hormone deiodination in the olfactory epithelium and retina of rainbow trout, Oncorhynchus mykiss, and sockeye salmon, Oncorhynchus nerka

Using low (0.5nM) substrate levels we determined the activities of thyroxine (T4) outer-ring deiodination (ORD), T4 inner-ring deiodination (T4IRD) and 3,5,3(')-triiodothyronine (T3) IRD activities in the olfactory epithelium (OLF) and retina (RET) of laboratory-held immature 1-year-old rainbow trout and immature 2.5-year-old sockeye salmon. In both species all three deiodination activities were detected in OLF and RET. For OLF, no particular pathway predominated and activities were similar to those of brain. For RET, T3IRD activity was greater than T4ORD activity and in sockeye RET T3IRD activity exceeded that of liver. Trout immersion for 6 weeks in 100ppm T4 increased plasma T4 levels 3-fold and plasma T3 levels by 50% and caused the anticipated autoregulatory responses in brain and liver deiodination ( downward arrow T4ORD, upward arrow T4IRD, and upward arrow T3IRD); OLF deiodination and RET T4ORD activity were unaltered but RET T4IRD and T3IRD activities increased dramatically. Two injections of a GnRH analogue (20 microgkg(-1)) into sockeye increased plasma T3 levels but not T4 levels and decreased RET T4IRD and T3IRD activities without changing liver, brain, or OLF deiodination. We conclude that in salmonids the main TH deiodination pathways occur in OLF but show no regulation by T4 or GnRH. In contrast, T3IRD activity predominates in RET and can be regulated by T4 and GnRH, suggesting that for RET plasma may be the major T3 source. These findings have implications for thyroidal regulation of sensory functions during salmonid diadromous migrations.

The effects of testosterone on upstream migratory behavior in masu salmon, Oncorhynchus masou

Effects of testosterone (T) on upstream migratory behavior in masu salmon, Oncorhynchus masou, were studied by use of artificial raceways. In Experiment 1, yearling precocious males castrated in August were implanted with a capsule of medical silicone tube containing 500 microg of T in September. Their upstream migration was recorded in an artificial raceway with castrated and sham-operated precocious males implanted with a capsule containing vehicle for 2 months. In Experiment 2, upstream migratory behavior of yearling immature parr implanted with a capsule containing T 500 microg or vehicle was observed from September through November. In Experiment 3, upstream migratory behavior of castrated, castrated + T 50 microg, castrated + T 500 microg, and sham-operated precocious males was observed from September through October. In Experiment 4, upstream migratory behavior of the control, T 50 microg-, T 500 microg-, and T 1000 microg-treated immature parr was observed from September through October. In each experiment, plasma and pituitary samples were taken from the fish that moved upstream and remained to measure levels of T, gonadotropin (GTH) II, thyroxine (T(4)), and triiodothyronine (T(3)) by radioimmunoassay. Administration of T caused increases in plasma T levels within the physiological range. In Experiment 1, the frequency of migration upstream was 0, 19.2, and 35.7% in the castrated, castrated + T 500 microg, and sham-operated groups, respectively. The frequency was higher in the castrated + T 500 microg and sham-operated groups than in the castrated group. In Experiment 2, the frequency was higher in T 500 microg-treated parr (22%) than in the control (2.6%). In Experiments 1 and 2, pituitary contents of GTH II in the T-treated groups and precocious males were higher than those in the castrated precocious males and immature parr. In Experiment 3, castrated + T 50 microg, castrated + T 500 microg, and sham-operated fish showed upstream migratory behavior, whereas castrated fish without T did not. In Experiment 4, the frequency was 5.7, 22.9, 17.1, and 28.6% in the control, T 50 microg-, T 500 microg-, and T 1000 microg-treated groups, respectively. In each experiment, plasma levels of T(4) in migrants were lower than those in nonmigrants, whereas plasma levels of T(3) did not show such changes. From these results, it is inferred that T is a factor influencing upstream migration in masu salmon.