Coadaptation of offspring begging and parental provisioning: A role for prenatal maternal effects?

Recent studies on birds have shown that offspring begging and parental provisioning covary at the phenotypic level, which is thought to reflect genetic correlations. However, prenatal maternal factors, like yolk testosterone, may also facilitate parent-offspring coadaptation via their effects on offspring begging and development. In fact, maternal effects are thought to adjust offspring phenotype to the environmental conditions they will experience after birth, which are in turn strongly dependent on the levels of parental provisioning. Using cross-fostering experiments in canaries, we tested the role of maternal effects on parent-offspring coadaptation from two different approaches. First, we analyzed whether females deposit yolk testosterone in relation to their own or their partner's prospective parental provisioning, measured as the rate of parental feeding to foster nestlings. Second, we investigated whether females deposit yolk testosterone in relation to costs they incurred when raising a previous brood, as this likely impinges on their capacity to provide parental care in the near future. However, from the results of both experiments we have no evidence that canary females deposit yolk testosterone in order to match offspring begging to the levels of care they and/or their partners provide. We therefore found no evidence that yolk testosterone facilitates parent-offspring coadaptation. In addition, our results suggest that the functional consequences of yolk testosterone deposition may relate to hatching asynchrony since it primarily varied with egg laying order.

Maternal thyroid hormones enhance hatching success but decrease nestling body mass in the rock pigeon (Columba livia).

Thyroid hormones (THs) - triiodothyronine (T3) and thyroxine (T4) - are essential for embryonic development in vertebrates. All vertebrate embryos are exposed to THs from maternal origin. As maternal TH levels are known to be essential to embryonic development, the natural variation of maternal THs probably represents a pathway of maternal effects that can modify offspring phenotype. However, potential fitness consequences of variation of maternal TH exposure within the normal physiological range and without confounding effects of the mother have never been experimentally investigated. We experimentally manipulated the levels of yolk T3 and T4 within the physiological range in a species in which the embryo develops outside the mother's body, the Rock Pigeon (Columba livia) eggs. Making use of the natural difference of yolk testosterone between the two eggs of pigeon clutches, we were also able to investigate the potential interaction between THs and testosterone. Elevated yolk TH levels enhanced embryonic development and hatching success, and reduced body mass but not tarsus length between day 14 and fledging. The yolk hormones increased plasma T4 concentrations in females but reduced it in males, in line with the effect on metabolic rate at hatching. Plasma concentrations of T3 and testosterone were not significantly affected. The effects of treatment did not differ between eggs with high or low testosterone levels. Our data indicate that natural variation in maternal yolk TH levels affects offspring phenotype and embryonic survival, potentially influencing maternal and chick fitness.

Dominance-related seasonal song production is unrelated to circulating testosterone in a subtropical songbird.

Circulating testosterone (T) is widely considered to play a key role in the production of sexual displays by male vertebrates. While numerous studies support a role for circulating T in promoting the production of song in male birds, this understanding is based primarily on evidence from seasonally breeding northern temperate species, leaving it unclear whether this mechanism generalizes to other regions of the world. Here we investigate whether variation in circulating levels of T can explain the marked within- and among-individual variation in male song performance observed in a subtropical population of the year-round territorial white-browed sparrow weaver (Plocepasser mahali mahali). Our findings reveal that both circulating T and male song production peaked at a similar time point, halfway through the population-level breeding season. However, while dominant males were more likely to sing and sang for longer than subordinate males, within-group paired comparisons revealed no dominance-related differences in circulating T. Moreover, comparisons both among and within individual dominant males revealed that song duration, syllable rate and proportion of time spent singing were all unrelated to circulating T. Together, our findings suggest that natural variation in male song production, at least in this population of white-browed sparrow weavers, is achieved principally through mechanisms other than variation in circulating T concentration. More widely, our results are in line with the view that male song production is not exclusively regulated by gonadally synthesized steroids.

Temperature-induced variation in yolk androgen and thyroid hormone levels in avian eggs.

Global warming has substantially changed the environment, but the mechanisms to cope with these changes in animals, including the role of maternal effects, are poorly understood. Maternal effects via hormones deposited in eggs, have important environment-dependent effects on offspring development and fitness: thus females are expected to adjust these hormones to the environment, such as the ambient temperature. Longer-term temperature variation could function as a cue, predicting chick rearing conditions to which yolk hormone levels are adjusted, while short-term temperature variation during egg formation may causally affect hormone transfer to eggs. We studied the effects of ambient temperature on yolk androgens (testosterone and androstenedione) and thyroid hormones (thyroxine and triiodothyronine) in great tits (Parus major) using data from unmanipulated clutches from a wild population and from aviary birds (ad libitum food) exposed to different experimental temperature treatments during five years. Both in the wild and in captivity, longer-term pre-laying ambient temperature was not associated with clutch mean yolk hormone levels, while the way androstenedione and thyroxine levels varied across the laying sequence did associate with pre-laying temperature in the wild. Yolk testosterone levels were positively correlated with short-term temperature (during yolk formation) changes within clutches in both wild and captivity. We also report, for the first time in a wild bird, that yolk thyroxine levels correlated with a key environmental factor: thyroxine levels were negatively correlated with ambient temperature during egg formation. Thus, yolk hormone levels, especially testosterone, seem to be causally affected by ambient temperature. These short-term effects might reflect physiological changes in females with changes in ambient temperature. The adaptive value of the variation with ambient temperatures pre-laying or during egg formation should be studied with hormone manipulations in different thermal environments.

Influence of mate preference and laying order on maternal allocation in a monogamous parrot species with extreme hatching asynchrony.

It is well established that in many avian species, prenatal maternal resource allocation varies both between and within clutches and may affect offspring fitness. Differential allocation of maternal resources, in terms of egg weight and yolk composition, may therefore allow the female to adjust brood reduction and to fine-tune reproductive investment in accordance with the expected fitness returns. The adaptive value of such maternal resource allocation is thought to be context-dependent as well as species-specific. We investigated the effects of female preference for her mate on the allocation of prenatal maternal resources in the budgerigar, Melopsittacus undulatus, a monogamous species of parrot that shows an extreme hatching asynchrony. We assessed mate preferences in a two-way preference test and allowed females two breeding rounds: one with the preferred and one with the non-preferred partner. We found no effect of preference on either latency to lay or clutch size, but females mated with the preferred partner laid eggs that contained significantly more yolk. Their eggs also contained significantly more androstenedione but not testosterone. Our results suggest that in this species, female preference may influence maternal resource allocation, and that the functional roles of each androgen in the yolk should be considered separately. In addition, we found a significant effect of laying order on egg and yolk weight as well as on yolk testosterone and androstenedione levels. These measures, however, did not change linearly with the laying order and render it unlikely that female budgerigars compensate for the extreme hatching asynchrony by adjusting within-clutch allocation of prenatal maternal resources.

Yolk concentrations of hormones and glucose and egg weight and egg dimensions in unincubated chicken eggs, in relation to egg sex and hen body weight.

Birds can manipulate offspring sex ratio under natural and experimental conditions and maternal hormones have been shown to be involved in this process. Studies also provided evidence for the presence of sex specific concentrations of yolk hormones in avian eggs. These findings led to the suggestion that yolk hormones could influence genetic sex determination in birds. However, in previous studies, yolk hormone concentrations and egg sex were studied in incubated eggs, although incubation of the eggs and embryonic development can alter yolk hormone concentrations and measured sex ratio. This study is the first to determine a wide array of egg components and hen body weight in relation to the sex of the egg in unincubated eggs. Egg parameters studied were yolk concentrations of testosterone, estradiol, androstenedione, progesterone, dihydrotestosterone, and glucose, and egg weight and dimensions. In addition, we studied the associations among all measured parameters. Associations were found between a number of yolk hormones (progesterone associated with testosterone, estradiol and androstenedione; androstenedione with testosterone; dihydrotestosterone with estradiol and androstenedione) as well as between yolk testosterone and egg length and egg weight. There were no significant overall differences between male and female chicken eggs in any of the measured egg parameters. However, there were a few interactions such as the interaction of egg sex with dihydrotestosterone and with hen body weight which predicted estradiol levels and an interaction of estradiol levels with egg width for predicting sex of egg. Their biological relevance need, however, further study.

Plasticity in metabolism of maternal androgens in avian embryos.

Mothers can influence offspring phenotypes by transferring non-genetic information to the young, which provides them with a flexible tool to adjust the developmental trajectory of the young in fluctuating environments. Mothers can differentially deposit their resources in the same reproductive attempt in relation to the offspring position in the sibling hierarchy. However, whether embryos from different positions can be plastic in their response to the maternal signals, potentially leading to a mother-offspring conflict, is yet unclear. We used Rock pigeons (Columba livia), that lay two egg clutches where maternal androgen levels in second laid eggs at oviposition are higher than in first laid eggs, and investigated the plasticity of embryonic metabolism of maternal androgens. We experimentally elevated androstenedione and testosterone levels in first eggs to that present in second eggs and measured the change in androgen levels and its main metabolites (etiocholanolone and conjugated testosterone) after 3.5 days of incubation. We found that eggs with increased androgens show a different degree of androgen metabolism depending either on the egg laying sequence or initial androgen levels or both. Our findings indicate that embryos have certain plasticity in response to maternal androgen levels depending on maternal signals.

Egg-mediated maternal effects in a cooperatively breeding cichlid fish.

Mothers can influence offspring phenotype through egg-mediated maternal effects, which can be influenced by cues mothers obtain from their environment during offspring production. Developing embryos use these components but have mechanisms to alter maternal signals. Here we aimed to understand the role of mothers and embryos in how maternal effects might shape offspring social phenotype. In the cooperatively breeding fish Neolamprologus pulcher different social phenotypes develop in large and small social groups differing in predation risk and social complexity. We manipulated the maternal social environment of N. pulcher females during egg laying by allocating them either to a small or a large social group. We compared egg mass and clutch size and the concentration of corticosteroid metabolites between social environments, and between fertilized and unfertilized eggs to investigate how embryos deal with maternal signalling. Mothers in small groups produced larger clutches but neither laid smaller eggs nor bestowed eggs differently with corticosteroids. Fertilized eggs scored lower on a principal component representing three corticosteroid metabolites, namely 11-deoxycortisol, cortisone, and 11-deoxycorticosterone. We did not detect egg-mediated maternal effects induced by the maternal social environment. We discuss that divergent social phenotypes induced by different group sizes may be triggered by own offspring experience.

Maternal steroids in egg yolk as a pathway to translate predation risk to offspring: experiments with great tits.

Exposure of mothers to risk of predation can induce phenotypic changes in offspring as shown in several species. We previously found that cross-fostered great tit (Parus major) chicks of females exposed to increased predation risk were smaller and lighter, but had faster wing growth than control cross-fostered chicks, possibly improving predator-escaping abilities. Here we examined the possible role of maternal steroids deposited in eggs as an underlying mechanism. We collected eggs from female great tits under either experimentally increased predation risk (PRED) or control treatments (CON) and analyzed the concentration of testosterone, androstenedione, and progesterone in the yolks. PRED eggs contained lower levels of testosterone than CON eggs, but levels of androstenedione and progesterone did not differ. The smaller size and mass of chicks found in the previous study may thus be explained by the lower testosterone concentrations, since yolk testosterone is known to boost growth and development. Alternatively, testosterone may act as a modulator of differential investment into morphological traits, rather than a simple growth enhancer, explaining lower body mass in conjunction with the accelerated wing growth. This could possibly occur concurrently with other hormones such as corticosterone.

Relationship between oxidative stress and circulating testosterone and cortisol in pre-spawning female brown trout.

Reproduction in vertebrates is an energy-demanding process that is mediated by endogenous hormones and potentially results in oxidative stress. The primary aim of this study was to quantify the relationship between oxidative stress parameters (antioxidant capacity and levels of reactive oxygen metabolites) and circulating testosterone and cortisol in a common and widespread teleost fish, the brown trout (Salmo trutta, L.). Results show that trout with higher testosterone levels prior to spawning have higher levels of oxidative damage at the time that they spawn (although by the time of spawning testosterone levels had dropped, leading to a negative relationship between testosterone and oxidative damage at that time). Cortisol levels were not directly related to oxidative damage or antioxidant capacity, but concentrations of this hormone were positively related to levels of fungal infection, which was itself associated both with lower antioxidant capacity and lower levels of oxidative damage. These results highlight the complexity of interactions between different components of the endocrine system and metabolism and suggest that caution be used in interpreting relationships between a single hormone and indicators of oxidative balance or other fitness proxies.

Effects of artificial dawn on sleep inertia, skin temperature, and the awakening cortisol response.

The effect of artificial dawn during the last 30 min of sleep on subsequent dissipation of sleep inertia was investigated, including possible involvement of cortisol and thermoregulatory processes. Sixteen healthy subjects who reported difficulty with waking up participated in random order in a control and an artificial dawn night. Sleep inertia severity was measured by subjective ratings of sleepiness and activation, and by performance on an addition and a reaction time task measured at 1, 15, 30, 45, 60, and 90 min after waking up at habitual wake up time at workdays. At all intervals, saliva samples were collected for cortisol analysis. Sleep electroencephalogram was recorded during the 30 min prior to waking up; core body temperature and skin temperatures were recorded continuously until 90 min after waking up. Subjective sleepiness was significantly decreased and subjective activation increased after waking up in the artificial dawn condition as compared with control, in which lights were turned on at waking up. These effects can be explained by effects of artificial dawn on skin temperature and amount of wakefulness during the 30 min prior to the alarm. Artificial dawn accelerated the decline in skin temperature and in the distal-to-proximal skin temperature gradient after getting up. No significant effects of artificial dawn on performance, core body temperature, and cortisol were found. These results suggest that the physiology underlying the positive effects of artificial dawn on the dissipation of sleep inertia involves light sleep and an accelerated skin temperature decline after awakening.

Gonadal steroid levels in rock pigeon eggs do not represent adequately maternal allocation.

Maternal hormones deposited in the egg can provide a powerful model for the study of maternal effects. The differential amount of maternal hormones in the yolk of freshly laid eggs is assumed to represent differential maternal allocation. However, some evidence suggests that these amounts do not reflect maternal allocation that in fact takes place before ovulation. We compared the amounts of a wide array of gonadal steroids and their metabolites in the yolk of pre-ovulatory follicles with those of freshly laid eggs of rock pigeons using mass spectrometry. We found that between the follicle and egg stages the levels of progesterone increase whereas androstenedione and testosterone decrease in which the strength of decrease was dependent on the laying order of the egg. For conjugated estrone the change between follicle and egg differed in direction for first and second laying position yielding a significant interaction effect. For conjugated testosterone the interaction did not reach but was close to significance. This extremely early steroid metabolism was not due to maternal enzymes in the yolk as indicated by incubation of pre-ovulatory yolks treated with proteinase-K, a protein digesting enzyme. The results have significant consequences for the functional and evolutionary interpretation as well as experimental manipulation of hormone-mediated maternal effects.

Nasal versus temporal illumination of the human retina: effects on core body temperature, melatonin, and circadian phase.

The mammalian retina contains both visual and circadian photoreceptors. In humans, nocturnal stimulation of the latter receptors leads to melatonin suppression, which might cause reduced nighttime sleepiness. Melatonin suppression is maximal when the nasal part of the retina is illuminated. Whether circadian phase shifting in humans is due to the same photoreceptors is not known. The authors explore whether phase shifts and melatonin suppression depend on the same retinal area. Twelve healthy subjects participated in a within-subjects design and received all of 3 light conditions--1) 10 lux of dim light on the whole retina, 2) 100 lux of ocular light on the nasal part of the retina, and 3) 100 lux of ocular light on the temporal part of the retina--on separate nights in random order. In all 3 conditions, pupils were dilated before and during light exposure. The protocol consisted of an adaptation night followed by a 23-h period of sustained wakefulness, during which a 4-h light pulse was presented at a time when maximal phase delays were expected. Nasal illumination resulted in an immediate suppression of melatonin but had no effect on subjective sleepiness or core body temperature (CBT). Nasal illumination delayed the subsequent melatonin rhythm by 78 min, which is significantly (p= 0.016) more than the delay drift in the dim-light condition (38 min), but had no detectable phase-shifting effect on the CBT rhythm. Temporal illumination suppressed melatonin less than the nasal illumination and had no effect on subjective sleepiness and CBT. Temporal illumination delayed neither the melatonin rhythm nor the CBT rhythm. The data show that the suppression of melatonin does not necessarily result in a reduction of subjective sleepiness and an elevation ofCBT. In addition, 100 lux of bright white light is strong enough to affect the photoreceptors responsible for the suppression of melatonin but not strong enough to have a significant effect on sleepiness and CBT. This may be due to the larger variability of the latter variables.

Maternal adjustment or constraint: differential effects of food availability on maternal deposition of macro-nutrients, steroids and thyroid hormones in rock pigeon eggs.

In oviparous species like birds, eggs provide the direct environment in which embryos are developing. Mothers may adjust different egg components in different ways in reaction to environmental cues either to adjust offspring development or because of constraints. In this study, we investigated the effects of food quality and quantity before and during egg laying on three different aspects of egg quality: macro-nutrients (egg and yolk mass), androgens (testosterone and androstenedione), and thyroid hormones (3,5,3'-triiodothyronine, T3 and l-thyroxine, T4), using the rock pigeon (Columba livia). As expected, egg and yolk mass were significantly reduced for the eggs laid under the poor-food condition, indicating a maternal trade-off between offspring and self in allocating important resources. We did not find any significant change in yolk testosterone or their within-clutch pattern over the laying sequence. This is consistent with the fact that, in contrast with nutrients, these hormones are not costly to produce, but does not support the hypothesis that they play a role in adjusting brood size to food conditions. In contrast, we found that T3 levels were higher in the egg yolks under the poor-food condition whereas the total T4 content was lower. This change could be related to the fact that iodine, the critical constituent of thyroid hormones, might be a limiting factor in the production of this hormone. Given the knowledge that food restriction usually lead to reduction of circulating T3 levels, our results suggested that avian mothers can independently regulate its concentrations in their eggs from their own circulation. The study demonstrates that environmentally induced maternal effects via the egg can be a result of a combination of constrained resources and unconstrained signals and that thyroid hormones might be an interesting case of both. Therefore, this hormone and the interplay of different maternal effects on the offspring phenotype deserve much more attention.

Acute and phase-shifting effects of ocular and extraocular light in human circadian physiology.

Light can influence physiology and performance of humans in two distinct ways. It can acutely change the level of physiological and behavioral parameters, and it can induce a phase shift in the circadian oscillators underlying variations in these levels. Until recently, both effects were thought to require retinal light perception. This view was challenged by Campbell and Murphy, who showed significant phase shifts in core body temperature and melatonin using an extraocular stimulus. Their study employed popliteal skin illumination and exclusively considered phase-shifting effects. In this paper, the authors explore both acute effects and phase-shifting effects of ocular as well as extraocular light. Twelve healthy males participated in a within-subject design and received all of three light conditions--(1) dim ocular light/no light to the knee, (2) dim ocular light/bright extraocular light to the knee, and (3) bright ocular light/no light to the knee--on separate nights in random order. The protocol consisted of an adaptation night followed by a 26-h period of sustained wakefulness, during which a 4-h light pulse was presented at a time when maximal phase delays were expected. The authors found neither immediate nor phase-shifting effects of extraocular light exposure on melatonin, core body temperature (CBT), or sleepiness. Ocular bright-light exposure reduced the nocturnal circadian drop in CBT, suppressed melatonin, and reduced sleepiness significantly. In addition, the 4-h ocular light pulse delayed the CBT rhythm by -55 min compared to the drift of the CBT rhythm in dim light. The melatonin rhythm shifted by -113 min, which differed significantly from the drift in the melatonin rhythm in the dim-light condition (-26 min). The failure to find immediate or phase-shifting effects in response to extraocular light in a within-subjects design in which effects of ocular bright light are confirmed strengthens the doubts raised by other labs of the impact of extraocular light on the human circadian system.

Short-wavelength attenuated polychromatic white light during work at night: limited melatonin suppression without substantial decline of alertness.

Exposure to light at night increases alertness, but light at night (especially short-wavelength light) also disrupts nocturnal physiology. Such disruption is thought to underlie medical problems for which shiftworkers have increased risk. In 33 male subjects we investigated whether short-wavelength attenuated polychromatic white light (<530 nm filtered out) at night preserves dim light melatonin levels and whether it induces similar skin temperature, alertness, and performance levels as under full-spectrum light. All 33 subjects participated in random order during three nights (at least 1 wk apart) either under dim light (3 lux), short-wavelength attenuated polychromatic white light (193 lux), or full-spectrum light (256 lux). Hourly saliva samples for melatonin analysis were collected along with continuous measurements of skin temperature. Subjective sleepiness and activation were assessed via repeated questionnaires and performance was assessed by the accuracy and speed of an addition task. Our results show that short-wavelength attenuated polychromatic white light only marginally (6%) suppressed salivary melatonin. Average distal-to-proximal skin temperature gradient (DPG) and its pattern over time remained similar under short-wavelength attenuated polychromatic white light compared with dim light. Subjects performed equally well on an addition task under short-wavelength attenuated polychromatic white light compared with full-spectrum light. Although subjective ratings of activation were lower under short-wavelength attenuated polychromatic white light compared with full-spectrum light, subjective sleepiness was not increased. Short-wavelength attenuated polychromatic white light at night has some advantages over bright light. It hardly suppresses melatonin concentrations, whereas performance is similar to the bright light condition. Yet, alertness is slightly reduced as compared with bright light, and DPG shows similarity to the dim light condition, which is a physiological sign of reduced alertness. Short-wavelength attenuated polychromatic white light might therefore not be advisable in work settings that require high levels of alertness.

Effects of artificial dawn on subjective ratings of sleep inertia and dim light melatonin onset.

The timing of work and social requirements has a negative impact on performance and well-being of a significant proportion of the population in our modern society due to a phenomenon known as social jetlag. During workdays, in the early morning, late chronotypes, in particular, suffer from a combination of a nonoptimal circadian phase and sleep deprivation. Sleep inertia, a transient period of lowered arousal after awakening, therefore, becomes more severe. In the present home study, the authors tested whether the use of an alarm clock with artificial dawn could reduce complaints of sleep inertia in people having difficulties in waking up early. The authors also examined whether these improvements were accompanied by a shift in the melatonin rhythm. Two studies were performed: Study 1: three conditions (0, 50, and 250 lux) and Study 2: two conditions (0 lux and self-selected dawn-light intensity). Each condition lasted 2 weeks. In both studies, the use of the artificial dawn resulted in a significant reduction of sleep inertia complaints. However, no significant shift in the onset of melatonin was observed after 2 weeks of using the artificial dawn of 250 lux or 50 lux compared to the control condition. A multilevel analysis revealed that only the presence of the artificial dawn, rather than shift in the dim light melatonin onset or timing of sleep offset, is related to the observed reduction of sleep inertia complaints. Mechanisms other than shift of circadian rhythms are needed to explain the positive results on sleep inertia of waking up with a dawn signal.

Time-of-day-dependent effects of bright light exposure on human psychophysiology: comparison of daytime and nighttime exposure.

Bright light can influence human psychophysiology instantaneously by inducing endocrine (suppression of melatonin, increasing cortisol levels), other physiological changes (enhancement of core body temperature), and psychological changes (reduction of sleepiness, increase of alertness). Its broad range of action is reflected in the wide field of applications, ranging from optimizing a work environment to treating depressed patients. For optimally applying bright light and understanding its mechanism, it is crucial to know whether its effects depend on the time of day. In this paper, we report the effects of bright light given at two different times of day on psychological and physiological parameters. Twenty-four subjects participated in two experiments (n = 12 each). All subjects were nonsmoking, healthy young males (18-30 yr). In both experiments, subjects were exposed to either bright light (5,000 lux) or dim light <10 lux (control condition) either between 12:00 P.M. and 4:00 P.M. (experiment A) or between midnight and 4:00 A.M. (experiment B). Hourly measurements included salivary cortisol concentrations, electrocardiogram, sleepiness (Karolinska Sleepiness Scale), fatigue, and energy ratings (Visual Analog Scale). Core body temperature was measured continuously throughout the experiments. Bright light had a time-dependent effect on heart rate and core body temperature; i.e., bright light exposure at night, but not in daytime, increased heart rate and enhanced core body temperature. It had no significant effect at all on cortisol. The effect of bright light on the psychological variables was time independent, since nighttime and daytime bright light reduced sleepiness and fatigue significantly and similarly.

Weak relationships between suppression of melatonin and suppression of sleepiness/fatigue in response to light exposure.

In this paper we examine the relationship between melatonin suppression and reduction of sleepiness through light by comparing three different data sets. In total 36 subjects participated in three studies and received 4 h of bright light either from midnight till 4:00 hours (experiments A and B) or from noon till 16:00 hours (experiment C). In experiment A (night-time light, partial illumination of the retina, pupil dilated) subjects were exposed to either 100 lx of ocular light on the temporal, 100 lx on the nasal part of the retina, or <10 lx of dim light on the whole retina. In experiments B (night-time light, whole retina, pupil not dilated) and C (daytime light, whole retina, pupil not dilated) subjects were exposed either to bright (5000 lx) or to dim light (<10 lx). Subjective sleepiness/fatigue and melatonin concentrations in saliva were assessed hourly in all three experiments. For experiment A, a significant suppression of melatonin due to nasal and temporal illumination of the retina was found, that was not accompanied by a detectable reduction of subjective sleepiness/fatigue. For experiment B we found a suppression of melatonin that was paralleled with a significant reduction in subjective sleepiness, but not in fatigue. During experiment C we found no melatonin suppression but a reduction of subjective sleepiness, but also no effect on fatigue. From these data we conclude that the effects of light on sleepiness/fatigue are not mediated by melatonin and that the influence of endogenous melatonin concentration on sleepiness/fatigue is restricted.