Ultradian and circadian CO2 emission variations in nocturnal and diurnal animals exposed to a light stimulus

1. Carbon dioxide emission (VCO2) has been continuously recorded in three laboratory animal species (Sprague-Dawley rats, Japanese quail, Hartley guinea-pigs) which differ by their nocturnal and diurnal activities. A 100 lux stimulus has been delivered at various time intervals. 2. A regular alternation of 12, 3 or 1.5 hr light (L) and darkness (D) gives VCO2 circadian and ultradian rhythms of 24, 6 or 3 hr periods, respectively, in quail and rats. 3. Such circadian and ultradian LD rhythms are not induced in all guinea-pigs. 4. The amplitudes of the VCO2 responses are greatest at D----L when the animals have a maximum diurnal activity and at L----D when their maximum activity is nocturnal. 5. Interactions between circadian and ultradian rhythms are seen in all LD experiments, as well as in continuous light (LL) or continuous dark (DD). 6. No more well-marked or even inverted VCO2 responses to the light stimuli may occur after several days of exposure to these LD alternations.     

Rats grouping and the circadian and ultradian synchronization of their carbon dioxide emission by a light-dark 12:12 h alternation.

Carbon dioxide emission (VCO2) was continuously recorded during 19 consecutive days in 25 Sprague Dawley young male rats placed in the same "respiratory chamber", grouped by 5 (G) and then separated (S). All rats were in controlled environmental conditions (20 degrees C temperature, humidity, ventilation, food and water ad libitum) and submitted to a light (100 lux)-dark alternation (LD 12:12). The curves obtained with the respiratory chamber CO2 concentration sampled every 20 minutes were analyzed for circadian periods, amplitudes, phases, ultradian peak oscillation intervals and amplitudes, and VCO2 time variations at L-->D and D-->L light transitions. Analysis of variance and t test show circadian amplitudes significantly (P < 0.001) higher (by 40.9%) than in S; moreover, ultradian peak amplitudes were higher in G than in S (by 78.0% in L and 105.8% in D). The circadian and ultradian (tau > 40 min) period intervals were not significantly different in G and in S. Circadian phase differences between L-->D and D-->L were significantly greater in S (by 50.3 min) but not in G. Light transitions did not significantly modify ultradian phases in G and in S. This data shows a better LD 12:12 synchronization in G than in S, resulting mostly from an increased respiratory amplitude modulation due to interindividual interactions.     

Effects of uncoupling of mitochondrial energy conservation on the ultradian clock-driven oscillations in Saccharomyces cerevisiae continuous culture

Protonophores have several different perturbative effects on dissolved O2 concentrations in continuous cultures of Saccharomyces cerevisiae. As well as uncoupling energy conservation from mitochondrial electron transport in vivo, they reset ultradian clock-driven respiratory oscillations and produce cell cycle effects. Thus, additions at low concentration (1.25 microM) of either m-chlorocarbonyl-cyanide phenylhydrazone (CCCP) or 5-chloro-3-t-butyl-2-chloro-4(1)-nitrosalicylanilide (S13) led to phase resetting of the 48 min ultradian clock-driven respiratory oscillations. At 2.5 microM CCCP or 4 microM S13, transient inhibition of oscillatory respiration (for 5 h) preceded synchronisation of the cell division cycle seen as a slow (9 h period) wave that enveloped the 48 min oscillation. At still higher concentrations of CCCP (5 microM), the cell division cycle was prolonged by about 7 h, and during this phase, the respiratory oscillation became undetectable. The significance of these observations with respect to the time-keeping functions of the ultradian clock is discussed.     

Effects of suprachiasmatic nuclei lesions on circadian and ultradian rhythms in body temperature in ocular enucleated rats

Abstract

To test the hypothesis that an oscillator located outside the suprachiasmatic nuclei (SCN) controls the circadian rhythm of body temperature, we conducted a study with 14 blinded rats, 10 of which receiving a SCN lesion. Body temperature was automatically and continuously recorded for about one month by intraperitoneal radio transmitters. Food intake, drinking and locomotor activity were also recorded. Periodograms revealed that 3 rats with histologically verified total bilateral SCN lesions did not exhibit any circadian rhythmicity. The 7 other rats appeared to have partial lesions. They showed shortening of period and severe amplitude reduction in all functions. Thus, no support was found for the hypothesis of a separate circadian ‘temperature oscillator’ located outside the SCN. Nevertheless, after large partial lesions body temperature showed more persistency than some of the other behavioral rhythms.

Ultradian rhythms in temperature persisted after partial and total lesions. Other functions showed parallel ultradian rhythms. In intact rats the ultradian peaks were restricted predominantly to the subjective night. After total lesions these peaks became more or less homogeneously distributed in time but more heterogeneously after partial lesions. So the SCN plays a role in the temporal structure of ultradian rhythms but does not generate them. Non‐24‐hour actograms showed instabilities of period and phase of ultradian rhythms. Intact and lesioned rats were similar with respect to the mean (about 3.5 hrs) and standard deviation (about 1.5 hrs) of ultradian periods in temperature. These features indicate that a mechanism outside the SCN is underlying ultradian rhythmicity, capable of generating short‐term oscillations. Two approaches, homeostatic sleep‐wake relaxation oscillations and multiple circadian oscillators, are discussed.     

Ultradian and circadian compartmentalization of respiratory and metabolic exchanges in small laboratory vertebrates

Carbon dioxide emission (VCO2) taken as an index of respiratory and metabolic exchanges, was continuously recorded during 4-30 consecutive days in 100 quail, 87 chicks, 347 rats, 665 mice and 70 guinea-pigs which were under controlled environmental parameters. Harmonic analysis, fast Fourier transform, chi-square periodograms, peak and trough intervals were computed with VCO2 values obtained with CO2 concentrations sampled every 20 min on the CO2 recordings. In LD 12:12 alternation, circadian rhythms were observed in all quail, chicks, rats and mice, but only in 80% of the guinea-pigs. Ultradian VCO2 rhythms, with periods which show statistically significant interspecies differences, were assessed. For each of the 5 species these computed periods, which were the same in LL and DD, were: 1.17 h for quail and chickens, 1.25 h for rats, 1.50 h for mice and 1.0 h for guinea-pigs. In LD 12:12 these periods were different during L and D in quail, chicks, rats and mice, but not in guinea-pigs. The amplitudes of these ultradian variations were, according to the species, 10-20% of their mean VCO2 levels. These ultradian rhythms persist in the absence (or masking) of circadian rhythms, e.g. in LD 12:12 in 20% of guinea-pigs and in LL in 87% of Japanese quail and in 23% of Sprague-Dawley rats. Moreover, these ultradian rhythms persist during starvation, locomotor activity restraint and ageing. These ultradian VCO2 cycles which are related to rest-activity variations appear to be basic physiological rhythms with a genetic origin.     

Societal synchronization in groups of rats or quail submitted to various lighting regimens

Groups of rats or of quail that had been previously synchronized in a light (L = 100 lux) dark (D) phase opposition (PO = LD and DL) were placed together in a L12:D12 or D12:L12 alternation or in continuous light (LL) or continuous darkness (DD). Emission of carbon dioxide (VCO2) which was continuously recorded in groups of individuals placed in respiratory chambers under controlled environmental conditions allows an index of their overall respiratory and metabolic exchanges to be found. In PO animals placed in LD or DL, the VCO2 circadian light dark synchronization comes back less quickly in rats than in quail, and the VCO2 variations at the light dark transitions (L-D and D-L) remain unchanged in rats, but are modified in quail. When PO animals are placed for 18 days in LL or DD, respiratory circadian rhythms disappear except in the grouped rats where they reappear after 4-5 days in DD.     

Differential elimination of circadian and ultradian rhythmicity by hypothalamic lesions in the common vole, Microtus arvalis

Effects of hypothalamic lesions on the ultradian and circadian organization of wheel running and feeding were studied in the common vole, Microtus arvalis. Circadian organization broke down within 30 days in continuous darkness in 24% of intact voles (n = 135). Ultradian rhythmicity of feeding (period 2-3 hr) persisted in constant conditions in all intact voles. Following lesions of the suprachiasmatic nuclei (SCN), circadian rhythmicity disappeared when lesions were complete (n = 8) or more extensive than 25% of the total SCN volume (n = 5). Absence of circadian rhythmicity was also found in animals with substantial lesions in the diencephalic paraventricular area (PVA) and in the retrochiasmatic area (RCA) and/or adjacent arcuate nucleus (Arc). Complete loss of ultradian and circadian organization occurred in eight voles with damage to the RCA and/or Arc. In three of these, the SCN was intact. The SCN is a likely candidate for a circadian pacemaker in voles (as in other rodents), while the loss of circadian rhythmicity following PVA and RCA/Arc lesions may be due to destruction of efferent pathways from the SCN. The RCA/Arc area is apparently necessary for the expression of ultradian rhythms. The intact SCN is neither necessary nor sufficient for the generation of ultradian rhythmicity.     

Wavelet-based time series analysis of circadian rhythms

Analysis of circadian oscillations that exhibit variability in period or amplitude can be accomplished through wavelet transforms. Wavelet-based methods can also be used quite effectively to remove trend and noise from time series and to assess the strength of rhythms in different frequency bands, for example, ultradian versus circadian components in an activity record. In this article, we describe how to apply discrete and continuous wavelet transforms to time series of circadian rhythms, illustrated with novel analyses of 2 case studies involving mouse wheel-running activity and oscillations in PER2::LUC bioluminescence from SCN explants.     

Ultradian, circahoral and circadian structures in endothermic vertebrates and humans

1. For more than 30 years many studies have been carried out concerning rhythms with periods approaching 24 hr (circadian rhythms). 2. The latter have been demonstrated as resulting from environmental 24 hr synchronizers (zeitgebers), but they usually persist in the absence of a 24 hr synchronization, which proves their endogenous nature. 3. Biological rhythms with periods less than 20 hr (ultradian rhythms) and particularly those approaching 1 hr (circahoral rhythms) have been determined: for motility, rest-activity, sleep phases, endocrine secretions and other physiological functions. 4. These ultradian and circahoral rhythms have been found in rodents, birds, monkeys and humans. 5. Existing at all stages of ontogeny, they have been proved to be endogenous and species and strain specific. 6. As these ultradian rhythms can be influenced by environmental factors and sometimes by circadian rhythms they are not truly periodic, so therefore cannot be computed by the usual processes of mathematical time analysis.     

Further evidence that the circadian clock in Drosophila is a population of coupled ultradian oscillators

We hypothesize that ultradian oscillators are coupled to yield a composite circadian clock in Drosophila. In such a system, period would be a function of the tightness of coupling of these oscillators, increasing as coupling loosens. Ultradian oscillations would become apparent under weak coupling or in the absence of coupling. A new technique for calculating signal-to-noise ratio (SNR) for biological rhythms to characterize their precision has yielded support for this hypothesis. SNR of rhythms of the allelic series of mutations at the period (per) locus of Drosophila melanogaster were compared. Per(o) was the noisiest, grading through perL, per+, and pers, the least noisy. SNR decreases significantly with increasing period in pers, per+, and perL; per(o) typically has multiple ultradian oscillations and the lowest SNR. At least 70% of perL individuals also exhibit ultradian periodicities.     

Extracellular long-term recordings of the isolated accessory medulla,the circadian pacemaker center of the cockroach <Emphasis Type="Italic">Leucophaea maderae</Emphasis>, reveal ultradian and hint circadian rhythms

In the cockroach Leucophaea maderae transplantation studies located the circadian pacemaker center, which controls locomotor activity rhythms, to the accessory medulla (AMe), ventromedially to the medulla of the brain’s optic lobes. The AMe is densely innervated via GABA- and manyfold peptide-immunoreactive neurons. They express ultradian action potential oscillations in the gamma frequency range and form phase-locked assemblies of synchronously spiking cells. Peptide application resulted in transient rises of extracellularly recorded activity. It remained unknown whether transient rises in spontaneous electrical activity as a possible indication of peptide release occur in the isolated circadian clock in a rhythmic manner. In extracellular glass electrode recordings of the isolated AMe in constant darkness, which lasted at least 12 h, the distribution of daytime-dependent changes in activity independently of the absolute action potential frequency was examined. Rapid, transient changes in activity preferentially occurred at the mid-subjective night, with a minimum at the middle of the subjective day, hinting the presence of circadian rhythms in the isolated circadian clock. Additionally, ultradian rhythms in activity change that are multiples of a fundamental 2 h period were observed. We hypothesize that circadian rhythms might originate from coupled ultradian oscillations, possibly already at the single cell level.     

Physiology and regulation of biological rhythms in laboratory animals: an overview

Biological rhythms have been observed in practically all groups of laboratory mammals and at every level of physiological and behavioural organization. Biological rhythms are classified according to their period as ultradian (less than 24 h), circadian (approximately 24 h), infradian (greater than 24 h), and seasonal or circannual rhythms (approximately 1 year). This review outlines what is known about the neurobiology of biological rhythms in mammals and describes the hierarchical order in which ultradian, circadian and infradian rhythms are related to each other. The article does not attempt to catalogue every physiological variable showing rhythmical fluctuations in laboratory mammals. Rather, it focuses on the basic concepts of circadian rhythms and recent advances made in our understanding of the physiology of the internal clock controlling circadian and other biological rhythms.     

Oscillations in joint synchrony of reproductive hormones in healthy men

Negative-feedback (inhibitory) and positive-feedforward (stimulatory) processes regulate physiological systems. Whether such processes are themselves rhythmic is not known. Here, we apply cross-approximate entropy (cross-ApEn), a noninvasive measurement of joint (pairwise) signal synchrony, to inferentially assess hypothesized circadian and ultradian variations in feedback coupling. The data comprised simultaneous measurements of three pituitary and one peripheral hormone (LH, FSH, prolactin, and testosterone) in 12 healthy men each sampled every 10 min for 4 days (5,760 min). Ergodicity, due to the time series stationarity of the measurements over the 4 days, allows for effective estimation of parameters based upon the 12 subjects. Cross-ApEn changes were quantified via moving-window estimates applied to 4-day time series pairs. The resultant ordered windowed cross-ApEn series (in time) were subjected to power spectrum analysis. Rhythmicity was assessed against the null hypothesis of randomness using 1,000 simulated periodograms derived by shuffling the interpulse-interval hormone-concentration segments and redoing cross-ApEn windows and spectral analysis. By forward cross-ApEn analysis, paired LH-testosterone, LH-prolactin, and LH-FSH synchrony maintained dominant rhythms with periodicities of 18-22.5, 18, and 22.5 h, respectively (each P < 0.001). By reverse (feedback) cross-ApEn analysis, testosterone-LH, testosterone-prolactin, and testosterone-FSH synchrony cycles were 30, 18, and 30-45 h, respectively (each P ≤ 0.001). Significant 8- or 24-h rhythms were also detected in most linkages, and maximal bihormonal synchrony occurred consistently at ～0400-0500. Collectively, these analyses demonstrate significant ultradian (<24 h), circadian (～24 h), and infradian (>24 h) oscillations in pituitary-testis synchrony, wherein maximal biglandular coordination is strongly constrained to the early morning hours.     

Photoperiodic influences on ultradian rhythms of male siberian hamsters

Seasonal changes in mammalian physiology and behavior are proximately controlled by the annual variation in day length. Long summer and short winter day lengths markedly alter the amplitude of endogenous circadian rhythms and may affect ultradian oscillations, but the threshold photoperiods for inducing these changes are not known. We assessed the effects of short and intermediate day lengths and changes in reproductive physiology on circadian and ultradian rhythms of locomotor activity in Siberian hamsters. Males were maintained in a long photoperiod from birth (15 h light/day; 15 L) and transferred in adulthood to 1 of 7 experimental photoperiods ranging from 14 L to 9 L. Decreases in circadian rhythm (CR) robustness, mesor and amplitude were evident in photoperiods ≤14 L, as were delays in the timing of CR acrophase and expansion of nocturnal activity duration. Nocturnal ultradian rhythms (URs) were comparably prevalent in all day lengths, but 15 L markedly inhibited the expression of light-phase URs. The period (τ'), amplitude and complexity of URs increased in day lengths ≤13 L. Among hamsters that failed to undergo gonadal regression in short day lengths (nonresponders), τ' of the dark-phase UR was longer than in photoresponsive hamsters; in 13 L the incidence and amplitude of light-phase URs were greater in hamsters that did not undergo testicular regression. Day lengths as long as 14 L were sufficient to trigger changes in the waveform of CRs without affecting UR waveform. The transition from a long- to a short-day ultradian phenotype occurred for most UR components at day lengths of 12 L-13 L, thereby establishing different thresholds for CR and UR responses to day length. At the UR-threshold photoperiod of 13 L, differences in gonadal status were largely without effect on most UR parameters.     

Circadian and Ultradian Rhythmicities in Very Premature Neonates Maintained in Incubators

Six very premature babies (born at 26–28 weeks gestational age) have been studied in hospital for 11–17 weeks, while in intensive care and in an incubator. Apart from suffering occasionally from the neonatal disorders of haemolytic jaundice and ‘respiratory distress of the newborn’, the babies were healthy and developed normally. Initially, the babies were continuously fed intravenously, and the lighting in the ward was on continuously. Routine care was given round the clock. When their medical condition permitted it, the babies were moved in their incubator to an adjacent ward, where they took frequent (2–4 hourly) small meals by mouth, the lighting was dimmed at night, and routine care tended to be given more in the daytime. Hourly recordings of insulated skin temperature were taken throughout the study, and it is the detection of rhythmicity in these measurements that has been the subject of the present study. The methods used were Phase-weighted Stacks, Phasor Walkout and Power Spectral Analysis. These methods have previously been used mainly in geophysical studies, and their value is that they can detect weak signals in noisy data and do not assume a particular waveform of any signal. Circadian rhythmicity was found in all babies for much of the time that were in the constant environment provided by the incubator. Ultradian rhythms were sometimes present also, but they were shorter-lived, and showed a wide range of changing periods, generally in excess of 8 h. When the babies were being treated for jaundice or respiratory distress, there was a tendency for the circadian rhythms to become weaker and for a broader spectrum of ultradian periods to appear. Placing babies in the 12 h : 12 h light : dark environment provided by the ward, and instituting feeding by mouth, had, in most cases, only modest effects upon either circadian or ultradian rhythms. Thus, circadian rhythms continued (but generally with a period not exactly equal to 24 h), and ultradian rhythms, when present, often did not show periods that could be related easily to feeding or care-giving. These results are discussed in terms of evidence for endogenous and exogenous origins of the observed rhythms, and of theories that have postulated the relationship between circadian and ultradian rhythms. It is concluded that the results from the present analyses are difficult to reconcile with the view that circadian rhythms develop from interactions between ultradian oscillators. We suggest that they indicate a matu-ration of the circadian system as a consequence of increasing associations between the circadian elements that are present in the suprachiasmatic nuclei and in other oscillators of the circadian system. The new analytical methods used here also indicate that the results obtained from time-frequency analysis depend to some extent upon the method used.     

Time-Dependent Effects of Trichloroethylene on Motor Activity in Rats

Circadian variations in acute and subacute neurobehavioural effects of trichloroethylene (TRI: 1.2 g/kg i.p.) were investigated in the rat under a light: dark = 12:12 hr cycle. An acute effect of TRI evaluated by decreased muscle tone was maximal during the early dark phase (21:00). A subacute effect of TRI was evaluated by a continuous recording of spontaneous locomotor activity in the rat. The circadian rhythm in spontaneous locomotor activity was extensively impaired by the injection of TRI for three consecutive days. Spectral analysis of spontaneous locomotor activity showed that ultradian periods became more dominant than the circadian period, and the 1//fluctuation of the spectrum disappeared after the injection of TRI. The effect of TRI on the circadian rhythm in spontaneous locomotor activity was circadian-phase dependent, and the treatment of TRI at 09:00 provoked greater circadian rhythm impairment than that at 21:00. The mechanisms of the time-dependent effect of TRI on neurobehaviour are the subject of further investigation.     

Time-Dependent Effects of Trichloroethylene on Motor Activity in Rats

Circadian variations in acute and subacute neurobehavioural effects of trichloroethylene (TRI: 1.2 g/kg i.p.) were investigated in the rat under a light: dark = 12:12 hr cycle. An acute effect of TRI evaluated by decreased muscle tone was maximal during the early dark phase (21:00). A subacute effect of TRI was evaluated by a continuous recording of spontaneous locomotor activity in the rat. The circadian rhythm in spontaneous locomotor activity was extensively impaired by the injection of TRI for three consecutive days. Spectral analysis of spontaneous locomotor activity showed that ultradian periods became more dominant than the circadian period, and the 1//fluctuation of the spectrum disappeared after the injection of TRI. The effect of TRI on the circadian rhythm in spontaneous locomotor activity was circadian-phase dependent, and the treatment of TRI at 09:00 provoked greater circadian rhythm impairment than that at 21:00. The mechanisms of the time-dependent effect of TRI on neurobehaviour are the subject of further investigation.     

Ultradian, circadian and seasonal variations of plasma progesterone and LH concentrations during the luteal phase

The circadian variations in plasma progesterone (P) and LH concentrations were investigated in six women, aged 23-40 years. All were studied in the mid-luteal phase (7 +/- 2 days after LH mid-cycle surge). Experiments were conducted in autumn and in spring. Blood samples were obtained every 15 min for 24 hr. Plasma P and LH concentrations were measured by RIA. Each subject's time-series was analysed using three methods; visual inspection (chronogram), spectral analysis to estimate component periods of rhythms (tau) and cosinor analysis to quantify the rhythms parameters. Marked temporal variations in plasma P concentration were observed in each subject. The maximal variations over a 24-hr period, ranged between 13-58.5 mmol/l. Differences related to sampling time were statistically validated by ANOVA (p less than 0.00001). Significant harmonic periods were detected by spectral analysis but differed among subjects. In all subjects but one, a circadian rhythm was detected. The acrophase location was similar (about 0700 hr) in the four subjects studied in autumn, but ranged from 1940 to 0320 hr in those studied in spring. An ultradian rhythm with tau = 8 hr was also validated in six time-series with similar acrophases (about 0200, 1000, and 1800 hr). Cosinor analysis of pooled data revealed that the 24-hr, 12-hr, and 8-hr rhythms were statistically significant (p = 0.001) in autumn. algebraic sum of these three cosine functions yielded a circadian waveform with peak-times occurring near 0300 and 1130 hr and a trough-time about 2200 hr. In spring, the circadian pattern appeared quite different, and peak-times were found near 0700 and 2000 hr, and trough-times near 0300 and 1500 hr. Furthermore, the 24-hr mean of P was higher in autumn (28.9 +/- 0.4 nmol/l) than in spring (17.2 +/- 0.4 nmol/l), p from ANOVA less than 0.00001. The evidence for a similar circadian LH pattern is not as strong. Seasonal, circadian and ultradian rhythms characterize the physiologic time structure of plasma P concentration in mid-luteal phase.     

Ultradian Rhythm of Activity in Japanese Quail Groups under Semi-Natural Conditions during Ontogeny: Functional Aspects and Relation to Circadian Rhythm

The function of ultradian rhythms is not yet clearly elucidated. In particular, short-term rhythms are expressed during early ontogeny, especially in broods of precocial birds. We investigated the relationship between the clarity of the ultradian rhythm of the activity/rest cycle of a group of young Japanese quail (Coturnix japonica) and the level of social synchronisation and spatial cohesion between the birds within that group. The subjects were descended from two lines selected for either very pronounced rhythmic or arrhythmic circadian activity. We found a positive relationship between the clarity of the ultradian rhythm of the activity/rest cycle when birds were young and the clarity of the circadian rhythm of feeding activity when birds were older, but still immature. The temporal organisation of the behaviour of the chicks from these two lines was observed in outdoor aviaries, when they were 4, 8, 12 and 15 days old. The mean ultradian period expressed by groups of 12 chicks was variable, with a minimum of 6 minutes. The ultradian period lengthened regularly as chicks grew older, and reached approximately 40 min on day 15. The clarity of the ultradian rhythmicity of group activity was linked to the level of inter-individual social synchronisation and of spatial cohesion; the more pronounced the ultradian rhythms of a group, the greater the temporal and spatial cohesion of the chicks within the group. Moreover, these characteristics varied with the age of the chicks. Finally, chicks in the less rhythmic groups weighed less. These results stress the adaptive value of this temporal organisation strategy under natural conditions.     

Ultradian Rhythm of Activity in Japanese Quail Groups under Semi-Natural Conditions during Ontogeny: Functional Aspects and Relation to Circadian Rhythm

The function of ultradian rhythms is not yet clearly elucidated. In particular, short-term rhythms are expressed during early ontogeny, especially in broods of precocial birds. We investigated the relationship between the clarity of the ultradian rhythm of the activity/rest cycle of a group of young Japanese quail (Coturnix japonica) and the level of social synchronisation and spatial cohesion between the birds within that group. The subjects were descended from two lines selected for either very pronounced rhythmic or arrhythmic circadian activity. We found a positive relationship between the clarity of the ultradian rhythm of the activity/rest cycle when birds were young and the clarity of the circadian rhythm of feeding activity when birds were older, but still immature. The temporal organisation of the behaviour of the chicks from these two lines was observed in outdoor aviaries, when they were 4, 8, 12 and 15 days old. The mean ultradian period expressed by groups of 12 chicks was variable, with a minimum of 6 minutes. The ultradian period lengthened regularly as chicks grew older, and reached approximately 40 min on day 15. The clarity of the ultradian rhythmicity of group activity was linked to the level of inter-individual social synchronisation and of spatial cohesion; the more pronounced the ultradian rhythms of a group, the greater the temporal and spatial cohesion of the chicks within the group. Moreover, these characteristics varied with the age of the chicks. Finally, chicks in the less rhythmic groups weighed less. These results stress the adaptive value of this temporal organisation strategy under natural conditions.