Increasing the colonic temperature by up to 20 C did not cause. (Received 13 May 1970) associated with arousal and movement. It was lower than colonic

Transcription

1 J. Phyeiot. (1970), 211, pp With 6 text-ftgure8 Printed in Great Britain THE ROLE OF HYPOTHALAMIC TEMPERATURE IN THE CONTROL OF PANTING IN THE CHICKEN EXPOSED TO HEAT BY S. A. RICHARDS* From Wye College, University of London, Near Ashford, Kent (Received 13 May 1970) SUMMARY 1. In unanaesthetized chickens, the temperatures of the hypothalamus, colon and skin have been recorded in relation to the onset and cessation of thermally induced panting. 2. During control conditions, hypothalamic temperature showed fluctuations associated with arousal and movement. It was lower than colonic temperature by about 0.90 C but this difference generally decreased during exposure to heat. 3. When the birds were exposed abruptly to 400 C, or to ambient temperatures increasing gradually from 20 to 520 C, there was a delay of min and marked increases in both peripheral and central body temperatures before panting commenced. 4. Infra-red irradiation ofthe thorax and abdomen caused vasodilatation in the comb and toe before detectable increases in the deep body temperatures. Increasing the colonic temperature by up to 20 C did not cause panting until hypothalamic temperature was also raised. This inhibitory effect of normal hypothalamic temperature was enhanced by low ambient temperature. 5. Infra-red irradiation of the head increased hypothalamic temperature by up to 3.50 C and caused vasodilatation in the toe without changes in colonic temperature. Panting, however, did not occur so long as colonic temperature was within the normal range. The inhibitory effect of normal colonic temperature was again enhanced by low ambient temperature. 6. In anaesthetized chickens, selective heating of the head and body caused panting only after increases in both the hypothalamic and colonic temperatures. 7. Repeated exposure of birds to 400 C did not reduce the time delay before panting started. * Present address: Department of Physiology, Makerere University Medical School, P.O. Box 7072, Kampala, Uganda.

2 342 S. A. RICHARDS 8. It is concluded that panting in the fowl requires an increase in some extracranial deep body temperature as well as in that of the hypothalamus. INTRODUCTION The relative importance of central and peripheral thermal stimulation in initiating and inhibiting the response of panting has been debated for over a century. Most of the work on mammals has indicated an interaction of these two components in which the role of hypothalamic temperature is especially important. Two recent reviews have summarized the available information on the thermosensitivity of the hypothalamus (Bligh, 1966) and on its function in those animals which depend upon panting to lose excess heat (Richards, 1970). There has been very little work on the part played by specific body temperatures in the control of thermoregulatory mechanisms in birds. The efficient insulation provided by feathers might be expected to limit the influence of ambient temperature and hence of peripheral factors, although the unfeathered legs are known to be important in thermoregulation (Kahl, 1963; Steen & Steen, 1965) and the comb and wattles of chickens have often been considered important sites for physical heat loss. It is also possible that these extremities might contain receptors which could elicit panting reflexly as occurs with the poorly insulated scrotal skin of the sheep (Waites, 1962). The experiments described here show that such is not the case in the fowl where panting is always preceded by a rise in core temperature. Nevertheless, some influence of peripheral thermoreceptors cannot be ruled out, and it therefore seemed of interest to test the idea that panting might be established as a conditioned reflex by repeated exposure to heat, rather as Hammouda (1933) proposed for the dog. METHODS Birds. Seventeen mature hens (Thornbers '404' hybrids) weighing kg were used. They were kept at about 15/13 C (dry bulb/wet bulb) throughout the period of their use. Food was withheld for 24 hr before any experiment but water was available until about 1 hr beforehand. Preparation. Five birds had thermistors, and five had thermocouples, implanted into the anterior hypothalamus. Each fowl was anaesthetized with ether and the head placed in a rat-type stereotaxic instrument modified to hold the beak. The temperature probe was then implanted using conventional techniques. The coordinates (in mm) used for placement of the probe tip between the anterior commissure and optic chiasma were: A 8*5; D 5 0; R 0'5, based on the atlas of the chicken brain by van Tienhoven & Juhasz (1962). The wires from the thermistors were soldered to micro-miniature strip-connectors (221 series, Amphenol Ltd.) and the sockets fixed to the skull with acrylic dental cement (Dental Fillings Ltd.). For the thermocouples, 20 cm lengths of insulated wire were used; they were threaded through a single polyethylene tube, passed subcutaneously for about 5 cm and

3 HYPOTHALAMIC TEMPERATURE AND PANTING 343 then brought out and sutured lightly to the dorsal skin of the neck for soldering to the leads from the recorder. No antibiotics were given and there were no infections. The birds were allowed at least 2 weeks to recover before their use in experiments. Temperature&. Naked-bead thermistors of about 0-6 mm diameter (VA 3102, Mullard Ltd.), soldered to 'diamel' coated silver wire of 0-15 mm diameter (Johnson, Matthey Ltd.) were used for hypothalamic temperature (Thy). One wire was passed through a polyethylene tube of 0-61 mm outside diameter (PPV 10, Portex Ltd.); the other wire and bead were then attached to the tubing by dipping in thin Araldite (MY 753/HY 951, Ciba Ltd.). Brain thermocouples were similarly constructed from 36 S.W.G. copper and constantan wires. Colonic temperature (Ta) was measured using a precision thermistor (44005, Yellowstone Instruments) mounted at the tip of a 6 cm long probe and held in position by rubber bands over the back and under the wings. Surface skin temperatures were measured on (i) the posterior lobe of the comb (TCb), (ii) the middle phalanx of the third digit of the left foot (Tt) and (iii) the middle of the back between the scapulae (Tb), using thermocouples attached by squares of surgical tape. Colonic and skin probes were calibrated frequently over the expected range in a standard water-bath; brain probes were calibrated before implantation and checked again after sacrifice of the bird. Temperatures were written out continuously on an M8 recorder (Devices Instruments Ltd.) or at 36 see intervals on a D.E. 6- channel potentiometric recorder (Cambridge Instruments Ltd.). The range of the core temperatures permitted changes of 0.05 C to be detected during routine recording and 0.10 C differences for peripheral temperatures. Higher sensitivities were used to investigate changes in hypothalamic temperature. Comb and toe temperatures lower than 200 C were not recorded. Respiratory frequency (f ). Pressure changes in a balloon-type pneumograph placed around the keel of the sternum were detected by a strain gauge transducer (Ether Ltd.) and displayed on a D43 oscilloscope (Telequipment Ltd.) via a DC2D preamplifier (Devices). During normal respiration the frequency was counted over 30 see periods using a stop watch, but during polypnoea it was written out on the recorder. Results are expressed as breaths/min; an increase above 100 breaths/min was arbitrarily designated the 'onset of panting' in all experiments and a decline to below this frequency the 'cessation of panting'. Histology. Eight of the chronic birds were killed with an overdose of pentobarbitone sodium and the heads perfused via the common carotid arteries with 0-9 % saline followed by 10% formol saline. After further fixation, three of the brains were removed, frozen, and sectioned at 50,t in the frontal plane and stained with thionin. In each case the tip of the tract left by the temperature probe was within 1 mm of the desired location. The remaining five brains were then sectioned by hand in the sagittal plane and the tracts identified in the anterior hypothalamus. ExperimentM on anae8thetized bird. Three experiments were performed on fowls anaesthetized with pentobarbitone sodium (175 mg/kg intramuscularly), held in the stereotaxic instrument throughout, and supported in the upright position by a wire through the fused dorsal spinous process of the synsacrum. The common carotid arteries were previously dissected out to beyond the bifurcation anteriorly, exposing the junction of the occipital and external carotid arteries (Richards, 1967). Both vertebral arteries were ligated. The birds were injected intravenously with heparin (500 i.u./kg) and 20 cm lengths of polyethylene tubing (PP 90, Portex) inserted into the common carotids to divert the blood for warming or cooling by submersion of the coils in a water-bath. In this way, the temperature of the head could be varied independently of the rest of the body. For the limited purpose of these experiments it was not found necessary to control the temperature of the blood returning from the head.

4 344 S. A. RICHARDS Experimental procedure. Each of the unanaesthetized chickens was restrained in a wooden frame which lightly secured the neck, legs and wings, while the bird stood normally. The hens were accustomed to the procedure and would eat and drink freely when permitted to do so. The frame could be rapidly transferred between the experimental chambers used and the connecting leads passed through a small aperture in the door. The chambers were (i) a thermostated, air-cooled 'cold-room' for ambient dry bulb temperatures (Ta) between 8 and 200 C, and (ii) an incubator, either receiving air at laboratory temperature, C, or electrically heated for temperatures between 28 and 52 C. The humidity in the rooms was measured but could not be controlled. The rate of air movement was constant in the chambers at about 25 and 47 cm/sec respectively. Before surgery each bird was subjected to a standard test procedure so that the normal responses could be measured and any effects of the operation on the thermoregulatory performance later assessed. The test consisted of exposure for 1 hr in a neutral environment of about 15/130C (dry bulb/wet bulb), followed by 3 hr at 40/21 C and 1 hr of recovery at 15/13 C. Colonic temperature and respiratory frequency were monitored at 10 min intervals, or more frequently at the start of exposure and recovery periods. An identical test was later performed on each bird between 2 and 5 weeks after operation. In all subsequent experiments a preliminary control period of not less than 1 hr was observed during which there was no form of heat stress. The five body temperatures and ambient temperature were measured throughout the experiments and respiratory frequency noted every few minutes. Each bird was subjected twice for 1 hr periods to ambient temperatures increasing from about 200 C to a maximum of C, and then allowed to recover as the temperature declined to 20-25' C in a further min. They were also abruptly exposed once or twice each to 40/21 C for periods of 1-4 hr and observed again during recovery. Two 250 W infra-red bulbs were placed at cm from the birds in experiments where ambient temperature was kept constant and the effect of radiant heat observed. Areas of the body which were not to be heated were shielded with asbestos plates; the back and the legs and feet were never subjected to direct irradiation. Four birds were used in experiments which attempted to establish the onset of panting as a conditioned response. In these, only respiratory frequency was measured, together with the time delay before panting started when birds were subjected to 40/210 C. Six control observations were made on the naive birds at intervals of 3 or 4 days to avoid any possible conditioning or acclimation effects, and there were subsequently 100 exposures made on 54 consecutive days. At least 6 hr were allowed for recovery between two exposures on the same day. RESULTS The standard exposures to 40/210 C, before and after implantation of the temperature probes, gave no evidence of any changes in the responses of birds which could be attributed to surgery. Each of the ten operated fowls exhibited normal behaviour, including spontaneous eating and drinking, and tolerated exposure to the hot environments without distress. Normal hypothalamic temperature Under all control conditions, the temperature of the fowl's hypothalamus was found to be lower than that of the colon. Mean values from forty-two

5 HYPOTHALAMIC TEMPERATURE AND PANTING 345 experiments on the ten operated birds, measured after an undisturbed period of 1-2 hr at C, were C (± 0-06 S.E.) and C (± 0.06) respectively. Hypothalamic temperature was subject to fluctuations of up to 0.80 C, often clearly correlated with the bird's state of arousal, for example, following auditory or visual stimuli, or the sudden opening of the chamber door (Fig. 1). These increases did not occur immediately at the start of a stimulus, but generally after a delay of sec. Thus, stimuli of only a few seconds' duration usually caused little or no disturbance in brain temperature. The changes were usually independent of colonic or skin temperatures and they were not associated with thermoregulatory activities. Transient increases in hypothalamic temperature were observed during the initial excitement of birds as they were being prepared for an experiment, and these occasionally reached higher levels than those prevailing at the start of panting caused by heat exposure. Effects of exposure to high or gradually increasing ambient temperature Table 1 presents the mean values for the core and skin temperatures of fourteen experiments on the ten birds under control conditions and when subjected abruptly to 40/210 C. Figures for the experiments involving gradually increasing ambient temperature are not included because in these the rate of rise of temperature varied somewhat according to external ambient conditions. A representative experiment of each type is illustrated in Fig. 2. There was an increase in both skin and deep body temperatures before panting could be initiated by either procedure. The temperatures of the comb and toe rose steeply with ambient temperature, but these changes were not associated with any abrupt acceleration of respiratory frequency. Rather, the panting response occurred after a delay which averaged, in the ten operated birds, 31 min (+ 6-0 S.E.) following exposure to 40/210 C. The increment in hypothalamic temperature at the start of panting under these conditions averaged 1.10 C, but there were considerable differences between individuals as well as between the same individual on different occasions. The mean increment in colonic temperature was only 0.70 C, so that the visceral-cerebral gradient typically decreased during exposure to heat. There was no evidence that panting started at one particular hypothalamic temperature. The decline in respiratory frequency was rapid during the recovery phase and the cessation of panting generally occurred when both the colonic and hypothalamic temperatures were still considerably elevated above their control levels (Table 1; Fig. 2); on average, however, the colonic temperature was 0.30 C higher, but the hypothalamic temperature 0.40 C lower, than their respective values at the start of panting.

6 346 S. A. RICHARDS Effects of irradiating the thorax and abdomen with infra-red heat The idea that a strong peripheral stimulus might initiate the thermoregulatory responses in the absence of increases in colonic or hypothalamic temperature was investigated at various ambient temperatures by warming with infra-red irradiation either the two sides of the body beneath I 2 C 0co._L CLI cad - 0EL Time (min) Fig. 1. The relationship between hypothalamic temperature and arousal in the conscious chicken. At the top, the event marked 1 consisted of hammering on the wall of the chamber containing the bird; this caused mild excitement, as indicated by the irregular respiratory movements, and an increase in hypothalamic temperature (Thy). Event 2 involved opening the chamber door and hammering; this was followed by great excitement and a larger rise in hypothalamic temperature. When the door was closed, the chamber light was also switched off; intermittent excitement continued for almost 2 min, causing a gradual rise in colonic temperature (T). Despite this, hypothalamic temperature fell in the dark to the lowest level reached. the wings, or the breast and abdomen simultaneously. The head was sheltered from the effects of irradiation. It was found at all ambient temperatures that this procedure caused vasodilatation in both the comb and toe, as judged by the increase in

7 HYPOTHALAMIC TEMPERATURE AND PANTING 347 TABLE 1. Body temperatures of chickens in relation to panting induced by abrupt exposure to 40/2 10 C and stopped by recovery at about 200 C. Mean values ( ± S.E.) of fourteen experiments on ten birds Site Colon Hypothalamus Back skin* Comb skin Toe skin Control Start of panting (O C) (O C) 41*5 + 0*07 42*2 + 0* * *07 41* ± *0 + 1*08 36* * Eight observations only. End of panting (O C) *5 + 0*21 41*6 + 0*30 34* ±1F25 U 0 C" < 0. E 0 0 c' V tn) L 00. E f I I I3 I Time (min) Time (min) Fig. 2. The effect on body temperatures and respiratory frequency of exposing chickens abruptly to 40/210 C (A), and to gradually increasing ambient temperature (B). Representative experiments on two birds. Note the different temperature scales. Abbreviations: f, respiratory frequency; T., ambient temperature; Tb, back skin temperature; Tc, colonic temperature; Tlb, comb skin temperature; Thy, hypothalamic temperature; T,, toe skin temperature.

8 348 S. A. RICHARDS their temperatures, and that the effect frequently occurred before any significant rise in hypothalamic or colonic temperature (Fig. 3A, B). During irradiation, these peripheral temperatures were subject to wide fluctuations, especially at ambient temperatures below about 240 C. Panting could be produced by heating the thorax and abdomen at all ambient temperatures from 8 to 300 C, although never as a result of the raised skin temperature alone; the response occurred only after increases ^- 45 0o_ 0 Tb 4 I 0,W 3-5 J 30 E V) 5 C 4 -,4 O4.3 w 4.@ I0 E.0 W 411, U 25 a 20 A 10.-c 15 I in 5 0 Ta 10 C *0 f r Time (min) Ta 270C Time (min) Fig. 3. The effects, at different ambient temperatures, of irradiating the thorax and abdomen with infra-red heat. The arrows indicate the beginning and end of irradiation. Representative experiments on two birds. Abbreviations as in Fig. 2. in both the colonic and hypothalamic temperatures. Under these conditions, however, the increment in colonic temperature was greater than that of hypothalamic temperature (Fig. 3A), except at C ambient temperature, when infra-red heating elicited increases of similar magnitude in both of the core temperatures and a more rapid onset of panting (Fig. 3B). At the lower ambient temperatures (8-15 C), colonic temperature was sometimes elevated by more than 20 C above the control level before hypothalamic temperature began to rise significantly (Fig. 3A),

9 HYPOTHALAMIC TEMPERATURE AND PANTING 349 and in these circumstances panting did not occur despite the systemic hyperthermia. So long as the chicken's head was protected from the infra-red heat, there was an inverse relationship between ambient temperature and the increment in colonic temperature necessary to elicit panting (Fig. 4A). When the heating was switched off, a transient increase was sometimes observed in both toe and comb temperatures. The subsequent decrease, indicating vasoconstriction, was most rapid at low ambient temperatures, 30 A 40 B 0U A'20L. 0 s 0 0, 10, 15 Ambient temperature (0C) ~~~Ambient temperature (OC) Fig. 4. The relationship between ambient temperature and the increment in colonic (A) and hypothalamic (B) temperature at the onset of panting in response to infra-red irradiation of the thorax and abdomen and of the head respectively. The open circles in B indicate the rise in hypothalamic temperature in those experiments in which panting could not be elicited. Results of twelve experiments on ten birds. when all of the other body temperatures were falling simultaneously. Respiratory frequency often declined rapidly, although panting did not cease until after a detectable fall in the deep body temperatures. However, there appeared to be no characteristic level of colonic or hypothalamic temperature at which panting always stopped, even in the same bird on different occasions; in some instances these temperatures were above, in others below, their levels at which panting had started in response to infra-red heat. Effects of irradiating the head with infra-red heat Although panting appeared to be principally under central control, it was not clear to what extent the temperature of the hypothalamus was responsible. When, therefore, it was found that the chicken's brain temperature could be profoundly influenced by irradiating the head with infra-red heat, it seemed that the method might be used to separate the influence on thermoregulatory mechanisms of brain temperature on the one hand and extra-cranial deep body temperature on the other. Fig. 5 illustrates the results of experiments performed at different ambient

10 350 S. A. RICHARDS temperatures and in which the body as a whole was protected from the influence of irradiation. It was found possible to increase the temperature of the brain, as measured in the hypothalamus, by up to 3.50 C by infra-red irradiation without causing distress to the birds. In the lower ambient temperatures (8-15 C), raising the temperature of the skin of the head and of the hypothalamus sometimes caused an increase in the temperature of the toe before any change could be detected in the colon, but often would not elicit panting, even when hypothalamic temperature reached between 43 and 440 C (Fig. 5A). Raising the brain temperature in the neutral or warm environments ( C) always caused vasodilatation in the toe before any alteration in respiratory frequency (Fig. 5B); there was, however, no panting until colonic temperature showed at least a small increase. In five experiments, heating the head at low or moderate ambient temperatures was followed initially by a decrease in colonic temperature by as much as 0.60 C. In one of these cases panting was elicited at a colonic temperature lower than the control value by 0.20 C, but at a time when it had started to rise and had reached C above the minimum level reached (Fig. 5B). The increment in hypothalamic temperature necessary to evoke panting increased with falling ambient temperature (Fig. 4B). The decline in respiratory frequency when infra-red heating was switched off occurred quickly when colonic temperature had been only slightly raised and paralleled the rapidly falling hypothalamic temperature. When colonic temperature had been substantially increased as a result of the radiant heat, however, the fall in respiratory frequency lagged behind that of hypothalamic temperature and panting ceased only after a considerable decline in the temperature of the colon (Fig. 5C). The colonic temperature sometimes continued to increase for a few minutes after the heating period and then declined towards the control level more slowly than did hypothalamic temperature, and panting often ceased at a higher level of colonic temperature and at a lower level of hypothalamic temperature than that at which it had started. Peripheral vasoconstriction, as judged by falling toe temperature, occurred immediately after the end of irradiation in ambient temperatures below about 160 C, but less rapidly and with more fluctuation in the warmer conditions. Experiments on anaesthetized birds In three experiments the hypothalamic and colonic temperatures were recorded in anaesthetized chickens in which (i) the body was heated by infra-red irradiation while the hypothalamus was maintained at about its control temperature, or at a temperature at which panting frequency

11 HYPOTHALAMIC TEMPERATURE AND PANTING 351 would normally be maximal, by manipulating the temperature of the cerebral blood supply; or (ii) the blood supply to the head was warmed with the rest of the body kept at normal temperature or at several degrees above normal. It was thus possible to determine the influence of brain temperature or extra-cranial deep body temperature on respiratory frequency in conditions where these could be more precisely controlled than was the U* 45 40_ Tb = E-2 E' 25 _- Tt: j;\ 20- i tt Tc 41 U 40 Tb t Tt B fi E250 Ta -E, C " 150_ f f Time (min) -f. Ta 200C Time (min) Time (min) Fig. 5. The effects, at different ambient temperatures, of irradiating the head with infra-red heat. The arrows indicate the beginning and end of irradiation. Representative experiments on three birds. Abbreviations as in Fig. 2. case with the chronic birds. Each experiment involved several repetitions of the selective heating procedures, but only the results from the first manipulations on each bird are presented in Fig. 6. It became increasingly difficult to elicit a respiratory frequency of 100 breaths/min despite the introduction of a tracheal cannula to diminish the resistance to breathing. This may have resulted from the long-term depressant effects of the anaesthesia, as well as from the gradual deterioration of the heparinized birds by slow haemorrhage. When hypothalamic temperature was held at 400 C, raising the colonic temperature up to 46 C was ineffective in producing panting. When, however, hypothalamic temperature was maintained at about 450 C, respiratory frequency was unchanged so long as colonic temperature

12 352 S. A. RICHARDS was lower than about C, but rose steeply when it was increased further (Fig. 6A). Conversely, when colonic temperature was held at 41 C, raising the hypothalamic temperature to near 450 C produced only a small acceleration in respiratory frequency, although at a colonic temperature of 46 C, hypothalamic temperature had only to be raised to about 41U70 C to initiate an increased frequency of breathing which reached a maximum value at a hypothalamic temperature of about C (Fig. 6B) A.' 0 E,,, 100 _ * 0 * 0 Thy 45 C 'p 0 sd 50 * Thy400C S Colonic temperature (0C) Tc 460C 150 B * 0 I,, -c~~~~~n 100 _. ap *ff t Tc 410C o Hypothalamic temperature (0C) Fig. 6. The influence of brain temperature and of extra-cranial deep body temperature on the respiratory frequency in anaesthetized chickens. A, the effect of raising the deep body temperature at two controlled brain temperatures. B, the effect of raising the brain temperature at two controlled colonic temperatures. Results from experiments on three birds. Effects on panting of repeated exposure to heat Despite the dominance of central control, it seemed possible that frequently repeated exposure of birds to a hot environment might lead to a progressive reduction in the time delay before panting started, perhaps as a result of the pairing of stimuli as occurs with a conditioned reflex. The results of 100 repetitions of the standard heat exposure test on four unoperated birds are presented in Table 2. Under the conditions of the experiment there was no evidence of any conditioning effect on

13 HYPOTHALAMIC TEMPERATURE AND PANTING 353 the panting response. Rather, in birds 1 and 4, there were significant (P < 0.05) increases in the time delay which were presumably the result of acclimation to the hot conditions. TABLE 2. The effect of frequently repeated exposure to 40/210 C on the delay-time before the onset of panting Mean ( ± s.e.) of six Mean ( ± S.E.) of test control exposures exposures Bird (min) (min) * 2 20* n.s n.s ** n.s. Not significant at the 5 % level of probability. *=P < 005; ** = P < DISCUSSION Bilateral lesions placed between the anterior commissure and optic chiasma of chickens have been shown to cause disturbances in the regulation of body temperature and food intake (Kanematsu, Kii, Sonoda & Kato, 1967; Lepkovsky, Snapir & Furuta, 1968). The unilateral damage due to implantation of hypothalamic temperature probes in the present experiments, however, had no such effects and it may be concluded that the responses of the operated birds were not significantly different from those of intact birds. Hypothalamic temperature has been studied in many mammalian species (Bligh, 1966), and large fluctuations associated with behavioural reactions, but which do not influence thermoregulatory mechanisms, have been described in the rat (Abrams & Hammel, 1964), cat (Forster & Ferguson, 1952; Baker & Hayward, 1967), dog (Fusco, 1963; Hayward, 1968), pig (Baldwin & Ingram, 1968) and monkey (Rawson & Hammel, 1963; Hayward & Baker, 1968). In birds, however, there is much less information and only Saint Paul & Aschoff (1968) appear to have investigated fluctuations of this type. The present results confirm their observations and show that transient increases in hypothalamic temperature of up to 0 6 C occur in relation to spontaneous arousal, as judged by the opening of the eyes after a period of sleep, or to auditory and visual stimuli. Increases of as much as 0 8 C, and lasting for several minutes, were seen during behavioural excitement when the level reached was occasionally higher than at the start of panting during subsequent heat exposure. The hypothalamic temperature of the chicken was always lower than

14 354 S. A. RICHARDS colonic temperature by an average of 0.90 C during control periods at C. A mechanism which might account for this discrepancy is that proposed by Baker & Hayward (1968) for the sheep, in which the temperature of the hypothalamus is also below the rectal temperature (Hemingway, Robinson, Hemingway & Wall, 1966). They showed that cool venous blood returning to the cavernous sinus from the nasal mucosa and skin of the head reduces the temperature of the cerebral arterial blood by exchange of heat with the warm blood in the carotid rete. On anatomical grounds, this explanation is feasible in the fowl where the cavernous sinus receives blood from the superficial ventral ophthalmic veins and itself surrounds the inter-carotid anastomosis from which the hypothalamus is directly supplied (Richards, 1967, 1968). The colonichypothalamic difference generally decreased during panting in the heat, which is opposite to the trend observed by Hunter & Adams (1966) in the cat. They argued that the widening visceral-cerebral gradient demonstrated that convective and evaporative heat losses from the upper respiratory tract had a direct cooling effect on the ventral diencephalon. The greater increase seen in the brain temperature of the chicken during heat exposure would support the idea that most of the evaporative cooling in birds occurs from the air sacs rather than from the upper respiratory tract (Salt, 1964; Richards, 1970). The precipitous fall in the hypothalamic temperature of panting birds abruptly introduced into a recovery environment need not conflict with this view since the local effects of cool air in the nasal cavities and on the skin of the head would then be far more advantageous in terms of heat exchange than the warm air in the thoracic and abdominal sacs. The results of exposing chickens to high or to gradually increasing ambient temperatures indicate that the peripheral temperature changes were ineffective in promoting polypnoea; despite considerable increases in the skin temperatures, as measured on the extremities and on the back, panting did not occur until both colonic and hypothalamic temperatures were elevated above the normal range. This is in contrast to the dog (Hales & Bligh, 1969) and the calf and sheep (Bligh, 1957, 1959) where the onset of panting occurs before a detectable rise in deep body temperature. Bligh (1962) also observed a sharp contrast between the almost immediate initiation of panting in the sheep and the min delay before sweating commenced in man under similar hot conditions. Although it was difficult in the exposure experiments to separate the influence of peripheral and central temperatures as causal factors in the control of polypnoea, the consistency of the delay before panting occurred in the fowl argues strongly for the dominance of central control. This was also the conclusion reached by Randall (1943) after experiments in which

15 HYPOTHALAMIC TEMPERATURE AND PANTING 355 he raised the skin temperature to 450 C but cooled the carotid blood by means of a metal collar and circulating water. When the collar was removed, respiratory frequency increased at once from 34 to 150 breaths/ min with cloacal temperature at C, a similar level to that at which panting started in the present experiments. Further evidence that panting in the fowl cannot be initiated by reflexes from the periphery alone was gained by the experiments using infra-red irradiation. When a large area of the feathered skin was irradiated, there was no panting until after the temperatures of the colon and hypothalamus had started to rise. These results again contrast with those of similar experiments on the ox (Findlay & Ingram, 1961) in which respiratory frequency always increased before any detectable rise in brain temperature. In the pig, however, Baldwin & Ingram (1968) found in most cases that panting was preceded by a rise in the temperature of the hypothalamus, a result which is more surprising in view of the almost naked skin of this species. Vasodilatation in the chicken's comb, as judged by large increases in its temperature, did occur as a result of irradiating the body in the absence of changes in hypothalamic or colonic temperature; when the head was heated, similar temperature changes were seen in the toe without increases in colonic temperature. Although in the latter case it was not possible to dissociate the influence of the temperature of the brain from that of the skin of the head, these observations suggest that vasodilatation in the extremities of the chicken may be under a fine reflex control, whereas thermal panting is controlled more coarsely by the central nervous system as a result of an increase in the temperature of the blood. The interaction between central and peripheral thermal stimulation, as well as that between hypothalamic and extra-cranial deep body temperature, is also illustrated by the infra-red irradiation experiments. When the colonic temperature was raised by heating the thorax and abdomen, panting was inhibited by a normal hypothalamic temperature, an effect which was enhanced by low ambient temperature acting at the periphery. A similar inhibitory role for hypothalamic temperature was also observed during the recovery phase of most experiments when panting generally ceased at a higher colonic temperature but lower hypothalamic temperature than their respective values at the onset of the response. These results were not unexpected in view of the known importance of hypothalamic temperature in controlling the mechanisms of heat loss (Bligh, 1966), but more surprising was the finding that when the head alone was irradiated, raising the temperature of both the surface skin and the brain, panting failed to occur in almost all cases until there was a detectable increase in colonic temperature. The implication here

16 356 S. A. RICHARDS would seem to be that the facilitating influence of an increase in some extra-cranial deep body temperature is necessary to promote panting, a possibility that was further supported by the experiments on anaesthetized birds. The extent to which inhibition of panting in the chicken may have resulted from peripheral as well as central effects is difficult to assess on the available evidence. That low ambient temperatures can depress or inhibit the respiratory effects of hypothalamic heating has been shown by Ingram & Whittow (1962) and Baldwin & Ingram (1968) in the ox and pig, and a similar effect seems at least possible in the fowl. In any event, the present results do not necessarily conflict with the hypothesis of Hammel, Jackson, Stolwijk, Hardy & Stromme (1963) that panting occurs at a 'set' temperature in the hypothalamus which, however, is not fixed, but is a variable function of afferent nervous influences, including those from the periphery and extra-hypothalamic deep body receptors. Recent work by Rautenberg (1969a, b) has produced direct evidence of the latter type of receptors in birds. Using the pigeon, he showed that temperature regulation could not be explained in terms of peripheral thermosensitivity alone, and that the essential central receptors were probably located within the spinal cord. Changes in spinal cord temperature, induced by perfusing thermodes chronically implanted into the peridural space, caused thermoregulatory responses without changes in hypothalamic temperature. The threshold temperature for panting elicited by this means declined with rising ambient temperature in much the same way as the hypothalamic and colonic temperature thresholds in the fowl subjected to infra-red heat. The experiments which attempted to train panting as a conditioned reflex were based on the hypothesis of Richet (1898) which suggested that in the conscious dog panting was the result of reflexes initiated by heating of the cutaneous receptors, whereas in the anaesthetized animal it was subject to central control and required a raised deep body temperature. Hammouda (1933) claimed that the so-called reflex panting was really a conditioned reflex; he demonstrated that repeated exposure of a dog in a heated box reduced the delay in time before panting began, and that the response would eventually occur in the absence of heat. Similar experiments failed to accelerate the onset of panting in the fowl and it is, in fact, difficult to see what could act as the conditioned stimulus in experiments involving a 20 min delay before the unconditioned response. The absence of any conditioning effects should probably be regarded merely as further evidence that a raised core temperature is the obligatory stimulus for panting in the chicken.

17 HYPOTHALAMIC TEMPERATURE AND PANTING 357 I should like to thank Mr P. Whittam for help in conducting many of the experiments. The work was supported by a grant from the Agricultural Research Council. REFERENCES ABRAMS, R. & HAMMEL, H. T. (1964). Hypothalamic temperature in unanaesthetized albino rats during feeding and sleeping. Am. J. Physiol. 206, BAKER, M. A. & HAYWARD, J. N. (1967). Carotid rete and brain temperature of cat. Nature, Lond. 216, BAKER, M. A. & HAYWARD, J. N. (1968). The influence of the nasal mucosa and the carotid rete upon hypothalamic temperature in sheep. J. Physiol. 198, BALDWIN, B. A. & INGRAM, D. L. (1968). The influence of hypothalamic temperature and ambient temperature on thermoregulatory mechanisms in the pig. J. Physiol. 198, BLIGH, J. (1957). The initiation of thermal polypnoea in the calf. J. Physiol. 136, BLIGH, J. (1959). The receptors concerned in the thermal stimulus to panting in sheep. J. Physiol. 146, BLIGH, J. (1962). A comparison between the onset of panting in sheep and of sweating in man under comparable conditions. J. Physiol. 160, 8P. BLIGH, J. (1966). The thermosensitivity of the hypothalamus and thermoregulation in mammals. Biol. Rev. 41, FINDLAY, J. D. & INGRAM, D. L. (1961). Brain temperature as a factor in the control of thermal polypnoea in the ox (Bos tauru8). J. Physiol. 155, FORSTER, R. E. & FERGUSON, T. B. (1952). Relationship between hypothalamic temperature and thermoregulatory effectors in unanaesthetized cat. Am. J. Physiol. 169, Fusco, M. M. (1963). Temperature patterns throughout the hypothalamus in the resting dog. In Temperature, its Measurement and Control in Science and Industry, vol. 3, ed. HARDY, J. D., pp New York: Reinhold. HALES, J. R. S. & BLIGH, J. (1969). Respiratory responses of the conscious dog to severe heat stress. Experientia 25, HAMMEL, H. T., JACKSON, D. C., STOLWIJK, J. A. J., HARDY, J. D. & STROMME, S. B. (1963). Temperature regulation by hypothalamic proportional control with an adjustable set point. J. apple. Physiol. 18, HAMMOUDA, M. (1933). The central and reflex mechanism of panting. J. Physiol. 77, HAYWARD, J. N. (1968). Brain temperature regulation during sleep and arousal in the dog. Expl Neurol. 21, HAYWARD, J. N. & BAKER, M. A. (1968). Role of cerebral arterial blood in the regulation of brain temperature in the monkey. Am. J. Physiol. 215, HEMINGWAY, A., ROBINSON, R., HEMINGWAY, L. & WALL, J. (1966). Cutaneous and brain temperatures related to respiratory metabolism of the sheep. J. apple. Physiol. 21, HUNTER, W. S. & ADAMS, T. (1966). Respiratory heat exchange influences on diencephalic temperature in the cat. J. apple. Physiol. 21, INGRAM, D. L. & WHITTOW, G. C. (1962). The effect of heating the hypothalamus on respiration in the ox (Bos taurus). J. Physiol. 163, KAHL, M. P. (1963). Thermoregulation in the wood-stork, with special reference to the role of the legs. Physiol. Zool. 36, PH Y 2II

18 358 S. A. RICHARDS KANEMATSU, S., Kii, M., SONODA, T. & KATO, Y. (1967). Effect of hypothalamic lesions on body temperature in the chicken. Jap. J. vet. Sci. 29, LEPKOVSKY, S., SNAPIR, N. & FURUTA, F. (1968). Temperature regulation and appetitive behaviour in chicken with hypothalamic lesions. Physiol. & Behav. 3, RANDALL, W. C. (1943). Factors influencing the temperature regulation of birds. Am. J. Phy8iol. 139, RAUTENBERG, W. (1969a). Untersuchungen zur Temperaturregulation warme- und kalteakklimatisierter Tauben. Z. vergl. Phy8iol. 62, RAUTENBERG, W. (1969b). Die Bedeutung der zentralnervbsen Thermosensitivitit fur die Temperaturregulation der Taube. Z. vergl. Physiol. 62, RAWSON, R. 0. & HAMMEL, H. T. (1963). Hypothalamic and tympanic membrane temperatures in rhesus monkeys. Fedn Proc. 22, 283. RICHARDS, S. A. (1967). The anatomy of the arteries of the head in the domestic fowl. J. Zool. 152, RICHARDS, S. A. (1968). The anatomy of the veins of the head in the domestic fowl. J. Zool. 154, RICHARDS, S. A. (1970). The biology and comparative physiology of thermal panting. Biol. Rev. 45, RICHET, C. (1898). Dictionnaire de Physiologie, vol. 3, p Paris: Germer Bailliere. SAINT PAUL, U. VON & AsCHOFF, J. (1968). Gehirntemperaturen bei HUhnern. Pflugers Arch. ge8. Physiol. 301, SALT, G. W. (1964). Respiratory evaporation in birds. Biol. Rev. 39, STEEN, I. & STEEN, J. B. (1965). The importance of the legs in the thermoregulation of birds. Acta phy8iol. 8cand. 63, VAN TIENHOVEN, A. & JUHASZ, L. P. (1962). The chicken telencephalon, diencephalon and mesencephalon in stereotaxic coordinates. J. comp. Neurol. 118, WAITES, G. M. H. (1962). The effect of heating the scrotum of the ram on respiration and body temperature. Q. JZ exp. Physiol. 47,