metabolic rate of camels: effect of body …...metabolic rate of camels 343 table i. oxygen...

Metabolic rate of camels: effect of body temperature and dehydration’

K. SCHMIDT-NIELSEN,2 E. C. CRAWFORD, JR.,3

A. E. NEWSOME, K. S. RAWSON, AND

H. T. HAMMEL

Department of Zoology, Duke University, Durham, North Carolina

SCHMIDT-NIELSEN, K., E. C. CRAWFORD, JR., A. E. NEWSOME, K. S. RAWSON, AND H. T. HAMMEL. Metabolic rate of camels: effect of body temperature and dehydration. Am. J. Physiol. 2 I ~(2) :

341-346. 19670 -The effect on metabolic rate (oxygen consumption) of changes in body temperature and state of dehydration was studied in four camels. The data were subjected to multiple-regression analysis. The metabolic rate increased with rising body temperature as expressed by a mean Qro of 2.06. The metabolic rate decreased with increasing dehydration, reaching about 77 yO of the original value at 77 yO of the initial hydrated body weight. The rate of respiration (rate of breathing) increased with rising body temperature with a mean Qlo of 10.86. Th e rate of respiration was decreased to 640/ of the initial value when dehydration had reduced the body weight to 77 yO of the initial weight.

oxygen consumption; multiple-regression analysis; Qro; respiration rate; breathing rate

T HE BODY TEMPERATURE of camels may vary considerably. In the absence of heat stress the daily fluctua-

tions are about 2 C, but a dehydrated camel in a hot environment may display fluctuations in excess of 6 C. Under these circumstances the early morning temperature may fall to below 35 C, and during the maximum heat load of the desert day the temperature may reach 41 C. These variations in body temperature are of significance in the water economy of the camel as discussed in previous publications (9, I 2).

Received for publication 13 May 1966. l This study was supported by National Science Foundation

Grant G I 761 g and by National Institutes of Health Grant HE 02228.

2 Recipient of National Institutes of Health Research Career Award 1-K6-GM-21,522 from the National Institute of General Medical Sciences.

3 Present address: Dept. of Zoology, University of Kentucky, Lexington, Ky.

4 Present address: Dept. of Zoology, University of Adelaide, Adelaide, S.A., Australia.

5 Present address: Dept. of Biology, Swarthmore College, Swarthmore, Pa.

6 Present address: John B. Pierce Foundation Laboratory and Yale University, New Haven, Conn.

Since the highest body temperatures occur when the external heat load is maximal, it is of interest to know whether or not the increase in body temperature causes increased metabolic heat production, for this would add substantially to the heat load. It is well known that the metabolic rate of mammals in general increases with increasing body temperature (e.g., fever), and decreases with decreasing temperature (e.g., hypothermia and hibernation). If the metabolic heat production of camels increases to the same extent as in other mammals, i.e., with a Qlo of about 2 or 3, a 6 C increase in body temperature should cause an increase in heat production of some 50-100 %.

It was the purpose of this study to establish the effect of body temperature on the metabolic rate of camels. Since extreme body temperatures occur only in the dehydrated animal, it was desirable to determine separately the effects of temperature and the degree of dehydration on the metabolic rate. The body temperature fluctuations depend on the degree of dehydration, but the separate contributions of these two interrelated factors can be evaluated by a multiple-regression analysis. We found that an increased body temperature causes an increase in metabolic rate similar to that in other mammals, and that dehydration leads to a decrease in metabolic rate.

MATERIALS AND METHODS

Animal material. These studies were carried out in central Australia (Alice Springs, N.T., 23.4OS) on the one- humped camel (Camelus dromedarius). Camels were orig- inally introduced into Australia from Afghanistan and have since become well established as a feral species in the central deserts. We used four animals, two that were semidomesticated and two that were caught wild (Table I). All four animals were readily trained to submit to the experimental procedures, which consisted of weighing, temperature measurements, and determinations of oxygen consumption. They were fed on commercially avail- able hay and dried alfalfa, and were kept in the open, either in pens or tethered near the laboratory.

Weighing. The animals were weighed suspended in a canvas sling, using a temperature-compensated Chatillon dynamometer which permitted readings to 0.5 kg. Sur-

341

by 10.220.32.246 on May 28, 2017

http://ajplegacy.physiology.org/D

ownloaded from

http://ajplegacy.physiology.org/

342 SCHMIDT-NIELSEN ET AL.

face areas, when used, were estimated from Meeh’s equation, S(m2) = 0.1 . W(kg)0.67 (I).

Body temperature was measured before, during, and after all metabolic determinations as deep rectal or vaginal temperature, using either a deeply inserted (IO cm or more) thermocouple probe or a mercury thermometer. All instruments were calibrated to an accuracy of better than o. I C. To diminish the reaction to vaginal measurements a before a

small amount of procaine ointment was applied probe was inserted. Vaginal measurements were

preferred because, during advanced states of dehydration, the rectum was frequently filled with fecal material which impeded the insertion of probes to an adequate depth.

Oxygen consumftion rates were determined in an open system with the use of a Beckman model C-2 paramag- netic oxygen analyzer. The animal was fitted with a molded plastic mask, loosely sealed with a rubber dam. Loss of expiratory gas was prevented by operating con- tinuously with a slight negative pressure in the mask. The air was pulled through the system at a rate of about IOO

liters/min, accurately determined with a calibrated dry- gas meter (=t: I %). A constant fraction of the air stream was collected in Douglas bags for consecutive ro-min periods. Analysis of the mixed gases in one Douglas bag was carried out while the subsequent collection took place in the next bag. Previous calibrations and use of the same instrumentation indicated that the accuracy was better than ri= 3 %. Further calibration was not carried out in the field, except for the determination of the oxygen consumption rate in one of the expedition mern- bers whose metabolic rate was known to remain ex- tremely constant during numerous earlier tests. All metabolic rates are given as oxygen consumption STPD. Although insignificant, a correction was made for the volume change due to an assumed RQ of 0.85.

Xes$vkatory rates (breathing rates) were determined by visual observation and the use of a stop watch, when necessary counting for several consecutive minutes. No differences were noted between observations made with and without the metabolic mask, and both were used.

Computer analysis. Of the 342 determinations of metabolic rate only 332 were used because a few observations were incomplete. The covariant effects of body temperature and the state of dehydration on the metabolic rate were estimated by multiple-regression analysis, using an IBM 7072 computer. The computed equations were of the general form y = b ax GZ, which is well justified for the effect of body temperature (x) on oxygen consumption (JJ), but is an arbitrary choice for the effect of dehydration (c). Actually, the computer calculated the coefficients of the equation logy = b + x log a + z log c, and determined the standard errors and F values.

RESULTS

Metabolic rate. The oxygen consumption was determined on four camels in a state of normal hydration and in various degrees of dehydration. Due to limited time

I I I I I I I

mo- No. I ” PETE” _

,--:-- l

l 8

l e 00 .

=2.09

I 2oo 35

I I I I I I 36 37 38 39 40 41 42

RECTAL TEMPERATURE, “C

FIG. I. Total oxygen consumption in camel I in relation to rectal temperature. All observations plotted, regression line calculated by method of least squares. In this and the following figures the dependent variable is plotted on a logarithmic scale.

we chose to obtain more complete observations on two camels (one feral and one domesticated), and supplement this with a smaller number of observations on two addi- tional animals. The observations on each camel num- bered 158, 96, 39, and 39, respectively, giving a total of

3420 The data from the one camel (Pete) on which we had

158 observations are plotted in Fig. I. The oxygen consumption increased with increasing body temperature corresponding to a Qlo of 2.09. These determinations were made with the camel sitting quietly on the ground in the open, outside the laboratory building. The camel showed no obvious struggling, but was at times slightly restless which may explain some of the highest values. However, the data include all observations made, both in the normally hydrated animal and in various states of dehydration, and the variations therefore include the effect of dehydration. The effect of this variable was estimated separately, as will be described below.

MuZti@e rqression analyses. Since the data presented in Fig. I include the effect of dehydration as well as of body temperature, the entire set of observations was subjected to a multiple-regression analysis to separate these covar- iants. The computer determined the constants in equations of the form:

log 02 consumption = CO + G (% JW -I- C~(TB)

with the constants Co, Cl, and C2 for each camel listed in Table I.

In these equations % IW means percent of initial body weight, which indicates the degree of dehydration of the animal, expressed as percent of the initial, or hydrated, body weight, and TB is the body temperature in degrees C .

The equations can be used for the calculation of a

by 10.220.32.246 on May 28, 2017


ownloaded from


METABOLIC RATE OF CAMELS 343

TABLE I. Oxygen consumption in four camels in relation to degree of dehydration and body temperature

Camel No. co

I 3 Feral -I .gp6

2 9 Domesticated -0.7872

3 9 Domesticated -0.3745

4 3 Feral -2.2614

All camels -I .8756

cl =t= SE

+0.002571 rto.ooo6gm +0.0005166 rt0.0006621 -0.0001338 rto.001246 +0.006067 ~to.001247 +0.005103 =to .001366

F Value for Cl

F(3,‘55) = ‘34963

F(3,91) = 0.6089

F(3,36) = 0.01153

F0,36) = 23.6724

F(3,327) = 13-9571

cs =t SE

+oqjg8 rto.003137

-I-0.01737 So .002486 +o.oogg66 rto.01236 +0.03516

~to.01178

+0.03130

ho .006505

F Value for Ca

F(3955) = 131.5765

F(s,g=) = 48.8036

F(3,36) = o-6499

F(3,36) = 8.9076

F(3,327) = 23~583

Constants are given in the equation log 02 cons = Co + Ci(% IW) + C,(T&, where O2 cons = oxygen consumption; 70 1~ = dehydration expressed as percent of initial weight; and TB = body temperature.

TABLE 2. Estimated rates of 02 consumption (0, cons.) of camels, computed from equations given in Table I, for fuZZy hydrated animals at the indicated standard body temperatures (Std. TB)

Camel No. a

Estim. Init.

Bodybwt7 kg N c Std* F OC

Total 02 Cons., ml/min

e

02 Cons., ml/ min per kg

f

02 Cons., ml/ min per m2

g Qlo

h

I 295 -3 158 38.20 489 I .656 110.1 2.29

2 559 -8 94 38.12 844 I .508 124.0 1 -49 3 552.1 39 38 005 981 = l 777 145 05 (I .26)* 4 266.2 39 37 l 9= 477 I*792 115.0 2.22

All 407 l o 330 38.13 673 I l 654 122.3 2.06

Last column indicates the effect of change in body temperature as expressed by the Q 10. For statistical significance, refer to Table I. * Not significant.

ccnormal” oxygen consumption in the four camels, elim- inating the effect of dehydration and referring the value to a ‘cstandard” body temperature for which we used the mean of the body temperatures at which the observations were made. The result of this computation of the oxygen consumption in resting, fully hydrated camels is given in Table 2. Since the body weights of the camels differed considerably, the information is given not only as total oxygen consumption, but also as oxygen consumption per kilogram body weight and per estimated body surface area.

camels was dehydrated was to 77 % of the initial body weight (fully hydrated weight). Not all camels were exposed to equally severe, or the same mumber of, dehydration periods. The slope of the regression line is therefore determined primarily by those two camels on which most data were obtained, I and 2.

The regression line indicates that when dehydration had resulted in a decrease in body weight to 77 % of the original the metabolic rate had decreased to 76.5 70 of that in the fully hydrated animal. The time required to reach this degree of dehydration does, of course, depend on heat load, exposure to sun, work, etc., and on individ- ual differences between the animals. In these particular experiments (central Australia in February) the rate of water loss was such that 23 % of the body weight was lost after about 12 days of feeding on dry feed without any watering. In a hotter desert, such as the Sahara, the rate of dehydration would be faster in summer, while in win- ter the rate would be much slower.

Temperature eject. The regression lines for the relationship between oxygen consumption and body temperature, including all data for all four camels are given in Fig. 2. Since the animals had different body weights, these results have been calculated per kilogram body weight to make them more directly comparable. It will be seen that there are no gross differences between the animals, which all showed an obvious temperature effect on the metabolic rate.

The simplest way to express the temperature effect is by the Q10, as listed in Table 2, column h. It should be noted that the over-all Qlo for all camels is weighted in favor of camel I because this animal was represented by more determinations than the others. The Qlo is appar- ently lower in the two large domesticated animals (2 and 3) than in th e t wo young feral camels (I and 4), but the material is insufficient to conclude whether any real difference existed.

Eject of dehydration. The equation for the effect of dehydration on oxygen consumption, when the data from all camels are combined, gives the regression line shown in Fig. 3. The maximum degree to which any of these

Respiration rate. It has previously been established that the main mechanism for heat dissipation in the camel is sweating, rather than panting (I 0). However, there still is a noticeable increase in respiratory rate with increasing

body temperature. The increase is from about 7 cycles/ min at a body temperature of 35 C to about 30 cycles/

min at 41 C body temperature (see Fig. 4). Part of this increase should be due to the increase in oxygen consumption at higher body temperature, but if this factor

alone is considered, the respiratory rate should only increase to about I I cycles/min. Since tidal volumes and carbon dioxide in the blood were not determined, it re- mains uncertain to what extent the animal hyperventi-

by 10.220.32.246 on May 28, 2017


ownloaded from



3.0 , I I I I I I

_ -- 35 36 37 38 39 40 41

BODY TEMPERATURE, “C

FIG. 2. Oxygen consumption per kilogram body weight in relation to body temperature (rectal or vaginal) in four camels. Regression lines refer to normally hydrated animals as derived from computer analysis. Number of observations, camel 1-158, z-96, J-39, g-39; total-332 observations.

lated and what fraction of the total heat dissipation could be accounted for by respiratory means.

The multiple regression analyses for the information on respiratory rate gave the following equation for all camels combined :

log resp. rate = -3.6140 + 0.008502 (% IW) + 0.1036 (TB)

The regression line in Fig. 5 shows the relationship between respiratory rate and dehydration (as expressed by the decrease in body weight), with all figures referred to a standard body temperature of 38. I 3 C.

At a dehydration involving 23 % loss of body weight, the respiratory rate would be decreased by 36 %. This decrease is somewhat larger than the decrease in oxygen consumption (23.5 %) which results from the same degree of dehydration (see Fig. 3).

DISCUSSION

This investigation was undertaken to study the influ- ence of degree of dehydration and of body temperature on the metabolic rate of camels. It has earlier been established that camels can tolerate an exceptional degree of dehydration. While other mammals, if exposed to heat, are likely to succumb when the water loss corresponds to some I 2-15 % of body wt, camels have tolerated water losses exceeding 25 % of body wt without apparent ill effect (9, I I).

The body temperature of the camel can fluctuate daily by some 6 or 7 C, from about 34 to 41 C. However, such extreme variations occur only in the severely dehydrated animal when exposed to high environmental heat loads (I 2). The rise in body temperature during heat load is advantageous to the water balance of the camel in two ways. First, the increase in body temperature dur-

ing the day constitutes a storage of heat which can be dis- sipated at night without expenditure of water. If the body temperature were to be kept constant during the day, an equivalent amount of water would have to be evaporated. Second, as the body temperature is allowed to rise, the temperature gradient (and heat flow) from the hot environment to the cooler body surface is correspondingly reduced, and thus the external heat load and the use of water is reduced. Although it has not been possible to establish the exact amount, it has been estimated that the increase in body temperature cuts the amount of water needed to dissipate heat gain from the environment to about one-half (I 2).

It is well known that an increase in the body temperature of mammals leads to an increase in metabolic rate (heat production). The metabolic heat production of the camel also increases at high body temperature, and the internal heat production thus adds to the heat load when the external heat load is at its maximum. The reduction in external heat load achieved by permitting a rise in body temperature is thus partly offset by the increase in metabolic heat load.

The effect of a temperature increase on a physiological process is most conveniently expressed as the Q10, i.e., the increase that occurs with a temperature rise of IO C. The temperature fluctuations in mammals are usually so small that the Qro cannot be established with precision, but it is commonly stated that Qlo for man is about 2 or 2.5. This information is based on the work of DuBois, who reported an average Q-JO of 2.3 for fever in man (4). More recently a Qlo of 2. I was found in a human subject where an intracerebral lesion had destroyed the temperature regulation (2). Qro for dogs has been reported to be 2.3 between 38 and 23 C (7), and for rats 2.01 between 38 and 18 C (5). Th e reduction in metabolic rate which occurs in hibernating mammals can also be used for a calculation of Q 10, which usually is in the order of 2-4 (8). However, it could be argued that this is a special

I I I I I I I

300 I I I I I

100 95 90 DEHYDRATION, Oh*& INl:AL W:lGHT 70

FIG. 3. Total oxygen consumption in relation to dehydration in four camels. Computer analysis based on 330 observations, regression line estimated for a standard body temperature of 38.13 C.

by 10.220.32.246 on May 28, 2017


ownloaded from


METABOLIC RATE OF CAMELS

1 I I I I I I

35 36 37 38 39 40 41

BODY TEMPERATURE, “C

FIG. 4. Respiratory rates in four camels in relation to body temperature. Regression lines refer to normally hydrated animals as derived from computer analysis. Number of observations, camel I- I 48, z-94, 3-37, g-38; total-3 I 7 observations.

case because of the profound physiological changes in the hibernating animal.

If we now return to the Qlo for camels and examine the data for the animal where we have the most observations (camel I) we find that the Qro is 2.09. However, it must be noted that these data include various states of dehydration. The elimination of this variable by multiple-regression analysis gave a Q1o of 2.29 for the same camel when all data were referred back to the fully hydrated animal (see Table 2). Q$s for two of the other camels were lower than 2, but the data are not sufficient to establish these values as significantly different. This is due to the small number of observations, their varia- bility, and the narrow temperature range within which observations were made. It is worth noting that when all the data are combined for all four animals, the result is

Q 10 = 2.06. Therefore, within the reliability of our data, we cannot establish any difference in Q1o between camels and other mammals.

If the metabolic rates of camels were insensitive to the temperature rise (which could hardly be expected), their Qlo should, of course, be equal to I. Since the Qro actually was about 2, we must conclude that the camel is not unique in this respect, and that a modified Qro is not an adaptation which has been employed by the camel in its adaptation to desert life (I 0).

In the earlier study of the body temperature and its relation to the water economy of camels (I 2) it was not possible to give a complete account of the heat balance because the metabolic heat production in the dehydrated camel was not known. It was suggested that the metabolic rate might decrease in the dehydrated camel, but since this had not been demonstrated at that time, it was not included in the estimation of the heat balance. As a result, only an indirect estimate of the change in heat gain from the environment could be made for the hydrated and dehydrated camel.

345

Our present results show that there is indeed a reduction in metabolic rate due to dehydration when the observations are referred back to the same body temperature (Fig. 3). 0 ur camels, when dehydrated to 77 % of the initial body weight, had metabolic rates reduced to about 77%. (This does not mean that the dehydrated camel always has a lower metabolic rate than if it were fully hydrated, because during the day it will have a high body temperature which, due to the Qro effect, gives a rise in metabolic rate.)

It is interesting that the metabolic rate decreases as water is lost, although there should be no decrease in dry tissue weight, protein mass, etc. The camel continues to eat as water is lost (until seriously dehydrated), and when water is offered it will drink until its body weight has been restored to the initial value.

Re.s/G-ation rates. The respiratory rate of the camel increased about threefold as its body temperature increased from 35 to 41 C (Fig. 4). The change was gradual, and even at the highest rates there was no open-mouthed breathing similar to the panting of dogs.

In dogs when panting begins there is a sudden change in frequency from some 30 to 40 cycles/min to some 300- 400 cycles/min. It has been suggested that dogs pant at the resonant frequency of the respiratory system, thus reducing the muscular work (and the heat production) required for the high respiratory frequencies (3). The gradual change in the respiratory frequency in the camel indicates that a similar principle of a resonant frequency is not involved in the increased respiratory rates at high body temperature.

An increased ventilation may of course be expected with the increased oxygen consumption at high body temperature. However, the Qro for the increase in metabolic rate is about 2, whereas the Qro for the increase in respiration is about IO (mean for all camels Qro = 10.86). A decrease in tidal volume could possibly take

1

l E

I I I I ,

40- ALL CAMELS - \

% 30-

STD. TB ~38.13 C

E 0

a

r 20-

a E tn

fr” I I I I I

100 95 90 85 80 75 70 DEHYDRATION, O/o OF INITIAL WEIGHT

FIG. 5. Mean respiratory rate in relation to dehydration in four camels. Estimated from equations given in text for a standard body temperature of 38. I 3 C.

by 10.220.32.246 on May 28, 2017


ownloaded from



care of this discrepancy, but tidal volume was not ade- quately determined in our camels. If the tidal volume remained unchanged, there would be a change in blood COB, but again, we have no information on this point. We must therefore consider the respiratory frequency as a separate phenomenon. In this connection it is interesting to note observations made by Hardy et al. (6) on discharge rates from hypothalamic tissue that was cooled or heated. About one cell of five rapidly increased its discharge rate with temperature with a Qlo = 5-10, whereas a majority of the cells showed minimal tempera-

REFERENCES

I. BENEDICT, F. G. Die Oberflachenbestimmung verschiedener Tiergattungen. Ergeb. Physiol. Exftl. Pathol. 36: 300-346, 1934.

2. COOPER, K. E., AND D. N. Ross. Hypothermia in Surgical Practice. London: Cassell, I 960, p. 8.

3. CRAWFORD, E. C., JR. Mechanical aspects of panting in dogs. J. k&$d. Physiol. 17: 249-251, 1962.

4. DUBOIS, E. F. The basal metabolism in fever. J. Am. Med. Assoc. 77: 352-355, 1921.

5. FUHRMAN, F. A., G. J. FUHRMAN, D. A. FARR, AND J. H. FAIL. Relationship between tissue respiration and total metabolic rate in hypo- and normothermic rats. Am. J. Physiol. 201:231-234, 1961.

6. HARDY, J. D., H. T. HAMMEL, AND T. NAKAYAMA. Observa- tions on the physiological thermostat in homoiotherms. Science I 36 : 326, 1962.

7. HORVATH, S. M., AND G. B. SPURR. Effects of hypothermia on general metabolism. In: The Physiology of Induced Hypo-

ture sensitivity with Qlo = I - 2. This analogy between a high Qlo for respiratory rate and for certain hypothalamic cells points to a field for future investigations that should prove most interesting in clarifying temperature responses in the central nervous system.

The respiratory rate did decrease with dehydration of the camels, as shown by multiple-regression analyses, which eliminated the temperature effect from the computation. This decrease, which was about 36 % for a 23 % loss of body weight, was somewhat greater than the decrease in oxygen consumption (23.5 %).

thermia. Washington, D. C. : Natl. Acad. Sci.-Natl. Res. Council, 1956, publ. 45 I, p. 8-25.

8. MEYER, M. P., AND P. MORRISON. Tissue respiration and hibernation in the thirteen-lined ground squirrel, Spermophilus tridecemlineatus. Bull. Museum Camp. Zool. Harvard Coil. I 24.: 405-420, 1960.

g. SCHMIDT-NIELSEN, K. Desert Animals. Physiological Problems of Heat and Water. Oxford : Clasendon, 1964, p. I-277.

IO. SCHMIDT-NIELSEN, K., E. C. CRAWFORD, A. E. NEWSOME, AND K. S. RAWSON. The metabolic rate of camels. Federation PYOC. 22: 176, 1963.

II. SCHMIDT-NIELSEN, B., K. SCHMIDT-NIELSEN, T. R. HOUPT, AND S. A. JARNUM. Water balance of the camel. Am. J. Physiol.

185: 185-194, 1956. 12. SCHMIDT-NIELSEN, K., B. SCHMIDT-NIELSEN, S. A. JARNUM, AND

T. R. HOUPT. Body temperature of the camel and its relation to water economy. Am. J. Physiol., 188 : I 03-1 I 2, 1957.

by 10.220.32.246 on May 28, 2017


ownloaded from


metabolic rate of camels: effect of body …...metabolic rate of camels 343 table i. oxygen...

Documents