Content uploaded by Robert C Schroter
Author content
All content in this area was uploaded by Robert C Schroter on May 21, 2015
Content may be subject to copyright.
Desaturation of Exhaled Air in Camels
Author(s): K. Schmidt-Nielsen, R. C. Schroter, A. Shkolnik
Source:
Proceedings of the Royal Society of London. Series B, Biological Sciences,
Vol. 211, No.
1184 (Mar. 11, 1981), pp. 305-319
Published by: The Royal Society
Stable URL: http://www.jstor.org/stable/35545 .
Accessed: 12/01/2011 08:20
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .
http://www.jstor.org/action/showPublisher?publisherCode=rsl. .
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
The Royal Society is collaborating with JSTOR to digitize, preserve and extend access to Proceedings of the
Royal Society of London. Series B, Biological Sciences.
http://www.jstor.org
Proc. R. Soc. Lond. B 211, 305-319 (1981)
Printed in Great Britain
Desaturation of exhaled air in camels
BY K. SCHMIDT-NIELSENt, R. C. SCHROTERt AND A. SHKOLNIK?
t Department of Zoology, Duke University, Durham, N.C. 27706, U.S.A.
t Physiological Flow Studies Unit, Imperial College, London SW7 2AZ, U.K.
? Department of Zoology, Tel-Aviv University, Tel-Aviv, Israel
(Communicated by D. R. Wilkie, F.R.S. - Received 4 June 1980 -
Revised 4 August 1980)
We have found that camels can reduce the water loss due to evaporation
from the respiratory tract in two ways: (1) by decreasing the temperature
of the exhaled air and (2) by removal of water vapour from this air,
resulting in the exhalation of air at less than 100 % relative humidity
(r.h.). Camels were kept under desert conditions and deprived of drinking
water. In the daytime the exhaled air was at or near body core tem-
perature, while in the cooler night exhaled air was at or near ambient air
temperature. In the daytime the exhaled air was fully saturated, but at
night its humidity might fall to approximately 75 % r.h. The combina-
tion of cooling and desaturation can provide a saving of water of 60 %
relative to exhalation of saturated air at body temperature.
The mechanism responsible for cooling of the exhaled air is a simple
heat exchange between the respiratory air and the surfaces of the nasal
passageways. On inhalation these surfaces are cooled by the air passing
over them, and on exhalation heat from the exhaled air is given off to
these cooler surfaces. The mechanism responsible for desaturation of the
air appears to depend on the hygroscopic properties of the nasal surfaces
when the camel is dehydrated. The surfaces give off water vapour during
inhalation and take up water from the respiratory air during exhalation.
We have used a simple mechanical model to demonstrate the effectiveness
of this mechanism.
INTRODUCTION
In the preceding paper (Schmidt-Nielsen et al. I981) we reported that the
respiratory water loss was unexpectedly low in camels that were subjected to
severe water deprivation, and in fact so low that the results were incompatible
with the view that exhaled air is saturated with water vapour. We have previously
reported that the respiratory air of camels and several other large mammals may
be exhaled at or near ambient rather than body core temperature (Langman et al.
1978, I979). This cooling of the exhaled air results in substantial savings in water,
and we now report that camels can further reduce the respiratory water loss by
extracting water vapour from the exhaled air, resulting in the exhalation of air at
less than 100 % relative humidity (r.h.).
[ 305 ]
K. Schmidt-Nielsen, R. C. Schroter and A. Shkolnik
METHODS
We studied two mature camels (Camelus dromedarius) that were accustomed to
human handling. Both animals had been maintained in a healthy and well
nourished condition and had been watered regularly. One was a 6 year old female
that weighed 575 kg when normally hydrated; the other was a ca. 12 year old
female that normally weighed 450 kg. The animals were transported by truck
to kibbutz Qalia at the north end of the Dead Sea, where we conducted our study
in July 1979.
The ambient dry bulb temperature varied from a maximum of 35 to 40 ?C in
the early afternoon to a minimum around 26 ?C in the early morning at sunrise.
The ambient relative humidity was around 25% in the afternoon, increasing
to over 50 % at night, while the absolute humidity remained fairly constant
throughout the study at approximately 0.012 kg of H2O per kilogram of dry air.
The animals were confined to a corral of about 5 m x 10 m with virtually no
shade during the day. After becoming familiar with their surroundings they were
denied water but were fed dry hay and small amounts of fresh green dates. Both
camels were subjected to 16 days of continuous dehydration (except that 51 of
water were given to each animal on day 12). At the end of the 16 days of dehydra-
tion each animal was allowed water ad libitum. They consumed 92 and 1071
respectively (16 and 24 % respectively of their normal body masses).
The animals withstood the dehydration without apparent signs of undue stress.
After a few days without water they ate very little hay but took green dates when
these were offered. They urinated sparingly and defecated extremely dry faeces.
The camels were free to move about, but were brought down to the resting
('sitting') position when measurements were taken. During the daytime, especially
as dehydration progressed, they preferred to remain in the resting position
throughout the hot day.
Rectal temperature was measured with a copper-constantan thermocouple
probe inserted at least 15 cm. Temperature was read on a Wescor TH 60 digital
thermometer with the reading noted after the signal remained stable for more than
1 min. In dehydrated camels the rectum tends to become impacted with dry
faeces; this required use of petroleum jelly to facilitate insertion of the probe and
often several minutes elapsed before the reading became stable, although the
nominal time constant of the system was less than 1 s.
Respired air temperatures were measured with welded copper-constantan
thermocouples made from wire 0.125 mm in diameter. The small mass of the
welded thermocouple and the absence of solder yielded a low time constant (90 %
response in less than 0.4 s). The thermocouple was connected to a Wescor TH 50
thermometer with analogue output that was recorded on a four-channel rectilinear
Physiograph. The thermocouple was mounted in a plastic T-tube with an inside
diameter of 1.4 cm. One end of the long arm of the T-tube was inserted into one
306
Desaturation of exhaled air in camels
of the nares, where it remained, kept in place by the nasal flap. The animals
showed very little reaction to the T-tube and could breathe freely through and
around the tube.
The relative humidity of the respired air was measured with a solid-state un-
coated humidity sensor (Vaisala Oy, Helsinki) and recorded on the physiograph.
The sensor is a thin plate that was positioned in the centre of the airstream
through the T-tube, orientated parallel to the flow. To protect the sensor from
damage or deposition of nasal mucous, it was housed in a fine-mesh wire cage with
large free area to allow circulation of air. Air humidity is measured as the capacit-
ance of a thin polymer film, about 1 pm thick, positioned between two thin gold
electrodes, mounted on a small glass wafer (4 mm x 4 mm x 0.2 mm). The upper
gold electrode is an extremely thin, water-permeable, vacuum-deposited gold film,
about 10-2
~Lm thick. (The thickness of this electrode is a compromise between that
giving short response time and that giving low ohmic resistance loss in the elec-
trode.) The capacitance changes in proportion to the relative humidity of the air,
and the temperature dependence is negligible. In other words, the sensor measures
relative humidity without significant temperature dependence (the manufacturers
quote a value of 0.05 %
?C-1).
The humidity sensor responds very rapidly to changes in relative humidity
(90 % response in less than 0.5 s), provided that the air is not fully saturated.
When the sensor is exposed to air at 100 % r.h. its plastic film becomes saturated
and the output overloaded. The time taken to remove excess water when the
sensor is repositioned in a low humidity depends on the time of exposure to
saturated air. The exposure to saturated air encountered in our study was short,
and the removal time was about 0.5 s or less. The resulting overall time taken to
respond to respiratory air humidity (less than 1 s) was considerably less than the
duration of an inspiration or expiration (> 5 s in the dehydrated animal).
The humidity sensor responded very quickly to the change in humidity en-
countered when switching from inhalation to exhalation, with a time lag of about
s behind the temperature and lung volume signals. However, when the humidity
changed from 100 % to the low inspiratory level, there was a lag time of a little
less than 0.5 s (see figures 1 and 2).
The thermocouple tracings show a temperature profile that contains two arte-
facts. As the animal begins to breathe in, there is an initial dip in recorded tem-
perature; this is the result of rapid evaporation from the probe tip of moisture
deposited during exhalation. This initial dip disappeared when the animals exhaled
desaturated air. The other artefact is an initially high temperature during exhala-
tion, caused by the release of latent heat by moisture depositing on the probe,
which at that moment was in temperature equilibrium with the inspired air. Both
these artefacts, which only appeared when condensation occurred, affect only the
initial parts of the temperature profiles.
All thermometers were calibrated over the range 20-50 ?C against a pre-
307
K. Schmidt-Nielsen, R. C. Schroter and A. Shkolnik
calibrated mercury thermometer and agreed to within 0.2 ?C. They were regularly
checked against each other to ensure correct readings.
The humidity sensor was checked before, during and after each measurement
by suspending it inside a bottle with the bottom covered by 1-2 cm of Drierite
(0 % r.h.), a bottle containing a saturated solution of NaCl (75 % r.h.), and a bottle
containing distilled water (100 % r.h.). The sensor was also regularly exposed to
the ambient air and its reading compared with the relative humidity derived from
the wet-and-dry bulb readings. The readings remained extraordinarily stable and
suggested that the error was less than 3 % r.h.
On a number of occasions the respiratory frequency was also recorded on the
Physiograph, by means of a displacement strain gauge located around the ribcage
at a level just above the last rib.
RESULTS
Rectal temperature
When hydrated, the camels' rectal temperatures varied less than 2 ?C through-
out the day and night. On dehydration the variations in rectal temperature in-
creased with daytime values near 41 ?C and night-time values around 36 ?C. This
pattern is similar to those reported in earlier studies (Schmidt-Nielsen 1964;
Schmidt-Nielsen et al. 1967).
Ventilation
The respiratory frequency increased during the day, as the body temperature
rose, and decreased again at night. The most marked effect, however, was that of
progressive dehydration. For example, a fully hydrated camel had a frequency
of 9.3 min-l (body temperature 39 ?C), but after 15 days of dehydration the
frequency had decreased to 4.3 min-1 (body temperature 38.7 ?0). These
observations are similar to earlier results of Schmidt-Nielsen et al. (1967), although
in the present study the respiratory frequencies were considerably lower at all
degrees of dehydration when allowance had been made for the effect of body
temperature.
Minute ventilation or tidal volume were not measured, but the output of the
chest-wall strain gauge provided information on the pattern of ventilation. First,
movements of the chest wall were in phase with flow conditions in the nose.
Secondly, the strain gauge changes were virtually linear with time (figures 1, 2).
Although strictly the gauge indicates chest perimeter, the pattern suggests that
the animals were breathing with constant and similar inspiratory and expiratory
flow rates, without significant periods of breath holding. This pattern is different
from those usually observed in man and other animals.
308
Desaturation of exhaled air in camels
volume
temp.;/?
r.h(%) ,-1 s -x._/ i \1/~
I y , 1
n--in "out--1
I I I
I I I I I I i I I
309
-~ 2s
2-
time
FIGURE 1. Recordings of respiratory movements (top), temperature of respired air (middle)
and relative humidity of respired air (bottom) of a normally hydrated camel (rectal
temperature 37.5 ?C, ambient air temperature 30.0 ?C). The records were obtained in the
evening after sunset. Exhaled air was at 32.0 ?C and 100% r.h. (It is an inherent charac-
teristic of the otherwise linearly responding humidity sensor that at 100% r.h. it con-
tinues to absorb water vapour, with the record going off scale.)
[- in- out-?
volume /
I i
temp./?C
38
----~b--___C1------ ----
36 -
r.h.(%) 80-.
- 2 s K-
time
FIGURE 2. Recordings of respiratory parameters, similar to those in figure 1, obtained from
a normally hydrated camel in the middle of the day (rectal temperature 39.0 ?C).
Exhaled
air was close to ambient air temperature (37.5 ?C) and at 100% r.h.
310 K. Schmidt-Nielsen, R. C. Schroter and A. Shkolnik
Exhaled air temperature
When the camels were fully hydrated or only moderately dehydrated the ex-
haled air was fully saturated (figures 1, 2). In the daytime, when the ambient air
was hot and near body temperature, the animal breathed out air at a temperature
near to body temperature. At night, when the air temperature fell, the exhaled
air temperature also fell (as reported by Langman et al. I979).
I I I i t i I I I I ' I
4- -
2- A-
vQ * inhaled (ambient) temp. A
4 - -
Fiaup.E Exhale irectal temp.
b2 - A A
A
A
0 - A A A
= 4- 2-
"~S 6~ ?
6-
8-^ a ^
10-- -
I a t I I I I I I l
OOhOO 04h00 08h00 12h00 16h00 20h00 24h00
time of day
FIGURE
3. Exhaled air temperature of camels in relation to ambient and rectal temperatures.
Upper panel: the temperature of the exhaled air remained within a few degrees of that
of the inhaled (ambient) air. At night exhaled air temperature could fall below that of
the inhaled air. Lower panel: in the daytime the exhaled air temperature was close to
body temperature but in the early morning and evening it could fall to more than 10 ?C
below body temperature.
The manner in which the exhaled air temperature varied with inhaled (ambient)
temperature and rectal temperature can be seen in figure 3, in which these are
plotted in relation to the time of day. During the hot day, when the solar heat load
was considerable, the exhaled air temperature was close to rectal temperature and
slightly higher than ambient temperature. However, at night the exhaled tem-
perature fell to as much as 10 ?C
below rectal temperature and could even be lower
Desaturation of exhaled air in camels
than the temperature of the inhaled (ambient) air. Such low exhaled air tempera-
tures relative to ambient have previously only been observed in small mammals
(Jackson & Schmidt-Nielsen 1964; Schmidt-Nielsen 1972).
Exhaled air humidity
When the camels were moderately dehydrated they exhaled fully saturated air,
but as the dehydration period extended beyond about ten days the animals began
to exhale unsaturated air (figure 4). At the beginning of an exhalation that lasted
32 _ -in-- out--?
temp./?C 341
100
80-
r.h.(%) 60 _
40- --in-- out
-- 2
-t 2s ~-
i
I I I I I
I, I I I I
time
FIGURE 4. Recordings from a camel after 12 days of water deprivation (rectal temperature
40.1 ?C). The temperature of the exhaled air (31.0 ?C) was slightly below that of the
inhaled ambient air (31.5 ?C). The exhaled air was, at the beginning of each exhalation,
less than fully saturated, but reached 100% r.h. towards the end of the breath. (For a
note on the response characteristics of the humidity sensor, see figure 1.)
about 5 s, the air was well below 100 ? r.h., but towards the end of the exhalation
the relative humidity rose to full saturation. Exhalation of unsaturated air was
only observed at night or in the early morning, and not during the daytime, when
the animals were subjected to a heat load.
Obtaining records of unsaturated exhalation required very gentle and quiet
handling of the animals. Excessive restraint, sudden noise or other disturbances
caused the camels to switch to fully saturated exhalation. An example is shown in
figure 5, in which a nervous and rapidly breathing animal suddenly changed from
unsaturated to fully saturated exhalation. The change in humidity of the exhaled
air is reflected in the temperature trace, where the condensation and evaporation
artefacts suddenly appeared as the exhaled air was fully saturated.
As the animals became more severely dehydrated, the exhaled air became more
unsaturated, with a relative humidity around a fairly constant value of 75 to 80 %
(figure 6). Under the most severe dehydration studied (15 days) the ventilation
pattern changed somewhat. The inspiratory time remained similar to that in the
311
312 K. Schmidt-Nielsen, R. C. Schroter and A. Shkolnik
31 - in I
out
temp./?C 29
r.h. (?\
-
100 1 ~ L; W - \ X
60- -iniout
'I I I I I I I ~
'--I I I I I
I?
time
FIGURE 5. Recordings from a camel after 11 days of water deprivation (rectal temperature
38.7 ?C). The temperature of the exhaled air (30.0 ?C) was nearly equal to that of the
ambient air (29.5 ?C). The humidity of the exhaled air was initially below 100% r.h. but
suddenly changed to full saturation after a slight disturbance to the relaxed animal
(cf. text).
32 i --in out i-
temp./0C 3 ---
100
80-
r.h (%) 60- i
40- ? in out-- ?
-, ,
in -- 2s -
l l l l l
I I I I I I I I I I
time
FIGURE 6. Recordings obtained at night from a severely dehydrated camel after 13 days of
water deprivation (rectal temperature 38.9 ?C). The exhaled air was clearly unsaturated
throughout the period of exhalation of approximately 4 s.
less severely dehydrated animal, but the expiratory time increased markedly,
respiratory rate falling to 4.3 min-'. We do not know whether this was due to slow
exhalation or to an end-expiratory pause. The exhaled relative humidity, which
in the beginning of each exhalation was approximately 75 %, increased towards
the end of the exhalation. (The thermocouple record indicated a cooling artefact
at the beginning of the inspiratory limb that may have resulted from an end-
expiratory pause with air in the vicinity of the sensor becoming saturated with
moisture.)
Desaturation of exhaled air in camels 313
DISCUSSION
Cooling and desaturation of exhaled air
It is well known that the respiratory air of mammals and birds can be exhaled
at temperatures far below body temperature, but it has been assumed that the
exhaled air is always saturated with water vapour at the temperature at which it
leaves the nasal passageways. The observation that water vapour can be removed
from the exhaled air as it passes through the nose is new, and helps to explain the
extremely low respiratory water losses that have been observed in camels.
The temperature and humidity relations of the respiratory air of humans have
been the subject of numerous studies (see, for example, Ingelstedt I956). Many
investigators have been concerned with the heating and humidification of the air
on inhalation, processes that are nearly complete in the nasal or oral passageways
before the air reaches the trachea. In the lungs the air is at body temperature and
fully saturated (Cole i953), but it is exhaled at a somewhat lower temperature
because the exhaled air loses heat as it passes over the nasal mucosa. In the process
the air also gives up some water vapour and remains at 100 % r.h. This can be
restated as: during exhalation some heat and water are reclaimed in the nasal tract
(see, for example, Walker et al. 1961).
The cooling of the exhaled air and recovery of water is strikingly important in
desert rodents, such as kangaroo rats (Jackson & Schmidt-Nielsen 1964). Kangaroo
rats live on seeds and other dry plant material. Usually they do not drink, and they
do not seek out green and succulent plants to obtain water. Their total water
supply consists of the small amount of free water adsorbed in their dry food and
the water derived from oxidation of the food ('metabolic water'). This water must
suffice to cover all losses: by evaporation and in urine and faeces. Losses in urine
and faeces are minimized by excreting highly concentrated urine and very dry
faeces, and the respiratory water loss is reduced by cooling of the exhaled air and
the resultant recovery of water. The mechanism is simple. The surfaces of the
nasal passageways are cooled by the inhaled air and because of evaporation they
may be cooled to below the temperature of the inhaled air. On exhalation the air
that passes over the cool surfaces gives up heat, water recondenses, and the
exhaled air may even be below ambient temperature. For example, a kangaroo
rat that inhales air at 30 ?C and 25 % r.h. is able to exhale air at 27 ?C and 100 %
r.h. Under these conditions 54 % of the water that was added to humidify the air
on inhalation is recovered on exhalation (Schmidt-Nielsen, et al. I970).
The processes of heat exchange and water recovery in the nasal passageways
have been the subject of careful theoretical analysis (Collins et al. 197I), which has
confirmed that the water recovery under favourable circumstances may exceed
75 %. This recovery of water is an essential component of the water balance of
kangaroo rats and other small desert mammals and their ability to subsist without
free water.
K. Schmidt-Nielsen, R. C. Schroter and A. Shkolnik
The heat exchange in the nose of humans is less effective because of the much
wider passageways and a relatively small surface area. It has therefore been
assumed that in other large animals nasal heat exchange would be relatively in-
effective, but we have recently shown that in the giraffe and several other large
East African ungulates the respiratory air may be cooled substantially, resulting
in considerable recovery of water in the nasal passageways (Langman et al. 1979).
These measurements indicated that nasal recovery of water in giraffes can explain
their relative independence of drinking water on the semi-arid East African
savanna. In camels the recovery of water caused by nasal heat exchange might
reach 70 % of the potential respiratory water loss (Langman et al. 1978).
The main site at which the heat and water exchange occurs has been identified
as the turbinate structures of the nasal passages. The reasons why heat exchange
and therefore water recovery are unexpectedly effective in large animals are that
the surface area of their turbinate structures is much larger than in humans and
that the wall-to-wall distance (the width of the passageways) is much smaller than
in humans, thus permitting rapid heat exchange during the flow of air.
The turbinates of the camel have been studied by Arnautovic et al. (I970), who
found that their surfaces are covered with a moist secretion that serves to humidify
the dry inspired desert air. To obtain a quantitative measure of the area of the
turbinate system we studied the heads of two adult camels, one formalin-fixed and
the other deep-frozen immediately after death. A section across the turbinates
shows them to be an extremely elaborate scroll-like structure. The manner in
which they develop from the nasal entrance to the level of the olfactory sinus can
be seen in figure 7. The structure is, in fact, more complex than suggested by
Arnautovic et al. (1970).
Measurements of the perimeter of the turbinates in serial sections suggest that
the total surface area that is exposed to respired air exceeds 1000 cm2. The gap
between adjacent surfaces in the air passages is only 1-2 mm. The remarkably
large surface area for exchange may be compared with an area of approximately
12 cm2 in humans and in rabbits (Negus I958). In a study of rabbits in which
thermocouples were implanted in the turbinate system it was found that the
inhaled air, even when inhaled at 0 ?C, is brought close to body core temperature
as it passes over the turbinate system (Caputa 1979). Similar measurements have
not been made in the camel, but the large surface of the turbinates and the narrow-
ness of the passageways gives assurance that on inhalation the air is heated and
fully humidified to body core conditions.
In our measurements we found a conspicuous difference in the temperature of
the exhaled air between day- and night-time conditions. In the daytime the ex-
haled air is close to rectal temperature and at night the exhaled air was cool and
approached and might even be below ambient air temperature. During the cool
night this aids in water conservation while in the daytime the heat load makes it
necessary for the camel to dump heat. In part this is achieved by exhaling warm
air with a correspondingly high water content.
314
Desaturation of exhaled air in camels
The difference between day and night temperatures of the exhaled air must be
achieved primarily through changes in the blood flow within the turbinates. At
night the blood flow may be minimal, thus adding little heat to the cooled tissues,
and in the daytime a high blood flow will reheat the turbinates, the blood itself
being cooled in the process. This results in exhalation of air close to the tempera-
ture of the blood and thus to the animal's body core temperature.
15.3 11.6 8.0 4.3 cm
0 1.0 cm
FIGURE
7. Tracings of cross sections of the nasal turbinate passages of a camel, made at the
indicated distances from the anterior end of the naris.
The cooling of the blood that drains from the turbinates is probably of import-
ance in protecting the brain from overheating, as suggested by Baker & Hayward
(I968). The approximate blood flow to the brain of camels can be estimated to be
approximately 0.4 l/min, based on an estimated brain mass of 700 g (Brody i945)
and a minute blood flow of 0.5 1/kg brain, similar to that in humans (Folkow &
Neil 1971). If air is inhaled at 35 ?C (25 % r.h.) and exhaled at 38 ?C (100 % r.h.),
then the change in enthalpy of the air is 93 kJ per kilogram of dry air. If one
assumes that the ventilation volume of the camel is 40 /mrin (as measured in the
daytime in a 500 kg camel under heat stress), the amount of heat removed per
minute would be 3.42 kJ. This is sufficient to cool about 410 ml blood by 2 ?C.
The cool venous blood from the nasal region drains via the cavernous sinus, where
it cools the arterial blood flowing to the brain. Such cooling of the arterial blood
to the brain depends on the presence of a carotid rete, which the camel possesses
(Ask-Upmark I935), and its importance has been amply documented in sheep and
315
K. Schmidt-Nielsen, R. C. Schroter and A. Shkolnik
several other species (see, for example: Baker 1972; Baker & Chapman I977;
Taylor & Lyman I972). Thus, the exhalation of warm, saturated air in the day-
time can permit selective cooling of the brain relative to body core temperature,
while at night, when the surroundings are cool, this mechanism is not needed and
the role of water conservation takes over.
The ability of the camel to desaturate the exhaled air has not been observed
before, and no similar phenomenon has been reported for other mammals. How-
ever, the mechanism explains the observations of extremely low respiratory water
losses reported in the accompanying paper (Schmidt-Nielsen et al. I98I). The low
respiratory water losses were observed only in severely dehydrated camels, as was
the directly measured desaturation of the exhaled air. This fact helps to explain
the mechanism responsible for the desaturation of the exhaled air.
Mechanism for desaturation of exhaled air
Any hygroscopic surface will take up water when moist air passes over it, and
water vapour will be given off again when dry air passes over it. This characteristic
of a hygroscopic surface appears to be the basis for the mechanism in the camel,
an assumption that is supported by observations with a simple mechanical model
that effectively achieves the same result.
When a camel is reasonably well hydrated the turbinate surfaces are kept moist
by secretions from the numerous glands in the mucosal lining, and especially by
secretion from the lateral nasal gland. (This gland, also known as Steno's gland, is
absent in man but present in many other mammals, such as dogs, in which its
importance in nasal secretion and heat exchange is substantial.) The nasal gland
is of appreciable size in camels (Abdalla & Arnautovic I970), but its function has
not been studied in detail. In the well watered camel one can observe small quanti-
ties of fluid flowing from the nose within the split in the upper lip. However, as
dehydration progresses, the secretion subsides and the turbinate surfaces pre-
sumably dry up. These large areas will then be covered with a layer of dried
mucous containing salts from glandular secretions, epithelial cells, etc.
The key to the desaturation of the air is that the dry surfaces take up water
from the exhaled air and give off water during inhalation. The mechanism is
similar to the nasal heat exchange that has been demonstrated in a variety of
animals (Schmidt-Nielsen 1972), except that in this case water vapour is deposited
on exhalation and removed from the surface on inhalation.
To demonstrate that this mechanism can work, we designed a simple mechanical
model in which air of different humidities was made to flow in alternating direc-
tions (figure 8). Various materials were used for lining of the narrow space over
which air was passed. To avoid condensation anywhere in this system, we used air
of 90 % r.h. to represent exhaled lung air, and for simplicity 0 % r.h. to represent
inhaled ambient air. The system was kept at constant temperature to avoid com-
plications from temperature gradients along the system.
316
Desaturation of exhaled air in camels 317
We found that ordinary filter paper convincingly demonstrated the effectiveness
of the model. A single sheet of filter paper, 6.8 cm x 24.2 cm (164 cm2), was tacked
with rubber cement to a brass plate and inserted, leaving an air space of I mm.
Passage of air in alternate directions over this system at a flow rate of 0.6 1/min
gave the results suggested in figure 8. The frequency of 'breathing' was kept at
6 cycle/min, the 'inhalation' and 'exhalation' each lasting 5 s. When dry air
passed through the model (figure 8, top), it arrived at the 'trachea' at 85 % r.h.;
0%
5%
85%
2mm
300 mm
70 mm r
FIGURE
8. Mechanical model used to demonstrate the ability of a hygroscopic surface to give
off or absorb water vapour in analogy to the camel's nose (cf. text).
when 'exhaled' air at 90 % r.h. passed over the filter paper, it exited from the
'naris' at 5 % r.h. The results were similar when 10 or 100 mg NaCl were added
to the filter paper.
The ability of a single sheet of filter paper to alternately humidify and dry the
air that passes over it makes us believe that this principle is effective in the nasal
region of the camel, which has an area approximately ten times as large as that in
our model. The air flow rates used in our model were 0.6 and 1.2 l/min. In the
camels for which we reported observations in the preceding paper, the respiratory
minute volume was 10-42 1/min (measured at ambient temperature and pressure),
K. Schmidt-Nielsen, R. C. Schroter and A. Shkolnik
or some 10 to 40 times higher than in the model. We therefore conclude that the
ability of severely dehydrated camels to desaturate the exhaled air can be explained
by a simple exchange system, based on a hygroscopic surface and in principle
analogous to the well known exchange of heat in the nasal passageways.
Effect on water balance
The question of greatest relevance to the water balance of the camel is: how
much water is saved when the exhaled air is cooled and desaturated, compared to
the loss that would occur if the respiratory air were exhaled saturated at core
temperature? The potential loss is the amount of water added on inhalation; the
actual loss is less. The amount of water saved can be computed as (water recovered
on exhalation)/(water evaporated on inhalation), or
i mass H20 in sat'd air at core temp.- mass H20 in exhaled air
mass H20 in sat'd air at core temp.- mass H20 in inhaled air
The recovery of water depends on several variables, such as the camel's core
temperature, the temperature of the exhaled air and its humidity, and, of course,
the amount of water already present in the inhaled air. As an example we can
consider measurements on camels under actual desert conditions.
Consider the record shown in figure 6. The ambient air was at 30.5 ?C, 40 % r.h.;
this air therefore had a water content of 12.3 mg/l. The exhaled air was at 31 ?C,
75 % r.h., and had a water content of 23.8 mg/l. Had the air been exhaled at core
temperature, 38.9 ?C, and 100 % r.h., it would have had a water content of 47.9
mg/l. The recovery on exhalation caused by cooling and dehydration of the air
was therefore 68 % of the potential water loss. (If the air had been exhaled cooled
to 31 ?C but fully saturated, the recovery would have been 46 %.)
The ability of severely dehydrated camels to desaturate the exhaled air repre-
sents a considerable increase in the animal's ability to conserve water. In the
present study the animals exhaled air at 75 % r.h. as the lowest observed value,
but we do not know whether this is the limit that can be achieved. The humidity
of the ambient air at the site of our study was never very low, the water content
remaining at approximately 12 mg/l, which in the night-time made the relative
humidity increase to around 40-50 %. It is not known whether the inhalation of
air at lower ambient humidities can result in exhalation at less than 75% r.h.
It can be concluded that during the daytime the dehydrated camel copes with
the heat load both by storing heat (permitting its body temperature to rise) and
by dumping heat via exhalation at the expense of water conservation. Sweating
also occurs, but it is probable that the substantial increase in evaporation from
the nasal tract is of importance in keeping the brain temperature from increasing
to intolerable levels. At night, when the environmental temperature is low, the
dehydrated camel cools down, reduces the temperature of the exhaled air, and
desaturates this air. This reduces the respiratory water loss to a minimum, thus
conserving water.
318
Desaturation of exhaled air in camels 319
This work was supported by grants from the National Geographic Society, NIH
grants HL-02228 and 1-K6-GM-21,522, and the Center for Wildlife Research of
Tel-Aviv University. Ephraim Maltz of Tel-Aviv University provided excellent
help in this study.
REFERENCES
Abdalla, O. & Arnautovic, I. I970 Morphological study of the lateral nasal gland of the
one-humped camel. Acta anat. 76, 123-130.
Arnautovic, I., Abdalla, O. & Fahmy, M. F. A. I97o Anatomical study of the vomeronasal
organ and the nasopalatine duct of the one-humped camel. Acta anat. 77, 144--154.
Ask-Upmark, E. I935 The carotid sinus and the cerebral circulation. An anatomical,
experimental, and clinical investigation, including some observations on rete mirabile
caroticunm. Acta psychiat. neurol. scand. 6 (suppl.), 1-374.
Baker, M. A. 1972 Influence of the carotid rete on brain temperature in cats exposed to hot
environments. J. Physiol., Lond. 220, 711-728.
Baker, M. A. & Chapman, L. W. 1977 Rapid brain cooling in exercising dogs. Science, N.Y.
195, 781-783.
Baker, IM.
A. & Hayward, J. N. 1968 The influence of the nasal mucosa and the carotid rete
upon hypothalamic temperature in sheep. J. Physiol., Lond. 198, 561-579.
Brody, S. 1945 Bioenergetics and growth. With special reference to the efficiency complex in
domestic animals. New York: Reinhold.
Caputa, M. 1979 Temperature gradients in the nasal cavity of the rabbit. J. ther\n. Biol. 4,
283-286.
Cole, P. 1953 Further observations on the conditioning of respiratory air. J. Lar. Otol. 67,
669-681.
Collins, J. C., Pilkington, T. C. & Schmidt-Nielsen, K. 197I A model of respiratory heat
transfer in a small mammal. Biophys. J. 11, 886-914.
Folkow, B. & Neil, E. I97I Circulation. New York: Oxford University Press.
Ingelstedt, S. 1956 Studies on the conditioning of air in the respiratory tract. Acta oto-iar.
131 (suppl.), 1-79.
Jackson, D. C. & Schmidt-Nielsen, K. I964 Countercurrent heat exchange in the respiratory
passages. Proc. natn. Acad. Sci. U.S.A. 51, 1192-1197.
Langman, V. A., Maloiy, G. M. O., Schmidt-Nielsen, K. & Schroter, R. C. 1978 Respiratory
water and heat loss in camels subjected to dehydration. J. Physiol., Lcnd. 278, 35P.
Langman, V. A., Maloiy, G. M. O., Schmidt-Nielsen, K. & Schroter, R. C. 1979 Nasal heat
exchange in the giraffe and other large mammals. Resp. Physiol. 37, 325-333.
Negus, V. i958 The comparative anatomy and physiology of the nose and paranasal si,uses.
Edinburgh: E. & S. Livingstone.
Schmidt-Nielsen, K. I964 Desert animals. Physiological problems of heat and water. London:
Oxford University Press. (Reprinted in 1979 by Dover Press, New York.)
Schmidt-Nielsen, K. 1972 How animals work. London: Cambridge University Press.
Schmidt-Nielsen, K., Crawford, E. C. & Hammel, H. T. I98I Respiratory water loss in
camels. Proc. R. Soc. Lond. B 211, 291-303.
Schmidt-Nielsen, K., Crawford Jr, E. C., Newsome, A. E., Rawson, K. S. & Hammel, II. T.
1967 Metabolic rate of camels: effect of body temperature and dehydration. Am. J.
Physiol. 212, 341-346.
Schmidt-Nielsen, K., Hainsworth, F. R. & Murrish, D. E. 1970 Countercurrent heat ex-
change in the respiratory passages: effect on water and heat balance. Resp. Physiol. 9,
263-276.
Taylor, C. R. & Lyman, C. P. 1972 Heat storage in running antelopes: independence of brain
and body temperatures. Am. J. Physiol. 222, 114--117.
Walker, J. E. C., Wells Jr, R. E. & Merrill, E. W. 1961 Heat and water exchange in the
respiratory tract. Am. J. Med. 30, 259-267.
