physiological reactions of the human body to
TRANSCRIPT
PHYSIOLOGICAL REACTIONS OF THE HUMAN BODY TO VARIOUS ATMOSPHERIC HUMIDITIES
C.-E. A. WINSLOW, L. P. HERRINGTON AND A. P. GAGGE
Contribution No. 16 from the John B. Pierce Laboratory of Hygiene, New Haveq Connecticut
Received for publication April 3, 1937
Plan of studies. As in the earlier studies from this laboratory’ our observations were made on unclothed male subjects in a semi-reclining position, placed in a copper booth, so arranged that air temperature and radiation from the copper walls could be independently varied. In each experiment the heat interchange between the body and the environment has been analyzed into its five factors- metabolism, radiation, convection, evaporation, and storage -by methods described in detail in the papers cited.
In our earlier work, we maintained, in general, a relative humidity of 40 to 50 per cent of saturation while varying air temperature, radiation and air movement. In the present studies we have provided a uniform turbulent air movement of approximately 17 linear feet per minute (8-9 cm./sec.) and no artificial radiation was employed, so that air and wall temperatures assumed a common value: air (and wall) temperature and relative humidity were the variable factors.
Three subjects were used for the present experiments, whose physical characteristics were as follows:
SUBJECT HEIGHT METERS
WEIGHT KILOGRAMS
DUBOIS LINEAR AREA
SQUARE METERS
CONSTANT FOR
EFFECTIVE HEAT LOSS BY
CONVECTION AND RADIATION AREA, RADIATION FOR SQUARE METERS* AIR VELOCITT
17 FEET*
I IV
VII
2.13 1.58 18.0 1.53
i 1.02
I 12.2
1.92 1.63 18.5
General results. The present study is based on some 300 different experi- ments and to save space the results are presented in table 1 in the form
l Winslow, Herrington and Gagge, 1936a and b, and 1937; Gagge, 1936; Herrington, Winslow and Gagge, 1937.
2 The methods of computing the factors are described in our earlier papers.
288
REACTIONS OF HUMAN BODY TO ATMOSPHERIC HUMIDITIES 289
of group averages, each group including about six different experiments in which air temperature and atmospheric humidity were approximately identical. The headings of the columns and the method of computing the various factors are explained in a footnote to the table and in the text below.
It will be noted that air temperatures varied from 16.6OC. to 37.9”C., and relative humidities from 14 to 80 per cent of saturation. Heat losses by radiation and by convection were, throughout, of almost exactly equal magnitude and, of course, both types of heat loss increased with decreasing environmental temperature according to principles whose oper-
120 - 0
IO0 - SUBJECT I
ao- :
60 - g
40 - O 0
20 - 0 0
I *: - 5 SUBJECT IV 0% - 4 60- 0
Y 0 40
: 0 &
z - 0 CM
$ 20 - - 0 l - 0.
F 0, d 120 - 8 2 : 0
s 100 - SUBJECT
VII
0
w ao- b0 60 - 0.
40 - 0:
0 0- 20 - 00 e
0~“““““““““““““““““~ IO I5 20 25 30 3s 40 4s
TA IN “C
Fig. 1. Evaporative heat loss in relation to air temperatures. Solid circles, high humidity. Open circles, low humidity.
ation was made clear in our earlier studies. Heat loss by evaporation varied widely in a fashion to be discussed below.
Evaporative heat loss in relation to atmospheric humidity. In order to elucidate this point, we have plotted in figure 1 the data for evaporative heat loss for each of the three subjeets in relation to atmospheric tempera- ture. All groups of experiments with a relative humidity below 40 per cent are plotted as open circles and all groups with relative humidit,ies higher than 40 per cent as closed circles.
These graphs, first of all, bring out the major relations between evapora- tive heat loss and atmospheric temperature which we have discussed in
TABL
E 1*
Gr
oup
mea
ns
of
expe
rimen
tal
data
(S
ubje
cts
I, IV
, VI
I>
GROU
P I-l 2 3 4 5 6 7 8 9
IV-1
2 3 4 5 6 7 8 9 10
11
12
13
14
15
16
17
18
19
20
21
22
6 6 6 6 6 6 6 6 6 6 4 6 6 6 8 6 6 6 6 6 6 6
-. OC.
24.4
27
.8
30.6
34
.3
37.9
25
.2
27.6
30
.6
34.3
24.3
27
.5
30.4
33
.2
33.8
34
.2
34.4
35
.8
36.5
37
.0
16.6
21
.6
21.6
24
.9
25.3
27
.7
28.6
31
.0
31.1
33
.0
34.9
34
.1
TA
RH
TS
Tg
--
pm O
C. O
C.
cent
35
31.8
36.7
26
32
.913
7.3
22
33.7
36.6
17
34
.537
.0
17
35.4
,38.
4 75
32
.4/3
6.6
75
32.8
i36.
9 60
34
.9’3
6.8
78
35.6
37.0
32
31.0
137.
3 27
32
.537
.2
22
33.4
36.8
20
35
.436
.9
27
35.4
36.7
38
35
.236
.6
19
34.3
37.2
29
35
.4 3
7.0
25 1
35.7
36.
9 14
34
.6’3
7 4
64 ,
29 5
136.
0 44
‘3
1:9/
36:6
76
31
.7 3
6.6
48
33.1
36.7
72
31
.836
.9
75
33.5
36.9
52
34
.6l3
6.7
63
34.7
36.8
45
35
.6 3
6.9
52
35.1
36.6
42
13
3.93
7.1
80
135.
1j36
.5
kgm
.-cul
. --
ri 88
87
89
85
94
99
103 99
62
58
62
66
61
74
57
65
64
68
65
61
57
70
68
67
63
72
63
67
63
67
E R
C TR
UE
8
kgm
.-cal
. ---
hr. -31
-21
-54
-85
-116
-1
9 -2
7 -6
0 -7
8
kgm
.-C
al.
_I- F
i2
-43
-26 -1
22
-61
-44
-36
-11
kgm
.-cal
.
-$3
-46
-28 -2 23
-6
5 -4
8 -3
8 -1
3
kgm
.-cat
. kg
m .
-Cal.
lit
ers
hr.
Gi3
pgc
min
. 63
16
.4
7.8
22
11.9
8.
0 21
17
.8
8.4
-1
16.7
7.
8 -1
4 11
.2
11.6
51
16
.7
7.7
20
13.9
7.
3 31
32
.8
8.5
3 34
.4
7.5
-22
-29
-35
-55
-53
-57
-60
-75
-84
-75
-12
-15
-14
-20
-19
-17
-21
-51
-35
-42
-66
-51
-40
-28
-16
-13
-10 -6 1 3 1:
-71
-59
-57
-47
-38
-33
-35
-21
-26
-12
6 -6
-45
45
11.8
-3
2 31
12
.1
-19
8 13
.0
-15
17
35.1
-1
1 13
25
.4
-7
-4
32.6
1
1 12
.8
3 4
27.2
5
10
39.1
15
-2
2 10
.1
-89
107
16.9
-7
1 84
19
.4
-69
83
18.0
-5
6 53
21
.6
-42
31
12.4
-3
8 21
16
.6
-41
34
28.2
-2
4 24
29
.7
-30
28
42.2
-1
4 1
28.8
7
-10
10.8
-6
-1
4 31
.o
,
COND
UC-
LUNG
RE
SP.
SKIN
HV
AP.
TANC
E VE
NT.
EVAP
. EV
AP.
POW
ER
--- ~
.--_
kgm
.-cd.
kg
m.-c
al.
---
hr. 9 10
10 9 14 7 5 7 3
2 11
44
76
102 12
22
53
75
kgm
.-cal
. ---
m.t/
hr.
76
84
92
96
100 52
45
63
35
4.7
5.2
5.5
5.5
5.5
5.5
5.2
5.7
5.0
7.0 5.8
5.9
5.5
5.5
4.5
5.0
5.5
5.0
5.1
5.5 5.0
5 .o
6 6 7 7 6 5 6 6 5 9 7 7 5 6 4 4 5 4 5 5 4 2
16
74
23
81
28
88
48
101
47
93
52
77
54
94
69
94
79
90
66
97
5 63
8
74
9 57
14
75
15
50
13
55
16
73
47
58
30
79
37
71
62
59
49
32
--
- -_
____
- .--
.. ----
...
-. - - .
--_ _
_ _ _
.
co
8:
a FE
4 E
-- pe
r ce
nt
13 6 22
37
46
11
23
40
100 14
19
21
31
33
44
38
48
57
45 5 7 10
12
20
15
14
53
25
34
1 ’ 69
1.
00
VII-1
6
124.
3 32
13
2.21
37.2
2
6 27
.8
26
32.8
137.
5 3
6 30
.6
22
34.5
137.
2 4
6 33
.1
29
35.6
37.1
5
6 33
.8
36
35X3
7.1
6 6
34.1
17
35
.337
.6
7 6
35.8
22
35
.937
.3
8 6
35.8
35
34
.237
.2
9 6
36.7
26
36
.137
.2
10
6 37
.4
16
35.8
37.8
11
6
17.1
70
29
.736
.7
12
6 20
.1
55
32.1
36.7
13
6
22.1
78
32
.636
.8
14
6 24
.4
75
31.8
37.1
15
6
25.2
48
33
.236
.5
16
6 28
.1
70
33.2
37.2
17
6
28.6
52
35
.036
.7
18
6 30
.4
65
35.6
37.4
19
4
30.7
47
35
.937
.1
20
6 33
.1
70
35.8
37.1
21
6
33.6
45
35
.837
.0
22
6 34
.2
77
36.3
36.8
104 92
102 84
97
105 96
97
95
86
96
92
91
99
82
94
85
104 91
95
98
10
5
-26
-71
-75
-29
-45
-47
-42
-36
-37
-65
-23
-25
-76
-19
-21
-96
-11
-11
-109
-1
-1
-7
9 15
17
-1
19
5 6
-116
15
17
-2
5 -1
15
- 12
7 -2
5 -1
02
-113
-2
3 -9
6 -1
06
-24
-67
-70
-26
-73
-80
-32
-46
-48
-32
-59
-65
-41
-47
-48
-49
-47
-52
-63
-25
-28
-78
-21
-23
-69
-19
-20
- 68
17
.8
10.0
12
14
29
13
.3
7.8
9 20
13
22
.0
10.6
13
29
29
37
.8
9.3
11
54
19
49.5
10
.3
10
66
13
26.3
10
.5
13
83
15
40.0
10
.0
12
97
-50
7.9
10.6
10
69
13
50
.5
10.3
11
10
8 -2
19
.6
14.2
17
99
17
1 19
.8
9.3
10
15
148
26.8
11
.0
10
15
134
27.8
10
.9
10
13
62
15.9
7.
5 7
17
97
24.3
9.
0 10
16
32
16
.2
8.1
6 26
71
47
.1
9.3
8 24
32
38
.7
9.0
7 34
57
63
.2
10.0
10
39
21
47
.4
9.3
5 58
24
54
.4
9.9
9 69
3
101.
3 7.
5 2
67
81
87
97
99
86
104
100 70
93
103 57
72
62
51
77
52
77
63
90
48
48
35
1 -
9 12
16
28
40
42
51
51
60
50
14
11
11
17
14
26
16
28
23
63
75
100
* Th
e fir
st
colu
mn
give
s th
e su
bjec
t an
d th
e ex
perim
enta
l gr
oup.
Th
e dr
y bu
lb te
mpe
ratu
re
(TA)
an
d th
e re
lativ
e hu
mid
ity
(as
a pe
r ce
nt)
appe
ar
in t
he n
ext
two
colu
mns
. Th
e sk
in
tem
pera
ture
(T
s)
is a
mea
n fo
r th
e wh
ole
body
an
d re
pres
ents
a
weigh
ted
mea
n of
15
segm
enta
l ob
serv
atio
ns.
The
body
te
mpe
ratu
re
(TB)
is
mea
sure
d re
ctal
ly.
The
met
abol
ism
(M)
is m
easu
red
by t
he
rate
of
oxy
gen
cons
umpt
ion.
Th
e to
tal
evap
orat
ive
loss
(E
) is
der
ived
from
th
e ra
te
of w
eight
lo
ss.
The
radi
atio
n lo
ss
(R)
and
the
conv
ectio
n lo
ss
(C)
have
be
en
calcu
late
d fro
m
the
usua
l fo
rmul
ae
usin
g th
e co
nsta
nts
indi
cate
d in
a
prev
ious
ta
ble.
Th
e tru
e st
orag
e (S
) ha
s be
en
foun
d by
di
ffere
nce.
Th
e co
nduc
tanc
e (C
ond,
) re
pres
ents
th
e m
ean
heat
flu
x th
roug
h th
e sk
in
surfa
ce
per
degr
ee
of g
radi
ent
fall
to t
he
skin
te
mpe
ratu
re.
It is
cal
cula
ted
from
th
e eq
uatio
n;
M+S
E+
R+c
A (T
g -
Ts)
Or
A (T
g -
Ts)’
wher
e A
is t
he
DuBo
is or
to
tal
area
of
the
bo
dy.
The
heat
lo
st
by e
vapo
ratio
n in
th
e re
spira
tory
tra
cts
has
been
ca
lcula
ted
from
th
e lun
g ve
ntila
tion
assu
min
g th
at
the
exha
led
air
is
satu
rate
d an
d at
bo
dy
tem
pera
ture
. Th
e “s
kin
evap
orat
ion”
re
pres
ents
th
at
evap
orat
ion
takin
g pl
ace
on t
he s
kin
surfa
ce
alone
an
d is
mea
sure
d by
the
diff
er-
ence
be
twee
n th
e to
tal
evap
orat
ion
and
the
resp
ired
evap
orat
ion.
Th
e ev
apor
atin
g po
wer
of t
he
air
is p
ropo
rtion
al
to
the
heat
lo
st
to t
he
envir
onm
ent
by a
un
it ar
ea
of e
xpos
ed
liqui
d su
rface
at
sk
in
tem
pera
ture
. Th
e “p
er
cent
of
wet
tedn
ess”
re
pres
ents
th
e ra
tio
of th
e sk
in
evap
orat
ion
per
unit
area
to
th
e ev
apor
atin
g po
wer.
If
mea
sure
s th
e fra
ctio
n of
the
m
axim
um
poss
ible
sw
eat
secr
etin
g su
rface
s th
at
are
actu
ally
st
imul
ated
. Fo
r co
nditio
ns
I-9,
IV-2
2,
and
VII-2
2,
the
mea
sure
d to
tal
evap
orat
ion
per
hour
wa
s 93
, 61
, an
d 84
kilo
gram
-cal
orie
s pe
r ho
ur,
resp
ectiv
ely.
Th
ese
valu
es
would
, ho
weve
r, ha
ve
resu
lted
in a
per
ce
nt
of w
ette
dnes
s ov
er 1
00 p
er
cent
. Th
e va
lues
in
the
ta
ble
for
E, n
amel
y,
78,
51,
and
69,
have
be
en
calcu
late
d su
ch t
hat
the
per
cent
of
wet
tedn
ess
is
exac
tly
100
per
cent
. Th
ese
expe
rimen
ts
repr
esen
t sit
uatio
ns
in
which
th
e to
tal
surfa
ce
wate
r lo
ss i
s no
t a
mea
sure
of
the
ef
fect
ive
evap
orat
ion.
292 C.-E. A. WINSLOW, 1,. P. HERRINGTON AND A. I’. GAGGE
our earlier communications. Below a certain critical point (about 30°C. air temperature)3 evaporative heat loss is at almost a minimum level (15-- 25 kilogram-calories per hour) which corresponds closely to evaporation from the normal skin with no active secretion of sweat. Above this point, ctvaporation rises rapidly with further increase of operative temperature.
The second fact brought out by figure 1 is that the relative humidity of the atmosphere, within the limits studied, seems to have no appreciablr influence upon the magnitude of evaporative heat loss so long as physic.a.1 conditions permit an evaporative balance. The open and closed circles in the graphs show no divergence, both alike exhibiting the sa.me relation- ship to atmospheric temperature.
Physiological control of sweat secretio,n. It is, of course, obvious that this identical evaporative heat loss, at a given air temperature but with widely varying humidities, calls for a physiological explanation. If the surface of the body maintains the same physiological characteristics, there would necessarily be less evaporation with high relative humidity. The rate of evaporation from a unit surface of water is proportional to t,he difference between the vapor pressure of the liquid and the vapor pressure of the surrounding air and to the degree of air movement. The constancy of evaporative heat loss with varying relative humidity must be maintained, therefore, by changes in sweat secretion.
In a previous paper (Gagge, l937), a new physiological variable has bee11 described which gives a measure of the extent of wettedness on the total exposed surface of t,hc body. This variable is defined by t,hc cqua~tiol~
(w/i) = E/A + [c(Ts) - rh X I], (1)
where E is the total evaporation from the body; A is the IM3ois area: w represents that fract,ion of the body surface which, if cornpletlely wetted, would produce the observed tot,a,l evaporative E; ,u is the proportionalit!J factor depending on the physica. nature of the evaporat#ivc processes; E(&) and e(TA) are the saturation \rapor pressures at skin temperature and air temperature, respectively; (rh) is the relative humidit,y of the air in terms of a fraction.
The value for (wp) above is computed for the body as a whole. By sub- tracting the evaporative heat loss of the respiratory tract (computed from lung ventilation, see table 1) from the total evaporation, we may obtain :I value of (wp) for the skin surface alone.
It has been shown (Gagge, 1937) that (wp) has a maximum value of 28.5 kilogram-calories per square meter per hour per centimeter of Hg
3 In our earlier studies, the critical point appeared at 3l”C., since, for a stmadnrtl time of exposure short of equilibrium, a slightly higher To is required to reach the critical 5”s at which positive evaporative regulation begins when a combination of cold air and hot radiation is cmploycd.
REACTIOXS OF HUMAN
pressure for the skin surface* cent of the maximum possible
w = (E/A) t 28.5 [e(Ts) - rh X QA)],
BODY TO ATMOSPHERIC HUMIDITIES 293
If we then define w as unity or 100 per wetted area, equation (1) becomes
(2)
where E now is the net evaporation from the skin surface only. Therefore,
28.5 [c(TS) - r/Z x dTA)l or&
is defined as the evaporating power of the ambient air. It is directly proportional to the heat loss by evaporation from a square meter of liquid surface at constant temperature (T,J, when placed in an environment, of
W% 60
40
20
a01 SUBJECT VII 1
o’- l”“‘~‘~“~“‘~“““““-““‘f JO IS 20 25 30 33 40 45
TA IN t
Fig. 2. Percentage of maximum possible area of wetted skin surface (w) in rela- tion to air temperatures. Solid circles, high humidity. Open circles, low humidity.
dry bulb (TJ, and relative humidity (rh). The evaporating power, 6, and the area of wettedness as a percentage, w, both calculated by the above formula, have been entered in the last two columns of table 1.
In figure 2, this factor of “wetted area,” w, is plotted against air tem- perature for the high and the low atmospheric relative humidities. Here we see a quantitative expression of the factor by which constant evapora- tion is maintained at a given air temperature with either high or low rela- tive humidity. That there must be a change in sweat secretion to account for a constant evaporation with varying moisture-demands of the atmos- phere is, of course, obvious. The closeness of the quantitat,ive results
294 C.-E. A. WINSLOW, L. P. HERRINGTON AND A. P. GAGGE
obtained is, however, striking. At air temperatures under 26.7” for subject I, 29.4” for subject IV, and 30.6” for subject VII, the “wetted area” remains low-about 5 to 20 per cent-irrespective of atmospheric humidity. Above these critical points, however, the values for high and low humidity follow completely different lines. With low humidity, even at air temperatures of 37.8”, the wetted area only reaches 40 to 60 per cent of the maximum possible wetted area. With high humidity at 34.2”C., on the other hand, it rises sharply to 100 per cent.
There is obviously an extremely delicate and perfect mechanism at work which increases the discharge of sweat so as to exactly balance the de- creased moisture-demand of a humid atmosphere and thus to maintain evaporative heat loss at a desirable level.
Circulatory changes in the skin in relation to atmospheric humidity. Since secretion of sweat increases with air temperature and with air humidity, it might be expected that the cutaneous blood supply would increase under the same conditions. In a previous study (Winslow, Herrington and Gagge, 1937), we have shown that the conductance of the surface layers of t0he skin may be computed from the surface area of the body, metabolism, st,orage, rectal temperature and skin temperature, and represents heat flux through the skin in kilogram-calories for each degree Centigrade fall in temperature per square meter per hour. This factor depends upon the intrinsic character of the tissues of the individual (as modified by various amounts of fat) and upon the blood flow through the skin. The con- ductance appeared, on analysis of our data, to be materially higher (for a given air temperature) with high atmospheric humidity than with low atmospheric humidity.
General picture of relationship for all subjects. Since the reactions of the three different subjects are, in general, so similar we have combined the results for all three in table 2, computed for all groups of experiments con- ducted with approximately similar conditions of atmospheric temperature and humidity. Metabolism and heat loss by evaporation have been com- puted per square meter of body area to make the subjects comparable and values for conductance and wetted area are already on a basis of unit body surface.
The data are divided into six groups, from the standpoint of atmospheric temperature (l&6”-21.0”, 21.1”-26.6”, 26.7”-32.1”, 32.2’-34.9”, 3&O”- 37.7”, and 37.8’ and over) and into the usual two groups as to atmospheric humidity. As atmospheric temperature rises (both for low and high humidity) we observe the usual rise in body temperature, skin temperature, and evaporation ratle. Conductance and wetted area also increase steadily with air temperature, except for conductance at Oemperatures over 37.8’- a value based on a single group of experiments.
Comparison of the two halves of t(he table gives us, however, the sig-
REACTIONS OF HUMAN BODY TO ATMOSPHERIC HUMIDITIES 2%
nificant approach to our present problem. For air temperatures under 21.0” (group I), we have only high humidity conditions, and for air tem- peratures over 34.9” (groups V and VI), only low humidity conditions. For the three sets of air temperatures between 21.0’ and 34.9”, however, two sets of conditions are available-for low and high relative humidity, re- spectively, but with metabolism, air temperature, mean body temperature and mean skin temperature for the two humidity groups practically identi- cal (with a single exception as to skin temperature to be noted later).
TABLE 2
Mean comparative data for all subjects classified by atmospheric temperature and humidity
--._. -. - LOW HUMIDITY
Number experi- ments
18 24.39 33 47 28 29.11 24 44 52 33.83 25 46 42 36.44 24 47
4 37.94 17 40
TA
“C.
RN M
per cent kgm.-031. hr.
- ---------.---.- -
“C.
37.06 37.11 36.94 37.28 38.44
TS WA Cond. w
kg,m.-cal. “C. ~
m.z/hr.
- kgm.-cd. -------
m.z/hr./“C. per cent
32.17 14 33.28 19 35.28 39 35.39 53 35.3gi 54
15.0 12 15.0 16 27.4 37 27.7 52 11.2 46
QROUP NUMBER
I 16 II 50 III 56 IV 42
HIGH HUMIDITY
Number experi- ments
M Tl? Ts E/A Cond. W
kgm.-cal. hr.
.
18.17 63 47 23.78 64 45 29.06 60 47 33.89 63 45
*
- --
OC. OC.
36.44 30.59 36.72 32.33 36.94 34.56 36.89 35.39
I
kgm.-cd. -_-- m.z/hr.
11 12 20 36
_~____. kgm#.-cd. - -----_. -- _
mj.2/hr./“C.
21.2 19.4 33.0 53.2
per cent
10 13 26 74
Comparing the figures we note, first, that total evaporation rate is com- pletely unaffected by relative humidity of the atmosphere. The rates are 14 and 12 for group II, 19 and 20 for group III, and 39 and 36 for group IV-the first figure in each case being for low and the second for high atmospheric humidity. In the second place, wetted area is almost iden- tical for both humidities for group II (12-13 per cent) where secretion of sweat is minimal; while for group III the value rises from 16 with low humidity to 26 with high humidity; and for group IV from 37 wit#h low humidity to 74 with high humidity. Conductance of the skin is through-
296 C.-E. A. WINSLOW, L. P. HERRINGTON AND A. P. GAGGE
out higher with high relative humidity (15.0 against 19.4 for group II; 15.0 against 33.0 for group III, and 27.4 against 53.2 for group IV).
Figure 3, plotted from the data in table 2, brings out the close parallelism between the two factors of skin conductance and wetted area. This conductance is striking in view of the fact that these factors are computed from quite independent experimental data. Skin conductance depends on metabolism, storage, body temperature and skin temperature; wetted area on weight loss, skin temperature, air temperature, and atmospheric humidity. When, in addition, one recalls that air temperature and skin t,emperature were, in general, the same for both high and low atmospheric
70 -
60 -
““““r’*‘r”‘r”“‘.‘.’ 14 16 I8 20 22 24 26 28 30 32 34 36 38 40
TA IN ‘t
0 HIGH 0 LOW
HUMIDITY HuMlDlTY
Fig. 3. Mean conductance of body surfaces and percentage of maximum possible area of wetted skin surface in relation to air temperature (for experiments with all subjects studied under corresponding conditions).
humidities, it seems clear that the body responds very definitely to high atmospheric humidity by an increased blood supply to the skin and an increased secretion of sweat-a process rative heat loss at the desirable level.
Two alternatives suggest themselves as possible explanations of the mechanism of this process. It may be th .at slight rises of skin temperature do actually occur which stimulate sweat secretion and lead to a prompt return of skin temperature to normal. Or the moisture content of the
nicely adjusted to maintain evapo-
skin as influenced by evaporation may have a stimulating influence. Whether the fall in conductance and sweat secretion for group VI is
REACTIONS OF HUMAN BODY TO ATMOSPHERIC HUMIDITIES 297
significant we are not certain. As stated above, there was but one group of experiments under this condition, with an abnormally low metabolism.
Limits to the cooling power of the evaporative mechanism. The power of the body to adjust its heat losses in a hot environment has, as we have seen before, its definite upper limits and the data already obtained make it possible to calculate these limits with a reasonable degree of precision.
The heat interchanges of the body for a condition of equilibrium may be expressed as follows :
when M=E+ (R+C), M = Metabolism,
(3)
and E = Total evaporative heat loss,
CR + c> = Combined heat loss by radiation and convection.
From our previous studies (Winslow, Herrington and Gagge, 1937) we know that
where (R + c> = Ko(Ts - To), K 0 = Environment,al constant,
(4)
and T S = Skin temperature, T 0 = Operative temperature.
With wall and air of equal temperature and standard air movement (17 feet per minute), To is the same as TA (air temperature).
Furthermore, if we divide all factors by the DuBois area we can express our results per square meter of body surface.
Equation (3) then becomes, per unit area,
M I = E’ + Kof (Ts - TJ. (5)
We know from our previous work that M (metabolism per square meter of body surface) is 47 kilogram-calories per hour4 and that K o is 8.45. With these data, and transposing, we have
E’ = 47.0 - 8.45 (T,--TA). (6)
From equation (1) we have
Combining (6) and (7) and solving for the humidity (rh) we have
RH = 100/c(TA) c(Ts) - 47*o - “;;y - T*)]. (8)
Equation (8) describes the interrelationship between the four variables, Ts, TA, RH (as a percentage) and (wp), under all conditions where regula- tion may be maintained, i.e., when storage is zero.
4 Mean metabolism for young male subjects in semi-reclining experiments.
position as in our
298 C.-E. A. WINSLOW, L. P. HERRINGTON AND A. P. GAGGE
In a previous communication, one of us (Gagge, 1937) has shown that at the upper limit of regulation by sensible perspiration, (wp) has a maxi- mum and constant value of 29.9 kilogram-calories per square meter per hour per centimeter of Hg vapor pressure for the total body. If we assump that at the upper limit the skin temperature is 35.6”C., the mean tempera- ture corresponding to a loo-per cent wetted area in table 1 for our three subjects, we may now substitute these values in (8) and have
RH= lOO/E(TA) 4.33 - 47.0 - 8.45(35.6) - T,)
29.9 1 J which equation may simplify to
RH = lOO/E( TA)
From equation (9) one may compute for any air temperature the rela- tive humidity which will correspond to t#he upper limit of evaporational cooling. This has been done in table 3.
TABLE 3 Upper limits of evaporative regulation at 17 jt/min. air movement
DRY BULB TEMPERATURE
---
“C.
45.3 42.5 40.0 37.5 35.0 32.5
BELATIVE HUMIDITY
per cent 0
13 28 47 70
100
-- - WET BULB TEMPEBATURE
OC.
16.6 21.1 24.4 27.8 30.0 32.5
These figures correspond reasonably well with the findings of McConnell, Houghten and Yagloglou (1924) that saturated air at 32.2” is the “upper limit of man’s ability to compensate for atmospheric conditions” in still air. They are also in close accord with the reports of Cadman and Haldane (British Departmental Committee, 1909). Cadman states that at 29.4” wet bulb temperature, the body temperature invariably rises, while at 33.9O wet bulb, “one is in a terrible state”; Haldane, that at 31” to 32” wet bulb “in fairly still air the body temperature begins to rise, even in tlhc case of persons stripped to the waist and doing no work; and when air is saturated this rise continues until symptoms of heat stroke arise.”
SUMMARY
In the present paper we have: 1. Applied a new procedure for estimation of the total effective sweat
secretion from the human body surface.
REACTIONS OF HUMAN BODY TO ATMOSPHERIC HUMIDITIES 299
2. Demonstrated that the discharge of sweat exhibits a close adaptation to the evaportive moisture demands of the environments (as those de- mands are conditioned by both the temperature and the relative humidity of the atmosphere) so that evaporative heat loss from the body maintains the level necessary to dissipate the heat produced by metabolism.
3. Shown that the increased excretion of sweat associated with high atmospheric humidity at any given air temperature, is accompanied or preceded by an increased cutaneous blood supply-for which process the initial stimulus must presumably be peripheral rather than central, since body temperature does not change appreciably in the process while even skin temperature alters but slightly and temporarily.
4. Indicated on theoretical grounds the upper limits of temperature and relative humidity beyond which this regulative process fails.
REFERENCES
British Departmental Committee on Humidity and Ventilation in Cotton Weaving Sheds. Minutes of Evidence and Appendices. Home Office, Great Britain. 1909.
GAGGE, A. P. This Journal 116: 656, 1936. This Journal 120: 277, 1937.
HERRINGTON, L. P., C.-E. A. WINSLOW AND A. P. GAGGE. This Journal 120: 133,1937. MCCONNELL, W. J., F. C. HOUGHTEN AND C. P. YAGLOGLOU. Trans. Am. Sot.
Heat. and Vent. Engineers, 30: 167, 1924. WINSLOW, C.-E. A., L. P. HERRINGTON AND A. P. GAGGE. This Journal 116: 641,
1936. This Journal 116: 669, 1936. This Journal 120: 1, 1937.