THE INFLUENCE OF DISTRIBUTION AND ECOLOGY
ON THE THERMOREGULATION OF SMALL BIRDS

By

CHARLES GERALD YARBROUGH

A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF

THE UNIVERSITY OF FLORIDA

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE

DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA
1970

.^NIVERSITY OF FLORIDA

3 1262 08552 3131

ACKNOWLEDGMENTS

I have profited from many discussions v/ith David W.
Johnston and Brian K. McNab . Dr. IIcNab also provided the
data on the Black-throated Trogon. David Niles and S. A.
Rohwer of the University of Kansas kindly obtained the
Harris' Sparrov/s for me. The equipment and facilities of
the University of Florida were used throughout the study.
The work was done while in tenure of a National Science
Foundation Traineeship. I wish to credit my v/ife, Hazel,
for her constant encouragement.

11

TABLE OF CONTENTS

Acknowledgements ii

List of Tables iv

List of Figures V

Abstract viii

Introduction 1

Materials and Methods 5

Results 7

Discussion 28

Heat Loss and Heat Gain 28

Metabolism 30

Conductance 34

The Physical Model 36

Determination of the Level of Tu^ 39

Body Size Effects 44

Precision of Tj^ Regulation 46

Ecological and Distributional Relationships 51

Conclusions and Summary 56

Appendix A - Symbols and Expressions Employed 59

in This Paper

Appendix B - Scientific and Common Names for 61

Species Discussed in This Study

References 64

Biographical Sketch 68

111

LIST OF TABLES

Table 1 Parameters of the energetics of T, 22

regulation in sone small 'birds
Table 2 Energetic parameters of small birds 32

selected from the literature

IV

LIST OF FIGURES

Figure 1 Responses of body temperature and oxygen 9
consuinption to ambient temperature in
Spizella passerina and Zonotriohia albicollis .

Figure 2 Responses of body temperature and oxygen 9
consumption to ambient temperature in
Passerculus sandwichensis and Ammodramus
savannarum.

Figure 3 Responses of body temperature and oxygen 11
consumption to ambient temperature in
Melospiza melodia and Melospiza georgiana.

Figure 4 Responses of body temperature and oxygen 11
consumption to ambient temperature in
Pooeaetes gramineus and PassevelZa itiaaa.

Figure 5 Responses of body temperature and oxygen 13
consumption to ambient temperature in
Zonotrichia querula and Zonotrichia
Zeucophrys .

Figure 6 Responses of body temperature and oxygen 13
consumption to airJDient temperature in
Parula americana and Vevmivora pinus.

Figure 7 Responses of body temperature and oxygen 15
consumption to ambient temperature in
Dendroica pinus, Vermivova celata, and
Mniotilta varia.

Figure 8 Responses of body temperature and oxygen 15
consumption to ambient temperature in
Dendroica covonata and Pvotonotaria oitvea.

Figure 9 Responses of body temperature and oxygen 17
consumption to ambient temperature in
Wilsonia aitrina and Dendroica dominica.

Figure 10 Responses of body temperature and oxygen 17
consumption to ambient temperature in
Seiurus novehovacensis and Seiurus auro-
capillus .

Figure 11 Responses of body temperature and oxygen 19
consumption to ambient temperature in
Dendroica palmarum and Trogon rufus.

Figure 12 Responses of body temperature and oxygen 19
consumption to ambient temperature in
Sayornis phoebe and Geothlypis trichas .

Figure 13 Responses of body temperature and oxygen 21
consumption to ambient temperature in
Empidonax viresoens and Contopus virens.

Figure 14 Responses of body temperature and oxygen 21
consumption to aribient temperature in
Myiarchus crinitus and Ty rannus tyrannus .

Figure 15 The relation of basal metabolic rate to 27
body weight in some small birds.

Figure 16 The relation of thermal conductance to 27
body weight in some small birds.

VI

Figure 17 The relation of the observed thormo- 38
regulatory quotient (Mi-,/C)q to body
weight in some small birds.

Figure 18 A plot showing the effect of body weight 38
on thermoregulation data taker, from birds
used in this study and from the literature.

Figure 19 The relation of the extent of the thermal 41
buffer to the thermoregulatory quotient
expected from body size alone.

Figure 20 A plot of the extent to which body temper- 4 8
ature is buffered, as a function of the
combined ratios (M]-j/C)j- and (M\^/C) q.

Figure 21 A plot of the degree of sensitivity of body 50
temperature to changes in ambient temper-
ature (ATj^/AT^) as a function of body size
(Mj^/Og in some small birds.

Figure 22 The relation of body weight and the ratio 50
(Hb/C)j- to the lower limit of therraoneutral-
i ty .

Vll

Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of
the Requirements for the Degree of Doctor of Philosophy

THE INFLUENCE OF DISTRIBUTION AND ECOLOGY
ON THE THERI^IOPxEGULATION OF SI^LL BIRDS

By

Charles Gerald Yarbrough

June, 1970

Chairman: David W. Johnston
Major Department: Zoology

Data from the present study and the literature show
that the energetics of thermoregulation in small birds
are accurately described by the Newtonian model of heat
loss. The level of body temperature is closely correlated
with basal metabolic rate, thermal conductance, and body
weight.

Tropical and desert birds generally have lower relative
thermoregulatory quotients than expected (less than 1.0)
on the basis of weight alone, and cold-climate species have
high relative ratios. The relative ratio is a measure of
the impact of ecology and climate on thermoregulatory
capacity. Larger body weight can compensate for a low
relative ratio, within limits.

Bird data now available indicate that ecology,
particularly food habit, limits the distribution of birds
outside the tropics and deserts, and that thermoregulatory

viii

parameters are adapted primarily to climate.

Studies of the energetics of T^ reaulation in non-

D

desert tropical birds should give information as to the
impact of ecology on this process, in that the climatic
variable would be absent.

IX

INTRODUCTION

Thermal homeostasis and its energetic cost for
endothermic animals have been extensively documented over
the last 20 years. The Newtonian model of heat loss has
been used as a unifying method of looking at some of the
parameters of body temperature regulation. This relation-
ship is usually stated as:

dQ^/dt = C (T}3 - T^)
in which (dQ^/dt) -*- is the rate of heat loss, T^^ is body
temperature, T is the ambient temperature, and C is a
proportionality factor known as thermal conductance. In
thermoneutral or cooler ambient temperatures with no
radiant energy source, it is obvious that (dO-r/dt) - rate
of heat gain = metabolic rate (M) , if the animal is going
to hold T, constant.

The zone of thermoneucrality for an endotherm is a
range of T over v;hich 11 is constant and independent of
T . If the animal is postabsorptive and quiescent, M is
equal to the basal metabolic rate (I-']-,) over this T range.
In this zone, heat balance is maintained by physically
changing the conductance (e.g., fluffing or compressing the
fur or feathers) . Ambient temoeratures below the thermo-

1 All symbols and expressions used in this paper are defined
in Appendix A.

2

neutral zone result in an increase in M, and the pro-
portionality of this increase to the (T. - T-,) difference
is a measure of C. T^ above therraoneutrality also result
in increased M values, nainly because of muscular activity
associated with panting and the van ' t Hoff effect on
chemical heat production.

The energetics of mammalian T, regulation are better
understood than those of birds. This stems from a number
of reasons. Interest in avian thermoregulatory studies
developed rather slowly, major stimuli being the series of
papers by Scholander et al. (1950a, b, c) and studies of
hummingbird metabolism by Pearson (1950). Avian species
studied subsequently have not been as representative of
the diversity in size, ecology, and distribution v;ithin
the class as has been the case with mammals. In many
instances, avian thermoregulatory studies have been in-
com.plete because of the absence of precise measurements of
the 1^ that is being regulated. Failure to report precise

T, values for M measurements often will introduce bias
b

into calculations, and greatly reduce the further useful-
ness of the data.

The review of energetics and thermoregulation in birds
by King and Farner (1961) has been of considerable value in
giving direction and perspective to this area of investigation.
From the data then available, they derived an equation
relating M, and body v;eiglit (W) in small birds {aa. 100
grams or less) and another equation for birds larger than 100

3

grams. A standard equation relating conductance and W
in birds is given by Lasiev/ski' et al . (1967). Both of
these equations will be discussed later.

A few attempts have been made at analyzing thermo-
regulation in terms of environmental demands , especially
heat stress and evaporative cooling (e.g., Bartholomew
et al. , 1962; Calder, 1964; Dav/son and Fisher, 1969; Ligon,
1969) .

Scholander et al. (1950c) have placed the emphasis on
insulation for thermoregulatory adaptation to low T . It
is now apparent (for example. Hart, 1957) that the basal
metabolic rate may also be adaptive. The adaptive nature
of body temperature is equivocal.

From Nev7ton's heat loss relationship, one could
anticipate that in response to environmental stress
(thermal and/or nutritional) an animal could (1) behaviorally
or physiologically alter the (T, - T ) difference, (2)
physically change its conductance, (3) chemically alter the
rate of metabolism, or (4) achieve some suitable combination
of these features. The behavior of these parameters in
relation to ecology and climate has been investigated in
some mammals, particularly by McNab (1966b, 1969, 1970;
McNab and Morrison, 196 3) .

The relationships among thermal and ecological factors
involved in thermoregulation have not been critically
studied in birds. Ecological and distributional correlations
have been made with differences in Mj^ and/or C in only a
few species (Wallgren, 1954; Lasiev/ski and Dawson,

4

1964; Johnson, 1968; Ligon, 1968). No attempt has been
made to synthesize into a rational whole the relation-
ships that may be of general application in birds. McNab
(1966a) has suggested that T^ in birds is determined by
the ratio of Mj^ : C, much the same as in mammals. He has
also indicated (1969) that in neotropical bats the {11^3/^)0
ratio, and thus thermoregulatory capacity, is ecologically
determined, especially by the reliability of the food
supply.

The present study compares the behavior of thermo-
regulatory parameters in response to thermoneutral or
cooler T^ for representatives of three avian families
having different food habits. It v;as suspected, in viev/
of the great mobility of temperate zone birds, that food
habit might be more influential on distribution than on
the thermoregulatory capacity. These studies on sparrows,
warblers, and flycatchers were undertaken to explore this
possibility. Representative data on thermoregulation from
the literature have been incorporated in the analysis of
physical and ecological correlates of general application
in birds.

MATERIALS AND METHODS

All specimens used in this study were caught in mist
nets. Most of the flycatchers were "obtained in the coastal
plain of North Carolina, the Harris' Sparrows"'" near
Lawrence, Kansas, and the Black-throated Trogon along the
Rio Negro near the junction with the Rio Branco in the
state of Amazonas, Brazil. All other birds were caught in
the vicinity of Gainesville, Florida, during 1968-69. The
time of year in which measurements were made for each species
is indicated in Appendix B.

The Harris' Sparrows were maintained in captivity,
but all other specimens were caught during the day (usually
afternoon) and were used for metabolic determinations the
same night. Small birds were fasted for at least five hours
and large birds six to seven hours before data collection
was begun. Only data obtained between 2100 and 0500 hours
were included in the study. Specim.ens were kept in complete
darkness during the study interval. All data from birds
suspected of being in Zugunruhe were excluded.

Metabolic rate was measured at three different ambient
temperatures: 10°, 20°, and 30.5° C, or as otherwise indica-
ted in a few cases. An open flow system was employed, and

Scientific names of all birds used appear in Appendix B.

6

air flow rates were sufficient to maintain a concentration
of at least 20 per cent oxygen in the metabolism chamber.
This .chamber had a volume of one gallon, and was immersed
in a large, constant-temperature v;ater bath. VVhen a tem-
perature change was desired, the chamber was sv;itched to
another water bath preset at the desired temperature. An
interval of about two hours was allowed for the bird to
adjust to the new temperature before data were used.

The partial pressure of oxygen in the air stream was
measured by a Beckman G-2 Paramagnetic Oxygen Analyzer
after carbon dioxide and water had been removed. Oxygen
concentrations were recorded on a Honeywell strip chart
recorder.

When a satisfactory metabolic reading was attained,
the bird was quickly removed from the chamber and the pro-
ventricular body temperature was measured by a YSI tele-
thermometer. The bird could be removed and its T, measured

b

in less than one minute. Thus, each metabolic value has
a corresponding Tj^ reading.

RESULTS

Data obtained in this study from 29 species of small
birds are shown in Figures 1-14 and are summarized for
each species in Table 1. These figures depict, species by
species, the effects of different ambient temperatures on
body temperature and metabolic rate (as measured by oxygen
consumption) . Metabolic measurements made within the zone
of thermoneutrality are basal levels (Mj^) . At T^ below
the lower limit of this zone, oxygen consum.ption increases.
The slope of a line describing this M increase with
declining T^ is considered to be the thermal conductance (C)
of the species. General patterns of these responses are
indicated later in this section.

For all species studied, the data indicate that 30.5°C
is well v/ithin the zone of thermoneutrality. This is support-
ed by the fact that most of the metabolic records obtained
at this T, were quite smooth, whereas records taken at 10°
or 20° C usually showed oscillations between maximal and
minim.al values , presumably due to periodic bursts of shiver-
ing or other m.uscular activity. Thus, metabolic values
taken at 30.5°C are probably basal rates, and represent the
lowest constant level maintained for 15 minutes. For
metabolic measurements belov; thermoneutrality, the repeatable
minima are used. This procedure minimizes both I-L and C.

7

Fig. 1. Responses of body temperature and oxygen

consumption to ambient temperature in Spi zella
passerina and Zonotriehia albicollis . The lower
part of the graph shows the relationship betv/een
oxygen consumption (M) and ambient temperature
(T^) . Horizontal lines indicate the mean basal
metabolic rate for each species. The absolute
value of the slope of each line in the lower part
of the graph represents the mean thermal conduct-
ance for each species. The upper portion of the
graph gives the body temperatures (Tj-^) corresponding
to the M values belov;. The slanted line in the
upper part is a reference which equates the temper-
ature axes.

Fig. 2. Responses of body temperature and oxygen con-
sumption to ambient temperature in Passerculus
sandwichensis and Ammodramus savannarum. See the
legend of Fig. 1 for explanation.

r

i"
u

1 ;.■"•■

10

• «

c >

■rK^wrr'^'v g.]W-»"F'.'Jy^*r^i^~. Vi-— ^"'^J^ryw is' ' »^ "t anuwa**-"

l^^.-c)

^Spi^plla pasr.gfina |;

" /fonotrictna ?ji;lr;pjlis_ l'

tJ

•::jcii35S^«:^-j^' ffiris?^E^r?:^:niT..'xz?l3;'i-dEl*E'-!k

20

7.1 Cc )

30

40

f3"

[

!»

";: <

3iT!ijiBatreOT«j!nM!!B«2i=iCTvjmr!K=»c««ra»=s=air^

., ('.

^ ,<'

t!tl

139

.,' J3 b

':j3 ^

li:

i^ :.

ii*^Tnscj:-:£T.-^c'.':rr:r/fi^^\^

if

r

** A.nrnodramijs savannaruni..

■"^

"•^^

II, 20 i(] »--.^j.-.».^

Fig. 3. Responses of body temperature and oxygen

consumption to ambient temperature in Melospiza
melodia and Melospiza georgiana. See the legend
of Fig. 1 for explanation.

Fig. 4. Responses of body temperature and oxygen

consumption to ambient temperature in Pooecetes
gramineus and Passerella iliaca. See the legend
of Fig. 1 for explanation.

11

T3 Co

I

O 2

V°c)

Tare)

Fig. 5. Responses of body temperature and oxygen

consumption to ambient temperature in Zonotrichia
querula and Zonotrichia leucophrys . See the legend
of Fig. 1 for explanation.

Fig. 6. Responses of body temperature and oxygen
consumption to ambient temperature in Parula
americana and Vermivora pinus . See the legend
of Fig. 1 for explanation.

13

Ta (\)

o
8

20 30

Tare)

Fig. 7. Responses of body temperature and oxygen

consumption to ambient temperature in Dendroica
pinus , Vermivora aetata, and Mniotilta varia.
See the legend of Fig. 1 for explanation.

Fig, 8. Responses of body temperature and oxygen

consumption to ambient temperature in Dendroica
ooronata and Protonotaria citrea. See the legend
of Fig. 1 for explanation.

15

Tf,(°c )

iPoiidroica piriu?
□Verrnivo^a f.el;it3_

T^LSji^'iiLsXESsaihiiiiEdssEa;^^

Talc I

f

O !

- 2

J

f

l

tOcndroica coronala
-•-y-Prctonotaria citrea

TJ"c)

20 30 40

I

TaTr.

Fig. 9. Responses of body temperature and oxygen

consumption to ambient temperature in Wilsonia
citrina and Dendroica dominioa. See the legend
of Fig. 1 for explanation.

Fig. 10. Responses of body temperature and oxygen
consumption to ambient temperature in Seiurus
noveboracensis and Seiurus aurocapillus . See
the legend of Fig. 1 for explanation.

17

20

Ta Cc )

I

r

f

I.

■■■■J

10 JO

X Spiurus noveboracensis
• S. aurocaiiillus

Ji

I I

aJ

1 a ( \. )

Fig. 11. Responses of body temperature and oxygen

consumption to ambient temperature in Dendroica
palmarum and Trogon vufus . See the legend of
Fig. 1 for explanation.

Fig. 12. Responses of body temperature and oxygen
consumption to ambient temperature in Sayornis
phoebe and Geothlypis trichas . See the legend
of Fig. 1 for explanation.

19

r

\

>*sg'gTafjgtTr.^;:L-jjJtS3S?iasg?3g^aij.Tif-r;r-'t-

,}{

o

u

ttPenciroica palmgrumj
'^ Tro.g.nri fuUiS

TaCc;

7"

io

»

-4

a

«

K

.^sa

.■

•

■

K

u 4

^l>

■.»•!,

■i£

micjxTS:^rrtn-c-a«s:SiW;

?

• Sayomis phoebfl

X Gcothlypis tri.-has

lEJ

ni "z v^^sstDfsssijsnz:/: ^irr^

Th(°0

...;.^..

-:-^^--.rjT3;:j^!jErv^£;:S2-H::i£iilT2;

30 40

Ta (°c)

Fig. 13. Responses of body temperature and oxygen

consumption to ambient temperature in Empidonax
virescens and Contopus vivens . See the legend
of Fig. 1 for explanation.

Fig. 14. Responses of body temperature and oxygen

consumption to ambient temperature in Myiarchus
ovinitus and Tyrannus ty r annus . See the legend
of Fig. 1 for explanation.

'21

o

10

TaCi

I

i

1 .

1l •

••X

•••

-* Myj^rcliu'; Cf initus

10 ^0 30

Ta (°c )

1"

••a 4 1

'lO

"•{• 39
38

I

'^(•c)

37

22

■a

m

1

B

A

n

H

0

6

«

0

«

3

u
o

•H

V

b
V

c

V

(A

0.

^
H

J3

u
a,

Ul

o
I

o
I

9\ 00

(A

N O

a

F

p.^

CO

0

to

*

ta

o>

0

t.

t1

3

••4

.c

t^

•c

<X

^1

3

<»

o

0

o

O

4a

•t*

t)

t^

t.

w

c

3

«

u

o

u

0)

F

03

"»

o

*vt

(D

o

r

,

0

o

o

.

*

0.

a.

N

N

«H CM

J3

u
a

z

u

23

in in CM IN

fO <M M

(U

u

r-i ro in in lo
f*> n ^ in in

O

U

^ 00 in (Tt ID

r* \c m vD CO

:^

<N (N fN

o

3

c

c
o
o

01

o

o

o

«

'ri

c

•♦i

t--

o

■w

p

0

3

-u

«

«.

M

(1

«

r

•ti

F

o

o

«-i

c

t.

a

o

01

ff

3

•.i

r;

o

Q-

e

u

ti

O

Is

■y

•»j

■ii

O

m

a.

r

Jl

n

F

Q

c

><

t«

tj

to

24

^

H

u

^
^

o

3
C

c
o
u

^

H

u
a>
a
in

t.

iH

in

«-t

0

0

X

•i

X

,

>.

(U

»H

<-^

00

U)

a

^

03

O

M-1

m

m

O

in

in

o

o

o
1

o

u

o
1

c

n

S

(

•o

S

0

0)

S

0

■H

a»

>

00

■u

u

(N

00

u

-v

«

o

^

a)

00

■

*

§

(M

r-

00

U)

t

(U

+J

\

.

a

o

•>H

^

to

:^

o

0

\

u

e

<u

<u

\

\

.H

\

■H

o

^

T)

u

:^

:^

rH

n)

01

TJ

i-l

u

>

>

Q)

■o

o

\

«

0

u

>

u

1

0

M

4^

§

CJ

Ij

>

s

*J

VI

•D

^

u

3

<

at

XI

VI

m

0^

^

dl

X

E

0

J3

0

J3

r4

^

C

■^

0

c

0

r*

• —

p

C

i-i

to

10
0

§

II

II

II

CI

0

B

^

j:

a

(0

£

0

0

kl

4J

a

u

<

0)

II

u

U

u

k<

II

\

\

^

U

0)

0

§

df

^

2^

^

1

U)

;s

o

*-'

*^

>J

25

The mean M, for ten species of sparrows is 92.6 per
cent of that expected from the equation of King and Farner
(1961) for small birds, and C is 75.0 per cent of the value
expected from the equation of Lasiewski et al . (1967) (Figs.
1 - 5, 15, 16, and Table 1) . Of the families studied here,
the sparrows have the most northern v/intering distribution,
and contrast with the tropical Black-throated Trogon, which
has a M, only 77 per cent of that expected and a very high
C (161 per cent) (Figs. 11, 15, 16, and Table 1).

The mean M, for all v/arblers is 89.2 per cent of that
expected and C is only 72.4 per cent of the expected value
(Figs. 6 - 12, 15, 16, and Table 1). Eight species winter-
ing as far north as the southeastern United States have
lower M, and C values (85.8 and 66,5 per cent) than five
species wintering farther south (94.6 and 82.0 per cent).
These values indicate a considerable reduction in energy
expenditure.

The five species of flycatchers have a collective M,
which is 100 per cent of that expected. C is 83.6 per cent
of that expected, and the Acadian and VJood Pewee have a lower
C (75.5 per cent) relative to that expected than the larger
flycatchers (89.0 per cent) (Figs. 12 - 14, 15, 16, and
Table 1) .

Each metabolic datum is accompanied by a T, value
(Figs. 1 - 14) . The mean T, for sparrows and warblers in
thermoneutrality was identical (40.3 °C) , but v/as less than
that of flycatchers (41.1 °C) . Very small birds (<10 grams)
usually have Tj^ that are lower and more sensitive to cool
T^ than those of larger species.

Fig. 15. The relation of basal metabolic rate to
body weight in some small birds. The line is
a plot of the King-Farner (1961) equation for
small birds (log M]-, = log 7.29 - 0.341 log V7) .
Each point is the mean value for a species.

Fig. 16. The relation of thermal conductance to

body weight in some small birds. The line is
a plot of the equation of Lasiewski et al . (1967)
(log C = log 0.84 8 - 0.50 8 log W) . Each point
is the mean for a species.

27

• Warblers

O Flycatr.hcrs
X Cp;trro<i.s
A Trogon

oo.?0 -

X X

• vVa'blers
A Trogon

A_

i<i«.KKy*l'»i«»i..''.nAjJ^«r^»fclci-i

O.S 1.0

'•? I.J l.< 1.5 l.S l.I

lOG W

Mi Ill '■■ \Mt^^-^ »T»iJ^.^^J..r.w»J.....J .. ■■ J 1 J LrJi^.l iTVT—

' 10 IS 20

DISCUSSION
Keat Loss and Keat Gain

Heat loss in animals is an extremely complex process.
The factor C conceals a multiplicity of problems concern-
ing heat loss by radiation, conduction, convection, and
evaporation of water. The effective body surface area may
be temporally varied, peripheral and appendicular blood
circulation is subject to alteration, and the temperature of
the body surface from which heat is lost may be different
than that measured in the core. Some of these problems
have been discussed in detail by Burton (1934), Scholander
et al. (1950a, b, c) , Kleiber (1961), Bartholomew and Tucker
(1964), and McNab (1970).

It would be desirable to have a rigorous experimental
capability such that the physical parameters could be
measured which are necessary for using models such as
Fourier's law of heat flow. The ability to apportion properly
the responsibility for heat loss among radiation, convection,
and conduction as well as evaporation is also a worthy
aspiration. However, as indicated in the previous para-
graph, there are a number of reasons why this is impractical,
if not impossible, at present. It is important that we
begin to measure quantitatively the partitioning of heat
loss by birds. This partitioning is probably highly adaptive.

28

29

Knowledge of trends of performance by each avenue of heat
loss would allow us to understand the thermal and ecological
interrelationships of individual species. The recent
studies by Hamilton and Heppner (19 67) and Heppner (19 70)
on radiative energy exchange and its effect on the heat
budgets of dark and light birds are steps in this direction.

However, the details of the partitioning of heat loss
are overlaid by m.ore general thermo-ecologic patterns. In
the final analysis, it is the total rate of heat loss that
must be balanced by heat production. It is exactly at this
point that the Newtonian model is useful. Its employment
simply ignores the various reasons for heat loss, and con-
siders the totality of heat lost as a single parameter.

This approach, although simplistic, allows examination
of the relationship of metabolic heat gain (M) , and total
heat loss C (Tj^ - T^) , to biological and distributional
characteristics of a species, such as ecology and climate. It
will be seen that the available data fit this simple approxi-
mation remarkably well. As long as the physical limitations
of this concept are understood, I believe that it can be of
considerable heuristic value in a biological sense.

The Newtonian model v;as originally used to describe
the cooling of a body. Since a decline in temperature does
not generally occur in homoiotherms , it is sometimes argued
that this law is not applicable to such systems. The argu-
ment becomes rhetorical when the cooling expression is re-
written as a heat loss statement using the specific heat (c)
and body weight (W) :

30

dT^/dt = k (Tj^ - T^) (1)

dTj^/dt = dQ/dt • 1/cW (2)

.-. dQ/dt • 1/cW = k (Tj^ - T^) and (3)
dQ/dt = kcW (Tj^ - T^) or (4) .

dQ/dt = C (Tj3 - T^) (5)

where C is a composite proportionality factor commonly
known as thermal conductance.

Metabolism
Of the species of birds used in this study, only a
fev7 sparrows have been subjected to metabolic examination
by other investigators. Table 1 shows that sparrows
larger than 20 grams have l-l almost exactly as predicted
from W by the King-Farner equation, except for the Vesper,
which also has a lower T . King's (1964) data on the I"7hite-
crovmed Sparrow agree with those obtained in this study.
Rising's (1968) values for Harris' Sparrow are much higher
than expected (158 per cent) and were obtained by weighing
CO produced. Ily data for the White- throated Sparrow are
similar to those of Hudson and Kimzey (1964). The values
given by Lasiewski and Dawson (1967) for \'7hite-throated , Song,
and Fox Sparrows are considerably higher than those from
this study. This may be due in part to differences in pro-
cedures of taking data from the strip charts. Figure 15
shows the position of my data in relation to the King-Farner
curve for small ]jirds.

It will be noted from Table 1 that those sparrows
with K lower than expected usually also have lower Tj^ than
the other species. M, for most of the v;arblors are also

31

somewhat lower than expected, particularly in species winter-
ing in Florida or in very small species. This reduces
energy expenditure considerably and may be very important
for small insectivorous species wintering in cool climates.
The reduction in Mj^ is made possible in such species by a
comparatively better insulation. The two species of insect-
ivores that winter in the southeastern U. S. and still main-
tain relatively high Mj^ (Phoebe and Myrtle V7arbler) are
known to depend heavily on berries for food during winter
(Bent, 1942, 1953; Yarbrough and Johnston, 1965).

The tropical and desert species in Table 2 have sig-
nificantly reduced Mj^ values in most cases. This may
function mainly for reducing energetic demands (small
hummingbirds) or for moderating the heat load and evaporative
water loss (desert owls and goatsuckers). The species
which appear to take exception to such strategies are either
very small (Black-rumped Waxbill) or have exceptionally
high C values, even in view of their small body size
(Paradise Widowbird and Zebra Finch) .

Lasiewski and Dawson (1967) have statistically separated
standard metabolism curves of passerine and non-passerine
birds. No biological rationale is evident for this decision.
Indeed, there iriay be differences in n^ between passerines
and non-passerines; there probably are different curves for
penguins and hummingbirds, auks and vultures, etc. Non-
passerines have been inadequately sampled over the size
range from which most of the passerine data have been

32

ox

•-I

k

m

vo

a\

V)

(0

u-x

U)

-H

c

•-4

0^

f-i

's

:t

ffv

u

»

11

•H

<u

0)

tH

^

0

V)

■H

•H

VO

2

U

»^

•i

ON

<k

(n

Ul

k

Ox

01

<U

1

fl

»--i

vO

t-A

2

nj

»H

u

s:

a

o

<T»

*^

a

>J

fj

*4-t

■H

4^

•0

4^

*H

a

»

•0

•0

a

U

ei

b

c

Tt

*

T

n

Q>

c

c

«.!

v^

K

*

fl

<r

00

if

a>

'O

u

ID

m

H)

m

o

•o

O

V4

V0

vD

o

VO

a\

u

X

^o

§

k

■H

Q

o\

a\

<T\

<?^

r-t

•H

0

•H

•H

■.^

0

M

T)

r-t

•H

.-t

.-H

^

Tl

is

M

j;

Tl

l-t

•M

Ul

C

%

(/)

c

CO

«

o:

C

C

0

;i

(tJ

«

«

•.

«

M

•?

Ifl

■*

1

3

fl

«

0

j::

<K

rH

c

C

C

c

0)

(U

^

11

(U

U)

r

U)

4J

■f-4

0

0

0

0

<)

TJ

o

•M

•H

•w

•u

o

s

l4

lA

x:

!7*

tp

a>

0^

T-H

w

J2

Ul

in

Ul

Ul

Ul

10

<!)

13

o

■H

.-1

■H

(tJ

(Q

tJ

10

«

10

0)

■H

Q

a:

^

w

^J

.J

•J

kJ

u

>-3

ui

I-)

h]

vJ

If

S

o

(U

O
U)

§

u

n)

a

c

^
H

p- as ^

U

O

^

«
a

rH »H rH 1-4 O

O^

u>

C^

fM

in

r^

r*

fO

r-

■V

00

o

t-i

O

ro

M

n

CM

(N

(N

<N

t-t

o

tH

r-

•H

in

^0

<T>

i-t

.-*

CD

IN

in

f -t

*£•

o

U>

r-

a*

u-i
in

O
00

a

0

b-

O

3

o
a

t3>
O

3

K

«.
O

c

<3
•J

33

■o

•a

u

u

0

0

H

Eh

VJ5

■0

•O

<T\

C

C

1-1

nj

10

^

c

C

(1)

0

0

4J

m

w

j;

3

s

tp

li

10

o

O

Q

>

u

4-

o
u

u
o

o

3

c

•H

c
o
o

u
o

(0

a
o

p

o

^^

•

^

o

fH

, .

CO

tn

a.

rr

CO

o

(M

01

m

o

,

in

o

0

o

1

o

u

o

1

c

ro

S

1

■0

S

0

a)

3

0)

■H

a*

>

CD

4J

1-1

(N

CO

^

^r

a

0

T

o

00

•

•

c

^

r~

00

M

0

-p

fO

\

.

XI

o

■M

<-(

in

^

o

0

\

u

E

<u

G

X

\

rH

\

■H

3

•a

u

^

^

rH

re

fi)

■0

•

^-^

tH

u

>

.">

m

•a

o

\

re

o

^

>

1}

1

0

u

tM

§

Q)

u

>

?

jj

m

Ci

u

u

3

<

0)

XI

i;i

a

«n

V,

O

X

e

0

XI
0

0)

X3

<N

^

c

■H

o

c

0

1^

g

■o

14

a

c

c

re

a)

(3

II

II

II

d)

(U

e^

QJ

£

a

w

E

o

o

kl

4J

Dj

u

^

V

II

u

u

u

M

II

X

\

\

OJ

o

1

w

XI

jf

=^

g

m

z

u

*^

^^

^^

34

obtained. Perhaps even more important than body size are
the cliraatic and ecological factors impinging on the birds
in question. It might be expected that birds from the
tropics or arid areas with great heat loads, soaring or
gliding species, flightless birds, would have reduced
metabolic heat production v/hen compared to active foragers
contending with cool, cold or variable T, in the absence of
a capacity for torpor. Many more data are needed for
passerine and non-passerine birds having similar distri-
butional histories, ecological characteristics, thermal
environments, and body sizes before it can be said that
physiological responses are determined by taxonomy at the
ordinal level.

Conductance

As has been seen above, it is sometimes meaningless or
even misleading to view heat gain without evaluating heat
loss. I have already indicated that the procedure of data-
taking used here minimizes C. Data are available in the
literature for only two of the species studied by me.
King's (1964) C data on the White-crowned Sparrow are similar
to those reported here, but Rising's (1968) C values for
Harris' Sparrow are nearly double those in the current study.

The mean C for every species in the present study is
lower than that predicted by the equation of Lasiev^ski et al.
(1967) , except for the Black-throated Trogon (Fig. 16) .
This completely tropical species has a C on a par with the
Spotted Nightjar of Australian arid lands (Dawson and Fisher,
1969) .

35

It seems to be generally true that birds from the
tropics or hot deserts have relatively high C values, and
species that encounter cool v;eather have low C. However,
there are some prominent exceptions to this generalization.
Very small desert or tropical species, such as humming-
birds may not be able to tolerate increases in C (also
compare the Black-rumped Waxbill with the larger Paradise
Widowbird and Zebra Finch) . The smaller caprimulgids with
more temperate wintering distributions (Poor-will and
Pauraque) have relatively better insulation that the larger
Nighthawk and Spotted Nightjar that winter at lower
latitudes.

The small nocturnal owls have good insulation (Table
2). Of course, the Saw-v/het is found in cold climates, and
the Whiskered and Elf Owls are active during the cool
desert nights. The only small owl in Table 2 with a rela-
tively poor insulation is the Pygmy, which is largely diurnal
m its activity and thus is exposed to the high T of the

3.

southwestern U. S. and Mexico.

The importance of providing Tj^ measurements along with
M values is seen in the calculation of C. Some species
of birds probably do not change abruptly from physical to
chemical thermoregulation (Figs. 7, 9, 12, 13). Many species
also experience a reduction in T^ when T^ is below the thermo-
neutral range. These effects are additive, and under such
circumstances the slope of the regression line through the
data points does not represent C. Minimal C is the slope of

36

a line connecting the mean M value at a given T^ below
therir.oneutrality with the corresponding mean T^ at M - O.
The lower limit of thermoneutrality for purposes of calcula-
tion can be determined by translating this line to the right
on the X-axis to the point which is the Tj-, measured at
thermoneutral T^.

The Physical Model
The relation of U^ to C for birds in the current study
is compared to a line representing predicted values (Fig.
17) . It has been stated previously that this relationship
of heat gain (M) to the coefficient of heat loss (C) is a
better expression of the thermoregulatory capacity of an
animal than either parameter alone. Each of these factors
can probably be modified by selection and/or acclimatization,
but only with proper reference to the capacity of its com-
panion factor and body size. For example, the response to
an environmental alteration of C can be made only to the
extent that body size is not limiting and that the animal
can obtain an adequate food or water supply to operate at
the new M level. All this must be done within the frame-
v7ork that exists for m.aintenance and reproductive success.

Therefore, the observed thermoregulatory quotient
(Mj^/Oq is broadly composed of two components: {l^^^/C)^, the
quotient expected of a "typical" bird of the specified W,
and (Mj^/Oj,, an index of that variation from "typical"
that may be attributed to the adaptive impact of ecological
and climatic demands. The modified heat loss equation can

Fig. 17. The relation of the observed thermoregulatory
quotient (^^b^^^o ^° body weight in some small
birds. The line is a plot of the quotient expected
on the basis of weight from the equations of
King and Farner (1961) and Lasiewski et al . (1967),
log (Mu,/C)_ = log 8.59 + 0.167 log W.

Fig. 18. A plot showing the effect of body weight
on thermoregulation data taken from birds used
in this study and from the literature. Lines
have slopes equal to (M}-,/C)g. Data for the
White-tailed Ptarmigan {Lagopus leucuvus) are
from Johnson (1968) . Symbols are as in Fig. 20.
See Tables 1 and 2 for data and references.

38

1.4

«

- 1.6

■ I!

IS

-. — .

1".^

g

I.I

*

. o

10

I.C

■

- OS

1_X

■II ^la nil j 1 1 WH ^■■■■■■■ii^ iii,«i.Biimn 11^ iMii i> limm
0.5 1.0 1.1 I!

y 0

• Wd'blc'S
o f lycslciic
y ^ X *■>! Jf'Owi

X I A TfOKon

I.] 1.4 l.» I.E

IS 20

iC

^K^jsamBj^maiaA>/Hu V 'itA-*si^'.imt ir* — i i^ ■ !■■■ wi m^i ■ni» ■■■iwmLhwi \m
• (.3 0 4 0.{ ol IP

39

be rearranged and partitioned so that:

■^b = ^^/C^o + K and ■ (6)

Tb = (Mb/C)r (^%/C)e + K (7)

where K is the lower limit of thermoneutrality (McNab, 1969,
1970).

Since equation (7) has the general form y = ax + b,
the fit of available data to this model can be checked.
A biological check is also possible by rearranging equation
(7) so that:

Tj3 - K = (Mb/C)j- (M^/Oe (8)

where Tj^ measurements on the left side of the equation are
compared with metabolic measurements on the right. Graph-
ically represented, when {T-^ - K) is plotted against (Mj^/C) ,
a line from the origin having a slope of (Mv^/C) should in-
tercept the corresponding data point (Fig. 19) . In a more
dramatic way, when (Tj^ - K) is plotted against (Mj-^/C)j.
(Mj^/C)g (equation 8), the data should fall along a straight
line having a slope of 1.0. In Figure 20, which is a plot
of this relationship, the closeness of fit is evident.
Determination of the Level of T^

It is apparent that K is determined by Mj-j and C, both
by the weight factor [lA-^/C] q and the relative ratio (M^/C)^.
Therefore, adaptation to climatic and ecological conditions,
along with W, sets the lower limit of the energetic quantity
(T]3 - K) . This difference is the buffer between the bird's
regulated Tj^ and the necessity for increasing energy expendi-
ture. The magnitude of (Tj^ - K) is a good indicator of the

Fig. 19. The relation of the extent of the thermal

buffer to the thermoregulatory quotient expected
from body size alone. Numbers at the left
terminus of each line are slopes. The number
accompanying each point is the (Mj^/Oj. value. The
mean (Mb/C)j- value for encircled points is equal
to the slope of the line through the circle.
Symbols are as in Fig. 20. See Tables 1 and 2
for data and references.

41

o

o

it-:
I

■o 15 -

C-Vi^i^//! I I ! ! I I I ...1 J.^Jo

° // iO 12,.. V '" 16 '8 2°

V C /c

.. 8.59 W"-^"

42

extent of cold stress to v;hich a bird is subjected in its
habitat.

The question as to whether Tj.^ is a dependent or in-
dependent parameter is more complex. It is logical to
agree with Hammel's (1968) contention that an initial step
in the evolution of homoiothermy was necessarily the acquisi-
tion of an adequate insulation. Then, it becomes energetically
feasible to invest in a fairly high M, . The eventual thermo-
regulatory quotient and size of the endothermic avian stock
likely resulted in a (Tj^^ - K) that was satisfactory for the
thermal environment in which these animals lived. This
original T, would have been dependent on II , C, and W. As
the advantages of thermal homeostasis asserted themselves
(whatever they may be) , it is reasonable to assume that
temperature-specific neural, enzymatic, and other systems
also evolved. These systems could be more efficient if T-^
were precisely regulated. So, there must have been a
parallel evolution of the energetic machinery making endo-
thermy possible and a regulating system that would maintain
Tjj stability in the interest of physiological efficiency.

Ontogenetic evidence lends support to this speculation.
Nestlings of some altricial species of birds show a
chronological improvement in insulation, and at some rather
well-defined stage begin a rapid development of metabolic
heat production and thermoregulatory capacity (e.g., Dawson
and Evans, 1957, 1960; Yarbrough , 1970b). In cool environ-
ments, such nestlings first are able to regulate their Tj^
at a lower level (e.g., 31 - 33 °C) than that of the adult

43
(about 40 °C) . The regulated Tj^ increases with age until
the adult level is achieved. So, it would seem that the
regulatory mechanism becomes functional at a stage when
(\/C)o ^"^ ^^ reach an acceptable level, but the set point
of the thermostat (the reference temperature for the
regulating mechanism) is elevated as (Mj^/C) q and W increase.

For the original data presented here, as well as for
the data from the literature, equation (6) predicts Tj^
with an average error of only 0.3 °C. One might well have
expected experimental and mechanical error to exceed this
value. Thus, the level of Tj^ is very closely correlated
with the interaction of Mj^ , c, and W in birds, as it is in
mammals (McNab , 1966b, 1969). However, I feel that the
existing evidence does not warrant classification of Tw
as a completely dependent factor, except possibly in an
evolutionary sense. Of course, it is true that in thermo-
neutrality a given Tj^ depends on a particular balance of C
and M^, but the Tj^ set point may influence the level of M .
The distinction lies between physical and physiological
dependence. In the absence of the proper interrelations
among M-^ , C, and W, the thermostat is ineffectual. Under
usual circumstances the regulating mechanism acts as a fine-
tuning device within the limits prescribed by the thermal
characteristics.

It may be that the 1-^ set point is an integral part of
the evolutionary thermoregulatory complex, and, as a result,
is to some extent adaptive. (In this case, M^^ , C, W, and Tj^
might be more accurately termed interdependent.) Scholander

44

et al. (1950c) and most subsequent investigators have
concluded that Tj_^ is non-adaptive, but there is now some
evidence that mammalian T, levels may be capable of some
adaptive variation independent of !l , C, and W (McNab,
197C) . However, the avian class apparently shows much less
variation in T^^ than is the case in mammals. This reduces
any adaptive effect of T , and increases the significance of
ecological and climatic impact on M, and C.

Body Size Effects

The impact of V? on thermal relations is considerable.
This is largely due to the change in surface : volum.e ratio
with weight. Surface area (A) is theoretically the most
satisfactory unit for heat loss calculations, except that it
is practically impossible to get an accurate estimate of A.
This is especially true in birds, where one would have to
decide what surface area must be considered. For example,
what relation does A of the exposed parts (legs, head, beak)
bear to the feathered surface? Alterations in peripheral
circulation, particularly to the extremities, and postural
changes complicate definition of the effective surface area.
Use of kW^/^ as a value of A does nothing to solve these
problems; it simply obscures them. Therefore, it is con-
sidered that, at present, W is the most undeceptive unit
for discussing body-size relations in birds.

The relationship of fV to W in small birds (<100 grams)
has been described empirically by King and Earner (1961)
as following an equation equivalent to :

!'^ (cc02/g - hr) - 7.29 w'^-^^^l (9)

45
where W is in grams. Likewise, Lasiev^ski et al. (1967)
considered C to be related to "W (grams) by the equation:

C (cc02/g - hr °C) = 0.848 w-0-508 (^q)
which is similar to the C equation of Herreid and Kessel
(1967). Thus, both Mj^ and C increase as W decreases.
However, lA^ increases less rapidly with small size than
does C, and the thermoregulatory quotient increases as
W increases:

(Mj^/C)e = 7.29 wO . 341/o . 848 W-0.508 (n)
= 8.59 wO-167
where W is in grams. So, it can be seen that a species can
attain a given (Tj^ - K) either by adaptively changing
(r-lj-j/Oj. or by possessing a suitable body size (Mi^/C) . The
effect of W on (Tj^ - K) v/hen (Mj^/C)^ is constant is shown
in Figure 18.

McNab (1969) suggested that there is a "critical"
weight in mammals below which T, is dependent on W, but
above which Tj^ appears to be independent of W. Departure
of (M]^/C)j- from 1.0 can compensate for W changes, and thus
alter this "critical" weight. A "critical" weight is also
indicated for birds (McNab, 1966a, 1970). So far as can be
determined from published accounts and the data presented
here, Tj^ in birds appears to be W-dependent throughout the
size range, with the highest T,^ values occurring at about
20 grams. Above this size, Tj^ declines very gradually, and
below it, precipitously. Much of the data scatter is
probably due to the (M^/O^, effect. Apparently the "critical"

46

weight is also variable. In warblers it occurs at 10-11
grams, in flycatchers at about 15 grains, and at 18-20 grams
in sparrows. The significance of these apparent differences
is not now obvious, although it may be similar to the case
in mammals (McNab, 19 70) in which temperate species have
lower K values than tropical forms.

Precision of T-^ Regulation
The birds studied to date fall into two categories:

(1) those that regulate very well at moderate T at all times
until W (energy) loss results in weakening and death, and

(2) those that regulate very well at T^ of 10-30 °C but

can go into torpor in times of stress or inactivity (humir.ing-
birds, goatsuckers, some swallows and sv/ifts). Thus T^^ in
birds would superficially appear to be less sensitive to
T^ changes than is the case in m.any mammals. It should be
noted, however, that most of the species of birds which one
would suspect as being poor thermoregulators are found in
the tropics. The physiology of tropical birds is essentially
untouched.

Birds employed in the present study are adequate to good
thermoregulators. The change in T, per unit change in T^
(AT, /AT ) is a measure of the sensitivity of Tj_, to T^.
Figure 21 shows that there is little change in Tj^ over T^ from
30° - 10 "C. Although there is a greater change in the smaller
species of each family or those having a low (Mv,/C)^, the
poorest thermoregulator is precise when compared with values
for some mamjnals (McNab, 1966b, 1969). Values of ATj^/AT^ >
0.10 are found only in those species v/ith both very small

Fig. 20. A plot of the extent to which body temperature
is buffered, as a function of the combined ratios
(Mj^/Oj- and (Mb/C)g. The line is expected from
Newton's lav/ of heat loss. See Tables 1 and 2
for data and references.

48

24

©Warblers
0 flycatchers
X Sparrows
ATrogon
AHumminEbirds
FjlCaprimuIgids
I /f Small Finches

^C Vidua
Ik Crossbills
jn Jsys

i-' Small Owls
iX Evening Grosbeak

12 IC 20 24

j~ f — - '
28 34 36

m m

Fig. 21. A plot of the degree of sensitivity of body
temperature to changes in ambient temperature
(ATj-j/AT^) as a function of body size {11^/ C) q in
some small birds.

Fig. 22. The relation of body weight and the ratio
(MV)/C)-j- to the lov;er limit of thermoneutrality .
Arrows indicate the climatic zone v/hich is the
wintering limit for each general food habit type.
See the text for further explanation, and Tables
1 and 2 for data and references. Symbols are
as in Fig, 20.

0.0 2

0.05

O.IO

Ta = 30.5 »L

,1.15

IJI

1.36,

,1.16

1.28

1.3!
•U6

,W

>.3i..29

/OO^UB

1.21

.1.03

,1.10

,1.19

U3

,I.M

J.23

50

• Wtifblers

0 Flycatchers
X Sparrows

,1.45

0

.1.38

,109

1.12

JJJJ.X1-IJUJ^JJIJJJ^UJJJJJLXL11JJJJJ^

(»e = «•" -"■"

Food h^bil nofi - Itmiting

Wintering large, aerial insect ivores
with capacity for torpor

Nectar (eetJers -,

WinlRnng tcaging ins*-ctivore5

WmtBdiK gianivores
and carnivores

"i^n — I — r-T — i — I — 1 — I — I — I — r~"T — i C

orf~T^^

51

body size and low (Mj^/C)^., the Parula Warbler and the
Acadian Flycatcher.

-The amount of energy reserves (fat) is an important
factor in the precision of regulation. :7eight deviations
greater than five per cent from the species mean are
positively correlated with Tj^ deviations greater than
0.5 °C from the species mean. Such an analysis will account
for 83 per cent of the variation in warblers, 77 per cent in
flycatchers, 61 per cent in Zonotvichia sparrows, and 100
per cent in sparrows smaller than 20 grams.

Adults of small species v/ith altricial young probably
live near the limits of their energy reserves during the
breeding season, particularly during the nestling period.
This is an especially acute problem for insectivorous
species, if that season happens to have frequent inclement
v;eather. The frequency of thunderstorms during the nestling
period of 19 69 was abnormally high in eastern North Carolina.
Results of several metabolic tests of the small flycatchers
(particularly the Acadian) had to be discarded because the
birds were in swch poor nutritional condition. If Acadians
were captured before noon prior to nocturnal testing, some-
times they V70uld weaken and die by the following morning.
Even in a species that forages and f lycatches , such as the
Blackpoll Warbler {Dendroiaa striata) , body lipid content
may be almost completely exhausted during parts of the
breeding season (Yarbrough, 1970a).

Ecological and Distributional Relationships

Some of the v;ays in which ecology and climate may affect

52

the behavior of various thermoregulatory parameters have
been discussed above. It is my intent in this section to
summarize some of the general relationships that exist
between the physical model and patterns of avian dis-
tribution.

Consideration of all the factors in the heat loss
model accurately accounts for non-experimental variation
in the data (Fig. 20). The various species spread out
along the theoretical line, with very small birds or desert
and tropical species nearest the origin, and birds that are
larger or adapted to cold T farthest from the origin.

A more meaningful representation of this information
is given in Figure 22. The lowest thermoneutral T for a
species is connected with the origin. The data point is
placed where this line intersects a line between the
weight and (li^/C) characteristics for the species under
consideration. These thermoregulatory parameters can then
be related to climatic or latitudinal distribution. The
climatic zone which is the winter thermal limit for birds
having certain food habits is also indicated. The picture
could be more precise if we understood the impact of
ecological factors other than food habit on the distri-
bution of birds.

Lines A, 3, and C are empirical thermo-ecologic
boundaries for small birds up to about 100 grams in weight.
Movement from one zone to another is possible by changing W
or {H^/C)j.. For example, the birds in the tropical-desert

53

zone are not generally cold-stressed in nature and have
become adapted for coping with- heat stress and water scarcity.
This may involve being very small (small finches and Paradise
Widowbird) or having a low thermal index (T, - K) , such as
that of the Black-throated Trogon and some caprimulgids.
The capacity for torpor serves as a buffer to reduce energy
outlay in times of food deprivation and occasional cool
weather.

Birds wintering in the cold-temperate to subarctic
zone must have a high (M, /C)^ and cannot be smaller than
about ten grams. The thermoregulation of northern chickadees
{Varus hudsonicus and P. cinctus) should be exajnined from
this standpoint. Food habit is U3ually limiting in this
zone, since only herbivores and carnivores (including scaven-
gers, such as gulls and ravens) are found here. Steen
(1958) has found that some small, cold-climate species
can allow T, to drop as lov/ as 30 °C at night as an energy
conservation measure. He also suggests that under natural
conditions these birds avoid severe cold stress by means of
their roosting habits.

The majority of species used in this study fall into the
temperate zone, having K values between 22° and 25° C. It
is of interest to note that all the flycatchers studied,
except the Acadian, are thermally capable of wintering in
the temperate zone. This is an obvious case of avian dis-
tribution being limited by food habit (and possibly evolu-
tionary history) . The warblers that winter as far north as

54

the southeastern U. S. fall into zone C (temperate) , where-
as those that winter from the -Caribbean area southward
fall into the subtropical zone on the graph. The winter
range of the Pauraque does not extend as far north as that
of the Poor-will. Yet, on a thermal basis the Pauraque is
much better equipped to regulate at low T^. It would be
germane to know if the Pauraque can or does enter torpor as
readily as the Poor-will. The lower K-^ of the Poor-will
may indicate a greater external heat load than is true for
the Pauraque.

The three small owls of the southwestern U. S. and
Mexico that have been studied are distributed as one would
predict from Figure 22. The Saw-whet has thermoregulatory
characteristics that coincide with its actual cold-temperate
distribution. The Giant Hummingbird has the thermal capa-
bility for a subtropical distribution, instead of tropical
as is the case with smaller hummingbirds. Its actual dis-
tribution extends from the montane tropics southward into
Peru and Chile. Range extension of some hummingbirds beyond
the thermal limits suggested by their thermoregulatory
parameters is probably due to their capacity for torpor.

The present state of knowledge concerning T^ regulation
in birds does not allow conclusions to the effect that food
habits determine thermoregulatory capacity, as they apparent-
ly do in bats (McNab, 1969). Food habit mainly affects the
distribution of most birds studied to date. Species with
seasonal food problems usually migrate; thus, thermoregulatory

55

adaptation is unnecessary. There may be correlations of
food habit with lA^ and C in some species, but other factors
such as heat and water stress would complicate any analysis.

In birds, it appears that thermoregulatory parameters
are adapted primarily to climate, and are only indirectly
or secondarily related to food habit. Therefore, large
climatic differences must not be introduced into data that
are used to determine the impact of ecology on therm.oregu-
lation. Such data have not yet been gathered. Consequently,
the need is obvious for extensive studies on tropical birds
of all sizes, ecological characteristics, and taxonomic
affiliations.

CONCLUSIONS AND SUMflARY

The energetics of T regulation in small birds are
accurately described by the Newtonian model of heat loss.
A thermoregulatory quotient (Mj^/C)q which considers both
heat production and th6 coefficient of heat loss is a more
satisfactory indication of thermoregulatory capacity than
either parameter alone.

Climatic and ecological adaptations of these parameters
are indicated by (M^/C) , which compares (I%/C)q to the
value (1.00) expected from VI alone {Mj^/C)g. Tropical and
desert species generally have low (Mj-j/C)j-, and birds that
must tolerate cold climates have high values. Birds living
in cool climates may partially compensate, within limits,
for low (K, /C) by having an increased body size (Mj-,/C)g, the
capacity for torpor, or both.

Mj^ may be lower than expected in desert species

mainly to reduce the heat load and evaporative water loss.

Tropical, non-desert species may also have a lower Mj^ than

expected for other birds in order to reduce the heat load,

to conserve energy (if the food supply is unreliable) , or

simply because they are not faced with the need for a

higher rate of heat production. Some cool-climate birds

may also have a reduced M, to conserve energy. This is

b

particularly true if the birds are very small and dependent

56

57

on a variable food source (such as insects). In this case,
C is also reduced. It is suggested that M, differences
betv;een any two avian groups are actually correlated with
ecological and climatic differences in the birds sampled,
not taxonomy per se.

C is generally high in tropical and desert forms, and
low in cold-climate species. However, the effect of
environm.ental stress on M, may militate against this
strategy in some cases. Likev;ise, K, cannot be radically
altered if the environment precludes compensatory changes
in C. Mj^ and C must evolve as complementary thermal
parameters.

The level of Tj^ is very closely correlated with M, , C,
and W. Non-torpid birds are very precise thermoregulators ,
as compared to most mammals. A combination of very small
body size and a low (M. /C) may reduce the precision of T,
regulation, as may a drastic loss of W during extended food
deprivation.

Presently available information indicates that the dis-
tribution of avian species outside the tropics and deserts
is limi.ted by ecology, particularly food habits. Thermo-
regulatory parameters are primarily adapted to climate. Thus,
in order to determine the extent to vmich food habit can
affect thermoregulation, significant climatic differences
must be circumvented. The ecological impact on T, regulation
in birds must remain poorly understood pending investigation
of troDical faunas.

APPENDICES

APPENDIX A
Syinbols and Expressions Employed in This Paper

Symbol Description Units

W Body weight • Grams

M Metabolic rate cc02/g - hr

M, Basal metabolic rate cc02/g - hr

Mj^% Mj-,% expected from W by the King- Per cent

Earner (1961) equation:

R^ (cc02/g - hr) = 7.29 w~0-341
C Thermal conductance cc02/g - hr A°C

C% C% of that expected from W by the Per cent

equation of Lasiewski et at. (1967):

C (cc02/g - hr °C) = 0.848 w~0-508
(Mj^/Oq Observed thermoregulatory quotient "C
(Mv/C)g Thermoregulatory quotient expected °C

from W by the King-Earner and

Lasiewski et al. equations (see

above): (M^^/C) =8.59 W*^-!^"^
(Mvj/C)^. Relative thermoregulatory quotient None

(riv,/C)o/(Mb/C)e
Tj^ Body temperature ''C

Tg Ambient temperature ''C

59

60

c

APPENDIX A
(continued)

Symbol Description Units

ATj^/AT Change in body temperature per unit None
change in ambient temperature

K Lower limit of thermoneutral zone °C

T -K Extent of thermal buffering °C

Qj^ Heat loss cc02/g or

cal/g

t Time Hours

k Cooling constant 1/hours

2

A Surface area cm

Specific heat cc02/g °C or

cal/g "C

APPENDIX B

Scientific and ComiT^on Maraes for Species
Discussed in This Study

Species for Which Original Data Are Presented

Sparrows

Spizella passerina Chipping (W)

Ammodramus savannarum Grasshopper (W)

Melosipiza melodia Song (W)

M. georgiana Swamp (W)

Passerculus sandwichensis Savannah (W)

Pooeaetes gramineus Vesper (W)

Zonotrichia albicollis White-throated (VJ)

Z. leucophrys White-crowned (W)

Z. querula Harris' (W)

Paseerella iliaoa Fox (W)

Flycatchers

Ernpidonax viresaens Acadian (S)

Contopus virens Wood Pewee (S)

Sayornis phoebe Phoebe (W)

Myiarchus arinitus Crested (S)

Tyrannus tyrannus E. Kingbird (S)

-'- The symbols indicate the time of year when species were

captured and studied: (W) v/inter, (S) summer, (M) migrant,
not in Zugunruhe .

61

62

APPENDIX B
(continued)

Wood Warblers
Parula americana
Vermivora pinus
V. aetata
Mniotilta varia
Dendroica dominiaa
D. palmarum
D. coronata
D. pinus

Geothlypis trichas
Wilsonia citrina
Protonotaria citvea
Seiurus noveboracensis
S. auTocapillus

Trogon

Trogon rufus

Parula (S)
Blue-winged (M)
Orange-crowned (W)
Black-and-v;hite (W)
Yellow-throated (S)
Palm (W)
Myrtle (W)
Pine (W)

Yellowthroat (W)
Hooded (S)
Prothonotary (S)
N. Waterthrush (M)
Ovenbird (M)

Black -throated

Species for V7hich Data Were Obtained from the Literature

Otus triahopsis
Glaucidium gnoma
Micrathene whitneyi
Aegolius acadiaus
Phalaenoptilus nutballii
Nyotidromus alhiaollis
Chordeiles minor
Eurostopodus guttatus

Whiskered Owl
Pygmy Owl
Elf Owl
Saw-whet Owl
Poor-will
Pauraque

Common Kighthawk
Spotted Nightjar

63

APPENDIX B
(continued)

Eugenes fulgens Rivoli ' s Hummingbird

Lampornis clemenciae Blue-throated Hummingbird

Patagona gigas Giant Hummingbird

Vidua paradisea Paradise Widov/bird

Perisoreus canadensis Gray Jay

Cyanocitta cristata N. Blue Jay

Hesperiphona vespertina Evening Grosbeak

Loxia curvirostra Red Crossbill

L. leuooptera White-winged Crossbill

Taeniopygia oastanotis Zebra Finch

Estrilda troglodytes Black-rumped Waxbill

REFERENCES

BartholoKiew G. A, , Hudson J. W. & Howell T. R. 1962.
Body terepsrature , oxygen consumption, evaporative
water loss, and heart rate in the poor-will.
Condor 64, 117-125.

Bartholomew G. A. & Tucker V. 1964. Size, body temperature,
thermal conductance, oxygen consumption, and heart
rate in Australian varanid lizards. Physiol. Zool.
37, 341-354.

Bent A. C. 1942. Life histories of North J^jnerican
flycatchers, larks, swallows and their allies.
U. S. Navl. Mus. Bull. 179, 140-154.

Bent A. C. 1953. Life histories of North American

wood v;arblers. Ij . S. Natl. Mus. Bull. 203, 239-258.

Burtoi: A. C. 19 34. The application of the theory of
heat flow to the study of energy metabolism.
Jour. Nutrition 7, 497.

Calder W. A. 1964. Gaseous metabolism and water relations
of the zebra finch, Taeniopugia castanctis . Pkysicl .
Zool. 37, 400-413.

Dawson W. R. & Evans F. C. 1957. Relation of growth

and developm.ent to temperature regulation in nestling

field and chipping sparrows. Physiol. Zool. 30,
315-327.

Dawson W. R, & Evans F. C. 1960. Relation of growth and
de^'■elopment to temperature regulation in nestling
vesper sparrows. Condor 62, 329-340.

Dawson VJ. R. & Tordoff H. B. 1964. Relation of oxygen
consumption to temperature in the red and white-
winged crossbills. Auk 81, 26-35.

Dav/son W. R. & Fisher C. D. 1969. Responses to temperature
by the spotted nightjar {Eurostopodus guttatus) .
Condor 71, 49-53.

Hamilton VI. J., Ill n, Heppner F. K. .1967. Radiant solar

energy and the function of black homeotherm pigmentation:
an hypothesis. Science 155, 196-197.

64

65

Hainmel H, T. 1968. Regulation of internal body temperature.
Ann. Rev. Physiol. 30, 641-710.

Hart J. S. 1957. Climatic and temperature induced changes
in the energetics of homeotherms. Rev. Can. Biol.
16, 133-174.

Heppner F. H. 1970. The metabolic significance of

differential absorption of radiant energy by black
and white birds. Condor 72, 50-59.

Kerreid C. F., II & Kessel B. 1967. Thermal conductance
in birds and mammals. Comp. Biochem. Physiol. 21,
405-414.

Hudson J. W. & Kimzey S. L. 1964. Body temperature and
metabolism cycles in the house sparrow. Passer
domesticus , compared v/ith the white-throated sparrow,
Zonotrichia alhicollis . Amer. Zool. 4, 294-295.

Johnson R. E. 1968. Temperature regulation in the white-
tailed ptarmigan, Lagopus leuaurus . Comp. Bioahem.
Physiol. 24, 1003-1014.

King J. R. 1964. Oxygen consumption and body temperature
in relation to ambient temperature in the white-
crowned sparrow. Comp. Biochem. Physiol. 12, 13-24.

King J. R. & Farner D. S. 1961. Energy metabolism, thermo-
regulation, and body temperature. In Biology and
Comparative Physiology of Birds (Edited by Marshall,
A. J.), Vol. 2, pp. 215-288. Academic Press, New York.

Kleiber K. 1961. The Fire of Life. John Wiley, New York.

Lasiewski R. C, Hubbard S. H. & Iloberly W. R. 1964.
Energetic relationships of a very small passerine
bird. Condor 66, 212-220.

Lasiewski R. C. & Dawson V7. R. 1964. Physiological responses
to temperature in the common nighthawk. Condor 66,
477-490.

Lasiewski R. C. & Dawson W. R. 1967. A re-examination of
the relation between standard metabolic rate and body
weight in birds. Condor 69, 13-23.

Lasiewski R. C. & Lasiev/ski R. J. 1967. Physiological

responses of the blue-throated and Rivoli's humming-
birds. Auk 84, 34-48.

Lasiewski R. C. , Weathers W. W. & Bernstein M. H. 1967.
Physiological responses of the giant hummingbird,
Patagona gigas. Comp. Biochem. Physiol. 23, 797-813.

66

Ligon J. D. 1968. The biology of the elf owl, Micrathene
wkitneyi. Univ. Michigan Mus . Zool. Misc. Publ. 136.

Ligon J. D. 1969. Some aspects of temperature relations
in small owls. Auk 86, 458-472.

McNab B. K. 1966a. An analysis of the body temperature
of birds. Condor 68, 47-55.

McNab B. K. 196bb. The metabolism of fossorial rodents:
a study of convergence. Ecology 47, 712-733.

McNab B. K. 1969. The economics of temperature regulation
in neotropical bats. Comp. Biochem. Fhysiol. 31,
227-268.

McNab B. K. 1970. Body weight and the energetics of

temperature regulation. Submitted for publication.

McNab B. K. & Morrison P. R. 196 3. Body temperature and
metabolism in subspecies of Peromyscus from arid and
mesic environments. Ecol. Monographs 33, 63-82.

Misch M. S. 1960. Heat regulation in the northern blue

jay, Cuanocitta cristata hromia Oberholser. Physiol.
Zool. 33, 252-259.

Pearson 0. P. 1950. The metabolism of hummingbirds.
Condor 52, 145-152.

Rising J. D. 1958. The effect of temperature variation on
the metabolic activity of the Harris' sparrow. Comp.
Biochem. Physiol. 25, 327-333.

Scholander P. F. , Walters V. , Hock R. & Irving L. 1950a.
Body insulation of some arctic and tropical mammals
and birds. Biol. Bull. 99, 225-236.

Scholander P. F. , Hock R. , Walters V., Johnson F. & Irving

L. 1950b. Heat regulation in some arctic and tropical
mammals and birds. Biol. Bull. 99, 237-258.

Scholander P. F. , Hock R. , VJalters V. & Irving L. 1950c.
Adaptation to cold in arctic and tropical mamn.als and
birds in relation to body temperature, insulation and
basal metabolic rate. Biol. Bull. 99, 259-271.

Steen J. 1958. Climatic adaptation in some small northern
birds. Ecology 39, 625-629.

Veghte J. H. 1964. Thermal and metabolic responses of the
gray jay to cold stress. Physiol. Zool. 37, 316-328.

67

Wallgren H. 1954. Energy metabolism of two species of the
genus Emheviza as correlated with distribution and
migration. Aata Zool. Fenn. 84, 1-110.

West G. C. & Hart J. S. 1966. Metabolic responses of even-
ing grosbeaks to constant and to fluctuating temperatures,
Physiol. Zool. 39, 171-184.

Yarbrough C. G. 1970a. Summer lipid levels in some breeding
subarctic birds. Auk 87, 100-110.

Yarbrough C. G. 1970b. The development of endothermy
in nestling gray-crov/ned rosy finches, Leuaostiate
tephroaotis gviseonuaha. Comp. Biochem. Physiol .
33.

Yarbrough C. G. & Johnston D. W. 1965. Lipid deposition

in wintering and premigratory myrtle warblers. Wilson
Bull. 77, 175-191.

BIOGRAPHICAL SKETCH

Charles Gerald Yarbrough was born October 13, 1939,
at Lumberton, North Carolina. He was graduated from
Bladenboro High School (N. C.) in May, 1957. In June,
1961, he received the degree of Bachelor of Science with
a major in Biology from Wake Forest University. From 1961
until 1963 he was a teaching assistant and graduate fellow
at Wake Forest, where he received the Master of Arts degree
in Biology in June, 1963. He attended the University of
Michigan as a teaching fellow during the academic year
1963-1964. During the summer of 1964 he did ecological
research in the Canadian subarctic on a Chapman research
grant from the Arierican Museum of Natural History. He
was employed as Instructor in Biology at Campbell College
for two years, 1964-65 and 1966-67. During 1965-66 he
worked as an Interim Instructor in Zoology at the Univer-
sity of Florida. From September, 1967, until the present
time he has pursued his work toward the degree of Doctor
of Philosophy while in tenure of a National Science Founda-
tion Traineeship at the University of Florida. In 1968, he
worked as an avian ecologist at an Atomic Energy Commission
project on Amchitka Island, Alaska.

Charles Gerald Yarbrough is married to the former Hazel
Ruth Hill, and is the father of two sons. He is a member of

68

69

Sigma Xi , Phi Sigma, the American Ornithologists' Union,
the Cooper Ornithological Society, and the Wilson Ornitho-
logical Society.

This dissertation was prepared under the direction
cf the chairman of the candidate's supervisory committee
and has been approved by all members of that committee.
It was submitted to the Dean of the College of Arts and
Sciences and to the Graduate Council, and was approved as
partial fulfillment of the requirements for the degree of
Doctor of Philosophy.
June, 1970

Dean, College of Arts an

lences

Supervisory Committee;

Chairman 77

Dean, Graduate School

J7/"v^^

L~ ^. a

y6-«^

A

cy

Al-

.}