Content uploaded by W. R. J. Dean
Author content
All content in this area was uploaded by W. R. J. Dean on Apr 26, 2019
Content may be subject to copyright.
Birds inhabit all deserts, an unusual observation because they are mainly diurnal, non-fossorial, and have the high-
est mass-specific metabolism of all vertebrates. Many of the “adaptations” of birds for life in deserts are thought by
some to be intrinsic in all birds, and thus have been considered to be pre-adaptations. Nevertheless, recent evidence
suggests that desert birds do differ from birds of more mesic habitats, and physiological adaptations include reduc-
tions in basal metabolism and cutaneous water loss.Behavioural mechanisms that permit a desert existence include
nomadism, enabling them to move away from low resource patches to more productive patches, and the ability to
use features of their environment to escape from high temperatures. Desert birds can prevent overheating by using
rodent burrows and other shelters during the heat of the day. Birds in deserts include a high proportion of ants in
their diet, a resource not often exploited by birds of more temperate environments, and they also show seasonal or
opportunistic shifts in their diet, feeding on seeds when available, and on more succulent green plant matter, such as
leaf bases of grasses, after rain. Certain taxa, for example sandgrouse (Pteroclidae) and larks (Alaudidae), are main-
ly found in deserts. The adaptations in some species of larks are discussed in detail.
79
Adaptations of birds for life in deserts with particular reference to Larks
(ALAUDIDAE)
BY W. R. J. DEAN1AND J. B. WILLIAMS2
1DST/NRF Centre of Excellence at the Percy FitzPatrick Institute, University of Cape Town, Rondebosch, 7701,
South Africa. e-mail:lycium@mweb.co.za
2Department of Evolution, Ecology, and Organismal Biology, Ohio State University, 318 W 12th Avenue, 300
Aronoff Laboratory, Columbus, OH 43210, U.S.A
INTRODUCTION
Birds occupy the most inhospitable deserts on earth. In these
extreme environments they face two selective pressures,
extreme heat and aridity, but many of the adaptations of birds
for life in deserts have been thought to be “pre-adaptations”,
intrinsic to all birds (Maclean,1996). Evidence has accumulat-
ed that birds have evolved both behavioural and physiological
adjustments to the desert environment, and that previous
notions of “pre-adaptation” were based on insufficient data.
Many desert birds differ from birds of more mesic habitats,
particularly in coloration and behaviour, and less in diet, repro-
duction and moult. Behavioural adaptations include nomadism
(Dean,2004), the use of shelters to avoid the heat of the day
(Maclean,1996) and nocturnal courtship displays (Jensen,
1972). Physiological adaptations include low basal metabolism
and economy of water use (Anava et al., 2000; Williams, 2001;
Tieleman et al., 2002). Birds in deserts include a high propor-
tion of ants in their diet (Keith et al., 1992), a resource not well
exploited by birds of more temperate environments, and they
also show greater seasonal or opportunistic shifts in their diet,
feeding on seeds when available, and on more succulent green
plant matter, such as leaf bases of grasses, after rain
(Willoughby, 1971). Plumage coloration in desert birds,
thought to have evolved in response to predation, often closely
matches their substrate colour (Niethammer,1959), and does
not seem to play a significant role in thermoregulation. Desert
avifaunas are, to some extent, thought to be structured by pre-
dation and food resources. It has been suggested that raptors,
common desert inhabitants, have limited the diversity of other
species of birds to those adapted to the risks of living in an
exposed environment (Cook,1997). However, although a num-
ber of species of diurnal (Accipitridae, Falconidae and
Cathartidae), and nocturnal raptors (owls, Tytonidae and
Strigidae) occur in deserts, none of these are restricted to
deserts.
Using the Meigs (1953) definition of deserts, there are only
a few species of birds that occur primarily in desert habitats
(Serventy,1971). Many so-called desert species also occur out-
side deserts as well (Cowan,1997), and many birds in deserts
actually use “non-desert” habitats within the desert environ-
ment (Dean,2004). Avifaunas in Old World deserts are domi-
nated by coursers, sandgrouse, larks and chats, and by func-
tionally similar, but mostly unrelated, species in the deserts of
the New World (Maclean, 1996; Dean, 2004). This convergence
is due to the harsh environmental selective pressures to which
all birds in deserts have been exposed, and is manifested in sim-
ilar morphological, physiological and behavioural adaptations
in unrelated species (Maclean, 1996).
Our focus in this review is on the family Alaudidae, the larks,
and on the areas we know best, the Namib and Saudi Arabian
deserts. For each section we present an overview of desert birds
generally, and follow this with specific examples from the
Alaudidae. Lark species have penetrated deep into some of the
most hyper-arid regions on earth, occurring in sparsely vegetat-
ed stony deserts such as the Fezzan and Tibesti in southern
Libya (Guichard, 1955), the Rub’al Khali in Saudi Arabia
(Williams & Tieleman, 2000) and the Gobi in central Asia
(Tourenq et al.,1996a, 1996b). Adaptations in the desert-
dwelling species include lower basal metabolic rates and field
metabolic rates than larks of mesic habitats (Williams, 2001;
Tieleman et al., 2002, 2004), and lower daily fluctuations in fat
reserves than similar sized species of mesic habitats (Shkedy &
Safriel, 1991).
Transactions of the Royal Society of South Africa80 Vol. 59 (2)
DESERT AVIFAUNAS
Desert avifaunas (Cowan, 1997) are almost as difficult to define
as the environments that they occupy (Noy Meir, 1985). Based
on rainfall, we consider that any bird species that spends most
of its life in areas that receive < 250 mm mean annual precipi-
tation (MAP) to be a bird of arid regions, but that only those
species that occur mainly in deserts (sensu Meigs, 1953) to be
desert species.
Of the > 9730 bird species in the world, about 2500 occur in
arid to semi-arid areas, and of these 190 species also occur in
deserts (Dean, 2004). However, only 44 species world-wide
apparently spend most of their lives in hyper-arid environments
(Dean, 2004). Many of these species also move into more veg-
etated areas adjacent to deserts at times, or remain in the desert
at oases and drainage lines, i.e. in functionally “non-desert”
habitats within the desert environment, including wooded
drainage lines, rift valleys or oases, with few species in bare
dunes or on un-vegetated gravel plains (Dean, 2004), so there
are, in fact, very few species of desert birds. Of the 44 species,
14 are larks (Table 1), the highest proportion of “desert” species
in any bird family. However, in the Namib, only one species, the
dune lark Calendulauda erythrochlamys, is a resident largely
restricted to the desert; all the other species of larks in this
region, although common in the desert, or mainly occurring
within the desert, are nomadic or locally nomadic and use adja-
cent and more mesic habitats at times. In Saudi Arabia, in the
hyper-arid Rub’al Khali region, two species, greater hoopoe-
lark Alaemon alaudipes, Dunn’s lark Eremalauda dunni and
desert lark Ammomanes deserti are present all year round (JBW
unpubl data), but all three species may be locally nomadic or
nomadic in other areas (Keith et al., 1992). At least three other
larks, black-crowned Eremopterix nigriceps and grey-backed
sparrowlarks E. verticalis, and singing lark Mirafra cantans, are
not desert species per se, but are common in deserts, but also
occur in a wide range of more mesic savanna habitats.
Table 1. Larks (Alaudidae) that occur in deserts in parts of
their distribution ranges (adapted from Dean 2004). Only the
desert areas in which the species occur are given; many of the
species also occur outside deserts.
Taxonomically, 97 families of the >149 families of birds are
represented in deserts and semi-deserts, with Apodidae,
Strigidae, Caprimulgidae, Columbidae, Accipitridae,
Falconidae, Corvidae, Hirundinidae, Passeridae and
Fringillidae fairly strongly represented in most of the arid zones
of the world. With one exception ( the Alaudidae) phylogeny
does not provide insight into whether a species will be a
“desert” species or not. Some families, Laniidae, with 20 of 31
species, Passeridae, with 142 of 391 species and Fringillidae,
with 226 of 1015 species, are relatively strongly represented in
the arid zones of the world, but not all are desert species.
Accipitridae is well represented in the Asian and African arid
zones, but less species rich in the Australian, North and South
American arid zones, but similarly not all are desert species.
The Alaudidae has the highest proportion of species that inhab-
it deserts: of 96 species in the family, 85 are associated with
semi-arid or desert habitats (Dean, 2004). The family is also pri-
marily Old World, mainly African, where 76 species occur
(Dickinson, 2003). It is characterized almost throughout the dis-
tribution range of the family by a number of fairly narrow-range
endemics, markedly so in Africa (Keith et al., 1992). Of the 31
species that occur in southern Africa, for example, 24 species
are endemic or near-endemic to the region (Hockey et al., in
press). Larks differ from other oscine passerines in lacking an
ossified syringeal pessulus and in having scutes on the posteri-
or and anterior surface of the tarsi, and the hind claw is long and
nearly straight in almost all species. Their affinities are obscure,
and larks are generally thought not to be closely related to other
passerines.
Numbers of species, and abundance of birds in deserts is
low; patterns of species richness is strongly related to rainfall
amount, increasing rapidly with increasing rainfall (Maclean,
1996; Dean, 2004). Sparse vegetation in deserts strongly influ-
ences not only bird species richness but also the size and diet of
resident birds. Species that use desert habitats with sparse veg-
etation are relatively few, are usually small (< 50 g) and have a
mixed diet of invertebrates and plant parts, including seeds.
Nests or nest sites are usually not entirely dependent on vegeta-
tion, so many species nest on the ground or in crevices in rocks
(Dean, 2004). Conversely, some species of desert and semi-
Species Desert
Mirafra cantillans singing lark Arabia
Mirafra somalica Somali long-billed lark Sahel
Mirafra ashi Ash’s lark Sahel
Calendulauda erythrochlamys dune lark Namib
Eremopterix signata chestnut-headed sparrowlark Sahel
E. grisea ashy-crowned sparrowlark Sahara, Sahel, Arabia, India
E. nigriceps black-crowned sparrowlark Sahara, Sahel, Arabia
E. verticalis grey-backed sparrowlark Namib, Kalahari
Ammomanes cincturus bar-tailed lark Sahara, Arabia, Afghanistan, Baluchistan
A. phoenicurus rufous-tailed lark India
A. deserti desert lark Sahara, Sahel, Arabia, Baluchistan
Ammomanopsis grayi Gray’s lark Namib
Alaemon alaudipes greater hoopoe-lark Sahara, Sahel, Arabia, Afghanistan, Pakistan
Ramphocoris clotbey thick-billed lark Sahara, Arabia
Melanocorypha calandra calandra lark Iranian deserts, Afghanistan
Calandrella somalica rufous short-toed lark Sahel
C. rufescens lesser short-toed lark Arabia, Afghanistan
Eremalauda dunni Dunn’s lark Sahara, Arabia
Eremophila bilopha Temminck’s lark Sahara, Arabia
desert habitats are exceptionally large, with the largest birds of
all in African (common ostrich Struthio camelus and Somali
ostrich S. molybdophanes), and Australian deserts (emu
Dromaius novaehollandiae).
The presence of trees along drainage lines creates a more
structured plant community, and considerably increases the
species richness in desert areas (Cody, 1985, 1999; Maclean,
1996). In general, desert avifaunas show increasing species
richness and increasing residency with increasing vegetation
height and complexity (Dean, 2004). For example, the bird
species richness was significantly higher in wadis than adjacent
sparsely vegetated gravel plains in northern Saudi Arabia (van
Heezik & Seddon, 1999), and rifts and river valleys in the west-
ern Sahara have a richer, more sedentary avifauna than the
desert plains (Smith, 1965).
PLUMAGE AND PREDATION
The pale plumage of desert birds prompted some to argue that
the colour of desert birds is adaptive, helping to reduce heat
gain from solar radiation (Serventy, 1971; Maclean, 1996).
However, plumage colour is now thought to be adaptive in
another sense, that of reducing predation (Niethammer, 1959;
Willoughby, 1969; Cook, 1997). Colour of desert birds is fre-
quently related to soil colours (Serventy, 1971; Maclean, 1996).
In the Namib, at least 19 of 21 species of birds in hyper-arid
areas have “desert coloration” (pale sandy browns and greys)
and “generalised cryptic coloration” (dull browns and greys,
streaked or mottled), and only two species are conspicuously
coloured (Willoughby, 1969). The reduction in heat gain from
pale plumage colours would appear to be small (Willoughby,
1969, but see Serventy, 1971). Desert birds are sometimes black
or have black bellies. In the males of one of the conspicuously
coloured species, mountain wheatear Oenanthe monticola,
there are a number of plumage morphs, varying from all grey to
black-and-white; females are invariably plain dark brownish
black with a white rump (Maclean, 1993). Similarly, in one of
the common species of the Namib, the grey-backed spar-
rowlark, both sexes have black bellies, but are cryptically
coloured when viewed from above. It is possible that black bel-
lies, or black plumages reduce the cost of maintaining body
temperatures during early mornings, when ambient tempera-
tures are low. Black, black-and-white and dark brown are cer-
tainly common plumage colours in the Old World deserts
(Louw, 1993; Snow & Perrins, 1998). Black plumage, as in the
Chihuahuan raven Corvus cryptoleucus, absorbs a large amount
of solar radiation, but very little of this heat actually reaches the
skin surface. Air trapped between feathers and skin provides an
insulating layer (Louw, 1993). Dark-coloured birds can save
metabolic energy at low temperatures if they can expose them-
selves to solar radiation, but need to alleviate heat stress at high
temperatures by using shade and behavioural thermoregulation
(perching on stones in the wind, retreating down rodent burrows
and so on) (Serventy, 1971; Williams et al., 1999). Black
colours in the plumage do not negate the theory that plumage
colours are cryptic in desert environments; black plumages,
together with behaviour, may help to create disruptive patterns
and eliminate shadows and thus effectively conceal the bird
(Serventy, 1971).
Matches between soil colours and plumage colours are well
developed in the larks. Six lark species occur in the hyper-arid
Namib, of which three have plumage matching substrate
colours, and three have generalised cryptic coloration
(Willoughby, 1969). Similarly, the desert lark Ammomanes
deserti of the North African, Arabian and Indian deserts has
plumage colours ranging from pale sandy to very dark browns,
matching local substrates (Meinertzhagen, 1951). However,
although many argue that non-cryptic coloration in desert larks,
or desert birds in general, has been selected against by preda-
tion, no study has demonstrated or compared differences in pre-
dation rates between cryptically and non-cryptically coloured
species in desert environments.
Species matched in colour with small areas of soil are partic-
ularly vulnerable to climate change, if predictions of the effects
of climate change are true. In southern Africa, increasing tem-
peratures and sharply decreasing rainfall in arid areas (Midgley et
al., 2001) are expected to impact some bird species (Simmons et
al. 2004). Adaptable, generalist species may be able to cope with
climate change, but desert species, such as red lark Calendulauda
burra on red sands, and Gray’s lark Ammomanopsis grayi on
whitish to grey gravels, if shifted away from present ranges may
be disadvantaged on other substrate colours.
SURVIVAL
Physiological adaptations
Energy
Discussions of physiological adaptation to desert environments
typically include a description of adjustments of small mam-
mals, particularly kangaroo rats Dipodomys spp and the con-
centrating ability of their kidneys, arthropods and their imper-
meable cuticle, and even some arboreal amphibians that secrete
integumentary lipids to reduce cutaneous water loss (CWL),
whereas birds have been traditionally thought to be “pre-adapt-
ed” to the desert, or to lack any physiological specialization to
these harsh environments (Maclean, 1996). As one recent
Animal Physiology text suggests, “birds occur in deserts, but if
they need to drink, they can fly to watering places at some dis-
tance”, implying little selection on desert birds for physiologi-
cal specialization” (Hill et al., 2004). These notions are prima-
rily based on the early work of Bartholomew and his colleagues
who examined water loss of 13 species from the deserts of
southwestern North America and from nearby more mesic
regions; they concluded that birds in deserts have not evolved
unique physiological specializations that distinguish them from
mesic counterparts (Bartholomew & Cade, 1963). These
authors lamented the paucity of data on Old World species and
thought that these populations might show more conspicuous
physiological adaptations to arid conditions than their ecologi-
cal equivalents in the New World because Old World deserts are
geologically much older.
Because primary productivity in deserts is low, and thus food
resources scarce, selection should favour phenotypes that have
a reduced energy expenditure in these environments (Williams
& Tieleman, 2000). Several reports, subsequent to the work of
Bartholomew & Cade (1963), have hypothesized that arid-zone
birds may have evolved a reduced basal metabolic rate (BMR)
(Dawson & Bennett, 1973; Withers & Williams, 1990).
Selective advantages attributed to reduced BMR include lower
overall energy demand, lower total evaporative water loss
(TEWL), and lower endogenous heat production which would
have to be dissipated in a warm environment, often by evapora-
tive means. Tieleman & Williams (2000) compared the BMR of
21 species of birds from deserts with that of 61 species from
more mesic areas. Based on conventional least squares regres-
Adaptations of birds for life in deserts with particular reference to Larks 2004 81
Transactions of the Royal Society of South Africa
sion, and on regressions of phylogenetic independent contrasts
(Felsenstein, 1985), the analysis showed that, in general, desert
birds had a BMR 17-25% lower than non-desert forms (Fig. 1).
These broad-scale comparisons have been criticized because
species differ not only in environment but also in phylogenetic
history, diet and behaviour (Leroi et al., 1994); comparisons of
closely related species from different environments may provide
greater insights into how selection has influenced physiological
adjustments to environments without the complications of dis-
similar historical backgrounds. An examination of the BMR of
12 species of larks (Table 2) along an aridity gradient that
extended from The Netherlands to the hyper-arid deserts of
Arabia indicated that BMR decreased as the environment
became more arid (Fig. 2), a result consistent with the idea that
natural selection has reduced BMR in arid environments
(Tieleman et al., 2002). Interspecific phenotype-environment
correlations can indicate either genetic differences brought
about by natural selection or phenotypically plastic responses to
environmental conditions. In a separate study, Tieleman et al.
(2003) showed that adjustments in BMR could not be attributed
82 Vol. 59 (2)
Tab le 2. Species, mass, basal metabolic rates (BMR) and total evaporative water loss (TEWL) of larks (Alaudidae) from arid to mesic envi-
ronments. Data taken from Tieleman et al. (2003).
Species n Mass (g) BMR ± SD (kJ d–1)TEWL ± SD (g day–1)
Chersomanes albofasciata spike-heeled lark 20 25.7 29.1±4.98 25.0±3.33
Eremopterix nigriceps black-crowned sparrowlark 6 15.2 16.5±1.10 15.2±1.34
Eremopterix verticalis grey-backed sparrowlark – –15.1 – 15.1±1.31
Ammomanes deserti desert lark 6 21.5 20.1±2.46 21.5±1.60
Alaemon alaudipes greater hoopoe-lark 21 36.9 32.8±4.45 36.9±2.59
Melanocorypha calandra calandra lark 2 50.6 49.5±1.07 50.6±3.03
Calandrella brachydactyla greater short-toed lark 8 24.0 35.6±6.73 –
Calandrella rufescens lesser short-toed lark 27 23.6 31.6±3.08 –
Eremalauda dunni Dunn’s lark 22 20.9 24.7±2.61 20.5±1.69
Spizocorys starki Stark’s lark – – – 15.6±1.31
Galerida cristata crested lark 6 31.2 32.2±2.30 31.2±2.44
Lullula arborea wood lark 20 25.6 49.4±9.96 25.5±2.41
Alauda arvensis sky lark 29 32 62.4±8.43 31.7±3.47
Eremophila alpestris horned lark – 26.0 28.6 (0 SD) 26.0±2.08
Figure 2. Mass adjusted average basal metabolic rate (BMR) in 12
species of larks (see Table 2) as a function of environmental aridity
(data from Tieleman et al., 2003).
Figure 1.A basal metabolic rate (BMR) and 1.B. field metabolic
rate (FMR) as a function of body mass in desert and non-desert
species (data from Williams & Tieleman, 2000). PIC phylogenet-
ic independent contrasts.
to acclimation of adults to thermal environment, food availabil-
ity, or photoperiod. Hence, these physiological differences are
likely the result of genetic differences, although work is
required to investigate the role of developmental plasticity.
The finding that BMR, a laboratory measurement, is reduced
in desert birds gains further evolutionary significance if it trans-
lates to a lower overall energy expenditure in the field.
Evaluating the hypothesis that desert birds have reduced field
metabolic rate (FMR), when compared to non-desert forms (as
measured by doubly-labelled water), Tieleman & Williams
(2000) showed that FMR of desert species was 49% lower than
mesic species (Fig. 1). Again restricting their analyses to larks,
Tieleman et al. (2004) examined the FMR of adults feeding 5-
8 day old nestlings from arid and mesic environments and found
that mass-adjusted FMR of arid-zone species was 36-42%
lower than that of larks in mesic habitats.
There seems to be considerable support for the idea that nat-
ural selection has influenced the rate of living among desert
birds. Differences in metabolism of temperate zone birds have
been attributed to selection for the size of organs with high
metabolic intensity, such as the liver, kidney and heart, required
to maintain levels of energy expenditure during the period when
parents care for nestlings, the putative time of peak energy
demand (Daan et al., 1990). However, when variation in body
mass is taken into account, organ sizes do not differ among
larks from deserts and mesic regions (Tieleman et al., 2003).
The differences in FMR and BMR among these species may be
related to variation in tissue-specific metabolic rates, an avenue
for future research.
Water and water use
The hot conditions of many deserts coupled with the scarcity of
drinking water creates the potential for physiological problems
associated with dehydration; these difficulties are especially
acute for desert birds because they are mainly diurnal, non-fos-
sorial, and have the highest mass-specific evaporative water
losses of all terrestrial vertebrates. Williams (1996) collated
rates of total evaporative water loss (TEWL), the sum of cuta-
neous and respiratory water loss, for 102 species ranging in size
from hummingbirds (Trochilidae) to common ostriches. Using
both conventional least squares regressions and regressions
based on phylogenetic independent contrasts, his analyses
showed that arid forms had a lower TEWL than species from
mesic environments. In a study on 12 species of larks from arid,
semi-arid and mesic environments (Fig. 3), Tieleman et al.
(2003) confirmed that TEWL decreased along an aridity gradi-
ent and suggested that natural selection was the most likely
explanation for the lower TEWL in desert forms.
In order to gain insights into the workings of natural selection
in desert environments, it is desirable to know the mechanisms
responsible for the reduced TEWL in desert birds. Several candi-
dates have been proposed that might function to reduce their
TEWL; hyperthemia which increases the temperature gradient
between the body and environment, thus reducing the need for
evaporative cooling (Calder & King, 1974; Weathers, 1981), a
counter-current heat exchange system in the nasal passages that
lowers respiratory water loss (Schmidt-Nielsen et al., 1970), and
adjustment of the lipid structure of the skin to reduce cutaneous
water loss (Menon et al., 1996; Williams, 1996). The role of
hyperthemia and of counter-current water recovery in the nasal
turbinates in reducing TEWL has been explored (Tieleman &
Williams, 1999), but these factors could not account for differ-
ences in TEWL between desert and non-desert species.
Figure 3. Mass adjusted average total evaporative water loss
(TEWL) in 12 species of larks (see Table 2) as a function of environ-
mental aridity (data from Tieleman et al., 2003).
Desert birds could reduce their TEWL by decreasing the rate of
water loss through their skin compared to more mesic species
(Williams, 1996; Williams & Tieleman, 2000). Although early
investigators surmised that most evaporative cooling took place
in the respiratory passages (Mount, 1979), later work showed
that CWL is an important avenue of water loss, at least at ambi-
ent temperatures below body temperature (Bernstein, 1969;
Webster & Bernstein, 1987; Wolf & Walsberg, 1996). A study on
CWL and respiratory (RWL) as a function of Tain greater
hoopoe-larks and Dunn’s larks from the Arabian desert, and sky
larks and wood larks from temperate grasslands in The
Netherlands showed that the contribution of CWL to TEWL in
larks ranged from 50-70% at moderate Tas, but at high Tas, RWL
dominated water loss (Tieleman & Williams, 2002a). Surface
specific CWL at 25ºC was 29% lower in arid-zone species than
in mesic larks suggesting that a reduction in CWL was a primary
determinant of the lower TEWL in desert birds. When acclimat-
ed to different Tas for 3 weeks, 15ºC-acclimated greater hoopoe-
larks increased CWL by 22% compared with 35ºC- acclimated
birds, but other species of desert birds or temperate-zone larks
did not change CWL. However, even with the increase in CWL,
rates did not equal those in mesic species. This study is consis-
tent with the idea that larks from deserts have a reduced CWL at
moderate and low Tas, but provided no support for the notion
that at high Ta larks from arid regions rely more on CWL than
larks from mesic environments. Interspecific differences in
CWL could not be attributed to acclimatory responses to envi-
ronmental temperature and were most likely the result of genet-
ic differences due to natural selection.
If water loss through the skin is reduced in desert birds, then
we need to understand the possible mechanisms that would pro-
duce this result. The skin of birds is composed of a thin outer
non-vascular epidermis and a thicker inner vascularized dermis
(Dyck, 1985). One component of the epidermis is an outer
cornified layer of non-living cells embedded in a lipid matrix
called the stratum corneum, the layer that forms the barrier to
water vapour diffusion from the animal to environment (Blank
et al., 1984; Elias, 1981; Bouwstra, 1997). Evidence to support
the idea that the stratum corneum forms the barrier to water
vapour diffusion comes from work on mammals that showed
Adaptations of birds for life in deserts with particular reference to Larks 2004 83
Transactions of the Royal Society of South Africa
that topical applications of organic solvents increased water
vapour loss (Scheuplein & Blank, 1971), that site-specific lipid
content of the stratum corneum correlated with barrier function
(Elias, 1981), that water vapour diffusion directly correlated
with removal of lipids from the stratum corneum (Grubauer et
al., 1989), and that replenishment of lipids of the stratum
corneum, from which they had been removed, restored barrier
function (Elias & Menon, 1991). Working on non-domestic
birds, Haugen et al. (2003) examined the relationship between
lipids of the stratum corneum and CWL of species of larks
along a temperature-moisture gradient. Results indicated that
free-fatty acids, cholesterol, and ceramides, large lipid mole-
cules containing a sphingoid base linked to a fatty acid via an
amide bond, were the major constituents of lipids of the stratum
corneum, just as in mammals, for all larks examined. CWL was
reduced in larks inhabiting deserts, but data did not support the
hypothesis that birds from deserts had larger quantities of lipids
per unit area of skin. Instead, they found that larks from desert
environments had a higher proportion of ceramides, and a
smaller proportion of free-fatty acids in their stratum corneum,
an adjustment that apparently reduced CWL. Subtle changes in
the ratios of lipid classes can apparently alter the movement of
water vapour through the skin. These authors hypothesized that
desert birds have a higher proportion of ceramides in their SC
and lower proportions of free-fatty acids because this combina-
tion allows the lipid lamellae to exist in a more highly ordered
crystalline phase and consequently creates a tighter barrier to
water-vapour diffusion.
THE RELATIONSHIP BETWEEN LIFE HISTORY
AND PHYSIOLOGY
Assuming that life-history trade-offs in terms of parental effort
are the result of energy allocation (Drent & Daan, 1980; Bryant,
1988), the proportion of energy devoted to reproduction versus
that spent on growth and maintenance can be measured
(Reznick, 1985). Tieleman et al. (2004) used parental energy
expenditure and water flux in the field as proxy for parental
effort, relating these variables to BMR and TEWL, and investi-
gated variation in clutch size among larks along an aridity gra-
dient as an independent measure of parental effort. Their results
show that, in terms of energy expended, parental effort, clutch
size, number of clutches and nestling growth rates decreased
with increasing aridity (Fig. 4). Parent-offspring energy budgets
for desert larks were 261 kJ/day for greater hoopoe-lark and 164
kJ/day for Dunn’s lark, but significantly higher for larks in
mesic habitats, equal to 388 kJ/day for sky lark and 347 kJ/day
for wood lark (Tieleman et al., 2004). Taking into account mass
differences between desert and mesic larks, desert larks used
27% less energy to raise a brood, and used 28-51% less water
per gram mass than larks in mesic habitats (Tieleman et al.,
2004). Nestling survival in desert larks was lower; with
decreased parental effort and smaller clutch sizes, single broods
may have a lower fitness value for desert larks, suggesting that
the probability of adult survival is higher in this group
(Tieleman et al., 2004).
Data on comparative rates of survival between arid zone
larks and larks of more mesic habitats are scarce. The survival
of dune larks in the Namib is high; birds ringed as adults have
been recaptured almost 6 years later (n 3, Williams, 1992)
and another adult, ringed by JBW was recaptured at > 7 years
old (Mark Boorman, unpublished data). The only other data
from southern Africa are for sabota larks Calendulauda sabota,
a species that occurs in both desert and relatively mesic wood-
lands. For six sabota larks, ringed as nestlings in dry woodlands
and recovered at the same site, the oldest was aged 4.5 years,
one was aged 2.2 years and the remainder at just over one year
(SAFRING unpublished data).
Survival of sky larks in Britain was similar to the survival of
sabota larks in southern Africa; 88% of 201 ringed sky larks
were recovered within five years of ringing, although some sur-
vived to 10 years of age (Dougall, 1996). No comparative data
are available for survival of larks in arid zones in the Northern
Hemisphere.
Parental effort and energy expenditure in desert larks may
also vary in a single species according to environmental condi-
tions. Lloyd (1999) showed that mean clutch sizes in spike-
heeled larks, grey-backed sparrowlarks and black-eared spar-
rowlarks Eremopterix australis increased after a substantial
rainfall in arid Bushmanland in South Africa (Table 3). The low
84 Vol. 59 (2)
Figure 4. Growth constant (K),
clutch size, number of clutches and
daily survival of six species of larks
(black-crowned sparrowlark, bar-
tailed desert lark, greater hoopoe-
lark, Dunn’s lark, wood lark, sky
lark) (data from Tieleman et al.,
2004). Aridity index is from most
arid (1.5) to mesic (3.5).
Adaptations of birds for life in deserts with particular reference to Larks 2004 85
amount of rain during 1993-1995 was insufficient to stimulate
breeding in the mainly granivorous sparrowlarks, but was suffi-
cient for breeding in the mainly insectivorous spike-heeled
larks.
FOOD
Food resources in deserts are mainly seeds, some invertebrates
and some vertebrates; there are very few juicy fruits offered by
desert plants, so frugivory is not an option in most deserts with
the exception of deserts of the New World where fruits of cac-
tus are seasonally abundant. Seed-eating birds are common in
deserts, feeding on fresh seed production or on wind-blown
seeds (Maclean, 1996; Cook, 1997; Dean, 2004). Resources in
deserts are patchy in space and time, but seed crops and associ-
ated invertebrates may be large and concentrated in fairly small
areas, leading to large numbers of birds that feed on them, and
this in turn attracts predators that feed on the birds (Dean,
2004). It follows that birds that feed on plants that grow fast,
flower and produce large quantities of seeds after irregular rain-
fall events, must be nomadic to some extent (Dean, 2004).
The larks tend to be generalists, feeding on seeds and inverte-
brates (Willoughby, 1971; Keith et al., 1992; Hockey et al., in
press), and those species that are locally nomadic, or nomadic
over larger distances (see Movements below) are more granivo-
rous than species that are relatively sedentary (Dean, 2004).
Willoughby (1971) examined stomach contents of six species of
larks, three sedentary, or relatively sedentary, and three nomadic
species in the Namib. All ate seeds and insects, and all had vary-
ing amounts of green vegetation in their stomachs (usually basal
nodes of grasses, but may also have been stomach contents from
grasshoppers that had been eaten). Among the residents, only
sabota larks ate more seeds (60%) than insects, whereas dune
larks and spike-heeled larks had 68% and 84% insects respec-
tively in their stomachs. For nomadic species, however, seeds
were a major part of the food and amounts in stomachs were from
56% seeds in Gray’s larks, 77% seeds in Stark’s larks and 91% of
the food in grey-backed sparrowlarks (Willoughby, 1971).
Greater hoopoe-larks in Arabia feed on a wide range of ani-
mal material, including termites (JBW unpubl data) and small
lizards (Keith et al., 1992). Snails, ant-lions (Neuroptera) and
hard seeds all seem important in their diet (Keith et al., 1992).
Other larks, such as thick-billed larks Ramphocoris clotbey,
desert larks and Dunn’s larks in Arabian deserts similarly feed
on seeds and invertebrates, occasionally on small lizards (Keith
et al., 1992), but relative proportions of plant and animal foods
are not known.
One other food item is frequent in the diet of desert larks,
may be common to all desert larks, and quite likely common to
all larks. Ants (Formicidae) are eaten by all species, including
all the largely granivorous sparrowlarks for which detailed
stomach contents data are known (Keith et al., 1992).
BEHAVIOURAL ADAPTATIONS
Thermoregulation
Activity patterns to reduce heat gain or water loss in desert birds
fall into four broad categories: (1) minimizing activity, (2) using
shade to reduce absorption of solar radiation, (3) repeated
bathing if water available, and (4) soaring to reduce radiation
and ambient temperature (Maclean, 1996). Concentrating activ-
ity into the beginning and end of the day, rather than the hot
period in the middle of the day, serves to reduce heat gain. In
Saudi Arabia, greater hoopoe-larks are inactive during the mid-
dle of the day, whereas larks in more mesic habitats forage
throughout the day (Tieleman & Williams, 2002b). Similarly,
many bird species, not only larks, in the Namib remain rela-
tively inactive during the middle of the day, spending the hot
hours in shade (Willoughby & Cade, 1967). An example is the
dune lark, that roost overnight in clumps of a reedy desert grass
Stipagrostis salbulicola and begins intensive foraging immedi-
ately after emerging from the roost, at about 20 minutes after
sunrise (Cox, 1983). During late winter, the birds foraged until
late morning, when the ambient temperature was about 25º C
and sand surface temperature had reached about 45º C, there-
after entering clumps of the grass for a period of inactivity that
lasted until ca 16h00 when they again began foraging (Cox,
1983). However, in summer, with soil surface temperatures
reaching 62º C, the larks continued foraging without a break,
resting between foraging sorties in S. salbulicola clumps
(Safriel, 1990). Similarly, Gray’s larks and spike-heeled larks
forage out in the open throughout the day (Willoughby, 1971).
Between foraging bouts in the hot parts of the day, Gray’s larks
lose heat by perching on stones or twigs a few cm above the
sand surface, facing into the prevailing westerly winds, and
holding their folded wings away from the body to expose thin-
ly feathered sides (Willoughby, 1971). Both Gray’s larks and
spike-heeled larks forage around the burrow entrances of
ground squirrels Xerus inauris and other rodent burrows, and
use the burrows as thermal refugia (Willoughby, 1971). There
are no hard data on diurnal activity patterns for other lark
species in the Namib; Willoughby (1971) implies that both
Stark’s larks and grey-backed sparrowlarks are inactive during
the middle of the day, but does not give times when the birds are
active. Stark’s larks lose heat, or more correctly avoid gaining
heat, by crouching in the shade of stones, overhanging rocks or
vegetation (Willoughby, 1971). Similarly, grey-backed spar-
rowlarks avoid heat gain by resting in the shade of small plants.
Maclean (1996) notes that tractrac chats Cercomela tractrac in
the Namib use the same behaviour as Gray’s lark to lose heat,
perching a few cm above the ground and facing into the cool sea
breeze.
The use of burrows as thermal refuges in other desert
ecosystems has been reported by Williams et al. (1999). Dunn’s
Tab le 3. Clutch sizes ± SD (n) of arid zone larks in South Africa after different sized rainfall events. Data from Lloyd (1999). Rainfall was:
1993-1995 no rainfall event > 25 mm, Sep-Nov 1996 54 mm in July, Nov-Dec 1996 78 mm in November.
Species 1993-1995 Sep-Nov 1996 Nov-Dec 1996
Chersomanes albofasciata spike-heeled lark 2.0 ± 0 (13) 2.83 ± 0.39 (23) 3.3 ± 0.78 (10)
Eremopterix verticalis grey-backed sparrowlark 2.37 ± 0.57 (121) 3.07 ± 0.46 (124)
E. australis black-eared sparrowlark 2.51 ± 0.53 (69) 2.93 ± 0.33 (44)
Transactions of the Royal Society of South Africa
larks, bar-tailed desert larks, black-crowned sparrowlarks and
greater hoopoe-larks used burrows of the large herbivorous
lizard Uromastyx aegypticus as thermal refugia during hot sum-
mer days in the Arabian desert. Soil surface temperatures in the
area exceeded 60º C on most days, whereas burrow substrate
temperature was about 41º C during midday (Williams et al.,
1999). For one species, the greater hoopoe-lark, there are sig-
nificant advantages in using burrows; the birds can potentially
reduce their water loss by as much as 81% by sheltering in the
lizard burrows during the hottest periods of the summer day
(Williams et al., 1999).
Soaring, as a means of losing heat, does not apply to larks.
For large species, such as vultures and eagles, soaring is an eff i-
cient way of losing heat. Air temperature decreases at the rate of
ca 3º C for every 300 m in elevation (Cook, 1997), so if a large
bird is soaring at 2400 m it would be experiencing ambient tem-
perature well within its thermoneutrality zone.
ANTI-PREDATOR BEHAVIOUR IN DESERT LARKS
Although predation is thought to structure avifaunas in deserts,
there are few actual observations of predation on birds in
deserts. Avian desert raptors tend to be “generalists”, taking
whatever prey they can get (Cook, 1997) and it is likely that this
would also apply to small predatory mammals. However, the
sooty falcon in the Sahara Desert specialises on passage
migrants, mainly small warblers (Brown et al., 1982), and in the
Namib, the red-necked falcon Falco chicquera is fairly spe-
cialised, commonly taking Gray’s larks and grey-backed spar-
rowlarks (Jensen, 1972) and Stark’s larks (WRJD pers obs).
Maclean (1970a) recorded a red-necked falcon (once) feeding
on an adult spike-heeled lark in the Kalahari, and suggested that
other small birds of prey may be included as lark predators.
More generalised lanner falcons F. biarmicus in the Namib prey
on Gray’s larks, grey-backed sparrowlarks and Stark’s larks but
also take sandgrouse (Pteroclidae) and doves (Columbidae)
(WRJD pers obs).
Adult dune larks are wary when foraging, and Safriel (1990)
notes that when the birds foraging at midday spotted an insect,
they emerged from their thermal refuge and “moved their heads
as if inspecting the sky for predators” before darting forward to
seize the prey. Terrestrial predators are likely to be as important
as aerial predators; Maclean (1970a) notes that bat-eared foxes
Otocyon megalotis are probably the commonest nest predator in
the Kalahari, and that Cape foxes Vulpes chama may also
account for some predation. Both these mammals are also like-
ly to prey on adult larks. Snakes, particularly Cape cobras Naja
nivea, are potential predators of both adult and nestling larks.
Predation on black-crowned sparrowlarks was thought to be
“exceedingly important” in shaping adult breeding bird behav-
iour and distraction displays (Morgan & Palfery, 1987). A list of
potential predators given by Morgan & Palfery (1987) included
large lizards, snakes, small mammals and birds. Shkedy &
Safriel (1992) considered that both aerial and terrestrial preda-
tors important in the Negev Desert, and noted that crested larks
were more vigilant than desert larks and this behavioural differ-
ence potentially reduced predation rates on nestlings.
Falcons in the Namib and Kalahari wait at waterholes, taking
the prey species as they arrive to drink or as they depart after
drinking. Under these circumstances there are few tactics that
the birds can use to avoid predation. The sandgrouse, doves and
larks arrive and depart in flocks, but this only reduces the prob-
ability of an individual becoming prey (see Hamilton, 1971).
Siegfried & Underhill (1975) demonstrated experimentally that
flock-feeding or flock-drinking does increase the probability of
individual laughing doves Streptopelia senegalensis feeding or
drinking optimally, and of detecting and avoiding a predator.
Although Willoughby (1971) states that burrows serve spike-
heeled larks and Gray’s larks as (inter alia) predator-escape tun-
nels, his observations were based on the behaviour of wounded
birds. No study has demonstrated that birds consistently dive
down burrows when a raptor appears.
MOVEMENTS AND NOMADISM IN DESERT
LARKS
Resources in deserts are patchy in time and space (Noy-Meir,
1985). Animal communities similarly vary in abundance and
local distribution (Polis, 1991), and most desert bird species are
nomadic at scales ranging from locally moving around in the
landscape to movements over large distances (Dean, 2004).
Nomadism, and local movements by birds in deserts are an
adaptation to the variability, and not the severity, of desert
ecosystems (Rowley, 1974).
It is this variability, rather than the lack of rain, that strongly
influences whether or not bird species are nomadic. Rainfall in
arid ecosystems is both low and variable, and has a high co-effi-
cient of variation (CV) (Le Houérou, 1988). Rainfall in arid and
semi-arid ecosystems in Africa (Tyson, 1986; Seleshi &
Demaree, 1995; Plisnier et al., 2000; Holmgren et al., 2001),
South America (e.g. Lima et al.,1999) and Australia (e.g.
Nicholls, 1991) is also greatly affected by El Niño events, lead-
ing to prolonged droughts, and La Niña events, usually with
extensive wet periods and concomitant effects on ecosystem
functioning (Holmgren et al.,2001). Fewer bird species occur
in areas where the rainfall is both low and highly variable
(Dean, 2004).
There are exceptions to the general rule that birds must move
around in order to survive in deserts. The dune lark is resident
and sedentary in areas that are fairly barren, hyper-arid and sub-
ject to large inter-annual variability in rainfall (Safriel, 1990).
Nevertheless, the birds remain within small circumscribed areas,
and birds ringed initially as adults have been recaptured at the
sites where they were ringed up to five years later (Williams,
1992). In contrast, Gray’s larks wander over “large expanses” of
the desert and did not show any tendency to remain within a cir-
cumscribed range (Willoughby, 1971). Stark’s lark and grey-
backed sparrowlark in the same area were more nomadic,
appearing in very large flocks when grasses were beginning to
grow (Willoughby, 1971). Grey-backed sparrowlarks, in fact,
wandered further and more freely than Stark’s larks and were
observed in the most barren parts of the Namib sand sea
(Willoughby, 1971). Similarly, Maclean (1970b) recorded large
fluctuations in numbers of Stark’s larks, pink-billed larks
Spizocorys conirostris, grey-backed and black-eared spar-
rowlarks in the Kalahari. Other larks in the area, including east-
ern clapper larks Mirafra fasciolata,fawn-coloured larks
Calendulauda africanoides, sabota larks and spike-heeled larks,
were apparently more sedentary (Maclean, 1970b). In
Bushmanland, in the Northern Cape, South Africa, red larks
Calendulauda burra are resident and largely sedentary on bare
to thinly vegetated red sand dunes in an area that receives <100
mm mean annual precipitation (Dean et al., 1991).
Black-crowned sparrowlarks are said to be migratory in
Saudi Arabia (Morgan & Palfery, 1987), but the movements of
the flocks and their appearances appear to be somewhat
86 Vol. 59 (2)
Adaptations of birds for life in deserts with particular reference to Larks 2004 87
nomadic. According to Meinertzhagen (1951), thick-billed
larks Ramphocorys clotbey are nomadic throughout their range
in the Sahara and Arabia, and the desert lark, bar-tailed desert
lark and Dunn’s lark are nomadic in the Sahara, Arabian and
Iranian deserts. A list of lark species thought to be nomadic in
the warm deserts of Arabia is given by Dean (2004) (Table 4).
REPRODUCTION
Displays in desert larks
A feature of the nomadic Gray’s Lark of the Namib is that the
flight song – the distinctive full or advertising song in larks – is
performed in the dark. The birds give a burst of singing about
half an hour after sunset, and then a prolonged session of
singing for about two hours before dawn and ending at sunrise
(Willoughby, 1971). Dune larks, Stark’s larks and grey-backed
sparrowlarks all display during the day. Dune larks deliver their
advertising song from perches or in the air, but Stark’s larks and
grey-backed sparrowlarks most frequently sing in the air, and
usually simultaneously in large numbers (Willoughby, 1971).
The flight song in Stark’s larks is loud and strong, delivered by
individual birds for several minutes at a time.
It is tempting to suggest that the nocturnal song and display
of Gray’s lark is to lower the probability of diurnal predation;
birds in display are probably less vigilant than at other times.
Gray’s Lark is found in low densities in the Namib. Stark’s lark
and grey-backed sparrowlark occur in thousands, and the prob-
ability of an individual being taken by a predator would be
lower, so the birds can enjoy displaying during the day.
However, dune larks display during the day, and are probably
the least abundant bird in the Namib, suggesting that the timing
of displays is not modified by the risk of predation.
Furthermore, monotonous larks Mirafra passerina that occur in
some high rainfall years on the edge of the Namib in large num-
bers display during the day and night, so there is little support
for the nocturnal avoidance of predation theory.
Pair-bonds and territories are known for some of the resident
larks; in dune larks pair-bonds seem to be fairly long-term, and
territories are held throughout the year. Dune lark males
increase the frequency of aerial displays when females are
building nests (Boyer, 1988). In greater hoopoe-larks in Saudi
Arabia, males hold territories, with pair-bonds lasting the entire
year, although some females will switch male partners during
the winter period (JB Williams unpubl data). Pair-bonds in the
nomadic larks appear to be transient, but sparrowlarks, for
example, are almost always seen in pairs, suggesting long-term
pair-bonds. Courtship displays in the nomadic larks vary from
intricate aerial manouvres with complicated songs to perch dis-
plays with monotonous songs, to quite simple ground displays
with simple calls (Willoughby, 1971, Keith et al., 1992).
Cues for breeding in desert larks
Breeding in desert birds is stimulated by rainfall (Maclean,
1970a; Immelmann, 1970, 1971; Willoughby, 1971; Rowley,
1990), including larks (Lloyd, 2004) but what is uncertain are the
visual cues that all desert birds (particularly nomads) need to
indicate where resources are abundant, and where it may be a
suitable area to which to move and nest. It is very likely the pres-
ence of green vegetation that provides some of the stimulus to
trigger breeding, and not only abundant food. The presence of
suitable nest material may also provide some stimulus for breed-
ing, but it is most likely a combination of the availability of sev-
eral resources that trigger breeding. One resource in abundance
will not trigger breeding; in an experiment in Saudi Arabia in a
“drought year”, six pairs of greater hoopoe-larks were provided
with food, including mealworms, and water for over one month
in the field. Only one pair attempted to breed, building a nest that
they subsequently abandoned (JBW unpubl. data).
Movement by the birds towards rainfall events will not
ensure that there are resources immediately available for repro-
duction because there is a lag between rainfall and seed pro-
duction, and rainfall and the response of potential insect prey
for the birds. Fairly strong evidence for cues for when to breed
comes from nomadic larks in the southern African arid and
semi-arid ecosystems (Dean, 2004). Black-eared sparrowlark,
grey-backed sparrowlark, Gray’s lark, pink-billed lark, Sclater’s
lark Spizocorys sclateri and Stark’s lark all use the awns of
desert grasses (Stipagrostis ciliata, S. obtusa and S. uniplumis)
as nest lining (Keith et al., 1992, Dean, 2004). The birds do not
start nesting until there is a good supply of suitable awns, and
this is visually evident from the air – the awns from these grass-
es are drifted by the wind into whitish heaps at the bases of
grass tufts, termite mounds, or any other obstruction to move-
ment. The pale heaps indicate that the grasses have set seed.
Conditions that lead to flowering and fruiting by the grasses
encourage increased surface activity in harvester termites
Hodotermes mossambicus (Coaton, 1958), an important food
for nestling larks (Keith et al., 1992).
Tab le 4. Nomadic larks (Alaudidae) species in deserts of Saudi Arabia and the United Arab Emirates. Data from Bundy & Warr (1980),
Clarke (1980), Phillips (1982), Rahmani et al. (1994), van Heezik & Seddon (1999), adapted from Dean (2004)
Species
Mirafra cantillans singing lark
Eremopterix nigriceps black-crowned sparrowlark
Ammomanes cincturus bar-tailed lark
A. phoenicurus rufous-tailed lark
A. deserti desert lark
Alaemon alaudipes greater hoopoe-lark
Ramphocoris clotbey thick-billed lark
Melanocorypha calandra calandra lark
Calandrella rufescens lesser short-toed lark
Eremalauda dunni Dunn’s lark
Transactions of the Royal Society of South Africa
NESTS AND NEST SITES
Many species of desert birds nest on the ground. In the larks, all
species nest on the ground in sites that vary from sheltered
places next to grass tufts or shrubs to open, exposed sites in
gravel patches (Keith et al., 1992). Resident larks in the Namib
place their nests against shrubs or on the sides of small hum-
mocks of sand (Willoughby, 1971; Boyer, 1988), but locally
nomadic and nomadic larks generally nest in rather more open
sites, but frequently against a stone or small plant (Maclean,
1970a; Willoughby, 1971). Nest sites in all larks in the Sahara
and Arabian deserts are similar to those in the Namib, usually
placed at the base of a grass tuft or shrub (Keith et al., 1992;
Snow & Perrins, 1998). Lark nests are unremarkable, except for
Gray’s lark that builds a thick-walled cup in a stone-lined hol-
low in the ground, thought to provide insulation from the sur-
rounding hot substrate. The other feature common to all lark
nests in all desert ecosystems is a rampart of small stones
almost invariably placed around the nest, the function of which
is unknown. It is evidently important in the context of nesting in
the desert; larks in mesic habitats only rarely place stones
around the nest rim (Snow & Perrins, 1998).
CLUTCH SIZES
Larks in deserts have smaller clutch sizes than larks in more
mesic habitats (Shkedy & Safriel, 1992; Maclean, 1993;
Tieleman et al., 2004) but average clutch sizes do not differ sig-
nificantly between resident and nomadic larks in southern
African arid ecosystems (Dean, 2004). In one of the few studies
comparing clutch sizes in a single species in desert and more
mesic parts of its range, Herranz et al. (1994) found that Dupont’s
lark Chersophilus duponti had a mean clutch size of 3.6 (n 33)
in Spain, but 2.76 (n 75) in Algeria and Tunisia (Keith et al.,
1992). However, clutch size in several small granivorous nomadic
lark species is significantly smaller than similar-sized, locally
nomadic granivorous canaries and buntings (Dean, 2004). As
already noted, clutch sizes in desert birds, including larks, tend to
increase (in a single species)with increasing amounts of rainfall
in desert ecosystems in southern Africa (Lloyd, 1999).
Populations breeding in any one area when environmental condi-
tions are favourable (Maclean, 1970a), are usually fairly well syn-
chronised because there is only a brief pulse in food abundance
that follows rain. Net primary production in the Namib Desert
(Louw & Seely, 1982) and other arid areas in southern Africa is
low and mostly occurs immediately after rain (Rutherford, 1980;
Louw & Seely, 1982). The young larks are fed on insects, main-
ly harvester termites and grasshoppers (Keith et al., 1992).
Foraging effort in termites is related to grass growth, so their
availability, like the seeds, is directly related to rainfall. Grey-
backed sparrowlarks respond quickly to rain, and the duration of
their breeding season varies according to the amount of rain, the
smaller the rainfall event, the shorter the breeding season
(Maclean, 1970a, 1970b). Selective pressure would thus be
against an extended breeding season, and would favour a short,
efficient season. The combination of the rain, grass growth and
termite activity leads to a high degree of synchrony (Lloyd, 1999)
which in turn allows little time for the development of a large
clutch of eggs and the provisioning of a large brood of young
(Dean, 2004). Each egg represents an additional 24 hours spent at
the nest or in the breeding area (Williams, 1966).
Other environmental factors in deserts and intrinsic adapta-
tions in desert birds have also increased selective pressure for
the benefits of small clutch sizes. The combination of low food
availability and aridity, that select for decreased energy and
water conservation, high temperatures that limit daily activity,
and high nest predation, among an array of factors peculiar to
living in a desert environment, all select for small clutch sizes
(Tieleman et al., 2004).
NESTLINGS AND NESTLING CARE
There is a trend in southern African arid ecosystems for small
arid-zone granivorous passerines to have significantly shorter
nestling periods (14.2 ± 4.9 days) than mesic-zone passerines
(17.5 ± 3.7 days) (Siegfried & Brooke, 1989). All desert larks
have nestling periods that are shorter than larks of mesic habi-
tats (Keith et al., 1992), and in all species nestlings leave the
nest before they can fly (Keith et al., 1992; Maclean, 1993). In
the sparrowlarks (Eremopterix spp) and Stark’s lark
(Willoughby, 1971), the nestling period may be as short as
seven days. Parental care may be divided; Morgan & Palfery
(1987) suggested that more than two young cannot be cared for
satisfactorily by the adult black-crowned sparrowlarks in Saudi
Arabia. Once the young had been led from the nest, each was
under the care of one parent, and that there was no further con-
tact between the two halves of the brood after the first day out
of the nest. There are no comparative published data on post-
nestling parental care in other desert larks.
If each adult in a pair cares for only one nestling, this may
confer an advantage to the nestling in that it then has the undi-
vided care of one parent and, perhaps, a greater chance of sur-
vival. Selection would thus favour an optimal clutch size of two
eggs or less. Sclater’s lark, with a full clutch of one egg (Lloyd,
1997), has evolved to the point where time spent on egg-laying
and incubation nestling has been minimized, and parental care
of the chick has been maximized. In this respect, co-operative
breeding, or pairs with helpers, is known for at least two species
of southern African desert larks; spike-heeled (Steyn, 1988) and
Gray’s lark (Boix-Hinzen & Boorman, 2003; Desmasius, 2003)
have been recorded breeding with one extra-pair helper. This
would increase parental care in situations where food is limited.
BREEDING SUCCESS IN DESERT LARKS
There are few comparative data on the breeding success of larks
in arid regions. In one of the few studies the breeding success in
larks in the Arabian desert was generally low compared to larks
of mesic habitats (Shkedy & Safriel, 1992; Tieleman et al.,
2004). In semi-arid Spain, Dupont’s larks had low breeding suc-
cess, and most losses were at the nestling stage; of 119 eggs in
33 nests, 114 eggs hatched, but only 18 young left nests
(Herranz et al., 1994).
In southern Africa, the production of young of resident and
nomadic larks in the Kalahari is low (Table 4). There are very
little comparative data for larks of mesic habitats in this region;
in the only study for which there is a large sample,
Winterbottom & Wilson (1959) found that red-capped larks
Calandrella cinerea nesting in a high rainfall area near Cape
Town, South Africa, had a mean production of 1.5 young/nest
from a clutch of 2 eggs and 2.0 young/nest from a clutch of 3
eggs. This is much higher than similar-sized species nesting in
desert areas (Table 5).
In the Namib Desert almost half the losses of eggs of dune
larks was thought to be due to predation (Boyer, 1988) possibly
by small mammals or birds. Similarly, eggs and young of sev-
88 Vol. 59 (2)
Adaptations of birds for life in deserts with particular reference to Larks 2004 89
eral species of larks in the Kalahari are lost to mammalian pred-
ators (Maclean, 1970a).
Predation on black-crowned sparrowlark eggs and nestlings
in eastern Saudi Arabia was recorded but not directly observed
(Morgan & Palfery, 1987). A list of potential nest predators
given by Morgan & Palfery (1987) included large lizards,
snakes, small mammals and birds. Breeding success in that
species, however, seems high; of 10 nests that were observed
from egg to nestlings that left nests at eight days old, the aver-
age brood size was 1.8 young, but recruitment to the population,
following further losses of nestlings, was thought to be only
0.18-0.36 young/pair into the population (Morgan & Palfery,
1987). Slightly further west and north, Shkedy & Safriel (1992)
studied breeding success and predation of nests of desert and
crested larks near Sede-Boqer in the Negev Desert (Table 5).
Desert lark nests had lower survival of eggs and nestlings (0.15)
than crested lark eggs and nestlings (0.24), with predation
thought to be important in reducing broods in both species. In
the Negev, crested larks lost 63.2% of eggs to predators, where-
as in Germany the same species lost only 25.1% of eggs
(Shkedy & Safriel, 1992). Predators of eggs and nestlings in the
Negev included domestic cats Felis catus, foxes Vulpes spp.,
white storks Ciconia ciconia, brown-necked ravens Corvus rufi-
collis, great grey shrikes Lanius excubitor, stone curlews
Burhinus oedicnemus and transient raptors (Shkedy & Safriel,
1992). Crested larks were more vigilant than desert larks and
this behavioural difference potentially reduced predation rates.
CONCLUSIONS
Desert birds face two selective pressures, extreme heat and arid-
ity. Current evidence suggests that birds have evolved both
behavioural and physiological adjustments to the desert envi-
ronment, and that previous notions of “pre-adaption” were
based on insufficient data. Many desert birds differ from birds
of more mesic habitats, particularly in coloration and behaviour,
and less in diet, reproduction and moult. Behavioural adapta-
tions include nomadism, the use of shelters to avoid the heat of
the day and opportunistic breeding. Physiological adaptations
include economy of water use and lower energy expenditure.
The larks (Alaudidae) are a family of birds well-adapted for
desert environments. Adaptations in desert species include
decreasing field metabolic rate, basal metabolic rate and reduc-
tions of cutaneous water loss with increasing aridity (Tieleman
et al., 2002, 2003, 2004; Tieleman & Williams, 2002a), and
decreased parental effort, clutch size, number of clutches and
nestling growth rates with increasing aridity (Tieleman et al.,
2004). Other adaptations include behavioural responses to the
problem of losing heat, nocturnal displays, quick responses to
breeding after rainfall events, short nestling periods, brood divi-
sion, and in some species, co-operative breeding to maximise
parental care.
ACKNOWLEDGEMENTS
This paper was presented at the colloquium on “Adaptations in
desert fauna and flora” Victoria West, 29-31 August 2004. The
attendance by WRJD at the colloquium was partly funded by
the German Ministerium für Bildung und Wissenschaft
(BMBF) under project number 01LC0024 (BIOTA – southern
Africa), and by the National Research Foundation (South
Africa). Financial support for research on the Alaudidae by
JBW was made available by the National Science Foundation
(USA) (Project no. IBN-0212092).
REFERENCES
ANAVA, A., KAM, M., SHKOLNIK, A. & DEGEN, A. A. 2000. Seasonal field
metabolic rate and dietary intake in arabian babblers (Turdoides squam-
iceps) inhabiting extreme deserts. Functional Ecology 14:607-613.
BARTHOLOMEW, G. A. & CADE, T. J. 1963. The water economy of land birds.
Auk 80:504-539.
BERNSTEIN, M. H. 1969. Cutaneous water loss in small birds. Condor
73:468-469.
BLANK, I. H., MOLONEY, J., EMSLIE, A. G. & SIMON, I. 1984. The diffusion
of water across the stratum corneum as a function of its water content.
Journal of Investigative Dermatology 82:188-192.
BOIX-HINZEN, C. & BOORMAN, M. 2003. Helping behaviour in Gray’s lark
Ammomanes grayi. Lanioturdus 36(3):2-3.
BOUWSTRA, J. A. 1997. The skin barrier, a well organized membrane.
Colloid Surface 123: 403-413.
BOYER, H. J. 1988. Breeding biology of the dune lark. Ostrich 59:30-37.
BROWN, L. H., URBAN, E. K. & NEWMAN, K. B. (Eds) 1982. The Birds of
Africa. Vol. I. London, Academic Press. 521 pp.
BRYANT, D. M. 1988. Energy expenditure and body mass changes as meas-
ures of reproductive costs in birds. Functional Ecology 2: 23-34.
BUNDY, G. & WARR, F. E. 1980. A check-list of the birds of the Arabian Gulf
States. Sandgrouse 1:4-49.
Tab le 5. Breeding success in terms of no. of young produced per nest attempt by larks in the Kalahari, South Africa. Species marked
(R) are resident.
No of young
Species Eggs (nests) /nest attempt Reference
Calendulauda africanoides fawn-coloured lark (R) 33 (13) 0.3 Maclean (1970b)
C. sabota sabota lark (R) 3 (1) 0 Maclean (1970b)
Chersomanes albofasciata spike-heeled lark (R) 69 (28) 0.3 Maclean (1970b)
spike-heeled lark (R) 63 (30) 0.56 Lloyd (1998)
Eremopterix australis black-eared sparrowlark 78 (37) 0.1 Maclean (1970b)
black-eared sparrowlark (105) 0.68 Lloyd (1998)
E. verticalis grey-backed sparrowlark 243 (114) 0.17 Maclean (1970b)
grey-backed sparrowlark (242) 0.54 Lloyd (1998)p
Spizocorys conirostris pink-billed lark 75 (37) 0.14 Maclean (1970b)
S. sclateri Sclater’s lark 25 (25) 0.2 Lloyd (1998)
S. starki Stark’s lark 12 (6) 0 Maclean (1970b)
Transactions of the Royal Society of South Africa
CALDER W. A. & KING, J. R. 1974. Thermal and calorific relationships of
birds. In: Farner, S. & King, J. R. (Eds). Avian Biology. London,
Academic Press. pp 259-413.
CLARKE, J. E. 1980. The avifauna of Shaumari Wildlife Reserve, Jordan.
Sandgrouse 1:50-70.
COATON, W. G. H. 1958. The hodotermitid harvester termites of South
Africa. Department of Agriculture Science Bulletin 375, Entomology
Series 43:1-112.
CODY, M. L. 1985. An introduction to habitat selection in birds. In: Cody,
M.L. (Ed.) Habitat selection in birds. London, Academic Press. pp 4-58.
CODY, M. L. 1999. Assembly rules at different scales in plant and bird com-
munities. In: Weiher, E. & Keddy, P. (Eds). Ecological assembly rules.
Cambridge, Cambridge University Press. pp 165-205.
COOK, W. E. 1997. Avian desert predators. Adaptations of Desert
Organisms. Berlin, Springer-Verlag. 128 pp.
COWAN, P. J. 1997. What is a desert bird? Bulletin of the British
Ornithologists’ Club 117:299-303.
COX, G. W. 1983. Foraging behaviour of the dune lark. Ostrich 54:113-120.
DAAN, S., MASMAN, D. & GROENEWALD A. 1990. Avian basal metabolic
rates: their association with body condition and energy expenditure in
nature. American Journal of Physiology 259:R333-R340.
DAWS O N, W. R. & BENNETT, A. F. 1973. Roles of metabolic level and tem-
perature regulation in the adjustment of western plumed pigeons
(Lophophaps ferruginea) to desert conditions. Comparative
Biochemistry and Physiology 44A:240-266.
DEAN, W. R. J. 2004. Nomadic desert birds. Adaptations of Desert
Organisms series. Berlin, Springer Verlag. 185 pp.
DEAN, W. R. J., MILTON, S. J., WAT K EYS, M. K. & HOCKEY, P. A. R. 1991.
Distribution, habitat preference and conservation status of the red lark
Certhilauda burra in Cape Province, South Africa. Biological
Conservation 58:257-274.
DESMASIUS, E. 2003. Fourth Gray’s lark ringed in Namibia. Lanioturdus
36(4):15-18.
DICKINSON, E. C. (ed) 2003. The Howard & Moore complete checklist of the
birds of the world. Princeton, Princeton University Press. 1039 pp.
DOUGALL, T. W. 1996. Movement and mortality of British-ringed skylarks
Alauda arvensis. Ringing & Migration 17:81-92.
DRENT, R. H. & DAAN, S. 1980. The prudent parent: energetic adjustments
in avian breeding. Ardea 68:225-252.
DYCK, J. 1985. Feather. In Campbell, B. & Lack, E. (Eds) A Dictionary of
Birds. Calton. Calton, T. & A.D. Poyser. pp 206-209.
ELIAS, P. M. 1981. Lipids and the epidermal permeability barrier. Archives
of Dermatological Research 270: 96-117.
ELIAS, P. M. & MENON, G. K. 1991. Structural and lipid biochemical corre-
lates of the epidermal permeability barrier. Advances in Lipid Research
24: 1-26.
FELSTENSTEIN, J. 1985. Phylogenies and the comparative method. American
Naturalist 125:1-15.
GRUBAUER G., ELIAS, P. M. & FEINGOLD, K. R. 1989. Transepidermal water
loss: the signal for recovery of barrier structure and function. Journal of
Lipid Research 30:323–333.
GUICHARD, K. M. 1955. The birds of Fezzan and Tibesti. Ibis 97:393-424.
HAMILTON, W. D. 1971. Geometry for the selfish herd. Journal of
Theoretical Biology 31:295-311.
HAUGEN, M., WILLIAMS, J. B., WERTZ, P. & TIELEMAN, B. I. 2003. Lipids of
the stratum corneum vary with cutaneous water loss among larks along
a temperature-moisture gradient. Physiological and Biochemical
Zoology 76:907–917.
HERRANZ, J., MANRIQUE, J., YANES, M. & SUAREZ, F. 1994. The breeding
biology of Dupont’s lark, Chersophilus duponti, in Europe. Avocetta
18:141-146.
HILL, R. W., WYSE, G. A. & ANDERSON, M. 2004. Animal Physiology.
Sunderland MA, Sinauer Associates. 769 pp.
HOCKEY, P.A.R., DEAN, W.R.J. & RYAN , P.G. (Eds) (in press). Roberts’ Birds
of Southern Africa 7th edn. Cape Town, John Voelcker Bird Book Fund.
HOLMGREN, M., SCHEFFER, M., EZCURRA, E., GUTIERREZ, J. R. & MOHREN,
G. M. J. 2001. El Niño effects on the dynamics of terrestrial ecosystems.
Trends in Ecology and Evolution 16:89-94.
IMMELMANN , K. 1970. Environmental factors controlling reproduction in
African and Australian birds – a comparison. Ostrich Supplement 8:193-
204.
JENSEN, R.A.C. 1972. Adaptation of birds to arid and semi-arid habitats in
central S.W.A.: a preliminary report. Mitteilungen Ornithologische
Arbeitsgruppe 8(3/4):1-10.
KEITH, S., URBAN, E. K. & FRY, C. H. (Eds) 1992. The Birds of Africa. Vol.
4. London, Academic Press. 609 pp.
LEHOUÉROU, H. N. 1988. Interannual variability of rainfall and its ecolog-
ical and managerial consequences on natural vegetation, crops and live-
stock. In: di Castri, F., Floret, Ch., Rambal, S. & Roy, J. (Eds). Time
scales and water stress. Proc. 5th International Conference on
Mediterranean Ecosystems. Paris, I.U.B.S. pp 323-346.
LEROI, A. M., ROSE, M.R. & LAUDER, G. V. 1994. What does the compara-
tive method reveal about adaptation? American Naturalist 143:381-402.
LIMA, M., MARQUET, P. A. & JAKSIC, F. M. 1999. El Niño events, precipita-
tion patterns, and rodent outbreaks are statistically associated in semi-
arid Chile. Ecography 22:213-218.
LLOYD, P. 1997. Hatched in hell ... Breeding strategy in Sclater’s lark. Africa
– Birds & Birding 2(3):14-15.
LLOYD, P. 1999. Rainfall as a breeding stimulus and clutch size determinant
in South African arid-zone birds. Ibis 141:637-643.
LLOYD, P. 2004. Eight to ten-day incubation and nestling periods among
Eremopterix sparrow-larks. Ibis 146:347-350.
LOUW, G. N. 1993. Physiological Animal Ecology. Harlow, Longman
Scientific & Technical. 288pp.
LOUW, G. N. & SEELY, M. K. 1982. Ecology of desert organisms. London,
Longman. 194 pp.
MACLEAN, G.L. 1970a. The biology of the larks (Alaudidae) of the Kalahari
sandveld. Zooligica Africana 5:7-39.
MACLEAN, G. L. 1970b. Breeding behaviour of larks in the Kalahari sand-
veld. Annals of the Natal Museum 20:381-401.
MACLEAN, G.L. 1996. Ecophysiology of desert birds. Adaptations of Desert
Organisms series. Berlin, Springer-Verlag. 181 pp.
MEIGS, P. 1953. World distribution of arid and semi-arid homoclimates. In:
UNESCO (Ed.) Reviews of Research on Arid Zone Hydrology. Paris,
UNESCO. pp 203-210.
MEINERTZHAGEN, R. 1951. Review of the Alaudidae. Proceedings of the
Zoological Society, London 121:81-132.
MENON, G. K., BAPTISTA, L. F., BROWN, B. S. & ELIAS, P. M. 1996.
Ultrastructural organization of avian stratum corneum lipids as the basis
for facultative cutaneous waterproofing. Tissue and Cell 21:83-92.
MIDGLEY, G., RUTHERFORD, M. & BOND, W. 2001. The heat is on… impacts
of climate change on plant diversity in South Africa. Cape Town,
National Botanic Institute. 9 pp.
MORGAN, J. H. & PALFERY, J. 1987. Some notes on the black-crowned finch
lark. Sandgrouse 8:58-73.
MOUNT, L.E. 1979. Adaptation to thermal environment. London, Edward
Arnold. 333 pp.
NICHOLLS, N. 1991. The El Niño/Southern Oscillation and Australian vege-
tation. Ve g etatio 91:23-36.
NIETHAMMER, N. 1959. Die Rolle der Auslese durch Feinde bei
Wustenvogeln. Bonner Zoologische Beiträge 10:179-197.
NOY-MEIR, I. 1985. Desert ecosystem structure and function. In: Evenari,
M., Noy-Meir, I. and Goodall, D.W. (Eds). Ecosystems of the World, 12
A. Hot deserts and arid shrublands. Amsterdam, Elsevier. Vol 12A: 93-
104.
PHILLIPS, N.R. 1982. Observations on the birds of north Yemen in 1979.
Sandgrouse 4:37-59.
PLISNIER, P. D., SERNEELS, S. & LAMBIN, E. F. 2000. Impact of ENSO on
East African ecosystems: a multivariate analysis based on climate and
remote sensing data. Global Ecology & Biogeography 9:481-497.
POLIS, G. A. (ed.) 1991. The Ecology of Desert Communities. Tucson,
University of Arizona Press. 456 pp.
RAHMANI. A. R., SHOBRAK, M. Y. & NEWTON, S. F. 1994. Birds of the
Tihamah coastal plains of Saudi Arabia. OSME Bulletin 32:1-19.
REZNICK, D. 1985. Costs of reproduction: an evaluation of the empirical evi-
dence. Oikos 44:257-267.
ROWLEY, I. 1974. Bird Life. Collins, Sydney. 284 pp.
ROWLEY, I. 1990. Behavioural ecology of the Galah Eolophus roseicapillus
in the wheatbelt of Western Australia. Chipping Norton, NSW, Surrey
Beatty & Sons. 188 pp.
RUTHERFORD, M. C. 1980. Annual plant production-precipitation relations
in arid and semi-arid regions. South African Journal of Science 76:53-
56.
90 Vol. 59 (2)
Adaptations of birds for life in deserts with particular reference to Larks 2004 91
SAFRIEL, U. N. 1990. Winter foraging behaviour of the Dune Lark in the
Namib Desert, and the effect of prolonged drought on behaviour.
Ostrich 61:77-80.
SCHEUPLEIN R. J. & BLANK, I. H.. 1971. Permeability of the skin.
Physiological Review 51:702–747.
SCHMIDT-NIELSEN, K., HAINSWORTH, F.R. & MURRISH, D.E. 1970. Counter-
current heat exchange in the respiratory passages: effect on water and
heat balance. Respiratory Physiology 9:263-276.
SELESHI, Y. & DEMAREE, G. R. 1995. Rainfall variability in the Ethiopian
and Eritrean highlands and its links with the Southern Oscillation Index.
Journal of Biogeography 22:945-952.
SERVENTY, D. L. 1971. Biology of desert birds. In: Farner, D.S. & King, J.R.
(Eds) Avian biology. London, Academic Press. pp 287-339.
SHKEDY, Y. & SAFRIEL, U. N. 1991. Fat reserves of an opportunist and of a
specialist species in the Negev Desert. Auk 108:556-561.
SHKEDY, Y. & SAFRIEL, U. N. 1992. Nest predation and nestling growth rate
of two lark species in the Negev Desert, Israel. Ibis 134:268-272.
SIEGFRIED, W. R. & BROOKE, R. K. 1989. Alternative life-history styles of
South African birds. In: Bruton, M. N. (Ed.) Alternative life-history
styles of animals. Dordrecht, Kluwer Academic Publishers. pp 385-420.
SIEGFRIED, W. R. & UNDERHILL, L. G. 1975. Flocking as an anti-predator
strategy in doves. Animal Behaviour 23:504-508.
SIMMONS, R. E., BARNARD, P., DEAN, W. R. J., MIDGLEY, G. F., THUILLER, W.
& HUGHES, G. 2004. Climate change and birds: perspectives and
prospects from southern Africa. Ostrich 75:295-308.
SMITH, K. D. 1965. On the birds of Morocco. Ibis 107:493-526.
SNOW, D. W. & PERRINS, C. M. 1998. The birds of the Western Palearctic.
Concise edition, Vol 2, Passerines. Oxford, Oxford University Press.
1697 pp + indices.
STEYN, P. 1988. Co-operative breeding in the spikeheeled lark Ostrich
59:182.
TIELEMAN, B. I. & WILLIAMS, J. B. 1999. The role of hyperthermia in the
water economy of desert birds. Physiological and Biochemical Zoology
72:87-100.
TIELEMAN, B. I. & WILLIAMS, J. B. 2000. The adjustment of avian metabol-
ic rates and water fluxes to desert environments. Physiological and
Biochemical Zoology 73: 461-479.
TIELEMAN, B. I. & WILLIAMS, J. B. 2002a. Cutaneous and respiratory water
loss in larks from arid and mesic environments. Physiological and
Biochemical Zoology 75: 590-599.
TIELEMAN, B. I. & WILLIAMS, J. B. 2002b. Effects of food supplementation
on behavioural decisions of hoopoe larks in the Arabian Desert: balanc-
ing water, energy and thermoregulation. Animal Behaviour 63: 519-529.
TIELEMAN, B. I., WILLIAMS, J. B. & BLOOMER, P. 2003. Adaptation of metab-
olism and evaporative water loss along an aridity gradient. Proceedings
of the Royal Society of London, Series B 270:207-214.
TIELEMAN, B. I., WILLIAMS, J. B. & BUSCHUR, M. E. 2002. Physiological
adjustments to arid and mesic environments in larks (Alaudidae).
Physiological and Biochemical Zoology 75: 305-315.
TIELEMAN, B. I., WILLIAMS, J. B. & VISSER, G. H. 2004. Energy and water
budgets of larks in a life history perspective: parental effort varies with
aridity. Ecology 85:1399-1410.
TOURENQ, C., AVIRMED, A., MONCHTUJAA, B. & FEH, C. 1996a. Observations
ornithologiques dans le sud-ouest de la Mongolie (Première partie).
Alauda 64:355-362.
TOURENQ, C., AVIRMED, A., MONCHTUJAA, B. & FEH, C. 1996a. Observations
ornithologiques dans le sud-ouest de la Mongolie (Deuxième partie).
Alauda 64:405-412.
TYSON, P. D. 1986. Climatic Change and Variability in Southern Africa.
Cape Town, Oxford University Press. 220 pp.
VAN HEEZIK, Y. & SEDDON, P. J. 1999. Effects of season and habitat on bird
abundance and diversity in a steppe desert, northern Saudi Arabia.
Journal of Arid Environments 43:301-318.
WEATHERS, W. W. 1981. Physiological thermoregulation in heat-stressed
birds: consequences of body size. Physiological Zoology 54:345-363.
WEBSTER, M. D. & BERNSTEIN, M. H. 1987. Ventilated capsule measure-
ments of cutaneous evaporation in mourning doves. Condor 89:863-
868.
WILLIAMS, G. C. 1966. Natural selection, the costs of reproduction, and a
refinement of Lack’s principle. American Naturalist 100:687-690.
WILLIAMS, J. B. 1992. Longevity of dune larks in the Namib Desert. Safring
News 21:17-18.
WILLIAMS, J. B. 1996. A phylogenetic perspective of evaporative water loss
in birds. Auk 113:457-472.
WILLIAMS, J. B. 2001. Energy expenditure and water flux of free-flying
dune larks in the Namib Desert: a test of the re-allocation hypothesis.
Functional Ecology 15:175-185.
WILLIAMS, J. B. & TIELEMAN, B. I. 2000. Physiological ecology and behav-
iour of desert birds. In: Nolan, V. & Thompson, C.F. (Eds) Current
Ornithology 16: 299-353. New York, Plenum Press.
WILLIAMS, J. B., TIELEMAN, B. I. & SHOBRAK, M. 1999. Lizard burrows pro-
vide thermal refuges for larks in the Arabian Desert. Condor 101:714-
717.
WILLOUGHBY, E. J. 1969. Desert coloration of birds in the central Namib
Desert. Scientif ic Papers of the Namib Desert Research Station 44:59-
68.
WILLOUGHBY, E. J. 1971. Biology of larks (Aves: Alaudidae) in the central
Namib Desert. Zoologica Africana 6:133-176.
WILLOUGHBY, E. J. & CADE, T. J. 1967. Drinking habits of birds in the cen-
tral Namib desert of South West Africa. Scientific Papers of the Namib
Desert Research Station 31:1-35.
WINTERBOTTOM, J. M. & WILSON, A. H. 1959. Notes on the breeding of the
red-capped lark Calandrella cinerea (Gmel.) at Cape Town. Ostrich
Supplement 3:289-299.
WITHERS, P. C. & WILLIAMS, J. B. 1990. Metabolic rate and respiratory
physiology of an arid-adapted Australian bird, the spinifex pigeon.
Condor 92:961-969.
WOLF, B. O. & WALSBERG, G. E. 1996. Respiratory and cutaneous evapora-
tive water loss at high environmental temperatures in a small bird.
Journal of Experimental Biology 199:451-457.