3-'

c ^l

BIOLOGY OF DESERTS

THE PROCEEDINGS OF A SYMPOSIUM

ON THE BIOLOGY OF HOT AND COLD DESERTS

ORGANIZED BY THE INSTITUTE OF BIOLOGY

EDITED BY J. L.C LOU DSLEY- THOMPSON

PUBLISHED BY THE INSTITUTE OF BIOLOGY
TAVISTOCK HOUSE SOUTH, TAVISTOCK SQUARE,

LONDON. W.C.I

1954

HAFNER PUBLISHING COMPANY
NEW YORK

FOREWORD

This volume contains papers read at a Conference on 'The Biology and Produc-
tivity of Hot and Cold Deserts', organised by the Institute of Biology, and held at
the Royal Institution during September 25th, 26th and 27th, 1952. The Symposium
consisted of six sessions devoted to various aspects of desert biology as follows :-

I Climate and Physical Environment
II Plant Ecology

III Entomology and Ecology

IV Economic Aspects

V Mammalian Physiology and Ecology : I
VI Mammalian Physiology and Ecology : II

It was opened by Dr Edward Hindle, F.R.S. President of the Institute. In
welcoming delegates from abroad, Dr Hindle mentioned that the United Nations
Educational, Scientific and Cultural Organisation had shown a special interest in
the conference, and had contributed toward the travelling expenses of speakers from
oversea. A meeting of the Unesco Arid Zone Committee took place immediately
after the symposium.

Most of the papers were naturally concerned with the scientific problems in-
volved in attempts to increase the productivity of deserts and arid zones to meet the
ever increasing needs of a hungry world. A synopsis of some of the chief topics
mentioned in the various discussions has been provided.

The publication of this volume has been assisted by a grant from Unesco. The
Council of the Institute of Biology wishes to record its gratitude to Unesco for this
assistance.

The Editor would like to extend a personal appreciation to Mr D. J. B. Copp,
General Secretary of the Institute of Biology and to Mr C. A. Ronan and Miss T. J.
Tippett of the Secretariat of the Royal Society.

J . L. Cloudsley - Thompson

Editor.

11

CONTENTS

Page

X^ UX >r W aJL U ••• ••• ••• ••• ••• ••• ••• ••• ••• ••• ••• ••• ••• 11

The geography of deserts. By Professor Frank Debenham, O.B.E. ... 1

The physical aspects of dry deserts. By Brigadier R. A. Bagnold, F.R.S. 7

Availability of under- ground water in hot deserts. By Professor F. W.

OIl(J lLOII ••• ••• ••• ••• ••• »*• ••• ••• ••• •■• ••• ••« ••• ^ J

Some bioclimatic observations in the Egyptian desert. By Dr C. B.

■V 1 X -^ X aI Ho*** ••• ••• •«• ••• ••• ••• ••• ••« ••• ••• ••• ••• XO

Plant ecological problems in increasing the productivity of arid areas.

By Dr H. Boyko 28

Modes 'contracte' et 'diffus* de la vegetation saharienne. By Professor

X lla iVit_/llVJU ••• ••• ••• ••• ••• ••• ••• ••• ••• ••• ••• ••• J^

The Bahrain Islands and their desert flora. By Professor R. D'O. Good 45

Hydro- economical types in the vegetation of Near East deserts. By

Professor M. Zohary 56

The occurrence of plant diseases in arid climates and their agricultural

significance. By Professor I. Reichert 68

Phytosociologie et mise en valeur des sols en Afrique du nord. By Pro-
fessor L. Emberger , 76

Les relations entre les zones desertiques et la pullulation des parasites
des plantes. By Professor P. E. L. Vayssiere

The desert locust and its environment. By Dr B.P.Uvarov, C.M.G., F.R.S. 85

Sur I'origine et le developpement des insectes nuisibles aux plantes

cultivees dans les oasis du Sahara francais. By Dr A.S. Balachowsky 90

Role des insectes sociaux dans les terrains du Sahara. By Professor

X^ • X-/ vT X 1 1 Q.X KJ. ••• ••• •■■ ••• ••• ••• ••• ••• ••« ••• >•• ••• XV/t

The microbiological formation of sulphur in Cyrenaican lakes. By K.R.

Butlin and J. R. Postgate 112

Forests, Aridity and Deserts. By Professor E.P.Stebbing ... ... ... 123

The influence of climatic factors on the reaction of desert shrubs to

grazing by oheep. By Professor H. C. Trumble and K. Woodroffe 129

Biological research and the productive transformation of steppe and

desert in the Soviet Union. By Dr S. M. Manton, F.R.S 148

Aspects of the ecology and productivity of some of the more arid regions

of southern and eastern Africa. By Professor J.Phillips, F.R.S.E. 156

Problems of physiology and ecology of desert animals. By Professor

F. S. Bodenheimer 162

80

111

71031

Domesticated animals inhabiting desert areas. By Dr N. C. Wright
Water conservation in small desert rodents. By Dr B. Schmidt- Nielsen
Heat regulation in small and large desert mammals. By Professor K

Ov-ilITiXClC * \\ Id S V O ••• ••• ••• ••• ••• ••• «•• ••• ••• ••• ••

Reactions to great environmental heat in animals. By Dr F. Marsh
Human adaptability to hot conditions of deserts. By Dr J. S. Weiner ..

Le peuplement humain du Sahara. By Dr E. Sergent

The physiological effects of cold environments on man. By Dr O. G

.C<UI10X III ••• ••• ••• *•• ••• ••• ••• ••• ••• ••• ••• ••• ••

Some aspects of human ecology in hot tropical regions. By Professor
Sir David Brunt, Sec.R.S

1^1 S^UoolOIlo ••• ••• ••• ••• ••• ••• ••• ••• ••• ■•• •«•• •••

Page

168

173

182
188
193
200

207

213
219

IV

THE GEOGRAPHY OF DESERTS

Professor Frank Debenham O.B.E.
(Cambridge)

This paper, being introductory to a series which deals with special aspects of
dbserts, can be little more than a review of where deserts occur on the world's surface
and why, with some reminders of how many and varied are the factors which may com-
bine to produce a desert.

In broadest outline the origin of an arid zone on land is simple enough, since it
is caused by an interruption or suspension of the exchange of water from sea to land
via the air in a certain area. They are in fact due to flaws in the fundamental cycle
on which all life on land depends, the cycle which raises v/ater by evaporation from
the sea, against gravity, moves it over the land, where it is precipitated, and then, in
part, brings it back to the sea again by gravity.

The primary factor in the distribution of deserts is therefore the scheme of world
circulation of air, since an on- shore wind can bring water from sea to land whereas an
off- shore wind cannot. On the continental scale, indeed, arid zones must occur on the
lee side of the land with respect to the prevalent winds.

That appears at first sight to be a neat explanation of the occurrence of deserts
on the western side of continents in the southern hemisphere round about the latitude
of the Tropic of Capricorn, but by itself it is by no means an adequate one. It leaves
out of account at least two major factors — first, the temperature of the sea which is to
yield the water as vapour and, second, the seasonal swing of the zone of Equatorial
pressure, which together vv^ith the polar areas appears to govern the circulation of the
atmosphere.

The temperature of the sea, again, is dependent on ocean currents, themselves
mainly due to prevalent winds far away from the actual arid zones. These, in the com-
paratively simple set of circumstances of the southern hemisphere, tend to bring warm
equatorial water to the eastern sides of the continents and cold polar water to the
western sides.

Added to that component we must note too that a continuous off- shore wind tends
to bring up cold bottom water to the surface of the sea, while an on-shore wind piles
up warmer surface water against the land. The cold water will extract water vapour
from the air in the form of fogs and mists, while the warm water on the other side will
give itself up to the atmosphere.

The argument as to the origin of arid regions is already becoming conplicated,
yet we have not mentioned several other factors which must bear on the matter, such
as the distribution of the land masses, the topography of the land itself, and the in-
cidence of those enormous eddies in the atmosphere which we are accustomed to call
'depressions' and which appear to upset the neat pattern of world circulation of air in
both plan and elevation. It might be wise therefore to leave the argument at that point
and view the occurrence of deserts from another and simpler angle.

An arid region is one where the precipitation is much less than the world average
and therefore where the amount of water as vapour in the air is much less than the
average also. This state of affairs can come about in two ways — either the air has
lost the water it originally had or else it never had a very good supply. To put this in
a rryare scientific way, the air over an arid region is far below saturation point, and it
may reach that state because its temperature has been raised after losing its original
supply or because its original temperature was low and therefore it never had a good
supply.

We may illustrate this by referring to the conditions on that narrow but very ex-
treme desert of the Namib which stretches along the coast of South-west Africa for
hundreds of miles. In the rainy season for Southern Africa, roughly from November to
Kiarch, the easterly winds from the Indian Ocean bring warm and nearly saturated air
inland. Forced up over the Drakensberg range and the plateau behind it is cooled and
parts with much of its vapour as rain. When it reaches the so-called Kalahari Desert
it still has enough vapour to produce some thunderstorm rain, and if it does not fall
there it will fall further west where the higher land of South - west Africa cools it. The
air then descends to the coast and warms up as it goes so that at sea- level it is so
far below saturation point that even the cold Benguela current can only produce an oc-
casional fog.

At other times of the year there are occasional drifts of air from the Atlantic in-
land over the Namib, but it has come from the cold surface current and again is so far
below saturation point that it cannot do better than an occasional mist. We should
note that this fogginess may be, and in the Namib often is, a very important factor for
the biology of that area; but nevertheless it remains a very severe desert.

We may therefore think of most deserts as largely due to their occupying vast
areas of rain- shadow, that is to say, areas which are on the lee side of land which
has robbed the air of most of its precipitable moisture. Narrow coastal deserts such
as those of South- west Africa and Chile add the effect of a cold current off-shore and
are more arid still. This simple explanation of the distribution of deserts will, no
doubt, be amplified by papers later in this symposium.

Yet no desert is completely and permanently without water, just as the air above
it is n^ver completely dry, and it is as well that biologists should realise just how
moisture does reach the ground to sustain such desert life as exists.

Dry air means clear skies and clear skies mean excessive insolation by day and
radiation by night — the two .processes which are mainly responsible for some degree
of precipitation. The rapid heating of the land by the sun by day induces rising cur-
rents of air, which are usually local in extent. On the smaller scale these produce the
familiar dust- whirlwinds which have so many curious names in different parts of the
world. On the larger scale these upward currents will take the air high enough to cool
down by adiabatic expansion and even reach dewpoint, so that clouds are formed —
usually of the cumulus type, since the release of heat within them still further accel-
erates the rate of ascent of air. These often produce rain, as their fuzzy under- sur-
faces show, but it is rain which rarely reaches the ground, and indeed one can see it

evaporating in wispy tails to the clouds as it descends. When it c'oes reach the ground
it tends to be torrential and brief, the thunderstorm type, and comparatively local in
extent.

Thus the two characteristics of desert rainfall are that it is accidental, being due
to a disturbance of equilibrium in the air, and that it is usually local in extent. In the
tropics it tends to be seasonal, precisely because the desert is in a rain- shadow area
and the air reaching it during the rainy season to windward is at least more saturated
than at other times of the year.

Returning to the heavy insolation which is the cause of this instability rainfall,
we must note that it is much more effective on dark bare soil than on light- coloured
soil with some vegetation covering. This fact appears to me to be of considerable im-
portance in the biology of deserts, though of course it applies rather to the semi-
desert where vegetation cover can occur than to the utter desert where it cannot.

In the western and drier parts of the Kalahari, for instance, the cover of low bush
and grass is quite considerable in the so-called rainy season. On the 'pans' how-
ever, which tend to occur in chains, the ground is either bare because of its salt con-
tent or has a short grass, kept shorter still by the herds of springbok. It was noticed
during my visit there that the thunderstorms tended to keep to the pans, that is to the
centres of the rising air currents. On one particular day a series of over a dozen
heavy thunderstorms passed along such a line of pans. At our camp, situated on a
sand- ridge half- a- mile from this chain, only one of the storms produced rain, though
it was nearly half- an- inch in half- an- hour. The natural deduction was that over the
pans the rainfall that day was very much greater, possibly several inches. This de-
duction was supported by the fact that the next day our lorries were badly bogged
crossing one of these pans, over which two days before we had driven at speed.

In any area of instability rainfall we have to be very cautious about accepting
rain-gauge data, but it will be doubly so if there is ground for suspecting that storms
choose their paths with some consistency, in the way outlined above.

If the tremendous insolation by day in a desert causes great vertical disturbance
in the atmosphere, the opposite occurs at night. The rapid cooling tends to cause an
inversion of temperature, so that air in contact with the ground becomes heavier and
remains there. All desert travellers are familiar with the experience of insupportable
heat by day and desperate cold at night. It is no exaggeration to say that a basin of
water outside one's tent may be frozen at six in the morning, thawed by eight and at
blood temperature by noon.

These rapid changes of temperature must obviously affect plants and animals in
the desert, but at the moment we are concerned chiefly with the yield of moisture so
caused. Measured in inches, even if that were possible, the total derived from frost-
rime and dew would not be impressive, but the fact that it is in immediate contact with
leaves and branches is no doubt of biological significance. Certainly it is the case
that some antelope, notably the springbok, derive all their water-supply from the dew
on the grass in the early morning. Later papers will perhaps assess the value of this
source of moisture, particularly in the case of plants.

We come now to a very vital problem in the physical geography of deserts, namely,
what becomes of such rain as does fall. The data we have on this important aspect of
arid areas are still very incomplete, nor have they been collected into any general
summary so far as I know. We know little as to the relation between evaporation, run-
off and absorption in deserts, which is hardly to be wondered at for not only are sta-
tions suitable for such observations very rare but the number of factors involved is so
large that data can rarely be of general application. It goes without saying that
evaporation must be high and that run -off on loose sand must be low or non-existent.
It is the remaining proportion which is so important and so difficult to measure.

We cannot here become involved in the figures available from French engineers in
North Africa, American engineers from work in their drier states, and British workers
chiefly in India, since they do not mean very much without accessory observations of
a very local kind such as the nature of the ground, the rate at which the rain falls, the
temperature of air and ground, the wind, and even the time of day. All we can say is
that in most deserts there is some portion of the rainfall which succeeds in passing
through the upper layers beyond the reach of plants, there to form a water-table which
will appear as seepage springs in an oasis or can be tapped by bores. Much of it may
remain in partially enclosed areas underground, to become brackish and to cement the
sand grains into a calcrete or a silcrete. The term 'fossil water' has been used for
such occurrences but it is not a very useful phrase. Obviously there must be critical
points as to the amount of rain, its rate of fall, and the other factors mentioned, below
which no water is conserved beneath the surface, and this critical point will vary with
each desert and again at different places within that desert.

It seems likely that the most critical factor may well be the rate at which rainfall
can seep downwards through the sand and soil, since it is a question of z race between
capillary action taking it back towards the surface and gravity leading it downwards,
beyond the reach both of plant roots and of the capillary rise, and the issue of the
race is largely dependent on the porosity. An accessory factor may be the imprison-
ment of air below the sodden layer after a storm; at least, that was my interpretation
of an observation we made in the northern, wetter, portion of the Kalahari. We were
boring with a hand- sampler, which in this case took us down 20 feet, and it was in
the middle of the rainy season. The first 3 feet produced damp, not saturated, sand
and then we suddenly ran into 10 feet of perfectly dry sand. At about 13 feet the sand
was again damp, and so continued to the limit we reached. From a nearby bore the
water-table here was at 40 feet. We took it that the zone of dry sand represented the
previous dry season and were surprised that six weeks of a rainy season, which on
average should have yielded about 8 inches, had here penetrated to only 3 feet. It was
the usual fine yellow Kalahari sand, which by experiment absorbed water at a much
greater rate than that.

An isolated observation of that kind is of little value, but nevertheless one seeks
for an explanation, just as one wonders how the reservoir at 40 feet or so can get any
significant replenishment each year under such a regime..

The hydrology of deserts must sound to the layman like a contradiction in terms,
yet it is the study of a desert's water- supply which is the basis of this whole con-

ference, since it is essential to all forms of life. It follows that the only absolute
desert from the biological point of view is that which has no reserve of water, no
means by which the rare rainfall can escape from instant evaporation. From that as-
pect the boulder- strewn rocky surfaces of the Sahara or of the Australian Central
Desert are more nearly absolute in their aridity than the moving sand-dunes, for the
latter will at least store rain within their mass, letting it our slowly to keep alive '■
those scant bushes in the troughs between the dunes.

This brings us to what, as a geographer interested in the practical application of
scientific knowledge, I regard as the most important consideration to come before the
participants in this conference. Even though deserts are, at the very best, but mar-
ginal land for the use of man, it behoves us to make what use we can of them. That
use, as we know, depends on the available reserves of water, but the investigation of
what reserves there are, is at present, a very expensive business.

Yet we know there must be some close relation between the available water and
those plants which are permanent occupants of an arid area, and they must therefore be
indicators in some degree of the underground water. The ecologist and the physical
geographer have the duty of finding out how far one can trust indicator plants, which,
after all, are best qualified to give us the information once we have wit to interpret
their message-
In the past when searching for sub- surface water in arid areas we have been ac-
customed to send water- engineers and geologists and even physical geographers. In
my opinion the ecologist is the more appropriate scientist for such a mission, since he
should be able to ask the question of the plants, which really do know the answer,
whereas the rocks and the sand carry no visible prool that there is water below the
surface.

To conclude on a still more practical note, I would suggest that the biologists
equip themselves more fully for this new duty by field- work directed especially to find-
ing out more about such indicator plants, particularly in those semi- arid regions which
could be put to better use than they are at present. I should like to quote the particu-
lar case of my own favourite 'desert', so miscalled, the Kalahari.

Even in the worst parts of it there is a cover of bushes and small trees which sur-
vive one- or two-year droughts. To prove by drilling that there are underground water
supplies would be a costly and haphazard means of investigation. I would rather em-
ploy a field ecologist who would make it his chief if not his sole purpose to establish
a relationship between sub- surface supplies of water and perennial plants.

The whole secret of life in arid regions is movement, a readiness and a freedom
to migrate. This is obvious enough for man and the larger animals who somehow find
out where the casual storms have occurred and move to the pastures so benefited. For
the less mobile small fry and the immobile plants it is a case of adapting themselves
to endure long dry periods in a state of dormant or suspended animation. This they do
in a myriad different ways, and perhaps the best examples of adaptation to environ-
ment are to be found in deserts. These adaptations however are in delicate balance
and it is for the biologist to study how far it is safe for man to interfere with them for

his own benefit. A very obvious example is the replacement in some Middle East arid
districts of the migrant gazelle, which only nibbles thorn- bushes, by the goat, which
eats them to the ground.

If semi-deserts are to become free ranching grounds the greatest care will have
to be taken as to stocking well below capacity, moving the cattle constantly, and use-
ing the principles of pasture management.

In the meantime the scientist, and particularly the biologist, must find out a great
deal more about desert ecology and the correlation betv/een the water resources and
the indigenous flora.

THE PHYSICAL ASPECTS OF DRY DESERTS

Brigadier R.A.Bagnold, F.R.S.
(London)

Cause and General Character of Dry Deserts

A desert can be defined as a region where the physical conditions are adverse to
human ecology, beyond some agreed stage. But there is no reason to confine ourselves
to human economy, and in any case we ought to think of the economy of desert folk
rather than of western civilisation. This kind of definition would be all right if the
physical conditions were uniform from one area of the region to another and from year
to year. But they are not. Hence for the proper study of the biology and productivity
of deserts we must have a clear idea not only of the general physical factors but also
of their variation from place to place and from year to year.

In what follows I shall include the more arid and the less economically inviting
desert conditions because I feel one can often see a particular important but narrow
part of a wide range of conditions in far better perspective after having seen the ex-
tremities of the range. And unfortunately very few biologists have had personal access
to the arid extremity. If I shall sin at times as a layman by straying into the realm of
biology, it will be for the same reason.

The primary cause of the great sub- tropical deserts is quite clearly meteorologi-
cal, though it is not heat but lack of moisture. Life can thrive in the very hottest
spots known. Desert regions lie beneath more or less permanent anticyclones where
the dry upper air descends to the ground. Atmospheric moisture is therefore low, rain-
fall is slight and precarious. The sun's radiation is for long periods unshielded by
cloud. Summer day temperatures are high. The downward seepage of water through the
sub- soil is so infrequent that salts tend to rise and accumulate at the surface in ex-
cess, through evaporation. The soil is dry for such long periods and to such a depth
that young replacement plants may not mature, and ultimately even deep-rooted plants
may not be able to exist. The resulting lack of vegetation cover allows a high rate of
erosion both by wind and rain.

The sub- tropical anticyclones like the trade winds are due to geophysical causes,
and so are quite unalterable by human agency. The fitful cloud and rainfall on their
borders depends on the degree to which disturbances whether local or from outside can
upset the general anticyclonic regime. This degree may vary from time to time but the
cores of the great deserts have most probably been relatively arid from far back into
geological time and must remain so into the future.

Factors Affecting the Availability of Desert Moisture

The simple measurement, or classification of desert conditions in terms of lack of
moisture is not possible. Too many factors enter. First one must be careful to dis-
tinguish those special areas which do not rely at all on the present day rainfall of the
region because they get adequate and reliable water some other way. Second we have

the factors introduced by sxirface relief, soil and geological structure, which cause
great variations from place to place in the amount of available water. Third we have
the probability of prolonged periods of cloud associated with the rainfall. Fourth and
perhaps most important of all we have the variability of rainfall from one year to an-
other. In some cases fog, frost -rime and dew may also be important.

Exclusion of Permanently Irrigated Areas

Life can luxuriate in the atmospherically driest spots on earth provided adequate
and reliable water is made available (and provided this water can also drain away).
The words adequate and reliable should here be emphasised. Such permanently irri-
gated spots have of course many specialised biological interests, but they are clearly
not themselves part of a desert, though they may be surrounded by desert. The clear-
est example is an area to which the supply comes direct by river or canal from else-
where beyond the desert. Another not so well recognised example is the oasis fed from
a large artesian reservoir beneath the desert. Here the supply comes not from else-
where but from elsewhen — from the fossil water of the rains of long ago. When geolo-
gical permeability limits the rate of supply and the sites are limited economically by
the pumping lift from the water table to the surface, the supply is virtually inexhaust-
able. Desert biology ought therefore to be confined to life that relies on the precarious
atmospheric moisture supplied from within the desert region itself.

Variations in Available Water Supplies from Place to Place

Effect of surface relief, soil and geology.

Because of the high evaporation light showers and dew ought to be excluded from
any estimate of the effective mean annual rainfall, except for those forms of life which
are specially adapted to absorb and store moisture very quickly. To what extent this
is possible appears to need a good deal more investigation.

Owing to surface dryness, lack of plant cover and to the fact that desert rainfall
in general is characterised by heavy and infrequent storms, run -off is high and local
sub- surface storage low. Hence we may get very strong contrasts in the available
moisture between the catchment grounds and the drainage lines which thread them. If
the geology is suitable we may get considerable storage in shallow underground pools
along the drainage lines, where water is preserved from evaporation and is near enough
to the surface to be directly available. In this case one finds narrow streaks of vege-
tation threading barren country. In other cases the run -off water flows too deep below
the drainage lines, or there are no impermeable rock barriers to hold it up. On the other
hand desert rainfall is markedly affected by changes in ground elevation- An isolated
group of hills a few hundred metres high may attract rainfall many times that over the
surrounding country. So if the drainage lines are highly permeable to some depth, we
may find life confined to the high ground and none below.

It is these places, where a good water supply is concentrated along the drainage
lines, but where not much is now directly available, that offer the greatest scope for
artificial improvement. Mislead by unfounded theories of very recent climatic change,

8

we are only just realising how much was achieved on these lines in ancient times with
no more rain than now falls and later destroyed. This is a matter for engineers and
geologists.

V/ind Erosion Deserts. Sand Surfaces

As one approaches the cores of the great desert regions and rainfall becomes less
and less frequent, a stage is reached at which erosion by wind has for ages exceeded
that by water. The landscape becomes lunar. Stony plateaux alternate with escarp-
ments, isolated hills, gentle isolated depressions and wide sand- covered plains.
Drainage lines fade out and disappear. Since concentration of run -off may be negli-
gible, variation in the available moisture from place to place now depends on local in-
creases in elevation which attract more frequent rain, and on the capacity of the gen-
eral surface soil to absorb and retain rain where it falls.

When the soil is suitable in this respect temporary grazing springing up from dor-
mant seeds is able to mature and seed itself from a single rain storm after several
years of drought. A limit is probably set to the period by the viability of the seed and
its physical durability under conditions of sand blast and strong solar radiation. The
best soil is undoubtedly blown sand which is relatively clean of fine dust particles
between the grains. Water can descend very quickly through this soil since its anti-
wetting property is low and its permeability high. Owing to capillary tension a given
charge of water applied at the surface of dry sand will sink to a certain depth and no
more, the depth being something of the order of eight times the immediate precipitation.
Water which has reached a depth of 20 to 30 cm. remains as a moist unsaturated zone
for several years because, sand being a very poor conductor of heat, the temperature
is constant and 'breathing' nil. The sand both above and below is dry. A sand accu-
mulation is a good desert soil for two other reasons. In a wind erosion plain it is the
only place where deposition can balance removal; and it produces the only sloping
surfaces capable of appreciable local run- off concentration. Hence one finds that the
most favourable vegetation sites, indeed the only sites, lie along the lower gentle
slopes of dunes.

I suggest that since blown sand has an almost constant composition and texture
it might well be used as a standard soil for the purpose of estimating from the presence
or absence of vegetation the mean useful rainfall of those areas for which no long-
period records are available.

Variability of Rainfall ftom Year to Year. Unreliability of Records

In temperate climates where adequate rain falls throughout the year we are accus-
tomed to some deficit in any year from the mean annual value, and we do not bother
about it. But as the mean annual value decreases towards a desert region, and be -
comes confined to one season only, the expected deficit does not obligingly diminish
in proportion. A stage is reached at which the probable deficit at any one place is of
the same order as the mean annual rainfall. Beyond this stage the probable rainless
period exceeds one year. Instead of asking 'have the rains been good this year?' we

begin to ask 'has it rained this year?' or even 'how many years ago did you have rain
here?'.

I have noticed a tendency for Western civilisation to limit consideration of deserts
to areas where some though inadequate rain falls every year, and to neglect the rest.
This is convenient for those who try to measure the degree of aridity in terms of mean
annual rainfall, but it leaves the general picture of desert biology, and even of human
desert ecology, very incomplete. I think we have been mislead by rainfall maps. Lack
of data forces them to lump all desert regions into one final omnibus category of say
250 m/m to zero mean annual rainfall. Whereas if we had enough data to spread this
category on a logarithmic scale we should see biology stretching out far beyond the
limit of annual rain.

For the more arid areas rainfall data is both inadequate and unreliable, and must
remain so until we have 10- year automatic recorders. For it is against human nature
to look conscientiously at an empty rain gauge for several years on end. By the time
rain does come the gauge has probably been put to some other use, or the observer is
elsewhere. It is the rule in some more rain- favoured countries for the gauges to be
stored during the dry season and put out on a fixed date, and it is not unknown for a
single widespread rain storm exceeding a whole year's mean to remain unrecorded, be-
cause it fell too soon. Moreover desert recording stations coincide with human habi-
tation which needs permanent water, i.e. with spots of least elevation. Hence their
recorded rainfall is probably considerably lower than elsewhere around.

Beyond the limit of annual rains the biological significance of mean precipitation
dwindles rapidly. I suggest that the dominant factor which takes its place is the mean
period between effective storms. I would define an effective storm as a fall of such
magnitude that some water remains availably stored in favoured spots such as sand,
mud pans and rock fissures after immediate surface evaporation has ceased.

The mean rainless period, in years, unlike the mean annual rainfall which needs
careful quantitative measurement under very adverse conditions, already exists as a
clear estimate in the minds of nomads. Their lives depend on it. And this estimate
could be extracted by careful questioning. A fair estimate of the quantity of rain from
an effective storn could also be made from descriptions of the degree of flooding.

It is just possible that the mean rainless period, which we could get, might be
linked approximately with the mean annual rainfall, which we cannot get, and the lat-
ter, though insignificant, as thus obtained indirectly, could then still be used for the
sake of continuity of the measuring scale. Various scrappy bits of information rather
suggest that the precipitation from a mean effective storm remains fairly constant from
one part of a given desert region to another, provided due allowance is made for the
effect of ground elevation. For N.E. Africa which includes the most arid areas in the
world I would put this constant at 15 to 20m/m. Allowing 50% run- off concentration
this figure agrees with the precipitation needed to soak sand to a depth of 20 to 30 cm.
Similarly a guess can be made of the proportion which effective rain bears to the total
rain. We might put this at \ and assume that % of the total rain falls as light showers
and can be neglected.

10

On these rough assumptions, if R is the mean annual rain as measured by a gauges
E is the effective mean, and T is the mean number of years or fractions of a year be-
tween effective storms, we have E = c/T, and R = 3E = 3c/T, where c is the effective
storm constant which I will take as 18m/m for the Libyan Desert. For the neighbour-
hood of Cairo, where R = 40m/m, we get T = 1.5 years, and we should therefore expect
patches of blown sand, away from the concentration in wadis, to become green most
but not every year, which is about right. The assumed constant for an effective storm
and the ratio of effective to total rain are of course very tentative, and need investi-
gation. But the general idea may prove useful in default of any other means of esti-
mating infrequent rain. There are I believe no permanently inhabited places in the
world where an effective storm has not occurred in living memory. And the experience
of travellers in the reriote interior of the Libyan Desert suggests that this applies
even here too. Odd bits of local information from this desert seem to indicate a general
figure for T between 30 and 50 years, reduced to 4 to 10 years for the few isolated bits
of high ground. Taking therefore a general figure of 35 years for T for the Libyan
Desert as a whole, we get a mean effective annual rainfall at the present day of half a
millimetre and a quite unmeasurable gauge figure of perhaps three times this.

Nomadic Life

Rain over a great desert region does not fall everywhere at the same time, or in
the same year. Nomadism depends on this fact. It enables a whole tribe to live per-
manently in an area where effective rain falls at any one place only once in two or
more years. An extreme case is that of the indigenous Libyan Desert Tibu who till
recently wandered in small groups across hundreds of kilometres of lifeless 30- 50
year country from one favoured hill spot of 4 to 10 year rain to another, with a few
sheep or goats and even with a cow. Wild nomad fauna such as addax antelope seem
to roam over the same rainfall range. We also have the semi -nomad, based on the
desert fringe, who in certain years migrates desertwards with his cattle, but without
water, for the grazing to be had off 3 to 5 year areas, and himself drinks nothing but
his cattle's milk for six months or more.

Civilisation seems to have overlooked the nomad way of life, even though it ex-
ports meat. Surely no other way could be persuaded to produce anything at all from
large areas of the world. But for some reason one never hears it suggested that nomad-
ism might be encouraged and may- be modernised. Better varieties of the specialised
herbage might be introduced gradually, better control of grazing, radio for the more
rapid.spread of news of rain elsewhere. Why not, if we wish seriously to improve the
productivity of deserts? As things are, nomadism tends to be discouraged as a politi-
cal nuisance. If traditional nomadism is allowed to die, as it is rapidly doing, for ex-
ample where oil - fields are being developed, the chances of re-creating this way of
life seem remote. Vast areas which can now produce and export at least some food
will then be permanently unproductive.

Effect of Small Long- Period Rainfall Changes

In extreme cases of aridity where the remembered rainless period approaches the
span of human life it is of course iirpossible to get at the real mean period. This

11

would need many centuries of records. Indeed on this time scale the mean rain regime
may never be constant. And a small climatic change would have a very marked biolo-
gical effect. Using the rough rainfall scale I have mentioned earlier, an increase of
8m/m only in the mean effective annual rainfall (say 24m/m by gauge) would make
nomad life possible over most of the now lifeless core of the Libyan Desert. Thirty
to fifty year country would get an effective storm every other year. This must roughly
have been the condition in Neolithic times, may -be until as recently as 2000 BC, over
the southern half or more of what is now dead land. Significantly one finds their camp
sites concentrated towards the sands.

Effect of Cloud and the Season of Rain. Effect of Vi/ind Direction

Excluding the monsoon deserts of Asia it is a general rule that the tropical fringe
of a dry desert gets summer rain whereas the temperate fringe gets its rain in winter.
But in spite of the higher temperature it seems easier for a general herb cover to re-
vive under conditions of infrequent summer than of winter rains. The likely explana-
tion lies in the more continuous cloudy period associated with the season of summer
rain. On the harsher temperate fringe the growth of occasional spring vegetation de-
pends markedly on the duration of the less frequent cloud periods after rain.

There is a general tendency too for the winds of dry deserts to blow across them
towards the tropics. This may affect the methods adopted by specialised plants to
maintain themselves within their desert habitat by seed transportation. Where the wind
is very uni- directional as in the Libyan Desert one notices that on the fringe nearest
the temperate zone the desertward wind is made use of and plants of the 'tumbleweed*
type abound, whereas on the tropical leeward fringe the seeds or even whole trans-
portable plants tend to be barbed, to enable nomad fauna to carry them against the
wind.

A more important wind effect is the carriage of loess- forming dust from the wind-
eroded desert core outwards to and far beyond the fringe. The quantity so transported
must be enormous. Good evidence exists^^) of a desert surface being lowered 23 metres
since mid- palaeolithic times — say 4cm. per century. It is interesting to speculate on
how much less fertile the surrounding lands would be without the benefit of the desert.

The Biological Limit

In most desert regions the biological limit is never reached. In the Libyan Desert
trees may live on purely local catchment in places specially favoured by shade and
underground storage where it is said to rain only once in 15 years, (mean effective an-
nual rainfall about Im/m perhaps). Jerboa have been found where no other local life
is apparent, but they seem limited to within say 50 km. downwind of seeding plants.
Maybe they get their moisture from dew. The most extreme ecology I know of is that
of the few hawks and snakes who live in utterly lifeless country where there is no lo-
cally produced nourishment at all. Their ecology must be based wholly on casualties
from trans -desert bird migrations. But this in a way is cheating.

(1) G.W.Murray, 1951, Geogr. J.. 117 (4), All- A'iA.

12

THE AVAILABILITY OF UNDERGROUND WATER IN HOT DESERTS

Professor F.W.Shotton.
(Birmingham)

This paper claims to be nothing more than a general survey of the problems and
possibilities of obtaining water from underground in the really dry and hot parts of the
earth's surface where, without it, human existence would be impossible. If its con-
clusions are rather pessimistic, their recording may nevertheless be desirable as a
counter to the optimism which is often expressed, usually in broad generalisations,
and which so frequently proves to be based on experience in semi-deserts where rain-
fall is by no means unimportant.

In full desert, vegetation is either non-existent or scanty and specialised. Often
such plants that exist spring into a short-lived period of abundance after the rare
event of rain following perhaps years of quiescence. It is obvious that to convert
such regions into productive areas on any useful scale, a regular supply of water must
be ensured. This is true even if special drought resistant crops are developed, for
these can only be expected to grow when they are being supplied with water. It is not
my intention to discuss here those cases such as the valleys of the Nile and the Eu-
phrates or that peculiar accident, the Fayum depression, which though truly desert in
climate, can draw irrigation water from large rivers. Situations of this sort are usually
fully developed and, even where this is not the case, the controlling factors of topo-
graphy and volume of the river's flow can clearly be assessed. Over most desert areas,
no gift of a large river is there for the taking, and any hope of increased productivity —
of productivity at all — lies in the development of underground water.

Water obtained by means of wells and pumps must, apart from any question of eco-
nomics, satisfy two conditions:-

(a) It must be produced in sufficient quantity and

(b) It must have a quality, judged by its dissolved constituents, acceptable to man
for his own drinking, for watering hig stock, and for the irrigation of his crops.

The second factor may be discussed first. Much information on this point has
been summarized by Dixey(^) and it is quite clear that the limits of quality for differ-
ent purposes may be broadly lain down, even if there is no general agreement on exact
figures for these limits.

The salts which are commonly found in important amounts in water are common
salt (sodium chloride), the carbonates and bicarbonates of calcium, magnesium and
sodium and the sulphates of the same elements. Not all can occur together, for some
are incompatible; and the carbonates and bicarbonates of calcium and magnesium, and
calcium sulphate, though important in producing 'hardness', are of such low solubility
that they do not in themselves affect the limits of potability. Sodium chloride is often
the dominant constituent and may conveniently be taken as the basis for assessing the
quality of a desert water — remembering always that the sulphates of sodium and mag-

(1) Dixey, F. 1950, A Practical Handbook of Water Supply, 2nd Ed., London : Murby.

13

nesium, with their purgative properties, cannot be present in quantity without seriously
affecting the direct use of the water by man and his domestic animals and that 'black
alkali' (sodium carbonate and bicarbonate), even in small amounts, is not acceptable
to plants at present grown as crops.

Ideas on the standard of water acceptable to man for drinking have change con-
siderably in recent years. We may now take as a fact that water with a salinity of
3000 parts by weight of NaCl per million of water can be drunk regularly by human
beings in a desert climate, that a figure of 4000 unaccompanied by important quantities
of other salts is acceptable, and that for short periods even a figure of 5000 is endur-
able. Domestic animals are even more tolerant of dissolved constituents than man
though there is no close agreement on the worst limits of quality. Thus to take three
examples from Australia, we find Jewell^^^ in Victoria, stating that 3000 parts of total
dissolved salts (not simply sodium chloride) per million is safe for working horses,
dairy cattle and pigs, and setting a normal limit of 7000 and an emergency one of
10,000 for grazing cattle and sheep. Jack(2)^ in South Australia states that horses will
thrive on water with 1 ounce of sodium chloride per gallon (6260 parts per million) and
sets the upper limits for living as 7800 for horses, 9400 for cattle and 15,600 for sheep

— unless magnesium sulphate is present, when the figures must be lowered. Edge-
worth David and Browne(3) giving figures expressed as total solids, set limits even
beyond those of Jack's — 8000 for horses in work, 13,500 for horses at grass, 14,000
for cattle and 19,000 for sheep. However we attempt to reconcile these somewhat dis-
crepant figures, it is apparent that man is rather less tolerant than are his herds, but
that both can drink water which, as will be seen later, is of a quality not infrequently
obtainable in deserts.

To obtain water of a quality suitable for crop - irrigation is a far more difficult
matter. There are certain salts — the alkali carbonates and bicarbonates (black alkali)

— which are only acceptable to plants in very small concentration. Figures of between
100 and 200 parts per million have been given as limits and such amounts are often
exceeded in desert waters considered to be of good drinking quality. Apart from these
special constituents, the total amount of dissolved solids in irrigation water must also
be much below the limits of drinking water. It is not that many plants are intolerant of
brackish water. I have myself seen date palms and tamarisk growing well in ground
water with 6000 parts of NaCl per million — a water which a man could not take; but
the process of irrigation in a desert climate is inevitably accompanied by evaporation,
with the gradual concentration of salts in the soil to a point where plant growth is in-
hibited. The practical limit that has been given for the total solids in irrigation water
is only 700 per million. No doubt some easing of the stringency of this figure is per-
missible in semi-deserts, where there is sufficient rainfall to leach out some of the
accumulating salts, but in this paper I am concerned primarily with true deserts. In
these, the high quality necessary for irrigation water is the controlling factor on pro-

(1) Jewell, N.R., 1927, Water for Stock. }. Agric. Victoria.

(2) Jack, P.L., 1914, Bull. Geol. Surv. S.Australia, No. 3-

(3) Edgeworth David, Sir T.W. & Browne, W.R., 1950. The Geology of the Commonwealth of
Australia. 2, 514-593.

14

ductivity. Unless man finds such water in quantity, he can neither grow regular crops
for their own sake nor as fodder for his herds. The truth of this is exemplified in the
Australian artesian basins, where very few of the thousands of boreholes are used for
irrigation schemes, because of the amount of dissolved salts. Consequently most of
the boreholes lie in the semi-desert, where rainfall, exceeding 10 inches a year, pro-
vides natural grazing. If long continued drought causes this to fail, water from bore-
holes may prevent the cattle dying of thirst but not of starvation.

If, therefore, the desert is to blossom, water of irrigation quality and quantity has
to be found. Underground water is dependent on rainfall, for we can discount juvenile
water in this connection. Probably no desert is completely rainless, but a low perco-
lation is accompanied by a paucity or absence of springs and that, it its turn, means a
slow underground movement of water with ample opportunity to dissolve salts from the
containing rocks. Along the Palestine coastal plain, which is not a true desert, we
can see this relationship between declining rainfall and increase in the mineralisation
of the water until at Rafah, on the Sinai frontier, it is not easy to obtain even satis-
factory drinking water. Before reaching this point we can see the increasing difficulty
of running satisfactory irrigation schemes. My experience of North Egypt and Libya
during the war convinced me that a water table could be found almost everywhere in
this desert but usually of such high salinity that a random well has small chance of
finding drinkable water and next to no chance of water of irrigation quality.

Although this taking up of salts in solution is controlled also by the characters of
the rock holding the water, I think it is a fair assumption that irrigation quality water
is not to be expected in a desert from its own local and limited rainfall unless excep-
tional conditions exist. Of a number of such conditions, two may be mentioned. The
first occurs when newly -percolated rain, making its way to the water table, finds dif-
ficulty in mixing with the general body of saline water. In the Western Desert during
the war many water points were established through this cause(^), with salinities from
200 to 2000 parts per million in a vast area where normally the salinity stood at 5000
or 6000 (i.e. unpotable) and exceptionally went up to 60,000. Characteristic of such
wells were the thin depth of good water (typically only a few feet), the very sporadic
distribution of these patches (undrinkable water could exist only 100 yards away), the
gradual tendency to become more saline with pumping and the small yield which rarely
exceeded a few hundred gallons an hour. Indeed, the smallness of yield is an inevi-
table corollary of the fresh water — a fissure or pore system open enough to give a
large yield would not permit the fresh water to remain unmixed with the salt in the
first place. Such wells, therefore, have no importance in irrigation prospects.

The second possibility to be discussed is that of perched water, where geological
structure causes the holding up of water above and quite separate from the main table.
The controlling factor is often a bed of shale or clay occurring as a lens or a fold be-
tween two aquifers. Such a structure has its limits and as rain joins it, there must be
an overflow from the perched position either as a spring (which is unusual in deserts)
or underground to the main water table. In this way, a one-way movement may be set

(1) Shotton, F.W., 1946, Vi/at. & U'a<. Engng. 49, 218-226.

15

up, leaching out the soluble salts from the upper containing rock until the water there-
in has little dissolved matter to take up.

Several examples of this type were developed in the Western Desert during the war,
the most notable being at Fuka(^), Its structure^ with a bed of clay separating two
limestones and folded into an elongate basin, has been fully described. From measure-
ments of exhaustion and replenishment after rain, it was calculated that 25,000 gallons
a day could be taken out without risk of failure and that the structure carried a reserve
equal to five years' supply to tide over any winters of low rain replenishment. Hence
there would appear here to be the possibility of a small irrigation scheme (but only of
about 20 acres, if we are to accept David's figures and allow also for use of the winter
rains) provided that the water was good enough. Actually the average of several ana-
lyses shows :

Total solids 1400 parts per million

Sodium chloride 750 "

Sodium carbonate 50 " " "

Sodium sulphate 160 "

Calcium (plus magnesium) carbonate 330 " " "

It is thus somewhat above the limit mentioned earlier, but in view of the fact that
there is here a winter rainfall of perhaps 6 inches which would tend to dissolve out
such salts as had been deposited in the soil during the preceding summer, an irrigation
scheme appears possible.

What must be emphasized, however, is the pitiful inadequacy of this, our only
spectacular example of perched water - a structure of 170 acres, capable of irrigating
20 acres, in a desert that was considerably if not exhaustively probed over perhaps
8000 square miles.

Wartime experience in the eastern Egyptian desert (Red Sea Hills), where rainfall
is extremely small and sporadic, showed that by careful attention to geology, aided by
geophysical measurements, underground reservoirs of drinkable water could be found •
Of 10 wells with drinkable water, 5 were of irrigation quality; but the yields were only
of a few hundred gallons an hour with a limited life, and so useless for irrigation
schemes.

I feel therefore, that it should be emphasised that in hot deserts, with their very
low rainfall, the derivation from this of underground supplies large enough and fresh
enough for irrigation must be an exceptional occurrence, the result of a combination of
geological accidents that can only occur very rarely.

There remains one other source of hope. Some deserts may have a geological
structure where a sedimentary formation, occurring at depth, eventually outcrops in an
area of normal rainfall beyond the confines of the desert. Provided this formation is
water- conducting and insulated from contamination with whatever higher saline water
exists, it may be entered by borings and good water may be obtained. Notable examples

(1) Shotton, F.W., 1944, The Fuka Basin. Roy. Engrs. J.. 107-9, 1946. Wat. & VJat. Engng
49. 257-263.

(2) Paver, A.L., 1946, V/at. & V/at. Engng. 49, 653-662.

16

of this effect are provided by the deep artesian wells of Tunis and Tripolitania, by the
Nubian Sandstone (Cretaceous) which outcrops in the Sudan but gives good water to the
depressions of Kharga and Dakhla, 650 miles to the north and, of course, by the vari-
ous artesian basins of Australia.

It is undoubtedly in such large-scale geological structures that the main hopes
for irrigating deserts lie. It would be well not to exaggerate those hopes. Artesian
supplies in a desert can only materialise in large and useful amounts (a) if the aqui-
fer outcrops outside the desert region, i.e. in an area of adequate rainfall ; (b) if the
aquifer has high permeability beneath the desert, so that there can be underground
transfer of water and the obtaining of high yields; (c) if the quality of the water re-
mains suitable for use in its long underground journey from the intake; (d) if the water
budget is balanced — i.e. the extraction is balanced by intake.

Mainly on the second ground, Du Toit held out little hope of artesian supplies from
the Karroo. In the Great Artesian Basin of Australia, much of the water is mineralised
as a result of its underground journey from intake to well, to the extent that although
it is acceptable to cattle and to man, it is not usable for irrigation.

The great African hollows of Kharga and Dahkla, where artesian springs were once
abundant, are notable examples of large-scale irrigation (mainly of date groves) from
artesian water of good quality in an area which is virtually rainless ; but attempts to
intensify that cultivation have been accompanied by a continuous demand for more and
deeper boreholes, with a steady lowering of water pressure and the drying up of springs.
The artesian water,here, then, is a slowly wasting asset on the present scale of culti-
vation. Such a picture does not encourage any belief in a spectacular increase in the
use of the Nubian Sandstone water in the Western Desert, even if low- lying areas can
be found where the water may be met at depths sufficiently shallow to allow economic
pumping. Nor is there much hope that this good quality water extends as far north as
the Egyptian and Libyan coast, in view of the bad water of the northern oases of Siwa
and Jiarabub and the saline water in a deep bore at Tobruk.

The deserts mentioned above are more fortunate than others which seem to have
no hope of a deep- seated supply. The great Arabian Desert, for instance, probably
has no suitable geological structure for artesian supply and even if it had, it could not
satisfy the necessary condition of an outcrop in a region of good rainfall beyond the
confines of the desert.

This paper is a very general survey, biassed perhaps by the deserts which I know
personally. It will have fulfilled its purpose if it sounds a warning against the opti-
mism which sometimes pervades discussions on the conversion of deserts to useful
productive land. Again it must be emphasised, however, that only full desert has been
under discussion. In those fringe areas (South Palestine and parts of Jordan are good
examples) where nature has provided at some period of the year an adequate rainfall
and yet turns the country to arid desert in the summer, there is every incentive to
search for underground water and to use it to balance out the irregularities of the rain-
fall. The search may often be long and difficult and the results must always conform
to the law that more water cannot be taken from the ground than soaks into it; but sub-
ject to those limitations, there is a future for parts of the semi-desert earth which
most of the true desert cannot hope to share.

17

SOME BIOCLIMATIC OBSERVATIONS IN THE EGYPTIAN DESERT

Dr C.B.Williams.
{Rothamsted Experimental Station)

I resided in Egypt from July 1921 to June 1927, and during this period I made three
short expeditions to the hilly desert to the south- east of Cairo to take observations on
bioclimatic — or what perhaps today would be called microclimatic conditions. The
object was to discover the range of temperature and other environment conditions avail-
able to animals with a power of choice, and of movement over short distances.

The results have already been published in technical and scientific bulletins of
the Ministry of Agriculture of Egypt (see bibliography at end), but as these are not easy
to consult in libraries, it was thought that a new summary might be useful to ecologists.

The locality chosen was in Wady Digla, a dried watercourse about twelve miles
south- east of Cairo and about seven miles from the nearest cultivation in the Nile Val-
ley. The wady (or valley) at the point chosen runs from east to west and is about 300
yards across at the top, about 80 yards across at the bottom and about 200 feet deep.
The rock is a pale brown limestone. Rain falls on an average not more than once a
year.

Three visits of eight days each were made in August 1922, March 1923 and Decem-
ber 1923, and on each occasion meteorological readings were taken, for seven consecu-
tive days, every hour from 5 a.m. to 11 p.m. and again at 1 a.m. In addition to these
three longer visits, many one or two day visits were made at all times of the year, and
on some of these temperature and humidity conditions were recorded.

Readings were taken in a variety of locations, including a Stevenson's screen in
the middle of the wady — in the shade of the rock on the south side of the wady — under
a large rock where it was just possible to crawl — on the plateau above the wady — at
different depths up to 30 cms. in a sand patch alongside the dried water course — at a
depth of about 75 cms. down a Jerboa burrow — in a bird's nest in a bush, and in an-
other hole in a rock- in two *ant- lion' pits one in the sun and one in the shade — and
at various depths in two caves. The records included black and white bulb tempera-
tures — wet and dry bulb temperatures with a sling-psychrometer — wind with two cup
anemometers — and evaporation with 'Piche' evaporometers.

Table I shows a summary of most temperature and humidity records (except those
in the caves) in each of the three periods of observation, and Fig. 1 shows diagrama-
tically the means and extremes in many of the habitats.

It will be seen that in August the black bulb thermometer reached 74°C (166°F),
the surface sand reached 58°C (136°F) and the temperature in a bush reached 44°C
(111°F), while the shade temperatures only reached 35°C (95°F) which was less than
the maximum of the sand at 20 cms. The August week was not exceptionally hot, and
shade temperatures 5° or even 10°C higher might well occur at this time of the year.
During this period a temperature of 44. 2°C (112°F) was recorded in the sand of an 'ant-
lion' pit, and at this temperature the ant-lion larva just below the surface immediately
captured an insect which was dropped into the pit.

18

pri iTv "<r

so CM

o

O C\

o

o

r^

\0

«^ O 1-1

\0

r-

O CS

.

. • •

J* Cs

C\ 00 00

-o o

1—1

O <N

<N

r^

cs

0(5

so so fN

1^

rO rO r^ O

—H

Cs

1-^ ~<r

a ^

r-j CM

(N

CM

(N w*

^H

.-4

.-4

f— ( 1—4

cs

00 -<r (N -<r

1-H

•g ro

\0 (N \o

r- r-

•^

vq ^

C\

1^

i^

<r\

CN SO so

SO

r--

<-> cs

u <N

• a •

• • •

3 c\

ro CS l~-

\r\ ^H

"O

U-\ ITN

ro

cs

f— t

1-H

1^ r- ro

fN

iO cs 1— 1 so

0(3

r^

tr\ rr\

2 -^

^ rr\

m

<N

(N — 1

1-H

1— (

rN

fN

1—1 1—1

00

r^ (N (Nl ><r

w

•

•

• • •

E

CO fs)
3 CM

tN — 1 SO

<r\ o

prv

o so

•<r

(3s

00

<D

ro PO "^r

00

, r~-

o

• • •

• •

•

1

.

«*CN

00 \o o

r- vo

1—1

o r--

^

<N

r-

vd

o cs so

so

(Nj -H so •-'

d

1

' (TN

(X)

9 ^

iri >r\ (N

1— 1 rCi

■<r

•<r CM

(N

1-^

ro

ro

ro fN

Cs

00 <N| -H so

(N

—4

<

4J

(9

•

6?

•

E

E

>*

N-^

Q

H

u

U

V

4>

Q

u

ai
D

w
oi

<

Max.
Max.
Min.

Min.
Rang

•

2

Max.
Min.

•

C

2

c

OS

(d
2

•

X

cd

2

Min.
Min.
Rang

2

D

X
(9

2

Max.
Min.
Min.
Rang

.

X
(9

2

.

X
(9

2

Mm.

Min.

Q
Z

r.l (0 ^ ^

(0 S

c/3
2

«

(«

§ §

(0

§

CO

•

(A

M 3 (0 §

w

•

n H to n

(A

§

n (0

Si V

Si

V u

ji

V

2

^

u

01 ^ V

>

St

V V Si V

a:

3

^

u

U -C

^ < 2 2 <: 2
<

U

<

2 2 <

2

U <

2 2 < 2

H-4

<

2 2 < 2

<

2

2 <

C/D

OS

o

1—1

o

<
J

O
QU

CQ
<

p

H

H

W

<

C/3

<!

<

ci

>

• ro
U CM

-H r~ (N

o -<r

V3

(N -^t

o

r^

>r\

HT

1 1

1

1

1 1 1

1

o

cs so O

r- <D

1> CN

<N o o

00 o

•-M

d r^

d

00

r-^

SO

1 1

1

1

1 1 1

1

CO

so ro (N

r-^ xA

Q ^

CM (N .-H

rsi

(N '^

1—4

\r\

\r\

1—1-^1—1

u 'T:

00 (NI o

(N ro

ro

r- -<r

00

^

iXi

-^

o r-.

\r\

o

-<r so ^

o

\r\

ro r-- lo

-H rvi

M <^

ea CN

PO (N| o

r- (N

rs

o o

r--

o

o

Cs

^ Cs

(N

1—1

so ^ d

r~^

cs

00 ro r4

aj so

2 --

fN (VI ■— 1

•—1

(N

(N I— 1

f— «

so

\r\

(N ^H

1— (

f— 1

rsi (V) _(

1— *

*J

« fN

1 1 1

1 1

cs

ir\ 1^

CN

o

■<r

•^

O (N|

-<r

fN

ro <N -H

<?s

1

1 I 1

1 1

«>Cn

1 1 1

1 1

\rs

■^ (Ni

1— (

(N

■^

(Nj

r^ \r\

•^

ro

rr, (N f<^

f-H

1

1 1 1

1 1

^'-

ro

ro CN

(N

1—1

r->

r^

ro ro

(Nl

fN

^ ^ fN

fN

U

0

Q

<!

O

u

^

u

(/5
w
a;

H

ax.
lax.
in.

ID
Ml

5 §

ta

•

C

CQ

•

(9

(9

GE RO
ax.

ax.

c

C

IN BU

iX.

ax.
in.

«

c

o

oi

z

X
(9

2
w

Sic

D MINIMUM
ean Min.
bs. Min.

<
w

EEN
bs. M
lean M
ean M

s. Mj
an R

w

Q

2

•

2 2

a c
(3 (a

2

(fl

c

2

(0

2

LAR
s. Mi
an M

2

c

C9

2

EST
s. Mi
an M
an M

2

!/5

H

CO

w

an M
an M
s. Mi

IX
2

W
H

Si lU

JD

(U lU

s

OJ

1^
u
<

JD

a;

g<2

Q

Z

4)

s

Z -Q OJ «j

Si

Z

J3

V i> s

« < 2 2
U

00

Z

< 2

<

2 2

<

2

<

2

2 <

c/5 < 2 2

Q

en

<

00

Q
2

<

2 2 <

O
oi

c^

CQ

D

CQ

5

O

19

OC

70

DECEMBER

MARCH

AUGUST

I

BLACK
BULB

op

170

160

150
140
130

120
110

100

90

80

70

60

50

40
OF

60

50

r BLACK
* BULB

I

WHITE
BULB

40

30

WHITE
BULB

R

W

T *

5 r

20

10

OC

S W

1 . " -

I

s

T W
;; T R
j: T

_A B_

NESTS

SAND

DECEMBER

CAVE

B

I

* : :

&

A i:

NESTS

CAVE

SAND

» -^ ■*" t CAVE

NESTS SAND
MARCH

AUGUST

Figure 1.

Wadi Digla near Cairo. Temperature Conditions in Desert.

S = Screen W = Shade in Wadi R = Under Rock. Nests, A = Bush B = in Hole in Rock

Sand at 1. 10. and 20 cms. Cave at Mouth, 5 and 12 metres in.

Fig. 2 shows the hour by hour changes in temperature in some of these locations
during five days in August. Apart from the great daily range of temperature in the
places exposed to the sun, there should be noted the low range but high average tem-
perature in the Jerboa burrow, and the low range and low average temperature in the
sand in the shade, as shown by the ant-lion pit. It is also interesting to note that the
maximum temperature at the end of the Jerboa burrow (which was estimated to be about
23 cms, below the surface) occurred about 9-10 p.m. and the minimum about 8-9 a.m.,
the former about 8 hours and the latter about 4 hours later than the corresponding stage
at the surface.

20

4 AUGUST

5 AUGUST

8 AUGUST

9 AUGUST 1

I 1 1

1 1

AIR SHADE '

1 1 1 ' ■ ' ' 1 ' ' 1

. « • • • • HULK UNUKK SI UlNa

ANTLION PIT IN SHADE

• e > • • BIRD'S NEST IN BUSH

....... ANTLION PIT IN SUN

45°

----. JERBOA BURROW j^go

OC

t

•

X

ft

• •
«
•
«

'.^

O

•
•
•

»

°F

1

*

4ni>

i

Q

/

»

«

•A

104"

4

1

\

•

2

•

\

•

•

IS

1
•

1

•

\

t

•

\

•
•

35°

»

•

•

t

A

95°

1

h

1

1

•

•

•
0

«

■

/

V ■

«

•

/

[Y

•/

\

\%

<

/

\''- 1

■(

\

>.

30°

1

«

/

>

•
•

•
«

1

*

•7

/

>

i

^^

.V

•

f

t

>

9 i

\

«

/

«

/■

V^,'

1

«

1

• /

\

i

■f'

V.

•

\

"...

/

•

r'

V

•

/

.-

»^

V

•
%

1

\

•
4

»

\

»v

/ -

•

1

*

L i -^

>

\>

-'1

'"

; '

\

•7

25°'

"t

\

/

u

A

7

770

\

y

•^

J

•^

■J

\

J

NO

ON

NO

ON

<^

NO

ON

NOON

4 8 12 4 8

4 8 12 4 8 1

4 8 12 4 8

4 8 12 4 8 1

A.M. P.M.

A.M. P.M. 1

A.M. P.M.

A.M. P.M. 1

Figure 2.
Hour by Hour Changes in Temperature

Fig. 3 shows the range of combined temperatures and relative humidities covere
by hourly observations in the shade on the south side of the wady, calculated from
sling- psychrometer readings. In August air temperatures ranged from 21° to nearly
37°C and relative humidity from 95% in the early morning to 17% in the late afternoon.
The range of humidity was slightly less (with distinctly lower temperatures) in March,
but reached almost to saturation in December. In this latter week there was, however,
a sudden change of air moisture resulting in an unusual range of combined temperature
and humidity.

For comparison with the extreme desert conditions in August there is shown the
range covered by 24 hourly records taken at sea in the Mediterranean one day in August.
The contrast speaks for itself.

As some insects and other animals are known to be active in the low light inten-
sity both at dusk and at dawn, the dusk and dawn records are shown individually in
Fig. 3, indicating the very wide difference between the cool damp morning and the warm
dry evening conditions.

As many animals burrow in the sand in deserts by day time, perhaps to escape ex-
treme conditions, special attention was paid to changes in temperature at different
depths in the flat topped sandy areas which were found here and there along the dried

21

Figure 3-
Combined Temperatures and Relative Humidities

water course in the centre of the wady. Fig. 4 A shows the gradually changing tem-
perature at different levels in the sand during 36 hours in August. Surface temperature
reached a maximum of 56*^ (133°F) at about 1 p.m.; at 5 cms. the maximum was 44°C
(111°F) at about 3 p.m.; at 10 cms. just over 40°C (104°F) at about 5 p.m.; at 18 cms.
36*^ (97°F) at about 7 p.m.; while at 28 cms. the temperature only ranged from about
23° to 24°C (73-75°F) with a maximum about midnight. The minima ranged from about
20°C (68°F) at 5 a.m. at the surface to 23°C (73°F) at 28 cms. about mid -day. Fig 4B
shows the temperature contours at different depths during the day and Fig. 4C shows
the movements of heat in the sand at different times, the surface heating during the day
and cooling during the night. The lines where the heat movement is momentarily zero
have been called the 'thermostatic lines* (see McKenzie Taylor and Williams 1924).

Fig. 5 shows how the observed changes in temperature at different depths in the
sand in August support the theory that the range of temperature at a depth x is given
by the formula R = Ra'^ where R is the range at the surface, and a is a constant for the
particular sand or soil. The figure shows above the observed maxima and minima, and
below the expected range calculated from the above formula with (3 = -01 3, and as 'cros-
ses* the observed ranges which fit extremely closely to the calculated values.

22

Figure 4.
Changing Temperatures at Different Levels in Sand

23

- 50OC

DEPTH IN CMS.
I I I I I 1 — I I I I I I I I I I I I — r—i — I I I I — I I I I

X 5 cms 10 15 20 25 cms

40

o

:. ^-^Z^J^V]^ :

SAND TEMPERATURES
AUGUST 1922

oO^C

40*^

DEPTH IN CMS,

Figure 5.
Temperature in Sand — Observed and Calculated Values

Microclimatic conditions were also investigated in two caves in the sides of the
wady, the first cave in August and March, the second in March and December, with a
few extra observations at other times of the year. The values for temperature and humi-
dity are shown in Table 11, and diagramatically in Fig. 6.

Fig. 6 also illustrates an interesting point with regard to relative humidity. It was
found that the vapour pressure inside the cave tends to come into equilibrium with that
outside. Vapour pressure is generally lower in winter than in summer, but the differ-
ence in temperature results that relative humidity is higher in winter than in summer.
Since at a depth of about 20 metres into one of these caves the annual temperature is
almost constant it follows that the relative humidity inside the cave tends to be higher
in summer than in winter, just the opposite to what is happening outside. It must be
remembered, of course, that these caves are surrounded by completely dry rock. Under
normal European conditions cave air tends to be quite saturated with moisture.

24

TABLE II 1

FIRST CAVE

TEMPERATURE °C

MOUTH 5 METRES 12 METRES 1

August (min. and^max.)

21.0-36.3 23.0-30.4 24.0-

25.4

range

15.3

7.4 1.4

March (min. and max.)

8.8-24.0 12.4-21.2 19.3-

21.4

range

15.2

8.8 2

1

December (min. and max.)

—

21.5-

23.0

range

^

- 1

5

SECOND CAVE

TEMPERATURE °C

MOUTH

5 METRES

15 METRES

25 METRES

March

12.0-25.5

19.3-22.3

23.2-23.4

24.2- (25.0)

May

20.0-29.0

22.3-24.3

24.0-24.2

24.0-24.0

September

22.5-33.0

25.4-26.3

24.5-25.0

24.0-24.2

December

10.8-10.8

19.0-23.0

23.2-23.6

23.7-24.0

Total range

22.2°C

7.3°C

1.8°C

1.3°C

RELATIVE HUMIDITY %

March

28-78

(40)- 54

33-35

35-38

May

25-62

37-52

42-47

37-38

September

34-88

50-58

57-60

55-56

December

36-56

33-36

30-39

38-43

Total range

63%

25%

30%

21%

VAPOUR PRESSURE in MM.

March

6.7- 8.2

(7.9)- 9.0

7.0- 7.3

7.9-(9.1)

May

7.6- 10.8

'8.3-10.2

9.4-10.3

8.4- 8.6

September

12.9-17.6

12.7-14.1

12.8- 14.0

12.3-12.4

December

5.4- 6.9

5.5- 7.5

6.2- 8.6

8.5-(9.5)

Total range

12.2 mm .

8.6 mm.

7.8mm.

4.5mm.

During the three periods of observation there was no general deposit of dew on the
ground, but one day in August there was dew on the wind gauge, and on twigs of vari-
ous dried up plants on the plateau above the wady. On this morning the dew point at
5 a.m. was 17°C and the surface soil 17.5°C, so that a very small further fall of tempera-
ture would have produced a general ground dew. It is interesting to note in this con-
nection that a desert plant, Reamuria hirtella J. and S. of the family Tamaricacea,
which was not uncommon in some spots along the wady, was found to be dripping wet
in the early morning whenever the relative humidity of the air was above 75%. This
was found to be caused by small crystals of sodium chloride on the surface of the plant
which absorbed moisture from the atmosphere above this relative humidity.

It will be clear from the above that within a distance of relatively few metres there
are available in this type of desert country a very wide range of temperature conditions

25

METRES INTO CAVE
2 4 6

2 4 6

METRES INTO CAVE

Figure 6.
Temperature and Humidity

26

from among which an animal can choose by little expenditure of energy. By burrowing
deep an animal can avoid the extreme heat of day — and by leaving the burrow at night
it can even escape the peak temperature below, as at a foot or so beneath the surface
there is a lag of about 12 hours in the time of maximum temperature. Almost as big a
choice is available in this hilly type of desert by moving into the more or less perma-
nent shade beneath the steep south side of the valley. An experiment was made one
day in August by artificially shading the sand where the temperatures were being mea-
sured. The maximum surface temperature under these conditions was 20°C (36°F) lower
than the previous day when the sun had been shining, while at 18 cms. deep the maxi-
mum (not reached till about 7 p.m.) was 5°C (9°F) lower than the previous day.

References

McKenzie- Taylor, E., & Williams, C.B., 1924. A Comparison of Sand and Soil Temperatures in
Egypt. Min. Agr. Egypt. Tech. and Scient. Bull. No. 40.

Williams, C.B., 1923. A Short Bio -climatic Study in the Egyptian Desert. I.e. Bull. No. 29-

Williams, C.B., 1924. Bio-climatic Observations in the Egyptian Desert in March 1923. I.e. Bull

No. 37

Williams, C.B., 1924. A Third Bio -climatic Study in die Egyptian Desert. I.e. Bull. No. 50.

27

PLANT ECOLOGICAL PROBLEMS IN INCREASING THE PRODUCTIVITY OF

ARID AREAS

Dr H. Boyko.*
(Jerusalem)

Approximately a third of the surface of the earth (Sears states 31%) is located in
arid zones, yet only a relatively small part of this area can be eliminated from this dis-
cussion on the basis of being climatically absolute desert. By far the largest parts of
the area can be classified as semi- deserts or man- made deserts. If we face the issue
of increasing productivity from the point of view of the plant ecologist, we must occupy
ourselves primarily with these regions.

Among the tasks of ecology, pasture ecology occupies a position of major impor-
tance. I deliberately use the term 'pasture' in order to avoid use of the word 'grass-
lands', for it is precisely the steppe and prairie regions which constitute the second
boundary to the scope of any topic, both objectively and geographically. Though these
may also be in an arid region, they are nevertheless always covered by a dense blanket
of vegetation, alive or dead, even during the dry season. Between the areas completely
under vegetative cover and those of absolute desert which is climatically caused, lies
the main area subject to discussion here. Israel provides a very good example. Driv-
ing through this small country, we pass within several hours through the majority of the
global vegetative zones located in the Northern Hemisphere, namely the forest belt, the
steppe belt, the semi-desert belt, and the climatically- caused absolute desert.

From the North - Mediterranean Laurel forest climax we pass through the Eu-\dedi-
terranean, semi- arid Quercetum cocciferae, and through the more arid sub -Mediterra-
nean forest association of Ceratonietum soliquae into that belt, which occurs globally
between the dense forest climax and the steppes, to which the above mentioned asso-
ciation is already a transition stage. This zone I call the arid border forest belt. In
south-west Asia it is divided into the Anatolian- Iranian Quercus Aegilops belt and the
Mauretanian- Iranian Pistacia mutica belt. These two intersect in Israel, from whence
they strike a wide arc round the Mesopotamian lowlands, the Pistacia further inland
than the oaks. Adjacent to this border forest belt is the Stipa steppe belt, followed by
semi-desert and desert. Analogous conditions occur all round the globe in both hemi-
spheres, with the possible exception of Australia. There we find a considerable amouni
of trees in a climatic zone where elsewhere we would expect only a treeless semi-
desert.

This surprising phenomenon may indicate that in Australia tall, woody species
have had a much longer geological period in which to adapt themselves to drought resis
tance than is the case in other continents. In the steppe regions of south-west Asia
perennial herbs predominate, after grasses, as in the Anatolian steppe, then the tall
feather grasses gain dominance, the cover becomes ever sparser, the bunches are far-
ther apart, and we pass from steppe to semi-desert. Here woody plants, bushes, and
bushy perennials predominate, and between them there grow bunches of low grasses of
varying density. In the spring, depending on the rainfall of the preceeding weeks,

• Chief Ecologist, Ministry of Agriculture and Development, Israel.

28

there are larger or smaller numbers of annuals, but for a short time only. Areas with less
than 100 mm. annual rainfall are to be designated as desert, and only on sand and in the
wadis is there somewhat denser vegetation. Under still more arid conditions, only in
wadis and oases do we find larger numbers of individuals in what can be called an as-
sociation.

Such is the natural picture where man has not interfered. This however is seldom
the case. The largest part of the border forest belt, has been converted to steppe, the
stQJpes to semi-desert, and semi-deserts into deserts. In the Orient, where over-
grazing has been the practice for millenia, and where the equilibrium between plant, ani-
mal and man has long been disturbed, this development is especially prominent. In or-
der to create a regeneration here, we must first determine the climax association of the
region under consideration.

The reconstruction of the original climax associations, and the mapping thereof, is
one of the most important tasks of the plant ecologist in these regions. Without this,
planning land- use cannot be on a sound basis. Another line of research is based on the
biological rxiles of climatic extremes. At a boundary of the distribution of a species or
of an association, the smallest variations on the environmental conditions are decisive
for its existence. Plant ecologists must therefore determine these ecological ampli-
tudes and the geographical boundaries on the one hand, and, learn to recognize the vari-
ations in the environment, and to measure them quantitatively on the other. Comparison
between the two phenomena will always yield most important and far reaching results.

Furthermore, when we wish to create a more dense plant cover, or to increase the
population density of a particular species, we must also take into account shifts in the
complex of factors in these boundary regions. I wish to present the example of the 'IE-
factor', the factor complex involving insolation and exposure. In the low latitudes, it is
surprising what large differences in micro- climate result from variations in the degree
of inclination on the same slope. Ashbel found that the difference in insolation between
a horizontal surface at the latitude of Jerusalem, and a 40° north facing slope is as
great as that between two horizontal surfaces, one located in Jerusalem, and the other
in Paris. Testing the results of my own investigations with regard to the decisive ef-
fect on vegetation, the same e^^erience has been made in U.S.A., Australia, etc. As a
result, we can pass through three distinct floral regions on a northern slope in the Wild-
erness of Judea east of Jerusalem, all within a few metres of each other. On a 25° slope
we find a Mediterranean flora, on horizontal surfaces a Saharo- Sindian, and on inter-
mediate slopes we find a transitional Irano - Turanian flora.

The whole Far Negev, (that is the part of Israel lying south of Beersheba), can be
considered to be an especially favourable area for experiments on all arid regions and
therefore my department there has set up three permanent observation stations, which
are complementary to each other both edaphically and climatically. One is situated in
the mountains near the ruins of the ancient city of Abdeh. Here we are attacking pro-
blems of pasture regeneration and the utilization under control of the torrential winter
floods. The second is at the oasis of Ein Ghadian in Wadi Araba, the deep trench bet-
ween the Dead Sea and the Red Sea; and the third is the most important Desert Garden
at Elath on the Red Sea, where my wife Dr Elizabeth Boyko is in charge.

29

Our plan for productivization of the mountainous part is, firstly, the regeneration of
the severely destroyed climax vegetation in the mountains and on certain sandy plains,
which could provide about 4-5 months pasturage a year following regeneration. One may
also assume a certain degree of self- sufficiency of the settlers in respect to grain,
vegetables, and fruit, if measures are taken to utilize the torrential winter floods in the
wadis. The latter problem is more technical, the former is purely geological. In this
area, which has been suffering from over- grazing for thousands of years, the theoretical
reconstruction of the climax associations depends in the first place on the finding of
good fodder plants, especially those rare species in inaccessible places.

Another important task is the determination of the palatibility of the various spe-
cies, since those of highest palatability are of course most in danger of total extenction.
On the other hand, unpalatable plants should also be studied in order to ascertain the
reason for their being shunned by grazing animals. Finally, the competitive power of
these plants must be evaluated and compared.

It would be helpful if small areas of such destroyed pasture regions that occur in
the Orient and in Africa, could be fenced off and protected from grazing for a time. In
a very few years important changes indicating the tendency of the natural succession
could be observed.

I should like to bring you another example from the Negev. The small shrub Arte -
misia herba alba is scorned by all animals. Consequently it has been able to conquer
large areas including all the mountains above the altitude of 450- 500 metres. As soon
as we eliminate grazing, however, the previously rare tall feather grasses multiply
rapidly, indicating a trend towards the actual climax condition, in which good fodder
grasses such as Slip a barbata, Stipa Szowitziana, certain Aristida spp., and others
dominate. It is clear that the vigorous fibrous root systems of these bunch grasses can
compete with the much weaker roots of Artemisia herba alba. The seeds of Stipa find
good conditions for germination in the bushes of Artemisia, and if they are not devoured
at an early age, may eliminate the Artemisia within a matter of years. Such studies of
succession lead to the reconstruction of the climax association, and only on the basis
of this can one begin to make serious plans for the conversion of man-made desert and
semi-deserts into pasture land.

In the days of the Palestine Mandate I had the opportunity to wander through these
remote areas on camel -back and I have therefore been able to observe their develop-
ment for a number of years. The most striking phenomenon is the increase of good fod-
der plants after the cessation of grazing. In the Tureibe region, a sandy area of about
30,000 acres, together with my assistant, Mr Tadmor, I carried out an exact statistical
study of their development. The unpalatable species remained constant in their popula-
tion density, while the highly palatable ones have increased in density approximately
ten times during the past five years, while there has been no grazing. I am referring
principally to Aristida plumosa, Danthonia forskahlei. Convolvulus lanatus, Argyrolo-
bium uniflorum, etc. Calligonum comosum, a large and valuable fodder bush, has not
multiplied during this time. As is the case with many trees and bushes, it apparently
needs one or more years of favourable conditions in order to germinate and to multiply
naturally. However, multiplication by the means of cuttings is within the range of eco-
nomic possibility.

30

Vi'e face ecological problems of an entirely different nature when we look for, and
even after we have found those desert plants which come under consideration as possible
sources of human food or of raw materials. Searching for, and finding them is only the
first part of the problem. Their introduction into commerical production is another com-
plicated problem. There are two ways in which the finding of such plants can take
place. One is the systematic comparison of useful plants from other countries with the
plant population of the country under consideration, and testing these plants for their
quality. The second, and in the end much more important way of discovering new trea-
sures from the plant world is by constant observation of nature, and investigations of
genetical relationships. A certain amount of intuition is of course always involved. In
my opinion, purely accidental discoveries are very rare indeed. The second way is much
more likely to lead to the discovery of new facts, since essentially the first only treats
material already known. Nevertheless, it should not altogether be neglected.

In Wadi Hafir I once observed a Bedouin woman who came to draw water from a
cistern. Most likely she had walked several miles in order to get there. The water
table had dropped, however, her rope was too short, and consequently her bucket did
not reach the water. I was about to lend her a rope of ours, when I saw that she went
to the nearest bush of Thymelea hirsuta and began to peel off the bark. With some of
this she made her rope longer, and hauled up the heavy buckets of water without diffi-
culty. I tested the tensile strength of the fibre myself, and found it to be surprisingly
great. I had a large sample collected immediately and its subsequent chemical and phy-
sical analyses showed that here was a valuable potential cellulose and fibre plant.

Let me take another example. It is generally known that the bulbs of Colchicum
species contain Colchicin, the demand for which is greater than the supply. Those Col-
chicum spp. from Israel that have so far been analyzed, contain a relatively large amount
of Colchicin. The plants occur scattered as single individuals however, not as Colchi-
cum autumnale which occurs on wet meadows in Europe in such masses as to endanger
the cattle grazing there. In such conditions the collection of the plant is simple and
inexpensive, and of value to the owner of the pasture. The situation is entirely differ-
ent in the semi-deserts and destroyed pasture lands of Israel, however, and in the Mid-
dle East in general. Here the plants generally occur as rare, or at least as scattered
individuals, and the digging of them from the hard soil is difficult, expensive and soon
destroys the stand altogether. I therefore collected the flowers of the plant and sent
them for chemical analysis, which incidentally was carried out by the sister of our late
President, Dr Anna Weizmann. The analysis showed that the flowers contain six times
as much of the chemical as do the underground parts. Consequently, the bulbs should
be used not for the extraction of Colchicin, but for planting in beds, and yearly collec-
tion of the flowers. Harvesting operations are thus reduced to a minimum of labour, and
the plant is saved from extinction. Perhaps most important, the plant has become a sub-
ject for breeding experiments in order to obtain strains with improved yields.

The final result of converting this wild plant into a domesticated one will no doubt
still require much ecological and economic research, about the results of which I am op-
timistic. With these examples I hope to show how diversified are the problems of the
ecologist, and how often he must combine both ecological and economic considerations
in order to find the best way to utilize the treasures that nature hides in our deserts.

31

I am convinced that a thorough analysis of our desert flora would uncover a number
of new sources of raw material. Already some are being exploited, and others are gain-
ing prominence in the thoughts of those who are concerned with such items. It must be
stressed, however, that all such efforts must be preceded by the work of an ecologist.

Let me now give a few examples of potential sources of various raw materials :

Oleagenous plants — Citrullus Coloquintus, Cucumis propbetarum, Cucurbita spp.
from the desert parts of the U.S. and Mexico, etc.

Cellulose and Fibre plants — Agave, Retama roetam, Tbymelea birsuta, Juncus
arahicus and the Haifa grass of North Africa, Stipa tenacissima.

Rubber plants — Astragalus species in south-west Asia, Guyaule from the semi-
deserts of northern Mexico, Acacias from north Africa, and many others.

Significant progress in the search for raw material plants from arid regions has
been made in Australia, and probably also in Russia. In this field, also, international
co-operation will prove itself to be fruitful. I need only point out that exchange of spe-
cies and varieties between arid countries and particularly between the southern and
northern hemisphere alone promises to have tremendous influence on the regeneration of
these lands.

So far I have deliberately not mentioned plants which grow only in oases. These
form a separate topic, and have already been discussed much more than actual desert
plants, both from the scientific and the economic point of view. Nevertheless, they
still present many ecological problems. Some of these are involved in the planning of
oasis economy; for example adjustment to the high salt concentration of the water, the
amount of available water, the fluctuation of the water table, the vertical zonation of
various plant associations with respect to the water table, and the economical possi-
bilities for agriculture. The details of the zonation of Ein Ghadian, an oasis in Wadi
Arabia are being measured and mapped by my department.

It is interesting that, though Juncus arabicus appears only in a rather small area in
dense stands, my assistants Rawitz and Tadmor have found specimens growing at a
height of 2.60m. above the water table in June! I have initiated the careful collection
of seeds and rhizomes from these individuals, since they exhibit an ecological ampli-
tude far above the usual, implying a much extended area for the possible cultivation of
this plant.

With the question of zonation above the water table, we enter another aspect of
plant ecology where the fields of ecology, hydrology and climatology meet. In conclud-
ing I would like to make a few remarks on these questions, since the possibilities for
their solution are as yet not widely enough known. In accordance with the introductory
nature of this lecture I cannot go into the details. Furthermore, these new methods have
been discussed at several international meetings during the last few years, and I had
the opportunity to refer to them only a few months ago at the UNESCO symposium in
Turkey. Here a group of new ecological methods, which enable us to measure quanti-
tatively certain geo- physical values by biological means is involved. Because of the
extraordinary importance of these methods, I should like to provide a summary of them.

32

We can differentiate between four different principal methods, all based on three
fundamental natural laws:

1. Liebig's Law of the Minimum

2. The Geo- ecological Law of Distribution (/. Ecol. 35)

3. The Biological Rules of Climatic Extremes (Pal. /. Bot. Rebovot Ser. 7)

Three of the methods are based on the surprising regularity of shifts in amplitude
in respect to shifts in climatic factors. These shifts in amplitude have long been
known, but only recently have they been analysed by statistical and mathematical me-
thods. Here nature shows us, and I must stress this again and again, that the vegeta-
tion of a region is a much more sensitive indicator of its climate than a collection of
meteorological data describing isolated single factors. The four methods provide the
key to the code we are attempting to read. Let us take the example of the Laurel tree.
In the graph we can see a geographical shift in its amplitude with respect to the IE -
factor, that is in relation to insulation and exposure. In areas with an annual precipi-
tation of 600- 700 mm., Laurus nobilis occurs only on very steep slopes with a small
insolation; that is, only on steep north, north-east, and north-west slopes. Between
the 700 and 800mm. isohyetals it occurs on much less steep north slopes, and even on
steep west and east slopes. Between 800 and 900mm. the Laurel occurs on south slopes
with a 5° slope, and at over 900 mm. of rainfall, the IE -factor ceases to be a factor in-
fluencing the geographical distribution of the species. This indicates that the plant is
already at the climatic optimum of its geographical distribution. This is a clear example
of the Geo -ecological Law of Distribution. This law, in abbreviated form, states, that
micro - distribution (that is the topographical distribution of a species or ecotype) is a
parallel function of macro -distribution or geographical distribution, since both are det-
.rmined by the same ecological amplitudes.

Next comes the method of geographical shifts in amplitude in relation to the depth
of the groundwater table, and finally the method of topographical shifts in amplitude in
relation to the IE -factor. The depth to the water table can be determined in a case
when it is not too far removed from the ground surface: also average precipitation.
Since records from rain gauges are almost always inadequate in arid regions the possi-
bilities of exact determination of isohyetals offered by these four methods are of special
significance.

My last mentioned example provides corroborative evidence, since I found out only
two years after the completion of ecological tests, that I had by chance conducted my
experiments in the vicinity of a rain gauge with a record of more than 20 years. After
considering the coefficients necessary to correct for sandy soil and elevation, the me-
thod of overlapping amplitudes indicated that average precipitation during the past 30-
40 years had been 130- 145mm. per year. Two years after the publication of these re-
sults, the record of a border station 2km. from the location of my test appeared, giving
a mean annual rainfall of 136.1mm. Since then I have had several other confirmations
of the accuracy of this method.

We are coming to recognize more and more that the vegetation of each region indi-
cates its climate with much greater accuracy and sensitivity than meteorological data.
The three fundamental laws and the four applied methods teach us to decipher the code in

33

which the book of nature is written. Since the prehistorical times man has felt that vege
tation is the most sensitive indicator of climate, but only now are we beginning to suc-
ceed in the decoding.

In less poetical form, we can say that these three fundamental laws and the four
biological methods derived from them supply us with quantitative solutions of geophy-
sical problems by plantecological means. These methods are well on their way to play-
ing an important role in the work of making the desert areas of the earth productive. For
in these areas we are very near to the limit of plant survival in general, and the plants
react therefore in a most sensitive way to the minutest changes in their environment.
Because of this it is much easier here than in the humid regions to exploit this sensi-
tivity for practical purposes. Nevertheless, in this field as in so many others, there
must be, of necessity, close international co-operation in order to reduce errors to a
minimum, and to apply practical results on a global scale.

One thing is certain, that symposia and discussions such as we are having here at
the Institute of Biology in co-operation with UNESCO, are the best way to reach this
objective.

34

MODES 'CONTRACTE' ET 'DIFFUS' DE LA VEGe'tATION SAHARIENNE

Professor Th. Monod
(Paris)

L'un des premiers naturalistes qui se soit aventure dans le Sahara central, le
regrette Conrad Kilian opposait des 1925 la 'flore des pays cretacico- tertiaires
sud constantinois' ou 'flore du Sahara arabe' a celle 'du massif central saharien'
ou 'flore du pays targui*.

Dans le Sahara arabe: 'presence de bastes etendues de paturages quasi per-
manents de Salsolacees ..., les eaux ne sont pas total ement centralisees dans les
lits d'oueds ..., conservation d'une certaine humidite diffuse partant, les oueds,
generalement larges et mal delimites quand il en existe, n'etant que legerement
plus humides (en surface) ,.., flore ... largement repandue, diffus ...'.

Dans le Sahara targui: 'vegetation persistante reduite ... en general a peu
pres au fond des oueds au dehors desquels on trouve le desert ..., des lits d'oueds
souvent en permanence tres humides, avec vegetation peu desertique conservee et
en dehors le desert (a moins de pluie recente, car alors il y a de I'acheb) plus ab-
solu souvent que le desert arabe, plus depourvu encore de plantes persistantes ...,
flore persistante ... reduite, concentree aux lits d'oueds en un reseau favorise (et
peu desertique)'.

J'insistais moi-m^me quelques annees plus tard (1931) sur les caracteres dis-
tinctifs des deux 'modes', le diffus et le contracte: 'Dans les regions a flore con-
tractee,c'est a dire tout le Sahara central a I'exception des parties hautes de I'A-
haggar, la vie vegetale est exclusivement et rigoureusement cantonnee dans les
oueds qui le caracterisent et lui imposent un trace extremement strict. La vege-
tation occuppe la les lignes d'un reseau aux mailles demesurees et parfaitement
steriles, entournees du grele ruban des oueds ... Si de ce Sahara central a vege-
tation contractee on va suffisamment loin vers le Sud a la rencontre des pluies
saisonnieres de I'hivernage, si I'on marche assez vers le Nord pour atteindre les
pays cretacico- tertiaires, si I'on pousse assez vers 1 'Quest pour toucher au rivage
atlantique, ou si I'on s'eleve assez haut sur les pentes de I'Ahaggar, on verra
alors, peu a peu, le vegetation s'evader de la prison des oueds et s'etaler sur des
surfaces de plus en plus vastes: de contractee elle est devenue diffuse'.

J'ajoutais: 'L 'influence des precipitations (pluie ou rosee) sur le caractere
diffus ou contracte de la flore saharienne me semble evidente puisque, pour un sub-
stratum identique, on voit la flore devenir, de contractee, diffuse avec I'altitude ou
la proximite des influences maritimes'.

Revenant peu apres (1932) sur la meme question, je precisais: 'Bien que, dans
le Sud Algerien, la limite entre le Sahara septentrional et le Sahara central coincide
indeniablement, grossissimo modo, avec un contour geologique, separant un pays
calcaire, cretacico- terti aire, d'un pays silico- cristallin, il ne semble nullement
que le passage de la flore septentrionale diffuse a la flore centrale contractee se
trouve conditionnee par la composition du sol. Le Tadmait, geologiquement tres

35

uniforme, appartient a la fois aux deux modes (par sa partie nord et sa partie sud),
les calcaires de la plage pre - tassilienne ont une vegetation rigoureusement con-
tractee, les cipolins du Tanezrouft meridional egalement, tandis que le mode diffus
reparait sur les sommets de I'Ahaggar, dans le Sahara atl antique, et dans le Sahara
sahelien, independamment de la composition petrographique des substrata'.

Rappelons enfin les observations d'un botaniste de profession B. Maire (1940):
'On sait que son climat (Sahara septentrional) presente encore une regularite rela-
tive, et qu'il re<^oit, bon an, mal an, quelques pluies, surtout hivernales, qui, bien
que souvent peu importantes, suffisent a I'entretien d'une vegetation permanente
sur tous les terrains (exception faite des sables mobiles et des substrata toxiques).
Cette vegetation permanente forme une steppe ordinairement tres lache, qui recouvre
a peu pres tout le pays, constituant une vegetation diffuse, qui a frappe les explora-
teurs du Sahara par son contraste avec la vegetation contractee qu'ils ont trouvee
dans le Sahara central ... Entre El-Golea et Fort-Miribel les conditions climati-
ques changent; les pluies deviennent de plus en plus tares et irregulieres, ce qui a
pour corollaire une modification progressive de la vegetation, elles ne sont plus
suffisantes pour entretenir la vie de plantes perennantes sur tous les terrains ... la
vegetation permanente tand a se localiser dans les depressions ..., I'acheb, d'autre
part, ne se developpe plus bien, en dehors des points a vegetation permanente, que
dans des stations rocheuses ou sableuses bien drainees, et permattant aux graines
d'echapper au balayage par les vents. Cette localisation de la vegetation est
caracteristique du Sahara central, et elle ne disparait qu'en altitudes elevees'.

'Le Tadmayt est une zone de transition; sa partie septentrionale, au Nord de
Fort-Miribel appartient encore, en partie tout au moins, au Sahara septentrional;
sa partie meridionale appartient incontestablement deja au Sahara central, bien
qu'on y trouve encore un certain nombre de plantes caracteristiques du Sahara sep-
tentrional'.

Notons, au passage, cette bipartition du Tadmait, surface homogene repartie
entre les deux modes pour des raisons apparemment climatiques. B. Zolotarevsky
et M. Murat, en 1938, formulaient quelques remarques d'un vif interet, soulignant le
fait que I'opposition modes diffus/contracte pourrait n'etre pas due a I'influence
exclusive du climat.

'On ne saurait nier la part importante qui revient aux brouillards et aux rosees
dans la repartition relativement reguliere de I'humidite que Ton observe au Sahara
septentrional et occidental cependant, la rarete des affleurements cristallins dans
le premier domaine semble aussi fortement responsable de la predominance du mode
diffus de la vegetation et surtout de la richesse moindre, comme le remarque Th.
Monod, de ses stations privilegiees.

'Dans le Sahara meridional, au contraire, les peneplaines et les massifs cris-
tallins, grace a leur impermeabilite, canalisent et localisent I'eau de pluie; ce
sont ces facteurs topographiques et edaphiques qui y determinent en premier lieu le
mode contracte de vegetation.

'Dans la partie sud du Sahara occidental on voit se superposer la vegetation
qu'on appellerait contractee, si elle existait seule, comme au Sahara meridional, et

36

la steppe a Salsolacees typique du Sahara septentrional. La premiere est favorisee
par la structure geologique et la topographie du pays, la seconde par les brouillards
venant de I'ocean ... La strate arborescente du Sahara est toujours contractee, la
strate suffrutescente est souvent diffuse et la strate des therophytes Test presque
toujours. Le mode contracte de la strate arborescente et en partie de la strate suf-
frutescente est provoque par les conditions topographiques et edaphiques, le mode
diffus, quand il se rencontre, est determine principalement par les conditions cli-
matiques'.

II ne sera peut-8tre pas inutile de revenir sur ces diverses conclusions, d'au-
tant plus interessantes qu'elles emanent de biologistes possedant une experience
personnelle etendue des regions en cause.

(1). Le mode diffus est determine 'principalement' par le climat, mais la nature
du sol peut en etre aussi 'fortement responsable' , les terrains cretacico - tertiaires
du Sahara septentrional favorisant la diffusion, le Precambrien (peneplaine et mas-
sifs) determinant 'en premier lieu' la contraction, a la fois semble - 1- il par leur
nature (edaphisme) et leur morphologie (topographie). — (Je n'ai pas 1 'impression
que la nature geologique du substratum soit en cause puisque I'on peut voir (a) le
mode diffus sur du Precambrien (bordure sahelienne, Sahara atlantique) ou sur des
gres {ibidem), done sur un substratum identique a celui des tassilis a mode con-
tracte du Sahara central, (b) le mode contracte sur les terrains les plus varies et,
frequemment, calcaires (Hamadas Safia et HI Haricha de Taoudeni, calcaires dolo-
mitiques de I'Adrar de Mauritanie, etc.).

Ce qu'il faut, par contre, reconnattre, c'est que, a I'echelle regionale, done
pour des conditions climatiques identiques, la tendance a la decontraction crott
avec le degre d'ensablement: dans I'Adrar de Mauritanie, si les versants abrupts,
talus d'eboulis, etc, sont nus ou presque, les plateaux greseux, horizontaux ou
moderement inclines, des qu'ils sont suffisamment ensables peuvent supporter une
vegetation diffuse (therophytes, hemicryptophytes, chamephytes).

(2) Le Sud du Sahara occidental presente une intrication, et comme une super-
position des deux modes, lies I'un au climat I' autre a la physiographie (edaphisme
+ topographie). — Le fait est, bien entendu, parfaitement exact et j'ai moi-meme
(1938) signale I'existence d'un mode diffus (interessant non seulement des hemi-
cryptophytes mais des phanerophytes) sur le plateau greseux de Chinguetti, done
sur les m^mes gres, exactement, qui supportent au Sahara central une vegetation
contractee.

Nous nous trouvons ici, a mon avis, sensiblement a la limite, fort imprecise
evidamment, de trois territoires bioelimatiques, comme d'ailleurs de trois domaines
floristiques, I'Adrar appartenant a la fois au Sahara mediterraneen et au Sahara afri-
cain*, avec des irradiations saheliennes, et il ne serait pas, a mon avis, surpre-
nant que la juxtaposition des deux modes ne puisse relever, ici encore, des seuls
facteurs climatiques.

*cf. Monod, Th. 1952. Contribution a I'etude du peuplement de la Mauritanie — Notes bota-
niques sur I'Adrar de Mauritanie, (Sahara occidental) Bull. IF AN. 14. (sous presse).

37

PLUIE TOTALE

TOTAL ANNUEL MOYEN EN MILLIMETRES

Fchelle hypsoTnctrit^ue

i5°^5<^^>^'0'

Projection conique conforme Echelle 1/20.000.000 aux latitudes 30^ et 60^

Figure 1.

Pluie totale au Sahara (moyenne annuelle en millimetres) d'apres J. Dubief et J. Lauriol, 1943,
Trav. Inst. Meteor. Phys. Globe Algerie, fasc. 4, C. 49. La limite entre le mode diffus et le
mode contracte passerait approximativement vers les isohyetes de 30 -50 mm (souvent entre 50

et 100 sur la bordure sahelienne).

38

(3) La strate absorescente est toujours contractee, la sufjrutescente souvent
diffuse, celle des therophytes presque toujours. — On doit savoir gre a MM. Zolo-
tarevsky et Murat d'avoir enrichi la notion des modes d'une distinction des types
biologiques interesses. II semble bien, en effet, que les modes concernent avant
tout la vegetation 'permanente', a 1 'exclusion des nappes de therophytes pouvant
se superposer aux vegetations diffuses ou se juxtaposer aux contractes. Le mode
diffus typique est — sauf au Sahel bien entendu — principalement constitue de
chamepjytes (et d'hemicryptophytes avec la 'steppe' extra- saharienne a Haifa)
alors que les phanerophytes caracterisent le mode contractee, en le colorant d'une
tonalite resolument africaine.

Ajoutons toutefois que non seulement au bord sud, avec le Sahel, mais meme
en situation saharienne sur certains plateaux ou certaines plaines argileuses ou
sablonneuses du Sahara occidental la 'decontraction' climatique ou edaphique, peut
interesser les arbres.

(4) Le mode contracte releve de facteurs non climatiques. — Je n'en suis pas
convaincu et expliquerai plus bas pourquoi.

Quand les modes sont typiques, ils sont nettement distincts.

(1) Mode diffus: pseudo-steppe ou savane desertique tres laches mais recouv-
rant a peu pres uniformement tout le pays, sans contraste tres brutal entre les oueds
et le teste du pays.

(2) Mode contracte: pseudo-steppe et savane desertique, limitees aux thal-
wegs, lignes de verdure incrustees dans un paysage denude.

Le mode diffus comprendrait les types suivants:

I Type marginal

(a) Nord: ex.: pseudo- steppe a Salsolacees, etc, de Ghardaifa— El Golea.

(b) Sud: ex.: fructicee a Calligonum commosum. Euphorbia balsamifera, etc.,
de I'Amonkrouz, Savane arbustive a Leptadenia Spartium, Panicum turgi-
dum, Aristida spp., Cenchrus biflorus, etc., de rAzaouad. La question se
pose de savoir si, ces formations etant en fait deja saheliennes, il existe
un type diffus marginal sud vraiment saharien.

(c) Atlantique: ex.: pseudo-steppe a Nucularia perrini du Tiris.

II Type altitudinal

Etages mediterraneens du Hoggar et du Tibesti.

Ill Type d'epandage

Il arrive que Ton observe, au debouche d'un oued en plaine, au sortir de la
montagne, ou parfois mSme fort loin de tout relief important en un point ou un
oued, a bout de course, etale largement et ses alluvions et son hunidite, des
zones d'epandage parfois tres vastes (maader, grara), sablonneuses ou sablo-
argileuses ou la vegetation, contractee plus en amont, deviant typiquement
diffuse.

Il arrive aussi, au moiiiS sur la peripherie, que la diffusion apparaisse liee au
sable encore mais alors que dans la cas du maader il s'agissait d'une nappe

39

sablonneuse enrichie en eau par un sous ecoulement, ici le revetement are-
nace, recouvrant par exemple une surface rocheuse, n'a recu que I'apport di-
rect des precipitations, suffisants pour nourrir, au moins temporairement (il
s'agit de therophytes ou d'hemicryptophytes plus au moins 'cycliques' et quasi
'reviviscents') une vegetation diffuse.

On pourrait done distinguer dans le type d'epandage deux sous- types:

(a) de maader (ou grara), a alimentation souterraine, indirecte; sables epais.

(b) de plateau ou de reg, a alimentation pluviale directe; sables en revete-
ment mince.

Dans le mode contracte, je serais dispose a distinguer:

I Type plcmitiaire

Le mode contracte n'est pas I'apanage exclusif des massifs aux oueds encais-
ses; il se rencontre, sous une forme moins frappant sans doute, mais
non moins typique, en plaine, et quelle que soit la nature geologique du sub-
stratum, sur une enorme surface s'etendant en latitude du Tadma'it (Sud) au
Sahel (montagnes exceptees: mode contracte encaisse + mode diffus) et vers
rOuest jusqu'a I'Adrar*. Il n'est pas de reg, de surface rocheuse sedimen-
taire ou de peneplaine cristalline qui ne developpe, si peniblement marque
soit- il, un chevelu hydrographique, mais la plus legere denivellation suffit,
sous le climat adequat, a emprisonner la plante sur les lignes memes du ruis-
sellement.

Ch. Sauvage avait note, au Sahara occidental (1949, p. 45- 47) que la pseudo-
steppe a Nucularia, Traganum, Salsola, etc., ne se trouvait pas sur des regs,
mais sur des zones d'epandages — parfois remarquablement plates et larges —
de certains oueds, ou I'on assiste evidamment a une decontraction de la vege-
tation.

II Type encaisse

C'est le cas exemplaire, et classique, I'oued entaille, souvent en canyon, et
jalonne d'un ruban de verdure relativement luxuriant. Intercale entre la diffu-
sion altitudinale et la contractic a planitiaire ce type peut etre separe de cette
derniere, dont ne le distingue en fait qu'une question de degre dans le volume
de la vegetation, par des zones d'epandage de mode diffus. Le passage direct
du type encaisse au type planitiaire ne s'observe que la ou le ravin, de peu
d'importance, ne provoque pas, a son entree en plaine, d'anevrisme de type
maader.

Ill Type de cuvette

Un etalement de I'humidite en milieu contracte declanchait, localement, une
diffusion. On ne s'etonnera pas qu'un enrichissement local de I'humidite en
milieu diffus ne provoque I'apparition de taches de contraction. Il semble
bien, en effet, que I'on puisse a juste titre regarder comme relevant encore,

♦La ou le substratum admet un ruisseliement organise, done a I'exclusion des immenses
surfaces dunaires, dont nous ignorons d'ailleurs encore le type de vegetation, il meme des
plateaux ensables a mode diffus.

40

fut-ce sous un aspect un peu aberrant, du mode contracte des vegetations de
cuvettes comme celles des dayars a Pistacia-Ziziphus du Sahara algerien, ou
comme celles des ^aras du Sahara espagnol septentrional.

Le tableau schematique des subdivisions proposees s'etablirait ainsi:

ORIGINE

Mode

climatique

physiographique

diffus

I. marginal IDl

(a) Nord : DIa

(b) Sud : Dlb

(c) atlantique : Die

II. altitudinal : D2

III. d'epandage : D3

(a) de mSader (ou grara)

(b) de plateau (ou reg)

contracte

I. planitiaire : CI
II. encaisse : C2

III. de cuvette : C3

Comme on le volt, ne relevent pour moi de la physiographie, en constituant des
types localises que les types D3 et C3.

Celui- ci concerne une contraction par concentration locale de I'humidite,
celui-la une decontraction par etalement de I'humidite, par exemple au debouche
en plaine d'un bassin versant. Expliquer la contraction de la vegetation tassi-
lienne, par exemple, par la topographie, ce serait admettre que le climat local,
dans I'hypothese (plaine, plateau non dra^ne) ou I'eau ne serait pas canalisee par
le reseau des oueds, permettrait I'etablissement du mode diffus et que c'est la
soustraction a I'ensemble de la surface, par I'ecoulement encaisse lineaire, d'une
part appreciable d'humidite qui denude les interfluves et peuple les thalwegs. Or
(1) la ou le plateau existe plus ou moins horizontal et (2) dans les plaines adja-
centes au relief (il ne s'agit pas de montagne vraie, bien entendu, celle- ci admet
climatiquement, le mode diffus) et apparemment de climat comparable, on ne voit
rien autre, qu'un mode contracte, tres appauvri sans doute, mais typique et dont la
vegetation parfois relativement exuberante ('sahariennement' pari ant!) des thalwegs
ne represente qu'un cas particulier. Entre le veritable boisement qui occuppe le
lit du canyon et le miserable 'ouedaillon' qui serpente sur le reg a peine souligne
par quelques touffes espacees de Graminees, il n'y a, a mon a vis, qu'une diffe-
rence de degre dans les resultats botaniques de la concentration de I'humidite,
mais aucune de nature: la morphologie peut concentrer plus ou moins d'eau dans
les oueds, et provoquer des variantes locales du mode contracte: I'ensemble parait
relever, quand meme, d'un type de climat impliquant la contraction.

II est a peine utile de rappeler que, si I'opposition entre la pseudo-steppe dif-
fuse du Nord des Territoires du Sud Algerien et la savane contractee des canyons
tassiliens est tres marquee, il existe entre ces aspects extremes route une serie
menagee de types intermediaires et puis des cas ou le diagnostic sera difficile.
Comment definir, par exemple, la vegetation de la Hamada de Tindouf, avec des

41

CO

o

>

a
-a

u

o

^^

o

CO

1

u

11

S)

4J

• -^

c

(0

a>

t)

w

u

<B

-o

J3
«1
C/3

3
«

v>
u
-a
o

E

M

<U

-o

c
o

c
o

u

• -4

"a,

U

^

•3 a

.•2 «J

(0

-a

u

a,

(0

a;
a,

ca

6

u

CO

i

u

V

0

>

^-^

•x

A

00

««

0

•

E

(0

i>

<u

c

M

a

4->

C3

10

1>

d

-a

■0

(0

w

c
0

V

a,

3

0

60

■^

II4

CA)

•— (

3

-0

0

C

a

u

D

c
u

(0

u
a,
3
o
U

42

Acacia tortilis disperses mais lies, localement, a la presence de petites cuvettes
tres plates. Si, comme, il le semble, cette disposition est comparable a celle des
dayas a Pistacia du Sud Algerien, on pourrait peut- etre classer dans C3 les boise-
ments de la Hamada de Tindouf, le reste du peuplement {Anabasis, etc) etant, la
ou il existe, diffus.

II n'est guere possible de definir de facon tant soit peu precise, et sur des
criteres directement applicables sur le terrain, les deux modes la ou ils peuvent se
trouver au contact et atypiques, par exemple dans des secteurs de transition. D'au-
tre part une 'bonne annee', pluvieuse peut jeter sur le pays le plus aride le mant-
teau polychrome d'une explosive floraison. La vegetation de I'Adrar mauritanien
dans I'hiver 1951- 1952, a la suite de fines pluies, etait dans une large mesure dif-
fuse. Mais, en msme temps, temporaire. Or I'essentiel de la distinction 'contrac-
tion' 'diffuse' doit porter avant tout, bien entendu, sur la vegetation permanente
(phanerophytes, chamephytes, hemicryptophytes) ou semi -permanente.

A propos des ecoulements en nappe en pays subarides, Cailleux distingue, de
la montagne a la plaine, 4 sections*.

(a) En nappes minces sur les versants.

(b) Concentration lineaire dans les thalwegs.

(c) En nappe au debouche des deltas dans la plaine.

(d) Legere tendance a la concentration, regroupement en ebauches de petits
torrents lineaires elementaires.

On ne saurait ne point etre frappe par le parallelisme de ces divisions avec les
les notres, les sections (a), (b), (c), (d), de Cailleux paraissant correspondre a nos
modes D2, C2, D3<2, et CI.

Sans doute ne devra- t-on point trop pousser la comparaison puisque

(1) les sections de Cailleux se succedent sur un profil d'extension limitee, et
a I'interieur d'un climat unique, a pluviosite (quantite et type) caracteristique dont
elles traduisent le mode d'action au sol.

(2) mon type D2 releve de causes climatiques plus varices que I'etalement de
lapluie en nappe sur 'les versants' et d'ailleurs, interessant de vastes regions ta-
bulaires ('tarsos' du Tibesti, etc.), elles- m^mes coupees de thalwegs, comprend a
la fois des versants, des surfaces et des oueds.

(3) L'ecoulement en nappes sur versants rocheux me parait devoir etre plus
frequent encore sur les pentes nues du domaine C2 que sur celles de D2, moins
glabres.

Il n'en est pas moins interessant de reconna^tre pour deux series differentes
de manifestations, type d'ecoulement, modes de vegetation, cette sorte de pulsa-
tion rythmee qui, loin d'etre 1 'apanage des choses de vie, a sans doute de bien
plus vastes implications.

En tous le cas, ici: a (nappe) - b (lineaire) - c (nappe) — d (lineaire) (Call eux)
et melodic plus etendue encore si on ne la limite pas aux reliefs et a leur piemont:
Dla - (C3) - Dlfl - CI - D3 - C2 - D2 - C2 - D3 - CI - Dl^.

Cailleux, A. 1950, Rev. Geomorphol. Dynamique, I, (6), 257, 261.

43

References

Kilian, C. 1925. Au Hoggar. Mission de 1922, Paris, pp. 190, 3 cartes, XVI pis.

Maire, R. Etudes sux la flore et la vegetation du Sahara central. (Mem. Soc. Hist. Nat Afr.
Nord., Mission du Hoggar, 2, Alger, 1933-40, 433p., 22 figs, 2 pis col., 22 pis, 2 cartes).

Monod, Th. 1931. Remarques biologiques sur le Sahara. Rev. Gen. Sc, XLD, no. 21, 609-
616.

Monod, Th. 1932. Mission saharienne Augieras- Draper, 1927-1928. Phanerogames Bull.
Mus., (2), N, no. 6, 756-774, 1 fig.

Sauvage, C. 1949. Le point de vue du botaniste in Monteil, V. et Sauvage, C. Contribution
a I'etude de la flore du Sahara occidental, 1, Inst. Hautes Et. Maroc, Notes & Doc, 5.

Zolotarevsky, B. & Murat, M. 1938. Divisions naturelles du Sahara et sa limite meridionale
in La vie desertique dans la region nord- tropicale de I'Ancien Monde, Soc. Biogeogr.
Mem. no. VI, 335- 350, 1 carte.

44

THE BAHRAIN ISLANDS AND THEIR DESERT FLORA

Professor R. D'O Good
(Hull)

The Bahrain Islands lie some 20 miles from the coast of Saudi Arabia, al^out half
way down the southern shore of the Persian Gulf, in the bight between Hasa and the
Qatar Peninsula, at approximately 26°N and 51°E. The group consists of Bahrain Is-
land itself, which has a length of about 30 miles and a maximum width of about 10 miles;
three much smaller islands (Muharraq, Sitra and Nabbi Salih), so close to Bahrain on the
north-east as to be virtually part of it; three other small islands (Umm Nasan, Jedda
and Raka) more detached in the north- west; and a few tiny islets. All but the last of
these are shown in figure L

THE
BAHRAIN
/ELANDS

JO 0

"Q

niLBS

Figure 1.
Sketch map of the Bahrain Islands showing Bahrain itself and its six satellite islands,
Muharraq (M), Nabbi Salih (NS), Sitra (S), Umm Nasan (UN), Jedda (J), and Raka (R), and its
three towns Manama (Ma), Muharraq (Mu), and Awali (A).

45

Politically the archipelago is an independent Arab principality ruled by a dynasty
established towards the end of the eighteenth century. The principality is in special
treaty relations with the British Government, which maintains a Political Agent in Bah-
rain and has lately transferred there the headquarters of its Political Resident in the
Persian Gulf. Bahrain is the centre of a pearl - fishing industry of great age; it has an
important oil- field; and it is becoming an increasingly significant centre of air and
other transport. It is also renowned for the great numbers of sepulchral mounds or tumuli
which cover considerable parts of its surface and which are generally thought to be bet-
ween three and four thousand years old. The population is mainly concentrated in the
two coast towns of Manama and Muharraq and at a recent cencus amounted to about
120,000.

Climate

The Bahrain Islands have a remarkable and somewhat notorious climate in which
the chief characteristics are high summer temperatures; scanty and irregular rainfall;
high relative humidity ; and rather persistent, though rarely very violent, wind. There
are recording stations at Muharraq Airport and at the Bahrain Petroleum Company
(BAPCO) desert town of Awali, and there is probably sufficient information available
for a detailed study of conditions, but for the present purpose they may be illustrated
by the following sample figures, based, unless otherwise stated, on the year 1947.

The mean annual temperature is about 80°F, or perhaps a little more, and the ex-
treme temperature variation during the year is about 70°F. Monthly figures are:-

Average

J-

F.

M.

A.

M.

J.

J.

A.

S.

0.

N.

D.

aver.

daily mean

65

65

73

82

90

92

90

90

85

82

82

74

80

absolute max.

80

78

91

108

116

117

114

111

103

100

96

93

101

absolute min.

48

47

55

64

72

76

76

74

68

67

64

47

63

The average annual rainfall for ten recent years is 2.46 inches but the total var-
ies greatly from year to year as shown by the 1946 figure of .15 inch and that of 1940
which was 5.53 inches, or nearly forty times as much. Rain falls on an average of about
20 days a year and there is practically none between late April and November, the mon-
thly figures for 1947 being :-

J. F. M. A. M. J. J. A. S. O. N. D.
.3 .6 .2 - .1 _ _ _ _ _ „7 .1

The relative humidity ranges during the year from 0 to 100% and in 1947 the aver-
age daily maximum was 83% and the average daily minimum was 25%. The lowest daily
maxima were 38% on 4 December and 54% on 3 June; the highest daily minima were 74%
on 19 March and 70% on 8 January and 18 February. 100% was reached on 21 occasions
in February, October and November (compared with 100 occasions, nearly all in the lat-
ter half of the year, in 1949), and the daily range varied from 4% on 19 March to 85% on
18 September.

There is a fairly constant light to moderate wind apparently throughout the year
which comes prevailingly from the northern quarters and especially from the north-west.
This wind is commonly called a shamal locally and tempers the climate, while the rarer
southerly winds tend to be oppressive. During the first seven months of 1947 there was,

46

at some time every day, a wind of at least 10 miles an hour, and on two or three occa-
sions the daily minimum was 20 miles per hour, but some calm periods were recorded on
most days. The absolute maximum rate during this period was just over 50 miles per
hour.

Bahrain Island
1. Structure and physiography.

Geologically Bahrain Island is a simple shallow elongated anticlinal dome of Eo-
cene rocks, which dip down- flank from 1 to 5 degrees, and which are covered peripher-
ally and unconformably by more recent deposits. On the north this peripheral extension
of the island is considerable and mainly of rocks of Miocene age with a maximum thick-
ness of about 150 feet but these are partially or entirely covered with even younger
superficial deposits. The Miocene — Eocene boundary runs through the north part of
Sitra, across Nabbi Salih to Adari and Barbar. In this part also the island appears to
be growing by the gradual elevation of fringing coral reefs. In the south the peripheral
extension is even greater and again consists largely of Miocene rocks, with a thickness
of some 90 feet, but these are completely covered with younger deposits. Here there
are no coral reefs and the island tapers abruptly at its south end to a sharp point of
small sand-dunes. On the west the peripheral belt, though similar, is much narrower
and there are no reefs. Here again there is evidence of recent uplift in the presence of
a raised beach. On the east the peripheral belt is lacking for a distance of some miles.
The highest existing point of the Eocene dome, the summit of the massif known as the
Jebel Dukhan, is about 450 feet above sea level.

Except for the peripheral deposits there are no major faults or unconformities in
the island and the simplicity of its anticlinal structure is complicated in only one im-
portant respect. This complication is that the whole central part of the island, com-
prising an area about 12 miles by 4, is a great shallow saucer with a slightly convex
floor, from which rises the Jebel Dukhan, and surrounded by a scarp cliff, called by the
petroleum geologists the Rim Rock, which averages about 50 feet in height. The Jebel
Dukhan is slightly west and north of the exact middle of the saucer and the floor is
rather lower in the south and parts of the west (where it is not more than 50 feet above
the level of the sea) than in the east and north (where it is about 100 feet above the
sea). Figure 2 shows diagrammatic sections of the island along the two main axes with
the vertical scale very greatly exaggerated.

In the northern part of the island there are in two places, near Buri and A\ Hisi,
lengths of other scarp cliffs very like those of the Rim Rock, and these are apparently
all that now remains of an outer scarp cliff.

The main strata of the Eocene rocks on the island are seven, namely, from above
downwards :-

1. White limestone - 0-150 feet thick

2. Orange Marl 30 - 50 feet thick

3. Brown crystalline limestone — Nummulitic limestone 100 — 150 feet thick

47

4. (a) Alveolina zone 30 — 50 feet thick
(b) Shark's tooth shales 6 — 8 feet thick

5. Chalky zone 110 — 220 feet thick

6. Central brown limestone nowhere completely exposed

The relation of these strata to one another and to the inner and outer scarps is
shown diagrammatically in figure 3.

Two explanations have been advanced to account for this remarkable structure of
saucer and scarps. Pilgrim, who made the first sketch of the geology of the island (see
Mem. Geol. Survey India, XXXIV, pt. iv, 1908) thought these features were formed soon
after the island was first raised, by the ordinary processes of sub -aerial denudation
operating on the various strata described above at a time when rainfall was much greatei
than it is now. Later the island first sank and then emerged again so that the saucer

Figure 2.
Diagrammatic sections across Bahrain Island from north to south and from west to east passing
through the Jebel Dukhan. The crosses indicate the Rim Rock or scarp of the central saucer.

Vertical scale greatly exaggerated.

Figure 3.
Ideal diagrammatic section, not to scale, across Bahrain Island along the line Buri, Jebel
Dukhan, Al Hisi to illustrate the geological structure of the island. The peripheral post-
Eocene deposits are stippled; the numbers are those of the six main strata, i.e. the white
limestone, the orange marl, the brown crystalline limestone, the Alveolina and shark's tooth
beds, the chalky zone, and the central brown limestone. AA is the outer scarp of which only
traces remain; BB is the Rim Rock or scarp of the saucer; and C is the summit of the Jebel.

48

previously formed remained as a lake, and it was the draining away of this water, main-
ly by way of the Zallaq gap (which is the only pass in the scarp cliff),Pilgrim suggests,
that the final configuration was attained. This explanation is a little complex and is
clearly based to some extent on the necessity of accounting for the sub -recent marine
shells which occur in places on the floor of the saucer.

A more recent suggestion is that the saucer has been formed, at least in part, by
foundering or 'slumping', that is to say by the shallow local vertical displacement of
strata as a result of the dissolving out of the salts in some of the lower beds and par-
ticularly of the anhydrite which occurs in considerable quantity in the chalky zone.

These theories need not be considered in detail here but it is appropriate to call
attention to two points.

Scattered over the surface of the saucer are various more or less isolated bluffs in
different stages of erosion, which on a small scale are analogous with the Jebel Dukhan
itself, and the outer of these at any rate are clearly detached portions of the Rim Rock,
and it is not easy to see why these should have been left projecting from what was, on
the first theory, a lake, or to account for them on the hypothesis of slumping. On the
northern flanks of the dome (outside the Rim Rock) there are two smaller depressions,
Umm Abdullah and Al Buhai, which seem clearly enough to be miniature replicas of the
main saucer, though without anything corresponding to the Jebel in the middle. At Al
Buhai the surrounding scarp cliff is continuous and there is no egress corresponding to
the Zallaq gap, and it is difficult to imagine how the material of this excavation can
have been removed by water erosion.

The essential difference between the two theories Is, of course, that according to
the first the filling of the saucer. has been entirely removed, while according to the
second it has merely been displaced in situ. A careful correlation of the beds of the
Rim Rock, of the saucer floor and of the Jebel, should therefore afford strong evidence
either for or against the hypothesis of slumping.

This Correlation has been made in detail by the geologists of the Bahrain Pdtroleum
Company and is demonstrated, by their courtesy, in figure 3. From this it will be seen
that the floor of the saucer is a true floor of denudation and therefore that the material
which must once have filled the saucer has been removed. It is also seen that the sum-
mit of the Jebel Dukhan consists of beds of stratum 3 (capped by resistant chert) and
that the summit of the anticline was therefore once higher by at least the thickness of
strata 1 and 2, which today have quite disappeared except here and there towards the
periphery of the dome.

On this evidence it seems certain enough that, although slumping may have occur-
red here and there on a quite local scale, it cannot be made to account for the saucer
as it is today, or for the formation of the outer scarp, and that a satisfactory explana-
tion for this remarkable physical structure is still to be sought.

Rather surprisingly subterranean water is plentiful in many parts of the island, be-
cause, both in the Eocene rocks as well as in the more recent deposits, porous and more
impervious beds tend to alternate, and these supplies can be tapped by shallow or ar-
tesian wells according to their depth. Most of the deeper water is said to be derived

49

from the rainfall of Central Arabia and even from regions further north-west, but some
of the shallower wells derive their supplies from local rainfall catchment. Unfortunatel]
all this water (except for one or two shallow wells) is highly brackish, that from the
north of the island commonly containing between two and three thousand parts per mil-
lion of dissolved salts, and that from the south as much as four thousand parts. It is
owing to this circumstance that despite the amount of water available for irrigation the
only crops which flourish on a considerable scale are dates and lucerne.

2. Natural Areas.

In the simplest terms Bahrain and its three most closely associated islands con-
sist of two natural regions only, those of the peripheral post- Eocene deposits, roughly
outlined by the 50 ft. contour line, and the central Eocene dome, but each of these is
further divisible.

With regard to the first various circumstances, of which the distribution and acces-
sibility of water is probably the most important, combine to make the northern part of
the area, and especially that north of the Miocene boundary, of much greater potentia-
lity as a human habitat than the rest of the island. In consequence nearly the whole
population is here and human exploitation of every sort diminishes very rapidly towards
the south. In latter years it is true that the building of the oil company's desert town
of Awali at the north end of the saucer has in some measure distorted this picture in
fact though not in theory, since its presence there has been made possible only by over-
coming the natural limitations of the site by purely artificial means.

The direction of the prevailing wind from the north-west adds a north-west to
south-east component to this southward gradient, partly by its effect on the distribu-
tion of rainfall, and partly by the accumulation of blown sand in its direction. As a re-
sult there is, in addition to a diminishing human gradient from north to south a dimini-
shing vegetational (fertility) gradient from north-west to south- east, an effect which is
particularly noticeable within the saucer.

The PERIPHERAL (post- Eocene) ZONE which, it will be recalled, is absent alon^
the central part of the east coast, can be divided into a northern cultivated area and a
western and southern almost uninhabited part. The boundary between the two in the
north-west of the island is not however clearly marked since there are areas of more or
less natural desert almost to the north coast, while there are scattered date gardens far
to the south. The cultivated area has several aspects ; the western and southern area
is more monotonous, its chief feature being an extensive shallow pan running north-
west from Mattala.

The CENTRAL (Eocene) DOME divides into several constituent parts in accordanc
with the physiography illustrated in figures 2 and 3, namely the FLANKS; the SAUCER
the CENTRAL PLATEAU and the JEBEL DUKHAN. The first of these is complex in
that it is locally double where the outer scarps occur and the white limestone is expose(
but on the main continuous flanks outside the saucer it is the brown crystalline lime-
stone that provides the surface, except for some local patches of orange marl. The sur-
face of the saucer is of beds of the chalky zone, modified to a varying degree by the
products of erosion of the central plateau and the Jebel. The central plateau is formed

50

of the central bcown limestone which, superficially at least, is not very different from
the brown crystalline limestone. Lastly the Jebel Dukhan consists of a base of chalky
zone beds, above which are the Alveolina — Shark's tooth series, and the brown crys-
talline limestone, this last capped by a bed of resistant chert.

Both the flanks and the saucer are conveniently subdivided again, the former into
four parts, north, south, east and west, and the latter into two, north-west and south-
east.

These various natural areas are shown in figure 4.

Figure 4.
Sketch map showing the natural areas of Bahrain Island, as follows:-

Peripheral Zone. Northern peripheral area (black). Western and southern area (^ite)
Central Dome. Flanks, divided into north, south, west and east (spotted). Saucer, divided
into north-west and south-east (diagonal lines). Central plateau (horizon-
tal lines). Jebel Dukhan (cross hatched).

51

3. The Desert Vegetation

The vegetation of Bahrain comprises three formations or major communities of spe-
cies of which two are local, one of these being largely adventive.

Most restricted in distribution is the halophytic vegetation of muddy shores, namely
salt marsh and mangrove swamp, which occurs especially on parts of the shores of the
deep inlet a mile or two south of Manama and more sporadically near Sitra.

More extensive but still local is the mainly adventive flora of the date and other
gardens in the cultivated northern part of the peripheral zone, where constant irrigation
is practised.

Everywhere else on the island the vegetation is one or other minor facies of a
single rich but highly selected desert plant community of the general North African —
Indian desert flora, — comprising something less than 200 species. Nowhere on the is-
land, with the two exceptions already noted, is climate, soil, altitude or any other in-
fluence sufficient to cause any real replacement or even important modification of this
general plant community.

The natural desert areas of Bahrain, that is to say the whole island outside the
cultivated part of the peripheral zone, displays well the three most prevalent kinds of
desert habitat, namely sandy or small -dune desert; stony and gravelly desert, or reg;
and rock exposures, or hamada, as well as one or two particular conditions.

A. Sandy desert, in which the substrate is so loose as to pile up to some extent at
least against the larger plants, is the characteristic condition of the peripheral zone
and occurs also here and there in the saucer, especially in the south-east, where wind-
borne material tends to accumulate. Its vegetation is often more considerable in bulk
than that of the other deserts because some of the individuals are larger (e.g. Lepta-
denia pyrotechnic a) but the species are fewer and of a more halophytic sort. In one di-
rection this type of desert passes towards plantless loose sand and in the other towards
consolidated salt flats such as are found north of Mattala and here and there elsewhere .

B. Stony and gravelly desert is also widespread and, to the eye at least, is the most
characteristic feature of the island, since it prevails in the more accessible parts, cov-
ering practically all the flanks of the central dome and parts of the central plateau. The
surface is of more or less consolidated sand or marl, thickly or even completely covered
with irregular and angular pieces of flint, chert or limestone, ranging from a quarter of
an inch to several inches in diameter, generally white or pale yellow but often black
and frequently modified in colour by grey -green or yellow lichens, which are a conspic-
uous feature of these deserts. This stony desert surface is presumably the result of
denudation in situ and it is here that the tumuli are so numerous, and from this material
that they are made. The vegetation is on the whole sparse, especially towards the tops
of the slopes, and there are often distances of many feet between neighbouring plants,
but it becomes much thicker wherever a slight cavity or consolidation (as, for instance,
wheel ruts) retains the intermittent rainfall a little longer than elsewhere. Such depres-
sions are especially numerous between the tumuli and here the vegetation is often al-
most closed. The surface of the tumuli resembles that of the plain and they tend to
bear a slightly depauperated mixture of the same species, with notably more individuals
on their northern sides.

52

C. Exposed limestone is found on the flanks where intermittent surface drainage has
scoured away some or all of the overlying sand and gravel, namely in the numerous small
wadis which are thickly distributed here, but occurs also in parts of the central plateau
where wind erosion has perhaps played a larger part. According to the steepness and
nature of the substrate occasional heavy rain may pass through these wadis almost in a
torrent, but elsewhere the flow may for various reasons be slower so that there may even
temporarily be pools of standing water. It is in these flatter wadis that the richest and
most luxuriant vegetation of the island is to be seen, in the form of thickets several feet
high composed of such shrubby plants as Zizyphus and Atriplex and various large herbs
and grasses. These wadis also show the largest collections of species, one of them
for example, south-east of the village of Ali, containing upwards of 90 species, or
about half the whole desert flora.

On the central feature of the island, the Jebel Dukhan, all three main types of des-
ert occur; sandy desert on parts of the north slopes, where wind has deposited material,
stony desert on the summits and flanks where the rock is not fully denuded and exposed,
and exposed limestone on the rest of the summits and on the flanks where slope and
other factors preclude any more superficial deposit. The Jebel presents two habitat
features peculiar to itself and its minor homologues, namely the presence of slopes
steeper than found elsewhere and locally even precipitous, and in consequence the oc-
currence also locally of shaded niches such as do not occur on the open desert. It is
interesting to note however that while various species attain an unusual size in such
favoured spots there are apparently no species peculiar to them. About 75 species
occur on the Jebel.

The parts of the saucer which do not bear typical sandy desert are for the most
part in something between this condition and stony desert but very notably on the west
side the white, marly beds of the chalky zone are bare, and here there are large patches
of vegetation of considerable local repute for horse and other grazing owing to the pre-
sence of 'sorrel' (apparently Emex spinosus) and certain other plants.

The Island of Umm Nasan

This has an area of about 9 square miles. It is used as a hunting preserve by the
Sheikh and is inhabited only by a few of his retainers.

Apart from some very local and shallow limestone exposures on the shore the whole
island is a flat sandy plain from which emerge two low isolated rocky hills of which the
larger is about 70 feet high. Owing to the extreme flatness of the rest of the island
these hills have a prominence out of all proportion to their size and are visible from a
considerable distance.

The sandy plain compares with the peripheral zone of Bahrain and is presumably
of similar age, and bears a well - developed community of about a dozen species all of
which occur also on the larger island. The hills each compare with the Jebel Dukhan
on a minute scale and the larger has 35 of the same species.

The Island of Jedda

This island, which is about a mile round, is used as a prison. It is a solid flat-
topped mass of limestone with practically no peripheral zone and is almost everywhere

53

surrounded by cliffs less than 100 feet high. The surface of the island is similar to,
and compares with, the summit of Jebel Dukhan as a plant habitat and has a flora of
about 40 species, all represented on Bahrain.

The Island of Raka

This small island is now a private estate with much cultivation and was not visited.

The Bahrain Desert Flora

The main feature of the desert flora of the Bahrain Islands is certainly its lack of
particularity. It may be that one or two of the more critical species will, on close ex-
amination prove to be peculiar to it but it is clear that there is virtually no endemic ele-
ment in the flora. Not only so but considerably more than half the species are gener-
ally described as having ranges which co^er at least the greater part of the whole North
African- Indian desert region, and the occurrence of these in Bahrain calls for no spe-
cial comment, because the islands lie close to the mainland of Arabia, which is one of
the chief constituent parts of this region. There is also a considerable number of spe-
cies which are usually regarded as characteristic of the western or 'Mediterranean'
part of this great region, and for many of these the Bahrain records probably extend the
known distribution considerably to the east. The remaining species, which do not
number more than about twenty, are geographically either Arabian or Persian- Indian,
that is to say they relate to one or other side of the gulf in which the Bahrain Islands
lie. In short, the Bahrain desert flora may be described as essentially an Arabian flora
in which the proportion of more widespread North African - Indian desert species is very
high.

Probably the most generally distributed species is Zygophyllum album which may
occur in almost any situation. Also particularly characteristic are Mesembryanthemum
nodiflorum, Heliotropium tuberculosum, Limonium axillare, Lycium persicum, Aspho-
delus tenuifolius, Aeluropus lagopoides, Sporobolus pallidas and Stipa tortilis.

Other common and prominent plants are:-

Reseda muricata ifloga spicata

Helianthemum kahiricum Richardea tingitana

Helianthemum lippi Senecio coronopifolius

Frankenia pulverulenta Glossonema edule

Spergularia diandra Cressa cretica

Fagonia ? bruguieri Arnehia hispidissima

Zygophyllum simplex Plantago coronopus

Erodium glaucophyllum Hemiaria hemistemon

Erodium laciniatum Sclerocarpus arabicus

Astragalus tribuloides Halopeplis perfoliatus

Medicago laciniata Salsola brevifolia

Trigonella stellata Andrachne telepbioides

Aizoon canariense Cyperus arenarius

Opophytum forskahlei Hyparrhenia hirta

Calendula aegyptiaca Cymbopogon schoenanthus

Launaea mucronata Koeleria phleoides

Launaea nudicaulis Schismus barbatus

54

Of the less common but particularly striking plants may be mentioned Ochradenus
baccatus, Leptadenia pyrotechnica, Aerva javanica, Rumex vesicarius and Calligonum
comosum, and the two parasites Cistanche lutea and Cynomorium coccineum. Small in-
dividuals of the date palm, Phoenix dactylifera, are numerous in the sandier deserts,
but it is difficxilt to determine their status.

The lack of endemism and the high proportion of widely distributed species in the
present flora of Bahrain suggests that it is a relatively new flora, in the sense that it
has not long been established in the islands, and this is of interest because it accords
with the impression gained from other sources also, such as the archaeological, that the
present state of the islands may be of comparatively recent origin. This again impinges
on the much wider problem of the age and history of the North African- Arabian desert
as a whole, which is not only one of the most fascinating questions of palaeogeography
but also one to which a satisfactory answer might be of the greatest significance in
that task of raising or restoring the productivity of the world's desert areas which is
such a pressing urgency of our time.

The foregoing account of the Bahrain Islands and their desert vegetation has been
prepared from material and specimens collected by the writer during a visit to the is-
lands early in 1950. His thank are due to the Royal Society for the generous grant which
made his visit possible, and he would also express his gratitude for the welcome and
help received from His Highness Shaikh Sulman; his Adviser, Sir Charles Belgrave; from
Sir Rupert Hay and other representatives of the British Government ; and from Mr E.A,
Skinner and others of the Bahrain Petroleum Company. For the identifications of most
of the plants mentioned he is indebted to Mr B.L. Burtt, late of the Royal Botanic
Gardens, Kew.

55

HYDRO- ECONOMICAL TYPES IN THE VEGETATION OF NEAR EAST

DESERTS

Professor M. Zohary

(Jerusalem)

This paper presents readily comparable data on the hydro- ecological behaviour
of the leading species of the most common plant communities representative of the
Near East deserts.

The Near East deserts comprise a vast trapezoid limited by the Syro- Palestine
mountain system to the west and by the Zagros mountains to the east. In the north
it merges into southern Anatoly and in the south it is bordered by a line drawn from
Suez Gulf to the Gulf of Aqaba.

The climate of these deserts is an extreme variety of the Mediterranean type,
characterized by mild to fairly cold and rainy winters and dry hot summers. The
mean monthly winter temperature never drops below 0°C. The bulk of the area (a-
bout 80%) is situated between the isohytes of 200 and 50mm and the monthly dis-
tribution of the rainfall is very unstable.

Although rather uniform in the physiognomy of its vegetation, the area under re-
view consists of two plant- geographical territories, the Irano- Turanian in the north
and the Saharo- Sindian in the south (Eig. 1938).

The observations and measurements recorded here were made mainly in the fol-
lowing plant associations: Association of Artemisia monosperma - Convolvulus
lanatus (on the eastern fringes of the coastal sand dunes); Haloxyletum articulati
(on sandy loess); Zygophylletum dumosi (on hammada); Acacietum tortilidis (in
runnels crossing sterile hammada); and Haloxyletum persici (on interior sand dunes
derived from Cretaceous Nubian Sandstone and crystalline rocks).

Pheno - Ecology

3y pheno- ecology I mean those seasonal changes in the plant organs which
affect, directly or indirectly, the water economy of the plant. An analysis of the
flora of the area concerned has led to the distinction of the following types (Fig. 1).

(a) Acacia type. Evergreen trees shedding the old leaves or green branches after
the formation of the new ones, so that defoliation never occurs. Time of leaf-
fall — summer. (Acacia raddiana, A. spirocarpa, Tamarix spp.).

(b) Anabasis type. Evergreen, articulate stem succulents producing new assimi-
lating branches in winter, while certain portions of older branches die back in
the summer (Anabasis articulata, Haloxylon articulatum, H. salicomicum, etc.).

(c) Retama type. Evergreen spartoids, shedding their leaves in early winter (Re-
tama roetan, Calligonum comosum, etc.).

(d) Lycium type. Wintergreen phanerophytes shedding all their leaves in midsum-
mer (Lycium arabicum, Anagyris foetida, etc.).

56

(e) Reamuria types. Chamaephytes which considerably reduce their transpiring
surface at the beginning of the dry period (e.g. Reaumuria palaestina, Salsola

villosa, Siiaeda palaestina, S. asphaltica, Artemisia Herba alba, Zygophyllum
dumosum and many others). This group is the most important among the per -
manent desert vegetation. Biseasonal annuals (e.g. Salsola autrani) are also
included here.

(f) Launaea type. Annuals, crypto — or hemicryptophytes finishing their life cy-
cles at the end of the rainy season. Very abundant.

8

u

i)

Q

a J- u

a,
<

V

a

3

-3

w

3

3
<

e

a,

C/3

V

o

U

O

jD

E

>
o

Xi

B
i)
u

V

Q

Acacia type

Anabasis type

Reaumuria type

Retama type

Lycium type

Launaea type

Filago type

Salsola Autrani type

Figure 1.
Pheno- ecological types in the Near East Desert Vegetation.

57

(g) Filago type. Ephemeral s finishing their life cycle long before the end of the

rainy season. Very abundant.

(h) Salsola type. Summer annuals starting to develop in spring and shedding their
large leaves in early summer, while retaining green, bract- like leaves up to
the end of summer.

The above analysis demonstrates the great phenological diversity of the vege-
tation, and the accordance between phenological events associated with surface
reduction of the transpiring body and seasonal decrease in the moisture resources
of the desert.

Morpho - Ecology

By this term I refer to formal and dimensional changes in the plant body, direc-
tly or indirectly associated with its water economy. This subject also includes
the study of life forms but viewed from another angle than that exposed by Raun-
kiaer. It is not the position and protection of the renovation buds that affect the
water ecology of the plant, but the dimensions of the transpiring organs regularly
lost by the plant in the critical season, that is most essential for the maintenance
of desert summer vegetation. In an unpublished paper, Orshansky (1952) has shown
that in the evergreen Zygophyllum dumosum the summer reduction of the transpiring
surface amounts to V3 of the total transpiring body. In other plants even much high-
er values have been found. The following morphological types have been distin-
quished in the local vegetation. (Fig. 2).

(a) Herbaceous whole- shoot shedders. This type comprises winter annuals, hemi-
cryptophytes and geophytes in which the whole plant or the epigaeous part
only dies away at the beginning of the dry season. This type comprises about
85% of the total flora.

(b) Phanerophytic summer leaf shedders. These include shurbs shedding the
leaves in midsummer (e.g. Lycium arabicum).

(c) Petiolate leaflet shedders. The leaf is composed of two leaflets borne on a
cylindrical leaf- like petiole, all succulent. In late spring the leaflets are
shed while the petioles remain physiologically active during summer (e.g.
Zygophyllum dumosum).

(d) Aphyllous leaf and branch shedders. This type comprises broom- like shrubs,
like Retama spp., Calligonuni comosum, etc., which shed their leaves in
winter and remain green the year round, but in summer a part of the last year's
branches dry up and break down. In this way a considerable part of the trans-
piring surface is removed from the plant.

(e) Aphyllous branch shedders {Ephedra type). They produce no leaves (except
scale- like ones); a considerable part of the green and brittle branches are
regularly shed in the dry season.

(f) Basiphyllous leaf shedders. At the start of the dry period the large winter
leaves crowded at the base dry up and die away, while the flowering shoots
develop small leaves, active during the whole s\xvnmQt.{Artemisia type.).

58

1. Whole- shoot shedders

4. Retama

8. Noaea

5. Ephedra

2. Lycium

6. Artemisia

•1

i

3. Zygophyllum

^

\

¥

7. Reaumuria

9. Anabasis

10. Haloxylon Persicum

Figure 2.
Morpho- ecological diversity among the dominant species in the Near East Desert Vegetation.

59

(g) Brachyblastic leaf shedders. This type comprises the majority of chamae -

phytes, the dominant life form in desert vegetation, though constituting a small
percentage only of the desert flora. The new shoots produce in the axils of
the winter leaves small bud- like branches (brachyblasts) which, after the shed-
ding of the winter leaves, are active during the whole summer {Reaumuria
type).

(h) Aesticladous leaf shedders. In this type the brachyblasts develop in summer
and consist of a spiny axis and minute leaflets, all dying away at the end of
the dry season (e.g. Noaea mucronata).

(i) Articulate shoot splitters. These are articulate evergreen stem succulents in
which considerable parts of the green 'skin' of the last year's stems are dry-
ing off and then split into rings and fall down. Herebelong Anabasis articu-
lata, Haloxylon articulatum, etc.

(j) Articulate branch splitters and shedders. As above but a part of the last

year's branches also break down and fall away in summer {Haloxylon persicum
type).

This way of morphological analysis renders more meaning to the life- form con-
cept. It shows that surface reduction of the transpiring body, achieved in various
ways, is most important for the permanence of the desert vegetation.

Transpiration

In order to obtain critical data on the transpiration behaviour of desert plants,
measurements of transpiration intensity of the leading species of the most typical
plant associations have been carried out throughout the whole year.

Transpiration has been measured by the rapid- weighing torsion balance (Huber,
1927), whereoy excised plant parts have been exposed for 2-4 minutes. In most
cases two or more parallel measurements with the shortest possible interval bet-
ween them were made every hour for each plant. Reference has been made to fresh
weight (see Huber, 1*^27; Walter, 1051; Hygen, 1951) and figures have been calcu-
lated to hourly averages (mg/g.h. ). In order to simplify the presentation of results,
data of a single summer day (in most cases August) and a typical spring day have
been chosen for each plant (Fig. 3).

The conclusions from own data and those obtained by Evenari & Richter (1937)
and Shmueli (1948) are:

(1) Desert plants vary considerably in their transpiration behaviour, both quan-
titatively and qualitatively.

(2) As to transpiration intensity two groups of plants can easily be distingui-
shed: those with a winter and those with a summer maximum. The former includes
the bulk of the permanent vegetation, while the second consists only of a few plants
with extraordinarily deep roots reaching sources of permanent moisture. The fact
that the dominant representatives of the permanent desert vegetation shows a con-
siderable summer decrease in transpiration intensity, is highly significant in the
water ecology of the desert.

60

250

500

lOOOmg/g.h.

-^ Tamarix Articulata

Reaumuria Palastina

■—' Lycium Arabic

um

Retama Roetam

■J Artemisia Monosperms

Zilla Spinosa

Retama Duriaei
■^ Nitraria Retusa
Salsola Autrani
Zygofyllum Dumosum
Haloxylon Salicornicum

-^ Ochradenus Baccatus
Calligonum Comosum

Salsola Villosa
Salsola Inermis

-^ Haloxylon Articulatum

3

Noea Mucronata

Arthrocnemum Glaucum

I

Anabasis Articulata

Artemisia Herba Alba

Acacia Raddiana

Acacia Spirocarpa

Atriplex Halimus

Haloxylon Persicum

0

250

500 lOOOmg/g.h.

Figure 3.

Spring (white) and late summer (black) transpiration rates of dontinant
species of the I'alestine desert vegetation.

61

(3) In regard to transpiration ranges, two groups can be distinguished: steno-
hydric plants with a rather narrow range of transpiration intensity (e.g. Haloxylon
salicornicum, Zygophyllum dumosum, etc.) and euryhydric plants with a wide range
of transpiration intensity (e.g. Artemisia monosperma, Zilla spinosa, Retama duriaei,
Calligonum comosum),

(4) Fig. 3 indicates the distinction of three main categories of plants according
to the summer values of the transpiration rate. These are megahydrics (high trans -
pirers) from 500mg/g.h. upwards, microhydrics (low transpirers) showing values up
to 350mg/g.h. and mesohydrics with values intermediate between both. It is clearly
shown that the micro- and mesohydric types are dominant among the permanent
desert vegetation while the megahydrics are rather exceptional. Indeed, only ex -
ceedingly deep rooting plants belong to the latter category.

(5) I do not agree with Stocker (1933) that there is no relation between habitat
and transpiration. Fig. 4 shows clear differences in the summer transpiration rates
between various plant communities. This difference is particularly striking when
for each plant community one or two dominants are chosen that display the highest
percentage of the permanent plant coverage, as shown in Fig. 4 (broad column).
Comparing various plant communities of the desert with those of the Mediterranean
region one finds striking differences between the two in late summer transpiration
intensities.

Osmotic Pressure of Cell Sap

As in transpiration so also in osmotic pressure desert plants are greatly hetero-
geneous.

Taking the data presented in Fig. 5 as a basis, at least three groups of plants
can be distinguished:

(a) Plants of hydro- or automorphous salines, distinguished by their high osmotic
pressure caused by the accumulation of soluble salts in the cell vacuoles. In
this group of plants the values range between 40 and 150 atm. In spite of the
high pressure they are all low transpirers.

(b) Plants with low or medium summer values distinguished by their high trans-
piration rate and their exceedingly deep roots reaching permanent sources of
soil moisture. These include Acacia spp. and Tamarix spp.

(c) The rest of the plants are true desert plants showing maxima of osmotic values
between 16 and 72 atm. Comparing these values with those available for Medi-
terranean maquis one finds no marked differences between these and the desert
plants. The following plants are particularly worthwhile mentioning: Medi-
terranean, (from Walter, 1951), Olea europaea (52 atm.), Rhamnus alaternus (37
atm.), Phillyrea angustifolia (60 atm.), Lonicera etrusca (53 atm.), Pistacia tere-
binthus (42 atm.). Desert, (my data), Artemisia monosperma (16 atm.), A Herba
alba (29 atm.), Zilla spinosa (17 atm.). Anabasis articulata (58 atm.), Haloxylon
persicum (56 atm.), Calligonum comosum (17 atm.), Retama roetam (26.atm.).

62

mg/g.h
2000

Desert

Mediterranean

Acacietum Tortilidis
^nabasidetosum

Haloxylonetum Persici

1000 ,

500

0
2000

1000.

500.

0
2000 J

Aitemisia Monosperma —

Convolvulus Lenatus

Association

1000.

500.

0
2000

Haloxylonetum Articulati

1000 .
500 .
0

Zygophylletum Dumosi

Cyperus Papyrus — Polygonum
Association

Scolmeto^- Prosopidetum

Artemisia Monosperma — Cyperus
Mucronatus Association

Poterietum Spinosi

Ceratonia Siliqua — Pistacia
Lentiscus Association

Figure 4.

Late summer transpiration rates of dominant species in typical
Mediterranean and desert plant associations of Palestine.

63

Salsola Inermis
Salsola Tetrandra

Suaeda Monoica
Suaeda Palestina
Zygophyllum Dumosum
Anabasis Articulata
Arthrocnemum Glaucum
HaJorylon Articulatum
Haloxylon Persicum

Salsola Autrani
Salsola Villosa

Nitraria Retusa
Salsola Rosmarinus
Lycium Arabicum
Noaea Mucronata

Tamarix Maris Mortui

Tamarix Tetxagyna

Artemisia Herba Alba
Retama Roetam

Ochradenus Baccatus

Acacia Tortilis

Zilla Spinosa
Calligonum Comosum
Artemisia Monosperma

50

100 Atm.
Figure 5.

Spring (white) and late summer (black) osmotic values of the dominant
plants in the Near East desert vegetation.

64

Discussion

A glance at the literature on this subject published during the last three de-
cades shows how inadequate and contradictory is our knowledge of the water eco-
logy of desert plants. From the results obtained one becomes doubtful whether the
older view on xerophytes, as expressed by Pfeffer (1897), Schimper (1898), 'farming
(I9I8) and others, and so strongly condemned by Maximov (1929) and his associates,
is to be rejected altogether. In their effort to bring evidence for the assumption
that xerophytes possess a higher intensity of transpiration than mesophytes, Maxi-
mov and his associates used a series of plants not critically chosen as xerophytes.
Among other plants that Maximov considered as xerophytes were the exceedingly
deep rooting Alhagi and Haloxylon ammodendron, the clearly mesophytic Portulaca
and Zygophyllum fabago, etc.

Our examinations clearly show that plants growing under extreme drought not
only show a very low transpiration intensity as compared with the less xerophytic
Mediterranean plants but also use pheno- ecological and pheno- morphological pro-
perties for further reduction of water loss. This is far from agreeing with the view
of Maximov on xerophytes.

Not less contradictory to Maximov's view is the fact that among the permanent
desert vegetation not a single plant has been found 'with a capacity of enduring
wilting without injury', a character so strongly assigned by Maximov to xerophytes.
\^hat is very striking in desert plant life is that plants of the permanent vegetation
are physiologically active the whole year round; some of them flower just at the
end of the dry season, some set fruits. None of them are so-called 'stop and wait'
plants.

The present study may thus supply substantial data for the reconsideration of
some aspects of the older view on the water ecology of desert plants. 'iXhere the
habitat is exposed to permanent or seasonal drought, the biseasonal vegetation is
exceedingly thrifty in its water expenditure, both in the rainy and the dry season.
This is well shown by the transpiration intensity values, in the phenological events
of the plants falling in the 'right time', and in their morpho- ecological behaviour
resulting in a considerable reduction of the transpiring body during the drought
period. The permanent vegetation is active the whole year round.

In the permanent desert vegetation of the area concerned the following hydro -
economical combined types may be distinguished:

(a) High transpiring, evergreen deep-rooters with transpiration increasing in sum-
mer (e.g. Acacia spp.).

(b) High transpiring, evergreen deep-rooters with transpiration decreasing in sum-
mer (e.g. Tamarix spp.).

(c) Low transpiring, biseasonal and surface reducing deep-rooters with transpira-
tion increasing in summer (e.g. Atriplex halimus).

(d) Low and medium transpiring, biseasonal and surface reducing spartoid deep-
rooters with transpiration decreasing in summer (e.g. Retama spp.).

65

(e) Low and medium transpiring biseasonal and surface reducing articulate deep-
rooters with transpiration decreasing in summer (e.g. Anabasis articulata,
Haloxylon articulatum).

(f) Low and medium transpiring, biseasonal, surface reducing, non- succulent flat-
rooters with transpiration decreasing in summer (e.g. Artemisia Herba alba).

(g) Low and medium transpiring, biseasonal, surface reducing, glyco- and halo-
succulent flat-rooters with transpiration decreasing in summer (e.g. Zygopbyl-
lum dumosum, Reaumuria palaestina).

Summary

(1) The Near East deserts constitute a more or less uniform entity in its clima-
tical and vegetational aspect but is heterogeneous from the point of view of plant
hydro - ecology.

(2) As moisture is the minimum factor, all features associated with hydro -
ecology are of supreme importance to plant life.

(3) There is a variety of morphological, phenological and physiological types
among the local vegetation, all reducing the amount of water expenditure lost
through transpiration.

(4) Of the various life forms the chamaephyte biseasonal are the most impor-
tant elements in th^ evergreen vegetation cover of the desert.

(5) The life form analysis in its conventional approach is of little significance
to hydro- ecology. But in the light of seasonal surface reduction of the transpiring
body it is hydro- ecologically very important.

(6) A variety of morpho - ecological types has been distinguished among the
permanent vegetation of the desert. In most of them seasonal surface reduction is
considerable.

(7) Both with regard to transpiration intensity and to osmotic pressure of cell
sap, various types have been distinguished in the vegetation of the desert.

(8) An attempt has been made to establish combined hydro- economical types
based on properties, associated with the water- economy of the plants.

References

Birand, H. A. 1938. Untersuchungen zur Wasseroekologie der Steppenpflanzen bei Ankara.
]ahrb. wiss. Bot. 87: 93- 172.

Eig, A. 1938. On the phytogeographical subdivision of Palestine. Palest. J. Bot. J. Ser. 1:
4-12.

Eig, A. 1946. Synopsis of the phytosociological units of Palestine. Palest. J. Bot. J. Ser. 3:
183-246.

Evenari (Schwarz) M. & Richter, R. 1937. Physiological - ecological investigations in the
wilderness of Judaea. J. Linn. Soc. Bot. 51: 333-381.

Harris, J. A. 1934. The physico- chemical properties of plant saps in relation to phyto-
geography. Minneapolis: Univ. Minnesota Press.

66

,Q>\CA/.

Huber, B. 1927. Zur Methodik der Transpirationsbestimmung am Standort. Ber. Deutsch.
Bot. Ges. 45: 611- 618.

Hygen, G. 1951. Studies in plant transpiration I. Physiol. Plant. 4: 57- 183-

Killian, Ch. & Faurel, L. 1933. Observations sur la pression osmotique des vegetaux deser-
tiques et subdesertiques de I'Algerie. Bull. Soc. Bot. Fr. 80: 775.

Killian, Ch. & Faurel, L. 1935. Etudes ecologiques sur les fluctuations de la pression os-
motique chez des psammophytes et quelques halophytes algeriens. Ann. Physiol. 11:
70,

Maximov, N. A. 1929. The plant in relation to water. (Transl. By R. H. Yapp). London: Allen
& Unwin.

Migahid, A.M. 1945. Binding of water in xerophytes and its relation to osmotic pressure.
Bull. Fac. Sci. Fouad I Univ. 25: 83-92.

Oppenheimer, H.R. 1951. Summer drought and water balance. /. Ecol. 39, 356- 362.

Orshansky, G. 1952. Ecological studies in the vegetation of the Near East Deserts. V.
Late summer water balance in a stand of Zygophylletum dumosi. (Unpublished)

Pfeffer, ft'. 1897. Pflanzenphysiologie. 2 Aufl. Leipzig: Engelmann.

Schimper, A. F. W. 1898. Pflanzen- geographie auf physiologischer Grundlage. Jena:
G. Fischer.

Seybold, A. 1929. Die physikalische Komponente der pflanzlichen Transpiration. Monogr.
Gesamtgeb. Wiss. Bot. Berlin: Springer.

Shmueli, E. 1948. The water balance of some plants of the Dead Sea salines. Palest. J.
Bot. J. Ser. 4: 117- 143.

Stocker, O. 1933. Transpiration und Wasserhaushalt in verschiedenen Klimazonen. 1. II. 111.
} ahrb. wiss. Bot. 75: 494-549; 78: 751-856; 81: 464-486.

Tadros, T. M. 1936. The osmotic pressure of Egyptian desert plants. Bull. Egypt Univ. Fac.
Sci. 1-35.

Walter, H. 1951. Binfuehrung in die Phytologie. III. Grundlagen der P flanzenverbreitung. I
Teil: Standort slehre. Stuttgart: E. Ulmer.

Warming, E. & Graebner, P. 1918. Lehrbuch der oekologischen Planzengeographie. 3 Aufl.
Berlin: Gebr. Borntraeger.

Weaver, J.E. 1919. The ecological relations of roots. Publ. Carnegie Inst. \^ash. 286.

Zohary, D. 1952. Ecological studies in the vegetation of the Near East Deserts. III. Vege-
tation map of the Central and Southern Negev. Palest. J. Bot. J. Ser. 5.

Zohary, M. 1940. Geobotanical analysis of the Syrian Desert. Palest. J. Bot. J. Ser. 2: 46-96.

Zohary, M. 1947. A vegetation map of Western Palestine. /. Ecol. 34: 1- 19.

Zohary, M. 1952. Ecological studies in the vegetation of the Near East Deserts. 1. Environ-
ment and vegetation classes. Israel Explor. J. 2 (4).

Zohary, M. & Feinbrun, N. 1951. Outline of vegetation of the Northern Negev. Palest. J.
Bot. J. Ser. 5, 96- 114.

Zohary, M. & Orshansky, G. 1949. Structure and ecology of the vegetation in the Dead Sea
region of Palestine. Palest. J. Bot. J. Ser. 4, 177- 206.

Zohary, M. & Arshansky, G. 1952. Ecological studies in the vegetation of the Near. East
Deserts. II Wadi Araba. Vegetation, (in press).

67

THE OCCURRENCE OF PLANT DISEASES IN ARID CLIMATES AND
THEIR AGRICULTURAL SIGNIFICANCE

Professor I. Reichert
(Rehovot, Israel)

INTRODUCTION

One of the great advances in plant pathology in the last thirty years is without
doubt the recognition of the responsibility of environmental factors in the occurrence
and distribution of plant diseases. The significance of these factors and especially
of climate, has been appreciated for a long time by plant growers, who take them into
consideration whenever plant introductions are made. Unfortunately, however, the
phytopathological side is still much neglected. The fact that plant diseases can be
introduced along with plants has not always been given the consideration that it de-
serves.

In another place (Reichert, 1950), I described phytopathogeographical methods
that enable plant growers to be more certain in predicting the reappearance of plant
diseases that may have been introduced with their hosts.

The application of these methods makes possible the exact planning of new im-
portations and may obviate diseases otherwise incurred by introducing plants from
foreign places.

It will be shown that the exothermic climate may be utilized within the frame of
phytopathogeographical planning for plant disease- control. The application of the
principles to be described will diminish to a great extent the affliction of crops by
diseases. It is known, that only a limited number of pathogens are able to establish
themselves and achieve their cycle of development in a xerothermic climate. This key
fact must therefore be utilized whenever introductions are made and the timing of sea-
sonal cultivations planned. The most important point that must be kept in mind in res-
pect to xerothermic climates is that a great part of the summer (the chief growing sea-
son for many crops) is either rainless or nearly so, and that only a relatively few
pathogens can surmount this obstacle and accomplish their cycle of development. This
fact has great agricultural significance.

Below, some examples taken from observations on plant diseases made in Israel
and other parts of the Middle East are quoted. The maladies mentioned are hampered
in their development in this area which is characterized by rainless summers, and a
winter precipitation (from November — December to March — April). In Israel this
diminishes in quantity from North to South and from West to East, thus, Safad in the
mountains of northern Galilee has an average of 744mm. of rain per year, whereas
Beersheba in the South (Negev) has 193mm. and Haifa on the coast has 6l2mm. There
is approximately 400mm. average yearly rain in Ain Harod in the great interior valley
of Je2yeel in the eastern part of the country average July temperatures range from 22.4 C
at Safad, 24.5°C at Beersheba, 25.5°C at Haifa, to 27.7°C at Ain Harod. Relative hu-
midities during July average 48% at Safad, 65% at Beersheba, 74%at Haifa and 56% at
Ain Harod. (Anon., 1938; Ashbel, 1949).

68

OBSERVATIONS

Cereals

Wheat. As the first example of a disease derived from a cool climate that sel-
finds adequate conditions for development in arid climates, we may mention Tilletia
tritici. It does not often occur in Israel, is especially rare in the interior valleys and
is to be found only when rain and ensuing cold weather follow closely upon the sowing
of the grain. The optimum temperature for seed germination is 25°C while that for ger-
mination of the spores is 10- 16°C. Since rain generally does not ^pear until some
time after sowing, the seedling usually has enough time in which to evade attack. Such
a case was observed in Palestine in 1924 (Reichert, 1928 a). The same conditions and
results are known in other xerothermic countries, such as southern India, southern Rus-
sia, the Spokane Valley in Washington, and others (Heald, 1933; Walter, 1950).

A second example is Claviceps purpurea, which is favoured by 'abnormally wet
seasons with reduced amount of sunshine' (Heald, 1933). Since arid regions are al-
ways dry during the blossoming period of wheat, no infection can take place here. Avi-
zohar (1947) was able artificially to produce this disease in Israel only by increasing
the humidity around the plant by artificial means. Similar environmental conditions
prevail in the distribution of the disease in other countries. (Heald, 1933).

Maize. One of the healthiest summer crops grown in Israel is maize, the only
diseases affecting the plant here being Puccinia sorghi and ilstitago zeae. They oc-
cur only near the sea or under irrigation ; conditions that raise the humidity to such an
extent that the pathogen can successfully attack the crop. This is particularly true in
the case of the rust. It is interesting to note that, although all the maize varieties
were originally derived from localities in the U.S. and were sown here without seed
treatment, they did not reveal any of the serious diseases recorded in other maize -
growing countries, e.g. Diplodia zeae, Sclerospora macrospora, and Physoderma zeae-
maydis. These three diseases require abundant and frequent rains throughout the grow-
ing period of the corn crop, conditions not to be found in arid regions. They are there-
fore also excluded from the arid south-western part of the U.S. (Heald, 1933; Walker,
1950).

Wheat and maize in our country have not been affected by bacterial diseases like
Pseudomonas translucens var. undulosa and P. desofvens, that cause great damage in
other countries.

Vegetables

A good example of how an arid climate can preclude or limit the occurrence of an
important vegetable disease is to be found in the case of 'black rot' of cruciferous
plants (Pseudomonas campestris). It is widespread in coastal regions of Israel (Rei-
chert, 1939 b), but its presence in the interior valleys is limited to early spring and
late autumn. In the U.S. a similar state of affairs exists, with the disease prevalent
only in the humid areas east of the Mississippi. It is very rare in the Rocky Mountains
and Pacific coast (Heald, 1933). Walker (1950) tells us that it is entirely excluded

69

from 'regions where the rainfall is very low during the period when seedlings are being
grown'. According to him, 'black rot' is little known in^rfie Puget Sound region in
Washington and in Pacific coast areas where summer precipitation is low. In these
regions, seed may be grown free from bacterial infection.

Solanaceous Crops. Characteristic, and of great agricultural import, is the mode
of occurrence and distribution of some important diseases attacking potatoes and to-
matoes in Israel. These crops are afflicted in warm, humid climates by three high-
humidity- loving diseases, - P seudomonas solanacearum, Phytophthora infestans and
Cladosporium fulvum. In Israel they are to be found only in certain areas. P. solana-
cectnim was found here only on potatoes and even then, only in the one year, 1947, in
the coastal strip during a particularly damp spring and late autumn (Littauer, Volcani,
& Temkin, 1926). Phytophthora infestans is considered in our country to be a serious
menace to the winter- sown potatoes because of the abundant rain and the mild tem-
perature that prevails in that season. But the opposite is the case in the late spring
and the autumn, and the occurrence of the disease at this time is almost nil or very
scarce in the interior valleys, in upper Galilee and in the dry Negev. It is interesting
to note that until ten years ago, the disease was rather scarce even in the coastal re-
gion, but owing to the introduction of overhead sprinkling it has been on the increase.
It seems to be a known fact in Israel that morning sprinkling is less conducive to
disease than that carried out late in the day, as the latter method extends the time in
which there is high humidity, and with it the possibility of infection. Even in the
coastal belt, though, the disease is halted during the months of April and May by the
drying sirocco winds (khamsin), and it renews its development in June.

Phytophthora infestans has been causing much damage to tomato crops within the
last years, and we are paying a great toll to this plant invader. It makes incursions
around the Sea of Galilee and in the coastal strip, but to a much lesser degree in the
interior valleys and the Negev. In the late spring it does not appear at all in the
Negev.

Cladosporium fulvum is a lover of high humidity (Walker, 1950) and therefore is
limited to the spring and autumn in the coastal area and is not to be seen in the in-
terior valleys.

Tomatoes are afflicted in our country and in many other Mediterranean countries
by the xerophilic plant disease Oidiopsis taurica, which attacks them heavily in the
interior valleys in the autumn, when the optimum relative humidity for the germination
of spores is less than 70% (Reichert, 1939b; 1949). Potatoes, in turn, are attacked
by an Oidium fungus which shows even greater xerophilic tendencies. Leaves are at-
tacked only in regions where humidity is lower than 50% (Reichert, 1949).

Plantations

Deciduous trees. Taphrina deformans is a serious malady of stone - fruit trees
in cooler countries. It appears in Israel chiefly on almond trees at higher altitudes in
early spring, since almond is an early bloomer. It occurs also to a lesser extent on
peaches. (Reichert, 1939c).

70

Another important disease of prunaceous trees in cooler climes is the 'scab' of
apple and pear {Fusicladium dentriticum and F. pyrinum). Only the local varieties of
these fruit trees, the early bloomers, fall prey to this malady, as they flower at a time
when rain and high humidity are frequent. European varieties, which form blossoms
and leaves three to six weeks later, when rain has already ceased and temperature
gone up, escape the disease (Pelberger, 1944),

Very interesting is the occurrence of 'rust' (Puccinia pruno - spinas ae) on pruna-
ceous trees, especially almond. Whereas it is very serious in humid climates and even
in the California coastal regions (Goldsworthy & Smith, 1931) due to the frequent fogs,
it makes its appearance in Israel in June, after the cessation of the 'khamsin' wind
season. In the interior valleys it is non-existant (Perlberger, 1943).

Another important disease of stone-fruits is 'brown rot' (Monilinia fruticola)
This species is a menace to fruit cultivation in cool, humid countries (Heald, 1933;
Walker, 1950), but does not appear in Israel, even under irrigation, for low soil mois-
ture and air humidity, combined with the high temperature, of the summer, impede the
attack upon the fruit blossoms.

Mediterranean trees. An interesting illustration of the controlling effect of xero-
thermic climate upon plant disease is provided by the diseases of some Mediterranean
trees. First among these is the downy mildew of grapevine (Plasmopara viticola) which
occurs in Israel only in the coastal region, and is checked in the interior valleys and
in the higher altitudes of the Jerusalem area and the upper Galilee (Reichert, 1927).
In valleys the appearance of the disease is hampered by the high prevailing tempera-
ture, amounting to 29- 29.9°C during the growing season, which is the maximal ger-
minating temperature of the conidia of the pathogen. In elevated localities, on the
other hand, the lower relative humidity and the scarcity of dew through wind action
precludes the occurrence of the disease (Reichert, 1927).

It might be worth mentioning that another important disease of the vine, anthrac-
nose (Cloeosporium amp el op ha gum) does not occur at all in Israel and neighbouring
countries. The reason for its absence here lies in its tropical origin ; it is prevalent
in the warm, humid climates, of the low- latitude Atlantic countries of Eurafrica, the
Western Caucasus, south-eastern North America and parts of the southern hemisphere.
It may be of interest to mention the fact that wine growers in our country had treated
vines against this disease for more than thirty years, having been influenced by French
instructors to whom the disease was a common thing. We were able to convince them
that the treatment was unnecessary, since the disease is non-existent here.

Another serious vine disease of cooler countries that does not occur in Israel and
the rest of the eastern Mediterranean area, nor in the arid parts of the U.S. and other
countries, is Botrytis cinerea. The high humidities required by this pathogen are not
present in these countries during the summer.

A characteristic disease of olives, Cycloconium oleaginum, which causes leaf-
drop, is to be found here chiefly in the coastal region since the interior valleys and
Galilee have not the high humidity that the parasite demands. When olive trees are
irrigated however, the disease becomes evident to a small extent even there.

71

In 1934, when visiting the vast plantations of Pistatia vera in the great steppes
east of Aleppo, we noticed that the trees were completely free of any disease, where-
as westward towards the seaport of Lattaquieh, we found them heavily infected by
Septoria pistaciae.

Citrus trees. The limiting effect of dry weather conditions upon plant disease
is particularly pronounced in the case of citrus diseases, and is evident in Israel and
adjacent countries. The arid conditions of this region exclude two diseases that in
warm humid countries are considered to be among the worst scourges of citrus. They
are 'scab', caused by the fungus Sphaceloma fawcetti, a near relative of the anthrac-
nose of vine, and 'canker', a bacterial disease caused by Phytomonas citri (Fawcett,
193^). Peltier & Frederick (1922) revealed the dependence upon warm and very humid
conditions of these diseases.

Another bacterial disease, Phytomonas syringae, the causative agent of 'blast' of
of citrus, develops in our country only when the autumn is particularly rainy and cool
(Reichert & Perlberger, 1928; Reichert, 1939 b).

Industrial Plants

Tobacco. Although this country has for many years been importing tobacco
seeds from U.S.A. and other countries that grow the crop under warm, humid conditions,
there has been no emergence of the very serious diseases that tobacco has been heir
to in the countries whence the seed came ; we have not found to date any of the downy
mildews such as Phytophthora nicotiana, Peronospora hyoscyamae, nor the various
bacterial diseases that affect tobacco. Even Bacterium solanacearum, which was men-
tioned above as having once attacked potatoes in the spring here, has failed to appear
on tobacco in Israel, The reasor! is that the crop here is grown during the dry summer.

Other instances of summer crops avoiding fungal attacks are provided by the sun-
flower (Helianthus annuus) and safflower {Carthamus tinctoria), both of which, in cool
damp countries are prey to Sclerotinia sclerotiorum. Safflower is attacked also by a
downy mildew, Plasmopara halsteadii. In Israel these diseases make no headway
whatever in the summertime, the optimal temperature for the germination of the asco-
spores of Sclerotinia sclerotiorum being 17°C : much lower than the average prevailing
temperature of the air during the growing period of the host. Besides, the parasite de-
mands a great amount of soil moisture for the fruiting bodies to discharge their spores.

DISCUSSION

The accumulated data regarding crop plants escaping disease when grown in hot,
arid, regions show clearly how important these regions, including steppe and desert
may become for agriculture. All these vast, neglected lands, hitherto considered as
'barren', may be transformed into productive, remunerative agricultural areas for grow-
ing crop plants since these are here less subject to the attacks of dangerous para-
sites than if they are grown in humid areas. Thus free of such troubles, the grower is
able to devote his energies to the control of the few xerophilic diseases remaining.
For instance, great areas of vineyards in the interior valleys of Israel, in the neigh -

72

bourhood of Damascus and in lower tgypt remain untouched by the standard maladies
that are usually met with in humid regions. The only serious disease remaining here
is powdery mildew, and this can be easily controlled. Citrus and other southern trees
escape the bulk of plant plagues if grown in xerothermic regions. This is especially
true of the leaf diseases usually concomitant with them in warm, damp climates. Thus
grapefruit groves in the interior valleys of Israel, as well as the large citrus planta-
tions in the vast western desert of Egypt are free from all leaf diseases. The same
may be said of cereals, particularly maize, and of many vegetables and industrial
plants, all of which, under these conditions, are quite free of hygrophilic diseases.

The above data show that, a distinction must be drawn between two types of xero-
thermic region. The first may be designated as interior areas, far removed from the
sea, which are characteristically free of certain diseases. The second type comprises
the area bordering the sea. It is naturally subjected to higher humidity and to humid
breezes, which modify the immunity found in the areas of the first category. For ex-
ample, the downy mildew of vine {Plasmopara viticola) which is prevalent, in a mild
form in the coastal plains of the Mediterranean, disappears in steppes and deserts far
from the coast (Reichert, 1927).

Another important factor in the controlling effect of arid climate on plant disease
must be recognised. This, is the difference between summer and winter periods. The
winter climate in a xerothermic region is cooler and more humid than that of summer,
and therefore may enhance the development of certain diseases that are suppressed in
the summer. Thus, Sclerotinia sclerotiorum attacks plants during the winter and not
during the summer because it has a low minimal temperature — 1°C and a low optimal
temperature — between 17 and 21°C. Decisive factors in the success of infection are
the amount of precipitation, the relative air humidity, dew, and last but not least, the
altitude of the locality.

As a good example, the distribution of the vine disease, Plasmopara viticola may
again be mentioned. This is limited to the coastal plains where sufficient humidity
and dew are present, whereas it disappears in the inland valleys where they are ab-
sent, as in the high mountains of Galilee, where freely blowing winds quickly dry up
any dew that is formed and diminish humidity of the air. The great importance of hu-
midity in arid countries in the propagation of the Plasmopara pathogen was demon-
strated during this study of the comparative distribution of the disease (Reichert,
1927).

Another important ecological point that must be emphasized is the duration of
night and the subsequent dew formation. In northern countries, parasitic fungi avail
themselves of the necessary dampness for infection both in the night and in the day-
time. In the dry regions however, these conditions prevail only during the night. It
has been demonstrated experimentally by a study of the ecological conditions on
downy mildew of cucumber {Pseudoperonospora cubensis) the extent to which dew for-
mation is essential for the development of diseases in dry regions. Vvhen dew was
prevented from forming on plants by means of canvas covers, no infection took place
(Goldsworthy & Smith, 1931).

73

The change from furrow irrigation to overhead sprinkling must also be considered
if one is to obtain a clear picture of the value of arid region farming. The general be-
lief in Israel is that the rapid spread of late blight (Phytophthora infestans) is due to
the abandoning of the old furrow method for the intensive use of overhead sprinkling.
In the case of the cucumber disease already mentioned, this has been experimentally
established. (Duvdevani, Reichert & Palti, 1946). Sprinkling changes the microclimate
of the plants by increasing the moisture on the leaf surface and the moisture over a
greater area of soil. Consequently a greater opportunity for parasitic activity presents
itself.

It is noteworthy that in arid regions, root and stem diseases are more prevalent
than leaf diseases, and may also appear in irrigated crops, but never to the extent
found in humid areas.

In xerothermic climate there are, therefore, two primary components which operate
in the inhibition of plant diseases — high temperature and low humidity. The relative
importance of each factor depends upon the origin of the disease. Uith those derived
from a cooler climate, as Tilletia tritici, Sclerotinia sclerotiorum, Phytophthora infes-
tans, etc. the high temperature is the limiting factor, but in the case of pathogens ori-
ginating from tropical countries, such as Bacterium colanacearum, scab of citrus
(Sphaceloma fawcetti), and citrus canker (Phytomonas citri), the low air humidity be-
comes decisive in the inhibition of the diseases.

References '

^^on, 1938 Government of Palestine. Meteorological Observ. :23pp.

Ashbel, D. 1949 Bio- climatic atlas of Israel. Met. Dept. Heb. Univ. : 151pp.

Avizohar- Hershenson, Z. 1947 The inoculation of rye with Claviceps purpurea as related to
environmental conditions. Palest. J. Bot. Rehovol.

Duvdevani, S. Reichert, I. & Palti, J. 1946 The development of downy and powdery mildew of
cucumbers as related to dew and other environmental factors. Palest. J. Bot. Reh. Ser. 5 :
127- 151.

Fawcett, H.S. 1936 Citrus diseases and their control. 656pp. New York.

Goldsworthy, M.C. & Smith, R.E. 1931 Studies on a rust of Clingstone peaches in California.
Phylofjath. 21; 133- 168.

Heald, F.D. 1933 Manual of plant diseases. 953 pp New York.

Jaczewcki, A. 1935 Bacterial plant diseases (in Russian) 750pp. Moscow.

Littauer, P., Volcani, Z., & Temkin, N. 1947 Brown rot of potatoes in Palestine. Palest. J. Bot.
Rehovot Ser. 6, : 2 19 -'20.

Peltier, G.L. & Frederick, V^'.G. 1926 Effects of weather on the world distribution and preva-
lence of citrus canker and citrus scab. ]. Agric. Res. 32: 147- 164.

Perlberger, J. 1943 The rust disease of stone fruit trees in Palestine. Bull. Agric. Res. Sta,
Rehovol. 34: 9pp. (In Hebrew with English Summary).

Perlberger, J. 1944 The occurrence of apple and pear scab in Palestine in relation to weather
conditions. Palest. J. Bot. Rehovot. Set. 4 : 157- 16 1.

Reichert, I. 1927 Downy mildew (Plasmopara viticola) of the vine in Palestine. Yedeoth, Tel
Aviv, 7-8 : 28pp. (In Hebrew with English Summary).

74

Reichert, I. 1928 a Comparative bunt resistance of wheat in Palestine. Agric. Exp. Sta. Tel
Aviv Bull 9: 29 pp.

Reichert, I.1939a Palestine: Plant disease of Citrus, Int. Dull. Plant Prot. Rome 13: 75-81.

Reichert, I. 1939 b Palestine: Plant diseases of vegetable crops. Int. Bull. Plant Prot. Rome,
13: 225-240.

Reichert, I. 1939c Palestine: Diseases of fruiting plants except citrus Int. Bull. Plant. Prot.
Rome. 12: 277-293.

Reichert, I. 1949 Climatic factors in the practice of fungicidal treatments in a Mediterranean
climate. Proc. 2nd. Int. Congr. Crop Prot. London, : 8 pp.

Reichert, I. 1950 A biographical approach to phytopathology. Proc. Int. Congr. Bot. Stockholm,
(in press).

Reichert, I. 1928 b & Perlberger, J. The Blast disease of citrus — a new citrus disease in
Palestine. Hadar, Tel Aviv, 1(4): 3-11.

Walker, J.C. 1950 Plant pathology, 699., New York.

75

PHYTOSOCIOLOGIE HT MISE EN VALEUR DES SOLS EN AFRIQUE DIJ NORD

Professor Louis Hmberger
{S\ontl>ellier)

L'Afrique du Nord Francaise a Linterieur de ses limites politiques est au 9/10°
aride ou semi - aride.

Les travaux que nous y avons entrepris sont nes des preoccupations des Gouver-
nements de nourrir une population qui s'accroit regulierement tous les ans.

Le probleme est, en effet, urgent: I'accroissement de la population est tel qu'il
faut tous les ans un supplement de 50,000 tonnes de cereales, chiffre base sur une
ration de 300 gr. par jour.

II est done necessaire de trouver des terres.

L 'aspect humain du probleme n'est pas le seal, on comprend qu'il ait aussi un
cote politique.

Les trois gouvernements de I'Afrique du Nord ont decide' la mise en valeur des
perimetres suivants :

En Tunisie: Vallee de la Medjerda, Tunisie centrale (pays de Sbeitla, Kasserine,

Cap Bon, littoral septentrional, Oasis du Sud.
En Algerie : Vallee du Chelif, region de Bone, Sahel d'Alger, environs d'Oran,

Hauts - Plateaux, confins algero- marocains meridionaux.
Au Maroc : Tadla- Beni Amir (pied occidental du Moyen Atlas meridional), Haouz

— pays d'El Kelaa, Doukkala (sublittoral), Gharb, pays de Berkane

(Maroc oriental)

Toutes ces regions sont arides ou semi-arides.

Comment le phytosociologue peut- il contribuer a la mise en valeur rationnelle du
sol?

On sait que, dans la nature les especes ne sont pas reparties sans ordre et d'une
maniere quelconque, mais qu'clles constituent des groufje merits ordonnes, hierarchises,
floristiquement definis, que nous appelons associations.

Or, la grande valeur pratique de cette notion est qu'a I'interieur d'un meme terri -
toire floristique, la meme association se retrouve chaque fois que les conditions eco -
logiques sont les memes. A chaque association correspond une ecologie determinee.

Sur route I'etendue d'une association les conditions ecologiques sont les memes.

II y a correspondance entre associations et milieux, d'une part, entre milieu et
vocation, d'autre part, done, aussi entre associations et vocations.

Etant donne cette solidarite, nous pouvons etablir une carte des milieux, si nous
cartographions les associations.

Nous avons ainsi entrepris, en Afrique du Nord, sur une grande echelle, des tra-
vaux de cartographie des associations.

Ces cartes sont done, en fait, des cartes de vocation des sols. Elles sont etab-
lies au 20,000°, parfois a des echelles plus grandes encore suivant les besoins.

76

Mais, dira-t-on, comment p rati que rnent, ces cartes des associations peuvent-elles
nous indiquer quelles sont les aptitudes agronomiques des sols qui les portent?

Nous procedons comme les geologues. Ceux- ci savent, par example, dans quelles
conditions tectoniques on troube le petrole. lis dressent done la carte geologique et
font des forages la ou les structures correspondent a celles qui ont ete observees.

Ainsi, nous etudions avec soin, en plus des associations, les essais de cultures,
reussis ou non, faits dans le perimetre dont nous levons la carte phytosociologique.
Si une culture a donne de bons resultats dans une association, il est clair que les
memes resultats peuvent etre obtenus partout ou la dite association existe.

En Tunisie meridionale, par exemple. Tune des questions posees etait de savoir
jusqu'ou il etait possible d'etendre les oasis existantes, mais devenues insuffisantes
pour la population.

Le probleme a ete resolu par la methode phytosociologique. Nous avons constate
que les Palmeraies etudiees etaient installees empiriquement par les Indigenes dans
des associations determinees. (Assoc, a Suaeda fruticosa et Salsola tetrandra, Assoc,
a Limoniastrum guyonianum et Halocnemum strobilaceum (sous- assoc. a Arthrocnemum)).
En cartographiant ces associations favorables au Dattier, nous avons pu definit les
limites d'extension possible et determiner des regions ou de nouvelles palmeraies pour-
raient etre crees, pourVu que I'eau necessaire soit disponible.

Si aucune experience n'a encore ete faite, soit dans une region donnee ou avec
une plante exotique interessante que I'on voudrait introduire, la methode est un peu
plus compliquee. On fera des essais dans les regions a climat susceptible de con-
venir a priori a la culture, comparable a celui du pays d'origine de I'espece envisagee.
Mais on fera les essais dans des associations determinees. Les resultats indiqueront
quelles sont les associations indigenes les plus favorables a la culture projetee.

L 'experience agricole prealable n'est meme pas toujours necessaire. Nous savons,
par exemple, que la plupart des cultures de plantes annuelles (ble, etc. ,) ne peu vent
etre faites immediatement dans certaines associations (Assoc, a Lithospermum fruti-
cosum, a Erica multiflora, etc. ,). Le sol de ces associations est intoxique par les
secretions des racines des principales especes qui la constituent. On peut cultiver
dans de tels sols, en fait d'annuelles, que des Legumineuses, dont les nodosites con-
tiennent des antitoxiques ; pour les autres cultures annuelles, il est necessaire d'at-
tendre jusqu'a ce que les pluies aient suffisamment lave les sols qui ont ete habites
par les associations toxiques.

Un autre exemple nous a ete donne par un grand domaine de Tunisie centrale.
Cette propriete a ete plantee d'Oliviers, d'Amandiers et d'Abricotiers, dans un sol qui
paraissait parfaitement homogene. Or, au bout de 10 ans, les arbres ont decline dans
une partie de la propriete, tandis qu'ils continuaient a vivre normalement sur le teste
du domaine. Nous avons procede a une etude phytosociologique et constate que la
propriete etait etablie dans une association, mais que sur certaines parcelles, cette
association existait sous forme d'un fades caracterisee par la presence tres dispersee
d'une Salsolacee (Salsola tetrandra). Or, ce facies indique qu'il y a du sel a plusieurs
metres de profondeur. Les arbres fruitiers, au debut de leur plantation, ont done pros-

77

pere normalement, mais au bout de 10 ans tous ceux qui etaient dans le facies a Sal-
sola, leur racines touchant le sel, ont deperi. Si un phytosociologue avait prospecte
ledomaine avant laplantation, il aurait pu delimiter avec precision les parcelles favor-
ables a 1 'arboriculture et celles ou seules des cultures herbacees, a enracinement
faible, etaient seules possibles.

Et le sol dira-t-on? La connaissance du sol est naturellement capitale aussi,
car il n'y a pas d'etude phytosociologique complete sans fiche pedologique, mais notre
experience nous permet de donner un avis formel sur un point: La prospection phyto-
sociologique doit preceder les recherches pedologiques. Etant donne qu'il y a corres-
pondance etroite entre les associations et la pedologie, les pedologues peuvent gagner
beaucoup de temps en prelevant les profils par associations, non par unite de surface
a prospecter. Un seul profil peut suffire pour toute la surface, meme tres grande, si
celle-ci n'est occupee que par une seule association.

Inversement, la diversite phytosociologique doit inciter les pedologues a multi-
plier les trous la ou ils serai ent rentes de n'en faire qu'un petit nombre, s'ils n'etaient
pas guides par la phytosociologie. Dans la vallee de la basse Medjerda (Tunisie), les
pedologues ont pu utilement diriger leurs prospections en tenant compte des groupe-
ments vegetaux, certains sols leur ayant echappe avant qu'ils eussent connaissance
de ces faits.

La methode exposee iciest generale, applicable a tous les cas, bien entendu aussi
a tous les pays arides ou semi- arides.

J'ai I'honneur de vous presenter deux cartes, I'une d'Algerie (Basse vallee du
Chelif), I'autre de Tunisie centrale etablies par mes coUaborateurs pour les Gouverne-
ment respectifs.

Je commenterai brievement la carte Tunisienne.

Elle est couverte par 27 Associations definies suivant la methode pratiquee a
MONTPELLIER. Les vocations economiques des surfaces couvertes par les asso-
ciations sont tres diverses ; en voici quelques exemples :

Les associations a Rosmarinus officinalis — Stipa tenacissima — Reseda papu-
losa et celle a Launaea mucronata — Erodium glaucophyllum ont une vocation fores -
tiere (Juniperus phoenica, Pinus halepensis), mais la foret sera plus difficile a in-
staller dans la 2° association que dans la premiere.

L'Association a Eragrostis papposa — Ziziphus Lotus - Artemisia campestris et
a Chrysanthemum coronarium — Peganum hamala sont excellentes pour 1' arboriculture
fruitiere (Amandiers, Oliviers, Abricotiers).

Une sous- association (a Stipa parviflora) de I'association a Eragrostis forme d'
excellents paturages.

L'Association a Aristida pungens et Rumex tingitanus var. lacerus convient a des
plantations d'Opuntia.

L'Association a Artemisia herba alba — Haloxylon tamarises folium (sous- asso-
ciation de Stipa parviflora), en melange avec I'association a Plantage lagopus — Sily -
bum eburneum est apte aux Cereales.

78

Dans I'association a Salsola vermiculata vai.villosa on peut encore planter des
Figuiers et des Amandiers, et des culture de cereales sont possibles; elle n'est pas
trop salee. Mais les surfaces couvertes par I'association a Limoniastrum - Salsola
CTUciata, souvent en contact avec la precedente, sont a eviter soigneusement. Ces 2
associations, surtout la premiere, peuvent etre utilisees aussi comme paturages.

Certains groupements, tels que I'association a Arthrocnemum glaucum et celle a
H alocnemum strobilaceum sont impropres a toute mise en valeur, en I'etat actuel de la
situation bien entendu.

79

LES RELATIONS ENTRE LES ZONES DESERTIQUES ET LA PULLU-
LATION DES PARASITES DES PLANTES

Professor Paul Vayssiere

(Paris)

A ma connaissance, la pullulation des parasites des plantes, cultivees ou
spontanees, dans ses rapports avec les conditions creees par les deserts et les
sub- deserts, n'a encore jamais fait I'objet d'observations d'ensemble d'autant plus
que, suivant I'espece des etres consideres et suivant la specialite du biologiste
interesse. Le role joue par les zones arides et semi- arides peut etre oppose et,
en tout cas, different.

Malgre, la complexite du probleme, je vais m'efforcer, par quelques exemples
frappants, choisis dans des groupes aussi eloignes que possible, de mettre en evi-
dence son aspect double et contraire. Toutefois, il me parait impossible, dans ce
simple expose, de separer les regions typiquement desertiques — deja si differentes
entre elles — des zones qui les entourent et qui ne sont qu'a demi arides. La,
I'Homme, grace a sa tenacite et parfois a son genie, a pu installer des cultures qui
constituent des ilots encercles d'aires couvertes d'une vegetation xerophile plus
ou moins spontanee, laquelle favorise la pullulation d'animaux phytophages suscep-
tibles de jouer, un jour ou I'autre, un role economique non negligeable. Pour etayer
son action I'Agronome doit done tenir compte des etres vivants qui evoluent et se
multiplient dans ces regions sub- desertiques, steppes ou savanes a ecologie si
complexe sous la dependance des facteurs climatiques, et surtout microclimatiques
en ce qui concerne les parasites des vegetaux,

Dans de nombreux cas, les surfaces arides, par les conditions memes qui les
caracterisent, constituent des barrieres infranchissables pour les parasites des
plantes, que celles-ci soient spontanees ou cultivees. Mais les moyens de trans-
port de plus en plus perfectionnes et rapides mis a la disposition de I'Homme ont

supprime cette protection naturelle de sorte que de tels parasites ont progresse
lentement d'oasis en oasis, le long des pistes avec les caravanes, tandis que
d'autres se trouvaient transportes par des voies plus rapides. Deux exemples pour
illustrer cette assertion: il est incontestable que le Ver rose de la capsule du
Cotonnier, le Pink Boll worm (Platyedra gossypiella Saund.) qui etait signale en
Egypte des avant 1910, n'a jamais pu, par ses propres moyens, traverser les regions
sahariennes pour se repandre dans les cultures cotonnieres de I'Afrique occidentale
bien que le genre Gossypium existe dans la plupart des oasis, II a fallu des en-
vois, plus ou moins clandestins, par poste ordinaire ou par la voie des airs, en Ni-
geria et en Afrique franqaise, de semences egyptiennes selectionnees, pour que ce
dangereux parasite puisse s'implanter dans les cultures au sud et a I'ouest du
Sahara. Les cochenilles specifiques du Dattier (Phoenix dactilifera) telles que
Phoenicococcus marlatti Ckll. et Parlatoria blanchardi Targ. ont, elles, progresse
lentement. De fortes presomptions situent leur pays d'origine dans le Moyen -
Orient, peut-etre dans la region de Sinai (oasis d'El Arish) ou Bodenheimer les a

80

etudiees (1924) sur des palmiers qui ne souffraient pas de leur presence, malgre une
une grande abondance d'individus; par le d^placement des caravanes, elles ont pen
a peu etendu leur aire d'habitat vers I'ouest du continent africain si bien que, a
I'heure actuelle, celle-ci se confond avec celle du Dattier? Les diverses etapes
de cette extension ont ete notees successivement, sur le territoire saharien fran-
cais, par L. Trabut, P. Vayssiere, A, Balachowsky.

En ce*qui concerne les parasites vegetaux, nous n'avons que tres peu d'obser-
vations precises sur le role utile joue par les deserts vis-a-vis des plantes cul-
tivees en particulier. Toutefois, il y a lieu de rappeler que, pour un grand nombre
d'especes de Cryptogames, les conditions des zones desertiques alterent leur pou-
voir de germination, done de propagation. C'est ainsi que Puccinia graminis re-
monte bien la vallee du Nil jusqu'a 60 km. environ au sud d'Assouan (observation
personnelle de M. Viennot- Bourgin en 1936) mais est incapable de parasiter les
Cerealeset les Graminees sauvages au-dela de ce point par manque d'humidite. On
ne le rencontre done pas au Soudan anglo- egyptien.

A cote du role utile que peuvent jouer les zones arides — role altere par les
transports: ce qui le fait de I'Homme — il en existe un autre, nefaste et particu-
lierement important, surtout pour les zones semi- arides: c'est qu'elles peuvent
servir de refuge, et en consequence de zones de multiplication, pour certains ani-
maux et vegetaux. Que ce soit pour repondre aux necessites de I'espece ou pour
suivre les lois de 1 'expansion, il arrive tou jours un moment ou ces etres vivants se
repandent sur les regions avoisinantes, et meme parfois lointaines et deviennent,
tres souvent, de veritables fleaux. On peut alors voir des especes phytophagps,
typiques ou occasionnelles, qui s'abattent sur les cultures et les detruisent
rapidement, ou s'adaptent a elles et les aneantissent progressivement, redui-
sant frequemment a la misere les populations de regions deja defavorisees. De
nombreux exenples peuvent etre fournis et le plus marquant, comme le plus connu,
est celui des Acridiens migrateurs, ou tout au moins de certaines especes d'entre
eux. Le plus caracteristique est, evidemment, le Criquet pelerin (Schistocerca gre-
garia Forsk.) dont les aires de multiplication (aires gregarigenes) se localisent, en
Asie et en Afrique, en bordure des immensites desertiques nord- equatoriales de
I'Atlantique au Pakistan et en Inde, en passant par I'Arabie et la Perse. Quant a
Locusta migratoria s.l, ne peut -on placer, sans discussion possible, les princi-
pales aires gregarigenes dans des zones semi- desertiques?; estuaires des grands
fleuves des bassins fermes en voie de disparition, en Russie meridionale et en Asie
pour Locusta migratoria migratoria, ancien delta du Niger pour Locusta migratoria
migratorioides. Dans le sud de Madagascar, les aires gregarigenes de Locusta mi-
gratoria capito se rencontrent dans une contree aride, avec un sol leger, sablonneux
expose a etre inonde par taches au cours des pluies; certains points favorables a
la constitution des foyers gregarigenes sont sur laterites dures.

Un autre Acridien, non typiquement migrateur, mais incontestablement gregaire,
Anacridium moestum malanorhodon Walk, joue lui aussi un role economique non ne-
gligeable en Afrique nord - equatoriale, dont il parait bien acquis (Morales, Roblot)
que I'aire permanente se trouve dans le Sahel, et dans la zone cotiere du Sahara
espagnol. Cette espece, sous certaines conditions, emigre parfois vers le sud.

81

D'autres Orthopteres, non gregaires, viennent egalement des zones desertiques
ou semi- desertiques pour s'attaquer aux cultures. G. 'de Lotto (1951) a constate
que, en Erythree, Pymateus viridipes depose ses oeufs dans les sols, incultes,
pierreux ou de sable compact; les larves, dont revolution dure environ 4 mois, res-
tent dans les zones semi-arides et ne sont done pas considerees comme nuisibles.
Mais les adultes, les femelles en particulier, qui vivent 7 mois, se deplacent vers
les cultures et defeuillent les Vignes, les Figuiers, les Cereales, etc.

Certains groupes d'Insectes, bien que phytophages, paraissent infeodes aux
regions desertiques; tels sont, chez les Coccides, genres Margarodes et Neomar-
garodes par exemple. La plupart des especes recoltees I'ont ete dans des etendues
arides ou elles vivent pendant une partie de leur existence sur les racines des
rares vegetaux xerophiles. Au cours de leur evolution elles ont un ou plusieurs
stades de resistance enkystes, bien connus sous le nom de perles de terre (ground -
pearl) ou perles du desert. Leur adaptation a des plantes cultivees peut se pro-
duire dans certaines circonstances. Ce fut le cas du Margarodes vitium Giard au
Chili et en Argentine. Cette cochenille, particulierement remarquable, a attire I'at-
tention du jour ou, vers 1890 ses attaques, tres serieuses, sur la Vigne furent con-
statees au Chili. Son origine fut alors recherchee et I'on constata qu'il s'agissait
d'un insecte polyphage precedemment observe dans des vallees incultes, loin de
toute Vigne. Les kystes etaient fixes en grand nombre sur 'les racines d'un arbris-
seau du genre Baccharis qui porte vulgairement le nom de "chirca" '(Valery Mayet)
(probablement Baccharis spinosa). Ce vegetal est un des rares qui se rencontrent
en lisiere de la pampa d'Atacama oil, actuellement, des periodes de 20 a 30 ans
peuvent s'ecouler sans que tombe la moindre averse!

Bien d'autres insectes phytophages trouvent dans les regions desertiques des
conditions favorables a leur puUulation: Kawiria Gabrieli Schiist., Tenebrionide
Platyopinae, sur le Saxaoul (Reymond) dans le desert de Kawir (Perse), Foleya
brevicomis Peyer, Tenebrionide Erodiidae, qui fut recolte par myriades dans I'Erg
occidental (Sahara) par A. Reymond, sur les epis de Drinn (Aristida pungens) en
fevrier - avril 1946 et 1947. II y a aussi une foule de Bostrychides, de Scolytides,
de Buprestides, qui passent de vegetaux indifferents a des plantes cultivees ou
seulement exploitees. Un Cerambycide, Polyarthron pectinicornis Fairm. puUule
dans les oasis sahariennes, en aout et septembre, aux depens du Dattier. Enfin,
une mention speciale doit etre faite des Scarabeides: Rhizotrogus, Annoxia, Phyl-
lopertha, Polyphylla, dont les larves trouvent un milieu permanent tres favorable a
leur developement dans les steppes circum- desertiques, et qui, sporadiquement,
commettent des depredations tres serieuses dans les plantations souvent tres eloig-
nees. C'est le cas de Polyphylla fulla qui, au Maroc, ravagent les jeunes Cedres
en coupant les racines (A. Reymond in litt.).

Des exemples comparables peuvent etre fournis pour des animaux superieurs
et H. Heim de Balsac a bien voulu m'en fournir qui interessent essentiellement les
regions sahariennes: alors qu'en Europe le grand Corbeau, Corvus corax, fuit
I'Homme, aux approches du Sahara, au contraire, il s'y est completement infeode et
la densite de I'espece y est meme fonction de I'importance de 1 'agglomeration. Le
vicariant desertique, Corvus ruficollis, se comporte de meme. II s'installe a proxi-

82

mite des campements, perchant sur les Chameaux et devorant les Tiques fixees sur
les parties genitales. Tous ces Corbeaux, de quelque espece qu'ils soient, sont de
grands destructeurs de Cereales et de Dattes.

En Afrique du Nord, les cultures de Cereales ont encore a souffrir des incur-
sions massives des Alaudides, tant sedentaires que migrateurs, et des Moineaux.
Ces derniers peuvent devenir un fleau pour les cultures: Le Maroc est particuliere-
ment alarme par leurs bandes (plus specialement de Passer hispaniolensis) qui de-
viennent migratrices ou erratiques et se repandent jusqu'au Sahara septentrional. A
Atar, le Moineau desertique, P. simplex et le Bengali, Oedemosyne cantaux, s'at-
taquent aux epis immatures de Mil, a tel point qu'il est necessaire d'entourer ceux-
ci de chiffons pour les preserver.

Parmi les Mammiferes il faut citer les Gerbillines, les Meriones qui, attires par
les cultures, deviennent, dans le sud-ouest marocain, commensales des habitations
et s'attaquent aux recoltes sur pied et engrangees et, meme, aux excrements hu-
mains. Les arganeraies ont a souffrir de I'Ecureuil Atlantocerus qui consomme les
fruits. Enfin, les Gazelles, Mouflons, Sangliers, causent des degats importants
aux jeunes plantations d'Opuntia inermis que Ton essaie de multiplier dans I'Anti-
Atlas et le pays Fekna. La plante est utile par ses fruits, mais les animaux sont
tres friands des raquettes qui constituent une reserve d'eau.

Ce qu'il importe de souligner, en regard de tous ces exemples qui pourraient,
aux especes pres, etre valables sous routes les latitudes, c'est que, nes dans des
regions ou les conditions de vie sont particulierement inclementes, ces animaux
semblent en etre devenus plus agressifs, done plus nuisibles, vis- a- vis des
plantes cultivees et des denrees qui sont elles-memes a la base de la vie humaine.

Malgre la brievete de cet expose, j'espere avoir reussi, a faire ressortir, I'in-
fluence, double et contradictoire, des deserts et semi-deserts sur les puUulations
nuisibles aux productions dont I'Homme tire sa subsistance. Mais il est un autre
aspect du probleme qu'il me semble impossible de passer sous silence: je veux par-
ler de Taction de I'Homme lui-meme dans la formation de ces zones arides et semi-
arides.

lei ce n'est plus I'aridite qui agit sur le comportement de I'Etre vivant, tout au
moins dans le sens direct qui nous occupe, mais c'est ce dernier, au contraire, qui
pese sur les possibilites du sol jusqu'a, parfois, creer le desert.

En effet, n'est- il pas normal pour les nomades et leurs betes, dans les regions
desertiques et semi- desertiques, de sejourner autour des points d'eau generalement
entoures d'un vegetation plus ou moins verdoyante? II en resulte un pietinement et
un pacage abusifs qui concourent a la disparition rapide de la couche vegetale et a
I'installation d'une zone sablonneuse et sterile s'etendant peu a peu au large du
point d'eau jusqu'a suppression complete de toute vegetation et impossibilite d'uti-
liser I'abreuvoir. Kachkarov et Korovine (Monod, 1942) citent I'exemple de I'Ari-
zona ou I'on a constate I'influence considerable du paturage excessif sur la steri-
lisation du sol: des espaces immenses couverts de Graminees et constituant des
paturages magnifiques ont ete transformes en fourres de Cactus, d'Agaves et d'au-
tres plantes epineuses inutilisables par le betail: un broutage exagere qui a

83

provoque la disparition progressive de la couvert'ore herbacee primitive devenue in-
capable d'eliminer la vegetation non comestible qui, finalement, I'a remplacee.

L'action nefaste de I'Homme s'exerca encore de bien d'autres manieres: de-
basement, feux de brousse, cultures intensives, etc. qui epuisent le sol et favori-
sent 1 'erosion. Mais c'est la une face du vaste probleme des zones arides et semi-
arides qui sort du cadre de cet expose. Je voulais seulement faire ressortir que,
s'il ne peut empecher certaines consequences de 1 'existence actuelle de regions
desertiques, I'Homme a non seulement le devoir de lutter centre ces consequences
mais, avant tout la responsabilite de ne pas concourir a en creer de nouvelles. De
plus, il lui incombe, tout d'abord, de fertiliser ces terres, dites semi -arides, qui
possedent encore quelques possibilites d'alimenter une vegetation, si maigre soit-
elle, avant de porter ses efforts sur celles qui en sont incapables pour quelque rai-
son que ce soit. Et une image me vient a I'esprit, qui est aussi un exemple: les
Hollandais 'repoussent la mer'pied 4 pied donnant des terres a I'agriculture mor-
ceau par morceau, ne s'attaquant a une part nouvelle que lorsque la precedente com
commence a repondre aux soins qui lui ont ete portes ... les deserts doivent etre
vaincus de la meme maniere, avec methode et tenacite, avec aussi une grande
patience!; alors certains fleaux des cultures — et je pense en particulier aux acri-
diens — ne trouveront plus les conditions qui favorisent leur puUulation, qui seule,
nous importe au point de vue economique et, en consequence, social.

References.

Balachowsky, A. 1952. Etude biologique des Coccides du Bassin occidental de la Medi-
terranee. Paris: Lechevalier, pp. 214.

Bodenheimer F.S. 1924. The Coccidae of Palestine. Bull. Inst. Agric. Nat. Hist. Tel Aviv,
1, 100.

Kachkarov, D. N., & Korovine, E. P. 1942. La vie dans les Deserts, edit, franc, par Th.
Monod Paris: Payot pp. 36l.

Lotto, G. de 1951. Osservazioni sulla biologia del Phymateus viridipes St. Riv: Agric.
subtrop., 45, 8- 18.

Mayet, V. 1895. Les cochenilles des Vignes du Chili. Rev. vitic. Paris, 1, 477-82; 2,
512- 16; 3, 557-62.

Morales Agacino, E. 1948. Algunos datos sobre la Langosta arboricola, Anacridium moes-
tum melanorhodon, en la zone meridional de Rio de Oro. Bull. Pathveg. Ent. agric, 16,
293-4.

Reymond, A. 1938. Resultats scientifiques d'un voyage en Asie centrale. Rev. Geogr.
phys. pp. 285.

Roblot, M. 1949. Etude sur Anacridum moestum subsp. melanorhodon. These d'Ingenieur
d' Agriculture aux colonies (dactylographiee). S.T.A.T. Nogent- sur- Marne.

Trabut, L. 1910. La defense contre les cochenilles ou autres Insectes fixes. Alger: Impr.
Agricole.

Vayssiere, P. 1926. Contribution a I'etude biologique et systematique des Coccides. Ann.
Epiphyt. 12, 197-382.

Vayssiere, P. 1951. Les bases ecologiques de la regeneration des zones arides (Introduc-
tion). Vn. int. Sci. biol. ser. B. No. 9, 5-9.

84

THE DESERT LOCUST AND ITS ENVIRONMENT

Dr B.P. Uvarov, C.M.G., F.R.S.
(Anti- Locust Research Centre, London)

The distribution area of the Desert Locust (Schistocerca gregaria Forsk.) coin-
cides with that of the hot Palaearctic deserts, although its swarm migrations may tem-
porarily extend beyond the limits of the latter, both southwards and northwards. From
the broad biogeographical point of view, therefore, the Desert Locust can be regarded
as a typical inhabitant of these deserts and it is of general interest to see to what an
extent the biology of the species is adapted to the desert conditions and, in particular,
whether such changes in the latter as may be induced by man, are likely to bring us
nearer to a solution of the Desert Locust problem which has a reasonable claim to be
regarded as the oldest problem of applied entomology.

The genus Schistocerca has a somewhat unusual geographical distribution, since
it includes some 80 species of South and Central America, with a few penetrating to
North America, while S. gregaria is the only species occurring in the eastern hemi-
sphere. It is fairly closely related to a South American swarming species, but cer-
tainly quite distinct from it.

Another feature of the genus Schistocerca is that, as far as is known, its Ameri-
can members are arboreal in their habits, and this, again, makes S. gregaria an apparent
exception and its partiality for deserts may suggest a great ecological and physiologi-
cal divergence from its congeners. However, when speaking of deserts, one should
bear in mind that perennial shrubs are an essential feature of many desert areas and,
in particular, of some sandy tracts where even something approaching sparse woods
can be found, where they have not been destroyed by man. Observations on the Des -
ert Locust, particularly in the adult stage, indicate its tendency for sitting on shrubs
or tall plants. This is particularly noticeable in swarms which normally roost for the
night on shrubs and trees. In this respect, therefore, S. gregaria has retained the
generic habit.

With regard to its food, while the Desert Locust feeds readily on a large variety
of plants including grasses (Bhatia, 1940), the latter play only a minor part in its diet
and the vast majority of its food -plants are either annual and perennial herbs, or
shrubs. In this respect, the Desert Locust offers a sharp contrast to the Migratory Lo-
cust (Locusta migratoria L.) which is predominantly a grass feeder (Kozhanchikov,
1950).

We see, therefore, that, ecologically, the Desert Locust cannot be regarded as a
typical insect of open desert, or of arid grasslands such as occur in the marginal des-
ert areas. In fact, leaving aside for the moment the occurrence of swarms which ex-
tend well beyond ecological barriers, the species is normally not encountered every-
where in the desert regions, but mainly in certain areas such as well overgrown sand
dunes on coastal plains, scrub belts along the beds of seasonal rivers and similar
habitats which represent 'ecological islands' in the desert.

85

The physiological requirements of the Desert Locust do not suggest a high degree
of adaptation to what is usually understood by desert conditions. The female locust
lays eggs preferably in sand, but not in dry sand. Sand which is dry on the surface
but m9ist underneath, is suitable. Other loose soils are also acceptable, provided
they are moist. In the absence of moisture, eggs may be laid on the surface of the soil
where they perish. A developing egg of the Desert Locust has very high humidity re-
quirements, since it needs to absorb more than its own weight of water for successful
development (Shulov, 1952). Therefore, the soil round the egg must preserve suffi-
cient free moisture for some 12- 15 days of the incubation period.

When young locusts (hoppers) hatch, they must have sufficient green food, usually
tender annual plants which rapidly spring up in sandy desert areas after a shower of
rain. Excessive heat and dryness during the hopper development, which takes 5-6
weeks, has been known to cause wholesale mortality of hoppers.

Adult Desert Locust are known to be able to survive for many months in condi-
tions of dryness, but for their sexual maturation they require either succulent green
food or high air humidity (Hamilton, 1950; Norris, 1952); in this respect, desert con-
ditions are definitely unfavourable for reproduction.

With regard to activity, it has been thought before that adult locusts, particularly
in swarms, would be most active in intensely dry and hot conditions and this would
cause them to migrate to more favourable habitats, but recent observations (Waloff,
1952) tend to dispel this idea, since flight activity of swarms appears to be more per-
sistent at higher air humidities than at the lower.

It would appear, on the whole, that the Desert Locust is far from well adapted to
general desert conditions, and the question arises how can the species not merely
survive, but be able, from time to time, to multiply in fantastic numbers.

The answer is to be sought in the fact that the widespread conception of a desert
is too generalised. It covers a great variety of landscapes, which provide desert ani-
mals with a wide range of habitats, some of them offering very favourable conditions
of life. In addition to this variety of conditions in space, there is a great seasonable
variation in all life conditions : a truly desert, lifeless area becomes covered by lush
annual vegetation almost immediately after a shower of rain.

The existence of such favourable ecological islands is the essential condition for
the existence of the Desert Locust. Since, however, many of such islands are only
ephemeral, they are clearly unable to support a stationary locust population. On the
other hand, an insect which is capable of moving from one favourable area to another,
has an excellent chance of survival, and the continuous existence of S. gregaria in the
desert regions is closely bound up with its migratory or, rather, nomadic habit.

Most biologists would be content to accept the beneficial value of migration as a
sufficient explanation of the migratory pattern, but it is possible now to offer a some-
what deeper analysis of this phenomenon.

Studies of Desert Locust migrations in relation to weather factors (Waloff & Rainey,
1951; Waloff, 1952) have shown that swarm displacements are closely linked up with
weather dynamics, and Rainey, (1951) put forward a well - documented hypothesis that

86

major swarm movements take place towards zones of convergence of air- masses which
are often associated with precipitation. The result is that locusts and rain are likely
to arrive in an area together, an occurrence which has been frequently noticed, before
its mechanism was understood. The importance of this for the maintenance and multi-
plication of the Desert Locust is obvious and appearances of large locust populations
after an area has received a shower of rain loses its element of mystery.

The coveregence hypothesis accounts also for the regular seasonal movements of
locust swarms between the areas receiving winter- spring rainfall and those subject to ■
monsoon rains (Waloff, 1946; Donnelly, 1947; Davies, 1952; Fortescue- Foulkes,
1952). The value of such movements for the species is obvious.

If we remember the ability of the adult Desert Locust either to mature and lay
eggs soon after becoming adult, or, in the absence of suitable conditions, to delay the
maturation for several months, the risk of losses of its popiilation through unfavourable
climatic conditions appears less serious than one might conclude from its physiologi-
cal requirements. There is no doubt that the instability of the environment and, in par-
ticular, the unreliability of rainfall in desert regions make the life of the Locust very
precarious, but its mobility, linked up as it is with weather dynamics, helps it to over-
come its physiological handicaps.

While these points are of general biological interest, they also have an important
practical bearing. The Desert Locust is unquestionably one of the most important in-
sect pests and its periodical swarming and invasions of fertile lands have always been
associated with the neighbouring deserts, which were blamed as the source of swarms.
Investigations of the last 20 years have not yet solved the problem of where and how
exactly swarms arise from scattered locust populations, but there is enough evidence
to state, in a general way, that swarm formation cannot occur in areas with persistent
extreme desert conditions. The maintenance of locust populations in the desert de-
pends on the existence of favourable ecological islands, be they permanent or season-
al, and such areas are dependent either on fairly regular seasonal rainfall or on run-
off of rain water from highlands along seasonal river-beds, or, finally, on artificial
irrigation. It is the latter which deserves our particular attention. There are already
some observations suggesting the importance of irrigation and cultivation for creating
or expanding habitats favourable for the Desert Locust. The land development scheme
at Abyan, Aden Protectorate, has lately become an area with a fairly persistent popu-
lation of Desert Locust and repeated efforts are required to keep it under control. Ex-
tensive cultivation areas in the Tokar delta on the Red Sea coast of the Sudan form a
classical locality for the Desert Locust and it has to be kept under regular observation
and control. In Tripolitania, the breeding by invading swarms in 1946 was mainly con-
centrated on reclaimed sand-dunes immediately adjoining cultivation (Brown, 1947).
On the Red Sea coastal plains of Eritrea, Saudi Arabia and Yemen, locusts are normal-
ly found in the sandy deltas of seasonal rivers where native cultivation is extensive,
if sporadic. Even in the heart of the desert, in the Fezzan, considerable breeding
populations of locust, were found in spring 1952 in alfalfa cultivated on run- off water
from the hills (K. Guichard, unpublished); and a similar observation was made in
Mauretania (Bruneau de Mire, 1952).

87

These examples suggest that reclamation of desert areas, which is most likely to
occur where the existing conditions already tend to create favourable locust habitats,
may not be an unmixed blessing, by making such habitats more permanent and provided
with a regular food supply for locusts. So far, this has happened only on a limited
scale, e.g. in Abyan and in Tokar, but if desert reclamation is to bring substantial
benefits, it has to extend to many more and to much wider areas, and the effects on
Desert Locust populations may well assume very serious proportions. Somewhat para-
llel cases are not unknown. The Rocky Mountain grasshopper (Melanoplus mexicanus
Sauss) can normally produce one annual generation in Arizona where it was not a seri-
ous pest until extensive irrigation and cultivation of alfalfa created a stable favour-
able habitat, making it possible for the grasshopper to produce several generations a
year; regular chemical control keeps the pest within limits, but at a considerable an-
nual cost (Uvarov, 1948).

It should not be concluded, of course, that desert reclamation is undesirable be-
cause it may encourage the locust, but this danger must be borne in mind when desert
development schemes are considered. It should be possible to provide safeguards a-
gainst undesirable consequences of irrigation and cultivation, but the need for such
safeguards must be realised in time.

The above considerations refer to reclamation of the desert itself, but the effects
of extension of cultivation in the marginal areas must also be mentioned with reference
to locust danger. At present, the possibility of keeping the Desert Locust under per-
manent control still remains theoretical, and extensive anti- locust campaigns are
necessary to prevent devastation of fertile regions by invading swarms. The strategy
of these campaigns aims at achieving maximum destruction of locusts at a season when
they are breeding in desert areas. To give a recent example, swarm breeding by the
Desert Locust in spring 1952 occurred over some 10,000 sq. miles of Arabian deserts;
large mechanised forces had to be used to control the infestation and some 9,000 hop-
per bands were exterminated. Swarms which would have arisen from these bands
would have spread over the fertile cresent of Middle East countries north of Arabia,
but not a single swarm was allowed to develop and crops were saved, although at the
cost of great efforts and very high expenditure. If crops were closer to the breeding
areas, it would have been extremely difficult to prevent their damage by hopper bands,
and the deeper cultivation penetrates in the desert, the greater are the chances of
serious losses during plague periods. Again, one should not argue against the exten-
sion of marginal agriculture, but it would be wise to realise the danger inevitably con-
nected with it.

The main general conclusion which may be suggested by considering the Desert
Locust in relation to desert reclamation, is that while the latter is certainly able to
increase crop producing areas, it would also increase the risks of losing the crops,
unless repercussions of reclamation on certain members of desert fauna, such as the
Desert Locust, are realised clearly and before it is too late.

The Desert Locust was taken in this paper mainly as a better known example of
desert fauna. There are many other members of that fauna which are also associated
with favourable ecological islands in the desert. An artificial increase of the number

88

and extent of such islands through reclamation will inevitably create a number of new
entomological problems, perhaps not as serious as that of the locust, but still deserv-
ing to be anticipated rather than merely overlooked or ignored.

P.eferences.

Bahatia, D. 1940 Observations on the biology of the Desert Locust (Schistocerca gregaria
Forsk.) in Sind-Rajputana desert area. Indian J. Ent. 2: 187-192.

Brown, E.S. 1947 The distribution and vegetation of egg -laying sites of the Desert Locust
(Schistocerca gregaria Forsk.) in Tripolitania in 1946. Bull. Soc. Fouad I Ent. 31: 287- 306,
4 pis., 2 figs, 3 tables.

Bruneau de Mire, P. 1952 Rapport de prospection en Mauritanie Orientale (A.O.F.) Bull. Off.
not. anti- acrid. 3: 1 pL, 1 fig, 5 tables.

Davies, D.E. 1952 Seasonal breeding and migrations of the Desert Locust {Schistocerca gre -
garia Forskal) in north-eastern Africa and the Middle East. Anti- Locust Mem. 4: 1-56, 13
maps, 2 figs.

Donnelly, U. 1947 Seasonal breeding and migrations of the Desert Locust {Schistocerca gre-
garia Forskal) in western and north-western Africa. Anti- Locust Mem. 3: 1-46, 19 maps,
Ifig.

Fortescue- Foulkes, J. 1953 Seasonal breeding and migrations of the Desert Locust {Schisto-
cerca gregaria Forskal) in south-western Asia. Anti- Locus t Mem. 5: 1-36, 14 maps, 1 fig-
Hamilton, A.G. 1950 Further studies on the relation of humidity and temperature to the develop-
ment of two species of African locusts — Locusta migratoria migratorioides (R.&F.) and
Schistocerca gregaria (Forsk.). Trans. R. ent. Soc. Lond. 101: 1- 58, 34 figs.

Kozanchikov, I.V. 1950 Fundamental features of food specialisation in the Asiatic locust.
Izv. Akad Nauk SSSR 4: 73-86, 4 figs.

Norris, M.J. 1952 Reproduction in the Desert Locust {Schistocerca gregaria Forsk.) in relation
to density and phase. Anti- Locust Bull. 13: 1-49, 8 figs, 14 tables.

Rainey, R.C. 1951 Weather and the movements of locust swarms : a new hypothesis. Nature,
Lond 168: 1057-1060, 2 maps.

Siulov, A. 1952 The development of eggs of Schistocerca gregaria (Forskal) in relation to
water. Bull. ent. Res. 43: 469-476, 8 tables.

Uvarov, B.P. 1947 The grasshopper problem in North America. Nature, Lond 160: 857-859.

Waloff, Z. 1946 Seasonal breeding and migrations of the Desert Locust {Schistocerca gregaria
Forskal) in eastern Africa. Anti -Locust Mem. 1: 1-74, 30 maps, 2 figs.

Waloff, Z. 1953 Flight in Desert Locusts in relation to humidity. Bull. euL Res. 43: 575-580.

Waloff, Z. & Rainey, R.C. 1951 Field studies on factors affecting the displacements of the
Desert Locust swarms in eastern Africa. Anti- Locust Bull. 9: 1-50, 1 map, 8 figs.

89

SUR L'ORIGINE ET LE DEVELOPPEMENT DES INSECTES NUISIBLES
AUX PLANTES CULTIVEES DANS LES OASIS DU SAHARA FRANCAIS

Dr A. S. Balachowsky
{Chef de Service A I'lnstitut Pasteur de Paris)

Introduction

Au Sahara, 1 'agriculture se concentre autour des points d'eau, car aucune
plante ne serait cultivable sans irrigation; ces points d'eau, lorsqu'ils sont suffi-
samment importants, constituent les oasis.

Sur I'ensemble du territoire saharien francais, les oasis sont tares et clairse-
mees, on peut les comparer a des ties isolees dans un vaste ocean (oasis d'El-
Golea, d'ln Salah, de Djanet, de Mauritanie, etc.) ou parfois a un 'archipel' com-
prenant des petits ilots rapproches (oasis du Touat, du Gourara, de la Saoura-
Zouzfana, Koufra, etc.) (Le terme d"archiper a ete utilise par P. de Peyerimhoff
pour designer I'ensemble des oasis de Koufra isoles dans le desert libyque). Cet
isolement n'exclut pas, comme nous le verrons plus loin, des rapports tres etroits
existant sur le plan faunistique, entre les oasis et le milieu desertique environnant.

Les oasis sont cultivees par des populations sedentaires qui sont pour la plu-
part d'anciens esclaves noirs ou 'harratines** originaires du Soudan dont les condi-
tions sociales et materielles n'ont d'ailleurs guere change depuis leur 'liberation'
mais, les populations nomades, presque toutes d'origine blanche (berberes, chleuhs,
touaregs, maures, chambaas) sont le plus souvent proprietaires des terrains et des
palmeraies. Le role des nomades est loin d'etre negligeable dans la vie des oasis,
car ce sont eux qui entretiennent les contacts entre les localites tres eloignees les
unes des autres, creent les echanges a travers le desert et transportent ainsi a des
distances considerables des rejets de dattiers (djebars), des plantes nouvelles
sous forme de grains, graines, boutures, etc. C'est de cette maniere que le plus
souvent les insectes nuisibles phytophages ont ete vehicules puis se sont accli-
mates dans les differentes regions habitees par I'homme au Sahara. Ceci est par-
ticulierement vrai pour les Cochenilles du palmier- dattier. (Cf. infra).

Les plantes cultivees en oasis sont toutes ou presque toutes introduites (y
compris le palmier- dattier) et la plupart d'entre elles n'ont pas une origine afri-
caine. Aucune plante cultivee ne possede une origine strictement saharienne, ex-
cepte quelques varietes de ble cultivees au Fezzan et a Koufra qui se sont differen-
ciees in situ mais dont la souche initiale doit etre recherchee ailleurs. Le palmier
-dattier (Phoenix dactylifera L.) constitue la culture essentielle des oasis ou il oc-
cupe actuellement pres de 100,000 ha au Sahara francais, L'importance des oasis
s'etablit suivant le nombre de dattiers en production qui s'y trouvent (ensemble des
palmeraies eViropeennes de I'oued Rhir: plus de 2,000,000; Ouargla: 1,000,000;
Figuig: 350,0000; Colomb- Bechar: 100,000; Ihrir et Ahrar (Tassili): quelques
centaines). II est evident que le nombre des palmiers est en rapport etroit avec les

* Les 'rachetes'.

90

possibilites d'irrigation. Sous les palraiers pousse toute une strate de cultures
fruitieres et vivrieres introduites des regions subtropicales et temperees telles cpie
le Grenadier, la Vigne, les arbres fruitiers, le Figuier, les Citrus, les Cereales
(ble et orge principalement), le Mais, le Sorgho, le Mil (Sud- Saharien), diverses le-
gumineuses (Feve, Luzerne, Haricot, Pois chiche, Oignon) et des Legumes varies.
On y trouve aussi le Tabac (varietes indigenes), le Cotonnier (vivace indigene), le
Kif, le Piment, des plantes a condiment, etc. Auguste Chevalier considere le Sa-
hara cbmme un centre d'Agriculture primitive. II est bien difficile d'admettre cette
theorie. Rien ne prouve en effet qu'il ait jamais existe au Sahara un centre d'Agri-
culture primitive (neolithique ou prehistorique) (archeologique de A. Chevalier)
comparable a ceux de I'Egypte, de I'Abyssinie, de I'Afganistan, de la Mesopotanie,
de rinde, de la Chine ou du Mexique. Tous les vegetaux. cultives au Sahara y ont
ete importes a une epoque relativement recente et il n'y existe aucune tradition
paysanne. Les travailleurs de la terre sont presque tous des descendants d'an-
ciens esclaves noirs venus du Soudan, soumis au travail force par des populations
blanches conquerantes (arabes, berberes, maures, peulhs, etc.) essentiellement no-
mades qui ont conserve jusqu'a nos Jours une veritable repulsion pour le travail du
sol. Tous les vestiges humains du Sahara, notamment ceux si nombreux du neoli-
thique recent, nous renseignent sur I'existence dans toute I'etendu desertique ac-
tuelle, de populations guerrieres ou semi- guerrieres qui vivaient de chasse, de
peche ou de coUecte de graines de vegetaux spontanes. Aucune trace d'Agricul-
ture n'y apparait. Le Sahara ne figure pas non plus parmi les differents centres
d'origine des plantes cultivees cites par Vavilow, 1949- 50, Chronica botanica, 13.

II est evident que les conditions ecoclimatiques qui regnent dans les oasis
sont differentes de celles du desert environnant. L'humidite y est nettement plus
elevee et plus constante (du fait de 1 'irrigation), la temperature plus egale, 1 'inso-
lation moins grande, notamment pour la strate cultivee sous les dattiers, le sol
plus meuble et plus riche en matieres organiques du fait de la culture. Ces condi-
tions permettent done le maintien d'une flore et d'une faune differentes de celles
existant dans le desert environnant; cependant, comme nous le verrons plus loirx, a
part quelques cas tres particuliers, I'influence du climat desertique proprement dit
elimine, meme des oasis, un tres grand nombre d'insectes nuisibles.

Les plantes cultivees dans les oasis hebergent toute une serie d'insectes nui-
sibles phytophages, mais I'interet economique de ceux-ci est tres inegal, il varie
d'ailleurs d'une oasis a 1 'autre, et cette faune ne revet nulle part un caractere de
rigoureuse homogeneite.

I. Les Differents Types D'insectes Phytophages des Oasis.

La faune des insectes nuisibles peuplant les oasis sahariennes est pauvre et
degradee si on la compare a celle existant ou nord et au sud du Sahara et vivant
sur des plantes cultivees similaires. D'autre part I'inventaire des especes nuisi-
bles a ete peu pousse et nous ne possedons encore que des renseignements frag-
mentaires sur 1 'ensemble de cette faune, etablis le plus souvent par des rapports
administratifs sans grande valeur scientifique, ou par des observations rapides ef-
fectuees par des voyageurs de passage. Les etudes detaillees et coordonnees

91

manquent, si bien qu'il est encore difficile d'avoir une idee precise sur la compo-
sition de la faune des especes nuisibles qui peuplent les oasis. Ce travail preli-
minaire n'est qu'amorce, un bon nombre de localites sahariennes n'ont encore ja-
mais ete visitees par les entomologistes.

Dans la mesure des connaissances actuellement acquises, on peut distinguer
parmi les insectes phytophages nuisibles vivant dans les oasis deux types d'ele-
ments d'origine nettement distincte comprenant, d'une part, les especes introduites
(la plupart cosmopolites) et, d'autre part, les especes adaptees d'origine nettement
saharienne.

(a) Elements introduits.

Ces elements sont constitues par une faune heterogene d'especes accidental -
lement introduites par I'homme a une epoque plus ou moins recente. lis compren-
nent principal ement des insectes cosmopolites vivant dans les denrees alimen-
taires stockees (grains, graines, farines, dattes, fruits et legumes sees, etc.) et
que Ton retrouve dans tous les pays; leur puUulation est generalement favorisee
pax de tres mauvaises conditions de conservation existant dans les entrepots saha-
riens (ce probleme preoccupe le F.A.O. qui a delegue une mission en Libye pour
etudier les moyens de protection des denrees alimentaires stockees contre les in-
sectes).

Parmi les especes les plus representatives de ce groupement, il convient de
citer Oryzaephilus surinamensis L. et O. mercator Fauv.* Carpophilus hemipterus
L. qui vivent dans les dattes; Tribolium confusum Duv., Sitophilus orizae L. dans
les farines et graines de cereales, pates alimentaires et diverses matieres amyla-
cees, les Dermestes, notamment D. frischi L., dans diverses matieres organiques.
Le Scolyte des noyaux de dattes Coccotrypes dactyliperda L. n'est pas saharien,
il vit dans les dattes immatures du Tell algerien et dans les graines d'autres pal-
miers, notamment de Phoenix canariensis. Il est tres commun dans les jardins de
la region mediterraneenne (Nord et Sud). Parmi les Lepidopteres, les Ephesthia
(farine) et les Myelois (dattes) sont les plus frequents. M. decolor Z. serait plus
specifiquement saharien et contaminerait les dattes mures sur les arbres pour se
developper ensuite dans les entrepots alors que M. ceratoniae Z. {— phoenicis
Dun.) est une espece cosmopolite vivant sur tous les fruits desseches (dattes, ca-
roubes, figues, abricots, etc.) (Real, 1948).

On pourrait ajouter a cette liste beaucoup d'autres especes d'interet secon-
daire.

En realite il existe relativement peu de vrais phytophages nuisibles d'origine
extra- saharienne introduits dans les oasis, et ceci est du en grande partie au clim-
ati saharien caracterise par une secheresse atmospherique et des maxima de tem-
perature tres eleves en ete, de grandes fluctuations journalieres et saisonnieres
empechant le maintien de nombreux phytophages originaires des regions non deser-
tiques du globe. D'autre part, les introductions par caravane necessitent des
transports de longue duree et le maintien d'une nourriture vivante fratche pour les

* Ce dernier apparait comme une simple forme du precedente (P. de Peyerimhoff).

92

insectes transportesj or ces conditions ne peuvent guere se trouver realisees que
pour les especes vivant sur des rejets de dattiers, des graines. des grains, des
bois vivants, des fruits desseches et autres denrees alimentaires stockees. Tous
les phyllophages, radicicoles floricoles, cecidogenes, mineurs de feuilles ou de
tiges, suceurs de seve, ne peuvent supporter les voyages de longue duree et se
trouvent elimines au cours des longs transports par caravane. II va de soi que les
oasis ayant un contact plus etroit et plus constant avec la civilisation, notamment
celles situees le long des lignes de chemin de fer (oasis du Nord) ou les grandes
pistes sahariennes automobiles, sont plus soumises aux introductions nouvelles
que les oasis eloignees, situees en dehors de toute voie de communication frequen-
tee. De meme, aujourd'hui, I'avion favorise les nouvelles introductions d'insectes.
Aussi le nombre des especes extra- sahariennes introduites est proportionnelle-
ment plus eleve dans les oasis de la bordure nord du Sahara en particulier les oa-
sis de la rive Sud du Sahara en contact avec les zones de cultures permanentes
soudanaises ou Ton trouve deja bien fixes divers elements cosmopolites n'existant
pas encore ailleurs dans le desert: presence de la Mouche des fruits (Ceratitis
cttpitata Wied. ) a Biskra, d'lcerya Purchasi Mask, a Biskra et Laghouat, de la
Teigne du poireau {Acrolepia assectella Z.), du Puceron noir des feves {Aphis fa-
bae Scop.), de divers Pseudococcus dans presque toutes les oasis Nord saharien-
nes et presahariennes du Sud de 1' Atlas. II existe des Pseudococcus indigenes au
Sahara, notamment Planococcus tuaregensis Balachw. que j'ai decrit du Tassili
(Amais) vivant sur Ficus scdicifolius var. teloukat Batt. et Trab.

Les Cochenilles du palmier- dattier. L'origine exacte du palmier- dattier
(Phoenix dactylifera L.) reste encore imprecise mais la majorite des botanistes
(dont Rene Maire) sont d' accord pour considerer la zone desertique orientale (Iraq,
Mesopotamie)comme sa patrie originelle. Sa culture au Sahara remonte a une epo-
que fort ancienne et pour certaines oasis du moins, bien anterieure a 1 'invasion
arabe.

Le dattier est parasite au Sahara par trois Cochenilles dont deux (Parlatoria
Blanchardi Targ. et Phoenicococcus Marlatti Ckll.) lui sont specifiques. P. Blan-
chardi est seul reellement nuisible; c'est une espece strictement desertique qui ne
peut se maintenir en dehors du climat saharien ou subsaharien. Son aire de repar-
tition coincide etroitement avec la zone de maturation naturelle des dattes. Sur le
littoral mediterraneen, ou le dattier est frequemment cultive comme arbre d'orne-
ment, on ne trouve la Cochenille nuUe part; il en est de meme pour le Tell et les
Hauts- Plateaux. A I'exception d'Inkermann et d'Orleansville dans la vallee du
Cheliff (Algerie) ou des dattiers contamines originaires de Biskra ont ete introduits
en 1928 et plantes le long de la gate (A. Perrin). Le climat de la vallee du Cheliff
est un des plus chauds du Tell algerien et caracterise par des maxima tres eleves
en ete (+ 40*^); ces conditions exceptionelles pour le Tell ont permis vraisembla-
blement le maintien permanent ou subpermanent de P. Blanchardi. La meme remar-
que s 'applique pour la rive Sud du Sahara ou, hors de la zone saharienne, le dattier
est indemne de Cochenilles (region de Garoua- Maroua, Nord- Cameroun et depres-
sion du Tchad). P. Blanchardi existe dans I'Air, I'Adrar des Iforas et dans cer-

93

taines oasis du Tibesti (Gourmeur). Nous ne I'avons pas trouvee sur les dattiers
de la region du Tchad ou elle a ete signalee dans le Borkou (mission antiacridienne
de 1934.

Au Sahara proprement dit, la Cochenille existe partout excepte dans les oasis
occidentales. Elle fait defaut encore dans la plupart des oasis rnarocaines (Bani -
Draa — Tafilalet), dans celle de la Saoura- Zouzfana, du Gourara (y compris celles
du Tinerkouk), du Taouat a 1 'exception de quelques localites ou son introduction
estrecente (Colomb- Bechar: 1920; certaines oasis du Touat:. 1912; Tata-Maroc:
1945). Cette aire de repartition confirme la theorie de I'origine orientale du dattier;
sa progression de I'Est vers I'Ouest au Sahara ayant ete plus rapide que celle de
la Cochenille qui a suivi son hote avec plusieurs siecles ou plusieurs dizaines de
siecles de retard. Mais, meme dans la zone d'invasion ancienne, certaines oasis
particulierement isolees et sans contacts avec la civilisation, restent encore in-
demnes de Cochenilles, comme c'est le cas pour la vallee d'Ahrar dans le Tassili
N'Ajjer. II est hors de doute que le role de rhomme fut preponderant dans ces in-
troductions et toute idee 'd'invasion naturelie progressive' de proche en proche
doit etre exclue etant donne la discontinuite de la repartition geographique du dat-
tier dans le Sahara et la specif icite rigoureuse de P. Blanchardi.

En ce qui concerne P hoenicococcus Marlatti Ckll., bien que son origine deser-
tique ne puisse etre mise en doute, son aire de repartition est beaucoup plus vaste,
car cette Cochenille a suivi le dattier un peu partout ou il a ete introduit, y com-
pris dans les nombreux pares, jardins, avenues, de la region mediteraneenne (Alger,
Tunis, Antibes, Elche, Palerme, etc.). P. Blanchardi et Ph. Marlatti ont ete intro-
duits aux Etats-Unis (Arizona, Californie du Sud) avec des djebars de 'deglet nour*
originaires du Sahara algerien et tunisien. Apres 30 annees d'efforts, les ameri-
cains ont elimine P. Blanchardi de leur territoire: Boyden, B.L. Eradication of Par-
latoria date scale in the United States {U.S. Dept. Agric. Mix. publ. No. 433 Wash,
D.C. 1941). Enfoncee dans les gaines foliaires, a I'abri de la lumiere et de I'inso-
lation, cette espece se trouve dans des conditions microclimatiques totalement dif-
ferentes de celles de P. Blanchardi, localise sur le feuillage et soumis directement
a une intense insolation. On est encore mal renseigne sur la 'qualite' de la lu-
miere du desert, la meme remarque s'applique pour les autres regions de I'Afrique.
II s'agit cependant la d'un facteur ecobiologique d'une importance considerable.
Nous Savons seulement que la lumiere du desert est riche en radiations appartenant
a la zone droite du spectre (bleu a ultra-violet). Cet habitat suffit a demontrer les
possibilites d'adaptation de P. Marlatti a des climats non desertiques.

Enfin une troisieme espece a ete signalee sur le dattier bien que sa presence
y paraisse accidentelle, c'est Pseudaspidoproctus hyphaenicus Hall, Margarodidae
decrit d'Egypte sur le palmier- doum (Hyphaene thebaica) (Hall, W.J., 1926). Cette
Cochenille est repandue dans diverses oasis d'Egypte et elle a ete trouvee dans
celle de Bendleia dans le Fezzan sur palmier- dattier, par F. Bernard (Rungs C,
1944).

Le palmier- doum est d'origine tropicale, il fait partie de la flore sahelienne
mais il a ete cultive autrefois au Sahara pour ses fruits et se retrouve a I'etat de

94

pieds isoles dans quelques oasis du Fezzan et du desert egyptien. Chevalier le
considere comme une relique de I'agriculture au Sahara ou sa presence doit etre
consideree comme tres ancienne et a sans doute precede celle du dattier. A notre
avis, la presence du Doum au Sahara constitue une relique d'un passe plus humide,
de i'epoque ou la flore sahelienne actuelle avait une extension continue et homo-
gene presque jusqu'en Afrique du Nord, a travers le Sahara actuel (quaternaire re-
cent). Le passage de la Cochenille du Doum.au Datteir s'est fait certainement in
situ, lorsque ces deux plantes ont ete en contact dans les oasis. II existe une 4^
espece specifique du palmier- dattier decrits de Mesopotamie, Asterolecanium phoe-
nicus Ram. Rao. EUe n'a pas penetre jusqu'ici en Afrique et son role economique
parait neglige able.

{b) Elements adaptes.

lis sont constitues par des especes phytophages saharienne vivant normale-
ment aux depens de la flore spontanee du desert. lis preexistaient done a la crea-
tion des oasis meme, mais celles- ci leur ayant apporte des conditions de vie plus
favorables (vegetation plus abondante, humidite plus reguliere, sol plus meuble),
leur pullulation a ete favorisee dans ces stations par la culture et il y a eu un phe-
nomene d'attraction. Cependant tous ces elements se retrouvent dans le desert pro-
prement dit en dehors des oasis, principal ement dans les lits d'oued, autour des
gueltas, dans les canons ou une humidite plus elevee se maintient en permanence.
Parmi ces collections d'insectes, il y a lieu de distinguer d'abord les Phytophages
polyphages a regime varie representes principalement par des especes aux moeurs
radicicoles ou subradicicoles appartenant a differents ordres ou families d'insectes:
Orthoptera, Scarahaeidae, Noctuidae, etc. En dehors de ces types on en trouve
d'autres a regime plus strict, specifique ou subspecifique, qui vivent dans le de-
sert aux depens de plantes de la meme famille botanique que celles cultivees en
oasis. Enfin, il existe une troisieme categorie d'elements particulierement interes-
sants vivant normalement sur la flore spontanee et qui se sont adaptes a la flore
cultivep lorsque celle- ci est apparue dans les oasis. Ces passages constituent de
veritables exemples d'allotrophie et demontrent d'une maniere suggestive le pro-
cessus de formation d'especes nuisibles aux depens de types sauvages consideres
jusqu'ici comme economiquement indifferents.

(i) Phytophages polyphages des oasis. — Nous passerons ici volontairement sous
silence tout ce qui se rapporte aux Acridiens migrateurs, le probleme acridien
n'etant pas strictement saharien. On trouve par centre dans les oasis de nom-
breux Orthopteres polyphages et parmi ceux- ci, il convient de citer divers Gryl-
lides et Gryllotalpides, notamment les Courtilleres (Gryllotalpa gryllotalpa L.
et Grylotalpa ajricana Beauv. ) et un gros grillon principalement repandu dans
les oasis du Sahara central et oriental, Brachytrypes megacephalus Lef.

Les Courtillieres devastent les jardins irrigues dans les oasis; on trouve dans
le Nord principalement G. gryllotalpa L. (grpsse Courtilliere) alors que dans le
Nord comme dans tout le reste du Sahara ainsi que sur I'ensemble du continent
africain on trouve G. africana Beauv. (petite Courtilliere qui a des moeurs pres-
que identiques(G. gryllotalpa, bien que d'origine palearctique, penetre profon-

95

dement dans le Sahara, Pasquier signale sa presence au Fezzan.). Ces espe-
ces se retrouvent dans les lits d'oueds sablonneux non cultives, autour des
gueltas dans les terrains humides et les sols legers, operant de la meme ma-
niere que dans les oasis. Quant au Brachytrypes, tres repandu dans les oasis
du Fezzan, c'est un tres gros grillon qui creuse de profondes galeries et devote
les plantes durant la nuit. (Pasquier 1951). Get insecte est egalement repandu
dans le Sahara soudanais. II existe en Tunisie, en Algerie (B8ne) et en Sicile
ou il constitue une 'relique' tropicale; signale des oasis de Touggourt et
Ouargla (Dr Jacquemin). On trouve egalement dans les oasis quelques Orthop-
teres se rencontrant normalement dans les lits d'oueds desseches sur des
plantes sauvages, principal ement les Graminees, et qui, dans les lieux cultives,
devastent le feuillage du Mil et du Mais, comme c'est le cas pour Euprepocne -
mis ploTOns Charp., acridien largement repandu sur le territoire africain, y com-
pris le Sahara et pour Poekilocerus hieroglyphicus Klug. Ce 'catoue' saharien
et soudanais de couleur jaune, aux ailes orange vif, est considere par divers
auteurs comme specifique du Calotropis procercu Au Tassili n'Ajjers nous I'a-
vons observe dans les lits d'oued, devorant des touffes de graminees alors que
les Calotropis faisaient completement defaut. Zonocerus variegatus L. le
'Catoue' d'Afrique tropicale ne depasse pas au Nord de la zone sahelienne.
Les larves de Scarabaeidae, notamment des Melolonthinae, Cetoniinae et Dynas-
tinae, se rencontrent frequemment dans les cultures sahariennes. Parmi les
premiers qui seuls sont veritablement phytophages et radicicoles, il y a lieu de
citer les Rhizotrogus (s.l.), mais ce genre si richement represente en Berberie
(64 especes) ne penetre guere dans le desert; il se rencontre seulement dans
les oasis septentrional es en bordure de la steppe. Comme I'a fait si bien res-
sortir P. de Peyerimhoff (1945), les Rhizotrogus nord- africains, presque tous
endemiques, sont en realite des especes des hauts plateaux algero- tunisiens
ou les ravages de leur larves s'exercent intensement parmi les cultures de Ce-
reales. Peu d'especes penetrent dans le desert proprement dit et le groupe se
rarefie aussi a I'Est ou il ne depasse par la Cyrena'i'que; aucune espece n'est
connue d'Egypt. Le genre est egalement assez pauvrement represente a 1 'Quest
(Maroc).

Les Dynastinae sont surtout representes par des especes ruderales vivant dans
le terreau et principalement dans celui qui s'accumule a la cime des palmiers a
la base des gaines foliaires. On trouve ainsi dans toutes les oasis Phyllog-
natus excavatus Forst. ('doudd' ou 'doudda' des Arabes) signale bien souvent a
tort comme nuisible; c'est beaucoup plus (dans les oasis) une espece detriti-
cole que reellement phytophage. Dans d'autres regions, les larves de Phyllog-
natus sont nettement phytophages radicoles. Cf. Balachowsky, A. et Mesnil, L.
1936. Les insectes nuisibles aux plantes cultivees, t. II, p. 659, Paris, De
meme, les Pentodon (P. deserti Heyden et P. bispinosus Kunt.) existent dans
tout le Sahara; leurs larves sont communes parmi les cultures des oasis (Hog-
gar, Fezzan, Koufra, Oued Rhir, Djerid, etc.). Files frequentent surtout les
terres riches en matiere organique et detruisent les plantes en les coupant au-
dessus du collet. Leurs degats sont toujours sporadiques et isoles.

96

Parmi les Cetoniinae il convient de citer les larves de Pachnoda Savignyi G. et
P. qui existent dans le Sahara central et soudanais (Tassili, Fezzan, Koufra)
vivant comme Phyllognatus excavatus dans le terreau de la cime des Dattiers:
cet habitat n'est certainement pas exclusif. Les adultes butinent les fleurs en
compagnie d'autres Cetoines, notamment d'Oxythyrea pantherina Gory et Tropi-
nota squcdida L.; elles sont frequentes sur les fleurs epanouies d' Acacia seyal
et d' Acacia raddiana,

Les Nocteules sont frequentes dans les oasis et leurs larves s'attaquent a
toute espece de plante cultivee, principalement aux legumes, sans avoir un re-
gime specialise. On trouve au Sahara des especes du g. Plusia, Prodenia, La-
phygma, Chloridea, etc., qui ont une aire de repartition tres vaste a travers le
continent africain. Elles se rencontrent non seulement dans les oasis, mais
aussi en dehors de celles-ci, en plein desert sur des plantes spontanees {Plu-
sia gamma L.; Laphygma exigua Hb.). Rhycia protophila Guen. est une espece
plus strictement nord- saharienne, nuisible dans les oasis du Sahara marocain
(Rungs). II y a lieu egalement de signaler des degats de Arenipses sabella
Hmps. (Gelechiidae) dont la chenille de 4cm de long, brunatre, s'attaque au
jeune regime de dattier avant sa sortie du spathe. II se produit une necrose
caracteristique pourrissant la fleur. J'ai observe cette espece a El - Golea en
avril-mai 1926 et elle est egalement commune a Timimoun et Adrar (Gourara-
Touat).

(ii) Elements specifiques ou subspecifiques. — Ce sont des phytophages a regime
plus strict, vivant au Sahara sur des plantes spontanees bien determinees ou
liees a une famille botanique; ils se sont adaptes en oasis a des plantes bota-
niquement voisines ou de la meme famille. L'on a affaire ici a de veritables
collections de phytophages passant sur la vegetation cultivee avec toutes leurs
cohortes de commensauz, parasites et satellites.

Les Pierides (P. rapae L. et P. napi L. en particulier) devastent frequemment
les cultures de Cruciferes en oasis (navets, raves, choux, etc.) et elles vivent
dans le desert aux depens de divers Cruciferes sauvages. Les Punaises des
Cruciferes, notamment les Eurydemna, obeissent aux memes regies.

Sur les Contonniers vivaces cultives en oasis on trouve toute la faune emigree
des Malvacees sauvages du desert, notamment des Malva et des Althea, particu-
lierement les Punaises Oxycarenus auxquelles viennent s'ajouter dans la zone
Sud - Saharienne des Dysdercus.

L 'etude des Aphides vivant au Sahara pose un probleme plus complexe depuis
que des travaux recents ont dissocie des especes telles que Aphis craccivora
Koch (= A. laburni des auteurs, non Kalt) considerees autrefois comme poly-
phages sur les Legumineuses et meme sur les plantes d'autres families.

Les plantes desertiques spontanees, constituant 1 'habitat des Aphides, forment
souvent des 'relais' permettant a I'espece non seulement de se maintenir, mais
encore de traverser le desert et d'occuper ainsi une aire de repartition continue
extremement vaste a travers le continent africain. C'est ainsi que Rhopalosi-
phum nymphaeae L., si commun en Europe et dans la region mediterraneenne

97

sur les plantes aquatiques les plus varices, se maintient au Sahara central sur
les Potamogeton ou nous I'avons trouve dans les gueltas du Tassili N'Ajjer
(Dider), loin de toute culture. II est probable que cette espece se maintient au
desert uniquement par la forme emigrante, car I'oeuf d'hiver est pondu sur des
Prunus dont il n'existe aucun representant parmi la flore desertique spontanee.
Le Potamogeton constitue done un veritable 'relais' permettant a Rhopalosiphum
nymphaeae d|avoir une aire de repartition continue depuis I'Afrique du Nord
jusqu'au Soudan. Aphis nerii F. vit sur Nerium oleander dans le Sud de 1 'Eu-
rope et le Nord de I'Afrique; on le retrouve sur cette plante au Sahara le long
des oueds, mais le Laurier ros6 n'existe pas partout dans le desert et se rare-
fie au fur et a mesure que Ton approche du Sahara central; 1 'espece passe
alors sur Calotropis procera qu'elle suit dans son aire de repartition vers le
Sud jusque dans la zone des savanes tropicales (region du Tchad). Les deux
especes que nous venons de citer ne sont pas reellement nuisibles, mais elles
constituent des exemples typiques valables pour d'autres especes qui peuvent
trouver des relais constitues par des plantes sauvages sahariennes, sur les-
quelles leur developpement est parfaitement possible. Un exemple analogue
nous est fourni par la Coccinelle du Melon (Epilachna chrysomelina F.) qui vit
dans la zone mediterraneenne sur diverses Cucurbitacees sauvages et cultivees,
occasionnant de serieux degats aux cultures de Melons et Pasteques dans le
Tell algerien. Cette espece a une aire de repartition tres vaste, jusqu'au Sou-
dan, et traverse le Sahara sur sa 'plante relais', la Coloquinte (Colocynthis vul-
garis Schred.) sur laquelle nous I'avons trouvee au Tassili, loin de toute cul-
ture. Chopard signale egalement sa presence dans I'Air sans preciser son ha-
bitat dans cette region et P. de Peyerimhoff au Hoggar et dans le Fezzan sur la
Coloquinte.

Lorsque ces 'relais' botaniques n'existent pas, I'expansion des phytophages
nuisibles dans le desert est soumise a des facteurs artificiels, dont le princi-
pal est le transport accidentel par I'homme.

L 'absence de 'plantes relais' pour certaines especes suffit a expliquer les la-
cunes considerables existant parmi la faune des Pucerons nuisibles dans les
oasis. C'est ainsi que le Puceron noir des Feves, Aphis fabae Scop., les
Pucerons des arbres fruitiers (Abricotier, Prunier), les Pucerons vivant sur la
Tomate, sur le Tabac, n'existent pas dans I'oasis de Djanet ni dans les autres
oasis du Sahara central que nous avons visitees a une saison favorable, alors
que ces insectes pullulent dans les oasis nord- sahariennes. Cette absence
fait ressortir non seulement I'importance du role joue par les 'plantes relais',
mais aussi que des facteurs atmospheriques tels que les vents violents qui
soufflent au Sahara ne suffisent pas a vehiculer les Pucerons a travers le de-
sert. Bien que des Aphides aient ete trouves jusqu'a 3,000 m d'altitude dans
les recherches effectuees sur la 'faune atmospherique' dans d'autres regions du
globe, le vent ne parait pas pbuvoir les vehiculer du Tell aux oasis du Sahara
central. Quant aux posibilites du maintien des Pucerons d'origine non saha-
rienne dans les oasis, elle est certainement possible, car bien des especes
sont susceptibles de vivre et d'evoluer sous les climats les plus varies. Un

98

grand nombre d'Aphides nuisibles puUulent dans les cultures de moyenne et de
haute fegypte dont les conditions ecobiologiques sont a peu pres identiques a
celles des oasis sahariennes.

(in)^lements recemment adaptes. — Dans cette derniere categorie, la plus interes-
sante a notre avis, se groupent quelques especes vivant normalement au Sahara
sur des plantes sauvages spontanees et qui se sont adaptees brusquement a
des plantes cultivees en oasis lorsque celles -ci se sont trouvees a leur con-
tact. Les elements 'recemment adaptes' constituent le fond de la faune des in-
sectes phytophages nuisibles de I'Afrique tropicale et equatoriale. Les ^ des
especes vivant actuellement en Afrique noire sur les plantes cultivees (elles-
meme presque routes introduites) proviennent d'adaptations recentes ou routes
recentes (certaines d'entre elles continuent a se produire a I'heure actuelle)
d'especes vivant primitivement sur des plantes sauvages de la savane ou de la
foret. Certaines de ces adaptations constituent de veritables exemples d'allo-
trophie et une demonstration suggestive de la formation sous nos yeux d'espe-
ces nuisibles aux depens de types n'ayant pas d'interet agricole. Les exem-
ples les plus typiques nous sont fournis par diverses Cochenilles Diaspidinae.

Saharaspis Ceardi Balachw, vit normalement sur les Ziziphus sauvages au Sud
de I'Atlas, mais il s'est adapte dans toute la zone Nord du Sahara aux arbres
fruitiers cultives en oasis, notamment a la Vigne, a 1 'Olivier, au Figuier, au
Murier, au Caroubier, etc. On le trouve depuis le Sahara marocain oceanique
jusqu'en Tunisie. Dans certaines oasis notamment dans cette de Tarjicht (oa-
sis pre'saharienne de I'Anti- Atlas marocain), nous avons trouve cette espece
attaquant avec vigueur des cepages de Vigne indigene cultivee, determinant le
dessechement des sarments. 5. Ceardi a ete trouve par les entomologistes ma-
rocains en dehors de la zone saharienne, notamment a Sale (env. Rabat) sur
Murier, ce qui prouve qu'il est susceptible, maintenant qu'il est adapte aux
vegetaux cultives, d'etendre considerablement son aire de repartition vers le
Nord.

Une autre espece que nous avons decrite du Hoggar, Aspidaspis Laperrinei
Balachw., vivant sur Olea- Laperrinei a 2,400m d'altitude se retrouve sur ce
meme Olivier et d'autres plantes dans I'etage mediterraneen du Tassili N'Ajjer,
notamment sur Myrtus nivellei et Nerium oleander { 1,400- 1,700 m d'altitude). II
s'agit done d'une espece polyphage indigene infeodee aux massifs du Sahara
central. (Cette espece a ete retrouvee recemment par Kaussari dans le Sud de
I'lran (Beloutchistan iranien) sur Calligonum sp.). Dans I'oasis d'lhrir et dans
celle de Djanet, nous avons trouve A. Laperrinei adapte sur des plantes culti-
vees, notamment a la Vigne, I'Abricotier, le Rosier et le Grenadier, dans le jar-
din de la Direction des Affaires indigenes. II s'agit la d'une adaptation toute
recente, ces plantes ayant ete introduites dans cette oasis il y a une dizaine
d'annees.

Sur la rive Sud du Sahara nous avons des exemples analogues avec Octaspidio-
tus Dallonii Balachw., Aspidiotini decrit de I'oasis de Gourmeur (Tibesti) vi-
vant sur Ficus salicifolius; cette espece a ete trouvee dans I'oasis de Myrriah

99

(environs de Zinder) par Remaudiere, sur des Goyaviers cultives (adaptation
recente).

Les trois exemples que nous venons. de citer demontrent que des especes con-
siderees jusqu'ici comme n'ayant aucun interet agricole sont en voie de devenir
nuisibles. II est probable que si ces insectes, qui sont intensement parasites au
Sahara par leurs ennemis naturels, etaient introduits dans d'autres regions du
globe, leur nocivite se trouverait considerablement accrue. lis possedent en eux-
meme un potentiel de nocivite permanent demontrant la necessite d'etudier bien da-
vantage la biologic des phytophages vivant sur les plantes spontanees en vue de
leurs possibilites de passage sur les vegetaux cultives. Une fois de plus ces pro-
blemes apparaissent comm^ intimement lies.

II Etat Phytosanitaire Des Oasis Sahariennes.

Dans la limite ou I'inventaire des insectes nuisibles a ete fait dans les diffe-
rentes oasis du Sahara franc^ais, I'etat phytosanitaire de celles- ci n'apparait nulle
part comme revetant un caractere reel de gravite, tout au moins en ce qui concerne
les insectes nuisibles. Exception fait pour les especes vivant aux depens des
denrees alimentaires stockees dont les ravages posent des problemes particuliers;
la nocivite des vrais phytophages est reduite, elle ne s'exerce jamais d'une ma-
niere generalisee a caractere epidemique entrainant des catastrophes economiques
ou sociales pour ces regions, deja tres desheritees en elles memes. En ce qui
concerne les Acridiens migrateurs, la seule espece qui vit reellement au Sahara est
le Criquet pelerin {Schistocerca gregaria Forsk.) qui traverse entierement le desert,
soit dans la direction Sud-Nord ou Sud- Est- Nord- Ouest, suivant que les essaims
partent de I'Atlantique ou de la mer Rouge. Ce sont les formes adultes qui, 3'a-
battant dans les oasis, provoquent par les annees d'invasion des degats conside-
rables en devorant toutes les cultures.

Les oasis presahariennes et celles situees en bordure de 1 'Atlas saharien sont
generalement beaucoup plus devastees que les autres du fait que les Criquets arri-
vent dans ces regions alors qu'ils sont deja a un age avance (Sauterelles jaunes)
et dans une phase d'alimentation intense. Les essaims venant directement du Rio
del Oro apparaissent au Sahara sous la forme jeune (Sauterelles roses) et ne s'ali-
mentent pas. Ces vols traversent en general 1 'Atlas pour effectuer leurs pontes, et
c'est a ce moment- la que I'espece se montre nuisible, tant a I'etat adulte qu'a I'e-
tat larvaire.

Quant au Criquet migrateur, Locusta migratoria L., ph. migratorioides , il ne de-
passe pas au nord la zone sahelienne soudanaise; c'est une espece presque exclu-
sivement graminicole. L 'ensemble de cette faune est faiblement agressive du fait
qu'elle est tenue en echec par de dures conditions ecologiques et aussi par un
parasitisme intense qui s'exerce partout au Sahara, aussi bien dans les oasis que
dans le milieu desertique proprement dit. Au Sahara, la vie a atteint presque par-
tout un equilibre stable et statique,

II est egalement peu probable que des introductions nouvelles se produisent,
excepte peut-etre pour quelques Aphides ou Coccides (cf. supra) en raison des

100

conditions ecobiologiques tres particulieres qui regnent au Sahara, rendant la vie
des phytophages non sahariens et leur adaptation, precaire ou impossible.

Un autre facteur favorable est constitue par I'isolement des oasis les unes par
report aux autres, il serait toujours possible de detruire ou de reduire les degats
d'une espece dangereuse au cas ou elle apparattr ait dans Tuned'elles; il serait ega-
lement facile d'empecher sa propagation par I'application de mesures phytosani-
taires elementaires. L'exemple de la progression tres lente de P. Blanchardi dans
le Sahara occidental est demonstratif a ce point de vue.

Telle qu'elle appara'i*t dans la limite des connaissances actuellement acquises,
cette faune presente des lacunes considerables, si on la coippare a celle qui vit
sur les memes vegetaux cultives, hors de la zone saharienne. Un nombre conside-
rable d'elements manquent, notamment de tres nombreux Pucerons nuisibles (cf.
supra), laplupart des Cochenilles nuisibles (toutes les Cochenilles des Agrumes
disparaissent au Sahara (El Golea, Mzaby oued Rhir)), des Thysanopteres, des
Coleopteres phytophages (Curculionidae, Chrysomelidae), dont certaines families
(Scolytidae) font meme totalement defaut, la plupart des Lepidopteres nuisibles non
polyphages, etc. Cet etat de choses fait ressortir a quel point cette faune est res-
tee a I'abri des introductions et des acclimatations. Certaines cultures, telles que
les Cereales (sur pied), ne possedent au Sahara pour ainsi dire aucun insecte para-
site important (en dehors des polyphages). Quant au nombre des endemiques saha-
riens nuisibles, il est egalement tres faible, les elements adaptes sont recents et
leur nocivite apparait encore comme peu accusee.

En dehors du climat saharien proprement dit, il existe certainement d'autres
facteurs qui contribuent a I'eliminatioii ou la limitation numerique des especes nui-
sible dans leur oasis.

Le 'rythme vital' presque exclusivement nocturne pour les especes saharien-
nes, est defavorable aux especes a 'rythme diurne', c'est- a- dire a la grande ma-
jorite des especes nuisibles phytophages. L 'insolation proprement dite avec sa
lumiere riche en radiations violettes et ultra- violettes influence certainement defa-
vorablement le developpement des oeufs ou des jeunes larves de nombreux phyto-
phages. De meme I'echauffement du sol en surface, surtout lorsqu'il est sablon-
neux et leger, ou il peut atteindre des temperatures critiques de mort des insectes
(+ 60 a 70°C) constitue en facteur eminemment prejudiciable a la nymphose de nom-
breuses larves de phytophages qui s'opere a une tres faible profondeur dans le sol.

Enfin, la grande majorite des insectes sahariens possede de longues diapau-
ses qui leur permettent de passer les periodes critiques ou defavorables meme si
elles se prolongent pendant plusieurs annees consecutives (Cochenilles - Margfl-
rodes). Il n'en est pas de meme pour la plupart des insectes phytophages introduits
dont I'ethologie n'accuse pas d'arrets de developpement, Ces especes se trouvent
done obligees d'evoluer dans des conditions tres defavorables, notamment pendant
la periode estivale chaude ou la vie est normalement tres ralentie au Sahara, sur-
tout pendant la phase diurne.

La pauvrete des phytophages sahariens apparait egalement pour les insectes
vivant aux depens des vegetaux spontanes, Ce phenomene est particulierement

101

accuse pour la faune des especes vegetales 'reliques* qui subsistent dans cer-
taines stations limitees ou tres limitees du Sahara, comme les temoins precaires
d'un passe plus humide ou plus frais. Olea Laperrinei Batt. et Trab., olivier sau-
vage de l"etage mediterraneen' du Hoggar et du Tassili, n'heberge pour ainsi dire
aucun phytophage en dehors d'une cochenille — Diaspidinae {Aspidaspis Laperrinei
Balachw.) rare et clairsemee dans les peuplements spontanes. Son 'correspondant'
mediterraneen, I'Oleastre (Olea europea L.) est habite par centre par pres d'une
centaine d'especes de phytophages specifiques ou subspecifiques de tous ordres.
II en est de meme pour Myrtus Nivellei du Tassili et du Hoggar qui n'est attaque
par aucun insecte, alors que son 'correspondant' mediterraneen, Myrtus communis L.
est tres parasite,

Cupressus Dupreziana Camus, le magnifique cypres du plateau de Tamrit
(1,750m) dans le Tassili (Sahara central), represente aujourd'hui par une centaine
d'individus presque tous millenaires ou plusieurs fois millenaires, n'est attaque
par aucun insecte xylophage ni phytophage comme nous avons pu le constater en
etudiant cette 'station relique' en mai 1949. Son 'correspondant' du Haut- Atlas
marocain, Cupressus sempervirens L. (= atlantica Gaussen) possede par contre
toute une faune de phytophages specifiques ou subspecifiques.

Si I'on etudie la biocoenose des 'emigres tropicaux' sahariens venus du sud,
on arrive a des conclusions identiques. Ficus salicifolius Vahl. ssp. Teloukat
Batt. et Trab. est un Ficus tropical qui remonte jusqu'au Sahara central (Tassili);
il est loin de renfermer la riche faune des Ficus tropicaux, c'est a peine s'il he-
berge trois ou quatre especes d'insectes dont un Lepidoptere mineur de tige, un
Aleyrodidae et une cochenille, Pseudococcini. Sur Balanites aegyptica Delile du
Sahara central nous n'avons rien trouve alors que cette plante heberge toute une
faune particuliere dans la zone sahelienne du Tchad et de I'A.O.F. Rien non plus
sur Salvador a per sic a L. en dehors d'un Aleyrodidae.

Les Acacias epineux du Sahara {Acacia raddiana Savi, Acacia seyal Delile)
forment aujourd'hui un reliquat degrade de la brousse sahelienne qui s'est etendue
autrefois beaucoup plus vers le nord. L 'etude de leur biocoenose fait appara'Jtre
une faune considerablement appauvrie par rapport a celle du Soudan, du Niger ou du
Tchad. Cet appauvrissement s'accuse au Sahara m^me, du sud vers le nord; dans
la Soura, le Djebel Bechar ou I'Anti- Atlas marocain, les Acacias ont perdu pres-
que toute leur riche faune originelle tropicale.

Tous ces vegetaux qui subsistent aujourd'hui dans des conditions differentes
de celles de leur milieu naturel ont r^idement perdu leurs insectes phytophages
specifiques ou subspecifiques qui n'ont pu resister au changement de climat ni
s'adapter aux dures conditions sahariennes. Dans certains cas, cette elimination a
ete totale, comme pour Cupressus Dupreziana Camus, qui est devenu une espece
vegetale 'azoique'.

II est probable que lorsque I'ecologie saharienne sera mieux connue, les fac-
teurs limitatifs jouant en defaveur des insectes phytophages nous apparaitront avec
beaucoup plus de clarte.

102

Quoi qu'il en soit, si les rendements sont faibles au Sahara et certaines
cultures tres deficitaires, cela ne tient pas specifiquement aux attaques
des insectes nuisibles, mais a d'autres causes beaucoup plus importantes. La
pauvrete organique du sol, la tres mauvaise qualite des graines de semence ne su-
bissant non seulement aucune selection par rz^port au milieu, mais souvent meme
aucun renouvellement, leurs tres mauvaises conditions de conservation, et enfin
leur plantation dans un sol constamment epuise, sont autant de facteurs qui con-
tribuent a la pauvrete des rendements sahariens, Certaines maladies cryptoga-
miques sevissent egalement avec intensite, notamment les Charbons des Cereales.
Dans certaines localites (vallee d'Ahrar), plus de 80% des epis sont charbonnes
alors que Ton ne trouve aucune attaque d'insectes.

Cette etude doit ^tre consideree comme un essai preliminaire, des conclusioas
definitives ne pourront intervenir que lorsque nos connaissances sur I'ensemble
des insectes nuisibles aux oasis sahariennes auront fait i'object de recherches
plus method! ques et plus approfondies.

References

Chevalier, A. 1938. Mem. Soc. Biogeographie, 6, 310- 11.

Hall, W.J. 1926. Techn. and Sc. Serv. Bull., 64, 1-5, id. Techn. Sc. Serv. Bull.. 72, 32.
Pasquier, R. 1951. Bull. Liaison Saharienne, No. 5.
Peyerimhoff, P. de. 1945- Ann. Soc. Ent. Fr., 114, 1-76.
Real, P. 1948. Rev. Path. Veg. Ent. Agr. France, 59-64.

Rungs, C. 1944. Mission francaise au Fezzan. Hemipteres Coccidae. Bull. Soc. H. N. A/.
Nd.

103

ROLE DES INSECTES SOCIAUX DANS LES TERRAINS DU SAHARA

Professor F. Bernard
(Alger)

Introduction:

Au desert, les paturages naturels, les cultures, les coi:structions humaines sont
abondamment peuples par les Fourmis et Termites du sol. lis existent partout: un
arbre isole, une touffe de plantes rencontree fort loin de tout autre vegetal, abritent
et nourrissent une ou plusieurs societes de ces Insects. II n'est pas exagere de
dire que leur influence au Sahara, souvent tres nuisible, est encore plus importante
que sous les Tropiques. En effet, la rare vegetation, particulierement vulnerable
apres une longue periode seche, est facilement achevee par les Termites, ou quel-
quefois protegee par des Fourmis insectivores.

Cependant, la distribution, le comportement et meme la systematique de ces
animaux si communs etaient encore, en 1944, relativement peu etudies. Depuis les
voyages classiques de Forel (1898) et de Lameere (1901) en Algerie, seules quel-
ques notes de Santschi (1910 a 1935) renseignent sur les Fourmis de Sud tunisien.
Pour les Termites, la faune de Tripolitaine et du Fezzan etait la moins mal connue,
grace aux travaux de Silvestri (1912 a 1924) et aux observations de Scortecci (1933
a 1939). En Egypte, au Moyen- Orient, en Asie Centrale, les investigations restent
surtout taxonomiques, a part I'excellente monographie de Menozzi (1933) sur les
Fourmis recoltees par Bodenheimer en Palestine.

Pendant et apres la seconde guerre mondiale, les missions en Afrique se mul-
tiplient, mais peu comprennent des specialistes dTnsectes sociaux. Signalons a
cet egard la randonnee d'Alger au Cameroun, consacree aux Termites, effectuee en
1948 par Grasse et ses collaborateurs. Les tres interessants resultats biologiques
obtenus sur les formes sahariennes occidentales ne sont pas encore tous publics.
Dans lapartie centrale du grand desert, grace au Gouvernement General de 1 'Algerie,
j'ai pu observer et recolter Fourmis et Termites, durant 5 mois en tout (Bernard,
missions au Fezzan (1944 et 1945) et au Tassili des Ajjer (1949)). Enfin, quelques
localites du Nord du Sahara et des Hauts Plateaux furent explores recemment par
Pierre (Beni- Abbes, Erg occidental), par HoUande (Hauts Plateaux, Bou Saada) et
par moi-meme (centre des Hauts Plateaux, sud tunisien).

Au total, une douzaine d'entomologistes ont examine sur le terrain les especes
sociales du Sahara. C'est relativement peu, si I'on songe a I'etendue a parcourir
et a la complexite des phenomenes a etudier. Toutefois, aujourd'hui, le role prati-
que et I'identite des types dominants peuvent etre precises. Bien qu'un dixieme
seulement du desert soit suffisamment connu, la repartition des formes les plus im-
portantes est assez vaste pour que les resultats ci-dessous aient un degre satis-
faisant de generalite.

Etat actuel des connaissances:

Les Guepes et les Abeilles sociales sont pratiquement negligeables, en raison
de leurs grands besoins d'eau et de fleurs, Meme dans les oasis, I'Abeille domes-

104

tique est inconnue car elle ne trouverait aucune fleur durant les trois quarts de I'an-
nee. La seule Guepe signalee est un Frelon: Vespaorientalis, banal en Asie
chaude, plutot rare au Sahara. II n'est abondant que dans les vallees du Tassili
n'Ajjer, massif exceptionnellement riche en sources et en lacs, ou il est represente
par sa variete Zavattarii, de couleur brun- chocolat. Peu agressif, il se montre
bienfaisant par les Insectes phytophages qu'il detruit. Les Touareg le nomment
An'kokar.

En 1902, les myrmecologues ne citaient du Sahara que 25 especes de Fourmis,
surtout recoltees dans le Sud algerien. Leur nombre s'est eleve a 41 en 1940, a 66
actuellement. Mais 15 Fourmis seulement ont une importance reelle pour la vie des
cultures et des sols. La plupart de ces Insectes communs, loin d'etre des sahariens
stricts, ont une large distribution dans le sud mediterraneen et I'Asie occidentale.
Au contraire, dans Tordre des Termites, habituellement tres hygrophile, les types
adaptes aux pays arides sont peu nombreux, et souvent nuls en regions humides.
Leur classification est delicate, et nous evaluerons tres provisoirement a 9 le stock
d'especes sahariennes connues, dont 2 seulement, tres nuisibles, jouent un role
economique capital dans 1 'ensemble du desert. Leur biologic ne commence a etre
elucidee que depuis 1948, et bien des points restent mysterieux.

Ces deux groupes montrent combien il faut etre prudent pour les conclusions
biogeographiques. Si les Fourmis sahariennes les mieux adaptees appartiennent a
des genres tropicaux tres evolues et recents (par exemple Monomorium et Acantho-
lepis), par contre les Termites xerophiles dominants representent des types archa'i-
ques, refugies dans les montagnes et les pays sees (Hodotermitidae et Psammoter-
mitnae).

Les Termites:

Tous les Termites communs du Sahara rongent tiges, racines et bois de con-
struction. Aucune plante locale n'echappe a leur atteinte, et meme les Asclepia-
dacees les plus veneneuses, comme le grand Calotropis procera, sont attaquees.
J'ai revu au Fezzan un fait deja observe par les chercheurs italiens: en plein reg de-
nude, sans aucun vegetal apparent, on trouve parfois des Psammotermes a quelques
centimetres dans le sol. lis subsistent la grace aux troncs subfossiles de I'an-
cienne flore disparue apres le pleistocene humide, notamment dans le bois des gros
Tamarix recouverts par les alluvions des inondations passees.

Leurs degats les plus manifestes concernent les troncs d'Acacias et de Pal-
miers, materiaux employes ici pour les habitatiois et le soutenement des puits. Au
bout de 4 a 12 ans, les troncs de Dattiers qui renforcent la paroi des puits s'effond-
rent, entierement mines par Anacanthotermes. Les poutres des maisons, particu-
lierement habitees par les Psammotermes, s'ecroulent aussi, mais plus lentement.
Le bordj militaire de Timimoun est ainsi entierement a rebatir. Moins apparentes,
mais encore plus nocives pour I'avenir de I'homme au Sahara, sont les destructions
operees dans les paturages naturels, loin des oasis. Les Termites mangent les tis-
sus vegetaux morts et alterent peu les parties vivantes. Mais, apres une suite de 4
ou 5 annees sans pluies (cas frequent au desert), les plantes spontanees. disse-
chees, sont achevees par les Psammotermes. Exemples: les paturages a Calligo-

105

num comosum du sud du Fezzan en 1944, ceux du centre du Tassili n'Ajjer en 1949.
Le Sahara occidental, a I'ouest du meridien d' Alger, ne parait pas changer beaucoup
de climat depuis 1900. Mais la partie orientale (Tunisie, Fezzan, Egypte etc.) est
en aridite accrue, de I'avis de nombreux specialistes. Beaucoup de points d'eau,
abondants il y a 50 ans, sont maintenanta sec, et divers paturages ont disparu.
Cette evaporation semble correlative du recul general des glaciers sur le globe,
elle est, en tous cas, aggravee par 1 'action des Termites sur la flore.

Sur le comportement de chaque espece je serai bref, car les observations plus
completes de la mission Grasse (1948) seront prochainement publiees. II suffira de
definir rapidement les principales differences ecologiques notees:

Anacanthotermes ochraceus (Burm.) puUule des les Hauts Plateau algeriens et
occupe tout le Sahara sauf sa lisiere sud. II remonte ^ plus de 2,000 metres en mon-
tagne. Ce gros Insecte a encore des yeux chez les ouvriers et les soldats (ces
dernier s relativement tares). Les ouvriers, tres fourrageurs, recoltent, en plus du
bois, des morceaux de tiges de Graminees et des detritus vegetaux. Chaque nid,
tres mal limite, est forme simplement de longues gaieties, peu profondes dans le
sol, pouvant depasser 200 metres de long. Aussi la reine est- elle tres difficile a
trouver: son premier et unique exemplaire connu a ete capture par Grasse en 1948.
Anacanthotermes ochraceus pxefere le sable et les rochers a I'argile, et supporte
bien les terrains sales. II manque souvent dans les paturages a sable argileux
eloignes des habitations, mais semble exister sans exception dans tous les oasis.
Une espece voisine: A. ^asmanni Sjostedt, peuple surtout la bordure nord du desert
et I'Atlas saharien. Ses sexues ailes sont noirs, et plus petits que les sexues
jaunes d'A. ochraceus. La biologic est tres analogue. Les Psammotermes sont des
Termites de faible taille, aveugles, montrant un grand polymorphisme des soldats,
tres nombreux, dont la longueur varie du simple au double dans un meme nid. Celui-ci
est aussi ramifieet long que chez Anacanthotermes, mais penetre plus profondement
dans le sol. La mission Grasse a prouve que depuis la termitiere, situee sur une
dune, les Insects sont capables de descendre de 10 a 30 metres jusqu'a la couche
aquifere, d'ou ils rapportent de I'eau dans leur bouche pour humecter les galeries.
Nous verrons ci-dessous un comportement comparable chez la Fourmi Acantholepis
jrauenfeldi.

En plus du nid permanent souterrain, il y a parfois une construction externe,
petite tour ou cube de 15 a 30 centimetres de haut. D'apres les notes de Scortecci
et les miennes au Tassili des Ajjer, cette tour existe surtout dans les lits d'oueds,
et servirait aux Psammotermes a s'elever au-dessus des alluvions, trop humides et
coUants apres les crues. En periode seche, ces habitacles saillants sont generale-
ment abandonnes.

Geographiquement, Psammotermes est nettement plus meridional que le prece-
dent genre: il manque au nord du parallele de Ouargla, mais atteint par contre la
savane seche du Soudan. Il ne depasse guere l,3000metres au Tassili, et n'est pas
signaledans I'Atlas. Mais, dans son domaine propre, il est tres ubiquiste, commun
dans les terrains argileux et les paturages isoles ou manque Anacanthotermes. Il
attaque peu les Palmiers, mais surtout les Graminees, les Calligonum, les Acacias

106

et les Tamarix. Seuls les lieux trop sales lui sont defavorables, ainsi que les ter-
rains a sable grossier. Une seule espece parait valable: P. hybostoma Desneux.
Mais une multitude de races geographiques compliquent son etude, comme d'ail-
leurs chez A. ochraceus. D'apres une etude inedite de G. Richard (1952), les soi-
disants 'ouvriers' de Psammotermes ne seraient que de jeunes sexues.

Un Amitermes, encore a 1 'etude, est parfois nuisible 9a et la. Enfin, pour se
limiter aux Termites communs, il y a au Sahara un Metatermitide sans Flagelles
symbiotiques; c'est le petit Microcerotermes (forme principale: M. eugnathus
Silv.). II abonde sur les Hauts Plateaux et dans les paturages du Nord, et se re-
trouve a Tamanrasset (Hoggar). Au sud de Colomb- Bechar, les plantes sauvages
sont recouvertes d'epaisses croutes de sable argileux par cet Insecte. Moeurs peu
connues, degats moins importants que ceux des genres precedents.

Les Fourmis

Les premiers observateurs des especes sahariennes: Forel et Lameere, ne
pouvaient preciser le role economique des Fourmis locales, dont le regime alimen-
taire demeurait souvent inconnu. De plus, la region alors exploree (Autour de
Biskra, Touggourt et Gharda'ia) ne representait qu'une tres faible fraction du desert.

Aujourd'hui, le comportement et la repartition des principaux types sont assez
decrits pour que Ton puisse faire un bilan approximatif de I'equilibre entre Four-
mis utiles et nuisibles. J'ai tente d'evaluer cette concurrence sur les divers
genres de terrain, selorhla methode des releves quantitatifs: Dans un biotope aus-
si homogene que possible par le sol, la pente et la flore, on denombre les four-
milieres presentes. Apres avoir trouve une cinquantaine de nids, le pourcentage
est etabli pour chaque espece. Un tel releve rapide prend une heure ou deux, et
c'est le seul procede commode durant les courtes haltes des caravanes. Le resul-
tat fournit une notion assez satisfaisante du peuplement, car chaque nid est 1 'unite
biologique pour les Insectes sociaux. La plupart des formes ont des terriers tres
visibles sur le sol desertique, grace aux deblais expulses. Les inconvenients cer-
tains de cette methode ont deja ete examines dans notre travail de 1948 sur le
Fezzan, et je n'y reviendrai pas ici.

Au total, 72 releves semblables furent pratiques j-usqu'a present au Sahara, de
0 a L800 metres d'altitude, et de la latitude de Tozeur (sud- tunisien) a celle d'El
Gatroun (Fezzan sud- est). Les facies etudies sont assez divers, et I'ecologie de
chaque Fourmis assez constante d'une region a 1* autre, pour permettre les deduc-
tions suivantes:

(1) Fourmis utiles et Fourmis nuisibles: Les seules formes reellement in-
sectivores (s'attaquant principalement aux Termites et aux Fourmis granivores
Messor) appartiennent au genre Cataglyphis Fbrster. Ce sont de grandes Fourmis
tres agiles, souvent a reflets argentes, sortant jour et nuit et chassant isolement.
Leurs 5 especes sahariennes (3 tres communes et 2 plus tares) peuvent etre con-
siderees comme bienfaisantes pour la vegetation. Un peu plus omnivores, les
petits Acantholepis (3 especes) paraissent toutefois surtout mangeurs d'Insects,
done utiles.

107

A I'extreme oppose, 15 especes environ, dont 3 abondantes partout, se mon-
trent franchement nuisibles, soit en protegeant les Homopteres a miellee sur les
plantes {Tapinoma simrothi Krausse, Crematogaster oasium Sant., Paratrechina
jaegerskjoeldi Mayr etc.), soit en recoltant une bonne partie des graines locales
{Messor aegyptiaca Em., Monomorium chobauti Em., et d'autres). Sur les Hauts
Plateaux, on evalue que les Messor detournent 10 a. 15% de la recolte de cereales.

Mais ces cas bien tranches ne concernent que 23 Fourmis sur les 66 connues
du desert. Les 43 autres, plus ou moins omnivores, compensent souvent, par les
animaux phytophages qu'elles detruisent, leurs propres degats, directs ou indirects,
a la vegetation. En voici des exemples: Monomorium salomonis (L.) leche rare-
ment le miellat des Homopteres et capture divers insectes, mais certaines de ses
races (surtout la sbsp. didonis Sant., tres commune) ont une preponderance de
graines varices dans 50% des nids ou davantage. Ces lignees granivores sont ob-
servees aussi chez plusieurs Pheidole et Tetramorium de la region.

(2) Influence du milieu sur le pourcentage de Fourmis utiles: Le tableau
ecologique ci-dessous n'a pas besoin de longs commentaires pour etablir I'effet
du terrain sur la faune. Apres divers essais de classement, les stations de rel-
ieves se groupent, assez naturellement, comme suit: En terrain sec (sable pur,
rochers, fortes pentes argileuses ou rocheuses) le rapport des fourmilieres utiles
a celles des especes franchement nuisibles varie de 6 a 28. Les Acantholepis
d'origine mediterraneenne {A. frauenfeldi Mayr., tres repandu, A. ajjer Bernard,
dominant au Tassili) jouent le role principal ici, sauf sur les rochers ou Mono-
morium salomonis abonde. Dans plus des 9/10 du desert, pauvres en couches aqui-
feres superficielles, il y a done une forte majorite de Fourmis protegeant la flore
contre les Termites. Les terrains sales occupent encore une vaste superficie dans
les depressions sahariennes. Les recherches au Fezzan montrent que le type de
sels, variable d'un point a I'autre (chlorures, sulfates ou carbonates y dominent) a
moins d'importance vis- a- vis des Fourmis qu'une propriete generale de ces sub-
stances: celle de retenir longtemps de I'eau. Sous la croute salee, seche et dure
en surface, il y a general ement un sable jaune, contenant de 5 a 50% de sels. Ce
milieu special est richement peuple, mais la plupart des especes nuisibles et des
Monomorium le tolerent mal. Acantholepis frauenfeldi reussit tres bien la encore,
et peut enrichir son nid en eau en remontant de la profondeur des boulettes salines
humides, phenomene vu a Mourzouk (Fezzan) et a Beni- Abbes (Algerie occidentale)
au cours de mes releves numeriques. Des formes hygrophiles des jardins, omni-
vores: Camponotus maculatus (Fab.) et Pheidole pallidula Nyl. s'ajoutent a cette
faune. Les types utiles restent ici en majorite, sauf da.is quelques palmeraies
trop ombragees, ou les Cataglyphis et Acantholepis ne trouvent pas la forte insola-
tion qui leur est necessaire.

La proportion des Fourmis nuisibles augmente beaucoup dans les terrains aqui-
feres peu sales (fonds d'oueds, jardins arroses ...). Cela tient a ce que les especes
utiles du desert, presque toutes xerophiles, resistent mal a I'inondation par les
oueds ou a I'arrosage. Au contraire, les Crematogaster et surtout I'envahissant
Tapinoma simrothi, genres d'origine tropicale, pullulent dans les jardins, ou ils
deviennent largement dominants. C'est encore plus vrai dans le Sahara du Nord et

108

^,^

<a

(0

c

^

to »;

§•2

to
to

0
u

ta

11 ^<u

3-e

CN 0 ^

(%I

6^

fl— 1

0

rM

\0 -H

6?

fN
fN

Csl

0

ea

Ul

d

O I-

0

ea

o 4;

3
° to

- w

^ a

to ^v

to -:5

i) V
— . to

.w to

U ■»; to

-^ 0 Vrt 1

s

^ CSI (N

rr\

•^ r^

•^

fN ^

6?
0

CN

0

*— *

0

ir\

Si «

1 =

" ^ *•
— < to — 1

F— (

\o

^ fN

f-H

CM

F-H

<2>

c (a

O J3

3 C '

to

c

u

• -^

+1 >< c,
ca 60 5

1-

ta

w

l-l
V

H

6?

fN

(—1

r-

fO ^H
fN

1^

•

0

0 CJN

6^

-^

r-

CN

rri

vH

2

2

2

0) u
3 SI

ST- •

rt

« -0 CO

<

= r.

to

a:

• 2

Vn ca -lu

to 3
'« w SI

6^

(a

6^

r- \o 0

\r\

^

ir\ r-

'«r

-H fN

6^

ir\

t— (

•— 1

0

fN

CO

i)

Ul

XI

tX)

fSI

>*

1— 1

»-<

<-<

f— 1

r^j

00

-0

e

2

lU c o

to

3

0

D

"o,

to to

O

PL,
C/5

-^ 4) to

2 to w

XI 00 o

'ta
to

w S 3

6?

V3 -<;t —I

f— I

6?

CM tr\

r-

Cs — '

1— 1

fN

0

fN

0

t— 1

"3

w

to

c

3 ^D
(J -T3

k-4

10

™ 3 ca
^ o

k4

11

6?

ITN \0 0

(Vl

^

OS cr\

i-H

m 0

6?

0

-<i'

0

0

NO

o to

1>

3

u

C

m

t— 1 I— 1

rOi

rr\

1—1

"<f

1— 1

H

VO

f<^

a-

CO

w

?; ''11

Q

O w t
to C 3

(0

Ul

HI

6?

0 "/> 0

6^

00 0

0

0 ^

6?

0

0

^

0

li-\

.2 m

to

4>

u

0

Ul

tvl

«-H

»-H

0

"^r

"•r

vd

O

-a t; "-a"

w 3

o

^4) 0) to

«

1)

U

w

to

3 3
ta Oi

fe?

0^0

0

6?

iri 0

^H

0 0

6^

0

0

r>j

■^

0

0

rr\

rCi

m

VD

I— 1

p

"S"^-:

"a

Ul

to

<

5 § 2

Ul

1)

-J
<

'■lU

H

<u A

6?

(30

000

1— i

6^

— 1 0

■— 1

0

0 0

6?

0

0

m

0

00

H

jT a,
c ■«
c <^

O V

a,

(0

2-5

_s

•

to

c
0

e
0

«

3
J5

>«4

•

a
to

w
3

ca

3

(A
• -4

3

Ji .0

0,3

U

•

"3

u
el

0 >^

0 E
a. 0
E <-•

C/5

0

X)

0

C
(0

ta c

-a c

3 >s

rt °
CO G

CO

w

1— 1

H

■-S ta

« to .S

"c C u
a, « s^

2 ta XI

to

a

V

"o

Ul

C

ta
u

CO

w

H

X

«

"Si

E 6

6 (9

0 •

M
V

V

"o
-a

x:

«

CO

1— 1

2

•

to
«d
E
0

c

a,
ca

a.
to

V

\r\

0

(0

to

0)

Ul
«J
u*

to
ea
00
0

ta
E

a

0

E

3

s

E
0

§

V

ut

3

w
0

a,

I— <

u u u

<

1— 1

l-H

2 2

CL

U (J

l-H

H

2

u

2

OS

109

I'Atlas, ou Tapinoma sort des oasis et grouille sur les alluvions ensoleilles, de
jour et de nuit, entretenant force Homopteres sur toutes les pi antes.

En resume, Acantholepis frauenfeldi, utile, et Nionomorium salomonis, indif-
ferent, sont de loin les 2 Fourmis dominantes, assurant souvent a elles seules plus
de 60% des nids. Leur importance diminue dans les sols humides et peu sales, ou
les omni votes NionomoTium gracillimum et Pheidole, les nuisibles tAessor, Cremato-
goster et Tapinoma I'emportent.

Procedes de Defense:

La lutte directe contre les Termites sahariens semble inoperante, a cause de
leurs galeries diffuses, longues de centaines de metres et du prix el eve des insec-
ticides efficaces. Sans pouvoir guerir les degats commis, mieux vaut prevenir de
futurs accidents par les memes methodes qu'en Afrique Noire: pas de bois dans les
constructions, poteaux impregnes de creosote, et surtout extirpation complete des
racines et des bois souterrains autour des fondations d'un batiment et sous elles.
Je renvoie sur ce point a la brochure de Grasse (Revue de Pathologic vegetale de
France, volume 23, 1936, fascicule 4).

La Fourmi la plus nuisible dans les oasis (Tapinoma simrothi) est passible
des memes moyens de lutte que le fameux Iridomyrmex americain: appats de Sucre
arsenic, places dans des pots accroches aux troncs d'arbres. Ce sucre a 2% d'arse-
niates empoisonne les reines et amene, au bout de quelques mois, I'extinction des
fourmilieres. Les Crematogaster et P aratrechina seront detruits en meme temps.

Reste le probleme, plus nouveau, de la protection des Fourmis utiles. Dans
la plupart des pSturages a chameaux, elles se trouvent deja en majorite. Autour
des cultures, elles continuent a prosperer si le sol demeure sale. Il faut done evi-
ter qu'un arrosage excessif, ou des fuites des seguia (canalisations des oasis)
viennent dessaler les emplacements incultes. Non seulement le sel favorise les
nids des Cataglyphis et des Acantholepis aux depens des Tapinoma et des Messor,
mais encore la lumiere solaire qu'il renvoie parait exciter le metabolisme et la
chasse de ces insectivores rapides, les plus agiles des Fourmis africaines.

Resume.

(1) L'importance economique des Termites et Fourmis parait encore plus grande
grande au desert que sous les Tropiques, car ils existent au pied de tous les vege-
taux, meme les plus isoles, facilitant le declin de ces plantes apres une longue
periode seche. Quelques Fourmis des genres Cataglyphis et Acantholepis protegent
au contraire la flore en devorant des Termites et des larves phytophages. On con-
nait aujourd'hui 9 especes de Termites au Sahara, dont 2 tres nuisibles, et 66 Four-
mis differentes, dont environ 8 manifestement utiles et 15 nefastes pour les cultures.

(2) L'ecologie des 2 Termites dominants est resumee. Psammotermes hybos -
toma Desneux existe partout, et fait ses principaux degats dans les paturages natu-
rels et dans les maisons. Anacanthotermes ochraceus (Burm.) manque parfois dans
les paturages et les maisons, mais supporte mieux le sel et le sable grossier; il
detruit notamment les constructions soutenues par des troncs de Palmiers.

(3) Un tableau de repartition quantitative des 14 Fourmis principales sur

110

divers types de terrains est donne. Generalement, les sols tres arides ou tres
sales avantagent les especes utiles. Les sols aquiferes peu sales, surtout s'ils
sont ombrages, favorisent les nuisibles, dont la plus commune est Tapinoma sim-
rothi Krausse.

(4) Les methodes de lutte contre les Termites seront essentiellement preven-
tives (extirpation de tous bois ou racines autour des futures constructions humaines).
Osntre Tapinoma et formes analogues, les appats Sucre s empoisonnes sont recom-
mandes. Enfin, il est possible de proteger les Fourmis utiles en evitant la dessa-
lure des sols incultes entourant les jardins.

References

Les references de presque tous les travaux cites se retrouveront dans 2 publications
recentes de I'Institut de Recherches Sahariennes de i'Universite d'Alger.

Bernard, F. 1948. Les insectes sociaux du Fezzan. Comportement et biogeographie,
Serie du Fezzan, 5, 87- 200.

Bernard, F. 1952. Les Fourmis du Tassili des Ajjer. Serie du Tassili, 1, 105- 190.

vCes volumes peuvent etre obtenus en ecrivant a I'Institut de Geographie de la Faculte
des Lettres d'Alger, ou a la librairie Lechevalier, 12 Rue de Cournon, Paris (6^).\

111

THE MICROBIOLOGICAL FORMATION OF SULPHUR IN CYRENAICAN LAKES

K.R.^Butlin and J.R. Postgate
{Chemical Research Laboratory, D.S.I.R., Teddington)

The importance of sulphate - reducing bacteria (Type name : Desulphovibrio desul-
phuricans) in the formation of non- volcanic deposits of sulphur has been recognised
by geologists and microbiologists for many years. It is thought that the organisms
reduced sulphates to sulphides, which were then oxidised to sulphur by chemical or
microbiological processes, or by a combination of both. Hunt (1915) attributed the for-
mation of the Sicilian deposits to bacterial sulphate reduction in marine conditions
similar to those existing in the Black Sea. Schneegans (1935) discussed the function
of sulphate reducers in the formation of some French deposits. The great Texas and
Louisiana deposits are said to be the result of microbiological sulphate reduction (Zo-
Bell, 1936), though the evidence is necessarily speculative. More direct evidence was
obtained by Subba Rao, lya & Sreenivasaya (1947) and Subba Rao (1951). They at-
tributed the formation of the sulphur in sulphur- bearing clays (27- 35% sulphur) in cer-
tain coastal areas of India to the catalytic oxidation of sulphide by atmospheric oxy-
gen in the presence of iron. They conducted field trials using cultures of sulphate re-
ducers isolated from the sulphur- containing clay, and demonstrated the formation of
sulphur in near- natural conditions. Murzaev (1950) discussed the microbiological pro-
duction of free sulphur in the muds of Russian lakes and suggested experiments to
stimulate its formation.

Mancuso (1939) studied the geochemistry of an area in the Libyan desert south
and west of El Agheila characterized by salt marshes and many small saline lakes. He
described one lake, Ain-el- Braghi, in some detail ; it differed from the other lakes
examined by him in being fed by a warm sulphur spring (32- 34°C). He noted the abun-
dant escape of hydrogen sulphide, which he attributed to the reduction of calcium sul-
phate by 'sulphur bacteria'. Some of the hydrogen sulphide was oxidised to finely -
divided sulphur which gave the lake a characteristic milky -white appearance. The
sulphur slowly settled on the bottom to form a deposit.

In January 1950 we received a report by Pinfold & Gee (1949) on sulphur- produc-
ing lakes about 20 miles south west of El Agheila. It seemed probable that a detailed
examination of these lakes would be of scientific interest from the point of view of the
processes involved in the formation of natural sulphur deposits, and might give infor-
mation and cultures useful in working out an industrial process for sulphur production.
We visited the area in May 1950 and examined four lakes {Chemistry Research 1950,).
1951).

The Sulphur- Producing Lakes

The four lakes examined were: (1) Ain- ez- Zauia, (2) Ain- el- Braghi, (3) Ain-
el - Rabaia, (4) Ain - umm - el - Gelud.

These lakes lie in a stretch of salt marsh running for about 30 miles in a S.E.
direction from a point 30 miles W. of El Agheila, which is in the S.E. corner of the
Gulf of Sirte 200 miles S.W. of Benghazi. We camped near Ain- ez- Zauia for two

112

nights. The other three lakes were visited all on one day and received only a cursory
examination.

General Description of Ain- ez - Zauia, The lake, reputed to be the most produc-
tive in the area, lay 20 miles S.W. of El Agheila in a long narrow salt plain. There
were two pools of unequal size connected by a swiftly flowing stream (Fig. 1). The
lake was supplied by a warm spring rising in the smaller section, for water of tempera-
ture 30- 52?C (shade temperature 16°C) flowed out of it in two streams, one into a mass

100 yards
(approx)

(—»—♦= direction of water -flow )

Figure 1.
Sketch map of Ain- ez* Zauia

113

of vegetation in the salt plain, the other into the larger section. In brilliant sunshine
(on May 3rci-4th, 1950), the main body of water reflected a vivid milky blue, though a
bottle sample was virtually colourless with a slight haze. The blue of the smaller
section was deeper and more vivid than that of the larger, probably owing to a greater
concentration of suspended particles of colloidal sulphur. Bordering the blue was an
uneven band of red gelatinous material, stretching in some places several yards from
the banks in shallow water. Bulbous formations (4" x 2") could be seen in this red
material and a few red masses were floating in the water. There was a pronounced
smell of hydrogen sulphide and some wind- blown sulphur was visible near the banks.
Salts were crystallising out round the edges of the lake.

Other observers have given descriptions differing from ours in important details.
Pinfold and Gee, who visited Ain-ez-Zauia on October 7 th, 1949, made no mention of
the red material bordering the pools, but reported numerous floating masses of a red
jelly-like material. It is remarkable, too, that Mancuso, who saw Ain - el - Braghi
several times in 1937, makes no mention of the red material, and describes the water
as milky- white; when visited by us on 6th May, 1950, it showed the same colour
characteristics as Ain-ez-Zauia. Clearly the appearance varied considerably with
the season and with the observer.

Detailed Examination of Ain-ez- Zauia. We carried a specially designed 'des-
ert laboratory' which contained sample bottles, apparatus for sampling water from dif-
ferent depths and bottom mud, solartions and apparatus for sulphide and pH determina-
tions and an alcohol lamp. Samples from different parrs of the lake were taken from
an inflated rubber dinghy. Most of the samples were stored for future examination at
Teddington ; others were examined microscopically beside the lake. The principal
results of the examination are given below.

(1) General. The water was saline and had a pH of 7.4. Its probable com-
position is given in Table 1. Noteworthy points are the presence of about 2% NaCl
and of a saturated solution of calcium sulphate, the latter derived from crystalline
gypsum, lumps of which could be detached from the bed of the lake. The presence of
boron was also mentioned by Mancuso. The organic content is low.

(2) Sulphide formation. The evolution of hydrogen sulphide was strong evi-
dence for the activities of sulphate- reducing bacteria. The concentration of sulphide
in bottom water samples was 108mg. HjS/l., and at the surface 15-20mg. HjS/l.
Microscopical examination showed that many vibrios resembling sulphate reducers
were present. Enrichment cultures of these organisms were subsequently obtained
from practically all water and mud samples, from which several pure strains were iso-
lated. Two of these strains were found to be unusual, in our experience, in being un-
able to utilise gaseous hydrogen for sulphate reduction, i.e. they contained no hydro -
genase enzyme (Adams et al.. 1951). It seems reasonably certain on this evidence
that sulphate reduction by sulphate- reducing organisms was taking place in the lake.
It is possible, however, that some HjS entered with the warm spring supplying the lake.

(3) The Coloured Gelatinous Material. This carpet-like material developed
prolifically in the shallow waters, in places stretching several yards from the banks
and sometimes appearing above the water level. It was composed of gelatinous mater -

114

TABLE 1

Probable Composition of Water from Ain-

ez- Zauia

Parts per
million

Calcium bicarbonate, as CaCOj

237

Calcium sulphate, as CaS04

2,613

Calcium chloride, as CaClj

880

Magnesium chloride, as MgClj

1,325

Ammonium chloride, as NH4CI

24

Sodium chloride, as NaCl

19,290

Potassium chloride, as KCl

630

Potassium nitrate, as KNO3

5

Silica as SiOj

•

70
25,074

Total solids

25,250

Organic matter, traces of phosphate and borate

176

PH

ca. 7.3

HjS

see text

Suspended solids (CaS04, SiOj and S)

2,816 p.p.m.

Phosphorus, as PjOs

1.1 p.p.m.

Boron, as B

5 p.p.m.

ial, red on the surface but with green and black matter underneath. In places it was
about Yi" thick. When pierced, the carpet released occluded gas. Microscopical ex-
amination at the lake side showed that it contained many sulphur granules, crystals
(?CaS04), many spiral bodies and a few protozoa; but the bulk of the material was an
amorphous mass of cellular material and sulphur granules. It was later recognised as
coloured photo synthetic sulphide- oxidising bacteria: Chromatium (red bacteria stor-
ing sulphur granules inside the cell) and Chlorobium (green bacteria which precipitate
sulphur outside the cell). These organisms require light for growth ; hence their mas-
sive development in the shallow water only. They were also detected in mud taken
from the deepest part of the lake (about 9 ft.) and in surface water samples.

Subsequently, pure cultures of both Chromatium and Chlorobium were isolated
using the techniques described by van Niel (1931). Old cultures of these organisms

115

formed zoogleal masses closely resembling the gelatinous material of the lake. It is
therefore reasonable to conclude that the latter consisted chiefly of masses of the two
photosynthetic sulphide- oxidising organisms. Both are obligate anaerobes. Their
significance in the production of sulphur in sulphide- containing waters, which are
essentially anaerobic, is discussed below.

(4) Microbial Population. In addition to the bacteria mentioned above and
despite the anaerobic environment, aerobic organisms were present in the lake, pre-
dominantly at the surface (Table 2). Fourteen morphologically distinct types of
aerobe, including a fungus, were isolated from the gelatinous material. Eight types
were isolated from a bottom sample (9 ft.) of water, three from half way and four from
a surface sample.

TABLE 2

Population of aerobic bacteria in samples

from Ain-ez-Zauia

Samples

of water from the middle o

f the lake were

plated

out on nutrient

agar

and

the

number of cc

•lonies recorded. Since there

was a delay due

to the

journey to England,

th

e fig

ures

probably do

not represent the true natural

Depth

Surface

half- depth {ca 1.5m.)

bottom {ca 3 m.)

population.

Count

(colonies/ml.)

181,000

52,000

57

Thiobacillus thiooxidans, an aerobic sulphur- oxidising organism, was not detect-
ted in water and mud samples, but it was present in samples taken above water level.
Thiobacillus thioparus was only sought in the gelatinous material and was not found.
Cellulose- decomposing bacteria were found only in one bottom mud sample, so were
probably not plentiful. No algae were isolated with the exception of a blue -green
alga from another lake, Ain-ei-Braghi. These Mycophyceae are often observed in sul-
phide-containing waters (Allen, 1952).

(5) Fish. Despite the high concentration of salts and sulphide, shoals of
small fish (1" - 2" long) were observed in the two streams leaving the smaller section
of the lake. They were later identified as belonging to the genus Cyprinodon, (Smithy
1952). Mancuso (1939) reports that Desio (1935) collected Cyprinodon from Ain-el-Braghi
in 1930.

(6) Sulphur. The vivid milky- blue appearance of the water by reflected light
was undoubtedly due to suspended colloidal sulphur particles, which could be seen as
highly refractive bodies under the microscope. A deposit of finely divided sulphur,

6" or more in depth, covered the bottom of the lake. This was removed annually by
local Arabs. The possible processes responsible for this sulphur formation are dis-
cussed below.

116

(7) Examination of Other Lakes. Ain-el- Rabaia (roughly circular, 100 yards
diameter) and Ain- el- Braghi (80 x 50 yards) were similar to Ain- ez- Zauia in smell
(H2S), colour (milky blue with coloured gelatinous material stretching 2-3 yards from
the banks) and in producing sulphur. Both were fed by warm springs, but we were un-
able to take the temperatures because both our thermometers were broken. According
to Mancuso, the temperature of Ain-el- Braghi varies between 32° and 34°C during the
year. Ain- el -Rabaia appeared to be cooler.

Ain- umm-el-Gelud was considerably larger (approx. lx% mile) and was differ-
ent in two important respects. It contained very little free sulphur, though there was
a pronounced smell of hydrogen sulphide. There was also none of the coloured gela-
tinous material round its borders, i.e. there had been no mass development of sulphide-
oxidising bacteria. It appeared likely that the non - production of sulphur was related
to the absence of gelatinous material.

Sulphate -reducing bacteria and sulphide -oxidisers were isolated from all three
lakes.

(8) Recovery of Sulphur. In the dry season the sulphur is scooped out by
hand dredges made of sacking into shallow earthen pans at the water edge. The sul-
phur is left to drain and dry for a week. It is afterwards transferred to higher ground
for further drying and is then collected into heaps for transport by lorry. The crude
product contains about 50% sulphur, most of the remainder being sodium chloride (20?0
and silica (12%). The total quantity recovered annually from three lakes is about 200
tons ; the total amount formed would be larger.

Laboratory Experiments

Our observations in Cyrenaica suggested strongly that the formation of sulphur in
the lakes was a microbiological process in which sulphate - reducing bacteria were as-
sociated with the coloured gelatinous material developing so prolifically in the shal-
low waters. We tested this hypothesis by experiments in an 'artificial lake*.

^Artificial Lake' Experiment. A 40- litre tank containing 'artificial lake medium'
(Table 1 supplemented with sodium lactate as organic source) was inoculated with
both the coloured gelatinous material and crude cultures of sulphate reducers from Ain-
ez- Zauia. The whole was incubated at 32°C with continuous illimiination from a 200-
watt bulb 18" above the water surface. At intervals the sulphide concentration was
renewed with saturated HjS- water. The pH was maintained at about 7.5. After 5 days
a red colouration appeared and sulphur deposition began. After a month a thick layer
of gelatinous material containing sulphur covered the bottom of the tank, closely resem-
bling that taken from the lake.

This experiment and others like it showed that sulphur formation could be induced
in an artificial lake medium (with sulphate as the sole sulphur source) by inoculating
with crude cultures of sulphate reducers and gelatinous material from the lake. Micro-
scopical and bacteriological examination of the gelatinous material had shown that it
consisted chiefly of the coloured photosynthetic sulphide- oxidising bacteria Chroma-
tium and Chlorobium. It was therefore very probable that sulphur formation in the

117

lakes was at least partly due to the combined action of two groups of micro- organisms-
(1) sulphate - reducing bacteria, which reduced the sulphate in the lake water to sul-
phide, and (2) photosynthetic sulphide- oxidising bacteria which oxidised the sulphide
produced in (1) to elemental sulphur.

Experiments with pure cultures. Experiments with crude cultures cannot be ac-
cepted as proof that specific organisms are responsible for what occurs. For final
confirmation of the hypothesis, experiments were carried out with pure cultures of sul-
phate reducers and of Chromatium and Chlorobium. All cultures originated from the
lakes.

Mixed pure cultures of D. desulphuricans + Chromatium and of D. de sulphuric ans
+ Chlorobium were prepared in various media, based on combinations of the media
used for the separate growth of the organisms and the composition of the lake water.
The cultures were incubated anaerobically at 32°C in an illuminated cabinet. No
source of sulphur other than sulphate was used. Growth of both pairs of bacteria oc-
curred in nearly all cultures. Those containing Chlorobium and sulphate reducers de-
posited a yellow layer of sulphur (Fig. 2). No such layer appeared in the Chromatium
cultures, but under the microscope the Chromatium cells were seen to be almost com-
pletely filled with sulphur globules. The best yield of sulphur (judged by inspection)
was obtained from D. desulphuricans + Chlorobium grown in the medium in Table 3.

These experiments show that elemental sulphur can be produced from sulphate by
the combined action of pure cultures of sulphate reducers and the photosynthetic green
and red sulphide oxidisers in a common medium. They provide evidence that some of
the sulphur in the Cyrenaican lakes was produced by a similar combination of sulphate
reducers and Chlorobium.

Reducing agent for sulphate reduction. The reduction of sulphate by D. desul-
phuricans requires a reducing agent, either hydrogen or an organic compound such as
lactic acid. The source of reducing agent for sulphate reduction in Ain-ez-Zauia was
not clear since the organic content of the water was low (see Table 1), though the pos-
sibility exists that the continuous supply of this organic material by the spring was
used for reduction. It was also possible that the coloured sulphide -oxidising bac-
teria, which can satisfy their carbon requirements by photosynthesis from COj, pro-
vided suitable organic matter for the sulphate reducers. To test this, mixed pure cul-
tures of D. desulphuricans + Chromatium and of D. desulphuricans + Chlorobium were
prepared with no carbon source other than NaHCOs, and incubated in light at 30°C.
In order to avoid false results due to carry-over of organic material in the inocula,
the mixed populations were sub- cultured at least three times. At each stage they
were inspected for sulphate reducers microscopically. In both cases the coloured sul-
phide oxidisers grew readily, and at each stage sulphate reducers were detected mi-
croscopically. The sulphate reducers were more plentiful in symbiosis with Chroma-
tium, and if thiosulphate was used in place of sulphide as a sulphur source their
presence could be detected chemically by blackening (FeS) after addition of a ferrous
salt.

118

Figure 2.
Formation of sulphur from sulphate by the combined action of pure cultures of sulphate-
reducing bacteria (D. desulphuricans ) and sulphide-oxidising bacteria ( Chlorobium ).

119

TABLE 3

Medium for microbiological sulphur formation from sulphate

D. desulphuricans (Strain El Agheila Z) + Chlorobium sp. were grown in light

at 30°C

g. /litre

distilled water

NajS04 3

Na hydrogen malate 1

NH4a 1

KHaPO^ 1

Mga,.6H,0 0.5

CaClj ' 0.1

NaO 10

NaHCO, 2

Yeast extract (Difco) 1

Trace element solution, 1ml.; pH 7.3

The trace element solution contained the elements below:

mg. /litre

Fe as Fea3.6H20 500

B as HaBOj 100

Zn as ZnS04.7H20 100

Co as Co(NO,)j6H,0 50

Cu as CuS04.5HaO 5

Mn as MnCl2.4HjO 5

Discussion

The available evidence suggests that the formation of sulphide in Ain-ez-Zauia
and in the other three lakes examined was most probably due to bacterial reduction of
sulphate. Sulphate reduction may also have occurred in the springs supplying the
lakes. There are at least five processes by which this sulphide could be oxidised to
sulphur.

(1) Atmospheric oxidation, which occurs in all sulphate- containing waters ex-
posed to air, was undoubtedly responsible for some of the sulphur formed in Ain- ez-
Zauia, but is too slow to account for the high sulphur yield. For example, Ain-umm-
el-Gelud produced hydrogen sulphide but formed very little sulphur though it was ex-
posed to atmospheric oxidation.

120

(2) Oxidation by nitrite formed by bacterial nitrate reduction (lya & Screenivasaya
1944,) may have occurred, but the low nitrate content of the water would not favour it.

(3) Oxidation by Th. thioparus, which is an obligate aerobe, could have yield-
ed sulphur at the air- water interface, but would not have occurred at lower (anaero-
bic) levels. Senez (1951) attributed sulphur formation in air by impure cultures of D.
de sulphur icons to this organism. Th. thioparus was not found in t"he samples
examined for its presence, but it is a fragile organism and may not have survived the
journey to England.

(4) Oxidation by Chromatium undoubtedly occurred in the lakes and was repro-
duced with pure cultures. However, Chromatium stores sulphur granules inside the
cell and would therefore not produce free sulphur unless lysis of the cell occurred.

(5) Oxidation by Chlorobium also occurred and was demonstrated with pure cul-
tures. As Chlorobium deposits sulphur outside the cell, oxidation by this organism
could account for much of the sulphur formed in the lakes.

The importance of Chromatium and Chlorobium in sulphur formation is supported
by three facts :-

(i) The insignificant production of sulphur in Ain-umm- el-Gelud corresponded
with an absence of the coloured gelatinous material. This suggests that the
coloured material, which consisted largely of Chlorobium and Chromatium,
played a key part in sulphur formation.

(ii) Chlorobium and Chromatium are obligate anaerobes and, subject to light being
available, would be active throughout the lake water. Aerobic sulphide oxi-
disers could only function at the surface.

(iii) Chlorobium and Chromatium synthesised organic matter from CO2 and sunlight
which would support growth and sulphate reduction by D. de sulphuric ans.

Thus there was probably a symbiosis between D. de sulphuric ans and the coloured
sulphide oxidisers, in which the sulphate reducers formed sulphide for growth of the
sulphide oxidisers, which in turn made organic matter photosynthetically for the sul-
phate reducers.

The formation of sulphur by this symbiosis is of considerable intrinsic scientific
interest. It also suggests that the larger sulphur deposits in nature may have been
laid down by a similar process. Its economic aspects are mostly obvious. No lakes
other than those in Cyrenaica are known to produce sulphur on a scale sufficient to
justify commercial exploitation. The production of about 200 tons annually is insig-
nificant in relation to the prevailing shortage of sulphur. Nevertheless it suggests a
method of augmenting sulphur supplies. Sufficient is now known of the physiology of
sulphate -reducing bacteria and the coloured sulphide- oxidising organisms to make it
clear that conditions in Ain-ez- Zauia are by no means optimal for the separate activi-
ties of these bacteria, though much more research is required before the best conditions
for their symbiosis and for maximum sulphur production are established. Even so, it is
possible that the addition of organic matter (e.g. vegetable waste) and phosphate might
increase the yield of sulphur. Unproductive lakes such as Ain- umm- el - Gelud might
be made productive by addition of necessary nutrients, which may be simple. More-

121

over, warm artesian springs containing sulphide and sulphate, and lakes in sunny cli-
mates might be induced to produce sulphur.

Present methods of harvesting the sulphur in the Cyrenaican lakes are very primi-
tive and give an impure product. The yield would be improved by better extraction
procedures, if it were economically feasible to use machinery in so remote an area.
The sulphur yield alone would probably not justify such an enterprise, but combined
with exploitation of the carnallite deposits at Marada, some 30 miles to the south, a
viable industry might be developed.

Summary Summary

The continuous deposition of sulphur in certain Cyrenaican lakes is attributed
mainly to the combined action of (1) sulphate- reducing bacteria, and (2) photosynthe-
tic sulphide - oxidising bacteria {Chlorobium and Chromatium). In laboratory experi-
ments, the latter synthesized organic matter for bacterial growth and sulphate reduc-
tion.

Some natural sulphur deposits may have been laid down by a similar process.
Sulphur production might be stimulated or induced in other lakes.

Acknowledgements

The authors are indebted to Miss M.E. Adams and Mrs M. Long for technical assistance,
to the Government Chemist for analysis of the water sample and to Professor J.L.B. Smith,
(Rhodes University, South Africa) for identifying the fish from the lake as Cyprinodon.

They gratefully acknowledge the help of officials of the British Administration in Cyre-
nalca in facilitating their journey to the lakes, the provision of transport by the Cyrenaican
Government and the loan of camping equipment by the Army. They also wish to record their
appreciation of the helpfulness and hospitality of the Arab assistants who accompanied them
on their journey.

This paper is published by permission of the Director of the Chemical Research Laboratory.

References

Allen, M.B. 1952 Arch. Microbiol. 17, 34.

Adams, M.E., Butlin, K.R., Hollands, S.J., & Postgate, J.R. 1951 Research. 4, 245.

Chemistry Research 1930. 1951 H. M. Stationary Of/ice, 85.

Desio, A. 1935 Missione Scient. R. Ace. d'ltalia a Cujra, 1, 369.

Hunt, W.F. 1915 Econ. Geol. 10, 543.

lya, K.K., & Sreenivasaya, M. 1944 Current Sci. 13, 316.

Mancuso, V. 1939 Ann, Mus. LibioStor. Nat. 1, 307.

Murzaev, P.M 1950 C. R. Acad Sci., U.R.S.S.. 72, 343-

van Niel, C.B. 1931 Arch. Microbiol. 3, 1.

Pinfold, E S. & Gee, E.R. 1949 Board of Trade Report.

Schneegans, D. 1935 Congr. Intern. Mines, Met., geol. Appl. 1, 351 (quoted in Amer. Chem. Abs.
1937, 31, 2562)

Senez, J. 1951 Vie et Milieu. 2, 5.

Smith, J.L.B. 1952 Ann. Mag. Nat. Hist. (12). 5, 888.

Subba Rao, M.S. 1951 Thesis for A.I.I.S., Indian Institute of Science, Bangalore.

Subba Rao, M.S., lya, K.K., & Sreenivasaya, M. 1947 Proc. 4th InL Congr. Microbiology, 494.

ZoBell, C.E. 1936 Marine Microbiology, 112. Mass: Chron. Bot. Co.

122

FORESTS, ARIDITY AND DESERTS.

Professor E.P. Stebbing
(Edinburgh)

The Man-made desert is a stern reality which so far has rarely been faced up to.
Man has been the enemy of the forest and of vegetation ever since he learnt to grow
crops for food and to pasture flocks and herds on the countryside. The term erosion is
widely used but very often without a real understanding of what is meant. It is easy
for the public to understand one type when they see a raging torrent tearing out the
heart of a hillside. But there are several different types of erosion which in the end
may result in the same catastrophe, namely the production of desert or arid conditions,
the methods being different.

I propose to confine myself here to one of the commonest and probably the oldest
of the types resulting purely from the acts of man. The most ancient type of cultivation
known to man is in nature and mainly covered with forest, being known in English by the
name of shifting cultivation. This type of cultivation from its very origin has many dif-
ferent names, even in the one country. For example in India there was a different name
for it in N. India, the Central Provinces, Bengal, Assam, Madras, Burma and Ceylon. In
Europe in olden times the same variation in nomenclature existed. But the method was
normally the same. A small piece of forest out of the surrounding mass was felled, the
material as soon as dry enough fired, and the ashes roughly spread over the ground thus
opened out and the seed of a crop sown. At the end of a few years, roughly three to
five, a dense weed growth supervened, or the soil decreased in fertility, or both. The
shifting cultivator then moved and repeated the operation in another piece of the forest.
When the world was young and the population small the forest was the enemy and to ob-
tain pasturage for increasing flocks the forest was fired to get rid of it. It was the in-
crease in population that gradually and imperceptibly had its effect on the forest. In
the temperate parts of the world and especially in the case of the conifers, great forests
were swept away. In the tropical and sub- tropical forests, however, where the forests
consisted of broad- leaved trees of many species intermixed, the forest did not neces-
sarily disappear but suffered a slow degradation. Valuable timber species susceptible
to fire disappeared but the forest remained. The habit of setting fire to the felled
material on the area to be cultivated resulted in other fires spreading into the surroun-
ding forest and the countryside during the hot season. This danger and damage is only
too well known to the modern day forester. The eventual degradation of the forest from
a fine dense high forest to a scrub, variously denoted scrub, bush or savannah, took
thousands of years and was in effect so imperceptible that it had passed unperceived —
in fact was lost in the past histories of the earlier nations who lived in what are now
deserts. Wars helped in this disappearance with the habit of a retreating army of set-
ting fire to the countryside to prevent pursuit. Sites of these old time nations are
known to us. It is suggested here that it was this degradation of the forests and the
drying up of the water supplies which led, in more than one instance, to the final dis-
appearance of the peoples. Time does not allow me to mention such instances. In many
parts of the world the shifting cultivator was therefore, to a certain extent nomadic. In

123

early times it was unnecessary for him to return to the piece of forest he had previously
cut down. In later times he returned but only after the lapse of a century or more when
a forest of second growth species had grown up. It was the increase in population and
the disappearance of so much forest which resulted in the shortening of this period of
return.

In British West Africa this method of cultivation is termed farming and the system
is more fixed, since the people are under the rule of numbers of Chiefs, each having a
fixed area of country with fixed villages under his jurisdiction. In British and French
West Africa between the sea and the southern Sahara and including East Africa and the
Sudan, this method of cultivation is in force over great areas of the so-called bush or
savannah where the forest, for it is still forest though much degraded, is considered
still to be good if it has a height of 30-40 feet and a corresponding density. The ser-
ious trouble is that the populations have greatly increased and with them the grazing
and pasturing herds. But the area of land under a Chief and his people remains a con-
stant. This results in a shortening of the fallow period and with a consequent more
open and shorter scrub growth which produces less ash at the burning and a poorer crop.
But more serious, an interference with the water supplies commences to make itself
felt. This may be summarised as follows ;-

(a) In the past the population of the regions now deserts, or on the way to become
deserts, was very small. The slow depreciation of the soil conditions caused
by a wasteful utilisation took many centuries to make its appearance. As is
known, the nomadic tribes moved away from an area becoming unproductive to
return to it, perhaps many years later, when it had recuperated. It was the ear-
liest form of 'crop rotation' or more correctly 'grazing rotation'.

(b) With the increase in populations and in consequence in the intensity of the mis-
use of the soil the migration from an exhausted area was not followed, on any
scale by a return at a later date. For those later migrations of more numerous
populations were not apparently undertaken till the water supplies had become
so unreliable and the soils so poor that neither the one nor the other were capa-
ble of supporting them.

(c) The decrease in rainfall supplies in springs, streams, rivers and wells precedes
the decrease in the rainfall. Here apparently lies one of the greatest stumbling
blocks to an appreciation in Africa of what is taking place under the misuse of
the soils in the upsetting of the balance of nature. The rainfall becomes unre-
liable or intermittent.

(d) In the past this stage probably continued through the lapse of many centuries.
The fluctuations waxed and waned over long periods. Later generations took
the fitful rainfall as something which had now become a climatic reality which
had to be put up with. Generations of people lived and died under these con-
ditions as they are doing today. Travellers studying the land and its people,
being misled by this fitfulness or intermittancy, reported that in parts the rain-
fall was improving and with this improvement the ground was becoming re-

124

covered with a vegetation after man had migrated from the area, or that records
together with the statements of local villagers showed that the intermittent rain-
fall was due really to climatic oscillations over which man had no control.

(e) This in spite of the more modern evidence upon the ground that the populations
— greatly swollen in numbers — were destroying the soil factors at an enhanced
rate, and that the rainfall fluctuations were of greater intensity; that no one,
scientist or African peasant, could predict the amounts of rainfall which would
occur in the year, or the times at which within certain periods, it would fall.
From the practical viewpoint of the administrator these are the points upon
which clarification is required and upon which concentration would appear
necessary.

Desiccation is a much debated term in Africa and elsewhere in the world. It is
held to be due primarily to the over- utilisation of the vegetation covering of the soil
under which productivity is reduced, the decrease of water supplies in the springs,
streams, rivers and wells, the sinking of the water table in the soil strata, and de-
creases in the rainfall. It may be due to (a) the presence of neighbouring deserts and
sand penetration; (b) erosion in varying forms through over utilisation of the soil ; (c)
a combination of (a) and (b), accompanied usually by dry hot or cold winds.

Lavauden (Les Forets du Sahara) does not use the word 'climate' in connection
with the process which he terms dessechment, by this meaning only the progressive
diminution of surface and subterranean water supplies. He does not discuss the rela-
tions existing between dis- afforestation and desiccation. Kennedy Shaw in consider-
ing this matter for Southern Libya says it is one of the present day increase of desert
conditions due entirely or largely to the acts of man'. In northern Nigeria desiccation
to whatever agency or series of agencies it may be subjected, is an accepted fact. Here
it may, it is suggested, be attributed to a combination of erosion sur place (for the more
level country) coupled with the lowering of the water table in the soil, the falling off of
the rainfall, and sand penetration from the Sahara.

In some parts of Africa desiccation aided by sand penetration, may be due, in
part, to blown sand from drying off river banks or diminishing lakes.

What is drought? An ordinary definition of 'drought' in the English language
would refer to months of dryness at periods when the ordinary average rainfall is re-
ceived. In Europe so far as records go, there appear to be years of wetter months fol-
lowed by years of drier ones, and we speak of 'drought' in its true sense — more or less
temporary climatic changes over which man can have little control. Can the word be
equally applied, or applied with its true significance, to the upsetting by man of
Nature's balance between the soil and its covering and the water supplies, with the re-
sulting dislocation in the regular average water supplies received in the rainfall and
from springs, streams, rivers and wells of the region? So far as we have knowledge and
evidence of the results attendent upon this intervention of man in Nature's balance it is
becoming more and more evident that periods of so-called drought will not be followed
by consistent wet periods, as has been sanguinely hoped in some quarters in connec-
tion with the major catastrophes facing man in certain parts of the world. Such hopes
are illusory. However the following proposition may be enunciated :- As a result of

125

erosion in one of its forms, water supplies have decreased either in moisture in the
upper layers of the soil owing to the lowering of the water table, thereby, affecting
wells; or decrease in, or cessation of, springs ; or disappearance of the water in
streams during the dry months of the year and lowering of the water level during the
same period in the smaller and larger rivers. The rainfall has also decreased in annual
amounts, though the amounts of such decrease may be slower in making their appear-
ance; but, more alarming, this rainfall has become capriciously intermittent in its sup-
plies — no man being able to forecast the amounts which will be obtained within the
year; or often, within limits, at what periods.

This is not 'drought' in the ordinary accepted sense of that word. I would term it
the 'Intermittent Stage' in rainfall supplies. It may be suggested that if the fact of the
intermittency of the rainfall, developing at a certain stage in the degradation of the soil
and its covering, be accepted as a factor of importance in this decrease in fertility, we
can start from a point at which we all are voicing the same position of affairs and can
commence, according to the different types of erosion, the business of combating the
danger. Lavauden (Les Forets du Sahara) "wrote'ln the middle of the Quaternary period, an
epoch which it is impossible to date precisely, the Sahara was a very humid region,
the fluvial system was of a particularly powerful type, allied without doubt to very
abundant precipitations. Today all these river beds are dry, and only the largest retain
underground water of which the amounts constantly diminish — slowly perhaps but in-
evitably — owing to the equilibrium existing between the precipitation and evaporation.
An important question is to determine at what epoch the dis- equilibrium between the
two commenced to make itself felt; in other words at what period desiccation commen-
ced to become seriously apparent.' This represents exactly what I term 'Intermittent
Rainfall '.

It may be asked 'How can this stage be recognised on the ground?' Examples are
only too plentiful. If we examine the regions bordering the southern edge of the Wes-
tern Sahara (British and French colonies) it will be found that a stage is reached in the
rainfall conditions where dependence upon them for crop production can no longer be
placed with ordinary confidence. For certain localities in Northern Nigeria and the
French Niger Colony the local population complain of violent winds which, blowing at
the beginning of the rainy season about May or June, bring blown sand on to their
fields. The millet crop is sown during the first rains. Should the previously normal
second rains arrive up to time, when the seedlings have shown above ground the roots
of the latter fix the sandy soil (it will be noted that there is already a sandy covering
blown from the adjacent Sahara Desert covering the normal soil surface) and the growth
of the crop proceeds successfully. Should the second and main rains not come up to
time, however, sand carried by the strong winds covers the seedlings and kills them.
The operation of sowing has then to be undertaken a second time and, maybe, a third or
fourth. Indeed cases are on record when the seed has been sown as many as ten times!
This example would appear to be a strong argument in favour of the postulate here ad-
vocated that a time arrives when the rainfall becomes intermittent and man can count no
further on its reliability.

It has become apparent that in some quarters opinions are held that there can be
no analogies between, say, the desiccation being produced in parts of Africa and the

126

conditions under which the Dust Bowls have arisen in America or the soil drift taking
place in Southern Australia. If, however, we trace all categories of erosion back to
their origin or commencement it is possible to show in most cases that in this over
utilisation of the resources available, with the consequent upsetting of Nature's bal-
ance, a stage was, with few exceptions, always rea-ched at which the factor which
governs all production and life, the water supplies, commenced to become intermittent.
One way or the other this stage must have made its appearance.

Under excessive utilisation of the soil then, the rainfall supplies in the region
fall into a delicate stage of oscillation,. It has been mentioned above that the first
decreases noticeable to man in the local water supplies show themselves in a lowering
of the water table in the soil owing to a decrease in the water in springs or their 'drying
up', including the wells, the drying up of streams in the hot months of the year, and the
lowering of water levels in the rivers. These partially precede the 'falling off in the
rainfall supplies. Man's life is short, his official service life shorter still, these gra-
dual manifestations occur almost imperceptably, though the speed is now- a- days much
accelerated, and long periods have passed before the balance delicately dropped on the
wrong side and man's chance of repairing the damage done is gone — for the desert or
conditions of aridity have won.

What of the present day? The world has been shocked at the appaling conditions
which have been produced in the Dust Bowls in the United States and Canada in a brief
half century of over- utilisation of the soil assisted by all modern developments. And
yet whether it is a result of fifty years or thousands of centuries the outcome is the
same and the dangers are the same whether it be on the southern edge of the Sahara
where sand penetration assists desiccation, the Creeping Desert in the Sudan, the level
country at the foot of rapidly eroding hill ranges, the periphery of the Dust Bowls in
America, or the confines of the soil drifts forming desert in Australia; and also the
danger to the neighbouring agricultural or pastoral country is the same, namely the ex-
tension of the existing destroyed regions over their boundaries owing to gradual further
desiccation, dust storms and so forth. It is in these neighbouring lands that the inter-
mittent rainfall stage has been reached and man is called upon to make his effort to
restore the balance of aforetime before it is too late.

Mr William Vogt is well known as the author of Road to Survival. He is a member
of the Chief Conservation Section of the United States Forest Service and has spent
nine years studying forests and forestry conditions in South America. He prepared a
memorandum for a Sub- Committee of the Forestry Section of F.A.O. which met in
Geneva in August 1947. His memorandum showed that against generally accepted
opinions amongst foresters, the Latin American Republics do not hold the enormous
forestry resources they were supposed to. He bases his reasons on the physical geo-
graphy of Latin America and the pattern of human settlement. The tropical lowlands
are not desirable for human settlement owing to prevalent diseases ; also a high pro-
portion of the area is unsuitable for agriculture because of excessively concentrated
rains that leach minerals from the soil, long blistering dry seasons that thoroughly
desiccate the vegetation once the forest has been cleared, and because high tempera-
tures tend to oxidise very rapidly organic materials in the soils. It is not at all un-

127

usual for a tract of land to pass from virgin forest to abandoned bush, when the low-
lands are cultivated, within a period of eight or ten years.

Dealing with the coffee and tea planter in Ceylon and Madras, Colonel Beddowe,
Conservator of Forests, Madras, wrote in 1876 'It must not be supposed that coffee is
at all a permanent cultivation — many deserted estates show that it is very little better
than the shifting cultivation of the hill man. It pays a coffee planter to take up a tract
of primeval moist forest on our mountain slopes for a few years; he gets bumper crops
the third, fourth and fifth years but denudation of the soil goes on rapidly and it does
not pay him to keep it up many years. Can we restore the grand old forest with all its
climatic influences? A thorny wilderness takes its place'. This was written 76 years
ago. Vogt has presented the same grave picture as happening in the world today. Popu-
lations, he continues, have therefore to concentrate above 700 metres. The best agri-
cultural land in the higher altitudes lies in the interment valleys but in part, because
there is an insufficiency of land and the best lands are in the hands of wealthy owners,
the mass of farmers are crowded on the slopes where they practice shifting cultivation
(milpa) which has resulted in a very high percentage of slopes throughout Latin America
becoming de- vegetated. The usual results are of loss of soils and aridity, with flash
floods with their consequent scouring action, interspersed with periods of little or no
water in the rivers. Mr Vogt continues 'The land settlement pattern in Latin America
has resulted in the extremely grave situation that there exists probably from twenty to
forty million displaced persons. They are displaced in the ecological sense, namely
that their present relationship to the land is destroying it at an accelerating rate, not
only in the highland areas where they live, but in the lower areas affected by river
flows. Many millions of acres of soil have become seriously eroded in Latin America
and, as Professor Stebbing has described in the case of the Sahara, deserts are on the
march. One of the worst instances of this land pathology is in St Salvador, where two
million people, increasing at the rate of forty thousand a year, have available for agri-
culture only about an acre per capita, and much of this land is of low productivity'.

How many displaced persons, using the term in Vogt's sense, are there at the
present moment in Africa? It is a natural query. V^e may also ask at what accelerated
pace, compared with the past, are the processes of degradation in erosion, desiccation,
sand penetration-the term varies with the conditions being produced — proceeding in
Africa today? According to French investigators the Sahara has advanced southwards
during the last three centuries at the rate of half a mile a year!

From the studies I have made I would record the opinion that a belt of country in
Africa betv/een the 13° and 15° parallels of latitude and stretching from French Senegal
in the west at EI Obeid and Kosti on the White Nile in the Sudan to the east, is at the
present day in the Intermittent Rainfall stage, and is still in a condition when man may
undertake operations to stop further degradation and the onward march of the desert.
It is impossible, nor is it necessary, to deal here with practical methods which could
be undertaken.

128

THE INFLUENCE OF CLIMATIC FACTORS ON THE REACTION OF
DESERT SHRUBS TO GRAZING BY SHEEP

Professor H. C. Trumble, and K. Woodroffe*
{V^aite Agricultural Research Institute, University of Adelaide)

The arid pastoral areas of southern Australia with a mean annual rainfall of less
than 10 inches are characterized by species of Atriplex and Kochia which occur ex-
tensively as low growing shrubs. These plants are well adapted to withstand rain-
less periods of long duration, and they provide reserves of feed in unfavourable sea-
sons. The shrub cover has been greatly depleted by grazing sheep for approximately
a century, and much land has been completely denuded. As viable seed of these
shrubs no longer occurs in significant quantity, natural regeneration of perennials
is rare on such country. Artificial seeding is costly and hazardous owing to low and
uncertain rainfall. Sheep - raising in these areas is now largely dependent on annual
herbage, and thus has become more subject to the risks of unfavourable seasons.

In other parts of the pastoral country, relatively undamaged associations of
perennial shrubs still occur. Fairly extensive areas carry a moderate to sparse popu-
lation of shrubs which produce seed in favourable seasons, making possible some
natural regeneration. Investigations of pasture management here should throw light
on the reaction of desert shrubs to grazing by sheep, and hence suggest more effec-
tive and permanent systems of utilization in arid regions. A previous investigation
(Osborn, Wood and Paltridge, 1932) has indicated that pastures of saltbush (Atriplex
vesicarid) may be improved under certain conditions of grazing.

Researches were commenced in 1941 at Yudnapinna Station, approximately 250
miles north-west of Adelaide; this centre is within the extensive North-west Pas-
toral District of South Australia. Investigations of the ecological factors concerned
with the grazing management of bluebush {Kochia sedifolia) and associated species
were undertaken on a long- term basis.

Climatic Factors

A meteorological station (Fig. 1) was established at Yudnapinna in October,
1938 and daily records of air temperature, relative humidity, wind mileage, free water
evaporation and rainfall have since been maintained. Rainfall had been measured
previously from 1885 onwards, thus giving an uninterrupted record to date of 67 years.
The mean annual rainfall is 7.92 inches with a range from 2.36 inches to 18.08 inches
per annum, and a modal frequency of between 5 and 6 inches.

Although the mean rainfall is fairly evenly distributed through the year, winter
rains are more frequent and dependable ; the mean monthly rainfalls for the period
May — August are slightly greater than for the remainder of the year. In summer the
precipitation tends to occur spasmodically as extremely fortuitous but heavy rains.

* The authors have been jointly responsible for the planning of the work described and the
analysis of the data recorded ; but the detailed investigations and most of the observations
made are attributable to Mr K. Woodroffe, and will be the subject of a subsequent paper by him.

129

Figure 1.
The Meteorological Station, Yudnapinna.

Summer temperatures are high, and the daily mean temperature from November to
March exceeds 70°F. The rate of evaporation from a free water surface is high during
summer, with a maximum of 14.5 inches in January; evaporation falls to a minimum
of 2.6 inches in June. (Table 1.)

Due to lower evaporation, winter rains are more effective for plant growth than
summer rains of equal magnitude and use has been made of the index P/E*^ (Pres-
cott, 1949) to define the minimum influential rainfall for each month. On the basis
of observations on plant response and monthly records of soil moisture in the root
zone, a monthly value of P/E^-^^ = 0.2 has been found to indicate the lower limit
above which soil moisture tends to become available, and new growth therefore pos-
sible. (Table 1.) Rainfall in excess of O.^EO-75' constitutes water available for trans-
piration, to which the amount of herbage growth can be related.

The pattern of influential rainfall for the period 1885- 1951 (Fig. 2.) indicates
clearly the predominance of favourable growing conditions during the restricted period

* For the month of June, the value of 0.41 derived from the above expression has been in-
creased to 0.5 to allow for the more frequent light falls of only a few points each in this
moDth.

130

/NFLUENT/AL fPA/NFALL

- /A/CHES PEP MONTH AVA/LABl-E FOP TPANSP/RAT/ON

885 1890 1895 1900 1905 1910 1915 1920 1935 1930 1935 1940 1945 1950

z

EFFECTIVE

INCHES

PER

EFFECTIVE

1885 1890 1895 1900 1905 1910 1915 1920 1925 1930 1935 1940 1945 1950

Figure 2.
Influential Rainfall — inches per month available for transpiration — Yudnapinna.

of May to August inclusive, and particularly in June. The expected occurrence of
monthly rainfall exceeding 4S.D ^'^^ (Table 1.) confirms the greater reliability of
favourable soil moisture conditions in winter, (Cornish, 1952). The monthly values
for influential rain have been summed to give a seasonal value; the mean influential
seasonal rainfall for 67 years is 2.85 inches and the mean number of months with
effective precipitation is 3.2 months per annum. Years with a seasonal value of less
than 2.0 inches of effective rain have been classified as drought seasons, and those
with a value greater than 2,0 inches have been classed as seasons favourable for
herbage production.

The amount of growth that can be made on limited quantities of available mois-
ture has been assessed in field and glasshouse studies of transpiration in relation to
growth. These indicate an average value for Kochia and Atriplex spp. of 3 cwt. dry
material per acre inch of water transpired, and this appears to be capable of suppor-
ting a sheep on 20 acres, or 32 sheep to the square mile for a period of 12 months.

MS.D^-'''^ is equivalent to 0.4E°-^5 (Prescott, et al., 1952).

131

1

ITi

r-

•

O

^

o

u

t-H

C\

(N

xr

o

D

•

•

•

•

•

Q

1— t

I-H

I-H

r-H

(N

•

rsj

^

rcv

>

r^

r-

I-H

(N

O

o

•

•

•

Z

pOl

o

I-H

I-H

t— I

r--

I-H

ir\

CN

•

O

■^

r^

4-»

u

00

00

•

o

O

o

O

fN

^0

CN

I-H

<vi

rO

^

I-H

00

00

w

00

-^

■^

00

I-H

Q<

•

HI

f-H

CN

NO

d

CO

t-H

1^

00

(N|

•

rH

CN

CN

ao

00

1^

I-H

NO

fN

<5

d

1— (

-<r

d

•

•<r

00

>,

CN

o

o

,_t

^

o

WN

■<'

"a

.

•

•

•

»— ^

o

I-H

f-H

CO

o

\r\

NO

f-H

a

00

00

r-

1^

a

00

cs

\r\

•<J*

ITN

C\

3

Gs

(Nl

•

fN

d

•

m

Cn

i-H

00

Cn

c
c

2

d

00

I-H

00

•

(N

•

NO

d

NO
(N

«

1— I

t-H

\r\

w

a

(N

(N|

CQ

•

i-H

00

•

00

00

o

<

3

<

1—4

m

NO

d

of

i-H

-O

00

CN

ca

•

00

fS

»-H

-o

cd

<V1

\r\

fS

I-H

o

(J

S

rvi

CN

•

d

•
I-H

•-H

I-H

r-

f-H

s

cs

fN

• -H

•

r^

xr\

rr>

*— <

^

(VI

fA

xr

fS

O

^1

(U

•

•

u<

ro

NO

rH

I-H

^H

I-H

t^

t-H

r<N

00

•

CN

1-H

(N

a

CN

\r\

\r\

\r\

o

CIJ

•

•

•

•

v-H

00

I-H

r-H

C

NO

I^

-o

w

rt

e

-o

■a

^— \

§

u

w

i^

^4H

IH

3

IL,

om s
hes)

o

O

0

^^

>> •>

^

(U

w

u

OJ

Ui r )

-H O

3

u

a

on f
(in(

M

babi
S.D.

«4H

o

<U

a

"-^

•-^ IH

o -^

4-1

c

4J

e

«-i

IH

«-k-i

c

vapor
orimei

(J

a

ated p
;eding

rt

fc^

(fl

h>

—1 V

3 U

c

§

c

c >

o

U M

(fl

(«

« z

u

—J 4>

(U

V

u

V ^

•

rt

2

2

2

2

u

O

CO

(N

CN

00

1

1

CN

ir\

CO

00

Cn

CO

I-H

I-H

>--•

■ ■

^^

^.^

— H

rt

ta

VIH

V4-I

c

c

rt

<a

u

u

^^

_,

a

rt

3

3

a

C

§

G
(«

c

a

rt

ea

4J

lU

2

2

132

Examination of the seasonal pattern indicates a tendency for seasons of high or
low rainfall to bunch together. The sequential nature of dry and wet winter seasons
is apparent; the May - August rainfall, and particularly the June rainfall, exhibit a
marked suggestion of periodicity. The annual rainfall of the Station, plotted as 10-
year running means (Fig. 3.) and extrapolated back prior to 1885 by correlation with
other records, and the notes of early, explorations, appears to indicate a long-term
repetitive pattern of 20- 25 years of increased rainfall, (e.g. 1870 — 1895) followed
by periods of 10-12 years of much lower rainfall (e.g. 1895 — 1905).

The importance of the climatic factors and particularly the rainfall sequence to
the behaviour of desert shrubs, and their utilization for sheep grazing, will be indi-
cated in the subsequent discussion.

Pasture Management.

Investigations of pasture management at Yudnapinna have been based essentially
upon a long- term grazing experiment in which bluebush (Kochia sedijolia) pastures
have been subjected to differential rates of stocking with sheep, on thirteen plots of
160 acres each since April 1941. The pasture (Fig. 4) consists of an open com-
munity of Kochia sedifolia with a mean density of 580 bushes per acre at the commen-

SHEEP POPULATION OF N.W. DISTRICT
IN RELATION TO RAINFALL AT YUDNAPINNA.
1870 1880 1890 I900 1910 1920 1930

I860

1940

1950

I ' I ' I 1 1 I I I I ' ' ' ' I ' ' ' I I ' ' ' ' I ' I ' ' I

I860 I870

1880 1890 1900

^ '*'' ^ '■■ ^ t ' ^ '' t '''■ I ''''!'■■' I '■'''■■■■ I ■

1920 1930 1940 1950

Figure 3-
Sheep population of N.W. District in relation to rainfall at Yudnapinna.

133

Figure 4.

Kochia sedifolia pasture, with myall trees {Acacia sowdenii); grass and other annuals
complete with the bluebush which is not growing vigorously.

cement of the experiment. Myall trees (Acacia sowdenii) and a number of shrubs of
minor importance are associated with the bluebush. During drought years the ground
between the bushes is devoid of plants, but in seasons of favourable rainfall, herbage
and grass provide a sparse to moderate cover. The soil is of an arid calcareous type,
with a brown loamy sand surface ; the texture increases gradually to a sandy clay at
about 3 feet. The effects of treatments have been measured primarily by weight esti-
mates (Woodroffe, 1941) of the amount of edible green forage present in the spring of
each year; trends in the numbers and production of bluebush have been used as the
main indicators of the results of grazing. Live weights and wool production of the
sheep have been recorded.

For the purpose of the present paper, discussion will be confined to five treat-
ments, viz :-

1. Control with no grazing.

2. Continuous light grazing at the rate of 24 sheep per square mile.

3. Continuous moderate grazing at the rate of 48 sheep per square mile.

134

4. Continuous heavy grazing at the rate of 72 sheep per square mile.

5. Intermittent very heavy grazing at a mean rate of 128 sheep per square mile;
average 90 per square mile 1941 - 1946 and 168 per square mile 1947- 1951.

The amounts of bluebush forage and the numbers of bluebush per acre over the
period, 1940 (before commencement of grazing) to 1951 are shown in Tables 2 and 3.
The weight of bluebush on all treatments fluctuated from season to season and was
related to the influential seasonal rainfall (Fig. 5) ; the green weight of edible blue

41 -42 43 '44 '45 '46 47 '48 49 '50 1951

1940 '41

42 -43

'44

•48

■45 46 '47

Figure 5.

Weight of Kochia sedifolia, estimated in September, on plots grazed
at different rates of Stocking — Yudnapinna.

Actual weights of forage are shown for the ungrazed control plot to indicate
seasonal fluctuations. For the grazed plots weights of forage have been
adjusted relative to the ungrazed plot to eliminate the direct effects of season,
and the mean of the ungrazed plot (broken lines) is shown for comparison.

135

i-H

o

o

o

o

o

l^

CN

-^

o

rsi

1^

CA

I-H
r—J

f-H
»-H

I-H

^H

f-H

O

O

o

o

o

o

\r\

NO

r-

^

NO

o

C\

o

CN

pr\

(N

r-

1—*

'"'

I-H

I-H

I-H

CN

o

o

o

o

o

-«f

(TN

NO

\r\

r-

tr\

C\

r-

CN

o

00

00

«

I-H

I-H

c

00

o

o

o

o

o

G

^

r^

o

rr\

o

I-H

CN
I-H

-<i'

m

1^

u~\

00

c

s

r--

o

o

O

Q

o

■<r

00

CN

(N

o

f-H

n

CN

CN

CN

m

CM

~^

(A

»— t

I-H

I-H

I-H

4-»

c
E

4^

NO

o

o

o

o

o

<a

^

l^

r<~\

I-H

r^

HT

c;

CN

^

r<-i

r-H

00

ro

Ul

»-H

I-H

I-H

(N)

I-H

CN

w

M)

C

N

u~\

o

o

o

o

O

■^

r^

r~4

I-H

o

Cn

CN

C\

r-

CN

r--

r-

«)

I-H

w
C
V

■<r

o

o

o

o

o

<U

-'J'

00

I-H

CN

(N|

NO

(N

Cn

.-H

ir\

ITS

•^

•^

(VI

'•V

a

ffi

o

rr,

o

o

o

o

o

<

•—4

•<r

o

(Nl

^

rr,

o

H

1-1

4.*

Cn

I-H

CN

00

CN

00

CN

cs

B

(N

o

o

o

O

o

^

I-H

•^

m

NO

t--

c

CN

-<r

r^

^

CO

(N

V

I-H

I-H

I-H

f-H

I-H

I-H

turf

&

<u

I-H

o

o

O

o

o

o

■^

o

00

CN

NO

-<r

Cn

m

o

(N

O-l

1^

•

*-4

^H

I-H

*"*

r-H

I-H

r^

o

o

o

o

o

o

Kv

■^

ITN

o

CN

CN

^

«iO

CN

NO

\r\

V-N

NO

NO

(9

r-H

bi

O

M-t

J3

tA

3

J3

a>

^^

^— ^

•-^

y-N

3

V

<u

u

<u

•-H

EQ

• ^

•-H

• -H

• >H

6

e

e

e

•

•

,

,

cr

cr

cr

cr

M

w

v>

CO

'v^

"^^s^

^>^

"^^

^

00

(^

00

60

rvj

■^

1^

r-4

c

I-H

^

u

o

>.

^J

>

m

u

rt

*-*-«

4-»

a

•h

V

-a
o

OJ

o

V

4H

ta

.— »

-C

ac

HI

(U

oi

z

J

2

X

>

136

1— I

o

o

o

o

o

.

lr^

rr\

ir\

r^

o

(>j

C\

NO

■^

^o

r^

fN

r-H

f-H

O

O

o

o

o

o

^r^

ro

ir\

00

rfN

o

C\

NO

•^

NO

r-

fN

»— (

p-H

CN

O

O

o

o

o

-<r

rri

u^

00

m

C\

C\

NO

^

NO

t^

rr\

i-H

I-H

00

O

o

o

o

o

-<r

00

CN

o

CO

CM

a\

NO

■^

1^

00

NO

f-H

I-H

o

• w

a,

f^

O

Q

o

o

o

ea

-^

m

00

■<r

(N

1^

c

C\

r-

l/N

o

I-H

r^

-a

^H

I-H

r-t

I-H

3
>-

\o

o

o

o

o

o

■<T

vy

u-\

r--

\r\

CN

c

CN

r~-

u-\

00

o

•^

4J

t-H

I-H

f-H

4-(

ca

u

u-\

o

o

o

O

O

fc4
4^

■<r

rO

o

NO

Cv

\r\

cs

I--

\r\

NO

\r\

(TN

60

i-H

C

'n

a

a

. "=t

O

o

O

o

o

-<r

CN

00

O

-<T

o

rO

C

CN

f-H

\r\

^

NO

ir\

i^

ffl

m

o

o

o

O

o

<

-a

■<r

(N

00

I-H

rr,

\r\

H

CN

NO

•<r

NO

\r\

\r\

a

1-H

o

u

tN

o

o

o

o

O

<fl

'^

00

o

r-

CN

o

C\

i-H

1^

■<r

U-^

^

u-\

a

M

3

<— *

o

NO

o

NO

O

R

a

^

CN

\r\

fo

u-\

■^r

'<1'

U

p— 1

3

3

O

o

o

o

o

o

o

■<r

f-H

CN

o

■<r

rM

W

CN

lr\

(N

ITS

-^

■<r

h

p-H

11

J3

3

z

,_,

^,^

^^

^^

ii

U

t)

V

*-H

• -^

• •^

e

G

B

B

•

•

,

•

cr

O"

O"

cr

w

(0

w

M

^^

■ ^

"^v^

^

00

rsi

00

60

fN

■<r

r-

fM

e

I-H

j2

u

o

>.

u

>

CO

V

a

<-Ul

4-»

V

-0

V

o

4-*

w

60

IH

US

ra

o

iJ

V

oi

Z

J

2

p^

>

137

bush in September on the control plot ranged from 14501b. per acre in 1946 to 4701b.
per acre in 194S, with a mean of 9701b. per acre over j2 seasons. The seasonal fluc-
tuation in the amount of all other forage was much greater; its green weight on the
ungrazed plot ranged from 23001b. per acre in 1946 to 151b. per acre in 1948.

It has been found possible by correlation and regression to eliminate the direct
effect of season on the weight of bluebush forage in the grazed plots, and to convert
these weights to a 'constant season' basis, represented by the mean of the control
plot. (Fig. 5)

After accounting for seasonal variation, the amount of bluebush on the lightly
and moderately grazed plots remained practically constant over the period 1940- 1945,
whereas there was a downward trend in the weight of bluebush on the heavily and
very heavily grazed plots. This trend was associated with heavy grazing of bluebush
during three seasons of low influential rainfall, including two drought years, 1943,
1944, followed by only moderate rainfall prior to September, 1945. All plots responded
to the favourable rains of 1946, but after eliminating the response due to season alone,
there remained an extraordinary increase in the weight of bluebush on the moderately
to very heavily grazed plots, which can only be ascribed to the more intensive graz-
ing with sheep at the higher livestock concentrations. There was a small but less
marked residual response on the lightly grazed plots. Since 1946, the seasons have
been unusually favourable, with the exception of 1948, and on the three plots sub-
jected to moderate to very heavy stocking the weights of bluebush forage, after some
recession from the peak of 1946, have fluctuated about levels considerably in excess
of their initial values. The high level has been maintained on the most heavily grazed
plot despite an increase in the mean stocking rate from 90 to 168 sheep per square
mile. The amount of bluebush on the lightly grazed plot has shown a gradual upward
trend over this period.

Part of the increase in production of forage by bluebush on the grazed plots, rela-
tive to the ungrazed plot, is undoubtedly due to the substantial increase in the num-
bers of bushes on the former plots from 1944 to 1946. (Table 3.) A heavy seeding of
bluebush occurred early in 1945, and large numbers of seedlings germinated on all
plots, including the control. The greater vigour of the seedlings on the grazed plots,
and particularly where grazing was heaviest, was most noticeable and accounted for
the larger number of young plants which became established on these plots. (Fig. 6)
flfithin the grazed plots, a greater increase in numbers of young plants occurred
towards the southern end of the plots. (Table 4.) Sheep graze into the prevailing
southerly wind, with the result that this part of a paddock is inevitably most heavily
grazed.

The increase in the weight of bluebush per unit area was not entirely due to in-
crease of numbers, and indeed, on all the stocked plots except those most heavily
grazed, the numbers have since fallen to levels comparable with the ungrazed plot.
The increased vigour and leafiness of the older bluebush on the heavily grazed plots,
compared with the ungrazed control, have been outstanding. Moreover, the larger
numbers of stock on the heavily grazed plots have been successfully carried in prime
condition, and with only a slight reduction in wool production per head.

138

This result can only have become possible through higher protein production per
unit of soil water available for plant growth. Increasing nitrogen concentration and
protein formation per unit of water transpired would have been favoured by heavier
defoliation (Trumble, 1952), and the greater concentrations of livestock might also
have tended to enhance nitrogen enrichment of the soil per se.

TABLE

4

Increase in numbers of bl

lie bush per

acre

from 1944 to 1946 over south*

srn and

northern

halves of plots

Rate of Stocking

Nil

Light

Moderate Heavy

Very heavy

Southern half 190

110

400 690

1500

Northern half 130

30

140 340

480

rtfia^i^.

*i<gt^^ia»

Figure 6.
Heavily grazed bluebush in foreground showing active regeneration of bluebush; moderately

grazed plot with less regeneration in background.

139

The improvement of shrub pastures of Kochia and Atriplex under moderate to
heavy but not too severe stocking has been frequently noted by pastoralists, and was
recorded in the investigation of the effect of grazing on Atriplex vesicaria (Osborn,
Wood and Paltridge, 1932). The greater vigour of grazed stands has been attributed
to pruning of the bushes leading to the production of young shoots. This probably
results in the conservation and more efficient subsequent use of soil water reserves,
but does not in itself account for the increased establishment of young plants. The
Chenopodeaceous shrubs are less palatable to sheep than much of the annual herbage,
and it is possible that part of the increase in number and production of shrubs may be
due to the control of competing annuals by grazing. In the present experiment, how-
ever, the weight of forage other than bluebush on all the grazed plots except the very
heavily stocked plot, has at least equalled that on the control plot over the period
1946- 1951.

Previous investigations with the perennial grass Phalaris tuberosa (Richardson,
Trvunble and Shapter, 1932) showed that repeated defoliation trebled the amount of
nitrogen in the edible above-ground portion of the plant compared with that con-
tained in previously ungrazed herbage. Approximately one half of this increase was
secured by the herbage at the expense of the nitrogen in the root system and other
portions of the plant not available to livestock. The remainder of the increase was
assumed to be the result of increased uptake of nitrogen from the soil.

A major factor responsible for the marked improvement of the bluebush on the
grazed plots relative to the ungrazed is considered to be the increase of soil fertility
through the grazing animal, a principle that is widely accepted on pastures in areas
of higher rainfall, particularly where phosphatic fertilizers are applied (Trumble and
Donald, 1938; Sears and Goodall, 1948), but which so far as the authors are aware
has not been recognized in arid regions.

The Fertility Status of Arid Soils

Although climatic factors are of paramount importance in determining the produc-
tivity of semi- arid and arid regions, it is now evident that the limiting role of soil
fertility must also be considered. The level of fertility is in part a reflection of pre-
vailing climatic conditions which limit the physical and chemical processes of soil
formation as- well as biological and microbiological activities. The breakdown of
rock minerals proceeds slowly, profile characteristics are generally not well devel-
oped, and soluble salts may accumulate; soil organic matter and soil nitrogen are in-
variably at a low level. In some cases the geological parent materials of the soils
are low in essential nutrients and the soils developed from them are correspondingly
deficient.

In the arid pastoral areas of southern Australia the principal soil groups are arid
red earths, stony tableland soils and arid calcareous soils. Atriplex vesicaria and
Kochia planifolia are the more important shrubs associated with the two former groups,
while Kochia sedifolia is associated with the arid calcareous soils. The soils of
the North-west Pastoral District have been developed on Cretaceous shales, and
Jurassic sandstones of low nutrient status (Jessup, 1951), whereas in the North-east
Pastoral District the soils are associated with richer Proterozoic and crystalline

140

Archaean rocks. Pedogenetic considerations, therefore, indicate a lower fertility
level for the north- west soils and this has been confirmed by practical observation
of herbage responses following rains and the stock fattening capacity of the two
districts.

As part of the present study, the productivities of a range of semi- arid and arid
soils were compared in a series of pot culture tests with that of a moderately fertile
red-brown earth soil from the Waite institute and representative of the better agricul-
tural areas of the State. (Table 5- Fig. .7)

The results indicate the low fertility of the arid soils and particularly of those
from the North- west District. Marked responses were obtained to soluble nitrogen
on all the arid soils although there was little response to phosphorus ; even with the
heavy application of nitrogen and phosphorus, however, the productivities of these
soils were considerably below that of the red -brown earth, indicating the possibility
of a further limiting factor or factors in the former soils.

One indication of the limitation of productivity imposed by deficient nutrient
supply is given by the wide differences in the transpiration ratio for Atriplex vesi-
caria in pot culture tests (series A) conducted on the respective soils: on red - brown
earth soil a transpiration ratio of 370 indicated a production of 5.4 cwt. dry forage per

Figure 7.
Growth of barley without fertilizer on various soils (left to right). Red- brown earth, U'aite
Institute; solonized brown soil, Pallamana; arid red earth, N.E. District; arid red earth, N.U'.

District.

141

c

u

60

0

t-<

w

c

CM

(TN

0

0

0

'c

■V

CA

VO

0 ~<r

kl

60

o

Z

fN

00

f<^

C

m

Z

a-

0
rs

i-H

12.4
10.5

a

c

CO

r^

X c

3

o

d

fN

d

s

CO

o

■^ ft)

00

fsi

(V)

r— 1

(A

z

"^ a

d

0

\r\

u^

0

0

r~-

rO

W\ XJ-

«

s

-C

«

q

\D

ro

ro

^

V

Cn

ITN

OS <^

I-I

1

Z

00

>r\

C\

•

a,

z

d

ro d

a.

(0

o

en

O

ex
w
o

J3

f^

■5
'9

*— < r— (

4.*

Cfl

a,

CO

a.

<3

Vi

ON

.— t

*— t

4^

a

0

u

>>

0

0

w^

0

3

S

S
3

cu

00

u-\

r-- (Ni

-C

(vi

rsi

fN
-0

d

\0 ir\

t. superpl
r acre.

V

/'-N

•

<

>%

V

q

r-

rr>

00

a

0

a

CO

"s

d

\6

i-H

r~^

CQ

"3

u

4-*

o

k4

a,

•

CQ

ir-^

\r\

0

^/^

i^

K.

PO

fO

u-^ 00

xr rt

0

z

00
1—i

00

rri

d

^

<— "1

z

I-H

00 ^

and
moni

a,

e

C/5

u

. S

c

• -4

60

10

% '^

CO

U

V

I-I

iri

•■4

O

t>0
(9

00

'I «

0)

3

dT o.

an

O

-0 "3

<

>

-a

1

C (0

«4^

c

41

4J

W

4-»

a

^J

w

W

w w

O

V

9

4j

o

U

3

0

CJ

U

(J 0

>

4-*

• •4

• W

• «i4

•-H .-S

. ^

b

o

4^

ta

4^

4^

ca

'11

4>J

ll

(d

"S

I-I I-I

4.* Ut

2 ^

■-4

>

o

3

o

(0

T3

(J
o

w
i)

E
H

(A

Q
W

CO

Q

•

1

(0
w

a

13

Q

W

1

CO

Q

fee'

c
S

<— <

(0

(0

Q

•
1

-E. Dis
-E. Dis

•
(0

4-»
0

a.

60
I-I

mall poi
cwt. an<

Q.

C

^

0-

Z

Z

^

a

z

z

CL

Z

Z Z

CO r^

CO II

fS

N—-

a

0 '^

(0

^ -0

HI
Cd

•—4
• -4

•—1
*>4

.^

^H

* v4

.^

^H

(J

•- «

o

0

0

0

0

0

'0

• -4

r^ Z

a.

w

(A

V)

CO

(0

(0

CO

"a,

1

1
o

•s

•s

0

CO

§

CO

3
0

a
0

CO

3

0

'CO

U

1

a
o

jQ

c

a
0

-a

V
N

a

u

13
0

u

13
u

u
_Q

-0
U
N

0

u

13
0

red earl
calcare

1

•

<

CO

CQ

(0

"0

o

-o

-o

"V

0

-0

T3

0

-0

-0 -a

1^

I-I

•-4

• -H

•«H

..-•

• -H .^^

I

^

-5

0
00

^

<5

5

■^

<5 -5

c^

00

•

o
Z

•

•

*

•

•

•

•

•

• Ti

.— )

(VI

ro

■^

»-H

fN

1/-N

■o

(Nl

^o

r- 00

O
C/5

142

acre inch of water transpired; by comparison, with material grown on north-west arid
red earth, a transpiration ratio of 570 indicated a production of only 3.5 cwt. dry
forage per acre inch of water transpired.

The fertility cycle under arid shrub vegetation

Comparatively little is known of the fertility cycle in arid soils, and any dis-
cussion is necessarily of a speculative nature. The soils of the North-west Pas-
toral District have low organic carbon and nitrogen contents of the order of .25 and
.025 per cent respectively (Crocker and Skews, 1941; Jessup, 1952). The processes
of breakdown and decay of plant remains are extremely slow, and despite a carbon
nitrogen ratio of 10 : 1, it appears that the mineralization of the soil organic matter
and particularly the production of nitrate is inhibited, partly by climatic conditions,
but perhaps also by unfavourable soil conditions including possibly the presence of
the root systems of perennial plants (Theron, 1951).

It appears legitimate, therefore, to regard the fertility cycle under natural con-
ditions as an almost closed system with the nutrients circulating at an extremely
slow rate, with a large proportion of the 'available' supply locked up at any given
time in living or dead plant material and partly decomposed remains. There is,
nevertheless, a small annual turnover of nutrients permitting limited production when
soil moisture conditions allow plant growth. Nutrients may accumulate in the soil
during a cycle of dry years when virtually fallow conditions prevail; then with the
occurrence of favourable rains a relatively lush growth occurs which gives a mis-
leading impression of fertility. But should another favourable season follow the
first the soils are incapable of sustaining production and very little growth is made.
This occurred at the Yudnapinna Station in 1947 following the previous good season
in 1946.

The impaqt of the grazing animal on the fertility cycle of arid shrub pastures

The introduction of grazing animals to shrub pastures leads to a marked accel-
eration in the fertility cycle by increasing both the amount of plant material returned
to the soil and its rate of decomposition. The rate of turnover of nitrogen in particu-
lar is accelerated because some 75 per cent of the amount ingested by merino sheep
is returned to the soil in urine, and rapidly becomes available to plants. A further
20 per cent, returned in the dung, undergoes more rapid decomposition than non-
ingested plant material. Provided conditions of rainfall are favourable, plant growth
is stimulated by the fertilizing effect of the grazing animal; and the cycle is repea-
ted at a rate depending on the stock concentration per unit area, on favourable tem-
peratures, and on availability of soil moisture.

It is estimated that the large- framed merino wethers employed in the grazing
management investigations under review (1501b. live weight) void annually 25- 30
pounds of nitrogen. On the most heavily grazed plot during 1946- 51 the amount of
nitrogen returned to the soil by the animal was of the order of 6- 71b. per acre per
annum. This represents a considerable increase of the soil nitrogen available, in
terms of the relatively low production level on a per acre basis. On the more heavily
grazed portions of the plot, the amount of nitrogen returned would have been corres-

143

pondingly greater. It is also possible that by the stimulation of greater microbio-
logical activity in the manured soil, the rate of decomposition of the soil organic
matter was hastened, leading to the release of further nutrient supplies.

The grazing animal can add little to the total store of fertility in arid soils in
the absence of important leguminous constituents of the pastures, but by increasing
circulation of nutrients, it may add greatly to the productivity of shrub pastures
during favourable seasons. This may also stimulate the regeneration of young shrubs
from seed. The drain on soil fertility by the removal of nutrients in animal products
does not represent a serious loss when spread over many acres, and insofar as nitro-
gen is concerned, would not amount to more than ^^Ib. per acre per annum in the
above example.

The improvement of soil fertility by the fertilizing effect of sheep accounts
satisfactorily for the observed effects on the shrub pastures in the grazing manage-
ment investigation. During seasons of low rainfall the pasture did not benefit from
the return of nutrients ; and the weight of forage per acre was depleted on the heavily
and very heavily grazed plots. During this period nutrients accumulated in the soil
in a fairly readily available form in the plots grazed at moderate to very heavy rates
of stocking. Under the favourable climatic conditions of 1946, the established shrubs
responded to the accumulated sheep manure. Young seedlings developed vigorously
and were able to become firmly established, leading to a large increase in bush den-
sity. Under the subsequent generally favourable conditions of 1947- 51, the circu-
lation of nutrients was maintained and the production of herbage continued at much
higher levels than on the ungrazed plot.

Hence the productivity of shrub pastures may be regarded as a function of graz-
ing intensity and effective seasonal rainfall; and high intensities of stocking can
only be maintained during favourable seasons, since the fertility cycle pasture —
animal — soil — pasture can only operate when there is sufficient soil moisture avail-
able for sustained plant growth. It is probable that repeated heavy defoliation of
shrubs during a run of drought seasons would lead to widespread mortality of the
bushes. This concept of grazing intensively the arid shrub pastures when seasonal
conditions favour their active growth makes possible substantial improvement in
practical management.

There is of course a limit to the number of sheep which may be carried on shrub
pastures during seasons of high effective rainfall. An equilibrium must be maintained
between growth and consumption of forage. The small size of the grazing areas em-
ployed in these investigations has permitted the support of sheep at rates greatly in
excess of the recognised carrying capacity of shrub pastures in the North-west Dis-
trict where the average stocking rate is 25 sheep per square mile (Jessup, 1951). The
mean size of paddocks in this area is, however, 32 square miles with an average of
only one watering point per paddock (Jessup, ioc. cit.). Under these circumstances,
the upper limit of carrying capacity beyond which permanent degeneration of the
shrub stand may occur is fixed by the stock concentration within a Hrnile radius of
the water, supply.

144

Sheep population in relation to the seasonal pattern

Efficient utilization of shrub pastures through intensive grazing during periods
of active growth can only be attained by fitting stock numbers to the long term rain-
fall pattern, which is the primary determinant of forage resources ; and by close sub-
division of pastures together with the provision of adequate watering places. In this
way, maximum production will be attained, stock losses avoided, and the pastures
either improved or maintained in a sound condition. If, on the other hand, the num-
bers of grazing animals are not so adjusted, pastures either may not be utilized
effectively, or may be permanently damaged in attempting to maintain high numbers
during a sequence of drought seasons.

The Yudnapinna rainfall records reflect the seasonal pattern for the North-
west District, and the record of sheep population in this area (Fig. 3) shows that
stock numbers have not been well adjusted to the broader rainfall trends.

The years 1870- 1951 may be grouped into five main periods characterized by
particular trends in rainfall, pasture status and livestock population.

First period (1870- 1895). The sheep population of South Australia increased
from the middle years of the nineteenth century to a maximum in 1890 (Davidson,
1938) largely by expansion to virgin pastures, including those of the North-west
District. In this area numbers rose during a succession of generally favourable
seasons from 374,000 in 1870 to 789,000 in 1882, a sheep density which the pas-
tures could not permanently support under the existing conditions of subdivision and
water supply. In the second phase of this period, from 1882- 1895, the number of
sheep maintained decreased to 618,000 by 1892, and to a still lower figure in 1895
for which no record is available. It is certain that considerable deterioration of the
pastures commenced during these years.

Second period (1895- 1905). The years of high rainfall prior to 1895 were fol-
lowed by a succession of dry seasons from 1895- 1905. Pastures had been gravely
overstocked; sheep were concentrated on the few permanent waters, and the adjust-
ment to drought conditions was enforced by sheep losses rather than by planned de -
stocking; numbers fell to 275,000 in 1905. Extensive damage to the perennial
shrubs occurred during this time, and their productivity in many cases was perman-
ently impaired.

Third period (1905- 1924). The third period from 1905- 1924 was on the whole
characterized by above average rainfall, but despite this, stock numbers tended to
remain at a depressed level of about 250,000 sheep; it seems probable that the pre-
vious degeneration of the shrub pastures was a major factor. This period may be
divided into three phases. The first phase from 1905- 1911 was characterized by
favourable rainfall; over, this period the sheep population rose from 275,000 to
376,000, due to temporary recovery of the forage resources. In the second phase
from 1911 to 1915 the degraded pastures were unable to provide forage during a
recession of rainfall, and stock numbers decreased to 132,000 in 1915. Only a limi-
ted increase to 278,000 sheep occurred in the third phase of exceptionally high rain-
fall from 1915- 1924, during which wartime labour shortage exerted some control. It

145

is likely that a considerable regeneration of shrubs occurred in this portion of the
third period.

Fourth period (1924- 1936). The sheep population increased from 278,000 to
314,000 during a sequence of dry years from 1925- 1929 inclusive, as a result of re-
generation of pastures during the previous stage. Numbers continued to increase to
540,000 sheep during the short term phase of high rainfall 1930- 1932, but receded to
338,000 with the resumption of the dry cycle to 1936. The general level of sheep
numbers rose from a ten year mean of 220,000 at the beginning of the whole period
to one of 400,000 at the end.

Fifth period (1936- 1951). The mean level of 400,000 sheep reached in the
previous stage has been maintained, with some fluctuations, during the recent period
from 1936- 1951, indicating that at this level sheep numbers are in equilibrium with
the present condition of the pastures. The general pattern of high rainfall over these
years is interspersed with two short phases of unfavourable seasons during which
the recession in sheep population was aggravated by wartime conditions. From 1945
- 1951, numbers of sheep rose from 318,000 to 478,000; during these favourable
seasons very marked regeneration of shrubs has occurred.

The pastures of the region are now at a stage when sound measures of manage-
ment related to foreseeable rainfall trends can lead to a considerable degree of im-
provement of both the capital resources and permanent productivity. On the other
hand, a repetition of earlier mismanagement may lead to another recession of the
pastures and widespread denudation. The investigations at Yudnapinna have indi-
cated that greatly increased productivity may be achieved by a closer subdivision of
pastures and the provision of better and more frequent water supplies.

Conclusions

The more effective utilization of arid and semi- arid pastoral areas depends
upon the recognition of four outstanding principles :-

(1) An analysis of the climatic resources, and especially rainfall, in the long-
term, with due allowance for evaporation rates and the inevitable groups of
drought years.

(2) The use of fencing to provide for appropriate grazing management.

(3) Multiplication of the points at which livestock can secure water.

(4) The adjustment of stocking rates to the variations of the forage supply which
are consequent upon the long-term climatic pattern.

These principles are of general application to arid regions and their adoption
could lead to a very substantial increase in the productivity of both hot and cold
deserts, provided (a) that there is a minimum quantity of moisture available for
transpiration of the order of 2 inches per year and (b) that growth is not inhibited
by low temperatures, of the order of monthly means of 40-45°F or less, during the
period over which moisture is available to plants of value for permanent grazing.

146

Summary

Perennial shrub species of Atriplex and Kochia originally dominated the arid
pastoral areas of southern Australia with a mean annual rainfall of less than 10
inches.

These pastures have been stocked with sheep for approximately a century ;
over- grazing, leading to denudation, has been marked. Some undamaged associa-
tions of shrubs remain, and some regeneration has occurred, but for the most part re-
covery of the pastures has been limited.

On both relatively undamaged and degraded country there is scope for improved
methods of utilization. A study of grazing management within the North- west Pas-
toral District of South Australia conducted over a period of 11 years is described.

The climatic factors, and in particular, the rainfall characteristics of the area,
have been analysed in the short and the longer term; and these have been related to
the reaction to grazing of Kochia sedifolia and to major trends in the sheep popula-
tion.

Marked improvement in the density and vigour of Kochia sedifolia resulted from
intensive stocking during favourable seasons. The reaction of this shrub to grazing
is explained in terms of an increased circulation of nutrients from the vegetation to
the soil via the grazing animal, and thence back to the plant. A reduction of com-
peting herbage, and more effective use of soil moisture by the shrub, as a conse-
quence of its pruning by sheep, are contributing factors.

The general application of the principles discussed in the paper may lead to
more effective utilization of the pastures associated with arid and semi- arid regions.

References

Cornish, E.A. 1952. Personal communication.

Crocker, R.L., and Skewes, H.R. 1941. Trans. Roy. Soc. S. Aust. 65, 44.

Davidson, J. 1938. Trans. Roy. Soc. S. Aust. 62, 141.

Jessup, R. W. 1951. Trans. Roy. Soc. S. Aust. 74, 189.

Osborne, T. G.B., Wood, J.G., and Paltridge, T. B. 1932. Proc. Linn. Soc. N.S.'W. 57, 377.

Prescott, J. A. 1949. J. Soil Sci. 1, 9.

Prescott, J. A., Collins, Joyce A., and Shirpurkah, G. R. 1952. Geographical Rev. 61, 118.

Richardson, A. E.V., Trumble, H. C, and Shapter, R.E. 1932. Coun. Sci. Industr. Res. Aust.
Bull. No. 66.

Sears, P. D., and Goodall, V.C. 1948. N. Zealand ]. Sci. Tech. 30, 231.

Theron, J.J. 1951. /. Agr. Sci. 41, 239.

Trumble, H.C. 1952. Adv. Agronomy, 4. Academic Press Inc. New York. (In press)

Trumble, H.C, and Donald, CM. 1938. Coun. Sci. Industr. Res. Bull. No. 116.

Woodroffe, K. 1941. J. Aust. Inst. Agr. Sci. 7, 117,

147

BIOLOGICAL RESEARCH AND THE PRODUCTIVE TRANSFORMATION OF
STEPPE AND DESERT IN THE SOVIET UNION

Dr S. M.Manton, F.R.S.
(London)

This communication does not concern any aspect of my own research but refers
to the many lines of work which are in progress connected with increasing the pro-
ductivity of the central Asian desert regions. In the summer of 1951 I accepted an
invitation to visit the Soviet Union to meet scientific colleagues and to see some-
thing of their work. I travelled as far as central Asia, where I saw much of the
speedy development of the country, of the modern laboratories in the Asian states and
of the work connected with irrigation and afforestation, and the control of factors
which limit animal and plant life over vast areas.

One million square miles of the Soviet Union is steppe and desert, and three
quarters of a million of these form the central Asian deserts. Rainfall of between 2
and 15 inches occurs in spring and autumn, but not in the summer, and dry desiccating
winds, at times of hurricane force, sweep from the Asian deserts westwards across
the south Ukraine.

A fifteen year afforestation scheme was started in 1948, a quarter of which had
already been completed. For hours I flew across the featureless steppe, now marked
by black stripes. Each of these consisted of ploughed land 25 -65 yards wide, and
carried lines of seeded or seedling trees separated by low growing crops such as rye
or clover which prevent weeds from smothering the young trees. Three thousand three
hundred miles of major tree lines are being planted. Watershed lines, each up to 370
miles long, will interrupt the stream-lined air flow of the lower atmosphere, substi-
tuting turbulence, which will reduce the desiccating power of the winds. Tree lines
are flanking river valleys for distances up to 700 miles, windbreaks are appearing
between farms, and around natural erosion scars, which will not only check further
erosion, but add humidity to the air and soil.

Twenty years of research has gone into establishing successful methods of culti-
vating trees under steppe conditions, as I saw on the research stations and in. the
field. Oak is being used to initiate the steppe forests because it develops a deep
root system in dry soil. An abundance of tree planting machinery is being employed,
but when acorns and not seedlings are planted, the appropriate mycorrhiza is added to
the soil along with the acorns. Laboratories are occupied with the grading of acorns
and other forestry work.

In 1950 hydro - electric developments were started which by 1957 will irrigate an
area of 70 million acres, 20 million of these lying in the Asian deserts, and thereafter
the acreage will increase still further. The newly made dam on the Don, with its half
a mile central spillway of steel and concrete and 8 miles of earth wings, is sending
water to the first quarter of a million acres of newly irrigated land this summer. Else-
where in European Russia dams are being built to retain most or all of the spring
flood water of the Dnieper and Volga. Six hundred miles of canals will carry water to

148

149

the South Ukraine and north Crimea, and the Volga is being converted into a series of
lakes up to 370 miles long, A canal of 375 miles is being made to carry a flow equal
to that of the Don for the irrigation of an area of semi-desert equal to England in area,
between the Volga and the Ural rivers. Besides canals, 44,000 water basins are being
created, and hydro- electric power from new stations (with an aggregate capacity of four
and a quarter million Tcilowatts ) will pump water on to the land and drive agricultural
machinery such as tractors, etc. The European schemes affecting the steppe will be
completed by 1956, and the desert irrigation is to be effective by 1957.

The deserts of central Asia are partly man made, but the whole region has been
drying up for thousands of years, although the cutting down of trees has worsened con-
ditions. In the 3rd- 4th millenium BC there were towns 12 miles across in the present
desert areas, and archaeological evidence shows the existence of past irrigation sys-
tems. At one time a branch of the Amu Darya river flowed across Turkmenia to the
Caspian sea. The fertility of much of the desert is seen in the small scale irrigated
areas of Turkmenia, and in the large oases, such as Tashkent, which supports a popu-
lation of 1,300,000.

Work has now begun on the building of a dam across the Amu Darya near its delta
to the Aral sea for the purpose of deflecting about 47% of its flow into a 683 mile canal
across Turkmenia to Krasnovodsk on the Caspian Sea. Another 746 miles of branch
canals will irrigate about three and a half million acres (5,000 square miles) of desert.
The middle section of the canal which will use the dried up bed of the Uzboi river, now
many feet thick in salt, will flood about 17 million acres (27,000 square miles) for
periods of from one to three months, so providing increased pasture for horses, cattle
and karakul sheep. The average daily flow on the canal will be about 6,500 million
gallons (4 times that of the Thames), but at times will be double this figure. All the
water at present will be used for irrigation, and none will pass on to the Caspian Sea.
A project of this magnitude has not been attempted anywhere in the world.

The associated scientific work serves two purposes. Firstly there is the prelimi-
nary work connected with the actual construction and maintenance of such a desert
canal, and secondly much work is in progress concerning the maximum productivity of
the land to be irrigated.

The preliminary work has been in progress on a considerable scale. By 1951 a
scientific base was in working order in the Kara Kalpak, with a newly built railway
line, and a feed canal to supply water for the workers, for hydraulic machinery and for
the local growing of crops. Three hundred scientists were at work there in the summer
of 1951 and their number had increased to 500 by the spring of 1952. In addition there
are many field parties, and a large scale photo survey has already been made from the
air.

The subsoil waters in the Kara-Kum desert are of greater importance than the sur-
face waters, and they are being investigated by parties of geologists, each numbering
about a dozen, and equipped with drilling apparatus. Underground fresh water streams
can in places be deflected to the surface, and in fact now wat^er the town of Krasnovodsk.
The salinity and rates of seepage of subsoil waters is being determined, as saline water

150

must on no account be raised to levels which would effect crops by seepage of water
from new canals. Seismological parties are studying the structure of geological strata
by measuring tremors caused by explosions.

Meteorological stations have been set up in many places in the Caspian lowlands
and Asian deserts to collect data which, among other things, may give information
concerning the origin of the dry desert winds referred to above, and assist in com-
bating sandstorms. A special laboratory for the study of sandstorms has been set up
in the Ashkhabad by the Turkmenian branch of the Academy of Sciences of the U.S.S.R.

Protection of both canals and of irrigated land in the desert against shifting sand
and wind are essential. Shelter belts of poplar trees can be quickly grown when
watered; they now form wind breaks round the fields of Tashkent and other oases, but
unlike the steppe forests, these poplar windbreaks depend on irrigation. Altogether one
and a quarter million acres of desert forest will be grown to shelter the canal and the
watered land; this work is co-ordinated by a special Ministry of Forestry for the con-
struction schemes. The Black Saxaul tree grows without irrigation in the desert, and '
was once more widespread in occurrence than at present, having been cut down for fire
wood. In the summer of 1951 seeds of this tree were collected, and in the spring of
1952 were sown over large areas by parties operating from camels (this traditional ship
of the desert is now very little used). The black saxaul sends its roots down to a
depth of 30 feet, thus fixing the soil as well as gaining moisture, and in 10 years it
can reach a height of 20 feet and a trunk girth of 1 foot. The controlled use of this
tree should provide substantial yields of timber within 25 to 30 years. Ash, white
acacia, apricot and mulberry will also contribute to the shelter belts when the water
comes, and smaller species of sand fixing vegetation and saplings are now being
planted on a large scale. For the protection of young trees wind screens made of reeds
have proved to be more satisfactory than solid ones. Impervious sheets hold back all
the sand, and produce dunes before becoming buried. The setting up of the screens
in sand is mechanised. Experiments are also in progress on the utilisation of a waste
product from industry which, when sprayed on to sandy soil, will immobilise the sur-
face, yet leave it permeable to rain and to vegetation.

There is much work connected with the irrigation of the desert which is of indirect
biological importance. Seepage outward from canals is being checked by packing with
clay, concrete or asphalt being prohibitively expensive. Intense rates of evaporation
will lead not only to the loss of high percentages of the water entering the Turkmenian
canal, but also to silt deposition and a tendency for the water to become saline. The
silt content of the canal water is estimated to reach 20- 25 million tons, and when de-
posited will be removed by electric excavators and suction dredgers, besides by silt
eliminators such as those already in use in Uzbekistan. These are floating instal-
lations which, by stream directing shields, allow only pure water to enter a canal,
thereby reducing the cost of cleaning out the canal to one tenth of its previous figure.
Again a special centre, the All -Union Scientific Research Institute for Hydro -tech-
nique and Amelioration, is dealing with this work.

The other side of the scientific work concerns the productivity of the watered
desert, and this is being prosecuted in laboratories as far apart as Moscow and Tash-

151

kent, and by field expeditions. Tashkent, for example, has been transformed during
the last 30 years into a modern city by irrigation, by the utilisation of power, and by
education. The Tashkent Academy of Sciences integrates the activities of 23 research
institutes comprising about 1500 full time research workers who are additional to those
working in the University, and most of them are of the Uzbek and other Asian races.

A soil survey is being made in considerable detail. Shifting sand has in parts
smothered fertile desert soils which are being reclaimed. A detailed knowledge of the
soil and subsoil is necessary both for the choice of the most suitable crop plants, and
for the decisions concerning methods of improving difficult types of soil. Research
stations are engaged on these problems. For example the clay plain north of Kizyl-
Arvat and patches of clay on the route of the canal were once considered unsuitable
for cultivation. Studies on the chemical and physical properties of these peculiar
soils has shown that when the texture can be improved they become fertile.

The areas to be watered by periodic flooding are greater than those to be irrigated.
Agronomists from the Institute for the Amelioration of Water and Marshes Economy are
selecting suitable grasses for growing on the new pastures, and the plant breeding
stations are taking steps to produce the seed in quantity. There are now 72 plant
breeding stations with 4,000 to 5,000 acres each, situated in different parts of the
Soviet Union. The animal breeders are endeavouring to improve the stocks of Karakul
sheep to graze these new meadows, and Turkmenian Tekin horses are already of high
quality. Increase in cattle is being pr6secuted by careful breeding and by modern
methods, and improvement in agricultural methods of fodder crop production. The
natural supply of fertilisers occurring in parts of the Turkmenian desert are being ex-
ploited in increasing quantity, fertilisers are added every six weeks for some plants,
and seven crops of lucerne, for example, are being harvested each year by these
methods.

The mechanisation of desert agriculture and the production of suitable varieties
of economic plants have made great strides in recent years. Problems of drainage are
being attended to, and are equal in importance to those of irrigation because increases
in soil salinity must be avoided. Yields of cotton per acre had trebled since 1932 in
the fields which I visited. The Tashkent cotton institute for example, employs more
than 40 scientists and over 200 other workers. Varieties of cotton suited to certain
localities are produced by hybridisation and by other means. A pre- sowing treatment
of the seed is providing a method of obtaining plants which are more resistant to saline
soils, and much of the desert which will be watered is saline. Two crops a year can
be harvested, and up to 30cwt. of cotton per acre can be raised in the irrigated parts
of Turkmenia. Two crops of wheat a year are also practicable. The plant breeding
stations near Ashkhabad and at Kara-Kala and Kara- Kalinskaya are engaged in pro-
ducing subtropical fruits suitable for cultivation in the areas to be irrigated, and large
scale production of cherry, apricot and peach trees is going ahead for planting on the
present arid wastes of Turkmenia. The Uzbekistan stations have already accomplished
much in the production of locally suitable varieties of hard fruit which I saw cropping
heavily. The first 5,000 acres of new land were irrigated in 1952 and on it experi-
mental crops are being grown.

152

In the summer of 1951 there were 22 expeditions of botanists, zoologists and soil
workers distributed in the Kara Kum desert, and a party of zoologists from the Turk-
menian Academy of Sciences has travelled about 2500 miles in the valleys of the
Atrek, Sumbar and Chandir rivers. I discussed the work of these parties with some of
their members and saw some of their material being worked upon in their laboratories.
An intensive 2 years of field work has preceded the actual building of desert dams and
canals, in the same way that the data collected by the pre-war scientific expeditions
to the region between the Volga and the Ural rivers has been utilised in planning the
irrigation in progress there. Each group working in the Kara Kum desert comprised 12
to 20 persons coming from all over the Soviet Union, besides from the young academies
of sciences and universities of the central Asian states.

Seeds of certain plants which are wanted for the new pastures and meadows were
being collected, and surveys were being made of the native plants and animals. A
look-out is always kept for wild varieties that can be turned to economic use, as were
the rubber bearing Scorzonera tau- saghys and Taraxacum kok- saghys, 'dandelions'
found in the Tien Shan mountains in 1930 and 1931. These two species now provide
the major part of the rubber crop of the U.S.S.R.

Ecological studies are stressed, and detailed work is carried on in selected
places, both virgin and in the oases. A few semi -permanent desert laboratories have
been set up for this work and for the soil analysis. I saw many cultures of soil micro-
organisms maintained in the Institute of Zoology at Tashkent.

A large field of work before the expeditions and the laboratory workers concerns
parasites and pests in general. The normal pests and predators of desert trees, shrubs
and plants which are about to be grown on a large scale are being studied, rodents as
well as insects, so that any enormous increase in the numbers of these organisms
arising from the altered balance of nature may be dealt with immediately by appropria s
measures, and wholesale destruction avoided.

Predators and parasites of domestic or potentially domestic animals are being in-
vestigated, and every opportunity is being taken to follow out the life cycles of flat-
worms axvd other parasites which inhabit two or more hosts. Information is being col-
lected concerning the species and habits of molluscan and other intermediate hosts.
The work associated with insect vectors of diseases of all kinds is as important here
as in other warm countries, and employs many persons. The incidence of malaria in
the oases is now low; I myself saw no mosquitos and I did not use the nets with which
I was provided. All slowly flowing irrigation channels dry out completely between
flooding which takes place every twelfth day for cotton. Gad-fly problems, they told
me, had been satisfactorily solved. Physiological work on domestic mammals occupies
many workers.

The most important crop to be raised in Turkmenia will be cotton, with much wheat,
rice, dates, olives, fruit and plants producing rubber and essential oils. Cotton and
rice will also be grown in the Ukraine for the first time. The anticipated yields from
the whole of the new irrigation schemes include, in millions of tons: wheat 8, sugar
beet 6, cotton 3, rice Vi, together with 2 million head of cattle and 9 million head of
sheep. This represents food for a 100 million persons, besides the industrial crops.

153

The deflection of so much water on to the land by the dam being built on the Amu
Darya river to supply the Turkmenian canal will inevitably lead to a lowering in level
of the Aral sea, and an increase in its salinity. This is welcomed up to a point be-
cause a reduction in subsoil water levels will make available large tracts of the fertile
delta of the Amu Darya for cultivation. The Caspian Sea will in time also be affected
by the diversion of so much water from the Volga on to the land.

The effects of irrigation of the new areas, which are equal to one -third of the
world's irrigated land, will also be to better the climate over an area estimated at some
300 million acres (an area larger than that of Europe), the temperatures will become
less extreme and the atmosphere more humid. The probable details of the climatic
effects of the schemes and the future water balance of the inland seas are the subjects
of much discussion in the Soviet Union.

The realisation of projects of this kind, which in scale approach those of natural
forces, is being carried out by mechanised navvying. In five to seven years about
4,000 million cubic yards of earth are being shifted — this represents about sixteen
times that moved for the Panama Canal — 25 million cubic yards of concrete are being
mixed, and thousands of tons of metal sections and equipment are being used. A
labour force of 200,000 persons is operating the machines for this work. Drag -line
excavators employ buckets of 18 to 32 cubic yards capacity, and their load can be dug
and dumped 130 yards away in a minute. Suction dredgers, having piled up the earth
wings to the Tsimlyanskaya dam on the Don, are now in use in the Amu Darya. Each
unit is manned by 10 engineers and it can operate down to a depth of 70 feet, churning
earth to a suction head from which it is removed by pipe for distances up to 3 miles,
and doing the work equivalent to about 10,000 to 15,000 men provided with picks and
shovels. Automatically controlled concrete mixing combines and many other machines
have been specially designed for the developmental schemes. Routes of communica-
tion are being developed, new towns are growing up and are being staked out in the
desert ready to receive the water when it comes in 1957. Krasnovodsk, at the Caspian
end of the canal, was once a desolate waterless place.; it is now a beautiful modern
city with an abundance of greenery, as in Tashkent.

This wide control over factors which limit life, and the productive development of
a large part of the potentially fertile central Asian deserts has been made possible by
detailed preliminary planning and by research of many kinds, ranging from purely bio-
logical problems to such matters as the properties of alloys and methods of construct-
ing dams which will ride earthquakes and not be destroyed by them (as are needed in
western Turkmenia). All these things are just as necessary as an ability to meet the
scale of the engineering requirements. But above all, it is the integration of the many
lines of work, directly of a biological nature and indirectly of biological significance,
that is leading to such immense productivity within a few years.

Only two great rivers traverse these Asian deserts; the Syr Darya is already used
almost to capacity for irrigation in Uzbekistan, and the Amu Darya, with an annual
flow of 10 and a half cubic miles, is being diverted in part across Turkmenia to water
20 of the 37 million acres of potentially productive land. Long term projects are now
being planned for the diversion of some 70 cubic miles of water annually from the

154

northwardly flowing Siberian rivers. Dams, canals and dried up river beds could carry
this water 2,500 miles to central Asia, and this would make possible the irrigation of
some 62 million acres of land for crops and 87 million acres for pasture, so satisfying
all central Asian needs for water. The more humid atmosphere would bring milder
winters which would allow agriculture to be carried out further north than is at present
possible. These projects could be started after 1957 when the Turkmenian canal is to
be finished, and the central Asian states could then support a population of 120 million
instead of the present 20 millions. Modern civilisation could inhabit the sites of
ancient communities, abandoned when the water supplies disappeared.

155

ASPECTS OF THE ECOLOGY AND PRODUCTIVITY OF SOME OF
THE MORE ARID REGIONS OF SOUTHERN AND EASTERN AFRICA

Professor J. Phillips, F.R.S.E.*
{Achimota, Gold Coast)

I Introductory Remarks.

As the purfHjse of the symposium is to study features of the ecology and pro-
ductivity of deserts it could, with' justice, be asked why I should deal with regions
and phenomena that, for the greater part, fall outside the conception and definition
of the desert proper.

II Objects.

My objects are:-

(1) To discuss briefly a few selected climatic regions of desert, sub- desert, arid
and sub - arid nature in relation to their actual and possible usefulness and the
threat of desiccation to which they are exposed.

(2) To touch on some of the ecological phenomena and problems in such regions.

(3) To record some of the major factors, processes and agencies presenting prob-
lems in the control of desiccation.

(4) To suggest ways and means of improving the productivity of some of the regions
simultaneously with the control of the march of 'desertification'.

My reasons are these: Even if we accept the view that climatically the whole
or portions of Africa and adjacent regions of Southern Europe, Persia and Arabia,
during more recent geological time, have little or no tendency toward increasing
aridity and that there is likely to be but slight change in this direction in the coming
tens of millions of years, we dare not close our eyes to the portents of a man- in -
duced desiccation or 'desertification' associated with the more arid U not truly
desert regions of Africa. More cogent even is an attempt to interpret the shape of
things to come if we accept as a working hypothesis that we are experiencing today
the ushering in of a phase of progressive aridity. If we agree with le Danois (1950)
that on a world wide scale deserts have increased within the past 3000 to 5000
years — and more particularly in the relevant regions of Asia, Eastern Europe and
Africa — and that the sands are driving into drier but not yet desert regions adjacent,
there is all the more reason for an emphasis upon the potentials and problems of
the regions facing growing danger of desiccation.

III Some of the Major More Arid Regions.

In the regions listed below serious local erosion and desiccation problems exist,
which, if not solved within reasonable time, are likely to increase the man-made
desert.

• Recently Chief Agricultural Adviser, Overseas Food Corporation (East African Groundnuts
Scheme) — Ed.

156

In a fuller account of the regions to be published elsewhere I refer briefly to
the major features of vegetation, productivity actual and potential, and the nature of
the threat already evident as the outcome of man's action.

(1) The Namib and the Great Nama Land Deserts

(i) The Namib
(ii) The Great Nama Land Desert.

(2) The Sub- Deserts of the Karroo and the Kalahari

(i) The Karroo
(ii) The Sub- Desert Kalahari.

(3) Certain Dry Tropical /Sub- Tropical Regions of Southern and Eastern Africa

(i) Arid Regions

(a) Arid portions of the North- Central Cape Province, South- Western and
Western Orange Free State and Extreme Western Transvaal.

(b) The Arid Kalahari

(c) The Arid Limpopo Region

(d) Arid Portions of North- Eastern Tanganyika and South East Kenya to-
gether with Arid country linking with the 'Somali' Region to the North

(e) The Arid 'Somali' Region,
(ii) Sub- Arid Regions

(a) The sub- arid Bushveld of the Transvaal

(b) Sub- arid regions in Northern Bechuanaland and Sotithern Rhodesia

(c) Sub- arid portions of Central Tanganyika

(i) Tribal Agriculture

(ii) East African Groundnuts Scheme: Kongwa Region,
(iii) Semi- Sub- Arid Regions.

IV Some Ecological Phenomena and Problems.

In the fuller record to be published elsewhere I deal briefly with the following
ecological phenomena and problems:-

(1) Habitat Factors

(i) Radiation
(ii) Humidity
(iii) Evaporation
(iv) Dew
(v) Rainfall
(vi) Edaphic factors.

(2) Biological and Ecological Phenomena

(i) Succession and development

(ii) Community and climax

(iii) Physiological and aut- ecological investigations
(iv) Physiology and reactions of animals

(v) The role of fire.

157

V Productivity Actual Und Potential of the Desert and Other Dry Regions To-
gether with some of the Problems Involved.

(1) Productivity.

(i) The deserts are of no actual and potential productive value, except very
locally for light browsing by nomadic to semi- nomadic hardy sheep and
goats based on drinking points and except for very limited irrigation of
suitable soils. The Namib is of much less value than the Great Nama
Land Desert for browse, local as this is even in the last named region.

For reasons of protection of adjacent regions of greater value the conser-
vation of the deserts, toward their margins is essential.

(ii) The sub • deserts Karroo and Kalahari are of value for grazing and brow-
sing. In the Karroo steady and widespread deterioration has followed the
mismanagement of the browse land, presenting a national problem. Nothing
less than a combined individual and national effort to reclaim the Karroo
veld can save this region from desolation, a desolation that would have
disastrous effects on regions adjacent because of the eastward and north-
ward march of the sub- desert. Local irrigation is actual and potential,
but demands much care in the use of water.

(iii) The arid regions are in imminent danger of increasing desiccation as the
outcome of mismanagement of livestock. These regions are of little crop
production value, apart from local peasant field husbandry, which demands
constant direction if its deteriorating influence is to be avoided.

(iv) The sub- arid region of the Transvaal Bushveld is in imminent danger due
to mismanagement of the natural grazing. Arising from the same cause,
there are marked signals of distress in portions of Central Tanganyika.

African peasant arable agriculture in Central Tanganyika requires particu-
lar guidance and the continued insistence on conservation measures if the
growing deterioration of portions of the country is to be stemmed in time.

(2) Some of the Problems Involved.

Matters of far-reaching significance in the maintenance and the development of
productivity in the arid and sub - arid regions are:-

(i) Where Tsetse- fly still exists on a large scale its attempted removal and
the efforts to introduce livestock to fly -free areas must be accompanied
by a policy and practice of herd control and pasture management, supply of
water points, conservation farming and informed and firm administrative
direction of the local people. Particularly disastrous is the consequence
of uncontrolled continuance of the widespread 'lobola* or purchasing of
wives by means of thriftless livestock — the status of a man being judged
not by quality of his livestock but by the number. In 1928 I sounded the
note to the Governor of Tanganyika of the day that because no government
in Africa at that time appeared prepared to shoulder these necessary con-
trolling responsibilities the 'fly' should be considered as the guardian of

158

much of the continent — against the ravages of erosion and desiccation. I
have since, on several occasions, reiterated the same thought. In the ab-
sence of measures providing for the conservation of country hitherto under
'fly*, overstocking would readily produce a far worse curse than the 'fly':
additional foci of desiccation. It is heartening to read that recently the
tribesmen in Sukumaland, Lake Province, have accepted a plan for the re-
duction of their stock by culling. If successful and if applied generally
this may have far-reaching influences on wise use of land freed of 'fly'.

Again I make a plea for the planned freeing of all such country from the
'fly* and its appropriate use and conservation on an organized basis.

(ii) Solution of the problems of economic and acceptable livestock reduction
and of the management of natural pastures in the drier portions under re-
view, is made all the more urgent because of the marked influence of
selective grazing and browsing and the witholding of fire upon the rapid
establishment of vast areas of thorny and other thicket and scrub, in which
Acacia and Dichrostachys frequently play a role. Nfillions of acres in South,
Central and East Africa are either in or advancing toward this condition —
its sole merit being that its impenetrability renders the ground it covers
safe from further trampling by livestock.

Measures for the rehabilitation of such areas, by means of scrub and thic-
ket thinning and removal accompanied by protection for a time and, later,
by systematic management of grazing, must be introduced if vast acreages
are not to be permanently lost to the use of man — otherwise a desert due
to erosion would be replaced by one of thorns!

(iii) Large-scale enterprises aiming at the ranching of the arid and semi- arid
regions, where such are Tsetse- free, will succeed to the degree to which
those responsible undertake rational preliminary survey of the potentiali-
ties, the provision of adequately distributed water points, the management
of the natural grazing and browse, the setting aside of reserve grazing or
fodder against hazard of drought and the consistent giving of attention to
pests and diseases.

Considerable potential there is in the Kalahari and in parts of Rhodesia
and Tanganyika for enterprises of the right kind properly planned and direc-
ted. Hasty, ill - directed action would end in disaster.

(iv) In the sub - arid regions — notably in Rhodesia and Tanganyika, large-scale
mechanized crop production should not be attempted, no matter how encoura-
ging the temporary successes on the now small - scaled scheme at Kongwa
or elsewhere in similar country may appear. The costs of clearing of vege-
tation, the preparation of the land and of periodic losses due to serious
droughts would not be justified. Nevertheless, it remains true. that on lower
lying, alluvial soil within the sub- arid areas small scale mechanized pro-
duction of Groundnuts and Sorghum may be economically worth while. In
such areas Maize should normally be considered uncertain, owing to the
marginal nature of the regions in terms of rainfall reliability.

159

(v) Selection and breeding of suitable exceptionally drought - hardy varieties,
particularly of Sorghum, Maize, Groundnuts and other legumes as well as of
Cotton, require consistent attention. Promising indications have been
yielded as the result of investigations in various of the drier regional
centres in South Africa.

VI The future.

It would be easy to prepare programmes for survey, research and administrative
action and to suggest resolutions for this Conference, drawing the attention of the
various Governments to the problems and threats already well known. This would be
a repetition of what has been covered in varying degree fairly recently at such offi-
cial gatherings as: the First Commonwealth Conference on Tropical and Sub- tropical
Soils (Harpenden 1948), which was attended also by representatives of certain non-
Commonwealth countries, the African Soils Conference (Goma, 1948) and the African
Regional Scientific Conference (Johannesburg, 1949). In addition, the creation of
the Inter- African Information Bureau for Soil Conservation and Land Utilization, in
Paris, as the outcome of the Goma Conference (African Soils, 1951), the setting up of
the Council for Scientific Research in Africa South of the Sahara and the establish-
ment by U.N.E.S.C.O. of the International Arid Zone Research Council — which in
turn has appointed a standing Advisory Committee on Arid Zone Research — augur
well for the provision of the requisite scientific, applied economic and administra-
tive information regarding all aspects of the threats of desiccation.

The selection by the above mentioned standing Advisory Committee on Arid
Zone Research of the two centres in Algeria — The Saharan Research Centre at Beni-
Abbes Oasis on the Qued Saoura in the Southern Algerian Territories, and the Beni
Ouif Saharan Biology Station in the Sud Ouranais midway between Colomb Bechar and
Ain Scfra ~ is an advance.

On the applied and economic sides the French are engaged in agricultural, con-
servation and other work in several of the desert, semi-desert and arid regions of
theii North and West African Empire.

A survey of action being taken in the territories mentioned reveals that the
Union of South Africa and Southern Rhodesia have special legislation and services
for combat of the causes and retardation of the processes of desiccation, while to a
lesser degree provision has been made in the British Colonial Territories such as
Northern Rhodesia, Nyassaland, Tanganyika, Kenya and British Somaliland. The
Portuguese have commenced in Mozambique, while for a part of Somalia there has
been a survey of aspects of the problem by a recent F.A.O. Mission.

From this it might be argued that all is well, that the authorities are aware of
the need for appropriate action and that such is indeed being either taken or serious-
ly planned.

Unfortunately, the scope and the degree of practical action does not fit the
dangers facing us. In part this is due to shortage of funds but rather more because
of the lack of staff with the necessary training and experience, while the want of

160

machinery and equipment is also a serious handicap. Beyond all this, however, are
the powerful forces of inertia on the part of the agricultural communities, European
and African, who fail to realise the seriousness of the situation.

Though counsels of perfection may appear, it is necessary to emphasize the
following once again to Governments and peoples.

(1) Magnitude of the dangers threatening us all.

(2) Need for more vigorous propaganda and education of all kinds and in all circles
from ploughman to parliamentarian.

(3) Urgency of training many more men at the various levels for undertaking work of
various kinds — scientific and other — against the menaces emerging from cur-
rent agricultural and related practices.

(4) The absolute necessity for attracting more men for the planning of reclamation,
conservation and rational land use — notwithstanding the increased national ex-
penditure involved in the offering of better stipends and careers.

(5) Close and frequent collaboration among the various States, so that matters of
, policy and practice may be the more readily studied and co-ordinated action

the more readily taken.

In retelling what is already known we must remember - in our endeavours in any
great matter — the sentiment that * ... it is not the beginning, but the continuing of
the same, until it be thoroughly finished, which yieldeth the true glory . . . ' We
must continue advising, guiding, stimulation and educating to the uttermost.

Meanwhile the desert is on the march. We must act so that this march does not
end in the 'Great South (— Central — Eastern — ) African Desert uninhabitable by
man'. (Drought Investigation Commission, S.A. 1923 — except for the words within
brackets!)

161

PROBLEMS OF PHYSIOLOGY AND ECOLOGY OF DESERT ANIMALS

Professor F. S. Bodenheimer
(Jerusalem)

The endocrine cycle of the reproductive glands in desert animals.

In all climates manifesting distinct seasonal contrasts the majority of
terrestrial vertebrates undergo a conspicuous annual cycle with regard to the sea-
sonal activity and the histological structure of the endocrine glands, especially of
the gonads. These seasonal changes occur in response to changes of external
stimuli, such as temperature, humidity, duration of day, etc., and their effect always
results in the birth of the young at the season of the most luscious vegetation. Ac-
cordingly, the rutting season is usually in the climatic autumn, and that of birth in
the climatic spring. This synchronization of the reproductive cycles into annual
cycles of climate and vegetation is doubtless of the greatest ecological importance.
It is such a conspicuous phenomenon that it can scarcely be overlooked. In many
cases this synchronization is fixed by heredity and is more or less rigid at least so
long as no counteracting external stimuli change the normal cycle. Thus we know
that sheep or deer transported from a moderate climate in the northern hemisphere to
a corresponding one in the southern hemisphere, where the winter corresponds to the
northern summer, adapt themselves within one or two seasons to the climatic cycle
of the new environment. Yet for the camel a transfer into the summer- rain regions
of the Sudan means an experiment which is rarely survived and still more rarely
leads to reproduction within the Sudanese cycle of precipitation.

The habitual seasonal cycle of reproduction is however often maintained for a
long time in a not entirely different climate where only the normal releasing external
stimuli are missing. Major Flower has published birth data for a number of species
of gazelles normally living in rather varied climatic conditions in N. E. Africa, which
with their offspring were kept for a long time in the Cairo Zoo. In Cairo rain is
practically absent. Fresh berseem - clover or lucerne (produced by irrigation) is fed
to the animals throughout the year, and the trend of temperature and of day -length
is more or less identical with that in their home countries. He showed that the
monthly birth incidence of gazelles in the Cairo Zoo remained for many years, in full
agreement with the seasonal rain cycles of their native regions, the peak of the
births usually following that of the rains by one month. The only typical domestic
animal of our deserts is the camel, whose reproductive endocrine cycle has recently
been studied in the Negeb, S. Israel, by R. Volcani. The camel, in contrast to many
other domestic animals, has preserved a pronounced rutting season from January to
March. Its pregnancy lasts 12 months, suckling 3 to 4 months, and the interval bet-
ween births is two years. Both birth and rutting seasons coincide with the season
of luxuriant vegetation on the margins of the desert. This is an extreme and most
remarkable adaptation to the desert environment with its short period of green vege-
tation. In consequence reproduction occurs only once in two years.

During the rutting season the female is 'on heat' for periods of 7 days with 20-
day intervals, until fertilization has taken place. The seasonal changes of the

162

ovary could not be followed, as females are not slaughtered at the age of fertility.
In the male, however, the testes and the epididymis show the following changes:

Testes

Epididymis

Month

Weight

Activity

Weight

Sperm in lumen

(grams)

(1-4)

(grams)

(1-5)

V-VII

66

1.0

17

0.7

vm-x

91

2.5

24

1.8

XI- XII

70

4.0

16

3.0

II -III

4.0

4.5

IV

96

1.5

25

1.0

The peak in the weight of the testes in autumn does not have the same signi-
ficance. The peak of 96 g in February- March is connected with the rutting season,
whilst that of 91 g in August- October is due to heavy hydration. During December-
March the diameter of the tubules is 183^ with 5-6 layers of germ cells in various
stages of differentiation with abundant mitoses. Many sperms are seen between the
Sartoli cells, their tails directed towards the lumen. The walls of the tubules are
highly acidophilous. The space between them is narrow, and their walls often touch
those of neighbouring tubules. The intersitial tissue is compressed, and its nuclei
are stained a dark colour by haematoxylin.

From March onwards, degeneration sets in; more and more of the tubules are
devoid of sperms and vacuolization becomes conspicuous. In May this degeneration
is well advanced; a few wall layers only are seen in the tubules and these are often
reduced to one. Sperms are almost absent, as are mitoses. The diameter of the
tubules is reduced to l3l(Ji, and the interstitial tissue is loose. In October, rejuve-
nation of the germ cells begins simultaneously with the disappearance of the dege-
nerated cells. The diameter of the tubules slowly increases, and an ever increasing
number of sperms is observed. From November to January, rejuvenation progresses,
reaching its annual peak in February. Observations on the epididymis and the
presence of sperms within its lumen agree with the seasonal trend of the testes.
The thyroid shows colloidal accumulation in extending lumina from July to Novem-
ber. Until February the cells remain flat, the nuclei are at rest and no further secre-
tion is observed. During February the cells of the glands become columnar and their
nuclei are reduced. From March and April onward, a slow preparatory accumulation
of colloidal secretion into the lumen begins. The seasonal peak of the thyroid close-
ly follows that of the testes.

The reproductive season of many species of small rodents in the desert also
seems to be concentrated in the short climatic spring, although prolonged breedings
of Meriones, Acomys, etc. have shown that under suitable environmental conditions
they are able to continue reproduction throughout the year. The limiting factor
seems to oe the composition of the food. We have shown for Microtus that in Israel,
certain factors in green plants may induce a duplication of fertility by raising the
number of eggs per ovulation, as well as by shortening the interval between two
consecutive pregnancies. Microtus is apparently unable to maintain fertility on a

163

diet of dry vegetation alone. Prolonged droughts bring it to the margin of total ex-
termination, except in certain favourable localities like the borders of swamps, irri-
gated fields, etc. In the typical desert rodents either the normal rate of ecological
destruction or the physiological resistance to dry food (largely replaced in the
desert by bulbs and succulent plants) must differ considerably from that of Microtus,
which is prevented from entering the desert at all. The seasonal and annual fluc-
tuations of the populations of the nocturnal desert mammals have not been studied.
Their enemies may be few, for there are not many nocturnal birds of prey, although
snakes are abundant.

The seasonal reproductive cycle of desert birds usually differs from that of the
birds of neighbouring regions. Most of our desert birds breed rather early (mainly in
March) according to the rainfall, and breeding ends in late March. The desert offers
sufficient food for the rearing of young only during the short period of late winter
vegetation. The total number of eggs and of broods per female per year is reduced as
compared with that in more favourable biotopes. Thus Oenanthe lugens only has up
to 5, compared with the 42 eggs of the equal sized Erythropygia in Mediterranean •
territory. The great heat reduces the size of the hunting area, and still more impor-
tant, it enforces upon desert birds many hours of additional rest at noon. The energy
spent in finding the same amount of food is also much greater than in more favoured
biotopes. These observations easily explain the lack of attraction which the desert
even at its most favourable season, exerts upon hibernating, aestivating or resident
birds beyond its borders. On the other hand, Heim de Balzac observed that on the
southern borders of the Sahara, a number of resident birds extend their breeding
area into the savanna, far beyond the limits of the desert vegetation. We must as-
sume that in that season only are conditions favourable for them in the savanna.
Mobile birds are able to utilize this situation for extending their range during the
short but vital season of nidification and reproduction, when the food situation is
less favourable in their permanent habitat, the true desert.

In reptiles a seasonal cycle of the gonads is also observed. Here the seasonal
occurrence of the main items of food as well as the historic origin of every species
are of importance. Whilst reptile- eating species may reproduce late in the season,
insectivorous reptiles breed early. The chamaeleon shows its historic affiliation
with the African savanna from which it originates, by its reproduction late in autumn,
when sycamore and charob trees are flowering. The reproduction of the few desert
amphibians depends entirely upon the incidence of heavy precipitation.

Among small aquatic invertebrates as well as in the big Phyllopods (Estheria,
etc.) a similar relation to rainfall prevails. The insects however show many paral-
lels with the terrestrial vertebrates of the desert. Diapause in any stage from egg to
mature adult, is one of the mechanisms of adaptation. It is induced and regulated
by seasonal changes of the secretory glands in response to unfavourable stimuli.
More attention should be paid to the secretory seasonal changes of the reproductive
glands in desert animals, as well as to the analysis of the stimuli which induce
these changes in every species.

164

Solar - radiation and colour pattern in desert animals.

In the bare and open desert landscape the intensity of solar- radiation, to which
every diurnal animal is exposed is high even if it is often less than on bare moun-
tains. For those animals that are active between late morning and late afternoon,
this radiation is in summer of the highest ecological importance. It is surprising to
find therefore, that no measurements are available on the transmission through the
integrement of rays from ultra-violet to infra-red.

Three types of colour patterns are apparent in desert animals:-

(1) Mammals and birds show prevalently buff, sandy, pale grey or spotted
colours and remain hidden during the day; or when they are diurnal, their chief ene-
mies are nocturnal, Buxton, Heim de Balzac and Morrison- Scott have thoroughly
destroyed the legend that this type of colouration is primarily protective*. We have
to be satisfied with the statement that this 'adaptive' colouration is primarily a phy-
siological effect of dry heat on the development of pigments.

(2) Many Orthoptera in particular, show a very intimate and complicated adapta-
tion to the colour of the soil on which they live, imitating the pattern of the pebbles
in their environment. Resting Eremiaphila and most Acrididae are usually not to be
discerned even by a searching eye. They appear very conspicuous however immedia-
tely they fly. Here we obviously have some kind of appreciation of the colour of the
environment immediately after the last moult via the eyes, the central nervous sys-
tem and endocrine mechanisms (probably connected with the cardiac glands). A
similar surprisingly close colour adaptation of the feathers exists in our common
desert larks {Ammomanes spp.), as well as in some other desert birds.

(3) The high proportion of black colouration in diurnal desert animals, was
apparently first pointed out by Buxton. This is a rather puzzling phenomenon, as it
seems to be a bad adaptation for desert animals and increases the absorption of
heat. Some of these black desert animals have black colouration in other biotopes
too, so that their blackness is no adaptation; but this only raises the question why
black elements prevail so much amongst the diurnal desert animals. Among those
are the Tenebrionid beetles which predominate in most deserts of which the noctur-
nal species (Blaps) are not less black than are the more common diurnal species.
The same can be claimed for the ravens, except that our desert ravens have perhaps
a less deep black nuance than have their cousins beyond the deserts. In both these
cases black is the common ancestral colour of the group. Wheatears and chats
(Oenanthe, Saxicola) are another group of prevalently black desert birds. Buxton
points out that their transformation from a buff to a black and white pattern can
easily be followed. Here we have apparently a definite adaptation to the desert.
The same is true for a number of insects such as the blackish races of Metacerus
and other grasshoppers.

We may also refer to an analogous condition in man. The tents of the Bedouins
of the desert are usually black (or dark brown) and their thick overcoats or abbayas

• c.f. Cott, H. B. 1940. Adaptive coloration in animals. London, p. 154 — Ed.

165

show the same colour. This empirical choice certainly has its reasons which we
are at present unable to recognise. It should also be pointed out that for all diurnal
terrestrial animals of the desert the strong radiation from the soil is an important
ecological factor. Their black colour may be the consequence of raised melanin pro-
duction as a reaction to certain parts of the solar spectrum, just as melanin is in-
creased by higher metabolic activity. The latter has been experimentally demon-
strated in the phases of the Desert Locust (Schistocerca gregaria), where the black
colour of the gregarious hoppers contrasts with the pale green of the hoppers of the
solitary phase. Superimposed on the effect of raised activity is the effect of the
hours of sun basking (i.e. of intensive exposure to solar radiation) which are greatly
prolonged, especially in the first stages of the gregarious hoppers as compared with
the solitary ones. Another mechanism produces adaptive black colouration among
the animals found in areas of black lava, (e.g. Agama stelliopicae), or on burnt or
otherwise blackish soil (as in many grasshoppers).

Of special interest is an internal black pigmentation in the peritoneum and
pleura of desert reptiles. Alpine climatologists stated some decades ago that alpine
lizards (Lacerta spp.) show black pigmentation of this kind. We have just begun to
pay attention to this phenomenon in our region, and we find it to be of common oc-
currence in lizards with diurnal summer activity, such as Acanthodactylus, Lacerta,
Agama, Eremias, etc. The deepest, velvet -black pigmentation of this type occurs
in the chameleon, which is especially exposed to sun radiation. In some species,
even the omentum and its fat show patches of black pigment. Before speculating
about these phenomena we must have information as to whether black pigmentation
is common to lizards and snakes in more northern climates. Some English species,
Lacerta vivipara and L. agilis possess a black peritoneum, whilst Anguis fragilis,
Vipera berus and Matrix natrix do not. The long exposure to strong solar radiation
of all diurnal desert animals between spring and autumn raises a number of questions
of physical physiology. The first and most important of these is to what degree the
rays of various wave- length in the solar spectrum are able to pierce the dead part
of the integument. We therefore undertook (in co- operation with Drs Halperin and
Svirski) some preliminary measurements on the transmission of rays through the in-
tegument of freshly killed insects and reptiles. The experiments were conducted
with a Beckman Quartz spectrophotometer. The results indicated that the actual
quantity of transmission depends primarily upon the intensity of the radiation.
Through the transparent wings of a dragonfly (Crocothemis) transmission in all ran-
ges of the spectrum is very high, even higher than for normal glass. Yet through all
other objects transmission in the ultra-violet range is nil, except perhaps for a slight
transmission through the white -scaled forewings of Pieris rapae.. Some slight trans-
mission is always observed in the higher range of the visible spectrum, whilst trans-
mission is always important in the infra-red range, where in all cases it reached a
maximum at 1200fi. This applied also to black insects. In the reptiles we had ex-
pected some ultra-violet transmission as we had regarded the black pigmentation of
the integument as a reaction against it. Instead, transmission of ultra-violet through
the skin of back and belly of Ophisops elegans and two other lizards {Chalcides
and Eumenes) was found to be absolutely nil. It was very low in the visible part of

166

the spectrum, and was least of all in the infra-red. The protection of the tissues
of the reptile body by the corneous part of the integument against penetration of any
kind of solar radiation is thus extremely efficient.

These measurements were made with the subject under investigation near the
exit slit of the spectrophotometer, at a distance of 4.5 cm from the window of the
phototube. In a second series the subjects were placed much closer to the photo-
tube (1cm distance). The ray that reached the object was thus much more concen-
trated. In consequence all readings gave higher transmission (up to 100%), but the
general picture remained the same. The readings in this second series gave approxi-
mately maximum values at right- angle incidence to the rays, which in nature would
be an extremely unusual occurrence. The first series is certainly a better illustra-
tion of what actually occurs in nature. Absorption of heat at the actual angles of
ray- incidence may be even lower than in the first series. The peak of transmission
at 1200/x in all the objects tested led to a further exploration of the 1000 to 14000/i
region, using Beckman I. R. 2 (sodium chloride optical apparatus). These results
show a second peak at 5400/i and deep depressions at 3000 and 6000/i' The first of
these may be produced by an OH- bond; the second may be due to the presence of
water.

From the results of the experiments it is perfectly clear, that the fur and fea-
thers of mammals and birds are quite sufficient to prevent any transmission of the
ultra- violet and visible rays of the solar spectrum to the integument. The degree
of transmission in the infra-red, low as it probably is, deserves further study. Yet
these qualities of fur and feathers are certainly no special adaptation of desert ani-
mals, but hold good for mammals and birds in all biotopes.

The study of these problems is still very much at its beginning. The physio-
logical consequences of the colour and structure of the integument, the importance
of the angle at which radiation meets the integument, etc. need much detailed re-
search. One point only is clear: the integument is normally fairly well protected
against penetration into its living tissues, irrespective of colour. For the time
being we would stress not the differences observed in our various objects, but the
striking similarity in the transmission trend through all colours and integumental
structures. The two peaks of transmission at 1200 and 5400/z, and the two depres-
sions at 3000 and 6000/i, as well as the general non - transmittance of ultra-violet
rays are probably due to certain biochemical components common to the integuments
of all animals.

167

DOMESTICATED ANIMALS INHABITING DESERT AREAS

Dr Norman C. Wright
{Scientific Adviser to the Ministry of Food, London)

I do not think I need apologise for limiting my paper to domesticated animals,
for such animals are essential to the economy of the human populations inhabiting
desert areas; they provide milk (often the only form of liquid available for human
consumption) ghee and cheese, meat, hides and skins for clothing, hair for tents and
other purposes, means of human and baggage transport, and even fuel in the form of
dried dung. These products are provided not only in sufficient quantity to meet the
local population's own needs, but are frequently available for export from the desert
areas. In spite of this, textbooks on animal ecology have almost without exception
failed to devote more than an occasional paragraph or two to domesticated anim.als,
and particularly to those inhabiting desert areas, while little work has been done on
the physiological reactions of such animals to their environments. This applies
equally to practically all classes of desert stock, — not merely to camels but to fat-
tailed and fat-rumped sheep, goats, yaks and donkeys. Our largest body of know-
ledge is in fact on cattle, which are not typical desert animals, though they are of
course widely found in the semi- arid areas bordering on deserts, such as East and
Central Africa and Northern India. For this reason I have not hesitated to draw on
cattle for certain illustrative material in this paper.

The effects of climate on the morphology of desert animals may be roughly
classed as direct and indirect. The direct effects are associated with environmen-
tal temperature (including solar radiation), with humidity, and with air movement.
These can affect body size and conformation, the skin structure and nature of the
coat covering, and possibly certain other properties (such as the subcutaneous fat
layer) which may affect the absorption or dissipation of heat. The indirect effects
are associated with the environmental vegetation and water supplies, and hence with
the nutrition of the animal. These may to some extent affect size ; they may affect
conformation in so far as this is influenced by the local deposition of nutrient re-
serves designed to tide over rainless periods; they may affect the mechanism of
food intake, and of water intake and conservation ; and they may be related to the
animal's facility for speed of movement.

It will therefore be desirable to summarise briefly the climatic environments to
which domesticated desert animals need to be adapted. Since such animals are too
large to avoid the extremes of climate in ways which are possible for smaller mam-
mals (e.g. by burrowing) it is necessary to take into account the full climatic en-
vironment of the open desert. For this purpose ordinary routine meteorological data,
however inadequate, is the only source to draw on.

In sub- tropical and tropical deserts (such as the Sahara) the outstanding fea-
ture is the uniformly high temperature. As one goes north-east across the Great
Palaearctic Desert region these consistently high temperatures are no longer found,
but they are replaced by two characteristics, — by extreme seasonal variations and
(particularly in the mid- regions at lower altitudes such as the Iraqi Desert) by

168

periods at extremely high temperatures, — exceeding in fact those of the tropical
regions. These high temperatures are paralleled as one goes further north-east or
as the elevation increases, by extremes of low temperature, — falling in the Iranian
Plateau to freezing point, in the Turkestan Deserts to figures well below freezing
point, and in the Gobi Desert of Mongolia to figures below zero degrees Fahrenheit.
It is clear therefore that animals inhabiting desert regions must exhibit marked
tolerance to heat, and in some areas, equally marked tolerance to cold. It is not, of
course, possible to give strictly comparable figures for solar radiation, but they may
be assumed to run roughly parallel with temperature.

Heat tolerance is, however, also affected by humidity, whether the animal loses
heat by sweating, by transudation or by respiration. Variations in humidity are most
conveniently shown by means of climographs, in which the dry bulb temperature is
plotted against the relative humidity. From a study of such climographs, one would
expect the need for heat tolerance to be low in the temperate areas of Europe; in a
tropical desert like the Sahara there will need to be a high heat tolerance, but there
will be opportunities for heat regulation by evaporative loss at the low relative
humidities ; in wet tropical areas, where high temperature and high humidity are
combined, the dissipation of heat will however be especially difficult. Desert areas
in intermediate climates show strikingly wide variations in environmental conditions,
from hot dry in the summer months to cold wet in the winter; clearly the animals in
such areas will have to possess wide adaptations to the direct effects of climate.
It is perhaps relevant at this point to note that, while few reliable records of air
movement are available, desert areas may normally be taken to involve free air
movement, while wet tropical jungle conditions are usually characterised by rela-
tively still air.

Turning to the indirect effects of climate (i.e. those concerned with vegetation
and water supplies), these are best reflected in precipitation curves. VThether rain-
fall follows the continental or the Mediterranean pattern, the outstanding features of
all desert areas are not merely their low total precipitation, but the long periods
during which there is virtually a complete absence of rainfall, — extending for as
much as six months out of the year. The only difference between semi- arid areas
and the more typical desert areas is the higher total precipitation in the former
during the rainy season, which will affect both the type of vegetation and the water
resources. But all are characterized by long rainless periods, with consequent vio-
lent fluctuations in herbage growth and therefore in facilities for natural grazing.

May I turn now to actual illustrations of the climatic effects on animal mor-
phology and adaptation. If in doing so I seem to rely too largely on teleological
argument, I hope it will be realised that this is inevitable in view of our present
meagre and non- quantitative knowledge of the subject.

The direct effects of climate on the size and conformation of animals have been
summarized in Bergmann's and Allen's Rules. The former postulates a larger body
size in the colder climates, — the effect of the larger body size being to reduce the
surface area available for the dissipation of heat. The latter postulates a parallel
lessening of the extremities in the colder climates. Put concisely, one would expect

169

a larger and more compact body in areas of low temperature, and a less compact
body with longer limbs and greater surface area in hot climates. As regards coat
covering, one would naturally expect a thicker coat — with correspondingly greater
insulation ~ in colder than in warmer climates, though this may of course be affected
by seasonal shedding.

These differences are well illustrated in the conformation and coat formation of
sheep as one goes from the northern latitudes to the equator, i.e. as the environ-
mental temperature increases. The Dorset Horn sheep is typical of our own tem-
perate area. The body is of fair size, the conformation compact, the legs short, the
neck stocky and the ears small. There is a thick and compact wool coat. This ani-
mal may be compared with the Kurdi sheep, typical of the Northern Iraq desert. Here
the body is still of fair size and the coat ample, as would indeed be necessary dur-
ing the cold seasons. But the legs are longer, the neck less stocky and the ears
large, — a phenomenon which is incidentally paralleled in Hamilton's work on ear
size in hares. The desert sheep of the tropics (for instance of Eritrea and the Sudan)
illustrate the extreme development of these various features. The body is far less
regular and compact, the legs exceptionally long, the neck elongated and the ears
large. Moreover, there is now a complete change in the nature of the coat, which in
place of wool consists of fine short hair, — a characteristic typical of sheep in all
hot tropical regions of low elevation. I say 'of low elevation' because at higher al-
titudes (and therefore at lower environmental temperatures) this generalization no
longer holds. The sheep of the Yeman plateau, which is at the same latitude as the
Sudan, are not only wooUed and otherwise comparable to those of the northern
deserts, but are in fact virtually earless.

Closely comparable changes in conformation and coat thickness are found in
desert goats. The Angora or Maraz goat — the origin of our Mohair supplies — which
is found in Turkey and the extreme north of Iraq, is relatively compact, with shor-
tish extremities, small ears, and a copious covering of fine hair. The Persian goat,
still well coated, is somewhat longer limbed and longer eared, with a less compact
body,- characteristics which are exaggerated (particularly as regards ear size) in
the Syrian desert goat. Here again, however, there is a dramatic change in both
form and coat in the tropical regions. As with the tropical desert sheep there is a
marked elongation of the body and neck, the legs are characteristically lengthened,
and the ears are almost uncomfortably large. Moreover the thicker coat of the nor-
thern type is again replaced by one of fine short hair.

Too much influence on length of leg should not, of course, be attributed to
direct climatic effects; apart from other reasons, animals in desert areas are charac-
teristically nomadic, and therefore require facilities for long and often relatively
rapid movement. This is perhaps best illustrated (if I may break my sequence) in the
length of leg of animals inhabiting northern deserts, — of which an excellent example
is the wild ass of the Gobi Desert. Donkeys are the most important class of equines
used by man in desert areas and their ability to cover long distances over waterless
tracts is one of their greatest assets, — not unconnected, probably, with their origin

170

May I now turn to cattle, which though not primarily desert animals, are widely
grazed nomadically in semi -arid areas. I should perhaps first refer to the Yak,
which while not classed as true cattle, are close relatives and are typical of the cold
Asiatic deserts of Mongolia and of the Tibetan plateau. This animal again shows all
the characteristics of cold resistant types, with relatively heavy and compact body,
small extremities and a thick insulating coat. The same may be said of our own
temperate cattle as illustrated in the Highland and Galloway breeds. It is instruc-
tive to compare these with their counterpart in the tropical desert areas. Here, how-
ever, contrary to Davidson's findings for North America, Bergmann's rule regarding
larger body size in cold areas no longer holds; cattle in desert and semi- arid areas
are among the largest in the world. The explanation may well lie in the fact that, in
such animals, the skin surface is very greatly increased by the abnormal develop-
ment of the dewlap and, in the male, of the sheath. Both developments are parall-
eled by a marked increase in the fineness of the coat in comparison with temperate
breeds. It would perhaps not be out of place to note at this point that the large size
and exaggerated skin areas of the cattle of the hot deserts is not found in the hot
but humid areas. Here the animals, while having normal sized extremities, are quite
definitely dwarfed.

Although I have no personal experience of the colder Asiatic deserts, it appears
that Bergmann's and Allen's rules are equally applicable to camels, — perhaps the
best known desert animal. Thus the two- humped Bactrian camel, common to the
cold deserts of Turkestan, is described as 'distinguished from its Arabian relative
not only by the presence of two humps, but by the facts that it is heavier, more com-
pact, and shorter in the leg, and that it has a heavier coat of longer hair'.

Camels form the obvious introduction to the second aspect of climate in rela-
tion to domesticated animals inhabiting desert areas, i.e. to the indirect effect of
climate via the vegetation and water supplies. One minor but none the less impor-
tant adaptation is that associated with the nature of the vegetation. Xerophyllous
vegetation is notably hard and frequently spiked or thorny ; yet the camel is able to
derive its nutrients from such material which, indeed, forms an important part of its
natural grazing. But the most striking feature of the camel, which makes it of spec-
ial value as a true desert animal, is its ability to store a reserve of fat in the hump
to provide energy (not water) to tide over rainless and therefore vegetationless
periods.

This feature is not confined to camels; it is less widely recognized but prob-
ably more important from the aspect of desert utilization as a feature of fat- tailed
and fat-rumped sheep. For some six months of the year deserts are (as I pointed
out earlier) normally devoid of rain. During much of this period the surface of the
desert is practically free from any vegetative growth. With the onset of the rains
the whole picture is transformed'by the growth of a thick carpet of annual herbage
plants. The duration of this vegetation is, however, relatively short, and within a
few weeks of the cessation of the rains it dries up and fragments, leaving only the
sparsest supply of grazing nutrients for stock. Nevertheless, owing to the wide
variations in the locality of rainfall, a flock may be able to secure more or less con-'
tinuous grazing by constant movement across the desert — the basis of nomadism.

171

as fast movers and their remarkable ability to make good their water deficit by drink-
ing at one time quantities which — weight for weight — would be impossible to man.

Reliance on a precarious rainfall does, however, involve the need for some
form of adaptation which will furnish nomadic stock with a mobile reserve of food.
Vihile in the camel this is located in the hump, in the sheep it is located in the tail.
The exact form of the tail varies with different types of sheep. In sheep inhabiting
the northern deserts it usually consists of twin lobes, the upper surfaces of which
form a continuation of the woolly coat; the under surface being bare of wool. In
tropical sheep the tail is long, rather than wide and lobed, but is still capable of
storing large quantities of fat. The tail of the fat- tailed sheep may, iii the lush
period, attain a weight of 151b. or more; in the lean period it will shrink to an
empty bag or 'rope' of small proportions. It would not, I think, be out of place to
draw attention to the fact that the existence of a f at - tail is in no way inimical to
high productivity, as shown by the fine udder development and good mutton confor-
mation of many desert sheep.

I have attempted in this paper to describe some of the more typical examples
of the adaptations of domesticated animals inhabiting desert areas. I have done
this with a two -fold object. First, I was anxious to stimulate interest in a field
which has in the past been much neglected by zoologists and physiologists, and
which not only merits increased attention on account of its economic importance but
furnishes a mass of unsolved but intriguing problems. We are, for instance, still un-
aware of the factors influencing tolerance to wide variations in environmental tem-
perature, apart from specific tolerances to heat and cold. And even in regard to the
latter, substantial progress has only been made with cattle, and little or no work
has been done on camels, fat -tailed sheep, goats and donkeys. We know little of
the mechanism causing local fat deposition, or of the reasons for the wide varia-
tions in tail and rump pattern. These and many other problems provide wide scope
for the research worker.

But my second reason for presenting the paper is that it justified, in my view,
an extremely cautious approach to any proposals (and many have been made) to in-
troduce into desert and semi- arid areas types of livestock which may in their own
environment be more productive but which are unsuited either to the climate, the
vegetation or the inevitable nomadic life of desert animals. I have indeed tried to
show that the existing desert animals are themselves capable of quite outstanding
production in spite of their harsh environment. If, as I firmly believe, such animals
are destined to continue to play an essential role in the utilization of deserts, it
will not be by their replacement by — or even their inter- breeding with — so- called
'improved' livestock. Rather must we look for improvement by the better selection
among the indigenous animals themselves and by the partial alleviation of the desert
environment, — through improvements in water supplies for stock in grazing areas,
through the increased practice of semi - nomadism, and - where this is impracticable -
through efforts to conserve fodder as an external reserve to reduce the demands on
the internal reserves of the animal itself. By such means — and only by such means -
can we hope not only to maintain but to increase the contribution of domesticated
animals to the desert economy.

172

WATER CONSERVATION IN SMALL DESERT RODENTS

Dr Bodil Schmidt -Nielsen
(Cincinnati, U.S.A.)

Animals inhabiting deserts have in the course of evolution acquired morpho-
logical and physiological characteristics which enable them to live £ind thrive in an
environment that is hostile and uninhabitable to other closely related forms.

Among rodents, specially adapted desert forms are found in all the major deserts
around the world. It is interesting that these rodents, although they belong to dif-
ferent families are similarly adapted to their environment.

In the north American deserts the kangaroo rats and pocket mice (Dipodomys
and Perognathus) of the family Heteromyidae are found. In the great Palaearctic
desert we find Gerbillus, Meriones and Dipodillus, belonging to the family Muridae;
and Jaculus, Dipus and Alactaga of the family Dipodidae. In South Africa, in the
Kalahari desert, the rodent Pedetes, family Pedetidae, is found; and in the Austra-
lian deserts the family Muridae is represented by Notomys. All these rodents have
several morphological features in common (Fig. 1) thus they are all adapted to a
bipedal saltatorial life with elongated hindlegs and reduced number of toes. Several
of them have cheek pockets (Heteromyidae)or gular pouch {Notomys fuscus and
cervinus) and they all have greatly inflated bullae auditivae. They are nocturnal
and stay in their underground burrows during the daytime.

Also with respect to their physiological adaptation to the environment do we
find striking similarities between these animals.

Personally we have worked primarily with kangaroo rats (Dipodomys) but we
have also had the opportunity to work with jerboas from Arabia and we found that
their water problem has been solved in the same way as the kangaroo rats' has.

It seems likely that the same would be true of many other desert rodents.

Can the desert rodents live without water?

When we started to investigate the water metabolism of the kangaroo rats the
first question to be answered was whether these animals can live entirely without
drinking water. Previous reports in the literature indicated that this was the case
but more accurate investigation was desirable to solve this problem. We kept kan-
garoo rats on diets of dry grain (rolled barley), without water. They maintained
their body weight, and some even gained body weight over a period of 2 months.

The next question was: Can the animals during a period of water deprivation
store their metabolic waste products to avoid spending water for excretion? This
is practiced by the lungfish which stores urea in its body when estivating in the
dry mud. The stored urea can amount to 2-4% of the body weight. Urea and elec-
trolyte concentrations were determined in the plasma of (1) kangaroo rats that were
freshly trapped, (2) kangaroo rats that had been on a moist xjiet (barley and water-
melon) and (3) kangaroo rats that had been fed on dry barley only from 2-8 weeks.
The same average urea and electrolyte concentration of the plasma was found in all

173

the groups and the concentrations were of the same order of magnitude as the con-
centrations found in other rodents, showing that there is no storage of waste pro-
ducts during water deprivation.

Figure L
Desert rodents from different parts of the world.

1. Gerbil from East Africa (From Buxton, 1923)

2. Jerboa from Egypt (From Buxton, 1923)

3. Kangaroo mouse, Nolomys. from Australia (From le Souef & Burrell, 1926)

4. Kangaroo rat, Dipodomys from North America (From Scientific Monthly, 69: 180, 1949.)

The third question was: Do the animals have a water storage that is gradually
spent during low water intake? To determine this the water content was determined
in animals that had lived on dry barley diet for varying periods of time. From Fig. 2
it is seen that the animals that had lived without water for 52 days had the same
average body water percentage as the animals that had been kept without water for
14 days only. There was no difference in percentage of body water between groups
of Heteromyids (kangaroo rats and pocket mice) on dry diet (barley alone) or wet
diet (barley and watermelon) while white rats and Neotoma (wood rat, family Crice-
tidae) had lower percentage of body water on dry diet than on wet diet.

Since the kangaroo rats maintained on dry barley without water (1) show no
weight loss, (2) excrete all of their metabolic waste products and, (3) do not get a
decreased percentage of body water, we can conclude that the animals simply

174

D.MERRIAMI

J J.

52 DAYS

(6 IND) 1

28 DAYS

(7 IND)

'

21 DAYS

7 IND)

14 DAYS

(3 IND) 1

L 1

-1 1 1

60%

62

64

66

66

70

72

74

Figure 2.

Percentage of water in the body of kangaroo rats which had lived on dry diet for different
periods of time. The two ends of each block represent respectively the highest and the lowest
value found for the group in question. The mean value is marked with a vertical line across

the block.

maintain water balance, which means that the intake of water on the dry diet is suf-
ficient to cover their needs for excretion. They can therefore be ejected to be
able to live for any length of time on the dry diet. ,

If we want to make an account for the water intake and water output of an ani-
mal we have:

Intake Output

Drinking water p„o,.^,^.-:«„/ skin

Evaporation Ij^^gg

Water in food Water in urine

Oxidation water

Water in faeces

Intake

Drinking water: For the kangaroo rat the drinking water can be disregarded. No
drinking water is available in their natural habitat except on rare occasions after a
heavy rain fall. Dew does not occur normally, often the relative humidity only
reaches about 40% at night. When kept in captivity on the experimental diet of dry
barley, no drinking water was given.

Wafer in food: The content of free water in the food, also called preformed water
(to distinguish it from the water formed by oxidation) is quite low when the diet
consists of dry barley. Grain is hygroscopic and its water content will therefore
vary with the relative humidity of the surrounding air. Determinations of the per-
centage of water in barley at different humidities gave the results in Table 1.

By storing seeds and other plant material in the more humid burrow, the ani-
mal can increase its water content somewhat. The kangaroo rats have large
storages of food in their burrows.

Oxidation water: In all animals water is formed by the oxidative metabolism. The
amount of water formed when a certain amount of any foodstuff is combusted can
easily be computed when we know the composition of the foodstuff. Table 2 shows

175

that 13.4 g of water is formed when lOOkcal of barley (corresponding to 25 g of dry
barley) is metabolized. This amount far exceeds the amount of preformed water at
all humidities and is therefore the main intake of water for the animals living on
dry grain. It cannot be increased in any mysterious way by desert animals as it
has sometimes been suggested. The desert rodents can, however, economize with
their water to a considerably higher degree than other mammals, as shall be shown
in the following.

TABLE 1

PREFORMED WATER

Water absorbed in pearled barley (lOOkcal) at various humidities

10%r.h. 33%r.h. 43%r.h.

76%r.h.

Gram of water per lOOkcal of pearled barley

(25g dry barley) 0.93 2.55 2.93

4.53

TABLE 2

OXIDATION WATER FORMED AND OXYGEN REQUIRED

WHEN PEARLED BARLEY

IS

METABOLIZED

Carbohydrate

Fat

Protein

Total

Gram of pure foodstuff per lOOkcal of

pearled barley

22.00

0.28

2.31

Gram of water formed per gram of

foodstuff combusted

0.56

1.07

0.40

Gram of water formed per lOOkcal of

pearled barley

12.20

0.30

0.92

13.4

Litre of oxygen used per gram of

foodstuff combusted

0.80

2.01

0.95

Litre of oxygen used per lOOkcal of

pearled barley

17.60

0.57

2.19

20.4

Output

Evaporation: The evaporation from the respiratory tract is proportional to the venti-
lation which again is proportional to the oxygen uptake. In man the exhaled air is
saturated with water vapour at a temperature of about 33°C which is slightly below
the rectal temperature. When completely dry air is inspired all the moisture that is
necessary to bring this air to saturation at 33°C must evaporate from the respiratory

176

tract. If we calculate the amount of water evaporated from the lungs of man in dry
air we arrive at 0.84 ml of water per ml of oxygen utilized. In kangaroo rats and
some other small rodents we measured the total evaporation (body surface and res-
piratory tract combined) in dry air, simultaneously with the oxygen uptake. The
results are listed in Table 3. The total evaporation in the kangaroo rats per ml oxy-
gen taken up is very low compared with the evaporation from the lungs alone in man.
In the white rat the total evaporation is approximately the same as from the lungs of
man.

TABLE 3

mg. HjO/ml.

O2 consumed

Dipodomys merriami

0.54

+

0.01

Dipodomys spectabilis

0.57

+

0.03

Perognathus baileyi

0.50

+

0.03

Rattus norvegicus, var. Alb., Albino rat

0.94

+

0.03

Mus musculus, var. alb., Albino mouse

0.85

+

0.03

Mus musculus. House mouse

0.59

Peromyscus crinitus, Canyon mouse

0.54

Cricetus aureus, Hamster

0.59

+

0.02

The explanation for the low evaporation in the kangaroo rats is that the ex-
pired air is saturated with moisture at the temperature of the nose which is about
10 C lower than that of the body. The white rats also have a low nose temperature
and low evaporation from the lungs. The higher total evaporation from the white
rats can probably be accounted for by a higher transpiration from other parts of the
body. The evaporation from the skin is negligible for the kangaroo rats. The eva-
poration from the lungs decreases with increasing amount of water vapour in the
inspired air.

Ue measured the relative humidity and temperature in the burrows of the kan-
garoo rats in order to determine how much water the animals would save by staying
in their underground burrow. The measurements were done by tying small micro-
climate recorders to the tails of kangaroo rats and releasing the animals in front of
their own burrows. The animal would carry the instrument to its nest chamber. The
instrument could then be dug out after the humidity and temperature had been recor-
ded for 10-12 hours.

The measurements showed that the absolute humidity in the burrows is about 3
to 4 times as high as the simultaneous humidity outside the burrow, which means a
considerable reduction in the amount of water evaporated from the respiratory tract.
The temperature in the burrow is much more constant than outside and always stays

177

below 30°C. The animal then by its nocturnal habits avoids the extreme heat of
the day and can therefore avoid spending water for heat regulation.

Urine: Water can be saved by increasing the concentration of the solids in the
urine. This is done to a high degree by the kangaroo rats. Table 4 shows the
maximum urine concentrations in man, white rat and kangaroo rats. The kangaroo
rat can excrete a much more concentrated urine than can man and the white rat.

TABLE 4
MAXIMUM CONCENTRATIONS OF ELECTROLYTES AND UREA IN URINE

Electrolytes Urea

Man 0.37N (2.2%) l.OM (6%)

Norway rat 0.60N (3.5%) 2.5M(15%)

Kangaroo rat 1.2 N (7%) 3.8M(23%)

With respect to electrolytes a kangaroo rat can excrete a urine that is twice as
concentrated as sea water. This brought up the old question whether sea water can
be utilized as drinking water by mammals. To test this kangaroo rats were fed on a
diet of soy beans. On this high protein diet, the animals cannot maintain water
balance without additional water. They were offered sea water to drink. A control
group was given soy beans and tap water. From Fig. 3 it is seen that both groups
of animals lost weight initially until they learned to drink, then they increased in
body weight until they reached a steady state. The animals on sea water were
doing just as well as the animals on fresh water. Sea water can then be utilized as
drinking water by the kangaroo rat.

Faeces: The faeces excreted by the kangaroo rat are very dry compared with the
faeces of the white rat. Determinations of the moisture content in the faeces of
kangaroo rats and white rats and determinations of the amount of faeces eliminated
when a certain amount of food was metabolized showed that the kangaroo rat looses
only 0.76g of water in the faeces when lOOkcal of barley is metabolized while the
white rat looses 3.4g of water when the same amount of food is metabolized.

Complete account for intake and output of water

With the information obtained above it is now possible to calculate at what
humidities in the surrounding air the kangaroo rat is able to maintain water balance.
Fig. 4 shows the result of the calculation. The calculation is based on the intake
and metabolism of 100 kcal of barley, corresponding to 25 g of dry pearled barley.
The ordinate gives the water intake and the minimum water output in grams per 100
kcal of pearled barley metabolized. The abscissa gives the humidity in the environ-
mental air.

178

+20

FRESH WATER

10 15 20

Figure 3.

Weight changes in adult kangaroo rats kept for 19 days on a diet of soy beans, given fresh
water, sea water, or no water for drinking. The weight changes are given in percentage of the

initial weight.

mg HgO/ 5 rug

/liter air

10 mg 15 mg

Figure 4.

20 mg

Kangaroo rats; Total water intake and total water output at various atmospheric humidities at

25°C.

Ordinate: Water intake and output in grams per 100 kilocalories of pearled barley metabolized.
Abscissa: Humidity in the environmental air.

179

mg HJOy
Av

liter air

White rats; Total water intake and total water output at various atmospheric humidities at

25°C.
Ordinate: Water intake and output in grams per 100 kilocalories of pearled barley metabolized.
Abscissa: Humidity in the environmental air. The point of intersection for the two curves
should not be taken too seriously in this graph because of inaccuracies in determining the

evaporation at higher humidities.

The lowest curve shows the evaporation from the animal. The evaporation at
zero humidity is calculated on basis of the determination of evaporation in dry air.
From Table 2 it is seen that 20.4 ml of oxygen is used per lOOkcal of barley. Then
the evaporation must be 11.0 g HjO. The evaporation decreases with increasing
water vapour content in the inspired air as shown by the sloping curve. The mini-
mum water loss through the urine is superimposed on the curve for evaporation. It
is calculated in the following way: Pearled barley contains 2.31 g protein per 100
kcalr which, when metabolized gives 0.79 g urea. The maximum concentration of
urea that kangaroo rats can excrete is somewhat above 20%. The minimum amount
of water required for the renal excretion of 0.79 g urea is therefore 3.4 g.. The water

180

loss through faeces, 0.76 g is again superimposed on the two curves giving the
minimum total water output when lOOkcal of barley are metabolized.

For water intake we have the oxidation water which, of course, is independent
of the environmental humidity. The amount of preformed water in the barley increa-
ses with increasing humidity. The top curve gives the total water intake when 100
kcal of barley are metabolized.

From the diagram it can be seen that the water intake exceeds the minimum
water output at all humidities above 2.2 mg HjO per litre air or 10% relative humidity
at 25°C. Below this value the water output exceeds the water intake, and the ani-
mals are in negative water balance.

Fig. 5 shows a similar diagram for white rats. It is seen that white rats have a
considerably higher water output and cannot be in positive water balance at any
humidities when they do not get drinking water with the barley.

The results in the diagrams were obtained by calculation. It was desirable to
check them by actual determinations of the animals' response to changes in the en-
vironmental humidity. A group of kangaroo rats was kept at different controlled
humidities at 25°C on a barley diet for periods around 10 days. The results showed
that the animals can maintain or. gain body weight at humidities above 10% relative
humidity. At 10% relative humidity and lower the animals lose body weight. )Xhite
rats were unable to maintain body weight even at 90% relative humidity when not
given additional water.

In its natural habitat with its relatively humid burrow the kangaroo rat will
have a certain margin of safety. The absolute humidities which are measured in
their burrows vary between 7 and 14 mg of water per litre air. The humidity outside
is considerably lower also at night. By its remarkable ability to conserve water
the kangaroo rat shows a high degree of adaptation to its arid habitat.

181

HEAT REGULATION IN SMALL AND LARGE DESERT MAMMALS

Professor Knut Schmidt -Nielsen
(Cincinnati, U.S.A.)

Desert rodents usually lead a nocturnal life and spend the day in underground
burrows. In this way they escape the excessive heat load that would be imposed by
high solar radiation, high air temperature and high ground surface temperatures.

The burrow temperature normally does not exceed 31*^, even on the hottest
day, as shown for example by Vorhies' investigations of the microclimate of kan-
garoo rat burrows.

Kangaroo rats, like other rodents have no regular sweat glands and do not
sweat. However, if they are exposed to high temperatures, it will be found that to
some extent these animals are able to keep the body temperature below that of the
environment by the evaporation of water.

In laboratory experiments it was found that the body temperature of the kan-
garoo rat will increase beyond the usual of 36- 37°C if the surrounding temperature
rose above about 35°C. A further increase in ambient temperature would lead to a
corresponding increase in body temperature, apparently without causing any physio-
logical reaction that would keep the body temperature from rising. However, if the
body temperature approached the lethal limit (around 42°C) a copious secretion of
saliva would occur, wetting the fur under the chin and throat, and evaporation would
keep the body temperature from rising further. This 'emergency heat regulation',
which is used only when conditions are critical to survival, will enable the animal
to keep its body temperature even slightly below that of the surroundings for a
short time.

There were differences in the reaction of different individuals, and it was found
that some kangaroo rats could survive at 43°C for at least 20 minutes in experiments
where white rats died at 39°C. The amount of water used for this evaporation is so
great that to continue for a long time would be impossible. The animals under the
conditions mentioned above had lost about 15% of their total body water, which we
know is not far from the 20% which is considered the limit for desiccation that can
be tolerated by mammals.

(Similar reactions of excessive salivation under heat stress have been found in
other animals, such as mouse, white rat, guinea pig, cat, swine, etc. For compari-
son, white rats were tested in the same experiments as kangaroo rats. It was found
that there was no apparent ability to lower the body temperature in the rats, and they
died at much lower air temperatures (39°C) than the kangaroo rats although the lethal
body temperature is nearly the same in the two species.)

The very high rates of evaporation in a small animal which uses water for heat
regulation is due to the fact that the relative surface area is greater in a small
body than in a larger one.

The amount of heat that should be dissipated in order to keep the body tempera-
ture constant in a hot environment equals the sum of the heat of metabolism and the

182

TABLE 2

The evaporation in different animals calculated from the assumption that the

water loss necessary to keep the body temperature constant under desert conditions

is 0.6kg. per m* body surface per hour. This value was observed in animals of 16-

96kg. body weight (see Table 1). The surface areas as used in Table 2 are calcu-

lated from the formula S = 0,1 x B"**', where S is body surface in square meter and B

is body weight in kilogram. This is an approximation only, but of sufficient accu-

racy for the considerations involved.

Body weight, kg

Surface m*

Total evaporation,
kg. per hr.

Evaporation % of
body weight per hi.

Camel

500

6.43

3.86

0.77

Donkey

95

2.11

1.27

1.33

Man

70

1.72

1.03

1.47

Dog

16

0.64

0.384

2.38

Jack -rabbit

2

0.159

0.0954

4.77

Kangaroo rat

0.1

0.0214

0.0128

12.8

Mouse

0.021

0.00753

0.00452

21.5

01 kg, Kangaroo rat

3

o

<u
Q.

I

>>
■o
o

CD

2 kg, Jock-robbit

^6 kg, Dog ^Man

32_ 95 kg, Donkey

100

200 300

Body Weight, kg

400

Gomel
— — o

500

Figure 1.
The curve shows the rapid increase in evaporation of mammals with diminishing body
size. The curve is based on the estimated hourly evaporation of mammals of differ
ent body size under desert conditions as given in the last column of Table 1.

183

heat gained from the surroundings by conduction and radiation. The heat gain from
the surroundings of a physical body is proportional to the effective surface area.
Also the metabolic heat of mammals is approximately proportional to the body sur-
face, and consequently the total heat to be dissipated is roughly proportional to the
surface area.

Table 1 gives a summary of rates of evaporation found under actual desert con-
ditions in a few mammals of different body sizes. It will be seen that the dog loses
body water at a rate which is more than twice as high as that of the donkey if cal-
culated on the basis of body weight. However, we have just found that the total
heat gain should be approximately proportional to the surface area of an animal, and
in the table we note that the water loss per surface area actually is nearly the same
in these animals, irrespective of their body size.

It would be permissible to extend this reasoning to animals of even smaller or
larger body size in order to estimate how much water should be evaporated in order
to keep the body temperature constant in a desert climate similar to that actually
ejqserienced in the observations given in Table 1. Such calculations are, of course,
very rough approximations, and give only an order of magnitude of the expected
rates of evaporation.

TABLE 1

The evaporation from donkey, man, and dog as observed in the daytime under
actual desert conditions in the South-western United States by different investiga-
tors. (Dill, Amer. J. Physiol. 19: 123, p. 377; Adolph, Ibid., p. 371; Dill, Ibid..
104, p. 36).

Body weight, kg.

Evaporation, % of body
weight per hr.

Evaporation, kg per m*
per hr.

Donkey (Dill)
Man (Adolph)
Dog (Dill)

96
79
16

1.24
1.41
2.62

0.573

0.60

0.657

The results of such calculations are given in Table 2. In the last column it
will be seen that an increase in body size from the donkey to the camel causes a re-
duction in the rate of evaporation to not quite half the value. On the other hand, in
small animals the rate of evaporation will increase rapidly with diminishing size.
Since the relationship is an exponential function, the rate of increase gives a loga-
rithmic curve as shown in Fig. 1. A mouse attempting to maintain constant body
temperature in the hot desert would have to use water in an amount exceeding 20%
of its body weight per hour. This amount of water loss is fatal, and here we find the
explanation for the fact that mammals of small body size usually do not sweat or in
other ways use water for heat regulation. If exposed to the heat for any length of
time there would be a choice of evaporation and death from dehydration, or no eva-

184

poration and death from heat. The desert rodents avoid this dilemma by leading a
nocturnal life and staying underground during the daytime. Only under exceptional
circumstances if the body temperature should rise close to a fatal level, will they
use water for heat regulation. Under these circumstances the water will last for a
short time only. The actual time of survival found in the experimental work described
earlier in this paper was close to that which can be calculated from the surface- body
weight relationship.

It is evident that the heat exchange between the environment and the body is
not as simple as assumed above. A very important factor is that the amount of heat
that reaches the body from the environment depends upon the surface insulation of
the body. In other words, the fur of an animal (or clothes in man) will cut down the
heat gain to an extent corresponding to its insulation value. This may seem para-
doxical, but it is nevertheless true that clothing in the desert reduces the heat load
and therefore is of advantage to the water economy. The value of clothing in man
has been clearly demonstrated under actual desert conditions by Molnar (Fig. 2).
However, water economy and greatest feeling of comfort do not necessarily coincide.

-100

-50

Environmental Heat Gain Calories per hour

+ 50 +100 +150 +200 +250 +300

+350

+400

^^~

1

— I 1 — •-

■

-^i 1

Nude
Clothed

Tropics at night
Tropics in 'Jungle'
Tropics in Laboratory

Tropics in Sun
Desert in Sun

Tropics at Night

Tropics in 'Jungle'
Tropics in Laboratory

im^l^

■^H

Des

Tro

ert in Tent
5ics in Sun

Desert in Sun 1

HHlHHIiiHiHH

1—

L

1

t 1

Figure 2.
This graph shows that tlie heat gain from a "hot" environment is much higher in the
"nude" man than in the "clothed" man- The graph is taken from V'oinar et al
{ Arner.J.Hyg. 44, p. 417). The "nude" men wore shoes and light shorts, the "clothed"
men wore light clothing including shirts and trousers. The difference between "nude"
and "clothed" is particularly significant in the desert sunshine.

185

It has been seen that even light clothing in man cuts down evaporation by a
major fraction of the total. The reduction in evaporation corresponds to the re-
duced heat gain from the environment, because we can assume that the metabolic
heat was the same. The insulation value of animal fur is considerably higher than
that of 'light clothing', and one can expect a considerable advantage in heat and
water economy due to the insulation of fur.

The advantage of increased surface insulation is of course limited. It is true
that an infinite insulation of the surface would reduce the heat gain from the en-
vironment to zero. This is of course not biologically possible, and furthermore,
there should remain means for dissipating the metabolic heat.

The only means of dissipating heat in an environment warmer than the body sur-
face is by evaporation of water. The heat is bound at the site of evaporation, and
here we will find some relations of importance to the effectiveness of evaporative
cooling.

A diagram of the animal surface is sketched in Figure 3. Water will appear in
the form of sweat on the surface of the skin. It will be seen that the water either
could wet down the fur and evaporate from the outer surface of the fur layer, or it
could evaporate at the skin surface and diffuse as water vapour through the fur to
the surrounding air. Since heat is bound at the location where water changes from
the liquid state to vapour, there is a considerable advantage if the sweat evaporates
at the skin surface without wetting the fur. The heat of evaporation will be taken
from the body as well as the outside air, and the amounts would be in reverse pro-
portion to the insulation of the layers between the source of the heat and the site of
evaporation. The fur layer between the site of evaporation and the hot environment
is a great advantage in reducing the amount of heat that reaches the site of evapora-
tion from the environment. However, if evaporation took place at the surface of the
fur, the fur layer would be a disadvantage by reducing the transport of heat from the
body and it would provide no insulation between the site of evaporation and the hot
environment.

It will now be clear that the most economical use of water for heat dissipation
includes the fur, and an increase in the insulation is advantageous as long as it
does not interfere with the dissipation of the water vapour. Furthermore, the eco-
nomy in the use of water will depend upon the ease with which heat is transported
from the body to the skin surface, i.e., the circulation in the skin and the insulation
value of subcutaneous tissues. An increase in the subcutaneous adipose tissue
would by its insulation properties directly disfavour an advantageous distribution of
the heat flow to the site of sweat evaporation. It can perhaps be assumed that there
would be reason to consider the distribution of adipose tissue in desert mammals
from this viewpoint. The thin skin and particularly the localization of depot fat in
e.g. the hump of the camel and the brahma cattle and the tail of the fat- tail sheep
may indicate the possibility that this distribution may have a value in the heat and
water economy as outlined above.

The principles outlined in this paper are an attempt to make it clear that active
heat regulation in desert animals of small body size is a nearly impossible propo-

186

sition because of the lafge quantities of water that would be required for evapora-
tion. They avoid the heat problems by underground life and nocturnal habits. On
the other hand, the larger animals cannot lead an underground life, but due to their
large size the problem of heat regulation is less severe. In the absence of exact
knowledge based on experimental work, a working hypothesis can be based on a
simple statement of the physical laws that govern heat exchange between the ani-
mal body and its surroundings.

in

>-
a:
<
a:

CD
<

Ul
QC

h-
<

a:
iij

Q.
UJ

AIR TEMP

AIR

'J-'f^W .*'•'■ !?■-■*-'■'' '■'•:'-'' '

frsrxtrsrs BODY TEMP

.SKIN:-:

BODY

Figure 3.
A simplified diagram of the temperature gradients at the surface of an animal when the am-
bient temperature is higher than the ix)dy temperature. The temperature gradients indicated by
the solid line ASB and the broken line AS' B, respectively, show in which direction heat flow
will occur under different circumstances. If sweat evaporates at the surface of the skin (S)
without wetting the fur the temperature gradients will be as given by the solid line ASB. If
sweat evaporates from the surface of the fur (S' ) the gradients will be as shown by the broken
line AS' B. In both cases heat flow to the site of evaporation from each side, but it is evi-
dent that much less water is required to maintain the gradients indicated by ASB than by
AS' B. In order to maintain constant body temperature the total heat flow along the gradient
BS must equal the metabolic heat. In the case of evaporation from the surface of the fur,
this gradient would have to be extended to S', requiring a lower temperature at S' than at S.
This lower temperature would further increase the steepness of the gradient AS', which
governs the heat flow from the air to the site of evaporation. The heat flow along the less
steep gradient AS (when water is evaporated at the skin surface) is much lower, and the steep-
ness (and the heat flow) will decrease as the insulation value of the fur layer increases. It
is further evident that a reduction of the insulating value of the skin itself (reduction of the
distance BS) will permit a steeper gradient to be set up for the heat flow from the body to the
surface, without a simultaneous increase in the steepness of the gradient AS. In other words,
the heat flow from the body to the surface is facilitated by a thin skin of low insulation value.

187

REACTIONS TO GREAT ENVIRONMENTAL HEAT IN ANIMALS

Dr Frank Marsh
(London)

The sun, we are told, is a shining example of an atom bomb. It is difficult for
people living in England — where the sun is rarely seen, and the weather is a
national joke — to have any idea of the severe trials undergone by travellers or in-
digenes in the Arabian Desert, or even in the cooler Sahara. The celebrated Wes-
tern Desert between Tripoli and Alexandria is probably cooler still; but this state-
ment may be disputed. 'Animal Life in Deserts* is a big subject treated scientifi-
cally in the classical publication of that name. The modern problem is to provide
living space for an ever increasing human world population, and also to provide
adequate nutriments for this human mass. The great deserts of the world, the
Sahara, the Arabian deserts, the Central Asian deserts, the deserts of California and
Mexico, the Australian deserts and the cold deserts at the poles are all being
thoughtfully surveyed by contemporary man. These great sterile wastes can all be
made fertile, green and productive by capital expenditure — as shown by Ritchie
Calder in 'Men Against the Desert' — and by the Tennessee Valley Authority in the
United States of America. Geologists and technicians know that 'wild cat' expendi-
ture of capital may provide very substantial returns for a relatively small outlay.
Valuable minerals and oil are nearly always found in barren, rocky or sandy wastes,
far from the outposts of civilization and subject to all climatic extremes, The fact
that some very valuable raw material is found even in the middle of the Sahara,
causes an immigration of technicians and their associated civil engineers, surveyors,
domestic and administrative staff and other parasites, who may include even a rude
medical or health service. This closely knit community will develop gardens,
bushes, shrubs and small trees, to mitigate the severity of the landscape, filter the
hot winds, provide some fresh vegetables and add to the amenities of clubs and
dwellings, however rudimentary. A little oasis will appear in a situation that — a
few years earlier — was nothing but a howling wilderness. This miraculous trans-
formation — for it is nothing less — is due to the patient spare time efforts of men,
and their devoted wives, with a desire for the amenities of life, but with no special
knowledge of desert reclamation, except what could be picked up as they went along.
The men who construct these commercial installations 'in the blue' are mercifully
free from many of the disorders of civilization; they do, however, risk a number of
unfamiliar disorders which are not absolutely confined to the brown tropics, but can
be described as very rare in temperate regions. One of the most dramatic and dis-
abling of these exotic afflictions of men is the syndrome often referred to as 'effects
of heat'. The effects of heat are, shortly, dehydration, high fever, affections of the
skin, with unconsciousness in the acute or hyperthermic cases, and lassitude, debi-
lity, faintness, malaise, slight fever, cramps, tetany, headaches, weakness and other
symptoms in the prodromal, sub- acute or chronic varieties of the disorder. Effects
of heat are preventable, and should be prevented in any settled community, but the
'wild cat' pioneers have none of the resources of civilization and take great risks.
Observations on the effects of heat on utterly unprotected personnel impelled me to

188

make a study of these acute manifestations in the hope of devising a rational, and if
possible, effective form of therapy. The mortality rate in human victims of hyper-
thermia or heat stroke is very high; effective therapy is an acute necessity : in War
perhaps even more than in Peace.

Our experimental animals were rabbits; they cannot sweat; the rabbit attempts
to cool its body — in a hot environment — by breathing quickly over the moist red
tongue and lips. In our initial experiments the emphasis was on respiration; after
exposure to the sun (138°F) the rectal temperature of an adult healthy rabbit reached
110. 0°F with respiration rate 125 per minute, carbon dioxide exhaled 230ml/sq.m./
min. and volume of expired air 2100ml/min. Ice was then applied to the whole of
the fur and the rectal temperature dropped to 104. 0°F, the respirations, carbon di-
oxide and volume falling to 120, 190, and 1700, but rising again to 140, 250, and
2700 at the end of the experiment. The animal made a complete recovery but gave
birth to two stillborn young a few hours later. As a result of a number of experi-
ments we found that if the respiratory activities were stimulated by a rising body
temperature, the animal tended to recover. At the peak of body temperature some
animals collapsed; one such with a rectal temperature of 112.2°F stopped breathing:
the carbon dioxide in the exhalations had increased as the body temperature rose,
but there was no compensatory increase in ventilation; the respiratory centre in the
brain appeared to have become relatively insensitive to increases in the carbon di-
oxide tension in the blood. Cyanosis was not observed in this animal, so the brain
cells may have been depressed by some other factor, mere heat or reduction of the
blood pH. This collapsed animal responded to ice applied to its fur, and behaved
normally for some hours, after its body temperature had been reduced, but late in the
evening he was discovered with a subnormal rectal temperature, pale ears, inaudible
heart beats, sighing respirations and very weak. The respirations had dropped to 60
per minute. This animal was painlessly destroyed to avoid further suffering and
portions of the body tissues were preserved for histological examination.

Another rabbit showed respiratory stimulation in the early stages of the experi-
ment, with respiratory depression near the peak of the body temperature (108.0°F) but
the respirations quickened again after cooling treatment. The ventilation increased
in response to the increased concentration of carbon dioxide in the exhalations —
and presumably, increased carbon dioxide tension in the blood — an indication that
the respiratory centre in the brain was sensitive and reacting normally; a good sign.
Yet another rabbit suffered a rise in rectal temperature to 113-6°F rather quickly and
died suddenly at the peak; before the crisis he had responded well. Still another
rabbit was taken up to a peak temperature (rectal) of 111.1°F respiratory stimulation
was shown until the crisis was reached, when there was a short period of depres-
sion, followed by further stimulation on cooling. This rabbit appeared to have been
successfully treated, and made a good recovery, which, however, proved only tem-
porary. Having spent the night in comfortable cool surroundings, he was found dead
next morning.

Provided the rectal temperature did not exceed lethal levels, the cardio -vas-
cular system seemed to adjust itself to the high body temperature during the acute
phases of heating up and cooling down. If the cooling process was delayed or omit-

189

ted, there were profound effects. But in some cases the acute phase was success-
fully negotiated and complete recovery seemed in sight, when a crisis of depression
occurred, of obscure origin and usually fatal. (This sequence of events is not un-
familiar in human cases of hyperthermia).

A very successful type of rabbit was a black female — No. 60 — exposed en-
tirely in the shade. Rectal temperatures were:- initial, 104. 1°F rising to 112. 7°F
and then, with cooling, falling to 101.0°F. Haemoglobin had an initial value of 105%
gradually falling to 92% at which level it remained for 28 minutes, then rose to 105%
at the rectal temperature peak and to 113% at the conclusion of cooling treatment.
Blood pressure was not estimated. Carbon dioxide and air volumes exhaled were ac-
cording to expectations. The heart beats of this rabbit were counted with a binaural
stethoscope (by tapping on a sheet of paper for ten seconds) on twelve occasions
during the experiment. Initially the heart beats were 288 per minute and loud, in-
creasing gradually with increase of body temperature to 350 beats per minute, loud,
continuing loud and rapid until just after the rectal temperature peak, when the beats
were 350 per minute and quiet. After cooling the beats were 300 per minute and
loud again. This rabbit recovered without complications.

Rabbit No 73 had a similar shade treatment, but is chosen because of the record
of blood pressure. Initially rectal temperature was 104. 2°F and blood pressure was
64.0 mm. Hg. rising to 80.0 mm. Hg. at rectal temperature 109.0°Fand falling to 42.0
mm.Hg. at the rectal peak of temperature (111.0°F). In the early stages of cooling
the blood pressure rose abruptly to 120.0 mm.Hg. and then fell, equally suddenly, to
20.0 mm.Hg. with recovery to 80.0 mm.Hg. about ten minutes after the cessation of
the cooling treatment. This rabbit also recovered completely, without complications.

Several rabbits developed cyanosis, apnoea, and almost inaudible heart beats
at the apex of rectal temperature, i.e. between 111.0°F and 114.0°F. In one case,
the rabbit became cyanosed, stopped breathing, and was treated by ice pack, with-
out avail. In this animal some three minutes after apparent death (cessation of res-
piration with unconsciousness) the heart was still beating at 80 beats per minute.

One rabbit of this group showed a blood pressure fall from 70.0 mm.Hg. initially,
to 40.0 mm.Hg. at rectal temperature 105. 6°F then, at rectal temperature 107. 8°F re-
covered to 70.0 mm.Hg. and at rectal temperature 109. 8°F was still 60.0 mm.Hg. Then
the blood pressure fell suddenly to 36.0 mm.Hg., made a jerky recovery to 65.0 mm.Hg.
at the rectal temperature peak (112. 0°F) and fell to 10.0 mm.Hg. in less than five
minutes. Ice pack treatment was unavailing, death occurred in a few minutes. With
sudden death at the peak of rectal temperature all the body systems, respiratory,
circulatory and nervous, seemed to be simultaneously depressed, probably from the
effects of heat on the nerve cells of the brain, including the respiratory and other
centres, and on the regulatory centre in the hypothalamus. This problem of brain
lesions will be approached later.

For the problem of delayed death some details of rabbit No. 22 will be con-
sidered. This female albino rabbit was restless and struggled throughout the ex-
posure, entirely in the shade. Initially haemoglobin was 95.0% at rectal temperature
104. 2°F when the blood pressure was 75.0 mm.Hg. The haemoglobin rose to 114% at

190

the rectal temperature peak (112.0°F) but the blood pressure fluctuated jerkily from
66.0 mm. Hg. to 90.0 mm. Hg. then from 76.0 mm. Hg. to 90.0 mm. Hg, at body tempera-
ture 111.2°F followed by a rapid fall to 34.0 mm. Hg. and with recovery to 66.0 mm.
Hg. at the conclusion of the cooling treatment. This rabbit responded to a rising
body temperature by increase of respiratory activities, ventilation being adequate
for the carbon dioxide levels in expirations until the body temperature 111.8°F was
attained. At this point there was respiratory failure, characterized by unresponsive-
ness of the respiratory centre at the body temperature peak and progressive deterior-
ation during the cooling treatment, there being no sign of stimulatory response for
the respiration during this phase. The blood pressure, however, responded favour-
ably, at the conclusion of cooling treatment.

It may be significant that the circulation appeared to be favourably influenced
by cooling treatment to which the respiration was unresponsive. There was no ap-
parent addition of fluid to the intra- vascular system during the phase of heating up;
and there appeared to be progressive loss of fluid from the circulation after the body
temperature 107 .8°F continuing steadily up to the peak of body temperature; haemo-
globin rise 19%. At the conclusion of the acute experiment, the rabbit was rather
dazed, but otherwise seemed in good condition. Examined at 9.30p.m. that day she
was found lying on her side, unable to stand or walk, breathing rapidly, heart rate 240
beats per minute, sounds quiet but audible. The rectal temperature was 88.6-°F (sub-
normal) the room temperature at this time was 96.8°F. The blood pressure was too
low to record. While we were attempting to obtain blood for a haemoglobin estima-
tion, the rabbit gave a number of violent inco- ordinate movements and died at
9.40p.m. At autopsy, performed immediately, the limbs and abdominal muscles were
very stiff, heart contracted and hard, petechial haemorrhages in the walls of the
small intestine, bladder distended with brownish fluid, ears very white, suprarenals
very pale in colour, abdominal viscera deeply engorged, fluid in the peritoneal
cavity under pressure, limb muscle white. The death of this rabbit appeared to be
due to some form of peripheral circulatory failure, emphasis being probably on
capillary damage rather than failure of the heart or vasomotor centres, except per-
haps as a secondary effect. The associated poikilothermia, however, may have been
due to damage to the controlling centres, since the respiratory centre became unres-
ponsive after the exposure, and showed no sign of improvement during cooling; even
though the blood supply to the respiratory centre — judging from the recorded blood
pressure — should have been adequate, in the absence of spasm of the cerebral
arteries and arterioles. (This suggestion is at variance with physiological dogma,
but we have several instances where the only explanation for central failure seemed
to be spasm of the cerebral arteries or arterioles.)

Our time is getting short, and I have outlined some of the problems involved in
the effective treatment of cases of heat stroke. Rapid cooling of the whole body
was the best treatment for the hyperthermic crisis; the delayed collapse in cases
successfully cooled was combated in a number of ways; very useful was an extract
of the cortex of the suprarenal gland, injected subcutaneously. The associated
damage to the central nervous system was investigated; we found focal ischaemic
areas scattered throughout the cerebral hemispheres, the cerebellum, and in some

191

cases the hypothalamus in autopsy material. Over thirty normal rabbits, submitted to
multiple episodes of sub- lethal hyperthermia and then autopsied and serial sections
cut of every brain showed no such lesions. Focal areas in the brains of men killed
by hyperthermia have been found by American observers; but the American material
was always haemorrhagic, in contrast to our findings which indicated ischaemia.
The Americans gave narcotic to their human cases before exposure to great heat.
Clinical signs of cerebellar damage have been reported by British observers in human
cases as a sequela of hyperthermia (heat stroke) and some American observers have
demonstrated destruction of the nerve cells in the hyperthermic nuclei in such cases.
There is a tendency for those patients not killed at once by their brain injuries, to
recover with suitable cooling therapy and supportive treatment for the circulatory
depression.

Nervous sequelae may occur in these recovered persons, and sometimes the
nervous changes clear up with a passage of time in a temperate climate: not all do,
however. There is not time to describe any further experiments; even if the patience
and enthusiasm of this exemplary audience could bear any more of this rather tech-
nical and unexciting chronicle. We have to thank Sir David Brunt for rescuing the
subject of heat effects from oblivion, and for rescuing it from obscurity by clothing
its scientific nakedness in mathematical expressions, which, though hardly the glass
of fashion or the mould of form, none the less are graceful and elegant interpreta-
tions. I have also to thank Dr J, L. Cloudsley- Thompson for the opportunity to give
expression to my views.

192

HUMAN ADAPTIBILITY TO HOT CONDITIONS OF DESERTS

Dr. J. S. Weinor
(Oxford)

The ten or so major hot desert regions — Sahara, Kalahari, Thar, Persian, Turan,
Gobi, Arabian, Mohave, Atacama, Argentinian — are of interest physiologically be-
cause of the combination of the following characteristics; the high air temperatures,
the high intensity of solar radiation, the occurrence of fierce hot winds, the low
humidity and the diurnal and seasonal fluctuations. Daytime dry bulb temperatures
in the shade may greatly exceed normal body temperature, maxima as high as 135°F
have been noted; mean daily maxima of 100°F and over occurring on 50 days or more
in the hot season, or of 95°F (i.e. very close to body temperature) on about 100 days,
are on record for the Sahara, Arabian, Colorado, Australian and Thar deserts. Such
temperatures imply not only the cessation of an appreciable convective heat loss
from the body but a large addition of heat to the body from the ambient surroundings ;
when winds prevail the convective heat load will be still further increased roughly
in proportion to the square root of the impinging air speed. Some gain of heat to
the body by conduction also occurs in the desert since surface temperatures of
150°F even 170°F may occur. Over and above these sources of desert heat there is
the high intensity of the sun's radiation, unhindered by atmospheric moisture and
added to by re- radiation from hot surfaces. The magnitude of the heat flow to the
body and its capacity for dealing with it are considered below. The physiological
severity of the desert depends on the intensity and duration of the daily hot spells
in the summer season and it is for such short exposures that most information is
available from physiological analysis. A certain amount of data has been obtained
at first hand by investigations in deserts. The most notable work is that of Adolph
(1947) and Dill (1938) in America and the more clinical work of Ladell (1944) and
Home & Mole (1950) in the Persian and Pakistan regions. The great bulk of our
data has been obtained in studies in artificially heated rooms. Nearly all these
studies give an insight into the effects of relatively short exposures but in few of
these have high radiant temperatures figured very much. Far less is known about
the effects of the extreme swings in temperature experienced in many desert areas
or of the cumulative effects over a season or a period of years. Some insight into
the nature of long-term effects is provided by a consideration of the physical charac-
teristics of the people indigenous to the desert.

1. Ethnology

Probably less than one per cent of the world's population endure desert climates.
Yet even this number and the variety of the peoples it represents furnishes obvious
evidence of the capacity of man as a species to. withstand the thermal rigours of such
environments. The principal hot arid regions provide an interesting ethnological
picture which can only be presented in outline here. The central Asiatic desert is
part of the territory of a mixed stock whose affinities become progressively more
Mongoloid as one passes across the Gobi, and increasingly Turki- Alpine in the
plains west of the Pamirs. In the Thar desert there is a complex of groups difficult

193

to classify — the Sodas and Khoras and the nomadic Udejas and some members of the
Bhils — representing on the whole elements akin to European varieties. The Middle
Eastern — North African deserts contain folk of Caucasian, i.e. European affinities
belonging principally to the Mediterranean variety; such are the Ruwala Bedouin and
the Tuareg. There is also in this N. African desert region a fringe of Negro peoples.
The Kalahari contains the Southern Bushmen, who, on blood group and other data,
should be considered of Negroid affinity. In the desert strip called the Namib, Negros
of the Ovambo tribe are to be found. In or near the South Western deserts of N.
America there are American Indians (like the Hopi) as also in the Atacama desert.
The Australian desert may be regarded as uninhabited in its main central area yet
there are many tribes such as the Arunta who endufe the rigours of a desert climate.
A handful of Europeans in Australia also experience quite severe desert conditions.

This sketchy description shows sufficiently that, where the desert heat load has
to be endured, the human species in all its varieties, Negro, Mongoloid, European,
Australian, has the physiological capacity to deal with it, even allowing that many of
the peoples mentioned have probably not come into desert conditions by choice and
that the populations are often nomadic and in many cases only semi -permanently in
these regions. The ethnological data gives strong a priori grounds for supposing that
the human physiological make-up does not itself constitute a primary or major bar to
the greater penetration and development of desert regions. Nevertheless, there is a
considerable adjustment needed, as will be shown, for existence in these conditions,
even for short periods, and a knowledge of these can contribute much to successful
and more extensive human activity in these parts of the world.

2. The Discomfort of Desert Conditions

Stimulated by the requirements of ventilating engineers, considerable investiga-
tion has been made in Europe and America on the relation between ambient condi-
tions and subjective impressions of warmth so that for individuals living in these
countries, these can be stated with some precision (Bedford, 1948). This is usually
and conveniently done in terms of the American Effective Temperature scale which
enables one to specify the subjective effect of any combination of wet bulb, dry bulb
and air movement (and radiation) as a single temperature. When the effective tem-
perature exceeds 70°F, many people in temperate climates move out of the 'comfort
zone' and above 75°F the majority will complain of the discomfort. Such figures refer
to people lightly clothed, sedentary, in the summer; in winter, tested in the same way,
these levels will be found to be too high, that is, there is an increased tolerance to
heat in the summer showing the existence of an acclimatization process. It is to be
expected therefore that individuals who have lived for long periods in hot climates of
the world would show a similar increase in subjective tolerance. Native-born white
Australians, according to the recent investigations of Drysdale (1951), can tolerate
without undue discomfort an upper limit of warmth as high as 80°F effective tem-
perature. At such temperatures the skin may be quite moist but there are few com-
plaints on this score. Results in other hot places (Iran, Singapore, India) generally
confirm these high limits though in some cases differences in clothing may have
been operative in the tests.

194

The severe shade conditions of deserts may now be viewed in the light of these
values. In the table, the desert effective temperatures are given for two levels of air
speed, fairly still air (30ft/min.) and air at 300 ft/min. (4m.p.h.), at two levels of
humidity, namely 20 and 30%, for individuals wearing light clothing.

Air speed :-

30 ft/min.

300 ft/min.

Dry Bulb : °F

95

105

115

95

105

115

R.H. 20% E.T. °F =

77

84

88

74

82

87

R.H. 30% E.T. °F =

79

85

92

76

79

90

It will be seen that up to dry bulb temperatures as high as 105°F desert condi-
tions yield effective temperatures not far from those which appear to be just tol-
erable for individuals acclimatized to hot climates. These figures, admittedly only
approximate, do enable one to evaluate more objectively the limits of tolerance to be
expected in desert conditions and in fact, despite the greatly increased heat flux the
body must cope with (see below) these limits are probably higher than usually thought
to be the case. No studies of indigenous people, however, appear to have been made
from this as from other points of view. The nature of acclimatization is also by no
means understood though, as we shall see, there are certain physiological changes in
the body which proceed in parallel with increased tolerance.

3. Capacity for Work

As indicated in the table, the desert on many days in the summer affords con-
ditions, even in the shade, more trying than the upper limit of the 'comfort zone* of
80 — 82°F effective temperature of acclimatized individuals. The heat load on the
inactive indoor individual is derived by convection and radiation from the surroun-
dings. Out of doors the direct and indirect solar heat load will be added to these.
Nevertheless it can be shown that there yet remains a fair margin to the body's capa-
city to maintain homeothermy even after dealing with the heat gain from the exterior.
This is the margin available for coping with the heat production of muscular work.
To understand how great this margin is likelyto be it is only necessary to consider
the maximum rate of cooling which the body is physiologically capable of developing.
In the desert conditions under consideration this is entirely dependent on evaporation
of water from the skin surface (that from the lungs adding only about 5 — 10%). It is
possible to predict the maximum cooling capacity of the body for any particular set of
desert conditions where the total surface area of the skin surface is regarded as ef-
fectively wetted and where a skin temperature of not higher than say, 97 or 98°F is
assumed. In fact, a physical body shaped like the human body, kept wet at this sur-
face temperature, could lose heat at the rate of about 500 or 600 Kcals/hr up to air
speeds of, say, 5m.p.h. In very severe desert temperatures (E. T. of 90°, air move-
ment 300 ft) the convective heat load impinging on the body might be of the order of
100 Kcals/hr, the radiation heat load on a man in the standing position might be about
150 Kcals/hr, so that 250 Kcals/hr remain for metabolism and work but this margin
would drop off rapidly as the air movement fell.

195

This simplified account indicates that the maximum ability of the human body on
physical grounds to maintain heat balance demands an output of about 1 litre/hr sweat
for a cooling equivalent to 500 Kcals/hr. (It is as if the provision of sweat glands
had been evolved to cope with heat loads as high as that required for life and acti-
vity in hot dry regions rather than for the lower heat loads of hot humid tropics). In
actual tests on Europeans this sweat rate (1 litre/hr) is indeed about the limit of
what the sweat glands can manage to maintain for 4—6 hours. It is not surprising
that physiological acclimatization, as we shall see, improves the performance of the
sweat glands and that efficient performance as well as breakdown in the desert for
the most part is a matter of water supply and water metabolism as abundantly illus-
trated by the work of Adolph and his colleagues.

4. Some Physiological Adjustments

The changes which we know occur in the heat regulatory system, the circulation,
the kidney and the endocrine system cannot be dealt with in any detail here. But it
is important to realise that short term exposure to heat induces circulatory effects
primarily to facilitate a greatly increased loss of heat from the surface, and as this
depends so much on sweating, there are consequent adjustments in water (and salt
balance) throughout the body. It is the regulation of these that calls for endocrine
activity by the posterior pituitary and the adrenal cortex — to mention only those
glands for which we have evidence.

Salt Intake Of these processes it is worth mentioning in a little more detail the
great capacity possessed by the human body for adjusting its salt loss to the supply
(Weiner & van Heyningen, 1952). It was first noted by Dill and his colleagues that
sweat of people living in the desert became progressively reduced in its salt content.
This is now known to happen only when salt intake is initially rather low. The kid-
ney in such circumstances immediately cuts down its salt concentration and output
and this is followed in a few days by a similar reduction in salt composition of
sweat. The normal sweat gland is in fact able to do considerable osmotic work in
producing a hypotonic fluid though this falls as high rates are approached. Many in-
dividuals can thus, after a period of adjustment, subsist in hot conditions on a
moderately low salt intake. There is however evidence of great individual variation
in this respect and the process may be attended by undesirable symptoms and a
lowered capacity for work. Salt imbalance is probably one of the commonest causes
for upset in hot conditions before acclimatization asserts itself. The evidence favours
the additional consumption of salt when sweating increases and water intake is cor-
respondingly high.

Sweating and Acclimatization Acclimatization proper to hot desert conditions
shows itself in an increased capacity to perform muscular work and a concomitant
improvement in bodily heat regulation, as shown by a progressive reduction in the
pyrexia induced by high heat loads. There is a concomitant increase in the sensi-
tivity io heat stimulation of the sweat glands as shown both by some increase in
sweat production for a standard heat load and by a more rapid response. The acclima-
tized man accumulates less heat and must therefore handle a greater heat loss, or to
put it another way, he purchases reduced discomfort and high efficiency by a lower

196

skin and body temperature and must therefore remove more heat by evaporation. The
earlier and greater output by the sweat glands of acclimatized individuals has been
noted in many laboratories to occur on the first 7-14 days of continuous or repeated
exposure. It may be detected even after a year of living in hot climates. In desert
conditions where sweat is so readily evaporated it certainly appears to be an adjust-
ment of real importance.

It should be repeated that we have no first hand data of the indigenous desert
peoples or on those who have lived long periods in desert regions.

5. Breakdown

Enough has been said to indicate that adjustment to severe desert conditions is
physiologically of a complex nature and it is not surprising that this may fail at dif-
ferent points. There is also great individual variation in liability to these failures of
physiological adaptation, at least in Europeans, for information on indigenous peoples
is lacking. Brief consideration of these disorders will serve to indicate the nature of
the physiological breakdown.

a. Circulatory failure with syncope occurs in most heat disorders but in a simple
form it is often the outcome of insufficient acclimatization and training for hard
work at high levels. In deserts it is often the sign of incipient dehydration or of
salt imbalance.

b. Dehydration resulting from lack of water in relation to heat load is by far the
most serious danger in deserts. The body's ability to deal with water shortage
is more limited than its capacity to compensate for salt shortage. The number of
days of survival in the desert can be readily calculated when water supply fails.
This has been dealt with in a thoroughgoing manner by Adolph and his colleagues
who have mapped these survival limits for all desert regions and for different
conditions of work allowing for movement either at night or day. There is no
evidence at all that individuals can be 'acclimatized' or 'hardened' to a low
water level. Enforced reduction in water intake does not reduce the deficit of
body water by sweating. Performance in the heat is better if water is taken con-
tinually. Fortitude when water is short is to a large extent a matter of morale.
The work of Adolph should be consulted for a vivid account of the key importance
of water supply not only to survival but to efficiency in day to day work in the
desert. Adolph describes many practical ways of reducing water requirements
such as the best use of shade and clothing.

c. Lack of salt may produce relatively mild effects such as undue fatiguability or
severe cramp of the abdomen and limb muscles. As already indicated, there is
good physiological compensation for low salt intake but it is likely that some
individuals are far less efficient in their ability to conserve salt by reducing its
concentration in urine and sweat. The desirability of supplementing salt has al-
ready been commented on.

d. Sweat gland fatigue can be demonstrated in laboratory studies but its exact re-
lationship to the apparently complete cessation of sweating seen in the failure
of heat regulation with hyperpyrexia, known as 'heat-stroke', is not clear. In

197

the condition known as 'anhidrotic heat exhaustion' there is evidence of actual
damage to the sweat glands as a result in many cases of pre- existing prickly
heat. It has also been claimed that injury to sweat glands is likely in the early
stages of sunburn. Such damage brings about a greatly reduced capacity to carry
out active work in the heat for the lack of sweating makes exertion very unplea-
sant and inefficient and circulatory failure is easily induced. This condition and
certainly prickly heat are both far more common in hot humid climates than in
desert conditions and could only occur in closed spaces with extremely poor
ventilation,

6. Desert and 'racial' selection

It has earlier been pointed out that representative groups of all the varieties of
mankind (whether these are classified serologically or morphologically) are to be
found in deserts, so pointing to the adaptiveness of mankind as a whole. Indeed,
laboratory tests have shown that short term acclimatization phenomena are similar in
Europeans, Asians and Africans. Nevertheless, this does by no means preclude the
possibility that these desert sub-groups have undergone distinctive changes as a
result of continuous residence in the desert. Evidence has accumulated that in re-
gions of high mean annual temperature some differences in bodily development and in
physique are encountered. D. F. Roberts in this laboratory has shown that peoples of
hot climates are of lower body weight and often in addition exhibit an elongation,
relative to trunk, of either upper or lower limbs or both, as compared with people in
temperate and cold climates. Probably 50% of the variance in these characteristics
is attributable to differences in mean annual temperature. These modifications in
physique will be recognised as being in line with Bergmann's and Allen's Rules and
it could be argued on physical grounds that they represent advantageous adaptations
to hot climates. However, the existing records relate overwhelmingly to peoples in
the hot humid regions and it is by no means clear whether desert peoples universally
exhibit these physical characters. Nor is it certain that such changes are neces-
sarily genotypic. Another effect of hot humid climates (which may be merely pheno-
typic) is a slowing down of the rate of skeletal maturation and sexual maturity. Again,
data for desert peoples is lacking.

The most striking of all 'racial' characters associated with desert people is of
course the steatopygia of the Kalahari bushmen. While there has been some specu-
lative discussion as to the role of this fat deposit as a fuel and water store, no
direct study has yet been made. Some of the Andamanese pygmy negritoes show
similar female steatopygia proving that the condition is to be found in humid tropics
as well as in deserts.

Yet another problem is the significance to be allotted to melanin pigmentation of
the skin and here again, owing to lack of adequate technique, no objective compari-
sons of skin colour amongst desert people can be quoted though one can be certain
that a great range of skin colour must exist. The general opinion that melanin depo-
sition is protective against ultra-violet light and of value in hot climates is streng-
thened by some recent experiments of Thomson (1951) in which he showed that in
Europeans sunburn can be associated with damage to the sweat glands. Another

198

pointer in the same direction is the greater incidence of rodent ulcer in white Austra-
lians which is presumably attributable to some special sensitivity to solar radiation.

7. Conclusion

This outline should serve to show that life in deserts is well within man's phy-
siological capacity as of his technology and organisational ability. Disorders occur-
ring purely as a result of the climate are to a large extent preventable and a matter
of fairly simple hygiene. The fact is that rules of living have to be acquired for
desert life as for other parts of the world and that this can be done efficiently is
obvious from the successful survival of the great variety of desert peoples. The rea-
lisation that this adaptability, as well as its limits, is being given a progressively
more exact physiological analysis should serve as an encouragement to the more in-
tensive interest in the development and amelioration of deserts.

References.

Adolph, E. F. 1947 Physiology of Man in the Desert. New York. Interscience Publishers, Inc.

Bedford, T. 1948 Basic Principles of Ventilation and Heating. London: H.K.Lewis & Co.
Ltd.

Dill, D. B. 1938 Life, Heat and Altitude. Cambridge: Harvard University Press.

Drysdale, J.W. 1951 Climate and Design of Buildings: Physiological Study No. 3.

Home, G.C. & Mole, R.H. 1950 Lancet, August 13, p. 279.

Ladell, W. S.S., Waterlow, J. C. & Hudson, M. F. 1944 Lancet, October 14 and 21, p. 1- 12.

Thomson, M.L. 1951 J.Physiol. 112, 31-42.

Weiner, J.S. & van Heyningen, R. E. 1952 Brit. J. industr. Med.. 9, 56-64.

199

LE PEUPLEMENT HUMAIN DU SAHARA

Dr Edmond Sergent
(Algiers)

Les prehistoriens nous enseignent qu'a I'age de la pierre un ciel pluvieux ar-
rosait le Sahara, de I'Atlantique au Tibesti, de 1* Atlas et des Syrtes au Niger et au
Tchad. Une vegetation abondante y nourrissait une faune tropicale, des elephants
et des hippopotames. Une population, que Ton a des raisons de croire de race
noire, I'habitait, Puis, vers la fin de la periode neolithique, les nuages reculerent
vers le Nord, le dessechement progressif du sol tua toute vie, crea le desert. Alors
les peuplades noires ont abandonne une terre devenue aride et ont emigre vers le
Sud, laissant d'innombrables temoignages de leur existence, armes et outils de
pierre taillee, gravures et peintures repestres, que Ton a actuellement la surprise
de decouvrir epars dans des regions nues et desolees, sous un ciel de fournaise.
La regression des Noirs au Sud du Tropique fut suivie plus tard de I'avancee de
Berberes blancs peu nombruex, venant des rivages mediterraneens.

On estime a 3 millions le nombre d'habitants eparpilles actuellement sur les
8 millions de kilometres carres que couvre le Grand- Desert, entre I'Atlantique et
le Tibesti, ce qui correspond a une densite demographique inferieure a la moitie de
I'unite, Dans les Territoires du Sud algerien, dont la superficie est d'environ
1,981. 000 Icm^, la population totale etait, en 1048, de 817,000 ames; la densite au
kilometre carre etait done de 0.4. Le Fezzan, dont on evalue I'etendue a 800,000
kilometres carres, compte 50,000 ames; la densite demographique y est done de
0.06 par kilometre carre.

Montesquieu a ecrit: 'Quand un pays est desert, c'est un prejuge de quelque
vice particulier de la nature du climat'. Le vice du climat saharien est d'etre un
climat d'extremes. Ses trois facteurs dominants sont une aridite extraordinaire, —
une grande chaleur estivale contrastant avec un froid relatif hivernal — des vents
impetueux.

On peut definir comme zone aride une region qui ne possede aucun cours d'eau
normal et qui recoit rarement de la pluie (moins de 100 millimetres par an, durant
une periode assez longue).

La seconde caracteristique meteorologique du Sahara est une temperature ex-
cessive et a variations brusques. Le thermometre marque souvent, pendant de
longues semaines, de mai a octobre, 50° et plus, jusqu'a 58°. Mais I'hiver est as-
sez froid: au centre de Sahara, a des altitudes qui ne depassent pas quelques cen-
taines de metres, on compte de 1 a 3 semaines de gelee par an, le minimum absolu
decendant a quelques degres au-dessous de zero. D' autre part, le rayonnement
nocturne, intense, cause des ecarts qui peuvent depasser 30° entre la chaleur acca-
blante du jour et la fratcheur de la nuit.

Apres I'aridite et les temperatures extremes, les vents contribuent a donner au
climat du Sahara son caractere de violence et de rudesse.

Les effets de climat saharien sur la nature sont d'une etrange brutalite.

200

L 'erosion fluviale au Sahara a ete tres considerable aux temps prehistoriques.
L'action de I'erosion eolienne, qui se poursuit de nos jours, est immense. Elle
decape le sol, le degrade.

Les ennemis des plantes au Sahara sont multiples: la rarete et I'extreme ir-
regularite des pluies, la secheresse de I'air et I'absence de rosee qui en resulte,
I'aridite du sol, les fortes chaleurs estivales et les froids hivernaux, I'insolation
intense, la frequence et la violence des vents. L'agriculteur ne peut faire que des
cultures irriguees, des jardins, qui sont les oasis, dans les points extremement
rares ou une eau souterraine affleurele sol ou se trouve a une profondeur accessible.
A I'ombre des dattiers, on cultive des arbres fruitiers, des legumes, quelques cere -
ales.

La faune saharienne comporte de nombreux genres et especes, mais les especes
y sont representees par un petit nombre d'individus, Le seul elevage qui reussisse
bien au Sahara est celui du chameau, qui se contente comme nourriture des plantes
ligneuses et epineuses des plateaux pierreux (les hamadas) et des sables (les ergs).

Le globe terrestre sera bientot surpeuple et trop petit pour le genre humain.
Les denrees alimentaires font de plus en plus defaut. Les matieres premieres com-
mencent a manquer. C'est pourquoi on pense a mettre en valeur le Sahara reste
vide, improductif, jusqu'a present. Ce sera la reconquete du Grand- Desert par
I'homme.

Une industrie pleine d'avenir au Sahara est celle des transports, car il peut
jouer le role d'une 'plaque tournante' entre le Nord, le Sud, I'Ouest et I'Est. Le
chemin de fer transsaharien, appele actuellement le Mediterranee - Niger, est com-
mence.

De plus, le Sahara procurera des bases precieuses a la navigation aerienne.

On a de grands espoirs de decouvrir des richesses minieres ou combustibles
dans son sous -sol. Quelques- unes sont connues. Les prospections se multi-
plient. Presque chaque anneede nouveaux gisements sont mis en exploitation.

Lorsque de grandes richesses minerales seront decouvertes, il faudra reunir
le nombre de travailleurs necessaires aux industries extractives. On devra alors
resoudre la question tres bien definie par Henri Prat: 'Dans toutes les zones
seches, le probleme de I'existence humain se pose ainsi: la population que Ton
peut faire vivre en un lieu donne est directement fonction de la quantite d'eau que
Ton peut fournir au sol. Tout doit done y etre subordonne au probleme de I'eau,
"facteur limitant" de toute activite humaine'.

La recherche des eaux du sous- sol dans le Sahara oriental est encore peu
avancee. II en est autrement dans le Sahara occidental, bien etudie depuis long-
temps par les savants francais. Il faut evoquer ici les etonnantes perspectives
d'avenir qu'a ouvertes le geologue J. Savornin qui, des 1927, a signale I'existence
d'enormes reserves d'eaux artesiennes exploitables surtout dans le Sud algerois.
Cette exploitation est commencee.

II est loisible, d' autre part, d'imaginer que les regions completement depour-
vues de n^pe phreatique pourront, dans I'avenir, recevoir, par des pipe- lines,
I'eau de lointains chateaux d'eau.

201

Enfin, il est permis de rever que le progres des inventions et la decouverte de
ressources energetiques nouvelles apporteront un jour la solution du probleme de
I'eau au Sahara, par la transformation du climat.

Sous reserve des etroites servitudes imposees actuellement par la penurie
d'eau, comment esperer reunir au Sahara le nombre d'hommes qu'exige sa mise en
valeur, les proteger contre un climat excessif, pourvoir a leur subsistance?

Trois eventualites peuvent etre envisagees:

(1) Implanter des colons de race blanche

(2) Implanter des colons de race noire

(3) Recruter la main- d'oeuvre necessaire parmi les habitants actuels du Sahara.

Ainsi se trouve pose le probleme de I'acclimatement des races humaines en
zone chaude et aride.

L'acclimatement n'implique pas seulement I'accoutumance de I'individu trans -
plante, mais encore la faculte, pour sa descendance, de se perpetuer, saine et
vigoureuse, dans une longue suite de generations, sans croisement avec les Indi-
genes, et en conservant tous les caracteres d'energie physique et morale de la
souche originelle. Sous 1 'expression de I'influence du climat, on a longtemps con-
fondu deux phenomenestres differents: Taction du climat proprement dit, c'est- a-
dire du froid et du chaud, de I'humidite et de la secheresse, des circonstances at-
mospheriques en un mot, et Taction des maladies regnantes. C'est par un abus de;
mot qu'on en^lobe, sous la meme expression d'acclimatement, Tadaptation aux
conditions physiques, surtout meteorologiques, d'un pays, et I'accoutumance a ses
maladies infectieuses.

Du point de vue pratique, la question de l'acclimatement se pose done en ces
termes: (a) une race peut- elle s'adapter a une plus grande chaleur on a un plus
grand froid que la chaleur ou le froid de la zone ou elle a vecu depuis des siecles?
Reponse: non, on ne se 'vaccine' pas contre le chaud ni contre le froid. Les tech-
niques modernes permettent seulement de se proteger contre les exces de la tem-
perature ambiante, par le 'conditionnement' du logement et du vetement; (b) peut-
on echapper a Taction nefaste des maladies exotiques? Reponse: oui, on peut,
et Ton pourra de mieux en mieux, se defendre contre les maladies regnantes, par
Thygiene, la prophylaxie, la therapeutique.

Les facteurs meteorologiques excessifs du climat saharien eprouvent directe-
ment la physiologic de Thomme. Les reactions au climat saharien de la race
blanche et celles de la race noire presentent des differences. II convient de les
considerer separement.

Une des fonctions les plus importantes de Torganisme consiste a maintenir sa
temperature normale. L 'action de la temperature du desert, qui va de la glace a
Textreme chaud, exige un bon fonctionnement de la regulation thermique. II y a
incompatibilite physiologique entre la surchauffe permanente a laquelle le Blanc
se trpuve soumis au Sahara et le bon fonctionnement de ses organes. Plus ou moins
et tot ou tard, Tappareil thermo- regulateur, excede, y fallit a son role. Les divers
systemes de Teconomie sont alors troubles. Heureusement, une transpiration pro-

202

fuse vient sauver I'equilibre thermique. La secretion de la sueur, acte reflexe qui
suit I'elevation de temperature, prend, au Grand - Desert, des proportions inusitees.
La ration d'eau necessaire est, par suite, fort elevee. Le chiffre minimum est de
5 litres par jour et par homme lorsque I'activite musculaire est restreinte et si Ton
n'est pas expose au soleil. En cas de travail un peu dur, il faut compter 10 ou 15
litres. D'autre part, les grandes transpirations soustraient a I'organisme du chlo-
rure de sodium, ce qui n'est pas sans inconvenients, en particulier pour la secre-
tion gastrique.

A la longue, les fortes chaleurs seches provoquent des perturbations dans les
fonctions digestives, mais c'est surtout sur le systeme nerveux du Blanc que le
climat saharien exerce une action profonde: action exaltante chez les ames de
qualite, action depressive qui peut devenir tres dangereuse sur les esprits qui man-
quent d'equilibre.

L'influence nefaste de la secheresse de I'air sur les enfants en bas- age est
bien connue. La mortalite des nourrissons blancs au Sahara est tres elevee. On ne
peut pas elever d'enfants blancs au Sahara. Les cimetieres y temoignent de I'in-
succes de quelques essais imprudents. La saison estivale surtout, qui dure de mai
a octobre, leur est fatale, ainsi qu'aux femmes blanches fatiguees. C'est pourquoi
le climat du Grand- Desert prohibe I'installation a demeure de families blanches.

11 ne faut pas se laisser tromper par le fait que des Berberes blancs, les Tou-
areg, sont fixes au Sahara depuis des siecles. En realite, si les Tquareg sont ar-
rives a survivre, au Desert, c'est parce que leur fatigue physique est reduite: ce
sont des pasteurs de troupeaux, des guerriers. lis ne se plient pas aux durs tra-
vaux de 1 'agriculture.

En conclusion, les families blanches europeennes ou nord- africaines sont in-
^tes a I'acclimatement au Sahara: les hommes blancs peuvent venir y travailler,
mais seulement dans les cadres. lis ne doivent pas y etre employes a des travaux
de force. lis ne doivent y etre que des agents d'autorite, de commandement, d'en-
cadrement, toutes personnes adonnees a un travail principalement intellectuel, et
qui ne comporte pas grande fatigue physique.

Enfin, les Blancs que I'on veut transplanter temporairement dans le Grand -
Desert doivent ^tre I'objet d'une selection attentive, portant sur les qualites phy-
siques et morales, etre installes dans des conditions de confort speciales, pour le
logement (qui doit etre 'climatise', au moins en ce qui concerne les cadres), le
vetement, la coiffure, et suivre des regies strictes d'hygiene et de prophylaxie.
Les heures de travail doivent etre bien calculees, et de longs conges en Europe ou
en Afrique du Nord prevus.

Le Noir resiste mieux que le Blanc a la chaleur et aux rayons solaires, a cause
de la pigmentation de sa peau et de sa retine et du developpement de ses glandes
sudorpares. Mais le fait majeur est sa faible resistance aux basses temperatures
hivernales, et aleursecarts brusques. II est plus sensible que le Blanc aux mala-
dies a frigore. Cette sensibilite au froid rend le Noir inapte a fonder des lignees
durables au Grand- Desert. La prehistoire nous donne une preuve de cet empeche-
ment: les nombreux Neolithiques noirs qui peuplaient le Sahara au tempsou il etait

203

pluvieux et chaud ont recule vers le Sud quand son climat est devenu aride, et frais
en hiver. Le Noir n'est pas fait pour le Sahara, Si on veut I'y employer, ce ne peut
etre que comme travailleur saisonnier, temporaire, sans sa famille. On devra le
vacciner contre la tuberculose par le vaccin B.C.G., et assurer sa surveillance
medicale reguliere.

Peut -on repeupler le Sahara en facilitant la multiplication et le developpement
des populations actuelles du Grand- Desert?

Deux sortes d'hommes vivent au Sahara: les hommes du palmier, c'est-a-dire
les cultivateurs, habitants sedentaires des oasis, qui sont, presque tous, des Ne-
groides, appeles Haratin — et les hommes du chameau, c'est- a- dire les pasteurs
nomades dans les vastes espaces, qui sont de race blanche.

E, F. Gautier a dresse un tableau impressionnant de la misere trop frequente
des NegroVdes oasiens: 'Chez eux, ce qui frappe le plus I'oeil, d'abord, c'est I'ab-
jection physique; ... la fievre et la faim ont sculpte d'effroyables anatomies; ...
Ces humbles echines de serfs font une impression de vie ralentie'.

Malgre leur misere, ces Negro'ides, adaptes a la vie sedentaire des oasis, y
elevant leurs families depuis des siecles, s'y livrant au dur travail de la terre, sont
lesseuls habitants du Grand-Desert qui peuvent fournir de la main- d'oeuvre pour sa
mise en valeur. Mais leur nombre est insignifiant au regard des immensites qui en-
tourent les archipels d'oasis. Leur accroissement numerique et leur developpement
physique dependront de la quantite d'eau et du bien-etre qu'on leur pro curera. II
sera necessaire avant tout d'elever leur niveau de vie, ce dont ces pauvres etres
sont incapables eux-memes.

A la difference des JSJegroides, sedentaires, les Blancs du Sahara, les 'hommes
du chameau', berberes (Touareg), ou arabes (Cha'amba, Maures), sont en errance
perpetuelle, de paturage en paturage dans I'immensite nue, avec leurs chameaux de
selle ou de bat, quelques chevres et quelques moutons a polls. lis font soigner
par des Haratin les dattiers qu'ils possedent dans les oasis. La paix francaise a
ruine la principale Industrie de ces nomades, qui consistait a prelever un tribut sur
les Oasiens et sur les caravanes. lis ne peuvent servir que comme 'gendarmes de
desert' ou entrepreneurs de transport. On ne peut pas compter sur eux pour fournir
des travailleurs du sol pas plus que du sous- sol.

11 n'y a pas de pathologic humaine speciale au Sahara.

La grande pandemie des pays chauds et humides, le paludisme, n'existe, au
desert chaud et sec, que dans les oasis. Les techniques antipaludiques issues des
decouvertes de A. Laveran et de R. Ross en ont facilement raison.

Le trachome est une grande plaie des oasis. II y atteint, encore aujourd'hui, la
grande majorite des nourrissons avant la fin de leur premiere annee. Le meilleur
moyen de lutte consiste dans une organisation de soins, locale et permanente.

Les Noirs sont tres sensibles a la tuberculose, dont les formes pulmonaires
evoluent tres vite chez eux. La seule methode de vaccination antituberculeuse par
le B.C.G. des populations dispersees du Sahara est la methode de Foley et Parrot,
qui consiste a vacciner par scarification cutanee, sans epreuve tuberculinique prea-

204

lable, tous les enfants au-dessous de 15 ans en bon etat apparent de sante — et
dans la repetition des memes seances tous les 3 ans, dans le meme lieu.

Les Noirs presentent une sensibilite tres grande aux pneumococcies, a la me-
ningite cerebro- spinale.

La bilharziose, qui est endemique dans le Sud marocain et dans le Sud tunisien,
comme en Egypte, a tendance a envahir d'autres oasis du Sahara occidental.

Les piqures de scorpions constituaient un redoutable danger dans beaucoup
d'oasis, ou les cas de mort n'etaient point tares. Le 'peril scorpionique' peut etre
ecarte depuis que I'lnstitut Pasteur d'Algerie prepare un serum antiscorpionique ef-
ficace.

En resume, le Sahara, pluvieux et habite a I'epoque paleolithique, s'est de-
peuple lorsqu'au Neolithique son climat est devenu aride et excessif. Actuelle-
ment, le Sahara, qui couvre 8 millions de kilometres carres, ne compte que 3 mil-
lions d'habitants: moins d'une demi- unite par kilometre carre. Comment le repeu-
pler?

(1) Par I' immigration de Blancs?

L'acclimatement de families blanches est impossible au Grand- Desert, parce
que les femmes et surtout les enfants ne peuvent pas supporter la chaleur, la se-
cheresse et les vents torides de I'ete, durant 5 mois.

Les hommes blancs, sans leur famille, peuvent vivre temporairement au Sahara,
a condition d'etre selectionnes, d'avoir du confort et d'observer les regies de I'hy-
giene. Les Blancs peuvent ^tre employes dans les cadres (Europeens et Nord-
Africains), et, jusqu'a un certain point, pour des periodes limitees, comme manoeu-
vres saisonniers (Nord- Africains celibataires).

(2) Par r immigration de Noirs?

L'acclimatement de families noires est impossible au Sahara, a cause des re-
froidissements hivernaux et des brusques ecarts de la temperature.

Les hommes noirs, sans leur famille, peuvent fournir des travailleurs saison-
niers (actuellement des manoeuvres ou des ouvriers), a condition d'etre sous sur-
veillance medicale.

(3) Par la multiplication des Indigenes actuels?

Les Oasiens Negro'ides, qui sont deja adaptes au climat, peuvent faire souche
au Sahara, mais leur existence est miserable. Leur multiplication sera proportion -
nelle a la quantite d'eau qui sera mise a leur disposition et conditionnee par le re-
levement de leur niveau de vie. lis peuvent fournir une main- d'oeuvre sedentaire.

Les nomades, chameliers Blancs, arabo- berberes, pasteurs errants et anciens
pillards, n'ont ni le desir ni la possibilite de se livrer a un travail manuel; tels
qu'ils sont, ils sont incapables de prosperer dans un pays pacific et police.

En conclusion, sous le climat actuel, le peuplement proprement dit du Grand-
Desert ne pourra ^tre realise que par les families des Oasiens Negroides, dans la
mesure ou I'eau vitale et une alimentation suffisante leur seront assurees. Pour la

205

mise en valeur et ['exploitation des richesses du sous- sol et des voies de commu-
nication, les Blancs de race pure et les Noirs de race pure ne peuvent ^tre au Sa-
hara que des travailleurs passagers, fournissant des cadres et de la main- d'oeuvre
temporaires, sans implantation de leurs families.

Pour diminuer les efforts physiques, epuisants au Desert, de la main- d'oeuvre,
dans les industries extractives, il faudra recourir le plus possible a la mecanisa-
tion, et, pour cela, disposer de ressources energetiques.

206

PHYSIOLOGICAL EFFECTS OF COLD ENVIRONMENTS ON MAN

Dr O. G. Edholm
(London)

The problem of mammalian, including human life in cold environments is essen-
tially a problem of insulation. Warm blooded animals, apart from periods of hiber-
nation, maintain a relatively constant deep body temperature of approximately 37°C
with an outer shell of tissue, the temperature of which depends on the thermal en-
vironment and the state of activity of the animal. The constancy of the internal
temperature depends on a balance of heat output and heat loss. Scholander, Irving
and their colleagues, who have recently published an important series of studies on
arctic animals, point out that there are three ways by which such animals might
develop mechanisms for survival in extreme cold. There might be a fluctuating deep
body temperature, varying according to external temperature. Their evidence strongly
discounts such a possibility: arctic animals, such as the fox or the dog, maintain a
constant deep body temperature of the same order as temperatures found in animals
inhabiting temperate or tropical 2»nes. There could be an increased heat production,
i.e. a high metabolic rate. Scholander and Irving do not consider that that is an im-
portant factor as the basal or resting metabolic rates in a large variety of arctic ani-
mals showed the same relationship to surface area as demonstrated by animals living
in tropical or temperate 2ones. The points fall close to the 'mouse- elephant curve'
constructed by Benedict many years ago .

The third mechanism consists of variatidn of the insulation of the deep body
temperature. This in turn depends on the thickness of the subcutaneous layer of
fat, the rate of blood flow in the skin and superficial tissues, the rate of production
of water on the surface of the body, and the thickness of the fur. In their experience
Scholander and Irving found that maintenance of a constant deep body temperature in
arctic animals depended essentially on adequate insulation, and this in turn was
largely due to the thickness of the layer of fur.

How does man adjust physiologically to life in cold environments? Is there any
evidence of relatively long term effects which suggest acclimatization to cold? It
should be made clear, at the outset, that the evidence so far is meagre, and there is
certainly no such dramatic effects as are observed when man is exposed to hot en-
vironments, as described by Dr Weiner.

The critical temperature for a nude man at rest is relatively high, about 27°C.
That means that body temperature is maintained without any change in metabolism,
down to temperatures of 27°C. Thereafter any further fall of environmental tempera-
ture will stimulate a'n increased metabolism. This may be compared with a critical
temperature of — 40°C for the arctic fox.

Metabolic rate, i.e. heat production or oxygen consumption, starts to increase
when the environmental temperature falls below the critical temperature. The in-
creased heat production is largely or possibly entirely due to shivering or other mus-
cular activity. The rise in metabolic rate may be very considerable and for short
periods can be as high as 6-7 times the resting or basal metabolic rate, i.e. up to

207

300cal./m^/hr. These high rates cannot be maintained for long periods and the ave-
average increment over a period of one hour is unlikely to exceed 150- 200 cal./m^/hr.

If the cold environment is maintained, shivering gradually diminishes as exhaus-
tion proceeds and the deep body temperature will begin to drop. When body cooling
is accelerated by immersion in cold water, it is found that shivering ceases at a
rectal temperature of approximately 32- 33°C, and metabolic rate declines thereafter
with rectal temperature. Conciousness is lost at a rectal temperature of 30°C and
death usually occurs at rectal temperature of 25°C, although survival has been re-
ported in one subject whose rectal temperature was below 20°C.

There are many other physiological changes which occur during acute exposure
to cold which can only be briefly summarised. The effect is to diminish heat loss
by increasing insulation, which is mainly effected by vasoconstriction in the skin
and underlying muscles. The reduction in blood flow leads to a fall in skin tempera-
ture, and the gradient of temperature from the deep tissues to the surface becomes
steeper. Body hair is vestigial inman^but the pilo-arrectores muscles attached to
the roots of the hairs contract and so produce goose flesh in the skin. This roughen-
ing of the skin surface increases the boundary layer of air in contact with the body
and so has a small effect on insulation. Water loss from the skin surface is greatly
reduced.

Acute exposure to cold also stimulates certain endocrine changes, similar to
those described by the term 'alarm reaction*. The main characteristic of this reac-
tion is the increased activity of the adreno- cortical mechanism.

The effects of long continued exposure to cold environments include vascular
and endocrine changes. As a result of peripheral vasoconstriction, there is a shift
of blood from the superficial regions of the body to the pulmonary and probably the
splanchnic areas. In addition, there is a gradual diminution of the blood volume,
owing to the loss of plasma with consequent haemoconcentration. The proportion of
red cells to plasma increases from a normal of 46% up to 52- 55%. During the period
of haemoconcentration there is a marked increase of urine secretion, and this diure-
sis represents a period of increased water loss.

The endocrine changes in man cannot be adequately described at present. In
laboratory animals who are exposed for long periods to temperatures of 0°C, there is
an increased activity of the thyroid gland, which is accompanied by a gradual rise
in the basal metabolic rate.

In laboratory animals there is also a hypertrophy of the cortex of the adrenal
gland, which can be diminished by increasing the amounts of ascorbic acid in the
diet.

The main physiological problem as far as man is concerned is whether acclima-
tization to cold takes place, in the sense that physiological changes occur which in-
crease tolerance for cold or improve survival in the cold. Such changes can be
clearly demonstrated in laboratory animals. Blair, for example, has recently com-
pared the responses of control rabbits with animals kept in the cold chamber for
many weeks. When both were exposed to very severe cold, the control animals all

208

had severe falls of body temperature whereas the cold adapted ones maintained a
steady body temperature. After removal all the controls developed frostbite, but
there was none in the treated animals.

Much work has been done in recent years to investigate the possible develop-
ment of acclimatization in man, but the positive findings are few. There have been
a number of studies on the Eskimos, who are the best example of people adapted to
life in severe cold.

The most comprehensive enquiry has been that of the Queen's University, King-
ston Ontario, under the direction of Dr Malcolm Brown. A group of workers have
spent several summers in Southampton Island, which is north of Hudson's Bay, lati-
tude 65°N, investigating physiological, nutritional, medical and social aspects, and
the work is still in progress.

The basal metabolic rate of the Eskimo is raised, averaging 30% above normal
values for the temperate zone (Hatcher, 1950). Similar raised B.M.R.'s in the Eskimo
had been recorded earlier by several workers, but the figures have not always been
accepted, as the conditions of measurement were subject to criticism. The work at
Southampton Island appears to be free of criticism as a number of repeat determina-
tions were made over a period of several weeks and the measurements were made in
the Eskimos' huts or tents after a period of at least 8 hours asleep and before arising
from bed. So it appears probable that there is a true increase in the B.M.R. in the
Eskimo.

There is no evidence as yet that other people who live in the north develop an
increased B.M.R. but not many comparable studies have been carried out. Such work
has been attempted on Antarctic expeditions, without any clear results indicating a
rise, but this may have been due to the difficulties of measurement. On the present
British North Greenland Expedition a physiologist will be carrying out regular deter-
minations, and as the members of the expedition will remain in arctic regions for at
least a year, it is possible that satisfactory evidence for or against an increased
metabolic rate in the cold will be obtained.

Peripheral blood flows were measured in the forearm at various temperatures in
the Southampton Island Eskimo and the values were closely similar to those obtained
in similar experiments in this country. There were, however, two possible exceptions
Water temperatures ranging from 10°C to 45°C were used in this country, 45*^ being
the highest temperature which can be tolerated for periods of two hours. The Eskimos
were unable to keep their arms in water at 45°C as blisters developed on their fin-
gers.

At the other end of the temperature scale, it appeared that average blood flow
in the Eskimo, when the arm was immersed in water at 10°C, was significantly higher
than in similar experiments in this country. Although these results are very sugges-
tive, it is clear that further experiments are needed.

There are many references in the literature on the Eskimo which indicate an im-
proved tolerance to cold especially in the hands, such as ability to handle cold

209

objects or to carry out manipulations which would be impossible for the white man.
There is also clear evidence by Mackworth that local adaptation to cold in the fin-
gers can be developed by continued exposure. Mackworth measured the duration and
degree of finger numbness by changes in tactile discrimination during and after ex-
posure of the bare finger at various temperatures and wind speeds. He carried out
his first experiment at Fort Churchill and found a significant difference between in-
door and outdoor workers; the latter have less and shorter impairment of finger
numbness with similar exposures than the former. As the results might have been
interpreted as indicating an ability to discriminate with fewer sensory clues by prac-
tice, i.e. a cortical rather than a local change, Mackworth also carried out experi-
ments at Cambridge. A number of subjects spent two hours a day in a room, the tem-
perature of which was kept at — 10°C. They wore ordinary seamen's clothing, with
bare hands. After two hours, one finger was exposed to a blast of cold air and tac-
tile discrimination was measured. The rest of the day was spent in normal activi-
ties outside the chamber.

During the first two to three weeks the finger numbness steadily diminished and
thereafter kept at a steady level. This result was not due to a learning factor as
shown in another experiment in which the subjects only spent one hour a day in the
cold room. No decrease in the numbness index was obtained under these conditions.
This experiment is an important one as it is the best objective evidence of signifi-
cant acclimatization to cold. Studies on the vasomotor changes are not yet complete.

The fishermen of Nova Scotia who habitually have their hands in cold water also
exhibit a degree of adaptation. Uhen the hands are plunged into the water, the nor-
mal individual suffers considerable pain, with a sharp rise of blood pressure. This
is the basis of the cold pressor test used in clinical medicine as a test of actual or
potential hypertensive subjects. The fishermen experience no pain and no rise of
blood pressure on immersing their hands in ice water.

Other factors which were investigated by the Queen's University group included
the nutrition of the Eskimo. It is commonly supposed that this dietary consists of a
high fat, high protein and low carbohydrate content. The difficulties of determining
the average diet is very considerable owing to the very wide daily and individual
variation both in composition and calorie value. Within a single week t^ie daily in-
take of one individual varied from 2,000-6,000 calories. On one occasion 80% of the
calories might be derived from fat, on others it might be as low as 10%. The raw
material of the food was available ad libitum, but it appears likely that the very er-
ratic feeding habits are related to the more normal situation in which food supplies
are dependent on successful hunting and hence are extremely irregular.

As a result of many dietary experiments in relationship to cold, it has been
shown that calorie requirements are considerably increased in the cold. In temperate
zones the diet of the soldier provides approximately 3,300 calories. At Fort Chur-
chill the rations issued yielded 5,000 calories. Part of this increased metabolic de-
mand is due to the hampering effect of arctic clothing, and it is not completely cer-
tain if there is a true metabolic increase apart from this.

210

High fat, high protein and high carbohydrate diets have been compared in ex-
periments with subjects who lived for long periods in cold chambers. Cold tolerance
was highest on the high fat diet, although the high carbohydrate diet was almost as
good. High protein diet was markedly inferior. Mitchell, Gluckman and their col-
leagues, who were responsible for these experiments, suggest that the high fat diet
may owe its value to the laying down of fat in the subcutaneous tissue.

The thickness of the subcutaneous layer of fat can be of considerable impor-
tance as regards insulation. The thermal conductivity of huma'n fat is from V2 - %
that of muscle. The difference in conductivity within the body may be even greater
as muscle is a much more vuscular tissue than fat. Recently my colleagues Dr Pugh
and Dr Hatfield have investigated the effects of immersion in cold water, and were
particularly interested in the performance of long distance swimmers. During the
last war considerable information was obtained of the survival at sea of shipwrecked
sailors. Molnar collected this information and his figures showed the time during
which survival was likely at various sea temperatures. His figures suggest that at a
water temperature of 15°C, there would be few survivors after five hours' immersion.
On the other hand, Channel swimmers may spend from 10- 20 hours in the water, and
measurements made last year during the race across the Channel showed that the
water temperature, except along the coasts, was approximately 15°C. Observations
were made on a number of the competitors in this race, and one volunteered for fur-
ther experiments. All the competitors examined were extremely fat with a subcu-
taneous layer up to three times that normally expected. Comparisons were made of
the rate of cooling of the Channel swimmer and control subjects. In well stirred
water kept at 15"^ normal subjects shivered violently and the oxygen consumption
rose up to seven times the resting rate. In spite of the violent shivering, rectal tem-
perature fell and one subject had to be removed after 40 minutes. The swimmer, on
the other hand, remained lying in the water reading a paper, with only very mild
shivering and quite comfortable. His metabolic rate was only doubled and there was
no fall in rectal temperature. From studies made of the tissue gradients it was clear
that the great difference was largely explained by the insulation of the subcutaneous
fat.

It is possible, therefore, that the high calorie, high fat diet, which is preferred
in cold regions, may owe some of its value to the increased development of subcu-
taneous fat.

The mechanisms which may be responsible for increased tolerance to cold, de-
veloped during long exposure to cold,can include a small rise in basal metabolic
rate, a changed distribution of blood permitting a greater constriction of peripheral
vessels, an increased insulation provided by fat, and a diminished local effect of
cold, the mechanism of which is as yet unknown.

The greatest and certainly the most important adaptation to cold environments
is not these relatively small physiological adjustments but learning how to live in
arctic conditions. The insulation required is provided by clothing, the absence of
cold injury is due to the avoidance of risks, and a knowledge of the conditions

211

likely to cause accidents. It is quite easy to detect early frostbite on exposed parts,
such as the cheeks or nose, and to rewarm such areas merely with the fur on the
back of gloves without any damage or discomfort.

Life in cold climates is perfectly tolerable once the rules are obeyed.

References

Hatcher, R. 1950. Personal Communication.

Hatfield, S. and Pugh, L. G. C. E, 1951. Thermal Conductivity of Human Fat and Muscle.
Nature, LoncL, 168, 918.

Mackworth, N. H. 1952. Cold Acclimatization and Finger Numbness. Medical Research
Council. Applied Psychology Research Unit Report, No. 173.

Mackworth, N. H. 1953. ]■ Appl. Physiol. 5, 533.

Mitchell, H.H., Glickman, N., Lambert, E.H., Keeton,, R. W. & Fahnestock, M. K. 1946. The
tolerance of man tocold as affected by dietary modification; carbohydrate versus fat and
the effect of the frequency of meals. Amer. J. Physiol., 146, 84-96.

Scholander, P. F., Walters, V., Hock, R., & Irving, L. 1950. Body Insulation of Some Arctic
and Tropical Mammals and Birds. Biol. Bull., 99, 225.

Scholander, P. F., Hock, R., Walters, V., Johnson, F. & Irving, L. 1950. Heat Regulation in
Some Arctic and Tropical Mammals and Birds. Biol. Bull., 99, 237.

Scholander, P. F., Hock, R., Walters, V. & Irving, L. 1950. Adaptation to Cold in Arctic and
Tropical Mammals and Birds in relation to Body Temperature, Insulation and B.M.R.
Biol. Bull., 99, 259.

212

SOME ASPECTS OF HUMAN ECOLOGY IN HOT TROPICAL REGIONS

Professor Sir David Brunt, Sec.R.S.
(London)

I have always been attracted to that aspect of physiology which deals with the
relation of man to his physical environment, by the hope of finding some logical
basis for the classification of the climates which occur in different parts of the
globe. In Fig. 1 below is reproduced a diagram* in which is given a tentative clas-
sification of climates, which I had hoped to test by comparison with data of times of
day and year when work of a specified degree of activity became impossible. The
line CC was assumed to be the limit to the right of which outdoor work would be try-

P C , A B

X>

40

50

SO

JO

BO

90 100 ^ US ao

TEMPERATURE *F.

m

140 150 leo iJO 190 190 200 no
(by courtesy of the Physical Society).
Figure 1.
AA. Heat- stroke limits for nude man resting in still air.
BB. Heat- stroke limits for nude man resting in air moving 200ft/min.
CC. Limiting conditions for clothed man resting in sunshine with about one- third of skin

wetted with sweat.
DD. Limiting conditions for clothed man walking 3 m.p.h. with about one -third of skin

wetted with sweat.
The broken line represents equivalent temperature 80°F. The figures 500 g, etc., indicate
rate of evaporation of sweat in grammes per hour for men of average size in order to main-
tain heat balance of the body.

* from Brunt, D. 1947. Some Physical Aspects of the Heat Balance of the Human Body,
Ptoc. Phys. Sac, 59 713.

213

ing, or in extreme cases impossible, but it has not been possible hitherto to test
the truth of this supposition. Through the courtesy of Mr D. A. Davies, Director of
the East African Meteorological Department, climatological data for his area, and a
statement of the hours of office work in certain parts of Africa, have been supplied
to me.

The statement concerning hours of office work is as follows:

A Uganda and Tanganyika (including Lake Area) - Normal office hours with lYi hr

lunch break.

B Kenya Highlands and East of Rift — Normal office hours.

C Rest of Kenya - Normal office hours with 2hr lunch break, except that sometimes

in Northern Province Area there is no afternoon work.

D Zanzibar and Pemba — No work in the afternoon.

The following stations were selected as characteristic of each of the four areas:

A Kitgun

B Nairobi, Nakuru

C Mombasa

D Chukwani

The data were represented in a diagram as in Fig. 2, the monthly mean of the
daily maximum temperature being plotted against the monthly mean of the daily mini-
mum relative humidity, which will be approximately synchronous, except that for
Chukwani the 15h mean values are plotted. It was thought that observations from
Area A (Kitgun) should be about marginal between conditions possible and impos-
sible for afternoon work, and that in Fig. 2 points representing area B should fall to
the left of those for Kitgun, and those representing areas C and D should fall mainly
to the right of those for Kitgun. Fig. 2 only shows observations for Kitgun, Nairobi
and Chukwani, the other stations selected for insertion being left out for the sake of
clarity. A number of places in area B (e.g. Nakuru) were represented in Fig. 2 by
points to the left of the strip covered by the Nairobi observations and so were not
retained in the final form of the diagram.

Area D, where no work is done in the afternoon, is represented by Chukwani in
Fig. 2. The points representing monthly values for Mombasa, where there is a 2hr
lunch break, fall so closely within the same area as those for Chukwani, that they
were omitted. The line CC which has been drawn in Fig. 2 is the line CC of Fig. 1,
and it appears to give as good a fit as can be expected for the marginal conditions
represented by the observations.

The view that CC is a boundary having some practical value is put forward in
the hope that either further confirmation will be available, or other observations are
available which show that this boundary requires revision. There must be some such
boundary for work of any specified degree of activity and it is an important matter
to obtain as close a specification as possible.

I will assume for the rest of this paper that I am correct in using CC as a boun-
dary such that conditions represented in the area to the right of CC will make it im-

214

so 85 90

TEMPERATURE °F

lOO

I05 IIO

Figure 2.
Monthly mean values of maximum temperatures (daily) and daily minimum of relative humidity

possible to do even light work out-of-doors. If the relatively frequent occurrence
for 3-4 hours per day of conditions hotter than correspond to the line CC is to be
avoided, the mean temperature of the hottest month should not exceed 75°F in a dry
climate, or 73°F in a damp climate. This rule may be taken as a rough guide.

At Beira on the Coast of Portuguese East- Africa, slightly north of latitude 20°,
the monthly mean values of minimum relative humidity vary between 59 and 63%, the
monthly mean maximum temperatures varying from 77.3°F to 89.6°F. All the months
from May to September are to the left of CC in Fig. 2, all the remaining months being
to the right of CC. Thus it is likely that normal office hours or light workout- of-
doors, would be possible from May to September inclusive, but would not be possible
during the remainder of the year.

Among the data which I received from Mr Davies from East Africa were hourly
mean temperatures and relative humidities for each hour of the day, and month of
the year, for Chukwani, From these data it is possible to represent, on such a dia-
gram as that shown in Fig. 2, the mean diurnal variation of the conditions for each

215

TEMPERATURE °F

Figure 3.
Mean diurnal curves for January and July

month of the year. Such a representation is shown in Fig. 3, for the extreme months
January and July, which are the hottest and coolest months respectively.

From Fig. 3 it may be concluded that at Chukwani in January light outdoor work
is likely to be impossible during the whole day, while even in July work is likely to
be impossible during the hours of the afternoon.

If such a diagram is drawn for each month of the year, for any place, we can
from these come to a conclusion as to the number of months of the year in which
work is possible during the day. From Fig. 3 it seems safe to conclude that Chuk-
wani is not a pleasant place for the white man to settle in.

Data for Khartoum are also shown in Fig. 2, and appear to indicate that in the
months from May to October active work is likely to be trying for the white man
during the afternoon. That this is true is confirmed by Mr J.F. Ireland, the Director
of the Sudan Meteorological Service.

The method outlined above in Figs. 2 and 3 can readily be applied to relate the
problems of white settlement to climatological conditions. My friend and former
pupil Dr S. P. Jackson, of the Department of Geography at the University of the Wit-
watersrand, has carried this method of analysis to its logical conclusion and has
given a map of Africa south of about latitude 12°N, on which is indicated those

216

217

parts of Africa which are unsuitable for European settlement (a) throughout the year,
(b) for eight months of the year, (c) for four months of the year, or (d) suitable for
more than eight months of the year.

When a diagram such as Fig. 3 shows that four hours of the day in a particular
month fall to the right of the line CC, it is concluded by Jackson that the place in
question is unsuitable for Europeans during that particular month. The work invol-
ved in drawing such a map as that shown by Jackson is completely straightforward,
and requires only reliable values of temperature and humidity for each hour of the 24
in the day. His map for regions south of the Sahara is shown in figure 4.

Dr Jackson's work is, as he himself states in the publication referred to*, only
a first approximation. It does not take account of winds, nor of the possible effects
of a long stay in a monotonous climate, which is regarded by some writers as having
a serious effect in leading to a loss of initiative and efficiency.

In the brief statement above, I have endeavoured to show the vital need to make
use of the fact that in some parts of the globe active outdoor work is only possible
during some hours of the day, or during some months of the year. There is a great
need to collate records of the hours when work is possible with the meteorological
records available, and possibly to institute new meteorological stations in regions
as yet unprovided with records. I should regard such collation as a most useful ad-
dition to the knowledge we have acquired by laboratory experiments on human sub-
jects in controlled atmospheric conditions.

I have asked that the African Regional Association of the World Meteorological
Organization should discuss this matter at their forthcoming session in January 1953,
with a view to considering what information they can supply. If any information can
be obtained by this means, or by any other means, I should be glad to do the work of
reducing and discussing such observations.

* Jackson, S.P. 1951. Chapter 1 in: Africa South of the Sahara. Oxford University Press.

218

DISCUSSIONS

Session I.
CLIMATE AND PHYSICAL ENVIRONMENT

Chairman Dr Edward Hindle, F.R.S.

In the discussion that followed the first session, Professor J.A. Prescott,
F.R.S. said that there was no nomadism in Australia, but that cattle moved to new
green areas after each thunderstorm. There were no fences in Northern Australia
to restrict their movements, and permanent water- holes were kept as a final re-
serve: Canning stock routes had been used only once since their establishment and
were difficult to maintain. The scattered nature and local distribution of rainstorms
had long been recognized by pastoralists engaged in cattle rearing on the tropical
margins of the Australian deserts, and played a part in determining the size of 'pad-
docks'. Cattle moved towards more favoured areas during the dry season and would
be stopped by fences.

Considerable experience had been gained in Australia on the use of water for
irrigation, stock and domestic purposes. In the neighbourhood of Adelaide water
containing 800 parts per million total salts was regularly used for irrigation and was
supplemented by 20 inches of rain falling in winter. Probably the longest record of
the satisfactory use of such water for irrigation came from Siwa Oasis where waters
containing 2,000 parts per million of salt had been in use. An extraordinarily effi-
cient drainage system had made possible an unusually permanent irrigated agricul-
ture. In Southern Australia the search for underground waters was of lively interest
to the Department of Mines and Geological Survey. The existence of overlying
saline ground waters was frequently observed, and attempts were made to avoid
mixing these with fresh water from lower levels.

Asked whether water could be de-salinized chemically, Professor F. W, Shot-
ton replied that the operation was costly, required skilled supervision and was
therefore not practicable, but Dr H. Boyko said that methods were being investigated
at Harvard and the Weizman Institute. Another speaker pointed out that in hot cli-
mates saline drinking water was desirable, and Dr N. Wright enquired about the ade-
quacy of geological knowledge. Professor Shotton agreed that such knowledge was
still inadequate and that the details were largely unknown. Although the quality of
underground water could not be determined in advance, geophysical methods could
increase the proportion of productive borings. Dr C. B. Uilliams asked about the
wells at Fuca, and Professor Shotton answered that there had been two native wells
there.

Mr J. Tosic said that a distinction must be drawn between 'free* and 'bound'
water in the analysis of desert soil samples, and Mr H. Green pointed out that irri-
gation was associated with a stable system of agriculture but that there were transi-
tional stages leading to nomadism. The flood waters of the Nile did not follow pre-
cisely the same course each year. The inland deltas of the Gask and Barak rivers
in the Anglo- Egyptian Sudan and Wadi Bana at Aden each year received violent

219

spates of water containing so much sediment that storage by means of dams was im-
practicable. Deposits caused a rise in the level of the river bed and this natural
rotation of the soil reduced weeds. The cultivator had the advantage of using the
equivalent of virgin land where the ground was watered only once in four years be-
cause pests were eliminated. Nomads noted the direction and duration of storms
before deciding where to cultivate. It was important to see that water was not con-
sumed by unwanted vegetation. Near Khartoum, where there were only a few inches
of rain annually, mosquito trees had been established on sand dunes by planting
them at shallow depth in moist sand and removing the inconspicuous weeds. Simi-
larly nomads guarded their lands from trespassing herds. Dr Williams then pointed
out that the frequency of the distribution of rainfall was on a logarithmic scale.

Professor J. F.Danielli asked why alkalinity was so serious and Professor
Prescott replied that sodium carbonate made the soil impermeable and no crops
could tolerate an alkalinity above pH. 10.0.

Session II
PLANT ECOLOGY

Chairman Dr B. T. Dickson

Professor F. W. Shotton asked whether the artesian water of the Bahrain Islands
originated from Central Arabia and Professor R. D. O'Good answered that such
was the local opinion. The water was believed to flow northwards towards the Per-
sian Gulf. Professor F. S. Bodenheimer said that between the times of aestivation
and hibernation there was a short, favourable period during which it would be fatal
for plants and animals to become active. Only in spring was the favourable period
long enough for development. Professor M. Zohary agreed, and added that the
Middle East Deserts belonged climatically to Africa rather than Asia.

Dr C. B. Williams enquired about the possible hygroscopic value of the salt
crystals that encrust many desert plants, but Professor D. Thoday said that plants
could not absorb water from them. Professor G. E. Blackman said Professor
Zohary's measurements were all of dry weight and asked why he had given no mea-
surement of transpiration from unit areas. The latter replied that surface measure-
ment in desert plants in the spring was open to many errors.

Session III
ENTOMOLOGY AND ECOLOGY

Chairman Dr J.W.Evans

Referring to Professor F. Bernard's paper Mr D. Wragge Morley said that the
more primitive ants were nearly always insectivorous while the more highly deve-
loped species were omnivorous and protected scale - insects, aphids and other harm-
ful plant- sucking insects. At the same time the more primitive ants, like Catagly -
phis could not compete with the social Monomorium. The latter and similar 'harm-

220

ful' ant species required conditions in which agriculture was possible while the
hunting Cataglyphis were not so dependent on well-established vegetation and
were therefore to be expected on the desert fringes and gave way to more social
species in agricultural areas. It was not however true to say that the ants which
protected plant- sucking insects were invariably harmful. The activities of ants in
turning over and aerating the soil might be of special importance in cultivated areas
near deserts. In Brazil for example where there were no earthworms, it had been
calculated that ants brought to the surface nearly a third as much soil again as that
brought up by earthworms in Europe.

Mr R. M. Elton referred to the importance of insects as human food and men-
tioned that he himself had sampled 43 species in Africa and Australia. Many such
as the witchetty grub contained a large proportion of moisture and their high salt
and glucose content enabled the natives to travel long distances on this diet in hot,
dry deserts.

Professor A. Balachowsky said that crows were never to be seen feeding on
date-palms in the Sahara, but that in their search for ticks they sometimes injured
camels and were therefore shot when seen on the backs of these animals.

Professor F. S. Bodenheimer emphasized the vulnerability of crops in oases
both to insects that changed their food habits, and to all the animals which attacked
plants for the sake of moisture.

Dr C, B. Williams pointed out that the North African desert was one of the
routes by which insect migrants travelled to Europe and that they bred along the
fringes of the desert. This area was also the main breeding- ground of many insect
pests whose numbers varied according to the rainfall. Similar conditions occurred
in North America.

Mr H. Green suggested that Professor L, Emberger had not sufficiently empha-
sized the skill required by an ecologist before he could safely interpret his obser-
vations and make them a guide in land use. Conditions were radically altered by
irrigation and fencing, pest control and the use of fertilisers or trace elements.
Consequently the ecologist's inferences involved a large subjective element of
skill and experience and an appreciable chance of error.

Dr H. Boyko also said that the interpretation of ecology to agriculture was
skilled work and often man obtained less from the land than it could support natu-
rally.

Session IV
ECONOMIC ASPECTS

Chairman Dr H.G. Thornton F.R.S.

In reply to a challenge for evidence that forest clearance resulted in reduced
rainfall. Professor E. P. Stebbing said that during their advance into India, Alex-
ander and his army had marched through vast areas of virgin forest where now only
desert was to be found. Professor J. F. V. Phillips added that during the last 150

221

years aridity had definitely increased in the eastern part of Cape Province and Nor-
thern Transvaal following the removal of evergreen forest. Although it could not be
proved that tree planting increased the rainfall or that forest clearance reduced it,
the availability of water was certainly increased by the presence of trees.

Commenting on Professor Stebbing's remarks, Dr A. S. Thomas said that a deep
humus layer was seldom to be found in tropical forest: indeed there was usually
more organic matter in grassland soils. He did not believe that fire had a dele-
terious effect on grassland - the worst factor was compacting of the soil surface by
stock animals. Tramping had produced desert -like conditions in Karamoza where
there was an annual rainfall of 25 inches. When tse-tse fly invaded the land how-
ever, and the stock went away, the vegetation soon recovered. He agreed with Pro-
fessor Phillips that tse-tse had a beneficial effect in preserving Africa.

Fire was a useful agent in the right place and at the right time, provided that
the ground was allowed to rest afterwards, said Professor Phillips: and much had
been learned about mechanized agriculture from experience in Tanganyika. The
removal of deciduous scrub at Kongwa had not resulted in a 'dust- bowl* or 'tennis
court'; and acres thrown back to nature had produced a crop of grass at the end of
a year. Africa needed a few years to rehabilitate herself he suggested.

The impossibility of countering the rape of the earth by overgrazing, when the
entire population was clamouring for food, was mentioned by another speaker with
experience of the problems in Somaliland Protectorate.

The paper by Professor H. C. Trumble and Mr K. Woodruffe was presented by
Professor J. A. Prescott, F.R.S. in the absence of the authors. In a short introduc-
tion, the latter said that the University of Adelaide possessed two field stations in
the semi- arid fringe to the southern margin of the Australian desert. Koonamore
dealt with the natural regeneration of native shrub steppe and was in charge of the
School of Botany. Yudnapinna had been endowed since 1938 for the special study
of pastoral management in this environment and was the responsibility of the Waite
Agricultural Research Institute. There had been pastoral occupation in the regions
for nearly one hundred years, and overstocking with sheep during drought periods
had resulted in an estimated loss of 80% of the original perennial shrubs. It was
expected that these studies by the Waite Institute would lead to a basis for the es-
tablishment of scientific principles of pastoral husbandry.

In the discussion following the paper, Dr H. Boyko enquired about competition
between bushes and grasses in areas where Atriplex and Kochia were the dominant
plants. This was a subject of very great importance in large areas of North Ameri-
ca. North Africa and South-west Asia. In North America Artemisia tridentata cov-
ered large areas as a result of overgrazing. He had recently seen large scale ex-
periments from Montana to Texas that were designed to establish methods for con-
trolling this undesirable shrub. In North Africa and South-west Asia, a related
species A, herba alba covered nearly the whole area between the isohyetals of 200
and 400 mm. from Morocco to Afghanistan where there were cool winters. The latter

222

species however had a root system providing much weaker powers of competition.
After the elimination of grazing for a number of years good fodder grasses could
easily compete with it. This was because the rainwater was absorbed by the fi-
brous root system of the grasses before reaching the deeper top -roots of the Arte-
misia,

Session V

MAMMALIAN PHYSIOLOGY* AND ECOLOGY I
Chairman Professor A.V.Hill, F.R.S.

Dr Edward Hindle, F.R.S. referred to the work of Dr H. B. Cott who had shown
that the dark colours of some desert birds were a protective mechanism, since these
were unpalatable. Professor J. F. V.Phillips drew a parallel between Dr N. Wright's
fat- tailed sheep and the 'fat- tailed' bushmen of the Kalahari and Karoo deserts.
He agreed that the selection of indigenous sheep and goats was most important.
Professor F. S. Bodenheimer said that a small school of thought held that a high in-
cidence of arterial sclerosis was related to large amounts of ultra-violet light
linked with excess vitamin D. Ultra-violet light penetration was greater in pale
than in dark skinned men, and white cattle suffered more greatly from fatigue in
South Africa than did black cattle.

Dr E. J.Moynahan said that there was no doubt that melanin production formed
an important part of the human protective mechanism against ultra-violet light. It
had been shown that radiation of short wave-length stimulated melanoblasts to
produce melanin and in addition ultra-violet light blackened pre-existing pigments
in the skin. The melanin was laid down to begin with as a supra- nuclear cap in
the cells of the malpighian layer of the skin. The thickness of the overlying hairy
layer was another important factor affecting ultra- violet light penetration: this
layer was thickened as a response to ultra-violet light, and was thicker in negroes
than in the skins of white races. Neither a high intake of vitamin D nor excessive
ultra-violet light played a part in causing arterio- sclerosis. The toxic effects of
excess doses of calciferol were mainly confined to the kidneys and were reversible.

In answer to a question about the productivity of Sudanese cows, Dr Wright
said that this was up to 1,000 gallons with an average of 350-400 gallons with
good feeding. Butter fat was up to 8% with a normal figure of about 6%.

The Chairman asked Dr Bodil Schmidt- Nielsen whether she had examined the
alveolar air of kangaroo- rats. She answered in the negative but said that the oxy-
gen dissociation curve of these animals was the same as that of white rats, as was
the oxygen and carbon dioxide content of the blood. In reply to another question
she said that fat storage in the camel's hump and elsewhere had a negligable effect
on water economy. The additional ventilation required for the oxidation of this fat
counterbalanced the metabolic water produced.

223

Session VI

MANdMALlAN PHYSIOLOGY AND ECOLOGY II
Chairman Professor J.F. Danielli

Dr J.S. Ueiner suggested that the discomfort zone might be higher than was
thought, for people could tolerate a good deal of sweating without discomfort. The
voluminous clothing of the Arabs kept out radiation and solar insulation, and saved
water. Uhen called on to do heavy work the Arab discarded most of his clothing.
Little was known of the effect on Europeans of continuous residence in tropical
climates.

Dr E. J.Moynahan said that there was a definite relationship between lack of
pigmentation and the incidence of rodent ulcer and other cancers of the skin. These
tumours occurred more frequently in white races living nearer the Equator. In mam-
mals with protective coats of hair, cancers were very rare.

Mrs G. E.C.Stone emphasized the need for aerial surveys of deserts and es-
pecially of arid sub- desert marginal regions as a framework on which to fit detailed
knowledge as it became available. Much geological survey work could be carried
out by means of aerial photographs, a method that saved a considerable amount of
time. At present photo- geologists concentrated on the areas in which oil and other
minerals were likely to be found; but Professor Prescott had already said that in
Australia water was considered the most important mineral. Photo- geological me-
thods might indicate the direction of an aquifer, and the survey of desert marginal
areas might assist protection against the extension of man- made deserts.

After the discussion, Dr Frank Malina spoke on behalf of UNESCO, and Pro-
fessor J. F. Danielli, Honorary Secretary of the Institute of Biology, summed up. He
said that from the papers presented at the symposium it was apparent that indivi-
dual deserts presented a multitude of different problems. Scientific investigation
must preceed development, but in most cases the major difficulties were social,
moral and political and presented problems of ethics rather than of science.

Set on Vari-typer at the Royal Society and printed by photolithography by
J. Smethurst & Co. Ltd, 35 Rothschild Road, Acton Green, W.4.