Content uploaded by Richard Cincotta
Author content
All content in this area was uploaded by Richard Cincotta on May 06, 2014
Content may be subject to copyright.
letters to nature
990 NATURE
|
VOL 404
|
27 APRIL 2000
|
www.nature.com
Acknowledgements
We thank the HSDP team for providing samples. We thank S. Simakin for ion probe
analyses of inclusions; L.V. Danyushevsky for the access to PETROLOG thermodynamic
modelling software; S.V. Sobolev for modelling phase compositions at high T-P;
P. Kelemen for providing unpublished data on Oman Gabbro; E. Macsenaere-Riester for
help with the electron microprobe analyses; F. Ku
Ènstler for preparing doubly polished
sections; and F. Frey, J. Eiler, L.V. Danuyshevsky, V. S. Kamenetsky, A. A. Gurenko and
S. R. Hart for comments that helped to improve the clarity of the manuscript. This work
was supported by Deutsche Forschungsgemeinschaft and the Russian Foundation of Basic
Research (A.V.S. and I.K.N.) and an Alexander von Humboldt award (A.V.S.).
Correspondence and requests for materials should be addressed to A.V. S.
(e-mail: asobolev@mpch-mainz.mpg.de)
.................................................................
Human population in the
biodiversity hotspots
Richard P. Cincotta, Jennifer Wisnewski & Robert Engelman
Research Department, Population Action International, 1300 19th Street, NW,
2nd Floor, Washington DC 20036, USA
..............................................................................................................................................
Biologists have identi®ed 25 areas, called biodiversity hotspots,
that are especially rich in endemic species and particularly
threatened by human activities. The human population dynamics
of these areas, however, are not well quanti®ed. Here we report
estimates of key demographic variables for each hotspot, and for
three extensive tropical forest areas1that are less immediately
threatened. We estimate that in 1995 more than 1.1 billion people,
nearly 20% of world population, were living within the hotspots,
an area covering about 12% of Earth's terrestrial surface. We
estimate that the population growth rate in the hotspots (1995±
2000) is 1.8% yr
-1
, substantially higher than the population
growth rate of the world as a whole (1.3% yr
-1
) and above that
of the developing countries (1.6% yr
-1
). These results suggest that
substantial human-induced environmental changes are likely to
continue in the hotspots and that demographic change remains an
important factor in global biodiversity conservation. The results
also underline the potential conservation signi®cance of the
continuing worldwide declines in human fertility and of policies
and programs that in¯uence human migration.
In 1988, ecologist Norman Myers introduced the term `biodi-
versity hotspots' to distinguish a global set of high-priority terres-
trial ecoregions for conservation2. Myers and others argue that,
because their 25 hotspots are high in species endemism and low in
pristine vegetation (,30% remaining), wise conservation invest-
ments in these ecoregions could help minimize future extinctions2±4.
Primatologist Russell Mittermeier subsequently developed a com-
plementary concept, the `major tropical wilderness areas'5. These
three areas of tropical forest (Upper Amazonia/Guyana Shield, the
Congo Basin, and the New Guinea/Melanesian Islands) are the most
pristine of all terrestrial ecoregions exhibiting a high degree of
species endemism. Together they cover 6.3% of Earth's terrestrial
surface, an area larger than the United States or China. Myers,
Mittermeier and others propose a strategy of conservation invest-
ments in these areas as a back-up strategy for the hotspot approach1.
By using mapped world distributions of humans (Fig. 1), various
census sources and ecoregional boundary data, we calculated
population density and growth rates for each of the biodiversity
hotspots and major tropical wilderness areas (see Methods).
We estimate that in 1995 population density in the hotspots was
73 people km
-2
, a ®gure 71% greater than that of the world as a
whole (excluding ice- or rock-covered land). We found that 16 of
the 25 hotspots (Fig. 2a) have population densities at or above the
world average (42 people km
-2
). According to our estimates, from
1995 to 2000, human population was still growing in all but one of
the hotspots (the Caucasus), with 19 of the hotspot populations
300
150–300
50–150
15–50
5–15
1–5
0–1
Wilderness areas
Biodiversity hotspots
Population density (km–2)
Figure 1 World population density (1995) and the 25 biodiversity hotspots (outlined in
red, numbered), and three major tropical wilderness areas (outlined in green, lettered).
Hotspots: (1) Tropical Andes; (2) Mesoamerica; (3) Caribbean; (4) Atlantic Forest Region;
(5) Choco
Â-Darie
Ân-Western Ecuador; (6) Brazilian Cerrado; (7) Central Chile; (8) California
Floristic Province; (9) Madagascar; (10) Eastern Arc Mountains and Coastal Forests of
Tanzania and Kenya; (11) West African Forests; (12) Cape Floristic Region; (13) Succulent
Karoo; (14) Mediterranean Basin; (15) Caucasus; (16) Sundaland; (17) Wallacea; (18)
Philippines; (19) Indo-Burma; (20) Mountains of South-Central China; (21) Western Ghats
and Sri Lanka; (22) Southwest Australia; (23) New Caledonia; (24) New Zealand; and (25)
Polynesia and Micronesia. Major tropical wilderness areas: (A) Upper Amazonia and
Guyana Shield; (B) Congo River Basin; and (C) New Guinea and Melanesian Islands.
© 2000 Macmillan Magazines Ltd
letters to nature
NATURE
|
VOL 404
|
27 APRIL 2000
|
www.nature.com 991
growing more rapidly than that of the world as a whole (Fig. 2b).
Although population growth rates were, in general, highest in the 19
hotspots wholly within developing countries, growth rates in the
hotspots within developed countries were in most cases substan-
tially higher than the worldwide average for developed regions
(0.3% yr
-1
).
In 1995, nearly 75 million people (1.3% of world population)
were living within the three major tropical wilderness areas,
representing an average density of about 8 people km
-2
. (Area
boundaries enclose several major cities.) These areas are experien-
cing population growth at a rate of 3.1% yr
-1
, which is more than
twice the global rate.
If population numbers are examined in isolation of other factors,
the three hotspots with the most elevated risks, as assessed by high
human population density, are the Western Ghats/Sri Lanka,
Philippines and Caribbean hotspots. Choco
Â-Darie
Ân-Western Ecua-
dor, Tropical Andes and Madagascar head a list of hotspots facing
elevated risks on the basis of rapid population growth alone.
Notably, the latest hotspot analysis by Mittermeier et al. concludes
that the Philippines, Caribbean and Madagascar hotspots appear to
be the highest-priority of these ecoregions on the basis of their
extreme endemism and the intense packing of species into a much
reduced area of original vegetation6.
Human population variables are imperfect indicators of risk
to biodiversity. Population density ®gures, for example, obscure
patterns of population distribution within areas. Roughly 90% of
the population of the southwest Australia hotspot lives in and
around Perth, a single metropolitan area covering less than 2% of
the ecoregion; however, such uneven distribution does not negate
risk to biodiversity. There is considerable evidence of the capacity of
urban populations to alter ecosystems, which are sometimes more
than 100 km away, through demand for wood fuel (principally in
developing countries), water, food (including wild foods), waste
disposal and recreation (mostly in developed countries)7,8.
Another problem is that disturbance caused by humans can occur
in the absence of widespread human settlement. This is the frequent
result of over-logging, burning, grazing, mining and commercial
hunting that have extracted or degraded natural resources, abetted
biological invasion or polluted soil and water resources9. Population
density remains low, for example, in the most arid hotspot, the
Succulent Karoo, which experiences heavy grazing and the over-
harvesting of its ¯ora for the international trade in ornamental
plants.
Population growth rates can also be misleading indicators of risk
to biodiversity. Because growth rates are calculated as the annual
percentage change in a population, low rates of growth in dense
populations add more individuals than much higher rates of growth
in sparse populations. Population growth rates mask spatial dis-
tributions of growth and the trend of that rate. And both density
and growth rates hide the culture, af¯uence and technology of the
numbers of people they represent.
Despite these caveats, however, population trends in the biodi-
versity hotspots and major tropical wilderness areas indicate a high
risk that habitats will continue to degrade as ecosystems dominated
by humans expand and species become extinct in the world's most
biologically diverse terrestrial regions. Results of the analysis also
suggest that, whatever species conservation strategies ultimately
emerge, conservation scientists and policymakers should take
human population dynamics into account. Especially relevant are
trends and potential changes in population growth, density and
migration, and the social and economic factors known to in¯uence
population variables. One hopeful sign for the conservation of
biodiversity is that declines in human fertility are gradually slowing
population growth worldwide. M
Methods
Population density
Weestimated population densities for biodiversity hotspots and major tropical wilderness
areas (ecoregions) using the Gridded Population of the World, 1995, a geographic
information systems (GIS) layer developed by geographers at the National Center for
Geographic Information and Analysis, University of California, Santa Barbara10. The
authors of this layer call attention to numerous sources of potential error in these data,
including extrapolation from census-year estimates to 1995 projections, the mapping of
census geographical boundaries and census estimates themselves. For census data in most
industrialized countries, demographers regularly assume an error (most often an under-
count) of less than 3% of the actual population11. Errors exceeding 10% occasionally occur
in censuses in the poorest countries, particularly those experiencing political instability.
Moreover the populations enumerated vary from country to country, with some countries
including, for example, military personnel living outside the country.
Population growth
We partitioned hotspots into countries and sub-national political divisions (provinces or
states), used available growth rates or census and projection data to determine the growth
of each unit, and then calculated average growth rates for the composite ecoregion. Data
on a provincial level were used in ecoregions covering parts of Argentina, Australia, Brazil,
Bolivia, Colombia, China, Ecuador, France, India, Indonesia, Mexico, Panama, Peru,
South Africa, Spain, Turkey, United States and Venezuela. Where provincial data were
unavailable or unnecessary (where the entire country fell within the hotspot), country
population growth rates were obtained from estimates generated by the United Nations
Population Division12. These United Nations data are also the source for 1995 world
population density and 1995± 2000 world population growth rates.
Received 20 October 1999; accepted 17 February 2000.
1. Myers, N., Mittermeier, R. A., Mittermeier, C. G., da Fonseca, G. A. B. & Kent, J. Biodiversity hotspots
for conservation priorities. Nature 403, 853±858 (2000).
2. Myers, N. Threatened biotas: hotspots in tropical forests. Environmentalist 8, 178 ±208 (1988).
-0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
C
B
A
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
Population growth rate (% yr
–1
)
World population
growth rate,
1995–2000
0 50 100 150 200 250 300 350
C
B
A
25
24
23
22
21
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
Population density (km
–2
)
World population
density, 1995
a
b
Figure 2 Human population densities (a) and annual growth rates (b) in the 25 global
biodiversity hotspots (1± 25; see map for names and locations, Fig. 1)and the three major
tropical wilderness areas (A, B, C).
© 2000 Macmillan Magazines Ltd
letters to nature
992 NATURE
|
VOL 404
|
27 APRIL 2000
|
www.nature.com
3. Myers, N. The biodiversity challenge: expanded hot-spot analysis. Environmentalist 10, 243± 256
(1990).
4. Mittermeier, R. A., Myers, N., Thomsen, J. B., da Fonesca, G. A. B.& Olivi eri, S.Biodiversity hotspots
and major tropical wilderness areas: approaches to setting conservation priorities. Conserv. Biol. 12,
516±520 (1998).
5. Mittermeier, R. A. in Biodiversity (eds Wilson, E. O.& Peter, F. M.)145 ±154 (National Academy Press,
Washington, DC, 1988).
6. Mittermeier, R. A., Myers, N., Robles Gil, P. & Mittermeier, C. G. Hotspots: Earth's Biologically Richest
and Most Threatened Ecosystems (Cemex, Mexico, D.F., 1999).
7. Repetto, R. The ``Second India'' Revisited: Population, Poverty, and Environmental Stress Over Two
Decades (World Resources Institute, Washington, DC, 1994).
8. Myers, N. in PopulationÐ The Complex Reality (ed. Graham-Smith, F.) 117±135 (Royal Society,
London, 1994).
9. Stedman-Edwards, P. The Root Causes of Biodiversity Loss: An Analytical Approach (Worldwide Fund
for Nature, Washington, DC, 1997).
10. Tobler, W., Deichmann, U., Gottsegen, J. & Maloy, K. The Global Demography Project Tech. Rep. No.
95±6 (National Center for Geographic Information Analysis, Univ. California, Santa Barbara, 1995).
11. Newman, J. L. & Matzke, G. E. Population: Patterns, Dynamics, and Prospects (Prentice-Hall,
Englewood Cliffs, 1984).
12. U.N. Population Division World Population Prospects: the 1998 Revision (United Nations, New York,
1998).
Acknowledgements
We thank A. Bornbusch, D. Blockstein, F. Meyerson, R. Mittermeier, N. Myers and
D. Sperling for comments on the manuscript, and K. Sebastian and M. Bartels for solving
numerous GIS problems encountered during this research.
Correspondence and requests for materials should be addressed to R.P.C.
(e-mail: cincotta@popact.org).
.................................................................
Identi®cation of sleep-promoting
neurons in vitro
Thierry Gallopin*²³, Patrice Fort²³, Emmanuel Eggermann*³,
Bruno Cauli§, Pierre-Herve
ÂLuppi², Jean Rossier§, Etienne Audinat§,
Michel Mu
Èhlethaler*& Mauro Sera®n*
*De
Âpartement de Physiologie, Centre Me
Âdical Universitaire, 1 rue Michel-Servet,
1211 Gene
Áve 4, Switzerland
²Neurobiologie des Etats de Sommeil et d'Eveil, 8 avenue Rockefeller, 69373, Lyon,
cedex 08, France
§Laboratoire de Neurobiologie et Diversite
ÂCellulaire, CNRS UMR 7637, ESPCI,
10 rue Vauquelin, 75005, Paris, France
³These authors contributed equally to this work
..............................................................................................................................................
The neurons responsible for the onset of sleep are thought to be
located in the preoptic area1±3 and more speci®cally, in the
ventrolateral preoptic nucleus (VLPO)4±6. Here we identify
sleep-promoting neurons in vitro and show that they represent
an homogeneous population of cells that must be inhibited by
systems of arousal during the waking state. We ®nd that two-
thirds of the VLPO neurons are multipolar triangular cells that
show a low-threshold spike. This proportion matches that of cells
active during sleep in the same region6. We then show, using
single-cell reverse transcriptase followed by polymerase chain
reaction, that these neurons probably contain g-aminobutyric
acid (GABA). We also show that these neurons are inhibited by
noradrenaline and acetylcholine, both of which are transmitters
of wakefulness3,7,8. As most of these cells are also inhibited by
serotonin but unaffected by histamine, their overall inhibition by
transmitters of wakefulness is in agreement with their relative
inactivity during waking with respect to sleep6. We propose
that the reciprocal inhibitory interaction of such VLPO
neurons with the noradrenergic, serotoninergic and cholinergic
waking systems to which they project5,9,10 is a key factor for
promoting sleep.
Intracellular recordings in slices revealed only two cell types
within the VLPO. Of 102 recorded cells, most (n= 70, 68.6%)
were characterized by a potent low-threshold spike (LTS)11 (asterisk
and inset in Fig. 1a, LTS cells) that was calcium dependent, as it
persisted in tetrodotoxin ( TTX, 1 mM) and was eliminated (n=3)
by nickel (200 ±500 mM). However, we found no evidence for an
intrinsic rhythmicity driven by the LTS11 in these cells. The second,
less numerous cell type (n= 32, 31.4%) lacked an LTS (Fig. 1b, non-
LTS cells) and was usually characterized by a more or less prominent
recti®cation apparent upon depolarization from a hyperpolarized
level (Fig. 1b, arrow). Basic membrane parameters, such as resting
potential, membrane input resistance and action potential width
did not differ between the two cell types. Injection of the intracel-
lular tracer neurobiotin into VLPO neurons indicated that whereas
both cell types were medium-sized (LTS cells, n= 14; mean large
diameter 6s.d., 19.1 62.0 mm; mean small diameter, 13.46
1.3 mm; Fig. 1c, d; non-LTS cells, n= 6; 21.3 63.1 mm versus
11.8 61.3 mm, respectively; Fig. 1e, f), their shapes and dendritic
arbours were completely different. All LTS cells were triangular
(Fig. 1d) and multipolar (mean number of primary dendrites: 3.0
60.0, n= 14), whereas non-LTS cells were fusiform (Fig. 1f) and
bipolar (1.8 60.4, n= 6).
The high percentage (68%) of LTS cells in the VLPO matches that
of cells active during sleep in this region4,6 and indicates that the LTS
cells may correspond to these sleep-active cells. To test this proposal
we measured the effects of noradrenaline, an important transmitter
of wakefulness3,7,8, and found that 18 out of 20 LTS cells (Fig. 2a, c)
were hyperpolarized by noradrenaline (two were depolarized),
whereas all (n= 8) non-LTS cells were depolarized (Fig. 2b, c).
These results indicate that the LTS cells in the VLPO should be
inhibited during waking, when noradrenaline is preferentially
released3,7,8, and thus are well suited to correspond to the sleep-
active cells recorded in vivo1,12,13. Non-LTS cells, in contrast, are not
well quali®ed for that role and will not be considered further here.
The results described above were obtained from intracellular
recordings using sharp electrodes. We wanted to test whether
VLPO cells are inhibited by noradrenaline in a condition closer to
the in vivo situation, that is, with minimal perturbation of the cells'
properties. We therefore used infrared videomicroscopy14 to record
extracellularly from VLPO triangular multipolar neurons (Fig. 3a)
in a loose-attached cell con®guration15. All neurons (n= 9) tested
in this way were inhibited by noradrenaline (Fig. 3b, c). We then
tested whether neurons inhibited by noradrenaline were also
inhibited by acetylcholine, another important transmitter of
arousal3,7,8; in every case (n= 5), these neurons were inhibited by
both transmitters (Fig. 3d, e). In addition, the effects of both
transmitters were postsynaptic, as they persisted (n=2)inahigh
magnesium (10 mM)/low calcium (0.1 mM) solution.
We also investigated the two other transmitters (serotonin and
histamine) usually associated with arousal3,8. Serotonin (100 mM, n
= 10), like acetylcholine, inhibited the majority of cells (7 out of 10)
previously inhibited by noradrenaline (Fig. 3f, g) and excited only a
minority (3 out of 10). Both effects persisted (respectively, n=2
and n= 1) in a high magnesium/low calcium solution. In contrast,
histamine (100 mM, n= 5), which was also tested on neurons
inhibited by noradrenaline, had no inhibitory or excitatory effect
(not shown).
To establish the possible functional role of the LTS cells we needed
to identify their neurotransmitter. Most of the VLPO cells, retro-
gradely labelled from the histaminergic tuberomammillary
nucleus5, the noradrenergic locus coeruleus9or the cholinergic
magnocellular preoptic nucleus10, are immunoreactive to glutamic
acid decarboxylase (GAD) and thus contain GABA. We investigated
the expression of GAD in LTS cells using single-cell reverse tran-
scriptase followed by polymerase chain reaction (RT±PCR)16± 19.In
addition to GAD65 and GAD67, the synthesizing enzymes for
GABA, we examined the expression of choline acetyltransferase
© 2000 Macmillan Magazines Ltd
