JOURNAL OF EXPERIMENTAL
ZOOLOGY (MOL DEV EVOL) 291:130–168 (2001)
Evolutionary Challenges of Extreme
Environments (Part 2)
TALBOT H. WATERMAN*
Department of Molecular, Cellular and Developmental Biology, Yale
University, New Haven, Connecticut, 06520-8193
Despite a plethora of theories, basic laws of nature seem elusive in biology even though they have
usually been considered the ultimate goal of physics and chemistry (Waterman, ’68). Perhaps their
evasiveness in biology stems from the quite special entities with which it deals. Living beings
have many complex components, dynamically interconnected in multiple ways. These are clearly
rather different from those of a falling apple, radiation propagating through space, or the combustion of glucose in a flask filled with oxygen
gas. Some scientists suggest that biology is more
like engineering, because it often uses the laws of
physics and chemistry to explain living material
and its information systems (Hengeveld and
Walter, ’99). Whether this is true or not, biology
and engineering often have mutually rewarding
roles as in cybernetics and robotics (Ritzmann et
Even so, quite a few broad rules or laws about
life have been proposed. Arguably, the broadest
and most persistent of such laws (Kleiber’s Law)
This essay is a more technical and detailed version of the last
chapter of the author’s book about extremophile animals, Animal
Frontiers, to be published by the Yale University Press. Some additional material has been drawn from earlier chapters to make this
part of the book stand on its own. Part one of three (Waterman, 1999)
focused on currently productive ways to study the evolution of animals living on the environmental frontiers. This second part concentrates on relevant long-term evolutionary trends and their relation
to natural selection in extremophiles. The last part will discuss evolution and the environment, including the frontiers, as well as sources
of phenotypic variation, evolutionary rates, and extinction as potential components of extremophile evolution.
Part 1 of this discussion already showed that the challenges of the
essay’s title are twofold. Biologically, potential animal extremophiles
have been frequently challenged for at least 500 million to 600 million years to maintain their fitness in environments that demanded
greater hardiness and more stress avoidance than they had previously experienced. Failure to meet such environmental challenges
obviously would block a species or its group from becoming more
extremophilic despite the currently steep deterioration of global habitats. Professionally motivated biologists are also challenged to extend and integrate the rather scattered and sparse existent data on
extremophile evolution, as well as to analyze the mechanisms responsible and their ultimate relevance to the rapidly changing biosphere and its future.
Part 3 will discuss sources of phenotypic variation, rates of evolution extinction as a component of evolution, and extremophiles’ future.
© 2001 WILEY-LISS, INC.
deals with relative growth and quantitatively predicts the relation between metabolic rate and size
in major groups from microbes to elephants (Smil,
2000). Despite much data and thought, a widely
acceptable explanation of this impressive generalization remains to be agreed upon (Dodds, et
al., 2001). Quite often, such biological rules are
soon forgotten, or frequently rejected, sometimes
with jeers, by second thoughts of others. Yet some
of them have considerable staying power.
For instance, life’s vigorous persistence in an
unstable and often highly stressful world may depend quite typically on two pairs of remarkable,
seemingly contradictory, traits:
• diversity and unity
• flexibility and stability
On the one hand, the exuberant diversity of the
millions of different species and kinds of living
things, plus their innumerable component organs,
cells, genes and special protein molecules, would
seem to contradict any notion of underlying uniformity.
On the other hand, all living organisms are
built of the same chemical elements and do function basically in the same way, subject to the classic laws of thermodynamics. In other words, they
all share a remarkable unity, particularly in the
nature of cytoplasm, the anaerobic core of energy
metabolism, the basic genetic code, and the drives
to survival and self-replication. This oneness of
life imposes limits and constraints on evolution
that often seem to be overlooked by biologists.
In addition, flexibility and stability also appear
quite opposite. Yet they are complementary and
necessary aspects of life in a world with many
kinds of habitats, constantly changing on shorter
*Correspondence to: Talbot H. Waterman, Department of Molecular, Cellular, and Developmental Biology, 902KBT, Yale University,
P.O. Box 208203, New Haven, CT 06520-8193.
Received 22 August 2000; Accepted 12 December 2000.
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
or longer scales of time and space. Diversity and
flexibility (plasticity, evolvability) are obviously
central to our preoccupation with evolution. Yet
they are critically interrelated with the unity and
stability that have been responsible for life’s persistence for billions of years.
Another more controversial aspect of life relates
to its adaptedness. Observers of nature have for
millennia noticed that animals and their environments seem to match each other, often to an extraordinary degree. Such correlations are often
particularly dramatic in extremophiles. Desert
animals, polar animals, deepsea animals, and high
mountain animals, for instance, are usually notable for a variety of structural, functional, and
behavioral features closely correlated with the
stressful aspects of their extreme habitats. These
correlations with the environment have long been
called adaptations and have often been considered
as basic characteristics of life.
In the 1960s and 1970s, the Darwinian belief
that this pervasive match between organisms and
their environment arose mainly if not exclusively,
by natural selection, was widely accepted by biologists (Amundson, ’96). Even so, the interdisciplinary research necessary to prove such a causal
link was scarce, difficult to carry out, and, in fact,
not widely pursued despite considerable speculation that fine-tuned adaptation was indeed the
rule. This state of affairs was vigorously denounced by Gould and Lewontin in 1979 (see
also Pigliucci and Kaplan, 2000) amid architectural and literary flourishes, and for some
readers, potential links with the “sociobiology
wars” (Brown, ’99; Sterelny and Griffiths, ’99;
This outspoken indictment criticized weakly
documented evolutionary “adaptationism” and the
irresponsible “adaptationists” who practiced or
accepted it. Whether from guilt as charged, or
sheer vulnerability, a topic of central interest to
biology, seemingly at one blow became a pejorative term not to be mentioned in respectable company. Some 20 years later adaptation and its
degree of precision are still matters of controversy
(Weibel et al., ’98). Yet there are sustained signs
that the subject may be making a substantial
comeback (Rose and Lauder, ’96; Bijlsma and
Loeschcke, ’97; Givnish and Sytsma, ’97; Van de
Vijver et al., ’98; Koslowski, ’99).
Part of the problem, as is so often the case,
was rooted in terminology. For instance, following Darwin strictly, any of an animal’s features
that increase its fitness, but have not evolved
through natural selection, should not to be considered adaptive (Rose, ’96). But more general
definitions are commonly used by scientists and
engineers referring to the adaptation, Darwinian or not, of many complex systems, both biotic
and nonbiotic: for instance, sophisticated robots
(Ritzmann et al., 2000; Frank, ’96; Givnish and
Sytsma, ’97; Auyang, ’98).
As is evident from this essay, the author, as an
emeritus comparative physiologist who worked
mainly on underwater vision and orientation
(Waterman, ’81; Waterman, ’82; Waterman, ’97),
was used to a far broader definition of adaptation
(Slobodkin and Rapoport, ’74) than one based
strictly on its evolutionary dependence on natural selection. Adaptation and accommodation in
eyes, as well as many other kinds of physiological
regulation, acclimatization, behavior, learning, reproductive and developmental patterns, phenotypic plasticity, symbiosis, human culture and so
on, are surely parts of the usual match between
organisms and their environments (Frank, ’96),
not to mention the fitness of the environment itself (Henderson, ’13). Clearly, adaptation as Darwin conceived it is not the only factor in the whole
Another broad biological rule of some interest
here is Malthus’s Law. It was suggested in part l
of this essay to be an emergent universal property of life arising from its complex system properties and providing a critical internal driving
force relevant to animal evolution (Koslowski, ’99).
If so, this “law” might explain how animals are
impelled into becoming extremophiles. Malthus’s
Law states that living things tend to reproduce
themselves indefinitely until their numbers reach
or exceed the limits of the ecological resources they
require, including living space.
The resulting Malthusian population pressure
to expand any given animal’s range was suggested
as a likely mechanism bringing animals to the
frontiers of extreme environments and constantly
challenging their capacity to survive greater
stresses. In addition to this increase in biological
numbers that creates crowding pressure for expansion, several other possible evolutionary trends
were mentioned briefly in part 1. Despite some
risk of resurrecting old controversies, these deserve further discussion here as probable trends
followed by innovative forms of life beyond the
pioneering prokaryote extremophiles, such as
deepsea hyperthermophilic chemoautotrophs,
which still flourish two billion or more years after their origin.
Since life began, fossil and geological evidence
demonstrates that organisms’ individual size, complexity, and taxonomic diversity have all increased
dramatically overall. Along with numbers of individual creatures that have ever lived, these evolutionary “growth factors” may seem somewhat
redundant. Life’s defining properties no doubt include reproduction, which increases number;
growth, which increases size; and evolution, which
often increases various kinds of diversity. Both
growth and evolution also typically include increased complexity that is often positively correlated with size (Bell and Mooers, ’97).
All four trends would seem to contribute to
crowding and a need for expansion because they
usually require more space and additional resources. The obvious circularity of this argument
suggests that evolution has had a net direction,
even though it is usually considered to be unprogrammed. Moreover, these predominant, mainly
phenotypic, increases were, in detail, irregular and
often locally or temporarily reversed. Evolutionary regressions, both structural and functional,
were sometimes drastic, as in many parasites; in
addition, extinctions were sometimes massive and
perhaps inevitable (part 3).
In so far as they may seem nonDarwinian, the
four positive trends cited remain largely problematical (Futuyma, ’98). Theoretically, such changes
may be emergent properties of complex self-organizing living systems. Probably the most spectacular and familiar emergent feature of animals is
the basically self-programmed development of the
single-celled “simple” zygote into its specific kind
of many-celled complicated adult (Wolpert et al.,
’98; Peterson and Davidson, 2000).
Although considered by some biologists to be
suspiciously metaphysical, intrinsic system properties may provide biological evolution, including
that of extremophiles, with an internally driven
direction and/or a terminal state (Auyang, ’98;
Csányi, ’98). These could act in addition to genetic drift and the usually accepted Darwinian external selective forces acting on variation. They
must also be taken into account when analyzing
evolvability as well as the role of chance in evolution discussed later in this article.
Overall evolutionary gains in number, size, complexity, and diversity of life seem to have prevailed
from the earliest microorganisms to the much
later origins and subsequent evolution of the many
celled plants, fungi, and animals (Szathmary and
Maynard Smith, ’95; Baldauf, ’99). This appearance of multicellular organisms bypassed limitations to the usual minute size of single cells.
Interestingly, the typical size increase in manycelled animals took place mostly through increases
in the number of cells per individual rather than
through further increases in cell size, with some
exceptions such as nerve cell processes that have
to reach the body’s periphery from the central nervous system.
Dendrites and axons may be one or two meters
in length in a large dinosaur or a giraffe. However, cell numbers and sizes have themselves been
components of local evolution by particular groups
of animals as diverse as nematodes and amphibians. Also, the genetic and hormonal controls of
cell size and cell number differ so that, at least in
Drosophila and mammals, organ or animal size
can depend either on cell growth or cell division
(Montagne et al., ’99). However, among invertebrates of many kinds increased cell size, along
with cell numbers, may be largely responsible for
the sometimes major increases in body size correlated with larger genome sizes, discussed later in
part 3 (Gregory et al., 2000).
More broadly, the reality of a tendency of increasing animal size among the outer branches of
higher phyla (McMahan and Bonner, ’83; Jablonski, ’96; Gould, ’97) has been a topic for considerable discussion and controversy. A number of
biologists have even suggested that the abrupt
appearance of most metazoan phyla as fossils
around 550 million years ago resulted from a massive size increase plus a sudden burst of body plan
innovation (Davidson et al., ’95; Fortey et al., ’97;
also discussed later in part 3). Size increase and
the other trends cited have been considered by
some other biologists as largely intuitive (McShea,
’96) and unsupported or even confounded by the
facts. Others consider them too biased by humancentered notions of cultural “progress” to be scientifically acceptable (Ruse, ’96).
With regard to size, the earliest extremophilic
bacteria were unquestionably far smaller than the
first single-celled eukaryotes, including protozoans.
The pioneer many-celled animals in turn were
clearly smaller than many of those that evolved
later, even though many kinds of the smallest early
creatures, such innumerable bacteria and archaea,
obviously continue to flourish in the. modern world.
Conversely simpler wormlike invertebrates such
as flatworms and nematodes are notably smaller
than cephalopod giant squids and crustacean spider crabs although some marine jellyfish and the
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
human tapeworm achieve one sizable dimension.
The longest tapeworm, for instance, can match the
blue whale’s length at about 30 m.
Yet the biggest invertebrates in turn are clearly
outsized by vertebrate whale sharks and many
fossil reptiles. In any case, the largest animal
known ever to have evolved on earth is the blue
whale, still living in the world ocean. The most
massive individuals of this whale weigh in at over
150 tons, compared with 50 to 60 tons estimated
for the largest dinosaurs known so far. The whole
whale family to which the blue whale belongs is
only about 12 million years old (Carroll, ’97). If
so, that record for largest animal size has been
set rather recently, geologically speaking. Size increases in smaller branches of invertebrate and
vertebrate evolutionary trees, are quite common,
but comparable evolutionary size decreases may
be significantly less so (McMahon and Bonner,
’83). Generation time is usually positively correlated with size so that size increases are often
inversely correlated with the rate of evolution
However, an exceptional evolutionary trend toward notably smaller size has occurred repeatedly
in a variety of different kinds of marine invertebrates among the meiofauna. These tiny creatures,
less than a millimeter in diameter, live between
sand grains. Also miniaturization of vertebrates
is a striking local trend in newts (Hanken, ’99),
frogs, hummingbirds, and others (Miller, ’96).
Yet minimum vertebrate adult sizes are far
larger than the invertebrate ones. The smallest
vertebrates (certain bony fish) are clearly much
larger than the smallest nematodes, insects, or
As a result, the tiniest mature fish is much too
big to live between sand grains. In turn the smallest mammals and birds are larger than the smallest fish. The lightest adult simian primate, the
pigmy marmoset, is a lot larger than that littlest
fish (Genoud et al., ’97) or even the smallest shrew.
The smallest chimpanzee, or human, is substantially larger than that marmoset.
SIZE AND EXTREMOPHILES
Among extreme faunas, most arid land animals
are small in size, and large desertophiles are in a
minority (Degen, ’97). Several factors may explain
this relation. As a rule, the numbers and diversity of the faunas in extreme environments are
reduced, drastically in some cases. The marginal
productivity of environmental frontiers is a likely
major factor in this (part 1). However deepsea soft
sediments seem exceptional in the richness of
their fauna, mostly small in size (Grassle and
Maciolek, ’92; Van Dover and Trask, 2000). Ants,
beetles, or small lizards require trivial actual
amounts of water and food, compared with most
large mammals. Some smallish desert frogs and
toads during estivation store modest, but significant, amounts of water available briefly during
desert rainy spells. Early in dormancy the urinary bladder, lymphatic system, and even the coelom are distended with urine-like fluid (Warburg,
’97). In their own arid emergencies Australian aborigines learned to dig out and drink the water
from such local anurans.
In various deserts, seemingly minor water
sources, such as dew, water vapor, and fog, may
provide adequate external water to many kinds
of small animals. A Namibian Desert tenebrionid
beetle, for example, adds to its other, more usual,
water sources by condensing fog on the back of
its body (Nicolson, ’80). Despite the area’s severe
scarcity of rain and ground water, such fog-basking depends on frequent bouts of nocturnal fog
along the coastal severely desert strip of the
Namibian Desert that borders the eastern South
Many other animals also benefit from this
minute supply of water. Quite a few insects and
some spiders and scorpions in the Namib make
use of fog water formed on the sand surface. One
other beetle there, Lepidochora, is remarkable because it digs little trenches in the sand to collect
fog condensate and dew. Special water- and
watervapor-sensitive receptors, clearly useful in
locating minor sources, have been identified on
desert scorpion legs (Gaffin et al., ’92).
Lizards and snakes in the Namib may also imbibe fog condensed on their body surface or dripping from desert plants. Even sizable mammals,
such as jackals, have been seen to lick condensate
from rocks (Bothma, ’98). Yet elephants, giraffes,
antelopes, and black rhinos still sparsely inhabit
the northern Namibian desert. The survival of such
large plant-eating mammals, exceptional in so arid
a place, apparently depends on their detailed
learned knowledge of just where and when to locate precious pockets of scarce water and food. Also
a few individual cheetahs, leopards, and lions may
hunt in this desert with the lions sometimes feeding on fur seals at the coast (Louw, ’93).
In the stressfully arid coast of Oman bordering
the Arabian Sea, dew and fog create a situation
somewhat similar to that in the Namibian coastal
desert (Spalton, ’99). But in this case certain
desert trees, shrubs, and grasses can use traces
of water in fog and dew to maintain some photosynthesis-driven growth, despite droughts that
may last two or three years. As a result, the large
Arabian oryx (hunted to extinction in the wild but
reintroduced), an ibex, and two species of gazelles
can maintain viable populations there dependent
on the water and minimum protein thus available there in plant food. These animals also benefit from digging out underground water-storage
tubers produced by some of the desert plants. Recall, too, that dew has been ingeniously exploited
by humans for agriculture in the Negev desert apparently back to prehistoric times.
A variety of insects elsewhere can even absorb
water vapor from a sufficiently saturated atmosphere without fog. For instance, in the North
American southwest, the sand-diving desert
cockroach, Arenivaga, collects moisture from the
air by using a pair of curious little balloon-like
bladders everted below the jaws. Apparently
their specialized outer surface is wet with a secretion that condenses water vapor present in
the air and thus makes liquid water available
for transfer into the insect’s mouth. Other insects can also absorb water vapor directly near
the rectal end of the intestine.
Scale, as epitomized by Kleiber’s Law, surely
has a pervasive influence on animal physiology,
behavior, and ecology (Schmidt-Nielsen ’84). Measured as body mass, animal sizes range over as
much as eight orders of magnitude between
minute rotifers or midges and whales. The resulting differences due to size have remarkably large
effects on most aspects of the animals’ lives.
Within the mammals, for instance, the metabolic
rate of a gram of muscle in an Etruscan shrew,
the smallest of mammals, is about 100 times that
of one gram of elephant muscle and may be 400
times that of a gram of muscle in a blue whale,
the largest of mammals. Even so, the whale overall needs perhaps 50 thousand times as much oxygen as the shrew even though the cetacean’s
weight is about 20 million times that of the smallest mammal.
In addition, diversity has been reported to be
inversely related to the animals’ size (mass) for
many kinds of animals, ranging from marine invertebrates, to insects and all vertebrate classes,
including mammals (Gardezi and da Silva, ’99).
Also because generation time is often shorter in
small animals, their evolutionary rates tend to be
faster. The wide extent and strong influence of
animal sizes greatly increases the spectrum of ani-
mal types potentially available to evolve into various extreme niches.
However, an important common feature of extreme environments is also relevant here. Food
scarcity and the threat of starvation are at least
seasonal in most frontier environments, ranging
from the polar regions to subtropical deserts. This
means that animals’ actual metabolic rates at rest
and in activity are often important limiting parameters in extremophile fitness. Hence there are
major tradeoffs needed between size and regulatory complexity in extremophiles. Despite the remarkably uniform relation between size and rates
of metabolism, the rates themselves may differ
greatly with the kind of animal and its particular
activity levels as well.
Usually, the component species of animal communities and ecosystems include many small species, and successively fewer medium and large
types, culminating in a few large predators. Basically this must depend on larger species’ need for
a greater part of the total local resources and the
decreasing efficiency of usable energy transfer
with added links in the food web. Extreme environments tend to have limited food webs, because
both numbers and diversity may be severely reduced by high stress levels. Also in most severe
environments primary production, largely through
photosynthesis, is curtailed by shortage of water
and light as well as extreme temperatures that
also directly stress animals. In addition green
plants are themselves stressed on land by soil
quality and in the sea by lack of essential substances such as nitrate, phosphate, and iron. This
in turn will limit the animal community that depends on plant production for its essential energy
In simple food webs specific items may be crucial in sustaining the whole extreme ecosystem,
such as the abundant shrimp-like krill in the
Southern Ocean around Antarctica (Reid et al.,
’99). These feed on phytoplankton, and are a pivotal element in the diet not only for fully aquatic
animals, such as fishes and squids, but also for
large populations of diving birds and mammals,
including baleen whales. The long Antarctic night
should present krill with a starvation-stressful
overwintering problem, in spite of which these
crustaceans have flourished in enormous numbers
(Hofmann and Lascara, 2000). In the deep sea,
the rain of food from productive near-surface waters (Christensen, 2000) and the diurnal vertical
migration of epipelagic animals is crucial to almost all deepwater animals. Modest changes in
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
some seemingly remote component of such large
scale ecosystems may have disastrous consequences for certain extremophiles in the food web.
Ectotherms, including invertebrates and most
fishes, amphibians, and reptiles, normally have
low metabolic rates although minute flying insects, such as Drosophila, and some fast swimming fishes, have high active rates. Among
vertebrates, there is a marked difference between
amphibians and reptiles compared with birds and
mammals. The endotherms have resting metabolic rates 5 to 10 times greater than the ectotherms and perhaps 50 times greater energy
requirements when vigorously active. Energy food
and oxygen needs will escalate directly with these
rates, other things being equal.
Yet for animal extremophiles, large size may
be maladaptive for desert ants and rodents, but
of positive advantage for others, such as the
Namibian rhinoceros, the Arabian oryx or dromedary, and Bactrian camels. Yet large, medium,
and small animals have all evolved to flourish in
extreme environments, even though large size demands more of the scarce frontier items per individual. Fossil evidence indicates that large
animals are more susceptible to extinction than
smaller ones (Hoffmann and Parsons, ’97) implying that they are at greater risk under stress.
Because metabolic rates per unit weight are
lower in large animals, one animal weighing a kilogram could be more efficient in using resources
than a swarm of smaller ones weighing altogether
the same amount (Griffiths, ’92). On the other
hand, reduced population numbers may, by acting as bottlenecks, jeopardize survival because of
genetic stresses (Landweber and Dobson, ’99).
Where scarce food or water occurs in local areas
distant from one another, as in desert oases, large
size is advantageous because locomotion in large
animals can cover more territory and obtain more
mileage per unit fuel than in small ones (McMahon and Bonner ’83). As a result, camels and
antelopes demonstrate how such interacting factors may have played out in deserts; yaks and llamas do the same for high altitudes.
Range-effective locomotion also favors birds and
mammals, such as many in arid parts of Africa,
that have to migrate long distances seasonally to
escape or mitigate intolerable local shortages of
food and water (Waterman, ’88). The same relation also holds in the Arctic. Snowy owls, crows,
ptarmigan, polar bears, musk oxen, and reindeer/
caribou are among numerous arctic birds and
mammals that actively inhabit high latitudes
above the Arctic Circle throughout the year.
Because of their high metabolic rates and insulation, these endotherms, including both herbivores and carnivores, are far more capable of
long-distance rapid geographic movements than
most terrestrial ectotherms. Seeking scarce food
they tend in winter to wander widely in longitude as well as south over the tundra. If prey is
scarce, snowy owls may occasionally reach quite
surprisingly distant lower latitudes well outside
their usual subarctic range limits. Also, largesized predators and prey may benefit because
greater size, up to a moderate point, can increase
running speed as in gazelles and cheetahs (Alexander, ’89).
Note also that among breath-holding divers in
pursuit of prey, the largest living penguin (the
emperor penguin) and the largest seals, the Northern and Southern Hemisphere species of elephant
seals, regularly dive furthest and longest (for their
respective families: Spheniscidae and Phocidae)
into the deep sea. Typically, more work effective
locomotion, lower metabolic rates per gram and
larger oxygen storage capacity favor diving prowess in the larger species.
However, the largest whale, the blue whale, filter feeds on zooplankton, more abundant nearer
the surface and in shallower water. Consequently
its normal diving behavior is modest. Yet the
toothed whales, which include dolphins and the
sperm whales do pursue prey at deepsea depths.
Repeated rapid descents and ascents are made to
400m to 600m by some dolphin species and to as
deep as 1200m by the sperm whale (Berta and
In addition, large- and medium-sized high-latitude terrestrial endotherms that do not seasonally leave their extreme habitat, are often quite
capable of maintaining viable internal body temperatures in the face of frigid polar conditions
Some mammals can also avoid winter stresses by
denning and by hibernation, a special kind of dormancy. Interestingly, classic hibernation, discussed
later in this article, occurs only in some small
mammals, mostly at subpolar temperate latitudes.
Even some moderate-sized high-latitude mammals
such as the arctic fox, ermine, wolverine, lemming,
and arctic hare, are nonhibernators. Yet they flourish in the subarctic and arctic winter.
Important responses to this thermally severe
life are often reflected in thicker and denser fur,
as well as heat-conserving regulation of blood cir-
culation to the skin and appendages. Some species, such as the collared lemming, may even seasonally lose as much as 50% of their body mass,
apparently to reduce the total maintenance metabolism they need (Nagy et al., ’95). Yet reindeer, ptarmigan (Blix, ’89), and polar bears are
also known to reduce their metabolic rates substantially in severe winter weather, largely by
minimizing locomotor activity and thereby energy
needs, but they do not become dormant.
The large hoofed mammals, musk oxen and reindeer/caribou, are plant eaters feeding all winter on lichens, mosses, and various other plants
that are then sparse, but usually protected against
freezing by snow cover. Massive long-range treks
are usually required to find enough nutritious
plant material in the coldest weather. Such geographically long-range high-latitude movements
are hardly feasible for small animals such as rodents or ants.
In addition, large size is often an advantage in
surviving both high and low environmental temperatures. Transfer of heat into or out of an
animal’s body depends on the temperature gradient between inside and out, on the thermal resistance of the body surface (whether bare skin, or
skin underlain by blubber or covered with feathers or fur) as well as on the area of the body surface (Schmidt-Nielsen, ’97). According to simple
geometry, the surface area of solids of similar
shape varies as the square of their linear size
while their volume varies as the cube of linear
size. This means, among other things, that large
animals heat up more slowly in a hot environment and cool off less quickly in the cold. On the
other hand, in large animals, the surface area for
evaporative cooling is smaller relative to the volume to be cooled.
Also, a large animal, such as a dromedary, can
take up much more heat for a given rise in body
temperature than a small one. In fact, the camel
and some other desert endotherms relax their
thermoregulation to warm up a few degrees above
normal by day and cool off through dry heat loss
at night. In this way the radiative heat gain is
reduced by the weaker heat gradient between the
hot surround and the animal’s hyperthermic body
temperature. Also, the increased body temperature conserves the water otherwise needed for
sweating to maintain the normal body temperature strictly.
Although they lack sweat glands, birds in the
desert tend to lose precious water through the skin
and, as mammals do, by respiration (Tieleman and
Williams, ’99). Most birds are diurnal and do not
burrow, so they tend to experience the desert’s
heavy heat load fully. Also, their metabolic rates
and body temperatures are somewhat higher than
those of mammals. This reduces the safety margin between their normal levels and lethal levels.
As a result, thermoregulation must be closely controlled for their survival. Like mammals in a hot
environment, they can conserve water effectively
by allowing a small rise in body temperature. This
works well for small and medium-sized birds but
apparently not over long periods (more than an
hour) for large ones.
In the cold, similar relations mean that for a
large animal, its body temperature falls less for a
given loss of body heat (calories) to the environment. These size relations no doubt contribute to
the fitness of yaks, llamas, ibexes, and bighorn
sheep in high-altitude habitats; musk oxen, polar
bears, and emperor penguins at high latitudes;
sperm whales and elephant seals in deepsea dives,
as well as of large antelopes and camels in deserts,
Generally large species of birds and mammals are
less common at low elevations in the tropics and
subtropics than they are at moderately higher elevations and latitudes.
The prospects for large animals in hot deserts
are obviously related to how they cope with heat
and scarce water stresses. This depends not only
on size but also on whether they are ectotherms
or endotherms, an element of complexity discussed
further later in this article. To ease the heat load
of hot deserts, large mammals, such as rhinos and
elephants may rest at midday on bluffs and other
elevated points where cooler and stronger breezes
often blow (Reardon, ’86). Medium-sized animals,
including leopards, may avoid heat by sheltering
in caves or ravines as well as under the shade of
trees or shrubs.
Shelter in extreme habitats
Seeking shelter in a broad sense is a widely
used way of living in an extreme environment
without bearing the full brunt of its stresses. As
just mentioned, it usually is limited to small or
moderate-sized terrestrial animals. For instance,
a desert spider has developed conspecific aggression because suitable shelter against which to construct their webs is scarce (Riechert and Hall,
2000). In most cases shelter is used daily, seasonally, or occasionally to avoid periods of potentially lethal high or low temperatures, extreme
dryness (Adamczeszka and Morris, 2000), and intense short wave solar irradiation, as well as
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
shortage or absence of food and the attacks of
predators. Although small birds and mammals
under arctic snow and desert ants in their underground nests may remain active and feeding,
decreased activity, slowed metabolism, and remarkable resting stages are often dramatic aspects of such stress avoidance (discussed later in
this article). Although referred to elsewhere in
this essay, a more coherent account of the frequent importance of shelter in extremophile biology seems in order here.
In the nests of ants and the burrows of small
mammals, temperatures are usually thought to be
significantly cooler or warmer than surface air temperatures in daytime deserts or high latitudes. Yet
relatively few detailed and well-controlled measurements have been made of the oxygen and carbon
dioxide levels as well as temperatures and moisture in various desert and high-latitude animal shelters. Available evidence implies that in subtropical
deserts severely stressful conditions are often
present even in the burrows (Walsberg, 2000) and
probably in frigid high latitudes even under snow.
In geographic regions subject to year-round permafrost (French, ’99), burrows in the ground and
caves are usually unavailable for animal shelter.
Great areas with perennially hard frozen ground
include all of Antarctica plus the South Orkney
and South Shetland Islands as well as most of
the high Arctic lands except northern Scandinavia
and Finland (warmed by the Gulf Stream). In substantial parts of the Arctic Ocean’s margins even
the sea bottoms are frozen solid with permafrost,
a remnant of earlier geological periods. Yet rather
bizarre thaw lakes occur widely in Arctic permafrost basins of various sizes and depths (Fogg, ’98).
Their animal inhabitants may be limited to one
copepod species and one fish species. Interestingly,
even very high mountains, except those located
at high latitudes, lack permafrost if they have a
daily warm up by sunlight.
An animal burrow or snow shelter can be important, for instance, in the storage of seeds or
other food by desert ants and rodents to tide over
periodic or occasional shortages. Unless they have
stored body fat for energy, or seeds, grass or tubers, collected when available, animal activity
must be reduced or stopped to conserve energy in
such shelters. In some cases, such as in lemmings,
voles, and pikas, edible plant material, plentiful
in the summer, may be gathered together, sometimes dried out, and stored for use during the winter. Arctic foxes also may, similarly, cache their
prey or parts thereof.
Although high-latitude terrestrial stresses may
be present all year, as they are in central Antarctica and Greenland, they mostly peak only during the local winter. Then the percentage of
individual animals that survive may be marginal
in the most extreme locations. In Antarctica there
are no vertebrate herbivores, or indeed any native terrestrial vertebrates at all. In contrast,
plant-eating hares, reindeer, musk oxen, and others flourish in the Arctic as mentioned. Yet even
in this less frigid north polar area, these herbivores as well as the carnivorous polar bear and
arctic fox are often subjected to seasonal near or
In addition to burrows, other sorts of shelters
allow some small animals living at high-latitude
and high-altitude to survive disabling cold without tolerance of or resistance to freezing. They can
do so by retreating into reliably frost-free refuges
locally present in the otherwise inhospitable habitat. A variety of these retreats, such as under bark
or deep in decaying wood, provide livable places
in which either active or dormant stages of a variety of rather minute animals can overwinter.
Temperatures near 0°C may prevail when it is
much colder outside.
Deepsea animals in open water have no equivalent to the shelter offered by burrowing to marine
bottom animals and many terrestrial types. Yet at
least one benefit of such shelter is offered in the
camouflage/bioluminescent system of many midwater shrimps, squids, and fishes of modest size
(Herring et al., ’90). During the day, light organs
and their sometimes elaborate related optical systems (Denton, ’90) tend to make these creatures
invisible, or nearly so, to their predators and potential prey. Less optically elaborate but functionally comparable camouflage, sometimes seasonal,
is common in the white fur and plumage of ermine, snowshoe hare, arctic fox, polar bears, snowy
owls, and ptarmigan. Yet crows are common at
snowy high latitudes and show no hint of winter
change in their jet-black feather color.
Although endotherms are the main concern of
the preceding discussion, ectotherms also can survive and flourish at high latitudes and high altitudes. In addition to cold and scarcity of food, these
habitats also usually lack liquid water and are particularly stressful for insects (Sinclair, 2000). Yet a
surprising number of invertebrates, including insects (Strathdee and Bale, ’98) and some fishes,
are active in or on snow and ice as well as flying
in air or swimming vigorously at temperatures
close to freezing. Rather unexpectedly the cold but
quite sheltered subenvironment under high arctic
snow (discussed previously for endotherms) is also
inhabited by many insects and various other
arthropods. In most cases the body temperature is
essentially that of their close surroundings.
As a result, they can function actively only
within a rather narrow band of cold temperatures.
A remarkable variety of ectotherms are active at
temperatures a few degrees above and below 0°C.
The ice worm (Mesenchaetraeus solifugus) offers
a striking example. This small oligochete, a remote relative of the familiar earthworm, actually
lives in and on subarctic glacial ice as, for instance, high on Mt. Ranier in Washington State.
There, at temperatures close to 0°C, it feeds on
windborne pollen and microorganisms.
Also a number of small insects and spiders, like
iceworms, manage to be active and reproduce at
near-freezing temperatures on permanent glaciers
on high mountains. Although warming by sunlight
is often thought of as a prerequisite for sustained
activity at such marginal temperatures, a number of high-latitude beetles and spiders take shelter during the day and emerge to feed at night.
However, their activity is limited to nights when
air temperatures do not fall too low. Clearly such
ice-living animals, even though they lack the ability to raise their body temperature above that of
the surroundings, have been able to keep their
enzymes and membranes functioning at icy temperatures either by preadaptation or evolutionary
adjustments (Leather et al., ’93).
A quite different group of worms, the nematodes, are common at high latitudes both as parasites of other animals and also free-living in
tundra. Surprisingly, one Arctic species has been
proved to survive substantial amounts of intracellular ice formation (Womersley et al., ’98). This
has usually been thought to be lethal. Certain
high-latitude nematodes may stay frozen eight
months in the year. Some sort of below-zero hardiness is not unusual in high latitudes in both
terrestrial and marine invertebrates.
In polar marine environments large numbers
of minute flatworms, nematodes, rotifers, crustaceans, and other animals actually live within the
Arctic pack ice (Gradinger, ’99). In Antarctica,
nematodes and rotifers apparently are absent in
this habitat, even though they flourish, along with
tardigrades, in some of the continent’s severe cold
desert valleys (Treonis et al., ’99) and in alpine
soil up to 3000 m or so (Thaler, ’99). Various mi-
croscopic algae flourish abundantly in the ice pack
environment and provide ample forage for herbivorous invertebrates there. In addition, larger animals, such as amphipods, krill, and some fishes
with antifreeze in their tissues, live actively swimming among the sludge ice and at the water–ice
interface. This is particularly daunting because
contact with ice crystals tends to seed internal
ice formation in a supercooled organism.
Ice-tolerant ectotherms allow, or specifically encourage by seeding, ice formation in their bodies,
usually only extracellularly. Terrestrial arthropods, frozen in this way, may survive extreme temperatures down to –100°C. Many temperate- to
high-latitude marine intertidal animals, such as
barnacles, are also freeze-tolerant, as are some
overwintering frogs and turtles. Alternatively,
freeze-resistant (freeze-intolerant) ectotherms
avoid internal freezing by supercooling.
Without antifreeze, this requires ridding the
body of small particles, including bacteria and fine
bits of food in the gut, that tend to seed ice induction. Collembola may survive in a chill coma
without freezing at temperatures of –30°C or even
–45°C (Eisenbeis and Meyer, ’99). Note that teleosts with antifreeze develop a much more modest supercooling temperatures in accord with the
freezing point of seawater at a nominal –1.86°C
or at –1.97°C near the surface around Antarctica.
Antifreeze proteins, mentioned several times
elsewhere in this essay, curiously lower the freezing point of body fluids without changing their
melting point. Apparently they act by coating tiny
incipient ice crystals, thus blocking their further
growth. In this way they can depress the freezing
point of body fluids and tissues far more than their
usual colligative effects could account for. In insects antifreeze proteins are stored within fat cells
during spring and summer, then activated and released into the hemolymph in the fall and winter.
Accumulation of sugars, such as trehalose, polyols,
such as glycerol, that may reach concentrations
of three to five molar, and amino acids (collectively
termed cryoprotectants) as well as substantial dehydration also typically contribute to the total cold
hardiness of terrestrial arthropods.
Aquatic thermal stresses are generally far less
severe than aerial and terrestrial surface habitats. Usually, deepwater temperatures, while cold,
are not extreme as are the icy near-surface waters in polar regions (Somero, in Pörtner and
Playle, ’98). With a few local exceptions, deep
ocean temperatures in recent times range monotonously within a few degrees above zero centigrade.
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
Deepsea ectotherms (invertebrates and most
fishes) must function effectively at steady temperatures just above 0°C and at high pressures
depending on depth (as mentioned in part 1).
Obviously, protein function, metabolic reaction
rates, and so on, must be geared to these factors.
Yet marine fishes living in high-latitude near-surface waters need antifreezes (Bargelloni et al., ’94)
to prevent internal ice from forming in their blood,
which freezes, unlike that of invertebrates as a
group, at higher temperatures than seawater does
(Jia et al., ’96). Strong differences between the
concentrations of osmotically active substances in
the various tissues are responsible for this important distinction between most marine invertebrates and vertebrates generally.
Cold-blooded animals can avoid the metabolic
slowdown inherent in low internal temperatures,
by somehow acquiring or retaining the heat that
they need to be more active. If so, their body
temperature may be significantly higher than
that of their close surroundings. Although most
invertebrates, fishes, amphibians, and reptiles
are cold at cold temperatures, there are a number of fascinating exceptions to this broad rule.
Despite their basic vulnerability to low temperatures, some of these animals can maintain high
levels of activity, and even significantly raise
their body temperatures above their cold surroundings.
There are two known kinds of such warm-up
devices. One depends on the animals’ internal
heat production, a natural, and thermodynamically inevitable, byproduct of metabolism. The
other depends on heat acquisition from some external heat source in the environment and can
significantly warm the blood of many cold-blooded
types. It typically depends on behavior that promotes the transfer of external heat into the body,
usually by controlling body surface exposure to
sunlight. Thermal adjustment by this means is
common for terrestrial animals both in ectotherms, such as insects and lizards and also in
birds and mammals that use it where possible to
control heat load or radiative cooling. The physical principles which permit such external heat acquisition are usually not applicable in aquatic
Reptiles, such as snakes and lizards, are well
known to seek out and bask in warm sunlight as
a means of increasing their internal temperature
by absorbing radiant energy. At night they also
may keep warm on previously sun-heated rocks
or asphalt highways. The resulting increase in
body temperature could sharpen their ability to
escape predators or speed up their digestion, their
growth, or their reproduction compared with the
rates that would prevail without warming. All of
these could facilitate their invasion of colder habitats than would otherwise have been possible. Locusts, grasshoppers, and some other insects also
regulate their temperature behaviorally by seeking or avoiding exposure to sunlight. Behavioral
thermoregulation of this kind is practiced almost
universally by terrestrial animals both cold- or
warm-blooded (Jacobs, ’96).
The alternative, internal way of warming up
cold-blooded animals usually requires a high level
of metabolism rare in invertebrates and the lower
vertebrates (Block, ’94). Yet leatherback turtles
Dermochelys coriacea, with low metabolic rates,
can maintain their body temperature at 25.5°C
in cold seawater (Paladino et al., ’90). This ability, which allows them to swim north of the Arctic Circle, is due to the species’ large size and the
effective thermal insulation of their shell. Unlike
this turtle, nearly all warm cold-blooded animals
have high metabolic rates dependent on strong
muscular activity mainly related to flying in many
insects and swimming in a few fishes. In most
cases the flight or swimming muscles provide the
required internal heat source.
For instance, actively flying insects are also exceptional in their internal capacity to raise their
body temperature above that of a cold environment. Some moths, for instance, regularly fly in
falling snow at air temperatures near zero (Heinrich, ’93). Winter moths of the family Noctuidae
are particularly interesting. They can fly at air
temperatures down to about 0°C but at lower temperatures become inactive, sheltered under leaves
and snow. Many species of such winter moths actually spend the summer in a dormant state as
pupae (Heinrich, ’93). More familiar temperate
latitude moths, of course, overwinter as pupae.
The key to this seeming “warm-bloodedness” for
most insects involved lies in their muscle tissue,
particularly the powerful flight muscles (Block,
’94). These muscles sustain the highest known
rate of energy turnover in any animal. Yet insect
flight muscle is only about 10% efficient in mobilizing mechanical energy from chemical energy,
so that 90% of the energy used appears directly
as heat. Hence heat produced just by shivering
can raise an insect’s body temperature to a level
permitting lively activity otherwise impossible at
cool or cold air temperatures. For instance, a number of insects, such as nectar feeding sphinx moths
are like old fashioned piston engine aircraft that
must warm up before taking off at low air temperatures.
Usually though, such insect warm-ups can be
effective only if the outside temperature is 15°C
or warmer. Yet, in contrast, certain noctuid moths
seem to have flight engines that can be warmed
up from a considerably lower winter starting level,
provided that they do not actually freeze. One remarkable geometrid moth can even start to fly
without warm-up at air temperatures no higher
than –2°C to 0°C. Their unusual ability involves
thermal changes in wing muscle biochemistry that
shift its most effective temperature range to an
overall lower level.
More generally, active insect fliers, such as individual honeybees, have special heat exchange
systems even in temperate environments. These
allow the insects to regulate the distribution and
level of their body heat to make their flight as
efficient as possible. Otherwise overheating would
threaten them during their normal summer
flights. In cool weather honeybees cannot fly unless the temperature of their thorax, where the
flight muscles are, is 27°C or over. The actual body
temperatures achieved by these warm-bodied insects are comparable to the 35°–40°C levels
steadily maintained by birds and mammals.
Bumblebees, common in the Arctic, and dragonflies, as well as a number of moths, also produce
enough heat with their flight muscles to reach
high internal temperatures that they can regulate over short periods (Heinrich, ’93).
Insect social warmth
Although ants usually reach their peak of diversity and ecological importance in warm climates, a few species of the genus Leptothorax
flourish at surprisingly high latitudes (Heinze et
al., ’96). One species has been found in the steppes
close to the Siberian cold pole and another occurs further north than any other ant in North
America. Survival of their small overwintering
colonies has been recorded down to –20°C and
internal antifreeze compounds may protect the
ants to even lower temperatures. Their social behavior allows them to survive winter low temperatures more successfully as a colony than as
individual ants, presumably due to food sharing.
Group behavior also keeps overwintering honeybees, for instance, from freezing in the cold
(Seeley, ’95). These bees cannot feed in cold
weather and ordinarily do not fly then. In winter
the colony shrinks to about 60% of its summer
population of worker bees, but if 15,000 or more
are present, they can keep their nest well above
lethal low temperatures. Together they do so with
heat produced by their flight muscles contracting
isotonically. The sustained social thermoregulation by a honeybee colony can maintain its nest
temperature fixed within 1°C over a whole day.
Even when outside temperatures fall to –30°C or
less, the core area of the nest can be kept above
10°C. Below 8°C to 10°C, honeybees become torpid and die in a day or two.
While heating the nest (and keeping themselves
warm) as a group, the workers come together in
a roughly spherical cluster. It is rather loosely organized when the amount of heat needed is modest but may contract to a much more compact ball
if the temperature within the nest falls toward
zero. Then the core honeybees in the cluster
produce most of the heat while the outer ones
facing head inward form an insulating shell,
comparable to fur, blubber, or feathers, which
greatly slow heat transfer in various vertebrates. Clustering, which also occurs in quite a
few species of cold-stressed mammals and birds
such as musk ox and penguins, greatly increases heat conservation, in the case of this
social insect by a factor of perhaps 20–30 times
over that for individual bees.
Actually, the honeybee’s way of surviving may
be rather precarious. Its low resistance to cold
may, perhaps, be inherited from tropical evolutionary origins, where low temperature stress was
rarely encountered. Among other things, their
overwintering is energy-expensive. As much as 20
kg of honey may be needed by a colony to keep
warm during the winter. Even in the moderate
latitudes of New England, the winter survival rate
of first-year honeybee colonies is low.
Honeybees’ overwintering shows two surprising
parallels with the winter denning of bears. In both
quite unrelated animals, a high body temperature
is maintained and reproduction takes place during a time when environmental temperatures are
below freezing and external food is not available.
Curiously, the much more closely related bumblebee survives winter quite differently than the honeybee by becoming dormant rather than keeping
warm socially. The bumblebee actually has a more
northerly range limit than the honeybee, commonly extending as it does into the tundra well
above the Arctic Circle. Yet in bumblebees the
summertime large colony may dwindle down by
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
midwinter to only a few dormant fertilized females
who seed the next generation when spring returns.
Surprisingly, a few large fishes can maintain internal temperatures warmer than cool or cold seawater. For example, large tunas as well as lamnid
and alopid sharks are able to hold their body temperature several degrees above that of the water
in which they are swimming. This is quite unexpected because the high heat capacity of the surrounding water rapidly conducts heat away from
a warmer body, as human swimmers and divers
well know (de Vries and van Eerden, ’95).
Normally in fishes such heat loss is almost total across the large efficient exchange surface of
their gills. For proper respiration a rapid current
of aerated water must circulate over them. The
warm-blooded tunas and sharks avoid the usual
rapid heat loss to cool or cold water by means of
a special heat exchanger in the swimming muscles
(Block et al., ’93). The exchanger in question functions rather like that in the feet of birds, such as
sea gulls, exposed to cold water. Numerous parallel small arterial and venous blood vessels making close contact and containing blood flowing in
opposite directions, are arrayed within the fish’s
centrally located swimming muscles.
This system acts to conserve body heat by passive heat flow from venous blood, warmed by the
muscles, into cooled arterial blood coming from
the gills. This in turn lowers the heat gradient
across the gills and so substantially reduces the
overall heat loss. Effectively, this arrangement
dynamically provides “insulation” analogous to
that of fur or feathers in terrestrial mammals or
birds. Field data show that in the bigeye tuna
(Thunnus obesus), the effect of the heat exchanger
may be adjusted to match the fish’s thermal needs
(Holland et al., ’92). Interestingly, muscle heat conservation in tunas may get out of hand and damagingly overheat the fish. What fishermen call
“heat burn” in tunas apparently results from the
violent struggles of stressed individuals caught on
hand-line hooks (Block, ’94). Presumably potential sashimi gets “cooked” as a result.
The normal success of such a heat-conserving
device is no doubt aided by the large size of the
fish concerned, such as a giant tuna. It is also
correlated with the continuous vigorous heat-generating swimming of such fishes that drives rapid
water currents into the open mouth and out over
the gills. This capacity to raise their whole body
temperature, exceptional in fishes, allows the few
species concerned to remain vigorously active, migrating and feeding, in colder, deeper water or at
higher latitudes than they otherwise could.
Ultrasonic telemetry has shown that large bigeye tuna in French Polynesia at dawn can follow
the sonic scattering layer, presumably including
the fish’s prey, from its nighttime location in the
top 100 m down to its daytime level at 500 m or
more (Dagorn et al., 2000). The tuna remain in
such depths during the day, with some interruptions, and then return toward the surface as the
scattering layer moves up several hundred meters
in the water column at dusk. The interruptions
mentioned involve several daytime excursions upward into shallow depths. These apparently are
necessary responses to two stressful aspects of the
For one thing the fish’s thermoregulatory system seems overloaded by the temperature decrease with depth that may reach 10°C or more.
So the brief trips up into warmer water presumably serve to warm the fish’s body temperature
back to a higher level. Also they may serve to pay
off an oxygen debt caused by the presence of a
stressful oxygen minimum common at intermediate depths, particularly in the Pacific (Dagorn et
al., 2000). If so, this deep diving fish behavior
shows a remarkable similarity to that of diving
mammals and birds that must return, in their
case, to the water surface periodically to breathe
(as discussed in part 1).
Another group of bony fishes, the billfishes,
which include the swordfish and marlin, as well
as a single species of mackerel, have a different
heating system specifically for the eyes and brain
(Block, ’94). The heat-producing organs, located
in these fishes’ heads, are modified eye-movement
muscles just beneath the brain. They have lost
much of their contractile machinery and have amplified the membrane systems normally involved
in triggering muscle contraction. As in many other
heat controlling systems, a capillary countercurrent heat exchanger is present and acts to retain
within the head much of the heat produced. As
with the tunas’ body warmers, these head warmers apparently allow the fishes to expand their
vigorous activity into cooler waters. Swordfish, for
instance, can regularly swim down to mesopelagic
depths of 600 m or more.
In response to sustained stress, many kinds of
animals survive by conserving energy or increasing energy efficiency, particularly because over-
coming stress itself usually requires extra work
(Parsons, ’99). Although some large mammals,
such as bears, share such slowdowns, a few small
hibernating ground squirrels and innumerable
small ectotherms have evolved the most drastic
ways of aiding passive survival by this means.
[Further examples of such parallel or convergent
evolution (Hodin, 2000) will be given in part 3.]
A large fraction of the living material and
biodiversity in terrestrial extreme environments
usually consists of many small-scale invertebrates
flourishing in the soil. Nematode worms and
minute arthropods, such as mites, springtails, and
ants are dominant faunal elements there (Thaler,
’99). As in the typical problem of terrestrial shelter in extreme environments, modest to small size
favors the availability of security for inactive and
Energy conservation in extremophiles may occur variously in response to low temperatures at
high latitudes and altitudes, to hypoxia at high
altitudes, to low humidity in semiarid environments, to osmotic challenges and/or to the
semistarvation typical of the most stressful environments. This generalized antistress response
can be observed most obviously in animals’ reduced physical activity and absence of unnecessary behavior. Physiologically, conservation is
reflected in a reduction of the metabolic rate
(Guppy and Withers, ’99; also part 1).
For instance, oxygen consumption rates for 10
species of benthic decapod crustaceans decreased
with their depths of maximum abundance between
200 m and 750 m despite the characteristic absence of a temperature gradient with depth in
their western Mediterranean deep-water, habitat
(Company and Sardà, ’98). The amplitude of such
reductions may, in various cases, be modest or
powerful in many marine, freshwater and terrestrial species. These include desert, temperate, and
high-latitude free-living nematodes, earthworms,
mollusks, tardigrades, crustaceans, insects, and
other invertebrates, as well as a few fishes, amphibians, and reptiles.
When stressed, they can partially shut down
their metabolism, sometimes to a remarkably
small fraction of its normal resting level (Womersley et al., ’98). Torpor and hibernation in birds
and mammals, diapause in many invertebrates
such as insects, and estivation and overwintering
in a number of vertebrate ectotherms, such as
lungfish and desert toads and frogs (Warburg, ’97),
are all examples of elaborate metabolic shutdown.
Some nematodes (Womersley et al., ’98), as well
as tardigrades and Artemia resting eggs, remarkably, can enter anhydrobiotic states with immeasurably low metabolic rates and with the capacity
to survive severe desiccation and high temperatures for long periods.
In some cases the scarcity of food and low temperatures at high latitudes slow down the life
cycles of animals there. As a result, maturation
may take two or more years instead of the one
growing season usual at lower latitudes. Cave animals, living in the dark, usually with sparse food,
also tend to have slowed life cycles and reduced
fecundity (Culver et al., ’95). Comparable attenuated life cycle may also occur at high altitudes for
reasons similar to those cited for high latitudes.
For instance, in a species of the alpine snail,
Arianta postembryonic development to maturity
takes five years in a population at 2600 m elevation, while at 1220 m only two years are required
(Thaler, ’99). Also larvae of some salamanders living in mountain ponds overwinter for one or two
years before structural metamorphosis occurs, instead of the normal annual developmental pattern at lower altitudes (Iwasaki and Wakahara,
’99). Interestingly, the molecular metamorphosis
of larval hemoglobin to adult hemoglobin occurred
in first-year overwintering larvae despite their
delayed morphological maturation. In contrast,
differences in rates of development between two
copepod species in a small high (nearly 1800 m)
Austrian lake have been interpreted as niche differentiation (Luger et al., 2000).
But frogs with wide altitude and latitude ranges
in their distributions have a number of life history features (some genotypic, others phenotypic)
that are correlated with increasing stresses in
their environment. In Rana temporaria, for instance, fecundity decreases, but age at maturity,
size at maturity, and longevity apparently all increase with altitude up to 2300 m, near the upper limit of its range in the French Alps (Miaud
et al., ’99). There the short activity season between
late spring thaw and early fall freeze of alpine
lakes and ponds is only three months.
Despite lower than normal temperatures, the
rate of development is increased in this case, and
the time to metamorphosis from tadpole to froglet
shortened (McDiarmid and Altig, ’99). This is similar to, but less extreme than, the greatly speeded
up development and larger size at metamorphosis of some desert anurans dependent on temporary pools of water to complete their life cycle
(mentioned in part 1 and Warburg, ’97). With the
seasonal or sporadic arrival of rain, arousal from
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
dry season dormancy is typically coupled with
rapid egg-laying and accelerated development of
aquatic larval stages in ephemeral pools of water
as also in desert dragonflies (Corbet, ’99).
In any case, the behavioral, physiological and
life cycle changes that allow many animals to
avoid strong stress by copping out are alternatives to the ways in which other species or types
may develop by directly submitting to or opposing the stress. Such direct fitness increasing responses include a whole arsenal of molecular
alterations (Szilágyi and Závodsky, 2000) special
protective proteins, and numerous structural and
regulatory mechanisms discussed elsewhere in
this essay. Remarkable correlations between lowered metabolic rates, increased stress resistance,
and slower aging are considered in part 3.
With regard to increasing complexity as a trend
in evolution, there is much to be said provided
definitions can be agreed upon (Lewin, ’92; McShea, ’96; Sanderson and Hufford, ’96; Pettersson,
’96). Complexity, like size, greatly affects the diverse ways in which animals could function in extreme environments. Abstractly, complexity can be
assayed by considering components in a system:
• How many are there?
• Are they the same or different?
• Are they hierarchically arranged one within
the other or not?
• What are the components’ dynamic interactions, including signaling and information
Practically, both definitions and useful quantitative measures of complexity are elusive, although
a number of formulas have been proposed based
on morphological, physiological, or genetic data.
For instance, one can compare the functional Shannon information content of the genomes to be
evaluated (Adami et al., 2000). The resulting analysis implied that there is a Maxwell Demon-like
force driving evolution toward greater complexity.
With regard to morphology, analyses have been
developed for the degree to which various animals
have evolved all possible structural patterns that
might be available to them (McGee, ’99). For instance, on the basis of 21 hard skeleton structural
criteria (including only three or four elementary
complexity measures) the Burgess Shale fauna
(part 1) was found already in the Cambrian to
have exploited more than 80% of the estimated
available morphospace (Thomas et al., 2000).
Also, analysis of the appearance of novel morphological character states in animal fossil history indicates that novelty appears to have been
exhausted fairly early in the evolution of long
phylogenetic branches (Wagner, 2000). Based on
a large number of trilobite, mollusk, echinoderm,
and vertebrate groups, these data imply that, like
molecular characters, anatomy must have a limited number of possible states. Obviously, this
complicates the identification of homologies and
imposes major problems for deriving phylogenies,
for systematics and for understanding the constraints or possible directionality of the evolutionary processes (discussed later in this article).
Even the simplest prokaryote or animal is, of
course, a highly complex organized entity with
hundreds or thousands of components elaborately
interacting. Animals are open-ended, self-organizing, homeostatic, self-replicating, work-doing, entropy-producing systems. As such, they are far
from thermodynamic equilibrium. They also contain substantial internal information about themselves, as well as about their environmental
niches. These features are inherent in our definition of life (part 1) and so, no doubt, have characterized living things from their beginning.
A number of distinct steps or hierarchies can
be identified in this overall process of complexity
increase (Maynard Smith and Szathmary, ’85;
Pettersson, ’96; Arthur, ’97). As with size, it is
maximum and mean complexity that increases,
while many of the simpler and less complex types
have continued to flourish all along. To begin with,
the evolutionary step from prokaryotes to singlecelled eukaryotes involved major increases not
only in the size of their cells but also in the cells’
complexity, as well as that of their component molecules (Lake and Rivera, ’96) (but see Forterre
and Philippe, ’99, for a different polarity).
Archaea and bacteria share a scarcity of structural variations along with a wealth of metabolic
patterns that are, among other things, important
to biologists for identifying the various kinds of microbes (Postgate, ’94). In contrast, the eukaryotes,
which include both single-celled and many-celled
animals as well as fungi (Cavalier-Smith, ’98b) and
plants, are marked by an exuberance of structural
variations on which their species are mainly based.
Eukaryotes seem far more species-rich than microbes, even though the nature of their respective
species may be quite different. Animals have far
more structural complexity, than bacteria and
archaea. This starts at the molecular level, as mentioned, and continues to the intracellular domain.
A membrane-enclosed nucleus (Lamond and
Earnshaw, ’98) containing a number of linear chromosomes and their genes became an early feature of almost every eukaryotic cell as did a
variety of other cell organelles and programs
(Netzer and Hartl, ’97). Thus nucleus, nucleolus,
the Golgi, the endoplasmic reticulum, the cytoskeleton, meiotic sex, exocytosis, endocytosis and
mitotic cell division, among other things, became
characteristic of eukaryotic cells although the evolutionary steps involved are poorly known (Roger,
’99). Also, the amount of genetic material (nucleic
acids) in each cell greatly increased as prokaryotes evolved into eukaryotes. For instance, the
nematode Coenorhabditis has more than 20 times
as much DNA in each of its cells than does E. coli
in its one cell. The prokaryote–eukaryote transition was a major one and no doubt was also made
in a number of stepwise changes.
Some primitive eukaryotes experienced another
major increase in complexity; they acquired mitochondria, perhaps two billion years ago (Philippe
and Andoutte, ’98). Almost certainly, this happened through symbiosis with a particular kind
of bacterium that came to live within eukaryotic
cells (Margulis, ’96). The resulting organelles,
which still contain some of their own old prokaryotic genes, are the powerhouse for mobilizing energy in most well-developed nucleated cells (Vellai
et al., ’98). Consequently they are central to nearly
all that animals can do, such as grow, move, reproduce, and evolve as extremophiles. The chloroplasts of green plants were also acquired in a
similar fashion from a kind of photosynthetic bluegreen bacteria.
These two major complicating steps in evolution seem crucial for the later history of all multicellular organisms. Note that such giant steps
are quite distinct from the gradual changes typically involved in Darwinian and neo-Darwinian
speciation. They are less like results of modest
gene mutations than they are like those of substantial, less orthodox gains, losses and exchanges
of genetic material not uncommon in microbes
(Doolittle, 2000). Note, too, that an intracellular
nontransient symbiont, or its genes, is presumably a Lamarckian acquired character that can
be inherited, like a mitochondrion.
However, in both animals and plants, symbiosis
provides a pervasive means of cooperation among
different kinds of organisms. In a sense this gives
the resultant superorganism the potential advan-
tage of a second full set of substantially different
genes acquired from another kind of organism.
Obviously, such partnerships may support major
evolutionary changes. Countless examples can be
cited at various levels (e.g., Moran and Telang, ’98;
Nardon, ’99; Ruby and McFall-Ngai, ’99).
For instance, the far-reaching importance of the
symbiosis of reef building corals and other rather
simple invertebrates with intracellular green algae
has apparently evolved independently a number of
times. It has long been studied in some detail
(Maruyama et al., ’98). Much more recently discovered, several kinds of invertebrate extremophiles
living around deepsea geothermal vents (Desbruyères and Segonzac, ’97) have used such foreign
species’ cooperation to tap an abiotic energy source
not usually available to animals (Vetter, ’91.)
This involves the capacity of certain bacteria
to oxidize hydrogen sulfide or methane flowing
out in hot vent or cold seep water and to use the
energy released to reduce carbon dioxide to organic compounds the bacteria need for growth and
reproduction. This light-independent system of
primary production has allowed some deepwater
extremophiles to acquire food, but not oxygen, independently of photosynthesis. That photon-requiring process, which releases oxygen, can
usually occur only in the upper 100 m or so of
the sea because of the rapid absorption and scattering of sunlight penetrating seawater. Yet
chemoautotrophic symbiosis has been reported
deeper than 7300 m in the Japan trench (Fujikura et al., ’99).
Typically the symbiotic types in such benthic
communities are accompanied by a surprisingly
varied fauna of bacterial mat and detritus feeders, scavengers, and predators (Sarrazin et al., ’99).
Present day deepsea vents seem to be ephemeral,
but associated faunas of such habitats may have
existed in Paleozoic and Mesozoic times (Shank et
al., ’99). However, molecular estimates of divergence times imply that recent members of deep
vent faunas are all less than 100 million years old
and for some types less than 20 million years old.
Some deepwater vent animals filter-feed on the
chemotrophic bacteria suspended in the vent water and thus obtain both the energy source as well
as the organic raw materials they need in the
usual way. Other animal types in the vent community have developed an intimate symbiotic relation with closely similar bacteria, which they
cultivate within the cells of their gills or a special
internal organ (Wittenberg and Stein, ’95). Oxygen and reduced sulfur or methane are picked up
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
from the environment by the host animal and
transported by its blood to their internal bacteria. The latter, in turn, like the free-living ones,
can oxidize the reduced substrate to generate ATP.
Part of this energy can be used to drive the host
animal’s metabolism that may also depend on organic compounds produced by the bacteria. One
deepsea hydrothermal vent mussel, a species of
Bathymodiolus living on the Midatlantic Ridge,
apparently relies for its nutrition on both sulfur
and methane oxidizing bacteria that it harbors
internally (Pond et al., ’98).
In the large tube-worm, (adults 2 to 3 m long)
the mouth, anus, and whole digestive tract are
quite absent when this best-studied deep hydrothermal vent animal is mature. Hence, this worm
depends entirely on its in-house sulfur bacteria
for energy as well as for most, if not all, of the
necessary organic raw materials in its own diet.
At the same time, the worm must provide raw
materials needed by its symbionts and eliminate
their waste products (Goffredi et al., ’99). The microorganisms themselves differ from their free-living relatives found in the surrounding seawater.
Hence, living within the worm may be necessary
for their survival, too. Yet, each individual seems
to acquire its symbionts from the surrounding seawater (Nelson and Fisher, 2000).
This mutually required collaboration between
bacteria and the adult worm raises some puzzling
questions. In such intimate interdependence, are
the two quite different organisms involved fused
into one? If so, is that one a distinct new entity,
like the many species of lichen that result from
various pairs of fungi and algae living cooperatively together? A new species so produced would
provide a distinct and rapid evolutionary step toward greater complication. Some lichens, for instance, are extremophiles that serve as the last
animal forage available to herbivores, such as
yaks and musk oxen, at the frontiers of extreme
high altitudes overall and high arctic latitudes.
An Antarctic sponge may provide a remarkable
case where an Antarctic stress has induced a symbiotic diatom, typically cooperative at low latitudes, to become a parasite (Bavestrello et al.,
2000). Presumably because of darkness under surface ice and the prolonged polar winter, the green
alga concerned, instead of producing extracellular polysaccharides that nourish the sponge, themselves take up some of their host’s metabolic
intermediates to prevent the diatom from starving in the dark.
The evolution of many-celled animals immediately
allowed a complexity increase through the subsequent specialization of many cell types along with
a resultant overall increase in body size (Bell
and Mooers, ’97). Multicellularity also permitted the emergence of a division of labor that
organizes cells into tissues, tissues into organs,
and so on, to various levels of social organization (Sendova-Franks and Franks, ’99). Specialization and the division of labor in turn could
increase efficiency (fitness).
Also, the number of different cell types in an
individual animal may provide another, rather
simple but significant, gauge of an animal’s complexity. Like life itself (part 1), cells are remarkable for their basic unity as building blocks and
at the same time for their marked diversity of
structure and function. Counts of cell types clearly
indicate a persistent increase over evolutionary
time (Valentine, ’95). For example, according to
• The earliest many-celled animal may have
had as few as 2 cell types, 600 million years
or more ago
• Early arthropods may have had 55 cell types,
520 million years ago
• Early amphibians may have had 150 cell
types, 320 million years ago
• Humans at present have at least 210 cell
types and probably many more, depending
As animal size and complexity increased during evolution, a growing degree of regulation was
required. The escalating numbers, kinds, and interrelations of the organism’s components had to
be integrated and controlled to assure their vital
survival and reproduction. Cooperation and the
resolution of conflicts were important requirements of evolution at a wide range of levels
(Keller, ’99). Of course, the first cellular life already needed a plasma membrane and genome to
regulate their existence as well as a cybernetic
energy-yielding metabolism to permit work to be
done (Morowitz, ’92). Yet when many-celled eukaryotes evolved, the need for coordination became
much more complex. Both their internal maintenance and regulation as well as their sensitivity
and appropriate responsiveness to the relevant
features of the external environment and its
changes require capable control systems.
At every level, from signal protein molecules to
familial and social behavior, control and regula-
tion gradually became pervasive and interactive
functions. Between molecules and behavior numerous important intermediates have evolved
widely. For instance:
• intercellular communication, as with tight
• multiple cascades of gene action; gene networks; proteomics (Pandey and Mann, 2000)
• rhythms and biological clocks
• sense organs, both internal and external
• endocrines, pheromones and neurohumors
• immune systems
• nervous systems, autonomic and systematic
• locomotor systems
• whole animal behavior
Indeed, so elaborate and effective have such stabilizing mechanisms become in the more complex
metazoans that how leeway can be found for more
than minor evolutionary innovation is a serious
problem (as discussed in part 1). It possibly could
explain the fact that no major new body plan
(phylum) has evolved for hundreds of millions
of years (Erwin, ’99). To test this effectively,
much more needs to be learned about bauplän
genes (Tautz and Schmid, ’99). However, fish
did evolve into tetrapods and tetrapods into
whales, as already documented as post-Cambrian macroevolution.
Self-regulation has to be a defining feature of
cellular life because the living organism is locally
quite different from its nonliving surroundings.
But the organelles and other elements involved
in maintaining a steady state have multiplied and
grown in complexity as metazoans evolved. Basically, organisms must freely acquire essential raw
materials, including energy, from the environment
and discharge waste and excess byproducts externally. Movement of particular ions and molecules both into and out of each cell as well as
selective impermeability to others are central
functions of plasma membranes. These organelles
also respond to both internal and external molecular signals vital to the animal’s survival and
reproduction. These plasma membranes, because
their proper functioning is quite sensitive to
stresses, such as temperature. or pressure, are
usually critically involved in survival in extreme
habitats (Viarengo et al., ’99).
In prokaryotes the obvious importance of the
plasma membrane for these basic functions goes
back to the beginning of cellular life (Morowitz
’92). With the much later evolution of animals,
the numbers and specialized diversity of such cellular organelles has greatly increased the organism’s complexity. Also, specialization of tissues,
organs, germ layers, and functional systems, such
as the circulatory system, have emerged on a more
complex level to stabilize the animal in many
ways. The skin, capillaries, gills, trachea, lungs,
GI tract, and kidneys are typical dynamic elements that function to maintain the internal
steady state of the organism as different from and,
to a considerable extent in terms of fitness, independent of its surroundings.
The modern roots of these regulatory ideas
reach back to Claude Bernard’s notion of the fixity of the milieu interieur, conceived around 1870.
Walter Cannon’s homeostasis and Norman Wiener’s cybernetics were important 20th-century contributions. Well-developed self-regulatory abilities
allow an animal to remain fit by counteracting
and repairing, as by blood clotting, wound healing, and regeneration, the many effects of environmental stresses affecting it, rather than just
passively enduring or behaviorally avoiding them.
Obviously internal stability is particularly critical for extremophiles because their frontier habitats test the limits of their self-regulation as well
as their capacity for effective stress resistance and
Such regulators include, as major types, the genetic, respiratory, circulatory, excretory, neurosensory, neuromuscular, central nervous, endocrine,
and immune systems plus hundreds or thousands
of protein signals and their receptors (Kliewer et
al., ’99; Ray, ’99). Some such stabilizing mechanisms
were present in the earliest animals, but their number, complexity, and scope have greatly increased
during historic evolution. Remember, however, that
both stability and flexibility are inherently antagonistic properties of life at many levels including
the genome (Becskei and Serrano, 2000).
As a rule, the presence of internal body temperature control, cell volume regulation, stabilization of water content, balancing specific ion
distributions (Hazon et al., ’97) and so on, broadens the animal’s tolerance of corresponding otherwise stressful environmental challenges in
changing or new habitats. For example, the lack
of adequate osmoregulatory mechanisms in the
whole phylum of echinoderms would seem to be a
critical factor in this major group’s failure to expand out of their sole marine aquatic habitat.
Even so, various quite distinct classes of this phylum are prominent benthic types down to extreme
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
One striking example of such internal self-regulation relates to buoyancy in mid-water deepsea
animals. Buoyancy is a challenge to all pelagic
organisms larger than microscopic or not small
enough for extended surface areas (parachute effect) and turbulence to keep them in place. Buoyancy for others becomes physiologically more
difficult in deep water because of interactions with
the rise of hydrostatic pressures with increasing
depth. Underwater pressures are high compared
with the atmosphere because water is about 800
times denser than air at sea level. As a result,
every 10 m of water depth adds one atmosphere
of pressure to that in air just above the water
Marine animals, as already mentioned, are subject to pressures ranging from two atmospheres
at 10 m depth to over 1000 atmospheres in the
deepest trench. The average pressure in the ocean
as a whole is about 380 atmospheres. Such pressures do not have grossly obvious biological effects, except on gas-filled cavities such as lungs
or swim bladders (Macdonald, ’75). However, because liquids and solids are relatively incompressible, the pressure at 10,000 m depth would reduce
a given liquid volume, including an animal’s living tissue, only by about 4% of that at the surface. As a result, an animal without any internal
gas-filled cavities will be just slightly reduced in
volume in the deepest seas but certainly not
squashed. High pressure is surely stressful for life,
but its influence, except on gases, seems to lie
mainly at molecular, biochemical, and membrane
levels (Somero, ’93).
Most pelagic animals, whether plankton or nekton, normally occur over rather limited depth
ranges. To remain in place within such water layers, neutral buoyancy can obviously conserve
swimming energy otherwise needed to counteract
gravity (Power, ’89). In between floating and sinking, when an animal’s average density just equals
that of the medium, it will, of course, not rise or
sink spontaneously, but will merely be suspended
at whatever depth it happens to be. At the same
time compression due to depth reduces the volume of unsupported gas-filled parts of the body,
and hence decreases the animal’s buoyancy with
Accordingly, to maintain neutral buoyancy, gas
pressure in such a float must be adjusted to that
in the immediate environment. Regulating buoyancy in this way is a widespread capacity among
midwater animals. Several intriguing mechanisms
have evolved for this purpose. Even organisms as
watery as jellyfish and ctenophores fairly quickly
respond to changes in the density of their medium
by osmotically altering their own density.
Yet most animal tissue, especially something as
heavy as calcified bone, is denser than water. As
a result, neutral and positive overall buoyancy are
possible only if lighter body components, such as
ammonium ions or blubber, act as floats to counterbalance the weight of heavier ones. Internal
spaces containing gas are the least dense floats
that animals can use to buoy themselves up. The
gas-filled swim bladders of many bony fishes and
a pelagic octopus, and the gas-filled spaces of
siphonophores and of the pearly nautilus are striking examples.
A major physical effect of pressure is to compress any gas present to an extent proportional
to its intensity (Boyle’s law). A liter of air at the
sea surface would contract to one milliliter in the
deepest ocean trench. Increasing depth strongly
compresses an animal’s lungs, gas-filled swim
bladder or other air-filled cavity. Such a collapse
would usually lessen or block normal function or
otherwise damage the animal. As a result, the
lungs of diving mammals, such as seals or whales,
are believed to be virtually collapsed and hence
useless for respiratory gas exchange at depths
greater than about 100 m. Even so, many species
regularly dive far deeper than that without apparent stress.
Remarkably, the gas-filled swim bladders of
most bony fishes are fully able to counteract external pressure changes (and hence circumvent
Boyle’s law) by actively secreting and reabsorbing
gas, mostly oxygen, into and out of the bladder’s
closed space. As a result, swim bladder volume,
and hence its lifting force, can remain unchanged
over a range of depths. Essential for maintaining
nearly neutral buoyancy, this remarkable organ
lets the fish stay at a given depth or rise or fall in
the water column without the effort of continual
swimming. Obviously this is valuable asset for
energy-poor deep-water pelagic life.
The swim bladder in some species can secrete
oxygen from the blood into the swim bladder
against a pressure gradient of up to 100 atmospheres (equivalent to a water depth of about 1000
m). Both grenadier fish (macrourids) and rose fish
(the percoid Sebastes) have gas-filled swim bladders, despite living at habitat pressures of more
than 100 atmospheres.
Furthermore, other fishes with gas-filled swim
bladders have been caught living as deep as 7000
m, which would require about 700 atmospheres
in the float to match the outside water pressure.
This remarkable feat depends on some unique
properties of bony fish hemoglobin (Mylvaganam
et al., ’96) and on respiratory gas transfer from
blood to swim bladder in a small specially organized part of the circulatory system. To raise such
great internal pressures, a length of closely parallel arterial and venous capillaries forms an efficient countercurrent exchange mechanism.
This can maintain a strong partial pressure gradient between ordinary arterial gas and the level
needed for transfer to the gas gland. In addition,
the system uses a recurrent multiplier to gradually build up increasingly high levels of respiratory gas in the capillary blood as it approaches
the gas gland. In some species that peak level apparently allows gas pressure in the swim bladder
greatly to exceed a 100-atmosphere limit. The
length of the countercurrent capillary system is a
major factor in that accomplishment. In a number of deepsea fishes its length increases roughly
with their usual habitat depth.
To float upward in the water without having to
swim, the volume of gas in the bladder must be
increased. To do so, the gas gland allows oxygen
at high concentrations in its capillaries to diffuse
into the float. To sink passively downward, a second capillary area in the swim bladder wall becomes active and allows oxygen under pressure
inside to diffuse back into the blood stream and
so deflate the float. To maintain neutral buoyancy
at different depths the pressure of gas in the swim
bladder must be adjusted to match that of the surrounding water, so that the gas float volume does
Some mesopelagic fish, such as moderately
deep-living lantern fishes, have swim bladders
filled or invested with oil or fat instead of with
gas. The oil and fat are less dense than fish blood
and thus serve as floats. At the same time they
are relatively incompressible as liquids and solids and so provide constant buoyancy over a wide
range of depths. In this way the fishes avoid the
need to secrete and reabsorb large volumes of gas
to adjust buoyancy for daily vertical migrations
of several hundred meters.
Another internal control mechanism, physiologically somewhat similar to the teleost swim bladder, but used for excreting highly concentrated
urine by desert rodents, is discussed later in this
essay. Parallels lie in using countercurrent exchangers by the mammalian kidney and the
buildup of a steep osmotic concentration gradient
that limits the maximum urine concentration.
Floats in the form of oil droplets occur in various types of plankton, such as copepods, as well
as in the large livers of pelagic sharks and in more
generally in the body of the Antarctic ice fish and
the lobe-finned fish Latimeria, all of which lack a
swim bladder. Typically deep pelagic animals decrease their density by having reduced and less
mineralized shells, carapaces, and skeletons than
their benthic relatives (Company and Sarda, 2000)
and generally substituting less dense materials
in their structure.
Ordinarily, animals’ genetic systems are primary
overall controls that also program development.
They themselves show remarkable evolutionary
growth in complexity and the scope of its regulation (Osawa, ’95; Nagy, ’98). Yet, the “universal
genetic code” which functions efficiently in nucleic
acid replication and protein synthesis (Freeland
and Hurst, ’98) was already established by the
time of the earliest known bacteria and has remained largely unchanged throughout two or
three billion years of evolution, like a living fossil, as will be discussed in part 3.
Overall, complexity usually is positively correlated with the number of genes but not the total
quantity of DNA in the genome (Graur and Li,
2000). A major complexity increase occurred in the
vertebrates during their origin from earlier chordates and subsequent history. Comparing nematodes, fruit flies, and tunicates with vertebrates
show a large evolutionary increase in estimated
protein-coding nuclear genes from about 15,000
in the nonvertebrates cited to 70,000 or more in
the vertebrates (Martin, ’99). This substantial coding gene increase has been attributed by some biologists largely to genome duplications in the early
history of vertebrates (Greer et al., 2000).
Yet an analysis of the phylogeny of 35 vertebrate gene families, variously active in a wide variety of levels and functions, indicates that this
massive increase in genetic complexity seems to
have occurred mainly through the accumulation
of numerous small-scale piecemeal changes (Martin, ’99). No evidence was found for a few largescale coherent events.
The phylogenetic occurrence of Hox genes also
shows significant evolutionary growth in complexity at least in their numbers and clustering. Available data are, no doubt, still rather sparse but
according to a count in the late 1990s, the bilater-
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
ians may have more than twice as many Hox genes
as the prebilaterians (sponges and cnidarians).
Also, the deuterostomes, including echinoderms,
hemichordates, and vertebrates (Bromham and
Degnan, ’99), may have somewhat more than the
protostomes, including annelids and arthropods
(Valentine et al., ’99). In addition, questions have
been raised about whether Hox gene clusters, apparently characteristic of the triploblasts, are even
present in the diploblastic phyla (Davidson and
Ruvkun, ’99; Holland, ’99).
Among chordates, amphioxus has a single Hox
gene cluster, as do the protostomal arthropods and
the pentaradial deuterostomal echinoderms (Mito
and Endo, 2000). Primitive jawless fishes, such
as lampreys, have three. Tetrapod vertebrates,
from amphibians to mammals, including humans,
have four such clusters, each on a different chromosome, suggesting to a number of scientists a
sort of octoploidy. Yet multiple clusters probably
arose by individual gene duplication or by duplication of chromosome sections containing the
genes (Wolpert et al., ’98; Martin, ’99).
Curiously, teleost fishes have various numbers
of Hox clusters with eight presumed to be the basic number, implying further evolutionary duplications (Longhurst and Joss, ’99). Possibly the
high number of teleost Hox clusters may be causally related to their substantially greater number
of species than the other vertebrate classes (Gregory and Hebert, ’99). Since the amphibians arose
from early jawed fishes, the number of Hox clusters in lungfish or coelacanths seems crucial to
understanding the core evolutionary pattern. In
the Australian lungfish, Neoceratodus Longhurst
and Ross identified the same four clusters, (A, B,
C and D) characteristic of tetrapods.
This is consistent with the notion that the further duplications and later losses of Hox gene clusters in teleosts took place after their split with
the lobe-finned fishes. Yet the exuberant array of
teleost Hox genes seems surprising in view of the
apparent lack of much differential growth along
their anteroposterior axis. Does this mean that
the high numbers indicate that axial differentiation of teleosts is more complicated than it seems
in this species-rich group? Or does it mean that
teleost Hox genes have at least in part evolved to
control other functions than those usually attributed to them?
Another important aspect of the increasing complexity in animals’ evolution emerged in the so-
cial relations that appeared among individuals of
a species, although its scope was long obscured
by the traditional isolation of scientific disciplines,
(Wilson, ’75; Gadagkar, ’97). Breeding groups, such
as families or swarms, obviously have strong, but
often overlooked, effects on population biology
(Sugg et al., ’96). Cooperation among individuals
within a species in families, schools, flocks, herds,
prides, gaggles, and eventually in societies, gave
rise to unprecedented new evolutionary prospects
(Parrish and Hamner, ’97).
This, of course, had already achieved remarkable
levels in the early emergence of colonial invertebrates such as corals, siphonophores, bryozoans,
and salps. Such ideal intraspecies cooperation
seems possible only in quite simple organisms. Yet
strongly social insects have evolved repeatedly,
mainly in bees, wasps, and ants, which are all in
the Order Hymenoptera (Wilson, ’75). In such insect groups individual survival depends closely on
their colony’s survival. The termites are the only
other kind of strongly social insects. However,
some beetles and other insects as well as certain
spiders have less strongly developed social behavior (Choe and Crespi, ’97).
Colonial animals and strong insect social systems
(Seeley, ’95) seem to form a sort of superorganism.
Birds and mammals have also evolved various degrees of sociality, but they are less far-reaching
than in the invertebrates mentioned. Individual
survival in the warm-blooded vertebrates is less
closely bound to their group’s survival. Finally
in humans and other primates, social organization, including families (Davis and Daly, ’97),
has increased strongly again. Its evolution has
led to the development of learning, externally
stored memory, and more effective communication systems.
To date, these culminate in language, stored
records, problem solving, exploration, self-awareness, and human culture in general, including
myth, ritual, art and science (Maynard Smith, ’99).
Such developments have evolved much faster than
traditional genetics-based evolution, and the resulting changes can be spread widely with remarkable speed (Bonner, ’88). Even so, phylogenetic
methods can be effectively applied to some of these
topics such as the evolution of human languages
(Gray and Jordan, 2000). Certainly behavior and
social structure can be crucial for animal fitness
in extreme environments as in desert ants, emperor penguins, and musk oxen, not to mention
humans. However, the extension by sociobiologists, of genetic determinism to the evolution of
human behavior has been bitterly controversial
(Sterelny and Griffiths, ’99).
Extremophiles and complexity
As in the case of size, animal extremophiles run
a wide gamut of complexity from sponges and
cnidarians to numerous kinds of insects, crustaceans, cephalopods, and vertebrates in various farfrom-average environments. Despite seemingly
endless capacities for sustained evolutionary
trends, even the most evolvable organisms may
be unable to change in certain directions, like echinoderms confronted with fresh water. Increasing
complexity may also conflict with some other evolutionary tendency that acts to constrain it. This
could have been the case for ammonoids, extinct
cephalopods well known from extensive fossils, including 588 genera that lived during a period of
more than 140 million years in the Paleozoic and
early Mesozoic (Saunders et al., ’99).
Ammonoid history was marked by a strong
overarching trend for increasing complexity in the
sutures joining the septa in the animals’ chambered shell to its outer wall. Yet this well-documented and taxonomically important anatomical
detail reversed its basic trend and became more
simple during three extinction episodes in the
group’s overall history. Apparently the factors increasing extinction rates during these severely
stressful episodes were selectively more destructive at shallower depths than in deep water.
Presumably, ammonoids began moving into the
depths to avoid this threat of destruction. But this
increased the hydrostatic pressure on the spiral
chambered shell, gas-filled for buoyancy, rather
like that of the living, but not closely related,
pearly nautilus (Wells, ’99). As a result, simple
septal sutures became more effective than complex ones, because the latter are more vulnerable
to collapse under high pressures. Avoiding extinction in shallow water would seem to have won
out in this conflict, accounting for the observed
decrease in suture complexity.
On a much larger scale the insects seem to have
been blocked somehow from subsurface pelagic
marine waters and the deep sea broadly, as part
of their striking near-absence from most of the
marine environment. Even so, a number of aquatic
types, such as water bugs and whirligig beetles,
are familiar in fresh water, as are water striders.
This last group also has open ocean surface species and many coastal marine ones. Insects, such
as springtails, mosquitoes, caddis flies, and
midges, are common in the upper tidal zone, man-
grove swamps, salt marshes, and even saline
lakes. Hence salinity per se would not seem to be
Yet most of the great volume of the world ocean
has no insect inhabitants whatever, shallow or
deep. Explanations are speculative, but perhaps
commitment to a tiny air-filled tracheal respiratory system, limiting them to shallow depths
(Maddrell, ’98), was a factor in this strange failure of the most speciose animal taxon to invade
the world’s largest living space, including the deep
sea. Perhaps, too, insects’ rather late evolutionary
arrival was influential, several hundred million
years after the ocean had been richly colonized by
numerous other animal groups, such as polychaetes, mollusks, crustaceans. echinoderms and
No doubt, the explanations and consequences are
different for amphibians, which also are nonmarine except marginally for one or two frog species.
In contrast to insects and amphibians, cephalopods
and echinoderms are all obligatory marine animals
and hence precluded from freshwater and terrestrial extreme habitats. The special eggs of reptiles,
birds, and mammals (collectively, the amniotes)
have, among other things, allowed them to invade
arid terrestrial frontiers. But rather unlikely types
without such eggs including a few crabs, frogs, and
toads also live successfully in various severe
deserts (part 1 and above), by undergoing periodic
sheltered inactive phases having strongly reduced
metabolic rates during dry periods that may last
8 to 10 months per year or longer. One Australian
species of desert frog in the genus Cyclorana can
survive as many as five years of dormancy (Warburg, ’97). Also exceptional are some high-latitude
amphibians and reptiles that in winter undergo a
prolonged inert subzero diapause, and a few that
are even freeze-tolerant.
Despite being ectotherms, reptiles may also succeed well in extremely hot desert environments.
Typically they tolerate considerably greater heat
stress-induced changes in their internal environment that would be lethal to birds and mammals
(Bradshaw, ’97). Also reptiles’ metabolic rates are
so low that their energy food requirements are
significantly less than those of endotherms. Their
water needs are also modest because, like insects
and birds, they typically excrete nitrogen as uric
acid that, being relatively nontoxic to process, can
strongly spare urinary water loss.
Reptiles’ dry and water-impermeable skin reduces water loss by evaporation as does the waxed
chitinous exoskeleton of desert insects. Recall, too,
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
that some reptiles and birds have salt glands that
can reduce salt loads with out sacrificing much
water (Schmidt-Nielsen ’97). Yet mammals, particularly many small desert rodents, excrete nitrogen as urea, which usually requires considerable
water loss. Even so, mammals can conserve water because of their kidneys’ unique ability to produce strongly hypertonic urine, as discussed
Such generalizations from comparative data reinforce the idea that stressful environments can
be responded to in many different ways. Even animal types that have not yet evolved many complex physiological self-regulatory mechanisms, can
commonly survive on the environmental frontiers.
What some ecologists refer to as trade-offs can be
made at a number of alternative levels and by
various means. Human beings seem an obvious
exception since they can penetrate and at least
briefly survive the most extreme habitats in the
world, such as the highest mountains (West, ’98).
Yet humans’ sustained survival in extreme habitats is only made possible by behavioral or social
means such as bringing oxygen, water, and food
into extreme environments, using insulated clothing, and setting up shelter, along with heating or
air conditioning and protecting themselves from
high or low pressures, as needed (Marriott and
Carlson, ’96). All of this survival gear can be marshaled nearly instantaneously compared with the
thousands or millions of years needed for any comparable Neo-Darwinian animal adaptations. Such
behavioral, social, and cultural assets for survival
are surely supported by remarkably complex
A naked human being, without clothes, water,
food, or other support, is actually far from an effective extremophile. Such an individual dies
rather quickly on any of our four major habitat
frontiers. Its metabolism, its hardiness, and most
of its means of self-replication fail in the most extreme habitats. Space travel, which requires longterm self-maintenance without outside access
aside from take-off stocks of fuel, food, and so on,
plus solar power in flight, brings the difficulty of
practical solutions to such problems into sharp focus. Despite much research, feasible bioengineering and social protocols remain to be worked out
for viable interplanetary travel and the colonization of the moon or other potential extraglobal
home sites. All of which obviously require sustained survival under extreme, usually strictly lethal, conditions. The whole prospect faced by
various space agencies of terraforming severely
hostile extraterrestrial environments is indeed
challenging (Fogg, ’95).
Increasing diversity in the kinds of organisms
on earth has, like increasing size and complexity,
been a continuing powerful trend in their long evolutionary history (Magurren and May, ’99). At the
level of total species, genera or families, maximum
and mean diversity of animals have greatly increased overall from the end of the Precambrian
to recent times. But like life’s growth in size and
complexity, its diversity changes are many-layered
and have a number of special features. Morphology, physiology, genetic, nervous and endocrine controls, behavior, lifestyle, as well as family, group
and population patterns—these are all diverse elements already mentioned under complexity. Here
animal taxa at various levels from species to the
animal kingdom are the central topic.
In any case, diversity, like complexity, presents
problems of definition and measurement beyond
simple counts of individual phenotypes or taxa.
Diversity, also, since the late 20th century, has
taken on urgent components of public policy, ethics, and natural resource ecology and economics
because of its central place in the conservation
debates (Abe et al., ’97). Even at the headcount
level, diversity’s overall rate of increase (discussed
in more detail below) has varied markedly over
time and has even been repeatedly reversed during major periods of extinction. (Mass extinctions
and rates of evolution, including bursts of species
and higher taxa will be discussed in part 3).
In addition, the taxonomic diversity of animals
and its changes with evolution is quite unevenly
spread among the various phyla and their subgroups. One phylum, for instance, has only a
single species whereas another has more than a
million. Why, for instance, have some phyla, such
as the priapulid and poganophoran worms, survived but remained undiversified and minor over
some hundreds of millions of years? Both the
worm phyla mentioned have only one closely-conserved body type and only a few, solely marine,
species. Among more than 30 generally recognized
animal phyla, nine have only 100 species or fewer
(Margulis and Schwartz, l998). The fossil record
for a number of these groups implies that never
in their history have they diversified notably either at the species level or higher.
In contrast to such weakly diversified phyla,
quite a number, particularly the arthropods, have
gone to the other extreme, with estimated species
numbers reaching into the millions. Identified arthropod species, even excluding insects, number
well over 500,000; mollusks number over 110,000,
and the chordates, which include the vertebrates,
45,000. Among the diversity-rich monophyletic
arthropods, the probably paraphyletic crustaceans,
for instance, can be subdivided into many groups
to separate the shrimps, krill, lobsters, crabs, barnacles, wood lice (isopods), beach fleas, and so on
(Spears and Abele, ’98). Each of these subgroups
can be separated into smaller ones until the species level is reached. Current opinions suggest that
malacostracan crustaceans, at something like an
isopod level, may have given rise to insects (Abzhanov and Kaufman, 2000) and independently to
other arthropod types (Brusca, 2000).
Other phyla, highly prolific in an evolutionary
sense, include nematode worms, annelid worms,
mollusks, echinoderms, and chordates (Margulis
and Schwartz, ’98). Of these, the species numbers cited for the nematode roundworms range
from about 20,000 currently identified to many
times that number still unknown (Malakhov, ’94).
About one-third of these are parasites in plants
and animals; the rest are free-living in a wide
range of environments, including various stressful frontiers.
Species are widely used as basic units in Linnaean systematics, population genetics, biodiversity, ecology, evolution, and comparative
biology. Yet some researchers consider species to
be quite an arbitrary term with little “real” meaning and in need of replacement (Bachmann, 2000).
Darwin himself seems to have been rather ambiguous on this point (Depew and Weber, ’95).
Even now, quantified species discrimination has
rarely been practiced (Wiens and Servedio, ’99).
Yet for the classic Linnaean systematic biologist
and many others, the species is taken to be a
stable and distinct kind of animal (de Quieroz,
’98), quite generally used as a measure of biodiversity (Claridge et al., ’97).
But even among biologists who accept species
as key evolutionary entities, the concept of species has been defined variously (Harrison, ’98).
Systematists, paleontologists, ecologists, and
population biologists all have rather distinct definitions. Consequently, several different meanings
of “species” must be acknowledged when discussing evolution (Gosling, ’94). Definitions often become even more troublesome when dealing with
higher taxa in classification, such as genera, families, orders, and so on. These groups tend to be
less clearly definable than species and vary con-
siderably in character for different major animal
groups, typically studied independently by specialists (Nielsen, ’95; Cavalier-Smith, ’98b).
If we refer here to speciation as grist for evolution, we are referring to the structural and functional differences between various kinds of rather
similar evolving animals, whether or not reproductive isolation has been, or even could be, tested.
Yet we do know that persistent changes in animal populations, heritable over generations, require some reproductive isolation, even if it may
be partial or temporary. In fact the genetics of species formation may be taken to be the genetics of
reproductive isolation (Coyne and Orr, ’98). Some
sort of genetically controlled traits that at least
partially block gene exchange must be involved
in such isolation.
The factors that promote speciation and hence
the species richness of particular animal groups
are still uncertain (Price, ’96). Ecologically, species diversity and productivity are correlated, but
explanations are still needed (Waide et al., ’99).
Even the genetics of species formation remains
in a rather elementary state (Via and Hawthorne,
’98; Gavrilets, ’99). Underlying this ambivalence
are the more than 20 different concepts of species
(Mayden, ’97) and widely varied estimates of species importance, mentioned later in this article.
Also complicating discussion are the quite distinct
ways of identifying taxonomic units used by the
Linnaean systematists (most of the many species
counts and other references to species in this essay fall into this category) and the phylogenetic
cladists (Nixon and Wheeler, ’90).
Later discussions of rates and mechanisms of
evolution will expand on this topic (part 3). Particularly interesting is the question of whether
stresses themselves can accelerate or positively
affect the direction of evolution (Hoffmann and
Hercus, 2000). Various earlier hypotheses about
speciation remain mostly untested. Comparison
of evolutionary sister groups, identified by cladistic analyses, may offer a promising future
approach to this problem (Barraclough et al.,
’98). For instance, data on plant-feeding by insects and on sexual dimorphism in birds show
that rates of genetic change and speciation rates
can be correlated.
As might be expected, there is also considerable
correlation between species diversity and adaptive versatility among various animal groups.
However, some exceptions are clear. For instance,
tardigrades are known from the early Cambrian
Chengjiang fauna (Brusca, 2000, but their cur-
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
rent morphological and species diversity are minor, with only 600 or so recent species (Kinchin,
2000). Yet they seem ecologically remarkably versatile, occurring among other places in the deep
sea, in hot springs, in highly stressful Antarctic
desert valleys, and on glaciers up to 5600 m in
Nepal (Thaler, ’99).
As discussed in part 1 of this essay, animal diversity at the phylum level has usually been
thought to have already reached its peak by the
early or mid-Cambrian and then apparently declined (Mayr, ’91). If so, phyla seem to be exceptions to the usual trend toward continual
increases in taxon diversity over time. Certainly
all of the major animal phyla, except apparently
flatworms, have been known as fossils dating
back to the late Proterozoic or early Paleozoic
(Valentine et al., ’99).
Certain of the minor sparsely diversified
phyla, such as the comb jellies (Ctenophora) and
priapulid worms, neither expanded from the sea
into other major habitats nor evolved widely even
into various sub types of marine environments.
Instead they have remained, since the Cambrian
at least, as a few species of marine “conservative,”
yet distinctive, major types. More recently proposed phyla, of which there are a few, seem to be
increasingly minute, poorly known types not yet
found as fossils (Brusca, 2000).
The most versatile and diverse present day phylum, the arthropods, apparently has been so for
at least 600 million years (Brusca, 2000). As a
result, much of the variation needed for animal
speciation, as wells as their evolvability itself, was
concentrated in one class of the most diverse phylum in earth’s history. That class of arthropods,
the insects, far outnumbers all other arthropods
together and constitutes about half the total
known number of species of recent animals (Kristensen, ’98). Such remarkable diversity has stimulated considerable speculation about how it came
to be (Fortey and Thomas, ’98).
The beetles, as mentioned earlier, include about
300,000 known species, although they form only
one order among the 25 to 30 orders of insects.
The fruit fly family to which Drosophila belongs
has about 3,200 named species. In ancient times,
insect rates of evolution apparently were not exceptionally high, but as discussed later in this article, their extinction rates were quite low. Insects’
early evolutionary success, no doubt, was also tied
to the then-burgeoning evolution of early land
plants, including seed-bearing conifers and cycads
(Kenrick and Crane, ’97).
Although usually associated particularly with
flowering plants, insects’ great diversification began well before those green plants evolved (Labandeira and Sepkoski, ’93). Instead, the typically
inverse relation between size and diversity, just
cited, may have been involved in insect species
profligacy. In turn, the insect tracheal respiratory
system, unsupported by effective circulatory transport, must impose a likely restraint on size, as well
as on insects’ severely restricted marine occurrence, mentioned below. However, long after insect origins, plant-eating beetles, co-evolving with
flowering plants, may account for perhaps 100,000
new species in that insect group just during the
Cretaceous and early Tertiary (Farrell, ’98).
More broadly, a number of insect features seem
crucial for their great evolutionary success on land
including extensive extreme habitats (Gullan and
• Acquisition of a waterproof external skeleton
made of chitin. Especially because of their
small (but not minute) size, water conservation is critical for insects to function well in
any aerial-terrestrial environment not saturated with water vapor.
• A largely impervious exoskeleton, which
helps insects survive in extreme habitats
with water shortages.
• The special excretory system of insects, which
also seems particularly relevant to terrestrial
life and water conservation. Insect urine, during water shortage, can be eliminated as a
paste or even a dry powder, mostly of uric acid.
As just mentioned, this provides a clear fitness gain by minimizing water needs in
deserts or high on mountains.
• Development of an effective respiratory system not dependent on a heart and circulation. It functions mainly by passive diffusion
of gaseous oxygen and carbon dioxide within
a fine branching tree-like structure of tiny
air-filled tubules that reach all active tissues.
This so-called tracheal system works very
well on a small scale but, presumably because gas diffusion is only fast enough over
very short distances, has kept insects quite
small. Some fossil dragonflies had wingspreads of half a meter, but we know few details about their flight or respiration (Wooton
et al., ’98). Perhaps a sufficiently sustained
pulse of higher than usual percent of oxygen
in the atmosphere and greater overall air
densities may have permitted such giants to
flourish for a limited period only (Dudley, ’98;
Harrison and Lighton, ’98).
• Tracheal respiration is also limited in its effectiveness to breathing in air because respiratory gases diffuse so slowly in water that
trachea would not work if water-filled. Aquatic
insects mostly breathe at the surface or take
bubbles of air down with them, features that
limit them to shallow water depths, as cited.
• The remarkably prolific evolution of modern
insects has also been attributed to flight that
allows them quickly to exploit varied and extended habitats despite the animals’ small size.
Another insect feature particularly effective in
periodically or occasionally extreme environments
is their widespread ability to enter a resistant dormant state. During extreme stress, most ectotherm survival requires, as in many insects, being
inactive and dormant as well as being freeze-tolerant or freeze-resistant in extreme cold at high
elevations and latitudes (Fogg, ’98; Sømme, ’99).
Dormancy in ectotherms typically reduces metabolism markedly and may strongly heighten resistance to stresses. The dormancy concerned may
occur in various phases of the life cycle from egg,
to larva, to adult. It permits survival during the
more or less temporary absence of otherwise viable temperatures, supplies of water, oxygen, or
food. During seasonal stresses at moderate to high
global latitudes and elevations, dormancy is crucial (Chapin and Körner, ’95; also mentioned in
part 1 and later in this article).
Clearly, some diversity of both animals and
available environments is basic to the evolution
of extremophiles. Habitat diversity for the large
taxa also varies widely, as already mentioned.
Consider, for example, the six animal phyla (following Nielsen’s ’95 analysis) that have more than
10,000 living species. Among these only the flatworms (Phylum Platyhelminthes) have scarcely
moved into extreme environments except as parasites of extremophile hosts or minute soil types.
The other five species-rich groups, arthropods,
mollusks, vertebrates, nematodes, and to a lesser
extent, annelids (with about 15,000 species), have
shown substantial evolutionary versatility. Yet the
starfish, sea urchins, and other echinoderms remained solely in their marine environment. Others, such as arachnids, just moved from the sea onto
the land (Dunlop and Webster, ’99) whereas mollusks stayed substantially in the sea, where they
are mainly deployed today. But, except for certain
taxa such as cephalopods, some mollusks also radiated into freshwater as well as onto the land.
All echinoderm and most mollusk extremophiles
are restricted to the deep seas. One clade of the
Aplacophora, a poorly known class of cylindrical,
rather strange spiny worm-like mollusks, has
mostly been found at depths between 40 m to
nearly 6,000 m, with most of the species diversity
below 200 m. At least 250 species have been described and perhaps 1,000 seem likely to be found
(Scheltema, 2000). Could they be candidates for
an evolutionary species burst? Mollusks are also
extraordinary because they evolved the octopuses
and squids, which are among the most active and
complex-behaving animal types with numerous
deepsea species (Pörtner et al., ’94)
Cephalopods are rather unexpected relatives of
slugs, oysters, and abalones. Many species in this
class are prominent bathypelagic or benthopelagic
carnivores. For instance, the giant squid Architeuthis is the largest invertebrate, a fast-swimming,
powerful deepsea predator, a proper model for sailors’ tales of sea monsters (Lordan et al., ’98). The
prominence of squid beaks in the stomachs of deep
diving whales proves that deepwater squids are a
major element in some cetacean diets. Above all,
the nematodes and arthropods are outstanding as
versatile phyla that have strong representation in
deserts, high latitudes, high altitudes, and the deep
seas as well as plant and animal parasites. The
vertebrates are close behind them but have not
evolved true parasites.
Birds and mammals, particularly, with their
well-developed thermoregulation and high-level
energy budgets can in many cases manage to keep
vigorously active under extreme stresses. Transhimalayan flight paths and the deep diving programs of certain large fishes and various large
marine mammals are discussed elsewhere. Aside
from a few minor taxa, many marine species in
nearly all major animal phyla have effectively invaded the deep sea. This did not require a change
in medium but only involved spreading into a
huge, but likely stressful, habitat always adjacent
to more benign inshore and near-surface pelagic
Between the “push” of phenotypic diversity, population pressure or chance and the “pull” of environmental opportunity (Schluter, ’98) lies the
keystone of Darwin’s theory of evolution: natural
selection (Futuyma, ’98). Natural selections acts on
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
phenotypic variations in living organisms in the
field so that some unfit variants are eliminated and
other fit variants persist (Price and Yeh, ’99). Since
the earliest stressful and tentative biotic times
more than two billion years ago, selection of some
sort must typically have favored the persistence of
accidental complex quasi-stable cooperative selfreplicating systems that were potentially alive and
the disappearance of those that were not.
An additional feature of natural selection is that
it acts ultimately on the reproductive rate of individual animals (Bell, ’97). That rate may depend
indirectly on various organisms’ size, morphology,
and physiology. Yet these in turn are basically determined not only by the animals’ genes but also
in complex ways by many other influences, such
as environmental signals (Van Buskirk et al., ’97;
Tollrian and Harvell, ’99), diet, acclimation, parasites, and learning. These apparently casual influences sometimes can affect the phenotype
dramatically (Huey and Berrigan, ’96) in ways
that may significantly affect natural selection
(Greene, ’99). Also, an organism’s everyday evolutionary path may be partly shaped, sometimes decisively by chance factors (discussed in part 3).
The basic course of Darwinian evolution is set
by natural selection interacting with inherited
phenotypic variations in a given population responding to environmental change of whatever
sort (Rollo, ’94). Sexual selection (Cunningham
and Birkhead, ’98) is a variant of ecological selection (Schluter, ’98) and may strongly facilitate speciation (Boake, 2000), as suggested to account for
the cichlid fish species burst and retreat in Lake
Victoria (upcoming in part 3).
Depending on circumstances, selection can stabilize a population (thus greatly slowing or blocking evolutionary changes in phenotype), or it can
move the population toward a state of higher fitness, or it can divide the two “tails” of a population’s
variation, such as the smallest and the largest individuals, into two new populations (Bell, ’97). On
the other hand, natural selection can also function
in extreme environments to move a population in
an evolutionary direction opposite to that expected.
Instead of conserving energy in response to cold
stress, for example, an endothermic taxon’s capacity for metabolic work and thermogenesis may be
increased in a way that would allow it to thrive
at higher latitudes or altitudes, provided that adequate energy food is available. In such a case,
exploitation rather than conservation is the word.
An interesting example has been reported for deer
mice that live at about 3800 m on White Moun-
tain in California (Hayes, ’89). In addition to the
increasing cold stress with altitude, these small
rodents, like other montane mammals, face reduced maximum metabolic output because of the
decrease of oxygen partial pressure with elevation (West, ’98).
To study this, the maximum aerobic capacity in
local wild deer mice was measured in a mountain
laboratory’s refrigerated wind tunnel for a substantial number of individuals collected by trapping on three separate occasions. The animals
were first measured, marked, and then released
to their natural environment. Two months later
the survivors were recaptured and tested again
for their maximum aerobic capacity. Strong directional selection for an increase in this factor was
statistically significant in the largest of the three
data sets. This appears to be an important instance of Darwinian selection for an extremophile.
One gene has been associated with maximum
aerobic capacity in deer mice, but other components of the overall control mechanism remain to
During life’s long history, natural selection has
evoked substantial amounts of apparently parallel or convergent evolution (will be discussed under “chance” in part 3). These are widely evident
at the level of orders and smaller taxa, at the
nucleic acid level within organisms (cited previously), as well as at the enzyme (Kreitman and
Akashi, ’95) and perhaps gene levels of evolution
(Goodrich et al., ’97; see also part 1). For instance,
rapid parallel evolution through natural selection
has been reported at the speciation level for stickleback fishes (Rundle et al., 2000) and at the level
of geographic clines for a species of Drosophila
(Huey et al., 2000). In the latter case, the genetic
mechanism of the parallel geographic size increase with latitude is different in the two fly
clines, one from Europe and the other from North
An extremophile example within what is usually considered a single species occurs in the evolution of some cave animals. In the Mexican
characin fish Astyanax fasciatus, both subterranean
cave-living and nontroglobitic surface populations
of the same freshwater species are well known.
Cave forms of this fish occur in eight groups of limestone caves. Some of these populations, estimated
to have been cave inhabitants for perhaps a million years, are as adults pale, pigmentless, and
blind, with much reduced eyes and visual centers
in the brain. Imperfect eyes form in the embryo,
but apoptosis of the early lenses apparently sig-
nals regression in much of the rest already present
(Yamamoto and Jeffery, 2000).
A different cave population, believed to have
been trogloditic for a much shorter time, has moderately developed eyes, intermediate between normal above-ground fish and the severely reduced
blind types. Such variable structural features of
these cave fishes, as well as certain behavioral
traits, are apparently all under genetic control.
As a result, hybrids between the above ground
Astyanax and its various cave-adapted populations
have provided remarkable insight into the genetics and fitness-related evolution of complex organs
such as eyes (also cited in part 1; vision and eyes
in the deep sea will be discussed in part 3).
Less well known is the amphipod crustacean
Gammarus minus that has repeatedly and independently adjusted to cave life after invading different cave systems in various limestone (karst)
basins in middle Atlantic and middle central
United States (Culver et al., ’95). Compared with
surface-dwelling populations, each such invasion
resulted in resident populations with compound
eyes having a reduced number of facets and underlying ommatidia, a shift in body pigment from
brown to blue or white, as well as lengthening of
the antennae and other paired appendages.
At the same time, the optic centers in the central nervous system were reduced in cave populations and the antennal centers enlarged,
presumably representing a shift in sensory
modes correlated with cave darkness. All of these
features showed heritable variation in general
and directional changes in the cave habitat. Curiously, extreme changes of this sort, such as reduction in ommatidial number from 25–30 to
3–4, were observed only in a few isolated large
caves in West Virginia and Virginia. In addition,
evidence for the reversal of these cave correlations was found in populations of Gammarus living in collapsed cave areas open to daylight. This
suggested that the genes involved were still
present in the dark cave forms and that their
expression could be resumed over generations
with exposure to day light.
In contrast to such labile features, any random
tendency to change or displace stable features
that are essential for animal survival must be
curbed either by selection or by the organism’s
self regulatory systems (Bell, ’97). Presumably
this constraint underlies the classic conservatism
of mutation rates in particular proteins and
nucleic acids (and their component amino acids
and nucleic acids), a conservatism correlated with
their inherent functional importance in the organism. Alternatively, to retain their essential
presence, their genetic control may be shifted to
new or alternative mechanisms, as may be the
case for the stasis of some living fossils, discussed
in the forthcoming part 3. While somewhat reminiscent of the old, rather fusty, idea of archetypes,
these complex-system characteristics actually can
be described in more contemporary and plausible
terms (Van de Vijver et al., ’98), including developmental and evolutionary modules.
The characters on which natural selection may
act range from parts of molecules to large-scale
components of anatomy, development, life history
and behavior. Despite some hubris on the part of
a few molecular biologists and genetic engineers,
genes, as fragments of information, ordinarily do
not surface directly in the phenotype. Hence individual organisms that carry the genetic information and, no doubt, higher levels, such as social
groups that interact cooperatively or antagonistically between themselves and the environment,
comprise the units selected (Reeve and Keller, ’99;
Weiss and Fullerton, 2000).
Darwin originally conceived of natural selection
largely as related to highly productive ecosystems,
such as tropical lowlands, where crowded inhabitants must compete, tooth and claw, within and
among species for essential available resources. In
such abiotically nonstressful resource-rich environments, populations seem to operate well below the
carrying capacity of the environment. Accordingly,
positive evolutionary factors tend selectively to favor fecundity and population growth. Yet natural
selection reins in organisms’ tendency toward reproductive exuberance and, as far as possible,
molds their inherent diversity to match available
habitats including extreme frontiers.
Natural selection may be considered as the complex outcome of an increase in an animal’s or a
population’s fitness at one or more levels. Eventually it results in greater reproductive power, if
that is not already limited by some other factor,
such as aging, that reduces reproductive capacity and ultimately individual survival. In addition, natural selection, under severe stress, may
reduce the number of healthy offspring of less robust animals so far that their death or extinction
results (Gems, 2000). But for any given organism, the emergent opportunity of moving into the
potential living space of a new uninhabited extreme environment is reduced by the organism’s
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
vulnerability to the stresses there and any inability to respond to them effectively.
In the evolution of extremophiles, natural
selection’s evolutionary force acts on populations
so that they can survive and reproduce under
various abiotic conditions that would have previously been stressful or even lethal. Yet to reach a
sustainable steady state, this positive force for
growth and expansion must be contained by an
opposing force that consistently eliminates the
less fit and the unfit.
Darwin himself recalled that a serendipitous
first reading of Malthus’s essay on population
(1798) provoked him to think about what forces
must prevent animals from propagating until they
are literally piled up one on top of the other
(Depew and Weber, ’95). He realized that offspring
typically differ from their parents and siblings
from one another. As a result, some are fitter to
survive than others, where fitness is ultimately
measured by the number of viable offspring produced by the organism or taxon (Kozlowski, ’99).
Selection in extreme environments
Survival of the fittest in Darwin’s sense may
not be directly relevant to life in extreme environments. Yet the evolution of fitness in such habitats is exactly the point of departure of this essay.
Even so, it has been argued that nonsurvival of
the unfit, including their failure to reproduce and
their ultimate extinction, is a more appropriate
maxim than the classic one. In either case the
survival to maturity of one male and one female
for sexually reproducing animals and at least two
viable offspring, one of each sex is the theoretically lowest limit for continuity. More practically,
a minimum population number of reproductive individuals of both sexes is needed for likely sustainability and evolution.
By definition, such evolution moves potential
extremophiles away from the middle-of-the-road,
sometimes to become specialists, but more often
to become hardy generalists (Parsons, ’99). For
instance, extremophiles flourish with reduced energy-food supply in general, with increasing hydrostatic pressure at increased water depth, with
decreased temperatures at higher latitudes and
altitudes, with increased temperatures near hydrothermal vents and so on, as already discussed.
But also the numbers of individuals and taxon
diversity decrease to an ultimate limit where no
taxa or individual animals survives more than
briefly, let alone reproduces. Presumably the limit
of survival and zero fecundity are usually imposed
by starvation-inducing low primary productivity
as well as by extremely severe abiotic factors, such
as anoxia at extreme elevations, acting directly
on the animal.
In contrast to mild highly productive habitats,
extreme environments near the Malthusian limit
have resources that are reduced to a minimum.
As a result competition may be largely secondary
to mere survival as in the interior of severe continental deserts or the Greenland and Antarctic ice
caps (Labropoulou and Kostikas, ’99; Parsons, ’99).
The polar bear on the Arctic ice sheet and the
Bactrian camel in central Asian deserts provide
extreme examples in which coexistence rather
than competition would seem to mark the sharply
limited animal communities (Takeshi, ’99).
Also, the high Andean habitats of two camelids,
the guanaco and the vicuña, overlap between 3000
m to over 4000 m without apparent interspecies
competition (Lucherini et al., 2000). Communities
of the birds and fish of northern Canadian lakes
seem to have evolved with minor predation and
competition between them (Paszkowski and Tonn,
2000). In any case, evolution’s usually assumed
central dependence largely on biotic factors, such
as competition and predation, may have been exaggerated by Darwin and by many ecologists after him (Hengeveld and Walter, ’99).
Note that in some less stringent frontiers having clusters of closely related animal types,
interspecies competition for scarce resources no
doubt does occur more widely. Examples are evident among certain fishes (Haedrich, ’97) and
crustaceans (Company et al., 2000) in the upper
1,000 to 2,000 m of the deep sea as well as among
species clusters of desert darkling beetles (Sømme,
’95), desert lizards (especially in Australia; Vitt
and Pianka, ’94) or desert rodents worldwide (Kelt
and Brown, ’99). Yet competition is known to be
important in some deserts for certain taxa, but
not for others.
On the other hand, intraspecies competition in
the form of cannibalism may erupt in an extreme
environment if survival becomes desperate (Polis,
’81). For instance, tadpoles of some desert toads
and frogs begin to eat one another if the temporary pool of water on which their accelerated larval development depends begins to dry up before
they can reach metamorphosis (cited in part 1;
Greene, ’99). Obviously a few individuals (sometimes carnivorous oversize phenotypic morphs)
might survive in a small puddle, while none may
survive in a writhing mass of hundreds or thousands. Experiments have shown that poorly nour-
ished treefrog tadpoles from an ephemeral drainage ditch reached metamorphosis more quickly and
grew to greater size when fed conspecific tadpoles
than did well-nourished, but not cannibalistic, controls (Babbitt and Meshaka, 2000).
Cannibalism is common, also, in Arctic char, the
salmonid, living close to the limits of survival as
the only fish species in some high-latitude lakes in
northern Svalbard (Hammar, 2000). In addition,
desert scorpions indulge in intraspecific prey. As
in the further instance of the arctic fox, such behavior can be considered as a homeostatic mechanism keeping calories within a population and
regulating the population’s size within the reduced
carrying capacity of a stressful environment.
Presumably cannibalism is under genetic control but is facultatively triggered by environmental, or more likely, internal signals, such as
hunger. More generally, cannibalism is remarkably common in many animals ranging from flatworms to mammals, stressed by food scarcity.
With some limitations, cannibalism can increase
individual fitness, and probably group fitness of
nearly starving extremophiles (Polis, ’81). Under
such conditions noncannibals starve to death.
However, Darwinian fitness over time and space
is usually some sort of unspecified integrated longterm concept, rather than an abrupt quantum leap
(den Boer, ’99). For one thing, the animal itself
varies considerably from moment to moment, with
time of day, asleep or awake, hungry or satiated,
and so on. Also populations and species are made
up of individuals whose many properties, including number and size, vary over space and time.
The environment, too, is patchy in space and variable over time; extreme spatial focus occurs in
desert oases and in deepsea sulfurous hot springs.
As a result, an individual animal’s overall fitness
fluctuates with stress levels. Also in populations,
individuals vary slightly, too, so that effective natural selection is probably considerably weaker than
in a simple diagrammatic model. Quite often, too,
the individual victims of predation and accidents
seem to be selected independently of their classic
genotypic or phenotypic fitness.
Limits to selection
Most evidence amply supports natural selection’s
central role at the phenotypic level. Yet a number
of factors, both internal and external (Arthur,
2000) limit its action. Because natural selection:
• does not respond to neutral evolutionary
• but does over time efficiently eliminate negative phenotypes with decreased viability and
• supports the persistence of potentially positive adaptive variations, including those relevant to extreme environments, and
• acts on traditionally overlooked internal factors such as development and other regulatory
controls, mainly independent of the environment (Arthur, ’97). Such internal, largely
nonDarwinian factors are discussed below.
• requires for adaptation, as it does for speciation, some reduction in gene flow between the
normal and the stress resistant populations
(Riechert and Hall, 2000).
The notion that most mutations are neutral
seems, like the notion of evolution by chance, at
odds with the Darwinian view that natural selection is the pervasive motor of evolution (Bell, ’97).
Current analysis indicates that a purely neutral
theory cannot even account for all of molecular
evolution (Kreitman and Akashi, ’95). The increase
in genetic variation with population size, predicted
by that theory, is not observed. For instance,
nucleotide variation in humans was found to be 4
to 10 times lower than in three Drosophila species. Yet fly and human protein variations were
about the same despite the presence of far more
individual flies. At present the role of natural selection in the evolution of DNA seems to be minimally, or at least poorly, understood. Clearly
natural selection is more relevant to protein molecules, such as visual and respiratory pigments,
than it seems to be to nucleic acids.
However, rare mutations may be beneficial and,
as proved in massive experiments with bacterial
cultures, can become fixed by natural selection in
populations that exhibit a sort of punctuated evolution (Torres et al., ’96; see also Coyne and
Charlesworth, ’96; Elena et al., ’96). Yet the neutralists argue that weak or no natural selection
could have the opposite effect by keeping a species orthodox. Neutral mutations could do this by
lowering variation, rather than allowing a species to wander away from the established middle
of the road.
Instead, only when ecological or biological
changes open up new or freshly available habitats or conditions could natural selection activate
evolution in a long-static species that has accumulated many previously neutral features. In this
way a surge in evolutionary change could occur.
Such a possible mechanism is related to the no-
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
tion of “preadaptation” (Futuyma, ’98) in which
some cryptic feature, such as tolerance to cold
(Margesin and Schinner, ’99), may be already
present in a tropical animal. Yet the potential
asset is not obvious or put to the test until the
climate cools or the animal migrates to high altitudes or higher latitudes.
Preadaptation, or perhaps more aptly recruitment, also is frequently relevant at the molecular and genetic levels (Le Guyader, ’96). For
instance, a number of ancient proteins long used
in heat shock responses or as certain enzymes in
intermediary metabolism were recruited, perhaps
by gene duplication and modification, for structural use as crystallins in the lenses of various
vertebrate and squid eyes. Apparently, the molecules concerned “happened” to be transparent to
visible wavelengths of light and otherwise suited
to the visual optic needs of various animal groups
A number of comparable switches in function
are likely among UV-B screening pigments that
are widespread in exposed organisms ranging
from the earliest types of prokaryotes to fungi,
plants, and animals. Various pigments serve passively to reduce radiation damage in exposed organisms (Cockell and Knowland, ’99). Animals in
most deserts as well as at high latitudes and altitudes usually are strongly exposed to such potentially harmful rays. They may also be important
in the upper 50 m of the open ocean generally.
In high alpine or Arctic freshwater ponds zooplankton exposed to high summer light and UV-B
intensities typically are bright red from carotinoids
that act both as light-shielding filters and as antioxidants (Hessen et al., ’99). Daphnia in such clear
water ponds occurs as a morphotype with a black
melanin pigment screen in the carapace. An alternative melanin-free morphotype occurs in less transparent, or shaded, ponds.
Another example relates directly to extremophile
biochemistry. As mentioned in part 1, Antarctic
notothenioid fishes year-round have antifreeze glycoproteins in their blood that prevent internal ice
formation at seawater temperatures, modestly below 0°C. Apparently, the genes controlling antifreeze synthesis evolved from those classically
responsible for trypsinogen (precursor of the enzyme trypsin) synthesis. In the process. much of
the trypsinogen molecule was discarded, numerous repeats of a retained threonine-alinine-alinine
tripeptide were generated, and two sugars attached
to each threonine (Chen et al., ’97a,b; Cheng and
Available evidence implies that notothenioid
antifreeze evolved just once in their history and
did so between 5 and 14 million years ago (Eastman and Clarke, ’98). Such a transformation of
an old molecule to a new function seems rather
like the more upscale conversion of a terrestrial
tetrapod limb into the flipper of a seal or whale
(part 1). Interestingly, quite independent evolution of a closely similar antifreeze molecule in the
Arctic cod (cited in part 1) took place by a different mechanism that did not involve trypsinogen
(Chen et al., ’97b). This implies that some biochemical constraint was involved here (Hodin,
QUESTIONING NATURAL SELECTION
Natural selection, although a key element in
Darwinian evolution, is nevertheless a topic of disputed importance among biologists. However, welldocumented examples of evolution, observed in
nature and in experiments in both the laboratory
and the field, prove that natural selection can act
dramatically and quickly to change the phenotypes of microorganismal, plant, and animal populations (Bell, ’97; Amzallag, 2000; discussed
further in part 3). Even the extensive changes involved in tetrapod and whale evolution (part 1)
appear from the fossil evidence to have proceeded
by cumulative modest Darwinian steps. In contrast, notions of macroevolutionary jumps, or
abrupt transitions, are implicit in the probable
effects of regulatory gene mutations (part 1). Also,
such major steps could perhaps account for some
of the frequent “missing links” in the fossil record
as well as in punctuated evolution (Schwartz, ’99;
further discussed in part 3).
Some major differences of professional opinion
depend on the level at which natural selection is
believed mainly to act (Sober and Wilson, ’98).
Molecules, genes, individual organisms, populations, species, higher groups, ecosystems, and even
a global ecosystem (Gaia?) (Huggett, ’99) or some,
or all of these together, have been taken by various biologists to be effective units in evolution.
Obviously, such a wide spectrum of opinions weakens the probability of reaching a broad consensus
(Williams, ’92; Depew and Weber, ’95; Jablonski
et al., ’96; Claridge et al., ’97). Most often, species
are taken to be the key units of selection.
More broadly, various biologists emphasizing
neutralist, punctuationalist, antiadaptationist,
haphazardist, and other points of view, have raised
critical questions about some basic Darwinian beliefs. For instance, molecular biologists generally
embrace the neutral mutation theory (Kimura, ’91;
Skibinski, 2000) and believe that natural selection
has a small, or even insignificant, role in molecular evolution and hence presumably in phenotypic
changes (Rollo, ’94). Yet the working connection
between genotypic and phenotypic evolution, no
doubt, depends only in part on natural selection.
To different degrees in various cases (Ridley, ’97),
it also involves chance drift of mutations to fixation or loss and on numerous factors that affect
and control gene inheritance and expression
(Fernandez and Hoeffler, ’99).
In the late 1990s a number of techniques became
available for exploring the relations of phenotypes
to genotypes (Streelman and Kocher, 2000). These
may be relevant variously at the levels of the genome, the transcriptome, or the proteome. Single
gene analysis is rarely adequate. Yet microarrays
of DNA, for instance, can be used as sensors to measure the activity of a thousand or more genes at
once (Marshall, ’99; Lockhart and Winzeler, 2000),
Such results are beginning to be correlated with
different kinds of cells and organisms engaged in
various types and levels of activity, including development and disease. Also, these data could
demonstrate changes in response to extreme
stresses, as well as any distinctive patterns of gene
expression (Niehrs and Pollet, ’99) characteristic of
species, such as camels, long established as extremophiles. One barrier to applying these methods to
evolution, especially to that of frontier animals, lies
in the importance of a detailed linkage map on
which to locate genes likely to be involved.
So far such information is limited to relatively
few mesophilic animals and some prokaryotic
extremophiles including hot spring and deepsea
species. Another hurdle, Streelman and Kocher
argue, is that data scans at both the genome and
transcript levels are needed (plus their linkage to
the phenotype) (Vukmirovic and Tilghman, 2000)
as well as functional tests preferably in the field.
The scope of this challenge is suggested by a DNA
micro array study of the effects over 250 cultured
generations of yeast partly starved for glucose
(Ferea et al., ’99). By the end of the experiment,
several hundred genes, clustered into groups related to metabolism, respiration, and glycolysis,
had changed their expression. The responses of
Drosophila (for which a large amount of the
required molecular data is available) to multigenerational starvation, mainly studied at the phenotypic level, will be discussed in part 3.
During the 1980s and 1990s, probability, various physical and chemical constraints on evolu-
tion, and emergent properties of complex systems
(Kauffman, ’93) gained some support as evolutionary mechanisms at the expense of fitness, adaptation, and natural selection (Depew and Weber,
’95). Yet, the neo-Darwinians and ecologists, as
well as some molecular and developmental biologists, seem often to have underestimated the probable significance of such inherent alternative vital
elements for evolution.
Most dramatically, these are components of the
basic unity of life mentioned near the beginning
of this part of this essay. Many of these are not
just “frozen accidents” (Weiss and Fullerton, 2000)
but seem largely dependent on the nature of the
universe. Also, the evolutionary consequences of
physiological self-regulation appear not to have
been factored in effectively. Such rather neglected
features may be quite distinct from the natural
selection-responsive factors usually considered relevant to evolution. They tend, in fact, to define
life’s essential components plus its emergent complex-system properties. Some of these inherent elements were in place in the solar system long
before living beings arose.
Thus the basic physics and chemistry of life are
remarkably the same as for nonlife. In their crucial energy handling, organisms are subject to
laws of thermodynamics (Prigogine and Stenger,
’84) comparable to those of Willard Gibbs for reversible abiotic chemical reactions. For its substance, life is made up of the same chemical
elements as the rocks, the sea, and the atmosphere, but in a highly selective pattern. The
properties of the particular chemical elements essential for life (initially, mostly carbon, hydrogen,
and oxygen, along with nitrogen, phosphorus, sodium, potassium, chloride, iron, and a few others) and of simple inorganic molecules (mainly
water and carbon dioxide) seem to be fundamental in all prokaryotes and eukaryotes through
phylogenetic time (Henderson, ’13).
Some of life’s complex inherent properties, such
as the core of its energy metabolism, have been
firmly in place since the early prokaryotes (part l).
Of its three major components in animals, the citric acid cycle has been proposed as the universal
ancestor of intermediary metabolism (Morowitz et
al., 2000). Also, the lipid plasma membrane, enclosing each cell, was a necessary component from
the start of cellular life. Even rhodopsin, the photon-driven proton pump, familiar in vision, was apparently already present in ancient purple bacteria
(Edman et al., ’99). Certainly since animals evolved,
many such ancient vital features up to now seem
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
to have been largely insulated from any ordinary
action of natural selection and the environment.
At the level of complex organic molecules, extensive analyses of single-stranded RNA indicate
that clear nonrandom biases in their nucleotide
composition (not sequence) are largely independent in widespread data sampling of many functions and phylogenetic affinities (Schultes et al.,
’97). Such remarkable broadly occurring close
similarities in nucleic acid composition probably
depend in this case, on biophysical constraints,
such as specific molecular folding (known to be
important for extremophiles) that may augment
thermodynamic stability (Schultes et al., ’99a).
These data imply that more than half of RNA’s
secondary structure relates to self-organization
and less than a third is available to be acted on
by evolutionary selection, with the remaining balance accounted for by random mutations (Schultes
et al., ’99b). Failure to allow for such substantial
limitations may explain some of the inconsistencies in phylogenies derived just from nucleic acid
sequences or even from coded proteins (part 1;
Foster and Hickey, 2000). Selectively analyzing
extremophiles in Schultes and colleagues’ massive
data sets for possible correlation of their RNA composition with their specific stresses may provide
some interesting leads.
Despite problems and controversy, considerable
additional evidence continues to pile up to support
the relevance of natural selection to most levels of
evolution (Culver et al., ’95; Givnish and Sytsma,
’97; Graur and Li, 2000; Huey et al., 2000; Rundle
et al., 2000). These include phenotypic molecular
changes, as in the hemoglobin of high-altitude birds
(Butler, ’91) and the quite special hemoglobins of
deepsea vestimentiferan worms inhabiting hydrothermal vent areas (Zal et al., ’98). Yet the time
may be ripe for a fresh formulation that will effectively integrate the burgeoning molecular data into
a new comprehensive theory of evolution (Carroll,
2000). The rapid, multiple divergence of some large,
as well as some small scale, evolutionary changes
are clearly challenging (forthcoming in part 3).
Among other things, the present day relationship
between Linnaean species and contemporary species defined by molecular and population genetics
or cladistics needs to be resolved.
The basic definition of the gene and its relation to development and evolution no doubt
needs some revision or replacement (Beurton,
’98; Keller, 2000). For instance, despite the oneon-one conservatism of classic coding genes as
units, much of the active genome functions in
complex interacting networks in which gene expression, suppression and release from suppression play critical roles (Nagy, ’98; Burton et al.,
’99). In addition, organization into compartments
or modules (Hartwell et al., ’99; Raff and Sly,
2000), involving control genes, and signaling systems, may provide the flexibility needed for
Also, the relation between genotype and phenotype, discussed further in part 3, is in need of aggressive research. How a genotype can change
substantially despite an apparently stable phenotype (phenogenetic drift), and how apparently conservative genes, such as Hox, can switch from one
classic function to a quite new and different one
(Eizinger et al., ’99; Abzhanov and Kaufman, 2000)
are challenging events. They surely attest to our
rather shallow understanding of such crucial relations (Weiss and Fullerton, 2000). This would
seem to be an example of the dilemma already
cited more generally: how can the basic unity of
life be maintained despite a built-in antagonistic
tendency toward diversity of many kinds that includes the wide spectrum of the world’s extremophiles. Perhaps the drifting genotype is a given
feature of life comparable to the Malthusian population pressure. Not the least of the problems involved is to deal scrupulously with the subtleties
of homology that are crucial for tracking evolution
at various levels from molecules to behavior
(Nilssen, ’96; Abzhanov et al., 1999).
In the third and last part of this essay the influence of several major factors on extremophile evolution will be discussed—genetic and possible
extreme environmental sources of animal variations, rapid bouts of diversification, as well as
chance and extinction.
Abzahnov A, Kaufman TC. 2000. Crustacean (malacostracan)
Hox genes and the evolution of the arthropod trunk. Development 127:2239–2249.
Abzahnov A, Popadic A, Kaufman TC. 1999. Chelicerate Hox
genes and the homology of arthropod segments. Evol Dev
Adamczewska AM, Morris S. 2000. Respiratory gas transport,
metabolic status, and locomotor capacity of the Christmas
Island Red Crab Cecareiodea natalis assessed in the field
with respect to dichotomous seasonal activity levels. J Exper
Adami C, Ofria C, Collier TC. 2000. Evolution of biological
complexity. Proc Natl Acad Sci USA 9:4463–4468.
Alexander RM. 1989. Optimization and gaits in the locomotion of vertebrates. Physiol Rev 69:1199–1227.
Amundson R. 1996. Historical development of the concept of
adaptation. In: Rose MR, Lauder GV, editors. Adaptation.
San Diego: Academic Press, p 11–53.
Amzallag GN. 2000. Connectance in Sorghum development:
beyond the genotype-phenotype duality. BioSystems 56:1–11.
Arthur W. 1997. The origin of body plans: a study in evolutionary developmental biology. New York: Cambridge University Press.
Arthur W. 2000. The concept of developmental reprogramming and the quest for an inclusive theory of evolutionary
machanisms. Evol Dev 2:40–57.
Auyang SY. 1998. Foundations of complex-system theories
in economics, evolutionary biology, and statistical physics.
New York: Cambridge University Press.
Babbitt KJ, Meshako WE Jr. 2000. Benefits of eating conspecifics: effects of background diet on survival and metamorphosis in the Cuban treefrog (Osteopilus septrionalis).
Bachmann K. 1998. Species as units of diversity: an outdated concept. Theory BioSci 117:213–230.
Baldauf SL. 1999. A search for the origins of animals and
fungi: comparing and combining molecular data. Am Nat
Bargelloni L, Ritchie PA, Patarnello T, Battaglia B, Lambert DM, Meyer A. 1994. Molecular evolution at subzero
temperatures: mitochondria and nuclear phylogeines of
fishes from Antarctica (suborder Notothenioidei), and the
evolution of antifreeze glycopeptides. Mol Biol Evol 11:
Barraclough TG, Barclay MVL, Vogler AP. 1998. Species richness: does flower power explain beetle-mania? Curr Biol
Bavestrello G, Arillo A, Calcinai B, Cattaneo-Vietti R,
Cerrano C, Gaino E, Penna A, Sara M. 2000. Parasitic diatoms inside Antarctic sponges. Biol Bull 198:29–33.
Becskei A, Serrano L. 2000. Engineering stability in gene
networks by autoregulation. Nature 405:590–593.
Bell G. 1997. Selection: the mechanism of evolution. New
York: Chapman and Hall.
Bell G, Mooers AO. 1997. Size and complexity among multicellular organisms. Biol J Linn Soc 60:315–363.
Berta A, Sumich JL. 1999. Marine mammals: evolutionary
biology. San Diego: Academic Press.
Beurton PJ. 1998. What are genes today? Theory Biosci
Bijlsma R, Loeschcke V, editors. 1997. Environmental stress,
adaptation and evolution. Basel: Birkhauser Verlag.
Blix AS. 1989. Arctic resignation: winter dormancy without
hypothermia. In: Malan A, Canguilhem B, editors. Living
in the cold II. Colloque INSERM/John Libbery Eutotext Ltd,
Block BA. 1994. Thermogenesis in muscle. Annu Rev Physiol
Block BA, Finnerty JR, Stewart AFR, Kidd J. 1993. Evolution of endothermy in fish: mapping physiological traits on
a molecular phylogeny. Science 260:210–214.
Boake CRB. 2000. Flying apart: mating behavior and speciation. BioScience 50:501–508.
Bonner JT. 1988. The evolution of complexity. Princeton, NJ:
Princeton University Press.
Bordenstein SR, O’Hara FP, Werren JH. 2001. Wolbachiainduced incompatibility precedes other hybrid incompatibilities in Nasonia. Nature 409:707–710.
Bothma JP. 1998. Carnivore ecology in arid lands. Berlin:
Bradshaw SD. 1997. Homeostasis in desert reptiles. Berlin:
Bromham LD, Degnan BM. 1999. Hemichordates and deuteros-
tome evolution: robust molecular phylogenetic support for a
hemichordate + echinoderm clade. Evol Dev 1:166–171.
Brown A. 1999. The Darwin wars: how stupid genes became
selfish gods: London: Simon and Schuster.
Brusca RC. 2000. Unraveling the history of arthropod
biodiversiification. Ann Mol Bot Gard 87:13–25.
Burton RS, Rawson PD, Edmands S. 1999. Genetic architecture of physiological phenotypes: empirical evidence for coadapted gene complexes. Am Zool 39:451–462.
Butler PJ. 1991. Exercise in birds. J Exp Biol 160:233–262.
Carroll RL. 1997. Patterns and processes of vertebrate evolution. New York: Cambridge University Press.
Carroll RL. 2000. Towards a new evolutionary synthesis.
Trends Ecol Evol 15:27–32.
Cavalier-Smith T. 1998a. A revised six-kingdom system of life.
Biol Rev Camb 73:203–266.
Cavalier-Smith T. 1998b. Neomonada and the origin of animals and fungi. In: Coombs GH, Vickerman K, Sleigh MA,
Warren A, editors. Evolutionary relationships among protozoa. Dordrecht: Kluwer Academic Publishers, p 375–407.
Chapin FS III, Körner C, editors. 1995. Arctic and alpine
biodiversity: patterns, causes and ecosystem consequences.
New York: Springer-Verlag.
Chen LB, DeVries AL, Cheng CHC. 1997a. Evolution of antifreeze glycoprotein gene from a trypsinogen gene in
Antarctic notothenioid fish. Proc Natl Acad Sci USA
Chen LB, DeVries AL, Cheng CHC. 1997b. Convergent evolution of antifreeze glycoproteins in Antarctic notothenioid
fish and Arctic cod. Proc Natl Acad Sci USA 94:3817–3822.
Cheng CHC, Chen BJ. 1999. Evolution of an antifreeze glycoprotein. Nature 401:443–444.
Choe JC, Crespi BJ. 1997. The evolution of social behavior in
insects and arachnids. New York: Cambridge University
Christensen JP. 2000. A relationship between deep-sea benthic
oxygen demand and oceanic primary production. Oceanol
Claridge MF, Dawah HA, Wilson MR. 1997. Species: the units
of biodiversity. London: Chapman and Hall.
Cockell CS, Knowland J. 1999. Ultraviolet radiation screening compounds. Biol Rev 74:311–345.
Company JB, Sardà F. 1998. Metabolic rates and energy content of deep-sea benthic decapod crustcaeans in the western Mediterranean Sea. Deep-Sea Res I 45:1861–1880.
Company JB, Sardà F. 2000. Growth parameters of deep-water decapod crustaceans in the Northwestern Mediterranean
Sea: a comparative approach. Mar Biol 136:79–90.
Company JB, Cartes JE, Sardá F. 2000. Biological patterns
and near-bottom population characteristics of two pasophaeis decapod crustacean species, Pasiphaea sivado (Risso,
1816) and Pasiphaea multidenta Esmark 1816, in the northwestern Mediterranean Sea. Mar Biol (submitted).
Corbet PS. 1999. Dragonflies: behaviour and ecology of
Odonata. Ithaca: Comstock Publishers.
Coyne JA, Charlesworth B. 1996. Mechanisms of punctuated
evolution. Science 274:1748–1749.
Coyne JA, Orr HA. 1998. The evolutionary genetics of speciation. Philos Trans R Soc Lon Ser B Biol Sci 353:287–306.
Csányi V. 1998. Evolution: model or metaphor? In: Van de
Vijver G, Salthe SN, Delpos M, editors. Evolutionary systems: biological and epistemological perspectives on selection and self-organization. Dordrecht: Kluwer Academic
Publishers, p 1–12.
Culver DC, Kane TC, Fong DW. 1995. Adaptation and natu-
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
ral selection in caves: the evolution of Gammarus minus.
Cambridge: Harvard University Press.
Cunningham EJA, Birkhead TR. 1998. Review: sex roles and
sexual selection. Anim Behav 56:1311–1321.
Dagorn L, Bach P, Josse E. 2000. Movement patterns of large
bigeye tuna (Thunnus obesus) in the open ocean, determined
using ultrasonic telemetry. Mar Biol 136:361–371.
Davidson DH, Ruvkin G. 1999. Themes from a NASA workshop on gene regulatory processes in development and evolution. J Exp Zool Mol Dev Evol 285:104–115.
Davidson EH, Peterson K, Cameron R. 1995. Origin of
bilaterian body plans: evolution of developmental regulatory mechanisms. Science 270:1319–1325.
Davis JN, Daly M. 1997. Evolutionary theory and history. Q
Rev Biol 72:407–435.
Degen AA. 1997. Ecophysiology of small desert mammals, Berlin: Springer-Verlag.
den Boer PJ. 1999. Natural selection or the nonsurvival of
the nonfit. Acta Biotheoret 47:83–97.
Denton EJ. 1989. Light and vision at depths greater than
200 metres. In: Herring PJ, Campbell AK, Whitfield M,
Maddock L, editors. Light and life in the sea. New York:
Cambridge University Press, p 127–148.
Depew DJ, Weber BH. 1995. Darwinism evolving: systems
dynamics and the geneology of natural selection. Cambridge,
MA: MIT Press.
De Quieroz K. 1998. The general lineage concepts of species:
species criteria, and the process of speciation. In: Howard
DJ, Berlocher SH, editors. Endless forms: species and speciation. New York: Oxford University Press, p 57–75.
Desbruyères D, Segonzac M. 1997. Handbook of deep-sea hydrothermal vent fauna. Plouzane, France: Editions
DeVries J, van Eerden MR. 1995. Thermal conductance in
aquatic birds in relation to the degree of water contact, body
mass, and bodyfat: energetic implications of living in a
strong cooling environment. Physiol Zool 68:1143–1163.
Dodds PS, Rothman DH, Weitz JS. 2001. Re-examination
of the “3/4-law” of metabolism. J Theor Biol 209:9–27
Doolittle WF. 2000. Uprooting the tree of life. Sci Am
Dudley R. 1998. Atmospheric oxygen, giant Paleozoic insects
and the evolution of aerial locomotor performance, J Exp
Dunlop JA, Webster M. 1999. Fossil evidence, terrestrialization and arachnid phylogeny. J Arachnol 27:86–93.
Eastman JT, Clarke A. 1998. A comparison of adaptive radiations of Antarctic fish with those of nonAntarctic fish. In:
diPrisco G, Pisano E, Clarke A, editors. Fishes of Antarctica:
a biological overview. Milan: Springer-Verlag Italia, p 3–26.
Edman K, Hollert P, Royant A, Beirhali H, Pebay-Peyroula
E, Hadju J, Neutze R, Landau EM. 1999. High-resolution
X-ray structure of an early intermediate in the bacteriorhodopsin photocycle. Nature 401:822–826.
Eisenbeis G, Meyer E. 1999. Ecophysiological and morphological features of glacier-dwelling Collembola. In: Margesin
R, Schinner F, editors. Cold-adapted organisms: ecology,
physiology, enzymology and molecular biology. Berlin:
Springer-Verlag, p 127–218.
Eizinger A, Jungblut B, Sommer RJ. 1999. Evolutionary
change in the functional specificity of genes. Trends Genet
Eldredge N. 1998. The pattern of evolution. New York: W.H.
Elena SF, Cooper VS, Lenski RE. 1996. Punctuated evolu-
tion caused by selection of rare beneficial mutations. Science 272:1802–1804.
Erwin DH. 1999. The origin of bodyplans. Amer Zool 39:
Farrell BD. 1998. “Inordinate fondness” explained: why are
there so many beetles. Science 281:555–559.
Ferea TL, Botstein D, Brown PO, Roenzweig RF. 1999. Systematic changes in gene expression patterns following
adative evolution in yeast. Proc Natl Acad Sci USA 96:9721–
Fernandez J, Hoeffler JP, editors. 1999. Gene expression systems: using nature for the art of expression. San Diego:
Fogg GE. 1998. The biology of polar habitats. Oxford: Oxford
Fogg MJ. 1995. Terraforming: engineering planetary environments. Warrendale, PA: Society of Automotive Engineers.
Forterre P, Philippe H. 1999. Where is the root of the universal tree of life? BioEssays 21:871–879.
Fortey RA, Thomas RH, editors. 1998. Arthropod relationships. London: Chapman and Hall.
Fortey RA, Briggs DEG, Wills MA. 1997. The Cambrian evolutionary “explosion” recalibrated. BioEssays 19:429–434.
Foster PG, Hickey DA. 2000. Compositional bias may affect
both DNA-based and protein-based phylogenetic reconstructions. J Mol Evol 48:284–290.
Frank SA. 1996. The design of natural and artificial adaptive systems. In: Rose MR, Lauder GV. Adaptation. San Diego: Academic Press, p 451–505.
Freeland SJ, Hurst LD. 1998. The genetic code is one in a
million. J Mol Evol 47:238–248.
French HM. 1999. Past and present permafrost as an indicator of climate change. Polar Res 18:269–274.
Fujikura K, Kojima S, Tamaki K, Maki Y, Hunt J, Okutani
T. 1999. The deep chemosynthesis-based community yet discovered from the hadal zone, 7326 m deep, in the Japan
Trench. Mar Ecol Prog Ser 190:17–26.
Futuyma DJ. 1998. Evolutionary biology. Sunderland, MA:
Sinauer Associates, Inc.
Gadagkar R. 1997. Survival strategies: cooperation and conflict in animal societies. Cambridge, MA: Harvard University Press.
Gaffin DD, Wennstrom KL, Brownell PH. 1992. Water detection in the desert sand scorpion, Paruroctonus mesaensis
(Scorpionida, Vacjoxidae). J Comp Physiol A 170:623–629.
Gardezi T, daSilva J. 1999. Diversity in relation to body size
in mammals: a comparative study. Am Nat 153:110–123.
Gavrilets D. 1999. A dynamical thoery of holey adaptive landscapes. Am Nat 154:1–22.
Gems D. 2000. An integrated theory of aging in the nematode Coenorhabditis elegans. J Anat 197:521–528.
Genoud M, Martin RD, Glaser D. 1997. Rate of metabolism
in the smallest simian primate, the pygmy marmoset
(Cebuella pygmaea). Amer J Primatol 41:229–247.
Givnish TJ, Sytsma KJ. 1997. Molecular evolution and adaptive radiation. New York: Cambridge University Press.
Goffredi SK, Childress JJ, Lallier FH, Desaulniers NT. 1999.
He ionic composition of the hydrothermal vent tube
worm Rifta pachyptila: evidence for the elimination of
SO2/4- and H+ and for a Cl-?HCO3 shift. Physiol Biochem
Goodrich J, Puangsomlee P, Martin M, Long D, Meyrowitz EM,
Copland G. 1997. A polycomb-group gene regulates homeotic
gene expression in Arabidopsis. Nature 386:44–51.
Gosling EM. 1994. Speciation and species concepts in the ma-
rine environment. In: Beaumont AR, editor. Genetics and
evolution of aquatic organisms. London: Chapman and Hall,
Gould SJ. 1997. Cope’s rule as physiological artefact. Nature
Gould SJ, Lewontin RC. 1979. The spandrels of San Marco
and the panglossian paradigm. A critique of the adaptationist programme. Proc Roy Soc London B205:581–598.
Gradinger R. 1999. Integrated abundance and biomass of
sympagic meiofauna in Arctic and Antarctic pack ice. Polar
Grassle JF, Maciolek NJ. 1992. Deep-sea richness: regional
and local diversity estimates from quantitative bottom
samples. Am Nat 139:313–341.
Graur D, Li WH. 2000. Fundamentals of molecular evolution. Sunderland, MA: Sinauer Associates, Inc.
Gray RD, Jordan FM. 2000. Language trees support the express-train sequence of Austronesian expansion. Nature
Greene E. 1999. Phenotypic variation in larval development
and evolution: polymorphism, polyphemism, and developmental reaction norms. In: Hall BH, Wake MH, editors. The
origin and evolution of larval forms. San Diego: Academic
Press, p 379–400.
Greer JM, Puetz J, Thomas KR, Capecchi MR. 2000. Maintenance of functional equivalence during paralogous Hox gene
evolution. Nature 403:661–666.
Gregory TR, Hebert PDN. 1999. The modulation of DNA content: proximate causes and ultimate consequences. Genome
Gregory TR, Hebert PDN, Kolasa, J. 2000. Evolutionary implications of the relationship between genome size and body
size in flatworms and copepods. Hered 84:201–208.
Griffiths D. 1992. Size, abundance, and energy use in communities. J Anim Ecol 61:307–315.
Gullan PJ, Cranston PS. 1994. The insects: an outline of entomology. London: Chapman and Hall.
Guppy M, Withers P. 1999. Metabolic depression in animals:
physiological perspectives and biochemical generalizations.
Biol Rev 74:1–40.
Haedrich RL. 1997. Distribution and population ecology; In:
Randall DJ, Farrell AP, editors. Deep-sea fishes. San Diego: Academic Press, p 79–114.
Hammar J. 2000. Cannibals and parasites: conflicting regulators of bimodality in high-latitude Arctic char, Salvelinus
alpinus. Oikos 88:33–47.
Hanken J, Wake DB, Freeman HL. 1999. Three new species
of minute salamanders (Thorius: Plethodontidae) from
Guerrero, Mexico, including the report of a novel dental
polymorphism in urodeles. Copeia 1999(4):917–931.
Harrison JF, Lighton JRB. 1998. Oxygen-sensitive flight metals in dragonfly Erythemis simplicicollis. J Exp Biol
Harrison RG. 1998. Linking evolutionary pattern and process: the relevance of species concepts for the study of speciation. In: Howard DJ, Berlocher SH. Endless forms:
species and speciation. New York: Oxford University Press,
Hartwell LH, Hopfield JJ, Leibler S, Murray AW. 1999. From
molecular to modular cell biology. Nature 402 Supp:C42–C52.
Hayes JP. 1989. Field and maximal metabolic rates of deer
mice (Peromiscus maniculatus) at low and high altitude.
Physiol Zool 52:732–744.
Hazon N, Eddy FB, Glik G, editors. 1997. Ionic regulation in
animals. Berlin: Springer-Verlag.
Heinrich B. 1993. The hot-blooded insects: strategies and
mechanisms of thermoregulation. Cambridge, MA: Harvard
Heinze J, Foizik S, Hippert A, Hölldobler B. 1996. Apparent
dear-enemy phenomenon and environment-based recognition
cues in the ant Leptothorax nylanderi. Ethology 102:510–522.
Henderson LJ. 1913. The fitness of the environment. New
Hengeveld R, Walter GH. 1999. The two coexisting ecological
paradigms. Acta Biotheoret 47:141–170.
Herring LJ, Campbell AK, Whitefield M, Maddock L, editors.
Light and life in the sea. Cambridge: Cambridge University Press.
Hessen DO, Borgeraas J, Kessler K, Refseth UH. 1999. UVB susceptibility and photoreception of Arctic Daphnia
morphotypes. Polar Res 18:345–352.
Hodin J. 2000. Plasticity and constraints in development and
evolution. J Exp Zool Mol Dev Evol 288:1–20.
Hoffmann AA, Hercus MJ, 2000. Environmental stress as an
evolutionary force. BioScience 50:217–226.
Hoffmann AA, Parsons PA. 1997. Extreme environmental
change and evolution. New York: CambridgeUniversity
Hoffmann EE, Lascara CM. 2000. Modelling the growth dynamics of Antarctic krill Euphausia superba. Mar Ecol Prog
Holland KN, Brill RW, Chang RKC, Sibert JR, Fournier DA.
1992. Physiological and behavioural thermoregulation in
bigeye tuna (Thunnus obesus). Nature 358:410–412.
Holland PWH. 1999. The future of evolutionary developmental biology. Nature 402 Supp:C41–C44.
Huey RB, Berrigan D. 1996. Testing evolutionary hypotheses
of acclimation. In: Johnston IA, Bennett AF, editors. Animals and temperature: phenotypic and evolutionary adaptation. New York: Cambridge University Press, p.203–231.
Huey RB, Gilchrist GW, Carlson ML, Berrigan D, Serra L.
2000. Rapid evolution of a geographic cline in size in an
introduced fly. Science 287:308–309.
Huggett RJ. 1999. Ecosphere, biosphere, or Gaia? What to
call the global ecosystem, Global Ecol Biogeog 8:425–431.
Iwasaki F, Wakahara M. 1999. Adaptable larval life histories
in different populations of the salamander Hynobius
retardatus, living in various habitats. Zool Sci 16:667–674.
Jablonski D. 1996. Body size and macroevaluation. In:
Jablonski D, Erwin DH, Lipps JH, editors. Evolutionary
Paleobiology. Chicago: University of Chicago Press.
Jablonski D, Erwin DH, Lipps JH, editors. 1996. Evolutionary paleobiology: in honor of James W. Valentine. Chicago:
University of Chicago Press.
Jacobs LF. 1996. The economy of winter: phenotypic plasticity in behavior and brain structure. Biol Bull 191:92–100.
Jia Z, DeLuca CI, Chao H, Davies PL. 1996. Structural basis
for the binding of a globular antifreeze protein to ice. Nature 384:285–288.
Kaufman SA. 1993. The origins of order: self-organization and
selection in evolution. New York: Oxford University Press.
Keller EF. 2000. The century of the gene. Cambridge, MA:
Harvard University Press.
Keller L, editor. 1999. Levels of selection in evolution.
Princeton, NJ: Princeton University Press.
Kelt DA, Brown JH. 1999. Community structure and assembly rules: confronting conceptual and statistical issues with
data on desert rodents. In: Weiher E, Keddy P, editors. Ecological assembly rules: perspectives, advances, retreats. New
York: Cambridge University Press, p 75–107.
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
Kenrick P, Crane PR. 1997. The origin and early diversification of land plants: a cladistic study. Washington, DC:
Smithsonian Institution Press.
Kimura M. 1991. A neutral theory of molecular evolution: a
review of recent evidence. Jpn J Genet 66:367–386.
Kinchin IM. 2000. On the significance of a “minor” phylum
(the tardigrada) in the context of a constructivist view of
knowledge. Perspect Biol Med 43:243–251.
Kliewer SA, Lehmann JM, Willson TM. 1999. Orphan nuclear
receptors: shifting endocrinology into reverse. Science
Kozlowski J. 1999. Adaptation: a life history perspective.
Kreitman M, Akashi H. 1995. Molecular evidence for natural
selection. Ann Rev Eco Syst 26:403–422.
Kristensen NP. 1994. The groundplan and basal diversification of the hexapods. In: Fortey RA, Thomas RH, editors. Arthropod relationships. London: Chapman and
Hall, p 381–293.
Labandeira CC, Sepkoski JJ. 1993. Insect diversity in the
fossil record. Science 261:310–315.
Labropoulou M, Kostikas, I, 1999. Patterns of resource use
in deep-water decapods. Mar Ecol Prog Ser 184:171–182.
Lake JA, Rivera MC. 1996. The prokaryotic ancestry of eukaryotes. In: Roberts DM, Sharp P, Alderson G, Collins M,
editors. Evolution of microbial life. New York: Cambridge
University Press. p 87–108.
Lamond AI, Earnshaw WC. 1998. Structure and function in
the nucleus. Science 280:547–553.
Landweber LF, Dobson AP, editors. 1999. Genetics and the
extinction of species. Princeton, NJ: Princeton University
Leather SR, Walters KFA, Bale JS. 1993. The ecology of insect overwintering. New York: Cambridge University Press.
Le Guyader H. 1996. Preadaptation et biologie moleculaire:
vers un renouveau du concept? Bull Soc Zool Fr 121:39–44.
Lewin R. 1992. Complexity: life at the edge of chaos. New
Lockhart DJ, Winzeler EA. 2000. Genomics, gene expression
and DNA Arrays. Nature 405:827–836.
Longhurst TJ, Joss JMP. 1999. Homeobox genes in the Australian lungfish, Neoceratodus forsteri. J Exper Zool.
Lordan C, Collins MA, Perales-Raya C. 1998. Observations
on morphology, age and diet of three Architeuthis caught
off the west coast of Ireland in 1995. J Mar Biol Assoc U K
Louw GN. 1993. Physiological animal ecology. New York:
Lucherini M, Birochio D, Martinelli C, Legato AM. 2000. Spatial niche overlap between vicuñas and guanaco. Act Theriol
Luger MS, Schabetsberger R, Jersabek CD, Goldsmith A.
2000. Life cycles, size and reproduction of two coexisting
calanoid copepods Arctodiaptomous alpinus (IMHOF, 1885)
and Mixodiaptomous laciniatus (Lilljeborg, 1889) in a small
high-altitude lake. Arch Hydrobiol 148:161–185.
MacDonald AG. 1975. Physiological aspects of deep sea biology. New York: Cambridge University Press.
Mackenzie D. 1999. New clues to why size equals destiny.
Maddrell SHP. 1998. Why are there no insects in the open
sea? J Exp Biol 201:2461–2464.
Magurran AE, May RM, editors. 1999. Evolution of biological diversity. New York: Oxford University Press.
Malakhov VV. 1994. Nematodes: structure, development, classification, and phylogeny. Washington, DC: Smithsonian Institution Press.
Malthus TR. 1798. First essay on population.
Margesin R, Schinner F, editors. 1999. Cold-adapted organisms: ecology, physiology, enzymology and molecular biology. Berlin: Springer-Verlag.
Margulis L. 1996. Archaeal-eubacterial mergers in the origin
of Eukarya: phylogenetic classification of life. Proc Natl Acad
Sci USA 93:1071–1076.
Margulis L, Schwartz KV. 1998. Five kingdoms: an illustrated
guide to the phyla of life on earth. 3rd ed. New York: W.H.
Marriott BM, Carlson SJ. 1996. Nutritional needs in cold and
high-altitude environments. Washington, DC: National
Marshall E. 1999. Do-it-yourself gene watching. Science
Martin AP. 1999. Increasing genomic complexity by gene duplication and the origin of vertebrates. Am Nat 154:111–128.
Maruyama T, Ishikura M, Yamazaki S, Kanai S. 1998. Molecular phylogeny of zooxabthellate bivalves. Biol Bull
Mayden RL. 1997. A hierarchy of species concepts: the denouement in the saga of the species problem. In: Claridge
MF, Dawah HA, Wilson MR. Species: the units of biodiversity. London: Chapman and Hall.
Maynard Smith J. 1999. Conflict and cooperation in human
societies. In: Keller L, editor. Levels of selection in evolution. Princeton, NJ: Princeton University Press, p 197–208.
Maynard Smith J, Szathmary E. 1995. The major transitions
in evolution. Oxford: WH Freeman.
Mayr E. 1991. One long argument: Charles Darwin and the
genesis of modern evolutionary thought. Cambridge, MA:
Harvard University Press.
McDiarmid RW, Altig R, editors. 1999. Tadpoles: the biology
of anuran larvae. Chicago: University of Chicago Press.
McGee BG Jr. 1999. Theoretical morphology: the concept and
its applications: New York: Columbia University Press.
McMahon TA, Bonner JT. 1983. On size and life. New York:
Scientific American Library.
McShea DW. 1996. Metazoan complexity and evolution: is
there a trend? Evolution 50:477–492.
Miaud C, Guyetant R, Elmberg J. 1999. Variations in lifehistory traits in the common frog Rana temporaria (amphibia: Anura): a literature review and new data from the
French Alps. J. Zool 249:61–73.
Miller RJ, editor. 1996. Miniature vertebrates: the implication of body size. New York: Oxford University Press.
Mito T, Endo K. 2000. PCR survey of Hox genes in the crinoid
and ophuroid: evidence for anterior conservation and posterior expression in the echinoderm Hox gene cluster. Mol
Phylogenet Evol 14:375–388.
Montagne J, Stewart MJ, Stocker H, Hafen E, Kozma SC,
Thomas G. 1999. Drosophila S6 kinase: a regulator of cell
size. Science 285:2126–2129.
Moran NA, Telang A. 1998. Bacteriocyte-associated symbionts
of insects: a variety of insect groups harbor ancient prokaryotic endosymbionts. BioSci 48:295–304.
Morowitz HJ. 1992. Beginnings of cellular life: metabolism
recapitulates biogenesis. New Haven: Yale University Press.
Morowitz HJ, Kostelnik JD, Yang J, Cody GD. 2000. The origin of intermediary metabolism. Proc Natl Acad Sci USA
Mylvaganam SE, Bonaventura C, Bonaventura J, Getzoff D.
1996. Structural basis for the root effect in haemoglobin.
Nature Struct Biol 3:275–283.
Nagy L. 1998. Changing patterns of gene regulation in the
evolution of arthropod morphology. Am Zool 38:818–828.
Nagy TR, Gower BA, Stetson MH. 1995. Endocrine correlates
of seasonal body mass dynamics in the collared lemming
(Dicrostonyx groenlandicus). Am Zool 35:246–258.
Nardon P. 1999. Symbiosis as an example of an acquired character: Neo-Lamarckism or Darwinism. Bull Soc Zool Fr
Nelson K, Fisher CR. 2000. Absence of cospeciation in deepsea vestimentiferan tube worms and their bacterial endosymbionts. Symbiosis 28:1–15.
Netzer WJ, Hartl FU. 1997. Recombination or protein domains facilitated by co-translational folding in eukaryotes.
Nicolson SW. 1980. Water balance and osmoregulation in
Onymacris plana, a tenebrionid beetle from the Namib
Desert. J Insect Physiol 26:315–320.
Niehrs C, Pollet N. 1999. Synexpression groups in eukaryotes. Nature 402:483–487.
Nielsen C. 1995. Animal evolution: interrelationships of the
living phyla. New York: Oxford University Press.
Nilssen DE, Osorio D. 1994. Homology and parallelism in
arthropod sensory processing. In Fortey RA, Thomas RH,
editors. Arthropod relationships. London: Chapman and
Hall, p 333–347.
Nilsson DE. 1996. Eye ancestry: old genes for new eyes. Curr
Nixon K, Wheeler QD. 1990. An amplification of the phylogenetic species concept. Cladistics 6:211–223.
Osawa S. 1995. Evolution of the genetic code. Oxford: Oxford
Paladino FV, O’Connor MP, Spotila JR. 1990. Metabolism of
leatherback turtles, gigantothermy, and thermoregulation
of dinosaurs. Nature 344:858–860.
Pandey A, Mann M. 2000. Proteomics to study genes and genomes. Nature 405:837–846.
Parrish JK, Hamner W. 1997. Animal groups in three dimensions. New York: Cambridge University Press.
Parsons PA. 1999. Can biodiversity and genetic variability
be conserved in hostile environments? In: Wasser SP, editor. Evolutionary theory and processes: modern perspectives:
papers in honour of Eviatar Nevo. Dordrecht: Kluwer Academic Publishers, p 273–282.
Patterson BD. 1999. Contingency and determinism in mammalian biogeography: the role of history. J Mammalogy
Pazkowski CA, Tonn WM. 2000. Community concordance between the fish and aquatic birds of lakes in northern
Alberta, Canada: the relative importance of environmental
and biotic factors. Freshw Biol 43:421–437.
Peterson KJ, Davidson EH. 2000. Regulatory evolution and
the origin of bilaterians. Proc Natl Acad Sci USA 97:4420–
Pettersson M. 1996. Complexity and evolution. New York:
Cambridge University Press.
Philippe H, Adoutte A. 1998. The molecular phylogeny of
eukaryota: solid facts and uncertainties. In: Coombs CH,
Vickerman K, Sleigh MA, Warren A, editors. Evolutionary
relationships among protozoa. Dordrecht: Kluwer Academic
Publishers, p 25–56.
Pigliucci M, Kaplan J. 2000. The fall and rise of Dr. Pangloss:
adaptationism and the Spandrels paper 20 years later.
Trends Ecol Evol 15:66–70.
Polis GA. 1981. The evolution and dynamics of intraspecific
predation. Annu Rev Ecol Syst 12:225–251.
Pond DW, Dixon DR, Bell MV, Fallick AE, Sargent JR. 1997.
Occurrence of 16:2(n-4) and 18:2 (n-4) fatty acids in the
lipids of the hydrothermal vent shrimp Rimicaris exoculata
and Alvinocaris markensis: nutritional and trophic implications. Mar Ecol Prog Ser 156:167–174.
Pörtner HO, O’Dor RK, Macmillan DL. 1994. Physiology of
cephalopod molluscs: life style and performance adaptation.
Basel: Gordon & Breach.
Postgate J. 1994. The outer reaches of life. New York: Cambridge University Press.
Power JH. 1989. Sink or swim: growth mechanics and zooplankton hydromechanics. Am Nat 133:706–721.
Price T. 1996. Exploding species. Trends Ecol Evolut 11:314–
Price T, Yeh P. 1999. One hundred years of natural selection
in the wild. Endeavour 23:145–147.
Prigogine I, Stenger I. 1984. Order out of chaos, man’s new
dialog with nature. New York: Bantam.
Raff RA, Sly BJ. 2000. Modularity and dissociation in the
evolution of gene expression territories in development.
Evolut Dev 2:102–113.
Ray LB. 1999. The science of signal transduction. Science
Reardon M. 1986. The besieged desert: war, drought, poaching in the Namib desert. London: Collins.
Reeve HK, Keller L. 1999. Levels of selection: burying the
units-of-selection debate and unearthing the crucial new issues. In: Keller L, editor. Levels of selection in evolution.
Princeton, NJ: Princeton University Press, p 3–14.
Reid K, Barlow KE, Croxall JP, Taylor RI. 1999. Predicting
changes in the Arctic krill, Euphausia syperba population
at South Georgia. Mar Biol 135:647–652.
Ridley M. 1997. Evolution. Oxford: Oxford University Press.
Riechert SE, Hall RF. 2000. Local population success in heterogeneous habitats; reciprocal transplant experiments completed on a desert spider. J Evolut Biol 13:541–541.
Rinkevich B. 2000. A critical approach to the definition of
Darwinian units of selection. Biol Bull 199:231–240.
Ritzmann RE, Quinn RD, Watson JT, Zill SN. 2000. Insect
walking and biorobotics: a relationship with mutual benefits. BioSci 50:23–33.
Roger AJ. 1999. Reconstructing early events in eukaryotic
evolution. Am Nat 154:S146–163.
Rollo CD. 1994. Phenotypes: their epigenetics, ecology and
evolution. London: Chapman and Hall.
Rose MR, Lauder GV. 1996. Adaptation. San Diego: Academic
Rose MR, Nusbaum TJ, Chippindale AK. 1996. Laboratory
evolution: the experimental wonderland and the Cheshire
Cat Syndrome. In: Rose MR, Lauder GV, editors. Adaptation. San Diego: Academic Press, p 221–241.
Ruby EG, McFall-Ngai MJ. 1999. Oxygen-utilizing reactions
and symbiotic colonization of the squid light organ by Vibrio
fischeri. Trends Microbiol 7:414–420.
Rundle HD, Nagel L, Wenrick Boughman J, Schluter D. 2000.
Natural selection and parallel speciation in sympatric sticklebacks. Science 287:306–308.
Ruse M. 1996. Monad to man: the concept of progress in evolutionary biology. Cambridge, MA: Harvard University Press.
Sanderson MJ, Hufford L, editors. 1996. Homoplasy: the recurrence of similarity in evolution. San Diego: Academic
Saunders WB, Work DM, Nikolaeva SV. 1999. Evolution of
EVOLUTIONARY CHALLENGES IN EXTREME ENVIRONMENTS
complexity in paleozoic ammonoid sutures. Science 286:
Sarrazin J, Juniper SK, Massoth G, Legendre P. 1999. Physical and chemical factors influencing species distributions
on hydrothermal sulfide edifices of the Juan de Fuca Ridge,
northeast pacific. Mar Ecol Prog Ser 190:89–112.
Scheltema AH, Schander C. 2000. Discrimination and phylogeny of solenogaster species through the morphology of
hard parts (Mollusca, Aplacophora, Neomeniomorpha). Biol
Schluter D. 1998. Ecological causes of speciation. In: Howard
DJ, Berlocher SH, editors. Endless forms: species and speciation. New York: Oxford University Press, p 114–129.
Schmidt-Nielsen K. 1984. Scaling, why is animal size so important. New York: Cambridge University Press.
Schmidt-Nielsen K. 1997. Animal physiology: adaptation and
environment. 5th ed. New York: Cambridge University
Schopf JW. 1999. Cradle of life: the discovery of the Earth’s
earliest fossils. Princeton, NJ: Princeton University Press.
Schultes E, Hraber PT, LaBean TH. 1997. Global similarities
in nucleotide base composition among disparate functional
classes of single-stranded RNA imply adaptive evolutionary convergence. RNA 3:792–806.
Schultes EA, Hraber PT, LaBean TH. 1999a. Estimating the
contributions of selection and self-organization in RNA secondary structure. J Mol Evol 49:76–83.
Schultes EA, Hraber PT, LaBean TH. 1999b. A parameterization of RNA sequence space. Complexity 4:61–71.
Schwartz JH. 1999. Sudden origins: fossils, genes and the
emergence of species. New York: John Wiley & Sons, Inc.
Seeley TD. 1995. The wisdom of the hive. Cambridge, MA:
Harvard University Press.
Segerstråle U. 2000. Defenders of the truth: the battle for
science in the sociobiology debate and beyond. New York:
Oxford University Press.
Sendova-Franks AB, Franks NR. 1999. Self-assembly, selforganization and division of labour. Phil Trans R Soc Lond
Shank TM, Black MB, Halanych KM, Lutz RA, Vrijenhoek
RC. 1999. Miocene radiation of deep-sea hydrothermal
vent shrimp (Caridea: Bresiliidae): evidence from mitochondrial cytochrome oxidase subunit I. Mol Phylogenet
Sinclair BJ. 2000. Water relations of the freeze-tolerant New
Zealand alpine cockroach Celatoblatta quinquemaculata
(Dictyoptera; Blattidae). J Insect Physiol 46:869–876.
Skibinski DOF. 2000. DNA tests of neutral theory; applications in marine genetics. Hydrobiol 420:137–152.
Slobodkin LB, Rapoport A. 1974. An optimal strategy of evolution. Q Rev Biol 49:181–200.
Smil V. 1997. Cycles of life: civilization and the biosphere.
New York: Scientific American Library.
Sober E, Wilson DS. 1998. Unto others: the evolution and
psychology of unselfish behavior. Cambridge, MA: Harvard
Somero GN. 1993. Adaptation to high hydrostatic pressure.
Annu Rev Physiol 54:557–577.
Somero GN. 1998. Adaptation to cold and depth: contrasts
between polar and deep-sea animals. In: Pörtner HO, Playle
RC, editors. Cold ocean physiology. New York: Cambridge
University Press, p 33–57.
Sømme, L. 1995. Invertebrates in hot and cold arid environments. Berlin: Springer-Verlag.
Sømme, L. 1999. Cold adaptations in terrestrial invertebrates.
In: Margesin R, Schinner F, editors. Cold-adapted organisms: ecology, physiology, enzymology and molecular biology. Berlin: Springer-Verlag, p 137–164.
Spalton JA. 1999. The food supply of Arabian oryx (Oryx
leocoryx), J Zool 248:433–441.
Spears T, Abele LG. 1998. Crustacean phylogeny inferred from
18S rDNA. In: Fortey RA, Thomas RH, editors. Arthropod
relationships. London: Chapman and Hall, p 169–187.
Strathdee AT, Bale JS. 1998. Life on the edge: insect ecology
in Arctic environments. Annu Rev Entomol 43:85–106.
Sterelny K, Griffiths PE. 1999. Sex and death: an introduction to philosophy of biology. Chicago: University of Chicago Press.
Streelman JT, Kocher TD. 2000. From phenotype to genotype. Evolut Dev 2:173–200.
Sturmbauer C, Meyer A. 1992. Genetic divergence, speciation and morphological stasis in a lineage of African cichlid
fishes. Nature 358:578–581.
Sugg DW, Chessar RK, Dobson FS, Hoogland JL. 1996. Population genetics meets behavioral ecology. Trends Ecol Evol
Szathmary E, Maynard Smith J. 1995. The major evolutionary transitions. Nature 374:227–232.
Szilágyi A, Závodszky P. 2000. Structural differences between
mesophilic, moderately thermophilic and extremely thermophilic protein subunits: results of a comprehensive survey.
Tautz D, Schmid K. 1999. From genes to individual developmental genes and the generation of the phenotype. In:
Magurran AE, May RM, editors. Evolution of biological diversity. Oxford: Oxford University Press, p 184–201.
Thaler K. 1999. Nival invertebrate animals in the east Alps:
a faunistic overview. In: Margesin R, Schinner F, editors.
Cold-adapted organisms: ecology, physiology, enzymology
and molecular biology. Berlin: Springer-Verlag.
Thomas RDK, Shearman RM, Stewart GW. 2000. Evolutionary exploitation of design options by the first animals with
hard skeletons. Science 288:1239–1242.
Tielemann BI, Williams JB. 1999. The role of hypothermia
in the water economy of desert birds. Physiol Biochem Zool
Tollrian T, Harvell CD. 1999. The ecology and evolution of inducible defenses. Princeton, NJ: Princeton University Press.
Tokeshi M. 1999. Species coexistence: ecological and evolutionary perspectives. Oxford: Blackwell Scientific Publishers.
Treonis AM, Wall DH, Virginia RA. 1999. Invertebrate
biodiversity in Antarctic dry valley soils and sediments. Ecosystems 2:482–492.
Valentine JW. 1995. Late precambrian bilaterians: grades and
clades. In Fitch WM, Ayala FJ, editors. Tempo and mode of
evolution. Genetics and evolution 50 years after Simpson.
Washington, DC. Nat Acad Sci, p 87–107.
Valentine JW, Jablonski D, Erwin DH. 1999. Fossils, molecules and embryos: new perspectives on the Cambrian explosion. Development 126:851–859.
Van Buskirk J, McCollum SA, Werner EE. 1997. Natural selection for environmentally induced phenotypes in tadpoles.
Van de Vijver G, Salthe SN, Delpos M, editors.1998. Evolutionary systems: biological and epistemological perspectives
on selection and self-organization. Dordrecht: Kluwer Academic Publishers.
Van Dover CL, Trask JL. 2000. Diversity at deep-sea hydrothermal vent and intertidal mussel beds. Mar Ecol Prog
Vellai T, Takacs K, Vida G. 1998. A new aspect to the origin
and evolution of eukaryotes. J Mol Evol 46:499–507.
Vermeij GJ, Dudley R. 2000. Why are there so few evolutionary transitions between aquatic and terrestrial ecosystems?
Biol J Linn Soc 70:541–554.
Vetter RD. 1991. Symbiosis and the evolution of novel trophic
strategies: thiotrophic organisms at hydrothermal vents.
In: Margulis L, Fester R, editors. Symbiosis as a source of
evolutionary innovation. Cambridge, MA: MIT Press, p
Via S, Hawthorne DJ. 1998. The genetics of speciation:
promises and prospects of quantitative trait locus mapping: In: Howard DJ, Berlocher H, editors. Endless forms:
species and speciation. New York: Oxford University
Press, p 352–364.
Viarengo A, Mancinelli G, Burlando B, Palfoli I, Marchi B.
1999. The SR Ca2+ ATPase of the Antarctic scallop Admussium colbecki: cold adaptation and heavy metal effects.
Polar Biol 21:361–375.
Vitt LJ, Pianka ER, editors. 1994. Lizard ecology: historical
and experimental perspectives. Princeton, NJ: Princeton
Vukmirovic OG, Tilghman SM. 2000. Exploring genome space.
Wagner PJ. 2000. Exhaustion of morphologic character states
among fossil taxa. Evolution 54:365–386.
Waide RB, Willig MR, Steiner CF, Mittelbach G, Gough L,
Dodson SI, Juday GP, Parmenter R. 1999. The relationship
between productivity and species richness Annu Rev Ecol
Walsburg GE. 2000. Small mammals in hot deserts; some generalizations revisited. BioScience 50:109–120.
Warburg MR. 1997. Ecophysiology of amphibians inhabiting
xeric environments. Berlin: Springer-Verlag.
Waterman TH. 1968. Systems theory and biology—view of a
biologist. In: Mesarovic MD, editor. Systems theory and biology. New York: Springer-Verlag, p 1–37.
Waterman TH. 1981. Polarization sensitivity. In: Autrum H,
editor. Handbook of sensory physiology, V. VII/6B. Heidelberg: Springer-Verlag, p 281–469.
Waterman TH. 1982. Fine structure and turnover of photoreceptor membranes. In: Westfall JA, editor. Visual cells in
evolution. New York: Raven Press, p 23–42.
Waterman TH. 1988. Animal navigation. New York: Scientific American Library.
Waterman TH. 1997. Photobiology of the deep twilight zone
and beyond (a review). Soc Photo-Opt Instr Eng 1963–Ocean
Waterman TH. 1999. The evolutionary challenges of extreme environments (part 1). J Exper Zool Mol Dev Evol
Weibel ER, Taylor CR, Bolis L. 1998. Principles of animal
design: the optimization and symmorphosis debate. New
York: Cambridge University Press.
Weiher E, Keddy P, editors. 1999. Ecological assembly rules:
perspectives, advances, retreats. New York: Cambridge University Press.
Weiss KM, Fullerton SM. 2000. Phenogenetic drift and the
evolution of genotype-phenotype relationships. Theor Pop
Wells H. 1999. Why the ammonites snuffed it. Mar Freshw
Behav Physiol 32:103–111
West JB. 1998. High life: a history of high-altitude physiology and medicine. New York: Oxford University Press.
Wiens JJ, Servedio MR. 1999. Species delimination in systematics: inferring diagnostic difference between species
Proc R Soc Lond B 267:631–636.
Williams GC. 1992. Natural selection: domains, levels, and
challenges. New York: Oxford University Press.
Wilson EO. 1975. Sociobiology. Cambridge, MA: The Belknap
Press of Harvard University Press.
Wistow G. 1993. Lens crystallins: gene recruitment and evolutionary dynamism. Trends Biochem Sci 18:301–306.
Wittenberg JB, Stein JL. 1995. Hemoglobin in the symbiontharboring gill of the marine gastropod Alviniconcha hessleri.
Bio Bull 188:5–7.
Wolpert L, Beddington R, Brookes J, Jessell T, Lawrence P,
Meyerowitz E. 1998. Principles of development. Oxford: Oxford University Press.
Womersley CZ, Wharton DA, Higa LM. 1998. Survival biology. In: Perry RN, Wright DJ, editors. Physiology and biochemistry of free-living and plant parasitic nematodes.
Wallingford: CaB International, p 271–302.
Wooton RJ, Kukalova-Peck J, Newman DJS, Muzon J. 1998.
Smart engineering in mid-Carboniferous: how well could
Palaeozoic dragonflies fly? Science 282:749–751.
Yamamoto Y, Jeffery WR. 2000. Central role for the lens in
cave fish eye degeneration. Science 289:631–633.
Zal F, Leize E, Lallier FH, Toulmond A, Van Dorsselaer A,
Childress JJ. 1998. S-Sulfohemoglobin and disulfide exchange: the mechanisms of sulfide binding by Riftia pachyptila hemoglobins. Proc Natl Acad Sci USA 95:8997–9002.