¿ìè¶ÌÊÓƵ

EVOLUTIONARY biologists tend to fall into two feuding groups. The conservatives,
who employ updated versions of traditional techniques, are the morphologists.
Concerned with the form and structure of organisms, they wave crumbling
bits of bone in threatening gestures at their opponents. In their turn,
the geneticists or molecular biologists thrust gels full of blotchy bands
under the noses of the morphologists. Both believe that they are taking
the most appropriate approach to reconstructing the course of evolution,
but the results of their endeavours are rarely compatible.

The debate is readily caricatured. Bones and teeth, the morphologists
contend, are the functioning structures of whole organisms, those creatures
whose past we want to know. You cannot understand the evolution without
understanding the biology of the animal, they argue. Geneticists counter
by observing that the molecules and their proxies, the gels or genetic sequence
data, are the genes (or their immediate products) themselves. Surely the
molecules must give a clearer picture of the changes in the genome, they
argue; surely it is time for evolutionary biology to use the tools of human
biology.

Further confusion comes from using the same technical terms to mean
different things. ‘¿ìè¶ÌÊÓƵs would sooner use each others’ toothbrushes
than their nomenclature,’ says Thomas Pollard of Johns Hopkins Medical School
in Baltimore. The prime example is the term homology, which to morphologists
means a similarity of structure so intricate and detailed as to imply recent
common descent. To a geneticist, homology in DNA sequences simply mean strings
of bases that are more-or-less the same.

The lines are drawn. The techniques and the knowledge base upon which
they rely are disparate; personal resentments are sometimes pointed. With
the exception of a few peace-loving souls who maintain that each approach
has something to offer, most scientists live firmly in one camp and rarely
listen to those in the other.

The crux of the debate is the old tension between phenotype (the morphologists’
purview) and genotype (the geneticists’). Natural selection presumably acts
on both, so why should studies of two ends of the same spectrum yield different
answers? There are many theoretical and pragmatic reasons for the divergence.
The first, and perhaps most important, is that the phenotype is clearly
not just a direct reflection of the genotype. When a morphologist looks
at a complex trait, such as mandible shape or age at sexual maturity, environmental
variables such as the availability of appropriate foods at the time of the
study, or even during development, are potent influences on its expression.
Female elephants, for example, generally become sexually mature at about
11 years but, if environmental conditions are poor due to drought or other
factors, maturity may be delayed until as late as 18 years.

The opportunities for thwarting the genetic potential of an organism
are not limited to such large-scale and impersonal influences as climatic
or environmental changes. The genes of other individuals, especially mothers,
or those who function as mothers, may be potent influences on the expression
of genes as phenotypes too. In a recent paper, William Atchley of North
Carolina State University and his colleagues comment, with a classic understatement:
‘We lack understanding of the genetic basis for evolutionary divergence
in morphology.

Another reason for the difference in answers may be that the two groups
are looking at different issues. The morphologist is often concerned with
understanding why and how two species diverged and differentiated; part
of the issue is thus changing adaptations. The geneticist, on the other
hand, is often more interested in when and by how much the two have changed.

Even when both approaches are used to deduce a simple branching diagram
of relatedness, or a phylogeny, the answers differ. It is rarely possible
to test the accuracy of different methods against a known phylogeny, so
the evaluation of techniques often rests upon a priori convictions. One
of the few tests conducted so far was reported by Walter Fitch and Atchley
in the mid-1980s. They used 10 genetic strains of mice developed in laboratories
that recorded both the actual phylogeny and the time of divergence. They
studied 97 genetic loci in the molecular analyses, 15 measurements of the
lower jaw, and seven life history traits, such as litter size, birth weight,
and adult body weight, in the morphological analyses.

Their results were striking. Simply put, the molecular data yielded
the correct phylogeny, whereas the morphological data did not. This difference
might be attributed to the fact that many more molecular loci were used
in the analysis than morphological landmarks. However, Atchley’s more recent
studies show that the concordance between the known and predicted phylogenies
improves little or actually deteriorates if more morphological points are
used. More probably, it is a matter of scale. As Atchley comments: ‘It is
largely the difference between looking at structural genes and regulatory
±ð±ô±ð³¾±ð²Ô³Ù²õ.’

The structural genes – coding for proteins and the like – accumulate
many small, apparently neutral mutations that are the shuffling footsteps
of genetic evolution. In contrast, very small genetic changes in regulatory
genes can produce great leaps in the timing and rate of development of bones
or guts or muscles, significantly changing what an animal looks like. Perhaps
the morphological changes that these strains of mice had undergone in a
mere 70 years were trivial, certainly too little to produce a species divergence.
It is not really surprising, Atchley maintains, that morphological measures
of divergence might be insensitive to changes that approximate evolutionarily
meaningless, random fluctuations within a species.

If molecular data are to be more useful for deducing phylogeny – a contentious
point not all biologists accept – then how useful are such data in addressing
other types of evolutionary questions?

A fascinating paper on just this issue, by Robert Wayne and Blaire Van
Valkenburgh of The University of California at Los Angeles and their colleagues,
appeared later in 1989 in the Journal of Heredity. This team took what may
be the first truly integrated approach to solving a classic evolutionary
problem: explaining how two similar species living in the same place avoid
competition. The standard mechanism invoked is ‘character displacement’,
an exaggeration or enhancement of ecological differences when species co-exist.
Somehow, the non-neutral, genetic changes that occur within two co-existing
species must get translated into behavioural or ecological differences that
lessen the competition. But how does this happen? And what would the molecular
approach have to contribute to such a question? The team’s findings make
clear the immense advantages of cross-fertilisation from different fields
of evolutionary biology.

Van Valkenburgh is a morphologist specialising in the evolution of guilds,
or groups of animals with the same general ‘occupation’. She has been studying
how carnivore guilds evolving together in the same area divide up the available
ecological resources. For evidence, Van Valkenburgh looks at differences
in body size, and in the detailed structure of teeth, skull and limbs. One
of the communities Van Valkenburgh has analysed, the carnivore guild inhabiting
the Serengeti Plains of East Africa, shows a puzzling anomaly. This guild
includes three jackals – Canis adustus, the side-striped jackal, C. mesomelas,
the black-backed jackal, and C. aureus, the golden jackal – that are very
similar in size and habits. They differ most obviously in the colour of
their coat.

‘Jackals just stood out as oddballs,’ Van Valkenburgh says. ‘You just
don’t see animals that similar in size and that similar in adaptation co-existing.
I couldn’t explain it. I thought maybe they hadn’t enough time to diverge.’

She discussed this enigma with Wayne, who was working at the time with
Stephen O’Brien at the National Cancer Institute, using molecular approaches
to study the genetic divergence of canids, the group that includes dogs,
wolves, jackals, and foxes. To create a truly integrated approach to the
jackal problem, they also enlisted Pieter Kat, a geneticist at the National
Museums of Kenya, and two ecologists, Todd Fuller and Ward Johnson, from
the Minnesota Department of Natural Resources and Iowa State University,
respectively.

To help them to understand what was going on with the co-existing jackals,
they decided to look at two other sets of canids that were also sympatric,
each set being two or more species that inhabit the same area. They already
had useful data on three canids that are sympatric over much of the western
US: the coyote, Canis latrans, the red fox Vulpes vulpes, and the grey fox
Urocyon cinereoargenteus. Van Valkenburgh’s data indicate that these North
American canids differ widely in body size, and the structure of their teeth
and skulls reflect differences in their diets. Wayne’s studies of the canids’
enzymes and chromosomes concur: he concludes that this trio are only distantly
related to one another.

Would those sorts of differences be found among other sympatric groups
of canids? In Patagonian Chile, the team studied and trapped two sympatric
foxes: the culpeo fox, Dusicyon culpaeus and the grey fox, D. griseus. There
were no canids at all in South America before the Panamanian isthmus closed
between 1 and 2 million years earlier, so they knew that some ancestral
fox must have migrated from North America once the isthmus had closed and
had then diverged into these two species. This knowledge served as a check
on the reasonableness of the date of divergence predicted by the molecular
analysis they carried out. Finally, the three East African jackal species
were themselves trapped and studied.

The team studied diet, use of habitat, various aspects of the ecology
and behaviour of the animals they captured in Africa and South America,
and took blood samples and many physical measurements before releasing them.

‘Ecologists rarely take blood samples for genetic analysis,’ Wayne remarks,
‘but I don’t understand it. It’s so easy to do, once you’ve already captured
an animal to put on a radiocollar or weigh it. The samples can be used either
to check the genetic relatedness of specific individual animals who are
being studied – which would be invaluable information for behaviourists
– or for larger-scale studies like ours. All you have to do is stick the
sample in a freezer until someone can deal with it. Why anaesthetise an
animal twice?’

The researchers analysed the mitochondrial DNA (mtDNA) in the blood
samples to give a measure of the time since the species diverged. Unlike
nuclear DNA, which is a mixture of the mother’s and father’s DNA, mtDNA
is passed intact from a mother to her offspring. In turn, her daughters
pass it on intact to their offspring, and so on. So, changes in mtDNA document
the mutations that accumulate in the females of an evolving lineage.

Work by Douglas Wallace at Emory University and the team of Rebecca
Cann, Mark Stoneking and Allan Wilson, then all at the University of California
at Berkeley, suggests that the percentage of changes in the mtDNA between
two forms can be used as a rough clock to determine divergence times. Cann
and her colleagues calculate that, on average, mtDNA will accumulate divergences
of only between 2 and 4 per cent in 1 million years.

What Wayne’s group found was that in North and South America the sympatric
canids exhibit classic signs of ‘character displacement’. The overlapping
species differ by a factor of two or more in body size and their skulls
and teeth reflect different dietary strategies. For example, the two South
American foxes weighed an average of 10.2 kilograms (culpeo fox) and 3.8
kilograms (grey fox), which means they would probably take different-sized
prey. Their canine teeth are adapted for different strategies of taking
prey and their cheek teeth suited for different diets. Because the mtDNA
of the two foxes differs by only 1 per cent, Wayne’s team concludes that
these marked morphological differences could have evolved in 250,000 to
500,000 years.

The jackals told a very different story. Their weights are completely
overlapping, with an average for each species between 6 and 7 kilograms.
Further, Fuller’s ecological study shows only minor differences in food
preferences and daily activity patterns and their skulls and teeth reveal
extremely similar adaptations. The surprise was that the mtDNA indicated
that the jackals diverged from one another between 2.3 and 4.5 million years
ago. In short, the jackals showed minimal character displacement despite
the fact that they had had 10 times as long to evolve differences as the
South American foxes.

If recency of divergence did not explain the similarity among the jackals,
what did? It was not that the jackals had only recently become sympatric,
because the fossil record shows that they occupied the same geographic area
at least 2 million years ago. The behavioural and ecological differences
are too subtle to be a fully effective means of avoiding competition.

Peaceful coexistence

In the end, the team proposed two explanations for the lack of apparent
character displacement among jackals. First, food, which is usually a limiting
resource in cases of competition, may not be an acute problem. The abundance
and unusual diversity of potential prey for jackals in East Africa may blunt
the knife edge of competition. Were food in shorter supply, the three jackal
species might well be unable to co-exist for long.

Secondly, the most common mechanism for avoiding competition, divergence
in size, may be an impossible option for these jackals. Although canids
obviously have the genetic potential to develop a wide range of body sizes
– as domestic dogs demonstrate so dramatically – these jackals are in an
unusual situation. The Serengeti Plains boast one of the densest concentrations
of different predators anywhere in the world. There are at least 20 carnivores
smaller, and seven larger, than the jackals in East Africa. Further, the
fossil record shows that this ‘packed’ guild of predators has a long evolutionary
history going back several million years. In contrast, in Patagonian Chile
where the foxes live, there is only one larger and four smaller species
of carnivore. Before the Panamanian isthmus closed, the predator guild of
South America was even sparser. So the foxes had more evolutionary ‘room’
to diverge than the jackals do.

Beside providing a striking explanation for an evolutionary problem
concerning jackals, this study shows much more. It reinforces the notion
that morphology reflects the genotype only to the extent that ecological
factors permit. In the case of these three species of jackals, it is not
just the habitat but the entire guild of sympatric carnivores that constrains
the morphological evolution.

Neither the molecular data nor the morphological and ecological data
would be as revealing if they were considered in ignorance of the other.
An ecologist, knowing nothing of the divergence data or quantitative measures
of morphology, might postulate that the small-scale differences in the use
of habitat or ‘activity patterns’ were somehow sufficient to avert competition.
It would be a rare and long-term ecological study, indeed, that would highlight
the entire carnivore guild as a potent evolutionary force shaping the jackals’
adaptations. A morphologist, ignorant of the molecular data, might postulate
– as Van Valkenburgh did initially – that the species simply had not had
enough time to diverge. A geneticist faced with only this information might
simply conclude that the species were really very different – which they
aren’t – and, indeed, might fail to perceive that there was a problem to
be explained at all.

In the 1940s and 50s, the ‘new’ evolutionary synthesis, which incorporated
population genetics into evolutionary thought, was forged. Maybe it is time
to try for a new, new synthesis, that will integrate whole organism and
molecular biology into a powerful new tool with which to examine our planet
and its history.

Pat Shipman is a science journalist based in Maryland in the US.