David Reznick talks with
ScienceWatch.com and answers a few questions about
this month's Emerging Research Front Paper in the field of
Article: Effect of extrinsic mortality on the
evolution of senescence in guppies
DN;Bryant, MJ;Roff, D;Ghalambor, CK;Ghalambor,
Journal: NATURE, 431 (7012): 1095-1099 OCT 28 2004
Addresses: Univ Calif Riverside, Dept Biol, Riverside, CA
Univ Calif Riverside, Dept Biol, Riverside, CA 92521
Calif Inst Arts, Sch Crit Studies, Valencia, CA 91355
Colorado State Univ, Dept Biol, Ft Collins, CO 80523 USA.
This paper was a comparative study of aging in fish derived from natural
populations that experience differences in mortality rate. I looked at
populations that either do or do not live with predators and have shown in
earlier work that those that live with predators suffer higher mortality
rates. Evolutionary theory predicts that they should have a higher rate of
My study is probably cited because it is a good example of testing
predictions of evolutionary theory in natural populations, but more so
because the outcome of the experiment was the opposite of what the
classical evolutionary theories predict. My results thus provide a good
incentive to consider the much larger diversity of theory that is now
available for how the aging process should evolve, since some of these new
ideas make predictions that are consistent with what I found.
Does it describe a new discovery, methodology, or
synthesis of knowledge?
"My research shows in various ways
that evolution is a fast process that can
have large impacts on our day-to-day
The results are a new discovery in the sense that they are probably the
best available documentation of the evolution of the aging process in a
fashion that departs so dramatically from the predictions made by Medawar
and Williams during the 1950s.
Would you summarize the significance of your paper in
Two classic papers, one by Peter D. Medawar ("An unsolved problem of
biology," H.K. Lewis & Co., London, 1952) and one by George C. Williams
("Pleiotropy, natural selection, and the evolution of senescence,"
Evolution 11: 398-411, 1957) made similar predictions for how and
why lifespan should evolve. They predicted that organisms which experience
low extrinsic mortality rates, meaning a low risk of mortality attributable
to external factors like predators or disease, will evolve longer
lifespans. These predictions have prevailed through all subsequent
discussions of lifespan and why some organisms live longer than others.
We tested these predictions in populations of guppies from natural
environments, in which they experience very different mortality rates,
because they either do or do not live with predators. Our results were the
opposite of what these classical theories predict. Guppies derived from
sites where they co-occur with predators, and where we had already shown
they experience much higher mortality rates, lived longer and reproduced
more successfully into advanced ages than guppies from sites where they do
not live with predators and have much greater life expectancies.
How did you become involved in this research and were
any particular problems encountered along the way?
I am an evolutionary ecologist and I began my research on guppies some 30
years ago. I chose them because prior research had demonstrated that there
were natural differences among populations in the risk of predation. They
live in mountain streams on the Caribbean island of Trinidad that are steep
and have many waterfalls; the waterfalls are often barriers to the upstream
dispersal of predators but not guppies, so I can often find populations
found close to one another in very similar habitats that have very
different mortality rates. I also chose guppies as they are a good organism
for laboratory research because they are so easy to rear and breed.
I wanted to apply predictions from evolutionary theory to the study of
natural populations and to do experiments in nature. I have done so,
sometimes treating streams as if they were giant test tubes by introducing
guppies or predators to a section of stream where they were not previously
found. In this way I have been able to manipulate mortality rates and study
the evolution of life history traits, including age at maturity, the number
of offspring produced, offspring size, and reproductive effort.
I have also been able to quantify the intensity of selection and rate of
evolution in nature. My results were among the earliest to show that the
rate of evolution that we see in real time can be many thousands, even
millions, of times faster than had been inferred as rapid evolution in the
fossil record. My colleagues joke that my choice of study has followed my
own development, since offspring size and maturity came first, then
reproductive effort and finally senescence.
Where do you see your research leading in the
Our unusual results suggested that there were environmental interactions
with the presence of predators that were modifying how guppies evolved in
response to predators. For example, while predators kill guppies, they also
lower population density and increase resource availability for the lucky
survivors. It is this sort of interaction that can account for our results.
We now have a very well-funded, multi-investigator grant to study the
interaction between ecological and evolutionary processes that was
suggested by the results of this paper.
"This paper was a
comparative study of aging in
fish derived from natural
populations that experience
differences in mortality
For the record, the study of such interactions is an emerging
subdiscipline. Ecology and ecological theory are traditionally done with
the implicit assumption that the organisms that participate in ecological
interactions do not evolve. The reason this is done is that it was assumed
that the rate of evolution is so much slower than the rate at which
organisms interact that it could be safely ignored.
My results show that this is not the case. This means that interacting
organisms can change their environment and that the changing environment
can, in turn, alter the kind of selection that these organisms are exposed
to, such that the outcome is not what would be predicted from traditional
ecological theory. The reason these interactions are of possible importance
is that including them can improve ecology as a predictive science.
Do you foresee any social or political implications for
My research shows in various ways that evolution is a fast process that can
have large impacts on our day-to-day lives. This is not new, since we
already know about the evolution of insecticide or antibiotic resistance,
but this work shows that the effects are more pervasive. For example, my
work has helped to show that the commercial exploitation of fish as a
source of food has almost certainly caused the fish to evolve, even over
the course of decades. The kinds of evolution that have occurred have in
turn caused them to be less useful as a source of food because they mature
at an earlier age and smaller size and often have lower growth rates.
With regard to aging, we have discovered natural populations that differ in
the rate of aging. They provide excellent material for studying the
genetics of the aging process that is different from the sorts of organisms
that are now used for this purpose. We currently study the genetics of
lifespan and aging in lab lines of model organisms like mice, fruit flies,
nematodes, and yeast that are the product of long-term laboratory culture
and are often screened for new mutations that affect lifespan.
We can instead offer organisms that have naturally evolved differences in
lifespan, which means that these differences have evolved under the
constraints of success in the real world. The differences in conditions may
mean that the genetic mechanisms that cause longer lifespan are entirely
different from those that are discovered in laboratory studies.
Characterizing them may then give us very different ideas about how and why
aging occurs. No one has pursued work on guppies from this perspective, but
perhaps that lies in the future.
David N. Reznick, Ph.D.
Professor of Biology
University of California, Riverside
Keywords: extrinsic mortality, evolution of senescence in
guppies, predictions made by Medawar and Williams during the 1950s,
mountain streams, Trinidad, waterfalls, barriers to the upstream
dispersal of predators, interaction between ecological and evolutionary