Nature Podcast

This is a transcript of the 24th January edition of the weekly Nature Podcast. Audio files for the current show and archive episodes can be accessed from the Nature Podcast index page (, which also contains details on how to subscribe to the Nature Podcast for FREE, and has troubleshooting top-tips. Send us your feedback to


The Nature Podcast brought to you by Bio-Rad's 1000-series thermal cycling platform - when you rethink PCR, you think about how easy it can be... see how at

End Advertisement

Kerri Smith: This week, they don't call it the Advanced Research Projects Agency for nothing.

Charles Herzfeld: Almost everything we did was an improvisation of some sort. We were really at the edge of the known and with a foot well into the unknown.

Kerri Smith: As DARPA turns 50, we get the inside story from a former director.

Adam Rutherford: And a 30-year-old theory of sex determination is finally tested with some reptilian research.

Dan Warner: There are extremely hyperactive skiddish little lizards, not easy to catch in the field, they are very quick, often times run bipedally.

Adam Rutherford: This is the Nature Podcast. I'm Adam Rutherford

Kerri Smith: And I'm Kerri Smith. First this week, a hoard of old data helps researchers find out about Old Man River. Here's Charlotte Stoddart.

Charlotte Stoddart: The Mississippi river system drains two-thirds of the United States. As its numerous tributaries wind their way to the river's mouth in the Gulf of Mexico, they carry with them bicarbonate ions, an alkaline form of inorganic carbon. The bicarbonate comes from atmospheric CO2, which becomes a weak acid when dissolved in soil water. In turn, the acid reacts with minerals in the soil to produce bicarbonate, which rainwater washes into streams and rivers. A new study published in Nature this week finds that over the last 40 years, the amount of bicarbonate in the Mississippi has increased because of changing agricultural practices in the river's watershed or drainage basin. The researchers looked at tens of thousands of bicarbonate measurements spanning the last one hundred years. I called author Peter Raymond at Yale to find out how they got their hands on this impressive flood of data. Nature 451, 449–452 (24 January 2008)

Peter Raymond: I first became aware of the data set on a visit to Yale by Eugene Turner. He mentioned in sort of passing that he had thought he had seen some old records of alkalinity in some of the water treatment plants. Many cities when they take in river water, they add lime to protect their pipes which corrode over time, so if you add lime, you've to dig up your streets less, so in order to do this, they take a little Mississippi river water every day to measure its bicarbonate concentration. So I begged and pleaded with Eugene, it didn't actually take much for him to revisit one of these sites and lo and behold! The data set was in old data sheets archived in boxes sitting in a warehouse and we were able to go in and rescue that data and then the rest of the data is from the United States Geological Survey. So it's really a remarkable data set and we feel pretty luck to have saved it because we're not sure how much longer it would have remained in this warehouse.

Charlotte Stoddart: What did you find out when you looked at the data?

Peter Raymond: The first striking piece of evidence was that the concentrations of bicarbonate began to rise in the 1940s and rise pretty quickly in the 60s and 70s to concentrations much higher than they were for the first 40 or 50 years with the data.

Charlotte Stoddart: I understand that over this period there was also an increase in rainfall and more rainfall can also mean that more bicarbonates gets into the river system. So how much is increased rainfall, the reason why we see increased bicarbonate in the river system and how much is it due to other factors?

Peter Raymond: Right well, this was very surprising to us, so there is basically two ways you can increase the amount of water coming out of river and the first is to increase the amount of water coming into the watershed, so increased rainfall. And this is occurring in many places around the world and it is due to global warming, but the other way that you can increase discharge in a river, regardless of what happens to precipitation is, you can change some watershed factor that's important to routing water to a river. And what we found here in the Mississippi was evidence that agricultural practices have done this, so things like tile drainage and changes in crop productivity which have been drastic in the Mississippi in the last 60 years are causing a greater percentage of the rainfall that hits the watershed to make it to the river.

Charlotte Stoddart: Right, so in the last 40 years we have seen changing agricultural practices and these have increased the amount of bicarbonate in the river system and tile drainage is a kind of plumbing system for fields, so it is a method of removing excess water from agricultural land and this might be one of the kinds of practices that's responsible.

Peter Raymond: Yes, these agricultural-type practices have altered the hydrology, which probably isn't a massive surprise; I think, what's surprising is, is the scale that we're reporting. It's almost like finding a new river coming out of the corn belt from the United States.

Charlotte Stoddart: So, what are the consequences of all this additional bicarbonate in the river system?

Peter Raymond: Perhaps the most relevant thing currently is that bicarbonate does originate from the atmosphere. Globally this flux is relatively small, but it's still big enough to make it on to the global budgets that are produced and it is also one of these net long-term important fluxes. It is one of the fluxes that, if we were to stop pumping CO2 into the atmosphere would ultimately bring atmospheric CO2 concentrations back down to something recognizable in, say a 1000-year time scale.

Charlotte Stoddart: So over a very long period of time, this could be a way of, sort of mopping up the extra CO2 that is in the atmosphere from human emissions.

Peter Raymond: Once in the ocean, bicarbonate is a pretty safe place to put the atmospheric CO2 where it will reside for thousands to hundreds of thousands of years. On shorter time scale say the next 100 years in the problem we face, it's not really that important but on longer time scales it's a very important flux.

Kerri Smith: Peter Raymond from Yale University. And from old data sets to getting older ourselves. None of us really like to think about it, but we are all ageing. Demographers studying the effects of age on population size, usually just look at people's ages, now a team led by Wolfgang Lutz of the World Population Program at the International Institute of Applied Systems Analysis in Austria have used some new measures of population ageing that take into account how long people have left to live. Both new and old measures show that the speed of ageing is ramping up, meaning that population growth is slowing down. There is an 88% chance that growth will stop completely by the end of this century. I spoke to Wolfgang Lutz and first asked him to explain why ageing matters. Nature advance online publication (20 January 2008)

Wolfgang Lutz: Our society is organized very much along age, whether we enter kindergarten or school, this is all by the age group we are in. The same with the pension system when we come towards the end of our productive life, age greatly matters, but age does not have a constant meaning throughout human history. Somebody who was 60 years old in 1900 for instance was considered an old man or an old woman, today we don't consider a 60-year-old particularly old, we think more of an 80 or 90-year-old person as being old. So the meaning of age has changed with the increase in life expectancy and with the change of society.

Kerri Smith: And have demographers been changing their measures of ageing along with this or are there differences in the methods we've traditionally used and what you've done here?

Wolfgang Lutz: Well, almost universally age is defined and measured as time since birth and it is only very recently that we think of new measures that would also somehow incorporate the increasing life expectancy and these measures would not only take the time since birth into consideration, but also the expected time to death.. The problem is of course that we don't know at an individual level how long will I live and how long will you live, but on the population level we do have life tables; we do have statistical information that give us an average for the remaining life years and we can make adjustments for future improvement in life expectancy. And if we have these adjusted measures of ageing then the speed of ageing is somewhat slower because at the same time life expectancy is expanding in the future,. But, no matter what ageing indicators we are using, all these indicators show that over the coming 2 or 3 decades we will see an acceleration in ageing, which poses all kinds of challenges to our society and to policy makers.

Kerri Smith: And so in your new paper then you have factored in this increase in life expectancy but your figures for and your predictions for how quickly we are all going to age have stayed roughly the same as what you predicted before.

Wolfgang Lutz: Well, in our paper we actually addressed two issues, one is the growth in the world population and here the big news, 5 years ago has been in another paper in Nature published in 2001 was called 'The End of World Population Growth', here the main message was that indeed we do see end of growth, we do see beginning of some population decline at the global level towards the end of this century, but this end of world population growth comes at the expense, if you want of significant population ageing because the increasing life expectancy which is a very positive development together, we have declining fertility besides in the change in the age composition of the population that will lead to a higher proportion of elderly people and this has been the focus of our article how this speed of population ageing varies over time and across regions.

Kerri Smith: And what kind of timeframe are we looking at for those kinds of things, obviously it varies from region to region.

Wolfgang Lutz: Well in Europe it is the next 2 decades where the speed of ageing will be most rapid and here it is an immediate policy challenge. In other regions, the Middle East or Africa, it is still population growth that is the main policy challenge and ageing related concerns are of secondary order. If we look to Japan or other countries such as China that already have very low fertility now, the speed of ageing will be a major policy challenge also over the 2 to 3 decades.

Kerri Smith: And how then should we be preparing for these kind of changes in the population?

Wolfgang Lutz: Well, generally change that is very rapid is more difficult to cope; it is more challenging than slow change. So this is why policy makers need to be aware of when ageing will be the most rapid, so that they can make appropriate decisions ahead of time. These are decisions with respect to our pension system, the healthcare system and all other infrastructure, societal institutions that are sensitive to changing age distributions.

Adam Rutherford: Wolfgang Lutz. Time for the Podium now and this week Harold "Skip" Garner of the University of Texas, Southwestern Medical Center reveals the extent of ethically dubious goings on in the world of science publications.

Harold Garner: To paraphrase Dickens, "it is the best of times; it is the worst of times" in a world of biomedical publications. Productivity as measured by scholarly publication rate is in an all-time high, yet high-profile misconduct cases remind us that some of these publications are not to be trusted. The question is how many and which ones. The three major sins of modern publishing are duplication, co-submission and plagiarism. Some forms of repeat publication are both ethical and valuable to the scientific community such as clinical trial updates, conference proceedings and errata. The most unethical practices involve substantial reproduction of another study without proper acknowledgment. If such duplicates have different authors then those involved may be guilty of plagiarism. Papers with overlapping authors may represent self-plagiarism. With few exceptions, the repeated publication of the same results by those who conducted the research is ethically questionable. It artificially inflates an author's publication record, it places undue burden on journal editors and reviewers and it is expressly forbidden by most journal copyright rules. Monitoring trends is important for the health of the scientific literature, as the number of peer-reviewed journals has multiplied the perceived odds of unethical publications escaping detection has improved. Fortunately, new text searching algorithms plus electronic indexes and full text electronic manuscripts are making easier to detect unethical publications. One such text-searching tool is eTBLAST. It is available free online for authors, editors and referees to assess the novelty of work before or after publication. Using eTBLAST we have recently identified over 71,000 highly similar records, all of which have been deposited in a publicly available online database called Déjà vu. We estimate there could be more than 200,000 such duplicates in MEDLINE. Considering only those duplicates in Déjà vu where we've manually inspected the full text of both articles we found 73 new plagiarism candidates. Telling apart legitimate and illegitimate duplication is beyond an automated algorithm, so these cases must now be examined by an editorial board or university ethics committee. One argument for duplicate publication is that it makes significant works available to a wider audience especially in other languages, but only 20% of the manually verified duplicates in Déjà vu represent translations into another language. Another common belief is that the occasional referee may stall a publication to give themselves time to duplicate and publish the same results first. The Déjà vu database contains many pairs of highly similar abstracts with overlapping authors that appear in the same month; all apparently acts as simultaneous submission to multiple journals. The general lack of duplicates with different authors appearing in rapid succession suggest that deliberate scooping is either rare than feared or that the perpetrators do a good job of concealing it. So what can be done to police shady publication practices? Journal editors had more frequently used the new computational tools to detect duplicates and advertise that they do so. It is odd that high schools and universities regularly use automated text matching systems but journals do not. We hold our children up to a higher standard than we do our scientists. It would be fairly simple to fold these tools into electronic manuscripts with machine systems making it a ubiquitous aspect of the publication process. Then the fear of having some transgression exposed in such a public and embarrassing manner would be a very effective deterrent. Like Dicken's Scrooge, the spectre of being haunted by dubious publications past might be enough to get unscrupulous scientists to change their ways.

Adam Rutherford: Skip Garner on plagiarism in science. Skip has written a commentary on that subject in this week's issue and as with all the stories from the show that's available on our website Nature 451, 397–399 (24 January 2008)


End jingle

Kerri Smith: This week Nature nods in the direction of the US military's research arm, which celebrates its 50th anniversary this year. Here's Nature's chief commissioning editor Sara Abdulla.

Sara Abdulla: In 1957 the Sputnik launch blind-sided the US. In response, President Eisenhower setup what's now known as DARPA -- The Defence Advanced Research Projects Agency in January 1958. For the past 50 years, this small organization has revolutionized warfare and civilian life bringing us the Internet, stealth aircraft, seismic systems and much more. DARPA is considered a paragon of government innovation. Today, some question if the agency is still relevant, others are planning to ape its famously nimble approach in areas such as energy and homeland security. In this week's Nature in a centre feature explore the past, present and future of this unique organization and its copycats. I spoke to the authors of both articles starting with Charles Herzfeld, DARPA's director in the mid 60s. I asked him what the agency was like on the inside. Nature 451, 403–404 (24 January 2008) ; Published online 23 January 2008 Nature 451, 390–393 (2008)

Charles Herzfeld: The main thing about DARPA in the 60s was that it was very exciting, very demanding almost everything we did was an improvisation of some sort. We were really at the edge of the known and with a foot well into the unknown and everything moved very fast.

Sara Abdulla: And what were the major things that happened while you were Director.

Charles Herzfeld: Well, there were a number of things, the ballistic missile defense program was going very well and began to have a real impact. The Counter Insurgency Program which was mostly in Vietnam and Thailand was doing well and at that time the war in Vietnam was not going that badly and we had made our successes in Thailand. We helped the Thais very significantly defend their Northeast against the insurgency.

Sara Abdulla: So the ARAPANET is probably DARPA's most famous success. Tell us about a few others that you think are the highlights of the past 50 years of the agency's work.

Charles Herzfeld: I think clearly that the Arpanet had the most impact on society in general. One has to remember of course that the Arpanet had to leave to the Internet and other DARPA activity to become fully functional. The Arpanet was pretty weak in many ways. The second most important thing I would say that was an off shoot of the Counter Insurgency Program was the beginning of precision warfare where we found technology that enabled us to hit smaller targets accurately and reliably. Just as with the Arpanet, the internet that took several steps and the first round failed completely. Everything went wrong, but we did it again and a year later, it worked fine.

Sara Abdulla: How would you say that DARPA has evolved since Eisenhower founded it in 1958?

Charles Herzfeld: The main change was that it has become more orderly, so to say, which means somewhat less ambitious, it has become more supervised and is also in my view at least tends to limit originality. If people are worried that they have to defend their ideas before an unfriendly audience, they get very cautious.

Sara Abdulla: Journalist Sharon Weinberger has been investigating how the agency is faring now and the plants effort to clone it elsewhere. I started by asking her how she described today's DARPA.

Sharon Weinberger: Well DARPA today is still very much, what I believe former defence secretary Bill Perry called it "The Jewel of the Pentagon." What it's still very good at is moving funding, very quickly to, sort of, centres of excellence to the best scientists arguably to the best universities and to the most innovative companies because it doesn't have to go through a very bureaucratic process.

Sara Abdulla: And are there conversely types of research or types of policy influence that these days it struggles to achieve?

Sharon Weinberger: Well, I think one of the biggest differences between DARPA now versus DARPA perhaps 20 or 30 years ago is the question of its relevance to the Pentagon. You know during the Vietnam war it was the lead agency for counter insurgency it was the 'go to' agency if you had a problem, now I think the feeling from many in the Pentagons is just still a best place for science and technology innovation, but it may be that science and technology is no longer at the top of the list, but there is a feeling in some circles in the Pentagon that DARPA has perhaps isolated itself; sometimes on purpose to protect itself bureaucratically to protect its researchers, to protect its independence. But the cost of isolating yourself is that people in the Pentagon find other ways to get things done. I think perhaps that's the dangerous, sort of, thin ice that DARPA skates on and there isn't a perfect solution.

Sara Abdulla: Despite DARPA's slightly waning influence, there are moves to emulate its daring approach to innovation, what's your view Sharon of the planned DARPA like research agencies, IARPA for intelligence and others for energy and homeland security.

Sharon Weinberger: We've already seen a lot of criticism of this proposal, people saying that well congress and agencies are looking for an easy solution and they say because DARPA worked for defence and it worked very well, lets do it for X, Y and Z, but the conditions are different. So for example let's look at homeland security ARPA, already that agency is sort of distancing itself from the DARPA model saying well, you know we don't have the money; we don't have 3 billion dollars, we have may be 60 million; so we can't do very big projects. And but it is more than that. They are looking at very very short-term solutions. They don't necessarily fund the big ideas out there and so it's very different, much smaller in scope. For the IARPA I think there is a little bit higher hopes for that and one of the reasons is that there is a perceived similarity to DARPA and that there is a customer. For DARPA, they're producing things the military services will use. IARPA the hope is that they will fund things that the intelligence community could use, but there is also a great deal of criticism of it and then perhaps the most controversial of these kind of ARPA-cation is the ARPA energy, where people are saying you know, where is the similarity to DARPA? You know for DARPA or even an IARPA there is an end customer. For energy, you have this maze of private companies, of customers, of government and it is not clear, are you going to be picking winners in the economy, who are you going to transition it to? And another criticism is perhaps the question for energy is not one of a technological silver board solution, but a question of economy, what energy solution, be it solar, or something else, is economical to produce.

Sara Abdulla: So I asked former director Charles Herzfeld what advice he has for any DARPA copycat.

Charles Herzfeld: To start over, I think you really have to be clear what the very large problems are, that an organization has to solve. The second thing is you need really brilliant bright people who are very imaginative and rather gutsy. Wading into some of these swamps is dangerous and you've got to get people who are willing to take risks. The third thing is absolutely mandatory, I think, is new money and then finally you have to have the guts and the skill to defend it against sniping from within their organization because that will come for sure and there will be a lot of naysayers. I can spend that money better; look how they're wasting it. Everyone will say that and some of them will be right and many of them will not be and you've got to protect the new child until it can fight its own fist fights.

Kerri Smith: Charles Herzfeld and Sharon Weinberger on DARPA and the clones, which although it sounds like it is not a 70s pop act!

Adam Rutherford: Finally this week, striking evidence for a theory of sex determination that stubbornly defied testing since the 1970s. In many species, reptiles in particular, the sex of the offspring is not determined by their chromosomes but by environmental factors such as temperature. In the Australian Jacky Lizard, females are produced at high and low temperatures, whereas an intermediate temperature results in a mixed brood, but why would natural selection favour this method over genetics which invariably gives equal numbers of males and females. The theory has been around for 30 years, but it has taken brand new study by Dan Warner and Rick Shine at Iowa State University to empirically prove the model. I spoke to Dan. He started by explaining how he tweaked the sex of the Jacky Lizard offspring to see how their fitness is affected by the temperature of their clutch. Nature advance online publication (20 January 2008)

Dan Warner: During development, there is an important enzyme that's responsible for converting testosterone into oestrogen and that enzyme is called aromatase and a lot of studies have shown that aromatase activity is very temperature-sensitive, so this enzyme plays a very key role in sex determination. So what I did was, apply an aromatase inhibitor to the eggs and what this does is block the conversion of testosterone to estradiol and thereby producing male offspring even at female-producing temperatures and that allowed me to decouple the confounded effects of sex in incubation temperatures. I could look at how temperatures influence offspring fitness in a sex-specific, sort of, way.

Adam Rutherford: So tell us a little bit about the theory behind this, because this has been longstanding for 30 years.

Dan Warner: In 1977 Eric Charnov and Jim Bull proposed their model to explain when natural selection should favour environmental sex determination over genetic sex determination. And in terms of temperature dependent sex determination, which is a form of environmental sex determination, their model simply states that natural selection will favour temperature-dependent sex determination, when developmental temperatures during incubation will have a differential impact on the fitness of sons versus daughters. So for example temperatures that produce sons should be optimal for males and temperatures that produce daughters should be optimal for females and if this is indeed the case, then temperature-dependent sex determination should enhance parental fitness overall, because temperatures at which eggs develop at, will produce the sex that benefits most from those incubation temperatures.

Adam Rutherford: So your results fit perfectly with the theory, I take it.

Dan Warner: That's right. They fit exactly what the theory had predicted. It was a pretty amazing finding.

Adam Rutherford: Okay and so why has it taken so long to test the theory empirically?

Dan Warner: Well, there are numerous obstacles had to be overcome. The first difficulty in testing the model is that we need to evaluate how a variety of incubation temperatures influence fitness of both males and females. I mean, to do this we need to produce both sex across a wide range of incubation temperatures to compare males at one temperature, for example, to males at another temperature. So the fact that sex is determined by temperature, this is an obvious problem, so that's why we had to use very hormonal manipulation to the eggs to produce the wrong sex at a given temperature, for example. Because this technique wasn't always there, we had, you know, it took some time before we can actually test the model with the proper experimental design. But the second major difficulty is that most species with temperature-dependent sex determination are say turtles and crocodilians. These species live extremely long life spans over 60 years, sometimes over 100 years and they mature at a very late age. So it makes it very difficult for researchers to go out there and measure the lifetime fitness of the individuals. It wasn't until recently that temperature-dependent sex determination has been discovered in a lot of short-lived lizards that live in Australia. So these species that were recently discovered to have temperature-dependent sex determination make great models for investigating how incubation temperatures influence reproductive fitness throughout their lifetime. Another aspect that brought this study to another level was the fact that actually measured reproductive fitness whereas past studies that have addressed this Charnov-Bull model have always measured some sort of trait, phenotypic trait of the offspring that they assume to be associated with fitness, such as the body size of the offspring or may be that how fast they run, for example. What I did was, I did the incubation experiments, I released all the lizards into these field enclosures, raised them up for 3 years they can reach maturity within 1 year after they hatch. They started reproducing their eggs then I incubated those eggs of the second generation. Then I used genetic markers to assign those second generation offspring back to the fathers, back to the parents, so I was able to actually measure how many offspring each female produced and how many offspring each male sired. So I had a direct measure of reproductive success.

Adam Rutherford: And how was it actually working with these lizards? What are they like to handle?

Dan Warner: They are extremely hyperactive skiddish little lizards, not easy to catch in the field, they are very quick, often times run bipedally when they are running away from you they are extremely quick. The study was very labour intensive. Obviously I am raising over 200 lizards from eggs all the way to adults, so there is always the constant worry that some disease might wipe out my experimental populations or some feral cat might get in there and eat the entire population. It made life interesting working on these lizards.

Adam Rutherford: Sounds like a lot of fun you had.

Dan Warner: Yeah. It was a lot of fun.

Adam Rutherford: Dan Warner, the lizard king. That's it for this week's show.

Kerri Smith: Our Sound of Science brings you a common genetic technique as you've never heard it before. Courtesy of our sponsors Bio-Rad, here is an ode to PCR, I'm Kerri Smith.

Adam Rutherford: And I'm Adam Rutherford. This is the Nature podcast. Thanks for listening.

[Sound of Science]

There was a time when to amplify DNA,
You had to grow tons and tons of tiny cells.
Then along came a guy named Dr. Kary Mullis,
Said you can amplify in vitro just as well.
Just mix your template with a buffer and some primers,
Nucleotides and polymerases, too.
Denaturing, annealing, and extending.
Well it's amazing what heating and cooling and heating will do.
PCR, when you need to detect mutations.
(chorus -. detect mutation)
PCR, when you need to recombine
(chorus – recombine)
PCR, when you need to find out who the daddy is
(chorus – who the daddy is)
PCR, when you need to solve a crime
(chorus– solve a crime)


The Nature podcast is sponsored by Bio-Rad, at the centre of scientific discovery for over fifty years, on the web at

End Advertisement