Podcast: Gravity’s big G and the evolution of babies

[Jingle]

Host: Adam Levy

Welcome back to the Nature Podcast. This week, we’ll be finding out about the evolution of adorable mammal babies.

Host: Shamini Bundell

Plus, new tests to try to pin down the strength of gravity. I’m Shamini Bundell.

Host: Adam Levy

And I’m Adam Levy.

[Jingle]

Interviewer: Adam Levy

What do puppies, kittens and baby humans have in common? Yes, that’s right - they’re ridiculously cute. But aside from that, they are also baby mammals. Mammals split from reptiles at least 300 million years ago, and while we know a fair bit about how we evolved from our scaly common ancestor to our cuddly present form, we don’t know a lot about the evolution of our babies.

Interviewee: Eva Hoffman

Along the line leading to modern mammals, we don’t have a lot of concrete information on how the reproductive strategies evolve.

Interviewer: Adam Levy

This is palaeontologist Eva Hoffman. Mammals’ reproductive strategy - in other words, how we have babies - is pretty special. We mammals are nourished as infants by lactation from mammary glands, where the name mammal comes from, in fact. And for the most part, we are born as live young rather than hatching from eggs. We’re also born in small groups. Most humans are born in a litter size of just one, of course. In this week’s Nature, Eva is presenting fossil evidence from an ancient mammal relative. This approximately 180-million-year-old find may help us understand a little better how our adorable young came about. I called Eva up and we started out talking about what’s been missing from our understanding so far.

Interviewee: Eva Hoffman

Mammals are very, very different in this way, from the primitive condition that we see still in many living reptiles, and we don’t know how that came to be. We haven’t had even the skeletons, even the shape of babies or of eggs of fossil mammals.

Interviewer: Adam Levy

So, you now have a find which fills in, to some extent, this gap. Can you explain where and how this was come across?

Interviewee: Eva Hoffman

The remains were collected almost 20 years ago by a University of Texas expedition. The specimen was not a particularly impressive one at first. There were some bones of an adult member of the species Kayentatherium wellesi lying on the hillside. It wasn’t until almost ten years later that a preparator noticed that among the large bones of the specimen that had been collected were additional tiny jaws and teeth. When those specimens were very quickly CT scanned, it was determined that these tiny jaws and teeth were a member of the same species as the larger individual. Because they were the same species, except on order of magnitude smaller, it was inferred that these had to be a mother or an adult found in association with her babies. But still, the CT scans at that point, they were still of low enough resolution that it was difficult to visualise the tiny, delicate, fragile baby bones.

Interviewer: Adam Levy

So, this mother and babies aren’t actually mammals, but they are early ancient mammal relatives, and so could in theory help us understand how mammal babies evolved. When you came to investigate this find with modern, high-resolution CT scanning techniques, what did you uncover?

Interviewee: Eva Hoffman

Well, the two major discoveries were that first, there were dozens and dozens of these tiny babies. So, I counted at least 38 of the babies, which is more than twice the mammal with the largest number of babies which is the tailless tenrec. And the idea is that probably there was little maternal investment of energy in each of these babies, but there’s a large number of babies in the hope that a few of them would survive to adulthood. And then in association with that, the other major discovery was that even though this specimen is very close to the origin of modern mammals, what we see in the babies is that their skulls are very similar in shape to the skulls of the adult, which is something which we see in reptiles which also have small brains, but is very dramatically transformed in mammals, which start out with round globular heads as babies. We don’t see that here. Instead, we have the primitive, the ancestral form of growing a head, again very close to the origin of mammals.

Interviewer: Adam Levy

Is this surprising given what we know about mammals’ history?

Interviewee: Eva Hoffman

I think it is surprising that in an animal that’s very closely related to mammals and has been modified in many ways to look like more like mammal and shares many characters with modern mammals, is still reproducing and developing in this primitive way.

Interviewer: Adam Levy

It’s such an interesting find. I mean, have you personally come across something which is like this before?

Interviewee: Eva Hoffman

I think nobody has, and one of the reasons that it’s so surprising, or one of the things that’s so surprising about it is that of course these are baby bones, they’re tiny and fragile, but preserved here, not as a flattened mush but in three dimensions, the bones of the babies which are so tiny.

Interviewer: Adam Levy

Could you describe what these animals look like?

Interviewee: Eva Hoffman

I would have to describe Kayentatherium as a mishmash of modern mammals. It was pretty large, so it’s size was about the size of a medium dog. I would say that in overall appearance they were probably sort of a possum-like, so kind of semi-sprawling, a kind of generalised looking mammal-like creature with a long tail, and they would have chewed on vegetation and been gnawing on things, most likely.

Interviewer: Adam Levy

Okay, so we know what this ancient mammal relative looked like, and what it’s babies looked like, but what about the questions about mammal reproduction that we still don’t have answers to?

Interviewee: Eva Hoffman

One example is that we think that these babies were either still in their eggs or they had just hatched. Of course, we don’t think of mammals laying eggs, even though the most primitive living mammals do lay eggs, and so one gap that needs to be filled in is how and why mammals went from egg-laying to viviparity which is giving birth to live young. A second thing that we’d like to know about how mammal reproduction evolved is when lactation and suckling arose. Probably somewhere around here in the mammal family tree, but we don’t yet know where.

Interviewer: Adam Levy

That was Eva Hoffman, who was at the University of Texas when she was working on this study, and is now at Harvard in Cambridge, in the US. Give Eva’s study a read over at nature.com/nature.

Host: Shamini Bundell

Still to come in the news, a new reproducibility study turns its sights on the major journals, including Nature. That’s at the end of the show. But right now, it’s time for the Research Highlights, read this week by Anna Nagle.

[Jingle]

Interviewer: Anna Nagle

Ever gone for a ride on a bubble? No, me neither. But now researchers have created capsules that can propel through water thanks to the buoyancy of oxygen bubbles. These so-called ‘proto cells’ are only about 0.3 millimetres across, so you’d have to be quite a lot smaller to catch a lift in one, but they are able to perform some impressive feats. 225 of them together were able to lift the weight of a dialysis bag in a tank of water. The capsules are made out of DNA and clay. Inside, they contain an enzyme that converts hydrogen peroxide into oxygen and water. Exposing them to hydrogen peroxide creates a trapped oxygen bubble, propelling the capsule to the surface of the water tank. Float on over to Nature Chemistry for more.

[Jingle]

Interviewer: Anna Nagle

Children undergoing treatment for brain cancer may find it harder to form new memories. Researchers spoke to children who had received radiotherapy and chemotherapy for a particularly common type of tumour. They asked them about two memories – one from the previous month, and one from the distant past. They found that the kids could recall the distant memory just as well as a control group. But the children provided many fewer personal details for the recent memory. This suggests that the treatment may be suppressing new neurons from growing in the hippocampus. For the full study, head over to the Journal of Neuroscience.

[Jingle]

Host: Shamini Bundell

It might sound bizarre to anyone who’s ever dropped a hammer on their toe or fallen out of bed, but gravity is actually incredibly weak. And it’s not only the weakest of the fundamental forces of the Universe, but it’s proven to be the most difficult to precisely measure. For over two centuries, scientists have used clever instruments to try and put an exact number on the gravitational constant - known as Big G. Sadly, the highest and lowest figures from labs across the globe still remain 0.5% apart - a big number, in the precise world of physics. This week, Jun Luo at the Huazhong University of Science and Technology or HUST, along with his team, have thrown new, precise measurements into the mix, with not one, but two experiments, each with slightly different experimental setups. The hope is that eventually the community can use these new measurements to untangle the cause of the international discrepancies. Stephan Schlamminger from National Institute of Standards and Technology in the US, spoke to Nature’s very own Big G, or Geoff Marsh as he’s usually known, to give his take on the new results and what they mean for science.

Interviewee: Stephan Schlamminger

So, Big G is the Newtonian gravitational constant and this is a constant in the law of gravity and it describes how strong gravity really is. Gravity is always between two bodies - we call it in physics an interaction. So, there’s always two partners that take part in the interaction. And for most of our daily experience, one of the partners is the Earth, and the Earth is so large and so massive that the force is actually quite large. But you can also have forces between two laboratory-sized objects or two wine bottles on a dinner table a metre apart, then the force will be roughly the weight of a human cell, and of course it’s very difficult to measure these weak forces in the presence of the large force that we have downward towards the Earth.

Interviewer: Geoff Marsh

So, it sounds like this kind of experiment demands really impressive craftsmanship, and that’s what we’re here to talk about today, is this team in China that have designed not one, but two experiments to measure gravity. Tell me about this team and what it is that they’ve managed to do.

Interviewee: Stephan Schlamminger

Yeah, I think you’re right. I mean, it’s an absolutely amazing human feat to pull of these two measurements, and I think the team has been working on the gravitational constant at least since 2000, probably 20 years or so or at least 18 years that they have been working really hard to measure gravitational constant. So, the experiment that they used is called a torsion balance, and what it is, it’s just like a plumb bop. It’s a thin string with a weight at the end, and in their case the weight was a rectangle. And this rectangle rotates now back and forth because the string twists. So, they have done two experiments. In one, they measured how fast it rotates back and forth, so the string had to twist in order to do that. On the other one, they used a feedback system to make it such that the string would not twist. And that’s the main difference between these two experiments - that one has the properties of the string, this fibre, in the result; the other one doesn’t.

Interviewer: Geoff Marsh

Okay, so if the twisting of the string is what’s been confusing the results, then the two experiments would give very different results presumably. But if they matched each other, then probably the string isn’t what’s being causing all the inconsistent numbers people have been seeing. So, what happened when they finally did compare the two?

Interviewee: Stephan Schlamminger

The results, they agree reasonably well. I mean, there’s not perfect agreement, it could be better, but it’s okay. I think the more troubling fact is that the new result, this agrees with an older result they published in 2010.

Interviewer: Geoff Marsh

And what about other results from around the world?

Interviewee: Stephan Schlamminger

The Big G measurement suffers from a problem, and that problem is that we have about 16 measurements worldwide right now, and the best measurement is the newly reported measurement from the group at HUST. It has an uncertainty of 11 parts per million, so this is the world record, 11.6. But the scatter between the other published value, the largest and the smallest number, they differ by about 500 parts per million, so this is, what is it, a factor of 40 or so larger than this published result, and that’s something we need to understand as a community.

Interviewer: Geoff Marsh

So, what we’re saying is this new result from HUST has the lowest level of uncertainty about its measurements of G, but when we compare that number with the other numbers from previous laboratories around the world, the scatter is so great that there’s clearly something that we’re not understanding.

Interviewee: Stephan Schlamminger

That is correct.

Interviewer: Geoff Marsh

What are we not understanding? You must have a hypothesis.

Interviewee: Stephan Schlamminger

I’m glad you asked. I mean, in principle there’s two possibilities I would say. So, it could be a mundane, technical effect, right. So, we don’t yet understand a property of these experiments, like, you know, where’s the masses, or whatever the property is of the fibre. But the more exciting property is that we don’t understand gravity, and there’s something fundamental about gravity that we have not built in to our measurements. I mean, we had the Newtonian picture of gravity and then general relativity came along, and general relativity has passed all the tests with flying colours. There’s nothing that leads us to doubt general relativity, so we think it’s the right theory, but with all these theories, you can never prove them, you can only falsify them. So, if you have experimental data that is inconsistent with the theory, you may be able to falsify and bring doubt on to this. And the Big G data right now is inconsistent and it may be due to a mundane, technical problem that we all don’t understand, or it could be new physics.

Interviewer: Geoff Marsh

Do you think people are going to continue persevering for more and more precise measurements of G?

Interviewee: Stephan Schlamminger

So, I think right now, the situation in Big G is this: that we are over-measuring and under-understanding. So that means we have 16 data points, there’s 2 new data points that have been added and the data points are really precise and it’s a great achievement, but we still don’t understand the whole picture. So, I think if you now would just add another data point, I don’t think that would clarify the situation. I think at this point, what we have to do is we have to go back and look especially at the experiments that are outliers and say why are they different or why do these experiments produce different numbers and try to understand this. You know, we have to catch up with the experiments by understanding these different results that they produce.

Host: Shamini Bundell

That was Stephan Schlamminger from the National Institute of Standards and Technology in the US, speaking to Geoff Marsh. Find the study and Stephan’s News and Views at nature.com/nature.

Interviewer: Adam Levy

Finally this week, it’s time for the News Chat, and reporter Holly Else joins us in the studio. Hi Holly.

Interviewer: Holly Else

Hi Adam.

Interviewer: Adam Levy

Now, for our first story for the week, we have a new reproducibility study. This is something we’ve spoken about a lot on the podcast, but how is this new study different?

Interviewer: Holly Else

Yeah, so this is a study that looked at reproducing social science findings, and specifically those that have been published in the high-impact journals Nature and Science. So, it’s slightly different from the previous reproducibility projects which looked at psychology research and there’s one ongoing about cancer research.

Interviewer: Adam Levy

So why just these two big journals?

Interviewer: Holly Else

Well, the argument is that these are the top journals and so these are the findings that potentially have the most impact, might get reported in the media for example, or be used to inform policy. So, it’s important to understand whether they are correct.

Interviewer: Adam Levy

These are all, you know, quite involved studies. It must be a lot of work to actually go and reproduce not just one of them, but a whole bunch of these studies.

Interviewer: Holly Else

Yeah, it’s a huge amount of work. I think they had five teams working around the world to do this particular reproducibility study. You know, other ones that are going on at the moment - there’s one around cancer, finding 50 most significant cancer findings, trying to reproduce those. They actually announced just a few months ago that they’d be reducing the size of their ambition there and go from 50 to 18 because they’re finding it so difficult to reproduce studies, and it takes so much time and costs so much money.

Interviewer: Adam Levy

And in this new reproducibility study can you give some examples of the kinds of studies they were aiming to reproduce?

Interviewer: Holly Else

Yeah, so one of the ones that they looked at was whether there’s any correlation between analytical thinking and your religious scepticism, and the other one looked at exam performance.

Interviewer: Adam Levy

Okay, so those are the kinds of studies they were looking at, but when they came to trying to reproduce the studies what did they find?

Interviewer: Holly Else

So, they found that they could replicate the study in 62% of cases.

Interviewer: Adam Levy

And is that a lot? How does that compare to other reproducibility studies?

Interviewer: Holly Else

So, that’s higher than the psychology reproducibility study from 2015, which found only a third to half of the 100 studies they looked at were reproducible. Now this is a much smaller sample, so it’s only 21 studies of which they’ve managed to replicate - 62%.

Interviewer: Adam Levy

So, a higher percentage, but it still suggests 2/5 might not stand up to scrutiny.

Interviewer: Holly Else

Yeah, that’s exactly right. And what they did do when they were reproducing each of these studies, was increase the sample size to try and make sure that the outcome they had, they were confident in.

Interviewer: Adam Levy

And one thing about this study that I’ve not seen, I don’t think, in any other reproducibility studies, is that they used something called a betting market.

Interviewer: Holly Else

Yeah, this is something that’s really interesting. So, they assembled a panel of up to 80 researchers for each paper, and gave them shares where they could bet on whether they thought each finding would be reproducible or not. And they found that overall, this group were able to judge with some degree of accuracy which studies were reproducible and which weren’t.

Interviewer: Adam Levy

And going forwards, apart from maybe having betting markets, what can be done to improve things, to improve the number of papers getting published that end up being reproducible?

Interviewer: Holly Else

Yeah, so one of the things they actually did for this particular piece of research, was go back to the original study authors and check the protocol to make sure that they were actually doing exactly what they had done. That’s something everyone can do to make their work more reproducible - you know, have a clear protocol that’s easy for someone who hasn’t been involved in the original research to follow.

Interviewer: Adam Levy

Let’s turn to our second story of the week, and the British Heart Foundation have announced a new funding pot. The plan with this funding pot is to give it to just one research project, but this is a huge amount of money.

Interviewer: Holly Else

Yeah, that’s right. So, this is £30 million, which is about US$39 million, going to one specific team who are going to address some project related to cardiovascular health, and what’s known as the Big Beat Challenge.

Interviewer: Adam Levy

And do we know anything more about the kind of project they’re looking for at the moment?

Interviewer: Holly Else

Well, we know that they want it to be something revolutionary and something that can really impact on people’s quality of life.

Interviewer: Adam Levy

Would that mean any particular discipline or are they kind of open to all different walks of science?

Interviewer: Holly Else

Open to anybody really, as long as they are tackling cardiovascular disease. And I think the project really wants to encourage multidisciplinary teams that involve experts from all different walks of life, whether that be artificial intelligence or economists, psychologists.

Interviewer: Adam Levy

What concerns are there about handing out funding in this way?

Interviewee: Holly Else

Well obviously, it’s a huge pot of money, and for it just to go to one team there’s a danger that it could go to somebody who is leading a team that is particularly well-established in a field, and sort of has a tried and tested line of research. Whereas perhaps a more junior researcher might have a more revolutionary idea but perhaps doesn't have the skills and experience to write the right proposal to get it funded.

Interviewer: Adam Levy

And I guess it’s always riskier for a funding agency to give money to a group who hasn’t done that much work previously.

Interviewer: Holly Else

Yeah, but they are saying that this is a “high-risk, high-reward” process, so this they’re really looking for those risky ideas. So, let’s see, I guess, if they put their money where their mouth is.

Interviewer: Adam Levy

Is it unusual then for the risk to involve this kind of amount of money?

Interviewee: Holly Else

I think so, yeah. I mean, there’s previously been a similar kind of grant in the States, where Google Life Sciences which are now Verily and the American Heart Association teamed up together to give a US$50 million prize, and then AstraZeneca actually added a further US$25 million, and then went to one team in the States who were looking at cardiovascular health. But the different with the British Heart Foundation grant is that it’s actually entirely open for anybody to apply with any idea they might have.

Interviewer: Adam Levy

And then when will we actually hear who’s getting this huge amount of money from the British Heart Foundation?

Interviewee: Holly Else

Well, applications open towards the end of 2018 and I guess then time will tell.

Interviewer: Adam Levy

Holly, thank you for joining us. For more on those two news stories and for the all the latest science news head on over to nature.com/news.

[Jingle]

Host: Shamini Bundell

That’s it for another Nature Podcast. But, if haven’t already, make sure to tune into Backchat – a behind the scenes look at how we report on science. This month’s show looks at the strengths and weaknesses of audio versus print, as well as other topics that we’re slightly less biased on. Until next time, I’m Shamini Bundell.

Host: Adam Levy

And I’m Adam Levy. Thanks for listening. [Jingle]