海角大神

When it comes to evolution, bigger really is better

Stanford researchers have compiled data that seems to support Cope's rule 鈥� the notion that evolution 'favors' size.

We鈥檝e seen the future of marine life, and it鈥檚 big.

After conducting an extensive study on size, a Stanford research team found that, over time, marine animal lineages generally evolve to be larger.

The team, which included research scientists, undergraduates, and high school interns, amassed mountains of data under Stanford paleobiologist Jonathan Payne. The hulking dataset they compiled spans 542 million years, and includes five of the major phyla and over 17,000 genera 聽鈥� about 75 percent of all marine genera in the fossil record, and nearly 60 percent of all animal genera to have ever lived. They described Thursday in Science.

鈥淪ize evolution had never been studied at that taxonomic or temporal scale before,鈥� Dr. Payne says, 鈥渟o it was unknown whether size tended to increase or not across the entire marine fauna. Several previous researchers had speculated or assumed that the average size had increased, but no previous research group had the data to conduct a rigorous test of the claim.鈥�

Cope鈥檚 rule, named after 19th century American paleontologist Edward Drinker Cope, is the notion that animal lineages tend to increase in physical size over evolutionary time. Although Cope himself never suggested it, the theory bearing his name has been contested since its proposal. Payne says his research seems to support Cope鈥檚 rule.

鈥淭he average animal in the oceans today is 150 times larger in biovolume than the average animal in the oceans during the Cambrian, 540 million years ago,鈥� Payne says. 鈥淧rior to our study, it was unknown whether there had been size change and, if so, in what direction or by how much.鈥� 聽

鈥淲e [also] found that the size increase did not result from universal selection toward larger size,鈥� Payne adds. 鈥淩ather, the classes that were already larger early in the evolution of animal life have diversified differentially across evolutionary time. In other words, our data suggest that larger size favors diversification, rather than that larger sizes are favored in all populations.鈥�

While the overall increase in marine animal size is pretty much indisputable, some scientists argue that size is not a matter of 鈥渁ctive selection,鈥� but a result of random, non-selective mutations 鈥� an concept known as neutral drift. In other words, neutral drift could cause some lineages to grow in size, but only by chance 鈥� that doesn鈥檛 necessarily mean evolution 鈥渇avors鈥� size. The neutral drift argument is supported by evidence from bird and insect populations, who have not grown in size as Cope鈥檚 rule postulates.

鈥淚t is possible that Cope's Rule applies mainly to marine animals,鈥� Payne admits. 鈥淯nderstanding the underlying causes better will be critical to determining whether or not we should expect animals in other environments to exhibit the same patterns.鈥�

But Payne says that, at least in this case, neutral drift isn鈥檛 his culprit.

鈥淥ur data are not compatible with that scenario 鈥� they require an active evolutionary process that favored animals with larger sizes.鈥�

Payne says his dataset shows evolutionary growth so significant that it couldn鈥檛 have happened by chance. If the impressive sample size wasn鈥檛 enough, Payne鈥檚 findings found additional support from a surprising source 鈥� virtual reality evolution.

To strengthen their research further, Payne鈥檚 team used a model that could simulate millions of years of evolution in under 10 seconds.

鈥淥ur model of evolution was a fairly simple one,鈥� Payne says. 鈥淚t included genera that in any given time interval could survive into the next interval, survive and give rise to an offspring genus with a size slightly modified from the parent genus, or simply go extinct. We ran this evolutionary branching process for 540 million years about 10,000 separate times to see the range of results that one could expect assuming either neutral drift, a lower bound with neutral drift, or a process favoring size increase.鈥�

鈥淭his is a huge improvement over the early paleontological simulations run by David Raup in the 1970s,鈥� postdoctorate and co-author Noel Heim says. 鈥淭hey couldn't do nearly as many runs, and each run only had about 100 time steps. Computers have come a long way.鈥�

But if evolution follows certain rules, and can even be simulated, can it be predicted?

鈥淚f evolution does follow directional trends,鈥� Payne says, 鈥渢hen this implies some degree of predictability. However, it is also important to keep in mind that living animals in our dataset span 14 orders of magnitude in size. In other words, the largest animal is 100,000,000,000,000 times larger in volume than the smallest one, and they are both around today. So being able to predict the average size of animals in 100 million years does not mean that I can tell you the size of any particular species.鈥�

鈥淎lso, the trend that we see in our data stands out because we are able to observe it over so much time,鈥� Payne adds. 鈥淚t is not useful for predicting the sizes of species in 100, 1000, or even 1 million or 10 million years, because the changes over those timescales would be very small relative to the changes induced by other factors.鈥�

Size isn鈥檛 the only trait thought to evolve directionally. Payne hopes the data collected by his team can be used in future studies of evolution.

鈥淐omplexity is another trait that has been argued to increase over evolutionary time,鈥� Payne says, 鈥渨hether measured as the number of body parts, cell types, neural complexity, or via other metrics. However, there are no comprehensive data at the scale of our body size study that are currently available to address this hypothesis.鈥�

鈥淎nother geographic pattern in body size is a tendency for organisms to be larger near the poles than near the equator,鈥� Payne adds, 鈥減erhaps due to the colder temperatures. If one were to match our size data to information on the geographic distribution of genera, one could assess whether or not animals near the poles in the oceans tend to be larger, smaller, or the same size as animals near the equator. These types of geographic patterns are topics that we hope to address with our dataset in future studies.鈥�

And as we continue to learn the rules of evolution, we come closer to understanding our universe.

鈥淥ne question that has intrigued biologists ever since Darwin's time is whether evolution is largely random or whether it is predictable and directional,鈥� Payne says. 鈥淪tephen Jay Gould famously asked whether the world would look similar if we could rewind the tape of life to the Cambrian and then let it run again. He spend much of the second half of his career arguing that a lot of evolution is less directional than we we tend to believe 鈥� that people often see patterns where there are none.鈥�

鈥淚n this case, it appears that there really is a pattern. As we come to better understand the underlying causes, it will be easier for us to predict whether or not we should expect it in terrestrial systems as well 鈥� or even on other planets, if we were to find life on them.鈥�