Climate, Cloud Forests, and Cotton Rats

The Cricetidae is the single largest family of mammals in terms of the number of species, at least according to the current count from the American Society of Mammalogists. While the name translates as "hamster family", the great majority of species are not themselves hamsters. In fact, there are five subfamilies of cricetid: the hamsters, the voles, and no fewer than three with members that basically look like mice or rats... even though the true mice and rats belong to the second-largest family, the Muridae.

The largest of these subfamilies is the Sigmodontinae, consisting of mouse and rat-like animals primarily native to South America, although one species lives as far north as Virginia, and several others reach Arizona and New Mexico. The group is named for the S-shaped pattern on the molar teeth and was originally coined as the genus name for the cotton rats in 1825.

There are over four hundred known species of sigmodontine, with more being discovered all the time, and it's reasonable to ask why that number should be so high. Partly, it's down to a successful body plan, the same thing that helped the true mice achieve their similar dominance. The sigmodontines, however, also had an advantage that their northern counterparts, the neotomines (deer mice and packrats), did not: South America.

For millions of years, South America was an island continent, cut off from the outside world much as Australia is today. Few animals could reach it from elsewhere. It did have rodents, in the form of the guinea pigs and their relatives, but what it didn't have was anything much like mice. That changed as North America got closer.

The first sigmodontines may have crossed into South America four million years ago, when the Panamanian land bridge formed. It's also possible that it happened a little earlier, since Central America was a chain of islands for a while, potentially close enough to allow small land animals to be swept from one to the other in tropical storms. But, either way, when they got to the new continent, they found it ripe for the taking.

They rapidly diversified, moving into habitats where there was nothing else like them to act as competition. Across South America, there are jungles, pampas, mountains, swamps, deserts, and grassy plains. All of these could provide new lands for the incoming rodents, separating them off from their neighbours by geography or by specialisation to a particular environment. The number of species rocketed from what must have been a small initial sample to the great range we have today. Sigmodontines live just about anywhere, the result of a great - and rapid - success story.

The very rapidity of this diversification creates difficulties in determining the details of how it happened. With so many species and larger groups being roughly the same age, it's difficult to determine what order things happened in, and therefore how the details of geography and environment shaped their evolution. Over the last decade or so, there have been at least fourteen different patterns proposed, based on genetic studies, analysis of body form, and so on. Only one of these seems to be supported by a more recent study, and even that was only able to look at 11% of the species and around 40% of the generic diversity.

That's actually not too bad, considering the huge number we're talking about. But scientific techniques advance, and new species have their genetics sampled. It's not just biological science, either; advances in geological science can improve our picture of how things like the climate changed over millions of years, allowing us to match evolutionary events with the ecological changes that drove them. All of which results in the latest analysis, and the largest so far, including 38% of the known species and 84% of the genus-level diversity. It won't be the last word, but it's a move in the right direction.

When we do these sorts of studies, we analyse the genetic differences between different species, calibrating them with fossils of known age to determine not just in what order, but when, splits between species or larger subgroups occurred. By checking to see where the species live, we can also get some idea of where their common ancestor is likely to have lived, adding this information to the picture.

One of the things we want to know is when and where the very first split within the group occurred, effectively allowing us to pinpoint its origin. In this case, the study showed that the sigmodontines originated 10.5 million years ago, in the Late Miocene. Moreover, it seems have happened in South America. This is significant because it had often been thought that the group originated in the north before the crossing, partly due to the oldest known sigmodontine fossils coming from Mexico, and partly because the few North American species, such as the hispid cotton rat, belonging to the oldest branch in the family.

The reasoning had been, therefore, that these North American species were descendants of the sigmodontines that had never left their home. But the new study, including more cotton rat species from the south, shows that this is unlikely, and that the northern species descend from some group that crossed back again much later. That Mexican fossil, for example, dates to about five million years after the study suggests the group first appeared, so says little about its ultimate origin.

This, of course, favours the "island-hopping" theory about the arrival of the group in South America, rather than them having crossed over by land alone. However, it does leave a long gap between their putative arrival and the beginning of the rapid diversification, which took off in earnest, according to this new study, 6.6 million years ago, at least four million years after the apparent colonisation of the continent.

Why the wait? Well, 6.6. million years ago is in the middle of a cooling period as the Miocene came to a close and lines up very closely with the collapse of the Pebas Mega-Wetland. This was a vast series of lakes, or possibly just one single body of water, that existed for much of the Miocene east of Ecuador and northern Peru, connected by a long channel through present-day Colombia to the Caribbean. This effectively turned the northern part of the Andes into a peninsula, separated from the Amazon by a great bay, or at least an area of swamps and wetland. 

So, it could be that the early sigmodontines were trapped on this peninsula, and only expanded beyond when the climate became drier and changes in geography helped drain the wetland. When that happened, it seems that the sigmodotines emerged into the rest of the continent in a great burst, heading into areas that changing weather patterns were making wetter and more forested.

This fits with the time that cloud forests, long present in the north of the continent, and in Central America, began to reach further south as the central Andes increased in height. Other groups of organisms also underwent expansions at the same time, including hummingbirds and wild gentian flowers. Unlike these, and unlike their closest relatives, the Central American tylomyines, the sigmodontines took advantage of the expansion of the cloud forests to reach beyond them, colonising new lands in the south.

Even so, the study suggests that, at this time, the rodents lived in comparatively warm, damp forests, even if they had moved down slope into the Amazon and the woodlands of what are now northern Argentina and Uruguay. Beginning in the mid-Pliocene, around 4.7 million years ago, that began to change as individual lineages began to undergo their own smaller bursts of diversification, one at a time, expanding into less forested habitats. A subgroup called the leaf-eared mice were among the first, entering the semi-arid parts of the central southern continent in the Gran Chaco region. This continued until about 3.1 million years ago, by which time the whole of the continent had been colonised, as far as Tierra del Fuego, where the olive grass mouse is the most southerly species of wild mammal in the world.

Again, this was a time of shifting climates, in this case, towards drier weather. Many other mammal species died out at this time, as grasslands took over from forests. This it seems, gave the sigmodontines another chance to expand, if more slowly and sporadically than before, partly due to being forced to adapt to changing conditions, and to potential competitors dying out and ceding their territory when they failed to do the same.

The sigmodontines, it seems, were in the right place at the right time.

[Photo by Yamil Hussein, from Wikimedia Commons]