Sunday, 23 January 2011

Have Children and Die

For many invertebrates, reproduction is an "all or nothing" affair. They mate once, produce a huge burst of offspring, and then die. Mayflies are a particularly extreme example, but there are a great many others. Vertebrates are almost always different; assuming they don't die for some other reason, most females will give birth more than once during their lifetime. Perhaps the best known exception is the salmon, which struggles its way upriver to its spawning grounds just once before it dies.

The almost complete absence of an "all or nothing" approach to reproduction among mammals means that, for them, raising a litter is always going to be something of a trade-off. The mother has to invest a lot of effort on her young; first she has to nourish them in her womb, and then provide them with milk, before we even consider teaching them to hunt, or whatever else might be needed before they can travel off on their own. But, on the other hand, she can't invest too much energy in them, because then she herself will suffer. A female mammal has to ensure not only that her children reach the point when they can survive on their own, but also that she survives to raise further litters. There's inevitably a balance to be reached between those demands.

But... well, what when there isn't? Animals don't live forever, and the more litters you have, the less likely that you're going to survive to have another one anyway. Something called "terminal selection theory" therefore suggests that mothers should invest more energy in looking after later litters than earlier ones. From an evolutionary perspective, once you've had all the litters you're going to, there isn't much to lose by investing everything you can in the last one.

There doesn't really seem to be a great amount of evidence that mammals actually do this, though. That's partly because it's a fairly complex process, with a lot of confounding factors. It's true that the first litter of many mammal species tends not to do so well as the later ones, but that can often be put down to parental inexperience, or that younger mothers may be smaller and not so fit as they become in later years. Long-lived species, which tends to mean the larger ones, may have several litters throughout their life, making it unclear in advance which one is likely to be the last one. It may be difficult for a mother to control the amount of energy she expends on her developing offspring while she's still pregnant, in contrast to animals that look after eggs. In some social animals, there may even be advantages to surviving into post-reproductive life, passing on your knowledge of the environment to your grandchildren (humans are the most obvious example here, but it also seems to be true, for instance, of elephants).

With all of these reasons to muddy the waters, its not surprising that terminal selection either doesn't happen in mammals, or at least, is very difficult to spot. The best place to find it would be in some animal that doesn't live very long, doesn't have to do anything much once its finished weaning its offspring, and, ideally, has more control than usual over how much energy it expends on its young. Enter the brown antechinus (Antechinus stuartii).

You might think, from the picture, that this is some sort of mouse, but it's not. Instead, its one of several species of small marsupial that happen to look remarkably mouse-like. It feeds on insects and other small invertebrates, and belongs to the same family as most other marsupial carnivores.

The fact that antechinuses are marsupials is important, because marsupials spend very little time actually pregnant. Instead, most of the development of the young takes place in the pouch, with the mother supplying them with milk, something that may well be easier for her to control than nutrition in the womb. The brown antechinus (there are also nine other species) is unusual among mammals in that it doesn't give birth to many litters during its life. Indeed, most individuals really do breed only once. Unlike salmon, however, it is at least possible for them to survive to breed again, and a significant minority of females do manage a second litter. The males, incidentally, seem to have no such luck; they get to enjoy a single three week mating season, and then simply die off, their life's ambition completed.

Even in placental animals, the maximum size of a litter is usually dictated by the number of available teats, and this is even more so in marsupials, where the young are physically attached to the teat for much of their development. In the case of the brown antechinus, that maximum is eight (or occasionally nine - mammals of a given species don't always have the exact same number of teats). Unsurprisingly, in an animal that only breeds once or twice, the litters are usually pretty close to that maximum size, and that's a significant cost for the mother. By the time the young are weaned, they're already over half the weight of the mother - which means, when you consider how many of them there are, that, for at least some time, the mother is providing milk for offspring whose combined weight may be four or five times her own. A fairly considerable commitment by anyone's standards, so its perhaps unsurprising that most mothers die (presumably from exhaustion) as soon as the young start fending for themselves.

A study published last week in PLoS One compared the litters of brown antechinuses that bred only once with the first and second litters of those that bred twice. One of the first things that was apparent was that the young of mothers that bred only once grew significantly faster than those in the first litter of mothers who later went on to breed again. That could be due to the second group of mothers being less fit, and not able to provide as much milk for their offspring, but that seems unlikely, since it wasn't possible to predict in advance, just from the mother's weight which group she would fall into. Instead, it seems that those mothers who went on to have two litters had exhausted themselves less with the first one, increasing their own chances of survival.

As one might expect, all the mothers put on weight as they became pregnant, and generally kept this weight up as long as the young were in the pouch. Those mothers having a second litter put on less weight, but only because, being older, they started out rather heavier. Once the young really began to grow, however, but before they were fully weaned, the mothers all dramatically lost weight, evidently diverting all of their resources to milk production. The "terminal selection" theory would predict that mothers having their second litter would expend far more energy the second time around, there being no chance for a third anyway, and this was exactly what happened. These mothers lost something like a third of their body weight in a little over a month, and then promptly died once their young were weaned.

That that sudden death wasn't due to old age was demonstrated by one individual that failed to breed in her second year. Reaching the grand old age of three, she did eventually manage to breed the following year, and then demonstrated the same pattern of putting dramatic effort into raising her second litter, and dying once they were safe.

Perhaps surprisingly, the mothers that had two litters had no more children reach adulthood than the great majority, who had only one. The relative lack of investment in their first litter meant that those offspring grew more slowly, and were more likely to die, while the second litters tended to be slightly smaller. All in all, it doesn't seem that, in this species, there is much advantage in having a second litter, which is probably why around 80% of mothers don't bother.

Its worth noting that one of the few other studies to show this effect in a mammal was also on a marsupial - in that case, a brushtail possum. Marsupials, it seems, find it easier to control the amount of milk they give to their offspring than placental mammals. Additionally, young placental mammals that are still in the womb have a surprising amount of control over their mother's metabolism, demanding nutrients for themselves without her being able to do much about it. That's a problem marsupials don't have to face, and may explain why the very few mammals that do have an "all or nothing" approach to raising young are all short-lived marsupials.

[Picture from Wikimedia Commons]

Sunday, 16 January 2011

The Origin of the Cloven Hoof

Foot of a giraffe
The shape of the feet in mammals falls into one of three categories (ignoring those that don't really have feet, such as dolphins). Plantigrade animals, such as humans, walk on the soles of the feet, with both the heel and the toes touching the ground. This might seem the obvious way of doing things, but its actually fairly unusual among large mammals. More common is the digitigrade stance, which we can see in, for example, cats and dogs. Here, the animal effectively walks on tip-toe, with the ankle held well clear of the ground. The part of the foot in between the ankle and the balls of the feet is generally elongated and narrow, so that the ankle looks rather like a backward-pointing knee.

This pattern is thought to allow the animals to move more quickly, making the limbs longer and more flexible, with an extra joint that can bend in the direction of motion. This is usually more important in the hindlimbs than the forelimbs, because they are the ones that push against the ground to propel the animal forwards, while the forelimbs are more important for braking. As a result, there are a few animals, such as raccoons, that have evolved digitigrade hindlimbs, but never got around to doing the same with the forelimbs.

The third possible stance is called unguligrade. These animals go even further than the digitigrade ones, standing only on the very tips of their toes, like a balancing ballerina. Such animals are usually hoofed, and an obvious example is the horse. Yet, while the evolution of horses and their single hooves has been described in many places, the origin of the cloven hoof is perhaps less well known. This, despite the fact that are far more animals with cloven hooves than there are species of modern horse. But then, for the same reason, the story is also rather more compliacted.

The group that these animals belong to (the "even-toed ungulates") includes ten families, although, admittedly, not all of them actually have hooves. These families can be placed into five broader groups, as follows:

  Other cloven    Chevrotains     Hippos
 hoofed animals                   etc.
      ^               ^            ^         Pigs
      |               |            |         etc.
      2               |            |          ^       Camels
      -----------------            |          |        etc.
              |                    |          |         ^
              4                    |          |         |
              ----------------------          |         |
                         |                    4         2
                         |                    |         |
                         ----------------------         |
                                    |                   4
                                    |                   |

When it comes to fossil animals, it's often difficult to determine exactly how they walked, since skeletons are rarely found in a perfectly articulated form. When they are mounted for museums, a fair degree of guesswork has to go into how the bones all fit together, and sometimes those guesses are going to be wrong. Fortunately, with mammals, unlike older and arguably stranger, fossils, like those of dinosaurs, its possible to compare the animals with living forms that look fairly similar. Comparing such things as the proportions and shapes of the limbs can reveal a lot of information about how mammals long vanished from the world moved and lived when they were still alive. An article recently published in the Journal of Vertebrate Palaeontology used this approach to examine some of the details of how the cloven hoof evolved.

Like the ancestors of horses, the ancestors of cloven-hoofed animals originally had five toes. Among their living relatives today, this is only true of the two species of hippopotamus, an animal whose large size makes it very different from its distant ancestors. Indeed, the living animals whose feet most resemble those of these early creatures appear to be the dogs.

For the most part, their feet were fairly typical of other early mammals, although the thumb/big toe was already quite small, and some of the bones do have a similar shape to those found in dogs (and, to a lesser extent, cats). It seems likely that, just as dogs are fast running animals because they want to catch their dinner, these early ancestors of the cloven hoofed animals would have been fast running because they wanted to avoid becoming dinner.

If this is right, then the first animals of this group would have lacked many of the distinctive features that the group has today. Their toes probably weren't strong enough to support their body weight without some assistance from the balls of the feet; in other words, they would have been digitigrade rather than resembling modern hoofed animals. They may even have had pads on the soles of their feet, as dogs do, although that's the sort of thing that can be difficult to know for certain.

These animals, then, would already have been quite good at running. The next step was the loss of the thumb/big toe. This was clearly a useful adaptation, because it appears to have happened more than once. We can tell this, because hippos, which still have all five toes, are not the most primitive members of the group. This means that the evolution of a four-toed pattern must have happened at all of the points marked '4' on the diagram above - and this ignores some extinct groups that don't appear to be the direct ancestors of anything around today.

In addition to the disappearance of the thumb, the index finger/2nd toe and little finger/5th toe also become much shorter. At the same time, the other two toes become stronger, able to take more of the weight of the animal, and the muscles that, in most mammals, move the toes apart are replaced by a tough ligament. There are also changes in the structure of the wrist or ankle to accommodate the changing arrangement of muscles.

The most obvious modern example here is the pig, which has a true cloven hoof, but also has two additional toes on each foot that do not reach the ground. So it should be at this stage that fossil mammals evolved from a more typical digitigrade stance to something at least much closer to the unguligrade one. At least one group of extinct mammals, one thought to be related to the ancestors of camels, does quite closely fit this description.

Some other extinct four-toed animals in the group, however, are a little different. Their feet look more like those of their primitive ancestors, with only some of the changes found in their living four-toed counterparts. Their joints are not quite the hinge-like shape found in pigs, although they are certainly more so than in dogs, and some of the other changes also seem less extreme. It would seem that this represents the actual change from digitigrade to an unguligrade stance, a stage reached with the three-toed Mesohippus in the evolution of horses. It seems likely - although far from certain - that these animals still had a foot pad, and not hooves, and that (unlike pigs) they used all four toes to walk.

Its obviously tempting to assume that these animals were the transition between the 5-toed and "more evolved" animals, such as pigs. There's probably some truth in that, but it can't be the full story, because at least some of these groups never evolved into anything else. Perhaps the best known such group, the Protoceratids, survived with more or less this arrangement for around 40 million years - about two thirds of the total timespan since the extinction of the dinosaurs. So its obviously a pretty effective way of moving about, something that works well in its own right, without being just a necessary transition to anything 'better'.

Nonetheless, in three groups, the outer two toes became more radically reduced, leaving only two true toes - the middle finger/3rd toe and the ring finger/4th toe - on each foot. Two of these groups are shown in the tree above, marked with a '2', but the other is extinct, and exactly where they belong really isn't clear.

Quite what happened to the other two toes varies quite a bit between different groups, but in all cases at least some of the bones disappear altogether. For example, in goats, only the very last bone in each toe survives, as a sort of small dewclaw at the back of the foot, while in camels, the toes vanish entirely. At the same time, the limbs became longer, especially between the ankle ("hock") and the foot, allowing a further increase in running speed. The shape of the joints also changed, to allow for the newly vertical posture of the foot, and to create hinge joints more suitable for fast running.

Again, the pattern isn't entirely neat. While there is a lot of similarity between the way that the cloven hoof formed in the three groups, there are also some differences, reflecting the fact that this happened more than once, instead of being a simple linear progression. In the case of camels, for example, the animals lost the unguligrade stance of their ancestors, and went back to being digitigrade, with a foot pad instead of a hoof. Flat splayed feet are, after all, more use in a sandy desert than sharp hooves.

[Picture from Wikimedia Commons]

Sunday, 9 January 2011

Bat Swarms of Colorado

When we think of bats, the environment that first springs to mind is that of a cave. While there are a number of bats that roost in trees during the day, a great many are, indeed, cave-dwellers. Because caves aren't all that common in most places, bat colonies within them can be huge, which creates something of a problem when they all try to fly out in the evening. But, if you watched a bat cave throughout a year, in most cases, the activity level would change radically as the seasons progressed.

This is largely because most bats eat insects. In temperate climates, insects aren't around much during the winter; in many cases they've died off and left their eggs to hatch the following spring when the weather has improved. Any animal that relies heavily on eating insects to survive has to deal with this scarcity somehow. Insect eating birds, such as swallows and thrushes, cope with the problem by migrating in the winter, heading to warmer climes where insects are still common. There are some bats that do exactly this - for example, the hoary bat migrates annually between Canada and the southern US - but more commonly, they deal with the absence of food by hibernating.