Whispering Bats on the Hunt for Nectar

Southern long-nosed bats (L. curasoae)

There are likely three things that most of us think of when we think of bats: they fly, they're nocturnal, and they can echolocate using inbuilt sonar. In fact, only the first of these is universally true of bats; there are (and never have been, so far as we can tell) any flightless bats in the way that there are flightless birds. The other two statements, while broadly correct, are not absolutely true of all living bat species.

Indeed, the bats are the second largest order of mammals, in terms of number of species, after the rodents. This, as might be expected, leads to a remarkable variety in their habits and ecology, even if much of it is not immediately obvious to the layman. There are, under the most common present scheme, no less than nineteen different families of bat, and while many other mammal families have names that can easily be rendered in regular English ("cats", "deer", "marmosets", and so on), those of bats are have to rely on more obscure terms ("funnel-eared bats", "sheath-tailed bats", etc.)

Among this great, if often unrecognised, variety of bats, the family that is usually said to have the greatest internal variation between its species is the Phyllostomidae, or leaf-nosed bats. These bats are found only in the Americas, typically south of the US, and perhaps the most obvious indication of their variety is in their differing diets.

As is true of bats in general, most leaf-nosed bats eat insects, which they catch with the aid of the echolocation. Most, but not all. The second largest category are those leaf-nosed bats that eat fruit. These are nothing like the giant fruit bats (or "flying foxes") of the Old World, but are actually fairly typically sized bats, eating things like figs or Cecropia fruit. And then there are others, some of them predatory on animals the size of small birds, some feeding on leaves, and yet others including the famous vampire bats.

And one subfamily within the group, the Glossophaginae, consists of species that feed primarily on nectar. These bats have long, thin, tongues that they can extrude into flowers to get at the nectar. Their tongues are further adapted with ridges and tiny bumps that help the bats lap up and swallow their liquid food. They include, among others, the brilliantly named tailed tailless bat (Anoura caudifer)... which, ironically, isn't even the only tailless bat to have a tail. But I digress.

What's not entirely clear about these bats is how, exactly, they find flowers with nectar. There are clearly a number of options, and there's some support for all of them - they probably use a mixture of techniques, after all. Perhaps the most immediate thought, at least to we humans, is that the bats simply look for the flowers. True, they are nocturnal, but, despite their misplaced reputation for blindness, bats do have fairly good night vision, and certainly enough to actually see the flowers are, at least during moonlit nights. We do know of insect-eating bats that, for example, use vision to help find moths, so this is quite reasonable, even if it's more likely an aid than their primary means of locating flowers.

A second option is scent. The flowers that the bats feed on rely on them for their pollination, just as other flowers rely on bees or hummingbirds. This means that it is in the interest of flowers to attract bats to feed on them, and it has long been obvious that, whereas flowers pollinated by birds or bees tend to be brightly coloured, those pollinated by bats tend to be white (making them more visible at night) and strongly scented. One study on Pallas' long-tongued bat (Glossophaga soricina), for example, showed that they are attracted to compounds such as dimethyl sulphide given off by their preferred flowers. So scent undeniably plays a role, although one might expect that it's of most use for finding the bushes that the flowers are on, rather than narrowing down the source of the odour to a particular source.

Which, given that these are bats, leaves echolocation. The problem here is that the echolocation calls of nectar-feeding bats don't seem too different from those of bats that eat insects or other types of food. Although, perhaps they don't need to be. With very few exceptions, all leaf-nosed bats, regardless of diet, seem to have a fairly similar call structure. Specifically, their calls are very short, with a clear pitch and multiple harmonics, and they tend to be unusually quiet, compared with those of other bats. The latter, which has given rise to the nickname of "whispering bats", is likely an adaptation to using calls in fairly cluttered environments. If the call was too loud, then it would bounce off far too many other things in the general vicinity, producing lots of confusing echoes. This feature would, for example, be useful for bats that have to find flowers (or, for that matter, fruit) amidst a bush, or lots of other closely packed vegetation.

All that remains then, is for them to have some means of distinguishing the flowers from the background, so as to be able to home in on exactly what they're after. Here too, the flower can help, and it has been shown that the sort of flowers that rely on bats for pollination tend to have a particular shape that is good for reflecting sound, and, moreover, doing so in particular directions that allow the bat to sense which way the flower is facing, and distinguish it from the background.

 Putting this together, and adding the fact that bats are, indeed, attracted to flowers of the right shape, the general pattern is clear. Even on a dark, moonless night, nectar feeding bats can detect flowers from a distance by their scent, and move in closer to use their echolocation for more precise manoeuvring. This is, unsurprisingly, the same technique as that used by Neotropical fruit bats, and the fact that these particular bats are after flowers, rather than fruit, seems to make little difference.

A new study, however, adds new information to this broader view, by examining exactly how bats approach flowers using their echolocation, and how well they can adjust their behaviour to suit new circumstances.

The study used lesser long-nosed bats (Leptonycteris yerbabuenae), a species native to Mexico, Guatemala, and Honduras, and which also migrates to Arizona and southern California during the southern months. Unlike many other nectar-feeding bats, they inhabit semi-arid scrub and grassland, where many of the flowers they feed on are on cactuses, rather than in the dense understory of forests. This, presumably, makes echolocation of the flowers somewhat easier than it might be for other species.

The bats were placed in large, darkened, flight cages, along with flowers of cardón, or elephant cactus, a species known to be pollinated by bats, and acoustically visible plastic hemispheres that could be filled up with sugar water. The intention of the latter was to see how well the bats could adapt to this unusual source of nectar... but, as it turned out, they just weren't interested. They'd fly up close, decide it wasn't a real flower, and then head away again, thus never realising that there was anything tasty inside.

So they aren't very flexible in their behaviour (or can't be fooled, depending on how you chose to look at it), but what about the real flowers?

These, they loved. They'd fly about in a search pattern, get close enough to decide that, yes, these were flowers, then fly up to them from underneath, aiming their heads (in the dark, remember) precisely towards the position of the flower's opening and then dart in to lap up the nectar, hovering like a hummingbird as they did so. Throughout this time, they were clearly echolocating. To begin with, as they searched for food, their calls were single and widely spaced, as we would expect while they were navigating their way about. Once they had identified the flowers, they switched to series of brief calls, getting shorter and closer together as they homed in on their target, before finishing with a long buzzing series of very rapid calls immediately before they placed their head in the flower.

This implies that, on the approach run, they were using highly directional sonar, the pulses coming closer together to avoid confusing their echoes as they got near to the target. They could also evidently figure out exactly how the flower was positioned on the cactus branch, fine tuning their flight behaviour accordingly, even though their general approach pattern - always coming in from underneath - never varied. The rapid buzz right at the end probably gives them the exact orientation of the nectar-bearing tube of petals, and it's interesting that this hasn't been observed in nectar feeding bats before, although it is known from some of their insect-eating relatives. It may be that the shape of the flowers, and the possibility of them being surrounded by sharp spines, encouraged them to do this; these were wild-caught bats, and would already have been familiar with flowers of this type.

It's hard to know for sure how much scent, or even vision, played a role in their behaviour, since the cages were not completely pitch-dark. But it's evident that echolocation was important, and used in quite a subtle and sophisticated way. Bio-sonar is useful for detecting insects in flight, but it is also useful for finding food even on targets that don't move.

[Photo from the US Fish and Wildlife Service, in the public domain.]