Paige studied a population of eastern pipistrelles in central Illinois. In high summer, the bats often roost in trees. But in spring and autumn, the bats linger in caves, emerging in large numbers only on the nights when many insects are on the wing.

The bats’ roosting sites deep underground are remarkably stable environments – temperature, humidity and air currents stay virtually constant. But barometric pressure in the caves varies in line with the pressure outside, and Paige suspected that this was the bats’ hunting cue.

To test his theory Paige counted the number of bats emerging from a cave and compared it to the relative abundance of flying insects at different barometric pressures. As the pressure increased, the number of insects fell and fewer bats came out of the cave. Paige does not think the insects are responding directly to changes in air pressure but rather tend to fly more often in warmer weather, when the pressure tends to be lower. Exactly how bats could detect changes in air pressure is unknown, but one possibility is that they use a structure in the middle ear called the Vitali organ. This is found only in bats and birds – and in birds it is believed to measure air pressure.

Paige also captured some bats placed them in a chamber in which the barometric pressure could be varied. The bats roosted quite happily within these chambers, allowing Paige to measure how their uptake of oxygen varied with pressure. At higher barometric pressures, the roosting bats respired more rapidly (Functional Ecology, vol 9, p 462). This “metabolic tracking” of air pressure seems to take some effort, as out-of-condition animals were unable to do it.

Paige was surprised by his discovery. “It seems, at first, counterintuitive,” he says. Paige had suspected that the bats would respire more rapidly when the air pressure was low, as they prepared themselves for a night’s hunting. The fact that they do the opposite, Paige believes, may help them conserve energy.

Although the bats fly to their cave entrances on any night when the air pressure is low, they only venture outside when it is not raining, as few insects fly when it is wet. By reducing their metabolic rate immediately before their flight to the cave entrance, Paige suggests, the bats account for the possibility that this effort may be wasted. “It’s a bet-hedging strategy,” he argues. Why the bats do not simply keep their metabolic rates uniformly low all the time they are roosting, however, remains a mystery.

Before it can reproduce, a colony of bumblebees must switch from producing only sterile female workers to producing males and queens as well. Queens are produced when workers give certain larvae extra supplies of food during a critical stage in their development. Male bumblebees, meanwhile, are produced from unfertilised eggs laid by the queen bee. Zoologists have assumed that the timing of the change to produce males and queens was decided by the colony’s queen. Jacqueline Shykoff and Christine Müller of Basel University in Switzerland set out to test this idea.

Queens and males do not forage for food, so for a colony to reproduce successfully it must have a worker force large enough to collect enough food to support the queens and males. In principle, there are two main ways in which a bumblebee queen could assess whether her worker force is large enough for the colony to begin breeding: she could monitor the ratio of workers to larvae, and begin to lay unfertilised eggs only when this ratio passes a critical value; or else she could simply begin producing males after she has laid a set number of fertilised eggs, and gamble that sufficient workers have survived to raise her sons.

Shykoff and Müller captured queen bees as soon as they emerged from hibernation, and let them set up colonies inside transparent hives. These colonies either had to forage for their food in the field, or were kept in the lab and provided with food.

The researchers found that there was no set ratio of workers to larvae at which the colonies began to produce males and queens. They also found that workers from the colonies who had to forage in the open air were more likely to die than those kept under cosy laboratory conditions. These colonies reproduced later on average than those held inside. This indicated that bumblebee queens do not simply begin reproducing after they have laid a set number of eggs (Functional Ecology, vol 9, p 106).

Shykoff and Müller do not rule out the possibility that the queen decides to begin producing males on the basis of a more complicated combination of cues. But it is more likely that the workers are making the decision, they say. After all, the workers may be in a better position to judge whether the time is right for reproduction, given that they are the ones who must go out and forage for food.

If this is the case, bumblebee queens may produce both fertilised and unfertilised eggs all of the time. Worker bees have been shown to be able to distinguish between different kinds of eggs, and in most bumblebee colonies some eggs and larvae are not reared. So Shykoff and Müller suggest that the workers may simply neglect any unfertilised eggs laid by the queen until they decide that the time is right to begin producing males.

The spider, Parawixia bistriata, lives in the vegetation of savannas and usually spins a small, fine-mesh web every evening at sunset. These webs catch the small flies which form the spider’s staple diet. However Christina Sandoval, who is now at the University of California, Santa Barbara, noticed that during September the spider occasionally spins a much larger, wide-mesh web. Unlike the smaller webs, the wide-mesh webs are spun at various times during the day.

Sandoval found that these wide-mesh webs were produced only when termites, were swarming nearby (Functional Ecology, vol 8, p 701). The termites normally emerge from their nests in September to swarm and set up new nests. The larger webs were presumably more effective at snaring these larger insects.

Previously, researchers have assumed that the differences in web construction within spider species are due to environmental factors such as wind, rain or the amount of space available. Parawixia, however, seems to be a highly specialised predator which adjusts its web size and the time of spinning according to the availability of specific prey.

Carlos Roberto Fonseca studied the Amazonian rainforest tree Tachigali myrmecophila. Hollows in its leaves are occupied by a stinging ant, Pseudomyrmex concolor. Although the tree provides no food for the ants, the insects nevertheless mount 24-hour patrols attacking all creatures that disturb the tree, whether they are other insects or mammals. The tree has a very low growth rate and spends most of its life as a sapling in the rainforest understorey. It reproduces only once, just before it dies.

The ants do not eat the leaf-eating insects they capture but prey on a colony of coccids they keep inside their nest. These produce honeydew from the phloem of the tree, and this is the ants’ main source of energy. The tree feeds the ants only indirectly, and in the past some biologists have suggested that this particular ant-plant relationship is not mutually beneficial.

But by conducting an experiment in which some trees were kept ant-free for 18 months, Fonseca was able to show that the ants’ presence is indeed beneficial to the tree. He found that trees without ants suffered at least twice as much leaf loss as those with ant colonies. The overall effect of removing the ants was to shorten the life of a tree’s leaves by about 50 per cent and to slow their rate of growth, presumably by reducing the leaf area able to carry out photosynthesis.

Fonseca believes that the cost of ant protection to rainforest trees is probably greater than that to acacias because they have to support the coccids as well as the ants. However, in the poorly lit rainforest understorey, the loss of leaf area may be more serious than in more open habitats. So for Amazonian trees, recruiting ant protection seems to be a cost worth paying and is probably less than the cost of photosynthesis loss due to leaf-eating insects.

Leticia Aviles of Harvard University is studying the spiders, and believes
that they may belong to a new species. They are lynx spiders of the Oxyopidae
family and Tapinillus genus (Biological Journal of the Linnean Society,
vol 52, p 163).

The case for the spiders’ inclusion in the category of near-social creatures
rests in part on their communal life in the web. These nests – not the classic
flat webs but three-dimensional designs woven around the ends of tree branches
– accommodate groups that include adults of both sexes and juveniles of
different ages.

But in another respect, the Tapinillus species is unique among spiders.
Alloenzyme analysis shows that apart from a few males, all the individuals
in a nest are siblings, the offspring of a single pair. This suggests parallels
with the social insects.

With one female producing all the young, as with ants, the spider’s
behaviour may be closer to that of colonising social insects than to that
of any other known spider. However, Aviles cautions against jumping to such
a conclusion until further work has determined the roles of those males
and females in the nest which do not appear to reproduce.

The spider will capture and eat by itself insect prey that are smaller
than it is. Larger prey, however, are dealt with by a group of adults acting
together, who then feed on it communally.

Young spiders and the egg sacs are kept in the interior of the nest.
The juveniles are allowed to feed on prey only when the adults have finished.
Adults of both sexes cooperate to keep the nest repaired.

The orthodox view is that bats are a distinct mammalian order, the Chiroptera,
which can be divided into two suborders. The large fruit bats and flying
foxes are grouped in the Megachiroptera while the others, small insect-eating
bats and vampires, are placed in the Microchiroptera. Many zoologists believe
that this classification reflects the evolution of all bats from a common
ancestral form.

However, recent studies have suggested that the ‘megabats’ are in a
sense ‘flying primates’, sharing a number of features with humans and the
apes. Now Arnd Schreiber, Doris Erker and Klausdieter Bauer of the University
of Heidelberg have looked at the proteins in the blood serum of megabats
and primates and found enough in common to suggest a close taxonomic relationship
between the two groups (Biological Journal of the Linnaean Society, vol
51, p 359).

If the German biologists are correct, it means that flying mammals arose
twice during the course of evolution and that the similarities between
the two kinds of bats reflect adaptations to their way of life rather than
to a common ancestry. The German researchers are careful to point out that
their data do not prove this theory but are compatible with it.

An alternative explanation is that the microbats have evolved more rapidly
and this could be the reason why they now appear to differ considerably
from the megabats. The bat controversy seems set to continue for some time
before being resolved.

Now Malcolm Edmunds and Robert Dewhirst of the University of Central
Lanca-shire in Preston report experiments which appear to do just this (Biological
Journal of the Linnean Society, vol 51, p 447). They used four types of
artificial prey designed to resemble fat, green caterpillars such as those
of hawk moths, and exposed them to wild birds on several garden lawns. The
birds included house sparrows, starlings, blackbirds, dunnocks, and blue
and great tits. The artificial prey were made of pastry and coloured dark-green,
light-green, countershaded (dark-green above and light-green below), and
reverse-shaded (as countershaded but upside down).

The researchers found that the light-green prey were most often taken,
followed by the reverse-shaded prey. The dark-green prey were only about
half as popular as the light-green, while the countershaded prey were only
occasionally eaten.

Edmunds and Dewhirst believe that this is because the countershaded
prey are better camouflaged, so the birds cannot detect them as easily as
they can the other types. It seems that being two-toned is not enough, as
the reverse-shaded prey were taken almost as often as the light-green type;
it appears to be necessary to have the darker shade on top.

According to Tim Birkhead of the University of Sheffield and Anders
Muller of Uppsala University, the reasons are remarkably similar to those
that make sperm banks so useful to people who breed domestic animals. The
ability to store sperm gives females the chance to be fertilised by the
best male whenever he happens to be available, regardless of whether that
is also the best time for young to be brought into the world. Reptiles which
store sperm can reproduce even in years when they do not meet a male (Biological
Journal of the Linnean Society, vol 50, p 295).

The record at the moment is held by the Javan wart snake (Acrochordus
javanicus), which can store sperm for seven years. Several other snakes
and some turtles are known to store sperm for four or five years. Birds
can do so for weeks. The avian record is 117 days, for the domestic turkey
(Meleagris gallopavo). Human females store sperm for only five days, and
domestic sheep and pigs for a mere two.

The main effect of nature’s sperm banks, say Birkhead and Muller, is
that they allow the females to exercise a considerable degree of sexual
selection. Being able to collect sperm from favoured males when convenient
and store it for future use enables the females, over the generations, to
influence the evolution of the species.

Linda Whittingham and Raleigh Robertson of Queen’s University at Kingston,
in Ontario, studied red-winged blackbirds (Agelaius phoeniceus) nesting
on two marshes: one surrounded by woodland where food was abundant and the
other by agricultural land where food was scarce.

The researchers found that at the woodland marsh, few males fed their
young, whereas all the males at the agricultural marsh fed at least one
of their broods. Whittingham and Robertson concluded that the males help
the females only when the females cannot collect enough food on their own
(Journal of Animal Ecology, vol 63, p 139).

However, being relieved of the duty of feeding does not necessarily
make life easier for the males. Given the opportunity, male red-winged blackbirds
are polygynous – they mate with more than one female. In habitats in which
a female can raise a brood on her own, competition among males for mates
increases. In the marsh by the woodland, the biologists discovered that
the successful males had more than twice as many mates as those in the agricultural
marsh.

So although female red-winged blackbirds that nest in a good habitat
may well have to raise their chicks without help, they are more likely to
have mated with fitter males. The researchers found that more chicks were
fledged in the woodland marsh, where they would have fitter male parents
and ample food supplies.

Simon Landhausser and Ross Wein of the University of Alberta studied the recolonisation of burnt areas of forest in the tundra near Inuvik in Canada’s Northwest Territories. They compared the trees recolonising sites that had been severely burnt several years earlier with similar unburnt sites (Journal of Ecology, vol 81, p 665). They found that in the recolonised areas two coniferous species, the spruces Picea mariana and Picea glauca, were less common than in unburnt sites, while two deciduous trees – a birch, Betula papyrifera, and a poplar, Populus balsamifera – were more common. These trees had also extended their range into burnt, previously treeless, tundra areas which the spruces had failed to colonise.

Landhausser and Wein believe that the deciduous trees have extended their range as the growing season has become slightly warmer and drier over recent years. They predict that global warming will allow deciduous trees, especially those with effective ways of dispersing seeds over long distances, to become more abundant and to invade the tundra a little further after each fire. Warmer and drier conditions could make wildfires more common, so accelerating the trend.