Home - Back to bats - Northern bats use vision
Talk on vision in microbats: 13th Int Bat Research Conference, Poland 2004
Johan Eklöf
(no slides)
Visual prey detection in vespertilionids
Abstract
Behavioural tests on optomotor responses establish visual acuity thresholds
in five species of vespertilionid bats. Three species of Myotis, which are aerial-hawking
bats, responded only to a stripe pattern equivalent to 5 degrees of arc. Myotis
daubentonii, responded to 2.5-degree stripes, northern bats (Eptesicus nilssonii)
showed reactions to 1 degree and the gleaning brown long-eared bats (Plecotus
auritus) responded down to 0.5 degrees of arc.
We also investigated the ability of brown long-eared bats and northern bats
to make use of visual cues when searching for food. The long-eared bats were
tested using petri dishes containing mealworms that were subjected to different
levels of illumination. We presented four individuals with different sensory
cues: visual cues, sonar cues and a combination of these. The bats preferred
situations where both sonar cues and visual cues were available, however, visual
information was more important than the sonar cues.
The northern bats were studied in a field situation when searching for ghost
swifts (Hepialus humuli) among clutter. We presented bats with white and dark
moths mounted on top of steel wires and found a significantly higher attack
frequency on white ones. This suggests use of vision. When we reduced the size
of the spread moths by cutting the wings, the preference for white individuals
disappeared at 4 cm, indicating that this is the bats’ visual acuity threshold.
We hypothesize that northern bats, at least in the initial search phase, use
visual cues as a complement to sonar to detect stationary ghost swifts.
Talk
Good morning
My talk today is about something that is not very well studied in microchiropteran
bats – vision.
Many people think it is a very strange idea indeed to study vision in bats because
they are supposed to be blind and at least in English, we have the expression:
as blind as a bat.
However, as you and I know, bats are of course not entirely blind, after all
they do have eyes and some of them relatively big eyes as well.
And I will hopefully give you an idea of what bats can see…
As the title of my talk implies, I will focus on vespertilionid bats and if and how they use vision in prey detection…but I will also give you a very very short review of other studies on vision…
Microchiropteran bats would not be what they are if it wasn’t for echolocation, and indeed, it is the bats most important sense seen from an evolutionary point of view.
But still there are some disadvantages with using sonar.
It is loud…and can alert prey
It doesn’t reach very far…
…and can be difficult - if not impossible - to use in a cluttered environment.
Therefore, bats do use other senses,
like sonic hearing as we have heard in previous talks,
smell of course…
…but what about vision?
One of the most recent and perhaps the most remarkable and exciting finding
on bat vision is that bats can see ultra violet light. York Winter and his colleagues
found that nectar feeding bats can detect UV and thereby use vision to find
certain flowers. As many flowers and also insects reflect ultra violet light,
this has the potential to be a very import finding, ecologically and evolutionary
and I hope to see more added to that story in the future
Some of the more important of the earlier studies on vision in microbats are
made by Suthers and Wallis.
For example, by examining eye physiology and characteristics of the lenses
in the several different bats, it was concluded that bats do see relatively
well at long distances, beyond the reach of echolocation, and therefore, ...which
make perfect sense..use vision when commuting between feeding sites and when
migrating,
But for detecting stuff at close range, echolocation is still a better tool.
But also at close ranges, bats may use vision. If the light levels allow it,
bats are able to avoid larger obstacles using vision alone.
And in several studies on escape behaviour it was found that vision is important
when bats escape from roosts, as they seem to be guided by light from outside.
However, the time of the day and amount of light seem to decide which of the
two senses vision or echolocation the bats defer to.
The only study so far that shows that microbats can hunt by using vision alone is made by Bell in 1985, when he found that Macrotus californicus can glean prey from the ground using nothing but its eyes to detect the prey
It has also been discussed whether other bats, such as megadermatids and emballonurids
also can use vision to find prey, either by using a bright sky to backlight
prey so to speak...or when gleaning prey from surfaces. After all, these bats
do have relatively large eyes and also resolving power good enough to see prey
sized objects, but if they use it or not has not yet been shown.
So, how well do bats see?
This slide is supposed to show you the visual resolving power in different
groups of bats. Many bats have a visual acuity around and a little less than
1 degree of arc, meaning that two objects need to be separated by 1 degree to
be seen as two objects and not 1.
Worth noticing is that among the gleaning phyllostomids at the top of the table
we find Macrotus californicus, which has the best visual acuity of all bats
...at least studied so far, and is, as I said, the only bat found to be able
to use vision only when foraging.
I will hereafter focus on vespertilionid bats, which is a large group of bats
and they show a great variation in not only ecology, life style and ways of
foraging, but also in visual resolving power.
From around 0.25 degrees in pallid bats, to over 4 degrees in some myotis species.
And just as a comparison, I can show you acuities of some other mammals.
Humans in normal lighting has an acuity of 1 sixtieth of a degree, and does
not fit into this scale.
So, can vespertilionids use vision when searching for food items?
The first of my own studies I will present to you is made entirely in the field
together with Jens Rydell, and some of you may have seen some of these slides
before, but I think they are worth showing again.
In late June in Sweden, ghost swifts – which are silvery white, large
moths, perform lekking display behaviour over hay fields. These moths are entirely
deaf but hide from bats by hovering close to the grass,
thus trying to be protected in the clutter.
But still it doesn’t take very long after the first moth have appeared
till a bunch of northern bats show up at the field site.
Northern bats are among the most common in Sweden and are normally aerial hawkers,
but every year, at least some of them take advantage of this smorgasbord of
moths and catch ghost swifts among the grass panicles.
And as the moths are very conspicuous, and shining white, we wanted to test
if the bats can use vision to detect them, as for example seagulls do, whish
also prey on these moths.
So we caught a bunch of ghost swifts and mounted them on top of sticks, and
then put them in the field among the other flying moths,
and just waited for the bats to attack them…
And they did, and this is very spectacular indeed, as the nights are very bright
this time of the year in Scandinavia, you can actually see the bats catching
their prey just a meter in front of you.
We always put out two moths next to each other, one mounted the right way,
dorsal side up, but the other one we turned around, because underneath they
are dark brown, almost black. We wanted to know if the bats attacked the white
one or the dark one.
The moths have a wingspan of around 6 cm, but by cutting the wings we presented
the bats with different sizes of moths. From 6, 5, 4 and down to 3 cm
And every night we counted the number of attacks on each colour and size, and
we shifted the positions after every attack.
And theses are our results.
On normal sized moths there was a clear significant difference between the attacks,
and also on the 5 cm moths.
But when the moths were as small as 4 cm we no longer found a significant difference
and at three cm there was no difference at all.
It seems as northern bats can detect the moths using vision as they prefer the
white ones, although their visual acuity is not good enough to detect the ones
that are too small.
And throughout the study, the bats were constantly echolocating. Perhaps vision eases detection from a distance, but echolocation is needed when zooming in on the prey. – and this of course, is just a hypotheses –
But the interesting now is that we know how large a prey the bats can detect.
That is 4 - 5 cm. And we also know that the bats always fly at 3 to 4 meters
above the field before diving almost vertically towards the moths.
Then it is quite easy to calculate that the visual acuity must be around 0.8
degrees.
Now of course, I wanted to know if this number was a good measurement of the
visual acuity.
And to be able to compare the result to previous studies I used an old fashion
device to test the bats for optomotor responses.
That is responses to moving patterns by moving their head and eyes.
So, I caught bats and put them in a glass cylinder inside a revolving drum.
On the inside of the drum i put striped patterns of different sizes. When spinning
the drum around the bat, the bat moves its head if it can see the pattern. When
the stripes are to narrow for the bats to recognise them as stripes, it does
not respond.
This is a rough but still an effective estimate of a bats visual resolving power
and has been used in many different groups of animals, from fish to horses.
The northern bats responded to the pattern equivalent to 1 degree of arc, which is close enough to the 0,8 degrees we estimated from the moth catching experiments.
I also tested some other vespertilionid bats,
Brown long eared bats showed reactions to the 0,5 degree pattern, which is relatively
good by bat standards,
and then I also tested a bunch of Myotis-species of which Daubentons bats showed
the best responses, of 2,5…the others, for example whiskered bats reacted
only to the 5 degree pattern.
To put this another way or to illustrate the size of objects detectable by
these bats relative to each other, i´m showing this slide.
In the beginning I also mentioned another vespertilionid which has an even better
resolving power than long-ears, the pallid bat, Antrozous, which have an acuity
of 0.25.
Now, I was interested in finding out if Plecotus, with its relatively good
acuity, actually can use its vision to detect prey.
After all they do have rather big eyes for being vespertilionids and they also
fly close to vegetation and experience problems with clutter, although they
perhaps use mainly passive listening to deal with that.
Nevertheless, to test their ability to use vision, I set up another experiment, this time together with Gareth Jones in Bristol and it was carried out indoors with captive bats, as opposed to the northern bat experiment.
We placed 2 feeding bowls of glass on the floor, inside two other ones and
then put some food in it. The food consisted of mealworms.
We then put two other bowls next to the first ones but this time with food in
between the bowls. Then we added some light to it, and got four different feeding
situations.
In the first the bats could find and get the food either by using vision or
sonar or both. In the second only echolocation cues were available, in the third
the mealworms were behind glass but clearly visible and the fourth served as
control and the mealworms were not detectable by sonar or by vision.
And finally we let a bat in, and every time it landed at one of the bowls we
shifted the bowls to different on beforehand randomised positions.
And this is what we found:
Perhaps as expected, the most popular bowl was the one that provided both visual
and echolocation cues. But maybe not as expected the vision bowl was visited
much more frequently then the bowl providing only sonar cues.
My conclusion is that long eared bats can use vision to detect mealworms, and
perhaps they even prefer using vision over sonar if possible
but to what extent they actually use this in the field remains of course unanswered.
And this leads us to the final conclusions, which are – if I may simplify
things a bit – that brown long eared bats can detect objects of at least
1 cm at a distance of a meter
whereas northern bats need a 2 cm large object to identify it at the same distance,
based on their visual acuity and behavioural performances.
And both species use their eyes when it is possible to do so
as a complement to other senses
even though their visual acuity is just average among bats – at least
in the case of the northern bats.
Thank you ..(I think there is plenty of time for questions)