Content uploaded by Gadi Katzir
Author content
All content in this area was uploaded by Gadi Katzir on Sep 19, 2019
Content may be subject to copyright.
The quality of an image formed on a vertebrate’s retina is
largely determined by the cornea and the lens. In air, the cornea
is the principal refracting agent of light rays and is responsible
for approximately two-thirds of the refractive power of the eye,
which, in humans, amounts to approximately 40diopters
(40D). This is because the cornea is bordered on its inner
surface by the aqueous humor, with a refractive index of 1.33,
and on its outer surface by air, with a refractive index of 1.0.
The cornea of birds plays an important role in accommodation
(Schaeffel and Howland, 1987; Glasser et al., 1994). In the
chicken Gallus gallus, changes in corneal curvature account
for up to 9 D of the total 15–17D of accommodation (Schaeffel
and Howland, 1987). The focusing power of the cornea in air
is a function of its curvature, and corneal accommodation is
brought about by curvature changes: the more curved it is, the
greater its power (Howland et al., 1997).
The refractive power of the cornea underwater is virtually
lost, as the media bathing its inner and outer surfaces (the
aqueous humor and water, respectively) are of similar
refractive indices. In a submerged eye, the lens becomes the
sole agent for accommodative adjustments and must provide
for the refractive power lost by the cornea if image quality is
to be retained (Fernald, 1990; Land, 1990; but see Pettigrew et
al., 2000). The lenses of fish (Fernald, 1990), amphibians
(Mathis et al., 1988), penguins (Sivak, 1980) and seals (Sivak
et al., 1989) tend to be spherical with an internal gradient of
refractive indices. This allows for the continuous refraction of
light within the lens itself and not merely at its surfaces
(Fernald, 1990; Sivak et al., 1989). These lenses exhibit a high
level of correction for spherical aberrations (Fernald, 1990;
Land, 1990; Sivak and Millodot, 1977).
In all vertebrate classes, there are species that perform
visually guided motor tasks in both air and water. Eyes that are
better adapted for terrestrial vision and are emmetropic (i.e. in
focus) in air tend to be hyperopic (i.e. far sighted) underwater,
while eyes better adapted for aquatic vision and are
emmetropic in water will tend to be myopic (i.e. near sighted)
in air. If retinal image is to remain sharp in both media, the eye
must cope with large changes in external refractive indices
(Martin, 1998; Sivak and Millodot, 1977; Howland et al., 1997;
Howland and Sivak, 1984; Land, 1990; Fleishman et al., 1988;
Glasser and Howland, 1996).
The corneas of penguins (Sphenisciformes) and albatrosses
(Procelariiformes) are relatively flattened and have refractive
powers lower than those of avian species of comparable eye
size (Tables 3, 4). Such corneas suffer relatively little loss of
power when submerged. Penguins, with corneal refractive
powers of 11–30D (Sivak and Millodot, 1977; Howland and
Sivak, 1984) are emmetropic in air and slightly hyperopic in
water, which is well within the compensatory power of the lens
(Sivak, 1976; Sivak et al., 1987; Howland and Sivak, 1984;
Table 3). Seals have flattened corneas and, in common with
penguins, make use of spherical lenses (Sivak et al., 1989).
Several other bird species that are pursuit-divers have
strongly curved corneas. Such corneas have a high refractive
power in air, and, indeed, pronounced capacities for lenticular
833
The Journal of Experimental Biology 206, 833-841
© 2003 The Company of Biologists Ltd
doi:10.1242/jeb.00142
In great cormorants (Phalacrocorax carbo sinensis),
corneal refractive powers, determined by
photokeratometry, ranged between 52.1diopters (52.1D)
and 63.2D. Photorefractive reflexes, determined by
infrared video photorefraction, indicated that in voluntary
dives the cormorants accommodate within 40–80ms of
submergence and with myopic focusing relative to the
photorefractor attained when prey was approximately one
bill length from the plane of the eye. Underwater, the
pupils were not constricted and retained diameters similar
to those in air. These results support previously reported
capacities of lenticular changes in amphibious birds yet do
not fully correspond with earlier reports in terms of the
coupling of iris constriction with accommodation, and
time course.
Key words: keratometry, IR photorefraction, lens, cornea,
accommodation, refractive power, amphibious vision, great
cormorant, Phalacrocorax carbo sinensis.
Summary
Introduction
Corneal power and underwater accommodation in great cormorants
(Phalacrocorax carbo sinensis)
Gadi Katzir1,* and Howard C. Howland2
1Department of Biology, University of Haifa, Oranim, Tivon 36006, Israel and 2Department of Neurobiology and
Behavior, Cornell University, Ithaca, NY 14850, USA
*Author for correspondence (e-mail: gkatzir@research.haifa.ac.il)
Accepted 13 November 2002
834
accommodation are observed in these species, indicating their
capacity to compensate for corneal loss of power during
dives. Accommodation in these species involves considerable
changes in lens curvature and is associated with highly
developed muscular mechanisms. In his pioneering studies,
Hess (1909, 1913; cited in Glasser and Howland, 1996)
demonstrated that, in cormorants (Phalacrocorax sp.), the lens,
when stimulated electrically, undergoes pronounced changes
in shape. The changes result in the lens being literally squeezed
into, and partially through, the rigid iris by the ciliary muscle.
The front surface of the now strongly curved lens produces a
region of high refractive power (>60D). Subsequent studies
have verified the extent of lenticular accommodation (e.g.
Sivak et al., 1977; Levy and Sivak, 1980; Table 3), although
there is still no agreement as to the precise muscular
mechanisms involved.
Two important aspects have been left open in studies of
lenticular capacities in pursuit-diving birds. First, a prevailing
assumption to date is that, when diving, pursuit divers such as
cormorants and mergansers (Mergus spp.) keep the retinal
image sharply focused. However, this assumption has not been
verified experimentally to date, while examples from other
vertebrate groups indicate that pursuit and capture of fish is not
necessarily coupled with high visual acuity. Thus, while otters
(Amblonyx cinerea cinerea; Schusterman and Barrett, 1973;
Balliet and Schusterman, 1971) and sea lions (Zalophus
californianus; Schusterman and Balliet, 1970) retain similar
grating acuity in air and in water, crocodiles (including
Gavialis, which feed exclusively on fish) do not accommodate
underwater (Fleishman et al., 1988), implying that crocodiles
can manage with blurred images.
The second aspect relates to methods of experimentation on
which conclusions on accommodation have been drawn. In most
experiments, drugs or electrical stimulation were employed to
elicit accommodation, or conditions of submergence were
achieved by forcibly holding the birds’ head underwater (e.g.
Goodge, 1960; Sivak et al., 1977; Levy and Sivak, 1980;
Table 3). To the best of our knowledge, states of accommodation
during voluntary dives of birds have been recorded, to date, in
penguins only (Howland and Sivak, 1984; Sivak et al., 1987)
and not in bird species that are said to have curved corneas and
may employ pronounced lenticular accommodation.
In the present study, we aimed to determine, in great
cormorant, P. carbo sinensis, (1) the refractive power of the
cornea in air, and thus the extent of compensatory power
required by the lens upon submergence, and (2) the capacity
to accommodate and the refractive range when freely diving.
Materials and methods
Seven great cormorants (Phalacrocorax carbo sinensis L.;
Cramp and Simmons, 1977) were tested for underwater
accommodation (infrared photorefraction), and five of
these seven birds were tested for corneal curvature
(photokeratometry). The birds were hand-reared and kept in a
large outdoor aviary in the Hula Valley, Israel. They were fed
on live and frozen fish and dived regularly for food in a large
pool. The birds’ age, sex and mass at the time of testing are
provided in Table 1.
Photokeratometry
The photokeratometer used to determine corneal curvature
was that described by Howland and Sayles (1985; also
Howland et al., 1997). It consisted of a 35mm Nikon SLR
camera, with an f/1.2, 55mm Nikkor lens mounted on a 31mm
extension tube, and operated at full aperture to minimize depth
of field. Eight light sources (the tips of fiberoptic light guides,
1.5mm diameter) were embedded in an aluminum ring, 75mm
in diameter, at the radii of a 67.5mm circle around the optic
axis, and the ring was mounted on the camera’s objective lens.
The proximal tips of the light guides were held in a bundle
directly in front of an electronic flash.
For calibration, we used a set of 10 steel ball bearings of
various diameters. Each ball was measured to the nearest
0.05mm using vernier calipers, and its photograph
(Kodakchrome/Ektachrome 100/200 ASA) was taken with the
photokeratometer mounted on a tripod. The focus of the
camera lens was set at infinity, yet, because of the extension
G. Katzir and H. C. Howland
Table 1. Corneal radii and corresponding refractive power
Mean right eye Mean right eye Mean left eye Mean left eye
Age Mass corneal radius corneal power corneal radius corneal power
Bird* (years) (kg) (mm) (D) (mm) (D)
F1 1 1.39 5.51 60.6 5.40 61.8
F2 1 1.63 5.87 56.9 5.92 56.4
F3 4 1.66 5.85 57.1 5.89 56.7
M1 4 2.40 6.31 52.9 6.67 50.1
M2 3 2.10 6.47 51.6 6.15 54.3
M3 5 1.95 – – – –
M4 4 2.26 – – – –
Corneal radius is provided as the mean of four measurements at 0°, 40°, 90° and 135° for the two slides used. M3 and M4 were tested for
infrared photorefraction only.
*F denotes female, M denotes male.
835Underwater accommodation in cormorants
tubes, the actual focus was at 150mm. In taking the
photographs, the camera-to-ball distance was adjusted for the
sharpest image. For each ball, the distances between opposite
reflections of the eight keratometeric reflection points were
determined with a measuring microscope accurate to 1.0µm.
This resulted in four measurements along the 0°, 45°, 90° and
135° meridians. The mean of the four measurements was
calculated and we then regressed the ball bearing diameters
against the mean reflection distances measured on the film
plane. We used this regression equation to estimate the corneal
radii (corresponding to half of the diameters of the calibration
ball bearings). To determine the dioptric power, F, of a cornea
(measured in diopters), the following equation was used:
F=337.5/R, where Ris the corneal radius (measured in mm).
This equation expresses the power of the human cornea as a
function of the radius of its first surface (Borish, 1995) and is
frequently applied in animal work.
The bird to be tested was held by one investigator, while
another investigator photographed each of the bird’s eyes with
the hand-held photokeratoscope. The room was lit by two 60 W
incandescent bulbs, positioned 2.5m above the bird. In taking
the photographs, the camera-to-bird distance was adjusted for
the sharpest image. Each eye of each of the five cormorants
was photographed 12 times. Pronounced eye movements and
rapid flicking of the nictitating membrane resulted in a
proportion of the slides being unsuitable for analysis. For each
eye of each bird, the two slides that provided the sharpest and
best-centered images of the photokeratometric light reflections
were used for extracting the values of the distances between
opposite reflections along the four meridians.
Infrared photorefractions
Photoretinoscopy was performed on the cormorants to
measure their natural accommodation with the use of an
infrared (IR) video photoretinoscope (Fig. 1). The
principles underlying the retinoscope are detailed
elsewhere (Schaeffel et al., 1987). In brief, the IR
retinoscope is based on a light source adjacent, and
eccentric, to a video-camera lens’ axis that projects
light rays parallel to the camera’s axis and records
the reflection from the fundus. IR is used to
minimize disturbance to the animals. The reflected
light appears as a crescent in the pupil, and the
position of the reflex indicates the sign of the
defocus relative to the camera. In hyperopia, the
reflex appears at the top of the pupil, while in
myopia the reflex appears at the bottom of the pupil.
The amount of defocus (D) may be obtained from
the size of the reflex: D=E/(2×A×DF×R), where Eis
the eccentricity of the light source, Ais the distance
of the camera to the eye, DF is the dark fraction in
the pupil and Ris the pupil radius (all dimensions
in meters). To improve the precision of the
measurements, light sources at five different
eccentricities are employed in a row, consecutively
providing five different crescents. Due to the high
mobility of the birds’ head and eyes, no attempt was made to
verify the amount of defocus by the use of correction lenses.
Filming was conducted when the bird was 1.2–1.4m from the
camera lens, and the horizontal distance of the eye to the glass
wall was approximately 5 cm. This provides an optical distance
(distance in air + distance in water/1.33) of approximately
1.0m.
Seven cormorants were tested over two consecutive days
(Table 1). A single bird was allowed into the test room from
its home cage and was encouraged to climb a short sloping
gangway, leading to a test aquarium. The room was lit by four
100W lamps and by indirect daylight from the open door. The
aquarium (80cm×40cm×50cm; length × width × height) was
kept three-quarters full of water, and the bird could perch
comfortably on its narrow side. One investigator then held a
small fish (Tilapia sp. or carp Cyprinus carpio) at the side
farthest from the bird and moved it to attract the bird’s
attention. He then dipped the fish and kept it underwater. The
bird would submerge its head, search for the fish and capture
it. Often, the fish was held against the outside of the aquarium
glass wall, and, if the bird attempted to capture it from within,
it was rewarded with a fish. The cormorants were acquainted
with this procedure and were continuously rewarded for
climbing the gangway and searching for fish in the
experimental aquarium. During the week preceding the tests,
they were fed daily in this manner.
As the cormorant climbed the gangway, the second
investigator, positioned so as to view the aquarium’s long axis
and to be level with the water surface, filmed the bird. Filming
was with the video camera held by hand, with the filming axis
perpendicular to the plane of approach of the bird. Filming was
continuous and conducted for as long as the bird searched for
fish. The termination of a test session was when the bird left
the aquarium and attempted to get back to the home cage.
A
B
IL
S
E
Image
plane Principal
plane Focal
plane Principal plane
of the camera
Fig. 1. Scheme of the optics of the infrared (IR) photoretinoscope used in this
study (based on Schaeffel et al., 1987). L, light source; S, black metal shield
covering the lower half of the camera’s lens; E, eccentricity (the distance of the
light source from the upper margin of the black shield); I, the highest ray above
the optical axis; A, the distance from the eye to the camera. Here, in a myopic
eye (focal plane at B), a real image of the light source is created at the image
plane, in front of the retina, and a blurred spot appears on the retina. Reflected
light entering the pupil from the back will refocus in the focal plane of the eye
(B) in front of the camera and subsequently diverge. Due to the shield, only rays
emerging from the bottom part of the pupil will be detected by the unvignetted
part of the aperture.
836
Video films (Sony 8mm) were digitized, and selected
sequences were captured using Adobe Premier 6. From these
sequences, the states of accommodation were determined.
Results
Photokeratometry
For each eye of each of the five cormorants, 2–3
photographs (with sharp light reflections and that were well
centred) were used (Fig. 2). Corneal radii, obtained from the
photographs, ranged from 5.4mm to 6.6mm. These radii
correspond to corneal powers of approximately 62D and 50D,
respectively (Table 1). No astigmatism was detected, and the
refractive power of the females’ corneas was higher than that
of the males’. Although it cannot be statistically verified, this
phenomenon is most probably related to body size rather than
gender.
Photorefraction
All cormorants readily climbed onto the side of the test
aquarium and took fish from the hand, both in air and
underwater. They searched for fish underwater and attempted
to capture them, even when the fish were presented beyond the
aquarium’s glass wall. Approximately 300 sequences of
accommodation footage were acquired for the five cormorants
tested for both air and water.
Photorefraction in air
In most recorded sequences, the eyes appeared to be in a
state of hyperopia, while emmetropia and myopia were
observed less frequently. An example of a state of myopia,
with a bird holding a fish halfway down its bill immediately
prior to swallowing, is depicted in Fig. 3B. Mean pupil
diameter in air during myopia and hyperopia did not differ
(9.9±0.9mm and 9.7±1.1mm, respectively). Mean refractive
states ranged from –0.44D to +0.45D (Table 2).
Photorefraction in water
The eyes appeared to be focused hyperopically relative to
the camera, while emmetropia and myopia were observed less
frequently. Mean pupil diameter did not differ between myopia
G. Katzir and H. C. Howland
Fig. 2. Photokeratometry. The light points are the reflection off a
cormorant’s cornea of the eight light sources in the photokeratometer
ring. The distance between opposite points of reflection is a measure
of the corneal radius of curvature. Scale bar, 6mm.
Fig. 3. (A) A cormorant holding a fish in its bill. The reflected
infrared light crescent at the bottom of the pupil indicates a state of
myopia. (B,C) States of myopia (light crescent at the bottom of the
pupil) observed underwater during capture attempts by two
individuals.
837Underwater accommodation in cormorants
and hyperopia (8.7±2.3mm and 8.9±0.7mm, respectively).
Mean refractive states ranged from –0.56D to +0.50D
(Table 2). Most sequences of a state of myopia occurred when
the target (fish) was close to the plane of the eye, one to two
bill lengths (approximately 6–12cm) away (Fig. 3B,C).
Discussion
We have found that the corneas in great cormorants are
curved and may provide up to 62D of refractive power in air.
We have further demonstrated that, while freely submerging
their head and searching for underwater prey, the cormorants
were well accommodated and could reach approximately –2D
in a state of myopia, relative to the camera at an optical
distance of 1.00m, for example, when focusing on a fish close
to the plane of the eye. Accommodation achieved by the great
cormorants thus exceeded 64 D, with transitions between states
of accommodation being frequent and rapid. The double
crested cormorant Phalacrocorax auritus was also found to be
emmetropic in air and in water (Sivak et al., 1977). Because
double crested cormorants are similar in size or smaller than
great cormorants, it is expected that they would exhibit a
corneal refractive power greater than 60D, yet the mean
corneal refractive power obtained was approximately 34D
only (table 2 in Sivak et al., 1977).
The demanding nature of amphibious vision in mammals
(e.g. Ballard et al., 1989) and birds (e.g. Glasser and Howland,
1996) has attracted attention for nearly a century (Glasser and
Howland, 1996). Hess (1909, 1913) first elucidated the
capacity of the highly pliable lens of a diving bird (cormorant)
to undergo dramatic changes in shape through the
exceptionally well-developed iris and ciliary muscles. Hess
(loc. cit.) concluded that these changes compensated for the
approximately 60D of corneal loss of power caused by
submergence. Subsequent studies indicated that, in penguins,
the corneas tend to be flattened and a penguin’s lens must
consequently accommodate for 10–30D lost during eye
submergence (Howland and Sivak, 1984).
In pursuit-diving bird species (but not penguins) tested to
date, the corneas seem to be curved and there exists a capacity
for marked change in lens shape through the action of highly
developed intraocular muscles. Thus, in the hooded merganser
(Mergus cucullatus), red headed duck (Aythya americana),
double crested cormorant and black guillemot (Cepphus
grylle), the ranges of accommodation span 40–80D (Table 3),
while in non-divers [e.g. pigeon (Columba livia) and chicken
(Gallus gallus)], the total range of accommodation is
approximately 10–20D. A high range of accommodation is
also found in the dipper (Cinclus mexicanus), a passerine that
captures insects underwater (Goodge, 1960). The capacity of
these species to accommodate underwater has been supported,
to date, by anatomical and histological characteristics of the
intraocular muscles and by changes in the lens shape elicited
by stimulation of intraocular muscles.
The ciliary muscle in birds comprises three muscle fiber
groups: the anterior muscle group (Crampton), the posterior
muscle group (Brücke) and the internal muscle group (Müller).
The Crampton muscle group is responsible for enhancing the
steepness of the central cornea and thus increasing the corneal
power, the Brücke group reduces the lens radius of curvature,
while the Müller group affects both cornea and lens.
Comparing four avian species, Pardue and Sivak (1997) found
that, in the hooded merganser, the majority of the ciliary
muscle fibers are in the posterior and internal fiber groups,
suggesting predominance of lenticular accommodation. By
contrast, in the pigeon, kestrel (Falco sparverius) and chicken,
the majority of muscle fibers are in the anterior muscle group,
suggesting emphasis on corneal accommodation. Hooded
mergansers also exhibit an especially large number of muscle
fibers in the peripheral iris, the region that is supposed to be
responsible for lens squeezing (Glasser et al., 1995). These
authors conclude that mergansers have the largest structures
associated with lenticular accommodation of the species
studied.
Accommodation in the hooded merganser is achieved by the
pressure of the malleable lens against the rigid iris plate, thus
resulting in the bulging of the lens through the pupil (Levy and
Sivak, 1980). The iris sphincter muscles presumably aid the
Table 2. States of accommodation and pupil size in cormorants in air and in water
Air Water
Hyperopia Myopia Hyperopia Myopia
(10) (10) (6) (14)
Refractive state (D)
Mean +0.45 –0.44 +0.50 –0.56
Range +1.00 to +0.11 –0.89 to –0.14 +0.89 to +0.25 –1.79 to –0.26
S.D. 0.33 0.07 +0.22 0.41
Pupil diameter (mm)
Mean 9.70 9.90 8.86 8.71
Range 8–11 9–12 8–10 –
S.D. 1.06 0.88 0.69 2.34
Numbers in parentheses indicate the total number of video frames analyzed for the seven birds (1–3 frames per bird).
838 G. Katzir and H. C. Howland
Table 3. Optical parameters and performance of avian species*
Common name Pupil
Species name Source Cornea Accommodation constriction†
Black guillemot Sivak et al., 1978 Radius: NA RS: Air=0 D, Water=+5.0 D. NA
Cepphus grylle Power: NA Methods: retinoscope, forced submergence.
Effective index: 1.370 N=6 birds.
Common goldeneye Sivak et al., 1985 Radius: NA Range: 66.8 D, peaks in approx. 0.4 s. 4→3.4
Bucephala clangula Power: 74.5‡Methods: electric elicitation of iris.
Effective index: NA N=1 bird.
Great cormorant Present study Radius: 6.7–5.4 RS: Air=+0.45 D; Water=–1.79 D. Not apparent
Phalacrocorax carbo Power: 50.1–61.8 Range: 64 D.
sinensis Effective index: NA Methods: voluntary dives. N=7 birds.
Double crested cormorant Sivak et al., 1977 Mean radius: 9.3 RS: Air=0 D; Water=+3.0 D to +36.0 D. NA
Phalacrocorax auritus Mean power: 34.8‡Methods: retinoscopy, forced submergence.
Effective index: 1.37 Anaesthetized underwater. N=3 birds.
Sivak et al., 1978 Radius: NA RS: Air=0 D, Water=+3.0 D. NA
Power: NA Methods: forced submergence.
Effective index: 1.371 N=2 birds.
Dipper Goodge, 1960 Radius: NA RS: Air:=–2.0 D to +1.5 D; Water=+40 D NA
Cinclus mexicanus Power: NA to +44 D; induced: 46–48 D.
Effective index: NA Methods: forced submergence. N=2 eyes.
Hooded merganser Levy and Sivak, 1980 Radius: NA RS: Air=0 D; induced=50 D. 3→1
Mergus cucullatus Power: NA Methods: nicotine sulfate, excised eyes. In tens of
Effective index: NA N=4 eyes. seconds
Sivak et al., 1978 Radius: NA RS: Air=+1.25 D; Water=+7.0 D.
Power: NA
Effective index: NA
Sivak et al., 1985 Radius: NA Range 78.3 D; peaks in approx. 0.5 s. 3.8→3.1
Power: 78.5‡Methods: electric stimulation of iris. 4.6→3.9
Effective index: NA N=4 eyes. 3.9→3.2
Mallard Levy and Sivak, 1980 Radius: NA RS: Air=+1.0 D; induced –5 D.
Anas plathyrhynchos Power: NA N=8 eyes.
Effective index: 1.373
Sivak et al., 1978 Radius: NA RS: Air=0 D; Water=45.0 D. Unpublished
Power: NA N=2 birds. data
Effective index: 1.373
Sivak et al., 1985 Radius: NA Range 2.7 D. 4.7→3
Power: 76.5‡Methods: electric elicitation of iris.
Effective index: 1.375 N=2 birds.
Red headed duck Levy and Sivak, 1980 Radius: NA RS: Air=0 D; induced 50 D. 5→1
Aythya americana Power: NA Methods: nicotine sulfate on excised eyes. In minutes
Effective index: NA N=4 eyes.
Sivak et al., 1985 Radius: NA Range 16 D; peaks in 0.15 s. 2.9→2.3
Power: 83‡Methods: electric elicitation of iris.
Effective index: NA N=1 bird.
Wood duck Sivak et al., 1985 Radius: NA Range 6.5 D; peaks in 0.3 s. 4→2.7
Aix sponsa Power: 74.5‡Methods: electric elicitation of iris.
Effective index: NA N=1 bird.
Gray headed albatross Martin, 1998 Radius: 14.5 NA NA
Diomedea chrysostoma Power: 23
Effective index: NA
*Data exclude those of penguins.
†Pupil constriction (mm) is during accommodation. RS, refractive state; NA, data not available.
‡Corneal power (measured in diopters) was calculated as F=337.5/R, where R(measured in mm) is corneal radius. Effective corneal
refractive index was measured in meters.
839Underwater accommodation in cormorants
formation of a rigid ring or plate against which the lens is
pushed to create lenticonus (Sivak and Vrablic, 1982). Studies
in the cormorant, double crested cormorant, dipper and red
headed duck (Hess, 1909, 1913; Goodge, 1960; Sivak et al.,
1977, 1985; Pardue and Sivak, 1997) lend support to the
existence of an iris accommodative mechanism capable of
producing dramatic lens changes in these species. Such
phenomena were not observed in non-diving species.
Curved corneas coupled with strong muscular mechanisms
have been demonstrated in otters as well as in several species
of pursuit-diving birds. Sea otters (Enhydra lutris; Murphy et
al., 1990), with a corneal refractive power of 59D, are nearly
emmetropic in both air and water. Based on their findings that
the iris musculature, meridional ciliary muscle and corneal
scleral plexus are highly developed in this species, Murphy et
al. (loc. cit.) concluded that sea otters rely on a powerful
accommodation mechanism for underwater vision. Similar
conclusions were drawn for the Canadian river otter (Lutra
canadensis; Ballard et al., 1989), although in this species the
states of accommodation while diving were not recorded.
Several studies point to a marked constriction of the pupil
while accommodating. Thus, Levy and Sivak (1980) reported
a fivefold decrease in pupil diameter in the red headed duck
and a threefold decrease in the hooded merganser during
accommodation stimulated with nicotine sulfate (Table 3).
Reduction of pupil aperture increases image quality and may
also be part of the formation of a rigid plate by the iris, against
which the lens is pressed. However, in the present study, we
observed only a slight constriction of the pupil (Fig. 3;
Table 2), suggesting that the coupling of pupillary changes and
Table 4. Optical parameters and performance of penguins
Common name
Species name Source Cornea Accommodation
Blackfoot Sivak, 1976 N=2 birds, 1 eye each. RS: air: emmetropic
Spheniscus demersus Mean radii: 14.8, 16.3 water: mean = +18.5 D.
Power: 22.8, 20.7* N=4 birds, both eyes each.
Gentoo Sivak and Millodot, 1977. N=2 birds, 1 eye each. RS: air: –0.5 D to –1.75 D
Pygoscelis papua Radii: 22.6, 22.1 water: +9 D to +13.0 D.
Power: 14.9, 15.3 Method: water-filled goggles.
Howland and Sivak, 1984 N=NA RS: air: mean = –0.4 D (9 eyes)
Radii: NA water: mean = +0.2 D.
Power: NA N=3 eyes.
Humboldt Sivak et al., 1987 N=4 birds. RS: air: emmetropic
Spheniscus humboldti Radii: 10.10–10.76 water: emmetropic.
Power: 33, 31.3* N=7 birds, both eyes each.
Martin and Young, 1984 N=NA RS: air: myopic
Radii: 11.33 water: emmetropic.
Power: 29.9*
King Sivak and Millodot, 1977. N=2 birds, 1 eye each. Refractive error: air: mean = –1.25 D
Aptenodytes patagonica Radii: 29.3, 30.3 water: mean = +10.0 D.
Power: 11.5, 11.1 Method: water-filled goggles.
Howland and Sivak, 1984 N=NA RS: air: emmetropic
Radii: NA water: emmetropic.
Power: NA Photorefractive limits: water 0.35 D.
N=4 eyes.
Magellanic Howland and Sivak, 1984 N=2 eyes Photorefractive limits:
Spheniscus magellanicus Mean radii: 11.5 air: mean = –0.1 D (N=5 eyes)
Mean power: 29.3 water: mean = +0.17 D (N=12 eyes)
Rockhopper Sivak and Millodot, 1977 N=3 birds, 1 eye each. Refractive error:
Eudyptes cristatus Radii: 18.5, 18.9, 19.8 air: mean = –1.0 D,
Power: 18.2, 17.8, 17.0 water: mean = +9.0 D.
Method: water-filled goggles.
Howland and Sivak, 1984 N=6 eyes Air: emmetropic; water: emmetropic.
Mean radii: 11.1 Photorefractive limits:
Mean power: 30.4 air: –0.25 D (N=10 eyes); water: 0 D
(N=2 eyes).
*Radii only are provided.
Corneal power (measured in diopters) was calculated as F=337.5/R, where R(measured in mm) is corneal radius.
RS, refractive state; NA, data not available.
840
underwater accommodation should be further investigated.
The initial large pupil diameter seen here was most probably
related to the dim illumination in the room, because, under
bright light, pupil diameters are less than 2mm (G. Katzir and
H. C. Howland, personal observation).
Certain important areas of amphibious vision are, however,
in need of further investigation. One area pertains to the actual
refractive power lost upon submergence in pursuit-diving
birds. Most studies imply that approximately 60D of corneal
refractive power is lost upon submergence, yet data on corneal
curvature and refractive power in air are usually lacking
(Tables 3, 4; but see Sivak et al., 1977, 1985). While the results
in the present study indicate that, indeed, the corneal refractive
power of the great cormorant may exceed 60D, both intra- and
interspecific differences are expected as a function of eye size.
Another area is related to the manner of eliciting
accommodation. Two approaches have been used to determine
states of accommodation: (1) by recording naturally occurring
changes in accommodation from a distance and (2) by
stimulating intraocular muscles. The former method, using
photorefraction and retinoscopy, was mostly employed in
studies of penguins (e.g. Howland and Sivak, 1984; Sivak and
Millodot, 1977; Sivak et al., 1987; Table 4), while the latter
was used predominately for other pursuit-diving species
(Table 3). Thus, studies that demonstrated changes in the lens
during accommodation were performed by electrical or
chemical stimulation of the eyes of anaesthetized or dead birds
(Hess, 1909, 1913; Goodge, 1960; Sivak et al., 1985; Levy and
Sivak, 1980). Simulation of submergence was obtained by
forceful holding of birds’ heads underwater (Goodge, 1960;
Sivak et al., 1977, 1978). Such methods may have yielded
results somewhat different from naturally occurring
phenomena. For example, the results in the present study
indicate that, in the great cormorant, transitions from hyperopia
in air to myopia underwater, i.e. over 60D, may occur within
40–80ms (1–2frames), while such changes obtained by
electrical or chemical stimulation were an order of magnitude
longer (Sivak et al., 1985). Moreover, while states of
emmetropia or myopia underwater were retained for only tens
of milliseconds in the present study, those achieved through
electrical stimulation (Sivak et al., 1985) were in the order of
hundreds of milliseconds.
Finally, visual performance underwater must be considered.
Amphibious animals that actively pursue fish underwater are
assumed to retain a sharp image on the retina. However, neither
the eyes of crocodiles (Fleishman et al., 1988) nor those of the
amphibious snakes studied by Schaeffel and Mathis (1991)
were well accommodated underwater, and, while clawless
otters (A. cinerea cinerea) are emmetropic in both air and
water, their acuity in either media is not high (Schusterman and
Barrett, 1973). Thus, during underwater pursuits, animals may
‘make do’ with blurred images or make use of non-visual
information. Moreover, visual acuity, the capacity of the visual
system to extract detailed information, is determined not only
by the eye’s optics but also by the underlying neural structures
(retina and brain), neural processing and the environment.
Compared with air, the underwater light environment has much
more deleterious effects on acuity, due to turbidity, pronounced
attenuation and chromatic absorption of light (Loew and
McFarland, 1990). To date, there are relatively few behavioral
studies on visual acuity, the performance of visually guided
tasks (Schusterman and Barrett, 1973) and the effects of the
underwater light environment in amphibious animals and none,
to the best of our knowledge, in pursuit-diving birds.
We are deeply indebted to the late Yaki Sternfeld, who, as
Dean of the Faculty of Sciences at the Haifa University and
while critically ill, made travel funds available to G.K., thus
enabling the testing of the birds. We are grateful to Ruth
Almon, Ido Izhaki, Tamir Strod, Ron Hoy, Tong Li, Bob
Wyttenbach, Elke Buschbeck, Birgit Ehmmer, Brandon
Loveall and Damian Ellias for their help. The research was
supported by grants NIH NEI EY02994 to H.C.H. and the
Israel Ministry of Science to G.K.
References
Ballard, K. A., Sivak, J. G. and Howland, H. C. (1989). Intraocular muscles
of the Canadian river otter and Canadian beaver and their optical function.
Can. J. Zool. 67, 469-474.
Balliet, R. F. and Schusterman, R. J. (1971). Underwater and aerial visual
acuity in the Asian “clawless” otter (Amblonyx cineria cineria). Nature 234,
305-306.
Borish, I. M. (1995). Clinical Refraction. Third edition. Chicago: Professional
Press.
Cramp, S. and Simmons, K. E. L. (1977). The Birds of the Western
Palearctic – Handbook of the Birds of Europe, the Middle East and North
Africa: Vol. I. Oxford: Oxford University Press.
Fernald, R. D. (1990). The optical system of fishes. In The Visual System of
Fish (ed. R. H. Douglas and M. B. A. Djamgouz), pp. 45-61. London:
Chapman and Hall.
Fleishman, L. J., Howland, H. C., Howland, M. J., Rand, A. S. and
Davenport M. L. (1988). J. Comp. Physiol. A 163, 441-443.
Glasser, A. and Howland, H. C. (1996). A history of studies of visual
accommodation in birds. Q. Rev. Biol. 71, 475-509.
Glasser, A., Murphy, C. J., Troilo, D. and Howland, H. C. (1995). The
mechanism of lenticular accommodation in the chick eye. Vision Res. 35,
1525-1540.
Glasser, A., Troilo, D. and Howland H. C. (1994). The mechanism of corneal
accommodation in chicks. Vision Res. 34, 1549-1566.
Goodge, W. R. (1960). Adaptations for amphibious vision in the dipper
(Cinclus mexicanus). J. Morphol. 107, 79-91.
Howland, H. C., Howland, M., Guinta, A. and Cronin, T. W. (1997).
Corneal curvature and refraction of central American frogs. Vision Res. 37,
169-174.
Howland, H. C. and Sayles, N. (1985). Photokeratometric and photorefractive
measurements of astigmatism in infant and young children. Vision Res. 25,
73-81.
Howland, H. C. and Sivak, J. G. (1984). Penguin vision in air and water.
Vision Res. 24, 1905-1909.
Land, M. F. (1990). Optics of the eyes of marine animals. In Light and Life
in the Sea (ed. P. J. Herring, A. K. Campbell, M. Whitfield and L. Maddock),
pp. 149-166. Cambridge, UK: Cambridge University Press.
Levy, B. and Sivak, J. G. (1980). Mechanisms of accommodation in the bird
eye. J. Comp. Physiol. A 137, 267-272.
Loew, E. R. and McFarland, W. N. (1990). The underwater visual
environment. In The Visual System of Fish (ed. R. H. Douglas and M. B. A.
Djamgouz), pp. 1-44. London: Chapman and Hall.
Martin, G. R. (1998). Eye structure and amphibious foraging in albatrosses.
Proc. R. Soc. Lond. B 265, 665-671.
Mathis, U., Schaeffel, F. and Howland, H. C. (1988). Visual optics in toads.
J. Comp. Physiol. A 163, 201-213.
Murphy, C. J., Bellhorn, R. W., Williams, T., Burns, M. S., Scheffel, F. and
G. Katzir and H. C. Howland
841Underwater accommodation in cormorants
Howland, H. C. (1990). Refractive state, ocular anatomy and accommodative
range in the sea otter (Enhydra lutris). Vision Res. 30, 23-32.
Pardue, M. T. and Sivak, J. G. (1997). The functional anatomy of the ciliary
muscle in four avian species. Brain Behav. Evol. 49, 295-311.
Pettigrew, J. D., Collin, S. P. and Fritches, K. (2000). Prey capture and
accommodation in the sandlance Limnichthyes fasciatus (Creediidae;
teleostei). J. Comp. Physiol. A 186, 247-260.
Schaeffel, F., Farkas, L. and Howland, H. C. (1987). Infrared
photoretinoscope. Appl. Optics 26, 1505-1509.
Schaeffel, F. and Howland, H. C. (1987). Corneal accommodation in the
chick and pigeon. J. Comp. Physiol. A 160, 375-384.
Schaeffel, F. and Mathis, U. (1991). Underwater vision in semi-aquatic
European snakes. Naturwissenschaften 78, 373-375.
Schusterman, R. J. and Balliet, R. F. (1970). Conditioned vocalizations as
a technique for determining visual acuity thresholds in sea lions. Science
169, 498-501.
Schusterman, R. J. and Barrett, B. (1973). Amphibious nature of visual
acuity in the Asian “clawless” otter. Nature 244, 518-519.
Sivak, J. G. (1976). The role of the flat cornea in the amphibious behaviour
of the blackfoot penguin (Spheniscus demersus). Can. J. Zool. 54, 1341-
1345.
Sivak, J. G. (1980). Avian mechanisms for vision in air and water. Trends
Neurosci. 12, 314-317.
Sivak, J. G., Bobier, W. R. and Levy, B. (1978). The refractive significance
of the nictitating membrane of the bird eye. J. Comp. Physiol. A 125, 335-
339.
Sivak, J. G., Hildebrand, T. and Lebert, C. (1985). Magnitude and rate of
accommodation in diving and nondiving birds. Vision Res. 25, 925-933.
Sivak, J. G., Howland, H. C. and McGill-Harelstad, P. (1987). Vision of
the Humboldt penguin (Spheniscus humboldti) in air and water. Proc. R.
Soc. Lond. B. 229, 647-472.
Sivak, J. G., Howland, H. C., West, J. and Weerheim, J. (1989). The eye
of the hooded seal, Cystophora cristata, in air and water. J. Comp. Physiol.
A165, 771-777.
Sivak, J. G., Lincer, J. L. and Bobbier, W. (1977). Amphibious visual
optics of the eyes of the double crested cormorant (Phalacrocorax auritus)
and the brown pelican (Pelecanus occidentalis). Can. J. Zool. 55, 782-
788.
Sivak, J. G. and Millodot, M. (1977). Optical performance of the penguin
eye in air and in water. J. Comp. Physiol. A 119, 241-247.
Sivak, J. G., and Vrablic, O. E. (1982). Ultrastructure of the intraocular
muscles of diving and nondiving ducks. Can. J. Zool. 60, 1588-1606.
