A 10230 Volume 92 Number 8 August 2005 Organ der Gesellschaft Deutscher Naturforscher und Ärzte Organ der Hermann von Helmholtz Gemeinschaft Deutscher Forschungszentren 1 3
Naturwissenschaften (2005) 00 DOI 10.1007/s00114-005-0010-0 SHORT COMMUNICATION Graham R. Martin Nigel Jarrett Phillip Tovey Craig R. White Visual fields in Flamingos: chick-feeding versus filter-feeding Received: 7 April 2005 / Accepted: 9 May 2005 C Springer-Verlag 2005 Abstract In birds, the position and extent of the region of binocular vision appears to be determined by feeding ecology. Of prime importance is the degree to which vision is used for the precise control of bill position when pecking or lunging at prey. In birds that do not require such precision (probe and filter-feeders), the bill falls outside the binocular field, which extends above and behind the head, thus providing comprehensive visual coverage. Flamingos Phoenicopteridae are highly specialised filterfeeders. They employ a unique technique that does not require accurate bill positioning in which the inverted head is placed between the feet. Feeding flamingos often walk forwards with the head pointing backwards. Here we show that in Lesser Flamingos Phoeniconaias minor visual fields are in fact the same as those of birds that feed by precision pecking and that feeding flamingos are blind in the direction of their walking. We suggest that this is due to the requirement for accurate bill placement when flamingos feed their chicks with crop-milk, and possibly when building their nest. We propose that chick-feeding may be the ultimate determinant of visual field topography in birds, not feeding ecology. Introduction In birds, the position and extent of the region of binocular vision appears to be determined by feeding ecology (Martin et al. 2004). Of prime importance is the degree to which vision is used for the precise control of bill position when pecking or lunging at prey (Martin and G. R. Martin ( ) C. R. White School of Biosciences, The University of Birmingham, Edgbaston, Birmingham, B15 2TT UK e-mail: g.r.martin@bham.ac.uk Tel.: +44-121-414-5598 N. Jarrett P. Tovey Wildfowl and Wetlands Trust, Slimbridge, Gloucestershire, GL2 7BT UK Katzir 1999). In birds that do not require such precision (probe and filter-feeders), the bill falls outside the binocular field, which extends above and behind the head, thus providing comprehensive visual coverage (Guillemain et al. 2002). In vertebrates with frontal eyes (e.g. primates, including humans, and some owl species), binocularity is associated with stereopsis: the perception of relative depth at close quarters that results from the simultaneous combination of information from the two eyes (McFadden 1994). However, in lateral eyed vertebrates, stereopsis has not been demonstrated and a possible neural substrate has not been found; hence, the function of binocularity in lateral eyed forms is not clear (Casini et al. 1993; Frost et al. 1994; Martin and Coetzee 2004). It has been proposed that in such forms binocularity may simply be the result of having part of each eye s monocular field extending across the sagittal plane to ensure that a section of each eye s visual field looks in the direction of travel (Martin and Katzir 1999). This arrangement provides a radially symmetrical portion of the optic flow-field in each eye. From this, information can be extracted that accurately specifies both direction of travel and time to contact the surface to which an animal s head (or in pecking birds the bill) is moving (Lee 1994; Davies and Green 1994). That each eye can act independently in the specification of such information is indicated in birds by the finding that Pigeons Columba livia can peck as accurately with one eye as with two (Zeigler et al. 1993). Forward placement of the eyes to provide such symmetrical optic flow-fields in each eye is, however, at the expense of comprehensive vision. This is because the monocular field of a vertebrate eye is rarely more than 180 in diameter (Martin 1994) and therefore any forward placement of the eyes inevitably results in a blind area to the rear of the head. Thus, the topography of the complete visual field is likely to be the result of selective pressures that have ensured sufficient binocularity for the control of movement towards targets while at the same time preserving more comprehensive visual coverage above and to the rear of the head. The latter is clearly beneficial for the detection of
predators or for the observation of others in a social group (Martin 1984). It is certainly clear that comprehensive visual coverage can be beneficial. Species with a blind area to the rear of the head spend more time in vigilance behaviour and less time feeding than species with comprehensive vision (Guillemain et al. 2002). When a Lesser Flamingo Phoeniconaias minor feeds with its head inverted, the bill floats on the water surface (del Hoyo 1992). The upper mandible is raised slightly and water is pumped into the mouth from around the bill tip and out at the base of the bill opening through a fine filter. This filter retains the blue-green algae and diatoms that are the Lesser Flamingos sole diet (Kear and Duplaix-Hall 1975). This feeding technique does not require the accurate visual control of bill position. We predicted that flamingo visual fields would match closely to those found in species whose feeding is controlled by non-visual cues, for example, filter-feeding ducks (e.g. Mallards Anas platyrhynchos and Shovelers A. clypeata) (Guillemain et al. 2002) and long-billed species that probe for invertebrates in soft substrates guided by tactile cues (Scolopacidae) (Piersma et al. 1998). In these birds, there is comprehensive vision about the head, including binocularity above and to the rear. In addition, the bill falls outside, or at the very periphery of, the visual field. Our aim was to determine whether such visual field topography is also present in Flamingos and thus to investigate further the general behavioural and ecological factors with which visual fields in birds are correlated. Methods Visual fields were measured in three adult Lesser Flamingos from the captive breeding colony held by the Wildfowl and Wetland Trust, Slimbridge, Gloucestershire, England. Visual field parameters were determined using an ophthalmoscopic reflex technique, which has been used in a range of birds of different phylogeny, ecology and feeding techniques and which readily permits interspecific comparisons (Martin and Katzir 1995; Martin and Katzir 1999; Martin et al. 2004). The procedures used were performed under guidelines established by the United Kingdom, Animals (Scientific Procedures) Act, 1986. Results The positions of the visual field margins in each of the birds were within 5 of each other at all elevations and the mean width of the frontal binocular field and of the blind sector above and behind the head are shown as a function of elevation in the median sagittal plane in Fig. 1. The frontal binocular field is narrow (maximum width 10 )but vertically long (90 ); the bill intrudes into the binocular field in the lower half. There is a blind sector both above and behind the head with a maximum width of 28 Discussion A precision-pecking visual field in a filter-feeding bird Given the specialised filter-feeding technique of Flamingos it is surprising to find that their visual field exhibits all of the principal features found in birds in which the bill is positioned with precision under visual control when feeding. Thus, the binocular field is narrow and vertically long, restricted to the frontal sector, with the bill positioned in the lower half (Fig. 1a and c). In addition, there is a relatively wide blind area above and to the rear of the head. Furthermore, the limits of the binocular field at the elevation of the bill are defined by the bill itself, suggesting that flamingos can see their own bill tip. All of these features are characteristic of birds that forage by precision-pecking or by lunging at evasive prey e.g. eagles (Accipitridae) (Martin and Katzir 1999), hornbills (Bucerotidae) (Martin and Coetzee 2004), herons (Ardeidae) (Martin and Katzir 1994), ostriches (Struthionidae) (Martin and Katzir 1995), penguins (Spheniscidae) (Martin 1999), starlings (Sturnidae) (Martin 1986) and pigeons (Columbidae) (Martin and Young 1983). It is also clear that when flamingos are feeding with the head inverted there is a blind area both in front of the bird (i.e. above its head) and behind (i.e. below the head), with respect to the direction of movement (Fig. 1b). Thus, a feeding flamingo walks forward in the direction in which it cannot see. We suggest that the characteristic head sweeping movements of feeding flamingos (Kear and Duplaix-Hall 1975) are a form of vigilance behaviour that allows scanning of these blind areas. Functionally, this is equivalent to the vigilance behaviour seen in those birds that interrupt feeding to raise their heads and scan in order to view the blind areas above and behind (Fernández-Juricic et al. 2004). However, in these birds scanning entails a break from feeding and therefore has a clear cost (Guillemain et al. 2002). In flamingos, this vigilance behaviour does not have such a cost, since the filtering of surface water continues while the head is swung around in the plane of the water surface. The function of binocularity in flamingos Why do flamingos have a frontal binocular field that encompasses the bill, as in precision-pecking birds? We propose that the answer lies primarily in the chick-feeding behaviour of flamingos. Those ducks and shorebirds that are filter or tactile feeders, and have comprehensive vision, produce precocial chicks that self-feed (Carboneras 1992; Piersma 1996). Flamingo chicks are also precocial but they are not able to self-feed. Adoption of the flamingo filter-feeding technique must await the development of the specialised bill apparatus and self-feeding does not occur until birds are 10 12 weeks old. Both parents feed their single chick on a liquid secretion, crop-milk, direct into the chick s bill with individual feeding bouts lasting up to
Fig. 1 Visual fields and feeding in Lesser Flamingos. (a) The width of the binocular field (shaded green, positive values) and the width of the blind area (shaded blue, negative values) as a function of elevation in the median sagittal plane. The region in which the bill defines the limits of the frontal field, and hence can be seen by the bird, is indicated by orange shading. (b) Head posture adopted by a feeding Lesser Flamingo. The head is inverted and the bill just enters the water surface. Yellow shading indicates the position of the blind area above and behind the head and shows that a feeding flamingo cannot see directly ahead of itself when moving forwards. (c) Perspective view of an orthographic projection of the binocular field as projected 20 min (del Hoyo 1992). Such chick-feeding would seem to require precise visually guided bill positioning so that crop milk is accurately dripped into the chick s mouth (Fig. 1d). We propose that this is the key reason why Lesser Flamingo visual fields are like those of precision-pecking birds, rather than those of filter-feeding birds. Another possibility for this precision pecking type of visual field in Flamingos could lie in the extent to which the bill is used in nest building. Flamingos build a conically shaped nest of mud, usually in shallow water (del Hoyo 1992). This protects the eggs and chicks from fluctuating water levels. The nest is made by scraping mud into position with the bill and this could require the birds to see the region around their bill tip. The nests of tactile feeding ducks and shorebirds are, however, usually simple depressions on the ground that do onto the surface of a sphere surrounding the bird s head. The grid shows conventional latitude and longitude at 20 intervals and the median sagittal plane of the bird s head is in the plane of the equator (which is vertical). The head is in the same posture as depicted in a and is in shown in frontal view in the drawing to the right of the grid. The frontal binocular field is vertically long but narrow with the bill tip projecting in its lower half. (d) Adult Flamingo feeding a chick (8 10 weeks old) by dripping crop milk directly into the chick s open bill. The bill tips meet precisely and the shaded area indicates the section of the binocular field in which the adult can see its own bill tip (drawing traced from a photograph) not appear to require the precise manipulation of materials with the bill. Binocular field width If the previous interpretation is correct then it is noteworthy that the flamingos precision bill positioning when chickfeeding is achieved with a binocular field whose width is approximately half that of pecking and lunging birds. With a10 maximum width it is similar to that seen in Mallards and Woodcocks which do not use their binocular field for the precise control of bill position, and probably use binocularity primarily for the control of flight. Factors that determine binocular field width are unknown but an important
parameter may be the velocity of approach towards, and therefore the time to contact, a target (Davies and Green 1994). High velocity and short approach times may require a higher redundancy of information to be extracted from the potential optic flow-field and this is best achieved with a larger forward facing sector of each eye s visual field i.e. wider binocularity. While flamingos need to place their bill precisely when chick-feeding, this does not have to be done at the kinds of velocity associated with pecking (Jäger and Zeigler 1991). Thus, there may be a trade-off between precision, velocity of approach and binocular field width. Chick-feeding versus filter-feeding We suggest that chick-feeding may be the ultimate determinant of visual field topography in flamingos, not feeding ecology. The requirements for building an elaborate nest with the bill may also be an important function of this visual field topography. We suggest that comprehensive vision of the kind seen in other filter-feeding birds may be possible only because these species do not need to see precisely what they are doing with their bill when both feeding themselves and when feeding their chicks, and possibly when constructing their nests. Acknowledgements We thank Terry Hornsey and staff of the Suffolk Wildlife Park, Kessingland, Suffolk, UK, for assistance with this work References Carboneras C (1992) Family Anatidae (ducks, geese and swans). In: del Hoyo J, Elliot A Sargatal J (eds) Handbook of the birds of the world, vol 1: Ostrich to ducks. Lynx Edicions, Barcelona, pp 536 628 Casini G, Fontanesi G, Bagnoli P (1993) Binocular processing in frontal-eyed birds. In: Zeigler HP Bischof H-J (eds) Vision, brain, and behavior in birds. MIT Press, Cambridge, Massachusetts, pp 159 171 del Hoyo J (1992) Family Phoenicopteriformes (flamingos). In: del Hoyo J, Elliott A Sargatal J (eds) Handbook of the birds of the world, vol. 1: Ostrich to ducks. Lynx Edicions, Barcelona, pp 508 527 Davies MNO, Green PR (1994) Multiple sources of depth information: an ecological approach. In: Davies MNO Green PR (eds) Perception and motor control in birds: an ecological approach. Springer-Verlag, Berlin, pp 339 356 Fernández-Juricic E, Erichsen JT, Kacelnik A (2004) Visual perception and social foraging in birds. TREE 19:25 31 Frost BJ, Wylie DR, Wang YC (1994) The analysis of motion in the visual systems of birds. In: Davies MNO Green PR (eds) Perception and motor control in birds: an ecological approach. Springer-Verlag, Berlin, pp 248 269 Guillemain M, Martin GR, Fritz H (2002) Feeding methods, visual fields and vigilance in dabbling ducks (Anatidae). Funct Ecol 16:522 529 Jäger R, Zeigler HP (1991) Visual field organization and peck localization in the pigeon (Columba livia). Behav Brain Res 45:65 70 Kear J, Duplaix-Hall N (1975) Flamingos. Poyser, Berhampsted Lee DN (1994) An eye or ear for flying. In: Davies MNO Green PR (eds) Perception and motor control in birds: an ecological approach. Springer-Verlag, Berlin, pp 270 291 Martin GR (1984) The visual fields of the tawny owl, Strix aluco L. Vision Res 24:1739 1751 Martin GR (1986) The eye of a passeriform bird, the European starling (Sturnus vulgaris): eye movement amplitude, visual fields and schematic optics. J Comp Physiol A 159:545 557 Martin GR (1994) Visual fields in woodcocks Scolopax rusticola (Scolopacidae; Charadriiformes). J Comp Physiol A 174:787 793 Martin GR (1999) Eye structure and foraging in King Penguins Aptenodytes patagonicus. J Comp Physiol A 141:444 450 Martin GR, Coetzee HC (2004) Visual fields in Hornbills: precisiongrasping and sunshades. J Comp Physiol A 146:18 26 Martin GR, Katzir G (1994) Visual fields and eye movements in herons (Ardeidae). Brain Behav Evol 44:74 85 Martin GR, Katzir G (1995) Visual fields in ostriches. Nature 374:19 20 Martin GR, Katzir G (1999) Visual field in short-toed eagles Circaetus gallicus and the function of binocularity in birds. Brain Behav Evol 53:55 66 Martin GR, Rojas LM, Ramirez Y, McNeil R (2004) The eyes of oilbirds (Steatornis caripensis): pushing at the limits of sensitivity. Naturwissenschaften 91:26 29 Martin GR, Young SR (1983) The retinal binocular field of the pigeon (Columba livia): English racing homer. Vision Res 23:911 915 McFadden SA (1994) Binocular depth perception. In: Davies MNO, Green PR (eds) Perception and motor control in birds: an ecological approach. Springer-Verlag, Berlin, pp 54 73 Piersma T (1996) Family Scolopacidae (sandpipers, snipes and phalaropes). In: del Hoyo J, Elliot A, Sargatal J (eds) Handbook of the birds of the world, vol 3: Hoatzin to auks. Lynx Edicions, Barcelona, pp 444 487 Piersma T van Aelst R, Kurk K, Berkhoudt H, Maas L (1998) A new pressure sensory mechanism for prey detection in birds: the use of principles of seabed dynamics? Proc R Soc Lond B Biol Sci 265:1377 1383 Zeigler HP, Jager R, Palacios AG (1993) Sensorimotor mechanisms and pecking in the pigeon. In: Zeigler HP, Bischof H-J (eds) Vision, brain and behavior in birds. MIT Press, Cambridge, Massachusetts, pp 265 283