Study

ISSN: 0006-3657 (Print) 1944-6705 (Online) Journal homepage: http://www.tandfonline.com/loi/tbis20

Why do Numenius have decurved bills?

N. C. Davidson , D. J. Townsend , M. W. Pienkowski & J. R. Speakman

To cite this article: N. C. Davidson , D. J. Townsend , M. W. Pienkowski & J. R. Speakman (1986) Why do curlews Numenius have decurved bills?, Bird Study, 33:2, 61-69, DOI: 10.1080/00063658609476896

To link to this article: http://dx.doi.org/10.1080/00063658609476896

Published online: 24 Jun 2009.

Submit your article to this journal

Article views: 194

View related articles

Citing articles: 10 View citing articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=tbis20

Download by: [University of Aberdeen] Date: 20 November 2015, At: 08:22 Bird Study (1986) 33, 61-69

Why do curlews Numenius have decurved bills?

N.C. DAVIDSON 1* , D.J. TOWNSHEND 11- , M.W. PIENKOWSKI 2 and J.R. SPEAKMAN 3 ' Department of Zoology, University of Durham, South Road, Durham DM 3LE, UK 2 Chief Scientist Directorate, Nature Conservancy Council, Northminster House, Peterborough PE11UA, UK "Department of Zoology, University of Aberdeen, Tillydrone Avenue, Aberdeen AB9 2TN, UK

The functions of the long, decurved bill of the Common are compared with those of the straight bill of the Bar-tailed . Comparison is also made with the dimensions of other of curlews and . We argue that decurvature is adapted to a technique of prey capture in which the bill tip follows a complex three-dimensional search path, and that the long bill of the Common Curlew is adapted to the intact removal of long prey (e.g. worms) from mudflats. The evidence indicates that strong selection for bill form operates on the wintering grounds.

he bill morphology of a bird is closely cor- foraging behaviour of several species of shore- T related with its feeding habits. However in in north-east Britain (e.g. Pienkowski species that occupy several habitats or take dif- 1973; Evans et al. 1979; Davidson 1980; Tow ns- ferent foods at different times of year, it can be hend 1981, 1982, 1985; Pienkowski et al. 1984; difficult to identify to which, if any, of several Speakman 1984a). In particular, we compare purposes the bill form is chiefly adapted. Thus the foraging behaviour of the Common Curlew Owens (1984) suggested that the decurved bill and Bar-tailed Godwit Limosa lapponica, since of the Common Curlew Numenius arquata had both of these species probe deeply into arisen by selection favouring either a greater sediments to capture polychaete worms, search-arc for detection of prey beneath the especially ragworms Nereis divers icolor and

Downloaded by [University of Aberdeen] at 08:22 20 November 2015 surface of mudflats in winter (an idea earlier lugworms Arenicola marina. We then compare proposed by Burton 1974) or easier capture of the dimensions of the Common Curlew with insects amongst long vegetation on the breed- those of other curlews Numenius spp. and of ing grounds. godwits Limosa spp. and consider briefly the Owens (1984) suggested that careful obser- incidence of bill decurvature amongst other vation of foraging behaviour may yield clearer groups of birds. We argue that decurvature is evidence than he could provide for the func- an adaptation for probing along complex tion of the Common Curlew's decurved bill. pathways and to aid extraction of worms with- We present such evidence here, drawn from out breaking them, and that the very long our observations during the last 12 years of the decurved bill of the Common Curlew is chiefly Present addresses: *England Headquarters, Nature an adaptation for probing along such pathways Conservancy Council, Northminster House, Peter- deep into mudflats. This differs from Owen's borough PE1 lUA, UK; +Nature Conservancy view. We also comment on several points of Council, Archbold House, Archbold Terrace, Owens' arguments which are not supported by Newcastle NE2 1EG, UK. the available evidence. 62 N.C. Davidson et al.

RESULTS AND DISCUSSION Burton (1974) pointed out that decurvature aids vertical insertion of the bill tip towards prey Prey capture techniques of Common Curlews detected whilst walking across the mudflats, and Bar-tailed Godwits on mudflats and that when the birds are foraging visually the bill tip may have to be moved a shorter Common Curlews and Bar-tailed Godwits distance to the point of insertion into the foraging on mudflats can detect prey both by sediment. When foraging by touch, Common touch and by sight. In tactile foraging, both Curlews can probe the surface some distance in species walk over the flats frequently inserting advance of their feet (see Fig. 1(ii)) whereas the tip of the bill into the sediment, and some- Bar-tailed Godwits, which likewise insert the times then probing deeply—probably only bill tip vertically, must do so much closer to after detection of the prey (Townshend 1982; their feet. Since some prey become less active, Cramp & Simmons 1983; pers. obs.). The exact and move deeper, when a shorebird walks way in which prey is detected is unknown: it across the mud (Goss-Custard 1970), detection may be by direct contact, by detection of of prey in advance of the feet may be of con- vibrations from the worm, or by detection of an siderable benefit. The ability of prey to detect occupied burrow by taste, as contributes to the passage of a implies that Owens' prey location in alpina and (1984) suggestion, that the decurvature confers C. alba (van Heezik, Gerritsen & an element of surprise through a sideways Swennen 1983; Cramp & Simmons 1983). In attack on a worm in its burrow, is improbable, visual foraging, the deep probe is not preceded since the bill tip of the Common Curlew ends by a shallow peck at the same place (pers. up closer to the feet than does the bill tip of a obs.). Bar-tailed Godwit. A decurved bill is more The major difference between the species lies readily manipulated than a straight bill along a in their technique of deep probing for prey. complex capture path (following a burrow Bar-tailed Godwits plunge their straight bills system) within the sediment, since a change in rapidly deep into the sediment, almost in- direction can be achieved by a rotation of the variably vertically downwards. Thereafter, the head or the neck, rather than requiring a lateral bill may be vigorously worked up and down, movement of the entire head and neck. This and the bird may pivot round the hole with the rotation is illustrated in Owens (1984), albeit bill deep in the mud, perhaps to get a better when the Common Curlew has probed almost grip on its prey. This rapid plunge probably to its fullest extent with the bill totally helps to catch prey such as lugworms that can immersed. retreat beyond the reach of the bill (see Smith & Evans 1973). In contrast, after detecting a Removal of prey from the substrate worm a Common Curlew probes deep into the mud more slowly than does a Bar-tailed God- Common Curlews always, and Bar-tailed wit. The bill may initially follow a simple path Godwits usually, must remove a worm from

Downloaded by [University of Aberdeen] at 08:22 20 November 2015 into the mud but more often during repeated the substrate before swallowing it (Burton probes at a site the bill tip follows a complex 1974, 1986). However, worms are fragile and route which involves movements both in line break easily. Breakages are disadvantageous with and perpendicular to the axis of the bill. because it takes longer to grasp and remove Unlike Bar-tailed Godwits, Common Curlews several small pieces of worm than a whole one seldom pivot round the probe site with the bill and because pieces of broken worm are some- deeply inserted. The probing of Common times left in the burrow after the bird moves Curlews is described further in Burton (1974) on. This is particularly so for lugworms, which and Townshend (1982). are pulled out 'tail first', since any breakage leaves behind the much thicker head end, which contains 65% of the energy content of Advantages of a decurved bill for foraging on mudflats the worm (Smith 1975). When pulling a worm from the sediment, A decurved bill has several advantages over a the risk of breakage may be higher if the worm straight bill for such foraging on mudflats. is not pulled vertically clear of the mud or if the

Curlew bills 63

a) Bar-tailed Godwit

)?

Figure 1. Techniques used by (a) a Bar-tailed Godwit and (b) a Common Curlew to pull a worm of similar length (relative to the bill) from the mud. Drawings (i) to (iii) show the normal sequence. Drawing (a (iv)) illustrates the difficulty of pulling a long worm vertically from its burrow with a straight bill (see text). Points A and B indicates the points at which the risk of fracture is increased if the angle between worm and substrate, or worm and bill, changes during extraction.

angle between worm and bill changes. It is thus the bill angle increasingly towards horizontal

Downloaded by [University of Aberdeen] at 08:22 20 November 2015 best to pull worms vertically from the sedi- whilst the pull is made (Fig. 1(iii)), thus increas- ment. Bar-tailed Godwits almost invariably, ing the risk of breakage of the worm where it is and Common Curlews usually (except in very grasped by the bird. In contrast, the decurved soft mud), withdraw the bill from the insertion bill of a Common Curlew allows a vertical grip hole so that the part of the bill at the mud to be maintained throughout the vertical surface is vertical (Fig. 1). Fig. 1 shows that a removal of a worm of the same length. straight-billed bird such as a Bar-tailed Godwit Both Bar-tailed Godwits and Common must raise its head relatively higher above the Curlews often break worms during capture ground to do this than must a bird with a attempts. For example, Bar-tailed Godwits decurved bill, if the worm is a long one. In- feeding on the polychaete worm Nerine cir- deed, a Bar-tailed Godwit probably cannot ratulus on a sandy beach broke at least 53% of raise its head high enough to extract worms worms longer than 4 cm (i.e. longer than half much more than the same length as its bill (see the bill length) when extracting the worms Fig. 1(iv)). To pull a long worm vertically from from the sand (N. C. Davidson, unpublished). its burrow, a Bar-tailed Godwit must change Similarly, Smith (1975) reported that 13% of 64 N.C. Davidson et al.

Table 1. Breakage of worms by Common Curlews and Bar-Tailed Godwits feeding on the same mudflat within Seal Sands, Teesmouth

Sept Oct Nov Dec Jan Feb Mar

Common Curlew Worms extracted 276 7 40 78 55 79 134 Percentage broken 16 0 8 15 29 38 10 Bar-tailed Godwit Worms extracted 0 0 32 41 235 12 0 Percentage broken 25 24 30 25

Worms were considered broken either when two or more worm fragments were removed or when removal of a worm fragment was followed by further probing at the same site.

lugworms captured by Bar-tailed Godwits removed from the substrate before they are were tails only. Pienkowski (1973) noted that swallowed. Common Curlews appeared to break a smaller A straight bill is more suitable mechanically proportion of ragworms than did Bar-tailed for rapid thrusting in the substrate, as by Bar- Godwits. He considered that this accounted tailed Godwits (Burton 1974). On estuarine for the apparently higher proportion of old sediments, this generally restricts Common worms (more than 1 year old) recorded in the Curlews to foraging on soft muds, whereas Common Curlew's diet than in the Bar-tailed Bar-tailed Godwits can use a variety of sedi- Godwit's: broken large worms were some- ments from soft muds to firms sands. How- times recorded as young (0-1 year old) worms ever, Common Curlews also forage extensively if their thickness could not be seen. When on rocky shores (e.g. Tay & Orkney Ringing feeding on ragworms on the same small mud- Groups 1984) and pasture fields (Davidson flat within Teesmouth, both species broke 1980; Townshend 1981). Perhaps they can use many worms (Table 1). the ability to orient a decurved bill along a complex path, in the same way as we have Other advantages and disadvantages described for probing within mudfiats, to search the complex three-dimensional struc- We have described some advantages of using a ture of a rocky shore and amongst the matted decurved bill to capture worms within mud- grass roots on fields. flats. There are, of course, also disadvantages. In particular a decurved bill is structurally Dimensions of curlews and godwits weaker than an equivalent straight one (Burton

Downloaded by [University of Aberdeen] at 08:22 20 November 2015 1974) and so cannot be pushed into firm sub- Common Curlews are much larger than Bar- strates. A Common Curlew's bill is heavily tailed Godwits, so being able to catch large reinforced to withstand the high stresses on it. prey is important if they are to get enough food As a consequence of this reinforcement, the to survive in midwinter. Since Common tongue is very short, so that prey grasped with Curlews have absolutely longer bills than Bar- the tip of the bill cannot be moved to the tailed Godwits they can certainly probe deeper mouth, using the tongue, whilst the bill is into the mud and so obtain more and larger within the substrate. Common Curlews must worms especially in midwinter when worms remove their bill from the substrate before the are deeper in the mud (Muus 1967; Dugan prey can be moved to the mouth by head jerk- 1981). The difference in bill length might be a ing (Burton 1974, 1986). Godwits have some consequence only of Common Curlews being bill thickening and tongue reduction, but they bigger than Bar-tailed Godwits. To examine can move prey up the bill with the tongue to this, we have compared mean bill length some extent (Burton 1986), although our (exposed culmen) with mean wing length observations suggest that most worms are (maximum chord) in all Numenius and Limosa Curlew bills h`,

species, using data from Prater, Marchant & length) will be equal to 1.0 on the log scale Vuorinen (1977) and Cramp & Simmons large species have disproportionately large (1983). Wing length is highly correlated with bills, the slope will be greater than 1.0. The overall body size and mass in many shorebirds slope was estimated by calculating the slope of (Davidson 1983, and unpublished). Separate the first principal axis of the relationship (Sokal means were used for different races and sexes & Rohlf 1981). where possible, since the differences between Fig. 2 shows that godwits have longer bills these are often as large as those between than curlews of the same size. However larger species. The data were log-transformed before species of Curlews have relatively larger bills analysis because this leads to easy inter- than smaller species, since the slope of the line pretation: if bill length is proportional to body is significantly greater than 1.0. The same is size over the range of species, the slope of the true of godwits although the rate of increase of line relating bill length to body size (wing bill length with body size is only two-thirds

250 300 350 Downloaded by [University of Aberdeen] at 08:22 20 November 2015 200 wing length (mm)

Figure 2. The relationship between bill length and body size (measured as wing length) for godwits (open symbols) and curlews (closed symbols). Circles are males, squares are females and triangles are samples that were not subdivided by sex; lines join the two sexes of a subspecies or species. Numbers refer to species and subspecies as follows: 1 Black-tailed Godwit Litnosa limosa, a L. 1. islandica, b L. 1. limosa, c L. 1. 1nelaneuroides; 2 L. haetnastica; 3 Bar-tailed Godwit L. lapponica, a L. 1. lapponica, b L. 1. baueri; 4 L. fedoa; 5 Numenius minutus; 6 Whimbrel N. phaeopus, a N. p. phaeopus, b N. p, variegatus, c N. p. hudsonicus; 7 Slender-billed Curlew N. tettuirostris; 8 Common Curlew N. arquata, a N. a. arquata, b N. a. orientalis; 9 Long-billed Curlew N. americanus; 10 Eastern Curlew N. tnadagascariensis; 11 N. borealis; 12 Bristle-thighed Curlew N'. tahitensis. Data are from Prater eta!. (1977) and Cramp & Simmons (1983). The slopes of the first principal axes are shown for curlews (solid line) and godwits (broken line). 95% confidence limits .for the slopes are: curlews 2.556-

3.496, godwits 1.317-4.228. The regression equations are: Y = 7.186x 10 -9( 2- 9599 for curlews and

Y = 1.0308 x 10 -3X 2. " 93 for godwits, where Y is bill length (mm) and Xis wing length (mm). 66 N.C. Davidson et al.

that of curlews. In both genera, the larger The tarsus length of some godwits, notably females tend to have disproportionately larger the Black-tailed Godwit L. limosa and Marbled bills compared with the males. Godwit L. fedoa (but not the Bar-tailed The problems of extracting long worms with Godwit), is greater than that of similar sized a long bill (as outlined above) will be less severe curlews. However, this seems not to be an for birds with long legs, since then the bill can adaptation for foraging on exposed mudflats be raised higher above the substrate. To but for foraging in deep water (Wishart & Sealy examine this we compared mean tarsus length 1980; Cramp & Simmons 1983). Likewise, the (from Prater et al. 1977; Cramp & Simmons longer legs of female Bar-tailed Godwits allow 1983) with mean bill length (Fig. 3). If leg length them to forage in deeper water than males, increases in proportion to bill length, the slope which feed more often on exposed mudflats. of this relationship will be 1.0. Such an increase This is important in cold weather, when a in tarsus length occurs in godwits (Fig. 3): the much higher proportion of prey are accessible slope of the relationship does not differ signi- to Bar-tailed Godwits in water than on nearby ficantly from 1.0. However, in curlews the mud (Smith & Evans 1973). In contrast, slope of the relationship is significantly less Common Curlews seldom forage in deep than 1.0, so the increase in bill length in large water. Foraging visually in water is difficult species is not accompanied by a proportionate because of diffraction, except when aiming increase in tarsus length. Similar trends are from vertically above. The straight bills of evident between sexes within most species and godwits permit this but the decurved bills of subspecies. curlews do not. Downloaded by [University of Aberdeen] at 08:22 20 November 2015

50 100 150 200 bill length (mm) Figure 3. The relationship between tarsus length and bill length for godwits and curlews. Symbols are as for Fig. 2. 95% confidence limits for the slopes of the first principal axes are:

curlews 0.396-0.548, godwits 0.104-10.700. The regression equations are: Y = 7.6805 X°. 4698 for curlews and Y = 0.6818 X' ('° for godwits, where Y is tarsus length (mm) and Xis bill length (mm). Curlew bills 67

Although long legs can aid the extraction of notably Curlew Calidris ferruginea, long prey, they have disadvantages. Strong and C. maritima. All winds cause increased heat loss and buffeting forage in winter by gentle probing into (which prevents accurate location and capture complex substrates, Curlew and of prey). This may select against long legs in Dunlins on mudflats and Purple Sandpipers on species, such as Common Curlew, Bar-tailed rocky shores. Bill decurvature may aid such Godwit and Hudsonian Godwit L. haemastica, foraging in the same way as we have described that feed chiefly on exposed mudflats and for Common Curlews. sandflats (Davidson 1981). Effects of wind on The three main foraging habitats of Common foraging shorebirds were reviewed by Evans Curlews (mudflats, rocky shores and grass- (1976). Buffeting is known to be a particular lands) are complex, like those of other birds problem for long-legged species: Redshanks with decurved bills. Whilst decurvature in the totanus find less food in strong winds bills of Common Curlews can be interpreted as and during gales they leave the mudflats on an adaptation primarily to locating insects which they usually feed and seek sheltered amongst vegetation (Owens 1984), this does feeding sites (Davidson 1982; Speakman not explain the variation in length and curva- 1984b). Marbled Godwits find less food when ture between different curlew species, since all feeding in exposed rather than sheltered sites forage for invertebrates (and berries) on the during gales (Wishart & Sealy 1980). breeding grounds in a similar way, amongst These comparisons of bill length and leg similar types of vegetation. Furthermore, length indicate that the decurved and long bill similar foods are taken by godwits on their of the Common Curlew allows the intact breeding grounds (Hale 1980; Cramp & removal of large worms from mudflats. The Simmons 1983). curvature permits bills to be very long without Short bill length in curlews seems to be the need for the legs to be as long as would be associated with foraging on during necessary with a straight bill, so that the effects winter. Firstly, small curlews that forage of buffeting are reduced. The thrusting feeding chiefly amongst vegetation, especially the technique used by Bar-tailed Godwits requires Little Curlew N. minutus and formerly the a straight bill but the effects of buffeting may Eskimo Curlew N. borealis (Cramp & Simmons have restricted any evolutionary increase in leg 1983), have relatively short bills (Fig. 2). length that would allow the vertical removal of Secondly, Townshend (1981) showed that it large worms intact. was chiefly male (i.e. short-billed) Common Curlews and not females, that moved from General functions of decurvature mudflats to forage on pastures as temperatures fell in autumn. Strongly decurved bills occur in a number of Owens (1984) argues, quoting Hale (1980), groups of birds. The common feature appears that since wintering shorebirds have a super- to be a foraging method involving gentle abundance of food at most times except during

Downloaded by [University of Aberdeen] at 08:22 20 November 2015 probing along precise but complex routes. For excessive cold, selection for mOrphological example, wood-hoopoes (Phoeniculidae), characters is unlikely on the wintering grounds especially the Scimitar-bill Phoeniculus and so may be strongest on the breeding cyanomelas and Abyssinian Scimitar-bill P. grounds. The case for selection operating on minor, probe for insects in fissures and crevices the wintering grounds cannot be so easily dis- of bark, and also into the narrow openings of missed: it is precisely during periods of severe ant-galls (Davidson 1976; Davidson & Ligon winter weather, when energy demands are 1987). Treecreepers Certhis (Certhiidae) also high and food scarce (Evans 1976; Davidson probe into bark crevices; Hoopoes Upupa epops 1981), that any selection- is likely , to by (Upupidae) probe fissures in the ground for strongest. Such selection would operate even if insects; honey-creepers (Drepaniidae), honey- severe weather occurred only every few years: eaters (Meliphagidae) and sunbirds (Nectarini- Common Curlews, like many shorebirds (see, idae) with decurved bills probe into the corollas e.g. Davidson 1981; Davidson & Evans 1982; of flowers. Pienkowski et al. 1984), can have difficulty in Several other have decurved bills, meeting energy requirements in severe 68 N.C. Davidson et al.

weather, and Chapman (1889) and Bolam Davidson, N.C. (1976) The evolution and system- (1912) noted that Common Curlews often atics of the Phoeniculidae. Bull. Nigerian Orn. Soc. became very emaciated and died in extremely 41,2-17. cold weather. Furthermore, Evans & Pien- Davidson, N.C. (1980) Seal Sands Feasibility Study. kowski (1984) found that even in mild winters, Report to Cleveland County Council and the mortality of Common Curlews and several Nature Conservancy Council. Cleveland County Council, Middlesbrough. other shorebirds was at least as high in winter Davidson, N.C. (1981) Survival of shorebirds as on migration and breeding grounds com- (Charadrii) during severe weather: the role of bined. It is thus likely that there is strong nutritional reserves. In Feeding and Survival selection for bill morphology in Common Strategies of Estuarine Organisms (Eds N.V. Jones Curlews in winter. Such selection has been & W.J. Wolff), pp. 231-249. Plenum Press, New found for Oystercatchers Haematopus ostra- York. legus (Pounder, Pounder & Greenwood 1971; Davidson, N.C. (1982) Changes in the body con- Swennen & Duiven 1983; Swennen et al. 1983). dition of Redshanks during mild winters: an Bill decurvature in curlews may have evolved inability to regulate reserves? Ringing Migration, 4, initially as an adaptation to foraging, 51-62. Davidson, N.C. (1983) Formulae for estimating the and Burton (1986) considers that the small lean weights and fat reserves of live shorebirds. short-billed species (which still feed chiefly on Ringing Migration, 4, 159-166. grasslands) resemble this ancestral stock. Davidson, N.C. & Evans, P.R. (1982) Mortality of However our evidence indicates that the bill Redshanks and Oystercatchers from starvation in decurvature is now used by the Common severe weather. Bird Study, 29, 183-188. Curlew for foraging in a variety of habitats by Davidson, N.C. & Ligon, J.D. (1987) Phoeniculidae: complex three-dimensional probing, and that Wood-hoopoes. In The Birds of (Eds E.K. the increased length of the Common Curlews' Urban, C.H. Fry & S. Keith), Vol. 3. Academic bill has evolved for such probing on mudflats. Press, London (in press). Dugan, P.J. (1981) Seasonal movements of shore- birds in relation to spacing behaviour and prey availability. PhD thesis, University of Durham. ACKNOWLEDGMENTS Evans, P.R. (1976) Energy balance and optimal Our ideas on the bill form of curlews have been foraging strategies in shorebirds: some impli- aided by stimulating discussions with col- cations for their distributions and movements in the non-breeding season. Ardea, 64, 117-139. leagues at the University of Durham and else- Evans, P.R., Herdson, D.M., Knights, P.J. & Pien- where, especially Dr P.R. Evans, Dr P.J. kowski, M. W. (1979) Short-term effects of rec- Dugan, S. Gregory, Dr P. Monaghan, Dr N. lamation of part of Seal Sands, Teesmouth, on Metcalfe and Dr A.G. Wood. We thank Dr P.R. wintering waders and Shelduck. Oecologia, 14, Evans, Dr P.N. Ferns, H.Y. Siman and J.D. 183-206. Uttley for helpful comments on the manu- Evans, P.R. & Pienkowski, M.W. (1984) Population script, N. Aebischer for advice on computing, dynamics of shorebirds. In Behavior of Marine and S. Carter for help with drawing figures. , Vol. 5 (Eds J. Burger & B.L. 011a), pp. Downloaded by [University of Aberdeen] at 08:22 20 November 2015 83-123. Plenum Press, New York. Goss-Custard, J.D. (1970) Dispersion in some over- Social Behaviour in REFERENCES wintering wading birds. In: Birds and Mammals (Ed. J.H. Crook), pp. 3-35. Bolam, G. (1912) Birds of Northumberland and the Academic Press, London. Eastern Borders. H.H. Blair, Alnwick. Hale, W.G. (1980) Waders. Collins; London. Burton, P.J.K. (1974) Feeding and the Feeding Muus, B.J. (1967) The fauna of Danish estuaries and Apparatus in Waders. British Museum (Natural lagoons. Distribution and ecology of dominating History), London. species in the shallow reaches of the mesohaline Burton, P. J.K. (1986) Curlews' bills: some anatomical zone. Meddr. Danm. Fisk.-og Havunders, 5, 1-316. notes. Bird Study, 33, 70. Owens, N.W. (1984) Why do Curlews have curved Chapman, A. (1889) Birdlife of the Borders. Gurney & beaks? Bird Study, 31, 230-231. Jackson, London. Pienkowski, M.W. (1973) Feeding activities of Cramp, S. & Simmons, K.E.L. (1983) The Birds of the wading birds and shelducks at Teesmouth and Western Palearctic, Vol. 3. Clarendon Press, some possible effects of further loss of habitat. Oxford. Rep. Coastal Ecology Research Station (The Nature Curlew bills 69

Conservancy). Department of Zoology, University ostralegus; a dynamic adaptation to specific forag- of Durham. ing techniques. Neth. J. Sea Res. 17,57-83. Pienkowski, M. W., Ferns, RN., Davidson, N.C. & Swennen, C. & Duiven, P. (1983) Characteristics of Worrall, D. H. (1984) Balancing the Budget. In oystercatchers killed by cold-stress in the Dutch Coastal Waders and Wildfowl in Winter (Eds P.R. Wadden Sea. Ardea, 71,155-159. Evans, J. D. Goss-Custard and W.G. Hale), pp. Tay & Orkney Ringing Groups (1984) The shore-birds 29-56. Cambridge University Press. of the Orkney Islands. Tay & Orkney Ringing Pounder, B., Pounder, S. & Greenwood, J.J.D. Groups, Perth. (1971) Hard times and the disadvantage of de- Townshend, D.J. (1981) The importance of field formity. Scott. Birds, 6, 278-279. feeding to the survival of wintering male and Prater, A.J., Marchant, J.H. & Vuorinen, J. (1967) female Curlews Numenius arquata on the Tees Guide to the Identification and Ageing of Holarctic estuary. In Feeding and Survival Strategies of Waders. British Trust for Ornithology, Tring. Estuarine Organisms (Eds N.V. Jones and W.J. Smith, P.C. (1975) A study of the winter feeding Wolff), pp. 261-273. Plenum Press, New York. ecology and behaviour of the Bar-tailed Godwit, Townshend, D.J. (1982) The use of intertidal habitats Limosa lapponica. PhD thesis, University of by shorebird populations, with special reference Durham. to Grey Plover (Pluvialis squatorola) and Curlew Smith, P.C. & Evans, P.R. (1973) Studies of shore- (Numenius arquata). PhD thesis', University of birds at Lindisfarne, Northumberland. 1. Feeding Durham. ecology and behaviour of the Bar-tailed Godwit. Townshend, D.J. (1985) Decisions for a lifetime: Wildfowl, 24, 136-139. establishment of spatial defence and movement Sokal, R.R. & Rohlf, F.J. (1981) Biometry. 2nd edn. patterns by juvenile grey plovers (Pluvialis W.H. Freeman, San Francisco. squatorola). J. Anim. Ecol. 54, 267-274. Speakman, J.R. (1984a) The energetics of foraging in van Heezik, Y.A., Gerritsen, A.F.C. & Swennen, C. wading birds (Charadrii). PhD thesis, University of (1983) The influence of chemoreception on the Stirling. foraging behaviour of two species of Sandpiper, Speakman, J.R. (1984b) Why do Redshank stop Calidris alba and Calidris alpina. Neth. J. Sea Res. foraging when it is windy? Wader Study Group 17 47-56. Bull. 42, 15. WishaA, R.A. & Sealy, S.G. (1980) Late summer Swennen, C., De Bruijn, L.L.M., Duiven, P., Leo- time oudget and feeding behavior of Marbled pold, M.F. & Marteijn, E.C.L. (1983) Differences Godwits ',Limosa fedoa) in southern Manitoba. in bill-form of the Oystercatcher Haematopus Can. J. Zool. 58,1277-1282.

(MS received 11 February 1985; revised MS received 16 November 1985) Downloaded by [University of Aberdeen] at 08:22 20 November 2015