The Condor89:602-609 0 The Cooper Ornithological Society 1987

EFFECT OF GROUP SIZE AND LOCATION WITHIN THE GROUP ON THE FORAGING BEHAVIOR OF WHITE IBISES

DANIEL R. PETIT Department of Zoology and UniversityMuseum, Universityof Arkansas,Fayetteville, AR 72701

KEITH L. BILDSTEIN Department ofBiology, Winthrop College,Rock Hill, SC 29733 and The Belle W. Baruch Institutefor Marine Biology and CoastalResearch, Universityof South Carolina, Columbia, SC 29208

Abstract. We studiedadult White Ibises (Eudocimus albus) on a South Carolina salt marshto determinethe effectsof socialgrouping on the birds’ foragingbehavior. White Ibises on our study site fed almost exclusivelyon fiddler crabs(Uca spp.). Four social categorieswere recognized: (1) centraladults in flocksof 2 15 birds,(2) peripheraladults in flocksof 2 15 birds,(3) adultsin small flocksof five or fewerbirds, and (4) solitaryadults (singletons).We useda pairedsampling scheme to comparethe behaviorof centralbirds with the behaviorof birds in the otherthree socialgroupings. Although peripheral adults did not differsignificantly from centraladults in numberof steps,number of crabscaptured, or number of captureattempts, they lookedup more often and for longerperiods of time than did centraladults. Behavior of solitaryibises was similar to that of ibisesin small flocks,but both forageddifferently than centraladults in largeflocks. Birds in the centerof largeflocks took fewersteps, probed more frequently,and scannedthe surroundingsless oftenthan birdsin the othertwo socialgroupings; there were no differencesin capturerates. Thus,White Ibisesused two distinct types of foraging strategiesdepending on flock size and their position within the flock. Ibises in small flocks, singletons,and, to some extent, ibises on the edgesof large flocks, stepped quickly to capture fiddler crabs before the crabs could retreat into burrows. Centrally-located ibises in large flocks were unable to use this foraging technique becausethe surroundingmembers of the flock created a disturbancethat caused the fiddler crabsto remain in their burrows. These birds, therefore, probed into crab burrows and found their prey by tactile means. Our resultssupport the predator-protectionadvantage of feeding within a flock independent of the feeding-efficiencyhypotheses. Key words: White Ibis: Eudocimusalbus; foraging; flock size; ciconitformes;vigilance; salt marsh:behavior,

INTRODUCTION ering spot, such as a roost, a bird might increase Optimal foraging theory predicts that animals the likelihood of foraging in an area containing should forage in such a way as to maximize the sufficient food resources. Also, an individual net rate of energy intake (for a review, seeKrebs might increase prey intake by copying the for- et al. 1983) becausethis will ultimately maximize aging techniques of successful flock members reproductive output. Within this context, several (Krebs et al. 1972), or simply by foraging at a hypotheses have been proposed to account for faster rate (Abramson 1979, Jennings and Evans why birds feed in closeproximity to one another. 1980). Members of flocks might experience in- The two most widely acceptedbenefits of group creased protection from predators presumably foraging are: (1) increasedprobability of locating becausethe increased number of eyes can detect or exploiting suitable food resources(Ward and an approachingpredator earlier than a single pair Zahavi 1973, Krebs 1974) and (2) decreasedvul- of eyes (Pulliam 1973, Elgar and Catterall 198 1) nerability to predators(Page and Whitacre 1975, and becauseearly detection increasesthe chance Kenward 1978, Caraco 1979). Either by watch- of escape. Thus, an individual can increase its ing where other individuals are feeding or by foraging efficiency because of decreased vigi- gaining information on food patches at a gath- lance. Obviously, these two hypothesized bene- fits of group foraging do not have to be mutually exclusive (cf. Abramson 1979). Studies simul- * Received 8 September 1986. Final acceptance 18 taneously testing the two hypothesized advan- February 1987. tages of flocking have produced conflicting re-

PO21 WHITE IBIS FORAGING BEHAVIOR 603 suits (e.g., Smith 1977, Kenward 1978, Jennings and 20,000 White Ibises breed on Pumpkinseed and Evans 1980, Fleischer 1983). For example, Island about 5 km from the marsh. Up to several some studies have shown that birds foraging in hundred White Ibises typically feed on the marsh flocks decreasedtheir time looking up while in- study area at any one time (Bildstein 1983). On creasing their foraging rate (Powell 1974, the marsh, ibises usually feed in groups of from Abramson 1979, Goldman 1980, Jennings and 2 to 75 individuals and almost exclusively on Evans 1980). In other investigations, however, fiddler crabs(Uca pugilator and U. pugnax, Hen- the decreasein time spent vigilant by flock mem- derson 1981). Although other studies have de- bers did not affect the foraging successas com- scribedWhite Ibises as tactile, nonvisual foragers pared to solitary individuals or birds in smaller (Kushlan 1978, 1979), on North Inlet marsh, groups (Smith 1977, Fleischer 1983). These dis- ibises used both tactile and visual methods for crepanciesmay reflect, at least in part, individual detecting prey. variation in foraging efficiency as a function of We used spotting scopesto observe the for- flock size and position within the flock as well aging behavior of ibises from an 18.5-m tower as speciesand prey type differences. located on the marsh edge. Data were collected On the other hand, increasing flock size may throughout the day, but most recordswere taken decreaseforaging successbecause of food deple- between 08:OO and 14:O0. Typically, ibises fed tion (Goss-Custard 1969, Smith 1975, Horwood on the higher elevations of the marsh where and Goss-Custard 1977), fighting over food Spartina was less than 30 cm high. These areas (Goss-Custard 1977a, Fleischer 1983) alteration of the marsh were usually covered by standing of search paths (Goss-Custard 1976), decreased water for only a few hours before and after high prey availability becauseof disturbance by other tide. We restricted our observations to periods flock members (Goss-Custard 1970, 1976) or when water had not inundated the high Spartina the increased likelihood that as flock size in- marsh because during these times fiddler crabs creases,some birds will be displaced into less return to their burrows (pers. observ.). preferred areas (Zwarts 1976, Goss-Custard To assessif foragingbehavior and capture rates 1977b). of adult (third year and older) ibises differed with In addition to potential differencesin foraging respectto flock size or position within the flock, efficiency based on the size of the flock, position we made sequential, 4-min, paired observations within the flock can affect an individual’s rate of of birds in four social situations. One observer food intake. Birds on peripheries of flocks have paired (within 5 min) the observation of an in- been shown to spend more time scanning the dividual in one social group (defined below) with surroundings as compared to other group mem- an observation of an individual in a different bers (e.g., Jennings and Evans 1980). Age (Bild- social group. We used this paired sampling stein 1983) and dominance status (Morse 1967, scheme to compare birds under similar condi- Russell 1978) also can influence where a bird tions, thereby minimizing the effectsof time of forageswithin a group, thereby affecting its for- day, weather, season, and prey abundance and aging behavior. availability. Most ibises, regardless of social In this study we examine the costsand benefits grouping, used the same areas of the marsh for of flock foraging by documenting differences in foraging. We limited our observations to adults the foraging behavior of White Ibises (Eudoci- because,on North Inlet marsh, the foraging be- mus albus) foraging in flocks of various sizesand havior of recently fledgedjuveniles and second- at different positions within the flocks. We then year birds (nonbreeders)differs from the foraging discussour results in light of current flock for- behavior of third-year and older birds (breeding aging theory. adults) (Henderson 198 1, Bildstein 1983). Young (< 2 years) ibises sometimes foragedwithin adult STUDY AREA AND METHODS flocks, but they rarely comprised > 10% of the From June to August 1983 and 1984, we watched flock (see also Bildstein 1983). We recognized White Ibises foraging on the North Inlet marsh four social categories:(1) central adults in flocks near Georgetown, South Carolina. Our 65ha of I 15 birds, (2) peripheral adults in flocks of study site is dominated by Spartina alterniflora, 115 birds, (3) adults in small flocks of five or and contains mud flats, oyster reefs, and tidal fewer birds, and (4) solitary adults (singletons). creeks (see Christy et al. 198 1). Between 10,000 Paired focal individuals were always > 100 m 604 DANIEL R. PETIT AND KEITH L. BILDSTEIN

m CENTRAL ADULTS 81 0 PERIPHERAL ADULTS m CENTRAL ADULTS IN FLOCKS OF b 15 IN FLOCKS OF L 15 (N=43) 0 ADULTS IN FLOCKS OF $ 5 (N.29)

FIGURE 1. Comparisonof foraging behavior of White Ibisesin the centerof large(2 15 birds) flocks FIGURE 2. Comparisonof the foragingbehavior of White Ibisesin the center of large (2 15 birds) flocks and on the nerinheriesof largeflocks. N representsthe number of pairedobservations and asterisksindicate and ibises in small (55 birds) flocks. N representsthe number of paired observations and asterisksindicate significant(P 5 0.05) differencesbetween the groups significant (P 5 0.05) differences between groups for for that variable.Bars equal + 1 SD. that variable. Bars equal + 1 SD. apart, except for central and peripheral individ- ing foraging bouts, we used the equations of Fe- uals within the same flock. Ibises were consid- dak et al. (1974) and Wiens and Dyer (1977) to ered central if they were part of the central 50% calculate the energetic costsincurred by ibises in of the aggregation, and peripheral if they were the different social categories.We estimated that not. Distinction between central and peripheral ibises travelled at 1 km/hr while foraging, and ibises was not possible for small flocks. All of that each step took 0.75 sec. These values were the birds observed were readily visible as they estimated by determining the distance travelled foragedin short Spartina. Each bird was watched by some focal individuals over a given time in- for 4 min, during which time we dictated into a terval and by recording the number of steps/ cassettetape recorder the number of steps, the second taken by ibises while actively foraging. number of probes into crab burrows or probes To formulate a mean time budget for each group, at surface crabs, the number of Uca captured, we multiplied the average number of steps/min- and the number of times and the total amount ute by 0.75 set to yield the number of seconds of time the bird looked up and scanned the sur- of each minute spent moving. For the remainder roundings. From the tapes we derived eight for- of the minute, the ibises were assumed to be aging variables: number of steps/minute, num- expending energy equal to the existence meta- ber of probes/minute, number of crabs captured/ bolic rate (Kendeigh et al. 1977). Our estimates minute, number of times the bird looked up/ of travelling speedand time budgets should pro- minute, number of seconds spent looking up/ vide some indication of the relative energetic minute, probe to capture ratio, and step to probe costsfor individuals in different socialgroupings. ratio. We also computed the number of steps/ This, in turn, may be useful in assessingcosts minute during the time actually spent foraging and benefits of foraging in flocks. However, be- (i.e., adjusted number of steps),which took into cause these are rough estimates, caution should account(i.e., subtracted)the amount oftime spent be used when interpreting the results of our en- looking up. The time individuals spent scanning ergy expenditure analysis. the surroundings was inversely proportional to To determine if an individual foraged simi- the time spent foraging because the ibises that larly irrespective of its position within a flock, we watched did not noticeably engage in other we followed seven White Ibises which were ini- behaviors (e.g., preening). tially either on the edge or in the center of one To compare estimatesof energy expendeddur- of the large flocks. We followed these birds for WHITE IBIS FORAGING BEHAVIOR 605

m CENTRAL ADULTS IN FLOCKS OF ,,S cant (0.10 > P > O.OS),central adults probed 0 SINGLETONS (N=25) I more often, had a higher probe to capture ratio, and took fewer steps between probes than did peripheral ibises. The number of steps,adjusted steps, and rate of prey capture were similar be- tween the two groups (Fig. 1).

CENTRAL ADULTS IN LARGE FLOCKS VS. ADULTS IN SMALL FLOCKS (N = 29) White Ibises in small flockslooked up more often and spent more time scanning the surroundings than did their counterparts in large groups.Birds in small flocks steppedmore often and took more adjusted steps/minute than did central ibises in larger aggregations.Ibises in small flocks took more steps between capture attempts than did birds in large flocks. Ibises in large flocks probed FIGURE 3. Comparison of foraging behavior of significantly more frequently, and probed more White Ibises in the center of large (2 15 birds) flocks often between captures compared to ibises in and ibises foragingalone. N representsthe number of paired observations and asterisksindicate significant small flocks. Capture rates were similar between (P 5 0.05) differencesbetween groups for that variable. birds in large flocksand birds in small flocks (Fig. Bars equal + 1 SD. 2). CENTRAL ADULTS IN LARGE FLOCKS up to 25 min and noted when a bird changed VS. SINGLETONS (N = 25) position within the flock. We compared the eight Singletons looked up more often and for longer foraging variables between the two data sets(i.e., periods of time than did central birds. Although center vs. periphery) as paired samples. number of steps was not statistically different, Data were analyzed with Wilcoxon’s matched- adjusted steps were significantly higher for sin- pairs signed-ranks test. All testswere two-tailed, gletons as compared to central adults in large and we considereda significant difference to exist flocks. Singletons also took more stepsbetween if P 5 0.05. probes, whereas central ibises probed more fre- RESULTS quently and had a lower successrate per probe than did singletons. Capture rates were similar CENTRAL VS. PERIPHERAL ADULTS IN between the two groups (Fig. 3). LARGE FLGCKS (N = 43) The only significant differences in foraging be- ENERGY EXPENDITURE OF IBISES havior were that peripheral ibiseslooked up more IN DIFFERENT GROUPS often and for longer periods of time than did Even though adults in small flocks, singletons, central birds. Although not statistically signifi- and, to some extent, peripheral adults, stepped

TABLE 1. Comparisons of prey intake, time budgets, and energy expended by White Ibises foraging within different social situations on the North Inlet marsh, Georgetown, South Carolina.

Energy expended % Time (kcalhin) Number % Difference Energy expended comparison preylmin m intake Walking Stationary Walking Stationary % d&rence

Central” 1.8 51% 49% 0.074 0.042 Peripheral 1.8 0.0% 52% 48% 0.077 0.040 +0.9% Central 1.6 39% 61% 0.056 0.052 Small flock 1.7 +6.2% 51% 49% 0.075 0.041 +7.4% Central 1.6 43% 57% 0.062 0.048 Singleton 1.9 + 18.7% 52% 48% 0.076 0.040 +5.4%

s See text for descriptions of groups. 606 DANIEL R. PETIT AND KEITH L. BILDSTEIN more often than ibises foraging in the centers of diture, the constant rate of Uca captured across large flocks, they did not expend significantly groups in our study should be viewed with cau- greater amounts of energy (Table 1). The largest tion. Our results, however, suggestthat there was increase(7.4%) was between birds in small flocks little difference in energy expenditure associated and central birds in large flocks, and it amounted with the different foraging techniques. to only 0.48 kcal/hr, or 4.9 kcal (5 to 6 Uca, J. W. Johnston, pers. comm.) per foraging day DISADVANTAGES ASSOCIATED WITH (=10.25 hr; cf. Kushlan 1977a). This amounts FLOCKING to only 3.0% of the daily energy expenditure of That centrally-locatedibises did not increasetheir 165 kcal/day for ibises breeding in a similar cli- capture rates despite devoting more time to for- mate (Kushlan 1977a). aging than did other ibises suggeststhat there are disadvantagesassociated with foraging among a COMPARISON OF INDIVIDUALS large number of birds. CHANGING POSITIONS Prey depletion (Zwarts 1980) appeared to be Results from observations on the seven individ- insignificant as we did not record a decreasein uals that changed position in a flock are consis- individuals’ capture rates over the 4-min obser- tent with the paired observations of central vs. vation period, nor over the time span that a flock peripheral adults in large flocks.Both the number was on the marsh (unpubl. data). These results of steps and adjusted steps were approximately may reflect the fact that ibis flocks moved over 40% higher at peripheral positions, while pe- the marsh and, therefore, may not have spent ripherally-located ibises had lower probing rates enough time in a location to appreciably deplete (central x = 11.3/min vs. peripheral K = 8.61 the local fiddler crab numbers. min) and probe to capture ratios (central K = 9.2 Another suggesteddisadvantage, fighting over vs. peripheral x = 5.4) than central birds. Cen- food (Goss-Custard 1977a, Silliman et al. 1977), trally-located ibises looked up less frequently also probably was not associatedwith the larger (central x = 0.8/min vs. peripheral R = 1.5/min) ibis flocks on our study site. Only during a brief and for less time (central x = 1.3 sec/min vs. 2-week period during late July 1984 did we ob- peripheral x = 3.4 sec/min). They also took fewer serve any interference or aggressionbetween for- steps between probes as compared to peripher- agingconspecifics, and even then, encounterswere ally-located ibises (central K = 2.6 steps/probe very infrequent ( < 0.5/hr). vs. peripheral K = 5.3 steps/probe).Capture rates Alteration of search paths and the prediction were similar between the two groups (central K = that as flock size increasesso doesthe probability 1.5 Uca/min vs. peripheral K = I .8 Uca/min). that some birds will be forced to forage in less- All the above comparisons, except capture rates, preferred areas (Zwarts 1976, Goss-Custard were statistically different (P I 0.05, Wilcoxon’s 1977b) were not specifically examined in this matched-pairs signed-ranks test). study. Although we do not believe that thesetwo potential disadvantages were important factors DISCUSSION here, one would have to quantify search paths ADVANTAGES ASSOCIATED WITH FLOCKING (e.g., Smith 1977) and know the dynamics of the Birds that foraged in the center of large flocks resource patches used by ibises to properly ad- spent less time vigilant and more time foraging dress these hypotheses. than other ibises. An increased amount of time Depressedavailability of surface crabs due to spent in search of food is presumed to provide activity of birds appears to be the major disad- an animal with a greater rate of energy intake vantage to White Ibises foraging in the center of (e.g., Murton 1971, Powell 1974, Caraco 1979); large flocks. Goss-Custard (1970) showed that this was not the case in our study. None of the Common Redshanks (Tringa totanus) walking comparisons demonstrated an advantage, in on the surfacecaused invertebrates to retreat into terms of prey intake, to ibises foraging in the burrows. We believe that this was the proximate center of large flocks, despite the greater pro- causeof the lower than expectedcapture rates of portion of time central birds spent searchingfor ibises located centrally within large flocks. We fiddler crabs. As different types of locomotion noticed crabs withdrawing to their burrows in (Tucker 1975) and activities (e.g., Schartz and responseto our presencewhen we walked on the Zimmerman 197 1, Mugaas and King 198 1) are marsh. This responseof fiddler crabs may have associatedwith varying degreesof energy expen- created the dichotomy in foraging behavior be- WHITE IBIS FORAGING BEHAVIOR 607 tween the central birds in large flocks and the hocks, more data are needed to confirm these other three groups. White Ibises foraging in the observations. center of large aggregationstended to be tactile White Ibises foraging in the center of large foragers(i.e., they probed into fiddler crab bur- flocks spent lesstime being vigilant compared to rows in search of prey). This contention is sup- other ibises, but they gained no benefits in terms ported by their increased number of probes, the of prey intake. By altering their foraging behav- fewer steps between capture attempts, and the ior, individuals in eachsituation managedto catch greater number of probes between captures. On prey at the same rate. Specifically, central ibises the other hand, ibises foraging alone, ibises on in large flocks spent more time foraging to offset the edgesof large flocks, and ibises in small flocks the depressedavailability of fiddler crabs on the tended to forage visually, i.e., by chasing crabs surface. On the other hand, ibises in the other on the surface and capturing them before they three situations were more vigilant, but were able retreated into burrows. Our data also support to capture crabs on the surface.Thus, White Ibis- this notion, as individuals in the latter three es’ foraging behavior changed in responseto the groups took more steps (frequently ran) and pattern of prey availability (behavior) (see also probed less than their centrally located counter- Kushlan 1978). In foraging flocks where mem- parts. Ibises on the peripheries of large flocks, bers are not feeding on highly mobile prey, such ibises in small flocks, and ibises foraging alone as fiddler crabs, the costs of flock foraging may were the first to reach undisturbed areas and, not be so great becausethe “prey” do not dras- thus, were able to locate and capture crabs on tically alter their behavior in responseto the high the surface. density of birds. Most studies that have docu- Our results do not support the hypothesesthat mented increased food intake of flock members by foraging within a flock an individual increases have dealt with birds feeding on grain, seeds,or prey intake by copying the foraging behavior of relatively immobile invertebrates during the successfulflock members or that the increased winter (e.g., Powell 1974, Abramson 1979, Car- protection gained by flock members allows them ace 1979, Goldman 1980, Jennings and Evans to procure more resources.Ibises feeding in the 1980). Alternatively, most studies which have center of large flocks, however, do gain an ad- not shown an increase in prey intake for flock vantage of increasedpredator protection without members were conducted on birds that were for- suffering lower feeding rates. Thus, our results aging on mobile prey, such as active insects, fish, support the predator-protection hypothesis (e.g., and crabs (e.g., Smith 1977, Pleischer 1983, this Pulliam 1973) independent of feeding-efficiency study). Thus, one proposedbenefit of foraging in hypotheses. a flock, increased prey intake, may vary from Why then do ibises sometimes forage on the speciesto speciesand according to the prey type edgeof large flocks,in small flocks, and by them- and its availability. Certainly, a single explana- selves?Some studies have suggestedthat social tion for the advantage of flock foraging cannot (Morse 1970, Russell 1978) or sexual (Peters and be extended acrossall taxa or even within a given Grubb 1983) dominance plays a role in deter- speciessubjected to different environmental in- mining where an individual forages.The shorter fluences. bill of females (Kushlan 1977b) might preclude them from reaching the bottom of some fiddler ACKNOWLEDGMENTS crab burrows, thus making them less efficient We thank P. C. Frederick, T. C. Grubb, Jr., C. Marsh, foragers in a situation where they must locate K. E. Petit, L. J. Petit, J. G. Robinson, K. G. Smith, and an anonymous reviewer for their helpful com- crabs by tactile means (i.e., in the center of large ments on the manuscript. Our studiesof White Ibises flocks). Therefore, birds on the edges of large have been supported by an NSF-LTER grant to the flocks,in small flocks,or foraging solitarily (those Baruch Institute. This is contribution No. 675 of the birds that hunt prey visually and capture prey Belle W. Baruch Institute for Marine Biology and from the surface), may have been females. Our Coastal Research,University of South Carolina. results, however, suggestthat this is not the case, as the seven focal ibises generally changed their LITERATURE CITED foraging behavior when they changed positions ABRAMSON,M. 1979. Vigilance as a factor influencing within the large flocks. Although our preliminary flock formation among curlews Numenius ar- results suggestthat sexual and social factors do quata. Ibis 121:213-216. not affect where adult White Ibises forage within BILDSTEIN, K. L. 1983. Age-related differencesin the 608 DANIEL R. PETIT AND KEITH L. BILDSTEIN

flocking and foraging behavior of White Ibises in KREBS, J. R., M. H. MACROBERTS,AND J. M. CULLEN. a South Carolina salt marsh. Colonial Waterbirds 1972. Flocking and feeding in the Great Tit (Pur- 6:45-53. us major): an experimental study. Ibis 114:507- CARACO,T. 1979. Time budgetingand group size: a 530. test of theory. Ecology 60:6 18-627. KREBS, J. R., D. W. STEPHENS,AND W. J. SUTHERLAND. CHRISTY, R. L., K. L. BILDSTEIN,AND P. DECOURSEY. 1983. Perspectivesin optimal foraging, p. 165- 1981. A preliminary analysisof energy flow in a 21 6. In A. H. Brush and G. A. Clark [eds.], Per- South Carolina salt marsh: wading birds. Colonial spectivesin ornithology. Cambridge Univ. Press, Waterbirds 4:96-103. Cambridge. ELGAR,M. A., AND C. P. CATTERALL.198 1. Flocking KUSHLAN,J. A. 1977a. Population energeticsof the and predator surveillancein House Sparrows:test American White Ibis. Auk 94: 114-l 22. of an hypothesis.Anim. Behav. 291868-872. KUSHLAN,J. A. 1977b. Sexual dimorphism in the FEDAK,M. A., B. PINSHOW,AND K. SCHMIDT-NIELSON. White Ibis. Wilson Bull. 89:92-98. 1974. Energy cost of bipedal running. Am. J. KUSHLAN,J. A. 1978. Feedingecology ofwading birds, Physiol. 227:1038-1044. p. 249-296. In A. Sprunt, J. C. Ogden, and S. A. FLEISCHER,R. C. 1983. Relationships between tidal Winkler [eds.], Wading birds. Natl. Audubon Sot. oscillations and Ruddy Tumstone flocking, for- Res. Rep. No. 7, New York. aging, and vigilance behavior. Condor 8522-29. KUSHLAN,J.-A. 1979. Feeding ecology and prey se- GOLDMAN,P. 1980. Flocking as a possible predator lection in the White Ibis. Condor 81:376-389. defensein Dark-eyed Juncos.Wilson Bull. 92:88- MORSE,D. H. 1967. Foragingrelationships of Brown- 95. headed Nuthatches and Pine Warblers. Ecology GOSS-CUSTARD,J. D. 1969. The winter feeding ecol- 48:94-103. ogyofthe Redshank Tringu totanus.Ibis 111:338- MORSE,D. H. 1970. Ecologicalaspects of somemixed- 356. speciesforaging flocks of birds. Ecol. Monogr. 40: GOSS-CUSTARD,J. D. 1970. The responsesof Red- 119-168. shank(Tringu totanus[L.]) to spatial variations in MUGAAS,J. N., ANDJ. R. KING. 1981. Annual vari- the density of their prey. J. Anim. Ecol. 39:91- ation of daily energy expenditure by the Black- 113. billed Magpie: a study of thermal and behavioral GOSS-CUSTARD,J. D. 1976. Variation in the disper- energetics.Stud. Avian Biol. 5. sion of Redshank Tringu totanus,on their winter MURTON, R. K. 1971. Why do some birds feed in feeding grounds.Ibis 118:257-263. flocks?Ibis 113:534-536. GOSS-CUSTARD,J. D. 1977a. The ecologyofthe wash. PAGE,G., AND D. F. WHITACRE. 1975. Raptor pre- III. Density-related behaviour and the possibleef- dation on wintering shorebirds.Condor 77:73-83. fects of a loss of feeding groundson wading birds PETERS,W. D., AND T. C. GRUBB,JR. 1983. An ex- (Churadrii). J. Annlied Ecol. 14: 721-739. perimental analysisof sex-specificforaging in the GO&-CUSTARD,J. D.- 1977b. Predator responsesand Downy Woodpecker, Picoidespubescens. Ecology prey mortality in Redshanks, Tringa totanus,and 64:1437-1443. a preferred prey, Corophium volutntor.J. Anim. POWELL,G. V. N. 1974. Experimental analysisof the Ecol. 46:21-35. social value of flocking by starlingsin relation to HENDERSON,E. G. 1981. Behavioral ecology of the nredationand foraeina.Anim. Behav. 22:501-5 15. searchingbehavior of the White Ibis (Eudocimus PUL~IAM,H. R. 1973.-O; the advantagesof flocking. albus).M.Sc.thesis, Univ. South Carolina, Colum- J. Theor. Biol. 38:419-422. bia. RUSSELL,J. K. 1978. Effects of interspecific domi- HORWOOD,J., AND J. D. GOSS-CUSTARD.1977. Pre- nance among egrets commensally following Ro- dation by the oystercatcherHuematopus ostrule- seateSpoonbills. Auk 95:608-610. gus (L.), in relation to the cockle, Cerustodermu SCHARTZ,R. L., AND J. L. ZIMMERMAN. 1971. The edule(L.), fishery in the Burry Inlet, South Wales. time and energy budget of the male Dickcissel J. Applied Ecol..14: 139-158: (Spiza americana). Condor 73165-76. JENNINGS.T.. AND S. M. EVANS. 1980. Influence of SILLIMAN,J., G. S. MILLS,AND S. ALDEN. 1977. Effect position in the flock and flock size on vigilance in of flock size on foragingactivity in wintering Sand- the starling, Sturnus vulgaris.Anim. Behav. 28: erlings. Wilson Bull. 89:434-438. 634-635. SMITH, J. N. M. 1977. Feeding rates, search paths, KENDEIGH,S. C., V. R. DOL’NIK, AND V. M. GAVRILOV. and surveillance for predators in Great-tailed 1977. Avian energetics, p. 129-204. In J. Pi- Grackle flocks. Can. J. Zool. 55:891-898. nowski and S. C. Kendeigh [eds.], Granivorous SMITH, P. S. 1975. A study of the winter feedina birds in ecosystems. Cambridge Univ. Press, ecology and behaviour of the Bar-tailed Godwij Cambridge. (Limosa luponicu). Ph.D.diss., Univ. Durham, KENWARD,R. E. 1978. Hawks and doves: factors Durham, United Kingdom. affectingsuccess and selectionin Goshawk attacks TUCKER,V. R. 1975. The energetic cost of moving on Woodpigeons.J. Anim. Ecol. 47~449-460. about. Am. Sci. 63:413-419. KREBS. J. R. 1974. Colonial nestingand socialfeeding WARD, P., AND A. ZAHAVI. 1973. The importance of strategiesfor exploitation of foid resourcesin the certain assemblages of birds as “information Great Blue Heron (Ardeuherodius). Behaviour 5 1: centres” for food finding. Ibis 115:517-534. 99-131. WIENS,J. A., AND M. I. DYER. 1977. Assessingthe WHITE IBIS FORAGING BEHAVIOR 609

potential impact of granivorous birds in ecosys- ZWARTS, L. 1980. Intra- and inter-specific competi- terns, p. 205-266. In J. Pinowski and S. C. Ken- tion for space in estuarine bird species in a one- deigh [eds.], Granivorous birds in ecosystems. prey situation. Proc. XVII Int. Omithol. Congr. Cambridge Univ. Press, Cambridge. (1978):1045-1050. ZWARTS, L. 1976. Density-related processes in feed- ing dispersion and feeding activity of teal (Anus crecca). Ardea 64: 192-209.

Lovebirds, Cockatiels, Budgerigars: BEHAVIOR AND EVOLUTION by J. Lee Kavanau University of California at Los Angeles Guided by unique, exhaustive laboratory studies of psittacine behavior, the author has deduced an insightful and remarkably comprehensive scenario of many aspects of the origin and evolution of birds and the roots of their behaviors....must reading for all interested in avian behavior and reproduction ...... a wealth of information and concepts that will be of interest to all evolutionists. Professor E.G. Olson

Feb., 1987 1,005 PP., 7 indices, 1,455 refs. $69

l A broad analysis and synthesis of avian evolution guided by new- ly discovered, relict reproductive behavior of small parrots that yields insights into ancestral habits of nesting and egg care.

l Courtship, pair bonding, coitus, nesting, egg recovery, incuba- tion, allofeeding and brooding of young, and familial interactions.

l Laboratory care, general biology, morphology, reproductive phys- iology, endocrinology, neurology, pathology, developmental biolo- gy, ecology, biogeography, paleological correlates, and compari- sons with reptiles and other vertebrates. Prepay only, check or money order. Add U.S. $4 for postage and, handling within, or $6 outside, continental U.S. In California add $4.50 for sales tax. Mail orders only. Science Software Systems, Inc., Box 241710, Los Angeles, CA 90024