Foraging Behavior and Foods of the Light-Footed Clapper Rail ’

SHORT COMMUNICATIONS 959

rarely feeds the female at the nest (pers. observ.). Per- I thank C. D. Piggott and P. Orriss for permission hapsif both sexesincubated or the male fed the female to study blackbirds in the University Botanic Garden, on the nest there would be less selective advantageto and J.N.M. Smith and an anonymous reviewer for having thermally tolerant embryos; conversely, intol- comments on the manuscript. The study was funded erant embryos may necessitatebiparental incubation by an Australia-U.K. Commonwealth Scholarship. or male incubation feeding.

The Condor90~959-962 0 The Cooper Ornithological Society 1988

FORAGING BEHAVIOR AND FOODS OF THE LIGHT-FOOTED CLAPPER RAIL ’

RICHARD ZEMBAL AND JACK M. FANCHER U.S. Fish and Wildlife Service,24000 Avila Road, Laguna Niguel, CA 92656

Key words: Clapper Rail; foods; foraging; Rallus of the reserve is about 304 ha, including a 11l-ha salt longirostris levipes; salt marsh. marsh(California Department of Fish and Game 1984). Descriptions of foraging behavior, presented here, Severalreports are available on the foods and foraging summarizeapproximately 180 hr of visual contactwith of subspeciesof Clapper Rails, Rallus longirostris(Bent feeding Clapper Rails during the period March 1979- 1926, Orr 1939, Moffitt 1941, Sibley 1955, Ohmart August 1987. In addition, several hours of foraging and Tomlinson 1977, Heard 1982) but nothing has were captured on video tape, mostly in January and been publishedfor the Light-footed Clapper Rail (Ral- February 1983. The tape of 11 foraging bouts by four iuslongirostris levipes). Because, the Light-footed Clap- rails, totaling 46.3 min of feeding, was detailed enough per Rail is considered endangered, such information to allow excellentmonitoring ofall their activities. The could be critical for effective management of the hab- motions and foraging success(swallowings) of these itat and this bird’s recovery (U.S. Fish and Wildlife birds were quantified: Service 1985). A total of 49.85 g of materials regurgitatedbv Lieht- The Light-footed Clapper Rail occursin California’s footed Clapper Ra& was collectedm the springbf 1979 coastal marshes from Santa Barbara County south to and of 1986. Of the total, 77% was collected at Upper the Mexican border and in northwestern Baja Califor- Newport Bay, 15.2% from the Seal Beach National nia, Mexico(Bent 1926, AOU 1957). Annualcensusing Wildlife Refuge in Orange County, 4% from the Ken- of the rails in California since 1980 revealed a high dall-Frost Reserve in Mission Bay, San Diego County, count of 277 pairs distributed in 19 marshes in 1984 and 3.8% from the Tijuana Estuary National Wildlife (Zembal and Massev 1981. 1985a).followed bv a crash Refuge on the Mexican border. The pellets and frag- to 142 pairs in 14 marshes in 1985. As of the spring ments were separated under a dissecting microscope of 1987, the state population had only poorly re- and recognizableremains were identified. covered. We have been investigating the habits of this en- RESULTS AND DISCUSSION dangered race since 1979 (see Zembal and Massey Over 90% of the time that Clapper Rails were observed 1983a. 1983b. 1985b. 1987: Massev et al. 1984: Zem- foraging,they hunted in the marsh vegetation, execut- bal et al. 1985; Massey and’Zemba1 1987). Herein are ing numeroussurface gleans and usuallyshallow probes. presented our observations on the known foods and They appeared to be hunting by sight and erratically foraging strategies. changedtheir direction of travel in responseto cues, presumablymovements by potential prey items. Con- STUDY AREA AND METHODS sequently,this generalforaging behavior involved a lot ForagingClapper Rails were observed at Upper New- of abrupt turns and direction reversals,and the foraging port Bay, an ecologicalreserve of the California De- birds often covered only very small areas. partment of Fish and Game in Orange County. Upper The movements involved in 11 bouts of foraging Newport Bay is a largelyunmodified embayment, sub- revealed rates of foraging that varied nearly eightfold ject to unconstrainedtidal influence (tidal range of 2.6 (Table 1). Higher intensity foraging was regularly ob- m) and winter freshwater storm flows. The total area served during peak foraging hours in the late evening. Rates of l,OOO-2,000gleans and probes per hour appeared to be commonly sustained.Lower frequencyof feeding motions with more travel by a rail was typical ’ Received 18 April 1988. Final acceptance7 July of “crabbing,” or the exploration of burrows, partic- 1988. ularly alongcreek banks, for crabsand other largerprey 960 SHORT COMMUNICATIONS

TABLE 1. Foraging movements and successof four Light-footed Clapper Rails.

Surfacegleans Success’ Maximum Duratmn Direction distance (set) n /mm n % %&? reversals moved lm)

121 439 67 33.2 28 41.8 3 0 0.2 238 - 17 4.3 9 52.9 148 6 8 314 - 55 10.5 18 32.7 113 5 10 433 439 76 10.5 32 42.1 177 8 10 280 439 38 8.1 15 39.5 162 9 10 414 440 114 16.5 43 37.7 121 5 48 439 17 21.3 6 35.3 14 ; 1 134 440 14 6.3 4 28.6 70 4 2 266 440 7.4 6 18.2 103 6 5 240 439 :: 7.5 12 40.0 105 6 5 287 439 60 12.6 16 26.7 43 4 3 ’ A successfulforaging motion was followed by an obvious swallowingby the rail.

(see below). The average rate of foraging for the 11 of the crabs and other foods taken by adults was fed bouts was about 675 gleans/hr. to the youngsters.However, even very young rails for- The overall success(number of swallows) for all 11 aged somewhat; snapping up dipterans seemed to be bouts extrapolates to approximately 250 morsels of one of their first pursuits. tiny prey in an hour. The lower successrates were Chicks sometimeslurked along the edge of the vege- associatedwith crabbingor casuallyforaging rails, and tation until an adult found food but were observed as the higher rates were by birds that found food concen- often in the close company of a foraging adult. Young trations. Consequently, an average percent success birds, as old as about 6 weeks, were fed in part by varying from the high 20s to the low 40s appeared to adults. The behavior of one youngsterof about this age be common for moderatelyto intenselyactive foragers. and an adult indicates how some foraging strategies Rails engagedin standard foraging behavior, glean- may be learned. The adult hunted along with the im- ing and shallowlyprobing the marsh, often moved from mature at its heels. When the adult froze with its head one bonanzato the next. As illustratedby the first entry slightly above the substrate, ready to strike, the im- in Table 1, concentrations of food were often quite mature raced in and snappedup a spider from the spot near one another, requiring little travel by a rail. Con- concentratedupon by the adult. centrationswere often discoveredat the basesof shrubs One rail was observed taking two different longjaw or under small bits of cover that the rail would toss mudsuckers(Gillichthys mirabilis) from the mudflat. aside. These concentrations were worked vigorously The fish were too bulky to swallow at first, so they were by a bird and appeared to be caches of small, very jabbed, shaken,and tossedabout until finally massaged mobile organisms that scattered quickly when their to a shape that could be swallowed whole. cover was thrown or nudged away. Three differentrails were observedeating ribbed horse Clapper Rails observedforaging for periods of l-3.5 mussels (Ischadium demissum)on the mudflat. One hr would almost invariably crab for a few minutes bird pulled flesh from an open but intact mussel and during that time. After a rail had swallowed hundreds swallowed six times before moving on. Two others of tiny prey items, too small for us to identify from walked along a dense algal mat over shallow water just 10 m away, it would walk into a tidal creek and gleaningand probing. Both encounteredmussels; one hunt the bank. A bird would searchburrows and probe worked on three of them. The musselswere not visible several of them, often lunging into one neck-deep,and initially on the surface; a rail, after probing around, within minutes one or more crabs(Puchygrapsus crus- would suddenlypull one up with vigoroushead shaking sipes, Hemigrapsus oregonensis,and Uca crenulata) and the bird would swallow something. In the process, would be eaten. Smaller crabs were swallowed whole; the mussel was cast to the side, it releaseda squirt of medium-sized crabswere dismemberedby shakingand water, and closed. The rail would return to the mussel, the parts swallowed. Larger crabs were dismembered, probe it for a short time, and then move on. This the appendagesswallowed, but the body was broken behavior was repeated on all three mussels,not one of open and only the flesh consumed. which was successfullyreopened and entirely con- Other lessfrequently observedforaging strategies in- sumed. cluded mudflat foraging, fishing, and scavenging.The The abundance of this nonnative mussel in San Clapper Rails we observed were mostly within marsh FranciscoBay was considereda potential threat to Cal- vegetation or along its edge. The rails were warier and ifornia Clapper Rails (R. 1. obsoletus)following obser- moved more quickly when away from cover but did vations of rails stuckin closedmussels (DeGroot 1927). venture onto the mudflats regularly. This was most We rescuedan emaciated and easily capturedrail with apparent in the summer and fall, when family groups a musselclamped tightly to one toe in Upper Newport were observed on the mudflats and algal mats in the Bay. We have also observed many rails with partly evenings. Crabbing was commonly observed of rails missing toes. Moffitt (194 1), in contrast, analyzed the on the flats along the central drain creeks. A portion contents of 18 stomachs of California Clapper Rails SHORT COMMUNICATIONS 961

and found the ribbed horse mussel to be the single Two of the pellets collected at the ditch contained a most abundant food item. He specifically mentioned total of 75 seeds,representing as many as 25 fruits of mussel shell fragments and apparently deduced that the elderberry (Sumbucusmexicana). A small number the flesh had been consumedas well. We have no evi- of elderberry trees overhang the ditch. This latter ob- dence, so far, that the Light-footed Clapper Rail is servation was unusual.Plant remains were rare in the capable of breaking the thick valves of this mussel. pelletsand other than the elderberry,totalled three very Perhaps small ones are swallowed whole. small unidentified seedsand several cordgrass(Spar- Clapper Rails were observed fishing seven times. tina foliosa) seeds.Only three Clapper Rails were ever They paddled around vigorously, partially submerged observedwith certainty feeding on plants. Two, while in pursuit, and occasionallydove. One rail took four incubatingon upper marsh nests, pulled off the tips of California killifish (Fun&/us parvipinnis) in 17 min pickleweed(Sulicorniu virginica) stems and swallowed from a small pond. Another took six fish from the same them. Another moved from one broken stem of cord- pond in 30 min. A rail was also observed once wading grassto the next, extracting and swallowing pith from through a very small pond feeding on tadpoles of the the stems. Pacifictree frog(Hylu regilu).The bird took 19 tadpoles Clapper Rails may eat a wide variety of foods that in 10 min. Fishing may be more common than our we have been unable to identify specifically.Most of observations would indicate. Small pools left by re- a rail’s foraging is done in dense vegetation, often in ceding tides often strand abundant and easy prey but cordgrass,and usually out of view. When a bird was most such locations were screenedfrom our view by finally in view, most of what was consumed during dense vegetation. many hours of observationfrom just a few meters away Scavengingwas observed three times and involved was too small to identify. These tiny items are probably rails feedina on dead mullet (Munil cephaks).One of soft-bodied and totally digested,so that pellet analysis the birds f&l for 40 min on the same-carcass.These does not aid in the identification of what may be nu- birds may have been eating insect larvae as well as merous species.Castings do occasionallyyield an ad- decayingfish flesh. ditional food item, however, that happensto be partly The materials regurgitatedby Clapper Rails and col- undigestible. In addition to the items already men- lected in 1979 included 18 intact pellets. The average tioned, other known foods included amphipods, gar- size and weight of these were 1.30 cm x 1.05 cm and den snails (Helix sp.), crane flies (Tipulidae), grass- 0.52 g. Thus, all of the regurgitatedmaterials we col- hoppers or crickets (orthopterans, from foraging on lected representedabout 96 average-sizedcastings by upland slopes),and perhapsmice (Microtus culifornica weight. and Mus musculus).We have seen mice being carried The composition of the materials collected in 1986 by several rails, mostly during high tides. Mice have was very similar to that of the 1979 samples.The 1979 also been found packedinto nestingmaterials. A small materials, representing 75.7% of the total, were col- enough mouse would most certainly be swallowed lected in 24 samples,mostly from different nests. Fre- whole. quency of occurrencewas determined for the recog- In summary, crabs are certainly important in the nizable components per sample whereas the 1986 diet. Birds observed foraging take them regularly and materials were pooled by marsh. Crab remains were they are an extremely largemeal comparedto the usual the most abundantlyrepresented item; they were found items seen swallowed. Snails may also be important; in 19 of 24 samples.One pellet contained evidence of they are relatively large, abundant in the marsh, and at least five crabs. Pachygrupsuscrassipes and Hemi- well-representedin pellet samples, although it is un- grapsusoregonensis were the two more common species. usualto observeone being eaten. There are also several Although the sizesof dactylsfrom crab chelipedsfound very important dietary items that have not been iden- in the pellets rangedin length from 2.5 mm to 12 mm, tified specifically.They are quite small, some are quite most were 4 mm to 8 mm long, from mostly small to mobile as well, and they are eaten by the hundreds medium-sized crabs. during a long foraging bout. These tiny morsels are The next most abundant remains were of California mostly surface-gleaned,often from the basesof plants horn snails (Cerithideucalzfirnica), found in 10 of 24 and regularlyfrom under dried piecesof algaeor debris samples,and salt marsh snails (Melampus olivaceous), that the rail flies out of the way. Whether the small found in seven samples. One pellet contained the re- invertebratesof the salt marsh are consumed in direct mains of 2 1 horn snails.Another held evidence of eight proportion to their relative abundancesis unknown. Melumpus.A few of the larger individuals of both snail Heard (1982) examined the contents of 187 stom- specieswere intact and undigested. achs of five subspeciesof Clapper Rails from the east Other organismsrepresented in the samplesincluded and gulf coastsof the United States.He concludedthat crayfish (Procumburussp.) (in one sample), beetles (in the subspecieshe studied ate basically the same kinds one sample), isopods (in two samples), and decapods of foods: crabs; snails; and to a lesser extent, insects; (in one sample). polychaetes;bivalves; fishes; and limited plant mate- About 22% of the 1986 sampleswere collectedalong rials. The contents of 16 stomachs of two other sub- a freshwaterditch at the upper edgeof Upper Newport speciesfrom westernMexico were analyzedby Ohmart Bay. These materials,as in 1979, were composedmost- and Tomlinson (1977); the major recognizablefoods ly of the fragmented exoskeletonsof crayfish. There were crabs,snails, and insects.These latter authorsalso were also 2 1 tiny cup-shapeditems of unknown source examined the contents of 16 stomachs of the Yuma but high calcium carbonatecontent (perhapsthe white Clapper Rail (R. 1. yumunensis)and found crayfish, stones mentioned by Ohmart and Tomlinson 1977) isopods, insects, and mollusks to be most abundant. and pieces of beetle elytra, as in 1979. These findingsand Heard’s (19 82) view of the Clapper 962 SHORT COMMUNICATIONS

Rail as an opportunistic omnivore that occupiesa rel- MASSEY,B. W., R. ZEMBAL,AND P. D. JORGENSEN. atively broad feeding niche within marsh ecosystems, 1984. Nesting habitat ofthe Light-footed Clapper mesh well with our observations. It is doubtful that Rail in Southern California. J. Field Omithol. 55: food is a limiting factor for the Light-footed Clapper 67-80. Rail in marshesthat have not been badly degraded. MOFRTT,J. 1941. Notes on the food of the California Clapper Rail. Condor 43:270-273. BarbaraW. Massey provided observationsthat con- OHMART.R. D.. ANDR. E. TOMLINSON.1977. Foods tributed substantiallyto our understandingof Clapper of westem’clapper Rails. Wilson Bull. 89:332- Rail foraging and made suggestionsthat greatly im- 336. proved this paper. We thank the California Depart- ORR, R. T. 1939. Fall wanderings of the Clapper ment of Fish and Game for storagespace and access Rails. Condor 4 1:15 l-l 52. at Upper Newport Bay and the U.S. Fish and Wildlife SIBLEY,C. G. 1955. The responseof salt marsh birds Service and U.S. Navy for accessto the Seal Beach to extremely high tides. Condor 57:241-242. National Wildlife Refuge. C. T. Collins and J. R. Gus- U.S. FISH AND WILDLIFESERVICE. 1985. Recoverv tafson are acknowledgedfor continued support. P. D. plan for the Light-footed Clapper Rail. U.S. Fish Jorgensenand M. Pruett-Jones collected some of the and Wildlife Service. Portland. OR. pellets for analysis.Our work was partially supported ZEMBAL,R., J. M. FANCIER,C. S. NORD~Y,AND R. J. by statetax check-off funds for researchon endangered BRANSF~ELD.1985. Intermarsh movements by and threatened wildlife, made available by the Cali- Light-footed Clapuer Rails indicated in Dart fornia Department of Fish and Game through a con- through regular censusing.Calif. Fish and Game tract with California State University, Long Beach. 71:164-171. LITERATURE CITED ZEMBAL,R., AND B. W. MASSEY. 1981. A censusof the Light-footed Clapper Rail in California. West. AMERICANORNITHOLOGISTS UNION.’ 1957. Check- Birds 12:87-99. list of North American birds. 5th ed. Port Citv ZEMBAL,R., AND B. W. MASSEY. 1983a. To catch a Press, Baltimore, MD. Clapper Rail-twice. N. Am. Bird-Bander 8:144- BENT,A. C. 1926. Life histories of North American 148. marsh birds. U.S. Natl. Mus. Bull. 135. ZEMBAL,R., AND B. W. MASSEY. 1983b. The Light- CALIFORNIADEPARTMENT OF FISH AND GAME. 1984. footed Clapper Rail: distribution, nesting strate- Upper Newport Bay managementplan. Calif. Dept. gies, and management. Cal-Neva Wildl: Trans. Fish and Game, Long Beach, CA. 1983:97-103. DEGROOT,D. S. 1927. The California Clapper Rail, ZEMBAL,R., ANDB. W. MASSEY. 1985a. Distribution its nesting habits, enemies and habitat. Condor of the Light-footed Clapper Rail in California, 291259-270. 1980-1984. Am. Birds 39:135-137. HEARD,R. W. 1982. Observationson the foods and ZEMBAL,R., ANDB. W. MASSEY. 1985b. Function of food habits of Clapper Rails from tidal marshes a rail “mystery” call. Auk 102:179-I 80. along the east and gulf coastsof the United States. ZEMBAL,R., AND B. W. MASSEY. 1987. Seasonality GulfRes. Rep. 7:125-135. of vocalizations by Light-footed Clapper Rails. J. MASSEY,B. W., ANDR. ZEMBAL. 1987. Vocalizations Field Ornithol. 58:41-48. ofthe Light-footed Clapper Rail. J. Field Omithol. 58:32-40.

The Condor90:962-964 0 The Cooper Ornithological Society 1988

EGGSHELL DIFFERENCES BETWEEN PARASITIC AND NONPARASITIC ICTERIDAE ’

H. RAHN,LINDACIJRRAN-EVERETT,ANDD.T. BROTH Department of Physiology,State Universityof New York at Buflalo, Buffalo, NY 14214

Key words: Eggshellthickness: eggshell mass; para- thicknessfor both parasiticand nonparasiticspecies of sitic icterids;Icteridae. the family Icteridae are largely lacking and present measurementsof several parasitic Molothrus species In their recent report Spaw and Rohwer (1987) state which showedsignificantly greater shell thicknesscom- that good comparative data on the relative eggshell pared with 17 nonparasitic representatives.These au- thors suggestthat the thicker eggshellmay be an ad- aptation to parasitism, making eggsmore resistant to I Received 22 April 1988. Final acceptance20 July damage by the host species.In this report we call at- 1988. tention to and review additional data from the tables