January1984] ShortCommunications 175

new buffer systems,or better staining regimes may formed in the Cornell Laboratoryfor Ecologicaland reveal the expression of enzymes not seen in our Evolutionary Genetics. study. Rather, the total score for each tissue over all the taxa we examined is more informative. There- LITERATURE CITED fore, a quantitative comparisonwas made by assign- ARISE, J. C., J. c. PATTON, & C. F. AQUADRO. !980. ing a ! to all tissueswith faint expressionand a 2 to Evolutionary geneticsof birds: comparativemo- those with good expressionand then totalling these lecular evolution in New World warblers and numbers for all enzymes in each tissue(see bottom rodents. J. Hered. 7!: 303-3!0. of Table !). Again, feather pulp was more informa- tive than blood but less informative than internal BAKER,M. C. !981. A musclebiopsy procedurefor use in electrophoreticstudies of birds. Auk 98: organs. 392-393. There are only two serious limitations to using BARROWCLOUGH,G. F., & K. W. CORBIN. !978. Ge- feather pulp for electrophoreticanalysis. First, the netic variation and differentiation in the Paru- birds must either be molting when sampled or be lidae. Auk 95: 691-702. subjectedto plucking to stimulate feather regrowth HARRIS, I-I., & D. A. HOPKINSON. 1976. Handbook of and then recapturedat a later date. Second,in small enzyme electrophoresisin human genetics.New birds such as Bobolinks, warblers (Parulidae), or York, American Elsevier. chickadees(Parus spp.), only tiny amounts of tissue MAY, g., J. E. WRIGHT, & M. STONEKING. !979. Joint can be obtained from the flight and tail feathers,ne- segregation of biochemical loci in Salmonidae: cessitatingthe useof two or more feathers.Mengden resultsfrom experimentswith Salvelinusand re- and Stock (!976) point out, however, that small quan- view of the literature on other species.J. Fish. tities of tissuemay be put into cell culture to increase Res. Board Canada 36: !!14-!126. the amount available; such tissue cultures may also MINGDEN,G. A., & D. STOCK.!976. A preliminary be stored for indefinite periods of time. Alternative- report on the application of current cytological ly, other forms of electrophoresisthat require smaller techniquesto sexing birds. Intern. Zool. Year- quantitiesof tissue(e.g. celluloseacetate) may be per- book !6: 136-14!. formed when only a few loci need to be examined. RYTTMAN,I-I., & I-I. TELGELSTROM.!981. Low degree Especiallyuseful to field workers, the tissueis con- of isozyme variation within and between Her- veniently "pre-packaged,"and may be placed, whole ring Gull (Larusargentatus), Lesser Black-backed or with the upper shaft removed, directly onto dry Gull (Larusfuscus) and their Britishand Swedish ice or into a portable liquid nitrogen flask. In terms subspecies.Hereditas 94: !61-!64. of easeof samplecollection and minimizationof stress SHERMAN,I•. W. !98!. Electrophoresisand avian ge- to birds, we suggestthat feather pulp is a viable al- neological analysis. Auk 98: 419-422. ternative tissuefor non-destructivesampling in avi- S•BLE¾,C.G. !96!. The electrophoreticpatterns of an electrophoreticstudies. avian egg-white proteins as taxonomic charac- We are grateful to Steven Bloom,Irene Brown, Mike ters. Ibis 102: 2!5-284. Denison, Tom Gavin, JoshHamilton, and Jerry Wald~ SMITH, J. K., & E.G. ZIMMERMAN. !976. Biochemical vogel for their help in obtaining the birds. Scott geneticsand evolution of North American black- Camazine and Paul Sherman made many helpful birds,family Icteridae.Comp. Biochem. Physiol. commentson earlier drafts of the manuscript.Fund- 53B: 319-324. ing was provided in part by a Sigma Xi Grant-in-Aid of Researchto J. E. Marsden. This work was per- Received14 February1983, accepted25 July 1983.

Food-nicheRelationships Between Great Horned Owls and Common Barn-Owls in Eastern Washington

RICHARDL. KNIGHTl AND RONALDE. JACKMAN2 •Departmentof WildlifeEcology, University of Wisconsin,Madison, Wisconsin 53706 USA, and 2681749th NE, Seattle,Washington 98115 USA

GreatHorned Owls (Bubovirginianus) and Common ships from our analysisof 622 Common Barn-Owl Barn-Owls(Tyto alba)have recentlybecome sympat- pellets,from 6 nestingand 2 roostingsites, and 234 ric in the PacificNorthwest (Stewart !980, Smithand Great Horned Owl pellets, from 4 nesting and 3 Knight !98!) and provide an opportunity for exam- roostingsites, collectedbetween October 1977-June ining resourcepartitioning between two membersc•f .- 1979•in EsquatzelCoulee, Franklin County, Wasl•-._ the samefeeding guild. We quantifieddiet relation- ington.The studyarea and its raptorpopulations are 176 ShortCommunications [Auk,Vol. 101

described in Knight and Smith (1982). Fitch (1947), (Kritzman 1977: 107) is temporally unavailable for Rudolph (1978), and Jaksi• and Y/tfiez (1980) have the strictly nocturnal Common Barn-Owl (Marti 1974, examinedfood-niche relationships between these two Rudolph 1978). Common Barn-Owls are the fourth specieswhen sympatricin California and Chile, pro- most important prey item, in terms of total prey bio- viding the opportunity for comparisons. mass, in the diet of Great Horned Owls. Prey remains in pellets were identified to species MPS of Great Horned Owls (œ _+ 2 SE = 55.24 _+ level, and mean fresh weights were obtained from 9.31 g, n = 872) was significantly greater (P < 0.001) the Bureauof Land Management (1979). The follow- than MPS of Common Barn-Owls (27.44 _+ 1.70 g, ing parametersfor eachowl specieswere calculated: n = 2628). This was the result of Great Horned Owls (1) mean prey size (MPS) (Herrera and Jaksi• 1980), preying on larger prey items such as rabbits and (2) food-niche breadth (Levins 1968: 43), and (3) food- Common Barn-Owls. The food-niche breadth of Great niche overlap (Pianka 1974: 2142). The numbers of Horned Owls was 4.12, considerablygreater than the prey speciesin each owl species'diet were used as 2.28 of Common Barn-Owls, suggestingthat Great entries for computationof niche breadthsand'over- Horned Owls apply a more uniform predation pres- laps. sure over a wider range of animals.Food-niche over- Small mammals were the major prey of Great lap was 0.97 (of a maximumof 1.00). This is largely Horned Owls and Common Barn-Owls at the Es- due to the importance of northern pocket gophers quatzelCoulee study area, both in percentageof total and GreatBasin pocket mice in both owl species'diets. prey items (91.1% and 97.8%, respectively) and in On our study area, these two speciesshow a high percentageof total prey biomass(77.9% and 98.3%) degree of ecologicaloverlap in several traits other (Table 1). Mammals compriseda lower percentageof than diet. Strongestevidence of habitat overlap is the total prey biomassin the diet of Great Horned Owls presenceof Common Barn-Owls in the diet of Great than of Common Barn-Owls (X2 = 162.34, P < 0.001). Horned Owls (Herrera and I-Iiraldo 1976). Addition- Two speciesof mammals, northern pocket gopher ally, Common Barn-Owls and Great Horned Owls (scientific names given in Table 1) and Great Basin showed few or no differences in nest-site structure, pocket mouse,comprised over half of the total prey mean nest height, mean nest-structureheight, land- items in the dietsof both owl species(50.1% for Great use patterns near nesting sites,and distancesof nest Horned Owl and 69.3% for Common Barn-Owl), as sites to human activity (Knight and Smith 1982). well as a major portion of total prey biomass(42.2% Nesting Great Horned Owls exhibited regular and and 73.9%). Rabbits were the third most important Common Barn-Owls random intraspecific distribu- prey species,in terms of biomass,in the diet of Great tions. Their interspecific pattern was random, how- Horned Owls, although they were not found in ever, indicating no interspecific territoriality. There Common Barn-Owl pellets. BecauseCommon Barn- were indications of ecological differences between Owls are smaller than Great Horned Owls (Marti the two species. Common Barn-Owls may utilize 1974), they may have difficulty in capturing rabbits habitats not frequented by Great Horned Owls, as (Jaksi• and Y/tftez 1980). Common Barn-Owls do oc- evidencedby differencesin prey species(e.g. house casionallyeat rabbits (Fitch 1947,Marti 1974);there- mouse).Common Barn-Owlsare strictly nocturnaland fore, an alternative explanation for the absenceof hunt mainly on the wing, whereas Great Horned rabbits as a Common Barn-Owl prey item in our study Owls are more crepuscular and hunt primarily by may be that the largely crepuscularNuttall cottontail flights from perches(Marti 1974, Rudolph 1978). Fi-

TABLE1. Prey items from 234 pelletsof Great Horned Owls and from 622 pelletsof CommonBarn-Owls in EsquatzelCoulee, eastern Washington. Common names are in parentheses.

Great Horned Owl Common Barn-Owl Number Total prey Number Total prey Weight of prey biomass of prey biomass Prey (g) (%) (%) (%) (%) Mammals Sylvilagusnuttallii 514 1.3 12.4 -- -- (Nuttall cottontail) Spermophiluswashingtoni 172 0.1 0.4 -- -- (Washington ground squirrel) Thomomystalpoides 200 7.7 29.7 5.44 39.3 (northern pocket gopher) January 1984] Short Communications 177

TAIlLE 1. Continued.

Great Horned Owl Common Barn-Owl Number Total prey Number Total prey Weight of prey biomass of prey biomass Prey (g) (%) (%) (%) (%) Perognathusparvus 15 43.1 12.5 63.85 34.6 (Great Basin pocket mouse) Dipodomysordii 53 1.4 1.4 0.53 1.0 (Ord's kangaroo rat) Reithrodontomysmegalotis 11 1.5 0.3 3.39 1.3 (western harvest mouse) Peromyscusmaniculatus 19 16.9 6.2 13.77 9.5 (deer mouse) Neotoma cinerea 310 0.3 2.1 0.04 0.4 (bushy-tailedwood rat) Microtus montanus 30 13.3 7.7 8.11 8.8 (montane meadow mouse) Rattussp. 300 -- -- 0.08 0.8 (rat) Mus musculus 17 0.7 0.2 1.75 1.1 (house mouse) Unidentified Microtinae 50 4.7 4.5 0.84 1.5 Mustelafrenata 178 0.1 0.4 (long-tailed weasel)

Birds Anasplatyrhynchos 1,185 0.1 2.6 (Mallard) Phasianuscolchicus 1,138 0.1 2.5 (Ring-necked Pheasant) Fulica americana 654 0.1 1.5 (American Coot) Charadriusvociferus 104 0.1 0.2 (Killdeer) Columba livia 332 -- -- 0.04 0.5 (Rock Dove) Tyto albaa 603 0.9 10.7 (Common Barn-Owl) Sturnusvulgaris 79 -- -- 0.08 0.2 (European Starling) Sturnellaneglecta 96 0.1 0.2• 0.04 0.1 (Western Meadowlark) Unidentified Fringillidae 26 0.3 0.2 0.23 0.2 Unidentified Passeriformes 56 1.1 1.2 0.27 0.5

Other Unidentified snake 207 0.3 1.4 Cyprinuscarpio 583 0.1 1.3 (carp) Unidentified Scorpionidae 1 0.3 tr b Stenopelmatussp. 2 4.0 0.2 1.45 0.1 (Jerusalem cricket) Unidentified Scarabaeidae 1 0.9 tr b 0.11 tr b Unidentified Locustidae 1 0.2 tr b

Total 872 99.9 2,628 99.9 Remainsfound beneath Great Horned Owl nestingand roostingsites; not foundin pellets. tr = trace amount; 0.01 g. 178 ShortCommunications [Auk,Vol. 101

TABLE2. Comparisonof three trophic statisticsfor the Great Horned Owl and Common Barn-Owl.

Mean prey Food-niche size (g) breadth Corn- Com- mon Great mon Great Food- Barn- Horned Barn- Horned niche Geographicarea and latitude Owl Owl Owl Owl overlap Study Central Washington,USA, 46øN 27 55 2.28 4.12 0.97 This study Northern California, USA, 41øN 30 28 1.98 1.92 0.99 Rudolph (1978)a Central California, USA, 37øN 52 98 2.95 8.17 0.24 Fitch (1947) a Central Chile, 33øS 123 266 3.98 6.90 0.48 Jaksie and Yafiez (1980) • We calculatedtrophic statistics using Table 1 in Rudolph (1978)and Tables2 and 7 in Fitch (1947).Weights of prey items were obtainedfrom both papersand the Bureauof Land Management(1979). nally, Great Horned Owls, becauseof their greater SouthAmerica is in agreementwith the findings of size and strengh, are able to utilize a wider variety Herrera and Hiraldo. MPS and food-niche breadth of potential prey. for both owl speciesincrease from Washington to The almost complete food-niche overlap between Chile, suggestingthat owls in temperateshrub-steppe the two speciesin our study and the low level of areasprey on smaller more numerousprey than do spatial and temporal segregationmay be the result owls in mediterranean habitats. For both owl species of their recent sympatry,which probably has been there are increases of over four-fold in MPS between causedby the expansionof irrigated farming in the Washingtonand Chile. Indeed, the MPS of Common region (Smith and Knight 1981:24). Alternatively, it Barn-Owlsin Chile is over twice as large as the MPS may reflect the normal situation wherever the two of Great Horned Owls in Washington. This is all the speciesoverlap. A comparisonof trophic parameters more surprisingas the adult body weight of Com- between thesetwo specieswhere they have coexisted mon Barn-Owls decreasesfrom north to south (Jaksie for differing periods of time is revealing. In areasof et al. 1982). This is in sharp contrast to the usual recent sympatry, such as northern California and relationshipsbetween predator and prey sizes(Wil- Washington (Stewart 1980), there is almostcomplete son 1975). In conclusion, we are unable to attribute food-niche overlap, whereas in areas where both the high dietary overlap between the Great Horned specieshave coexistedlonger, such as central Cali- Owl and the CommonBarn-Owl in our study solely fornia and central Chile, overlap values are one-half to recentsympatry. Further studiesof trophic rela- to one-quarter as great (Table 2). Thus, high food- tionships among New World owl speciescommuni- niche overlap values are not always characteristicof ties and prey populations are needed to clarify our associations between Great Horned Owls and Com- understandingof the coexistencemechanisms of owls. mon Barn-Owls,lending supportto the recent sym- For critical comments and important suggestions parry explanation.Perhaps the high diet overlap de- we are most grateful to Fabian M. Jaksi•, Susan K. tected in northern California and Washingtonis not Knight, William J. Mader, Carl D. Marti, Gordon H. the result of recent sympatrybut is insteada char- Orians, Thomas W. Sherry, Karen Steenhof, Chris- acteristicof Bubo-Tytopairs in temperate shrub-steppe topher H. Stinson,and Stanley A. Temple. We thank areas.Likewise, low diet overlap may be character- Sievert A. Rohwer (Thomas Burke Memorial Mu- istic of owls in mediterranean ecosystems.Central seum, Univ. Washington), and Gordon D. Alcorn and California is the climatic/physiognomic analog of Ellen B. Kritzman (Puget SoundMuseum of Natural central Chile, both of which are mediterranean-type History, Univ. Puget Sound) for allowing us to ex- ecosystems(Thrower and Bradbury1977). Herrera and amine museum specimens. J. Trumand and T. W. Hiraldo (1976) found high dietary overlap in owl Pietsch assisted in invertebrate and fish identifica- communitiesin middle and northern Europe and tion, respectively.Herb Camp kindly allowed us ac- minimal overlapin a mediterraneanowl community. cess to his land. These differences were associatedwith a tight clus- tering of owl speciesfeeding on microtineswith high LITERATURE CITED populationlevels in middle and northern Europeand fewer owl species with wider niche breadths de- BUREAU OF LAND MANAGEMENT. 1979. Snake River pending on less abundant small mammals in the birds of prey specialresearch report. Boise,Ida- mediterranean community. ho, BLM. Evidence presented in Table 2 for coexistingGreat FITCH,H. $. 1947. Predation by owls in the Sierran Horned Owls and Common Barn-Owls in North and foothills of California. Condor 49: 137-151. January 1984] ShortCommunications 179

HERRERA, C. M., & F. HIRALDO. 1976. Food-niche MARTI, C. D. 1974. Feeding ecology of four sym- and trophic relationshipsamong European owls. patric owls. Condor 76: 45-61. Ornis Scandinavica 7: 29-41. IVIANKA,E. R. 1974. Niche overlap and diffuse com- -, &:F. M. JAKSIQ.1980. Feeding ecology of the petition. Proc. Natl. Acad. Sci. USA 71: 2141- Barn Owl in central Chile and southern Spain: a 2145. comparativestudy. Auk 97: 760-767. RUDOL?H,S. G. 1978. Predation ecologyof coexist- JAKSI•,F. M., & J. L. YXI•IEZ. 1980. Differential uti- ing Great Horned and Barn owls. Wilson Bull. lization of prey resourcesby Great Horned Owls 90: 134-137. and Barn Owls in central Chile. Auk 97: 895- SMITH, D. G., & R. L. KN•CHT. 1981. Winter popu- 896. lation trends of raptors in Washington from ß R. L. SEIB,&: C. g. HERRERA. 1982. Predation Christmas bird counts. Olympia, Washingtonß by the Barn Owl (Tyto alba) in mediterranean Washington Dept. Game. habitats of Chileß Spain and California: a com- STEWART,I2. g. 1980. Population trends of Barn Owls parative approach. Amer. Midi. Natur. 107: 151- in North America. Amer. Birds 34: 698-700. 162. THROWER,N.J. w., &: D. E. BRADBURY(Eds.). 1977. KNIGHT,R. L., &: D. G. SMITH. 1982. Summer raptor Chile-California mediterranean scrub atlas: a populations of a Washington coulee. Northwest comparative analysis. Stroudsburg, Pennsylva- Sci. 56: 303-309. nia, Dowden, Hutchinson, & Ross. KRITZMAN, E. B. 1977. Little mammals of the Pacific WILSON,D.S. 1975. The adequacyof body size as a Northwest. Seattle, Washington, Pacific Search niche difference. Amer. Natur. 109: 769-784. Press. LEVINS,R. 1968. Evolution in changing environ- Received8 February1983, accepted6 July 1983. ments: some theoretical explorations. Monogr. Pop. Biol. No. 2, Princeton, New Jersey,Prince- ton Univ. Press.

Broodedness in Bobolinks

THOMAS A. GAVIN Departmentof Natural Resources,Cornell University, Ithaca, New York 14853USA

Extensive quantitative information on the number femaleswere more likely to be double-broodedthan of broods raised per female bird per seasonis scarce younger females in this species(Kluijver 1951). The (Cody 1971). This knowledge is necessaryfor an ac- latitude at which a bird breeds apparently is related curate analysis (or modelling) of population dynam- to broodednessin two opposing ways. Lack (1968: ics or reproductive "strategies" of individuals. In 196) stated that "... most passerinebirds of high lat- particular, a comparison of the reproductive fitness itudes raise more than one brood each year .... "pre- of monogamousmales with that of polygynousmales sumably becausethe longer daylength reducesthe depends on knowing the reproductive successof time to fledging relative to that at lower latitudes. paired femalesßwhich is pertinent also to questions Ecologicalconditions, however, may be suitable for dealing with sexual selection, female choice, and nesting for a longer period of time at lower latitudes, polygyny threshold models. To obtain complete data and, therefore, more broodsper female may be raised on reproduction, it is necessarythat individual fe- in a year than at higher latitudes (e.g. doves, many males be marked and monitored throughout the passerines). breeding season. The Bobolink (Dolichonyxoryzivorus) is a polygy- Replacementof lost clutchesof eggs is common in nous, ground-nesting icterid that winters from No- birds (Lack 1968), but Lack (1968: 302) generalized vember to March in South America between latitudes that "most speciesof birds raise only one brood in a 8øS and 32øS (Engels 1969) and breeds in North year, becausethe time required for courtship, nest- American hayfields and meadowsfrom May to July buildingßincubation and raising the young to inde- between latitudes 40øN and 50øN. Recent studies of pendence is so long that a second brood could not populations of marked birds in Wisconsin (Martin normally be completed before the ecological condi- 1971), Oregon (Wittenberger 1978), and New York tionswhich permit breedinghave endedfor the year." (R. L. Kalinoski pers. comm.) reaffirmed the earlier Great Tits (Parusmajor), however, were found to raise conclusion of Bent (1958) that Bobolinks can renest two broodsper year more frequentlyin habitatwhere after nest failure but do not attempt a secondnesting food was more abundant (Perrins 1965), and older after fledging young from the first nest. An un-