VESTIGES OF A PROPRIETARY INTEREST IN NESTS BY THE BROWN-HEADED COWBIRD PARASITIZING T'HE KIRTLAND'S WARBLER

HAROLD F. I•½•AYF1ELD

SINCE the Brown-headed Cowbird, Molothrus ater, dependsexclu- sively on other speciesto incubateits eggs and to care for its young, and since its fleeting visits to lay or remove eggsare rarely observed, we might be led to supposeits nestingdrive to be fully extinguishedby this time in its evolutionaryhistory. However, I believe such is not quite the case. On the contrary, I believethe female cowbird in many instancesgives a continuing and discriminating attention to the nests in which it lays its eggs,and consequentlymight be said to manifest, in moderatedegree, a vestigialproprietary interest in those nests. I have never known the male cowbird to touch a nest. This impressionhas developedgradually over the yearsof my study of the Kirtland's Warbler, Dendroicakirtlandi[, a study that becamein part also a study of the cowbird, since the reproductivelives of these two birds are so closelyentwined. Watching nest buildingintently. The cowbird'sinterest in the nest of the Kirtland's Warbler, as has been noted with other host species, first becomesapparent during nest building. At this time the female cowbird sits on an exposedperch some distanceaway and watches for extendedperiods while the warbler works. The spanof attentionof the cowbird is far longer than would be required merely to locate the nest. In a higher animal we might interpret this behavioras vicarious participation,and the term may not be totally amisshere. Certainly, the cowbird'sinvolvement would seemto be deeperthan that of a mere onlooker,if it is broughtto ovulationas a resultof this experience,as Harm has suggested(1937: 207). Looking into the nest at intervals. Occasionally,when the ownersof the nest are not present,the female cowbird walks up to the nest and peersinto it. This may occurvery early, even before the nest is tint ished. Hann (1941: 220) in his study of the Ovenbird, Seiurus auro- capillus,concluded that the "cowbirdmakes regular trips of inspection to nests." Once at a nest of the Kirtland's Warbler that was almost finishedbut hadnot yet receivedits first egg,I sawa cowbirdwalk up to the nest,poke its headthrough the overarchingcanopy of grass,look for a momentinto the cavity, and then fly away. This occurredtwo hoursafter sunriseand thus muchlater in the day than the usualegg- laying visit.

162 Aprill MAYFIELD,Proprietary Interest in Nests 163 1961 J

On another occasion, in the semidarkness before dawn I saw a cow- bird xvalkto a uest,fly away, and then return 10 minuteslater, laying an egg at one of the txvovisits. The dim light and the groundcover pre- vented me from being sure whether the egg was laid iu the first or secondvisit. Then, on the followingmorning, whi'ch •vas still one day before the first warbler egg arrived, the cowbird returned and looked into the nest at seven, nine, and ten o'clock. Doubtless, most of the cowbird's attention to nests centers in the building and egg-laying periods--that is, before incubationbegins. At thesetimes the cowbird is not likely to be opposedby the hosts,since smallsongbirds typically speud a very smallpart of thesedays near the nest. Also such cowbird visits are not likely to be recorded, for it is not often that a humanobserver will be watchinga nest constantlyfrom concealment at these stages. Cowbird visits are probably less frequent after incubation be,gins. Now the nest is closely defended, and the cowbird is usually routed before it gets to the nest. Further, it appearshesitant in mannerand more easily repelled in these broad-daylightvisits than in the predawn, egg-layingvisits. Hann (1937: 202) saw a cowbird push a shrieking Ovenbird right out of its nest in order to enter and lay an egg'. But [ saw a female Kirtland's Warbler put an unresistingfemale cowbird to flight three times in txvodays when the cowbirdwalked up to within a meterof the uest. This nestheld young birds almost ready to leave. Dubois ( 1956: 286) saw a female cowbird strike with its bill into the •lest of a Song Sparrow, MeIospiza •eIodia, injure one of the four nestlingsand carry away another. This nest,like that of the Kirtland's X¾arbler,was on the ground. Dubois doesnot mention seeingthe adult sparrows: so the nest was presumablyundefended at the moment. I suspectthat molestationof nestsso late in the cycleis rare. In my examples,the cowbirds were not banded,and therefore my inferencethat repeatedvisits to the same nest were made by the same cowbird is not proved. However, this inferenceis supportedby the findingsof others--particularlyFriedmann (1929: 175), Nice (1937: t54), and Laskey (1950: 167)--that eachcowbird bas a definitehome range. Also the idea that a host nest is normallythe object of attention for one cowbird exclusivelyis supportedby the fact that I have never found more than one cowbird egg laid in a (lay in a l(irtland Warbler's nest, although 52 per cent of parasitizednests in my study received more than one cowbirdegg each (Mayfield, 1960: 148). Here I have not consideredegg-laying visits after the normaltime- during incubation, with nestlings present, and in abandonednests. 164 M^¾F•:•D,Proprietary Interest i,, Nests [ •/o•.U•8

These "late" cowbird eggs compriseabout 10 per cent of the total (Mayfield, 1960: 159), but I believe they indicate, not a continuing interest in nests,but an unpremeditateduse, as might occur if a cowbird ready to lay an egg were to find the intendednest destroyed. l/isiting the nest repeatedlyto removeeggs. The cowbird'scontinu- ing interest in the Kirtland's Warbler nest is shownby the fact that it removesmore eggs than it lays, even though someparasitized nests do not lose any eggs; that is, a nest losing any eggs often losesseveral, presumablyas a result of separatevisits by the cowbird. To get a measureof the total loss,it is not sufficientsimply to count eggsas they appearand disappear.Too manyare removedbefore they are seenby anyone. Therefore, I prefer to deducethe lossesby com- paring a sampleof parasitizednests with a sampleof neststhat have not beenmolested by cowbirds,as follows:

Eggs in Kirtland's Warbler Nests (142 nests)

Warbler eggs in 67 nests not parasitized 310 Warbler eggs per nest not parasitized 4.63 Warbler eggs in 75 parasitized nests 205 Warbler eggs per parasitized nest 2.73 Warbler eggs lost per parasitized nest 1.90 Cowbird eggs in 75 parasitized nests 125 Cowbird eggs per parasitized nest 1.67 1.90 Warblereggs lost per cowbird egg gained, 1.67 1.14 (Nests were included only if clutches were judged com- plete because seen on at least two days without increase.)

Removingeggs only from nestsit is using. It is significantthat the cowbird'sdrive to removeeggs is directedonly at nestsin which it has laid or will lay its own eggs. Even thoughthere are other Kirtland's Warbler nestsin the vicinity, these other nestsalmost never lose eggs without the destruction of the entire clutch at one time, a kind of destructionnot attributed to the cowbird. Therefore, the cowbird nor- mally seemsnot to seek eggs for themselvesbut takes eggsonly from nestsit is using for its own. Removingonly the hosts'eggs. The cowbirddoes not removeeggs at random, but is able to discriminate between its own and the hosts' eggs, some of which are treated as other birds might be expectedto treat "foreign objects." I have 13 definite instanceswhere cowbirds April1 M^¾r•;I•U Proprietary interest in Nests 165 1961 ] removedeggs fron• nestscontaining both cowbirdeggs and Kirtland's Warhler eggs,and each time the cowbird took only warbler eggs. In these instancesthe cowbirds were confronted with choices among a total of 23 warbler eggsand 19 cowbirdeggs. The oddsagainst 13 correctchoices without an error are 3,000 to 1 by chancealone. Probably the coxvbirddiscriminates on a basis of size rather than color or pattern. Both kinds of eggsare whitish and lightly speckled, but the warbler egg is smaller. At the nest of the Ovenbird the cowbird has more difficultyin distinguishingthe hosts'eggs, which are nearer the size of its own. But eveu here Hann (1937' 204) noted the lossof only four cowbirdeggs as against30 Ovenbirdeggs lost. The mean size of Brown-headedCowbird eggsin Kirtland's Warbler nestsis 20.9 by 16.5min. (N = 24); of Ovenbirdeggs, 20.3 by 15.6mm. (N: 48, Hann, 1937: 172); of Kirtland's Warbler eggs,18.1 by 13.9 min. (N = 154). Never emptyingthe nest completely. I have never known the cowbird to take the only egg presentin a Kirtland's Warbler nest. Long ago Burroughs (1887: 30) noted that the cowbird takes an egg only when the nest contained"txvo or more eggs," and Harm (1941: 212) found this to be true also in nestsof the Ovenbird. However, it is only fair to point out that few Kirtland's Warbler nestswere found so early and watchedso closelythat a first egg might not have been laid and re- moved unseen. This subtledistinction by the cowhirdmay have survivalvalue, be- causesongbirds are more likely to desertif the nest is emptiedthan if its contentsare merely reduced; on the other hand, an inattentivehost might not be aware of the lossof an only egg if it were replacedby a cowbird egg before the host'snext visit. Destroying several eggsin a crowdednest. I have a small number of examplessuggesting that a cowbirdremoving eggs from a crowdednest may create havoc in it. Not surprisingly, these instancesare few, becausethe co•vbirdis most active early in the egg-layingstage and becausefull clutchesare likely to be under incubationand therefore defended. (The Kirtlaud'sWarbler normallybegins incubation on the next-to-last egg.) In neststhat bad completesets of eggswhen cowbirdsvisited them to lay and to remove eggs,the resultswere as follows: 166 M^¾v•L• Proprietary' Interest in Nests f[ Vol.Auk 78

Cowbird Dama.qeto Full Ulntchesof Kirtland's Warbler Eggs (7 nests)

Warbler eggs Cowbird eggs Before cowbird damage 32 1' After cowbird damage 8 9 Cowbird eggs added, 8 Warbler eggs destroyed, 24

* One nest held a cowbird egg laid earlier.

The resultantset of eggsin eachinstance xvas a clutchof oneto four eggs,that is, aboutthe usualnumber to be foundin a nestat the usual time of a coxvbird's visit. Excluded from this table xvere four other full clutchesthat receivedadditional cowbird eggswithout loss.

SUMMARY

The Broxvn-headedCo•vbird still retains vestigesof a nestingdrive shoxvnin xvays suggestinga proprietary interest in the Kirtland's Warbler nestsit parasitizes,as folloxvs: 1. Watching nest buildingintently. 2. Looking into the nest at intervals. 3. Visiting the nest repeatedlyto removeeggs. 4. Removingeggs only from nestsit is using. 5. Removingonly the host'sejgs and not its oxvu. 6. Never emptyingthe nest completely. 7. Destroyingseveral eggs in a crowdednest.

LVrERA'rURE Ct'rED

BURROUghS,J. 1887. Birds and bees. Houghton Mifflin, Boston. 88 I)P. DUBols, A.D. 1956. A cowbird incident. Auk, 711: 286. Fm•;1)M^NN,H. 1929. The cowbirds. Charles C. Thomas, Springfield, Ill. viii + 421 pp. H^NN, H.W. 1937. Life history of the Oven-hird in southern Michigan. Wilson Bull., 49: 145-237. H^NN, H. W. 1941. The cowbird at the nest. Wilson Bull., 511: 211-221. L^sK•;v, A. R. 1950. Cowbird behavior. Wilson Bull., 62: 157-174. M^vVmlm, H.F. 1960. The Kirtland's Warbler. Cranbrook Inst. of Sci., Bloom- field Hills, Mich. 226 pp. Nxc•;, M. M. 1937. Studies in the life history of the Song' Sparrow, I. Trans. Linn. Soc. New York, 4: 1-242.

River Road, RFD, Waterville, Ohio.