THE BEHAVIOR AND REPRODUCTIVE SUCCESS OF THE BLUE-GRAY GNATCATCHER

RICHARD B. ROOT Museum of Vertebrate Zoology University of California Berkeley, California 94720 (Present address: Department of Entomology and Limnology and, Section on Ecology and Systematics Cornell University Ithaca, New York 14850)

Despite its broad distribution and relative spectrum of activities for a representative abundance, the Blue-gray Gnatcatcher, Polio- sample of breeding pairs. I visited each of ptila cam&a, has never been the subject of an the territories on the study plot two or three extended behavioral investigation. Our knowl- times daily except for short periods when I edge of this species, best summarized in Bent was away. On at least one of these daily visits, ( 1949), is based upon short-term observations I followed a member of the pair for about 20 at a few nests and isolated notes on other minutes, recording its movements and en- aspects of behavior. Nice (1932) has written counters with other gnatcatchers on a map of the most extensive paper on the biology of the area. I stayed with the bird for a longer the gnatcatcher (in the following text, “gnat- period when it was engaged in an unfamiliar catcher” refers only to Polioptilu caerulea) activity. that is based upon original observations. Our A special effort was made to find all of the information on the biology of other Polioptila nests on the study plot; every extant nest and spp. is even more fragmentary. fledgling brood was observed at least once a While studying the niche exploitation pat- day. It was usually possible to link the succes- tern of the gnatcatcher, I had an opportunity sive nests of each pair together, because gnat- to follow its general behavior closely on the catchers normally re-use material from their breeding grounds in California and the winter own previous nests ( discussed below ) . Most quarters in Arizona. The following behavioral of the later nests were found by following pairs observations are being published here because which were dismantling abandoned nests. they relate to subjects of greatest interest to Four adults were captured by using the ornithologists. My material on the habitat re- predator-decoy method (Root and Yarrow quirements and foraging ecology of the gnat- 1967). These, together with nestlings from catcher appears elsewhere (Root 1967). three broods, were marked with color bands and followed throughout much of the 1983 METHODS breeding season. On ,the basis of the almost I observed breeding gnatcatchers during lQSQ- daily sightings of each-bird, the succession of 1961, and 1963 at the Hastings Natural History nests, and the behavior of the banded individ- Reservation, Monterey County, California. uals, it was possible to piece together com- The most thoroagh observations were made in plete biographies for six breeding pairs during 1963 when I followed the population from the the 1963 season. arrival of the first birds in March until breed- Each of the gnatcatchers’ behaviors was ing activities ended in late August. My described in my notes until I was completely principal study area, where I spent more than familiar with the details and variability of its 1266 hours, was 56.1 acres (22.7 hectares) in performance. The principal behaviors were extent and contained large stands of oak then classified and noted whenever they were woodland and chaparral. Twelve pairs of observed. These data gave a rough measure gnatcatchers occurred on this plot in 1963. of the relative frequency of various behaviors. Additional observations were made in a Many of my observations were made in the variety of vegetation types at other localities oak woodland where the openness of the vege- in central California. tation afforded clear visibility. After a few Our best field data on passerine behavior weeks the birds became accustomed to my have often come from extended observations presence and permitted me to observe their at a few well-situated nests. In this study I undisturbed activities from as close as 15 have tried instead to follow the complete feet. The Condor, 71:16-31, 1969 1161 BEHAVIOR AND REPRODUCTION IN THE BLUE-GRAY GNATCATCHER 17

Each nest and group of fledglings was ob- checked by making further observations. The served for at least 10 minutes daily, and some- areas of these territories were estimated by times for periods of over an hour. I have drawing the final boundaries on an aerial consulted the detailed nest records kept at the photograph (which had been taken especially Reservation by John A. Gray, Jr., Robert to facilitate accurate mapping at the Reserva- Holdenreid, and Charles G. Sibley to fill out tion) and tracing the outlines with a planirn- my account of behavior at the nest. In addi- eter. Between 28 April and 5 May 1963, a tion, the field notes of John Davis, Jean M. period when territorial boundaries are well Linsdale, and others were used to document defined and relatively stable, these territories the seasonal occurrence of gnatcatchers at were found to average 4.6 (2.2-7.4) acres or the Hastings Reservation. 1.8 (0.9-3.0) hectares in extent. The winter behavior of the Blue-gray Before the beginning of a nest and during Gnatcatcher and the Black-tailed Gnatcatcher, the period when eggs are being laid, the male, P. melunura, was studied on three trips, total- often accompanied by the female, patrols the ing 24 days of observation, during 1962 and periphery of the territory frequently. Once 1963. Most of the winter studies were carried nesting is underway, the males’ patrolling ac- out on the floodplain of the Colorado River tivity is most pronounced during the half near Yuma, Arizona, and at three localities hour period that follows sunrise. While patrol- near Tucson, Arizona (for a more detailed ling, the male gives an advertising song which description of my study areas, see Root 1967). consists of a variable array of from four to eight short sibilant phrases (e.g., spee spuu RESULTS spee SW&). The advertising song of the TERRITORIAUTY gnatcatcher is not as loud as that given by most other passerine birds living at the Reser- Immediately after arriving on the breeding vation. grounds, each male gnatcatcher begins to de- When another male or pair is encountered fend a territory. At the Reservation, the first along a boundary, the males approach each males have been sighted as early as 24 Feb- other and engage in an intense series of ruary and as late as 30 March. During the vocalizations and posturing. The outcome of spring of 1963 a period of over a month 62 border incidents was observed in this elapsed between the arrival of the first resident study. Upon first recognizing an alien in- male on 6 March and the establishment of the dividual, the male usually assumes a tail- last territory on 18 April. This lack of spread posture: the tail is fanned to its fullest synchrony in territory establishment appears extent, displaying the white outer rectrices to be related to temporal differences in the which are partially concealed by black abundance of foliage anthropods on different rectrices under most circumstances. The tail territories ( Root 1967). is held horizontally to the long axis of the All of the gnatcatchers’ courtship, nesting, body and wagged from side to side. Often and foraging activities occur within the the males engage in short undulating flights boundaries of the territory. The territory is during which the spread tail is displayed. defended by the males and sometimes by the The advertising song is given frequently dur- female against all conspecific individuals. ing boundary disputes, but the most charac- Adult females and juveniles are tolerated on teristic vocalization given in these encounters the territory when they join the resident pair is a slow series of emphatic peeew calls which in harassing a predator (below). On three are accented on the first syllable. occasions, however, pairs were seen to leave When disputing males meet on a well- their nest vulnerable to the predators’ attack marked boundary, they move parallel to one in order to drive an adult male, apparently another with a distance of 40 to 150 feet attracted by the pairs’ mobbing calls, from the territory. separating them. Along poorly defined bound- I was able to determine the complete bound- aries, the territory is defended in a more vigor- ary of nine territories with reasonable ac- ous fashion. The males approach to within 30 curacy by following the males for prolonged feet of one another and continue to give the periods in the early morning and afternoon. tail-spread display and the advertising song The movements and territorial encounters of or peeew calls. In addition, one of the males these males over a period of at least four may occasionally assume an upright posture: successive days were plotted on a base map. the closed tail is depressed below the long This map furnished the basis for drawing axis of the body and the feathers of the lower approximate boundaries which were then dorsum and flanks are raised. Whispered 18 RICHARD B. ROOT vocalizations, including the long rambling series of warbles, whistles, and calls which are aVocal Exchange commonly assumed to function as the gnat- catchers’ “song” (Bent 1949) are sometimes ICombat and Chasing given. Since the upright posture and whis- pered vocalizations are associated with court- ship activities (below ), their occurrence in boundary disputes may be elicited by the presence of an opponents’ mate. In the most intense disputes, one male FIGURE 1. The frequency of territorial defense in flies directly at the other male. If this second the Blue-gray Gnatcatcher during 1963. Each column male retreats, it is chased as far as 70 feet, represents a period of 5 days. the pursuer snapping repeatedly with its beak at the fleeing male. Frequently, the second male flies up to meet its opponent in midair. defended. Those disputes involving combat In this event, the males, their breasts nearly that occur late in the season are associated touching, ascend in a vertical flight as high as with the shifts in territory boundaries which 40 feet above the ground, all the while snap- often accompany renesting activity. Starting ping at each other with their beaks. On six in late May, there is a gradual decrease in occasions, the disputants were observed to fall territorial behavior. to the ground where they remained locked in combat for several seconds before ascend- PAIR FORMATION AND COURTSHIP ing again in the vertical flight. The com- The pair bond was already complete when batants break apart at the apex of these verti- the three males which established territories cal flights and a short chase ensues. Between in chaparral were first discovered during my bouts of fighting, the displaying males ac- daily rounds of the study area in March, tively search for food. Any large prey item 1963. The bond in these pairs must have been which they capture is held conspicuously in formed before their arrival at the Reserva- the tip of the beak for several seconds before tion or, most likely, within the .%-hour period it is devoured. following the males’ settlement. In the oak The female usually takes a less active part woodlands, there was a delay between the in territory defense than does the male. Dur- males’ establishment of a territory and the ing a dispute, the female normally gives peeew formation of a pair bond, although unattached calls while remaining about 15 feet behind females were seen drifting through the study her displaying mate. On one occasion when area during this period. a pair trespassed deeply into another territory, Following their arrival unmated males fre- the two females fell to the ground where they quently inspect potential nest sites on the remained locked in combat for several seconds territory. During these inspections, the male while the males displayed to each other a few crouches in a tree fork, similar in configura- feet away. When the female is alone and tion to those in which nests are found, and encounters a trespasser, she usually assumes turns at least 180 degrees while looking at the the tail-spread posture and gives peeew calls. surroundings. On two occasions, I had an On three occasions, the defending female at- opportunity to observe the sequence of events tempted to chase the alien male until she was that immediately follows the arrival of a fe- joined by her own mate. male on the territory. In both cases, I had A characteristic feature of territorial be- been following a single male for several havior in the gnatcatcher is the high fre- minutes before he encountered a recently quency with which boundary encounters re- formed pair on the territory boundary. In sult in actual combat (fig. 1). Territory each instance, the female followed the un- defense is most intense during ,the period mated male onto his territory for a short immediately following the arrival of males. period before she returned to her original Once nesting is underway, the boundaries be- mate. The male responded to the females’ come relatively stable, and from late April presence by perching near her in the upright through June I have repeatedly observed posture (described above) and singing an gnatcatchers that respected an undefended elaborate whispered song which contained a boundary. The boundaries are patrolled in- variable array of warbles, whistles, and calls. frequently and only the areas surrounding the The female then followed the male to a few nest and favored foraging sites are regularly potential nest sites in which the male perched BEHAVIOR AND REPRODUCTION IN THE BLUE-GRAY GNATCATCHER 19

for several seconds in the upright posture and Date : April 24 25 26 27 26 29 continued giving the whispered vocalizations. Before the beginning of a nest and follow- Weather: l $ l Q 0 0 ing the destruction of a nest, the female is OF= 46O 47O 44O 53” 64O 64’ frequently seen following her mate. When No. of visits the female lags behind, the male returns to sing and posture near her. The male shows 0.5 0 0 5.0 6.0 6.5 0 several nests sites to the female by perching c? 2.0 0 0 9.0 2.0 10.5 in suitable forks and giving the whispered vocalizations. The female then crouches in zz 2.5 0 0 13.0 6.0 I%0 these same forks and, while treading with her feet, turns around and examines the surround- Sets. at Nest ings closely. Many of the behavioral com- ponents in this nest-inspection display are the ? 2 O O 3* “O 19’ same as those employed in nest building. The rjt 26 0 0 64 25 114 female may inspect forks that are not shown zz 29 0 0 102 135 305 to her by the male. Later the pair focuses its FIGURE 2. The relationship between weather con- attention on two or three prospective nest ditions and nest building tempo during 15 minute in- sites, where they trade places in the fork re- tervals. The data were taken at the same time each peatedly and exchange token nest materials. day at two nests where construction began on 22 Following the beginning of nest construction, April 1963. Degree of cloud cover is represented by the darkening of the circles, rainfall by the diagonal the male continues occasionally to give the lines. upright posture and whispered vocalizations when the female is near the nest. During early April, females which subse- forming all of the nest-building activities, al- quently disappeared were seen on four oc- though there is an indication (table 1 and casions to follow a male and inspect nest sites casual observations ) that the female does most with him for periods of more than an hour. of the final arranging of materials when the These observations suggest that at least some nest is near completion. The labor of build- females examine different territories before ing at nests before the fledging of the first they settle in an area and begin breeding. brood is shared about equally between the Other displays that probably function in sexes; during short observation periods at 18 courtship and maintenance of the pair bond different nests, males made 40 trips to the are occasionally seen throughout the spring nest while females made 47 trips (also see and early summer. These include a head fig. 2). Following the fledging of a brood, flagging display, in which the bird faces its the males do most of the construction on the mate and wags its head from side to side with subsequent nest; in the course of observations the beak partially opened; and begging, in at seven such nests, males made 28 trips as which the bird crouches, cocks the tail slightly compared with only 6 trips by females. The upward, and quivers the wings in front of the frequency and length of bouts of construction mate. The male sometimes reacts aggressively decreases during overcast or rainy weather toward the female by bill snapping and chas- (fig. 2). ing her for a short distance. Throughout the The nest cup is built of dry grasses and year, both members of the pair exchange speee plant fibers which are arranged roughly in a calls while they are foraging near one another. swirl, with the finer materials being situated During the period just before the beginning toward the interior. Often the twigs and small of the autumn migration, some adults are still branches that form a part of the nest site are traveling together in pairs while others seem incorporated into the cup. The nest is lined to be alone. On 23 August, two pairs, which with plant down and with feathers from several were without broods, were each followed for species of birds. The exterior is ornamented 20 minutes. When these birds were shot, they with crustose lichens affixed to strands of were found to be midway through the post- spider web and placed in the outer portions nuptial molt. The testes of the males were less of the cup. This ornamentation serves to than 1 mm in length. camouflage the nest under most circumstances, as the bark of the deciduous oaks (chiefly, NEST ARCHITECTURE Quercus Dmglasii), where most nests are The nest is built by both members of the pair found at the Reservation, is usually covered (Nice 1932). Both sexes are capable of per- with the same species of lichens. The oma- 20 RICHARD B. ROOT

mentation does not always closely match the found no evidence to suggest that pairs re-use surroundings of the nest, however, and nests building materials from nests other than their that are placed in chamise (Adenostomu own. Two pairs were observed to re-use ma- fascicuZutum) or small oaks having a sparse terials from two of their former nests in the lichen covering are often quite conspicuous. construction of a new nest. Chamberlin (1901) reports that pieces of Both members of the pair dismantle old burnt bark were used in place of lichens in nests. The birds always perch outside the an area that had recently been burned over. nest cup while tearing it apart, While a A nest that had no external ornamentation major portion of a new nest may come from was found in a desert willow (Chilopsis a previous nest, it always contains some new line&s) wash at the Joshua Tree National material. Monument in southern California (Miller and The re-use of nest materials that are ac- Stebbins 1964). cumulated in the first nest probably reduces The nest is anchored in place by “fingers” the amount of effort required to build sub- of spider web that extend as much as 15 mm sequent nests. The building period, the in- from the cup to where they are attached to the terval between the beginning of construction bark of the supporting branches. The ma- and the laying of the first egg, averaged 13.6 terials are so arranged as to lend a degree of (12-15) days at five nests where only new elasticity to the nest. Before egg-laying and materials were used, and 4.8 (3-6) days at again on the day of fledging, measurements five nests where building materials were were made of a nest that produced five young. drawn from a previous nest. Since all five of During this period, the inside diameter at the the new nests were built immediately follow- rim increased from 40 to 56 mm, the outside ing arrival in the spring, some of the differ- diameter increased from 57 to 68 mm, and the ences in the length of the building period may depth of the cup decreased from 37 to 32 mm. be related to differences in the reproductive Gnatcatchers gather nest materials at dis- condition of the birds. tances as much as 250 feet from the nest site. While building the nest, the gnatcatchers In obtaining these materials, the birds must always stand in the center of the supporting frequently engage in activities for which they fork or cup. During the first or platform stage appear to be poorly adapted. Gnatcatchers of construction, the birds attach a loose ac- have been observed having obvious difficulty cumulation of plant materials to the nest fork in tugging dead grasses free of tangled vegeta- with spider silk. At this time, the builders tion and in maintaining their perch on a tree frequently lean down to place silk on the bark trunk while prying lichens from the bark. The below the fork. When the platform is only inability of the gnatcatcher to hold an object a few millimeters high, lichens are placed on with the foot while pulling at it with the beak the exterior by wiping the beak along the would appear to be a major hindrance in platforms’ edge, where the lichens adhere to gathering nest material. the “sticky” spider silk that covers the outer When a nest site is abandoned, the gnat- margins. Thus the nest is camouflaged from catchers frequently tear the old nest apart and its very beginning. When there is a deep ac- re-use the building materials in a new nest. cumulation of material in the nest fork, the This nest-moving behavior has been reported gnatcatchers begin an additional series of from widely scattered localities throughout building activities. After dropping material the breeding range (Lloyd 1932; Hargrave onto the platform, the birds push their breasts 1933; Murray 1934). Of the 42 nests that I into the mass and while treading with their found in 1963, 23 were known to contain ma- feet, turn through an arc of as great as 180 terial from a previous nest. A banded female degrees. This serves to push up the walls of re-used building materials in six consecutive the nest. Once there is a shallow depression nests, and on one occasion moved the bulk in the platform, some materials are arranged of a complete nest to a new site which was around the rim by poking up and down with over 500 feet away. Materials are taken from the partially opened beak while turning nests that are abandoned following fledging around. After the walls of the cup are a few of the young as well as those that are deserted centimeters high, the rim is stroked with the during the construction and incubation pe- underside of the beak, neck, and tail while riods. Three nests that were overrun with the the bird is engaged in foot treading. The mesostigmatic mite Ornithonyssus sylviarum ornamentation of the exterior and some lining were not moved, even though the pair re- are added throughout the construction period. nested within a distance of only 150 feet. I As the nest nears completion, the gnatcatchers BEHAVIOR AND REPRODUCTION IN THE BLUE-GRAY GNATCATCHER 21

TABLE 1. The relationship between completeness covered for more than a few minutes. Both of the nest and building tempo.” members of the pair call to their mates oc-

No. of visits Sets. at nest casionally while sitting on the nest. Follow- ing the exchange of several calls, they fre- stage TillE ?dZ 0 $I:‘ quently trade places on the nest, although New platform 12:15 6 713 66 63 129 birds that have been silent for several minutes Shallow are also relieved from incubation. depression 10:50 7 10 17 146 126 272 Low cup 11:15 2 0 2 65 0 65 Full cup 11:46 7 0 7 118 0 118 DEVELOPMENT OF YOUNG AND PARENTAL CARE 1 Data gathered during ZO-minute observation periods at each nest on 1 May 1983 when the sky was clear and tbe tempera- ture was 85°F. Nest stages are described in the text. I was able to follow the complete develop- ment of the nestlings from the day of hatching to the day of fledging at only two nests. The spend longer periods at the nest (table 1) young remained 12 days in one nest and 13 arranging materials. days in the other. The post-hatching develop- Because of its adhesiveness (Eisner et al. ment in these broods was as follows: 1964), spider silk appears to be an essential ingredient of the nest. In addition to the First day, the nestlings are naked and their eyes are closed. When the edge of the nest is tapped, uses described above, spider silk binds the they frequently raise their heads and open their other materials into an elastic mass that can beaks in a gaping response. The mouth lining is be shaped by the simple treading and stroking bright yellow with two black spots on the tongue. movements. Second day, the developing flight feathers are evident as short pinfeathers. Both members of the pair give a few calls Third day, the developing pinfeathers of the body when approaching or leaving the nest. This tracts become evident. conspicuous behavior made it easy for me to Fifth day, the eyes are opened to narrow slits find nests during the construction period. when the nestlings are gaping and remain closed the rest of the time. Occasionally, the gnatcatchers exchange build- Sixth and seventh days, the flight feathers and ing materials with each other near the nest. feathers of the body tracts erupt at the tips of the feather sheaths. EGG LAYING AND INCUBATION Eighth day, the eyes remain open most of the Of the 20 complete clutches examined in this time. study, 2 contained 3 eggs, 11 contained 4 Ninth day, the nestlings give peet calls from the nest when the adults are nearby. eggs, and 7 contained 5 eggs. At three nests, Tenth and eleventh days, the nestlings begin to it was found that one egg is laid on each stand up while stretching their wings and preening consecutive day, probably in the early mom- their flight feathers. ing. Twelfth day, the nestlings climb onto the nest rim and nearby branches while stretching and preening, The incubation period, reckoned from the after which they return to the nest. date when the last egg was laid to the date when the last egg hatched, was 15 days at two The nestlings are brooded during the day- nests. time until at least the eighth day after hatch- On all nine territories for which I had ing. Brooding occurs most frequently on the sufficient data, both members of the pair en- first two days following hatching. Nice (1932) gaged in incubation. During short visits made observed a steady decline with increasing age in the morning to 20 different nests, the male of the nestlings in the per cent of the time was incubating on 30 occasions and the female spent brooding at two nests in Oklahoma. on 70 occasions. There was no obvious dif- The females do most of the brooding, although ference in the incubation pattern of those males occasionally sit on the nest for a short pairs that had previously fledged a brood. period. The male brings food to the brooding At a nest watched by Sibley (field notes) in female, which she then feeds to the nestlings, the late afternoon, the male incubated for 128 or occasionally eats herself. The female minutes and the female incubated for only mandibulates large prey items which the male 48 minutes. During this period, the birds sat brings for several seconds before feeding them continuously on the nest for an average of to the young. When the sun is high and 25.1 (7554.5) minutes. Similar incubation shining directly upon the nestlings, one of patterns at nests in Oklahoma and Ohio have the adults frequently shades the nest by stand- been described by Nice ( 1932). ing on the rim with the wings partially opened. Incubation occurs sporadically on the first The adults carry the fecal sacs that are day following completion of the clutch, but on produced by the nestlings at least 30 feet from subsequent days the eggs are rarely left un- the nest before dropping them. When a fecal 22 RICHARD B. ROOT

TABLE 2. Number of feeding trips made to nests gnatcatchers averaged 28.0 g in body weight, containing nestlings of different ages.” while the total weight of five gnatcatcher fledglings, the maximum size of a normal Aam%; Trips during 2 hours brood, was 29.3 g. nestlings Observation Nest in days Date period 0 d z During the four days before fledging, the 1 3 27 May 09:15-11:15 17 23 40 male made 39.1 per cent of the total feeding 14:00-16:OO 20 12 32 trips reported in table 3. At 15 other nests, 4 28May 09:15-11:15 14 24 38 where short-term observations were made, 14:00-16:OO 14 25 39 the males made 29.4 per cent (n = 136) of 2 12 22May 14:40-16:40 Ob 67 67 the feeding trips to first broods and 53.4 per There‘ were five nestlings in both nests. cent (n = 131) of the feeding trips to second b Female present on territory. broods. The males’ feeding rate appears to be adjustable, however; when the female is sac falls out of the nest, it is retrieved and brooding the young nestlings at these early carried away. On occasion, the adults eat the nests, the male brings most of the food. As a fecal sacs or feed them to the nestlings. result of this adaptability, the male may be Food is occasionally brought to the nestlings able to increase his role in feeding the brood from distances as much as 500 feet from the in the event of a severe food shortage. Such nest, but most of the adults ’ foraging activities a response has been observed in the Blue Tit are concentrated within a radius of about 200 (Parus caeruleus) in which a male nearly feet around the nest. The adults immobilize doubled his feeding rate at the nest after his all large prey that they obtain for the young mate disappeared (Arnold and Arnold 1952). by beating it against a perch. The prey is The fledging of a brood usually takes less carried to the nest in the tip of the beak and than a day. At one nest, the first fledgling usually placed head first into the nestlings’ left at lo:14 on 30 June, the second at 05:24 mouth. If a nestling has difficulty in swallow- on 1 July, and the remaining two at 06:28 on ing the item, the adult removes the prey and 1 July. At three other nests, all four or five repositions it in the nestlings’ mouth. Some- nestlings fledged within an 18-hour period times prey that is taken from the nestling is between my visits to the nest. On one oc- beaten and mandibulated by the adults before casion, at least three young left a nest during it is again fed to the young. Near the nest, the a 20-minute period when the adults were female occasionally begs the male for food, harassing a stuffed Screech Owl (Otus asio) which she immediately feeds to the nestlings. which I had placed about 30 feet from the Prolonged observations at nests show that nest. The adults do not seem excited when a vounger broods are fed less frequently (table fledgling leaves the nest. 2, and Nice 1932) and that, while the highest The post-fledging development of three feeding rates usually occur during the morn- broods that were observed in 1963 was as ing, the feeding tempo does not vary greatly follows: throughout the day (tables 2 and 3). First day, the fledglings are capable of flying Nestlings of the Brown-headed Cowbird distances of at least 20 feet at a time. They fre- ( Molothrus ater ), which is a brood parasite of quently fall or flap their wings in an effort to the gnatcatcher, are fed at a rate similar to maintain balance when landing. It may be that the aerial maneuvers, which are so characteristic of the that of a normal brood. On 7 July, a single gnatcatchers’ foraging repertoire, develop from these cowbird which fledged three days later was initial balancing movements. As a result of a tendency fed 12 times by the male and 15 times by the for the fledglings to follow the adults, the brood usually remains close together. They spend most of female between 08:55 and 0955. Two fledg- the time perched quietly in dense foliage where they ling cowbirds fed singly by separate pairs of engage in frequent bouts of preening. The fledglings

TABLE 3. Number of feeding trips to a nest during two hour periods at different times of day.”

Dates

27 June 28 June 29 June 30 June Observation period 0 d B P d z 0 d E 0 d z

05:15-07:15 31 29 60 38 27 65 45 23 68 52 27 79 09:15-11:15 35 27 62 25 20 45 62 24 86 47 23 70 13:15-15:15 21 23 44 29 18 47 28 7 35 16 27 43

*Observations were made at one nest containing four advanced nestlings. The first fledgling left the nest at lo:14 on 30 June, while the other three remained until the morning of 1 July. BEHAVIOR AND REPRODUCTION IN THE BLUE-GRAY GNATCATCHER 23

TABLE 4. The number of times an individual changes markedly between the first and fledeline is fed in relationshin^ to the individual fed second broods. During short periods of ob- immidiately preceding.” servation on eight different groups of fledg- FOUOWS lings, the males made only 9.4 per cent (n = Individuals A B C D E 53) of the feeding trips to first broods, and 84.4 per cent (n = 66) of the trips to second ,^ A 17 5 1 3 1 broods. The possible reasons for the sharp z8 B 2 9 2 4 1 0 c 2 1 11 1 1 decline in the females’ attentiveness in feed- 2 D 4 2 1 15 4 ing the late brood are unknown. La E 2 1 1 3 11 In late August, two adult males and a fe- male, collected while feeding fledglings, were Totals 27 18 16 26 18 midway through the postnuptial molt. a This brood had fledged the previous day. The observations were made by Sibley (field notes) between 08 :22 and 11:52 on 23 May 1938. MAINTENANCE BEHAVIOR Preening was rarely observed, probably be- do not forage for themselves, but instead, they beg cause birds engaging in this activity are rela- food from the adults by quivering their wings, gaping, tively inconspicuous. During the five bouts and giving a short series of peetcalls. of preening that I observed closely, the birds Second dav. the fledzlines have less difficultv in remained silent while perching for a few moving from-perch to perch: but still remain perched minutes on an exposed twig in the direct sun- in one place for periods of longer than 30 minutes. Often two or three members of the brood will perch light. Between intervals of preening the body so close together that they touch each other. plumage, the birds probe the rump and base By the fourth day, the fledglings are noticeably of the tail by reaching back over the closed more active in hopping about and in following the wing. The ventral surface at the base of the adults for short distances. Bill wiping is first seen on this day. tail is rotated upward and preened over the Fifth day, the young begin to peck at leaves and wing. The wing is partially opened and ex- twigs near their perch. This pecking does not seem tended to permit preening of the inner surface. to be directed at anything in particular, as one The head was scratched by placing the leg fledgling was seen to peck at its foot and at the same leaf on repeated occasions over a lOminute over the partially extended wing on the six period. Another fledgling recovered a perch from occasions when this activity was clearly ob- which it had slipped by executing a hovering served. The beak and the feathers at its base maneuver similar to that employed by the adults are frequently rubbed against the perch ( Root 1967 ) . By the ninth day, the fledglings are obtaining some During July and August conditions are very of their own food by gleaning. dry at the Reservation and the only free- By the thirteenth day, the young are capable of standing water occurs at a few isolated pools. performing all of the foraging maneuvers that are Gnatcatchers were never observed to visit characteristic of the adult repertoire. Still, the these pools. Dust bathing and anting were feldglings obtain the bulk of their food from the adults. The brood is still highly gregarious and never observed. follows the adults constantly. The maneuvers employed in capturing and By the sixteenth day, the young are striking out manipulating food are described elsewhere on their own frequently, but they are still fed oc- ( Root 1967). casionally by the adults until at least the nineteenth day after fledging. INTERSPECIFIC AGGRESSION Independent juveniles are occasionally heard The gnatcatcher responds aggressively toward to give a whispered rendition of a highly most vertebrates which venture to within variable sequence of calls, which may be about 20 feet of its nest. If the intruder is a termed a “sub-song.” small, nonpredatory species, the adult gnat- Sibley (field notes) found that the adults catchers attempt to drive it away with bill- tend to feed the same individual fledgling snapping and aerial attacks. Most species several times in succession (table 4). His respond immediately to the attack by fleeing, data, gathered on the day following fledging, in which case the gnatcatchers chase them for show that a brood of five young was fed a a short distance. The following species were total of 105 times in a 3%hour period, or 30 chased away from the vicinity of the nest by times per hour. I found that a brood of four gnatcatchers (number of encounters in paren- fledglings was fed at least 42 times in an hour theses ) : on their fifth day out of the nest. A fledgling cowbird was fed 28 times by a female gnat- Western Fence Lizard, Scdopolws occidentalis (2) Black-chinned Hummingbird, ArchiZochus catcher during a 2Bminute period. alexundri ( 1) The role of the sexes in feeding fledglings Nuttall Woodpecker, Dendrocopos nuttallii ( 1) 24 RICHARD B. ROOT

*Western Flycatcher, Empidonar difficilis ( 1) rels (Sciurus griseus) and several species of *Western Wood Pewee, Contopus sordidulus ( 1) birds approach to join the gnatcatchers in *Plain Titmouse, Parus irmrnatus( 11) Wrentit, Chamueu fasciuta (1) harassing the predator. The gnatcatcher holds House Wren, Troglodytes a&on ( 11) the tail slightly above the horizontal, often Bewick Wren, Thryomanes bewickii (4) partially spread, and wags it from side to side. *Western Bluebird, Sialiu mexicana (2) Gradually they move to within 5 to 10 feet Hutton Vireo, Vireo huttoni (4) Solitary Vireo, Vireo solitaries ( 1) of the predator and begin to fly at its head Orange-crowned Warbler, Vermiuoru celata ( 1) from various directions. In the most intense Black-headed Grosbeak, Pheucticus melrmocephu- encounters, matcatchers hover above the lm (1) predator and dive repeatedly at its head. Fre- Lazuli Bunting, Passerina amoena ( 1) quently the pnatcatchers strike the predator Lesser Goldfinch, Spinus psaltria (4) Lawrence Goldfinch, Spinus lawrencei ( 1) with their beaks. Owls are harassed through- Spotted Towhee, PipGo erythrophthalmus ( 1) out the breeding season, but the most intense Brown Towhee, Pipilo fuscus ( 1) encounters occur when the gnatcatchers have Oregon Junco, Bunco oreganus (4) a nest or fledglings. After a stuffed owl is Chipping Sparrow, SpizeUa pusserinu (2) Merriam Chipmunk, Eutumias merriami (2) removed, the pair continues to give bhew calls and to examine the spot where the owl The asterisk designates those species that are was placed at frequent intervals for a period known to supplant the gnatcatcher away of from 5 to 10 minutes. from its nest. Under natural conditions, the following The above list includes several species species were harassed by gnatcatchers at the whose food habits are similar to those of the Hastings Reservation (number of encounters gnatcatcher, sueqesting that interspecific ag- in parentheses) : gression around the nest might function to rattlesnake, Crotdus sp. ( 1) create some degree of hvperdispersion with Coopers’ Hawk, Accipiter cooperii (4) species that are potential competitors. Also Barn Owl, Tyto alha ( 1) included in the list are species which are Screech Owl, Otus asio ( 1) dominant to the gnatcatchers in encounters Pygmy Owl, Gluucidium gnoma (4) Scrub Jay, Aphelocoma coerulescens (78) that occur at a distance from any nest. The Yellow-billed Magpie. Pica nuttulli (2) pnatcatcher is particularly hostile toward the Western Bluebird fledgling, Sialiu mexicanu ( 1) House Wren, attackina: this species at distances The Scrub Jay, which is a relatively common of over 40 feet from the nest during all stages bird at the Hastings Reservation, is a serious of the nesting cycle. The House Wren is predator on the eggs and young of small birds, known to steal building materials and eggs such as the gnatcatcher, which build open from the nests of other small birds (Bent nests (Bent 1946). Gnatcatchers which do not 1948). The number of encounters with the have nests or dependent young either ignore other species is roughly proportional to their or harass jays only mildly. When gnatcatchers abundance on the study area. are aware of Scrub Jays within a radius of Throughout the breeding season, the gnat- about 150 feet from a nest or brood, the catcher responds to any large predator which adults fly over to begin their attack. The it encounters on the territory by engaging in a harassment display against jays is usually characteristic harassment display. This dis- effective: the jays appear to be obviously play was observed under optimal conditions distracted and often leave the territory within on several occasions when I placed either a a short period. Gnatcatchers have been seen stuffed Screech Owl (Otus mio) or a Homed to follow jays for distances of over 100 yards Owl (Bubo virgin&us) near the nest. and for periods of longer than 10 minutes. When the predator is first recognized from Nestlings crouch in the nest when the adults the nest, the gnatcatcher leaves silently and are harassing a jay nearby, while fledglings flies a short distance away. It then approaches quickly scatter when a jay approaches them. the predator and starts to give a rapid series Juvenile gnatcatchers usually ignore the Scrub of bhew calls. The bhew calls resemble the Jays which they encounter on their own, but peeew calls which are given during territorial they are attracted to the scene where adults disputes, but differ in being more emphatic, are mobbing a predator. in the increased tempo that the calls are de- The harassment display is not always effec- livered, and in the longer duration of the tive. Sibley (field notes) observed six Yellow- series. In apparent response to the bhew billed Magpies ripping apart a gnatcatcher calls, the first gnatcatcher is joined by its mate nest and eating the eggs while the adults dove within a short period. In addition, gray squir- repeatedly at the predators ’ heads. Once the BEHAVIOR AND REPRODUCTION IN THE BLUE-GRAY GNATCATCHER 25

Construction and Egg Laying

1 wms z Feeding Fledglings

I I I I March ’ April May June July ’ August

FIGURE 3. The seasonal distribution of various reproductive activities by six pairs of Blue-gray Gnatcatchers during 1963. nest was destroyed, the gnatcatchers ’ harassing tree was harassed almost continuously for activity subsided quickly. more than two hours by one or the other of A pair of gnatcatchers has been observed the adults. Since bluebirds are not normally to harass a female cowbird during the breed- harassed, this observation suggests that an ing season (Blincoe 1923). In the foothills of intruders’ behavior may be as important as the Sierra Nevada, a pair of gnatcatchers its color or configuration in eliciting the which were feeding nestlings harassed a Log- mobbing response. gerhead Shrike, Lanius ludovicianus (Myers 1907). Coopers’ Hawks and a stuffed Sharp- REPRODUCTIVE SUCCESS shinned Hawk (Accipiter striatus) were har- Of the 12 pairs of gnatcatchers that remained assed when they were discoverd on a perch. on the study area throughout most of the Gnatcatchers responded to accipitrine hawks 1963 breeding season, five were unsuccessful flying nearby, however, by giving strident in raising a brood to fledging age, three suc- alarm calls and fleeing to a dense clump of ceeded in raising one brood, and four SUC- foliage where they remained silent until after ceeded in raising two broods. The seasonal the hawk had left the area. pattern of reproductive effort for six of these A Western Bluebird fledgling which could pairs is presented in figure 3; I found every not fly well enough to flee when a pair of nest of these pairs that persisted for more gnatcatchers tried to drive it from the nest than two days. Two color-banded females 26 RICHARD B. ROOT

build open, elevated nests was 19.3 per cent (10.6-33.3 per cent) in a deciduous scrub biotope in Indiana. Thus the gnatcatchers’ reproductive success seems to fall within the normal range of variation for birds which nest in similar situations. The greatest losses (63.4 per cent) occurred during the egg-laying and incubation stages, while there was only a 26.7 per cent loss during the period when nestlings (both gnatcatchers and cowbirds included) were present. The exact cause for the failure of a nest was known in only a few cases. Nonetheless, in- direct evidence suggests that predation on the eggs was the major mortality factor. Several nests were found where part or all of the clutch was missing and the nest lining was torn up (when gnatcatchers dismantle an old nest, material is always removed from the A B C D outside first). Cowbirds, which were first Stage of Nest Cycle Completed seen at the Hastings Reservation in 1959, also contribute heavily to nest losses. Six of the 22 FIGURE 4. The stages of the nesting cycle success- broods found in this study consisted of a fully completed in 42 nesting attempts during 1963. single cowbird nestling. One brood of four Only nests that were first found during the early gnatcatcher nestlings which hatched eight stages of construction are considered. The stages are defined as follows: A. Construction of the cup com- days previously were found dead in a nest pleted. B. Incubation begun. C. Eggs hatched. D. which was heavily infested with the mesostig- Fledglings left. The black portion of the histograms matic mite Ornithonyssus sylviarum. The refers to nests known to contain young Brown-headed stomachs of these nestlings were filled with Cowbirds. partially digested food, and there was no evi- dence of internal parasites or attack by a predator. were each observed to construct seven nests In late May and early June of 1961, there during a single breeding season. Construc- was a population outbreak of the tent cater- tion of a new nest may begin as early as the pillar, Malucosomu constricta (Lasiocampidae), day following the fledging of the first brood at the Hastings Reservation. The late-instar and in at least one instance, the fledglings of larvae of this moth, which are not eaten by the the first brood were fed by the adults through- gnatcatchers ( Root 1966)) completely de- out most of the incubation period on a second foliated the oaks on a large portion of the clutch. study area. I made a thorough search of the My data on reproductive success (fig. 4) defoliated area between 11 June and 15 June, are presented in terms of the stages of the and was able to find only three gnatcatcher nesting cycle which each nest completes. A nests, all of them in late stages of construction nest was considered to be successful if all or or early incubation. Other pairs on the area only a part of the original clutch succeeded did not seem to be engaged in any reproduc- in completing a stage of development. Orni- tive activity. This situation suggests that the thologists usually calculate reproductive suc- decrease in food supply, which must have ac- cess in terms of the number of eggs which companied the defoliation of the trees, re- successfully complete each stage. Since gnat- sulted in a general failure of the nesting ac- catchers frequently desert their nests and there tivity normally underway at this season (Root is a high rate of nest predation, I followed 1967 ) . the fate of each individual in only those few nests which I could examine without disturb- BEHAVIOR IN WINTER ing the nest site. Gnatcatchers fledged from During the winter gnatcatchers are frequently 24.4 per cent of the nests which reached an seen foraging together in what appear to be advanced stage of construction in 1963 (fig. 4). pairs. On 13 different occasions, involving Nolan (1963) found that the average nest at least seven different “pairs,” two birds re- success, calculated in the same manner as the mained within six yards (sometimes much figure above, of six species of birds which closer) of one another for more than five BEHAVIOR AND REPRODUCTION IN THE BLUE-GRAY GNATCATCHER 27 minutes. Within a 45-minute period, two gnat- In Arizona, the gnatcatcher is frequently catchers moved a distance of over 150 yards seen foraging with Verdins (Auriparus flaui- together, exchanging frequent speee calls as ceps) , Black-tailed Gnatcatchers ( Polioptila they progressed. I frequently noted that the melanura), and Ruby-crowned Kinglets ( Reg- color of the lower dorsum of one bird was ulus calendula) during the winter months. duller in comparison with that of its partner. This association between P. caerulea and P. Both individuals in two such “pairs” were melanura is particularly interesting, since the shot; in each case the gnatcatcher with the two species possess similar morphological dull plumage proved to be a female while the adaptations and foraging repertoires, and brighter specimen was a male. These plumage would therefore be expected to be strong differences are very subtle and are not always competitors for food. These two species were obvious in museum specimens. On 29 January, seen foraging within 10 yards, twice within a gnatcatcher gave a whispered song, which three yards, of one another on eight occasions was indistinguishable from the courtship song, without displaying any aggression toward eight times during a 20-minute period. This each other. In two instances, the birds were bird was alone and had not yet achieved the followed for long enough to determine that males’ characteristic nuptial plumage. The there is obvious cohesion, a tendency for two close association between the sexes during the birds to travel in the same direction together, winter is of interest, since the late arrivals, between the two species. On 20 December, a and possibly the early arrivals, on the breeding pair of P. melanura and a single P. caerulea grounds are not paired. remained within 15 yards of each other while The social bond between wintering gnat- moving over 120 yards together during a catchers is not strict, however. On eight oc- period of 23 minutes. At the end of this ob- casions, single individuals were followed for servation, the male P. melanura began to give more than 15 minutes. Once two birds which a series of five-parted vocalizations and then had been foraging together separated and re- flew about five yards to supplant the P. mained apart (two observers involved) for caerulea from its perch. Soon after this en- at least 35 minutes. counter the two species slowly drifted apart. Gnatcatchers appear to remain on home A male P. melanura, which was previously ranges during the winter. I have been able, silent, was seen to supplant a P. caerulea on with the help of other observers, to follow the one other occasion. movements of the same individuals almost The gnatcatchers which were collected at continuously for as long as three hours in Yuma during the winter had moderate to vegetation dominated by saguaro ( Carnegiea heavy accumulations of fat. The mean body gigantea ) and palo Verde ( Cercidium micro- weight of 10 winter specimens was 5.6 (5.s phyllum) at Sabino Canyon and the San 6.1) g, while the average weight of I3 adults Xavier Reservation, both near Tucson, Ari- which were collected at the Hastings Reserva- zona. I returned to these same spots at dif- tion in July and August was 5.7 (5.4-6.0) g. ferent times of day throughout a period of a week and found that it was possible to cir- DISCUSSION cumscribe the areas within which gnatcatchers TERRITORIALITY occurred. I estimate that the gnatcatchers’ Territoriality is often considered to function home range in this biotope is at least 22 acres, as a form of contest competition (Nicholson which is larger than the normal breeding ter- 1957) giving dominant individuals exclusive ritory at the Hastings Reservation. In an access to a supply of limited requisites which isolated screw-bean (Prosopis pubescens) is sufficient for survival and reproduction. woodland near Yuma, Arizona, at least nine For birds, the supplies of food and suitable gnatcatchers (eight specimens taken) were nest sites are frequently held to be the most found within an area of 22 acres. At this critical requisites. That most gnatcatcher locality, the birds were spaced out, often as territories contained an abundance of nest obvious pairs, and each individual seemed to sites is indicated by the construction of as restrict its movements to an area of about many as seven nests in different locations three acres or less. Territorial defense of the within the boundaries of a small territory dur- winter home range was never observed. On ing a single breeding season. Similarly, the two other occasions, three gnatcatchers for- food supply on the territories seemed to be aged within 15 yards of each other for a few more than adequate. Breeding gnatcatchers minutes without engaging in any aggressive normally made their greatest demands on behavior. the available food supply when the adults 28 RICHARD B. ROOT were feeding young (Root 1967). During dense stands of chaparral and live oaks were this same period, .the adults ’ foraging area utilized alternately by fledglings from the became restricted to a portion of the territory adjacent territories on their first few days surrounding the nest and much of the area away from the nest (Root 1967). Similarly, used earlier was poorly defended. Such con- the adults from as many as three territories tractions in the size of the utilized territory shared the same chaparral stand (which had during the nestling stage have been reported been defended as a single territory in March) for a variety of bird species (e.g., Odum and during their postnuptial molt in August. Kuenzler 1955; Stefanski 1967). The simplest Finally, the “oversized” territories may func- inference suggested by these observations is tion as a reserve against exceptional years that the territory boundaries defended at the when food is temporarilv in short supply. beginning of each nesting attempt usually When the oaks were defoliated by tent cater- enclose a surplus of food and nest sites. This pillars in 1961 (above), the abundance of simple interpretation, however, is static and gnatcatcher prey must have been greatly re- fails to account for important changes in the duced (Root 1966). At this time, the density gnatcatchers’ habitat requirements. of adult gnatcatchers in the vicinity of the Defense during the period of territory study area appeared to be normal but most establishment (March-April) appears to set breeding activity had been halted and much the gnatcatcher density for the remainder of of the foraging was centered in the isolated the breeding season. After late April, no new stands of chaparral and live oaks that had territories were established although pairs escaped defoliation. When the deciduous oaks moved off the study area and were replaced began to develop new foliage, the gnatcatchers by pairs from adjacent areas. Within this rela- were able to resume nesting soon after. tively stable population, the defense of “over- These observations lead me to suggest that sized” territories provided sufficient space for territory size in the gnatcatcher is “ultimately” the gnatcatchers to alter their utilization of related (Lack 1954, 1966) to the supply of habitats in several adaptive ways. limited resources. The apparent surplus of The abundance of food in the immediate such requisites at any one time does not vicinity of the nest becomes progressively negate this interpretation. The gnatcatchers’ more important as the nestlings mature. It utilization of different habitats in the vicinity seems unlikely that the adults could deliver of its territory is flexible, changing in response small arthropods to the older nestlings at the to seasonal and yearly variation in the disper- observed rates (tables 2 and 3) and also sion of limited resources and according to the forage regularly in distant portions of the varied requirements associated with different territory; this may also be true for other small stages in its life history. If the territory size insectivorous species. As a result, nests are was just sufficient to contain the requisites best located where food is extremely abundant. used over a short period, the population could At the Hastings Reservation gnatcatchers ap- become so crowded that many individuals parently responded to a seasonal succession would be restricted to small areas, unable to in the abundance of foliage arthropods in meet their requirements as local conditions different habitats through an almost continu- varied. ous realignment of territory boundaries (Root Compared with other species that defend 1967). Thus in March and early April, chapar- territories of similar size, the gnatcatchers’ ral and live oaks were the preferred foraging territorial behavior seems to be less conspicu- habitats. When the supply of available ous and “ritualized.” The gnatcatchers’ ad- arthropods in these habitats began to decline vertising calls are soft (to the human ear) in late April, there was a gradual shift in and the males rarely display from exposed territory boundaries that resulted in most song perches. As a result, the frequent patrol- pairs having greater access to the adjacent ling observed during territory establishment deciduous oak woodland where food was more may be necessary to hold a large area. Such abundant. The large size of the initial ter- reliance on patrolling is probably related to ritories facilitated this shift by permitting nests the shifting pattern of habitat utilization; as to be relocated in the new optimal habitat pairs become occupied with nests, their on areas that the original defenders visited dominance over little-used portions of the infrequently. territory is reduced by their infrequent visits. The large territories also provided ample These areas can then be occupied with little access to special habitats that were used for opposition by adjacent pairs. only short periods. Thus the same, small, The prominence of territory in the gnat- BEHAVIOR AND REPRODUCTION IN THE BLUE-GRAY GNATCATCHER 29 catchers’ breeding ecology is reflected in the bers of the pair share about equally in con- courtship and predator mobbing behavior. structing the first nest when all of the materials Thus the females ’ activities immediately after must be gathered fresh (none has been cached arriving on the breeding ground suggest that in a previous nest) and adverse weather they are assessing the qualities of the terri- causes frequent delays (fig. 2). The nestlings tories as much as those of ,the males. On all of the first brood are fed mainly by the fe- three occasions when an alien male attempted male while the male continues to defend the to join mobs harassing a predator near the contracted territory. After the first brood nest of another pair, territory defense by the fledges, the female continues to do most of residents took precedence over further ag- the feeding, leaving the male free to build gression toward the predator. This observa- the bulk of the new nest. The male can begin tion requires further verification, perhaps by this nest as early as the day after the first the presentation of models at mobs formed brood fledges. The male then takes a more around predator decoys. nearly equal role in feeding the nestlings of the second brood; this may be correlated with NESTING POTENTIAL the summer decline in the abundance of Pairs of gnatcatchers can complete as many foliage arthropods (Root 1967). The inter- as seven nests in a season (fig. 3). Adapta- changeable roles of the parents is the major tions that increase the number of nesting factor that enables gnatcatchers to care for attempts, and thus the chances of leaving two successive broods at the same time (fig. successful progeny, have obvious advantage 3). Judging from the length of the breeding to a species that experiences such a high level season (fig. 3)) it may be possible sometimes of nest loss (fig. 4). The selective pressures for pairs to raise three successful broods a year involved are reflected in 1963, (an apparently at the Hastings Reservation. normal year) when five of the twelve pairs on the study area were unable to raise a SUMMARY brood to fledging age. The behavior of the Blue-gray Gnatcatcher The gnatcatchers’ prolonged nesting season was observed during four breeding seasons at the Hastings Reservation (late March to at the Hastings Reservation, in coastal Cali- late August) is the principal factor that per- fornia, and on the wintering grounds in south- mits pairs to build several nests. Within this em Arizona. The breeding season data are long season, the reproductive potential is fur- based on short, daily observations on each of ther increased by behavioral adaptations that the pairs nesting in a 56-l-acre study plot serve to compress the time interval between containing oak woodland and chaparral. successive nests. Both sexes are present when the first ter- The re-use of building materials from ritories are established at the Reservation; former nests appears to decrease the time spent the extreme dates for first arrivals are 24 in constructing later nests. As suggested by February and 30 March. Within a season, a Weston (in Bent 1949) and McCabe ( 1963), period of over a month can elapse between the original nest probably serves as a cache the arrival of the first residents and establish- for critical building materials that are difficult ment of the last territory. In the spring, ter- to obtain late in the season. Dead grasses, ritories average 4.6 acres. Territory defense used extensively in the nest cup, may be more is not highly ritualized; the males patrol the difficult, or dangerous, for gnatcatchers to territory boundaries frequently and close en- obtain from the ground after the new grass counters involving vigorous displays and com- cover has achieved its maximum growth in bat are common. May. Similarly, the re-use of building ma- During courtship, females follow the males terials reduces the need to gather such items and inspect potential nest sites where many as spider web which require much time displays are performed. Newly arrived fe- despite their availability throughout the breed- males consort briefly with different males. ing season. Finally, moving nest material per- Both members of the pair build the first mits pairs to establish new nests rapidly when nest although the female does most of the they lay claim to new territory (above) that final arranging of materials. Nest building is is temporarily vacant. curtailed during rainy, overcast weather. As The lack of marked sexual specialization early as the first day after a brood is fledged, in parental and nest building behavior in- the male starts a new nest on which he does creases both the responsiveness and efficiency the bulk of the construction. The nest cup, of the adults ’ breeding activities. Both mem- from its earliest stage of construction, is oma- 30 RICHARD B. ROOT mented with lichens. These lichens make the cover. Thus territory size may be ultimately nest more conspicuous in certain situations. related to the availability of resources. Ter- Spider silk, perhaps because of its adhesive ritoriality is a dominant activity that influences properties, is used extensively in the nest. many other aspects of the gnatcatchers’ be- Pairs usually move the materials from their havior. own abandoned nests to new building sites The re-use of nest material and the inter- that may be as far as 500 feet away. Materials changeable roles of the parents can be inter- are not moved from nests that have been preted as adaptations that increase nesting overrun by the mite Ornithonyssus sylviarum. potential and the populations’ responsiveness The incubation period is 15 days and the to environmental change. nestling period is 12-13 days. During the last ACKNOWLEDGMENTS days in the nest, broods may be fed as often as 43 .times an hour with the most active I am indebted to Frank A. Pitelka for his parent visiting the nest about once every two guidance and encouragement throughout this minutes for extended periods. Fledglings are investigation. William C. Dilger, Peter R. dependent upon the adults for most of their Marler, and Ruth M. Yarrow made helpful food until about the 16th day after leaving comments on the manuscript. I owe a special the nest. While both parents feed young debt to John Davis who, together with the throughout the season, the females assume other residents at the Hastings Reservation, the most active role with early broods, the created a working environment that was both males with fledglings of late broods. pleasant and stimulating. Joe T. Marshall, Jr., Adults chase most small vertebrates (at and his students at the University of Arizona least 22 species) that approach their nest and greatly facilitated my work in the Tucson area. engage in the intense harassment of predators Larry L. Wolf, Laidlaw 0. Williams, and encountered on any part of the territory. Edwin 0. Willis assisted me during certain During 1963, young fledged from 24.4 per phases of the field work. This research was cent of 42 nests that were first found during supported by a National Science Foundation the construction stage. Banded pairs made graduate fellowship and a grant from Francis as many as seven nesting attempts during the S. Hastings to the Museum of Vertebrate breeding season (late March to late August). Zoology. Some pairs were unsuccessful in raising a LITERATURE CITED brood to fledging age while others succeeded in raising two broods. The greatest loss of ARNOLD, G. A., AND M. A. ARNOLD. 1952. The nesting of a pair of Blue Tits. Brit. Birds 14: nests occurred during the egg-laying and in- 175-180. cubation periods. Probably predation by BENT, A. C. 1946. Life histories of North American Scrub Jays is the major factor responsible for jays, crows, and titmice. U. S. Natl. Mus., Bull. nest loss. A heavy infestation of the mite 0. 191. BENT, A. C. 1948. Life histories of North American sylviurum was apparently responsible for the nuthatches, wrens, thrashers, and their allies. death of one entire brood of nestlings. Six U. S. Natl. Mus., Bull. 195. of 22 gnatcatcher broods consisted of a single BENT, A. C. 1949. Life histories of North Amer- Brown-headed Cowbird. During 1961, all ican thrushes, kinglets, and their allies. U. S. nests apparently failed after a population out- Natl. Mus., Bull. 196. BLINCOE, B. J. 1923. Gnatcatchers attacked cow- break of tent caterpillars that defoliated the bird. Bird-Lore 25:253-Z% oaks. CHAMBERLIN, C. 1901. Some architectural traits of During the winter in Arizona, gnatcatchers the Western Gnatcatcher. Condor 3:33-36. seem to remain on home ranges and often EISNER, T., R. ALSOP, AND G. ETTERSHANK. 1964. Ad- hesiveness of spider silk. Science 146:1058-1061. travel as “pairs.” Frequently they forage in HARGRAVE, L. L. 1933. The Western Gnatcatcher close association with the Black-tailed Gnat- also moves its nest. Wilson Bull. 45:30-31. catcher; this is of interest because the two LACK, D. 1954. The natural regulation of animal species have very similar niche requirements numbers. Clarendon Press, Oxford. 343 & viii pp. LACK, D. 1966. Population studies of birds. Clar- and should be strong competitors. endon Press, Oxford. 341 & v pp. Viewed on a short-term basis, gnatcatcher LLOYD, C. K. 1932. The Blue-gray Gnatcatcher territories apparently contain a surplus of moves its nest. Wilson Bull. 44:185. MCCABE, R. A. 1963. Renesting of the Alder Fly- limiting requisites. The probable function of catcher. Proc. XIII Intern. Ornithol. Congr., the “over-sized” territory is to provide suf- Ithaca, p. 319328. ficient space for the birds to alter their pat- MILLER, A. -H., AND R. C. STEBBINS. 1964. The tern of habitat utilization in response to tem- lives of desert animals in Joshua Tree National Monument. Univ. of Calif. Press, Berkeley. 452 poral changes in the dispersion of food and & vi pp. BEHAVIOR AND REPRODUCTION IN THE BLUE-GRAY GNATCATCHER 31

MURIUY, J. J. 1934. The Blue-gray Gnatcatcher ROOT, R. B. 1966. The avian response to a popula- moving its nest. Wilson Bull. 46:128. tion outbreak of the tent caterpillar, Malacosoma MYERS, H. W. 1907. Nesting ways of the Western con&&turn ( Stretch). Pan-Pacific Entomol. 42: Gnatcatcher. Condor 9:48-51. 48-53. NICE, M. M. 1932. Observations on the nesting of ROOT, R. B. 1967. The niche exploitation pattern the Blue-gray Gnatcatcher. Condor 34: 18-22. of the Blue-gray Gnatcatcher. Ecol. Monogr. 37: NICHOLSON, A. J. 1957. The self-adjustment of 317350. populations to change. Cold Springs Harbor ROOT, R. B., AND R. M. YARROW. 1967. A predator- Svmn. Ouant. Biol. 22: 153-173. decoy method for capturing insectivorous birds. NOLAN, V., in. 1963. Reproductive success of birds Auk 84:423-424. in a deciduous scrub habitat. Ecology 44:305- STEFANSKI, R. A. 1967. Utilization of the breeding 313. territory in the Black-capped Chickadee. Condor ODUM, E. P., AND E. J. KUENZLER. 1955. Measure- 69: 259-267. ment of territory and home range size in birds. Auk 72: 128-137. Accepted for publication 23 October 1967.