Volume 6, 1996 British Columbia Birds ISSN 1183-3521

CONTENTS

GUIDELINES FOR AUTHORS 2

BOOK REVIEWS:

THE PATIENT PREDATOR: FORAGING AND POPULATION ECOLOGY OF THE GREATBLUE HERON ARDEA HERODIAS IN BRITISH COLUMBIA, by Robert W. ButlerReviewed by Keith Simpson . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 17

THE ABUNDANCE AND DISTRIBUTION OF ESTUARINE BIRDS IN THE STRAIT OFGEORGIA, BRITISH COLUMBIA, edited by Robert W. Butler and Kees VermeerReviewed by David B. Lank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 18

LAST OF THE CURLEWS, by Fred BodsworthReviewed by Martin K. McNicholl 19

BIRDER'S DICTIONARY, by Randall T. CoxReviewed by Martin K. McNicholl 19

WASHINGTON WILDLIFE VIEWING GUIDE, by Joe La TouretteReviewed by Lloyd Esralson 21

THE VISITORS GUIDE TO THE BIRDS OF THE ROCKY MOUNTAIN NATIONAL PARKS,UNITED STATES AND CANADA, by Roland WauerReviewed by Allen Wiseley 22

ACKNOWLEDGEMENTS AND EDITOR'S COMMENTS 24

DELAYED INCUBATION IN THE BLACK-CAPPED CHICKADEE, POECILE ATRICAPILLUSJo Ann MacKenzie. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 9

THE INFLUENCE OF WIND AND TIME ON CALLING RATE OF THE COMMON POORWILL(PHALAENOPTILUS NUTTALLII): CONSIDERATIONS FOR INVENTORYTrudy A. Chatwin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

FIRST HOODED ORIOLE RECORD FOR WESTERN CANADAAudrey Wallis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

MERLIN (FALCO COLUMBARIUS) PREYS ON FLYING DRAGONFLIESJohn M. Cooper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

MODEL SNAKE ELICITS NEST DEFENCE BEHAVIOUR BY WHITE-BREASTEDNUTHATCHES, SITTA CAROLINENSISJ. E. Bryan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

Volume 6, 1996 British Columbia Birds Page 3

THE INFLUENCE OF WIND AND TIME ON CALLING RATE OF THE COMMON POORWILL (PHALAENOPTILUS NUTTALLII): CONSIDERATIONS FOR INVENTORY

Trudy A. Chatwin British Columbia Ministry of Environment, Lands and Parks

2080 Labieux Road Nanaimo, B.C. V9T 6J9

Abstract -- I collected data on the timing of vocalizations by Common Poorwills at two Okanagan Valley sites during six June 1995 nights to determine whether or not call rates during the twilight period differed from those during the post-twilight period. My results showed that poorwills called more frequently after twilight, in contrast with previous findings in California and locally of greatest vocal activity by this species during crepuscular periods. Detection of poorwill calls decreased significantly when winds were above 6.4 km. per hour (strong breeze). From these findings, I suggest that inventory protocol for Common Poorwills should include recommendations that surveys begin one-half hour before the darkest part of the night (Nautical Twilight) and that surveys not be conducted on windy nights.

Key wo&: Common Poorwill, diurnal vocalization patterns, inventory protocol, Okanagan Valley, Phalaenoptilus nuttallii, wind effects.

The Common Poorwill (Phalaenoptilus nuttallii) is a cryptic, nocturnal insectivorous bird that is considered a rare or uncommon resident of high rolling prairies, semi-arid flats and rocky foothills in North America (Csada and Brigham 1992a). In Canada, its known breeding range is restricted to the arid southern interior north to Riske Creek in British Columbia and the Cypress Hills of Alberta and Saskatchewan, areas that continue to undergo habitat alteration from agriculture, urbanization and logging. In spite of scant information on population trends of poorwills in Canada (Csada and Brigham 1992b), this species was "downlisted" in British Columbia from a "red list" of candidate species for legal designation as "endangered" or "threatened" to a "yellow list" of species of regional concern by the British Columbia Ministry of Environment in 1995 (British Columbia Wildlife Branch and Habitat Protection Branch 1995).

Recent interest in determining the status and population trends of poorwills and the desire for standard, effective inventory methodologies led to the development of a preliminary manual for sampling goatsucker populations in British Columbia (Bender and Brigham 1995). This inventory manual recommends that surveying for poorwills should begin at civil twilight (when the sun is about 6' below the horizon) and end at nautical twilight (when the sun is 12' below the horizon) at points 400 to 500 m. apart. Detection of the distinctive l)oor-will" call, uttered primarily by territorial males, is used to provide an index of the relative abundance of the birds and to estimate the population in the area censused. Brigham and Barclay (1992) recommend recording moonlight conditions, temperature, wind speed and cloud cover so that their effects on calling activity can be considered.

Although few direct observations of foraging poorwills have been published (Bent 1940; Cannings et al. 1987; Brigham and Barclay 1992), these birds apparently usually

catch flying insects during brief sallies from a low perch or the ground. Radio-telemetry studies (Brigham and Barclay 1992) demonstrated that poorwills in the Okanagan begin feeding approximately 30 minutes after sunset and continue as long as there is sufficient light. The duration of feeding after nautical twilight depends on the amount of moonlight (Brigham and Barclay 1992). Since both sexes incubate eggs and share in brooding, and incubation shifts occur throughout the night (Csada and Brigham 1992a), energy constraints on both sexes should be similar.

Calling by poorwills likely functions as song, repelling other males and attracting potential mates (Csada and Brigham 1992a). From an energetic perspective, during courtship, incubation and brood-rearing, male poorwills could theoretically face a choice between spending time foraging during optimal conditions of visibility, and spending energy calling to defend their territories. However, diurnal calling patterns tend to coincide at least generally with temporal patterns of foraging (Brauner 1952; Brigham and Barclay 1992).

In this study, I collected data on calling rates by male poorwills during the period between civil twilight and nautical twilight and during the darker part of the night. I also collected data on the effect of windspeed on calling rates of poorwills.

MATERIALS AND METHODS

Study Site

The study was conducted in the Okanagan Valley near Oliver, British Columbia (49'14'40"N, 1 19'35'20"W, where poorwills are fairly common (Campbell et al. 1990). Data were collected simultaneously at two sites (Suzi Mine and Geology Camp) located about 1 km. apart nightly from 22-28

Page 4 British Columbia Birds Volume 6,1996

June 1995, except on ,25 and 26 June, when data were collected at only one (Geology Camp) site. During these nights, the moon was not above the horizon, and therefore could not affect calling or foraging rates.

Data Collection and Analysis

At each site, one or two observer(s) from a "Bats and Nocturnal Birds" course recorded the following data during each ten-minute sampling period between 21:00 and 00:30:

1) Number of individual poorwills calling, based on directions of calls.

2) Total number of poorwill calls. 3) Windspeed on the scale: 0 = 0 km./hour (still), 1 = 1.6-4.8

km./hour (light breeze), 2 = 4.8-614 kmhour (moderate breeze), 2.5 = 6.4-11.2 kmhour (strong breeze), and 3 = greater than 11.2 km./hour (windy). Windspeeds were estimated by using the Beaufort scale (Longstreth 1953).

We excluded those 10-minute periods before which poorwills were heard calling from analysis, but included those from all periods after the first poorwill called. For each 10- minute period, calls per bird were calculated by dividing the total number of calls counted by the number of birds heard calling. Time periods were classified as either twilight (when the sun is 12' or less below the horizon, between 21:30 and 23:OO as calculated for the Oliver area by the Dominion Astronomical Observatory, Victoria) or after nautical twilight (between 23:OO and 00:30).

A two-way analysis of variance (ANOVA) (e.g., Zar 1974) was used on the entire data set to test the effects of wind and time on the number of calls per poorwill. As the number of observation periods at higher windspeeds (2.5 or 3) was very small (1 3), and higher windspeeds generally reduce observers' abilities to hear birds (e.g. Robbins 1981), I eliminated these data from the data set and calculated a second two-way ANOVA to assess the effects of time and windspeed on calls per poorwill at windspeeds of less than 2.5.

RESULTS

Over the six nights, data were collected from a total of 105 10-minute observation periods. The mean number of calls per bird was 70.9f4.9 during twilight, whereas the mean number of calls per bird after twilight was 110.8f-9.0.

When all categories of windspeed were included, fewer calls per bird were detected at windspeeds of 2.5 and 3 than at other windspeeds (p<0.01; F=3.6; d.f.=4: Table I), confirming an impression I had earlier from experience in the Chilcotin area of British Columbia that strong winds reduced observer ability to detect poorwill calls. When all wind speeds were included, calls detected per bird did not differ at different times of night (p>0.05; F=2.3; d.f.=l: Table 2).

When data for windspeeds 2.5 and 3 were removed from the analysis, and the effects of the remaining three wind speeds and time of night were reanalysed (Table 3). no effect of wind on calling rate was indicated (p>0.05). However, the mean calling rate was greater after twilight (1 16.0 calls per bird per 10-minute period f 10; n=97), than during twilight (72.7 calls per bird per 10-minute period f 5; n=lOO)(p<O.Ol). Therefore, amount of calling differed significantly at different times of the night during the six-day observation period. This analysis showed no significant interaction effect between windspeed and time of night.

TABLE 1

MEAN NUMBER OF CALLS PER BIRD AT VARIOUS WIND SPEED CATEGORIES

Wind Speed Mean No. of Category' n2 CallsBird S.E.3

0 94 107.9 8.82 1 56 76.7 8.19 2 47 87.0 10.83

2.5 7 17.6 9.22 3 6 57.3 21.91

See Data Collection section of text for definitions. n =sample size. S.E. =standard error.

TABLE 2

COMPARISON OF WIND AND TIME ON POORWILL CALLING RATE, USING ALL DATA

Factor DF' F-Ratio2 P3

Wind 4 3.6 0.008 Time 1 2.3 0.133 Wind*Time 4 0.9 0.452

DF =degrees of freedom F =a test statistic P =Probability

TABLE 3

EFFECTS OF WIND AND TIME ON POORWILL CALLING RATE, USING WINDSPEED

CATEGORIES O , 1 , AND 2 ONLY

Factor DF' F-RatioZ P3

Wind 2 2.7 0.071 Time 1 10.6 0.001 Wind*Time 2 1.5 0.216 ' DF =degrees of freedom

F =a test statistic P =Probability

Volume 6, 1996 British Columbia Birds Page 5

DISCUSSION

My data indicated that overall calling by poorwills at two sites in the Okanagan over the six-day period in the third week of June was significantly greater after twilight than during twilight. These data contrast with previous observations of Brauner (1952) in California and of Brigham and Barclay (1992) locally, whose "impression was that the vast majority of vocal behaviour occurred in crepuscular periods."

Our data also contrast with the general pattern of maximum vocalization during crepuscular periods in other sallying caprimulgids (e.g. Wynne-Edwards 1930; Mengel and Jenkinson 1971; Mills 1986). Further studies are needed during different times of the breeding cycle and incorporating the pre-dawn calling period to see whether my findings were unique to a specific time andlor area or apply over a broader area and/or season. Our observations were undertaken during the incubation period of this species in the Okanagan (Cannings et al. 1987), when calling may be less than during courtship and territorial establishment, and may also differ from calling patterns during brood-rearing and after fledging. Since females at least occasionally call (Brauner 1952; Csada and Brigham 1992a), and both sexes incubate (Csada and Brigham 1992a), with periods when the eggs are left uncovered (Csada and Brigham 1992a), the apparent increase in number of calls during post-twilight in this study could represent periods when both birds are off the nest. Further studies of incubation patterns are needed to rule out or verify this and other possibilities.

In contrast to Bender's (1994) recommendation that inventories should begin one-half hour after sunset or civil twilight, my data suggest that point counts to inventory poorwill populations should begin one-half hour before nautical twilight and continue into the post-twilight period. In practice, this may involve dividing point count surveys for Caprimulgidae (Common Nighthawk, Chordeiles minor, and Common Poorwill) into separate surveys, where their breeding ranges overlap. Further observations during May and early July are needed to confirm or refute these recommendations on timing of surveys.

My results confirm the importance of excluding observations for calling poorwills during windy periods. Bender and Brigham (1995) recommend that inventory of both poorwills and Common Nighthawks not take place on nights with wind greater than about 8 km. per hour. I concur that wind seems to either decrease hearing of calls, or that poorwill calling rate decreases during winds of 6.4 km. per hour or greater.

ACKNOWLEDGEMENTS

I would like to thank fellow class mates in the 1995 Bats and Nocturnal Birds Field Course from the University of

Calgary, University of Regina and York University for assistance in collection of data. Ken Wright assisted with data entry. Risa Smith provided invaluable assistance with the SYSTAT data analysis program and statistics. The project was ably directly by R. M. Brigham. Alexander M. Mills and Ken H. Morgan provided valuable comments on the manuscript. However, most of all, the tireless and thorough editing of Martin K. McNicholl brought this paper to publication.

LITERATURE CITED

Bender, D. J. 1994. The effect of moonlight on nocturnal activity by the Common Poorwill (Phalaenoptilus nuttallii). Bachelor of Science Honours thesis, University of Regina, Regina.

Bender, D. J. and R. M. Brigham. 1995. Preliminary inventory manual for sampling goatsuckers (Caprimulgidae) in British Columbia. British Columbia Resources Inventory Committee, [Victoria].

Bent, A. C. 1940. Life histories of North American cuckoos, goatsuckers, hummingbirds and their allies. U. S. National Museum Bulletin 176.

Brauner, J. 1952. Reactions of Poor-wills to light and temperature. Condor 54: 152-1 59.

Brigham, R. M. and R. M. R. Barclay. 1992. Lunar influences on foraging and nesting activity of Common Poorwills (Phalaenoptilus nuttallii). Auk 109:315-320.

British Columbia Wildlife Branch and Habitat Protection Branch.1995. Amphibians, reptiles, birds and mammals not at risk in British Columbia. British Columbia Wildlife Branch and Habitat Protection Branch, Victoria Wildlife Bulletin No. B-74.

Campbell, R. W., N. K. Dawe, I. McTaggart-Cowan, J. M. Cooper, G. W. Kaiser and M. C. E. McNall. 1990. The birds of British Columbia. Volume 2. Royal British Columbia Museum, Victoria.

Cannings, R. A., R. J. Cannings and S. G. Cannings. 1987. Birds of the Okanagan Valley, British Columbia. Royal British Columbia Museum. Victoria.

Csada, R. D. and R. M. Brigham. 1992a. Common Poorwill (Phalaenoptilus nuttallii) . No. 32 in A. Poole, P. Stettenheim and F. Gill (Editors). The birds of North America. Academy of Natural Sciences, Philadelphia and American Ornithologists' Union, Washington, D.C.

Csada, R. D. and R. M. Brigham. 1992b. Status report for the Common Poorwill (Phalaenoptilus nuttallii: Caprimulgidae). Committee on the Status of Endangered

Page 6 British Columbia Birds Volume 6,1996

Wildlife in Canada (COSEWIC), Ottawa.

Longstreth, T. M. 1953. Understanding the weather. MacMillan, New York (reprinted by Collier, New York, 1962).

Mengel, R. M. and M. A. Jenkinson. 1971. Vocalizations of the Chuck-will's-widow and some related behavior. Living Bird 10:171-184.

Mills, A. M. 1986. The influence of moonlight on the behavior of goatsuckers (Caprimulgidae). Auk 103:370-378.

Robbins, C. S. 1981. Bird activity levels related to weather. pp. 301-310 in C. J. Ralph and J. M. Scott (Editors). Estimating numbers of terrestrial birds. Studies in Avian Biology No. 6.

Wyme-Edwards, V. C. 1930. On the waking-time of the Nightjar (Caprimulgus e. europaeus). Journal of Experimental Biology 7:241-247.

Zar, J. H. 1974. Biostatistical analysis. Prentice-Hall. Englewood Cliffs, New Jersey.

Common Poorwill. Photo by Mark Nyhof

Volume 6, 1996 British Columbia Birds Page 7

FIRST HOODED ORIOLE RECORD FOR WESTERN CANADA

Audrey Wallis 2072 Weiler Avenue

Sidney, B.C. V8L 1R4

Abstract -- I document the first British Columbian and second Canadian record of Hooded Oriole.

Key work: British Columbia, Icterus cucullatus, Hooded Oriole.

In the early evening of 6 May 1996, 1,noticed a bright yellow bird leave a hummingbird feeder and fly across the back yard at 2072 Weiler Avenue in Sidney, British Columbia. It flew to another hummingbird feeder which was in a more protected position in a mountain-ash (Sorbus sp.) tree. Light conditions were very good; clear and sunny. My husband, Bob Wallis, and I observed the bird through 7x50 binoculars.

The bird was blackbird-sized, with a bright yellow body, shading to a very subtle orangy tint on its head. The lack of red colouration on the head precluded a male Western Tanager (Piranga ludoviciana). The tail was completely black, lacking the yellow edges of adult male Bullock's Oriole (Icterus bullockii). The bib was also completely black and extended from the upper breast to the beak and eye. The black wings exhibited white wing-bars. These features fit that of an adult male Hooded Oriole 0. cucullatus) (Robbins et al. 1966).

The oriole moved through two small maples (Acer sp.) and a sumac (Rhus sp.), which are in close proximity to the mountain-ash that contains the feeder, and settled for up to three minutes on the feeder.

At 07:30 on 7 May 1996, the bird was observed again by us and birder Barbara R. Begg, who was able to confirm his call. The Hooded Oriole stayed throughout the daylight hours,

when over 30 birders, including Bryan Gates, David Stirling and Bmce Whittington, observed him moving through the trees and to the feeder. The weather changed frequently from showers to sunshine. The bird was in close proximity most of the time, being viewed from a window at a distance of 15 ft. (about 5 m.). Several photographs taken helped confirm identification, but were not suitable for identification. A drawing of the bird appears in Figure 1.

Colin Knecht, President of the Vancouver Island Cage Bird Society, has given assurances that to the best of his knowledge there is no record of Hooded Orioles as cage birds. Therefore, we can assume that this sighting is not of an escaped bird.

The normal range of Hooded Oriole along the western coast of North America extends north to northern California, but it has occurred in west-central Oregon (American Ornithologists' Union 1983). The only previous accepted Canadian record occurred in 1992 at Long Point, Ontario, where it was netted, banded and photographed (Boardman 1992; Bain 1993; see also colour photograph in Anonymous 1992). The Sidney occurrence has been accepted by the British Columbia Field Ornithologists Bird Records Committee (G. S. Davidson personal communication to M. K. McNicholl 13 November 1996). It thus represents the first documented record of this species in British Columbia, and the second in Canada.

- - - - - - - - - - -

Figure 1. Drawing of male Hooded Oriole at Sidney. British Columbia. 6-7 May 1996.

Page 8 British Columbia Birds Volume 6,1996

[Editor's note: Subsequent to this record, there have been four more reports of possible or confirmed Hooded Orioles in British Columbia. An unconfirmed bird was reported in a yard near Mundy Park, Coquitlam on 25 May 1996 (K. Elliott. 1997. Wandering Tattler 21 (1):s-12). It was considered "likely the same bird" as the subject of this note (J. Bowling. 1996. National Audubon Society Field Notes 50:321-327). Another Hooded Oriole (sex not stated) spent 19 July 1997 in a backyard at Esquimalt Lagoon (M. Bain and M. Holder. 1997. Birders Journal 6:159-161, 164-176). A male was reportedly observed at Blackie Spit, Crescent Beach on 19 September 1997 (M. Bain and M. Holder. 1997. Birders Journal 6:210-228). Much observed and much photographed was a male that appeared at a feeder in Terrace on 19 November 1997 and stayed much of the winter of 1997-1998 (Figure 2; C. Thomas and G. Thomas. 1998. BirdersJoumal 7:6-7). -M.K.M.]

ACKNOWLEDGEMENTS

Comments on an earlier draft by Gary S. Davidson and Mary J. Taitt helped improve the manuscript.

LITERATURE CITED

American Ornithologists' Union. 1983. Check-list of North America birds, sixth edition. Allen Press, Lawrence, Kansas.

Anonymous. 1992. Pictorial highlights spring 1992. American Birds 46:498-501.

Bain, M. 1993. Ontario Birds Records Committee report for 1992. Ontario Birds 11:46-63.

Boardman, R. 1992. Hooded Oriole at Long Point [,I Ontariola potential first for Canada. Birders Journal 1 :228-229.

Robbins, C. S., B. Bruun and H. S. Zim. 1966. A guide to field identificationhirds of North America. Golden Press. New York.

Figure 2. Male Hooded Oriole at Terrace, British Columbia, 1 February 1998. Photo by Jo Ann MacKenzie.

Volume 6,1996 British Columbia Birds Page 9

DELAYED INCUBATION IN THE BLACK-CAPPED CHICKADEE (POECZLE A TMCA PILL US)

Jo Ann MacKenzie 15341-21 Avenue

Surrey, B.C. V4A 6A8

Abstract -- A pair of Black-capped Chickadees undertook nesting in early July, late in the local breeding season of this species, in a suburban garden in Surrey, southwestern British Columbia. Two eggs hatched after a delayed incubation of 17-20 days, four to seven days longer than normal. Delayed incubation is discussed. The two young fledged during the afternoon, an unusual time of day.

Key words: Black-capped Chickadee, delayed incubation, Poecile atricapillus, southwestern British Columbia.

The Black-capped Chickadee (Poecile atricapillus) nests in a cavity dug by the pair in a rotten tree stub, but will also use abandoned woodpecker holes, natural cavities, or nesting boxes. Both sexes excavate a cavity, but the female constructs the nest, normally in three to four days (Bent 1946), with known extremes of two days to two weeks (Smith 1993). Upon completion of the nest, she may pause for a day or so before commencing egg-laying. She then lays one egg per day, usually early in the morning, until the clutch is complete (Odum 1941; Kluyver 1961). Clutch size in British Columbia ranges from one to nine eggs, usually five to seven (Campbell et al. 1997). Incubation is solely by the female, although some males develop a partial brood patch (Odum 1941; Kluyver 1961; Smith 1991, 1993). Incubation usually lasts 12-13 days, possibly rarely as short as 11 days (Peck and James 1987). The young leave the nest after another 16-18 days, usually early in the day (Odum 1941). The purpose of this note is to describe an instance of delayed (i.e., longer than usual) incubation in Black-capped Chickadees.

OBSERVATIONS

In mid-May 1995, I observed a pair of Black-capped Chickadees inspecting nesting boxes in a moderately well- treed suburban garden in Surrey, southwestern British Columbia (49' 3'N, 122' 48'W). On 22 May, a female was noted carrying moss into one of the boxes. Three eggs were in the nest on 9 June. The outcome of this nesting is not known.

On 2 July, two eggs were felt lying deep in the nest material in the same box. Two adults were observed entering and leaving the nest box on 13 July. One remained in the box 20 minutes before departing, behaviour indicating that incubation was probably taking place (Odum 1941; Smith 1993).

Both adults were delivering food to the box and removing fecal sacs on 27 July. On 1 August, one well-feathered nestling crouched motionless in a depression in the nesting material immediately adjacent to the open side. I could not see farther into the nest chamber, and I did not reach in to

ascertain the presence of additional nestlings, as the intrusion might cause the visible nestling to burst out prematurely (Smith 1993). On 5 August, voices of two nestlings could be distinguished giving "dee" calls, one loudly enough to be heard 8 m. away. During an observation period of several minutes, one adult foraged in a honeysuckle vine (Lonicem periclymenum) 3 m. from the box, and delivered unidentified food at 10-second intervals.

On 6 August, a cold (12' recorded in the yard of the nest- site), windy, wet (19 mm. of rain in Greater Vancouver area) day (Climate Information, Environment Canada, Vancouver, personal communication), one nestling was observed at the box entrance, putting its head out and calling loudly. No adults were noted during the observation period from 11:OO to 12:00, nor at any time that morning. "Looking out" behaviour of the nestlings indicates that fledging will take place within "a few days" (Smith 1991,1993). The nestlings were still in the box at 13:OO. No more observations were made until 15:00, when I noted that the young had fledged, later in the day than normal (Odum 1941; Smith 1993). Two chick voices were heard calling loudly from dense trees 11-12 m. from the nest box, but they were not seen. The next morning, no voices were heard.

A close inspection of the well-constructed nest on 13 August revealed one unhatched intact egg remaining near the bottom of the nest material.

Approximate hatching date($ of the two eggs that produced young can be calculated by backdating from the known fledging date of 6 August. A normal nestling period of 16-18 days indicates that the chicks must have hatched between 19 and 21 July. Since incubation normally begins with the penultimate egg in this species (Smith 1991, 1993), incubation should have begun on 1 or 2 July, depending on whether or not the third egg was already present when the nest was checked on 2 July. This gives an incubation period of 17-20 days, indicating that incubation was delayed or prolonged 4 to 7 days beyond the normal 12-13 days. Kluyver (1961) observed a prolonged incubation period of 15 days in a Black-capped Chickadee nest, and Peck and James (1987)

Page 10 British Columbia Birds Volume 6,1996

reported an instance of a 17Lday incubation period at one nest. Her brood patch may not have been fully in contact with the

eggs.

DISCUSSION

I do not know whether this was a true second brood (rare in Black-capped Chickadees -Smith 1993) or simply a replacement clutch. The early July laying date is near the late extreme of 12 July known for British Columbia (Campbell et al. 1997). There was no evidence in or,near the nest box of additional eggs that may have been predated or carried off.

The extended incubation period could have been due to a single period of egg neglect, faulty incubation behaviour or a combination of these factors. The cause of the incubation delay may have been due to the female's brood patch not being fully developed, resulting in insufficient heat transfer to the eggs. Chickadees often cover their eggs with nest material before leaving the nest, both during the laying period and sometimes during incubation (Odum 1941; Smith 1991). Further, incubating females are quite active on the nest, "frequently fidgeting with the eggs" (Perrins 1979; Smith 1991). The third egg, if present on 2 July, may have been worked so deeply into the nest that sufficient warmth did not reach it, or it may have been infertile, possibly left over from the earlier nesting.

In Kluyver's (1961) case of a 15-day incubation period, the nest box was unusually large (70 cm.' bottom area), and the nest loose. The nest cup enlarged during incubation, allowing the eggs to sink and scatter, precluding normal incubation. Three of the seven eggs hatched, but two days longer than usual. Our box presented no such problem.

Haftorn (1988) showed that under certain adverse weather conditions, an incubating female may desert her clutch for several hours at a stretch, and suggested that this is normal behaviour in small passerines. During the incubation period in question (about 1-21 July 1995), the mean temperature was 2' above normal (as published in Peace Arch News); there was no precipitation (Climate Information, Environment Canada, Vancouver, personal communication). Consequently, weather was not a factor.

The incubation rate varies with the heat supplied by the incubating bird, and may be suspended for some time without injury to the embryo (Wing 1956). Delayed incubation is known to occur among species that forage on unpredictable resources long distances from their breeding colonies and have long incubation periods, particularly smaller Procellariiformes. In the Fork-tailed Storm-Petrel (Oceanodroma furcata), egg neglect is dramatic; the incubation period is between 37 and 68 days, with some embryos surviving up to seven continuous days of neglect and 23 days of intermittent neglect (Boersma et al. 1980; Simons 1981; Campbell et al. 1990; Astheimer 1991). The Cassin's

Auklet (Ptychommphus aleuticus) has an average incubation period of 37.8 days, with a range of 37-42 days (Manuwal 1974; Campbell et al. 1990).

Delayed incubation of 42 days has also been documented for the Northern Goshawk Oiccipitergentilis) (Wallace and Mahan 1975), instead of the normal 35-38 days (Cramp 1980). Incubation in the Northern Bobwhite (Colinus virginianus) may be suspended for three weeks or more (Wing 1956).

Hatching time may vary considerably within a species, according to parental incubation, temperature of the nest or temperature of individual eggs within the nest (Ricklefs 1993), or by some combination of factors (Webb 1987). Some British tits delay incubation until several days after the clutch has been completed (Perrins 1979), especially in an early- season nesting (Gibb 1950). Ordinarily, in the Black-capped Chickadee, incubation begins with the laying of the penultimate egg in the clutch, but in a late-season nesting, incubation may begin some days earlier (Smith 1991, 1993).

During incubation, an egg loses weight due to loss of water vapour by diffusion through the shell (Rahn and Ar 1974). If incubation is delayed or prolonged too long, the embryo succumbs to dessication. The likelihood of this is much reduced in cavity nesters due to the insulative value of the nest chamber andnest material. The chickadees' egg-covering behaviour when leaving the nest during laying and sometimes during incubation would further retain nest humidity, retarding water loss, and making it possible for the embryo to survive a 4-7 day delayed incubation.

ACKNOWLEDGEMENTS

I thank Hue MacKenzie for assistance with the observations and R. Wayne Campbell for guidance and a helpful manuscript review of an earlier version of this manuscript. For further review and valuable comments, I thank Mary J. Taitt and Andr6 Desrochers. I especially appreciate the detailed comments and encouragement of Martin K. McNicholl. I also thank the Climate Information staff of Environment Canada for weather data.

LITERATURE CITED

Astheimer, L. B. 1991. Embryo metabolism and egg neglect in Cassin's Auklets. Condor 93:486-495.

Bent, A. C. 1946. Life histories of North American jays, crows, and titmice. US. National Museum Bulletin 191.

Boersma, P. D., N. T. Wheelright, M. K. Nerine and E. S. Wheelright. 1980. The breeding biology of the Fork-tailed Storm-Petrel (Oceanodmmafurcata). Auk 97:268-336.

Volume 6,1996 British Columbia Birds Page 11

Campbell, R. W., N. K. Dawe, I. McTaggart-Cowan, J. M. Cooper, G. W. Kaiser, and M. C. E. McNall. 1990. The birds of British Columbia. Volumes 1 and 2. Royal British Columbia Museum. Victoria.

Campbell, R. W., N. K. Dawe, I. McTaggart-Cowan, J. M. Cooper, G. W. Kaiser, M. C. E. McNall and G. E. J. Smith. 1997. The birds of British Columbia. Volume 3. University of British Columbia Press, Vancouver.

Cramp, S. (Editor). 1980. Handbook of the birds of Europe, the Middle East and North Africa. The birds of the western Palearctic. Volume 2. Hawks to bustards. Oxford University Press, Oxford.

Gibb, J. 1950. The breeding biology of the Great and Blue titmice. Ibis 92:507-539.

Haftorn, S. 1988. Incubating female passerines do not let the egg temperature fall below the 'physiological zero temperature' during their absences from the nest. Omis Scandinavica l9:97-llO.

Kluyver, H. N. 1961. Food consumption in relation to habitat in breeding chickadees. Auk 78:532-550.

Manuwal, D. A. 1974. The natural history of Cassin's Auklet (P~choramphus aleuticus). Condor 76:421-431.

Odum, E. P. 1941. Annual cycle of the Black-capped Chickadee -2. Auk 58:518-535.

Perrins, C. M. 1979. British tits. Collins, Glasgow.

Rahn, H. and A. Ar. 1974. The avian egg: incubation time and water loss. Condor 76:147-152.

Ricklefs, R. E. 1993. Sibling competition, hatching asynchrony, incubation period, and lifespan in altricial birds. Chapter 5, pp. 199-276 in D. M. Power (Editor). Current ornithology. Volume 11. Plenum Press, New York.

Simons, T. R. 1981. Behavior and attendance patterns of the Fork-tailed Storm-Petrel. Auk 98:145-158.

Smith, S. M. 1991. The Black-capped Chickadee: behavioral ecology and natural history. Cornell University Press, Ithaca, New York.

Smith, S. M. 1993. Black-capped Chickadee (Parus atricapillus). No. 39 in A. Poole. P. Stettenheim and F. Gill (Editors). The birds of North America. Academy of Natural Sciences, Philadelphia and American Ornithologists' Union, Washington, D.C.

Wallace, G. J. and H. D. Mahan. 1975. An introduction to ornithology. Macmillan, New York.

Webb, D. R. 1987. Thermal tolerance of avian embryos: a review. Condor 89:874-898.

Wing, L. W. 1956. Natural history of birds, a guide to ornithology. Ronald Press, New York.

Peck, G. K. and R. D. James. 1987. Breeding birds of Ontario. Nidiology and distribution. Volume 2: passerines. Royal Ontario Museum Life Sciences Miscellaneous Publication, Toronto.

Page 12 British Columbia Birds Volume 6,1996

MODEL SNAKE ELICITS NEST DEFENCE BEHAVIOUR BY WHITE-BREASTED NUTHATCHES, SITTA CAROLINENSIS

J. E. Bryan' Department of Psychology

Simon Eraser University Burnaby, B.C. V5A 1SG

Abstract -- Seven nuthatches of four pairs studied at their nests inspected models of snakes placed nearby without otherwise reacting to them initially. Six of these seven then flew away, gave alarm calls, andlor performed threat displays to the snake model. Threat displays included TaiLFan-Back-Ruffle, Wing Spread andlor Sway.

Key Words: Displays, Nest Defence Behaviour, Okanagan, Reactions to Snakes, Sitta carolinensis, White-breasted Nuthatch.

Many bird species use defensive behaviour to protect eggs and young from snakes (e.g. Nolan 1959; Jackson 1974; Gottfried 1979; James et al. 1983; Brown and Brown 1987; Walters 1990; Winkler 1991, 1992). White-breasted Nuthatches W t a carolinensis) direct threat displays towards some other birds and mammals (Kilham 1968; Stokes and Stokes 1983; Bancroft 1987), but such behaviour has apparently seldom been observed in response to snakes. An observation of White-breasted and Pygmy nuthatches (S. pygmaea) inspecting a live snake near a pond near Penticton, British Columbia led me to conduct experiments that indicated that nuthatches were more likely to inspect snake models than other novel stimuli, suggesting that they recognized snakes as significant environmental factors (Bryan 1998). Although none of these nuthatches exhibited threat displays towards model snakes away from the vicinity of nests (Bryan 1998), they might be expected to respond more strongly to snakes in the vicinity of nests if snakes were recognized as a threat. The purpose of this note is to document tests of this expectation with model snakes placed near nest holes and boxes.

METHODS

I studied four pairs of nesting White-breasted Nuthatches during May and June 1993 and 1995 in Ponderosa or Lodgepole pine (Pinusponderosa or P. contorts) forests of the Okanagan Valley near Penticton, B. C. (49' 30'N, 119' 38'W). To test the response of nuthatches to snakes at their nests, models were placed in the vicinity of the nests after the birds left to forage. The snake model was placed on the ground about 50 cm. from a nest in a hollow tree stump with its hole at ground level, and on top of the nest boxes used by the other three pairs. Models were left in place until reactions to them by returning birds were observed and documented, usually from both parents (4-60 minutes). To record behaviour of the returning birds, I sat about 15 m. from the nests in plain view of the nest site. My observations were

' current address: B. C. Environment, 3547 Skaha Lake Road, Penticton, B.C. V2A 7K2

Figure 1. White-breasted Nuthatch performing Sway display. The realistic model snake is shown above the nest box.

recorded with pencil and paper while watching the birds. After the initial observations were recorded, the behaviour of the third nesting pair was also photographed (Figure 1) and videotaped.

In my previous experiment, I found that a wide range of snake models elicited inspections (Bryan 1998). In three of the present experiments, I used a model like the most realistic of these, a 120 cm. rubber toy snake with brown spots on a tan background coiled in an approximate 16 cm. diameter circle (Figure 1). At these nests, the behaviour of returning birds was observed and recorded only on their first trip back to the nest after the snake model had been placed nearby.

Volume 6,1996 British Columbia Birds Page 13

At one of the nest boxes, I performed a slightly different experiment by first placing out a cruder model (a black-and- white striped rope) (Bryan 1998: Figure 1) and observing the reaction of the birds to this model on their initial return. When they left again, I replaced this crude model with the more realistic one, and observed their behaviour on their next two return trips to the nest.

My operational definition of "inspection" is that near a model the bird landed or stopped hopping down the tree and usually turned its head in the direction of the model. I used the terminology of Stokes and Stokes (1983) fot two basic threat displays, Tail-Fan-Back-Ruffle (TFBR) and Wing- Spread (WS) observed, and hereunder describe a variant of WS as Sway. In TFBR, "tail may be fanned, back feathers may be ruffled, and head may point down" (Stokes and Stokes 1983:133). In WS, "body and bill are pointed vertically, wings and tail are fully spread, and bird sways slowly from side to side. It may last only one or two seconds or it may be longer" (Stokes and Stokes 1983:133).

RESULTS

Both members of the pair that nested in the hollow stump arrived back simultaneously. Both inspected the model, although these inspections were not independent, as the birds could see each other. One bird then flew away, uttering an alarm call in flight for 1.5-2 minutes. Its mate followed.

One of the box-nesting pairs was tested with the more realistic model about 1.5 hours before sunset. The female returned with food, inspected the model from several perches above the model, and then began to utter a distress call. She continued calling for about 5 minutes, attracting a sparrow (sp.?) that flew over to the nest box and inspected the model. Eight minutes after returning to the nest site, the female nuthatch began TFBR and Sway (Figure 1) displays on the tree trunk and branches above the nest box. In Sway, the bird's wings are spread out from the body at various angles and heights (but not fully extended vertically as in WS). The bird's tail is spread as in TFBR, but the wings are extended farther away from the body. She moved up and down the tree trunk, but not past the snake model. She began another bout of distress calling, followed by the threat displays. I removed the model after she had displayed and called for a total of 35 minutes without going past the model into the nest hole. Although he had been returning to the nest with food for periods of 1-5 minutes just prior to the test, her mate did not return during the experimental period. Observations ceased 10 minutes after the snake was removed, even though the female had not yet returned to the nest. Both parents apparently returned to the nest later, as six days later at least one parent was nearby and the young were believed to have fledged successfully (M. Harris, personal communication).

The female of the other nest box tested with the more realistic model arrived first, landing on the trunk about 50 cm.

from the model. She watched the model and remained motionless for 5 minutes. The male then arrived, and landed on the tree trunk opposite the female. He looked at the model, then spread his tail and began displaying TFBR. After a few bouts of TFBR, he raised his wings over his head and began swaying from side to side in the WS, followed by Sway displays. He continued for several minutes before the female began doing the same displays. After displaying for about 8 minutes, the male hopped towards the model while continuing WS. Upon reaching the snake, he pecked it several times gently as though exploring it. At this point, I terminated the experiment by removing the model to reduce disturbance to the nesting nuthatches and to facilitate subsequent photography (Figure 1) and videotaping of their responses to the snake model.

At the nest box on which the cruder model was placed initially, the female arrived first with a beak full of food. She hopped down the trunk, paused for about 2 seconds to inspect the model, and then continued into the nest box. When the male returned also carrying food, he paused and inspected the model for a much longer time from the tree trunk. He then flew to a branch from where he looked at the model for several seconds before flying to the top of the nest box near the model, paused, then hopped down the side and into the hole. After I replaced the cruder model with the more realistic one, the male returned first, and inspected the model from the tree trunk for about two minutes. He then performed TFBR and Sway for a few minutes. Next, he flew up the tree to where he was obscured by branches for about one minute before flying back to the trunk and hopping past the realistic model into the nest box. On his next trip back, he hopped past the model without inspecting it, and entered the box. On her first encounter with this new, realistic model, the female hopped past it without pausing.

DISCUSSION

Although White-breasted Nuthatches have been observed to use TFBR and WS to threaten birds and mammals (Kilham 1968; Long 1982a; Stokes and Stokes 1983), Long's (1982a) observation of one full and one partial display induced by a Fox Snake (Elaphe vulpina) in Wisconsin appears to be the only previous record of such behaviours directed at snakes. Although my sample size is small, my observations show that both sexes exhibit distraction displays, but when both are present the male at least sometimes displays more intensely than the female, as also found by Long (1982a, 1982b). When similar tests were conducted near two nests of Pygmy Nuthatches (Sitta pygmaea), all returning individuals inspected snake models, but none performed threat displays (J. E. Bryan, unpublished data).

The nuthatches presumably inspected and threatened snake models because they recognized snakes as predators. Snakes are not mentioned as known predators of White-breasted Nuthatches in the reviews of this species' biology by Bent

Page 14 British Columbia Birds Volume 6, 1996

(1948) and Pravosudov andGrubb (1993), and I could find no documented instances of such predation. Nevertheless, they would presumably be as vulnerable to snake predation in the Okanagan as Pygmy Nuthatches, whose young have been eaten by Gopher Snakes (Pituophis melanoleucus) there (Cannings et al. 1987). Both these nuthatch species in the Okanagan build nests in tree cavities or nest boxes at similar heights (Pygmy: 0.8-21.5 m., mean 6.6 m; White-breasted: 1.5-21 m., mean 5.8 m.) and in similar habitats (Cannings et al. 1987). Such nests could be reached by snakes, particularly Gopher Snakes, which are very good climbers (Terres 1980:76l; Thompson and Turner 1980).

Swallow defensive behaviours do little to prevent snake predation at nests (Brown and Brown 1987; Winkler 1992), and it seems unlikely that nuthatches could deter snakes with alarm calls or threats. However, recognizing snakes as threatening probably has survival value. Threats and alarm calls could alert mates even if they were ineffective in deterring a snake.

ACKNOWLEDGEMENTS

M. Harris and L. Rockwell gave me access to nesting birds. I thank D. St. John, W. C. Weber, R. Howie, M. K. McNicholl, S. G. Sealy and M. J. Taitt for comments on the manuscript. C. B. Crawford provided a search of a bibliographic data base (BIOSIS 1969+).

LITERATURE CITED

Bancroft, J. 1987. Observations of White-breasted Nut- hatch. Blue Jay 45:172-174.

Bent, A. C. 1948. Life histories of North American nuthatches, wrens, thrashers, and their allies. US.

National Museum Bulletin 195.

Brown, C. R. and M. B. Brown. 1987. Group-living in Cliff Swallows as an advantage in avoiding predators. Behavioral Ecology and Sociobiology 21:97-107.

Bryan, J. E. 1998. Inspection of snakes by White-breasted, Sitta carolinensis, and Pygmy, Sittapygrnaea, nuthatches. Canadian Field-Naturalist 112:225-229.

Cannings, R. A., R. J. Cannings and S. G. Cannings. 1987. Birds of the Okanagan Valley, British Columbia. Royal British Columbia Museum, Victoria.

Gottfried, B. M. 1979. Anti-predator aggression in birds nesting in old field habitats: an experimental analysis. Condor 81:251-257.

Jackson, J. A. 1974. Gray Rat Snakes versus Red-cockaded Woodpeckers: predator-prey adaptations. Auk 91:342- 347.

James. D. K., L. Petrinovich and T. L. Patterson. 1983. Predation of White-crowned Sparrow nestlings by the Western Terrestrial Garter Snake in San Francisco, California. Copeia 1983:511-513.

Kilham, L. 1968. Reproductive behavior of White-breasted Nuthatches. I. Distraction display, bill-sweeping, and nest hole defense. Auk 85:477-492.

Long, C. A. 1982a. Comparison of the nest-site distraction displays of Black-capped Chickadee and White-breasted Nuthatch. Wilson Bulletin 94:216-218.

Long, C. A. 1982b. Nest-site distraction displays by birds with egg-like spots in the wings. Jack-Pine Warbler 60: 22-26.

Nolan, V. J. 1959. Pileated Woodpecker attacks Pilot Black Snake at tree cavity. Wilson Bulletin 71:381-382.

Pravosudov, V. V. and T. C. Grubb, Jr. 1993. White- breasted Nuthatch (Sitla carolinensis). No. 54 in A. Poole and F. Gill (editors). The birds of North America. Philadelphia Academy of Natural Sciences, Philadelphia and American Ornithologists' Union, Washington.

Stokes, D. W. and L. Q. Stokes. 1983. A guide to bird behavior. Volume 2. Little, Brown and Company, Boston.

Terres, J. K. 1980. The Audubon Society encyclopedia of North American birds. Alfred A. Knopf, New York.

Thompson, B. C. and C. L. Turner. 1980. Bull Snake predation at a Cliff Swallow nest. Murrelet 61:35-36.

Walters, J. R. 1990. Anti-predatory behavior of lapwings: field evidence of discriminative abilities. Wilson Bulletin 102~49-70.

Winkler, D. W. 1991. Parental investment decision rules in Tree Swallows: parental defense, abandonment, and the so-called Concorde Fallacy. Behavioral Ecology 2:133- 142.

Winkler, D. W. 1992. Causes and consequences of variation in parental defense behavior by Tree Swallows. Condor 94:502-520.

Volume 6, 1996 British Columbia Birds Page 15

MERLIN (FALCO COLUMBARIUS) PREYS ON FLYING DRAGONFLIES

John M. Cooper Sirius Environmental Research

1278 Laurel Road, R.R. 3, Sidney, B.C. V8L 5K8

Abstract -- A juvenile Merlin was observed to catch and eat one dragonfly and chase others while on migration. The importance of large insects in the diet of Merlins during migration warrants further investigation.

Key Word!: Diet, dragonflies, Falco cplumbarius, Merlin, migration, Vancouver Island.

The diet of the Merlin (Falco columbarius) consists mainly of small birds, such as sparrows, larks, swallows and shorebirds, with mammals, insects and reptiles usually constituting a smaller proportion (e.g. Fox 1964; Beebe 1974; Becker 1985; Bibby 1987; Sodhi and Oliphant 1993). Although one early study on the east coast of North America found a large proportion of insects in the diet of migrating Merlins (Allen and Peterson 1936), insects are usually reported as infrequent components of the diet. For example, breeding Merlins in Montana preyed mainly on small birds, but grasshoppers and moths comprised 4% and 1% of the diet respectively (Becker 1985). Beebe (1974) generalized that insects, "especially dragonflies," are taken occasionally on the wing. Most studies of diet have been conducted during the breeding season or in winter, whereas diets during migration are less well known (Sodhi et al. 1993). In this note, I document predation by a migrant Merlin on flying dragonflies.

On 10 September 1996 at 19:00, I observed a juvenile Merlin chasing an American Robin (Turdus migmtorius) through our open-wooded, semi-rural neighbourhood in North Saanich, about 25 km. north of Victoria, British Columbia. This was during the September peak of fall migration of Merlins in the Victoria area (Campbell et al. 1990:Appendix 1). At 19:05, I noticed the Merlin perched on a dead-topped Bigleaf Maple (Acer rnacrophyllum). The Merlin sallied out briefly, swooped at something in midair, and returned to its perch, whereupon it plucked the wings from a large dragonfly (probably Aeschna sp.) and swallowed the body. A few minutes later, the Merlin circled our yard, flew unsuccessfully at two flying dragonflies, and disappeared from sight.

Flying dragonflies of several species were quite evident in our neighbourhood during the warm (19' C), late summer evening, as were flying termites and several songbirds. After its unsuccessful pursuit of a robin, the Merlin apparently targeted dragonflies, as an abundant alternate prey.

Merlins are known to concentrate on particular prey species which are a) abundant, b) leave cover frequently, making

them vulnerable to aerial predation, and c) have a mass of between 21 and 40 g. (Sodhi and Oliphant 1993). Dragonflies seem to meet the first two criteria, but the largest species, Aeschna spp., weigh only a few grams (R. Cannings personal communication). The bird that I observed appeared to be hunting the largest dragonflies available. Oliphant (1974) reported an observation of an immature Merlin hawking dragonflies, of an unknown species, on the wing in Saskatoon in late summer (Oliphant and McTaggart 1977). That Merlin consumed the dragonflies in flight. My observation differed in that the Merlin returned to a perch and plucked the wings from the dragonfly before consuming it; a behaviour similar to how Merlins eat birds (Sodhi 1992). Palmer (1988) summarized a few previous observations of both behaviours. Dekker (1985) reported an immature in Alberta eating some in the air, and others on a perch during a single hunting sequence, although he did not indicate whether or not the latter were plucked. Bird (1932) found that the wings had been removed from Aeschna remains in the stomach of a Merlin shot in Manitoba, whereas the remains of other dragonfly species in the same stomach were intact.

Beebe (1974) considered the Merlin to be an exceptionally successful hunter, more so than other falcons. During its southward migration on the south coast of British Columbia, the Merlin has a multitude of migrant songbirds and large flying insects to choose from and likely preys opportunistically on species which are locally abundant and of appropriate size. Because little is known about Merlin diet during migration, the importance of large flying insects as a food source during fall migration may have been underestimated. This possibility is supported by Dekker's (1988:footnote to Table 2) comment that Merlins captured "numerous dragonflies" during migration around Beaverhills Lake, Alberta. Dragonfly "hawking" appears to be very common among young-of-the-year Merlins and other falcons (Oliphant and McTaggart 1977; L. W. Oliphant personal communication to M. K. McNicholl17 December 1996) and may be important practice in learning to hunt (Dekker 1985). Further studies on prey taken during migration would help determine the relative importance of insects in the diet of Merlins.

Page 16 British Columbia Birds Volume 6, 1996

ACKNOWLEDGEMENTS

The comments of Kenneth H. Morgan and Lynn W. Oliphant on an earlier draft helped improve the manuscript.

LITERATURE CITED

Allen, R. P. and R. T. Peterson. 1936. The hawk migrations at Cape May Point, New Jersey. Auk 53:393-404.

Becker, D. M. 1985. Food habits of Richardson's Merlins in southern Montana. Wilson Bulletin 97:226-230.

Beebe, F. L. 1974. Field studies of the Falconiformes (vultures, eagles, hawks and falcons) of British Columbia. British Columbia Provincial Museum Occasional Paper No. 17.

Bibby, C. J. 1987. Food habits of breeding Merlins Falco columbatius in Wales. Bird Study 34:64-70.

Bird, R. D. 1932. The Pigeon Hawk as an odonatologist. Entomological News 43:242.

Campbell, R. W., N. K. Dawe, I. McTaggart-Cowan, J. M. Cooper, G. W. Kaiser and M. C. E. McNall. 1990. The bird; of British Columbia. Volume 2. Royal British Columbia Museum, Victoria.

Illustrations by Roger Tory Peterson, from the collection of the Newfoundland Museum,

reproduced with their kind permission in Robie Tufts' Birds of Nova Scotia

on the Nova Scotia Museum of Natural History website

(http://nature.ednet.ns.ca)

Dekker, D. 1985. Wild hunters. Canadian Wolf Defenders, Edmonton.

Dekker, D. 1988. Peregrine Falcon and Merlin predation on small shorebirds and passerines in Alberta. Canadian Journal of Zoology 66:925-928.

Fox, G. A. 1964. Notes on the western race of the Pigeon Hawk. Blue Jay 22:140-147.

Oliphant, L. W. 1974. Merlins -the Saskatoon falcons. Blue Jay 32:140-147.

Oliphant, L. W. and S. McTaggart. 1977. Prey utilized by urban Merlins. Canadian Field-Naturalist 91 : 190-192.

Palmer, R. S. (Editor). 1988. Handbook of North America birds. Volume 5. Diurnal raptors (part 2). Yale University Press, New Haven, Connecticut.

Sodhi, N. S. 1992. Central place foraging and prey preparation by a specialist predator, the Merlin. Journal of Field Ornithology 63:71-76.

Sodhi, N. S. and L. W. Oliphant. 1993. Prey selection by urban-breeding Merlins. Auk 110:727-735.

Sodhi, N. S., L. W. Oliphant, P. C. James and I. G. Warkentin. 1993. Merlin (Falco columbarius). No. 44 in A. Poole and F. Gill (Editors). The birds of North America. Academy of Natural Sciences, Philadelphia and American Ornithologists' Union, Washington, D.C.