Territoriality and the Significance of Calling in the Lanyu Scops Owl Otus Elegans Botelensis

lbiS (2000) 142, 297-304

Territoriality and the significance of calling in the Lanyu Scops Owl Otus elegans botelensis

LUCIA LIU SEVERINGHAUS Institute of Zoologx Academia Sinica, Taipei, Taiwan, ROC

Territories of Lanyu Scops Owls Otus elegans botelensis overlap in both breeding and non-breeding seasons. Results of radiotelemetry showed that neighbouring owls do not use the shared areas of their territories at the same time. Frequent countersinging apparently permitted individuals to avoid potentially costly encounters with neighbours. Non- territorial owls can forage and rest in occupied territories. Experiments using decoys and playbacks showed that intruding owls were tolerated within either core or peripheral territories in all seasons if they remained silent, while calling intruders almost always incited threats or attacks even in autumn. Tolerating silent owls that are not competitors for mates or for nest sites appears to be an energy-saving territorial strategy.

Territoriality is a spacing system maintained by the al. 1987; Eastern Screech Owls Otus asio, Belthoff et behavioural patterns of a given species. The reward for al. 1993; Barred Owls Strix varia, Nicholls & Fuller territorial defence is frequently better survival or 1987). The overlapping areas are primarily used by one higher reproductive success. The territoriality of a bird in Eastern Screech Owls (Belthoff et al. 1993), species thus reflects important environmental and while in Barred Owls it is used only for short periods social constraints on the species. Models of the exter- (Nicholls & Fuller 1987). No study has reported nal conditions that favour territorial behaviour usually varying territory defence strategies of owls. consider the quality, abundance and distribution of The territoriality of the insectivorous Lanyu Scops resources in space and time, as well as the density of Owl Otus elegans botelensis is complex. They have competitors and their competitive abilities (e.g. overlapping activity areas so that they appeared non- Carpenter & MacMillen 1976, Myers et al. 1981, territorial at first (Severinghaus 1986). However, Davies & Houston 1984). border fights between neighbours suggest that activity Economical defendability is a major determinant areas may also be their territories. Territorial defence of territorial behaviour (Brown 1964), while the in this species is variable. Sometimes an intruder is difference in resource value to the contestants may tolerated while at other times it is expelled. This paper affect territorial strategies (Hammerstein 1981). describes the unusual territoriality of this species, Furthermore, the costs and benefits associated with and reports the results of a series of experiments territorial maintenance may affect the methods used conducted to explore the following questions: (1) in settling a contest (Harper 1991). Owls generally Does territorial aggression occur only at specific times? maintain all-purpose territories (Tawny Owl Strix (2) Is aggression aimed at only specific individuals? (3) aluco, Hirons 1985, Galeotti 1994; Little Owl Athene What stimulus may trigger aggressive responses from noctua, Finck 1990; Tengmalm's Owl Aegolius finereus, territory owners? Korpimaek 1987, 1988; Great-horned Owl Bubo virginianus; Spotted Owl occidentalis, Hunter Strix METHODS AND MATERIAL et al. 1995). With radiotracking, which permits careful documen- Study population tation of activity patterns within territories and determination of boundaries, the territories of neigh- The Lanyu Scops Owl is a small insectivorous owl bours in a number of species of owls have been found (weight about 125 g) found throughout Lanyu Island to overlap (Boreal Owls Aegolius funereus, Hayward et southeast of Taiwan (longitude: 12 1"5'E, latitude: 22"N). I have been studying this population since Email: [email protected] 1985, and more than 80% of the birds in selected areas

@ 2000 British Ornithologists' Union 298 L.L. Sevennghaus

are colour-marked. The core habitat for this owl is Silent owls were not seen enough times to map out mature \\-oods with tall, broadleaf trees while the their activity areas and are considered non-territorial. peripheral habitat contains small clusters of trees near villages, or next to agricultural fields rvhere human dis- Reaction to playback turbance is high (Severinghaus 1989). Habitat patches vary in size, in yegetation structure, in their number of In the lowland areas around Lanyu Island, 60 playback known tree cavities, and in o~vldensity. stations were selected, each roughly 0.5 km apart. The basic breeding biolog!, of the Lanyu Scops Otv1 Every month from Iuly 1990 to August 1991 taped \vas reported by Severinghaus (1 992). Young Lanyu male calls were played for 20 min at each station and Scops Owls enter the breeding population at t\vo years all the responses and reactions of owls were recorded. or older and can live more than 13 years. The pair bond may he maintained for scveral years, but mate change Radiotracking is frequent between years. The tenure of good nest cavities is short (Severinghaus unpubl. data). We used AVM radio-transmitters to track Lanyu Scops Male Lanyu Scops Ouds make a hoo-lzoo call, while Owls in 1990. Each transmitter weighed less than 5 g females also make a yap call. Oivls increase calling and lvas secured on the ow1 with a body harness (< 5141 frequent). in late December and early January when of bodyTveight). Effective battery life was 2 months. courtship and ncst-cavity searching start. Counter- The 24 owls we were tracking included one mated pair, singing hy neighbours is common from January until two 2-neighbour (both one male one female), a cdq’ July. These calling bouts are interspersed .rvith 3-neighbour (one male and a pair) group, and isolated periods of silence. Males also utter a distinct copulation individuals. The four groups of birds listed above were call, similar to that reported for Otits scops (Koenig respectively monitored simultaneously from two 1973). stations (White 8- Garrott 1990). Location readings ‘The Lvork reported in this papcr rvas conducted from were taken every 30 min throughout the night for I989 to 1996 inclusive. In general, t’ivo field assistants three nights each in June and July 1989. Locations of and tux) to four sunimer students assisttd with data these owls were marked on topographic maps. The gathering each year. maximum area polygon method was again used to determine territory sizes. Determination of activity areas Decoy and playback experiments 7he boundaries of om1 act]\ it) areas I\ ere determined in two locations bq the maximum area polygon I made life-size decoys with papier mache. In April method ahr mapping all the perch sites of each 1995, I selected nine known nest trees and did the ~olour-ringedindividual The tv o locations 1ver-e (1) a following experiments. (1) A decoy or a stuffed Lanyu 2 34-ha mature hardwood forest, and (2) a 3-ha Scops Owl specimen was placed on a perch or a rock coconut plantation with on11 dead trunk5 left Many of directly below the nest cavity or as close to that point the coconut trunks in the plantation had ca\ities in as possible, and about 3-5 m from it. The reactions of them 11 hich \$ere used b> breeding OM 15 in 1990, but any oivl to the decoy or the specimen were recorded unfortunately the coconut trunks I\ ere le\ elled and the for 25 min. (2) A tape of mixed female and male calls field con\ crted to agriculture in spi ing 199 1 \vas played for 25 min from a tape recorder placed In the torest 5tud) area, perch sites uerc determined next to the decoy or the specimen, and the reactions of through intensii c obser\~ationson threc nights el er) the oivl were recorded. In January and October 1996, month bh threc trained obserters froin Tul\ 1993 to the same test was rerun in the same territories using June 1904 In thc. plantation, plqbacks of owl calls only decoys. \!ere uitd in Apiil IW) The taped loice belonged The decoy expcriments were conducted in April ased inale rrcorded in 1989 from another 1995, January and October 1996, because Lanyu loc~itioii it \\a\ strange to all the 0~1spresent in Scops Owls’ reactions to playback were the most 1WO Pla\ bail, locations \\ere regular]) spaced at intense in April, weakest in autumn months (Fig. I), 20-in intti\dl\ 171 the plantation and the tape M~S while January is the very beginning of the breeding pld\eecl np to 30 minutes each time, depending on season. Test results were analysed by comparing r\ hcn m\ (>I\I icqxmdd The identit), sex, and tlpe of proportions in paired samples (Snedecor & Cochran re\poii

92000 British Ornithologists Union Ibis 142, 297-304 Lanyu Scops Owl territoriality 299

o Tree 251 00 20

0 ...... 0 Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun .... 1990 1991 .... 0 N 010m ...... Figure 1. Variations in owl response to playback of a male owl \- ...... 0 call at different times of year. Owls showed the most vigorous reactions in April and the least in autumn. +, Male approach Figure 3. Territories of marked Lanyu Scops Owls overlapped in the non-breeding season (Aug 1993 to Feb 1994). Other expla- calling; 3,male on speaker calling; 0,female approach calling; nations see Fig. 2. D, male nest guarding; 0, male approach silent; 0, female approach silent. Fig. 4) to a narrow strip (March-April 1990 along the R ES U LTS highway, Fig 5). This latter overlap was discovered during monthly tape playback. Some owls were seen Territory boundaries and radiotelemetry fighting and chasing each other during the spring. Yet, In core habitat, territories of marked individuals non-territorial owls were seen resting and foraging in a overlapped greatly both in the breeding and in the number of core and peripheral territories (as can be non-breeding seasons in 1993-94 (Figs 2, 3). In seen in Figs 2-4). No efforts by territory owners to peripheral areas, the degree of territory overlap varied exclude these owls were observed. from zero (April 1990 in the coconut plantation, Results of radiotraclung in 1989-90 also showed that during the breeding season neighbouring owls had overlapping territories (Table 1). Judging from the o Tree pattern shown between the pair and their male neigh- 0 Nest tree - Male owl bour (Fig. 6), the female overlapped more with the ...... Female owl neighbour than did her mate. The male neighbour was Unpaired owl - within the female’s activity area 23.1% of the time that he was located by radiotrackmg, but he was never there when either the male or the female was there. In 13 tracking records taken 30 min apart, the male neighbour was in the overlapping area on the eighth, ninth and eleventh time, while one or both members of the pair was there on the second and ninth time. Within the overlapping area, the locations of the neighbouring male were only 17.6 k 5.6 m from the previous locations of the pair, but the time they were there was at least 30 min apart. The mean distance between neighbours at any moment was significantly longer than that between the paired birds (Table 1, t-test, P < 0.0001). Figure 2. Territories of marked Lanyu Scops Owls in core I used the activity pattern of these three owls to habitat overlapped in the breeding season of 1994 (March- July). Territory boundaries were determined by maximum analyse further the interactions between neighbours. If polygon method based on perch locations. Unpaired owls were an owl’s utilization of its entire territory is 1, the prob- tolerated. ability that the overlapping area is used by this

02000 British Ornithologists’ Union, Ibis, 142, 297-304 300 L. L. Severinghaus

Figure 4. Territory boundaries of Lanyu Scops Owls in a peripheral habitat from March 1990 to April 1991. Vegetation or topography prohibited observation of territory boundaries in some cases. Unknown owl: owl that moved too fast for identification or unmarked individual.

individual should be equal to its proportion of the did not differ from that expected. However, the male entire territory The probability that two owls appear and his neighbour each used the overlapping area simultaneously in the overlapping area would be equal significantly more times than expected, but never at to the product of the two individual probabilities. the same time. The two males apparently actively These probabilities are used to calculate the expected avoided each other in their frequent visits to the over- frequencies for the four possible situations: both owls lapping area. in the overlapping area, owl A in/oivl B out, owl A out/o\vl B in, and both out. This is used to test the Decoy tests observed frequencies with a chi-squared test (Table 2). Although the female and her male neighbour never Test results showed that a silent decoy or stuffed used the overlapping area at the same time, the specimen aroused no aggressive response from either frequency with which they used the overlapping area paired owls in spring or resident owls in winter

Table 1. Distances between neighbour owls obtained from radiotracking

Between Case Min (m) Max (m) Mean t sd(m) Overlap (ha) Overlap (%)*

~- ~ ~~~~ ~~~~ Mates 1 1.7 110.1 35.9 i 34.8 0.69 90.7 2 15.0 75.0 33.2 i 18.4 0.40 62.6 Males 1 34.5 198.8 149.1 5 53.1 0.08 32.7 2 15.0 - - 0.31 39.1 Female + male 1 36.3 191.0 132.4 i 42.9 0.21 13.7 2 62.5 336.3 194.0 k 83.0 1.40 37.1

*Overlap as a percentage of the male's territory. When two males are involved, this number is the mean of the two percentages. In the second case of 'between males' category, the two owls appeared simultaneously in the area only once. Thus only one distance measurement was obtained.

g)2000 British Ornithologists' Union, Ibs, 142, 297-304 Lanyu Scops Owl territoriality 301

Table 2. Comparison of the frequency a pair of owls and their male neighbour used the overlapping area.

Only Only Both in female in neighbour in Both out

Observed 0 4 8 30 Expected 1.1 5.8 7.9 27.3 Chi-square = 1.927 df = 3 ns

Only Only Both in male in neighbour in Both out

Observed 0 4 12 30 Expected 0.28 2.5 5.06 38.18 Chi-square = 12.448 df = 3 P c 0.01

The probability that the overlapping area is used by an owl equals its proportion of the entire territory. The probability that two owls simultaneously appear in the overlapping area would equal the product of the two individual probabilities. Expected frequency is calculated using these probabilities against the total number of observations. Figure 5. Narrow territory overlap of Lanyu Scops Owls in peripheral habitat in April 1990. Boundaries were determined by playback of taped calls. Fighting occurred when both males reacted vigorously, either by repeatedly flying over the arrived at the same location. ‘intruder’ while calling, or by physically attacking the intruder (Table 3). In April, some pairs mated after (Table 3). When taped owl calls accompanied either a trying to expel the intruder. The owls did not react decoy or a stuffed specimen, most resident owls ddferently to the decoy or the stuffed specimen; thus results of decoy and specimen tests were pooled in malung analyses for April. In each season, territorial owners’ reactions to silent or vocal intruders differed significantly (Table 4, comparison of proportions in paired samples, P < 0.025).

DISCUSSION Lanyu Scops Owls are territorial, but not exclusively so. It is believed that less than 50% of the adult Lanyu Scops Owls breed, because of a shortage of nest cavities (Severinghaus, unpubl. data). Although it is possible that some of these owls are actually breeders and we fail to find their nests, the large number of owls

Table 3. Summary of territory owner‘s reactions to decoy and tape playback experiments.

Date Type N Aggression

Apr 1995 Decoy only Specimen only Figure 6. Relative position of a pair (YW, BW) and a single (S + Decoy + tape W) Lanyu Scops Owl based on radiotracking. Arabic numerals Specimen +tape refer to the sequence of positions during 3-4 June 1990. The Jan 1996 Decoy only same number represents the simultaneous locations of the Decoy + tape three owls. The overlapping portion of their territories was visit- Oct 1996 Decoy only ed by all during a night, but not at the same time. Other nights Decoy + tape show similar patterns.

@ 2000 British Ornithologists’ Union, Ibis, 142, 297-304 302 L. L. Severinghaus

Table 4. Analysis of the results of decoy tests by comparing Monkeys Alouatta palliata used dawn calls as a spacing proportions of paired data. mechanism in the overlapping part of their home Test results (frequency) ranges. Although Sekulic (1982) questioned the Lvisdoni of Howler Monkeys informing neighbours of Decoy onlyiwith tape Apr 1995 Jan 1996 Oct 1996 their position, it should be an adaptive strategy if frequent encounters lead to aggression each time. +I+ 0 0 0 Diverse species of birds can identify the calls of their +I- 0 0 0 mates and their neighbours (e.g. Wooller Weary -I- 2 1 0 1978, -/+ 6 6 a et al. 1987, Speirs & Davis 1991). Many owls appear to recognize their mates by their calls, as do Tawny Owls Chi square 4.17 4.17 6.13 (Galeotti & Pavan 1991), and Screech Owls (Cavanagh P< 0 05 0.05 0.025 & Ritchison 1987). Tawny Owls can differentiate Methods given in Snedecor and Cochran (1967) +I+ Owls neighbours from strangers and gauge their responses reacted to decoy both with and without tape, +/-, owls reacted appropriately to save agonistic energy (Galeotti & to decoy only but not when tape was played, -I-, owls did not Pavan 1991, 1993). The calls of Great-horned Owls react to either decoy or when tape was played, -I+, owls did not (Rohner 1997) have individually identifiable charac- react to decoy except when accompanied by tape teristics. Lanyu Scops Owls probably can differentiate the calls of their neighbours and mates from those of contcsting for trrritories each spring is a strong strangers, since the calls of individuals appear different indication that many 0x3 Is ha\ e no breeding opportu- on sonograms (Severinghaus 1986). Results of decoy nities I hq arc pre\ented from breeding because tests indicate that the calling of an unfamiliar Lanyu suitable habitat\ are saturated \\ ith donunant indn idu- Scops Owl is taken as an act of challenge to which nls, as found in J number of species (eg Broitii 1969, they react aggressively. Yet the same ‘owl’ can be 1 ckert & IVeatherhedd 1987, Arcese 1089) tolerated when resting silently. The non-territorial owls ’I‘hc population densitt of Idan\u Scops O\t 1 appears tolerated by territory owners were always foraging or to he iorrelatcd 111th the number of nest cavities resting silently. at ailablc in an arca (Se\ eringhaus unpuhl data) Contests are costly to territory owners even when Where ou 1 population densit) is high, breeding terri- oivncrs and floaters show what Maynard Smith (I 982) tories merlap greatl!, \I here cnil densit\ 15 lo\\, calls payoff asymmetry, uncorrelated asymmetry, or territories do not oxerldp Floaters hebeen seen competitive asymmetry. When floaters are many and toiaging and restin? in a number of territories The persistent, it may be unprofitable for owners to try to overlap in Lanyu Scops On1 territories is most likely a evict them from their territories (Brown 1982). The I esult of high competition pressure for nest sites, rather folloiving example testifies to this. One female owl than of food supplq as in the Dunnock Pziizella established a territory around two adjacent cavities in fnodtiliiny (Davies & Hartlei 1996) This is because in early spring of 1993 and 1994. Each year she fought populations where members arc food-limited, floaters off intruding females for more than three months, only probablt c annot forage regularl) in territories (Smith to be supplanted from both cavities just prior to egg- 1978) Eurtherniore, contests for tood are unlikely to laying. Her energy reserves had perhaps been reduced escalate into tatdl fighting in animals (Enquist & after repeated contests. Leimai 1990) I an\ u scops 1s fight for nest caities Signalling can reduce the costs of territorial mainte- \t hile tolerating the prewnic of silent toraging ouls nance (Ydenberg et nl. 1988). An opponent’s strength Radiotrachng hay sho\i n that although territories or abilities may be determined by assessing the oi erlapped, neighbouring Lamu Scops Chvls do not information containcd within the signals (Harper us(>the shared areas at the same time Furthermore, 1991). Vocalizations are particularly efficient signals thy trt quent countersinging of neighbouring males for nocturnal animals. Lanyu Scops Owls do not have pi-obabl\ ,illo\~\them to monitor each other’s posi- complex vocal repertoires. Their calls thus function tions Jnci to ,i\ old potcntiall\ costl) dggressii e both as a territorial proclamation and an advertisement t’ncountc’rs \uth as seen t\\icc duiing Lipe pld) bdck of their availability to potential mates. Silent owls are 1, hcn ti\ o iiialc ox\ 15 shou eJ up to expel the calling not overtly advertising their presence as mating candi- strangci aid ended up in a \igorous fight izith each dates, thus are not treated as competitors. other L’vc hae seen 01%Is fight \I ithout tape plat back Temeles (1990) found that in Northern Harrier Chi\?i\ (1L)O(l) suggvsted that lllantled Honler Circus cyaiieus differences in potential payoffs lead to

6 2000 British Ornithologists Union, Ibis. 142, 297-304 Lanyu Scops Owl territoriality 303

differences in territorial defence against neighbours Research Station on Lanyu Island made available to us and floaters. Lanyu Scops Owls live in rich tropical comfortable living quarters, and provided daily moral forests with few food competitors and apparently are support. This study has been supported by grants from under no food shortage pressure. Defending territories the Council of Agriculture, Taiwan. Dr Fushing Hsieh against foraging owls would be of no benefit. If silence helped with statistics. Dr Ian Newton, Dr Jerram also denotes weakness, and weak owls stand no chance Brown, and the reviewer's comments on an earlier ver- of winning a nest cavity, silent owls are not worth the sion of the manuscript greatly improved its quality. effort of expulsion. Since the purpose of aggression is to expel the competitor, tolerating silent owls should REFERENCES be an energy-saving territorial strategy. The fact that silent owls are ignored is not because Arcese, P. 1989. Territory acquisition and loss in male song Spar- they only lurk around in the periphery of defended row. Anim. Behav. 37: 45-55. Belthoff, J.R., Sparks, E.J. & Ritchison, G. 1993. Home ranges of territories, which is the case with the Great-horned adult and juvenile eastern screech owls: size, seasonal variation Owl (Rohner 1997). Non-territorial Lanyu Scops and extent of overlap. J. Raptor Res. 27: 8-1 5. Owls also use areas close to nest cavities. Some of them Brown, J.L. 1964. The evolution of diversity in avian territorial sys- have been seen repeatedly in the same locations, tems. Wilson Bull. 76: 160-1 69. similar to what Smith (1978) found with Rufous- Brown, J.L. 1969. Territorial behavior and population regulation in birds: a review and re-evaluation. Wilson Bull. 81: 293-329. collared Sparrow Zonotrichia capensis floaters. Brown, J.L. 1982. Optimal group size in territorial animals. Theor. Arcese (1989) suggests that the degree of habitat Biol. 95: 793-810. saturation should influence the tactics of floaters Carpenter, F.L. & MacMillen, R.E. 1976. Threshold model of feed- because it affects the probability of settling without ing territoriality and test with a Hawaiian honeycreeper. Science engaging in conflicts. Lanyu Scops Owls live under 194: 639-642. Cavanagh, P.M. & G. Ritchison. 1987. Variation in the bounce and very high population density. Floaters have adopted whinny songs of the eastern screech owl. Wilson Bull 99: the behavioural strategy of being quiet in territories, 620-627. which makes them tolerable to territory owners, Chivers, D.J. 1969. On the daily behavior and spacing of howling permits them to wait for opportunities to replace the monkey groups. Folia Primatol. 10: 48-102. owners, and perhaps cuckold the owners in the mean- Davies, N.B. & Hartley, I.R. 1996. Food patchiness, territory overlap and social systems: an experiment with Dunnocks Prunella time (Morton et al. 1990). One observation suggests modularis. J. Anim. Ecol. 65: 837-846. the last situation may exist in Lanyu Scops Owl. One Davies, N.B. & Houston, A.I. 1984. Territory economics. In Krebs, female territory owner disappeared before her young J.R. & Davies, N.B. (eds) Behavioural fcologx an Evolutionary hatched. A floater female that frequently perched Approach: 148-1 69. Sunderland, MA: Sinauer Associates. quietly near this nest took over incubation and the Eckert, C.G. and Weatherhead, P.J. 1987. Owners, floaters and competitive asymmetries among territorial red-winged black- rearing of young. Whether the second female actually birds. Anim. Behav. 35: 1317-1 323. laid some of the eggs in that nest is unknown. Enquist, M. and Leimar, 0. 1990. The evolution of fatal fighting. In conclusion, this study shows that vocalization Anim. Behav. 39: 1-9. plays an important role in the territoriality of Lanyu Finck, P. 1990. Seasonal variation of territory size with the Little owl Scops Owls. Whether territorial aggression is displayed (Athene noctua). Oecologia 83: 68-75. Galeotti, P. 1994. Patterns of territory size and defence level in rural does not depend on variation in the food resource base, and urban tawny owl (Strix aluco) populations. J. Zoo/. Lond. nor the number of owls present, but rather whether 234: 641-658. a competitor is displaying vocally. Galeotti, P. & Pavan, G. 1991. Individual recognition of male tawny owls using spectrograms of their territorial calls. €tho/. Ecol. Many people and organizations contributed to this Evol. 3: 113-126. Galeotti, P. & Pavan, G. 1993. Differential responses of territorial long-term study. For the results reported in this paper, Tawny Owls Strix aluco to the hooting of neighbours and I wish to thank especially Mr Liang Chieh-teh, Mr strangers. Ibis 135: 300-304. Chen Hui-sheng, Mr Chen Chun-chin, Mr Mark Hammerstein, P. 1981. The role of asymmetries in animal contests. Phillipart, Ms Kuo Yen-min, Mr Lo Hung-jen, Mr Anim. Behav. 29: 193-205. Chang Chien-chung, Mr Su Sheng-hua, Mr Chien Harper, D.G.C. 1991. Communications. In Krebs, J.R. & Davies, N.B. (eds) Behavioural Ecology, an Evolutionary Approach, 3rd Che-chung, Ms Fang Yun-ju for assisting with field- edn: 374-398. London: Blackwell Scientific Publications. work. Mr Chen Hui-sheng, Mr Chang Chien-chung, Hayward, D.G., Hayward, P.H. & Garton, E.O. 1987. Movements Mr Su Sheng-hua, Mr Chien Che-chung, Ms Lin Yi- and home range use by boreal owls in central Idaho. In Nero, hua, and Ms Fang Yun-ju assisted with data analysis R.W., Clark, R.J., Knapton, R.J., & Hamre, R.H. (eds) Biology and computer work. Mr Chien of Chung-shan Military and Conservation of Northern Forest Owls. Symposium Pro-

@ 2000 British Ornithologists' Union, Ibis, 142, 297-304 304 L. L. Severinghaus

ceedings: 175-1 84. Winnipeg, Manitoba. Snedecor, G.W. & Cochran, W.G. 1967. Statistical Methods. Ames, Hirons, G.J.M. 1985. The effects of territorial behaviour on the sta- 10: Iowa State University Press. bility and dispersion of tawny owl (Strix aluco) populations. J. Severinghaus, L.L. 1986. The Ecology of Lanyu Scops Owl Otus Zoo/. 6 1: 21-48. elegans botelensis. Taipei: Council of Agriculture. (in Chinese) Hunter, J.E., Gutierrez, R.J. & Franklin, A.B. 1995. Habitat config- Severinghaus, L.L. 1989. The Status and Conservation of Lanyu uration around spotted owl sites in Northwestern California. Scops Owl Otus elegans botelensis. In Meyburg, B.U. i? Chan- Condor 97: 684-4593. cellor, R.D. (eds) Raptors in the Modern World: 423-432. Berlin: Koenig, L. 1973. Das Aktionssystem der Zwergohreule Otus scops World Working Group on Birds of Prey. scops (Linne 1758). Adv. Ethol. Suppl. J. Comp. €tho/. No. 13. Severinghaus, L.L. 1992. Monitoring the population of the endan- Korpimaeki, E. 1987. Selection for nest-hole shift and tactics of gered Lanyu Scops Owl (Otus elegans botelensis). In breeding dispersal in Tengmalm's owl Aegolius funereus. J. McCullough, D.R. & Barrett, R.U. (eds) Wildlife 2001: Popula- Anim. Ecol. 56: 185-1 96. tions: 790-802. London: Elsevier Applied Science. Korpimaeki, E. 1988. Effects of territory quality on occupancy, Smith, S. 1978. The 'underworld' in a territorial species: Adaptive breeding performance and breeding dispersal in Tengmalrn's strategy for floaters. Am. Nat. 112: 571-582. owl. J. Anim. Ecol. 57: 97-108. Speirs, E.A. & Davis, L.S. 1991. Discrimination by adelie penguins Maynard Smith, J. 1982. Evolution and the Theory of Games. Cam- Pygoscelis adeliae between the loud mutual calls of mates, bridge: Cambridge University Press. neighbors and strangers. Anim. Behav. 41 : 937-944. Morton, E.S., Forman, L. & Braun, M. 1990. Extra-pair fertilizations Temeles, E.J. 1990. Northern harriers on feeding territories respond and the evolution of colonial breeding in purple martins. Auk 107: more aggressively to neighbors than to floaters. 6ehav. Ecol. 275-283. Sociobiol. 26: 57-63. Myers, J.P., Connors P. G. & Pitelka, F.A. 1981. Optimal territory Weary D.M., Lemon, R.E. & Date, E.M. 1987. Neighbor-stranger size and the sanderling: compromise in avariable environment. discrimination by song in the veery, a species with song reper- In Kamil. A.C. i? Sargent, T.D. (eds) Foraging Behaviour: Eco- toires. Can. J. Zoo/. 65: 1206-1 209. logical. Ethological and Psychological Approache: 135-1 58. White, G.C. & Garrott, R.A. 1990. Analysis of wildlife radio-tracking New York: Garland STPM Press. data. London: Academic Press, Inc. San Diego. Nicholls, T.H. & Fuller, M.R. 1987. Territorial aspects of barred owl Wooller, R.D. 1978. Individual vocal recognition in the Kittiwake home range and behavior in Minnesota. In Nero, R.W., Clark, Gull, Rissa tridactyla. Z. Tierpsychol. 48: 68-86. R.J.. Knapton, R.J. i? Hamre, R.H. (eds) Biologyand Conserva- Ydenberg, R.C., Giraldeau, L.A. & Falls, J.B. 1988. Neighbors, tion of Northern Forest Owls. Symposium Proceedings: strangers, and the asymmetric war of attrition. Anim. 6ehav. 36: 121-128. Winnipeg, Manitoba. 343-347. Rohner, C. 1997. Non-territorial 'floaters' in great horned owls: space use during a cyclic peak of snowshoe hares. Anim. Behav. 53: 901-912. Sekulic, R. 1982. Daily and seasonal patterns of roaring and spacing in four red howler (Aloualta seniculus) troups. folia Primatol. 39: 22-48. Received 15 October 1998; reviston accepted 28 April 1999

t) 2000 British Ornithologists Union /b/s 142, 297-304