A Long-term Phylogeographic Study of Magpies (Genus Pica):

Behavior, Ecology, and Evolution

 

CHOE Jae Chun

School of Biological Sciences, Seoul National University, Seoul 151-742, Korea

 

 

Abstract

 

With a purpose of elucidating the phylogeny of the genus Pica, behavioral ecology and molecular data were obtained and compared among the subspecies. For the former, we have investigated the breeding ecology of the black-billed magpies (Pica pica sericea) on the campus of Seoul National University since 1998. Breeding successes of 1998, 1999 and 2000 were 2.81.6, 3.31.7, 2.41.4 fledglings/nest (meanSD), respectively and were found to be influenced by the temperature during incubation. Timing of breeding seems to be determined by trade-offs between benefits and costs of early breeding. Possible benefits were chicks getting higher dominance rank by early fledging and additional breeding attempts of parents. Increased incubation load and rearing high-cost, male-biased broods were suggested as the possible costs. Territory size and distribution did not significantly differ among the years. Average territory size was close to 1 ha, and distance to the nearest active nest was approximately 93 m. Molecular sexing revealed that offspring sex ratio was male-biased in early nests and female-biased in late nests. Considering that early-fledged birds generally have higher dominance position than late-fledged ones, this result suggests that getting high dominance by fledging early may be more advantageous in males. Parallel to ecological data, mitochondrial DNA sequences were analyzed and molecular phylogeny was examined. Our molecular data suggest the paraphyly of the genus Pica, which was also supported by colonial breeding in North American and yellow-billed magpies. Although further investigation on molecular and ecological aspects should be conducted, we suggest that the current classification needs to be revised. A long-term phylogeographic study such as ours will be crucial to the understanding of behavior, ecology, and evolution of any species.

 

 

INTRODUCTION

 

The magpie is a member of the crow family, Corvidae, which consists of 102 species. Like the other members of Corvidae, magpies are very common in the Northern Hemisphere, ranging from Northeastern Asia to Europe (Vaurie 1959; Goodwin 1976). The genus Pica is comprised of 2 species, P. pica, the black-billed magpie, and P. nuttalli, the yellow-billed magpie. Pica nuttalli is restricted to a small area in California, whilst P. pica is found globally. The black-billed magpie is currently classified into 13 subspecies (or races), based on the geographic distribution and morphological traits (Vaurie 1959).

Until now, P. nuttalli has been considered as a distinct species mainly because of its morphological character, and it has been thought that behavioral and ecological similarities between P. nuttalli and North American black-billed magpie P. p. hudsonia are due to convergence (Fig. 1 a). However, more and more of behavioral and ecological evidence are being accumulated suggesting that P. nuttalli may be phylogenetically closer to P. p. hudsonia (Birkhead 1991; Fig. 1 b). This possibility was also suggested recently by Sibley & Ahlquist (1990) based on DNA hybridization data.

Still the phylogenetic relationship within the genus Pica remains unclear. Moreover, there have been only a few ecological reports so far and no biochemical study on Asian populations (examples of ecological studies include Takeishi & Eguchi 1994; Eguchi 1995; Eguchi & Takeishi 1997).

 

 

 

 

 

 

 

 

 


Figure 1.  Suggested phylogenies of the genus Pica. Evolution of colonial breeding was indicated by tick markers on branches.

 

 

Pica p. sericea is found found from Amurland, Manchuria, southern China to Indochina, Korea and Japan (Goodwin 1976), and widely distributed throughout Korea. Magpies are known as a bird species that successfully adapted to human habitations. In recent years, as human population increases and many forests are cleared in Korea, magpies have been expanding their distribution into urban areas. Although magpies are found very common in urban areas, almost no report has been made on the ecology of magpies except Lee & Koo (1986) and Koo & Kim (1986).

We launched a long-term monitoring project on the magpie population on the Kwanak campus of Seoul National University. Magpies on the campus of Seoul National University form a stable population where nearly 70 or more active nests are found annually. The birds are well accustomed to human presence, which permits close observation on their behavior and ecology.

The aim of this paper is to present our results of long-term research on both behavioral ecology of magpies and phylogenetic relationships among magpie subspecies.

 

 

MATERIALS AND METHODS

 

Field study

 

The black-billed magpie population on the campus of Seoul National University has been observed since 1998. Nest visits were made at least twice a week during the breeding season (from March to June). Although we observed mainly with binoculars in order not to disturb the birds, we also checked the status of nests by approaching the nest with rope, tall ladder, and/or cargo-crane when it was necessary.

Data on egg characteristics and nestling development were collected in 2000. During the incubation period, we measured the lengths of long axis, short axis, and weights of eggs. Because egg weight decreases as the embryo develops, we used volume index (Birkhead 1991) when comparing egg sizes. Nestling development was measured with tarsus lengths and weights of chicks.

When 14~20 days old, nestlings were marked with patagial wing tags and colored leg bands and all the measurements of body parts were made. Individuals can be recognized either by the two English letters written on wing tags or color combination of leg bands. All marked birds were observed subsequently.

In 1998 and 1999, territory mapping was conducted during the incubation and feeding period. In this period, territories are settled around nests and magpies defend their territories vigorously (Erpino 1968). Positions of the pair were marked on a map and territorial boundary was determined via maximum polygon method (Odum & Kenzler 1955).

 

Sex determination of chicks

 

We used molecular sexing techniques of Griffiths et al. (1998). This technique adopts amplification of different introns within CHD (chromodomain-helicase DNA) genes on Z and W chromosome.

Blood samples were collected by brachial wing vein puncture at least once when nestlings were 3~20 days old. Blood droplet was preserved in either 30~50STE buffer or 70% EtOH. The samples were kept in a deep freezer (-20). Genomic DNA was extracted with DNeasy Tissue Kit (QIAGEN) following the protocol of isolation of genomic DNA from whole nucleated blood. PCR reaction was conducted with P8-P2 primer pairs (Griffiths et al. 1998). When the PCR products were separated on 2.5% agarose gel, two bands appeared from sample of female while only one band from that of male.

 

Molecular Phylogeny

 

Partial nucleotide sequences of mitochondria tRNA-Leu, NADH dehydrogenase subunit 1 (ND1) and ND2 were used in our phylogenetic analysis. We used 19 samples of 11 taxa. Samples used in our analysis were 2 or 3 samples of Korean (Pica pica sericea), Kamchatkan (P. p. camschatica), European (P. p. pica), and North American black-billed magpies (P. p. hudsonia), yellow-billed magpies (P. nuttalli), and 6 other outgroups. Outgroups represent possible sister taxa and proximal outgroups to the genus Pica within the Family Corvidae (Cyanopica, Cissa, and Urocissa) as well as slightly more distant ones (Platylophus and Corvus). Korean samples were collected by us, while others were taken from the specimens at Burke Museum of the University of Washington, Museum of Zoology of the University of Michigan, Louisiana Museum of Natural History at Louisiana State University, and personal collections.

MtDNA was extracted from the heart and pectoral muscle using QIAamp Tissue Kits (QIAGEN) and from feather quills by the addition of dithiothreitol to the QIAamp protocol. The primer pairs used in PCR were L3827-H4644 for part of 16s rDNA, tRNA-Leu and ND1 and L5216-H5766 for part of ND2 (Sorenson et al. 1999). PCR amplifications were performed using Taq DNA polymearse and reaction buffer (500mM KCl, 100mM Tris-HCl, 1.0% Triton X-100, and 25mM MgCl2) from VIOGENE. Reaction mixtures were subjected to 35 cycles of denaturation at 94C for 30 s, annealing from 53 to 55C for 20 s, and extension at 72C for 80 s. PCR products were purified from the agarose gels using QIAquick Gel Extraction Kits (QIAGEN). Sequencing was done with the ABI 377 automated sequencer at the University of Michigan and the KAIST BioMedical Research Center. We manually reconciled automated base pair calls by comparing complementary strands using ABI's Sequence Navigator program. For phylogenetic analyses, we aligned the sequences with ClustalX (Thompson et al. 1994) and sought phylogenetic trees with both maximum parsimony (branch-and-bound option) and neighbor-joining method using PAUP ver 4.0b1 (Swofford 1998). Bootstrap was implemented with a branch-and-bound search with 500 replicates.

 

RESULTS

 

Breeding Success

 

Total 74, 80, and 88 active nests were found in 1998, 1999 and 2000, respectively, on the Kwanak campus of Seoul National University. Average breeding successes (measured as fledging successes) were 2.81.6, 3.31.7, 2.41.4 fledglings per nest (meanSD) in 1998, 1999, 2000, respectively and these were not significantly different (ANOVA, F2,105=0.02, P>0.05). Proportions of successful nests were 43.2% (32/74), 53.8% (43/80), 50.0% (44/88), respectively. Even if the proportion was higher in 2000 than in 1998, fledging success was lower in 2000 (Fig. 2). Repeat nestings were recorded on 7 breeding pairs in 1999 and 11 pairs in 2000. Excluding repeat nests, proportions of successful breeding pairs were 58.9%, 55.8% in 1999 and 2000 respectively.

 

 

 

 

 

 

 

 

 

 

 

 

 


Figure 2.  Annual differences in the breeding success (measured as fledging success) between 1998, 1999, and 2000. Solid circles with error bars denote mean (SD) number of fledglings per nest, and boxes denote proportions of successful nests.

 

The slight differences in the fledging success during the three years can be ascribed to climate condition (Fig. 3). Low fledging success in 2000 seems to be due to the low temperature during incubation compared to other years. While average precipitation was much higher in 1998 than in other years, it was not reflected in breeding success. It appears that temperature during incubation is crucial for hatching, while rainfall during feeding is not so important for chick survival.

 

 

 

 

 

 

 

 

 

 

 

 


Figure 3.  Climate changes in 1998, 1999, and 2000.

 

 

It is widely known in birds that breeding success within a population decreases as breeding season proceeds (Birkhead 1991). Magpie population on SNU campus also showed seasonal decline of breeding success in 1998 and 1999, although the decrease was not statistically significant (Fig. 4a and b). Unlike other years, breeding success did not show seasonal decline in 2000 (Fig. 4c). This implies that seasonal change in breeding success can be different from year to year. The increase in 2000 may be due to the fact that early nests failed because of the low temperature during the incubation period.

The most common cause of breeding failure was laying and hatching failure (Fig. 5; 48.3~61.5% of the total breeding failure). The frequency of hatching failure was especially high in 2000, which seems to be related to the unusually low temperature during incubation. Renesting was quite common (9.6% in 1999, 14.3% in 2000), although it is still less than in a Japanese population (18.8%; Eguchi 1995). The reason for renesting does not seem to be restricted to breeding failure, because four nests were repeated after successfully fledging their young in 2000. In those cases, however, repeat nests were not successful because incubation time was not sufficient due to time loss to feeding fledglings and repeat nests were abandoned eventually.

 

 

 

 

 

 

 

 

 

 

 

 

 

 


Figure 4.  Seasonal change in breeding success. (a) n=23, P<0.05; (b) n=31, P<0.01; (c) n=19, P>0.05.

 

 

 

 

 

 

 

 

 

 


Figure 5.  Causes of breeding failure. Note that half of breeding failure can be ascribable to laying or hatching failure.

 

 

Egg size was variable both within and among nests (Table 1). Within nests, larger eggs were more likely to hatch than smaller eggs (size as volume index, one-tailed t-test; t=2.76, df=73, P<0.01).

 

 

 

 

 

 

 

 

Territoriality

 

Black-billed magpies on SNU campus defended territories of approximately 1 ha and average distance to the nearest active nest was 93 m (Table 2). If we assume that the nest is located at the center of each territory and territory is round, then the distance to neighboring nest would be 56.4 m, which is quite smaller than actual distance. This calculation implies that quite a few territory boundaries were overlapped and territorial interactions occur frequently, which was confirmed by field observations and territory mapping.

Terrioriality showed no significant difference between 1998 and 1999 (Table 2; (a) territory size: two-tailed t-test, t=1.77, df=85, P>0.05; (b) distance to the nearest active nest: ANOVA, F2,165=0.043, P>0.05). Quite unexpectedly, little correlation was found between territory size and number of fledglings (in 1998, r= -0.30, P>0.05; in 1999, r=0.07, P>0.05). The only significant difference pertaining territoriality was that the territories containing mostly trees were significantly smaller than those with buildings or grass fields (one-tailed t-test; t=5.22, df=49, P<0.01). Even in that comparison, fledging successes were not significantly different between two types of territories (two-tailed t-test; t=0.68, df=49, P>0.05).

 

 

 

 

 

 

 

 

 

 

 


Offspring Sex Ratio

 

We compared male proportions in early (fledged before Mid-May) versus late nests (fledged after Mid-May). Male proportions in the early nests were significantly higher than those in the late nests (Fig. 6; one-tailed t-test; t=4.23, df=22, P<0.001). When pooled, overall sex ratios were not biased within a year (2-test; 2=2.16, df=12, P>0.05 in 1998; 2=1.41, df=12, P>0.05 in 1999).

 

 

 

 

 

 

 

 

 

 

 

 


Figure 6.  Comparison of male offspring proportions between early and late nests. Error bars denote standard deviations.

 

 

Regression analyses showed that weight increase was slightly higher in male chicks than in female chicks, although increase in tarsus length was similar in both sexes (Fig. 7). Our results accord with those of Birkhead (1991) who found that tarsus length was less variable at a given age and food availability.

 

Molecular Phylogeny

 

The nucleotide sequences of up to 1338 base pairs were determined. Mean base proportions were: 25.32% T, 30.85% C, 30.05% A, and 13.78% G. Sequence identity ranges from 50.0% (outgroup versus European sample) to 96.0% (between the European samples). 979 (73.2%) base pairs were invariant and 198 (14.8%) base pairs were parsimony-informative.

Analysis with PAUP found single most parsimonious tree of length 474, with consistency index 0.827 (Fig. 8). The same tree topology was also found by neighbor-joining method. The topology of the tree was so

Development in magpie chicks. Males were noted with closed squares and females with open circles. (a) tarsus, male: y=11.6∙ln(x)+11.8, r=0.83, P0.001; female: y=13.1∙ln(x)+8.25, r=0.81, P0.001; (b) weight, male: y=78.6∙ln(x)-70.2, r=0.90, P0.001; female: y=70.5∙ln(x)+52.2, r=0.82, P0.001

 

 

 

 

 

 

 

 

 

 

 


Figure 7.  Development in magpie chicks. Males were noted with closed squares and females with open circles. (a) tarsus, male: y=11.6∙ln(x)+11.8, r=0.83, P0.001; female: y=13.1∙ln(x)+8.25, r=0.81, P0.001; (b) weight, male: y=78.6∙ln(x)-70.2, r=0.90, P0.001; female: y=70.5∙ln(x)+52.2, r=0.82, P0.001

 

 

robust that no changes were noted with changes in parameters. High bootstrap values (66~100%) also support the branching patterns.

The phylogenetic relationships among the Palearctic subspecies were revealed, and European magpie was found to be closer to the Kamchatkan. Korean magpie formed a basal group of the genus Pica. The most noteworthy point is that P. p. hudsonia and P. nuttalli grouped together, by which the genus Pica is recognized as a paraphyletic group.

 

 

 

 

 

 

 

 

 

 

 


Fig. 8. A single most parsimonious tree found from branch-and-bound method. Numbers on the branches indicate numbers of characters support the branch/bootstrap value with 500 replications (tree length=474, CI=0.8270, HI=0.1730, RI=0.7892, RC=0.6527).

 

DISCUSSION

 

Breeding Ecology

 

Although breeding successes did not differ significantly among the three years, patterns of seasonal change in breeding success were different from year to year. These differences in seasonal changes appeared partly due to climate condition, especially temperature during incubation. Similar report was made by Crick & Sparks (1999), who found the temperatures in March and April affected laying dates in 17 bird species through an examination of long-term trends in laying dates. They also found that larger-bodied species seemed to show little response to temperature, which may explain statistical insignificance of differences among breeding successes of three years in our study.

Seasonal decline in breeding success observed in 1998 and 1999 is a fairly common pattern among birds including magpies and possible explanations includes: (i) young birds and/or poor-quality birds breed late, lay small clutches, and have low successes (Birkhead 1991; Goodburn 1991); (ii) repeat nests have low successes (Birkhead 1991); (iii) decline in food availability (Birkhead 1991); (iv) differences in territory quality (Högstedt 1980). From our results only, we cannot conclude any of these explanations are relevant in magpies. Concerning explanation (ii), we did not have any comparable data on repeated breeding attempts because magpies succeeded only once if they bred twice. Concerning explanation (iv), we found that territory size had no correlation with breeding success, although we had no direct data on territory quality. Recently, Eguchi (1995) added that the decline may be due to increased nest predation in late nests. However, we found no evidence of predation in our study population. In order to discern the factors affecting seasonal change in breeding success, measurements on territory quality and bird (parent) quality should be made.

It is quite surprising that territory size and fledging success appear to have almost no correlation. This suggests that territory quality such as abundance of terrestrial invertebrates is heterogeneous according to the vegetation types within territory, which is confirmed by significantly smaller sizes of territories with dense vegetation. It seems that magpies adjust their territory size according to food availability rather than area they can defend.

It appears that laying and hatching are the most critical factors in magpie breeding. A similar phenomenon was found in European magpies (Baeyens 1981). Difficulty in hatching may be related to insufficient incubation time due to frequent territorial interactions and low temperature during incubation.

Timing of breeding seems to be determined by trade-offs between benefits and costs of early breeding. Early-fledged young is presumed to achieve a higher dominance rank when flocks are formed among yearlings (Gerstell & Trost 1997). In addition, parents can increase their reproductive success through additional breeding attempts, if they are successful in repeat nesting.

Nonetheless, early breeding burdens parents with increased incubation load due to low temperature. Considering that fledging weight is higher in males, male chicks can be regarded as being more costly to rear than female chicks. Thus, the cost of early breeding is twofold; increased incubation load and rearing high-cost broods. However, early breeding pairs can enhance their reproductive success by rearing costly but rewarding males because early-fledged males are more likely to get higher dominance ranks, get territories of good quality, and eventually produce more offsprings of good quality, as was found in tree swallows (Whittingham & Dunn 2000).

Male-biased sex ratios in early nests of can be explained by early bird hypothesis which predicts that competition for breeding sites among dispersing males confers an advantage to early-fledging males (Smallwood & Smallwood 1998). In our study population, opportunities for a newly-formed pair to get a territory seem to be extremely rare and magpies were not observed to attempt breeding until the third year. Considering the exceptionally high breeding density at the study site (Table 2) and that European magpies were found to breed even in their first year (Birkhead 1991), we can easily imagine how difficult it is for young magpies to squeeze themselves in already tightly packed territories. Under this situation, it would be very important to achieve higher dominance ranks by fledging early. Although we do not have any data on sex-differential dispersal, male-bias in early nests suggests that it is more important for males to fledging early.

Developmental differences in weights between male and female chicks may be due to several factors such as aggressiveness in male chicks leading to decrease in food uptake by female chicks and/or male-biased parental investment. Comparison of developmental rates in mixed-sex broods versus single-sex broods and direct observation of begging intensity of chicks and parents feeding behavior would enable us to discriminate factors affecting differential development according to sex.

 

Phylogenetic Relationship

 

The phylogenetic relationships within the genus Pica revealed in our molecular analyses can be summarized as follows:

 

(1) The European magpie (P. pica pica) was close to the Kamchatkan (P. p. camschatica),  

(2) The Korean magpie (P. p. sericea) formed a basal group of the genus Pica,  

(3) The North-American races (P. p. hudsonia and P. nuttalli) were grouped together.

 

Thus, our molecular data do not support the monophyly of the genus Pica. As for now, the monophyly of the genus Pica seem to be supported by morphological traits of P. nuttalli and fossil records. However, the morphological traits of yellow bill, which was previously thought to be derived characters only occurring to P. nuttalli, were also found in European magpies and the genetic basis for the trait was suggested to be unstable (Birkhead 1991). Voous (1960) also supported the monophyly of the genus Pica based on fossil records. He suggested that that after ancestral Asian population migrated to North America, North American population became extinct during the last glacial period except the small relict area of California, and this small population diverged into the yellow-billed magpies and current black-billed magpies in North America was formed through recolonization from Asian population. According to his hypothesis, P. p. hudsonia should be closer to P. p. camschatica, and this subspecies closer to P. p. sericea, which would be quite different from our results. Our results suggest that the scenario based on fossil records needs to be thoroughly re-examined. Rather than extinction and recolonization of P. p. hudsonia, our results propose that P. nuttalli and P. p. hudsonia share their ancestor and recently diverged from each other.

This possibility also can be found in comparison of their ecological traits. When we compared territoriality of subspecies, we found that P. p. sericea and P. p. pica breed solitarily, while P. nuttalli and P. p. hudsonia colonially (Table 3). It seems that aggregated nesting were obtained somewhere between Asian ancestor and North American descendant population and colonial breeding was firmly established in P. nuttalli.

However, we cannot exclude the possibility that evolution rate of molecules may be different from that of morphology or ecology. Different rates of morphological and molecular evolution also have been documented in other avian species (Avise & Zink 1988; Zink et al. 1995; Miranda et al. 1997). The derived phenotypic features of P. nuttalli compared to P. p. hudsonia suggest the occurrence of rapid differentiation. A founder event followed by genetic drift and geographic isolation appears to have played a role in rapid morphological speciation in P. nuttalli.

Hypothesis on behavioral phylogeny could be drawn only after the data are accumulated through several years. In addition, both behavioral ecology and molecular data on other subspecies should be included in order to clarify the phylogenetic relationships within the genus Pica. We suggest that appropriate taxonomic decision should be made after a careful re-examination on the degree of divergence among the subspecies based on morphology, behavior and ecology and molecular data.

 

Table. 3. Subspecies comparison of territoriality (modified from Birkhead 1991).

Feature

Pica pica pica
P. p. hudsonia

P. p. sericea

P. nuttalli

Breeding unit

Single pair

Single pair

Single pair

Colonies of 5–30 pairs

Nest spacing

Even

Highly aggregated

Aggregated

Highly aggregated

Distance to neighboring nests(m)

100

80

90

50

Territory occupancy

Throughout year

Breeding season only

Throughout year

Throughout year

Timing of territory defense

Pre-laying

None?

Pre-laying

Pre-laying

Territorial interactions

Frequent

Infrequent

Frequent

Frequent

Territory function

Food supply,

Mate guarding

Nest defense

Food supply,

Mate guarding(?)

Food supply,

Mate guarding

Feeding areas

Within territory

Up to 400m from nest

Within territory,

Flock areas

Within territory,

Flock areas

 

 

REFERENCES

 

Avise, J. C. and R. M. Zink. 1988. Molecular genetic divergence between avian sibling species: King and Clapper rails, Long-billed and Short-billed dowitchers, Boat-tailed and Great-tailed grackles, and Tufted and Black-crested titmice. Auk 105: 516-528.

Baeyens, G. 1981. Magpie breeding success and carrion crow interference. Ardea 69: 125-139.

Birkhead, T. 1991. The Magpies. London: T&AD Poyser.

Crick, H. Q. P. and T. H. Sparks. 1999. Climate change related to egg-laying trends. Nature 399: 423-424.

Eguchi, K. 1995. Seasonal change in breeding success of the black-billed magpie Pica pica sericea. Japanese Journal of Ornithology 44: 73-80.

Eguchi, K. and M. Takeishi. 1997. The ecology of the black-billed magpie Pica pica sericea in Japan. Acta Ornithologica 32(1): 5-37.

Erpino, M. J. 1968. Nest-related activities of black-billed magpies. The Condor: 70: 154-165.

Gerstell, A. T. and C. H. Trost. 1997. Advantages of early versus late nesting in black-billed magpies Pica pica hudsonia: variation in productivity, sex ratio, and fledgling size. Acta Ornithologica 32(1): 39-44.

Goodburn, S. F. 1991. Territory quality or bird quality? Factor determining breeding success in the magpie Pica pica. Ibis 133: 85-90.

Goodwin, D. 1976. Crows of the World. London: British Museum.

Griffiths, R., M. Double, K. Orr, and R. Dawson. 1998. A DNA test to sex most birds. Molecular Ecology 7: 1071-1076.

Högstedt, G. 1980. Evolution of clutch size in birds: adaptive variation in relation to territory quality. Science 210: 1148-1150.

Koo, T. and J. Kim. 1986. Breeding ecology of magpie, Pica pica sericea Gould, in Korea. Nature Conservation 56: 37-48.

Lee, D. and T. Koo. 1986. A comparative study of the breeding density of magpies Pica pica sericea Gould, between urban and rural areas. Bulletin of the Institute of Ornithology, Kyung Hee University 1: 39-51.

Miranda, H. C., R. S. Kennedy, and D. P. Mindell. 1997. Phylogenetic placement of Mimizuku gurneyi (Aves: Strigidae) inferred from mitochondrial DNA. Auk: 114(3): 315-323.

Odum, E. P. and E. J. Kuenzler. 1955. Measurement of territory and home range size in birds. Auk 72: 128-137.

Sibley, C. G. and J. Ahlquist. 1990. Phylogeny and Classification of Birds. New Haven: Yale University Press.

Smallwood, P. D. and J. A. Smallwood. 1998. Seasonal shift in sex ratios of fledgling American kestrels (Falco sparverius paulus): the early bird hypothesis. Evolutionary Ecology 12: 839-853.

Sorenson, M. D., J. C. Ast, D. E. Dimcheff, T. Yuri and D. P. Mindell. 1999. Primers for a PCR-based approach to mitochondrial genome sequencing in birds and other vertebrates. Molecular Phylogenetics and Evolution 12(2): 105-114.

Swofford, D. L. 1998. PAUP: Phylogenetic Analysis Using Parsimony, Version 4.0. Beta Version 4.0b1. Sunderland: Sinauer Associates.

Takeishi, M. and K. Eguchi. 1994. Nest-site characteristics in the black-billed magpies Pica pica sericea. Japanese Journal of Ornithology 42: 53-59.

Thompson, J. D., T. J. Gibson, F. Plewniak, J. Meanmougin, and D. G. Higgins. 1997. The ClustalX windows interface: Flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research 24: 4876-4882.

Vaurie, C. 1959. The Birds of the Palearctic Fauna. London: Witherby.

Voous, K. H. 1960. Atlas of European Birds. London: Nelson. 

Whittingham, L. A. and P. O. Dunn. 2000. Offspring sex ratios in tree swallows: females in better condition produce more sons. Molecular Ecology 9: 1123-1129.

Zink, R. M., S. Rohwer, A. V. Andreev, and D. L. Dittmann. 1995. Trans-Beringia comparisons of mitochondrial DNA differentiation in birds. Condor 97: 639-649.