i
A Survey of Brown-Headed Cowbird Parasitism in North-central Florida and Mechanisms Explaining Cowbird Distribution
Project Director: Matthew J. Reetz1 Coauthors: Kathryn E. Sieving1 and Scott K. Robinson2 Date of Final Report: October 15, 2007 Project Number: SWG04-031
1Department of Wildlife Ecology & Conservation, University of Florida, P.O. Box 110430, Gainesville, FL, 32611-0430 2 Florida Museum of Natural History, 305 Dickinson Hall, Gainesville, FL 32611 i
ABSTRACT
Despite conservation implications surrounding recent brown-headed cowbird (Molothrus ater) expansion into Florida, little data currently exist on the species in the state. Our objectives for this project were to: 1) quantify parasitism rates on host species to determine the extent of parasitism and identity of host species parasitized, 2) experimentally test prevalence of egg rejection behaviors in common Florida songbird species, and 3) determine actual cowbird fecundity in north-central Florida. We surveyed parasitism of 32 host species breeding in Florida through nest monitoring and observations of family units with fledglings. Approximately 1.5% of >1000 nests and 6.25% of 272 fledge families contained a cowbird egg or young. The majority of fledge families containing cowbird young were of canopy nesting species (88.2%). Because observed parasitism rates were low across the host community, we tested egg rejection behavior of four common Florida songbird species. We experimentally parasitized 78 nests using eggs obtained from a captive cowbird colony. Cowbird eggs were accepted by hosts in 48 cases (61.5%). Brown thrashers were the only species to consistently eject the eggs (14 of 17 cases), while northern mockingbirds, northern cardinals, and red-winged blackbirds typically accepted experimental cowbird eggs. Finally, to separate observed from actual parasitism rates, we compared ovarian reproductive condition of breeding female cowbirds collected in Florida and Texas. Based on our reproductive assays, Texas and Florida cowbirds lay eggs from late April until the first week of July. Oviduct and ovary mass increased significantly at onset of breeding when follicles began to grow and oviduct tissue increased to allow development and passing of eggs. Average clutch sizes of cowbirds from Florida and Texas were 3-4 eggs, with a non-laying interval of 2-3 days. Fecundity does not appear to be limited in either Florida or Texas, where each breeding female produces ca. 35 eggs during the breeding season, an estimate consistent with cowbird fecundity measured elsewhere. Our data indicate that fecundity of Florida cowbirds is high yet parasitism of Florida songbirds appears to be limited in north-central Florida based on our parasitism surveys. Our experiments indicate that ejection behavior may not be retained in most species practicing it in other areas with longer exposure to cowbirds. This suggests that defense of nests does not limit parasitism in north-central Florida. Our data suggest that some species, particularly certain canopy nesting species, may be preferentially parasitized by cowbirds so that a few species may be bearing the brunt of the cowbird expansion into Florida. ii
ACKNOWLEDGMENTS
Invaluable field help was provided by Andrew Cook, Steve Daniels, Dan Dawson,
Elizabeth Bauman, Nia Haynes and Rebecca Hamel. We thank the Florida Department of
Environmental Protection, Danny Driver, David Armstrong, Joel McQuagge, Jesse Smalls, Steve
Coates and Bruce Christensen of the University of Florida, and Geoff Parks of the Gainesville
Nature Operations Division for access to field sites. We are indebted to Michael Avery, Eric
Tillman, Kandy Keacher and Michael Milleson of the USDA National Wildlife Research Center for support with trapping and maintenance of captive cowbirds. Brian Branciforte and Kate
Haley of the Florida Fish and Wildlife Commission and Monica Lindberg, Caprice McCrae,
Laura Hayes and Sam Jones of UF provided valuable logistical support. Research design of this project was improved through comments of various avian discussion groups. Research protocols
D539 and D954 were approved by the UF Institutional Animal Care and Use Committee.
Finally, this work was made possible by the Florida Fish and Wildlife Conservation
Commission‟s program, Florida‟s Wildlife Legacy Initiative, and the U.S. Fish and Wildlife
Service‟s State Wildlife Grants program (SWG04-031).
iii
TABLE OF CONTENTS
INTRODUCTION ...... 1
Objectives ...... 3
METHODS ...... 4
Nest Monitoring and Observations of Family Units ...... 4
Experimental Parasitism ...... 5
Cowbird Collection ...... 6
Reproductive Assays ...... 8
RESULTS ...... 13
Parasitism Frequency ...... 13
Experimental Parasitism ...... 15
Cowbird Collection ...... 16
Reproductive Assays ...... 17
DISCUSSION ...... 35
Patterns of Parasitism ...... 35
Maintenance of Defense Behaviors ...... 37
Cowbird Fecundity...... 38
Management Implications ...... 39
LITERATURE CITED ...... 42
APPENDIX A ...... 45 1
INTRODUCTION
The brown-headed cowbird (Molothrus ater) is among three brood-parasitic cowbird species in North America, and the only one native to the continent. It was originally restricted to the grasslands of the Great Plains region of central North America where it foraged with bison herds, parasitizing locally-breeding host populations as they moved. With the arrival of
European settlers that caused habitat fragmentation, destruction of bison populations, and establishment of permanent feeding areas, the cowbird began expanding in distribution. For example, until the 1800s, brown-headed cowbirds were found uncommonly in the eastern United
States, probably due to lack of availability of suitable feeding habitats among large unbroken tracts of forest (Mayfied 1965). However, with the increase in human populations, cowbirds began to appear in states successively farther east (Lowther 1993) and local records of cowbirds changed from rare or absent to common or abundant (Mayfied 1965). Presently, the breeding range of the brown-headed cowbird now includes the entire contiguous United States and much of North America (Sauer et al. 2006).
The brown-headed cowbird is a generalist obligate brood parasite, and is commonly cited as laying eggs in the nests of over 220 potential host species (Friedmann and Kiff 1985).
However, prior to human alteration of the landscape, cowbirds were naturally limited to host species with which they were sympatric during the breeding season. The spread of the brown- headed cowbird brought it into contact with a new and diverse suite of host species and populations, most of which lacked at least recent historical exposure to brood parasitism. For some songbird species, this led to significant population declines as cowbird populations steadily increased over the last 100 years (Brittingham and Temple 1983).
In many parts of North America, brood parasitism by the brown-headed cowbird is recognized as having a significant negative effect on the reproductive success of host species and
2 communities (Friedmann et al. 1977, Robinson et al. 1995). Effects of cowbird parasitism are particularly evident at local scales and with single species. For example, cowbird parasitism is thought to have resulted in the extirpation of local populations of Bell‟s vireo (Vireo bellii) in
California (Laymon 1987), and threatens extinction of species with limited ranges such as the
Kirtland‟s warbler (Dendroica kirtlandii; Mayfield 1992). Cowbirds limit reproduction of host species through removal or damage to host eggs and reduced hatching or fledging success of host young (although net reproductive success may not be reduced to due renesting; Ortega and
Ortega 2000). As a result, brown-headed cowbirds have become a regional management problem for particular songbird species and are often aggressively controlled (De Capita 1993).
Cowbird range expansion has been negatively correlated with human land use changes, particularly fragmentation of forests and the proliferation of livestock feeding areas (Ortega
1998) and grain-based agriculture. Expansion into new regions has particular implications for species without evolutionary exposure to brood parasitism. Patterns of cowbird spread and the corresponding incidence of parasitism have varied regionally. In the southeastern Figure 1. Breeding Bird Survey trend graph of U.S., brown-headed cowbirds (M. a. ater) Brown-headed Cowbirds in Florida, 1966-2005. were first recorded breeding in far western South Carolina in 1934 (Sprunt and Chamberlain
1970) but had spread across the state and were breeding near Charleston by 1965 (Cutts 1964).
Cowbirds were first recorded breeding in Georgia in 1945 near Augusta on the South Carolina border (Denton 1946) and in the Atlanta area in 1949 (Parks 1950). Cowbird populations have risen somewhat slowly (ca. 1.0 and 0.8% per year, respectively) since cowbirds first were
3 confirmed breeding in these states. Proportions of parasitized nests of common songbirds in the mid to late 1990s were at moderate levels (8.2-13.0%) in Georgia and South Carolina with some as high as 35% (Whitehead et al. 2002, Kilgo and Moorman 2003).
Brown-headed cowbirds were also first recorded as breeding in Florida during the mid-
1900s, when a fledgling was sighted in the panhandle (Monroe 1957) and had reached the north eastern coastal plain by the mid-1960s (Ogden 1966). Like in other portions of the Southeast, cowbird populations have grown somewhat slowly in Florida with a 1.9% increase per year between 1966-2005 (Sauer et al. 2005; Figure 1), yet are confirmed breeding in 50 of 67 counties
(Florida Fish & Wildlife Conservation Commission 2007). However, in contrast to other southeastern states, parasitism rates are extremely low across the host community (1.5%, see below) in north-central Florida where relative abundance of cowbirds is similar to other southeastern states (Sauer et al. 2005). These rates are surprising given that most of Florida‟s approximate 50 breeding songbird species have been parasitized and have reared young in other areas (Ortega 1998). Isolated records of parasitism of Florida breeding songbirds indicate that many species may be at risk to brown-headed cowbird expansion (see review in Cruz et al.
1998). Despite potential impacts of cowbird parasitism for native breeding Florida songbirds, little data are available on brown-headed cowbird parasitism in Florida.
Objectives
Specific objectives for the field component of this project are twofold: 1) quantify rates of parasitism affecting many Florida host species in north and central Florida where cowbirds are abundant and, 2) evaluate general mechanisms to explain current distributions of brown-headed cowbirds. Specifically, we examined the prevalence of egg rejection behaviors in Florida
4 species and compared fecundity of Florida cowbirds to a region where they are a conservation problem. To do so, we cooperated with federal (USDA), state (Florida DEP), private (Nature
Conservancy) and local (Gainesville Nature Operations) agencies.
METHODS
Nest Monitoring and Observations of Family Units
We surveyed nests from a diversity of potential hosts to determine natural parasitism rates in north-central Florida where brown-headed cowbirds are common. Nest searches were conducted daily in a variety of habitat types where brown-headed cowbirds occur. All sites were located within 25 km of Gainesville, FL (ca. 29º39‟ N, 82º19‟W; Figure 2) in Alachua and
Putnam Counties, and included pine flatwoods, mesic hardwood hammocks, bottomland forest, wet prairie, and shrubland/scrub habitats. Sites encompassed a gradient from urban (downtown
Gainesville) to rural (no or low nearby housing density), from semi-natural (e.g., city parks) to natural (e.g., managed state preserves), and from disturbed (e.g., regenerating rangeland) to quasi-natural (large forests in state preserves). Nests were monitored every 1 to 3 days. Nest data were supplemented from a nest monitoring project (Sieving & Contreras, unpubl. data) in
Myakka River State Park near the southwest Florida gulf coast in Sarasota and Manatee Counties
(ca. 27°14‟ N, 82°14‟ W).
There are usually little breeding biology data on canopy species such as red-eyed vireo
(Vireo griseus), northern parula (Parula americana), and pine warbler (Dendroica pinus) because nests are generally too high or difficult to monitor. Nests of other species such as
Carolina wren (Thryothorus ludovicianus) may also be under-represented in samples due to problems locating highly cryptic nests. Therefore, to better assess species parasitized by cowbirds, we performed observations of family units with fledglings. Monitoring of adults
5 feeding young outside of the nest is the main component of reproductive indices used to estimate reproductive success (Vickery et al. 1992) and are useful when nest monitoring is logistically infeasible (Bonifait et al. 2006). Families were found by watching foraging adults and following adults carrying food or by tracking fledgling begging calls. Fledglings of some species give more audible begging calls, but we attempted to listen for any call that might represent feeding of a fledgling or the fledgling itself. Families were followed for a minimum of ten minutes to determine if a cowbird fledgling was present. We assumed that cowbird fledglings were fed by the host adults that raised them, and not by a non-host species responding to begging calls.
Lower-nesting species tend to be better represented in samples of the entire community of birds, except those that are locally rare (e.g., orchard oriole Icterus spurius) or that are cryptic (e.g., common yellowthroat Geothlypis trichas). As a result, we conducted visual scans for adult birds in the mid- to upper canopy more regularly than for those lower to the ground, the fledge families of which were located opportunistically when searching for and monitoring nests.
Experimental Parasitism
To examine limitations on cowbird parasitism due to behavioral defenses, we imposed parasitism on select species using brown-headed cowbird eggs obtained from captive birds by placing them in nests of host species. Cowbird captivity protocol was developed from the literature and refined through consultation with numerous researchers. In September 2004, we captured 40 hatch-year cowbirds (20 male, 20 female) in Alachua County, FL using baited drop- down decoy traps. In March 2005, 10 pairs were separated for breeding in individual cages and fed a specialty diet that met caloric and dietary needs for reproduction (Holford and Roby 1993).
Each cage contained two or more open-cup or domed artificial nests and infrequently an inactive northern cardinal (Cardinalis cardinalis) nest removed from the field. Some artificial nests
6 contained a single wood artificial egg that was similar in size, appearance, and weight to the egg of a red-eyed vireo. To improve the probability that a captive female cowbird would lay eggs, cowbird pairs were housed with zebra (Taeniopygia guttata) and society finches (Lonchura striata domestica). These species bred readily in captivity and provided stimuli associated with active bird nests (e.g., nest-building, incubation behavior, warm eggs). Cages were monitored daily for the presence of cowbird eggs.
Eggs obtained from captive birds were used to experimentally-parasitize four primary species that are abundant during the breeding season in Florida. Species chosen are known to respond to cowbird parasitism through ejection or nest abandonment behaviors in other regions; brown thrasher (Toxostoma rufum), northern mockingbird (Mimus polyglottos), northern cardinal, and red-winged blackbird (Agelaius phoeniceus). A single experimental cowbird egg was placed in each host nest between the egg-laying stage and 2-3 days following initiation of incubation as determined by candling of eggs (Lokemoen and Koford 1996). Eggs were warmed slightly in the hand prior to being placed in the nest. Nests were visited daily and host responses
(e.g., ejection, acceptance) determined based on Rothstein (1975). Eggs were considered
„ejected‟ if the cowbird egg or fragments were found outside the nest, or if the egg was missing with no apparent decrease in numbers of host eggs. Nests were considered „abandoned‟ if contents were undisturbed with no sign of nest activity, or if a new layer of material covered the eggs. If no change occurred on subsequent days and the nest remained active, the egg was considered „accepted‟. No cowbirds were artificially produced by the experiment.
Cowbird Collection
We compared fecundity of breeding cowbirds in north-central Florida with those from east-central Texas (Figure 2). Cowbirds breeding in Florida are the subspecies M. ater ater
7 which tend to be larger (Lowther 1993) than those breeding in Texas, which are typically the dwarf cowbird subspecies M. a. obscurus (Summers et al. 2006). Texas specimens were trapped or shot at Fort Hood in Bell and Coryell Counties (ca. 31°11‟ N, 97°40‟ W) between April and
July, 2005 by the Nature Conservancy according to protocol used by Summers et al. (2006).
Cowbird control has been conducted at Fort Hood since 1988 (Eckrich et al. 1999) due to impacts of parasitism on native breeding birds. Cowbirds were trapped using a series of baited decoy traps typically placed near free-ranging cattle or shot from within 30 meters using a 20- gauge shotgun, full choke, number 9 lead shot and 2-3/4 inch shotshells. Birds were shot between 0800-1100 CST when foraging in groups on the ground. Potential local breeders were distinguished from migrant females based on morphometric, phenotypic and physiological characters (Summers et al. 2006).
Females were captured in Florida from April-July, 2006 and 2007 in baited drop-down decoy traps used to capture cowbirds in the non-breeding season. In 2006, two trap sizes were used. Large traps (2.4 x 2.4 x 1.8 m) were borrowed from the USDA Wildlife Research Center,
Florida Field Station and were constructed of wood and wire mesh. Medium-sized traps (1.7 x
1.4 x 1.4 m) were based on the “pick-up trap” design of the Texas Parks and Wildlife
Department and were primarily constructed of one-inch hardware cloth. Small traps (1.45 x 1.2 x 1.2 m) were used in 2007 and were similar to the design of medium-sized traps. Cowbird decoys placed in traps typically consisted of one female and one or more male cowbirds collected in September 2004. Traps were checked daily between 0900-1100 or 1400-1600 EST and females were euthanized humanely on site via cervical dislocation according to IACUC- approved protocol. Trapping effort ranged from one to eight open traps per day at up to three sites. In 2006, trapping was conducted at the University of Florida (UF) Dairy Research Unit
8
(DRU), Beef Research Unit (BRU), Horse Teaching Unit, and Santa Fe Beef Research Unit in
Alachua County, FL. In 2007, trapping was conducted at the same sites in addition to the UF
Large Animal Clinic and Beef Teaching Unit (BTU) in Alachua County.
With the onset of the breeding season in May, the number of female cowbirds trapped often drops significantly (Beezley and Rieger 1987, Summers et al. 2006). However, shooting of females in feeding areas allows a steady collection of specimens throughout the breeding season
(Summers et al. 2006). Trapping of cowbirds may also be biased toward individuals of poorer condition (Dufour and Weatherhead 1991). Therefore, trapping was supplemented in 2007 with collection of individual birds using pellets and a Crosman 664 Powermaster pneumatic air rifle equipped with a Tasco 4 x 32 scope. Birds were shot opportunistically between 1400 and 1800
EST at DRU, BRU, and BTU sites in or near fields where they foraged with cattle and other cowbirds. We used playback of female chatter calls in some cases to improve shooting efficiency as recommended by Rothstein et al. (1987). All collected birds were labeled and frozen until analyzed.
Reproductive Assays
All females collected were analyzed in the Ornithology Range of the Florida Museum of
Natural History in Gainesville, FL. Breeding condition of females was assessed using pre- ovulatory and post-ovulatory follicles (POFs) of the ovary (Scott and Ankney 1983). Pre- ovulatory follicles represent developing ova that are subsequently released into the abdominal cavity and taken into the oviduct for egg production. After ovulation, POFs remain attached to the ovary and are reabsorbed over time. The condition of the oviduct was examined for presence of an oviducal egg or signs that an egg has passed (straight vs. convoluted). Ovaries were examined with a dissection probe, #5 tweezers, and Bausch & Lomb Stereo Zoom 4 (7-30x
9 magnification) dissecting microscope because macroscopic counts of follicles may be unreliable
(Arnold et al. 1997). We recorded the number and size of pre-ovulatory follicles (ones that could potentially produce the yolk of an egg) in each specimen. The widths of pre-ovulatory follicles were measured to the nearest 0.01mm using a Mitutoyo Absolute Digimatic caliper (model
CD6:CS). Round, orange-yellow pre-ovulatory follicles were distinguished from degenerating or atretic follicles which are irregular in shape and cream-colored. Pre-ovulatory follicles generally grow quickly and sequences of developing follicles may represent eggs of the same clutch with distinguishable sequence breaks representing separate clutches. K-means hierarchical cluster analysis (maximum 10 iterations) using SPSS (2003) was used to group follicles into categories A, B, or C based on size (sensu Payne 1965, Scott 1978). A follicles are those that would likely have ovulated on the next day following collection, B follicles two days from collection and so on. Although smaller pre-ovulatory follicles (e.g., D follicles) were sometimes present in birds in the sample, some follicles are known to regress before ovulation
(Payne 1976). However, Scott (1978) found only 3 cases of atresia in 69 follicles >5mm indicating that larger follicles (i.e., A and B) are unlikely to regress before releasing the yolk.
Therefore, analysis was limited to three clusters to improve the probability of including only follicles that would reach ovulation.
We also recorded the number and size of all recognizable POFs which are flat and oval- shaped and have an open slit (stigma) through which ova are released into the abdominal cavity during ovulation. Only those follicles with ruptured stigma were considered POFs. POFs were distinguished from burst atretic follicles (Pearson and Rohwer 1998) by examining color and other differentiating characteristics (Scott and Ankney 1983). Burst atretic follicles are ruptured but typically contain a cream-colored residue that is lacking in POFs. The area of each flattened
10
POF was estimated as the product of width and maximum lengths measured to the nearest
0.01mm.
Ovaries, oviducts, and oviducal eggs were removed and weighed to the nearest 0.0001g using a Mettler Toledo Analytical Balance (model AB54-S). Average start and end dates of the cowbird breeding season were determined separately for Texas and Florida based on presence of enlarged oviducts, large pre-ovulatory follicles (i.e., A follicles) and oviducal eggs. Collected birds found in reproductive condition with enlarged oviducts (>0.7g), oviducal eggs, or the presence of one or more large pre-ovulatory (A follicles) or post-ovulatory follicles are
“Breeding”. All other birds are considered Non-breeding birds. Breeding birds that were collected with a developing egg anywhere in the oviduct are called “Laying” birds while all other
Breeding birds are called “Non-laying” (sensu Scott and Ankney 1980).
Shapiro-Wilk tests of normality indicated that body mass, ovary mass, and oviduct mass variables from each collection location were normally distributed (P>0.15 for all). Therefore, measures of reproductive condition of Breeding birds were compared between Florida and
Texas, and Laying and Non-laying cowbirds using two-tailed t-tests assuming unequal variances in SPSS (2003). Change in body mass over the course of the breeding season at each collection location was analyzed using simple linear regression. Finally, due to change in size of reproductive organs with the beginning and end of the breeding season, we used regression analysis to test if mass of oviducts and ovaries of Breeding birds differed between Florida and
Texas. PROC GLM (SAS Institute Inc. 2003) with Type IV sum of squares was used to fit a general linear model to the data because it allows for both continuous and categorical predictor variables. The response variables in two models were oviduct mass and ovary mass with collection week and location (Florida and Texas) as predictor variables in both models. A
11 quadratic regression function was applied to the week variable in both models due to the observed growth and regression curve of reproductive organs. Residuals plots were used to make sure there was a good fit between the model and the data.
Clutch Size Estimates
Average clutch size for brown-headed cowbirds has been estimated based on sequences of developing pre-ovulatory follicles and regressing post-ovulatory follicles (Payne 1965, Scott
1978). However, some developing follicles may regress prior to ovulation (Payne 1976), released ova may get reabsorbed before entering the oviduct (Meyer et al. 1947) and post- ovulatory follicles may be difficult to distinguish from burst atretic follicles (Payne 1965,
Lindstrom et al. 2006). As a result, clutch size can be overestimated when relying on assumed follicle sequences (Davis 1942, Payne 1965, Scott 1978, Arnold et al. 1997). Clutch size can also be underestimated using this method if post-ovulatory follicles are reabsorbed quickly
(Arnold et al. 1997). As a result, we estimated clutch size using numbers of post-ovulatory follicles observed in Breeding birds (Scott and Ankney 1983).
This method relies on two clutch size estimates based on numbers of post-ovulatory follicles in: 1) Laying birds and 2) Non-laying birds. First, the total number of eggs that Laying birds would have laid in their current clutch is estimated. This number is derived by finding the total number of clearly-identifiable POFs in the sample population of Laying birds, which should represent the average midpoint of laying sequences. That is, a sample of 10 Laying cowbirds with an average clutch size of 5 would lay 50 eggs and have 50 POFs. At the midpoint of their sequences they should have 30 POFs. Thus, this midpoint number can be used to determine the total number of eggs the sample would have laid by multiplying it by 2, and subtracting the
12 sample size (i.e., (30 x 2)–10=50). Since size breaks in regressing follicle sequences indicate distinct clutches (Payne 1976, Hannon 1981), we only included POFs that were clearly in the same laying sequence.
Second, clutch size of Non-laying birds was estimated. Non-laying birds do not have oviducal eggs by definition, so they are likely in between clutches. Therefore, recent clutch sizes were estimated based on total number of POFs divided by the sample size. In some cases, Non- laying birds did not have clearly identifiable POFs so were assumed to have laid a clutch earlier.
This assumption is supported by the observation that all Non-laying birds showed evidence of egg production (e.g., enlarged and convoluted oviducts). Because it is assumed that these birds had the same frequency distribution of clutch sizes, the contribution of these birds was estimated as the same proportion of POFs recorded in all other Non-laying birds. For example, if a sample of 50 Non-Laying birds included 40 birds with 60 POFs (proportion=1.5 POFs/bird) the remaining 10 individuals without identifiable POFs would be estimated to have 15 POFs (10 birds x 1.5). To estimate average clutch size, the sum of 1) total number of eggs that Laying birds would have produced and 2) the estimated number of POFs of Non-laying birds, were divided by the total number of birds in the sample. Estimates were performed separately for
Florida and Texas samples.
Inter-clutch interval (number of days between distinct clutches) was estimated using the proportion of Non-laying birds in a particular portion of the non-laying period (Scott and Ankney
1979). For example, some birds without an oviducal egg may have finished laying their clutch on the morning before collection. These birds would have large POFs that are less than a day old indicating ovulation occurred on the day prior to collection. This would represent the first day of the non-laying period for these females. Similarly, large pre-ovulatory follicles (A follicles) in a
13 cowbird without an egg in the oviduct would represent the end of the non-laying interval since ovulation was likely to occur on the following day. The reciprocal of the proportion of Non- laying birds in each of these stages should represent the inter-clutch interval. For example, in a sample of Non-laying birds with an inter-clutch interval of two days, roughly ½ of the birds should be in one of the two non-laying days. These two non-laying states were used to estimate the non-laying interval in Florida and Texas female cowbirds.
Fecundity Estimates
Average annual cowbird fecundity in each study area was determined as the product of daily fecundity and duration of the cowbird breeding season (Scott and Ankney 1980). Daily fecundity was calculated using the proportion of Breeding birds collected that had an egg in the oviduct (Payne 1973). Average dates of the breeding season in Texas and Florida were estimated as the first and last days when approximately half of birds had enlarged oviducts, large pre-ovulatory follicles or oviducal eggs. Only Breeding birds that were collected between respective breeding season dates in Texas and Florida were included in fecundity estimates. We assumed that all birds had an equal probability of being collected regardless of whether or not they were carrying a developing egg. Scott & Ankney (1979) found that the proportion of collected birds carrying an egg was not affected by variables such as collection location, time of day or method, among others.
RESULTS
Parasitism Frequency
From March-August, 2004-2006, we monitored the parasitism status of 35 potential host species breeding in Florida (Appendix A). Surveys included monitoring of 1120 nests of 29
14 potential cowbird host species. Data from Sieving & Contreras (unpubl. data) provided an additional 76 nest records for three species and 26 records for three additional species not monitored in this study (32 species total). Only ~1.5% (18) of nests of eight host species contained a cowbird egg or nestling. The most records of parasitism were from northern cardinal nests (6) but constituted only 2.3% of cardinal nests monitored. Other species parasitized with at least five nest records were blue-grey gnatcatcher (Polioptila caerulea; 20%), hooded warbler
(20%), blue grosbeak (Guiraca caerulea; 14.3%), white-eyed vireo (Vireo griseus; 10.5%), eastern towhee (6.1%), Bachman‟s sparrow (Aimophila aestivalis; 0.9%) and northern mockingbird (0.2%). One of three (33%) common yellowthroat nests found was parasitized.
The only yellow-throated vireo (Vireo flavifrons) nest that was found contained a lone cowbird nestling. Cowbirds fledged successfully from five nests (1 each of yellow-throated vireo, blue- grey gnatcatcher, common yellowthroat, Bachman‟s sparrow, northern cardinal). Bachman‟s sparrow nests have known to be parasitized (Ortega 1998), but this is one of only a handful of reports of this sparrow species raising a cowbird to fledging. It also represents the first case of
Bachman‟s sparrow being parasitized in Florida (Reetz et al., MS submitted).
Because of inherent difficulties in finding nests in the canopy, the sample mostly includes species nesting closer to the ground. Of the nests of mid- or canopy species able to be monitored, 1.9% (2 of 108) compared to 1.5% (16 of 1088) of lower-nesting species‟ nests were parasitized. When northern mockingbirds are not included in the sample, 2.4% (15 of 638) of lower nests were parasitized.
To better sample parasitism on the entire suite of potential host species, we monitored 272 fledge families of 20 species, including three species (Carolina chickadee Poecile carolinensis, red-eyed vireo, northern parula) for which no nest data were available. Seventeen (6.25%) of
15 these families contained a cowbird fledgling (Appendix A). Families of mid- or canopy-nesting species were more likely to be found feeding a cowbird young than lower-nesting species (Log
Odds Ratio=3.16, CI=5.25-106.15). Indeed, of the 17 parasitized families monitored, 15
(88.2%) were of mid- or canopy nesting species, and 19.74% (15 of 76) of mid- or canopy- nesting families observed were parasitized. For example, 3 of 5 (60%) summer tanager (Piranga rubra), 7 of 18 (38.9%) northern parula, 3 of 9 (33%) blue-grey gnatcatcher, 1 of 4 (25%) red- eyed vireo, and 1 of 14 (7.1%) pine warbler families monitored were observed feeding a cowbird fledgling. Northern cardinals were the only lower-nesting species that was observed feeding a fledgling cowbird (2 of 93 families or 2.2%).
Experimental Parasitism
Ten pairs of captive cowbirds produced 34 eggs in 2005 and 122 eggs in 2006 (mean mass=2.88g, range 1.16-3.54g). The most productive female produced 39 eggs in 54 days in
2005 with a run of 13 eggs in a row. Eggs obtained from the colony were used to artificially parasitize 75 nests of four common cowbird hosts and one nest each of three additional species
(Table 1). It was determined that at least some experimental eggs were viable because three eggs that were accepted into nests subsequently hatched, but 47% (27 of 58) of accepted eggs did not hatch. All cowbirds from hatched eggs were depredated in the nest. All other experimental eggs were ejected by hosts, abandoned in nests, or depredated.
Brown thrasher was the only species tested that consistently ejected cowbird eggs placed in their nests while all other species typically accepted cowbird eggs. Northern mockingbirds rejected eggs in only 15% of cases, while northern cardinals and red-winged blackbirds accepted eggs in 95% and 100% of cases, respectively. For a limited number of thrasher nests, we made observations of behaviors following artificial parasitism. Egg ejection typically occurred within
16 twenty minutes after the return of the female, and some females that eventually ejected the egg initially sat on the mixed clutch.
Cowbird Collection
A total 142 female brown-headed cowbirds were collected from Texas between 22 March and 6 July, 2005. The majority (N=94) of these specimens were collected by shotgun. A total of
71 and 63 female cowbirds were collected in Florida between 23 April and 12 July, 2006 and
2007, respectively (Figure 3). Total effort (traps open x days open) was 146 trap days in 2006 and 463 in 2007. In general, more females were caught per trap day in large traps (0.415) than in small and medium traps combined (0.061; Figure 4). All 2006 specimens were collected in traps, the majority (59%) being trapped in July when a higher proportion of individuals were no longer in breeding condition. Trapping success was low from April to May in both years but remained low in June only in 2007 (Figure 4). Trapping success increased from May to June in
2006 but the majority of these birds (74%) were collected after June 21 when some birds were in declining or non- reproductive condition (see below). Therefore, supplementing collection of females with shooting of individual birds in 2007 resulted in a more balanced sampling of females from each collection month (Figure 5).
During collection of brown-headed cowbirds, we collected four specimens of shiny cowbirds (Molothrus bonariensis), three female and one male. These contribute to only a handful of reported cases of shiny cowbirds in north-central Florida, all occurring since 2000.
One of the females collected also had a developed ovary with three large pre-ovulatory follicles and a single potential post-ovulatory follicle. The oviduct was also large (ca. 1g) and convoluted indicating that it may have passed an egg. Because there are few diagnostic features that distinguish females of shiny from brown-headed cowbirds (see Pyle 1997), tissues from the
17 specimen have been sent to the Bell Museum in Minneapolis for genetic analyses to confirm species identification. If confirmed as a shiny cowbird, it would represent only the second confirmed case of shiny cowbirds breeding in Florida.
Reproductive Assays
Reproductive assays were conducted on all female cowbird specimens collected. Of 142
Texas females collected, 79 were called Breeding birds due to various characteristics indicating mature reproductive condition. Of Breeding birds, 40 were found with an egg in the oviduct and were called Laying birds, and the remaining 39 were called Non-laying birds. 134 specimens were collected in Florida and 56 were called Breeding birds. Of Florida Breeding birds, 29 were
Laying birds and 27 were Non-laying birds.
It was not possible to test for any effect of year on reproductive measures due to small sample sizes for part of the breeding season in Florida in 2006. However, there was no difference (independent samples t-test, P>0.20 for all) in the number of pre- and post-ovulatory follicles, body mass, or mass of oviducts and ovaries of samples collected during 1 June to 21
June in 2006 (N=8) or 2007 (N=16). These variables were also not significantly different when samples from 2006 were paired with randomly-chosen samples from similar collection dates ( 2 days) in 2007 (paired samples t-test, P>0.30 for all, N=8). Therefore, all data presented for
Florida birds represents pooled data over both collection years.
Ovary measurements ( SE) of Texas cowbirds collected between 22 March and 6 April showed no signs of breeding season development (26.89 1.21mm2, 0.027 0.002g, N=40).
Oviducts were similarly small and undeveloped during this time (0.043 0.009g, N=7). For both Florida and Texas females, growth and maturation of ovarian follicles led to a dramatic (ca.
18
14-fold in Texas) increase in mean ovary mass in mid- to late-April (Figure 6). Similarly, oviducts of collected specimens increased significantly (ca. 28-fold in Texas) in mid- to late
April (Figure 7). Mean ovary and oviduct mass remained fairly consistent from May to early
June, although there was much individual variation due to the dynamic nature of ovarian follicle growth and yolk release during ovulation (Figures 8 to 11). On average, laying birds (with an oviducal egg) in Florida had significantly heavier ovaries (t=2.480, P=0.017) and oviducts
(t=3.429, P=0.001) than Non-laying birds (Figures 8, 9). However, Breeding birds without oviducal eggs still had large oviducts (i.e., mean mass >1.0g), probably because they recently completed clutches or were preparing to ovulate. Laying birds in Texas had significantly larger ovaries (t=2.091, P=0.041), but not oviducts (t=1.263, P=0.212) than Non-laying birds (Figures
10, 11). Ovaries and oviducts began to regress in the second week of June as resource allocation to reproduction was likely waning with the end of the cowbird breeding season. There was no significant difference between Breeding females from Texas and Florida in ovary mass (t=0.805,
P=0.422) over the course of the breeding season (Table 2) but oviducts were significantly heavier in Texas than Florida females (t=-2.304, P=0.023). Because growth and regression of reproductive organs roughly followed a bell-shaped curve, we applied a general linear model to the data to determine if oviduct and ovary mass differed between Florida and Texas cowbirds as a function of week of the breeding season. The regression model indicated that collection location was not a significant predictor of ovary mass (parameter estimate=0.037, F=0.88,
P=0.350; Figure 8). However, the model indicated that collection site was a significant predictor of oviduct mass (parameter estimate=-0.129, F=5.17, P=0.025) with Texas females having heavier oviducts than Florida females.
19
Body mass of Florida females declined significantly from 23 April until 12 July (R2=0.161,
P<0.001, N=110). Mass of Breeding birds in Florida tended to decrease over time (Figure 12) but not significantly (R2=0.024, P=0.263, N=53) especially when the mass of the oviducal egg was removed from body mass (R2=0.006, P=0.585). Overall body mass decline from April to
July was likely due to mass differences between pre-breeding season birds prepared for reproduction and post-breeding birds with regressing reproductive organs. Mean mass of
Breeding females (R2<0.001, P=0.990, N=58) and all cowbirds collected in Texas between 19
April and 6 July (R2=0.001, P=0.826, N=79) did not change significantly with time (Figure 13).
Similarly, no body mass change was observed over time when oviducal egg mass was removed from body mass of Laying birds (R2=0.012, P=0.429). Although the eastern subspecies of cowbird is generally larger than the dwarf cowbird of Texas (Lowther 1993, Pyle 1997), mass of females did not significantly differ between Breeding Florida and Texas cowbirds (t=-1.113,
P=0.268). Similarly, wing chords of Florida breeders also were not significantly different from those of Texas cowbirds (Table 2) collected at Fort Hood during previous breeding seasons
(Summers et al. 2006).
Increase in ovarian mass was associated with the development of pre-ovulatory follicles in late April. Oocytes (immature ova) were generally ca. 1.0mm in diameter in March and early
April whereas Breeding birds had one or more large orange-yellow yolky follicles representing potential eggs. A follicles (those likely to ovulate the next day) were as large as 10.50mm in diameter in Texas and 9.89mm in Florida. Sequences of developing follicles were determined using cluster analysis. For Florida birds, A, B, and C follicles were those that were larger than
8.57, 5.85, and 3.39mm diameter, respectively. Similarly, Texas birds cluster cutoff values for
A-C follicles were 8.72, 6.38 and 3.37mm diameter, respectively. These values are similar to
20 those reported by Scott (1978) who found that the three largest pre-ovulatory follicles clustered at >8mm, and ca. 6 and 4mm. Earliest and latest A follicles were found in cowbirds collected on
26 April and 28 June in Texas and 26 April and 27 June in Florida.
All but four individuals had at least one post-ovulatory follicle (POFs), although birds without POFs were still clearly in reproductive condition. All Laying birds had one large POF that was associated with the egg in their oviduct and which released the yolk one day prior to collection (Day -1). Laying birds had as many as four POFs that showed a graded series in size.
Mean areas ( SE) of the sequence of POFs from largest (Day -1) to smallest (Day -4) in Laying birds in Florida were 22.89 0.72, 13.01 0.47, 7.80 0.44, and 6.57 1.50mm2, respectively.
In Texas birds mean areas of POFs in this sequence were 25.89 0.85, 25.42 0.73, 12.01
0.75, and 10.43 1.38mm2. Day -2 POFs that were identified as likely being part of the same sequence showed ca. 40% decrease in size from the most recent follicle. This indicated that
POFs regressed rapidly and that follicles older than four days would likely be difficult to identify. Furthermore, the three most recent POFs (Days -1, -2, and -3) of Laying birds were significantly larger in Texas than in Florida (t-tests, P<0.01 for all; Figure 14). POFs from Days
-1 and -2 were ca. 13% and 18.5% smaller in Laying Florida specimens. This is likely due to differences in collection protocol since all Texas specimens were collected before 1100 hours and Florida specimens were collected as late as 1800 hours, allowing more time for regression of follicles during the day.
Clutch Size Estimates
Clutch size and inter-clutch interval estimates were determined using the sample of
Breeding birds from each collection location. In Texas, 40 Laying females had between one and
21 four readily identifiable POFs and most commonly three (Table 2). The mid-point of the laying sequence for these birds was estimated as 108 total POFs. Two Non-laying birds were excluded from counts due to the poor condition of their ovaries. A total of 29 Non-laying females had 66 recognizable POFs (2.3 POFs/bird). The remaining 8 Non-laying birds without POFs were therefore estimated to contribute 18 POFs (8 birds x 2.3). Based on these figures, average clutch size of Breeding birds in Texas is estimated as 3.38 eggs per female (Table 2).
In Florida, one Laying bird was excluded from clutch size estimates because the ovary was damaged when collected. A total of 28 Laying females also had between one and four POFs, with a mode of three (Table 3). These females produced 83 POFs. Twenty-three Non-laying birds had 46 distinct POFs (2.0 POFs/bird) so that the remaining four birds without POFs were estimated to have 8 POFs. Average clutch size of Breeding birds in Florida is estimated as 3.49 eggs per female (Table 4).
Inter-clutch intervals were estimated using two methods that use the proportion of Non- laying birds with: 1) size A pre-ovulatory follicles and 2) POFs large enough to have ovulated an egg on the day before collection. In Texas, 13 of 37 Non-laying birds had size A pre-ovulatory follicles that would likely have ovulated on the next day. Of 27 Non-laying birds in Florida, 12 had size A ovulatory follicles. The reciprocals of these proportions results in estimates of inter- clutch interval of 2.85 and 2.25 for Texas and Florida, respectively.
For method 2, we assumed that any Non-laying bird with a POF larger than the smallest
POF in a Non-laying bird ovulated on the previous day. Since POFs regress approximately 50% in 12 hours following ovulation (Scott and Ankney 1983), we used the smallest POF of a Laying bird collected in the afternoon to establish a cutoff size for POFs in Non-laying birds. These values are 18.82mm2 and 15.01mm2 for Texas and Florida. Twenty of the 37 Non-laying birds in
22
Texas had POFs that indicated that the last day of ovulation was one day prior to collection. Of
27 Non-laying Florida birds, 14 had large POFs. The reciprocals of these two proportions results in inter-clutch interval estimates of 1.85 and 1.92 days for Texas and Florida, respectively.
Therefore, the average non-laying period was approximately 2-3 days in Texas and 2 days in
Florida.
Sequences of egg-laying in five captive Florida cowbird females used in the experimental parasitism portion of this research resulted in a mean estimate ( SE) of clutch size (3.68 0.57 eggs) and inter-clutch interval (1.80 0.23 days) similar to those estimated above for wild- breeding populations.
Fecundity Estimates
Average fecundity of cowbirds was calculated based on the method of Scott & Ankney
(1980) using length of the breeding season and daily laying rate An average start and end date to the laying season was estimated using the presence of enlarged oviducts, large pre-ovulatory follicles and oviducal eggs. Of the eight Florida birds collected from April 23 to 29, three had oviducal eggs and three of the five non-laying birds had large POFs and mid-sized (B) pre- ovulatory follicles and were therefore in breeding condition. The earliest bird that had an oviducal egg was collected on 26 April but had three distinct POFs indicating the ovulation probably began as early as 23 April. We were not able to collect an adequate sample of birds prior to this date. The earliest specimen was collected 23 April and had no POFs but some developing follicles. Therefore, we assume that the average female begins breeding during the final week of April. Therefore we estimate an average start date to the cowbird breeding season in Florida to be 26 April.
23
Females were collected in Florida as late as 17 July but the last female with an oviducal egg was collected on 6 July. This female is likely an outlier because only one of the remaining
17 females collected in the first week of July had post-ovulatory follicles that indicated laying of an egg could have occurred in early July. One of 46 birds collected after this date had a single small post-ovulatory follicle, indicating that laying had likely ceased after the first week of July.
To assign an average end date to the cowbird breeding season in Florida, we estimated the last date when ca. 50% of the birds were in breeding condition. From 25 June to 1 July, only one of 18 birds collected had an oviducal egg. Of the remaining 17, six were likely still in breeding condition due to the presence of pre-ovulatory follicles larger than 5mm in ovaries with little or no atresia. Therefore 38% of birds were still in laying condition. From 18-24 June, four of seven birds (57%) were in breeding condition (one with an oviducal egg). Central dates for each of these weeks are 21 June and 28 June respectively. If it is assumed that the percentage of birds in breeding condition decreases steadily between these dates, the first date when less than 50% of birds were in laying condition is estimated to be 24 June (48.9% of birds).
In Texas, birds were collected beginning in March and the first bird with an oviducal egg was collected on 26 April. The presence of two large POFs indicated the ovulation could have begun as early as 24 April. However, of the other six birds collected during 23-29 April none had ovaries with developing follicles so were not in breeding condition. Indeed some of these birds may have been migrating north to other breeding areas so it is not clear whether or not they were local breeders. Because the sample from Texas does not include sufficient sample sizes of identified breeders from mid-late April to early May to deduce an average start date, we estimate a similar start date for Texas birds as those from Florida. Reproductive behaviors of brown- headed cowbirds begin as early as late March but few cowbirds are collected with an oviducal
24 egg until mid-April (S. Summers, pers. comm.). Since the start date estimates when an average bird is laying (i.e., 50% of the population), 26 April represents at least a conservative estimate of the beginning of the breeding season in Texas.
The latest oviducal egg-bearing female captured at Fort Hood was collected 12 July, but for most years the latest is the second week of July, typically around 9 July (S. Summers pers. comm.). Consistent with these observations, the last Texas specimen analyzed that had an oviducal egg was collected on 6 July. From 25 June to 1 July, four of nine birds collected had an oviducal egg. Of the remaining five, none were still in breeding condition because they lacked developing follicles. Therefore 44% of birds were still in laying condition. From 18-24 June, 10 of 16 birds were in breeding condition, seven of which had an oviducal egg. Using central dates and the assumption of a steady decrease in breeding condition, the first date when less than 50% of birds were in laying condition is estimated to be 26 June (49.6% of birds).
In sum, we estimate the average breeding season length of 60 days from 26 April to 24
June in Florida and 62 days from 26 April to 26 June in Texas. Both starting and ending dates are earlier than those reported for brown-headed cowbirds breeding in southern Ontario (Scott and Ankney 1980) but the estimated breeding seasons for Florida and Texas are 4-6 days longer.
This is consistent with the phenomenon of decreasing avian breeding season length with increasing latitude (Baker 1939, Johnston 1954).
Nine birds from Florida and seven birds from Texas that were in reproductive condition were excluded from estimates of daily laying rate because they were collected after the average breeding season end dates. Of excluded birds, two Florida and five Texas females had developing eggs in their oviducts. Daily laying rate was estimated as the proportion of Breeding birds with eggs in the oviduct (Payne 1973). In Texas, 40 of 71 Breeding birds collected
25 between 26 April and 26 June had oviducal eggs resulting in an estimated daily laying rate of
0.56 eggs per female per day. In Florida, 27 of 47 Breeding birds collected between 26 April and 24 June resulting in a daily laying rate of 0.57 eggs. Using these rates and the respective lengths of the breeding season, it is estimated that each female brown-headed cowbird in both
Texas and Florida lays an average of 35 eggs in a breeding season (TX: 62 days x 0.56 eggs/day
= 34.93, FL: 60 days x 0.57 eggs/day = 34.47). Because these estimates only include Breeding individuals, they only represent average fecundity for birds in reproductive condition and not fecundity of the entire population of birds present in the breeding area.
We assumed that the average daily laying rate was consistent over the course of the breeding season because small sample sizes for many weeks precluded analysis. However, daily laying rate of brown-headed cowbirds has been found to be consistent during each week of an average cowbird breeding season (Scott and Ankney 1980: Table 3).
Potential Biases
Mass of Breeding cowbirds trapped in Florida was significantly lower than shot birds
(34.70 vs. 38.50g, t=-5.498, P<0.001). Ovaries (0.3250 vs. 0.5217; t=-0.3838, P<0.001) and oviducts (1.0253 vs. 1.4126, t=-4.495, P<0.001) were also significantly lighter in trapped birds.
However, wing chord (t=0.013, P0.990) and tail length (t=-0.767, P=0.448) was similar in trapped and shot birds. Therefore, differences between trapped and shot birds may not reflect a condition bias (e.g., Dufour and Weatherhead 1991), but a sampling bias. Shot birds were more likely to have an oviducal egg than trapped birds (Log Odds Ratio=2.76, CI=3.63-68.41) so that trapping may bias toward individuals in between clutches rather than birds in poorer condition.
It is also possible that the protocol for shooting birds was biased toward individuals in laying condition. 17 of 20 Breeding birds shot during the Florida breeding season had oviducal eggs. If
26 birds were only collected using this method, the fecundity estimate would have been 51 eggs/female/season. With trapping alone, the estimate would be 22 eggs. Trapping or shooting methodology, when used alone, may bias fecundity estimates that rely on the presence of oviducal eggs.
27
Tables and Figures
Table 1. Results of experimental parasitism from 2005-2006 for four common species that are known to eject cowbird eggs or abandon parasitized nests. Brown thrashers were the only species to consistently eject cowbird eggs while the other species generally accepted brown-headed cowbird eggs experimentally placed in their nests.
Nests Species Parasitized Accepted Ejected Abandoned % Accepted Brown Thrasher 17 2 14 1 11.8 Northern Cardinal 19 18 1 0 94.7 Northern Mockingbird 20 17 3 0 85.0 Red-winged Blackbird 19 19 0 0 100 Other species* 3 2 0 1 66.7 *Eastern Towhee (Abandon), Loggerhead Shrike, White-eyed Vireo
Table 2. Summary of means (N, 95% confidence interval) of morphometric data taken from Breeding female Brown-headed cowbirds collected in Florida and Texas, 2005-2007 and results of statistical comparisons (independent samples t-tests). Data for wing chord length of Texas birds are taken from Summers et al. (2006).
Ovary Mass Oviduct Mass Body Mass (g) Wing Chord (mm) Florida 0.41 1.20 36.0 94.98 (54, 0.35-0.46) (55, 1.11-1.29) (54, 35.14-36.86) (62, 94.82-95.14) Texas 0.39 1.38 36.61 94.95 (74, 0.33-0.45) (73, 1.29-1.46) (59, 35.95-37.26) (959, 94.34-95.56) P-value 0.422 0.023 0.268
28
Table 3. Estimate of clutch size of 2005 Texas cowbirds using number of post-ovulatory follicles (POFs) clearly identified to have a ruptured stigma. Laying birds are those with oviducts containing a developing egg while Non-laying birds are individuals in breeding condition (presence of enlarged oviducts, pre-ovulatory follicles, or post- ovulatory follicles) but without an oviducal egg. Method derived from Scott & Ankney (1983). Number of POFs for 10 Non-laying birds without clearly identifiable POFs (?) is estimated using the proportion of POFs in all other Non-Laying birds (66/29=2.3 POFs/bird).
Laying Birds Non-laying Birds Number of Est. Clutch Number of No. POFs Birds Total POFs Size Birds Est. POFs 1 2 2 ? 8 18 2 13 26 1 8 8 3 20 60 2 7 14 4 5 20 3 12 36 4 2 8 Total 40 108 37 84
Clutch size = [(108 x 2) – 40] + 84 = 3.38 eggs 40 + 37
Table 4. Estimate of clutch size of 2006-2007 Florida cowbirds using number of clearly identifiable post-ovulatory follicles (POFs). Number of POFs for Non-laying birds without clearly identifiable POFs (?) is based on the proportion of 2.0 (26/23) POFs/bird in all other Non-laying birds.
Laying Birds Non-laying Birds Certain Number of Est. Clutch Number of Total POFs Est. POFs POFs Birds Size Birds 1 1 1 ? 4 8 2 4 8 1 9 9 3 18 54 2 6 12 4 5 20 3 7 21 4 1 4 Total 28 83 27 54
Clutch size = [(83 x 2) – 28] + 54 = 3.49 eggs 28 + 27
29
Figure 2. Location of brown-headed cowbird research sites from April-August, 2004-2007. Site 1 = Fort Hood, Texas, Site 2 = north-central Florida. Site 1 included collection of female cowbirds in 2005 and Site 2 included cowbird collection, parasitism surveys and experimental parasitism experiments from 2004-2007.
45 42 40 35 2006 30 2007 27 25 21 19 20 17 15 10 6
5 1 1 Number of Females Collected . Collected Females of Number 0 April May June July
Figure 3. Number of female cowbirds collected per month at sites in north-central Florida in 2006 and 2007. Sample sizes are given above each bar.
30
Figure 4. Mean number of female cowbirds collected per trap day at sites in north-central Florida from April to July, 2006-2007. Total trap days were 146 and 463 in 2006 and 2007, respectively. Total trap days were 392 and 217 for small/medium and large traps, respectively. Trap efficiency data presented by trap size includes both 2006 and 2007 samples.
25 Shot 1 20 10 Trapped 10 15
10 20
5 9
6 7 Number of Females Collected . Collected Females of Number 0 April May June July
Figure 5. Number of female cowbirds collected per month in Florida by trap and by pellet gun from April-July 2007. Sample sizes are given for each collection method.
31
0.70 Florida 5 0.60 2 18 Texas 6 7 3 0.50 1 0.40 3 11 8 8 19 0.30 8 Mass (g) Mass 7 2 7 0.20
0.10 2 3 18 21 12 5 34 0.00 4/15 4/29 5/13 5/27 6/10 6/24 7/8 Week Beginning
Figure 6. Mean weekly mass ( SE) of ovaries of all Florida and Texas female cowbirds collected between 15 April and 15 July, 2005-2007. Samples sizes for each week are given next to each point.
2.00 1.80 5 (FL) 9 19 1.60 3 6 1 3 1.40 1 18 11 1.20 6 4 7 8
1.00 2 6 Mass (g) Mass 0.80 8 Florida 0.60 5 2 Texas 0.40 18 21 34 0.20 2 0.00 4/15 4/29 5/13 5/27 6/10 6/24 7/8 Week Beginning
Figure 7. Mean weekly mass ( SE) of oviducts of all Florida and Texas female cowbirds collected between 15 April and 15 July, 2005-2007. Sample sizes for each week are given next to each point.
32
0.80 0.70 3 4 0.60 2 6 5 3 0.50 3 2 0.40 1
Mass (g) Mass 0.30 2 4 1 0.20 Laying 5 5 1 0.10 Non-laying 3 0.00 4/15 4/29 5/13 5/27 6/10 6/24 7/8 Week Beginning
Figure 8. Mean weekly mass ( SE) of ovaries of Breeding brown-headed cowbirds collected in Texas, 2005. Laying birds are those with an oviducal egg while Non-laying birds are in reproductive condition but lack an oviducal egg. Sample sizes for each collection week are given next to each point.
2.00 1.80 3 4 1.60 3 6 5 1 1.40 3 2 1.20 1 2 1.00 2
Mass (g) Mass 0.80 4 0.60 5 3 Laying 5 0.40 1 0.20 Non-laying 0.00 4/15 4/29 5/13 5/27 6/10 6/24 7/8 Week Beginning
Figure 9. Mean weekly mass ( SE) of oviducts of Breeding brown-headed cowbirds collected in Florida, 2006-2007. Laying birds are those with an oviducal egg while Non-laying birds are in reproductive condition but lack an oviducal egg. Sample sizes for each collection week are given next to each point.
33
0.70 Laying 8 0.60 6 Non-laying 1 3 8 3 0.50 1 3 2 0.40 8
0.30 2 Mass (g) Mass 0.20 4 6 4 0.10 2 2 11 11 0.00 4/15 4/29 5/13 5/27 6/10 6/24 7/8 Week Beginning
Figure 10. Mean weekly mass ( SE) of ovaries of Breeding brown-headed cowbirds collected in Texas, 2005. Laying birds are those with an oviducal egg while Non-laying birds are in reproductive condition but lack an oviducal egg. Sample sizes for each collection week are given next to each point.
2.00 3 1.80 6 10 1 1.60 3 1 4 8 1.40 3 1.20 2 3 8 2 3 1.00 1 6 1
Mass (g) Mass 0.80 10 0.60 Laying 0.40 Non-laying 0.20 0.00 4/15 4/29 5/13 5/27 6/10 6/24 7/8 Week Beginning
Figure 11. Mean weekly mass ( SE) of oviducts of Breeding brown-headed cowbirds collected in Texas, 2005. Laying birds are those with an oviducal egg while Non-laying birds are in reproductive condition but lack an oviducal egg. Sample sizes for each week are given next to each point.
34
Figure 12. Mass change of female brown-headed cowbirds collected in Florida between 23 April and 12 July, 2006-2007. Mass of all birds decreased significantly over time (hashed line, R2=0.16, P<0.001, N=110). Mass of Breeding birds did not significantly change over the course of the breeding season (solid line, R2=0.024, P=0.263, N=53).
50.0 All Birds 45.0 Breeding Birds
40.0
35.0 Female Mass (g) Mass Female
30.0
25.0 4/15 4/29 5/13 5/27 6/10 6/24 7/8
Figure 13. Mass change of female brown-headed cowbirds collected in Texas between 19 April and 6 July, 2005. Mass of all birds (hashed line, R2=0.001, P=0.826, N=79) and of Breeding birds (solid line, R2<0.001, P=0.990, N=58) did not change significantly over time.
35
30.0 * 25.0 Texas 40 Florida 20.0 26 * 15.0 34 * 10.0 18
POF Area (mm2) 25 3 5.0 22 4
0.0 -1 -2 -3 -4 Collection Day
Figure 14. Mean area ( SE) of post-ovulatory follicles (POFs) of brown-headed cowbirds collected in Texas and Florida from April to June, 2005-2007. Areas presented include only birds with eggs in their oviducts. Day -1 represents POFs associated with oviducal eggs and which released yolk one day prior to being collected. Asterisks indicate comparisons between Florida and Texas that are statistically significant (t-tests, P<0.01). Sample sizes are given with each bar.
DISCUSSION
Patterns of Parasitism
Extremely low parasitism rates of nests on potential host species breeding in Florida were observed. Brown-headed cowbirds have been identified as “generalist” hosts because they are known to parasitize over 220 species (Friedmann and Kiff 1985). Because parasitism often affects many or sometimes most potential host species in a community, parasitism may not be host-specific because eggs are spread among hosts according to their abundance ('shotgun' hypothesis; Rothstein 1975). In this case, „good‟ hosts (those able to successfully fledge cowbird young) are parasitized as frequently as „poor‟ hosts. However, host-specific characteristics can affect the survival probability of cowbird young so that selection may favor parasitism of hosts successful at raising cowbirds. As a result, parasitism of nests may be biased toward a few select
36 species ('host selection' hypothesis; Rothstein and Robinson 2002). Host selection is supported by data showing nonrandom parasitism frequencies of a suite of potential hosts (Woolfenden et al. 2003). In areas where cowbirds are abundant, individuals are likely to employ both host selection and shotgun tactics because competition for resources (nests of „good‟ hosts) is more intense.
For successful expansion of cowbirds into new regions such as Florida, the availability of at least some suitable hosts that accept cowbird eggs and appropriately nourish juveniles is critical. Suitable hosts are often parasitized more frequently than those that reject eggs (Sealy and Bazin 1995) or feed young a diet inappropriate to cowbirds (Middleton 1991), but patterns of parasitism across the host community have not been fully addressed in areas where cowbirds recently expanded. In a newly colonized area such as Florida, low initial cowbird densities should favor host selection because of low competition for nests. In this case, the availability
(abundance/diversity) of quality hosts may represent a limiting factor to reproduction.
Overall, there were very few nests that were parasitized (~1.5%) despite cowbirds being common at all study sites. Approximately 2% of northern cardinal nests that were monitored were parasitized. This may be a conservative estimate of parasitism because cowbird and cardinal eggs are sometimes difficult to distinguish. Furthermore, cardinal eggs can also show intra-clutch variation in color (Kinser 1973), making identification of a cowbird egg problematic.
However, approximately 2% of cardinal fledge families contained a cowbird fledgling which was consistent with nest parasitism data. Our data support host selection because certain species, particularly those that nest in the canopy, appear to have been preferentially parasitized. Canopy species tended to be parasitized at a higher frequency than nests of lower-nesting species.
However, specific nest preference is difficult to determine without a large dataset of canopy
37 nests or experimental tests of cowbird preference, both of which are not logistically feasible.
Observations of fledge families support this pattern since the majority of parasitized families were of canopy-nesting species such as northern parula and blue-grey gnatcatcher. Therefore, these and other canopy-nesting species probably account for a disproportionate amount of recruitment of juvenile cowbirds in north-central Florida.
Maintenance of Defense Behaviors
It is possible that the observed low frequency of parasitism was due to cowbird eggs being removed by adults or nests being abandoned before the active nest could be monitored. Indeed, one inactive white-eyed vireo and one northern cardinal nest were found with a single cowbird egg and no host eggs. It is not known if these were abandoned nests, or if the cowbird egg was laid in a depredated nest but it is not possible to fully account for this effect on parasitism frequencies. However, of the hundreds of active nests monitored during the nest-building and egg-laying stage, no incidences of natural egg rejection or abandonment were recorded.
Furthermore, results from this project show a relatively low prevalence of egg ejection behavior of three of four common potential hosts experimentally parasitized.
Rejection and abandonment behavior does not appear to be retained in some species in
Florida as it is in other regions with longer exposure to cowbirds. Despite only recent expansion of cowbirds to the southeastern coastal plain, many species display anti-parasitism behaviors
(Whitehead et al. 2002), perhaps due to inflow of genes from regions where cowbirds have historically bred (Rothstein 1990). For example, northern cardinals breeding in South Carolina abandon 50% of parasitized nests (Whitehead et al. 2002). This pattern was not observed in this experiment where northern cardinals accepted eggs in 95% of nests parasitized. Many of these nests were parasitized before completion of the clutch and one nest was parasitized upon nest
38 completion and before a host egg was laid, yet the female cardinal accepted the cowbird egg and subsequently laid a full clutch of her own. While fossil evidence suggests that brown-headed cowbirds may have existed in the Southeast U.S. in the late Pleistocene (Hamon 1964), birds presently breeding in Florida probably have limited evolutionary „knowledge‟ of cowbirds.
„Ejector‟ (e.g., mockingbirds, thrashers) and „deserter‟ (cardinals) species are among the most abundant songbird species breeding in north-central Florida but do not appear to limit overall cowbird reproduction through common anti-parasitism behaviors.
Cowbird Fecundity
Brown-headed cowbirds have the highest fecundity of any passerine in North America.
Females produce an average of 40 eggs (Scott and Ankney 1980) and as many as 77 eggs per nesting season (Holford and Roby 1993). In fact, cowbirds are the only known songbird species not to show regression of ovaries following the completion of a clutch (Lewis 1975).
Furthermore, cowbirds start new clutches in as little as 1 day (Scott and Ankney 1980) and lay eggs in about 40 seconds, an order of magnitude faster than other icterids (Sealy et al. 1995).
These adaptations contribute to the brown-headed cowbird‟s colonization capabilities.
Although cowbird fecundity is high in many parts of their range, factors such as host, food, or calcium availability can physiologically limit reproduction. For example, brown-headed cowbirds have shown high fecundity in areas where grain-based agriculture is widespread (e.g.,
Illinois) probably because food sources are accessible and abundant. Where agricultural practices are not grain-based (e.g., Florida), cowbird food resources may be relatively lower. In the West Indies, lower abundance of shiny cowbirds on Martinique than St. Lucia may be due to differences in food rather than nest availability (Post et al. 1990). Breeding cowbirds also must obtain necessary calcium for eggshell formation from ingestion of mollusc shells (Ankney and
39
Scott 1980) or secondarily through insect and seed consumption, although the latter two are poor sources. In areas with calcium-poor soils, songbirds may produce inferior eggs or desert nests, leading to fewer nestlings per nesting attempt (Graveland and Drent 1997). Any variation in supplemental calcium availability may help explain regional differences in observed parasitism rates (Holford and Roby 1993). However, calcium is probably readily available to birds in
Florida due to the prevalence of exposed limestone containing calcium carbonate.
Our results indicate that brown-headed cowbird reproduction in north-central Florida is not physiologically limited by food or nutrient resource availability. Our estimate of fecundity of cowbirds breeding in north Florida is as high as that for cowbirds breeding in Texas, where nest parasitism has led to critical declines of some species of native songbirds (Eckrich et al.
1999). It is also similar to estimates in other parts of brown-headed cowbird distribution (Payne
1973, Scott and Ankney 1980). Therefore, out data indicate that individual female brown- headed cowbirds breeding in Florida lay as many eggs as female cowbirds in regions where they have caused serious population declines and have threatened extirpation of some species.
Management Implications
This project is relevant to conservation and wildlife management in Florida for a number of reasons. First, we studied a species that has represented a serious conservation concern in many states, but which has been little studied in Florida. Second, this project contributes extensively to a limited reproductive dataset for many Florida songbird species. Third, the results generated by this project identify potential influences on brown-headed cowbird reproduction and distribution, and host species or communities at particular risk to parasitism.
Lower-nesting species such as northern cardinal, northern mockingbird, brown thrasher, and Carolina wren are probably poor host species to cowbirds due to anti-parasitism behaviors,
40 nest site choice, provisioning of young, or high predation rates. In north-Florida, these species constitute a relatively higher proportion of the overall bird abundance than in other regions where there is a high diversity and abundance of songbird hosts (e.g., emberizids, parulids, turdids, etc.). Other north-Florida species such as indigo bunting (Passerina cyanea), yellow- breasted chat (Ictera virens) and orchard oriole are quality hosts and are moderately parasitized in other southeastern states where they are common (Whitehead et al. 2002, Kilgo and Moorman
2003). However, north Florida represents the southern extent of their breeding range and they are found in very low densities. As a result, the data suggest that canopy species are likely to experience higher parasitism rates because many of these species are abundant (e.g., pine warbler, northern parula) and are thought to be quality hosts. Despite the logistical difficulties in locating and monitoring nests of canopy species, particularly in Florida where Spanish moss
(Tillandsia useneoides) and other epiphytes provide significant nesting cover, we stress the need for nest data of canopy species to determine parasitism rates and to identify any specific species at risk.
Data from this project indicate that parasitism by brown-headed cowbirds is not limited by behavioral responses of hosts given that the species that commonly exhibit them did not do so in this experiment. Canopy species such as those observed feeding cowbird fledges typically accept cowbird eggs laid in their nests in other parts of their range. This suggests that brown- headed cowbirds choosing to lay eggs in nests of canopy species will not be limited by adaptive responses by these „preferred‟ hosts. We also observed that cowbird fecundity in north-Florida may be quite high. If an average female is indeed laying 35 eggs per season as data suggest, and the abundant lower-nesting species are not being parasitized, the few canopy species breeding in
Florida may bear the brunt of cowbird parasitism. Populations of quality canopy hosts should be
41 closely monitored in the next decade to determine if management of cowbirds or host species is necessary.
Finally, we recommend that trapping of cowbirds be supplemented by shooting to avoid any biases misrepresent the reproductive status of the local cowbird population. In addition, fecundity estimates based on oviducal analyses should be performed only using a sample of birds collected via a variety of methods.
42
LITERATURE CITED
Ankney, C. D., and D. M. Scott. 1980. Changes in nutrient reserves and diet of breeding brown- headed cowbirds. Auk 97:684-696. Arnold, T. W., J. E. Thompson, and C. D. Ankney. 1997. Using post-ovulatory follicles to determine laying histories of American coots: implications for nutrient-reserve studies. Journal of Field Ornithology 68:19-25. Baker, J. R. 1939. The relation between latitude and breeding season in birds. Proceedings of the Zoological Society of London 109A:557-582. Beezley, J. A., and J. P. Rieger. 1987. Least Bell's vireo management by cowbird trapping. Western Birds 18:55-61. Bonifait, S., M.-A. Villard, and D. Paulin. 2006. An index of reproductive activity provides an accurate estimate of the reproductive success of palm warblers. Journal of Field Ornithology 77:302-309. Brittingham, M. C., and S. A. Temple. 1983. Have cowbirds caused forest birds to decline? Bioscience 33:31-35. Cruz, A., W. Post, J. W. Wiley, C. P. Ortega, T. K. Nakamura, and J. W. Prather. 1998. Potential impacts of cowbird range expansion in Florida. Pages 313-336 in S. I. Rothstein and S. K. Robinson, editors. Parasitic Birds and their Hosts: Studies in Coevolution. Oxford University Press, New York. Cutts, E. A. 1964. Brown-headed cowbird egg collected on the South Carolina coast. Chat 29:115. Davis, D. E. 1942. The number of eggs laid by cowbirds. Condor 44:10-12. De Capita, M. E. 1993. Brown-headed cowbird control on Kirtland's warbler nesting areas in Michigan, 1972-1993. North American Workshop on the Ecology and Management of Cowbirds, Austin, TX. Denton, J. F. 1946. The breeding status of the cowbird in Georgia. Oriole 11:24-27. Dufour, K. W., and P. J. Weatherhead. 1991. A test of the condition-bias hypothesis using brown-headed cowbirds trapped during the breeding season. Canadian Journal of Zoology 69:2686-2692. Eckrich, G. H., T. E. Koloszar, and M. D. Goering. 1999. Effective landscape management of brown-headed cowbirds at Fort Hood, Texas. Studies in Avian Biology 18:267-274. Ehrlich, P. R., D. S. Dobkin, and D. Wheye. 1988. The Birder's Handbook. Simon & Schuster Inc., New York. Florida Fish and Wildlife Conservation Commission. 2007. Florida's breeding bird atlas: a collaborative study of Florida's birdlife. September 1. http://www.wildflorida.org/bba. Friedmann, H., L. F. Kiff, and S. I. Rothstein. 1977. A further contribution to knowledge of the host relations of the parasitic cowbirds. Smithsonian Contributions to Zoology 235:1-75. Friedmann, H., and L. F. Kiff. 1985. The parasitic cowbirds and their hosts. Proceedings of the Western Foundation of Vertebrate Zoology 2:225-304. Graveland, J., and R. H. Drent. 1997. Calcium availability limits breeding success of passerines on poor soils. Journal of Animal Ecology 66:279-288. Hamon, J. H. 1964. Osteology and paleontology of the passerine birds of the Reddick, Florida, Pleistocene. Florida Geological Survey Geology Bulletin 44. Hannon, S. J. 1981. Postovulatory follicles as indicators of egg production in blue grouse. Journal of Wildlife Management 45:1045-1047.
43
Holford, K. C., and D. D. Roby. 1993. Factors limiting fecundity of captive brown-headed cowbirds. Condor 95:536-545. Johnston, R. F. 1954. Variation in breeding season and clutch size in song sparrows of the Pacific coast. Condor 56:268-273. Kilgo, J. C., and C. E. Moorman. 2003. Patterns of cowbird parasitism in the Southern Atlantic Coastal Plain and Piedmont. Wilson Bulletin 115:277-284. Kinser, G. W. 1973. Ecology and behavior of the cardinal, Richmondena cardinalis (L), in southern Indiana. Dissertation. Indiana University, Bloomington. Laymon, S. A. 1987. Brown-headed cowbirds in California: historical perspectives and management opportunities in riparian habitats. Western Birds 18:63-70. Lewis, R. A. 1975. Reproductive biology of the white-crowned sparrow (Zonotrichia leucophrys pugetensis Grinnell). Temporal organization of reproductive and associated cycles. Condor 77:46-59. Lindstrom, E. B., M. W. Eichholz, and J. M. Eadie. 2006. Postovulatory follicles in mallards: implications for estimates of breeding propensity. Condor 108:925-935. Lokemoen, J. T., and R. R. Koford. 1996. Using candlers to determine the incubation stage of passerine eggs. Journal of Field Ornithology 67:660-668. Lowther, P. E. 1993. Brown-headed cowbird (Molothrus ater). Account 47 in A. Poole and F. Gill, editors. The Birds of North America. Birds of North America, Inc., Philadelphia, PA. Mayfied, H. 1965. The brown-headed cowbird, with old and new hosts. Living Bird 4:13-28. Mayfield, H. F. 1992. Kirtland's warbler (Dendroica kirtlandii). Account 19 in A. Poole and F. Gill, editors. The Birds of North America. Birds of North America, Inc., Philadelphia. Meyer, R. K., C. Kabat, and I. O. Buss. 1947. Early involutionary changes in postovulatory follicles of the ring-necked pheasant. Journal of Wildlife Management 60:94-107. Middleton, A. L. A. 1991. Failure of brown-headed cowbird parasitism in nests of the American goldfinch. Journal of Field Ornithology 62:200-203. Monroe, B. L., Jr. 1957. A breeding record of the cowbird for Florida. Florida Naturalist 30:124. Ortega, C. P. 1998. Cowbirds and Other Brood Parasites. University of Arizona Press, Tucson. Ortega, J. C., and C. P. Ortega. 2000. Effects of brown-headed cowbirds and predators on the nesting success of yellow warblers in southwest Colorado. Journal of Field Ornithology 71:516-524. Payne, R. B. 1965. Clutch size and numbers of eggs laid by brown-headed cowbirds. Condor 67:44-60. Payne, R. B. 1973. Individual laying histories and the clutch size and numbers of eggs of parasitic cuckoos. Condor 75:414-438. Payne, R. B. 1976. The clutch size and numbers of eggs of brown-headed cowbirds: effects of latitude and breeding season. Condor 78:337-342. Pearson, S. F., and S. Rohwer. 1998. Determining clutch size and laying dates using ovarian follicles. Journal of Field Ornithology 69:587-594. Post, W., T. K. Nakamura, and A. Cruz. 1990. Patterns of shiny cowbird parasitism in St. Lucia and southwestern Puerto Rico. Condor 92:461-469. Potter, E. F., and G. T. Whitehurst. 1981. Cowbirds in the Carolinas. Chat 45:57-68. Pyle, P. 1997. Identification Guide to North American Birds, Part I. Columbidae to Ploceidae. Slate Creek, Bolinas, CA.
44
Reetz, M. J., E. Farley, and T. A. Contreras. In submission. Records of Bachman‟s Sparrow (Aimophila aestivalis) Raising Brown-headed Cowbirds (Molothrus ater) to Fledging. Submitted to Wilson Journal of Ornithology, October 2007. Robinson, S. K., F. R. Thompson, T. M. Donovan, D. R. Whitehead, and J. Faaborg. 1995. Regional forest fragmentation and the nesting success of migratory birds. Science 267:1987-1990. Rothstein, S. I. 1975. An experimental and teleonomic investigation of avian brood parasitism. Condor 77:250-271. Rothstein, S. I. 1990. A model system for coevolution: avian brood parasitism. Annual Review of Ecology and Systematics 21:581-508. Rothstein, S. I., and S. K. Robinson, editors. 2002. Parasitic Birds and Their Hosts: Studies in Coevolution. Oxford University Press, New York. Rothstein, S. I., J. Verner, E. Stevens, and L. V. Verner. 1987. Behavioral differences among sex and age classes of the brown-headed cowbird and their relation to the efficacy of a control program. Wilson Bulletin 99:322-337. SAS Institute Inc. 2003. SAS/STAT User's Guide, version 9.1. SAS Institute Inc., Cary, NC. Sauer, J. R., J. E. Hines, and J. Fallon. 2006. The North American Breeding Bird Survey, DRAFT Results and Analysis 1966-2005. Version 6.2.2006. USGS Patuxent Wildlife Research Center, Laurel, MD. Scott, D. M. 1978. Using sizes of unovulated follicles to estimate laying rate of the brown- headed cowbird. Canadian Journal of Zoology 56:2230-2234. Scott, D. M., and C. D. Ankney. 1979. Evaluation of a method for estimating the laying rate of brown-headed cowbirds. Auk 96:483-488. Scott, D. M., and C. D. Ankney. 1980. Fecundity of the brown-headed cowbird in southern Ontario. Auk 97:677-683. Scott, D. M., and C. D. Ankney. 1983. The laying cycle of brown-headed cowbirds: passerine chickens? Auk 100:583-592. Sealy, S. G., and R. C. Bazin. 1995. Low frequency of observed cowbird parasitism on eastern kingbirds: host rejection, effective nest defense, or parasite avoidance? Behavioral Ecology 6:140-145. Sealy, S. G., D. Neudorf, and D. P. Hill. 1995. Rapid laying by brown-headed cowbirds Molothrus ater and other parasitic birds. Ibis 137:76-84. Sprunt, A., Jr., and E. B. Chamberlain. 1970. South Carolina Bird Life. South Carolina University Press, Columbia. SPSS Inc. 2003. SPSS® Base 12.0 for Windows® User's Guide. SPSS Inc., Chicago. Summers, S. G., R. M. Kostecke, and G. L. Norman. 2006. Efficacy of trapping and shooting in removing breeding brown-headed cowbirds. Wildlife Society Bulletin 34:1107-1112. Vickery, P. D., M. L. Hunter, Jr., and J. V. Wells. 1992. Use of a new reproductive index to evaluate relationship between habitat quality and breeding success. Auk 109:697-705. Whitehead, M. A., S. H. Schweitzer, and W. H. Post. 2002. Cowbird/host interactions in a southeastern old-field: a recent contact area. Journal of Field Ornithology 73:379-386. Woolfenden, B. E., H. L. Gibbs, S. G. Sealy, and D. G. McMaster. 2003. Host use and fecundity of individual female brown-headed cowbirds. Animal Behaviour 66:95-106.
APPENDIX A. Summary of nests and fledge families for 32 species monitored in the study areas, 2004-2006. U: species nesting in the mid-story to canopy, L: species nesting low or on ground. Nesting category based on field observations and Ehrlich et al. (1988). Species totals for each nesting category are given for nests/families monitored. Species with an asterisk include supplemental data from Contreras & Sieving (unpubl. data, 2000-2004). Number of supplemental nests (number parasitized): GCFL = 12, BLJA = 8, COYE = 2 (1), EATO = 43 (3), BACS = 11 (1), EAME = 3.
Unparasitized Parasitized Unparasitized Parasitized Common Name Latin Name Category Nests Nests Families Families Eurasian Collared Dove Streptopelia decaocto U 3 Mourning Dove Zenaida macroura U 42 Great-crested Flycatcher Myiarchus crinitus U 12 Acadian Flycatcher Empidonax virescens U 14 1 Loggerhead Shrike Lanius ludovicianus U 5 1 White-eyed Vireo Vireo griseus L 17 2 25
45 Yellow-throated Vireo Vireo flavifrons U 1 5
Red-eyed Vireo Vireo olivaceus U 4 1 Blue Jay* Cyanocitta cristata U 9 Carolina Chickadee Poecile carolinensis U 1 Eastern Tufted Titmouse Baeolophus bicolor U 2 Carolina Wren Thryothorus ludovicianus L 10 12 Blue-grey Gnatcatcher Polioptila caerulea U 4 1 9 3 Eastern Bluebird Sialia sialis L 44 3 Wood Thrush Hylocichla mustelina U 3 Northern Mockingbird Mimus polyglottos L 449 1 28 Brown Thrasher Toxostoma rufum L 105 7 Northern Parula Parula americana U 18 7 Pine Warbler Dendroica pinus U 3 14 1
Appendix A. Continued. Unparasitized Parasitized Unparasitized Parasitized Common Name Latin Name Category Nests Nests Families Families Prothonotary Warbler Protonotaria citrea L 1 Common Yellowthroat Geothylpis trichas L 2 1 2 Hooded Warbler Wilsonia citrine L 4 1 7 Yellow-breasted Chat Icteria virens L 1 Summer Tanager Piranga rubra U 1 5 3 Bachman‟s Sparrow* Aimophila aestivalis L 10 1 Eastern Towhee Pipilo erythrophthalmus alleni L 54 3 12 Northern Cardinal Cardinalis cardinalis L 254 6 93 2 Blue Grosbeak Guiraca caerulea L 6 1 Indigo Bunting Passerina cyanea L 2 3 46 Eastern Meadowlark Sturnella magna L 3 0
Red-winged Blackbird Agelaius phoeniceus L 81 Common Grackle Quiscalus quiscala U 4 Boat-tailed Grackle Quiscalus major U 2 Orchard Oriole Icterus spurius U 2 3 House Finch Carpodacus mexicanus L 29 TOTALS 35 1178 18 253 17 Mid- or Canopy Species 17/10 106 2 61 15 Lower Nesting Species 15/10 1072 16 192 2