TY - JOUR
AU - Brush,, Timothy
AB - Abstract During 2002–2003, we studied the breeding ecology of the Altamira Oriole (Icterus gularis), a permanent resident that builds pendulous nests in remnant tracts of Tamaulipan brushland in the Lower Rio Grande Valley, Texas. We found 76 active oriole nests, seven of which were reused for second broods, for a total of 83 nesting attempts. Nearly 20% of the breeding individuals in our sample were subadults (second-year orioles). Using a microvideo camera, we were able to estimate incubation and nestling periods of 12.5 and 15.5 days, respectively. Fifty-nine percent of nests fledged at least one young, with successful nests averaging 2.3 fledglings. Failed nests were all intact, indicating that predators entered through the small opening in the top of the nest. Six nests fledged Bronzed Cowbirds (Molothrus aeneus), although two of these nests produced orioles as well. Vegetation analysis suggested that orioles preferred the tallest trees in the study sites in which to place their nests. A greater number of fallen logs was also a predictor of nest sites, which agrees with previous studies suggesting that orioles prefer open woodlands and edges. Formerly vast, dense forests in the Lower Rio Grande Valley have degraded into open woodlands, perhaps benefiting Altamira Orioles during the last 50 years. However, because of the oriole's preference for tall trees, continued degradation of forested tracts may not be beneficial to this species. Resumen Durante 2002 y 2003, estudiamos la ecología reproductiva de Icterus gularis, un residente permanente que construye nidos colgantes en los sectores remanentes de matorrales de Tamaulipas en la cuenca baja del Río Grande, en el estado de Texas. Hallamos 76 nidos activos, siete de los cuales fueron reutilizados por una segunda camada, para un total de 83 intentos de nidificación. Casi el 20% de los individuos reproductores en nuestra muestra fueron subadultos (individuos del segundo año). Usando una cámara de vídeo en miniatura, pudimos estimar el período de incubación en 12.5 días y el período de crecimiento de los polluelos en 15.5 días. El 59% de los nidos criaron por lo menos un polluelo volantón y los nidos exitosos criaron en promedio 2.3 polluelos volantones. Todos los nidos que fracasaron estuvieron intactos, lo que podría indicar que los depredados entraron al nido por la abertura pequeña en la parte superior del nido. Seis nidos criaron ejemplares de Molothrus aeneus, aunque dos de estos nidos también criaron ejemplares de I. gularis. El análisis de la vegetación sugiere que, en los sitios de estudio, I. gularis prefirió los árboles más altos para construir sus nidos colgantes. Un mayor número de troncos caídos cerca del nido predijo también la ubicación de los sitios de nidificación, lo que coincide con estudios anteriores que sugieren que I. gularis prefiere bordes y áreas arboladas abiertas. Los bosques, antes extensos y densos, de la cuenca baja del Río Grande, han sido degradados transformándose en áreas arboladas abiertas, quizás beneficiando a I. gularis durante los últimos 50 años. Sin embargo, debido a la preferencia de I. gularis por los árboles altos, la degradación continua de los bosques puede no ser beneficiosa para esta especie. Introduction The Lower Rio Grande Valley of Texas supports a highly diverse flora and fauna and represents the northernmost range limit of many tropical species (Clover 1937, Blair 1950, Oberholser 1974, Jahrsdoerfer and Leslie 1988). Since the 1920s, the Lower Rio Grande Valley (consisting of Cameron, Hidalgo, Willacy, and Starr counties) has undergone a massive landscape transformation from a mosaic of subtropical evergreen forest, riparian woodland, and chaparral (collectively referred to as Tamaulipan brushland) to agricultural fields and urban developments. An estimated 95% or more of the original native brushland habitat in the Lower Rio Grande Valley has been cleared (Marion 1974, Jahrsdoerfer and Leslie 1988). Following the elimination of major flooding in 1953 with the completion of Falcon Dam, plant communities in some protected areas shifted from subtropical evergreen forest to shorter, denser thorn forest and thorn scrub, accompanied by shifts in avian communities (Gehlbach 1987, Brush and Cantu 1998). Remnant brushland communities have likely also been affected by severe freezes (Lonard and Judd 1991). As a result of these events, many local brushland birds, such as the Plain Chachalaca (Ortalis vetula), White-winged Dove (Zenaida asiatica), Audubon's Oriole (Icterus graduacauda), and Hooded Oriole (I. cucullatus), declined considerably during the twentieth century (Oberholser 1974, Brush 2005). Today, as the landscape continues to change, declining populations of other formerly common species such as the Altamira Oriole (I. gularis) are of conservation concern. The Altamira Oriole is on the Texas Organization for Endangered Species watch list as potentially threatened or endangered in the United States (Texas Organization for Endangered Species 1995), and it is a U.S. Geological Survey Species at Risk (U.S. Geological Survey 2000). It inhabits subtropical and tropical lowland forests from the Lower Rio Grande Valley southward throughout much of Mexico and into Nicaragua (Dickey and van Rossem 1938, Sutton and Pettingill 1943, American Ornithologists' Union 1998). It builds a long (35–50 cm), conspicuous, pendulous nest, 4–15 m high in semiopen forests and riparian woodlands. Because it is placed on the tip of a small flexible branch, the nest is thought to be inaccessible to many predators (Sutton and Pettingill 1943). In South Texas the nest is typically placed on the northwestern side of a tree's canopy, presumably to avoid damage from prevailing southeasterly winds (Pleasants 1981). Although originally collected by Frank B. Armstrong in Brownsville in 1897 (Oberholser 1974), Altamira Orioles in South Texas were not detected by most early naturalists (Sennett 1878, 1879, Smith 1910, Friedmann 1925). The first nest was found in 1951 (Grimes 1953), and Altamira Orioles rapidly became the most common nesting oriole at Santa Ana National Wildlife Refuge and Bentsen–Rio Grande Valley State Park (Oberholser 1974). Populations of Hooded Orioles and Audubon's Orioles decreased during the same period, likely due to heavy brood parasitism by Bronzed (Molothrus aeneus) and Brown-headed (M. ater) Cowbirds, although there are no conclusive data that link parasitism to the declines of these species (Sealy and Underwood 2004). Altamira Orioles have been confirmed to raise Bronzed Cowbirds (Brush 1998), but may avoid high levels of nest parasitism because of the enclosed nest and because they can eject cowbird eggs (Hathcock 2000). During the 1980s and 1990s, Altamira Orioles declined at Santa Ana and other nearby sites (Hathcock and Brush 2004, Werner 2004). Continued habitat degradation is considered to be the leading cause of recent declines in Altamira Oriole populations in the Lower Rio Grande Valley, although few data exist describing breeding success, levels of nest predation and parasitism, foraging requirements, and overwinter survival (Brush 1998). In this study, our objectives were to enhance our knowledge of the breeding ecology of the Altamira Oriole by: (1) determining nesting phenology and nesting success, (2) examining factors limiting nest success and productivity, and (3) describing nest-site selection. Methods Study Area During July and August 2001, we chose most study sites by locating areas that supported large numbers of orioles, or that had recent records of nesting. All study sites were in Hidalgo County, Texas, which is considered to be in the “mid-valley” of the Lower Rio Grande Valley (Jahrsdoerfer and Leslie 1988). Study sites included most of the largest tracts of native Tamaulipan brushland in southern Hidalgo County (Werner 2004; Fig. 1). Climate is semiarid and subtropical (Jahrsdoerfer and Leslie 1988), with an average yearly rainfall of 56 cm peaking in May–June and September, and average high temperatures exceeding 35°C in summer (National Climatic Data Center 2003). Figure 1 Open in new tabDownload slide Sites searched for Altamira Oriole nests in the Lower Rio Grande Valley, Texas, 2002–2003. Site names are: (1) Santa Ana National Wildlife Refuge, (2) Marinoff tract, Lower Rio Grande Valley National Wildlife Refuge (LRGVNWR), (3) Gabrielson tract, LRGVNWR, (4) Anzalduas County Park, (5) Madero tract, LRGVNWR, (6) Madero residential (private land with houses), (7) El Morillo Banco tract, LRGVNWR, (8) Bentsen–Rio Grande Valley State Park, and (9) La Joya tract, LRGVNWR. Figure 1 Open in new tabDownload slide Sites searched for Altamira Oriole nests in the Lower Rio Grande Valley, Texas, 2002–2003. Site names are: (1) Santa Ana National Wildlife Refuge, (2) Marinoff tract, Lower Rio Grande Valley National Wildlife Refuge (LRGVNWR), (3) Gabrielson tract, LRGVNWR, (4) Anzalduas County Park, (5) Madero tract, LRGVNWR, (6) Madero residential (private land with houses), (7) El Morillo Banco tract, LRGVNWR, (8) Bentsen–Rio Grande Valley State Park, and (9) La Joya tract, LRGVNWR. Vegetation at the study sites has been characterized as “mid-valley riparian woodland” and “mid-delta thorn forest” (Jahrsdoerfer and Leslie 1988). The delta region consists of ancient floodways, or resacas, alternating with upland areas, creating a mosaic of different plant communities. Dominant trees included cedar elm (Ulmus crassifolia), sugar hackberry (Celtis laevigata), Mexican ash (Fraxinus berlandieriana), Texas ebony (Chloroleucon ebano), tepeguaje (Leucaena pulverulenta), and mesquite (Prosopis glandulosa). Common shrubs included granjeno (Celtis pallida), brasil (Condalia hookeri), Texas persimmon (Diospyros texana), colima (Zanthoxylum fagara), lotebush (Ziziphus obtusifolia), and la coma (Sideroxylon celastrinum), often forming impenetrable thickets. Plant communities at the study sites are further described by Vora (1990), Brush (1999), and Lonard and Judd (2002). Nest Searching and Monitoring We conducted all nest monitoring during 2002 and 2003. Nests were located either by following orioles that showed nesting behavior (Martin and Geupel 1993) or by randomly searching for the large, conspicuous nests during times of the day when oriole activity was low. We searched all sites except La Joya, Marinoff, and Madero Residential (Fig. 1) from mid-March until the end of August of both years. La Joya was searched from mid-May until the end of July of both years because of logistical constraints. Marinoff and Madero Residential were searched via automobile. We focused nest-searching and monitoring efforts at Bentsen and Santa Ana because of their large size and abundance of orioles during preliminary surveys. We canoed the entire 9.5 km stretch of the Rio Grande at Santa Ana once per season (25 May 2002 and 26 June 2003) to find additional nests along the river. Our field effort was slightly greater in 2003 because of one additional field assistant. We recorded locations and nest-tree species of all completely built nests, including nests that appeared to be inactive (inactive nests usually indicated a nearby breeding pair, because nests from previous years nearly always fall down during the winter months). A nesting attempt was defined as an oriole nest in which at least one egg was laid, even if the nest was reused after a previous clutch had been laid in the same nest. Most nests were monitored every 3–5 days, with more frequent visits during transition periods (e.g., hatching, fledging). We used standard procedures during nest visits to minimize human disturbance (Martin and Geupel 1993). At each nest, we recorded the age of the male and female in the breeding pair as adult or subadult. Second-year Altamira Orioles have mostly yellow, olive, and brown “subadult” plumage until the second prebasic molt, after which they attain the characteristic bright orange and black “adult” plumage (Dickey and van Rossem 1938, Pyle 1997). Although second-year birds are capable of breeding, we refer to them as subadults in this paper. We assumed that females did most or all of the nest-building and incubation (Brush 1998). Nesting attempts were considered successful if at least one oriole chick fledged, and failed if this did not occur. When we could not find fledglings near an empty nest, the young were assumed to have fledged if, during the previous visit, the nestlings were 12 days old or begging very loudly and the nest was still intact during the final visit. We made most nest observations from a distance using binoculars, although determination of the nesting stage was facilitated by using a nest-inspection camera in nests that were accessible and below 11.2 m (Proudfoot 1996, 2002). Altamira Oriole eggs were distinguished from cowbird eggs according to descriptions and illustrations in Baicich and Harrison (1997). For nests that were inaccessible to the camera, the presence of cowbird nestlings was determined by standing close to the nest and listening for cowbird begging calls. The begging calls of Bronzed and Brown-headed Cowbird nestlings are much harsher and more persistent than those of Altamira Oriole nestlings. Cowbird nestlings continue to beg loudly long after the adult has left the nest to forage, whereas oriole nestlings generally do not (SMW, pers. obs.). Although no orioles were color-banded, we were able to follow multiple nesting attempts by the same female or breeding pair with a moderate level of certainty. This is because Altamira Orioles are solitary nesters and maintain exclusive breeding territories (Pleasants 1977), and the presence of subadults in this breeding population allowed for a higher probability of identifying individuals than if all breeding birds had shown adult plumage. Furthermore, several Altamira Orioles at Bentsen displayed unusual plumage patterns (SMW, unpubl. data), allowing for a greater degree of certainty that some individuals were resighted. The assumption of solitary nesting allowed us to estimate yearly productivity per nesting female. We defined a breeding pair as a pair of birds occupying a consistent territory, which was estimated as birds were followed to and from their nests. Inactive nests found in areas not regularly searched were also used to calculate the number of breeding pairs, or territories, at the study sites. Inactive nests were generally assumed to have been built by the same pair if they were within 300 m of each other. Nesting Success Determining Lengths of Nest Stages We visited easily accessible nests every 1–2 days to obtain accurate estimates of when incubation began, when eggs hatched, and when nestlings fledged. Final lengths of the incubation and nestling periods were estimated to the nearest half-day (Martin et al. 1997). Nest Survival We calculated daily nest survival rates, standard errors, and survival probabilities following Mayfield (1961, 1975) and Johnson (1979). We followed Mayfield's (1961, 1975) and Martin et al.'s (1997) suggestions for determining exposure days. We used the average clutch size (four) and the lengths of the incubation and nestling stages to calculate dates for laying, start of incubation, and hatching when field observations were less accurate. A single nest success value that combined laying, incubation, and nestling stages was calculated. For failed nests we calculated the failure date as the half-way point between the last confirmed active date and the date on which the nest was confirmed not active. We excluded nests for which the stage at failure was unknown and could not be estimated. Altamira Oriole Clutch Size and Productivity We determined the average number of eggs laid for nests inspected within five days of the start of incubation (to minimize partial predation effects). We calculated the number of fledglings both from successful nests and from all attempts with known outcomes. Habitat Measurements At each oriole nest, we took two groups of measurements representing different scales: (1) nest-placement variables that described placement of the nest within the tree, and (2) nest-site variables at a larger scale that described the vegetation around the nest. Many of the measurements were derived or modified from BBIRD methodology (Martin et al. 1997). Nest-placement Variables For all nests, we recorded tree species, nest-tree height, diameter at breast height (dbh) of the nest tree, nest height at the nest opening, azimuth from the nest-tree trunk to the nest (“trunk-to-nest angle”), orientation of the nest opening (“nest-opening angle”), and horizontal distance from the nest-tree trunk to the nest. Heights below approximately 8.5 m were measured with a telescopic pole, and heights above 8.5 m were measured with a clinometer and tape measure. Nest-site Variables We measured vegetation at the nest site within a 0.04 ha circular plot (James and Shugart 1970, Martin et al. 1997) centered on the nest (“nest plot”). We used a paired, random-plot design to identify features of the vegetation that were more likely to be associated with oriole nests. The center of the other 0.04 ha circular plot was located in a random compass direction and at a random distance between 20 and 50 m from each nest (“unused plot”). We limited the distance to 50 m to avoid placing the unused plot in a distinctly different habitat such as a grassland or wetland. Nest-site variables were measured in all nest plots and unused plots. We recorded the number and dbh of small trees (dbh 15–30 cm), large trees (dbh >30 cm), and snags (dbh >15 cm, height >1.4 m), and the number of fallen logs (diameter >15 cm and length >3 m). Canopy cover was measured using a concave densiometer at the center of each plot. Within each plot, we placed four 10 m transects in the cardinal directions projecting from the center of the plot. We placed a 7.6 m telescopic pole at point intervals of 2 m along the transects and counted the number of vegetation contacts on the pole, or “hits” (Wiens and Rotenberry 1981, Braden 1999), in each 1 m vertical layer. Thus, the total number of points sampled with the pole in the plot was 21 (five points per cardinal transect and one center point). Hits above 7.6 m were estimated after obtaining nest and tree heights, usually with a clinometer. The tallest vegetation within 10 cm of the pole at each point was recorded as the maximum height variable for the plot. The vertical profile was divided into three strata: ground layer (0–1 m), shrub layer (1–3 m), and tree layer (>3 m). The tallest vegetation was usually around 10 m in height and rarely exceeded 15 m. For each vertical layer, pole hits were used to calculate foliage frequency (the sum of the points with foliage hits divided by 21 points on the plot) and foliage density (the number of foliage hits summed at all 21 points). We calculated variation in vegetation height across the plot, where height variation  =  (maximum vegetation height – minimum vegetation height)/mean vegetation height (Wiens and Rotenberry 1981). Vertical structural diversity was calculated among the three vegetation layers as: vertical structural diversity  =  1/Σ(pi2), where pi is the proportion of foliage hits in vertical layer i on a plot (Hill 1973, Braden 1999). Statistical Analyses Nesting Success We used program CONTRAST (Hines and Sauer 1989) to compare daily nest survival rates among sites and between adult and subadult breeding pairs (i.e., a breeding pair composed of one or two subadults). We used ANOVA to compare the number of orioles fledged per breeding pair among sites, and a Kruskal-Wallis test to compare the number of orioles fledged per nest among study sites, because ANOVA assumptions were not met. Nest-placement and Nest-site Selection We calculated the mean trunk-to-nest angle and mean nest-opening angle at nests, and obtained mean daily wind direction during April, May, and June, 2002–2003, from the weather station at McAllen-Miller International Airport (National Climatic Data Center 2003). We used Rayleigh's test to test the null hypothesis that each of these distributions was random (Zar 1996). Trunk-to-nest angles that were nonrandomly distributed were then compared with wind direction using a Watson-Williams test with an F ratio. We also used Watson-Williams tests to compare nonrandomly distributed trunk-to-nest angles between successful and failed nests and nonrandomly distributed nest-opening angles between successful and failed nests. The successful vs. failed comparisons included one angle per nesting attempt. Thus, some angles were counted twice because of two nesting attempts in the same nest. We used matched-pairs logistic regression (MPLR) to explore habitat preferences for nest placement (Hosmer and Lemeshow 2000). First, mean differences in nest-site variables between nest plots and unused plots were calculated. To select variables for the multivariate MPLR, we entered each of the 14 nest-site variables into a univariate MPLR and retained the variable if the likelihood ratio test was significant at P < 0.25 (Hosmer and Lemeshow 2000). We then checked for collinearity among the selected variables. Highly correlated (|r| > 0.60, P < 0.001) variables were entered into separate multivariate analyses (Beck and George 2000, Chase 2002). We used a backward elimination method, starting with a full model and eliminating variables when they did not significantly contribute to the model (likelihood ratio tests, P < 0.10). We then checked for linearity in the logit and for plausible interaction terms in the reduced models. Goodness-of-fit was assessed with a residual analysis (Hosmer and Lemeshow 2000). We compared the final models using Akaike's information criterion corrected for small sample sizes (AICc) per Burnham and Anderson (2002). Nesting-outcome Habitat Differences We used binary logistic regression with a likelihood ratio statistic to compare mean habitat variables between successful and depredated nests. We chose variables for a multivariate binary logistic regression model in the same manner as used with the MPLR for nest-site selection. All nesting attempts were included in this analysis, including renesting attempts in the seven reused nests. For these comparisons, we considered each nesting attempt an independent datum, assuming that the outcome for each reused nest was not dependent on the outcome of the first attempt. We measured vegetation once at each nest, so reused nests had the same habitat measurements as their ‘original’ nest. We used SPSS for Windows versions 11.0 and 12.0 (SPSS 2001, 2003) for all statistical analyses, except Rayleigh's test and daily survival rate differences. An alpha level of 0.05 was used for all tests unless noted otherwise. Means are presented ± SE. Results During 2002 and 2003, we located 89 completed oriole nests in approximately 55 territories in the study area. Thirteen of the 89 nests were either inactive or likely active but not checked regularly. Of the remaining 76 nests, seven were reused for a second clutch. Thus, we monitored 83 total active nesting attempts (hereafter referred to in the nesting success context as ‘nests’ for simplicity). The proportions of adults and subadults in the breeding population studied were similar in both years. For the 53 breeding pairs that we identified at nests, 82% of individuals were adults in 2002, and 81% were adults in 2003. Over both years, the female adult-to-subadult ratio was 3.4:1, and the male adult-to-subadult ratio was 5.6:1, although these ratios were not significantly different (𝜒21  =  1.0, P  =  0.32). Seventy-two percent (n  =  38) of the breeding pairs were composed of two adults, 6% (n  =  3) were adult females paired with subadult males, 13% (n  =  7) were subadult females paired with adult males, and 9% (n  =  5) were subadult females paired with subadult males. Nesting Phenology The first dates of nest-building were 7 April in 2002 and 28 March in 2003. The latest completed nest of either year was begun on 2 July 2002. The first oriole eggs of each year were laid on 30 April 2002 and 20 April 2003, and the last active dates of each year were 13 August 2002 and 7 August 2003, both of which were fledging events. We knew the exact date of the start of incubation, or the hatching or fledging date, or combinations of the three events for 14 nests. The exact stage lengths were as follows: incubation, 12.5 ± 0.3 days (range: 11–14 days, n  =  10); nestling, 15.2 ± 0.3 days (range: 14–16 days, n  =  7); and incubation plus nestling, 28.1 ± 0.6 days (range: 26–31 days, n  =  8). Final lengths after considering the rest of the nest data, rounded to the nearest half-day, were 12.5 days for incubation, 15.5 days for nestling, and 28.0 days for incubation plus nestling. Nesting Success We were able to determine the outcome of 80 of 83 nesting attempts. Of the 31 failed nests, seven (23%) failed during egg-laying, 17 (55%) failed during incubation, and four (13%) failed during the nestling stage. We were uncertain of the stage at failure of the remaining three failed nests. Six oriole nests (7%) fledged Bronzed Cowbirds, fledging one to three cowbirds each. We excluded six nests from the Mayfield analysis due to uncertain outcomes or unknown stages at failure. The nestling period appeared to be the most successful stage (Table 1). Table 1 Mayfield daily survival rates for Altamira Oriole nests in the Lower Rio Grande Valley, Texas, 2002–2003. Daily nest survival values of grouped rows were compared with program CONTRAST. Asterisks indicate significantly different survival rates (P < 0.05). Open in new tab Table 1 Mayfield daily survival rates for Altamira Oriole nests in the Lower Rio Grande Valley, Texas, 2002–2003. Daily nest survival values of grouped rows were compared with program CONTRAST. Asterisks indicate significantly different survival rates (P < 0.05). Open in new tab Clutch Size and Productivity For the 30 nests that we inspected within the first five days of the onset of incubation, clutch size was 3.9 ± 0.2 eggs (range: 2–6). Because the camera was better at detecting eggs than nestlings, we knew the exact number of fledglings for only 14 of the 49 successful nests. This number was 2.4 ± 0.3 (range: 1–5, n  =  14), which was similar to the overall mean of 2.3 ± 0.2 (range: 1–5, n  =  49). Adult pairs produced 2.4 ± 0.2 fledglings (n  =  41), and pairs with one or two subadults produced 1.6 ± 0.3 fledglings (n  =  8). For the 47 breeding pairs with 80 nesting attempts of known outcome, the mean number of orioles fledged per nest was 1.4 ± 0.2. There was no difference in the number of orioles fledged per nest among sites (Bentsen: 1.5 ± 0.2, n  =  39; Santa Ana: 1.1 ± 0.3, n  =  21; other sites: 1.5 ± 0.4, n  =  20; Kruskal-Wallis 𝜒22  =  1.8, P  =  0.4). For yearly output per breeding pair, there was a trend toward higher fecundity at Bentsen (2.8 ± 0.4, n  =  21) versus the other sites (Santa Ana: 2.3 ± 0.7, n  =  10; other sites: 1.8 ± 0.5, n  =  16), but this difference was not statistically significant (F2,44  =  1.3, P  =  0.29). The maximum number of successful broods by a single pair during a season was two. Twelve nesting pairs each fledged two broods during this study. All but one of these pairs were composed of two adults (the exception was a subadult female paired with an adult male). The maximum number of clutches laid by a single pair in one season was four. Seven nests were reused, all by adult pairs. Two of the nests were reused after failed attempts. Factors Affecting Nest Success and Productivity Three of the 31 nest failures were caused by the nest falling from the nest tree (two broken nest branches and one fallen nest tree). None of the other 28 failed nests appeared to have any structural damage (e.g., ripped open), and no nests were abandoned with intact eggs or live chicks. We did not observe any predators enter nests, but we observed 11 nest entries by Bronzed Cowbirds, which are known to pierce host eggs and other cowbird eggs (Carter 1986, Peer and Sealy 1999). Only five of these entries were into nests containing eggs, and we could not confirm if the cowbirds pierced any of the eggs. We documented 17 partial oriole clutch losses from 12 different nests where the female continued to incubate the remaining egg(s). One Bronzed Cowbird egg was present in one nest during these inspections. We saw Bronzed Cowbird eggs in six nests, three of which remained active and two of which had been abandoned the day before. At two nests Bronzed Cowbirds eggs were removed between nest visits, although we never observed orioles actually removing eggs from any nests. Of the six oriole nests that fledged Bronzed Cowbirds, two also produced one and two oriole fledglings, respectively, from the same brood. Two (33%) of the six successfully parasitized nests had a subadult as one member of the breeding pair, which was similar to the proportion of subadult pairs (28%) in the sample population. Nest-site Selection Nest Placement Altamira Orioles built their nests in 12 different tree species, on low-voltage power lines, and on a television antenna on a house (Table 2). In 2003, 11 nests were located within 5 m of a nest from the previous year. Seven of these nests were in nearly exactly the same location as the 2002 nest. An interesting case occurred when a subadult female built a nest in 2003 on the same branch on which an adult female built a nest in 2002. Table 2 Tree species associated with overall number of nests found and successful or depredated nesting attempts by Altamira Orioles in the Lower Rio Grande Valley, Texas, 2002–2003. Parentheses indicate number of nests found in dead trees included in preceding total. Number of nests found does not necessarily equal number of nesting attempts due to inactive nests, reused nests, or uncertain outcomes. Open in new tab Table 2 Tree species associated with overall number of nests found and successful or depredated nesting attempts by Altamira Orioles in the Lower Rio Grande Valley, Texas, 2002–2003. Parentheses indicate number of nests found in dead trees included in preceding total. Number of nests found does not necessarily equal number of nesting attempts due to inactive nests, reused nests, or uncertain outcomes. Open in new tab Mean nest height was 8.8 ± 0.3 m (range: 4.1–14.0 m, n  =  67), mean nest-tree height was 12.7 ± 0.3 m (range: 7.7–18.4 m, n  =  63), mean nest-tree dbh was 32.0 ± 1.1 cm (range: 8.8–55.6 cm, n  =  63), and horizontal trunk-to-nest distance was 5.4 ± 0.2 m (range: 2.3–10.5 m, n  =  63). Mean trunk-to-nest and nest-opening angles were nonrandom and northwest in orientation. Mean wind direction was from the southeast and was also nonrandom. The trunk-to-nest angle was 316.5° (Rayleigh's z  =  39.8, P < 0.001, n  =  63) and the nest-opening angle was 311.7° (Rayleigh's z  =  33.4, P < 0.001, n  =  67). The mean daily wind direction was 134.3° (Rayleigh's z  =  116.6, P < 0.001, n  =  162). The nest-to-trunk angle of 136.5° (the opposite of 316.5°) did not differ from the mean wind direction (Watson-Williams F1,223  =  0.2, P > 0.25). Differences Between Nest Plots and Unused Plots Nest plots generally had slightly lower foliage frequencies than unused plots at heights below 8 m, but taller vegetation (Fig. 2). Of the 14 nest-site variables considered for the nest-site selection MPLR, 10 were significant at the P  =  0.25 level in univariate analyses (Table 3). The best final MPLR model indicated that canopy cover, number of logs, and maximum height were the best predictors of nest-site selection (Table 4). There were no significant plausible interaction terms in any of the models. Figure 2 Open in new tabDownload slide Foliage frequency (percentage of 21 points for each vertical layer on a plot that had foliage vegetation contacts, or “hits,” against a vertical pole) for nest plots (black bars; n  =  66) and unused plots (gray bars; n  =  63) of Altamira Orioles in the Lower Rio Grande Valley, Texas, 2002–2003. Error bars represent 2 SE. Frequency of foliage hits were similar at most strata between nest plots and unused plots, but nest plots had greater amounts of vegetation at greater heights. Figure 2 Open in new tabDownload slide Foliage frequency (percentage of 21 points for each vertical layer on a plot that had foliage vegetation contacts, or “hits,” against a vertical pole) for nest plots (black bars; n  =  66) and unused plots (gray bars; n  =  63) of Altamira Orioles in the Lower Rio Grande Valley, Texas, 2002–2003. Error bars represent 2 SE. Frequency of foliage hits were similar at most strata between nest plots and unused plots, but nest plots had greater amounts of vegetation at greater heights. Table 3 Summary of mean ± SE differences in nest-site variables between paired nest plots and unused plots for Altamira Orioles (n  =  63) in the Lower Rio Grande Valley, Texas, 2002–2003. Likelihood ratio test statistic (LRS 𝜒21) and P-values are from univariate 1:1 matched-pairs logistic regression (MPLR). Asterisks denote significance for inclusion in the multivariate MPLR (P < 0.25). Open in new tab Table 3 Summary of mean ± SE differences in nest-site variables between paired nest plots and unused plots for Altamira Orioles (n  =  63) in the Lower Rio Grande Valley, Texas, 2002–2003. Likelihood ratio test statistic (LRS 𝜒21) and P-values are from univariate 1:1 matched-pairs logistic regression (MPLR). Asterisks denote significance for inclusion in the multivariate MPLR (P < 0.25). Open in new tab Table 4 Final matched-pairs logistic regression (MPLR) models describing nest-site selection by Altamira Orioles (n  =  63) in the Lower Rio Grande Valley, Texas, 2002–2003. Log(⁠ ⁠)  =  maximized log likelihood function of the full model vs. the null model; K  =  number of parameters; Δi  =  difference in Akaike's information criterion corrected for small sample size (AICc); and wi  =  Akaike weights. Only models with ΔAICc ≤ 10.0 are listed. Open in new tab Table 4 Final matched-pairs logistic regression (MPLR) models describing nest-site selection by Altamira Orioles (n  =  63) in the Lower Rio Grande Valley, Texas, 2002–2003. Log(⁠ ⁠)  =  maximized log likelihood function of the full model vs. the null model; K  =  number of parameters; Δi  =  difference in Akaike's information criterion corrected for small sample size (AICc); and wi  =  Akaike weights. Only models with ΔAICc ≤ 10.0 are listed. Open in new tab Habitat Differences Between Successful and Depredated Nests All but two of the nest-placement and nest-site variables were similar at successful nests and depredated nests. Successful nests had more associated downed logs than depredated nests (4.6 ± 0.6, n  =  46, vs. 2.2 ± 0.6, n  =  25; 𝜒21  =  7.7, P < 0.01). Successful nests were also in areas with greater shrub-layer foliage density than depredated nests (81.5 ± 10.0, n  =  46, vs. 51.5 ± 6.6, n  =  25; 𝜒21  =  4.6; P < 0.05). Interestingly, shrub-layer foliage density at successful nests approached the value of shrub-layer foliage density of unused plots (Table 3). Most (19 of 30) nesting attempts made in cedar elm were successful, but there were few if any trends relating nest-tree species with nest outcome (Table 2). Discussion Presence of Subadults in the Breeding Population This is the first documentation of such a large number of breeding subadult Altamira Orioles, small numbers of which were first noticed in 1996 by TB (unpubl. data). Earlier studies either made no mention of breeding subadults or stated that subadults did not breed (Dickey and van Rossem 1938, Sutton and Pettingill 1943, Skutch 1954, Pleasants 1977, 1981, 1993, Gehlbach 1987), although breeding by second-year birds of other oriole species is well documented (Scharf and Kren 1996, Rising and Williams 1999). In the Lower Rio Grande Valley, the number of Altamira Oriole nesting pairs composed of one or more subadults has increased during the last ten years (Werner 2004; TB, unpubl. data). Subadult pairs successfully fledged young, albeit at a lower rate than adults. Subadult and adult pairs appeared to be equally prone to raising cowbirds. Further research is needed on subadult breeding in Altamira Orioles throughout their range. Incubation and Nestling Periods Our estimate of 12.5 days for the incubation period is slightly shorter than Pleasants' (1993) estimate of 14 days. Corman and Monson (1995) noted similar incubation (12–14 days) and nestling periods (14 days) for the ecologically similar Streak-backed Oriole (Icterus pustulatus) breeding in Florida. Nesting Success The vast majority of nest losses appeared to be predator-related, and survivorship was highest during the nestling stage. Hathcock (2000) obtained similar survival rates for nestlings, including no losses to predation during this stage. Identification of nest predators was lacking in this study and remains a poorly understood area of Altamira Oriole nesting ecology (Brush 1998, Hathcock 2000, Werner 2004). Cowbird Parasitism and Predation Records of successful parasitism by Bronzed Cowbirds of Altamira Oriole nests are extremely rare and were mostly anecdotal before the 1990s (Dickey and Van Rossem 1938, Carter 1986). Brush (1998) reported successful cowbird parasitism of Altamira Orioles in 1996 and 1997, and Hathcock (2000) found one successfully parasitized nest out of 22 active Altamira Oriole nests (5%). During 2002–2003, we found a similar proportion of successfully parasitized nests (six of 80, or 8%). The large number of partial clutch losses (17 eggs in 12 different nests) observed in this study could have resulted from egg-piercing by Bronzed Cowbirds followed by egg-removal by an oriole before our subsequent inspection. Hathcock (2000) observed only one partial clutch loss out of 25 oriole nests monitored during incubation. Further data are needed on partial clutch losses and why some Altamira Orioles appear to reject cowbird eggs, while others accept them (Sealy and Underwood 2004). Renesting Nearly one-fourth (12 of 47) of nesting pairs attempted multiple broods. It is probably energetically advantageous for Altamira Orioles to reuse nests for second broods, as occurred in 9% (seven of 76) of the total nesting attempts in this study. The pendulous nests often deteriorate over the course of a nesting cycle, and the condition of the nest may be a key factor in determining whether it is reused for a second clutch. Altamira Orioles' propensity to nest in the same tree for two to three years (Brush 1998, this study) could help habitat managers target specific nesting sites for conservation. Nest-site Selection Downed logs were significant predictors of nest sites as well as successful nests. This variable in combination with maximum vegetation height suggests that orioles selected areas with larger trees and potentially higher levels of recent tree mortality. Pleasants (1977) also noted that Altamira Orioles at Santa Ana often used the largest trees that were emergent over the forest canopy. Because many of the study areas were considered thorn forest or thorn scrub during 2002–2003 (i.e., vegetation was shorter and denser than what was described historically), orioles may have had few choices in selecting sites to build their pendulous nests. The patchy remaining stands of cedar elm and hackberry, many of which also had high levels of dying trees, were likely the most suitable habitat. The most important trees in our study (cedar elm and sugar hackberry) differ somewhat from those in a study by Hathcock and Brush (2004) at Santa Ana, who found that the most frequently used nest-trees were black willows and Mexican ashes located along the Rio Grande or other wetland edges. While many nests at Santa Ana in our study were also located near wetlands, most nests at other sites were located away from the river (Werner 2004). Pleasants (1977) found that Altamira Orioles commonly nested in tepeguaje, but this tree species was not observed to support any nests at Santa Ana during 1993–1999, possibly due in part to diminished numbers of this species after several catastrophic freezes and drought (Hathcock and Brush 2004). In contrast, we observed 10 total nests in tepeguaje, including three at Santa Ana. This species may be important to orioles in the future if its numbers increase without disruption from freezes or drought. Brush (1998) characterized Altamira Orioles as an “edge” species, and they prefer open, often secondary-growth woodlands with scattered trees throughout their range (Dickey and van Rossem 1938, Skutch 1954, Howell and Webb 1995). The patchy woodlands of the Lower Rio Grande Valley in the late twentieth century may have been ideal habitat for the orioles. The apparent range expansion appears similar to that observed by Dickey and van Rossem (1938:526) in El Salvador, where “the clearing of the forest on mountain slopes has permitted both the mimosa and, following it, the [Altamira] orioles to reach 4500 feet on both the volcanoes of Santa Ana and San Salvador... far above the normal ranges of either.” However, if suitable habitat in the Lower Rio Grande Valley continues to diminish, the long-term potential for this species to exist there remains in question. Future Research and Conservation Priorities Results from this study show that Altamira Orioles nested in the largest trees in the study sites, and that some orioles were susceptible to cowbird parasitism. Despite these findings, much remains unknown about this species, such as specific causes of nest predation and egg loss, and overwinter survival and ecology. It is unclear whether some aspects of Altamira Oriole breeding biology (e.g., breeding subadults) in this population are typical of populations elsewhere, because of the limited number of studies done to date. Maintaining Altamira Oriole populations in the rapidly urbanizing Lower Rio Grande Valley will likely require the continued presence of large trees and the water to support them. Artificial flooding at Santa Ana has been successful at sustaining large trees, some of which were used by orioles in this study. Forested areas at Bentsen have degraded substantially in recent years, and many nest-trees used by orioles in this study were unsuitable or had fallen down by 2007 (SMW, unpubl. data). A flooding program similar to the one at Santa Ana could increase nesting habitat for Altamira Orioles at Bentsen. Several nests in our study were located in restored woodlands at La Joya and El Morillo Banco tracts that were planted in 1992 and 1993, respectively (C. Best, U.S. Fish and Wildlife Service, unpubl. data), suggesting potential restoration benefits for orioles (Werner 2004). The U.S. Fish and Wildlife Service aims to reestablish a continuous wildlife corridor along the Rio Grande and reduce the insularity of the refuge tracts (U.S. Fish and Wildlife Service 1997). Long-term efforts should be made to determine how Altamira Oriole populations respond to this restoration in the face of continued habitat degradation. Acknowledgments We thank the U.S. Geological Survey/U.S. Fish and Wildlife Service Species at Risk Program and the Texas Parks and Wildlife Conservation Action Grant program for funding this research. Personnel at the Santa Ana/Lower Rio Grande Valley National Wildlife Refuge complex and Bentsen–Rio Grande Valley Sate Park were very generous in providing access to the study sites and lodging for field personnel. Linda Drabek, Veronica Pedro, and Jon Dale provided invaluable field assistance. Comments on the manuscript from R. Douglas Slack, Fred Smeins, Keith Arnold, Kathryn Purcell, Brian Peer, and Kevin Omland were extremely helpful. We also thank David Blankinship, Mitchell Sternberg, Jeff Rupert, Dorie Stolley, Stephen Winter, Chris Best, Patty Alexander, Glenn Proudfoot, and Kathy Granillo for advice and support during the project. Literature Cited American Ornithologists' Union 1998 . Check-list of North American birds. 7th ed Washington, DC American Ornithologists' Union . Baicich , P. J. , and C. J O. Harrison . 1997 . A guide to the nests, eggs, and nestlings of North American birds. 2nd ed San Diego, CA Academic Press . Beck , M. J. , and T. L. George . 2000 . Song post foraging site characteristics of breeding Varied Thrushes in northwestern California. Condor 102 : 93 – 103 . Google Scholar Crossref Search ADS WorldCat Blair , W. F. 1950 . The biotic provinces of Texas. Texas Journal of Science 2 : 93 – 117 . WorldCat Braden , G. T. 1999 . Does nest placement affect the fate or productivity of California Gnatcatcher nests?. Auk 116 : 984 – 993 . Google Scholar Crossref Search ADS WorldCat Brush , T. 1998 . A closer look: Altamira Oriole. Birding 30 : 46 – 53 . WorldCat Brush , T. 1999 . Current status of Northern Beardless-Tyrannulet and Tropical Parula in Bentsen-Rio Grande Valley State Park and Santa Ana National Wildlife Refuge, southern Texas. Bulletin of the Texas Ornithological Society 32 : 3 – 12 . WorldCat Brush , T. 2005 . Nesting birds of a tropical frontier: the Lower Rio Grande Valley of Texas College Station, TX Texas A&M University Press . Brush , T. , and A. Cantu . 1998 . Changes in the breeding bird community of subtropical evergreen forest in the Lower Rio Grande Valley of Texas, 1970s–1990s. Texas Journal of Science 50 : 123 – 132 . WorldCat Burnham , K. P. , and D. R. Anderson . 2002 . Model selection and multimodel inference: a practical information-theoretic approach. 2nd ed New York Springer-Verlag . Carter , M. D. 1986 . The parasitic behavior of the Bronzed Cowbird in south Texas. Condor 88 : 11 – 25 . Google Scholar Crossref Search ADS WorldCat Chase , M. K. 2002 . Nest site selection and nest success in a Song Sparrow population: the significance of spatial variation. Condor 104 : 103 – 116 . Google Scholar Crossref Search ADS WorldCat Clover , E. U. 1937 . Vegetational survey of the Lower Rio Grande Valley, Texas. Madroño 4 : 41 – 72 . 77 – 100 . WorldCat Corman , T. , and G. Monson . 1995 . First United States nesting records of the Streak-backed Oriole. Western Birds 26 : 49 – 53 . WorldCat Dickey , D. R. , and A. J. van Rossem . 1938 . The birds of El Salvador. Field Museum of Natural History Zoological Series 23 . Friedmann , H. 1925 . Notes on the birds observed in the Lower Rio Grande Valley of Texas during May, 1924. Auk 42 : 537 – 554 . Google Scholar Crossref Search ADS WorldCat Gehlbach , F. R. 1987 . Natural history sketches, densities, and biomass of breeding birds in evergreen forests of the Rio Grande, Texas, and the Rio Corona, Tamaulipas, Mexico. Texas Journal of Science 39 : 241 – 251 . WorldCat Grimes , S. A. 1953 . Black-throated Oriole (Icterus gularis) nesting in Texas. Auk 70 : 207 . WorldCat Hathcock , C. R. 2000 . Factors affecting the reproductive success in hosts of the Bronzed Cowbird (Molothrus aeneus) in the Lower Rio Grande Valley, Texas. M.Sc. thesis Edinburg, TX University of Texas–Pan American . Hathcock , C. R. , and T. Brush . 2004 . Breeding decline and nest-site shift of the Altamira Oriole at Santa Ana National Wildlife Refuge, Texas. Southwestern Naturalist 49 : 33 – 38 . Google Scholar Crossref Search ADS WorldCat Hill , M. O. 1973 . Diversity and evenness: a unifying notation and its consequences. Ecology 54 : 427 – 432 . Google Scholar Crossref Search ADS WorldCat Hines , J. E. , and J. R. Sauer . 1989 . Program CONTRAST: a general program for the analysis of several survival or recovery rate estimates. U.S. Fish and Wildlife Service, Fish and Wildlife Technical Report 24. Hosmer , D. W. , and S. Lemeshow . 2000 . Applied logistic regression. 2nd ed New York John Wiley and Sons . Howell , S. G. , and S. Webb . 1995 . A guide to the birds of Mexico and northern Central America New York Oxford University Press . Jahrsdoerfer , S. E. , and D. M. Leslie Jr . 1988 . Tamaulipan brushland of the Lower Rio Grande of South Texas: description, human impacts, and management options. U.S. Fish and Wildlife Service Biological Report 88(36). James , F. C. , and H. H. Shugart . 1970 . A quantitative method of habitat description. Audubon Field Notes 24 : 727 – 736 . WorldCat Johnson , D. H. 1979 . Estimating nest success: the Mayfield method and an alternative. Auk 96 : 651 – 661 . WorldCat Lonard , R. I. , and F. W. Judd . 1991 . Comparison of the effects of the severe freezes of 1983 and 1989 on native woody plants in the Lower Rio Grande Valley, Texas. Southwestern Naturalist 36 : 213 – 217 . Google Scholar Crossref Search ADS WorldCat Lonard , R. I. , and F. W. Judd . 2002 . Riparian vegetation of the lower Rio Grande. Southwestern Naturalist 47 : 420 – 432 . Google Scholar Crossref Search ADS WorldCat Marion , W. R. 1974 . Status of the Plain Chachalaca in south Texas. Wilson Bulletin 86 : 200 – 205 . WorldCat Martin , T. E. , and G. R. Geupel . 1993 . Nest-monitoring plots: methods for locating nests and monitoring success. Journal of Field Ornithology 64 : 507 – 519 . WorldCat Martin , T. E. , C. R. Paine , C. J. Conway , W. M. Hochachka , P. Allen , and W. Jenkins . 1997 . BBIRD field protocol Missoula, MT Montana Cooperative Wildlife Research Unit, University of Montana . Mayfield , H. F. 1961 . Nesting success calculated from exposure. Wilson Bulletin 73 : 255 – 261 . WorldCat Mayfield , H. F. 1975 . Suggestions for calculating nest success. Wilson Bulletin 87 : 456 – 466 . WorldCat National Climatic Data Center [online] 2003 . McAllen Miller International Airport. <http://www.ncdc.noaa.gov/oa/climate/stationlocator.html> (2 December 2003). Oberholser , C. H. 1974 . The bird life of Texas. Vol. 2 Austin, TX University of Texas Press . Peer , B. D. , and S. G. Sealy . 1999 . Parasitism and egg puncture behavior by Bronzed and Brown-headed Cowbirds in sympatry. Studies in Avian Biology 18 : 235 – 240 . WorldCat Pleasants , B. Y. 1977 . Ecological factors affecting the breeding dispersion patterns of some North American orioles (Icterus). Ph.D. dissertation Los Angeles, CA University of California . Pleasants , B. Y. 1981 . Aspects of the breeding biology of a subtropical oriole, Icterus gularis. Wilson Bulletin 93 : 531 – 537 . WorldCat Pleasants , B. Y. 1993 . Altamira Oriole (Icterus gularis). In Poole , A. , and F. Gill , editors. The birds of North America, No. 56 Philadelphia, PA The Academy of Natural Sciences . Washington, DC and The American Ornithologists' Union . Proudfoot , G. A. 1996 . Miniature video-board camera used to inspect natural and artificial nest cavities. Wildlife Society Bulletin 24 : 528 – 530 . WorldCat Proudfoot , G. A. 2002 . Two optic systems assist removal of nestlings from nest cavities. Wildlife Society Bulletin 30 : 956 – 959 . WorldCat Pyle , P. 1997 . Identification guide to North American birds. Part I Bolinas, CA Slate Creek Press . Rising , J. D. , and P. L. Williams . Bullock's Oriole (Icterus bullockii). In Poole , A. , and F. Gill , editors. The birds of North America, No. 416 Philadelphia, PA The Birds of North America, Inc . 1999 . Scharf , W. C. , and J. Kren . Orchard Oriole (Icterus spurius). In Poole , A. , and F. Gill , editors. The birds of North America, No. 255 Philadelphia, PA The Academy of Natural Sciences . Washington, DC and The American Ornithologists' Union . 1996 . Sealy , S. G. , and T. J. Underwood . 2004 . Accepters and rejecters of cowbird parasitism in the New World orioles (Icterus spp.). Ornitología Neotropical 15 : 331 – 347 . WorldCat Sennett , G. B. 1878 . Notes on the ornithology of the Lower Rio Grande of Texas from observations made during the season of 1877. Bulletin of the U.S. Geological and Geographical Survey of the Territories 4 : 1 – 66 . WorldCat Sennett , G. B. 1879 . Further notes on the ornithology of the Lower Rio Grande of Texas, from observations made during the spring of 1878. Bulletin of the U.S. Geological and Geographical Survey of the Territories 5 : 377 – 440 . WorldCat Skutch , A. F. 1954 . Life histories of Central American birds. Pacific Coast Avifauna No. 31. Smith , A. P. 1910 . Miscellaneous bird notes from the Lower Rio Grande. Condor 12 : 93 – 103 . Google Scholar Crossref Search ADS WorldCat SPSS 2001 . SPSS for Windows: release 11.0.1 Chicago SPSS Inc . SPSS 2003 . SPSS for Windows: release 12.0.0 Chicago SPSS Inc . Sutton , G. M. , and O. S. Pettingill Jr . 1943 . The Alta Mira Oriole and its nest. Condor 45 : 125 – 132 . Google Scholar Crossref Search ADS WorldCat Texas Organization for Endangered Species 1995 . Endangered, threatened, and watch list of vertebrates of Texas, fourth revision. Publication 10 Austin, TX Texas Organization for Endangered Species . U.S. Fish and Wildlife Service 1997 . Lower Rio Grande Valley and Santa Ana National Wildlife Refuges Interim Comprehensive Management Plan Alamo, TX U.S. Fish and Wildlife Service . U.S. Geological Survey 2000 . Species at risk: pre-proposal package, fiscal year 2000 Reston, VA U.S. Geological Survey . Vora , R. S. 1990 . Plant communities of the Santa Ana National Wildlife Refuge, Texas. Texas Journal of Science 42 : 115 – 128 . WorldCat Werner , S. M. 2004 . Breeding biology and habitat associations of the Altamira Oriole and Northern Beardless-Tyrannulet in the Lower Rio Grande Valley, Texas. M.Sc. thesis College Station, TX Texas A&M University . Wiens , J. A. , and J. T. Rotenberry . 1981 . Habitat associations and community structure of birds in shrubsteppe environments. Ecological Monographs 51 : 21 – 41 . Google Scholar Crossref Search ADS WorldCat Zar , J. H. 1996 . Biostatistical analysis. 3rd ed Upper Saddle River, NJ Prentice-Hall . Author notes Present address: BioResource Consultants, P.O. Box 1539, Ojai, CA 93024-1539. E-mail: smwerner@yahoo.com © The Cooper Ornithological Society 2007
TI - Breeding Ecology of the Altamira Oriole in the Lower Rio Grande Valley, TexasEcología Reproductiva De Icterus Gularis De La Cuenca Baja Del Río Grande, TexasAltamira Oriole Breeding Ecology
JF - Condor: Ornithological Applications
DO - 10.1093/condor/109.4.907
DA - 2007-11-01
UR - https://www.deepdyve.com/lp/oxford-university-press/breeding-ecology-of-the-altamira-oriole-in-the-lower-rio-grande-valley-ED0NKPDP7f
SP - 907
VL - 109
IS - 4
DP - DeepDyve
ER -