Research and Management of the Brown-headed Cowbird in Western Landscapes.—Michael L. Morrison, Linnea S. Hall, Scott K. Robinson, Stephen I. Rothstein, D. Caldwell Hahn, and Terrell D. Rich (Eds). 1999. Studies in Avian Biology, Vol. 18. 312 pages, 12 maps, 94 tables, 80 figures. ISBN 1-891276-06-9. Cloth, $18.00.—This collection of 33 papers and 3 summaries by 89 authors is a subset of 67 presentations given at a symposium held at California State University, Sacramento, California, 23–25 October 1997, and sponsored by the Partners in Flight Research Group, U.S. Bureau of Land Management, and U.S. Geological Survey Biological Resources Division. Much work on cowbird–host interactions has been accomplished in central and eastern North America. This symposium was organized to bring together the results of current work and the perspectives on potential affects of cowbirds on hosts in western landscapes. Although preceding it in publication of papers, it follows the lead of a symposium held in Austin, Texas in 1993 (Smith et al. 2000). In large part, this volume was quickly completed through determination and reciprocal reviews by the contributing authors and editors.

The collection of papers is divided into three sections. The first section deals with cowbird ecology and factors affecting cowbird abundance and distribution. The second addresses effects of brood parasitism on host reproductive performance and discusses the potential criteria for taking management action. The third section focuses on some aspects of cowbird control.

The authors of the summary chapters (Hall, Robinson, Rothstein, and Smith) present what is depicted as emerging perspectives and consensus, underscoring heterogeneity in cowbird distributions and their potential effects on host populations. The summary chapters also include the perspective that, although cowbirds may have significant influence on the success of parasitized nests, their effect on most host populations may be relatively minor. Under that perspective, concerns for cowbird control may be overrated and should be limited to some endangered species or local populations. In addition, alternatives to cowbird trapping and removal need more consideration, particularly at the landscape level. Rich, in the preface, also acknowledges the need for a better understanding of proximate relationships between cowbirds and cattle and spatial relationships between host parasitism patterns and landscape features of the West.

In many ways, the volume represents the current state of cowbird research and management. Some papers depict clear advances in our understanding of cowbird distributions and the relationships of cowbirds to their hosts. Collectively, the volume incorporates a great amount of data on cowbird biology, a testimony to the current level of activity. Its weighting to western landscapes also presents an emerging slant on cowbird–host biology that is unique to this volume. However, although the volume contains some of the best current research and some expansions of perspective, it also includes some significant analytic and conceptual limitations.

The initial section may be the volume's greatest strength. Those papers generally deal with issues affecting cowbird distribution at landscape levels and how habitat features may be related to high or low rates of parasitism. Robinson develops a respectable summary of that section, providing previous perspectives, then presenting hypotheses and predictions such that they can be evaluated with the results of the contained papers. Tewksbury et al. may provide one of the strongest analytical algorithms and its results at the local landscape level. Young and Hutto give an expansive evaluation of cowbird abundance at a broader landscape level, whereas Staab and Morrison identify specific habitat features that are related to higher parasitism levels. Several papers examining cowbird abundance and parasitism intensity show limited effects on hosts in shrubsteppe habitats (Van der Haegen and Walker, Farmer, and Ellison).

The second section, summarized by Smith, however, may be more problematic than enlightening. Smith attempts to isolate consequences of parasitism on host populations from consequences on host individuals or host nests. He acknowledges the lack of reliable estimates for “key demographic parameters needed to model the impacts of parasitism reliably.” Few studies deal with intrinsic growth rates. Under that handicap, he develops an argument that, even when the cost to some host individuals is high, the cost of parasitism to host populations or host communities is relatively small; only isolated populations of a few select species with limited distributions may be under significant threat from parasitism. From his interpretations, he develops some management guidelines for deciding when to implement local cowbird-control programs.

Two basic problems are evident in that section and in the cowbird literature at large. First, there are several prevalent analytical misconceptions; many statistics are accepted with little or no consideration as to whether the results were products of biological processes or artifacts of sampling design or analysis, or both. For example, almost all estimates of parasitism level are given as percents of observed samples, with the failure to recognize lessons of Mayfield (1961, 1975) indicating that samples of passerine nests are generally incomplete and nonrandom. The same principles that would cause one to apply the Mayfield approach for evaluating nest predation also apply to nests that are parasitized and abandoned. Because nest sample data will be most deficient in nests that fail early in their nesting cycles, a disproportionate number of nests that were parasitized and abandoned will be missed. The incomplete and nonrandom nest sample issue incorporates itself significantly in summary analyses that use studies with biased statistics (e.g. Lorenzana and Sealy, Whitfield and Sogge).

Several papers attempting to deal with seasonal fecundity overlooked renestings, or assumed, implicitly or explicitly, that females of their study species nested only once during their breeding seasons (e.g. Greene; Hahn et al.). Greene's general algorithm for assessing intrinsic growth rates was perhaps most respectable in the volume. However, his claim of limited renesting for an “intensively monitored” group of Lazuli Buntings (Passerina amoena) appeared to be generated from a small and perhaps opportunistic sample of nests (Table 3); that may have created the pessimistic conclusions regarding the threat of cowbirds to bunting populations. In contrast, the failure of Hahn et al. to consider renestings of hosts in deriving an upper bound of only 8.16 eggs laid per female cowbird in acceptor host nests may have created the authors' relatively benign conclusion for cowbird effects on hosts. The assumption of one to three renestings per host female would adjust estimates of Hahn et al. to 16–35 eggs per female cowbird. Their egg estimation generated speculation of much higher parental investment by cowbirds in their eggs, concepts empirically unsupported in their manuscript. Overall, the analytical misconceptions in this section do not help clarify the issue regarding effect of cowbirds on their hosts.

The second underlying weakness of this section, as well as in other parts of the volume, are misconceptions incorporated into design and interpretation of analysis that are actually removed from population models and population parameters. For example, Robinson et al. develop an analysis design combining potential host species that eject cowbird eggs laid in their nests with those that abandon a significant proportion of parasitized nests. They reason that both of those host groups cause a significant percent loss in cowbird eggs laid, resulting in a detrimental effect on cowbird populations. However, the parameter of interest in population models is young produced per female. Five successful eggs out of 10 produced per female (50% egg success) is as equally successful a strategy as producing 5 young from 20 eggs (25% egg success). Attention to population models would not have created that functionally misleading analysis design proposed by the authors. Their conclusion that habitats containing species that abandon parasitized nests can serve as sinks for cowbirds is numerically undemonstrated. They fail to consider the compensatory strategy cowbirds have of laying more eggs, and they also fail to explain how cowbirds became abundant enough in those habitats so that their numbers could subsequently decline.

In another example, DeGroot et al. failed to consider dispersal in their analysis design contrasting host communities between areas only 5–10 km apart that were either trapped or untrapped for cowbirds. Because untrapped sites are easily within expected dispersal distances of both hosts and cowbirds in trapped areas, treatments are not discrete and comparisons are not independent.

Although the second section struggles with problems mentioned above, it also assembles some useful approaches to the study of cowbird and host population dynamics. With relatively minor caveats, the fundamental model structure of Citta and Mills, and the basic algorithm of Greene, provide some highlights for taking a population perspective.

The third section contains a small collection of papers on efficacy of local cowbird-control programs. Eckrich et al. (recognizing a less than pure statistical design) provide an honest attempt to extract value from the large data set amassed at Fort Hood, Texas. Clotfelter et al. provide a credible site-scale perspective on the effects of burning on the use of Red-winged Blackbirds (Agelaius phoeniceus) as hosts by cowbirds.

The summary by Hall and Rothstein confronts the efficacy of local cowbird removal in increasing host populations. However, the authors operate under some of the same constraints listed above by interpreting statistically weak treatments tangential to the context of demographic and population models. For example, they assume that estimates of parasitism are accurate, and that parasitism was reduced to very low levels in several Willow Flycatcher (Empidonax traillii) studies (neither are necessarily true; see above). The low response in flycatcher numbers caused the authors to conclude that factors other than cowbird trapping were more important to flycatcher reproduction, rather than considering the possibility that the cowbird trapping was ineffectively implemented.

Hall and Rothstein digress somewhat from local cowbird-control removals to potential landscape alternatives. Sorely overlooked in that treatment was a critical appraisal of broad-scale control options. In fact, data and analyses of several studies elsewhere in this volume suggest potential cowbird effects on host communities and the value of broad-scale control options when using population perspectives (e.g. Citta and Mills, Greene, DeGroot et al., Peer and Sealy), but were not considered by Hall and Rothstein. In addition, the authors also include a treatment of a number of less science-based issues (e.g. rights of cowbirds, excessive money spent on trapping, profit motives of trappers) related to the development of scientifically sound management practices without recognizing that these can often complicate scientific evaluation.

Although this inexpensive volume contains some very good papers, there are hidden costs of treading a minefield of analytical and conceptual traps (thus wasting time and conservation dollars). This volume should be in all university libraries, and can be of use to researchers and decision makers for cowbird management, but with the caveat that little should be taken for granted. As researchers solidify analytical issues and take on more structured population-based perspectives, the understanding of cowbird–host processes and appropriate management considerations should improve substantially. This volume will still likely expand the general perspectives on this path.