The first six articles in this issue of JABES consist of papers that were presented at a symposium on the Study of Resource Selection Using GIS Data. The symposium was held in Snowmass, Colorado on Saturday, September 27, 1997, in conjunction with the 4th Annual Conference of The Wildlife Society. The symposium was organized by Grey Pendleton, Alaska Department of Fish and Game, and Lyman McDonald, West, Inc., Cheyenne, WY. It was sponsored by the Biometrics Working Group of The Wildlife Society. Len H. Carpenter was the 1997 Program Chairman for the Conference. Lyman McDonald, an Associate Editor of JABES, served as a Co-Editor of JABES with respect to these papers. Lyman received a lot of assistance from J. Richard Alldredge, Washington State University, and Bryan F.J. Manly, University of Otago, New Zealand, who are both Associate Editors of JABES. These folks took the responsibility for getting these papers reviewed and revised, and then the final versions were sent to me. The seventh paper was not presented at the conference, but it fits in very nicely with the previous papers. The last paper in this issue was not presented at the symposium, but was ready to be published in an issue of JABES.

Geographical Information Systems (GIS) are becoming widely available, making vast quantities of data available for the study of important biological issues. The first reaction of many biologists to a GIS is that sampling and statistical issues can finally be ignored. The reaction upon trying to test hypotheses with collected data is that the statistical and sampling issues have just been increased by an order of magnitude. Mixing data on some variables from a GIS with limited sample data on other variables is just one example of the kinds of problems that may be encountered. Hypotheses concerning resource selection (or resource use) based on GIS data are among the first to be considered by field biologists. This symposium provided a forum to (1) clarify the statistical issues for use of data from a GIS in resource selection studies, (2) explore the various approaches and sometimes differing viewpoints which researchers have envisioned for modeling resource selection, (3) explore the almost infinite possibilities for the study resource selection at different scales when using a GIS, and (4) provide specific examples of the kinds of models available, different scales of study, and interpretation of results for resource management.

The first article in this issue is authored by J. Richard Alldredge, Dana L. Thomas, and Lyman L. McDonald, and is devoted to a broad introduction to the problems and models which been proposed for study of resource selection. Analytical methods are compared with respect to underlying assumptions, and suggestions for future development are given. The second article, authored by David L. Otis, looks at the potential effect of spatial pattern of the collection of habitat patches on the habitat selection process and explores the adaptation of quantitative techniques and modeling concepts used by landscape ecologists to the study of wildlife habitat selection.

The third article by Marti L. McCracken, Bryan F.J. Manly, and Madeleine Vander Heyden presents an adaption of the discrete-choice studies that are commonly used by economists and sociologists to study the process of resource selection by animals. Applications are illustrated using a portion of a study on habitat selection by female American black bears in the Central Oregon Cascades. Next, Grey W. Pendleton, Kimberly Titus, Eugene DeGayner, Craig Flatten, and Richard Lowell review the use of compositional analysis and GIS for study of habitat selection by goshawks in southeast Alaska. This analytical approach addresses several issues commonly criticized in habitat use studies, including use of appropriate sample units and dependence of habitat proportions. The fifth article, by Wallace P. Erickson, Trent L. Mcdonald, and Robert Skinner, illustrates modeling techniques for resource selection studies using data from a study of winter habitat selection by moose on the Innoko National Wildlife Refuge in Alaska. Then, Steven T. Knick and John T. Rotenberry evaluate habitat selection by black-tailed jackrabbits in Idaho using GIS data and the Mahalanobis distance statistic in a region of shrubsteppe in southwestern Idaho.

The last article in this set was not presented at The Wildlife Society symposium; however, Nairanjana Dasgupta and J. Richard Alldredge's paper was submitted for publication to the JABES at the time this issue was being prepared. They develop a method, based on the maximum of the multivariate chi-square distribution, to model dependent behavior as evidenced by sightings of animals together. Methods are illustrated with data, from part of a study of the ecology of sharp-tailed grouse in eastern Washington.

I am confident that all who take the time to read these papers will discover valuable information that will help statisticians when they are faced with a need to analyze data that arises from the use of Global Information Systems.