The level of interest in small-footprint LIDAR use in forest assessments is enormous, as indicated by the growing body of research published recently on the subject. Little attention has been focused on assessment of the detectability of tree and forest parameters under different mission and forest settings. This paper puts forth the premise that, for effective use in forestry, a better understanding of the underlying sampling theory must be known to develop the most efficient solutions to operational use of LIDAR. Examples of the issues are presented from LIDAR data sets being used in on-going research in the Department of Forestry at Mississippi State University.