The most extensive timber type on the Choctawhatchee National Forest, 30 miles east of Pensacola, Florida, is longleaf pine, Pinus palustris. The stands consist of a sparse growth of 8 to IO mature and overmature longleaf pines to the acre with an understory of turkey oak, Quercus catesbaei, and bluejack oak, Q. cinerce. Understocking of pine, while resulting in part from past logging practice and frequent fires, may also be attributed to natural causes, among which insect attacks, drought, and poor site have played no small role. In addition to causing the gradual depletion of the veteran trees, these adverse conditions have made natural replacement inadequate, so that the present forest is on the decline (fig. i). As the Choctawhatchee region is typical of approximately five million acres in the state of Florida, and as timber growing appears to be one of the most logical means of utilizing this large acreage of sandy land, the importance of a study of reforesting these lands with desirable tree growth has been recognized. The most widespread soil of the region is classed as Norfolk sand, 'deep phase. This sterile sand, in places 6o feet deep, is 96 to 98 per cent silica. Because of the small amount of fine material present, the normal moisture holding capacity is approximately only 5 per cent. Drainage is excessive over most of the area. The average annual rainfall is 65 inches, and is fairly evenly distributed throughout the year with the exception of late spring or early summer when dry periods of 3 to 6 weeks duration may occur. That fires have been common in this region is attested by the numerous turkey oak sprouts. A sparse ground cover exists, consisting of wire grass, Aristida sp., other herbaceous plants including a variety of legumes, and the gopher apple, Chry'soblalwns obloiigifolius. Pine reproduction is rarely present in sufficient density to insure normal stocking. On abandoned fields, however, reproduction is much more abundant, indicating that climatic and edaphic factors are probably not the only ones unfavorable to the regeneration of longleaf pine. One phase of the work on the natural reproduction of longleaf pine in this region is a study of the roots of trees varying in size from seedlings to mature trees, the older trees being included as a means of judging whether root competition from these individuals is a factor in seedling survival. Although this