This paper was presented to the Fifth British Empire Forestry Conference in London, 1947, and is reprinted here with the kind permission of the Director of Forestry. Average data accumulated chiefly from dissection of from twenty to thirty stems from each of five different sites are used in an attempt to show that variations in wood density in pine stems, hitherto regarded as fortuitous, are actually systematic. Analysis of the data shows that whereas aggregate wood density, or the mean density across entire transverse sections, is in normal circumstances in direct correlation with rate of growth, that of wood formed many particular year is directly proportionate to a function of age. The influence on density of aspect, of edaphic factors, with special reference to moisture, and of seed selection, is also briefly discussed; and a method of research based on the density/age relationship for use in further attempts at isolation of environmental determinants, suggested.