We have used archival FUSE data to complete a survey of interstellar HD in 41 lines of sight with a wide range of extinctions. This follow up to an earlier survey was made to further assess the utility of HD as a cosmological probe; to analyze the HD formation process; and to see what trends with other interstellar properties were present in the data. We employed the curve-of-growth method, supported by line profile fitting, to derive accurate column densities of HD. We find that the N(HD)/2N(H2) ratio is substantially lower than the atomic D/H ratio and conclude that the molecular ratio has no bearing on cosmology, because local processes are responsible for the formation of HD. Based on correlations with E(B-V), H2, CO, and iron depletion, we find that HD is formed in the densest portion of the clouds; the slope of the logN(HD)/log(H2) correlation is greater than 1.0, caused by the destruction rate of HD declining more slowly than that of H2; and, as a sidelight, that the depletions are density dependent.