The titanium Knight shift was measured at room temperature as a function of hydrogen concentration in TiH/sub x/ and as a function of temperature for TiH/sub 2/. In contrast to hexagonal Ti metal, the close-lying /sup 47/Ti and /sup 49/Ti resonances were resolved in TiH/sub 2/ even in the tetragonal phase. The shift in TiH/sub 2/ was found to be temperature dependent, varying from (0.252 +- 0.001)% at room temperature to (0.312 +- 0.001)% at 180 K. It was consistent with previously measured hydrogen-spin-lattice-relaxation times and with accepted band-structure models. The accuracy of the hydrogen-concentration dependence of the Knight shift was much lower due to vacancy-induced quadrupolar interactions and the resultant inability to resolve the /sup 47/Ti and /sup 49/Ti resonances. The Knight shift was only slightly concentration dependent over the entire range, having a value of about 0.25%, but jumping suddenly to 0.55% at x = 1.5. This was accompanied by a widely separated doubly peaked resonance line, We speculate on the cause of this jump and the resultant line shape.