We have measured the solid state H-1 and H-2 Zeeman relaxation rates R at room temperature in two methyl-deuterated samples of 1,9-dimethylphenanthrene. The H-1 dipolar rate Rd (d for 1H dipolar and 9 for 9-methyl group) and the H-2 quadrupolar rate R-q(1) (q for H-2 quadrupolar and 1 for 1-methyl group) were measured in 1-(trideuteriomethyl)-9-methylphenanthrene. The H-1 dipolar rate R-d(1) and the H-2 quadrupolar rate R-q(9) were measured in 1-methvl-9-(trideuteriomethyl)phenanthrene. Models are developed for both R-d(m) and R-q(m) (m = 1 or 9) due to methyl group rotation. In a large class of simple dynamical models for spin relaxation in the solid state, the ratios R-q(m)/R-d(m) are independent of the dynamics (except for the mass difference between the H-1 and H-2 nuclei). In the present case. these ratios are also independent of in. In addition, the ratios R-k(9)/R-k(1) (k = 1d and q) are dependent only on the activation energy for methyl group rotation and another related dynamical parameter, but they are independent of the interaction being modulated. Because many parameters factor out, these ratios can be predicted with few or no adjustable parameters, depending on the sophistication of the theoretical model. The agreement between theory and experiment is good, even for the simplest theoretical models. These agreements give one a greater confidence in the models for quadrupolar and dipolar relaxation, but particularly the latter, which, because of spin diffusion, are more difficult to test.