3583the tunneling step (the second step). The correlation of In kH with In K suggests that it is the third step which mainly shares ratelimiting character with the second, rather than the first. In the nonhydroxylic solvents translational and rotational solvent motions, which are responsive to bulk viscosity," are important components of the heavy atom reorganization steps. We suggest that the hydroxylic solvents disable the pretunneling configuration or stabilize the posttunneling configuration, by hydrogenic motions which are faster than molecular rotation and unrelated to bulk vis~osity.'~ In the theoretical section we have shown that the observed pattern of isotope effects could be produced by the competition of the tunneling step with either or both of the spontaneous heavy atom reorganization steps.Other things being equal, eq 2 anticipates that high-viscosity solvents, which give lower limit values for k H / k D , should also give lower values of kH' than Solvents which give upper limit values of k H / k D . In the lower limit cases an additional bottleneck has appeared. We recognize, of course, that the solvents might also influence k in more conventional, quasi-thermodynamic ways.4Some of those influences are also reflected in K. Those may be removed by using kH' instead of k H . In fact, all the lower limit solvents give significantly smaller values of kH' than do solvents which give larger values of k H / k D although the differences are not large. Since solvent viscosity, in general, is correlated with solvent density, the isotope effects and rates cited in Table I correlate with the latter to about the same degree as they do with the former. This raises the possibility that the kinetic quantities are responding to internal pressure, a quasi-thermodynamic solvent pr~perty,'~ rather than to viscosity, a dynamic property. At the moment we see no rigorous way to exclude this possibility. However, it seems unattractive because it seems likely that an increase in internal pressure would increase the rate of our reaction, since it almost certainly has a negative volume of a~tivation.'~ Similarly, the present results do not require that the second stage be a tunneling process. They strongly suggest only some sort of two-or threestage process, in which solvent characteristics determine the rate-limiting stage. Without this we see no way that two very different sets of isotope effects could be generated with such small ranges of K and kH. The suggested model accounts for the observations and seems to be in general accord with theory and previous work on this reaction. Further tests are in progress. (33) Bauer, D. R.; Alms, G. R.; Brauman, J. 1.; Pecora, R.Abstract: Oxazine 1 perchlorate (A) and oxazine 170 perchlorate (D) were selected to investigate the influence of cyclodextrins (CDs) on the dyes' aqueous solution aggregation equilibria. Through equilibrium processes, these days form inclusion complexes with j3-and y-CD. The amphiphile hexane sulfonate can be coincluded with D in y-CD. We report determinatio...