diet an ordering temperature of 28 ± 5 K. An equivalent calculation for a more compressed lattice with the same nearest-neighbor spacing as the dense-solid phase but with molecular axes allowed to tilt out of the film plane gave approximately the same ordering temperature, 9 the closer spacing being compensated by the formation of a rickrack configuration of the molecular axes. That a model which treats the surface film as an idealized, two-dimensional structure gives results in good agreement with the experimental observations is testimony to the essentially twodimensional character of the transition.Finally, we have to address the question of what are the ordered-state molecular configurations of the registered and dense-solid films. This would be a difficult problem even if we were dealing with a three-dimensional solid. If that were the case, we would expect to be able to deduce the ordered configuration from the intensity profile of the diffraction pattern, 10 assuming sufficient diffraction data could be collected. But there is too much background scattering from the substrate (particularly at larger scattering vectors) to make such an approach feasible with overlayer films. The only realistic alternative is to attempt to compare the observed intensities with those calculated assuming ordered configurations suggested by theory. We find, unfortunately, that either the two-or four-sublattice configuration can satisfactorily reproduce the observed diffraction profile of the registered-solid phase; the data available are not sufficient to allow us to choose between the two suggested alternatives. One other possibility remains, namely, to search for the extra super lattice reflections expected if the ordered state is the four-sublattice rather than two-sublattice configuration. As it happens, this too appears to be impractical; a calculation of the relevant structure factors makes it clear that none of the extra reflections could be detected with neutrons. Thus we are forced to conclude that, while we can study the onset and development of orientational order in the nitrogen overlayer, we cannot, at least at this time, determine the ordered configuration.We wish to express our appreciation to W. Press for a number of helpful suggestions and to M. L. Klein for making available to us the results of unpublished computations.The oscillatory flow of superfluid s He through an 18-/rni-diam orifice has been studied. A clearly defined critical velocity is seen in both the A and B phases. For velocities greater than the critical value, the superflow involves an excess dissipation. The temperature dependence of the critical velocity is reported for temperatures near the superfluid transition (T/T c >0.9).The possibility of dissipationless flow is one of the mechanisms of critical velocities has proven the most dramatic aspects of the superfluid state. a valuable test of our understanding of the nature The study of the limits of stable superflow and of the superfluid state for both superconductors 1332