an appendix to ReL (4). Recently, the present authors have given a theory of dislocation drag in the normal state based on the Boltzmann-equation formalism [Ref. (lO)]. The results indicated that the temperature-independent scattering result of Holstein is probably correct for close-packed metals. However, for bcc metals the electronic drag was found to be proportional to the electrical conductivity. As it has not yet been possible to derive this latter result from a scattering theory, we will take the dislocation drag in the normal state as that given by Eq. (17) in this paper. At any rate, one might hope that the ratio of the drag in the superconducting state to that in the normal state will not be greatly changed, whatever the form of the normal state drag coefficient. We hope to treat this question in more detail in a future paper. 20 The results for a t and a 2i which are simply the attenuation ratios of a high-frequency phonon, are quite interesting in themselves and will be given in a separate paper. The temperature dependence of the energy gap in all these calculations has been taken as that given by B.Magnetic measurements of superconducting transitions of very pure isotopically separated bulk samples of Ga and Zn reveal a large "paraconductive" effect. This is contrary to the prediction for such systems by Aslamazov and Larkin based on corrections to first order in the fluctuations in the system. It is suggested that this discrepancy may be explained by increased electron pair lifetimes due to the absence of isotopic and other scattering of phonons allowing higher order processes to be significant.According to the theory of Aslamazov and Larkin (AL), 1 the electrical conductivity of a superconducting material in the normal state increases as the temperature approaches T c because of fluctuations in the system (paraconductivity). This effect can be large for T-T c ~ 10~3 K in thin films and whiskers, but for pure bulk type-I superconductors, appreciable paraconductivity should be confined to the presently unobservable temperature range T-T c ~ 10~1 5 K. The theory predicts both a magnitude and a temperature dependence for the effect. Measurements on a variety of thin films 2 have shown rather good agreement with the theory for most substances measured. There were, however, substantial deviations from the theoretical magnitudes for measurements on low-resistivity Al films. 2,3 This Letter presents results of measurements on pure bulk type-I superconductors which disagree with the present prediction of the theory, and proposes a mechanism by which these re-sults, those on similar systems, and the Al film results may be understood.In magnetic measurements of the isotope effect in superconducting Zn 4 and Ga, 5 it was observed that some of the samples exhibited broad superconducting to normal (S-N) transitions in small magnetic fields. We have found that this paraconductivity is only observed in metallurgically and isotopically pure samples and in small magnetic fields (^1 Oe). We observed measurable paraco...