As a result particularly of experiments carried out by Miss Andrewes, Mr. Ramachandra Rao and myself at King’s College and by Dr. Rudberg in Stockholm, we now have much more information about these phenomena than we had when I discussed this question two years ago. The experiments of Whiddington as well as of Rudberg have removed much of the vagueness which was then necessary as to the nature of the first process which occurs when an electron strikes a solid. It is evidently either reflected without loss of energy, as in Davisson’s experiments, or it undergoes an inelastic collision of such a kind that it loses a substantial amount of energy. This is shown by the well-marked minimum in the energy distribution curve for the reflected and secondary electrons between the peak corresponding to the Davisson effect and lower energy values. In fact in important cases the density at this minimum is practically zero. This shows that in such cases there is no appreciable number of secondary electrons which have come back as a result of encounters with free electrons; since this process would give rise to a smooth and continuous distribution of velocities which should have an appreciable density at the place where the minimum occurs. The velocity distribution curves for the secondary electrons show, in addition to the two already mentioned, a third group having a low average energy. This group is closely connected with the group which has undergone inelastic collisions, since both groups are almost absent when the energy of the primary electrons is low. The electrons which form this group are to be regarded as the true secondary electrons originating in some way as a result of the excitation accompanying the inelastic collisions. This is very definitely supported by the fact that the energy loss in the inelastic collisions corresponds to definite level differences which are independent of the energy of the exciting electrons and at the same time the energy distribution in this third group is very insensitive to changes in the primary voltage. It is, in fact, very similar to the energy distribution among the photoelectrons of low energy which are generated when the soft X-rays which accompany these electron emissions fall on a metal plate. This similarity suggests that these photoelectrons and the third group of secondary electrons originate in a common mechanism or at least in mechanisms of very similar type. This is indeed required by the fact that the total secondary electron emission, i. e ., the sum of the three groups, and the X-ray emission both show a series of discontinuous increases at the same primary voltage. These discontinuities are very numerous and it is now abundantly clear that the great majority of them have nothing directly to do with the Bohr levels which govern the structure of the atoms of the bodies concerned.