18 FREILING, CROCKER, AND ADAMS not well understood. When the condensed and vaporized species are identical, as in the case of condensing mercury, it is usually near unity. When there is a change in species, as in the cases of P, condensing on P, and of AS,O, condensing on AS,O3, it can be extremely low. In these two cases!® it is less than 107", Apropos of the case of fallout, we have recently obtained evidence for low evaporation coefficients of alkali metals from fused oxides which indicates the likelihood of similarly low condensation coefficients. Whenthe condensation coefficient is incorporated into Fuchs’ equa- tion, one finds the condensation rate to be g Bi: 4 R+aA ahR?+R+) where S is the surface of the particle. If @ is very small (ahR? «R+X), surface resistance to mass transfer predominates and the condensation rate is _ It is instructive at this point to consider the case of a volume V of condensable molecules at concentration n; exposed to S square centimeters of condensing surface. If there is no reevaporation and first- order kinetics can be assumed, the condensation rate is Thi = adi F = Omi By from which few] n; =n; exp (— -— where n° is the initial concentration and 4vV. Vv (2zm)" w= oaS ~ aS kT is the mean molecular residence time in the vapor phase. If reevaporation can occur the rate equation of the previous para- graph must be modified. Again assuming first-order kinetics, this is most simply done in terms of the mean residence time on the surface t,. Letting (n,); denote the number of molecules per square centimeter of surface, we can write