Orthopositronium (o-Ps), formed on the cone of a channel electron multiplier by 400- eV positrons, is shown to diffuse quickly into an Mgo-coated evacuated chamber with en- ergies up to 0.8+ 0.2 eV. The decay rate of o-Ps is measured to be 7.09+0.02 @sec where the error is due principally to systematic effects. This result is interpreted to be the first direct measurement of the vacuum decay rate of the o-Ps. It is in disagreement with the theoretical value of 7.242+ 0.008 p, sec ' and previous experiments in gases as well. In a previous Letter' Gidley, Marko, and Rich (GMR) reported the decay rate of orthopositron- ium (o-Ps) in powders of SiO, to be a linear func- tion of the free volume density. The result of ex- trapolating the decay rate to zero density was 7.104+0.006 psec ', This result is (1.9+0.1)g below the theoretical value' of 7. 242 + 0.008 psec ' and more than 2/q below the results of two experiments in gases, 7.262+ 0.015 p.sec and 7.275+0.015 p.sec '. GMR noted that their result could plausibly be interpreted as the vac- uum decay rate. Ford, Sander, and Witten, ' after a detailed consideration of the possible interac- tions of Ps with the surface of SiO found no rea- son to doubt that the extrapolated value is the free-space decay rate. We report here the re- sults of a completely different experiment in which 0-Ps is demonstrated to be decaying in vacuum. The decay rate is found to be in agree- ment with the reported value of GMR. Our method for measuring the vacuum decay rate is based on the observation' that 0-Ps is ef- ficiently formed at or near the surface of a solid by slow positrons (F. & 1 keV) and then diffuses into the surrounding vacuum. In order to meas-