A MEMORANDUM recently circulated by Prof. C. F. Marvin, Chief of the U.S. Weather Bureau, raises the question of an appropriate unit of pressure, especially for meteorological usage. The measure of pressure by a barometric height, in millimetres or inches of mercury, even when reduced to standard temperature, is not an absolute statement at all, for its meaning depends on the local value of gravity. On the other hand, theC.G.S. measure of one megadyne per square centimetre, besides being absolute, happens to express quite closely the mean atmospheric pressure at about 100 metres above sea-level. The advantage that could be taken of this fact has long been obvious; it is referred to in early editions of Everett's “C.G.S. Units,“and so long ago as 1888 the adoption of the unit of pressure as one dyne per square centimetre, under the name of a barad, was recommended by a committee of the British Association. But nothing very definite followed; and Prof. Marvin gives the history, which is not without its moral, of the way in which the natural appropriateness artd utility of this unit re-noticed, reported-upon, or brought into partial use upon inconsistent systems by Guillaume, Bjerknes, and various others, including international committees.