-12- (3) The sampling rate must be such that the balloon flight time at altitude is less than about 2 hours since the difficulty of recovery increases with increasing distance of the impact point from the release point. This time limitation implies volume flow rates of the order of 500 cubic feet per minute assuming the collection efficiency is 25% or better. (4) The weight must be kept as low as possible, preferably less than 100 lbs to keep the balloon size within reasonable limits. (5) The collection efficiency must be definitely known. It should also be high so as to reduce volume flow requirements to a minimum, and should be independent of particle size, since the particle size distribution is unknown The Ash Can system has suitable weight and volume flow rate and is relati: inexpensive to manufacture and simple to operate, but it has a low filtration efficiency @reliminary calibration data indicate about 20 to 25% at the lowes: sampling altitude). Furthermore, exhaustive laboratory tests have shown tha! the collection efficiency varies strongly with particle size and air density. This system has the further disadvantage for the radiochemist that the tenth of a microgram or less of stratospheric dust which is collected, containing less than a billion atoms of Sr-90, is spread over 5 square feet of thick, loose-textured filter paper. One may feel some anxiety lest some of them be evaporated or otherwise lost under the rough treatment, such as ashing or dissolving in quantities of strong acid, which is necessary in order to prepa the bulky filter for chemical analysis. Some desirable methods of study such as autoradiography or electron micrography are rendered virtually impossible by the bulk and dilution of the sample. The AEC is sponsoring development work on several more advanced concepts: DOE ARCHIVES an improved filter system, an improved electrostatic precipitator and a rotating impactor. The advanced filter system, designed for higher eee