Fission-product bromine was separated rapidiy from fission-product iodine and the neutron activity of either species was counted. This was done by irradiating with thermal neutrons a solution of U/sup 235/ containing the reagents for transforming the unwanted halogen into nonvolatile form. Air containing some of the desired halogen as carrier is swept rapidiy through the solution and carries the halogen of interest to a tube which holds a number of glass bends wet with carbon tetrachloride. The halogen dissolves in the surface film while the air and rare gases are carried on out. The neutrons emitied from the tube are counted by a neutron counter after moderation in paraffin. The irradiation time is varied from a fraction of a second to several minutes. The sweeping process is generally started aboul 0.2 seconds after the end of irradiation and lasts for ~ 1.2 seconds, during which time about 0.5 liter of gas was transported through 4 ml of solution. All operations are carried out automatically by use of a timer and electrically operated valves. The neutron counts are scaled and recorded on tape with a pen recorder. The initial counting rates are normally in excess of 1000 per second and formore » some irradiations are as high as 15,000 per second. The decay of the neutron activities is followed to background, which is typically about 1.5 per second. The decay curves are analyzed into periods by conventional graphical techniques. For bromine, neutron activities are found having the following half lives: numbers is most likely to bu 54-second Br/sup 87/, 15.5second Br/sup 88/, 4.0-second Br/sup 89/ and 1.5- second Br/sup 90/. For iodine, half-lives of 24.4, 6, and 2.5 seconds were found. The mass assignment appears to be 24.4-second I/sup 137/, 6second I/sup 138/ and 2.5-second I/sup 139/. Decay data from unseparated samples are customarily analyzed into six components of which two are appreciably shorter in half-life than those found. Since seven are already separated, it appears that at least nine periods exist. None of the new activities reported may be considered to have relatively small yield, all being more abundant than the 54-second Br/sup 87/ . It is apparent thut measurements of half-lives, yields, and neutron energies which were done with unseparated samples should be reexamined in the light of the new periods. (auth)« less