This paper is a critical review of the "drop piate" method of determining thenumber of viable bacteria in fluids, together with a description of an experimentalcomparison of the "drop plate" with the more usual "pour plate" procedure.It is shown that counts on pure cultures of bacteria made by the "drop plate"method are some 7 /6 higher than those made on the same cultures by the "pourplate" method. It is also shorvn that the standard error of a series of countsmade by the "drop plate" procedure is slightly less than for those made by the