For seawater Samples from various environments, viable bacterial counts obtained by various plate count techniques were compared with each other and with direct viable counts and total counts. Of the three plate count methods, i.e., spread plate, filter and pour plate methods, the filter method gave the lowest number of colony forming units. Average nercentages of number of colony forming units obtained by the filter method to the number obtained by the spread plate method ranged from 5.2% for samples from Tokyo Bay to 26% for samples from open seawater of the Pacific Ocean and the South China Sea. The pour plate method using agar media of low gelling temperature yielded about 90% of the number of colony forming units obtained by the spread plate method. The colony forming units in the pour plate method decreased in number about 20% when the temperature of molten agar media was raised from 37°C to 40°C. The pour plate method with agar media of low geling temperature has an advantage over the spread plate method when water samples from open seas. in which low numbers of colony forming units can be expected, are to be examined.