Summary Laboratory experiments with transparent liquids in glass tanks were conducted to study the formation and movement of bubbles and the factors influencing effectiveness of agitation with air. Motion pictures were taken in two short horizontal cylindrical tanks, to determine flow patterns and allow measurements of bubble size and bubble and liquid movement. To represent blending as in standardizing or fortifying, observations were made of the time required for complete color change after addition of acid to indicator solution in the tank. To simulate incorporation of a cream layer, the minimum air-rate for dispersion of slightly buoyant plastic particles was determined. A cylindrical vessel, 11 in. in diameter and 24 in. long,placed either horizontally or vertically, was used in most tests. Important variables were tank shape, fraction of fill, air rate, air supply pipe arrangement, and the nature of the material being agitated. Air agitation was more effective in the vertical than in the horizontal tank. In both tanks, effectiveness decreased at low liquid depths. The influence of air rate was variable: In some cases an increase markedly increased effectiveness; in others, little or no improvement was obtained. For blending solutions of low viscosity in the horizontal tank, an air supply pipe with 12 holes uniformly spaced was less satisfactory. But for blending solutions of high viscosity and dispersing the plastic-particle suspension, this air-supply arrangement was most effective. In the vertical tank, a supply pipe with six holes uniformly spaced across the diameter gave the most satisfactory results under a variety of conditions. A single air-supply hole required the longest blending times when placed at the center of the vertical tank bottom, but favored rapid blending when placed one-third of the tank radius from the side.