Three of the most common fire retardant formulations-Phos-Chek XA, Fire-Trol 100, and Fire-Trol 931 (liquid concentrate), in addition to water-were dropped from a TBM air tanker under various conditions to determine the effect that thickening agents, wind-speed and direction, drop height, and aircraft speed would have on ground distribution patterns. Seventy-four drops were made over 820 cups in a grid system. The cups were collected, weighed, the concentration in gallons per 100 square feet computed and summarized, and a computer plot of the ground distribution patterns printed. Drop height and windspeed were consistently the strong variables in models that were used for predicting ground distribution patterns of all retardants. Covariance analysis of the models indicated that the greatest real differences existed between gum-thickened Phos-Chek XA and the remaining retardants. The area of effective coverage, the length of effective coverage, and retardant recovery, all tended to decrease for Fire-Trol 100, Fire-Trol 931, and water, in that order. Predicted values of effective areas as a function of height, wind, and concentration are calculated from the mathematical model for each retardant. Predictions of recovery are given as a function of height and wind. The greater total recovery and more concentrated patterns for Phos-Chek XA are attributed to a greater cohesiveness when subjected to airstream shearing forces. The result is larger mean droplet sizes when terminal velocity is reached. This phenomenon results in shorter drop times and less evaporation losses for Phos-Chek XA than for other materials. Within the range tested (93 to 127 knots), the effect that aircraft drop speed had upon ground distribution patterns was small and quantitatively fell within uncontrolled variations of the data. The maximum effective drop speed for the TBM is probably near the maximum safe drop speed of 145 knots. Maximum effective drop heights depend on the particular fire situation. However, assuming a particular effective concentration, the optimum height for any wind can be determined from prediction tables developed from the mathematical model for each retardant. In general, under low wind conditions (<6 m.p.h.), the optimum drop height is between 150 feet and 300 feet. Many drops under these conditions are currently being made at drop heights below this range; thus, it appears that some advantage in effectiveness and safety can be attained by raising drop heights under such situations. This study provides the basic data from which trade-off studies between retardant salt content and effective areas can be performed. Optimum retardant salt contents for both thickened and unthickened retardants can then be established. The basic data can also be utilized in drop mechanization studies designed to improve either the retardant solution or the delivery systems.