Florida Agricultural Experiment Station floated in at least three days before the plants are set. How- ever, if the weather should be unseasonably wet and cold in the growing season, some further side-dressing with nitrogen might be beneficial. It is suggested that this side-dressing con- sist of nitrate of soda or nitrate of potash on normal, unburned soils, or of sulfate of ammonia on neutral or slightly burned soils, applied at approximately 200 pounds to the acre. It should not be worked in deeply, because of the mass of roots near the surface of the soil. Assuming that the celery field is adequately mole-drained and that water does not stand on it, this side- dressing should be applied as soon as the water from the un- seasonable rains has subsided and the field is workable. MISCELLANEOUS SUGGESTIONS Celery should not be planted on raw land. The first year that sawgrass soils are broken it is better to plant them to some crop of low fertility requirements. Sawgrass decomposes rather slowly, and during the process of decomposition crops requiring large amounts of fertilizer will not do well. Although seedbeds are commonly raised two to three inches to permit rapid surface irrigation, it is neither necessary nor desirable to set the celery in the fields upon raised beds in the Everglades area. Celery is commonly planted in rows 30 inches apart, and with a four-inch spacing between plants. Under such conditions, there will be approximately 52,275 plants to the acre. The relatively close spacing usually will prevent pro- duction of too many extra large stalks, will encourage erect stalks with good hearts, and will give a high per acre yield. The distance between rows is commonly modified somewhat for convenience in spraying, depending upon type of equip- ment used. With regard to celery varieties, only the so-called self- blanching types are grown. Selected strains and especially bred varieties of celery are being introduced almost yearly by the seed houses. Some of these "specials" have done very well, and almost all of them are improvements over the old Golden Self-Blanching. In all of this experimental work the only filler used with the fertilizer mixtures was sawgrass peat soil. Consequently the filler may be ignored completely in a consideration of the active ingredients of the fertilizer mixtures. In the event that the grower mixes his own fertilizer, or has it mixed to his order,