Florida Agricultural Experiment Station This is a return of approximately 33% on the investment of the nitrogen in the fertilizer. A higher return per field crate (as in 1937-38, Table 1) would mean a greater difference be- tween the profits realized from the two treatments, and, con- versely, a lower return would tend to minimize the treatment differences from the economic standpoint. The other three comparisons shown in Fig. 4 are for one-year yields only and are, therefore, not so dependable as are the fig- ures for the five-year average. For the 1930-31 crop, mentioned briefly above, a total of only 2,000 pounds of fertilizer per acre was used, one-half before planting and one-half as a side- dressing. This was a good year for celery, as can be seen from the yields obtained. The same materials were used in the fer- tilizer as outlined in the discussion of the five-year yields, but with the amount of fertilizer used only one-half as large, the crop costs per acre were slightly lower, being $174.23 without nitrogen and $188.26 with nitrogen. Respective yields were 601 and 685 field-trimmed crates per acre, which give a growing cost per crate of 29.0c for the former and of 27.5c for the latter. It is obvious that the nitrogen added in the latter treatment proved an excellent investment, since' this treatment not only produced 84 more crates per acre but reduced production costs 11/c per crate. Assuming, as above, that the grower received a return of 50c a field-trimmed crate, the net profit per acre of celery for the year would have been $126.27 without nitrogen and $154.24 with nitrogen. The return ($27.97) from an in- vestment of $14.08 worth of nitrogen fertilizer in this case would be almost 200% for this year, with the celery selling at 50c per field-trimmed crate. The 1936-37 crop, Figure 4, has been already mentioned with respect to unfavorable conditions for production. Yields were very low with both treatments, the treatment containing nitro- gen yielding but five crates per acre more than the one without. This was the first crop grown in Area 3 and was fertilized with 2,000 pounds per acre before planting and another 2,000 pounds as side-dressing; formulas used were 0-6-12 and 3-6-12. The nitrogen was supplied in the latter as half sulfate of ammonia and half nitrate of soda, which lowered considerably the nitro- gen cost in comparison to that in Area 2 in the previous year already discussed, since the State Chemist's quotation on the nitrogen in nitrate of soda is 12.4c per pound (4). Other materials were the same as those used in Area 2. The field