If 7 1/2 gallons are to be applied, the applicator valve must be adjusted to deliver 1 pint in the time it takes the rig to move 181 feet. If the adjustment is not marked on the valve, four steps will be necessary to determine the correct opening: Mark off 181 feet. Count the number of seconds required for the rig to move 181 feet at operating speed. With the tank half full, hold a pint measure under the delivery tube, and by making several trial openings with the valve, determine the exact position required to deliver i pint in the measured amount of time. Mark that position on the valve handle so that the correct opening can be adjusted at will.

If convenient, a second valve may be installed just above to act as a cutoff. The machine should always be kept free of trash and grit. The liquids should be strained into the tank.

Making the necessary adjustments, preparing the soil, and selecting the appropriate time for treatments are the most important preparations. The rest is only a matter of carrying out a few simple procedures.

The first step in row fumigation is to mark off the 4-foot rows with a shovel plow, making a shallow furrow to remove all trash that might clog the applicator foot.

The second step: With the first row as a trial run, and moving at regular speed, hold the plow foot in the center of the row mark so that the liquid is buried 10 inches beneath the normal soil surface. Practice will determine when to cut the valve off and on when turning. It is a good plan to open the valve wide occasionally and close it immediately to clean out grit that might change the rate of flow. Keep the tank filled as near as practical to the same level. See that several inches of soil fall back into the applicator furrow immediately covering the liquid. Check the flow from time to time.

The third step is to follow a few minutes later with one turn-plow furrow on each side of the row to make a large bed. This bed should be considerably larger than usual to insure treating a wide strip. A disk lister may be used if it is set to make a large bed.

If the soil is subject to washing, it may be advisable to run out the middles with a sweep, so that drainage water will be confined to the middles and not wash over the beds. If the soil is cloddy and does not settle well after treatment, the beds may be rolled immediately afterward with a log or similar equipment. If the soil is in proper condition, rolling is neither necessary nor beneficial. Beds that are rolled must be opened again and reworked before planting. Under no condition should the beds be opened until 1 week after fumigation, however.

Our directions imply that the fumigant is to be applied in a single stream. There has been some question as to whether more effective row fumigation could be obtained by applying the fumigant in two streams 8 to 12 inches apart. The single and double-stream methods have been compared. While the double stream was slightly more effective, the differences were too slight to influence yield.

BOTH FERTILIZER and fumigant may be applied at the same time and in one operation with one- and two-row tractor rigs if the tobacco is to be set within 3 weeks. With the one-horse rig, fertilizer may be put in immediately after marking off the rows or in advance of the fumigator.

When fertilizer is applied 2 and 3 weeks before planting, however, it usually is advisable to put in only 60 to 75 percent in the row, and then to side-dress with the remainder 20 days after planting. If row fumigation is done 4 to 6 weeks before planting, it is best not to apply the fertilizer along with the fumigant. In that event, any time after 1 week following treatment, the beds may be carefully opened with a small shovel plow and the fertilizer applied and mixed in the usual way. Rows are then re-bedded; the second bed is made smaller than the first to lessen the danger of bringing in untreated soil from the middles. Opening the bedded rows in that way has not reduced effectiveness of treatment. In all operations between fumigation and planting, however, it is important to keep the row exactly in its original location.

Broadcast fumigation involves treating the entire field by applying parallel streams of fumigant about 12 inches apart. It usually is done best with a tractor rig equipped with a pump to force the fumigant into the soil at controlled rates. The rig may be equipped with 6 to 10 shanks, each with a delivery tube attached to the rear. It is necessary to have the field leveled or smoothed in order to prevent too shallow applications in low spots. If parallel streams are applied 12 inches apart, each shank must deliver 1 pint to each 272 feet of furrow for 20 gallons an acre, and three-fourths of a pint for 15 gallons. With broadcast treatments it is necessary to follow the applicator with a drag to level the soil and reduce evaporation loss.

Repeated tests in Georgia have shown that 7 1/2 gallons of 40 percent ethylene dibromide or 10 gallons of D D mixture in the row per acre gave more efficient disease control than twice those amounts broadcast. Broadcast treatments reduced disease from a 90 level of infestation in untreated soil to 42; row applications reduced the disease index to 27. The difference often was not enough to influence yields significantly, however.

In South Carolina the situation was reversed. Broadcast treatments repeatedly proved better than row applications. Disease was better controlled and yields were higher. The fact that extensive experimentation in Georgia showed that row treatments were more economical and effective, while similar work in South Carolina gave results that consistently favored broadcast treatments, indicates that generalization is not possible for all areas with respect to method of application.

The most effective broadcast rate of 40 percent ethylene dibromide was 15 gallons an acre, although 20 gallons sometimes slightly increased yields. The minimum effective broadcast rate of D D mixture was 20 gallons an acre. More than 20 gallons of either material did not increase disease control sufficiently to justify the additional cost. The higher rates sometimes damaged tobacco, particularly if they were applied a short time before planting.

Repeated tests with 7 1/2 gallons an acre of 40 percent ethylene dibromide and 10 gallons of D D mixture gave better disease control than 5 and 7 1/2 gallons, respectively, but the more efficient control was not reflected consistently in higher yields. Sometimes the 7 1/2- and 10-gallon rates stunted the tobacco, particularly if the soil remained wet after treating or the beds were rolled. Ten and fifteen gallons in the row caused still more damage and lowered quality as well as yields. The minimum amounts of 5 and 7 1/2 gallons an acre gave the maximum benefit per unit of fumigant and were least likely to cause damage.

An interval of 3 weeks or longer between fumigation and time of planting is desirable for both row and broadcast applications. Under ideal conditions and with minimum applications of ethylene dibromide, the period may be reduced to 2 weeks. A longer interval is required with D D mixture to be on the safe side; also,the higher the rates, the longer the waiting period. With row treatments the waiting period may be increased to 6 weeks without any great loss in efficiency. With broadcast treatments it is practical to apply the materials in the fall or, in areas of mild climate, in winter. Fall and winter applications usually have been safer than those made in the spring and as effective. But treatments in early fall or September in Georgia have failed to control root knot the following season. Late fall applications are better there.

Soil type has been a definite factor in determining the length of time between treating and planting. Light, sandy soils, particularly lowlands, require at least a week longer than the heavier soils. Wet, cold soils need as much as 6 weeks, particularly if D D mixture is used.

Growers are cautioned not to be misled by good disease control that may be obtained from heavy row applications made shortly before planting. Repeated tests have shown that even when plants are not visibly injured, such a practice often has resulted in depressed yield and lower quality.

A fumigation depth of 10 inches has been more effective in the soils tested than 6 inches or less. Root knot nematodes occur abundantly in upland Georgia soils between the 4- and 16-inch depths. Maximum concentrations are at 12 inches or so the zone where fumigation must be most effective. Soils that remain waterlogged several months of the year need more shallow treatment because nematodes may be nearer the surface.

The risk in shallow treatments of 4- to 6-inch depths is that much of the fumigant may be lost too soon by evaporation. The danger in deep applications of 10 to 12 inches in wet soils is that the fumigants may be retained in the soil too long and cause damage to crops. Comparatively deep applications are essential for row treatments because they permit a wider strip of soil to be sterilized.

One good treatment will remain effective long enough to protect one crop. More than that cannot be predicted. Some treatments put in during January in Georgia controlled root knot during early summer but offered no protection after the first of August. Sometimes second-year benefits are apparent but may not be important. Minimum annual treatments therefore are preferred in heavily infested areas. Residual protection cannot be guaranteed even by heavy treatments. Thus, 30 gallons an acre in South Carolina tests did not always give measurable protection the second season. Ethylene dibromide showed more second-year benefits than D D mixture.

To understand the immediate problem of quality arising from fumigation, one has to consider the materials used and their composition. One gallon of 40 percent ethylene dibromide mixture contains about 3 pounds of bromine. Thus, with 7 1/2 gallons an acre of the mixture in the row, the bromine application is 22.5 pounds. One gallon of crude D D mixture contains approximately 6 pounds of chlorine; 10 gallons an acre in the row means an application of 60 pounds of chlorine. With broadcast applications, and twice the amount of fumigant applied, the respective bromine and chlorine figures would be 45 and 120 pounds an acre. All the amounts may be dangerous from the point of view of quality.

Normally tobacco does not contain more than 0.01 percent of bromine. Both row and broadcast ethylene di-bromide treatments have increased the bromine content from 0.04 to 0.31 percent. The average was less than 0.20 percent.