fact that the water used in all of the pots contained a trace of lime.

'There is evidence that both phosphorus and potassium have an indirect value aside from their direct value as plant food for the alfalfa. This is the value of these elements to the bacteria. Bacteria themselves are living plants, and while they are microscopic in size they are almost infinite in number, and their multiplication and development are largely dependent upon the supply of available minéral elements of plant food. It will be observed that a yield of 2,080 pounds per acre (7) was secured without the addition of phosphorus. The yield became 3,520 pounds when phosphorus was applied (9). This increase of 1,440 pounds per acre may be due in part to the direct value of the phosphorus to the alfalfa and in part to its value in promoting the development of the bacteria and thus increasing the supply of nitrogen which the bacteria secure from the air and furnish to the growing alfalfa. Again, the addition of phosphorus to the combination, lime, potassium, bacteria, increased the yield from 1,120 (5) to 2,720 (8), an increase of 1,600 pounds per acre. The fact that the soil itself contained sufficient phosphorus to produce a yield of 2,080 pounds (7) tends to prove that the first 960 pounds of this 1,600 pounds' increase was due to the increased development of the bacteria resulting from the additional supply of available phosphorus; the remainder of the increase, 640 pounds, is probably due to both the direct and the indirect value of the phosphorus.

Without addition of potassium the maximum yield was 3,040 pounds (6); consequently the increase in yield from 1,600 pounds (4) to 2,720 pounds (8), resulting from the addition of potassium to the combination, lime, phosphorus, bacteria, was probably largely due to the increased development of the bacteria in the presence of a larger supply of available potassium.

The field experiments, which are described further on, give abundant evidence of the value of applications of lime in promoting the development of the alfalfa bacteria.

The photographic reproduction of the six pots receiving no artificial fertilizer, three of which were inoculated and three uninoculated, as they appeared on May 11th, less than three weeks after cutting off the crops discussed above, may be of interest (see Fig. 1).

Cuttings made from these six pots on May 21st just four weeks after the previous cuttings, gave the results shown in Table II.

TABLE II.-ALFALFA POT CULTURES; CUT MAY 21, 1902. WEIGHTS IN GRAMS PER POT AND POUNDS PER Acre.

It will be observed that the highest yield of the three uninoculated pots was 320 pounds per acre, while the average yield of the three inoculated pots was 1,600 pounds,-five times as great. These results only serve to confirm those secured from the preceding crops, and to show the value of the inoculation in a most conclusive manner.

When we remember that the twenty-four pots used in this series of experiments were all filled with the same kind of well-mixed soil, and that this soil is fairly representative of thousands of square miles of black prairie land in Illinois and adjoining states, that both series, of twelve pots each, were kept on the same table in the greenhouse, watered at the same times with exactly the same kind of water, and in every way treated exactly alike, except that one series was inoculated with alfalfa bacteria while the other series was not inoculated, then these most positive and conclusive results, as shown by the records of the experiment, including the color of the foliage, the height of the growing plants, the photographic reproductions, and the absolute yields per pot and rate of yield per acre, seem truly remarkable and appear to be of tremendous importance in solving the question, Why is alfalfa so commonly an unsuccessful crop on ordinary prairie soils?

FIELD EXPERIMENTS WITH ALFALFA.

An acre of ordinary slightly rolling black prairie land, capable of yielding 70 bushels of corn to the acre, was seeded with alfalfa in June, 1901. The soil was considerably better than ordinary cultivated soil (such as was used in the pot culture experiments).

Previous to 1895, it had been in pasture for at least eighteen years, and since 1897 it had been in meadow; thus, only three grain crops (corn in 1895, 1896, and 1897) were grown on this soil during the past twenty-five years. The field, which was 8 rods wide east and west, and 20 rods long north and south, exclusive of some border and division strips, was divided into two parts by a line running north and south, and into five parts by lines running east and west. The west part of the acre was inoculated with soil taken from an old alfalfa field in Kansas. The five divisions from north to south were fertilized as follows:

Plot No. 1-Check (nothing applied).

Plot No. 2-Lime.

Plot No. 3-Lime and phosphorus.

Plot No. 4-Lime and potassium.

Plot No. 5-Lime, phosphorus, and potassium.

The rates applied per acre were: 320 pounds of air-slaked lime, 320 pounds of bone-meal (containing 30 per cent. phosphoric oxide), and 160 pounds of crude potassium sulphate.

The infected alfalfa soil was applied at different rates of seeding on narrow strips running north and south on the west half of the field, the lightest application (320 pounds) being on the west side, and, on successive strips eastward, the rates of application were 640, 960, 1,280, 1,600, and 1,920 pounds, respectively. Each of these strips was about one-half rod wide. A good stand of young plants was secured, but a very heavy rain storm, which occurred on July 2nd, washed the soil somewhat, and, as the west side of the field was somewhat higher than the east side, it was reared that the bacteria might be carried over the east plots and thus inoculate the whole field to some extent, which afterward proved to be the case, particularly along the east side where the water stood for a short time. The southeast quarter of the field was the lowest part, and, although it was tile-drained, the water stood on it long enough to kill most of the alfalfa plants. Because of these occurrences, the results of the experiment are probably not so marked as they would otherwise have been.

During the summer of 1901, the alfalfa was clipped three times, the clippings being left lying on the field. During midsummer, the weeds seemed to grow faster than the alfalfa, but with each