cent of nitrogen, but it would be unfair to claim superiority for the former brand simply because it would cost more in the market. The difference in the cost of the two lies in the fact of the larger proportion of nitrogen in the former, and it would not follow that it would be wise to purchase the former simply because of its higher valuation. This is equivalent to saying that high valuations are not necessarily an indication of high agricultural value for use under given conditions. The agricultural value of a particular fertilizer to a given consumer is determined by the profits from its use, that is, its efficiency in promoting larger crop production on his land. It is entirely possible that a ton of acid phosphate costing $14 would return larger profits under some conditions than a ton of dried blood costing $30. The agricultural value, then, is determined by composition and not by cost, and the disadvantage, to some extent, of the system of valuations followed by certain experiment stations was that consumers confounded cost with value in their own practice.

The legal control which the various states have assumed over the fertilizer trade has become most efficient. Probably no industry in the country is more amenable to legal control than is the manufacture and sale of fertilizers. The benefits of such control are evident and are realized by both manufacturer and consumer. The manufacturer is defended against competition with goods that are sold on a fraudulent basis, and the consumer is guaranteed the purchase of materials fairly comparable with the composition that is claimed. Many instances could be cited of the final exclusion from the market of fraudulent mixtures which otherwise would have maintained more or less hold on the market for a long time, as for instance Poplein's Silicated Fertilizer, which was extensively exploited in the seventies, and later other mixtures of an equally inferior character. A very large proportion of the brands of fertilizers now in the market may be regarded as reliable and of fairly uniform composition. Moreover, the literature issued by the inspecting departments or officials is widely distributed among purchasers, who now have the opportunity of obtaining definite knowledge of what they are buying, which would not be possible without the existing inspection.

The purchase of commercial fertilizers.

The opportunities that are presented for the purchase of commercial fertilizers in the United States involve peculiar, and, to some extent, unsatisfactory conditions. In the first place, the number of brands offered in the market is very large, the brand names numbering among the hundreds. If these brands represented characteristic differences in composition, the situation would be more rational. They exist, however, because of what is regarded as necessary commercial expediency, the large number of manufacturers believing it to be essential to their business that they use trade names of a proprietary character. The trade names must be regarded in many cases as altogether unfortunate. They often imply facts which

do not exist and more often are of such a character as to relegate the fertilizer trade to the atmosphere of the patent medicine business. It is probable that twenty-five different mixtures at the outside would meet all the demands of the trade or of the farmers, and thus simplify very greatly what is to the purchaser a confusing situation.

Another condition that is unfortunate for the consumer, but for which he is largely responsible, is the presence in the market of a large number of low-grade mixtures, that is, low grade in the sense of carrying small percentages of valuable ingredients. An analytical review of the fertilizer trade in the state of New York for the year 1902 revealed the fact that 171 brands, or 25 per cent of all those in the market, were of a lowgrade character, carrying on the average only 1.2 per cent of nitrogen, 8.2 per cent of phosphoric acid and 2.6 per cent of potash. As suggested, these brands exist because of a demand on the part of the farmers for low-priced goods. To purchase such goods is mistaken economy.

A comparative study of the composition and selling prices of low grade and high grade goods revealed the fact that in the former the valuable ingredients had a pound cost over one-third higher than was the case with the latter. It is very evident that manufacturers cannot afford to mix a given quantity of valuable ingredients in an unnecessarily large tonnage and distribute this diluted, or low grade material, to consumers without increasing the cost over that which would attend more highly concentrated materials. To illustrate, the valuable ingredients of a ton of fertilizer carrying 1.2 per cent of nitrogen, 8.2 per cent of phosphoric acid and 2.6 per cent of potash, could be procured in 1,200 pounds of 14 per cent acid phosphate, 163 pounds of nitrate of soda and 140 pounds of muriate of potash, in all 1,500 pounds, showing that in sending out a low grade mixture freight is paid on 500 pounds of unnecessary material. If the nitrogen is furnished in 10 per cent dried blood, the increase in weight would be less than 100 pounds.

It is said by many farmers that they cannot afford to pay the price of high grade materials. In making this assertion it is forgotten that a smaller quantity of a high grade mixture will do the work of a larger quantity of the low grade. The expenditure should be based on the pounds of valuable ingredients obtained and not on the total weight of the mixture.

The expensiveness of commercial fertilizers to consumers is also increased by the fact that the trade is conducted largely on the credit system, and by means of agents who are abundantly distributed through the rural districts and who handle only small quantities of goods. A much better system of distributing plant-food to consumers would be the sale by the established merchants of cities and villages of standard materials to be sold under proper descriptive names, the prices to be based strictly on composition. This would place the purchase of commercial plant-food on a rational and much more economical basis.

For many years, the leaders in agriculture have advocated the home-mixing of fertilizers, that is, the purchase of chemicals and raw materials and the mixing of the same on the farm in such proportions as is believed by the farmer to best suit his needs. This system has not made rapid headway despite its many evident advantages, for reasons which are not altogether obvious. Home-mixing is entirely possible and practicable, because the chemicals and raw materials in the market are in efficient forms adapted to immediate use without any further manipulation. It is not true that the ingredients of a mixture of acid phosphate, potash salts and ammoniates have any greater efficiency when compounded by a fertilizer manufacturer than when compounded by a farmer. It has been shown, moreover, that the mechanical condition of the manufacturers' output is no better than that of the farmer's mixture.

There are certain advantages which pertain to home-mixing, especially when practiced by those who are using large quantities of commercial fertilizers. In the first place, market conditions have so far been such, even in the retail trade, that chemicals and raw materials have offered valuable fertilizing ingredients at less cost than the manufacturers' mixtures. This difference has been large. In the second place, by the purchase of chemicals and raw materials it is easily possible for the farmer to adapt the materials used to his needs, however unlike these needs may be for his different farm operations. In the third place, it is possible to so purchase chemicals and raw materials as to have entire assurance that they are of high quality. This is not true to the same extent of the manufacturers' mixed goods.

It is urged against home-mixing that farmers do not know what they should purchase, but it is clear at least that intelligent farmers are better qualified to judge of their needs than is the distant manufacturer who is only rarely well informed in farm practice, or the special needs of particular localities.

It is also asserted that much trouble attends the practice of home-mixing of chemicals and raw materials. It is certainly doubtful whether consumers who use only a few hundred pounds of commercial fertilizer can afford to take the trouble of purchasing the ingredients and mixing them themselves. The large farmer, using generous quantities of commercial fertilizer, can certainly afford to do so; and in any case farmers may wisely combine and purchase the necessary ingredients by the car-load. The mixing itself is a very simple matter, involving only a smooth barn floor and a shovel. It should be suggested in this connection that in mixing what is known as a complete fertilizer, that is, one containing compounds of nitrogen, phosphoric acid and potash, part of the nitrogen at least should be introduced in the form of organic material, or else the mixture should be put together only a brief time before it is used. If these precautions are not observed, the mixture is likely to "cake" and be undesirable for sowing either by hand or through a drill.

Literature.

It is impossible to give in this place a comprehensive bibliography of the literature on fertilizers. This literature is extremely varied and voluminous. Most of the experiment stations have issued bulletins of fertilizer analyses and tests. The experiment stations in Rhode Island, Massachusetts, New York, New Jersey, Vermont, Ohio and Kentucky have given special attention to the subject. The United States Department of Agriculture has also issued a number of valuable bulletins on fertilizers. Only a few references can be given: Fertilizers, by E. B. Voorhees; The Fertility of the Land, by I. P. Roberts; chapters in Storer's Agriculture; Fertilizers and Feeding Stuffs, by Bernard Dyer, London, 1903; Bone Products and Manures, by Thomas Lambert, London, 1901; Artificial Manures, by Alfred Sibson, London, 1901; Fertilizer Laws of the Various States, by T. Breyer and H. Schweitzer, Chemists' Pocketbook, 1893; Potash in Agriculture, results obtained in the United States, B. von Herff, Baltimore, 1893; Florida, South Carolina, and Canadian Phosphates, by C. C. H. Millar, London, 1892; The Phosphates of America, by Francis Wyatt, New York, 1892; Chemicals and Clover, or Farming with Concentrated Dung, by H. W. Collingwood, New York, 1892; Tabulated Analyses of Commercial Fertilizers from Samples Selected in Accordance with act of June 28, 1879, Pennsylvania State Board of Agriculture, Harrisburg, 1891; Nature and Origin of Deposits of Phosphate of Lime, by R. A. F. Penrose, United States Geological Survey Bulletin, No. 46, 1888; Agricultural, Botanical and Chemical Results of Experiments on the Mixed Herbage of Permanent Meadow, Conducted for more than Twenty Years in Succession on the Same Land, by Sir J. B. Lawes and J. H. Gilbert, in Rothamsted memoirs (see also Philosophical transactions, Pt. I, 1880, Pt. IV, 1882, Vol. 2, 1886); Chemical Conversion Tables for Use in the Analysis of Commercial Fertilizers, by F. B. Dancy and H. B. Battle, Raleigh, 1885; Talks on Manures, by Joseph Harris, New York, 1878; On the Connection between Manures Made on the Farm and Artificial Manures, a lecture by Sir J. B. Lawes, Haddington, England, 1877; A Practical Treatise on Pure Fertilizers and the Chemical Conversion of Rock Guanos, Marlstones, Coprolites, and the Crude Phosphates of Lime and Alumina into Valuable Products, by Campbell Marfit, London, 1873; Commercial Fertilizers, by Peter Collier, Montpelier, Vt., 1872.

Besides the publications referred to above, numerous valuable addresses before agricultural organizations and conventions have appeared from time to time in the reports of societies and of state departments and boards of agriculture. Publications of this character issued by the states of Connecticut, Maine, Massachusetts, New York and Pennsylvania, should be especially mentioned in this connection. German and French literature also present many results of investigations of plant-nutrition problems that are of a fundamental character and are important in American farm practice.

CULTURE EXPERIMENTS FOR DETERMINING FERTILIZER NEEDS

By Cyril G. Hopkins

In conducting experiments in the field or by pot cultures to determine what plant-food element limits the crop yield, special care must be taken to avoid indirect or secondary influences. Thus, in the selection of plant-food carriers, soluble salts and low-percentage materials should be avoided, if possible, because of the secondary effects which they may produce. This article deals with soil experiments, that is with experiments to determine what element or elements may be deficient in the soil. Experiments to determine the comparative value or availability of different carriers of the same element are properly called fertilizer experiments. [See also pages 134 and 365.]

Kinds of plant-food materials.

Nitrogen. As a nitrogen fertilizer, dried blood carrying 14 to 15 per cent of nitrogen is the most satisfactory form for experiments. It is a readily available and concentrated nitrogen carrier. It is not soluble in water and will produce no secondary effects unless its decomposition products slightly influence the liberation of other plant-food and affect the physical properties of the soil by the small addition of humus. These secondary effects are appreciable with less concentrated forms of organic nitrogen, as cottonseed meal.

Sodium nitrate is objectionable because it is a soluble salt and also because the element sodium, which is present in larger amount than the nitrogen, is a strongly alkaline base that, after the nitrogen has been taken up by the plant, largely remains in the soil and may correct soil acidity or displace and liberate equivalent amounts of potassium, calcium, magnesium, and the like, from the mineral constituents of the soil. These secondary effects of sodium nitrate are especially noticeable on legume crops, as clover, which is markedly benefited by the correction of soil acidity and by the liberation of mineral plant-food, while, if infected with the proper bacteria, it is practically independent of the supply of nitrogen in the soil. Since a good growth of clover is usually of great benefit to succeeding crops in the rotation, this indirect or secondary effect of sodium nitrate, which may be the means of saving the life of the plant at a critical period in unfavorable seasons, may exert a more beneficial influence on the entire crop rotation than the effect due to the nitrogen applied.

Ammonium sulfate is objectionable as a nitrogen fertilizer for experimental work because ammonium also acts as a strong base, becoming quickly, though temporarily, fixed in the soil and liberating mineral bases. Furthermore, as the ammonia nitrogen nitrifies and is taken up by the crop, the sulfuric acid radicle tends to increase the acidity of the soil until it may become injurious to certain crops, notably clover and barley.

Phosphorus. As a phosphorus fertilizer, steamed bone-meal, carrying 12 to 14 per cent of the ele

ment phosphorus, and very finely ground natural rock phosphate, which is as rich in phosphorus as steamed bone, are satisfactory carriers. The bone-meal when well-steamed and finely ground is readily available and contains less than 1 per cent of nitrogen, which is in organic form and probably no more effective than the small amount of soluble nitrogen usually contained in acid phosphate. Rock phosphate is not readily available and must be applied in larger quantities if substituted for steamed bone-meal.

Acid phosphate is objectionable because it contains a soluble, corrosive acid salt, H, Ca(PO4)2, and a much larger quantity of soluble calcium sulfate formed in the process of manufacture. Calcium sulfate (land-plaster) is a well-known and very effective soil stimulant, having power to liberate mineral plant-food from the soil, and thus for a time to increase the growth of plants, especially of clover and other crops requiring much potassium. High-grade acid phosphate contains but 7 per cent of the element phosphorus, being only one-half as concentrated as steamed bone or natural rock phosphate.

Slag phosphate is objectionable because it is a strongly alkaline material, carrying some caustic lime which corrects soil acidity and which also acts as an effective soil stimulant, both in liberating mineral plant-food and in promoting nitrification. Slag phosphate is also much less concentrated than steamed bone.

Raw bone-meal is not satisfactory because it carries too much nitrogen. It is also poorer in phosphorus and less readily available than steamed bone-meal.

Potassium.-There is no potassium fertilizer which is satisfactory for use in culture experiments for determining which element of plant-food limits the crop yield. Potassium sulfate carrying 40 per cent or more of potassium is the least objectionable, but even this is a water-soluble salt capable of disturbing to a greater or less extent the chemical equilibrium of the soil. It is well known that the solubility of tricalcium phosphate, is appreciably increased by small amounts of soluble potassium salts, doubtless because of the formation of some potassium phosphate. Sodium salts also increase the solubility of tricalcium phosphate in laboratory experiments and, while sodium chlorid when applied to the soil tends to liberate potassium, calcium, magnesium and others, it is evident that the increased crop yields resulting from applications of common salt are in some cases produced by the phosphorus thus made available rather than by the potassium liberated.

As a potassium fertilizer, the chlorid possesses the same objectionable qualities as the sulfate, and in addition it tends to form calcium chlorid which may become injurious to crops if it remains in the soil; or if it passes off in drainage water, as it is likely to do because of its great solubility, it reduces the calcium or lime content of the soil, thus hastening soil acidity.

Kainit is very objectionable as a carrier of potassium not only because it consists of soluble

salts of potassium, magnesium and sodium, but also because it contains only 10 per cent of potassium, so that four times as much kainit as potassium sulfate would be required for equal applications of potassium. In some cases the indirect effect of kainit (and possibly even of potassium sulfate or chlorid) is greater than the effect due to the potassium as plant-food.

Amounts of plant-food materials.

The amount of dried blood, steamed bone, and potassium sulfate to use in experimental work should be governed by the requirements of the crop for approximately maximum crop yields. Thus a crop of average corn, yielding 100 bushels of grain and three tons of stover per acre, will contain about 148 pounds of nitrogen, 23 pounds of phosphorus, and 71 pounds of potassium; while 1,000 pounds of dried blood, 200 pounds of steamed bonemeal, and 200 pounds of potassium sulfate, should supply 150 pounds of nitrogen, 25 pounds of phosphorus, and 80 pounds of potassium. As an initial application, that is for the first year, it is well to apply double the standard amount of steamed bone because it is less readily available than dried blood and potassium sulfate. For wheat or oats the nitrogen may be reduced to one-half and the phosphorus and potassium to two-thirds of these amounts. For a crop of a different class of plants, as sugar-beets, quite different proportions and amounts may be needed; while, for a rotation of crops including a liberal use of legumes and some farm manure, the application of commercial nitrogen should be entirely omitted and the amounts of mineral elements may be reduced; and, with liberal quantities of farm manure containing both solid and liquid excrement, the phosphorus may be reduced and the commercial potassium may be greatly reduced or entirely omitted.

Specific examples of culture experiments.

The following examples will illustrate systems of pot cultures and plot experiments and at the same time show the actual results secured on different soil types.

(a) Soil deficient in nitrogen.—Much of the rolling or sloping unglaciated hill land in the southernmost seven counties of Illinois has been under cultivation 75 to 80 years. This was originally a fairly productive soil, 25 to 30 bushels per acre of wheat having been common yields. Through loss of nitrogen by continuous cultivation in corn and other grain crops and by surface washing, the productive capacity has been so reduced that five bushels of wheat per acre is now the average yield on the worn soil.

A series of pot culture experiments conducted in 1902 gave the results shown in Table I. Common glazed earthenware, 4-gallon jars, 10 inches in diameter, are used. A half-inch hole is punched through the bottom and covered with copper gauze and a bunch of glass wool for drainage. The pot is half filled with soil collected 5 to 10 inches below the surface, and then filled with soil from the surface 5 inches with which the plant

The chemical analysis of this soil showed that it was very deficient in nitrogen, while moderately supplied with phosphorus and rich in potassium. The pot cultures gave results in accord with the chemical composition, except that potassium when added to nitrogen increased the yield from 26 grams (No. 3) to 33 grams (No. 7), but this increase was probably due not to the effect of potassium as added plant-food but rather to the additional supply of phosphorus made available by the soluble potassium salt.

The marked effect of nitrogen suggests, not that commercial nitrogen should be purchased and used on this soil, which would be too expensive to be profitable in general farming, but that liberal use should be made of legume crops as catch-crops or in rotation. As soon as the wheat was harvested cowpeas were planted in No. 2, and later in the season they were turned under as a catch-crop. In 1903 wheat was again grown on all the pots, and a second legume catch-crop was grown after the wheat on No. 2. This was repeated in 1904, and again in 1905, with the results shown in Table II:

less due not to a more abundant supply of nitrogen furnished but to the supply of phosphorus liberated in part from the legume crop residues and in part from the soil itself in contact with the decaying organic matter. (See Figs. 641 and 642.) These pot-culture tests have now been confirmed by field experiments.

(b) Soil deficient in phosphorus. The principal type of soil in more than 20 counties in southern Illinois (lower Illinois glaciation) is markedly deficient in phosphorus, moderately supplied with nitrogen, and fairly rich in potassium, as judged by chemical analysis in comparison with normal fertile soils. Table III gives the results of field experiments with wheat in 1904, on tenth-acre plots on the University of Illinois soil experiment field near DuBois, Washington county, Illinois:

Fig. 641. Pot cultures of wheat (1902). Soil deficient in nitrogen; commercial nitrogen applied to pot 3 (on right). See Table II.

added. It may be stated that an entirely independent duplicate series of plots on this experiment field produced very similar results.

Table IV gives three years' results in wheatgrowing on this same type of soil in a different system of plot experiments, conducted on fifth-acre plots on the University of Illinois soil experiment field near Odin, Marion county, Illinois. A fouryear rotation is being practiced, on four different series of plots, each crop thus being represented every year. No nitrogen has been added except by growing legume crops and catch-crops in a rotation of corn, oats, wheat and cowpeas. The legume treatment for 1903 consisted of a catch-crop of cowpeas grown after oats in 1902. The legume treatment for the 1904 wheat consisted of catchcrops of cowpeas in the corn in 1902 and after the oats in 1903. For the 1905 wheat crop the previous legume treatment has been a full crop of cowpeas turned under in 1902, and catch-crops of cowpeas in the corn in 1903 and after the oats in 1904. No legumes are grown on the check plot except in the regular rotation crop, and that is harvested and removed.

TABLE IV. WHEAT YIELDS FROM SOIL DEFICIENT
IN PHOSPHORUS.

Wheat bushels

per acreе

Legume, lime, phosphorus

10

22

36

Legume, lime, phosphorus, potas

7

11

25

16

The season of 1903 was very poor for wheat, 1904 was fairly normal, while 1905 was exception

20

dicated by the

yield on the un

33

treated plots.

A gain of 18 bushels per acre was produced by the phosphorus even when applied alone; and when phosphorus was applied after nitrogen the yield was increased from 11 bushels to 33 bushels per acre, making a gain of 22 bushels per acre under field conditions. Nitrogen increased the yield somewhat (4 bushels) and the increase became more marked (8 bushels) after phosphorus had been applied, as would be expected; but the gain of 9 bushels per acre (7 to 16) made by applying potassium alone, and the gain of 9 bushels (11 to 20) made by adding potassium to nitrogen, may rightly be considered as evidence that the soluble potassium salt has produced this increase not because of the potassium added as plant-food but rather by the phosphorus liberated from the soil. This evidence is strengthened by the fact that nitrogen and phosphorus without potassium produced as large a yield as when potassium also was

The soil is acid, and lime produces an appreciable increase in the growth of legumes. The legume crop residues not only supply nitrogen but as they decay they evidently liberate some phosphorus and ultimately abundant potassium for the wheat crop. It is noteworthy that while potassium produced a

Fig. 642. Pot cultures of wheat (1905). Same soil as described in Fig. 641, deficient in nitrogen. No. 1 (on left) received no plant-food; No. 2, received legume catch-crops: No. 3 (on right) received commercial nitrogen. See Table II.