always chiefly supplied by the atmosphere and may be supplied by the atmosphere exclusively. It is not necessary that it should be in the soil.

The nitrogen of the plant, which forms indeed a small proportion-two per cent. perhaps, as an average-of the dry plant, is an important ingredient, for without it vegetation cannot exist.

Some crops have the power of gathering nitrogen without any difficulty; and they not only supply themselves with it but they 'even cause its accumulation in the soil. There are other crops which are dependent upon artificial supplies of nitrogen, unless the soil be naturally very rich in this element-crops which, if we undertake to raise them continuously on the same field, presently begin to show that they lack something, while if we apply nitrogenous compounds as fertilizers, the growth is ensured. We do not know in full detail how plants acquire a sufficient supply of nitrogen from the atmosphere, but we conclude, with great probability, from the results of practice, that different plants draw on the natural supplies of nitrogen in a different way.

Let us consider how the structure and habits of two typical crops, wheat and clover, stand in relation to their power of assimilating atmospheric nourishment. In respect of foliage we cannot certainly say that the wheat plant or the wheat crop when full grown, exposes a less surface to the air than full grown clover, but we know that the leaves of wheat, as of all our cereals, maintain their green color and succulence during a much shorter time than is true of clover. In case of winter grain the period of leafactivity usually begins in October and ends shortly after heading out, in June, some weeks before the crop is harvested. Clover, on the other hand, is not arrested in its growth by any crisis of seed-production, but, when cut for hay, sends up new shoots, unfolds new leaves, and shortly yields an aftermath, its growth going on uninterruptedly all the summer and late into autumn, until checked by heavy frosts.

That the actual leaf surface of the clover crop, taking its duration into account, is much greater than that of the wheat crop, I do not doubt, because although the total weight of the harvested crops is, on the average, not very unlike when clover is cut for hay,*

the total amount of vegetable matter organized is much greater in case of clover than in that of wheat, as appears from the table on page 195, where clover roots are seen to constitute two-fifths (equal to six-fifteenths) of the entire plant, while the roots of rye, which doubtless do not differ much from those of wheat, are but one-fifteenth of the entire plant.

You see that the foliage and mode of life of these two classes of plants are very different for the purposes of gathering food from the atmosphere, and they must therefore be expected to leave the soil in a very different condition, because their roots remain there, and the material of those roots is gathered very largely from, the atmosphere; so that when we raise a grain crop we leave in the soil a small quantity of material taken from the air, but when we cultivate a deep-rooted plant which grows the season through, we leave a large amount of atmospheric matter in the soil.

Again, in ordinary culture some plants are permitted and required to reach a crisis of growth which others are not allowed to attain. This crisis is seed-production.

Our meadow grasses are of the same botanical order as the cereal grains; which means that all these plants are of the same great race and closely resemble each other in their most characteristic features. The noble wheat and the scoundrel quack are, in fact, brothers of the same family, both being of the genus Triticum. The latter is sometimes termed wheat-grass, as if in allusion to this brotherhood. There are two other grasses, vagabond members of the wheat family, living obscurely in this country. Barley and the oat have each two brothers of low degree-worthless grasses, living on salt or sandy shores, or on rocky hills, and unknown to the cultivator.

If wheat, instead of being allowed to ripen its seed, as is our universal practice, should be mown or fed off just before heading out, it would throw out new shoots and continue to grow the summer and autumn through, would come on the second year and deport itself as a perennial; would in fact, become grass in the usual sense of that word. Wheat is probably not hardy enough to make a good substitute for Timothy, but it is sufficiently so to justify our statement.

The reason why wheat under our culture is an annual is that the process of seeding exhausts the plant, and as a consequence it dies out naturally. It is the universal opinion among farmers that the meadow grasses are weakened very much by being al

lowed to go to seed. I have myself observed that where Timothy seed was raised the crop of grass the next year was very small, although the soil was excellent. The plants had suffered severely from being allowed to go to seed, notwithstanding Timothy has a bulb at the root, which should fortify it considerably against this strain, and a small seed, which renders the exhaustion less than is the case with our bread grains. The production of seed is thus a critical thing for the life of the plant.

Let us consider again for a moment, the mode of growth of our cereal grains. Sown in the spring, the plant comes up and grows, slowly at first but with increasing vigor, up to the time of "heading out". -a period of two months. Then the growth acquires its greatest intensity. It heads out, blossoms, and the seeds begin to form and ripen, and this whole process of seed production requires but about a month when the weather is favorable for its completion.

In actual trials with the oat plant, it has been found by Bretschneider and Arendt that a large share of the growth of the overground part of the plant occurs at the time of heading and blossom. Thus the former observed that out of 6,886 lbs. of the dry acreage yield of the oat, 3,099 lbs., or three-sevenths of the crop, were produced from June 19th to July 8th, i. e., in nineteen days; the total period of growth being one hundred and six days. Arendt found that three-eighths of the total dry produce of the oat grew in twelve days, 18th to 30th of June, the period of heading and bloom, and during the twenty-two days between June 18th and July 10th, nearly three-fifths of the growth took place. [How Crops Grow, p. 205, et seq.]

Before the seed is ripe the lower leaves begin to turn yellow, and show that their activity is diminishing or has ceased altogether; and the ripening of the plant takes place to a great extent, by the removal of matters which have been previously stored up in the stem, leaves and roots, into the seed. You may cut any of the grains at the root when the kernel is in the milk, and the seed will still ripen, and although, if you cut it too early, the kernel will shrink, it will be perfect in its parts and serviceable as seed grain.

It thus appears that the cereal plant grows from the soil and atmosphere until the seed arrives at a certain stage of development, then the activity of the roots and foliage decreases, the acquisition of food from external sources gradually diminishes,

until it ceases altogether, the plant concentrates all its energies upon the seed, all its juices flow thither, and the roots, as well as the leaf and stem, are exhausted in the effort. The seed grows,

not directly at the expense of the soil and atmosphere, as the plant has done, but at the expense of the plant itself. It is, indeed, true in all cases that the seed is formed from the plant itself; but there are plants which, while feeding the seed from themselves, are still active in gathering food from external sources; and there are other plants, like the cereals, which do not, at the same time that they are elaborating seed, gather food from outside sufficiently to maintain their individual life.

In contrast with the cereals, look again at the clover plant. This starts from a seed, grows vigorously, buds, blossoms, forms seed, and the seed ripens; but there is not that uniformity in the time of budding, flowering and ripening of clover that is noticed with wheat. In a field of wheat, if the catch has been good and everything as it should be, when one head is ripe all the heads in the field, practically, are ripe. Every stem heads out, blossoms and ripens about the same time. In the case of clover, you have a much greater diversity, especially when the soil is rich and the plant grows thriftily. If the soil is poor, you will have a nearer approach to uniformity. When yon are getting a large amount of foliage, you will find on the same plant ripe heads and buds. If you pick off the ripe heads the plant will still keep throwing out new buds. The process of flowering and ripening is a continuous one, and it does not affect the vigor of the plant to nearly the degree that happens to wheat. During all the period of the growth of the clover plant until the seeds are ripe, the roots are still active and the foliage still vigorous. The quantity of seed produced by the clover plant is much less, relatively to the weight of the plant, than the quantity of seed produced by the wheat plant, and the energies of the clover plant are relatively less occupied in ripening the seed than is the case with wheat.

You would therefore expect these very different plants to have a very different function in the rotation of crops.

An annual plant, again, one that is sown in the spring, or in the fall, perhaps, and is harvested within a year, other things being equal, will be different in its relation to the soil, from a biennial plant, which lives two years, or a perennial plant, which keeps along indefinitely. Now, our ordinary grains are annuals, as we cultivate them; the clover plant is a biennial more nearly

than any thing else. When it grows vigorously, it is usually spent in about the second year. We may not call it properly a biennial in a botanical sense, but in an agricultural sense it is a two-year old plant. We cannot depend, ordinarily, upon having much clover from the sowing of 1872, later than 1874, except as the result of self-seeding. Our natural grasses are perennial; they live, we do not definitely know how long. Their mode of propagation, besides from seed, is by root-suckers; the old root dies, but in the meantime it has propagated a numerous family, which succeeds it, and the race is kept up without trouble of sowing any seed or giving any attention to the matter at all. These distinctions make an obvious difference in the relation of the three kinds of plants to the subject of rotation of crops.

We have thus considered the plant itself, its roots, foliage, and manner of growth; now let us look more closely at what remains when the crop is removed. This matter came up incidentally, and a little out of order, yesterday, as I referred to the tables on the board. When I raise a crop and harvest it, I leave, of course, the roots in the soil, I leave the stubble on the surface. If each crop were taken out of the soil completely, root as well as branch, so that nothing of it were left in the field, the effect of any crop upon the soil would be measured simply by what we took away. But we leave a great deal in the soil. Ever since farming has been practiced, the value of what is left on and in the soil has been, to some extent, appeciated, but we have not known accurately the quantities or the relative proportion of those substances. We have known that clover leaves much more than wheat, but the precise relation we have not understood as we understand it now, and we do not understand it now as we ought to and as we shall understand it after further investigation. I referred yesterday to the table of Dr. Weiske, of Proskau, which gives the ingredients of the stubble and roots of various crops remaining on and in an acre of land after harvest. (See page 194.) This is the first, or nearly the first, exact experiment of the kind that has ever been made, and these observations must be repeated here and there, on different soils, before we can get entirely trustworthy data, to enable us to make a satisfactory calculation. Still, these first results will serve a very good purpose.

In the case of rye, for instance, you have 3400 lbs. of dry vegetable inatter remaining in the soil to the acre. Ordinary rye straw contains some fourteen per cent. of moisture. The vegetable