to the light and air, and those that are most vigorous usually get above and shade the others. Again, the supply of water and mineral food in the soil of a given area is limited; hence, plants that cannot get what they need are dwarfed and finally starved to death. In

the third place, injurious insects destroy an enormous amount of vegetation, the loss of cultivated crops alone from this cause being estimated at $700,000,000 annually. Frosts, dry seasons, heavy rains, and fungous diseases are other important factors in the life of many plants. And so if we were able to see what is actually going on in each square foot of the earth's surface, whether of forest,

field, or meadow, we should doubtless witness a life and death struggle for existence (1) between individual plants, of the same kind, (2) between individual plants of different kinds, and (3) between plants and animals.

Charles Darwin in his great book on the "Origin of Species," published in 1859, -a book which has doubtless influenced human thought more than any other book of modern times, closes his chapter on the "Struggle for Existence" with the following words: "When we reflect on this struggle we may console ourselves with the full belief that the war of nature is not incessant, that no fear is felt, that death is generally prompt, and that the vigorous, the healthy, and the happy survive and multiply."

129. The survival of the fittest. We have seen in our study thus far (1) that no two individual plants even of the same kind are exactly alike, (2) that enormous numbers of seeds are produced by plants, and (3) that there is inevitable competition or struggle for existence. The question, then, that confronts us is this: Which of the many competitors will survive in the

struggle, reach maturity, and finally reproduce themselves? Obviously those individual plants that vary from the rest in such a

1 Darwin's "Origin of Species," p. 72.

way that they can best adapt themselves to their surroundings. Let us see, for instance, why certain weeds like the dandelion are so common a nuisance on our lawns. In the first place these weeds have fleshy roots that reach deep down into the soil, thus helping the plant to get and keep a stock of moisture and food. In the second place the reserve supply of nutrition stored in these roots enables the plants to put forth leaves and flowers in early spring, and so to get a good start ahead of their competitors. Again, their short stems and tough leaves can be trampled upon without killing the plant. Insects and fungous diseases, for some reason, do not seem to attack them. And finally dandelions produce a large number of tiny seed-like fruits, each one of which is provided with a delicate tuft of hair which a puff of wind will carry for a considerable distance, thus insuring a wide dispersal of its seeds. In nature, then, plants like the dandelion, pigweed, and thistle have survived in the struggle for existence, because they are best fitted to their surroundings.

V. THE IMPROVEMENT OF PLANTS BY MAN

130. Artificial selection of favorable variations.

In the pre

ceding pages attention was frequently called to the fact that plants show a tendency to vary more or less from each other. Now it has been found that in a state of cultivation this tendency becomes even more pronounced. A watchful farmer will often find that in his cornfield one group of individuals ripens sooner than the rest, and so if he wishes to sell earlier corn, he selects and plants next year corn grains derived from plants that have varied in this direction. Again, he may notice that the ears on certain stalks are larger and

pen more kernels (see Fig. 52); these the crop-raiser who uses his brains would select for seed in order to increase his yield per acre. Variations in many other directions might be chosen by the successful farmer which would add immensely to the value of his crops. It is estimated that if every farmer were to select his seed carefully, the corn production in the United States, which at present is about $1,000,000,000, in a short time would be increased 10 per cent, which would add $100,000,000 to our annual income.

131. Artificial crossing of related species. Not only can man secure new varieties of plants by watching for favorable variations and perpetuating them from year to year, but he can actually be instrumental in producing new kinds of plants. This process is known as plant breeding. It depends fundamentally on the principles we learned in treating of cross-pollination in flowers. Let us illustrate plant breeding by the following account of the work which has been done for the U. S. Department of Agriculture by Dr. H. J. Webber of Cornell University.1

In the winter of 1894-1895 a heavy frost destroyed practically every orange tree in the northern and central part of the State

of Florida. The loss was over $75,000,000. The problem that confronted the orange growers of the State was that of starting their groves anew and if possible of preventing a repetition of such an experience by planting a more hardy kind of orange tree. Dr. Webber, in casting about for such orange trees, finally chose a type called the trifoliate orange (Fig. 56) often used for an ornamental shrub, and one that would not be killed by winters as far north as Philadelphia. The fruit of this tree, however, is small, bitter, and worthless for eating purposes. His task, therefore, was to combine the characteristics of a juicy, sweet-flavored fruit of the ordinary Florida orange tree with the hardy, cold-resistant character of the trifoliate type. He proceeded in this fashion:

1 See Year-books of U. S. Department of Agriculture, 1904, 1905, 1906.

From the flower-buds of one type of orange trees he removed all the stamens before blossoming time, and then covered the pistils with paper bags to prevent the visit of insects bringing pollen. A second set of buds on trees of the other type were likewise covered with paper bags to prevent possible mixing of pollen by insect visitors. When the stamens of one kind of orange blossoms and the

FIG. 57. Fruits of two parent plants (orange and trifoliate) at left. Six types of hybrid fruits (Rusk, Willits, 783, 771, 772, 767) developed by cross-pollination from the parent plants, all being good fruits except 767, which proved to be worthless. Compare seeds and pulp in various sections.

pistils of the other kind had matured, the bags were carefully removed, and the pollen of one variety was dusted over the pistil of the other (see 87). The paper bag was then replaced over the artificially pollinated pistil, and the latter left to ripen. Fruits