This is the whole of the story of the chinampas, the most fertile and beautiful little gardens upon the face of the earth. A correct picture of them would be poetry enough, without the addition of falsehood; for whether it is the rainy season or the dry season, it is always the same to them. They know no exclusive seed-time, and have no especial season for harvest; but blossoms and ripe fruits grow side by side, and flowers flourish at all seasons. As market gardens they are unrivaled, and to them Mexico is indebted for its abundant supplies. The evidence that Humboldt[35] produces in favor of floating gardens, viz., that he saw floating islands of some 30 feet in length in the midst of the current of rivers, amounts to little in this case; for every one that has traveled extensively in tropical lowlands has seen vegetation spring up upon floating masses of brush-wood. Where earth torn from the river bank is so bound together by living roots as to form a raft, it will always float for a little while upon the current, provided that its specific gravity does not materially exceed that of the water; and those grasses that flourish best in water will spring up and grow upon these islands. Peat, too, in bogs, will float and form islands, for the simple reason that it is of less specific gravity than water; and vegetation will also spring up on these peat islands. But all this furnishes no evidence that the invariable law of nature, which carries to the bottom the heaviest body, has been suspended at Mexico. Had the floating gardens been built in large boats made water-tight, they might have floated. But, unfortunately, the Indians had not the means for constructing such boats. Even timber-rafts would have become saturated in time, and sunk, as rafts of logs do if kept too long in the "mill-pond," waiting to be sawed into lumber. There is another law of nature, which must not be lost sight of, which is at war with the idea of a garden floating on a bed of rushes; and that is capillary attraction, which would raise particles of water, one by one, among the fibres of the rushes until the frail raft on which the earth rested was saturated; and still pressing upward, the busy drops would penetrate the superincumbent earth, moistening and adding to the specific gravity of the garden by filling the porous earth until it became too heavy to float, if it ever had floated.