botanists agree to this theory, we shall let the facts we have already brought forward stand in maintenance of its opposite, namely, that cultivated forms are derived from wild species often apparently very different; but at the same time it may be well to state, that in all probability some of the discrepancies of experimenters may have arisen from some confusion in the species operated upon.

In 1860 we gathered some seed of the Daucus maritima (sea-side carrot) at Bognor, which, on being sown in a prepared plot the following spring, certainly resulted in fairly succulent roots, which on being cooked were pronounced by our party of four to be excellent. While on this subject, it may be mentioned as not a little remarkable, that so many of our garden esculents should be derived from sea-side plants. Thus, probably carrot, but certainly celery, sea-kale, asparagus, and cabbage. This would seem to point to the fact that cultivation requires a complete change of the circumstances necessary to maintain a wild condition; and hence cultivated plants can only be kept up by the labours of a cultivator.

Now, as regards the sea-side carrot, we are after all inclined to the belief that it is the parent of the cultivated varieties, whilst, on the other hand, we view the Daucus Carota (the wild inland carrot) as a probable descendant from the cultivated or garden stock; and if this be so, the Daucus maritima is the original species from which both the wild and cultivated races have descended. Bentham, indeed, carries this view a little further, the following remarks tending to throw doubts upon the carrot in any form as being a true native. Under the heading of Daucus Carota he says:—

Probably an original native of the sea-coasts of modern Europe, but of very ancient cultivation, and sows itself most readily, soon degenerating to the wild form, with a slender root, and now most abundant in fields, pastures, waste places, &c., throughout Europe and Russian Asia; common in Britain, especially near the sea. Flowers the whole summer and autumn. A decidedly maritime variety, with the leaves somewhat fleshy, with shorter segments, more or less thickened peduncles, more spreading umbels, and more flattened prickles to the fruits, is often considered as a distinct species.

Seeing then that crop plants are derivatives from a wild stock, we can readily understand how the varying circumstances attendant upon the development of the former should tend to the production of varieties, and this merely as the result of the treatment of the fairly derived legitimate seed. If, again, we take these variations for the purpose of obtaining hybrids, we need not wonder at the infinite variety of sorts which can be brought about, but rather that any sort could be maintained in that trueness of character or in that state of permanency which we sometimes find to be the case.

CHAPTER III.

ON TRUENESS OF SORT IN ROOT CROPS.

The importance of trueness and purity of seed arises from the evenness of growth of a good genuine strain; while if this quality be wanting we have some parts of our crop growing well, whilst others get on but poorly. Thus a free-growing plant beside one over which it has got the advantage, maintains it for the most part through the whole period of growth. Again, some sorts are of value for being early, others for lateness of growth, and some kinds are better fitted for early than late sowing; if, therefore, we have a mixture in these respects, we may at least expect a partial failure; for whichever is best for our purpose, if mixed will be accompanied by those which are not so good. A want of trueness to sort may arise principally from the following causes:—

1st. Want of selection in seeding bulbs.

2nd. Hybridization.

3rd. A mixture of seeds.

1st. The propriety of selecting the specimens from which seed is to be grown is admitted by all: by the seedsman, who always advertises his turnip and swede seed, for example, as being “from selected bulbs;” and by the farmer, as this announcement is only made to induce him to buy. It is not only important that the roots should be selected, but that they should be stored and then planted in a fresh soil; for as these latter are among the cultivative processes by which sorts have been obtained, so should they be repeated in order to ensure a continuance of the induced condition. Seeding upon the same soil and in the same bed in which the seed is sown is hardly the way to keep up a form induced by cultivation, as this is exactly what would be done by the plants in a state of wildness.

In selecting roots for seeding, care should be taken to choose good-shaped examples, in which a clean unbranched bulb, not too large, with a small tap-root and a small top, confined to a single central bud; a branched root and a many-headed top being true signs of degeneracy. And no less so is neckiness in swedes and mangels, as well as a coarse corrugated skin in roots of all kinds.

Taking such points as these into consideration, how absurd must appear most of the huge mis-shapen roots to which prizes are usually awarded at shows, where the specimens are chosen for size, and trimmed up with the knife, to make them look more presentable. As an evidence of the mistaken principles upon which prizes are awarded to bundles of roots, let any one seed such examples, and we will venture to assert that such seed would produce a large proportion of degenerate examples, without affording so good a crop as would seed, from middle-sized but well-shapen specimens.

2nd. Some of the forms of roots, and more especially those belonging to the Brassicaceæ, such as turnips and swedes, seem to have a wonderful facility for hybridizing; and this not only to the extent of one sort of turnip with another, but sports may be caused by the fertilization of the turnip with rape and its congeners. Indeed, the hybrid with turnip and rape is doubtless the origin of the Swedish turnip; but there is reason to believe that mixtures may accidentally be made with such wild plants as charlocks and mustards, the growth of which in the vicinity of a seeding crop tends to the production of degeneracy. Seeding-patches, then, and the ground about them, cannot be kept too clean.

Again, if trueness be aimed at, there should be no mixture of sorts in seeding examples; all of the same kind should be selected for seeding-plots, as even one or two of a wrong sort may result in a very mixed sample, as it would seem that sometimes strange plants exert more than ordinary influence.

Of course, the putting seeding-patches of different sorts side by side is to be reprehended. If more than one sort be seeded in a season, it is advisable to place the patches as remote from each other as possible. And we would here remark, that, for seeding, the roots should, as a rule, be farther apart than when grown for bulbs, both in rows and in sets; as, if too close, the stems grow up thin instead of robust, and a smaller seed, with a tendency to the growth of smaller roots, will be the result.

3rd. Mixtures of seeds should be avoided for the reason assigned, that “sorts” do not usually grow evenly; and when one sees (as is by no means infrequent) a patch of swedes overshadowed by a mixture of some large early turnip,—the Tankard, for example, our crop of swedes will certainly suffer for it, even supposing the turnip to be as useful as the swede, which is seldom the case.

Mixtures, again, do not