In a previous set of experiments the Giessen professor had ascertained that the particular plants under observation grew equally well in all the varieties of soil in which they were placed, provided due care was taken to prevent the growth of intruding weeds. Having arrived at this result, Prof. Hoffmann next left the several plants to themselves, with a view of ascertaining how they would comport themselves, without assistance, against the inroads of weeds. The result was, that the weeds completely gained the upper hand, as might have been expected from their known habit. The species which held out longest was Asperula cynanchica. This plant, after having been grown in a bed for three years, and protected from weed-invasion by the use of the hoe, was then left to take care of itself. It held out for four years, but was ultimately elbowed out by the intruders. Acting on the principle of "set a rogue to catch a rogue," Prof. Hoffmann then set himself to observe the results of the internecine struggle between the weeds themselves, thinking that the ultimate survivors would perhaps prove to have special affinities for the soil in which they grew.
Thus left to themselves, the beds became so densely covered, that, in a square foot, the professor counted 460 living plants, and the remnants of many others, which had succumbed in the encounter. Every year, in July, the plots were examined, and every year the number of species was found to have diminished. Melilots, at first abundant, gradually disappeared; Artemisia vulgaris succumbed after two or three years; and so on, till at length only a few species were left, and these not only persisted, but slowly gained ground from year to year, and ultimately remained in possession of the plot. The plots under observation were 2 mètres 30 cents. long, 1 mètre broad, and all as nearly as possible under the same conditions, save that the soil was varied, in some cases consisting of the ordinary soil of the garden, in others of an admixture of lime, in others of sand, or of sand and lime, and so forth.
Of the 107 species under observation, all, or nearly all, found the most essential requisites of their existence equally well in all the varieties of soil; so that, other conditions being equal, the nature of the soil was indifferent. The species which remained victors, all the others being ultimately dispossessed, were Tritieum repens (couch), Poa pratenis, Potentilla reptans, Acer Pseado Platanus (sycamore), Cornus sanguinea, native plants; and Aster salignits, A. parviflorus, Euphorbia virgata, and Prunus Padus, derived from other portions of the garden.
It may, therefore, be inferred that the district in which these experiments were made would, in process of time, if no obstacle were afforded, become covered with meadows and woods—meadows in the low ground and woods in elevated places. Again, the experiments show that the survival of certain plants has not been influenced by the