The many seeds obtained from the above crossed and artificially self-fertilised flowers were sown out of doors, and two large beds of seedlings, closely adjoining one another, thus raised. This was the first plant on which I experimented, and I had not then formed any regular scheme of operation. When the two lots were in full flower, I measured roughly a large number of plants but record only that the crossed were on an average fully 4 inches taller than the self-fertilised. Judging from subsequent measurements, we may assume that the crossed plants were about 28 inches, and the self-fertilised about 24 inches in height; and this will give us a ratio of 100 to 86. Out of a large number of plants, four of the crossed ones flowered before any one of the self-fertilised plants. Thirty flowers on these crossed plants of the first generation were again crossed with pollen from a distinct plant of the same lot, and yielded twenty-nine capsules, containing on an average 55.62 seeds, with a maximum in one of 110 seeds. Thirty flowers on the self-fertilised plants were again self-fertilised; eight of them with pollen from the same flower, and the remainder with pollen from another flower on the same plant; and these produced twenty-two capsules, containing on an average 35.95 seeds, with a maximum in one of sixty-one seeds. We thus see, judging by the number of seeds per capsule, that the crossed plants again crossed were more productive than the self-fertilised again self-fertilised, in the ratio of 100 to 65. Both the crossed and self-fertilised plants, from having grown much crowded in the two beds, produced less fine capsules and fewer seeds than did their parents. CROSSED AND SELF-FERTILISED PLANTS OF THE SECOND GENERATION.