Ten flowers crossed with pollen from plants known to be distinct seedlings, yielded nine capsules, containing on an average 34.2 seeds, with a maximum of seventy-seven in one. Ten flowers self-fertilised yielded eight capsules, containing on an average only 13.1 seeds, with a maximum of twenty-five in one. This gives a ratio of 100 to 38 for the average number of seeds per capsule for the crossed and self-fertilised flowers. The flowers hang downwards, and as the stigmas stand close beneath the anthers, it might have been expected that pollen would have fallen on them, and that they would have been spontaneously self-fertilised; but these covered-up plants did not produce a single capsule. On some other occasions uncovered plants in the same greenhouse produced plenty of capsules, and I suppose that the flowers had been visited by bees, which could hardly fail to carry pollen from plant to plant. The seeds obtained in the manner just described were placed on sand, and after germinating were planted in pairs,--three crossed and three self-fertilised plants on the opposite sides of four pots. When the leaves were 2 or 3 inches in length, including the foot-stalks, the seedlings on both sides were equal. In the course of a month or two the crossed plants began to show a slight superiority over the self-fertilised, which steadily increased; and the crossed flowered in all four pots some weeks before, and much more profusely than the self-fertilised. The two tallest flower-stems on the crossed plants in each pot were now measured, and the average height of the eight stems was 9.49 inches. After a considerable interval of time the self-fertilised plants flowered, and several of their flower-stems (but I forgot to record how many) were roughly measured, and their average height was a little under 7.5 inches; so that the flower-stems on the crossed plants to those on the self-fertilised were at least as 100 to 79. The reason why I did not make more careful measurements of the self-fertilised plants was, that they looked such poor specimens that I determined to there them re-potted in larger pots and in the following year to measure them carefully; but we shall see that this was partly frustrated by so few flower-stems being then produced. These plants were left uncovered in the greenhouse; and the twelve crossed plants produced forty capsules, whilst the twelve self-fertilised plants produced only five; or as 100 to 12. But this difference does not give a just idea of the relative fertility of the two lots. I counted the seeds in one of the finest capsules on the crossed plants, and it contained seventy-three; whilst the finest of the five capsules produced by the self-fertilised plants contained only thirty-five good seeds. In the other four capsules most of the seeds were barely half as large as those in the crossed capsules. TABLE 6/91. Cyclamen persicum: 0 implies that no flower-stem was produced. Heights of plants measured in inches.