The ten tallest Westerham-crossed plants produced 111 ripe and unripe capsules, weighing 121.2 grains.  Therefore 100 of such capsules would have weighed 109.18 grains.

The ten tallest intercrossed plants produced 129 capsules, weighing 76.45 grains.  Therefore 100 of these capsules would have weighed 59.26 grains.

The ten tallest self-fertilised plants produced only 44 capsules, weighing 22.35 grains.  Therefore 100 of these capsules would have weighed 50.79 grains.

From these data we get the following ratios for the fertility of the three lots, as deduced from the relative weights of an equal number of capsules from the finest plants in each lot:—­

Westerham-crossed plants to self-fertilised plants as 100 to 46.

Westerham-crossed plants to intercrossed plants as 100 to 54.

Intercrossed plants to self-fertilised plants as 100 to 86.

We here see how potent the influence of a cross with pollen from a fresh stock has been on the fertility of plants self-fertilised for four generations, in comparison with plants of the old stock when either intercrossed or self-fertilised for five generations; the flowers on all these plants having been left to be freely crossed by insects or to fertilise themselves.  The Westerham-crossed plants were also much taller and heavier plants than the self-fertilised, both in the pots and open ground; but they were less tall and heavy than the intercrossed plants.  This latter result, however, would almost certainly have been reversed, if the plants had been allowed to grow for another month, as the Westerham-crossed were still growing vigorously, whilst the intercrossed had almost ceased to grow.  This case reminds us of the somewhat analogous one of Eschscholtzia, in which plants raised from a cross with a fresh stock did not grow higher than the self-fertilised or intercrossed plants, but produced a greater number of seed-capsules, which contained a far larger average number of seeds.

Colour of the flowers on the above three lots of plants.

The original mother-plant, from which the five successive self-fertilised generations were raised, bore dingy purple flowers.  At no time was any selection practised, and the plants were subjected in each generation to extremely uniform conditions.  The result was, as in some previous cases, that the flowers on all the self-fertilised plants, both in the pots and open ground, were absolutely uniform in tint; this being a dull, rather peculiar flesh colour.  This uniformity was very striking in the long row of plants growing in the open ground, and these first attracted my attention.  I did not notice in which generation the original colour began to change and to become uniform, but I have every reason to believe that the change was gradual.  The flowers on the intercrossed plants were mostly of the same tint, but not nearly so uniform as those