Some of these two lots of seeds were left on sand, but nearly all the self-fertilised seeds germinated before the others, so that I was forced to throw all away.  The remaining seeds were then sown on the opposite sides of four pots.  At first the self-fertilised seedlings were a little taller in most of the pots than the naturally crossed seedlings, and this no doubt was due to the self-fertilised seeds having germinated first.  But in the autumn all the plants were so equal that it did not seem worth while to measure them.  In two of the pots they were absolutely equal; in a third, if there was any difference, it was in favour of the crossed plants, and in a somewhat plainer manner in the fourth pot.  But neither side had any substantial advantage over the other; so that in height they may be said to be as 100 to 100.

19.  Dipsaceae.—­Scabiosa atro-purpurea.

The flowers, which are proterandrous, were fertilised during the unfavourable season of 1867, so that I got few seeds, especially from the self-fertilised heads, which were extremely sterile.  The crossed and self-fertilised plants raised from these seeds were measured before they were in full flower, as in Table 5/65.

Table 5/65.  Scabiosa atro-purpurea.

Heights of plants measured in inches.

Column 1:  Number (Name) of Pot.

Column 2:  Crossed Plants.

Column 3:  Self-fertilised Plants.

Pot 1 :  14 :  20.

Pot 2 :  15 :  14 4/8.

Pot 3 :  21 :  14. 
Pot 3 :  18 4/8 :  13.

Total :  68.5 :  61.5.

The four crossed plants averaged 17.12, and the four self-fertilised 15.37 inches in height; or as 100 to 90.  One of the self-fertilised plants in Pot 3 was killed by an accident, and its fellow pulled up; so that when they were again measured to the summits of their flowers, there were only three on each side; the crossed now averaged in height 32.83, and the self-fertilised 30.16 inches; or as 100 to 92.

20.  Compositae.—­Lactuca sativa. (5/19.  The Compositae are well-adapted for cross-fertilisation, but a nurseryman on whom I can rely, told me that he had been in the habit of sowing several kinds of lettuce near together for the sake of seed, and had never observed that they became crossed.  It is very improbable that all the varieties which were thus cultivated near together flowered at different times; but two which I selected by hazard and sowed near each other did not flower at the same time; and my trial failed.)

Three plants of Lettuce (Great London Cos var.) grew close together in my garden; one was covered by a net, and produced self-fertilised seeds, the other two were allowed to be naturally crossed by insects; but the season (1867) was unfavourable, and I did not obtain many seeds.  Only one crossed and one self-fertilised plant were raised in Pot 1, and their measurements are given in Table 5/66.  The flowers