592 THE FERTILISATION OF FLOWERS. [PART Iv. 
(1) In maize cultivated at Itajahy, the grains stand for the most part in 
twelve or fourteen rows, not unfrequently in ten, rarely in eight or sixteen, and 
very rarely in eighteen rows. In more than 100 ears which I bought for 
sowing in 1867 I found only one of eighteen rows. This ear produced next year 
205 ears, among which were twenty-two with eighteen rows, besides one which 
had eighteen rows in the upper part and twenty in the lower, and one which 
had eighteen at the top, twenty in the middle, and twenty-two at the bottom. 
Next year, 1868-69, I had 460 ears obtained from seeds out of ears with eighteen 
rows ; and of these 460, 18°2 per cent. bore eighteen rows, 4'4 per cent. bore 
twenty, and ‘2 per cent. bore twenty-two. In the following year, among ears 
obtained from seed out of ears bearing twenty-two rows, one ear occurred 
which bore twenty-six rows. 
(2) In an Abutilon from Capivary the normal number of styles isten. A 
seedling obtained from a flower with nine styles had, among 100 flowers, two 
with seven styles, and twenty-seven with eight ; while in 100 flowers upon the 
mother-plant there were only three with eight, and none with seven styles. 
“ (3) In Abutilon there occur sometimes, but very rarely in most species, 
flowers with six petals. 
“ From seed obtained from an hexamerous flower of the hybrid EF’, fertilised 
with pollen from a pentamerous flower of the variety /’, a plant was reared 
(EFF) which I watched for three weeks (August 17 to September 6, 1869) 
counting the petals upon all the flowers that appeared on it. It bore in the 
three weeks :— 
Flowers with 5 petals. A ; ‘ 145 
Ditto MLL lala : : ; . 103 
Ditto ee Sith ee : : pies 13 
“At the same time I watched another plant descended from pentamerous 
parents of the same varieties as those of the former (but obtained by pollen of 
the hybrid EF applied to the stigma of the pure species /), It bore in the 
three weeks :— 
Flowers with 5 petals : : : ; 454 
Ditto A MS ; : ’ é 6 
Ditto eee i P : , : 0 
“The simplest explanation of these facts seems to be, that each species has 
the property of varying within a certain amount: crossing of different in- 
dividuals so long as no selection is made in a definite direction, suffices to keep 
the middle point fairly constant, about which the variations oscillate ; and so 
the extremes also remain constant. But if the tendency to vary in either 
direction is aided by natural or artificial selection, then our middle point is as 
it were shifted to that side and the extremes of oscillation are also shifted 
towards the same side.” 
Since the first appearance of this work (1873) I have discussed 
in a series of essays the question of the Origin of Flowers and the 
1 # indicates the Abutilon called Hmbira branca by the Brazilians; F, an 
Abutilon from Pocinho (see No. 557); ZF implies the hybrid produced by pollen of 
F applied to the stigma of Z. 
