190 THE INHERITANCE OF CHARACTERS IN RICE, If 
The following figures, showing approximately the 9: 3: 5: 1 ratio 
expected, were obtained :— 
DARK FURROWS GOLD 
Piebald Self-colour Piebald Self-colour 
GIP GIp GiP Gin 
No. 1285 N .. | _—*969 278 283 97 
Oo ae32 1 anors: te RIE este 0? 
The piebald types are shown in Plate I, figs. 6 and 7. 
The effect of the piebald factor on the ripening gold type has not yet been 
noted. 
The factor T. 
The action of this factor is very similar to that of P, since it limits the 
colouring due to G, though to a smaller extent. In homozygous dark golds it 
produces t7pped gold, a type in which small green areas appear at the apex and 
base of the glumes (Plate I, fig. 8). The green areas vary somewhat in size 
and occasionally the lower one is absent, the base of the grain being entirely 
gold. In heterozygous golds, where the colour is lighter, the restriction of the 
gold colour is greater and much less regular, giving a very variable type, patchy 
gold (Plate I, fig. 9). 
The restriction of dark furrows is very irregular, the result being a con- 
siderable reduction of the pigment, which assumes a granular appearance, and 
its restriction roughly to the middle part of the grain. The type produced, 
granular dark furrows (Plate I, fig. 10), shows very considerable variation. 
The reduction of pigment is much greater in the Gg type, as with the golds, 
but it is not possible actually to separate two types corresponding to tipped 
and patchy golds. 
The factor 7 inhibits ripening gold almost entirely, giving ripening straw 
in which the glumes are slightly yellowish in early stages (Plate I, fig. 3), but 
ripen to ordinary straw-colour (Plate II, fig. 3). Occasionally afew grains 
show traces of gold in very small patches and there is some indication that 
this is more common in heterozygotes of the 7t type. The golden internode 
is not affected but persists in the ripening straw type as one of its distinctive 
e atures. 
The results given in Part I, pp. 82-84 and Tables IV-VI, are now 
easily explained. The original cross that gave the families of Table IV 
