76 THE INHERITANCE OF CHARACTERS IN RICE 



consisted very largely of natural crosses resulting from the cultivation, side 

 by side, of a large number of different varieties. 



In 1913, when this work was started, an examination was made of about 

 one hundred varieties growing on the Government Farm at Coimbatore. Some 

 of these were almost pure, containing very few plants differing from the type, 

 whereas others were so much mixed that it was impossible to distinguish a 

 definite type for the variety. In each variety a few type plants were selected, 

 together with any variations which were to be found. In selecting the latter 

 such plants as appeared to be unrelated to the type, due to accidental mixture 

 of other varieties, were neglected. 



Almost all these selected plants were selfed in the manner described below. 

 Their progeny, grown in 1914, showed that a large proportion of the aberrant 

 plants were heterozygous. The segregating families gave Mendelian ratios 

 for various characters and these were followed up in further generations. For 

 the sake of convenience the original aberrant plants which proved to be hetero- 

 zygous w^ill be referred to as Fi and their progeny as F2, F3, etc. 



In addition to this material a few definite crosses were made between 

 pure strains. 



II. PRELIMINARY CONSIDERATIONS. 



{a) Pollination. 



From the accounts pubhshed by observers in different places it would 

 seem that the details of normal polUnation vary considerably for different 

 localities. 



Knuth^ states that dehiscence of the anthers does not take place till 

 they are fully extended and in a pendent position. Akemine^ has found that 

 in a variety of North Japan self-fertilization takes place immediately before 

 the flowers open. Fruwirth and Van der Stok^ conclude that in Java 

 cross-fertilization occasionally takes place between neighbouring plants. 



Farneti* reports that in Italy rice is cleistogamous, the glumes remaining 

 closed throughout flowering. A number of freshly introduced and old accli- 

 matized varieties gave the same result and he concludes that cross-fertilization 

 is impossible. 



1 Knuth, P., Hand-book of Floiver Pollination, vol. Ill, 1909, page 521. 



2 Akemine, M., " Ueber die Blute und das Bluhen von Oryza satira,"' Lavdw. Ztsrh. 

 Nogyo-Sehai, 1910-11. 



3 Fruwirth, C, and Van der Stok, E., Die Znchhi7)g der Landw. Kidhirpflanzev, 

 Bd. V, p. 36. 



* FsiTneti,'RoAo]to,Atti dell Istituio Botanico dell Universita'di Pavia, Series II, vol. XII, 

 1913, p. 351. 



