82 



The Journal of Heredity 



AVENA BARBATA TRIFLORA 



A well-marked form of A.barbata. This species is found 

 throughout much of northern Africa, the Abyssinian 

 varieties being closely allied to it, if not identical. 

 (Figure 19). 



In 1854, Cosson (Bulletin dc la 

 Societe Botanique dc France, vol. 1, 

 p. 15) distinj^aiishcd an A. 'Jatua glahre- 

 scens which merj^^cs with A. hybrida 

 Peterm. (Flora Saxonica, p. 17, 1842). 

 These forms have been regarded as 

 hybrids of A. Jatua and sativa, but with- 

 out any proof. In the Sersou (altitude 

 1000 metres) I have found a link uniting 

 A. hybrida with A. sativa. 



This form (A. subunijlora) presents 

 the general characters of A. Jatua; l)ut 

 it is glabrous, the well indicated articu- 

 lation functions badly, the second 

 flower is most often abortive and the 

 lower flower has a long twisted and 

 geniculate awn (fig. 17, nos. 3-4) like 



the wild forms. In culture it main- 

 tains itself and presents at the same 

 time some awnlcss individvials which it 

 will he easy to fix. These transition 

 forms arc sometimes regarded as retmnis 

 of a cultivated race to a wild state; 

 this is possible, but it is certain that the 

 supposed reversions of A. sativa lead 

 toward A . Jatua. 



It is quite evident today that A. Jatua 

 has .su]:)i)lied the great majority of the 

 culti\-atcd oats brought together under 

 the name .4. sativa, and that it is still 

 possible to bring together the wliole 

 series of intennediatc fonns between the 

 wild type and the forms adapted to 

 cultivation. 



