128 Glume Length in Triticum Polonicum 



tinged. In this experi^ient the polonicum will be considered smooth, 

 as indeed it is in comparison with the other ; the classification was done 

 by eye, unaided b}' a lens. The second generation was plotted as a curve 

 (Chart, Figs. 2, 3, 4 and 5) in the same way as the Kubanka Polish 

 cross. Here again it was impossible to separate the pure longs from 

 the heterozygotes and statistically there is no sharp dividing line be- 

 tween the shorts and the heterozygotes (Chart, Fig. 5). However it 

 is fairly safe to say that the pure shorts are comprised among those with 

 a glume length between 9 and 13 mm. — though undoubtedly several of 

 13 mm. are poorly grown heterozygotes. 



The ratio of short glume to long is : 



Long and 

 Intermediate Short 



514 178 



Expectation ... 519 173 



There were 692 plants in this family, besides 39 which were not noted, 

 being too green at the time of harvesting to determine the colour. 



The second curve (Chart, Fig. 4) shows the total analysed into 

 those individuals which were felted like the short glumed parent and 

 those (shown by the dotted line) which were practically smooth, like 

 the Polish parent. Here again, it will be seen, the length of the glume 

 has acted as an inhibitor of pubescence. A study of the colour shows 

 this inhibiting nature even more clearly (Chart, Figs. 2 and 3). With 

 the single exception of one plant of 16 mm. glume length, all the fully 

 coloured individuals are between 8 and 13 mm. — among the shorts, in 

 fact. The proportion is : 



Tinged and Wliite Coloured 



129 49 



Expectation ... 133-5 44-6 



It was impossible to draw a really satisfactory distinction between the 

 heterozygous tinged or faintly coloured, and the colourless, as the faintly 

 tinged individuals were easily confused with stained whites, but the 

 pure blacks were easily classified. Nevertheless Chart, Fig. 3, shows 

 a curve of those individuals which were considered to be tinged. 



The point of interest which attaches to this curve is the distribution 

 of the tinged individuals ; it will be seen (Fig. 3) that they are not quite 

 evenly distributed among those with glume length varying between 

 13 and 20 mm., but that they occur with greater frequency at the 

 short end of the heterozygote cui-ve. The colour seemed to be quite 

 independent of the pubescence. To ascertain whether, among the 



