DISTINCTIONS IN CULTIVATED BARLEYS. 



15 



In length of leaf, the method is much less promising. Not only 

 is the probable error greater, but the measurement is unsatisfactory. 

 The leaves become so broken by whipping in the wind that speci- 

 mens which are entire at the tip are seldom found. An effort was 



S30 



S 6 7 <3 & Sb // /2 /3 



Fig. 8. — Composite curve showinj 



the width of leaves in millimeters in eight selections of 

 barley. 



made to overcome this difficulty by choosing an earlier stage of 

 development and thus utilizing the better protected leaves nearer 

 the ground. Although the extreme tendencies were not yet devel- 

 oped, the second leaf from the seedling was found to offer fewer 

 experimental difficulties. Such leaves were entire and the length 



/30 

 /20 

 //O 



too 



so 

 ao 



70 



60 



so 



40 

 30 

 SO 

 'O 



o 































































































































































































































































































































































































































































































/O // JS /3 /* 



.ae sa se ss s6 S7 



Pig. 9. — Composite curve showiDg the length of leaves in centimeters in six selections of 



barley. 



measurements accurate, but even then the width was much less 

 variable than the length. All measurements, consisting of 100 leaves 

 of each strain, showed a sharp curve in width, but a flat one in 

 length, the latter sometimes having two summits. Composite curves 

 are shown in figures 8 and 9. 



