14 



BULLETIN 137, U. S. DEPARTMENT OF AGRICULTURE. 



WIDTH AND LENGTH OF LEAVES. 



Any study of leaf dimensions must be statistical and therefore 

 difficult to report briefly. The obstacles to the use of such measure- 

 ments are twofold: The leaf varies with its nourishment and with 

 its exposure, and it is often damaged by the wind. In a study of 

 mature plants, the second leaf from the top being used in all cases, 

 the normal variation was found to be considerable. For instance, 

 at the same place in the same season the leaves of border plants were 

 from 1 to 2 mm. greater in width than those from the interior of the 

 plat, and the length of the leaves of such plants was from 2 to 3 cm. 

 greater. In Princess, one of the least variable varieties, the average 

 size of the leaves of the border plants was 13.7 mm. by 24 cm., and 

 of the interior plants 12.7 mm. by 23 cm. 



To be usable in breeding, a note must be reasonably easy to obtain. 

 To test the usefulness of this character, the first 25 of the 100 meas- 

 urements of each selection were tabulated, as shown in Table III. 

 With width of leaf, the experimental error is small, as width can be 

 determined quite accurately and the broadest part of the leaf is 

 seldom damaged. If the figures, then, are conclusive mathematically, 

 the method is practical. The probable error in the 25 measurements of 

 Princess is ±1.2. It thus fails to separate this variety dependably 

 from Kitzing and Proskowetz, its nearest relatives, or from the selec- 

 tion of deficiens, or Odessa. (See Table III.) .From the rest, how- 

 ever, the separation is clear enough to be significant. With the two 

 selections of Oderbrucker, the separation is sufficient to establish a 

 difference. In this case the two are closely related and the note 

 becomes serviceable. As a rule, the width of leaf is seldom a suffi- 

 cient basis for separation in closely related strains. Fortunately, 

 such differences are seldom unaccompanied by other points of va- 

 riance, and it is often the sum of several differences that serves to 

 distinguish individual strains. 



Table III. — Greatest, least, and average ividth and length of 25 leaves in each 

 of 13 selections of barley grown at St. Paul, Minn., in 1911. 



Pedigreed selection from— 



Leaf width. 



1 



,eaf length 





Greatest. 



Least. 



Average. 



Greatest. 



Least. 



Average. 





Mm. 

 15.0 

 15.5 

 16.0 

 17.5 

 20.0 

 22.0 

 15.5 

 20.0 

 20.0 

 20.0 

 15.0 

 22.0 

 16.0 



Mm. 

 12.5 

 11.0 

 12.5 

 14.0 

 14.0 

 15.0 

 12.5 

 16.5 

 16.5 

 15.5 

 11.0 

 17.0 

 11.0 



Mm. 

 13.2 

 12.7 

 13.7 

 15.5 

 16.7 

 IS. 7 

 14.3 

 18.5 

 18.3 

 17.8 

 13.7 

 18.7 

 13.0 



Cm. 

 28 

 28 

 32 

 23 

 27 

 28 

 26 

 26 

 26 

 25 

 22 

 25 

 28 



Cm. 

 20.0 

 20.0 

 26.0 

 17.0 

 18.0 

 20.0 

 18.0 

 18.0 

 19.5 

 20.0 

 14.0 

 20.0 

 23.0 



Cm,. 

 23.5 





23.7 





28.7 





19.2 





22.8 





24.3 





22.5 





22.6 





22.9 





22.8 





17.9 





22.0 





25.5 







