LEAF WIDTH AND SUCROSE. 



Correlation 1. 



The correlations thus far investigated will now be dealt with in succession, 

 comoiencinf^ with that between leaf width and sucrose. It is a well recog- 

 nised fact that, while more or less profound differences in the floral organs arc 

 found necessary for the classification of plants, it is to the minute difi'erenccs 

 in the vegetative organs that the plant breeder looks for indications of character. 

 Small differences in the width of leaves^ otherwise entirely similar, fall und«r 

 the latter category. These are easily marked in any batch of seedlings and, 

 from the fact that the broader leafed forms mors nearly approach the thicker, 

 richer, tropical can3s, thsy have attracted attention from the very beginning. 

 Between 1,100 and 1,200 seedlings have now been measured for leaf width at 

 crop time. In growing plants, as has been stated above, probably the best 

 method of determining the average width of the leaves in any seedling is to 

 observe the general a,ppearance of the plant at a short distance, and then to 

 measure what appears to be an average leaf carefully. This method is called 

 " eye measurement," but, at maturity, it has not been found possible to apply 

 it. The j[)lants have grown too high for accurate j udgment and the leafy shoots 

 are frequently battered by the weather. The following method has accord- 

 ingly been adopted. Ten healthy shoots are cut and laid out. In each of 

 these the broadest leaf is selected and measure:! at its broadest part, and an 

 average is theii struck between the ten measurements obtained. The factor 

 here dealt with is therefore less the average leaf width than the maximum, 

 the greatest leaf width of the seedling at crop time. The method of dividing 

 the seedlings into classes has been described in detail for the Kurnn series 

 and, in the table appended, summaries are recorded for all the batches of 

 seedlings measured. In this and the following correlations the classes are 

 arranged in ascending order, commencing with narrowest, shortest, thinnest, 

 etc., and passing regularly to the widest and so on. Chittan, Karun, Kaludai 

 Boothati, SaretJia and Cheni agree in showing a very distinct negative correla- 

 tioji between leaf width and sucrose in the juice. In the small class of Poovan 

 seedlings, this correlation is reversed, because of the errant position of the 



