BUTLER AND H^VFIZ. 153 



origin occurred. On May 30th the number of sound shoots was 

 counted in the trenches numbered 3 to 18, of which the odd numbers 

 were from healthy seed and the even from diseased, the result being 

 679 and 117 respectively. The crop was lost by a severe flood in 

 August, so that the final result cannot be given. 



In 1907, the experiment was repeated in a field not subject to 

 flooding, ten trenches being sown on March 7th with the varieties 

 Striped Mauritius and Red Mauritius. By the end of April the 

 results were as striking as in the previous year, but in May 

 many shoots withered in the trenches planted with healthy seed 

 of the Red Mauritius variety. This variety was taken from a 

 diseased field and was so generally infected that the ratoons nearly 

 all died out. It is probable that many of the apparently healthy 

 canes contained Colletotrichum, for, as will be shown below, while 

 reddening of the pith is a sure indication of disease, unless the canes 

 have been mechanically injured, absence of reddening does not 

 always imply freedom from it. 



In 1908, the cane selected was again Red Mauritius, which was 

 planted on March 6th. Germination was good in all the trenches. 

 The condition of four rows, Nos. 12 to 15, on May 30th, is graphically 

 shown below, the small circles each representing a sound shoot. 

 Rows 12 and 14 were from diseased seed, rows 13 and 15 from healthy. 



Besides these comparative experiments, the main crop of cane 

 grown on the Pusa Farm has been yearly supervised, so that setts 

 sho^\^ng red marks in the pith are not planted. The result has been 

 that, excepting the season 1907-08, which will be separately con- 



