BUTLER AND HAFIZ. 165 



in vertically elorigated streaks and, at the end of the 2nd week, 

 entirely surroimdod the node and had extended for an inch or two 

 above and below. After twenty days several inoculated roots were 

 examined at their origin from the stem and large quantities of hyphse 

 were found passing froni the root to the stem, not only along the 

 vascular tissues but also laterally into the stem parenchyma in all 

 directions. Fig. 5 shows the conditions at this stage. The stems 

 were split longitudinally a month after the inoculations and were 

 found entirely infected, the characteristic pith discoloration, with 

 transversely elongated white blotches, being well developed. 



Further inoculations were made on June 3rd, 1913, on the feeding 

 roots of well-established cane plants growing in large culture pots 

 in soil. The soil was carefully removed until the roots were exposed 

 and these were inoculated by sprinkling with a suspension of spores 

 from a pure culture, care being taken not to injure the roots. The 

 soil was then replaced. After sixteen days two of the inoculated 

 roots were examined by sectioning. Both showed a small area of 

 reddening 2 or 3 mm. behind the growing point. Penetration 

 was found to have occurred here (Fig. 4) and the liypha? were 

 extending freely in the tissues of the root. 



Out of many hundreds of canes affected with red rot examined 

 during the past ten years, we have met with a limited number of 

 cases where natural infection had occurred through some part of the 

 stem above ground and where the base of the cane was unaffected. 

 In February, 1910, a Khari cane was examined and was found 

 t(^ show definite symptoms of red rot in the 5th and 6th internodes 

 from the base, the lowest internodes being quite free from reddening 

 or hyphse. At the node between the two infected internodes there 

 was a broken shoot, distinctly reddened in its interior. Character- 

 istic hyphse of CoUetotrickum were found in this shoot and could ])e 

 traced from the broken surface through the bundles into the pith 

 of the stem. There were no other injuries, and infection had doubt- 

 less occurred through the broken shoot. A second cane of the same 

 variety showed a similar case of infection through a broken shoot 

 at the 5th node. Higher up several nodes showed slight infection, 



2 



