COBB S DISEASE OF SUGAR-CANE. 



53 



writer's inoculated plants some of these closed cavities contained as much as a teaspoonful 

 of the yellow slime (plate 4, and fig. 4). The organism also breaks out of the foliar bundles, 

 especially when the leaves are immature, occupies the intercellular spaces very generally, 

 making passageways between the cells, and comes to the inner surface of the leaf-sheaths 

 and on the softer inner sheaths to the outer surface also, as a copious slime. Whether it 

 exudes solely through stomata, or also sometimes makes openings of its own, is not known. 

 I have found it oozing from the stomata on leaf-sheaths so commonly as to make it unneces- 

 sary to consider any other method of reaching the surface (figs. 5, 6, 7). 



Figure 21 shows in cross-section 

 an early stage in the occupation of 

 the bundle and figs. 22, 23, in longi- 

 tudinal section, show later stages, 

 with much disorganization of the 

 tissues of the bundle, all three taken 

 from inoculated plant No. 6, at the 

 end of 3 months. When we reflect 

 that the bacteria were not intro- 

 duced into this stem directly, but 

 that the plant was inoculated solely 

 by means of a few needle-pricks on 

 the blades of two leaves, and con- 

 sequently that the bacteria had to 

 travel or grow through the vascular 

 bundles a distance of 4 or 5 feet to 

 reach the tissues here figured, the 

 short time required for the general 

 infection and the enormous multi- 

 plication of the organism are aston- 

 ishing. For a longitudinal section 

 through a slightly diseased cane 

 see fig. 24. 



The red stain, as already men- 

 tioned, is believed to be an oxidation 

 product closely connected with the 

 presence of the bacteria. Its greater 

 abundance in the bundles occurring 

 in the nodes is accounted for by the 

 falling away of the diseased leaves 

 and the entrance of air in larger 

 amounts than would be possible on 

 the falling away of normal leaves, 

 the diseased leaf- traces being unable 

 to heal over so as to exclude air. 

 That the stain should also occur in 

 much greater amount in the inter- 

 nodal bundles immediately below the nodes than in those immediately above the nodes is 

 accounted for by the downward movement of the leaf-traces as they enter the stem. For 

 evidence of the close connection between the red stain and the presence of the bacteria 

 consult figs. 12, 13, 25, and various statements under Etiology. 



Fig. 23.* 



*Fic. 23. Longitudinal section showing the xylem part of a bundle wholly destroyed by Baft, vascularum. From 

 plant No. 6. Slide 310 (13, left-hand section. For a detail see fig. 27. 



