VASCULAR DISEASES OF BANANA. 



169 



thoroughly sterilized by i : i ,000 mercuric-chloride water before taking some of the interior, 



with sterilized instruments, for cultures. 



No further opportunity has occurred for the study of this disease, but the writer has no 



doubt (Jan. 1909) that the Cuban banana disease, whatever may be thought of the Jamaican 



one, will be found to be due to Fusarium cubense which is not unlike those described by 



him from cotton, melons, cowpeas, and cabbage. 



Inoculations begun by the writer in July and August 1909 demonstrated the ability of 



the fungus to occupy the vascular bundles of the 

 banana and to penetrate long distances from the 

 point of inoculation (7 feet and more), with produc- 

 tion of the characteristic stain, but the experiment 

 was broken off before secondary signs appeared. 



Fig. 76.' 



F. 77. f 



During the construction of another hot-house the root-stocks remained for several months 

 dry on a bench and when finally planted again they grew without development of secondary 

 signs (4 years). 



*!' ic. 76. Enlarged cross-section of part of a banana fruit-stalk, showing white drops of Bacillus musae oozing 

 from the bundles. The white threads are spirals pulled out of the vessels by a dull knife. Photographed in Washing- 

 ton in 1910 from Trinidad material received from James Birch Rorer. 



fFic. 77. Large leaf of a yellow-fruited banana inoculated by the writer on the petiole December 21, 1910, 

 using needle-pricks and white bacterial slime, from fruit stalk shown in fig. 76. Photographed February 25, 1911, when 

 the whole leaf was shriveling and the vessels of the petiole bundles were filled with bacteria. About i/io natural size. 



