DISEASES OF SUGAR CANE 327 



Causation. 



No fungus or bacterium has been found to have any causative 

 connection with the disease. The evidence points to its associa- 

 tion with the affections caused by infective viruses, the existence 

 of which is estabhshed in the mosaic diseases of tobacco, tomato, 

 cucumber, bean, potato and other plants, and is strongly in- 

 dicated in diseases such as sereh, potato leaf-roll, and curly- 

 top of beet. (See Chapter III.) 



Communicability. 



Cuttings from affected plants invariably reproduce the disease, 

 and it is in this way that it has been distributed about the world. 

 The infection of healthy strains brought within the range of the 

 disease shows clearly that it is in some way transferable. Contact 

 alone does not produce this result, nor is there any evidence 

 that the infective principle resides in the soil. 



Healthy cuttings planted in fields or in pots from which badly 

 diseased plants have been recently removed are not infected. 

 The spread of the disease by infection is observed to be very 

 erratic, rapid at one time or place, slow or absent at another. 

 The evidence points to distribution by a carrier, probably an 

 insect. E. W. Brandes, in greenhouse experiments at Washing- 

 ton, has proved transmission of the disease from sorghmn to 

 cane and corn by Aphis maidis. Extensive experiments in Porto 

 Rico with all the common insects found on sugar-cane have, so 

 far as reported, had only inconclusive results, as the infections 

 obtained (with cane-fly, leaf-scale, yellow aphis, and mealy-bug) 

 were so few that chance infection is not certainly excluded from 

 their origin. No infections were secured in experiments made by 

 C. B. Williams, in Trinidad, with several insects, including the 

 frog-hopper (Tomaspis). Cane-fly and aphides are exceedingly 

 rare on sugar-cane in Trinidad. In that island there is little or 

 no evidence of infection in dry weather, but rapid extension of 

 the disease becomes apparent a month or two from the beginning 

 of the rainy season. 



Artificial infection tested in many ways has only proved 

 successful when juice expressed from affected plants, with pre- 

 cautions against oxidation, has been directly inoculated into the 

 tips of growing canes, and with this method positive results have 

 been difficult to obtain. 



Resistance and Susceptibility. 



In Porto Rico all the stcindard varieties, and all the local, 

 Barbados, and Demerara seedlings are attacked. The Bourbon 

 cane is exceedingly susceptible and the Striped (Rayada), White 

 Transparent (Crystalina), and YeUow Caledonia, all of which 

 are grown in quantity in the iMand, are severely injured. Certain 

 seedlings imported from and much favoured in Java (P.O.J. 



