26o 
Journal of Agricultural Research voi. xxiv, No. 3 
to 30 per cent of diseased plants. Fields of the so-called Texas-Ribbon 
variety (probably Gooseneck) contained from 0.5 to 15 per cent of af¬ 
fected plants, or uniformly less than the first-named sort under the same 
conditions. No infection was observed in a field of Early Amber, but 
this variety has proved to be susceptible under greenhouse conditions at 
Washington. A number of grain-sorghum varieties recently introduced 
from Africa were planted in a greenhouse at Washington exposed to 
natural infection, but they have showed no signs of the disease. About 
50 large stools of mosaic sugar cane were present in this house, and 
Aphis maidis was abundant. 
During the summer of 1921 a patch of pearl millet {Pennisetum glaucum ), 
sold under the name Georgia cat-tail millet, was planted for soiling pur¬ 
poses at the Sirup Experiment Station, Cairo, Ga. On July 15 this 
planting was noticed to be severely attacked by mosaic. More than 50 
per cent of the plants were affected and were noticeably smaller than 
healthy plants. This species also had been previously proved to be sus¬ 
ceptible to grass mosaic under controlled conditions at Washington. 
(See Table II.) 
FIELD OBSERVATIONS ON MOSAIC OF WILD GRASSES 
Collections of wild grasses affected with mosaic have been made in 
Louisiana, Georgia, and Florida. In all cases the plants were found in 
or near affected sugar-cane fields. That the mosaic appearing naturally 
in the field in these species is identical with sugar-cane mosaic has been 
verified experimentally under controlled conditions. (See Tables I and 
II.) A list of mosaic grasses collected with dates and locahties is given 
in Table VIII. 
Tabl^ VIII .—Collections of wild grasses affected with mosaic 
Species and locality. 
Date of collection. 
Species and locality. 
Date of collection. 
Syntherisma sanguinalis: 
Cairo, Ga. 
Sept II, 1919 
Oct 15,1919 
Sept 6,1920 
Sept 1,1920 
Paspalum boscianum: 
Cairo, Ga. 
Sept. II, 1919 
Aug. 16,1920 
July 27,1920 
July 29,1920 
New Orleans, La. 
Do. 
Marianna, Fla. 
Cairo, Ga. 
Chattahoochee, Fla.. 
Reno, Ga. 
Plaquemine, La. 
Mananna, Fla. 
Aug. 21,1920 
Aug. 14,1920 
Chaetochloa magna: 
Dade County, Fla... 
Dec. II, 1920 
Reno, Ga. 
July 31,1920 
Brachiaria platyphylla: 
New Orleans, La. 
Do. 
Aug. 11,1920 
Sept. 6,1920 
No complete record of individual observations on mosaic of wild 
grasses has been kept. The foregoing list is sufficient to show that mo¬ 
saic of these species is common near affected sugar cane under natural 
conditions. 
EXPERIMENTS ON SEED TRANSMISSION OF GRASS MOSAIC 
It seems appropriate to include in the present paper a brief account of 
experiments to determine whether the disease is transmitted by means of 
seeds, because of the relation of such transmission to the infection of one 
species with inoculum from a different species. In regions where the 
vegetative parts of annual grasses are completely killed during winter. 
