A OSU SP a ee 
ALLARD: MOSAIC DISEASE OF TOBACCO 451 
inoculation. Her results on this point, however, are somewhat 
open to question, since from the description of the plants one can 
not be sure that she worked with the true mosaic disease of tobacco. 
Jensen (63) (1903-1911) reports his investigation of the mosaic 
disease of tobacco in the Dutch East Indies. A number of mosaic- 
free plants were selected in the field, and the progenies of these 
kept separate each year. By continuous selection of mosaic-free 
plants within these lines from 1903 to 1907, Jensen states that 
there was a very noticeable increase each year in the percentage 
of mosaic-free plants. He was led to believe that careful selection 
afforded an efficient means of decreasing mosaic by esis 
strains naturally resistant to mosaic infection. 
Jensen describes and gives an illustration of a most remarkable 
tobacco plant found in 1899. With the exception of a single basal 
leaf which had developed to normal size, the growth of the leaves 
was limited to the development of the midribs alone. The plant 
was selfed but the entire progeny grown in 1910 was normal. It 
is very probable that this abnormal plant was an exceptionally 
severe form of mosaic. 
Experiments were carried on to determine if excessive quanti- — 
ties of plant food had any influence on the susceptibility of plants 
to mosaic. It was found, however, that mosaic occurred quite 
as generally in the fertilized as in the non-fertilized plants. 
Jensen (1911) states that attempts to develop mosaic-resistant 
lines from plants escaping mosaic infection in the field resulted in 
failure. At Kebon-Aroem, where mosaic is very prevalent, 93 
healthy plants selected from a badly mosaic field were inoculated 
with the filtered sap of mosaic plants. Not a single plant escaped 
infection. The conclusion was finally reached that mosaic-free 
plants selected in a field, must be inoculated with the virus of 
mosaic before their resistance to mosaic can be determined. 
In 1912 the writer (59) published briefly in Science additional 
facts concerning the mosaic disease of tobacco. It was shewn 
that the disease is communicable to practically all genera of the 
solanaceous family, that the disease is dependent upon specific 
infection, and that simply cutting back plants does not produce the 
malady. It was also shown that particular kiads of aphides may 
be active disseminators of the disease. 
