Dec. 10, 1917 Breeding Sweet Corn Resistant to the Corn Earworm 571 
the table qualities of sweet varieties in the hope that a sweet variety 
with some degree of immunity would thus be secured. 
Crosses were made between commercial varieties of sweet corn and 
southern varieties of field com. Sweet seeds were selected from the first- 
generation ears, and in the second generation plants with well-covered 
ears were chosen and propagated. The descendants of these plants have 
been found to be much less subject to injury from the corn earworm 
than commercial sweet varieties. 
The earworm resistance was tested in 1915 near San Diego, Cal., and 
in 1916 near Washington, D. C. In both seasons the series of hybrids 
as a whole was found to be less damaged by the corn earworm than 
commercial sweet varieties. There were also pronounced differences in 
the immunity of the progenies derived from the different F 2 generation 
ears. The close agreement between the extent of damage of first and 
second ears of the different progenies is taken to indicate that the con¬ 
stitution of the plant is an important factor in the immunity. An 
effort was made to determine the plant characters which give rise to 
this comparative immunity. 
That the factors concerned in immunity are inherited, and are, 
thus, capable of improvement is indicated by the correlation between 
the extent of damage in related progenies. The correlation between the 
average damage of parent and offspring was 0.66 ±0.09. 
Low damage was found to be significantly correlated with a number 
of morphological characters. For the most part these morphological 
characters were also correlated with one another. Of the characters 
measured, prolongation, or the extent to which the husks exceed the 
ear, was found to be the most closely correlated with low damage. The 
interprogeny correlation between prolongation and damage was —0.71, 
with a regression of 1.02 per cent in damage for each centimeter of pro¬ 
longation. Within the progenies the average correlation was —0.26 
and the regression 0.72. This difference between the inter- and intra¬ 
progeny regression is believed to indicate that the protection is in part 
due to other characters correlated with prolongation and not included 
in those measured. 
The thickness of the covering provided the ear by the 'husks was 
also found to be associated with low damage, but since only 5 per cent 
of the larvae that reach the ear do so by penetrating the husks here 
again the relation can hardly be a direct one of cause and effect. From 
the standpoint of worm resistance husk leaves are also shown to be an 
undesirable character. 
By recording the number of larvae in each ear it is possible to resolve 
the total damage into (1) number of larvae and (2) the average amount 
of injury done by each larva. 
