Mar. is, 1924 Wheat Scab , Corn Rootrot , and Crop Successions 869 
cases where practicable, however, sowing wheat after corn should be 
avoided. 
The data discussed pertain solely to scab or head blight, as only this 
phase of the disease can be seen and accurately estimated. No account 
has been taken of seedling blight and weak plants due to the parasite in 
the soil attacking wheat seedlings. Investigations by Dickson (4) and 
Johnson and Dickson (13) show the importance of seedling blight due to 
soil infestation with Gibberella saubinetii. Therefore, the economic im¬ 
portance of not bringing com and wheat together in a rotation is much 
greater than data on wheat scab (head blight) alone would indicate. 
However, it is nearly impossible to get an accurate estimate on seedling 
blight in a commercial field, and, therefore, this phase has not been con¬ 
sidered in this paper. 
CORN ROOTROT IN RELATION TO WHEAT-SCAB 
The losses resulting from com rootrot caused by Gibberella saubinetii 
can not be estimated as easily as can those resulting from wheat scab. 
It can be done, however, more advantageously on carefully controlled 
experimental plats than by an extensive survey of commercial fields. 
Fortunately, com disease investigations have progressed sufficiently (12 , 5 ) 
so that small quantities of com that are highly resistant to rootrot 
under central Illinois conditions can be obtained. This resistant com 
was used as a check on the ordinary susceptible com in all the experi¬ 
ments reported below. 
Comparative yields were considered the best index to the extent of 
rootrot. Extensive data by Holbert et al. ( 12 , 5 ) show that yields are 
inversely correlated with the amount of disease found on the germinator 
and with corn rot produced by pure-culture inoculations. The latter is 
also shown in Table X. The disease-resistant and disease-susceptible 
selections, between which comparisons were made, were both selected 
from the same strain of Yellow Dent com according to the methods 
described by Holbert et al.* The yields of two fields planted with the 
same kind of com but with different previous crops can not be compared 
directly because there doubtless would be a difference in soil fertility. 
But by comparing the yield from disease-susceptible com with the yield 
from disease-resistant com, in different crop sequences, a fairly accurate 
determination can be made. The experimental series reported below 
were conducted on adjacent fields in each case, and on similar soil of 
the same elevation and drainage. 
In 1920 an experiment was conducted to compare (1) the yields of corn 
grown on plats which had been cropped to clover during the two previous 
years and to oats in 1917 and (2) the yields from adjacent plats which 
had a crop of badly scabbed spring wheat on them in 1919, com in 1918, 
and oats in 1917. A diagram of the plats is given in figure 2. Each 
plat was 4 rows wide and 10 hills long. The hills were 42 inches apart 
each way and were planted by hand at the rate of 3 kernels per hill. 
The data as given in Tables VI and VII were obtained by omitting the 
outside hills of each plat. Thus each plat contained 2 rows of 8 hills 
each, on which data were taken, a total of 16 hills with a perfect stand 
of 48 plants. 
As shown in Tables VI and VII, the yields from disease-resistant com 
were very different on the two series of plats with different previous 
6 Holbert, J. R., and others, op. err. 
