1 86 



AMERICAN JOURNAL OF BOTANY 



[Vol. 9. 



disposition due to a too vigorous growth in early life. Biffen (1912), in his 

 studies on the inheritance in wheat of rust resistance to Puccinia ghimarum, 

 found that the rust is most virulent when a complete fertilizer is used, and 

 that the virulence of the disease decreases with a decrease in the amount 

 of fertilizer. Comparing the two principal types of asparagus soils in 

 California, the sediment and the peat, Smith (1905, p. 56) notes that aspar- 

 agus growing in the latter soil is considerably more damaged by the same 

 amount of disease. He comments that on peat formations, composed 

 almost entirely of vegetable matter and water, a very luxuriant, quick- 

 growing, tender, and succulent asparagus is produced. 



Zavitz (1913) reports some very interesting observations on the relative 

 susceptibility to rust of oats grown under conditions of varying thickness 

 of seeding. The experiment was conducted through each of four years, 

 using both large and small seed of heavy-stooling, medium-stooling, and 

 light-stooling varieties of oats, and planting the seed of each variety in 

 squares one, two, three, four, six, eight, and twelve inches apart. Table I 

 is adapted from the data presented by Zavitz, and presents the average 

 results of thirty-two tests made by planting oats at seven different distances 

 apart. The results are the averages for four years. 



TABLE i 



The greater amount of rust observed with the increased distance between 

 plants is best correlated with the increased luxuriance of growth exhibited 

 by these plants. The difference of a week in the time of maturing between 

 the most closely spaced and the most liberally spaced oat plants is hardly 

 sufficient to account for the difference in the amount of rust infection. 

 Observations were made at frequent intervals through the summer, and 

 a rust difference due only to difference in time of maturity would not have 

 shown up in this fashion in the data. A more logical explanation is the 

 increase in the amount of lodging which closely parallels the increase in 

 percentage of rust from the one-inch spacing to the six-inch spacing. But 

 from the six-inch spacing to the twelve-inch spacing the amount of lodging 

 decreases appreciably while the percentage of rust increases further, indicat- 

 ing that the increase in the amount of rust is independent of lodging. There 



