RELATION OF LOOSE SMUT TO YIELD OF BARLEY

Abstract

A simple technique was devised to inoculate barley with Ustilago nuda (Jens.) K. and S. in order to facilitate the study of the effect of loose smut on yield. Infected seed lots so obtained were mixed in different proportions with normal seed to provide at least six levels of infection ranging from nearly zero to the maximum. The infection levels were planted in replicated field test plots at three stations in each of two years.No significant correlations were obtained between yield of barley and the percentage loose smut in the 1939 trials. A direct linear relationship was obtained in 1940 between percentage incidence of loose smut and the reduction in yield. For every 1% increase in loose smut, yield was reduced 0.85% at St. Paul, Minn., 1.4% at Edmonton, and 1.2% at Fallis, Alta. These regression coefficients are not significantly different. Loose smut did not affect tillering noticeably. Similar infections of loose smut were obtained at all stations for comparable seed lots. Higher infections were observed as the rate of seeding was increased. The incidence of loose smut at Edmonton was the same on the basis of percentage smutted spikes as on the basis of percentage smutted plants.