Difficulty had been encountered in the determination of magnesium in ductile iron by point to plane methods because tiny eruptions in the burn caused irregularities and non-reproducible results on the vacuum spectrometer. Investigation showed that magnesium sulfide, formed from treatment magnesium and residual sulfur, floating slowly upward, was the probable cause. A teaspout ladle was used and, after pouring the molten iron into a can-like hand ladle lined with fireclay for the sample, gave iron free of these defects.
John R. Boyd, "Point-to-Plane Determination of Magnesium in Ductile Iron," Appl. Spectrosc. 24, 103-105 (1970)