Solidification processing has seen strong interaction between theory and practice, benefiting both fields. Solving important problems requires conceptualizing on vastly different size scales and extensive experimentation to justify theoretical approximations. Areas where theory and practice have advanced together include dendrite arm spacing, grain size control, columnar structures, eutectic-like in-situ composites, inclusion formation, macrosegregation, and non-dendritic structures. This lecture aims to convey three ideas about solidification processing. First, theory and practice have effectively interacted in recent years, leading to new industrial processes and solving industrial problems. At the same time, critical industrial problems have stimulated fundamental research that has changed our understanding of the field. Second, solving important practical problems requires conceptualizing on vastly different size scales, which is not as easy as it might seem. Processes at the liquid-solid interface must be visualized at the Angstrom level, while inclusions are at the micron scale, dendrite arm spacings are fractions of millimeters, and castings or ingots are on the scale of meters. Understanding structural features such as surface finish, shrinkage, and segregation requires understanding processes at the microscopic level. Third, many practical and fundamental problems in the field are too complex to be answered quantitatively with great rigor at present. Approximations and assumptions must be made before quantitative treatment is possible, and it is vital that experimental work be carried out to test the range of validity of these approximations. Fig. 1 shows solidification structures in binary alloys against a flat mold wall. The most common structure is equiaxed dendritic, with new dendrites forming and growing in the liquid-solid zone. Other structures include aligned columnar dendritic growth and spheroidal growth centers. At higher values of G/R, cellular and plane front solidification structures are obtained. If the alloy is single phase after equilibrium solidification, the resulting solid is a single crystal; if two phase, it is a two-phase in-situ composite. Another structure, amorphous or glassy, can be obtained at sufficiently high cooling rates. Fig. 2 shows dendritic structures at lower magnification, with each grain composed of tens or tens of thousands of dendrite arms. Fig. 3 shows shrinkage and macrosegregation in a killed steel ingot. Fig. 4 shows inclusions, such as oxysulfides and silica or alumina, that are entrapped or pushed by growing dendrites.Solidification processing has seen strong interaction between theory and practice, benefiting both fields. Solving important problems requires conceptualizing on vastly different size scales and extensive experimentation to justify theoretical approximations. Areas where theory and practice have advanced together include dendrite arm spacing, grain size control, columnar structures, eutectic-like in-situ composites, inclusion formation, macrosegregation, and non-dendritic structures. This lecture aims to convey three ideas about solidification processing. First, theory and practice have effectively interacted in recent years, leading to new industrial processes and solving industrial problems. At the same time, critical industrial problems have stimulated fundamental research that has changed our understanding of the field. Second, solving important practical problems requires conceptualizing on vastly different size scales, which is not as easy as it might seem. Processes at the liquid-solid interface must be visualized at the Angstrom level, while inclusions are at the micron scale, dendrite arm spacings are fractions of millimeters, and castings or ingots are on the scale of meters. Understanding structural features such as surface finish, shrinkage, and segregation requires understanding processes at the microscopic level. Third, many practical and fundamental problems in the field are too complex to be answered quantitatively with great rigor at present. Approximations and assumptions must be made before quantitative treatment is possible, and it is vital that experimental work be carried out to test the range of validity of these approximations. Fig. 1 shows solidification structures in binary alloys against a flat mold wall. The most common structure is equiaxed dendritic, with new dendrites forming and growing in the liquid-solid zone. Other structures include aligned columnar dendritic growth and spheroidal growth centers. At higher values of G/R, cellular and plane front solidification structures are obtained. If the alloy is single phase after equilibrium solidification, the resulting solid is a single crystal; if two phase, it is a two-phase in-situ composite. Another structure, amorphous or glassy, can be obtained at sufficiently high cooling rates. Fig. 2 shows dendritic structures at lower magnification, with each grain composed of tens or tens of thousands of dendrite arms. Fig. 3 shows shrinkage and macrosegregation in a killed steel ingot. Fig. 4 shows inclusions, such as oxysulfides and silica or alumina, that are entrapped or pushed by growing dendrites.