Combinations of fresh liquid microfiltration retentate of skim milk, ultrafiltered retentate and permeate produced from microfiltration permeate, cream, and dried lactose monohydrate were used to produce a matrix of 20 milks. The milks contained 5 levels of casein as a percentage of true protein of about 5, 25, 50, 75, and 80% and 4 levels of true protein of 3.0, 3.76, 4.34, and 5.0% with constant lactose percentage of 5%. The experiment was replicated twice and repeated for both 1 and 2% fat content. Hunter color measurements, relative viscosity, and fat globule size distribution were measured, and a trained panel documented appearance and texture attributes on all milks. Overall, casein as a percentage of true protein had stronger effects than level of true protein on Hunter L, a, b values, relative viscosity, and fat globule size when using fresh liquid micellar casein concentrates and milk serum protein concentrates produced by a combination of microfiltration and ultrafiltration. As casein as a percentage of true protein increased, the milks became more white (higher L value), less green (lower negative a value), and less yellow (lower b value). Relative viscosity increased and d(0.9) generally decreased with increasing casein as a percentage of true protein. Panelists perceived milks with increasing casein as a percentage of true protein as more white, more opaque, and less yellow. Panelists were able to detect increased throat cling and mouthcoating with increased casein as a percentage of true protein in 2% milks, even when differences in appearance among milks were masked.