ABSTRACTDoughs were sheared at a constant rate and heated in a cone‐and‐plate viscometer. As the temperature increased, the apparent viscosity first decreased, passed a minimum, and then increased rapidly presumably due to the swelling of starch granules. If the dough was heated more rapidly, the viscosity had its minimum value at a higher temperature. This is explained by the swelling of starch granules being a rate process. The present experiments explain why in previous work with heating prior to shearing, the minima in the viscosity versus temperature curves were found at unexpectedly low temperatures. The temperature at which in a standard amylograph experiment the viscosity begins to rise is a fairly good estimate of the temperature at which viscosity has its minimum value during baking. The agreement is due to compensating effects of the rate of heating and of the starch concentrat