When there is no interaction among population means in two-way design, the numerator of F ratio for interaction in data from an experiment expected to be:
a. About the same as MSw
b. Zero
c. About the same as the sum of the numerators for the two main effects
d. Dependent on the number of cells in the design