Reversing the direction of selection should reverse the change--each population would move back towards the mean (and beyond, if the reversed selection were maintained). Since the selection pressure would be similar, the rate of change would be expected to be similar, though the increased rarity of alleles for the opposite phenotype, due to the first 35 generations of selection, might slow the change at first.
Work Step by Step
Selection that works to move the phenotypes directionally one way should move them in the opposite direction once reversed. The rate ought to be similar if the selection regime is similar, just reversed in direction, though the alleles for the opposite phenotype may have become rare, slowly selection for the first few generations since it may be hard to find those alleles now being favored.