we can arrange things so that we test equal numbers of treatment A and treatment B subjects before and after lunch. Many physical variables will be balanced across conditions automatically as we assign our subjects to treatment conditions at random. Time of testing, weather conditions, and day of week are typically controlled in the way. Usually, we do not even think about these sorts of extraneous variables. As long as there is no systematic change in an extraneous variable, thing are fine. If we assign subjects to treatment conditions at random, we can be reasonably sure that we will not accidentally test all subjects receiving one treatment on a cool, comfortable day and all subjects receiving a different treatment on a hot, muggy day. We can, of course, easily improve control by using block randomization.