The methods of optimal design of experiments are considered for the regression problem when the observations are some averages of a response function over an interval whose length can be controlled. It is shown that the problem is closely related to the classical problem of Markov’s moments