In the first version of this paper (arXiv: 1003.2899v1, 15.03.2010) there is described first, traditional method of measuring the non-zero shift of interference fringe in the Michelson interferometer, operating on the effects of second order with respect to v/c, and are revealed hidden causes of the failure to measure the shift of interference fringe in the period from 1881 till 1960. It is shown that at the latitude of Obninsk within a 24-hour observation period the horizontal projection of aether wind velocity varies from 140 km/s to 480 km/s. The second version of this paper (arXiv: 1003.2899v2, 15.04.2010) is supplemented with a second method of finding the velocity of the aether wind -- through measuring the largest seasonal decrease in the ratio of the summer shift of the interference fringe to the winter one (equaled ~12%). It gave the same interval of values of the projections of the aether wind velocity as the first method. More than hundred years there persists a belief that Michelson-type interferometer can not be adjusted such as to detect effects of the first order with respect to v/c. Below I show that it is possible, and more successfully than on the interferometer of the second order. In contrast to the traditional approach, in the interferometer of the first order the light after splitting on a semi-transparent plate propagates in both arms to the reflecting mirrors in one optical medium, and returns after reflection from the mirrors through another optical medium. The shift of the interference fringe was found to be proportional to the difference of dielectric permittivities of these media. The horizontal projection of the Earth's velocity relative to luminiferous aether measured by the first order interferometer agrees with that found by the first two methods.