This articles focuses on the very difficult but increasingly important problem of measuring nucleation rates as opposed to only the onset conditions or limits of metastability. Past efforts are reviewed, and recent progress is described including a dramatic new advance consisting of a technique developed by the authors in recent years. This is a nucleation pulse method for experimental study of homogeneous nucleation in vapors. We highlight the specific features of the experimental setup and procedure allowing time-resolved measurements of nucleation rates J in the range 10(5) < J/cm(-3) s(-1) < 10(9) as functions of supersaturation S and temperature T. Homogeneous nucleation is induced in an expansion chamber by a single pressure pulse. Pressure, temperature, and vapor supersaturation are maintained uniform and practically constant during the nucleation period of about 1 ms defining the nucleation pulse. The number concentration of nucleated droplets as well as the condensational growth following the nucleation pulse is monitored by light scattering, permitting one to discriminate heterogeneously from homogeneously nucleated droplets. We illustrate the proper function of the technique with new measurements of homogeneous nucleation rates for 1-butanol from the vapor phase in the temperature range 225 < T/K < 265. The growth of the 1-butanol droplets formed was observed to be comparatively slow, ensuring well-defined constant conditions during the nucleation pulse. Furthermore, at the temperatures considered the equilibrium vapor pressure of 1-butanol is sufficiently high to permit its precise experimental determination. Since from past measurements we also know the surface tension of 1-butanol with sufficient accuracy, a quantitative comparison of the experimentally obtained nucleation rates with the predictions of theoretical models can be performed. The new feature for expansion chambers is that the individual J-S curves are obtained at selectable constant temperatures. From the slopes of these isothermal J-S curves, the molecular content of the critical clusters can be determined without reference to any specific nucleation theory. This allows a direct experimental test of the Gibbs-Thomson (or Kelvin) equation for very small droplets.