I've done quite a bit of research and testing into solarization and the main thing to consider is that the majority of Man Rays and Lee Miller's would have been as Pinholemaster pointed out, solarized when processing the film, not during the printing. In the bottom image, the model would have been shot against black, which was clear on the negative until the film was flashed with light during the dev. The flash would be enough to fog the clear film but not the midtones and highlights of the body. The film is then processed more, darkening the (once) black background, (clear on neg), which will then print as grey or white depending on the length of the flash. The dividing line between the highlights and flashed shadows remains clear, printing as black. (There is a very long technical explanation of what is happening to form this line, but I haven't got time to go on too much; am probably not technical enough to describe it accurately anyway!) Most of the solarizations I've done have been with Ilford Ortho processed in weak Tmax dev so it can be done under a safelight. (1+9) 2 mins dev, then put the film in a try of water under a spare enlarger, expose and then given a further two minutes. What is amazing is how much re-exposure you can give. A short one will make the background grey, while a long one makes it white. I've watched the half developed film be exposed to 2 minutes of light and the mid tones and highlights are not affected. Naturally lots of testing is required to get the right starting points. The neg is usually a little bit flat but then you can bump up the contrast in printing. However, the main thing to remember is to consider the tones in the original shot so the solarization will work. It's great fun!