I'm somewhat amazed by how many varied techniques there are for rotary shadowing, and how few seem to agree with what works for us. So, here is our method.

We shadow biological molecules from the connective tissue matrix, usually ranging in size from 16 to 300 kilodaltons. Many are linear but some are globular. We spray the molecules in solution with 70% glycerol. The other 30% is 100 microgram/mL of protein, preferably in a volatile buffer such as 1% acetic acid or 0.1M ammonium bicarbonate, pH 7.8. Other buffers can be used, but salt crystals can be a big problem.