Following experimental lesions, spinal cords of cats and rabbits were fixed with acid, neutral, and alkaline solutions. Staining was limited to a chromate-osmic (Marchi's) solution and a chlorate-osmic solution. The following conclusions were drawn: The presence of an acid in the fixative caused normal myelin sheaths to stain. This effect was reduced by washing tissues before staining, by adding acetic acid to the stain, or by employing a non-formalin fixative. It was, however, at no time entirely obviated. A study was made of the granular deposits which occur in nearly all Marchi preparations and which are especially confusing if very light backgrounds are obtained. The staining reactions of the granular deposits were very similar to those of degenerating myelin but some suppression of the granules was obtained by adding KCIO3 to the formalin fixative.