Spinal cord of cat and rabbit was stained, after experimental lesions, by variations of Marchi's method. The following conclusions were drawn: 1. The presence of an oxidizing agent (K2Cr2O7, NaIO3, or KCIO3) in the osmic acid solution is of primary importance and a preliminary oxidation in Mueller's fluid is unnecessary or even detrimental. 2. Acetic acid added to Marchi's fluid, accentuates the action of the oxidizing agent in restraining the staining of normal myelin. 3. Too high concentration of oxidizing agent or of acid may inhibit staining of degenerating myelin. 4. Marchi's and Busch's methods have been modified as follows: Fix one day in 10% formalin and transfer without washing to the staining mixture, either A or B. Staining mixture A: Marchi's fluid plus 1 to 3% glacial acetic acid. B: An aqueous solution containing KCIO3 0.25%, osmic acid 0.33%, and acetic acid 1%. Stain about one week. These methods worked on spinal cord and medulla, but cannot be recommended for brain. 5. The detrimental effects of long post mortem autolysis or of prolonged fixation in formalin may be counteracted to some degree by increasing the concentration of the acid in Marchi's fluid up to 5% or of the KCIO3 up to 0.4% in the modified Busch's fluid.