IN A SERIES1 of experiments it has been possible to demonstrate that the dopa reaction, which indicates the pigment-forming ability of melanocytes, may be clearly divided into two phases: (1) the spontaneous auto-oxidation of dopa solutions in a slightly alkaline medium without the action of an enzyme, and (2) the actual dopa reaction, which also takes place even more clearly and selectively in an acid medium. It is known also that the dopa reaction occurs, nonspecifically, in the cells of the myeloid granular series (leucocytes, myelocytes). The two reactions which may be obtained with dopa, the one with melanocytes and the other with granular leucocytes, show certain essential differences. The reaction in melanocytes varies according to the activity of the cell; it is relatively slow, inhibited by cyanides. It is apparently specific for dopa and for oxytyramine. It should be noted that Fitzpatrick and others2 have obtained