It is the purpose of this note to show that serum reactions may offer considerable aid in the differentiation of various species and sub-species of yeasts—a tedious and difficult task by current methods. Schütze (1) in 1902 found that anti-yeast sera in rather high concentrations caused non-specific precipitation when mixed with press juice: agglutination reactions, as carried out by him were also found to be non-specific. The tendency of many yeasts to flocculate spontaneously led me to abandon work on agglutination phenomena. But in repeating Schütze's precipitin experiments, it was found that while some yeasts, notably S. logos, yielded non-specific sera, others, particularly if the dilutions were carried quite far, gave nicely specific results. Ten authentic cultures were used, five of them races, or subspecies, of S. cervisiae, and therefore very closely related. These were first kept for nine months on the same medium, a cane molasses phosphate agar, grown aerobically with monthly transplants.