The color purple: from royalty to laboratory, with apologies to Malachowski

Abstract

The components of the blood stain, eosin and methylene blue, were introduced by Baeyer and Caro, respectively. Methylene blue was used primarily for detecting Mycobacterium tuberculosis until Ehrlich in 1880 mixed methylene blue with acid fuchsin to produce what he termed a "neutral stain," which allowed differentiation of blood cells. Eight years later, Chęciń ski changed the acidic component of the dye to eosin. Plehn subsequently altered the proportions of eosin and methylene blue to produce a greater range of red and blue hues. In 1891, Malachowski and Romanowsky independently developed stains composed of eosin and "ripened" methylene blue that not only differentiated blood cells, but also demonstrated the nuclei of malarial parasites. A number of "ripening" or "polychroming" techniques were investigated by different groups, but the aqueous dye solutions produced were unstable and precipitated rapidly. Subsequently, methanol was introduced as a solvent for the dye precipitate and techniques were developed that utilized the fixative properties of the methanolic solution prior to aqueous dilution for staining. This avoided the troublesome process of heat fixation of blood films. Giemsa further improved these techniques by using more controlled methods of methylene blue demethylation. In addition, he used measured amounts of known dyes and increased dye stability by adding glycerol to the methanol solvent. With the outbreak of World War I, it became difficult to obtain German dyes outside of Germany; during the World War II, it became impossible. In their effort to improve the inferior American versions of Giemsa's stain, Lillie, Roe, and Wilcox discovered that the best staining results were obtained using pure methylene blue, one of its breakdown products (azure B) and eosin. These three substituents remain the major components of the stain to this day.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 42 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.