Aflatoxin G, was originally identified as a compound emitting fluorescence under ultraviolet irradiation, and this emission has been widely used for identifying the toxin on thin-layer chromatographic (TLC) platesl. However, LIJLNSKV AND BJTLER~ recently reported that G1 is a blue fluorescent compound and that the green fluorescence is due to a yellow impurity superimposed on the blue G, spot. Furthermore, they demonstrated that blue G1 is toxic to ducklings. Moreover, ultraviolet adsorption datas-s available for G, toxin are inconsistent. The calculated ratios of absorbance (R) at 363 m,u to that at 265 rnp of G, from the reported data in methanol or in ethanol fluctuated from r.61 to 1.87. Since the 363 rnp band is the excitation band396 responsible for the fluorescence of the toxin, any substance absorbing in this ultraviolet region should change the fluorescence property. In view of these conflicting observations concerning the hue of fluorescence and .purity of G,, a careful re-examination of these points seems desirable.
During preparative separation of G, by TLC procedures, modifications of the toxin into a number of unknown derivatives occurred and thus hindered effective purikation. An antioxidant and a chelating agent were used in an attempt to prevent such undesirable chemical changes. This paper describes the application of 4-methyl-2,6-di-tert.-butylphenol (BEET) and ethylenediaminetetra-acetic acid (EDTA) in the purification of G,.