A small specimen of this unknown substance was isolated by ion exchange chromatography8 and further purified by paper chromatography. Its mobilities on paper chromatograms in four solvents (Table 1) and its spectra at five pH. values (Fig. 1) were found to be in excellent agreement with those of a synthetic specimen of N2-methylguanine4. Analogous to the case of N6-methyladenine and of 8-methylaminopurine5 was a small shift to longer wave-length of the absorption maxima of N2-methyl-guanine relative to those of guanine. To confirm further the identity of the unknown with synthetic N2-methylguanine, both were treated with nitrous acid. (To a suspension of 0-25 mgm. of substance in 0-25 ml. of 2 M acetic acid at 60, 0-05 ml. of 1 M sodium nitrite was added over a 10-min. interval.) The paper mobilities and ultra-violet absorption of the product from either specimen again coincided within experimental error (Table 1). The structure of the product is at present unknown. The substance produced by action of nitrous acid on N6-methyl-adenine has been tentatively formulated as an N6-nitroso derivative6. Table 1. PROPERTIES OF THE URINE ISOLATE AND OF N2-METHYL-GUANINE, AND OF THE PRODUCTS FORMED BY NITROUS ACID TREATMENTProduct from action of nitrous acid onProperty Urine Na- Nl- isolate methyl-guanine urine isolate methyl-guanineRF values in n-butanol-0 -6 Mammonia 0-25 0-25 0-04 0-04 n-butanol-aq. formic acid 0-27 0-27 0-50 0-492H 7 aqueous buffer0-45 0-46 2H 9 aqueousbufferf 0-45 0-45 Ultra-violet ab-sorption at pE. 2-1, max. (mn) 251, 280 251, 279 245, 292 244, 29039 H 21, min. (niyu) 228, 270 228, 271 234, 279 236, 278 pH 6 -2, max. (m^) 247, 280 247, 280 245 2452?H 6 -2, min. (m/j) 226, 266 226, 267 228 228 pTL 9-0, max. (m^u) 247, 280 246, 280 243 243pH 9 -0, min. (m^) 228, 266 229, 266 228 2290 -07 M phosphate buffer. t 0 -05 M borate buffer.The urine of all the normal human subjects examined contained the new compound. The average daily urinary excretion was 0-5 mgm. The urinary excretion of N2-methylguanine, and of the purines earlier described1, was not significantly changed after three days on a diet restricted to glucose and water, or after sterilization of the intestinal tract by ingestion of neomycin and succinylsulphathiazole. This indicates that they are products of endogenous metabolism in man, and not dietary or bacterial in origin. Fig. 1. Ultra-violet spectra of the urine isolate (lower curves) and synthetic N-methylguanine (upper curves), (a) in 1 M hydrochloric acid -; (fe)at pB. 2-1,----; (c)atpH6-2,; (d) at #H 9 -0,; and (e) in I M sodium hydroxide, This identification of N2-methylguanine as a urinary constituent seems to be the first time it has been recorded in a natural product. Of interest in this connexion is the recent demonstration of N6-methyladenine6 and of N6-dimethyladenine7 in bacterial and fungal sources. With the earlier identification of 7-methylguanine2 and 1-methylguanine1, the identification of N2-methylguanine brings the number of methylated guanines in human urine to three. A possible fourth such derivative is the substance designated1 'spot K\ which is still under investigation. The similarity of the spectra of this substance to those of 8-hydroxyguanine at a number of pH. values and other evidence suggest that it may be a methylated 8-hydroxyguanine. The possibility that methylated guanine derivatives may play some significant part, as yet unexplored, in human metabolic processes is indicated.We wish to thank Dr. Gertrude B. Elion, of the Wellcome Research Laboratories, for making available a specimen of synthetic N2-methylguanine. The investigation was supported in part by a grant-in-aid from the National Institute of Arthritis and Metabolic Diseases, National Institutes of Health.