In order to test the Fuller and Hodgson hypothesis that modification of the anticodon-adjacent base in certain tRNA's not only prevents mRNA base-pairing at that site but also increases the stabilization of a stacked conformation in the anticodon loop, we have examined the interaction between adenosine and its N6-isopentenyl derivative by means of model compounds. The synthetic 9-[3-(aden-9-yl)propyl]-6-(3-methyl-2-butenylamino)purine, Ad-C3-iPAd (IV), in which the adenine and N6-substituted adenine moieties are joined at the 9 and 9′ positions by a trimethylene chain, served as a useful spectroscopic model for assessing the base-base interaction free from the complicating features of the carbohydrate and phosphodiester groupings. The hypochromism for the model, which was determined in dilute aqueous solution and represents the decrease in integrated ultraviolet absorption intensity of Ad-C3-iPAd (IV) compared with equimolar 9-propyladenine (Ad-C3) and 6-(3-methyl-2-butenyl-amino)-9-propylpurine (V, iPAd-C3), was 17.9 per cent in neutral solution, 8.4 per cent in 0.1 N HCl, and 18.5 per cent in 0.1 N NaOH. Comparison with the per cent hypochromism calculated for the simple model Ad-C3-Ad (e.g., 16.5% in neutral solution) confirms the strong interaction in IV observed between uncharged plane-parallel adenine and N6-substituted adenine rings. The cause for changes in the absorption spectrum of Ad-C3-iPAd are discussed. The fluorescence and phosphorescence emission spectra of Ad-C3-iPAd in ethylene glycol-water glass at 80°K add considerable weight to the conclusion that there is a strong tendency for adenine and N6-(Δ2-isopentenyl)adenosine (I) to stack if this is permitted by steric considerations.