Circular dichroism (CD) curves are reported for poly dA, (pdA),, (pdA)z, poly A, ApAp, ApA, AMP, dApA, pdApA, A-2'4-rnethyl PA, and A-2'4-methyl PAP. Analysis of these curves indicated the presence of single CD bands at 228-230 mp and a t 278-280 mp in oligomers longer than dmucleotides. I n the case of dinucleotides and mononucleotides (from the literature, in addition to those studied here), the 230 mp CD of band appears but the 280 mp CD band does not. We ass@ the 230 mp band to a very weak w* transition a t this wavelength. From theoretical considerations, we show that the 280 mp band is not an exciton component of the strong u-u* transition a t 260 mp in adenine. We conclude that the 280 mp CD band must be assigned to a distinct absorption, not previously reported, which we suggest arises from an n-r* transition. The fact that the n a * CD band a t 280 mp is not seen in mononucleotides or dinucleotides is ascribed to solvation of the adenine ring by water, which shifts the band to shorter wavelengths. Therefore, only interior residues of oligomers have the 280 mp band, and the optical activity of a polymer cannot be computed from that of a dinucleotide, by using a nearest-neighbor approximation. The existence of this end effect has been tested, by taking it into account in computing the rotational strengths of the 278 mp n-* transition for several oligomers; it is pointed out that a more sensitive test of this end effect would require CD data for the oligo dA series of 3 to 5 residues. We speculate about the structural and optical differences between poly dA and poly A, and point out the need for a theoretical treatment of n-p* Cotton effects in polynucleotidea.