The polyphenolic compound 1, also known as cynarin, is of interest biomedically because of its hypocholesterolemic, hepatoprotective and antioxidant activity.1 The dicaffeoylquinic acids (DCQAs), in general, show a number of pharmacological actions, with the most important being anti-HIV activity. The DCQAs have recently been recognized as an important class of HIV-integrase inhibitors and, thus, prevent insertion of viral DNA into the host genome, and essential step in the replication of the virus. The DCQAs strongly and irreversibly inhibit HIV-1 integration in biochemical assays and block viral replication at non-toxic concentrations. Earlier studies2 showed that the peracetate derivatives of 3 and 4 displayed greater anti-HIV activity than the free compounds and that the activity of peracetyl-3 approaches that of AZT. Thus, there is considerable biomedical interest in the DCQAs and selected derivatives of the DCQAs. Obtaining quantities of pure 1 has been difficult. Extraction from plant material proceeds in low yield and provides an impure product. Synthetic methods provide pure 1, but are expensive and provide a low overall yield. The recent description of a facile and inexpensive method for the preparation of 1 from 23 provides a route to sufficient quantities of 1 for biological testing.1 Three papers have described some aspects of the mass spectrometry of caffeoylquinic acids and their analogs,2,4,5 but no reports have described the mass spectral analysis of 1 and 2, or their analogs. Because of the interest in caffeoylquinic acids, we have performed an ESI MS and ESI MS/MS study of 1 and 2 and their peracetate derivatives and the mass spectral patterns of 1 and 2 have been compared to each other and to the spectra of 3 and 4 and their peracetate derivatives. Methods: A sample of 1 was obtained following conversion from 2.1 Purity was checked by HPLC and MP and the structure confirmed by NMR. Compound 2 was isolated from natural sources and its structure and purity confirmed as for 1.3 The origin of 3 and 4 has been described.3 For mass spectral analysis the samples were dissolved in the mobile phase (MeOH/H2O; 1:1) and infused into a Finnigan Q LC/MS. Full scan spectra were acquired and major ions selected for collision. The collision energy used in the MS/MS studies were the same for each of the different samples. Results: Both positive and negative ion ESI can be used to differentiate 1 from 2. Both the free compounds and the peracetate derivatives show differences in fragmentation which can be used to identify the isomer. For the free compound, the most useful data is obtained using (-) ESI and performing a product ion scan of the [M-H]- ion; a large peak at m/z 335 (M-H-162-18) and the presence of a peak at m/z 307 (M-H-162-18-28) distinguishes 1 from 2. The most useful data for identification of 1-4 is obtained using the full scan (-) ESI spectra, without collision. Each of the compounds shows either significant differences in ion intensity or unique ions which uniquely identify the isomer being examined. Other spectral data can also differentiate each of the isomers. Conclusions: Both (+) and (-) ESI MS/MS spectra of compounds 1-4 and 1a-4a can be used to uniquely identify each of the isomers. Changes in structure are reflected in different preferred fragmentation pathways, but the details of the structural effect are as yet not known. While mass spectrometry can be used to differentiate pure samples of each of the isomers, LC/MS would be needed to differentiate the compounds from a biological matrix or synthetic reaction mixture. Additional studies are currently underway which will, hopefully, explain the different pathways operating in these, and related, compounds.