The earliest types of two-dimensional NMR experiments that received widespread attention were the proton-flip and the gated-decoupler experiments (Z-8). Concep tually those early experiments are rather simple to understand, but the use of those experiments in practical cases often is complicated by data storage requirements and sensitivity considerations. Only in the case where the spectroscopist is interested in determining the multiplicity of a certain carbon13 site is the use of the two-dimensional experiment straightforward, in such cases only a very low resolution of the carbon13 multiplet structure along the F, axis is necessary, and hence the data matrix can be kept small. The sensitivity will be fairly good in this application because the carbon13 intensity is spread over a maximum number of only four multiplet components. However, this application of two-dimensional J spectroscopy has been outdated by even more sensitive and experimentally more convenient “spectral editing” experiments (9-Z I). The use of the original 2D J-spectroscopy experiments for the exact determination of coupling constants requires a very fine digitization along the F, frequency axis; this leads to a very large size of the required data matrix, unless the acquisition time along the t2 axis is made very short, which in turn decreases the sensitivity of the experiment dramatically. The intensity of a certain carbon13 nucleus in this highF1-resolution application will generally be spread over a large number of multiplet components, which decreases the sensitivity even further. All problems mentioned above can be overcome by using a modified version of the proton-flip experiment in which only one proton is “flipped” by the proton pulse (which in this case is selective) in the center of the evolution period (12). This allows the accurate determination of all heteronuclear long-range couplings with this particular proton. In this communication we describe two experiments which rely on the same principle as the selective proton-flip experiment and which allow the accurate determination of either all direct ‘H-13C coupling constants by removal of the long-range splittings or the determination of all long-range couplings while removing all the direct ‘H-13C splittings. The experimental pulse scheme is set out in Fig. 1. As in the original proton-flip experiment (3, 4), broadband low power ‘H irradiation during the preparation period provides a NOE effect, enhancing the sensitivity, and high power proton noise decoupling is employed during the detection period. A carbon13 180” pulse is applied