signal at certain offsets. We have found that COM-Vm gives a much narrower bandwidth as compared to a previous work on the same composite pulse.[14] This is due to the fact that longer pulses as employed in our work (220μs instead of 70μs) are more adequate for 14N site with smaller CQ value (1.18 MHz for glycine [18]), thus leading to much narrower excitation bandwidth. Nevertheless, the coverage of the excitation profile resulting from COM-IVm is twice larger than that resulting from single pulse in both works. It should be reminded that for most 2D experiments, the carrier frequency of the indirect dimension is preferably set in the center of spectra in order to avoid the interference of artefacts arising from the imbalance of quadrature detectors. Therefore, such symmetric excitation at the horns could be employed for recording signals with distinct chemical shifts. This is particularly helpful for indirect detection of 14N species in biomolecules because amide and pyridine nitrogen species with well-defined chemical shifts often co-existed in these samples. Such advantage is demonstrated on the 1H-{ NOT DQ1 4 } D-HMQC spectrum of L-[U-13C]-histidine shown in Fig. 3, where the two large cross peaks corresponding to amide and pyridine nitrogen species arepresent. These two nitrogen sites exhibit a shift difference of 14.7 kHz. As a consequence, a symmetric excitation profile at 77.4 kHz should be optimum as long as the carrier frequency is set in the center of 14N dimension. This optimum excitation is realized by using COM-IIm or COM-IVm, but is impossible to obtain with two single pulses that lead to much