We have obtained spectra of the high-lying vibrational levels of the 1 X state of Liz [5] and have been applied to ultracold photoassociation spectra [2,3]. However, in those investigations the details of the potential are not accounted for, and the accuracy of the resulting long-range coefficients is only a few percent. We are able to extract a more precise value for the long-range potential coefficient, and therefore, the atomic radiative lifetime by fitting a model of the full interaction potential to the observed photoassociation spectrum. [7].Unfortunately, the theoretical value differs from the experimental one by more than 0.7%, or more than four standard deviations. Clearly, the resolution of this discrepancy is a high priority.The experiment uses a six-beam magneto-optical trap (MOT) [8] to produce an ultracold lithium vapor [9]. A 500-mW probe laser beam with a Gaussian beam waist (1/e radius) of 500 p, m is directed through and retroreflected back through the trapped atom cloud to induce photoassociation. As the probe is tuned to the red of the 2S»2-2P»2 atomic transition, it can excite colliding groundstate atoms into various high-lying vibrational levels of the molecular excited-state manifold. These newly formed diatomic molecules decay into a bound ground state of the molecule or into a dissociative continuum of two groundstate atoms that may gain enough kinetic energy to escape the trap. These combined processes result in an observable reduction in trap-laser-induced fluorescence, which is monitored by a photodiode. Vibrational levels up to 3 THz (100 cm ) red of the 2S»2-2Pi/2 atomic transition have been observed in this manner (Fig. 1) [10].