IN the preceding note, which the Editor of NATURE has been good enough to let me see in advance of publication, Schmidt, following Murakawa1, suggests that the frequent occurrence of perturbations in terms of both the arc and spark spectra of iodine indicate a nuclear quadrupole moment. This may be correct, but as I have shown that at least in the spark spectrum2 term interaction perturbation occurs, and in view of the complex electronic structures involved, it is very likely that the perturbations may be due to either of the above causes or possibly both, depending upon individual terms (that is, if a quadrupole moment is assumed to exist). This fact makes caution desirable and any calculation of the quadrupole moment can only be very approximate. Schmidt has been obliged to employ my measurements for the arc spectrum, and as I have previously pointed out, the structures in this spectrum are not so well resolved as in the spark spectrum. His calculation is therefore based upon a small deviation from the interval rule in which I was not able to prove whether more than one level was affected, so that the type of perturbation is uncertain.