Remarks on the multiplicity measure for $e^{+}e^{-}$ annihilation

A recent proposal of the Lund group suggesting a multiplicity measure for $e^{+}e^{-}$ annihilation is analyzed. It is shown that the results obtained for the multiplicity distributions in full phase do not change if one removes a questionable assumption of exact localization of branching products in rapidity (thus avoiding troubles for small rapidity bins). The predicted asymptotic form of the distribution is shown to differ from the negative binomial distribution.