A recently suggested modification of the g-index is analysed in order to take multiple coauthorship appropriately into account. By fractionalised counting of the papers one can obtain an appropriate measure which I call gm-index. Two fictitious examples for model cases and two empirical cases...