Recently, Norden et al.[1] compared four in vitro methods for measurement of synergy be-tween pairs of antibiotics. These were (A) the stan-dard checkerboard method,(B) calculation of an interaction index [2],(C) the construction of iso-boles [3], and (D) the determination of killing curves. The methods were compared using combinations of cephalothin and gentamicin tested on 22 strains of Klebsiella pneumoniae, and the results may be summarized as follows. Methods B and C invariably agreed; method A suggested a lower frequency of synergy than methods B and C, whereas method D suggested a much higher frequency. Antagonism was suggested in some cases by methods B and C but in no case by method A or D. Method D suggested synergy in some cases that met the criteria for antagonism of methods B and C. Thus there were substantial dis-agreements between the methods, except between methods B and C. These discrepancies require ex-planation.