Researchers may have failed to discover a deterrent effect from capital punishment, because almost all studies of capital punishment use annual rather than weekly or daily homicide data. If publicized executions deter homicides for a brief time only, the deterrent effect of capital punishment may not be detectable with yearly homicide data but only with weekly or daily data. This study examined homicide statistics for London (England) for 1858-1921 using the Great Britain General Register Office for these years. A standard casebook of notorious murderers was used to generate a list of heavily publicized English executions for the period under study. The number of homicides before, during, and after the week of each execution was determined. Findings shown that homicides drop significantly in the week of a publicized execution, and the more publicity given the execution, the more homicides decrease. These findings cannot be ascribed to the effect of weekday or seasonal fluctuations in the data or to linear trends, because these effects were corrected for in the selection and treatment of the control periods with which the experimental periods are compared. A long-term deterrent effect from executions is not evident from the findings. Tabular and graphic data and 41 references are provided.