In poliomyelitis patients who have received some antigen during the month prior to onset there is a high degree of correlation between site of paralysis and site of injection. Such cases tend to show a different distribution of paralysis and a more severe paralysis than do comparable children immunized 2 to 6 months previously, immunized in previous years or never immunized. The time interval between date of injection and onset of illness suggests some relationship other than chance. There is no evidence that this relationship is associated with any one antigen. There is no evidence that the influence of the antigen persists, or in any way conditions the response to poliomyelitis virus for longer than one month. Since immunization is an elective procedure, it can well be delayed during outbreaks of poliomyelitis; but there is no suggestion that the postponement should be more than temporary.