Calculations based on the mortality of whooping-cough before 1957 predict accurately the subsequent decline and the present low mortality. Notifications of incidence, though variable and incomplete, follow the same pattern of steady decline in the United Kingdom and are unaffected either by small-scale vaccination beginning about 1948 or by nationwide vaccination beginning in 1957. When valid comparisons can be made, attack-rates may be lower and complications fewer in vaccinated children, but allowance has to be made for overcrowding and socio-economic differences which may be more important as determinants of attack-rates. No protection by vaccination is demonstrable in infants. Adverse reactions and neurotoxicity following vaccinations were studied in 160 cases. In 79, the relationship to pertussis vaccine was strong. In 14 of these cases, reaction was transient but characteristic of a syndrome of shock and cerebral disturbance, which, in the other 65 cases, was followed by convulsions, hyperkinesis, and severe mental defect. It seems likely that most adverse reactions are unreported and that many are overlooked. Precise information about the efficacy and safety of this vaccine is lacking, because existing provisions, national and international, for epidemiological surveillance and evaluation are inadequate. The claim by official bodies that the risks of whooping-cough exceed those of vaccination is questionable, at least in the U.K.