We document, for a new data set, the existence of daily seasonality. The data set consists of the trades in four equities and two bonds on the Dublin stock exchange for the mid nineteenth century.

AMONG the subjects of public importance there are few which attract less attention and whose meaning is so little understood by the public as the simple weekly statement which enables a. comparison to be made in the prevalence of sickness and mortality during one period and another, or between one locality and another.It may be well therefore to take the opportunity offered by a meeting such as this to refer as briefly as possible to some of the principal features which must be understood before the simple mode of expressing them can be appreciated.The few observations I have to offer are made with the definite object of facilitating a correct appreciation by the intelligent layman of the facts and figures relating to the health of a city which are recorded week by week in the newspapers.The first fact to fix in the mind is that quite apart from the sanitary condition of a community, the number who will die during any given period is largely influenced by the ages of the members of that community. In other words, a division of the population into twelve age-groups shows that the number of persons who will die during a year out of every thousand in each of the different age-groups will vary widely.Tables were exhibited showing the actual number who die CMClI year in England and Wales out of every thousand living at each of twelve given age-periods. The tables showed that if, for example, we could conceive that the whole population of Liverpool consisted of persons between the ages of 25 and 35 the death-rate would be about 7 per 1,000; if, on the other hancl, we could conceive that it consisted entirely of people under 1 year of age the death-rate would be about 150 per 1,000, and this with absolutely no change whatever in the general state of municipal sanitation.Here, then, is the first elementary principle, viL. : that, irrespeeti',re of other conditions, the mortality varies enormously at different age-periods. Iienee it is self-evident that if the death-rates at each successive age-period be precisely alike in two towns, but in the population of one of them there be a mucli larger proportion either of very young or of very old persons than in the population of the other, the general deathrate will almost certainly be higher in the former than in the latter, inasmuch as the average mortality of the very young or at UNIVERSITE DE MONTREAL on June 7, 2015 rsh.sagepub.com Downloaded from