In the following year this new section was referred to a prelude to the establishment of a flourishing society (the Stat Society), which acknowledges itself the offspring of the Associ and which promises, by a procedure similar to that introduc the Association, to advance materially the greatly neglected s of British statistics.

From the original prospectus of the Statistical Society, pr at the end of the report of the third meeting of the B Association, we extract the following:

"THE STATISTICAL SOCIETY OF LONDON has been established (15 M 1834) for the purposes of procuring, arranging, and publishing calculated to illustrate the condition and prospects of society.'

"The Statistical Society will consider it to be the first and essential rule of its conduct to exclude carefully all opinions fro transactions and publications,-to confine its attention rigorously to —and, as far as it may be found possible, to facts which can be numerically and arranged in tables.

"The first operation of the Society will probably be to subdivide organise its General Council in such a manner as may enable that Co to deal conveniently with all the subdivisions of the subject-matter Those subdivisions will necessarily be numerous.

it.

"The whole subject was considered by the statistical section o British Association at Cambridge, as admitting a division into four classes:

"1. Economical Statistics.

"2. Political Statistics.

"3. Medical Statistics.

c "4. Moral and Intellectual Statistics.

"Economical Statistics comprehend, 1st, the statistics of the na productions and the agriculture of nations; 2ndly, of manufactures; 31 of commerce and currency; 4thly, of the distribution of wealth, or facts relating to rent, wages, profits, &c.

"Political Statistics furnish three subdivisions: 1st, the facts rela to the elements of political institutions, the number of electors, jurors, 2ndly, legal statistics; 3rdly, the statistics of finance and of nati expenditure, and of civil and military establishments.

"Medical Statistics, strictly so called, will require at least subdivisions; and the great subject of population, although it might classed elsewhere, yet touches medical statistics on so many points, tha would be placed most conveniently, perhaps, in this division, and wo constitute a third subdivision.

"Moral and Intellectual Statistics comprehend, 1st, the statistics literature; 2ndly, of education; 3rdly, of religious instruction ecclesiastical institutions; 4thly, of crime. Although fourteen subdivisi have now been enumerated, it is probable that more will be required."

upon the facts presented, or to draw conclusions from them, is not within the province of statistics; but is the business of the statesman and of the political economist. And, in order still further to exemplify the precise character and limits of statistics, the emblem chosen for the Statistical Society consisted of a wheatsheaf, with the motto "aliis exterendum." The statistical drudge was absolutely required to stifle his opinions; and to content himself with binding up his sheaves of wheat for others to thrash out!

It is satisfactory to know that, in the interests of science, these absurd restrictions have been necessarily disregarded in the numerous valuable papers that have appeared in the series of volumes of the Journal since issued by the Society. The statist usually sets out his results in numerical tables, because facts, especially when they exist in large numbers, are most compactly and clearly conveyed in tabular statements; and because he is not satisfied with giving deductions which may admit of question, but supplies the materials which every one may himself examine and compare. It is not true that statistics consist merely of columns of figures; it is required that all conclusions shall be fairly drawn from well-attested data, and shall be capable of mathematical demonstration. One of the principal objects of the statist is to ascertain the actual condition of his country and the causes of that condition, with a view to discover also the methods of improving it. The actual condition of the country is by no means obvious to the superficial observer, and the causes of the various phenomena which it exhibits lie; very deep, and can only be elicited with the aid of extensive investigations conducted with skill and judgment as well as with mathematical exactitude. The statistics of education, of crime, of pauperism, of labour, of health, of trade, of agriculture, of manufactures, and of every one of the details which enter into the survey of our national condition and prospects are closely connected one with another. At the same time, not only do they admit of being studied separately, but more true progress will be made by such a method of study. In these inquiries the statist should know what information is wanted, and how he will look for it, before he commences his examination of a mass of figures; his mind should be unbiassed by any preconceived opinions and ready to receive the suggestions which the study is sure to produce. He should be ready at once to abandon any hypothesis as as he finds it to be untenable and to subiect his materials

Statistics of pop

pendent inquiries cannot fail to elicit. should have his special attention, as it is closely connected many other subjects.

Population is the statistical element of greatest importa relates, above all, to the people, and a just appreciation welfare and their requirements. It would be vain to atte form social statistics of value without having as a bas results of a periodical census executed with every necessa caution, and with all the accuracy which is so indispe Other data have no real value, except so far as they relate numbers of the population. The classification according supplies the means of establishing useful tables of populat forming correct views on mortality, on the efficient resour the disposal of the State in case of necessity, and of fixi ratio between the useful portion which actively contribu the general well-being and the weaker portion which yet re assistance and support to become in its turn useful. The fication by professions, trade, and occupation, indicates the by which the population provides for its subsistence, and the legislator more particularly to concentrate his attenti the separate details of the complicated machine confided care. The classifications by civil condition, by educatio origin, &c., furnish the administration with no less va information to assure internal good order, and to facilitat execution of the laws. In fact, all questions which are conn with population deserve the greatest possible attention on the of the Government.

Medical statistics are closely connected with the subje population, and present the greatest difficulties to scientifi vestigation. The progress of medical science is necessarily and uncertain, for this reason, that the phenomena observed their origin to causes that are innumerable, and that, c quently, they seldom admit of being compared one with ano If there were a rigorous identity in all constitutions, then disease followed by a cure, and well observed, would be suffi to obtain a similar success every time that the same dis was reproduced in other individuals. But this perfect ide cannot reasonably be supposed to exist; and if it were absolu requisite to foresee all the cases which could be presented, so a

collect sufficient observations to verify all possible combinations, we should despair of ever arriving at anything satisfactory. Looking, however, at the broader lines of distinction, there does not really exist a very formidable number of essentially different cases, although the causes which influence one disease are in general both numerous and susceptible of variation. It is therefore only necessary to distinguish the more prominent causes which show a very decided individual influence, and to ascertain from a sufficient number of observations the degree of energy which belongs to each. In medical statistics relating to public health, in consequence of larger numbers, the facts are, in the aggregate, less complicated, and therefore more easily susceptible of a statistical analysis.

In the prosecution of medical science there are two distinct departments which may be conveniently separated. The successful cure of the patient depends on a compound event, viz.:-1, That the doctor shall discover the nature of the disease; 2, That he shall know the mode of treatment to be adopted. The former of these inquiries should have the precedence; diagnostics should always precede therapeutics. Investigation should first be directed to questions of diagnostics, or the distinguishing of symptoms. Such symptoms being given, what is the probability that the patient is affected by one malady rather than another? What is the probability of its duration, of its intermittence, and of its passing into a chronic state when it is abandoned to the operation of nature only? These questions, amongst others of equal importance, being once resolved, would constitute the surest basis for the study of the influence of medicine. In all these, as in other inquiries, the student should inform himself by observation, collect and analyse well-recorded facts, and render them as comparable as possible before seeking to discuss them with a view of deducing their relations, and methodically proceeding to the appreciation of

causes.

The sciences of observation have for their object the study of the natural bodies by which we are surrounded and of the laws connected with them. They all proceed in a similar manner, and the same rules are generally applicable to them. In all of them it is equally found necessary to commence by collecting wellobserved facts and classifying them systematically, so as to convey the required information. It is in weighing and appreciating the value and relative importance of the facts that science properly

their commencement have undergone the same phases; th formerly arts, for they were in the first stages necessarily to grouping, in a more or less successful manner, colled observations and facts belonging to a particular order of and it is by the comparison and discussion of these facts t afterwards became elevated to the rank of integral depa of science. If the subject of statistics presents itself in a manner and passes through the same phases, its future doubtful to those who can comprehend the sciences of obs in a philosophical point of view.

It should be remembered as a matter of general utility preparation and discussion of statistical tables, that verifications, by means of differences, should extend to the c of figures, as to the continuity of their progression. This is singularly facilitated by the use of diagrams. A simp enables us to appreciate at a glance a succession of numbers the most subtle mind would find it difficult to retain and co The facility we thus have of at once perceiving a series of and of recognizing the peculiarities of their progression i that we may at the same time follow another set of resul judge by the common inflexion of the lines the ratios e between the elements under comparison, and the common which may modify them simultaneously. Graphical represent cannot be too highly recommended when it is required to series of numbers influenced by common causes, and to general view of the modes of action of such causes.

It has already been stated, at the commencement of this that the scientific principles of statistics, properly consi enter somewhat extensively into the calculus of probab An account of this important connexion, exhibiting some higher mathematical applications, will perhaps comprise the interesting and most scientific part of the subject now under sideration. Before entering upon this account it may be useful to premise a short statement of the fundamental prin of the mathematical theory.

PROBABILITY is a numerical fraction expressing the val the degree of presumption in favour of the occurrence of a parti event, when the causes or conditions which influence the resul partly known and partly indeterminate. The word chance