borate, and expensive works on county history and topography as England. But, with few exceptions, their authors seem to have been totally unacquainted even with the elements of statistics, and to have thoroughly contemned what they did not understand. According to Mr Gough, incorrect pedigrees, futile 'etymologies, verbose disquisitions, crowds of epitaphs, lists of 'landholders, and such farrago, thrown together without method, 'unaccompanied by reflections, and delivered in the most un'couth and horrid style, make the bulk of our county histories.' (British Topography, vol. I. p. 21.) Perhaps this is rather rating them too low; but nothing certainly can be more miserably poor and deficient than the statistical notices interspersed throughout the great majority of these publications, with, in some instances, the exception of those that relate to prices. Our professed geographical works have not been of a class to make for their deficiencies; and down, indeed, to a very late period, the wretched compilations of Guthrie and Pinkerton were the only general works on geography to which the mere English reader could refer. Now, however, there are symptoms appearing of a better state of things. The translation of Malte Brun's Geography, Mr Murray's Encyclopedia of Geography, and the excellent articles on geographical subjects in the Penny Cyclopedia, will familiarize the public with improved models, and will, most likely, put an end to the trash that, under the names of systems, grammars, gazetteers, &c., has so long disgraced our literature.

up

But, as respects statistics, we have made comparatively little progress. Sir William Petty's Essay on Political Arithmetic,' published more than 150 years ago, is the latest native work, of a purely statistical and unofficial kind, that has appeared in England! Supposing a foreigner were to transmit to London an order for the best account of Great Britain, we do not know that the bookseller could do better than send him the last edition of Chamberlayne's State of Britain, published some time in the reign of George II. No Englishman ever asks for any such book, because he knows that none such exists. No doubt, had the public been at all anxious to procure a good descriptive or statistical account of the empire, the existing desideratum would have been supplied. But such works, and the taste for them, are alike unknown in England. If an individual living in Kent wishes to learn any thing of Northumberland, he has nothing for it but to go there; or to trust to the meagre, and, generally speaking, inaccurate details to be found in some Gazetteer or Encyclopedia; and those who wish to learn any thing of the arts, the military and naval force, constitution, &c., of

the empire, must consult at least some twenty or thirty different works.

We scruple not to say that this is discreditable to the country; and we are happy to observe that a conviction appears to be growing among intelligent persons that such is the case, and that the subject has latterly begun to excite considerable attention. A Statistical Society has been founded in the metropolis; and an office has been opened at the Board of Trade, for the collection and publication of official documents connected with statistical subjects. It would seem, therefore, as if it might be reasonably anticipated that something will shortly be done to bring us more nearly, in this respect, on a level with other countries; and that we shall no longer be the only civilized nation in Europe which has made no progress in this highly interesting

science.

But there are some pretty formidable obstacles that must be overcome, before we make much way in statistical enquiries. The nature, objects, and limits of the science are neither well understood nor defined; and the accurate information at our disposal, is much less complete and extensive than might be supposed.

By a statistical account of a country, we mean a work describing its situation and extent-its natural and acquired capacities of production—the quantity and value of the various articles of utility and convenience existing and annually produced in it-the number and classes of its inhabitants, with their respective incomes-its institutions for the government, improvement, and defence of the population; with a variety of subordinate and subsidiary statements and details. A science embracing so great a variety of objects is not easily defined or limited; nor is it in all cases possible to state absolutely what ought to be taken in, and what left out. It has many features in common with geography and politics; and embraces that sort of mongrel science that has sometimes been called political arithmetic. Some authors, and, among others, M. Say, have contended, that any description, however brief, of the territory of a country, was foreign to statistics, and belonged exclusively to physical geography. But this is evidently a most erroneous statement. How can we aquire any accurate or comprehensive knowledge of the capacities of production enjoyed by any country, if we be unacquainted with its situation, soil, climate, native products, and its facilities for bringing them to market, and exchanging them with others? So far from being alien to, this knowledge lies at the very foundation of the science; and any work which should omit it, however well it might be executed, would have no claim to be considered as a complete statistical

work. Those topographical details which are essential in a geographical work, would, however, be totally out of place in one devoted to statistics. One of the principal objects of the latter is, to exhibit the means and sources of the national wealth, its amount, and distribution; and these would not be in any respect promoted by indulging in topographical details. Hence, in statistical works, a short notice of the principal divisions of the country, with reference especially to its climate, soil, native products, agriculture, manufactures, and population, is generally sufficient. This much, however, cannot be dispensed with. It is the substratum on which all the rest of the building is to stand; and the completeness of the other and more elevated parts will generally depend more on the compactness and solidity of this than on any thing else.

Opinions differ as widely as to the mode in which statistical treatises should be drawn up, as with respect to the topics they ought to embrace. Some contend that their object is limited to an exposition of the state of a country, province, or place at some given period; and that, consequently, all historical and theoretical discussions should be rigidly excluded from them. But though the Germans have completed several works on this plan, it seems to us to be radically objectionable; and to be calculated to deprive the science of all that is most interesting and instructive. It presents us with a skeleton, instead of a living, animated body. It is of importance certainly to know the amount of population, the sum paid in taxes, the rate of wages, &c. in a country or district at any given period; but it is only by comparing these statements with others of the same sort, applying to different epochs, that we learn whether such country is advancing or retrograding; and it is only by comparing details peculiar to one country with those peculiar to another, that we learn in what respects they agree and differ, and that the attention of the politician and the economist is called to those circumstances that retard the progress of the one, and accelerate that of the other. Those who are fond of subdividing, call this comparative statistics, and contend, that it ought not to be mixed up with descriptive statistics. But to be really good, and possessed of interest enough to make it be read, a work must embrace both. They ought not to be so intermixed as to create confusion; but unless they be combined in due proportions, the principles on which the comparisons are made expounded, and the circumstances that produce the discrepancies pointed out, and their influence correctly appreciated, no details, however accurate, can be of much value. It is the easiest thing possible to pile figures on figures; but unless deduced from cor

rect data, they serve only to mislead; and they do this the more easily that they have a scientific air about them, and that most people shrink from the irksome task of examining whether tabular statements be correct or not. There is nothing, in fact, as to which one should be so sceptical as the greater number of statistical facts and details; or with respect to which a sound and searching criticism is so necessary. The reader would do well generally to look with suspicion and distrust on most statements, how imposing soever they may appear, unless he is informed of the sources whence they have been derived, and of the principles on, and the mode in, which they have been compiled.

Considering the vast mass of official returns and other papers annually printed by order of the House of Commons, and the number of private publications on subjects having relation to this science, no one not pretty familiar with such topics would easily imagine the extraordinary deficiency of our information as to some of the most important particulars respecting British statistics. Had the science met with any portion of that attention to which it has so many well-founded claims, these deficiences would long since have been pointed out, and efforts would, no doubt, have been made to get them supplied. At present, however, even the best work that could be compiled on British statistics, would be, in many respects, singularly incomplete. Instead of accurate, well authenticated data, we have frequently nothing to trust to, but vague conjectural estimates ; and are obliged to choose among conflicting, contradictory statements, where it may be no easy matter, perhaps, to assign any very satisfactory grounds for the preference given to one rather than to another. Nothing, indeed, is more common than to hear the most opposite allegations made in Parliament as to matters of great public importance; nor is there, in fact, a season in which the Legislature does not proceed to legislate in the dark,-interfering when it should do nothing, and doing nothing when its interference might be advantageous. Every one admits that it is impossible to legislate wisely for any country, or for any particular department of industry, or for any class or order of society, without being thoroughly acquainted with all the circumstances peculiar to such country, department, or class. And those aware of the ignorance prevalent amongst us on such matters, will not certainly be much surprised at the ill-success that has attended many measures from which much advantage was anticipated.

We are grossly deficient as respects some of the merest, but, at the same time, most important elements of statistics. Down to a very late period, the statements as to the area of England

and Wales differed from 28,000,000 to 46,916,000 acres! The first is the estimate of Sir William Petty; and being made when even the best maps were exceedingly inaccurate, we need not wonder at its being so very wide of the mark. But the other, which is as much above, as Petty's is below the truth, was framed by the late Mr Arthur Young. And if any thing could evince more strikingly than another, the low state of geographical and statistical science in this country, it would be the fact that Mr Pitt, who had every wish to be accurate, and every means of acquiring the best information at his command, adopted this estimate, and made it the basis of all his statements and calculations respecting the income tax. Its extreme inaccuracy was demonstrated by Dr Beeke, in his elaborate and excellent tract on the Income Tax: he showed that the area of England and Wales was under 38 millions of acres (38,498,572). * And it is now ascertained, by means of the operations under the Ordnance Survey and otherwise, that even Dr Beeke's computation was too high; and that the area of England and Wales is 57,812 square miles, or thirty-seven millions (36,999,680) acres. There is even yet, however, considerable doubt as to the correct measurement of the different counties. The areas of the parishes and hundreds in each English county, as given in the population returns, do not, when added together, in any instance correspond with the area assigned to the counties, taking them in the aggregate. The one estimate differs, on the whole, from the other, by no less than 746 square miles, or 477,440

acres.

The area of Scotland is less accurately known than that of England. Mr Templeman of Bury estimated it, inclusive of the islands, at 27,794 square miles. But, though this estimate has been adopted by Pinkerton and others, there can be no doubt that it is very materially within the mark. According to measurements deduced from Arrowsmith's map, the mainland of Scotland contains 26,158 square miles, and the islands 4,080; making together 30,238 square miles, or 19,352,320 statute acres. In so far as Scotland proper is concerned, we do not believe that this measurement involves any considerable error; but as respects most of the islands, it can be looked on as little better than a rough guess. The area given above includes 638 square miles of fresh water lakes. +

The first estimate of the area of Ireland, entitled to attention, was framed by Sir William Petty; who was employed to survey

Beeke on the Income Tax, p. 14, 2d ed.

Analysis of the Statistical Account of Scotland, vol. i. p. 71.