classes of subjects distinctly deflued, so as to avoid all danger of mis- 

 concei)tiou by the local ageocies for the collection of facts ; and that 

 they shoal, I be so constructed as to admit of modification to suit local 

 an«l national variations. The points of inquiry should be as few as pos- 

 sible and embrace only the points of pressing necessity. They should 

 be arranged into two general groups : 1st. The soil, its occupation by 

 specific crops, live stock, number and occupation of the inhabitants. 

 2d. The actual production, market-prices, transportation, wages, loans, 

 and transfers of real estate. The first group embraces the more perma- 

 nent and the second the more variable elements of agricultural produc- 

 tion. The decennial census should be taken in all countries the same 

 year, and its results published as speedily as possible. 



In regard to acreage of crops, data should be obtained from intelli- 

 gent residents. Tax-rolls give imperfect information, as the present use 

 of land is very different from what it was when these were originally 

 framed. The area of local inquiry should, as far as possible, represent 

 specific political subdivisions of the country. The schedules should 

 specify the amount of laud lying waste, woodland, land in pasture, in 

 meadows, under plow-culture, under spade-culture, in orchards and 

 fruit-gardens, in ornamental parks, in barn-yards, door-yards, roads, &c. 

 If a larger or smaller number of inquiries be used, they should be ar- 

 ranged so as to be capable of condensation or division into a uniform 

 international series. The relative acreage in different grains is a subject 

 of special importance, which has been very successfully treated in the 

 Duchy of Baden. There is even here considerable divergence between 

 the real and estimated areas, especially of small crops ; yet the general 

 results show the wisdotn of the method of gathering local statistics by 

 intelligent private individuals. France, Belgium, Austria, and Sweden 

 have made tolerably complete divisions of crops, and their schedules 

 approximate the true practical ideal. Austria, composed of diverse na- 

 tionalities, has had unexpected success in her method. 



Only a few countries publish authoritative statements of crop-yield; 

 hence, reliance must be placed on estimates of residents, which, though 

 only approximations, may, by careful discrimination between the intelli- 

 gent and unintelligent reports, between small and large farming, between 

 modes of culture and known results, be made to indicate results closely 

 approaching the actual truth. Discrimination should be made between 

 rich, medium, and poor soils, but political divisions cannot be ignored. 

 The heads of divisions in the schedules of crop-yield should accurately 

 correspond with those of area ; the land measures and weights should 

 be reduced to hectares and kilograms. 



Great inconvenience is felt from the lack of reliable statistics of 

 woodland and wood production. Estimates in this department should 

 be made only by men of experience, whose number is small. The points 

 of inquiry should be few and practical, embracing heaths, hedges, decid- 

 uous trees, (cultivated or not.) conifers, brushwood, &c. Timber valuable 

 for industrial purposes should be particularized. An annual enumeration 

 of all sorts of live stock should be taken at a uniform period of the 

 year. The maximum season is that in which the young are just born 

 and the old not disposed of; the minimum is the winter, when every 

 farmer reduces the amount of live stock to feed during the absence of 

 pastures. These extremes should, as far as possible, be equalized. This 

 class of statistics is generally defective. France and a few other coun- 

 tries have obtained satisfactory data. The schedule proposed by the 

 congress embraces each class of animals, discriminated according to 



