mercial conventions to hold annual meetings, Avhich poisoned the public mind it 

 thus controlled, by making it believe that the Union existed at the cost of 

 southern prosperity. It made no effort to set forth the foots teaching us that 

 every section was dependent on the other, and in that dependence lay the great 

 prosperity which had blessed the industry of every section. Thousands of dollars, 

 if necessary to this purpose, Avould have been wisely spent in the collection of 

 statistics which now would have saved millions of dollars, and lives of far greater 

 value than all these millions. With the great events of the day pointing to 

 these errors of the past, the nation should profit by their teachings. The sta- 

 tistics of every branch of industry must be collected if their mutual relations 

 are to be comprehended. Omit important portions, as is now done, and the sta- 

 tistician gropes among partial and doubtful statements in vain conjecture of 

 what is the extent and relations of the industry of the country. In a Union so 

 extensive, Avith productions so diverse, with forms of government tending so 

 strongly to that which is local against that which is national, the mutual rela- 

 tions of that industry must be made familiar to all. 



In March the Senate of the United States directed the Secretary of the 

 Treasury to collect statistics of the domestic commerce of the United States. 

 In his recent annual report that officer says : " The materials for a proper state- 

 ment of this internal commerce must be sought in reports of State commissioners 

 of statistics, of boards of trade, of railroad and canal companies, and occasional 

 or periodical publications relating to trade and business. This is a department 

 of statistics comparatively new and difficult of exploi'ation." 



To fully set forth the " intertwined" trade of the several States, an extensive 

 and connected system of statistics is necessary. Whilst the materials mentioned 

 by the Secretary are important to exhibit this trade in its aggregation, yet they 

 are too imperfect in their disconnexion to furnish sufficient materials for the un- 

 folding of our internal commerce. But they are all that can now be had. They 

 take no notice of much that is material. Instances are seen in the trade 

 between the western and cotton-growing States. Ohio, Indiana, Illinois, and 

 Kentucky supply the latter Avith horses and mules. So in the cattle trade 

 between the Avest and the east. Illinois is credited with the large supplies of 

 cattle it furnishes the eastern States and the packing houses, yet Indiana and 

 Ohio raise large numbers of these cattle. They are driA^en to the immense 

 cornfields of Illinois to be fattened. Whilst the slaughtering houses furnish the 

 statistics of the number of hogs, and their weight, that are packed in them, 

 they can give no account of the bacon and lard sold by farmers, Avhich not only 

 supply the consumption of localities where sold, but in many years form no in- 

 considerable portion to the general commerce of the country. Statistics should 

 be gathered from all occupations representing the industrial pursuits, and a cor- 

 respondence should be had Avith all, such as is held by this Department Avith the 

 agriculturists of the country. 



Again : by making a statistical bureau a part of the Department of Agricul- 

 ture, greater economy Avould be effected. The acts of Congress require the col- 

 lection of agricultural statistics by this Department, and such collection implies 

 the use of them in all investigations relative to the progress of agriculture, and 

 their publication. For this purpose the monthly and annual repots are pub- 

 lished. As has been seen, where the power to collect is lodged, there must exist 

 the poAver of investigation also, for the former must have direct reference to the 

 principles unfolded by the latter. 



The Agricultural Department, then, must collect these statistics, must consider 

 and publish them, and necessary to such consideration is a statistical library and 

 papers. It must have a statistical division, as it now has, and these must be to 

 an extent not much less than should be found in a bureau of statistics. Why 

 Lave two libraries, two bureaus in effect, two sets of agencies in collecting sta- 



