REPORT OF THE STATISTICIAN. 



365 



the basis of the comparison had been a full crop (say 100 of reports 

 of condition) these returns would have been substantially correct. 

 We are therefore forced to the conclusion that the reporters regu- 

 larly and persistently underestimate in comparisons with an "aver- 

 age" crop; that the real basis of their comparison is a full crop. 



In the Mark Lane Express for August 23 there is a statement of 

 crop returns made to that paper, accompanied by a record of returns 

 for ten years, making a serious under average for the whole period. 

 In only three years of the ten is there an ' ' over average " in wheat, 

 oats, beans, and peas, and only four in barley, while there are seven 

 under average years in all except barley. The number of reports or 

 "advices" in the aggregate of the ten annual returns is 16,418, of 

 which 33 per cent, are average, scarcely 20 over average, and 47 un- 

 der average. In every one of the crops the poor returns largely out- 

 numbered the good, as follows: 



Crops reported. 



Advices. 



Over 

 average. 



Average. 



Under 

 average. 



Wheat 



Barley 



Oats , 



Beans 



Peas 



Total 



3,833 

 8,763 

 3,751 

 2,597 

 2,474 



16,418 



851 

 735 

 454 

 379 



3,212 



1,106 



1,387 



1,410 



732 



800 



5,435 



1,934 

 1,525 

 1,606 

 1,411 

 1,295 



7,771 



The meaning of this is, the crops have been greatly "under aver- 

 age" in the aggregate for the whole period. Either this is an under- 

 estimate, or ten years is not a period sufficiently long to make an 

 average; yet it is a grave question whether a test of twenty years 

 would not show the same result. 



It has been a question on this side of the Atlantic whether the 

 yields claimed as averages for the several crops of Great Britain are 

 not too high. In this country there are acres that produce over 50 

 bushels each of wheat, while the average for the whole crop may not 

 be more than 12. It is known that there has been no verification by 

 census of the British product. It is made from estimates of rate of 

 yield, and all statistical experience in this part of the world shows 

 that any set of returns of yield will make too high an average. It is 

 feared that the high yields of Lawes and Gilbert's experimental plots, 

 and of other reported liberal yields on individual farms, have an 

 undue influence in estimates of averages. If a census of yield, like 

 that of acreage, should include every holding, and be returned from 

 actual measurement, it is very probable that it would seriously shat- 

 ter many fancied averages, as it has done in this country. 



CURRENT CROP STATISTICS. 



CORN. 



There is a constant tendency to enlargement of the area in corn. 

 It has been doubled in fifteen years, increasing at the rate of two to 

 three million acres annually. The reduction in values has little effect 

 in arresting extension of area of maize. It is realized that meats, 

 butter, and cheese are safer and more profitable surplus products than 

 the bulkier raw products of agriculture, at whatever rates such prod- 

 ucts have been or are liable to be sold. The uses of corn are vari- 



