V2S SECTIONAL TRANSACTIONS.— L. M. 



(e) Mr. W. W. Vaughan. 



Many of the misconceptions in the public mind arise from those who write novels 

 about the public schools, because they find themselves unable to make a living in 

 any other way. These critics know nothing of the modern public school and never 

 by any chance mention the musical and scientific societies or that great institution, 

 the O.T.C. So far from wishing to keep the public school spirit to themselves, public 

 school men desire notliing better than to extend it into every school and every section 

 of the community. They are sometimes told that they want to shut out the parent. 

 On the contrary thej' welcome the parents' co-operation. 



SECTION M.— AGRICULTURE. 



(For references to the publication elsewhere of communications entered in the 

 following list of Transactions, see page 449.) 



Thursday, August 5. 



1. Mr. J. A. Venn. — TJie Yields of British Crops; an inquiry into the 

 results attaching to past and present methods of crop estimating and 

 forecasting. 



In 1868 Lawes and Gilbert prepared the first estimates of wheat yields to cover a 

 series of years ; these commenced with 1852 and recorded- a figure for the United 

 Kingdom of 284- bushels, which unfavourable seasons subsequently reduced to less than 

 25 bushels during the 'seventies. Official statistics, inaugurated in 1884, indicate a 

 slow advance up to about 1910, when a level of 32 bushels was attained. Since then 

 there has been a loss of a bushel. These movements cannot be correlated with the area 

 under the plough or under ^^'heat, are in the main applicable to other cereals, but are 

 little evidenced in Scotland and Ireland, where all such yields are in excess of the 

 English level. 



Concurrently with the introduction of the official series, deviations from the normal 

 became less marked ; thus the Rothamsted figures had shown a mean range equivalent 

 to 14 per cent., while the official series at first indicated a deviation of 7 per cent., 

 which has steadily declined to the present level of below 5 per cent. Comparison with 

 other countries producing cereals under similar conditions afi'ords widely divergent 

 results, e.g. in Denmark the figure is 13 per cent, and in Germany 10 per cent. The 

 Rothamsted estimates were merged into an annual statement published by The Times. 

 Close comparison of these figures (prepared locally by a large body of experienced 

 agriculturists) with the official series indicates that in almost 80 per cent, of instances 

 they exceed the latter by substantial margins. Investigation of the statistical methods 

 respectively employed fails to account for this divergence. Further, the official figures 

 display chronological and geographical inconsistencies ; thus, although half the wheat 

 produced in the Eastern Counties has recently been shown to consist of the newer 

 varieties, there is no indication, officially, of increased yields from this area. 



These results appear to be due to conservatism on the part of Crop Reporters, and 

 are in no way attributable to the methods of amalgamation and weighting adopted by 

 the Ministry of Agriculture, whose statistical work is unassailable. Local under- 

 estimation of, perhaps, 5 to 10 per cent, in the case of cereals and 15 to 20 per cent, 

 in that of roots seems, however, to be operative. 



In common with those relatmg to other countries, official crop forecasts also show 

 a pessimistic trend, for, if the series issued at monthly intervals prior to harvest is 

 closely compared with the actual estimates of yields subsequently recorded, it emerges 

 that there is a tendency (a) seriously to understate the position, and (6) not to antici- 

 pate an above-average crop. Forecasts issued by The Times are free from this bias. 



The institution of ' Standard plots,' together with possibly the reorganisation of 

 the part-time personnel of the Ministry, might ameliorate both of the above tendencies, 

 viz., to forecast sub-normal crops and to understate actual yields. 



