RFXORDIXG AGRICULTURAL EXPERIMENTS. 419 



equi\'alent to the Daybook of the commercial man, and that 

 to make this practicable the records of these experiments must 

 be kept in a separate book or books. 



Bv using- a carbon-book, in which every experiment to. be 

 conducted at a certain Station is entered and numbered con- 

 secutively, the first object is attained. Full details of the 

 object of and method of each experiment can be entered there 

 in such a way that the carbon copy when torn out and 

 forwarded to the Station with a letter of transmittal, furnishes 

 full instructions to the Station officer. As a rule these can be 

 conveyed in a few lines so that a carbon book of lOO pages 

 mav list over a thousand experiments. 



It is desirable, and often essential to ' the success of, all 

 experiments, that the Station ofticer should understand clearly 

 the object and scope of the various experiments he is carrying- 

 on, and these should therefore be stated clearly and concisely 

 in the instructions. 



For purposes of convenience of reference, it will be desirable 

 to classify the experiments, in an index at the beginning or 

 end of the book. 



The Station Books. 



The carbon copies of the List of Experiments are filed in a 

 self-binding cover by the Station officer, in such a manner that 

 they can be added to from time to time as fresh experiments 

 are planned at the Head Office. 



Record Books. — Where i.ooo or more experiments are being- 

 conducted, the records cannot be conveniently contained in one 

 book, as it would be too large to take out to the plots. To use 

 a daily diary of records involves transferring them to the 

 record books at the close of the day; in the busy season this 

 takes too much time; they soon accumulate, and eventually this 

 work is neglected, with the result that the records are never 

 up-to-date. 



By classifying- the records into lo to 15 groups, a small book 

 of 100 pages can be used for each group, and this can easily be 

 carried to the plot when that particular group is being studied 

 or inspected. These books are lettered A, B, C, etc. 



Like the experiment list, the record books should have 

 duplicate pages and carbons. At the beginning of the season 

 each experiment is assigned to a separate page of the record 

 book of the class to which it belongs, and a cross entry is 

 made, the dominant letter and page of the record book being- 

 entered in the experiment list, against that experiment, and 

 in the record book the consecutive number of the experiment 

 is referred to. Reference is thus made perfectly easy, and it 

 is therefore unnecessary to detail in the record book 

 particulars of the reasons for the experiments or its outline. It 

 is only necessary to detail the actual facts of the expeViment, 

 e.g.. number of block and plot or field; character of soil; date 

 of sowing or planting; distance of sowing or planting; notes on 

 germination and growth; date of harvesting, yields; etc. 



At any time during the season, if an experiment has been 

 completed, or has failed, the carbon copy of the record can at 



