1899.] PUBLIC DOCUMENT — No. 38. 67 



It is important to point out that none of the crops in- 

 cluded above has in any year occupied the whole of the area 

 under experiment. Each year we have had some four or 

 five crops, and the areas in each have varied. The aJjove 

 figures are valuable, then, solely as a basis of comparison 

 between the several plots. 



Conclusions based on Results uj) to 1897, 

 The chief conclusions which seemed justified by the results 

 above o;iven are the followino; : — 



1. Sulphate of potash in connection with nitrate of soda 

 (Plot 5) has generally given the best crop. In those cases 

 where this has not been true, the inferiority of this com- 

 bination has usually been small. In one case only has it 

 fallen much behind, viz., with sweet corn, a crop which 

 makes much of its growth in the latter part of the season. 



2. Nitrate of soda (plots 2 and 5) has in almost every 

 instance proved the most valuable source of nitrogen, 

 whether used with muriate or the sul])hate of potash. 



3. The combination of sulphate of ammonia and muriate 

 of potash (Plot 1) has in every instance given the poorest 

 crop. This fact is apparently due, as Dr. Goessmann has 

 pointed out, to an interchange of acids and bases leading to 

 the formation of chloride of ammonia, which injuriously 

 aflects groM'th. 



Tlie Experiment in 1898. 



In the fall of 1897 the plots were ploughed, and rye sown 

 on all (without further manuring) as a cover crop, chiefly to 

 prevent soil washing. The gro^vth on Plot 1 (sulphate of 

 ammonia and muriate of potash) was sickly and fcel)le, but 

 no particular dillcrence was noticed between the other plots. 



Ghawje in Plan. 

 In view of the fact that market gardeners, in whose inter- 

 ests chiefly these experiments are being carried out, almost 

 invariably use large (juantities of sta1)le maiuire, and employ 

 fertilizers, if at all, simply to sup})lement the manure, it was 

 decided to make a change in the plan of the experiment, in 

 order that the conditions under which we are working may 

 more nearly conform to those of the average market gardener. 



