276 maine agricultural experiment station. i915. 



Discussion oi^ Re:sults oi^ Analysers of Fe;rtii.ize;rs. 



The errors due to the samphng and laboratory examination 

 effect, of course, the results of the analyses as reported. With 

 the best of care, different samples taken from the same goods 

 at the same time and examined by the samie analyst will show 

 slightly discordant results. These may be slightly too high or 

 slightly too low. A variation of two-tenths of a per cent in the 

 actual content may occur. That is, the examination of a sam- 

 ple of goods that actually carried 4 per cent nitrogen might 

 show a result as low as 3.8 per cent or as high as 4.2 per cent. 

 Hence when only a single sample is examined variations to 

 these limits would always be passed. If several samples of the 

 same goods are found to be all or nearly all only slightly low a 

 case could probably be maintained under the fertilizer law, pro- 

 vided the other constituents were not present in sufficient ex- 

 cess to indicate the sample taken did not fairly represent the 

 output. 



Nitrogen and potash are the most costly constituents of 

 fertilizers, and in examining the tables special attention should 

 be given to these constituents. When there is a marked defi- 

 ciency in the goods, the fact is noted in the tables. When 

 there is a shortage of more than .2 per cent in one constituent 

 and a corresponding overrun in others, no comment is made in 

 the tables. 



On the whole the fertilizers of 191 5 are fairly well up to the 

 guaranty. It is to be remembered, however, that these guaran- 

 tees are minimum and not average guarantees, and that com- 

 panies whose goods on the whole run close to their guaranteed 

 analyses may be manufacturing too close to their minimum 

 guarantees for safety to themselves or their customers. 



Formerly the lower figure in a guaranty was usually maixi- 

 tained with a good margin for safety. There seems to be a 

 growing tendency on the part of certain companies to manu- 

 facture too close to the minimum guaranty. Of course in 

 theory a fertilizer should never fall below the guaranty for 

 that is supposed to be its minimum. 



In quite a number of instances 19 14 goods were sampled and 

 the analyses are reported herewith. Wherever it is known to 

 the Station that these were 19 14 goods the fact is indicated in 

 the tables. 



