248 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



Here again there is a gain from the inoculated over the uninoculated 

 seed amounting in the aggregate to 13 per cent., but a considerable part of 

 this gain is more apparent than real, on account of the variation in the 

 number of plants of 1 Maincrop.' If this variety be ignored the gain is in 

 the aggregate only 3 per cent. 



These four plots may be compared as in the series on the cultivated 

 land, and we find that the smallest yield was on the untreated plot. 

 When the soil alone was inoculated there was a gain of 8 per cent, in the 

 number of pods, 2 per cent, in their weight, and 5 per cent, in the 

 weight of the peas they contained. When the seed alone w r as inoculated 

 there was a gain of 18 per cent, in the number and weight of the pods, 

 and 17 per cent, in the w T eight of the peas. When both seed and soil 

 were inoculated the gain over the untreated plot was 15 per cent, in the 

 number of pods, 13 per cent, in their weight, and 20 per cent, in the 

 weight of peas they contained. 



The small number of plants of 1 Maincrop ' on the untreated plot 

 interferes with the comparison, and if we leave this variety out entirely we 

 find the results are much more even (Table J). 



TABLE J. 



Total A N I > COMPARATIVE WEIGHTS OF PRODUCE OF THRFF VaRIF.TIFS ON 



Plots XIII.. XIV., XV., XVI. 





No. of 



Wt. of 



Wt. of 



Percentage difference 

 from untreated plot 





pods 



pods 



peas 



No. of Wt. of 

 pods pods 



Wt. of 

 pens 



Plot XIII. 





grs. 



grs. 







Soil only inoculated 



1,560 



14,376 



5,613 



0 -1 



-1 



Plot XIV. 













Soil and seed inoculated . 



1,645 



15,750 



6,260 



+ 5 +5 



+ 10 



Plot XV. 













Seed only inoculated 



1,622 



15.47S 



5,963 



+ 4 +3 



+ 4 



Plot XVI. 













Untreated .... 



1,558 



14.903 



5,667 







Taking these three varieties alone then, we find that when the soil 

 alone was inoculated there was a slightly diminished crop, while when 

 the seed alone or both the soil and seed were inoculated the return was 

 only slightly increased. 



T.i king these facts into consideration, and bearing in mind the fall off 

 in the soil towards the upper end of the series, we cannot but conclude 

 that here again there was no advantage derived from inoculation. 



