179 



.cwt. per acre) were satisfactory, the others less so ; and as yet there was 

 no decided effect from inoculation. By July 7 the plants, except those 

 inoculated with peasoil, where mostly yellow and unhealthy looking, 

 and generally no nodules were to be found on their roots ; only a few of 

 them were normally developed, and these had nodules. On the other 

 hand, the plants, both beans and peas, growing on the plots inoculated 

 with pea soil were strong, dark green, and healthy generally ; all of 

 them had numerous nodules on their roots. 



The effect of inoculation with pea soil was more marked on the burnt 

 portion of the field than on the other On the unburnt plots the num- 

 ber of stronger plants (beans and peas) increased (July 14) where no peat 

 soil, but 32 cwt.. of lime per acre, had been applied; but the majority of 

 the plants on these plots remained yellow, and without nodule forma- 

 tion. 



On July 28 the relative differences remained the same for the different 

 plots. 'iowards the end of August the horse beans were all attacked 

 by rust, seed production being thus much hindered, and the plants had 

 to be harvested before ripe, so as not to damage the next crop (winter- 

 rye). The weather had been very wet, but the delay in ripening may 

 have also beten due to the symbiosis, which would be in accordance with 

 observations made by Hellriegel and by Nobbe. 



The table below gives the total produce, generally averages of two or 

 more plots. 



Lime applied Inoculating Produce (lb.) Produce (lb.) 



.cwt. per acre. soil. on burnt soil. on soil not burnt. 



32 Not inoculated 103.14 166.59 



24 " " 111.72 



16 " " 48.69 



32 Pea sandy soil 237.14 255.02 



24 " 220.14 



16 " " 97.15 



32 sea mud 76.84 



32 lupin soil 96.49 



The results show that in absence of stable manure, peaty soil poor in 

 nitrogen, and not containing the micro-organisms necessary to enable 

 leguminous plants to utilise free nitrogen, may be cultivated after 

 inoculating with a suitable soil in the quantity of 16 cwt. per acre, 

 such inoculation being more essential and more productive of good re- 

 sults, in freshly burnt soils, than on land which has not been recently 

 burnt. The selection of suitable soil for inoculation is again shown to 

 be of great importance, lupin soil being, in these experiments, useless 

 when applied for peas and beans. Finally, the presence of su£B.cient lime 

 is clearly shown to be essential for the production of a good crop. Ap- 

 plications of 16 cwt. per acre were quite insufficient, and better results 

 were obtained with 32 cwt. than with 24 cwt. per acre on the peaty soil 

 on which these experiments were made. 



After the harvest, another field was inoculated by means of some of 

 the surface soil of the successful plots (i.e. those which had received pea 

 soil), and peas and beans were sown in it. The plants developed well, 



