AGRICUI.TURAL BOTANY, CHEMISTRY AXD PHYSIOLOGY OF PLANTS I43I 



2. IT 9 to 2.370 per cent of nitrogen in the dry matter, the latter only con- 

 tained from 0.724 to 0.924 per cent. The phenomena can also be interpre- 

 preted as a stimulation to over-production of flowers correlated with a scar- 

 city of nitrogen in the flowering branches. 



The present investigations were directed to a further study of the con- 

 tributing causes and more particularly to the part played by the " myco- 

 tropic " rootlets as nitrogen storers. These " mycotro])ic " rootlets or 

 rootlets or endotropic mycorhiza are caused by the presence of a symbiotic 

 fungus in the cortical parenchyma of the root ; thej^ can like normal rootlets 

 absorb nitrates from the soil solution M' means of hairs ; but where 

 as the nitrates absorbed b}^ normal rootlets are passed directly through 

 the cortical parench3'ma to the distributing vessels of the central cylin- 

 der, the nitrates absorbed by mycorhiza are immediately changed to com- 

 plex organic compounds for the benefit of the endophitic mycelium which 

 feeds on these substances and on the starch found in the adjoining cells. 

 A large number of experiments were carried out from which the follow- 

 ng results may be summarised : 



(i) In a soil containing little organic matter, whenever nitrates were 

 formed in small quantities, i. e. lower than the usual percentage in non- 

 manured soils, mycorhiza were found in large numbers, their development 

 varying inversely with the amount of available nitrogen in the soil. Also 

 there was always a proportional reduction in the growth of normal 

 rootlets. No nitrate reaction (i) was obtained from the mj'cotrophic roots 

 in these cases. 



(2) When nitrates were present in quantities up to but not exceeding 

 those usually present in a good non-manured soil, about one third of the 

 rootlets were transformed into mycorhiza. No nitrate reaction was ob- 

 tained from the mycorhiza. 



(3) When nitrates were present in large quantities, very few my- 

 corhiza were formed and those few gave a positive nitrate reaction. In 

 these cases the rapidity of growth of the normal rootlets prevented the fun- 

 gus infection to a large extent, and even where the fungus managed to 

 penetrate into the root and to develop, the mycorhiza was not able to 

 seize and fix all of the nitrate absorbed by the root. 



These facts demonstrate the effect of endotrophic mycorhiza on the 

 olive trees with regard to the interception of nitrates and identical results 

 have been obtained when the nitrogen was absorbed in the form of am- 

 monia. IMoreover once the nitrogen has been converted into organic 

 compounds only a minute portion of it ever goes back into circulation in 

 the host plant, even when the intercellular h^-phae which are rich in albu- 

 minoids are subjected to autolysis or digestion. 



The interception of nitrogen must accentuate the ill effects of a poor 

 soil on olive trees more especially as in such soils the development of 

 mycorhiza is particularl}- active. Practical trials have shown that on well 

 nianured soils only 30 to 40 per cent of the rootlets are transformed into 



(i) Nitrates were tested by the Mol'Kch iiuthod (£</.). 



