236 WHEAT SOILS OF TH1-: ALKX A NDRI A DIVISION. 



9. Brack (Alkali) salts have not been observed on an}- of 

 the lands represented by the Alexandria samples, with one ex- 

 ception. In no case, however, did analysis show any markea 

 ])roportion of snch salts in the samjiles of soil collected. 



10. The samples examined are not rich in organic matter, 

 and. from this cause and their rather sandy nature, they are not 

 very retentive of moisture. 



11. Bacteria are now generally supposed to be the active 

 agents in converting into humus for the requirements of plants 

 anv organic matter present in the soil, but without organic matter 

 even the presence of the necessary bacteria becomes nugatory. 



12. Of the usual constituents of plant food, phosphates in 

 particular are deficient, and potash next in order. For wheat 

 culture more nitrogen in the soils is desirable. 



13. Magnesia is present in all the soils in (|uantities sufficient 

 as a plant food, and not large enough to act injuriously. 



14. Comparing the proportions of plant food in the culti- 

 vated and in the virgin soils, there have been losses of magnesia 

 and lime during the .course of cultivation in almost every case, 

 but the soils still contain sufficient of these constituents to render 

 their addition unnecessary. Losses of potash and nitrogen were 

 less general, but these were not present in the soils in such relative 

 abundance, and for the mo.st part their disappearance from the 

 soil must be made good. Phosphoric oxide has markedly dimin- 

 ished in onlv four cases out of the ten, and in two of these cases 

 it could well be spared; most of the other soils are verv deficient 

 in phosphates. 



15. The causes of ina(k'(|uatc ])roduction, therefore, seem 

 to be — ( 1 ) the rather sandy character of some of the soils, con- 

 joined with their inherent ])overty in plant-food constituents; 

 (2) \\)v rcnio\al of some of those constituents by continuous 

 croi^ping without manure; and (3) the further losses caused by 

 the surface soil sutTering (le])letion in rcs])cct of silt and clav. 

 The moisture conditions of the soil have not been investigated. 



i^. ( )ut of. the ten localities investigated, only one is not 

 in immediate need of manuring of anv kind, in order to fit it 

 for wheat-production, liight re(|uire manuring with nitrogen, 

 five need potash fertilisers, and six need fertilising with phos- 

 ]ihates. 



17. The fertilisers that would be desirable are nitrate of 

 soda, sul])liate of i)otash and basic slag, in ])roportions suited to 

 the re((uirements of the respective soils for these fertilisers. 

 Under present conditions it would j^robably be necessary to use 

 kraal manuic and super])b()S])hate. with a to])-dressing of suir 

 phate of ammonia. em])loying a nitrogen-gathering catch cro]) to 

 provide the bulk of the nitrogen. 



