.'546 JOUENAL OF THE DEPARTMENT OF AGRICULTURE. 



In iN'atal the largest acreage is in New Hanover — 817 acres — 

 but the production of 2,595,900 lb. corresponds to a yield of only 

 3177 lb. per acre. The other large sweet potato districts in Natal 

 are Durban and Lower Tugela, for which the statistics are given 

 below : — 



Acres. Production. Yield per Acre. 



Durban 580 3,800,400 lb. 6553 lb. 



Lower Tugela ... 525 2,004,000 lb. 3817 lb. 



The only really large sweet potato districts in the Transvaal are 

 Barberton and Rustenburg, for which the figures are as follows: — 



Acres. Product imi. Yield per Acre. 



Barberton 901 3,994,200 lb. 4433 lb. 



Rustenburg 658 2,027,400 lb. 3081 lb. 



In the Division of George, the chief sweet potato growing district 

 of the Cape Province and of the Union, 874 morgen (1850 acres) were 

 under sweet potatoes in 1918, and produced 31,586 muids (4,737,900 

 lb.) of the tuber, equivalent to a rate of 2561 lb. per acre. It is a 

 matter of common knowledge that the soils of George are much poorer 

 than those of Oudtshoorn, but the relatively low yield of the former 

 Division is not accounted for by that fact alone. To a large extent 

 the method of cultivation is promiscuous and unscientific; with more 

 intensive methods the area under cultivation in this division should 

 obviously be capable of producing very much heavier crops, and as 

 an illustration of what can be accomplished in the Little Brak 

 District, Mr. Harold Robertson related the following: "From a 

 piece of ground, measuring 425 yards long and 80 yards wide, a yield 

 of 1800 50-lb. baskets of large tubers and approximately 450 baskets 

 of small tubers was obtained during last harvest ; that is say about 17 

 tons per morgen* in all. The soil was black alluvium and had been 

 irrigated before ploughing. It was manured with half a ton of 

 commercial " potato fertili^i^oi'," guaranteed of the following com- 

 position : — 



Phosphoric oxide: Total 8.0 pei' cent. 



CUtric-solublo 7.5 ,, ,, 



Nitrogen 3.5 ,, ,, 



Potash 6.0 ,, ,, 



Lime 16.0 ,, ,, 



The season was dry, and the crops had received no subsequent water 

 other than occasional showers. 



* 16,300 lb. per acre. The difference between this return and the low 

 census figures quoted above is both striking and significant, and seems to 

 imply that the latter figures are in general much below the actual yields. 

 Various causes seem to combine together towards this end. First of all, there 

 is a liability to omit from the estimation of the yield any produce not sold 

 off the farm but used for home consumption, and crops raised by bywoners 

 may quite easily also remain unrecorded. Next, a natural tendency exists 

 amongst agriculturists to over-estimate areas under cultivation. Then, too, 

 they frequently have a reluctance to state their full yields lest such candour 

 should prove the precursor of taxation. The farmer is apt^ moreover, to 

 include in his statements not only the areas actually sown during the preced- 

 ing season, but also fallow lands previously under the same crop. All these 

 circuiTistances contribute to lower the recorded yield per acre. 



