i68 su(;ar i'.ket in south afru a. 



work carried out subsequent to the previous publication. In 

 the Grahamstown branch of the Cape Government Laboratories 

 investigations had been begun in 1905, and, in addition to those 

 above mentioned, samples from Kingwilliamstown, Bedford, 

 Albany and Graaff-Reinet had been chemically examined, and 

 the practical growing of beet had been the subject of preliminary 

 investigation in the Molteno and Knysna districts as well as on 

 the Cape Flats, and at that stage the matter remained. Dr. 

 Nobbs, who had initiated the i)ractical experiments, left the 

 Cape Service, and his experiments came to an end. They had 

 shown all the more clearly that the country was capable of ]jro- 

 ducing good sugar beet, but the problem that still remained un- 

 solved was whether beet croi)s of high sugar content can be 

 grown in sufficiently large quantities to make sugar extraction 

 commercially profitable. 



The purpose of the present paper is to record further ex- 

 l)eriments that have been carried out subsequently to those pre- 

 viously mentioned. 



In September, 191 1, three samples of sugar beet grown in 

 the Eastern Districts were analysed, but yielded only small pro- 

 portions of sugar, being apj^arently not yet properly matured. 

 The figures were as follows : — 



During April, 1912, two further samples of sugar beet, 

 experimentally grown by the (Tovernment Agriculturist in the 

 Humansdoq:) District, were examined, and these, too, were 

 aj)parently immature, as the following results show: — 



Sugar. 

 Variety. per cent. 



Vilmorin's improved vvliite sugar beet 5.84 



Danish improved sugar beet ... 9.30 



Privately conducted experiments in sugar beet cultivation 

 in the Eastern Districts gave much higher figures. Sam])les of 

 beets were taken at intervals of a month, beginning about the 

 middle of April, 1912, and yielded the following percentages of 

 sugar : — 



Percent- 

 Sample taken in \])ril 16.64 



Sample taken in May 14.02 



Sample taken in June 17-46 



In each of these cases the analysis represents an average of 

 five to eight beets. 



Four months later analyses were made of beet from the 

 same source as the last, but grown in the winter season. Two 

 separate lots were analysed, and gave respectively 7.31 per cent, 

 and 8.93 per cent, of sugar, but another sample taken in Novem- 



