1905.] 



Improvement of Mangels. 



355 



The success of this method of selection is shown by the 

 following figures, and side by side with them are printed figures 

 for the percentage of dry matter in mangels : — 



Year. 



Sugar Beet. 

 Sugar in Juice. 

 Per cent. 



Year. 



Mangels. 

 . Dry matter. 

 Per cent. 



1 860-1 86 1 ... 



10-93 



1852 





1868-1869 ... 



ii*34 



1880-1884 ... 



10*97 



1870-1872 ... 



1 1 -8 b 



1885-1889 ... 



1 1 78 



1873-1874 ... 



12.65 



1890-1894 ... 



13*04 



1882-1883 ... 



13*60 



1895- 1900 



11-80 



1885 ... * ... 



14*00 



1902 



12*9 



1886 



1 5 *co 



1903 



ii*8 



1889 ... 



I5'°4 



1904 ... 



12*3 



These figures bring out several important points. For in- 

 stance, the steady improvement of the sugar-beet from about 

 1870, when chemical selection was established on a satisfactory 

 basis, is very apparent, and contrasts markedly with the con- 

 stancy in the composition of mangels during the last fifty 

 years. 



The sugar-beet has been selected for a definite purpose, and 

 great improvement for that purpose has been brought about. 

 The mangel has been selected also, but selection has been 

 made for such external characters as shape, colour, size, rather 

 than for improved chemical composition. The result is that we 

 now have many strains of improved shape, colour, size, and so 

 on, but the average percentage of dry matter remains much as 

 it was fifty years ago. 



The question now arises : Is it possible to improve the mangel 

 in quality, and, if so, how must the selection be carried out ? 



The first step, Messrs. Wood and Berry point out, is to 

 decide what particular quality we want to improve. Mangels 

 are grown almost entirely for food for cattle and sheep, and, 

 unfortunately, there is very little definite information to be found 

 as to the feeding value of the separate constituents of roots. It 

 would appear, however, that selection for high content of dry 

 matter would probably improve the feeding quality. 



A careful comparison was made of the various characteristics 

 (i.e., weight, dry matter, sugar, specific gravity of the juice, 

 colour of the juice, percentage of total nitrogen and of proteid 



F F 2 



