170 



Analyses of Ashes of Plants. 



from the bulbs at the extreme point of the latter, in such a way 

 however that the base of the leaves was always sufficiently com- 

 pact to hold them together when severed from the bulbs. At 

 this particular point, the vegetable matter partakes of the cha- 

 racter and consistency of both the bulb and the leaf; it may be 

 thought by some that the separation should have taken place 

 higher, by others lower in the plant ; and there is no doubt that 

 estimations by two different persons would not perfectly agree, 

 with whatever care they might be effected. We claim for our 

 results, however, the merit of uniformity. They are compara- 

 tively, if not absolutely correct. The quantity of turnip and 

 turnip-top weighed, varied from 30,000 to 100,000 grs., and 

 generally included three or four bulbs. In order that the extent 

 of variation may be seen, a table is here given of all the turnips 

 examined together, with the quantity of ash found in their bulbs 

 and tops. The swedes and hybrids are included with the turnips 

 (properly so called), because, as will be seen^ there is little or no 

 distinction between them, as far as mineral constitution is con- 

 cerned. (See Table I., following page.) 



It will be seen from this table, that the proportion of bulb to 

 top is very variable ; in one specimen (No. 86) bemg as 62 to 38 

 in 100 parts of the entire plant ; in another instance (Spec. 98), as 

 96 to 4. These numbers represent, however, the extremes, so far 

 as the specimens we have examined are concerned. Out of 30 

 specimens enumerated, 24 or 4-5ths range within the limits of 

 84 and 94 of bulb to 16 and 6 of top in the 100 parts. The 

 mean of all the specimens is 86 8 of bulb to 13*2 of top ; but if 

 those which exhibit the greatest deviation are omitted (Specimens 

 84, 86, and 94), we shall obtain for the remaining 27 specimens 

 a mean of 89 to 11. 



These numbers would indicate a very much smaller propor- 

 tion of top than is usually believed to obtain. As was before 

 said, we do not wish more reliance to be placed upon the estima- 

 tions than is their due ; but there are good reasons for believing 

 that they are not far wrong. Thus, 4 specimens of Skirving's 

 swedes (78, 79, 80, 81), grown at the same time on the same 



19. The appearance of the crop. 



20. The time at which it was collected. 



21. The yield per acre fin roots. 



perches being weighed tin tops. 



It is very desirable that precise information should be obtained with 

 regard to the crops intended for analysis. With this object, it is wished 

 that the questions above be answered 2i?> fully as possible. If, however, 

 any difficulty should arise, or any question appear objectionable, the an- 

 swer may be altogether omitted. 



