62 EXPERIMENTS UPON TURNIPS. [Appendix 



lieve is contaminated also with a considerable proportion of saline substances 

 artificially added to it. That it contains many substances useful to plants there 

 can be no doubt, and that it may prove a valuable manure is exceedingly pro- 

 bable, but imdcr its present name it can only lead to false deductions in expe- 

 rimental agricultui'e — and the use of it, therefore, in comparative tiiajs such as 

 these we are now discussing, ought to be avoided. It is only, as I have already 

 said, from the use of pure substances mixed in known proportions, iJiat valuar 

 ble, because undoubted, conclusions can be drawn. It is in vain co attempt to 

 eliminate the effects of diversity of soil and climate, if new causes of diversity 

 are introduced by tlie very substances with which our experiments are made. 



5°, Bones dissolved in muriatic acid. — The action of bones is not in general 

 exhausted in a single season. If they are in the state of fine dust, they decom- 

 pose more quickly and cease to act in a shorter space of time. By dividing 

 them still more minutely, or by solution in an acid, it has been thought that 

 their apparent efficacy might be increased. Mr. Fleming, in 1841, made some 

 experiments which seemed to justify this conclusion. In the present tables 

 other results are exhibited, which favovu the same opinion. I place togethei 

 here the results upon potatoes, as well s,s upon turnips, for the purpose of 

 comparison : — 



Bone^ust. Bones in rmtriatic add. 



These results, the only ones contained in our tables which can be compared 

 together, are both greatly in favour of the dissolved bones, in so far as the action 

 upon the first crop is concerned. It will require longer obsei-vation to deter- 

 mine in which form the same weight of bones will produce the-.more lasting 

 effects — and will be the more economical on the whole. 



6°. NP'Tate of soda. — The effect of 1 cwt. of this salt per acre upon the early 

 yellow turnip is very remarkable (p. 44), having given upwards of 27 tons of 

 bulbs, at a cost of 25s. It is to be regretted that no similar experiment is re- 

 corded upon the other varieties of turnip, either by Mr. Fleming or by Mr. Al- 

 exander. In the text (Lecture XV., p. 335 to p. 342) an abstract of all the pub- 

 lished results hitherto obtained by the use of nitrate of soda will be found in a 

 tabulated form. 



7°. Lime. — An interesting result in Mr. Fleming's first table may hereafter 

 lead to some satisfactory experimental determinations of the points considered 

 still doubtful in regard to the form in which, and the time when, lime may 

 be most efficaciously applied, in reference to the culture of particular crops. He 

 caused carbonate of lime and caustic (newly slaked 1) lime to be sown in the 

 drills without manure, and the effect upon the crop of Swedes was as follows : 

 Soil unmanured ..... 12 tons 5 cwt. 

 Carbonate of lime, 20 bushels . . 16 " 11 " 



Caustic lime, 50 bush els . . 11 " 8 " 



The immediate effects of lime applied in these two forms was veiy different — 

 the caustic lime lessened the turnip crop, while the carbonate increased it by 

 4f tons. This effect most probably arose from the lime, in its caustic state, 

 taking from the soil the carbonic and other organic acids from which the roots 

 in the early infancy of the plant would have derived a portion of their nourish- 

 ment, and thus retarding and stunting their growth. At all events the experi- 

 ment seems to indicate that lime ought to be in the state of carbonate— the mild 

 state — more or less entirely, if it is intended to benefit the crop to which it is 

 immediately applied. When mixed .with manure, however, where vegetable 

 matter abovinds in the soil, or where the lime is merely harrowed into the sur- 

 face — in all which cases it will readily become, in a great measure, saturated 

 with carbonic acid — the skilful fanner will understand that the deduction drawn 

 from the pre^ding experiment will not apply. 



