Plot

TABLE VI.-BARN FIELD MANGEL WURZEL.

Average Produce of Leaf per acre, and Proportion of Leaf to 100 Root-27 years, 1876 to 1902.

1

2 In 1895 and since, 500 lb. sulphate of potash also.

1 Omitting 1885 and 1901, when owing to drought, &c., the cross-dressings of nitrate

of soda and ammonium salts were not applied.

Proportion of Nitrogen Recovered to Nitrogen Supplied. 55

In good seasons when the crop of roots is large, the amount of leaf shows little corresponding increase, hardly more than would be accounted for by the comparative absence of blanks and missed plant which characterises a good season. In consequence the proportion of leaf to root is at its lowest in a good year, as may be seen from the figures given in Table II. on page 34, which shows the crop of both root and leaf in 1900. It is evident that once the plant has developed a sufliciency of leaf, the difference between a good and a bad season depends upon the rapidity with which the leaves can do their work of carbon assimilation from the atmosphere, for all the products of that action are at once passed on to the root and stored there, in the case of the mangel chiefly in the form of sugar. A good season does not involve any greater luxuriance of leaf than usual, just as, in a similar manner, plots which grow a small crop of roots because of the absence of alkali salts may yet possess a normal development of leaf.

Conclusion.

(11.) The amount of leaf grown varies with the amount of Nitrogen supplied, but is not much affected by the presence or absence of mineral constituents.

PROPORTION OF THE NITROGEN RECOVERED IN CROP TO THAT SUPPLIED IN MANURE.

In view of the large amounts of nitrogen applied in the manures, it is important to consider what proportion of it is recovered in the crop, in the case of each of the different nitrogenous compounds which are used. For Plots 4, 5, and 6, in each series, the proportion of nitrogen in the crop has been determined each year, so that an exact estimate can be formed of the amount recovered. In the case of the Plots 1 where dung is used, the determinations of the nitrogen in the crop have only been made for nine years, 1878-82, and 1897-1900, inclusive. In these latter cases an average proportion of nitrogen for the roots from each plot has been calculated from the proportions in the crops of the nine years when analyses were made, and this average proportion of nitrogen has been applied to the crop for the whole period, in order to obtain an estimate of the total quantity of nitrogen removed from each plot. The procedure is not strictly correct, but probably does not introduce any serious error, as both the crop and the proportions of nitrogen are less variable on these than on the other plots.

In forming an estimate of the nitrogen recovered, it has been customary hitherto in the Rothamsted publications (vide "The Rothamsted Experiments," Trans. Highland and Agricultural

Society of Scotland, 5th Series, Vol. VII., 1895, page 56) to deduct the nitrogen contained in the plot receiving no nitrogen from that removed in the corresponding plot receiving a nitrogenous dressing, and to assume that the excess only has come from the manure. It is doubtful if this is advisable, for the crop receiving no nitrogenous manure is being grown out of the reserves contained in the soil, together with the small amount of combined nitrogen which falls in the rain; the soil therefore on these plots is getting steadily poorer in nitrogen year by year, whereas there is no evidence which would lead us to suppose that the soil of the plots manured with nitrate of soda or ammonium salts is now any poorer in nitrogen than it was at the beginning, while the soil of the dunged plots is known to be very much richer. If the soil has suffered no loss of nitrogen, then the whole of the nitrogen removed in the crop must have been derived from the manure, and on that assumption the percentages given in the following Table are calculated. In calculating the amount of nitrogen recovered each year no account has been taken of the leaves, because they are returned to the soil and their nitrogen is not removed from the land; if both leaf and root were taken into account the recovery of nitrogen for any single year would be very much greater; indeed in some seasons when a big crop is grown more nitrogen is removed in the roots alone than was supplied in the form of manure. Table VII. shows the nitrogen supplied and removed from Plots 4, where there was a full supply of mineral manures, and from Plots 1 where dung was used with nitrogenous manures.

The results show that both the nitrate of soda and rape cake are very effective manures, about three quarters of the nitrogen they supply each year being recovered in the roots removed from the land; the ammonium salts, and ammonium salts mixed with rape cake are less effective, the recovery being between fifty and sixty per cent. of that applied. On the plots receiving dung, the proportion of nitrogen recovered at once becomes very much less, sinking to about one third of that supplied in the manure. It is known that there is a very large accumulation of nitrogen in the soil of these continuously dunged plots, though not sufficient to make up all the difference between the nitrogen supplied and that removed in the crop. Of the nitrogen unaccounted for, some has been washed as nitrate into the subsoil and some liberated as nitrogen gas by the agency of bacterial changes.

Thus, when dung and nitrate are used 115.8 lb. of nitrogen is recovered in the crop as compared with 63.3 lb. recovered from the dung when used alone; if we deduct the 63-3 lb., as due to the dung, from the 115.8 lb. we obtain 52.5 lb., which

Proportion of Nitrogen Recovered to Manure Supplied. 57

may be taken as the return from the nitrogen of the nitrate of soda when it is used in conjunction with dung. This amounts to 61 per cent. of the 86 lb. of nitrogen supplied, a proportion which compares favourably with the proportion recovered from nitrate of soda when used with a mineral manure only, if we take into consideration the fact that a much bigger crop is being grown with the two manures in conjunction than with either singly, and, as we have seen before, it is the smaller applications of manure which give the best proportionate returns.

Plots 4.—Superphosphate, sulphates of potash and magnesia, and common salt.

1 Average for twenty-one years, omitting 1876-7, 1885, 1895, 1901-2.

2 Average for twenty-two years, omitting 1876-7, 1885, 1901-2.

3 Percentage calculated from nine years, 1878-82, 1897-1900.

These results showing the large proportion of the nitrogen of nitrate of soda and other nitrogenous manures that is recovered, even when they are used in large amounts with dung year after year on the same land, lend no colour to the opinion that in ordinary farming there is likely to be serious loss of nitrogen by "denitrification" when nitrogenous manures and dung are used together. The soil is undoubtedly always suffering losses of nitrogen in the gaseous state through various

bacterial changes which can be roughly grouped together under the term "denitrification," and these losses will increase the higher the condition of the land becomes and the more easily decomposable nitrogenous bodies are present. But there is no evidence in the Rothamsted mangel experiments to support the view that specific and excessive loss will set in when dung and nitrate of soda, or other active nitrogenous manure, are used together.

THE CROP AND SEASON.

The mangel crop varies from year to year to a greater degree than do most of the other staple crops of the farm, this being doubtless due to its comparatively short period of growth and its dependence upon suitable weather conditions at the critical time of germination. If the subsoil has accumulated a reasonable supply of water in the winter and early spring it is generally recognised that a hot season, even accompanied by periods of drought after the plant has been established, is most favourable to the growth of mangels.

Attempts, however, to correlate the weight of the crop with the records of rainfall, temperature, and sunshine at Rothamsted, for the portion of the year during which the crop occupies the ground, have failed to give any results of general application, even when the observations are grouped into weekly periods from the date of sowing to that of lifting the crop. In the first place everything depends upon securing a good plant soon after the seed has been sown, and the critical conditions of tilth, moisture, temperature, and evaporation, which prevail about the time of seeding, are too minute and intricate to be capable of deduction from the ordinary meteorological observations. Even when a good plant has once been obtained it is difficult to trace from the records what have been the dominant factors in producing a large or small crop. For example, the best mangel crop at Rothamsted was grown in 1900, a year of comparative dryness, great evaporation, and much sunshine; the current year, 1902, has given a crop but little smaller, yet the character of the season has been entirely different, for it has been a cool summer, with much cloud, little sunshine, and frequent though small rains, except for one period of heavy rains towards the end of August and beginning of September.

Putting aside the question of failure of plant, even in the good seasons and on the dunged plots where germination is assured, no systematic connection can be traced between the weight of crop harvested and the records of temperature, rain, sunshine, &c. The meteorological observations made at present do not enable one to estimate the prevalence of what the farmer knows as "good growing weather."