PASTURE TOP-DRESSING IN CANTERBURY.

New Zealand Journal of Agriculture, Volume XXX, Issue 6, 20 June 1925, Page 393

PASTURE TOP-DRESSING IN CANTERBURY.

EXPERIMENTS AT SOCKBURN AND TAI TAPU, 1923 and] 1924.

F. E. WARD, H.D.A.,

Instructor in Agriculture, and A. W. HUDSON, B.Agr>,

B.Sc., Assistant Instructor in Agriculture, Christchurch.

In the more climatically favourable districts of New Zealand topdressing of pastures has been for many years an established practice, but, under Canterbury conditions of light rainfall and short-rotation pastures top-dressing has not generally been practised. In co-operation with the Canterbury Soils Improvement Committee, carefully planned experiments were arranged in 1923, and were continued by the Fields Division in 1924. The results of those undertaken in 1923 are now recorded, one experiment being continued into the second year.

MANURES USED.

The manures used were as follows, the quantity stated being per acre in each case :

Season 1923 : (1) Superphosphate (42-44), 2 cwt ; (2) superphosphate (42-44), 11 cwt., plus dried blood, | cwt. ; (3) basic super (40-43), 2 cwt. ; (4) basic super (40-43), T| cwt., plus dried blood, | cwt 4

Season 1924 : As in the case of the manures used in the wheat manurial experiments, results of which were published in the April Journal, it was decided to apply equal amounts of phosphate (2 cwt.) on all plots, the blood being an addition on those plots where it was used. The object in the selection of the manures was to test soluble phosphate (super) against the less soluble phosphate (basic super), and to note the effect of dried blood used in conjunction with these phosphates.

SEASON 1923 EXPERIMENTS

(1.) On Farm of T. Bloor, Sockburn.

The soil of this area is a rich silty loam. The paddock was sown down in September, 1921, with a mixture of 30 lb. perennial rye-grass and 5 lb. cow-grass. Hay was taken off in 1922, the paddock being then grazed until closed in October, 1923, for the top-dressing experiment. The manures were applied on 6th September, 1923, and the plots weighed early in December. Each of the plots was four widths of the drill and the length of the paddock. , The four manures mentioned and a control plot (no manure) constituted a series. There were four series, giving twenty strips in all. ,

On certain portions of the field distinct differences between. manured plots and controls could be observed before harvest, while on other parts no differences in growth were discernible.

Six areas of acre were weighed on each plot, giving twenty-four weighings in each treatment. The ratio of hay weight to green weight was not determined, but, in light of estimations made this year, it may safely be assumed that this would be about 1 to 3, and for purposes of evaluation of increases the hay weight is regarded as being one-third of the green weight. The table which follows shows the results obtained.

The results as given in the table show that only a small margin of profit can be claimed from the use of two of the manures. " These results, however, only record the effect on three months’ growth, and if the increase ,in hay pays for the manure it is highly probable that the benefits .will be quite considerable ultimately.' .

A study of the weighings taken twelve months : later in W. G. Macartney’s experiment showed a very definite belated effect from the manures of 1923 application. A consideration of the 1924 experiment on. this farm and shown later herein indicates that the application, of these manures has, paid handsomely. The lateness of sowing, coupled with the extremely dry conditions which prevailed during the growth of the hay crop, no doubt limited the effect of the fertilizers on the early growth. ; L?

(2.) Experiment on Farm of W. G. Macartney, Tai Tapu

The soil of this area varies from a clayey loam at one end to a silty loam at the other. Sowing of the pasture took place in the autumn of 1922, the mixture consisting of 20 lb. Italian rye grass, 20, lb. perennial rye-grass, and 6.1 b. cow-grass per acre. The manures for the experiment were applied early in September, 1923, and the plots weighed in December. The lay-out of the plots was similar to that in T. Bloor's experiment, and may be seen in the . accompanying diagram (Fig. 1— season 1923 plots). . .

7,,<; 1n this experiment ten areas of acre each were weighed in each strip, giving forty weighings in each treatment. Since the differences in yield which occurred are in no case significant, only the. green weights, per acre are shown. . In this case, however, the chances of super being better than control are. 22 to 1. We do not accept these odds as certainty, but regard them as indicating a good chance of the differencebeing a real one. .. ... ./, ■ .

The yields on the different treatments were as follows, the amounts stated being green .weight per . acre Control, 2-83 tons ; super, 3-26 tons ;.. super and blood, 2-94 tons ; basic, super, 2-73 tons ; basic super and blood, 2-83 tons. ; . V./A ■ ■ ciAA

These figures will be found very interesting when' the reader has perused the results of the weighings made on this area in 1924. From the 1923 figures, however, it must be concluded that in the first hay crop the application of the manures did not pay. The appearance of the paddock later in the season, when the subsequent crop of red clover was growing for seed, showed that the manures were having some effect. Mr. Macartney was of the opinion that distinct differences could be seen in places.

On 4th August, 1924, the field was visited by the writers. A casual examination. revealed the fact that certain plots had a much greater amount of clover stubble on them than the others. These plots proved to be controls, and so distinct was the difference that the edge of each control could be seen from one end of the paddock to the other. A closer examination showed a much better sole of rye - grass on the manured areas ; the grass looked much darker and more vigorous. No difference could be detected between the clover contents of the plots. The extra stubble on the control areas shows a decidedly smaller amount of grazing on these plots, indicating in all probability, less growth and a lower palatability. Undoubtably this is an important factor, and one the value of which cannot be readily determined.

Mr. Macartney having signified his willingness to give the field further trial, it was then decided to repeat the experiment as described below:—

SEASON 1924 EXPERIMENT ON FARM OF W. G. MACARTNEY

As indicated previously, a crop of red clover for seed was taken off the area after the hay crop of 1923. In 1924 the main constituents of the pasture were perennial rye-grass, suckling-clover, red clover, and white clover; and in patches creeping-fog and black medick had made their appearance. The plan of experiment shown in Fig. 1 was decided upon in order to try and ascertain the following points : (1) The belated effect of the 1923 dressing on the yield of 1924 ; (2) the effect of dressing for two consecutive years with the same manures; (3) the effect of the 1924 dressings on the controls of 1923 ; (4) the effect of the 1924 dressing on top of that of 1923. ■ As will be seen, the dressings of 1924 were made across those of 1923.

The manures were applied on 20th August, and the paddock was closed for the hay crop on that date. On 23rd October the plots were closely examined, but no differences between manured plots and controls could be distinguished. At this stage it appeared as though the results would be a repetition of those of the previous year.

The area was cut and weighings made on 2nd and 3rd December, at which time what appeared to be slight differences could be detected in places only. This is a noteworthy fact, and the results go to show that differences may be present without being easily seen.

METHOD OF MEASURING, CUTTING, AND WEIGHING PLOTS.

The reader is referred to the plan (Fig. 1) where,,sighting-poles are indicated by crosses. For measuring purposes the poles- were placed on a line which marked the end limits of ■ the plots to be weighed. It will be seen that by standing anywhere on this line (fortexample, line AB), and sighting in either direction (according to whether the

observer is one side or the other of the centre pole), it is possible to determine the end limits of the plot without measuring during the weighing. Every plot was marked out as indicated, and the area of each was 28 3 acre. By cutting through the crop at right angles to the line AB, and along the junction of the two middle plots, four sides were available on which to work. This arrangement is important, as it enables the farmer who is co-operating in the experiment to carry on his harvesting operations with as little delay as possible. In the experiment under review only two sides were worked, as the weighing staff was small, and the field contained an ample area outside the experimental portion on which cutting could be done while weighings were in operation. A straight cut was made along the edge of the first plot to be weighed, and cutting proceeded in the usual way, except that the mower ran empty along the headlands. As soon as the cut nearest the middle of each of the strips of 1924 dressing was reached, particular care was taken to make the cut a constant width throughout its length. The accuracy with which such a cut could be made was remarkable, the width rarely varying by more than 1 in. either side of the average width of 4 ft. 3 in. As soon as the cut was started, the weighers followed the machine. The grass on each plot C, D, E, F (see plan) was raked to the middle, rolled into a sheet, and weighed as shown in Fig. 2. No strip was cut until the previously cut one had been weighed. This method ensured the material being weighed under as uniform conditions as were practicable and before drying out took place.

METHOD OF ESTIMATING PROPORTION OF HAY WEIGHT TO GREEN WEIGHT.

■ Ten samples, each of 10 lb., were weighed to the nearest ounce immediately they were cut. The sampling was done indiscriminately so that a fair average could be obtained. These samples were dried separately under as uniform conditions as possible, and when they were considered to be as dry if not drier than ordinary hay each was again carefully weighed. The average percentage of hay was then calculated, the figure obtained being 36-8 dz 0-73 per cent. The figure following the T: sign is the “ probable error,” and may be regarded as a measure of the reliability of- the average with which it

numbers which are referred to when discussing the tables of results in the article. The figures in the middle of each rectangle are actual green weights per plot of a typical section of the field, given in order to indicate the variation which occurs among plots of similar treatment ; hence the necessity for a large number of weighings. The plots running across the page are those of the 1923 treatment, and those running from top to bottom are the 1924 treatment, It will be seen that certain plots (for example, where a control of 1924 crosses a control of 1923) have had no manure at all during the two years. These plots are referred to as double controls,” and Nos. 6 and 21 are examples. Other plots which have had treatment with, the same manure for the two seasons, as where super 1924 crosses super 1923 (for example, No. 2), or where basic super 1924 crosses basic super 1923 (for example, No. 13), are referred to as “double super or double basic super respectively. In every case comparisons have been made between plots situated alongside one another or . separated only by : one intervening plot. The crosses (x) indicate the . position of sighting-poles, the use of which obviated the necessity of measuring every plot during weighing.

is associated. By reference to a table of odds we know that it is .13 to 1 (here accepted as practical certainty) against any other similar set of samples varying from the one already obtained, by more than 2-7 times the probable error. It can be said, therefore, that the percentage of hay weight is 36-8 and certainly not less', than 34-88- — that is, 36-8 (0-73 x 2-7) of the green weight. . For purposes of evaluating' the increases we have taken the percentages of hay as being 34’8, this allowing a margin of safety. • The hay weights as calculated are shown in Tables 2 to 5. ■ ‘ Table 2.—Belated Effect of 1923 Dressings on Yields in 1924. Notes.—The manurial treatments are compared with double controls (i.e., plots which have received no manure at all during the two years) or with one another. For example, in the case of super, Plot 7 is compared with Plot 6 (see plan), and Plot 22 with Plot 21, Plots 7 and 22 having been dressed with super only in 1923, and Plots 6 and 21 having received no manure at all: ■ The number of plots compared is given in the second column ; for example, the yields and differences between super and control are estimated from thirty-two paired plots, &c.. Hints on reading this and following tables : When a control is shown between two manures both manures are compared with the same controls. When the manures are compared with one another directly it will be noticed that in some cases the yield for a certain manure is slightly different in one place from that shown in another. . This is due to . the fact that a different number of plots has been used to estimate the average difference in each case. For instance, where super and blood in Table 2 is compared with control, the yield and difference are estimated from twenty-four paired' plots. Where,' however, super and blood is compared directly with basic super and blood, and with super, thirty-two paired plots are taken. The number of plots compared is governed by the proximity to one another of such plots. Compared plots are bracketed with one another in the tables.

• This table shows that the manures applied in 1923 gave a sufficient increase in yield in the 1924 hay crop alone to leave a profit. When it is borne in mind that an increased growth must have resulted during the greater part of the time since the application of the manure there need be no doubt regarding the total increase in yield.

In this case the crop was light, and a strip was cut down the length of each plot and through the standing crop without undue damage. This method is easier if the crop is not too heavy, the mower being engaged in cutting the remaining portion of the plots when waiting for the weighing operations to catch it up. ' , ■ ■ . ■ . . '

The increases due to all the manures shown in this table are highly significant and paying ones, but the differences between the yields of the manured plots themselves are not significant. In this case the cost of the manure for the two years has been more than paid for by the increase in only one crop of hay. ■ When it is considered that this crop represents only about one-half of the total growth of grass for the 1924-25 season the results speak for themselves.

From this table it will be seen that the application of manure was a paying one in every case. The differences between super and basic super and between super and super and blood are not significant, but both basic super and super and blood when compared directly with basic super and blood show a slight increase, indicating that blood in combination with basic super had a depressing effect on growth. . , ,

These results show that the second dressing of 1924 was paid for in the single hay crop, except in the case of basic super and blood. Here again the effect of the blood in combination with basic super seems to be a depressing one. Blood in combination with super gave a slight increase over super alone.

SUMMARY.

W. G. Macartney’s Experiment.

(1.) In 1923 a season of extremely light rainfall manurial dressings of 2 cwt. per acre did not pay in the hay crop taken three months after the application of the manures.

(2.) Eleven months after application a line of demarcation between the treated and. untreated areas could be distinctly seen, but this difference was not apparent when the second hay crop was" taken.

(3.) Weighings taken fifteen months after treatment showed a significant increase in the case of every manure.

General Conclusions applying to 1924 Experiments

(1.) In every case, with one exception (basic super and blood, Table 5), the actual cost of manures at country stations had been more than paid for by the increase in one crop of hay.

(2.) If the growth of grass represented by one crop of hay is regarded as half the total growth for the year the net profit resulting from the application of the fertilizers is greatly increased above that shown in the tables. ' ....

(3 J The net profit resulting from the application of manures was always slightly in favour of super or basic super.

(4.) In one case only has the addition of blood to phosphate (super) caused an increase in yieldand that a very slight one. In the main its inclusion has not been profitable, and from the experimental data available the use of blood for top-dressing of pastures cannot be recommended.

(5.) In two cases the addition of blood to basic super has resulted in a lower yield than with basic super alone.

' (6.) Excepting in the case of basic super and blood, the application of manures in 1924 on the manured plots of 1923 has left a good margin of profit in the single crop of hay. The increases due to the second application, however, have not been so great as those due to the first dressing. With the aid of a damp growing season in 1924 thejresidual effect of the 1923 manures was very evident.

(7.) As instanced by the grazing during the winter and early spring of 1924, the manures have had a decided effect on the palatability of the pastures.

(8.) The enhanced fertility due to the more vigorous development of clovers which generally accompanies phosphate manuring must be considerable, although its actual value cannot be measured.

(9.) The belated effect of the 1923 dressings indicates that, owing to the possibility of dry-weather conditions, top-dressing. should be done in the winter (say, June) rather than in the early spring.

(10.) The amount of £5 per ton, on which the values of increases in hay are estimated, is purely a nominal figure, and should it be considered too great for this season’s crop the substitution of a reasonably lower figure will in almost all cases still show a handsome margin of profit.

The writers’ thanks are due to Messrs. W. G. .Macartney and T. Bloor for facilities afforded ,in carrying out th e experiments, also to Mr. W Carter, who rendered valuable assistance. Mr. E. J. Fawcett, 8.A., Assistant Instructor in Agriculture, Hastings, was very closely associated with the work in 1923. He compiled the results of the first season’s work, and his valued assistance. is duly acknowledged.

Seed-testing Practices —At the New Zealand Official Seed-testing Station, in order to conform with American and European practice, the germination percentages of mangolds, beets, &c., will in future be expressed as percentages of the number of balls which show sprouts, irrespective of the number of sprouts per ball. The discarded method gave the total number of sprouts per 100 balls.

Noxious Weeds Act Orders.— The Southland County Council has declared Californian thistle and ragwort not to be noxious weeds within that county ; the Portobello Road Board has similarly declared Californian thistle within its district. The following are recent declarations of plants as noxious weeds : Kiwitea Ccunty, hemlock; Carterton Borough, broom, elderbeny, ox-eye daisy, pennyroyal, St. John’s wort; Masterton Borough, hemlock and fennel.

. * Increase not significant. The application of the statistical method enables the investigator to estimate the chances of any differences which occur being real and due to the treatment and not to chance variation.. When a difference occurs, and the odds in favour of such difference being a real. one are greater than 30 to 1, the difference is here said to be “ significant.”. If a difference occurs with chances of less than 30 to lin its favour, such a difference is spoken of as being “ not significant.” To simplify this article the odds are omitted, and differences which are “ not significant ” are noted as such. Any non-significant difference must, be regarded as due to mere chance variation in the crop, or else an insufficient number of weighings has been taken to reduce the “probable error” sufficiently to permit the difference being measured with certainty, in which case the result must, be regarded as showing no difference.

* Super (42-44), £7 ss. per ton ; basic super (40-43), £6 15s. per ton ; dried blood (12 per cent, nitrogen), 13s. per hundredweight. -f Difference not significant. ■ , : '

* Total cost of both applications, 1923 and 1924. f Difference not significant.

Note. — this case, where the dressings applied in 1924 have , crossed a control plot of 1923, these are plots with only one dressing of manure. Comparisons are made between these plots and those having no manure (double controls), or between the manures themselves. Examples—(a.) Super, 1924 (Plot lin plan), is compared with its neighbouring double control (Plot 6 in plan), and so on. (6.) Basic super, 1924 (Plot 11 in plan), is compared with basic super and blood (Plot 6in plan). . ■ -

* Difference not significant. f The comparison of basic super and basic super and blood shows A difference in favour of the basic super, which cost 13s. 6d., as against £1 for basic super and blood.

* Here, as in Table 3, the addition of blood to basic super has had the effect of reducing the yield of the plots with that mixture to below that of basic super alone.

O . - • ' ■ Yield of Increase Value of Cost of ’■ ■ Treatment. Yield of Hay due to Increase at Manure per per Acre. Increase due to Manure. Value of Increase at £5 per Ton. Cost of Manure per Acre. Net Profit per Acre. Tons. Tons. .' - £■ s. d. . £ . s. d. £ s. d. Super ... 1-29 o-i3 * * ❖ Super and blood 1-38 0-22 ’ 12 0 0 17 5 047 Basic super . . ... 1'38 . 0-22 12 O • o 13 . 6 086 Basic super and blood 1'32 0-16 . o 16 o 0 16 8 Loss 8d, Control '• . .T16

Table 1. —Result of Top-dressings on Farm of T. Bloor, Sockburn.

Treatment. Number of Plots compared. Yield of Hay per Acre. Difference in Hay per Acre. Value of Difference at £5 per Ton. Cost of Manure per Acre at Country . Station.* -Net Profit per. Acre. ; Tons. Tons. £ s- d. £ s. d. 7' s. d. Super . . 32 ■ 2-635) 0-550 2 15 I 0 14 6 207 Control . . . . 32 . 2-085 f b -- Basic super . . . 32 2-515 ) 0-430 2130 0 13 6 1 19 6 Basic super and blood. . ■“ 24 " .2-396) 0-279 1 7 11 0 16 8 0113 Control 7 . 24 > 2-117 f 1 •• Super and blood ■ 24 ,2-430 ) 0-313 iii 4 0 17 5 0 13 11 Super . . . . 32 2-635 } t + Basic super ... 32 2*515 ) T 1 Super and blood ' ... 32 • 2-5I3 1 + • + ’ + Basic super and blood. . 32 2-516 ) T ’ 1 1 Super . . . . 32 2-635 .. 1 t Super and blood . . 32 ; 2-513 ) T ■ ’ T . T 7 w Basic super . . 32 2-515 + 4Basic super, and blood. . 32 2-516 r T . 1 . I

Treatment. Number , of Plots compared. Yield of Hay per Acre. Difference in Hay per Acre. Value of Difference at £5 per Tori.” Cost of Manure per Acre at Country Station.* Net'Profit per Acre. Tons, i Tons. £ s d. £ s. d. £ s- d. Double super 16 2-954) 0-873 4 17 3 I 90 383 Double control '■ .. 16 : 2-o8i J 1 •• Double basic super 16 2-878 J 0-797 3 19 8 170 2 12 8 Double basic super and' ■ 12 ■ ■ 2-680 1 0-535 2 13. 6 1168 0 16 IOblood ’ - y •p- > ■ Double control 12 2-145 J •• Double super and blood 16 2-803 1 0-662 362 1 18 5 179 Double control 16 .2-141 ) Double super 8 - -; 12 : -2-965 ■ • ’ X • Double super and blood 12 . 2-863 . ... . I, ,, , ■ 1 x ; ‘, \'. ’ 7 ' <? Double- basic super . ,. 16 ■ 2-877 t t t -■ Double basic super and l6 2-814 f - T I.:- ... ■blood Double super ■ .. ...... >' id 2-954 1 t J. - f t Double basic super ' . . 16 2-877 J T x .. • •• ■ . ■ - 16 2-877 .. -I-.. no ■ . - , J > ■. . . .: \ v -J • .... • •' 4 • ' - - ... k.'

Table 3. —Yields from Double Manures compared with Double Controls, and Double Manures compared with one another. (For explanation of terms “ double ” manure, &c., see note on plan. Example of comparison Double super, Plot 2, compared with double control, Plot 6, as shown on plan.)

Treatment. Number of Plots compared. Yield of . Hay per Acre. Difference in Hay per Acre. Value of Difference at £5 per Ton. Cost of • Manure per Acre at Country Station. Net Profit per Acre. Tons. Tons. £ s. d. £ s. d. £ s. d. -Super ... 16 -•934 1 0-853 4 5 4 0146 3 IO IO ■Control 16 2-081) 1 •• Basic super . . 16 2-816 /o-735 3 13 6 0 13 6 300 Basic super and blood. . 16. 2-66o \ 0-571 2 17 1 100 I 17 I •Control 16 2-089) \ ’ * Super and blood 16 2-869 f 0-780 3 18 0 1 1 0 217 0 Super ... 16 2-934 1 * * ❖ Basic super .. 16 2-816 j Basic super .. 16 ■ 2-816 \ 0-156 0 15 7t Basic super and blood .. . 16 2-660 /.. Super and blood 16 2-869 4 0-209 1 0 11 Basic super and blood .. 16 2-660 ? .. ■Super 12 2-863 v * • ❖ * Super and blood 12 2-933 J

Table 4. —Yields 'from Manures applied in 1924 compared with Double Controls.

Treatment. Number of Plots compared. Yield of Hay per Acre. Difference in Hay per Acre. Value of Difference at £5 per Ton. Cost of Manure per Acre at Country Station. Net Profit per Acre. Tons, Tons. £ S. d. ' £ s. d. 4 s. d. Super 64 2-8901 0-404 205 0146 i 5 11 Control .. 64 ■ 2-486 J 1 •• Basic super . . 64 2-866 J 0-380 I 18 0 0 13 6 146 Basic super and blood. . 64 2-788) 0-185 0 18 6 100 Loss of V is. 6d. Control 64 2-603) \ •• Super and blood 64 2-950 J o-347 1148 1 1 0 0 13 8 Basic super . . 64 2-866) 0-078 0 7 10* Basic super and blood. . 64 2-788 j Super and blood 48 2-997) 0-148 0 14 10 Super 48 2-849) r # e

Table 5. — Effect of Application of Manures in 1924 on those of 1923, compared with 1923 Application (1924 Controls) and with one another. (All plots falling on the controls of the 1923 experiment are eliminated, and in the case of super and control, Plots 2,3, 4, and 5 are compared with Plots 7,8, 9, and 10 respectively, &c.)