VARIATIONS IN THE PERCENTAGE OF BUTTERFAT IN MILK.

New Zealand Journal of Agriculture, Volume XXX, Issue 2, 20 February 1925, Page 77

VARIATIONS IN THE PERCENTAGE OF BUTTERFAT IN MILK.

A STUDY BASED ON NEW ZEALAND C.O.R. DATA.

( Continued.)

W. N. PATON,

Dairy

Division, Wellington

111. MONTHLY VARIATIONS. Monthly variations in test are fluctuations which occur as the lactation period advances. They are not to be confused with variations which occur from day to' day. Lactational or monthly variations are influenced by practically a different set of factors from those which affect daily variations. As it is the phase of test variations which concerns C.O.R. breeders most, it is hoped that this article will help to clear certain points for them. The subject has been treated as simply as possible, and as many of the results as space would permit have been represented by graphs. LACTATIONAL VARIATIONS DUE TO BREED. The difference in lactational variations due to breed are readily apparent in Graph 4 and in Table 7 from which the curves were plotted. The figures in the table represent the averages for all first-class records for the chief breeds up to 31st December, 1923, except in the case of the Jerseys ; with the latter figures for one year only are given.* The first most noticeable point is that the Jersey graph line is different in general conformation, inasmuch as it exhibits an upward trend from beginning to end, whereas for the other breeds a “ dip ” appears in the first half of each curve. In the case of the Friesians, Milking Shorthorns, and Ayrshires the length of this “ dip ” seems to bear some relation to the average test of the breed, as it will be noticed that it becomes less as we consider these breeds in the ascending order of their average tests. To illustrate this more clearly, points have been taken on these curves where they again reach the same point as indicated by the commencement of the curves, and these joined by the straight dotted line shown.f The fact that this line is a straight line may be merely accidental. Apart from the surprising and interesting nature of the result, it is possible that there is some significance attachable to it. Why does it apply only to these three breeds ? The following extract from page 1008 of the British Friesian Journal of October, 1924, may help to answer the question : “ Among the well-known breeds which are known to have their origin in the cattle of the Netherlands are the Hollandaise of Belgium, Flamande, Boulonaise, and Artesienne of France, Brittenburg and Oldenburg of

Germany, and the Kolmogorian breed of Russia. Eminent breed historians, such as Professor Low and Howard, tell us that these same Dutch cattle in the seventeenth and eighteenth centuries helped to form the foundation of the Teeswater breed, now known as the Shorthorn, and of the Dunlop, from which the Ayrshire has been developed by Scottish dairymen/’ It would be interesting to ascertain if the result obtained held good for the other breeds which are claimed to have their origin in the Dutch cattle. Of course, all curves would need to be obtained from records secured under similar conditions, otherwise the comparison would not be a true one. The other dotted line shown in the graph joins the points where the average tests would lie on the respective “ smoothed ” curves. The parabola obtained is interesting, as it is considered probable that curves for other breeds tested in New Zealand would cut this line at points corresponding to the. average tests for such breeds. In Table 8 each monthly test as given in Table 7 is divided by the average test and then multiplied by 100 — i.e., the lactational tests are expressed as percentages of the annual test.* By this method

all figures are reduced to one basis, and comparison can thus be more readily made. One striking point is that for the fourth test all breeds practically coincide. The Red Polls and Jerseys agree fairly closely from the sixth test onwards, and the Ayrshires and Friesians agree quite well right through except for the difference in dip. It is interesting to note that for the first lactational test the Jerseys commence at a considerably lower point than do the other breeds. ; . The next aspect worthy of consideration is the average total range of variation — i.e., the difference in the average of the highest and lowest tests for all cows of each breed. Table 9 gives the results of this analysis. The only point of particular interest brought out in this table is in regard to the average lowest test. With the exception of the Jerseys, it appears that the average lowest lactational test is practically a constant percentage of the average annual test. The average highest lactational test, therefore, is mainly responsible for the total range of variation. The total range of variation, however, appears to be a peculiarity of the breed, as it bears no apparent relation to the average test for the year.

LACTATIONAL VARIATIONS DUE TO TIME OF YEAR OF CALVING. In Graphs 5 and 6 it is readily apparent that the influence of the factor of time of year of calving on the lactational-test variations is considerable.* The first test and the number of records for each curve are given on the left and right respectively. In conformation the curves for similar periods for Ayrshires and Milking Shorthorns bear a striking resemblance, and this indicates that the effect on these two breeds is similar. . The dip in the curves becomes smaller up to the month of October, when it almost entirely disappears. The result is practically, a straight line with a uniform upward trend from beginning to end. ' The curves for the summer period exhibit a shallow dip and a “ falling away ” towards the latter end. This falling-away tendency can be noticed- slightly in the September and more distinctly in the October curves. However, the quantity of data available in the case of the Friesians is sufficiently large to show curves for each month of commencement.

The results are presented in Graph 7, in which the numbers on the right give the number of records included for each curve. From August to December that portion of the curves from the tenth to the twelfth tests has been gradually depressed, until for December the twelfth point is below that for the tenth. From August to October the dip becomes less, until for October it is very slight. However, the dip occurs again to a greater extent in the November curve, with the depth of the dip increasing up to January. In the December curve a flattening-out of the portion from the ninth to eleventh tests will be noticed as compared with the same portion in the November curve. In January we have a two-dip curve, the second dip making its appearance at the eighth test. The February curve has the second dip at the ninth test, while for March we have two pronounced dips separated by a high point at the fifth test. From April 'to July the two dips are gradually, transformed into one long dip by the gradual depression of the high point at the fifth test. May, June, and July curves climb very steeply from the eighth test onwards, June having the greatest gradient. . = Average lactational tests have been run . out for the four chief breeds, grouping separately spring, summer, autumn, and winter cal vers, as shown in Graph 8. Considering the graphs for Friesians, Milking Shorthorns, and Ayrshires only, a marked similarity is noticeable in the respective curves for cows commencing in different seasons of the year. The spring curves have first a small dip, and then a gradual upward trend right to the end. The summer curves are shaped something like the letter “ S ” lying on its side and considerably straightened out, the final regression of the test being well borne out. The autumn curves are “ two-dip ” curves, the dips being separated at the fifth test. In the case of the Ayrshires there are really three dips, but this is considered unimportant owing to the very small number of records for this curve. It is satisfactory, however, to observe that the characteristic of the fifth point is nevertheless quite apparent. All the winter curves have long and deep dips, with fairly sharp “ risings ”at the ends. The Jersey graph presents what appears quite a different position from those of the other breeds. As there is no dip in the first portion of each of the curves, comparison at first seems difficult. However, if we compare the respective curves from the seventh test onwards it is obvious that the agreement for the four breeds is quite close. As in the case of the . Milking Shorthorns and Ayrshires, autumn calvers are poorly represented numerically in the Jerseys, and the curve for this reason is rather irregular. Nevertheless the fifth test ranks high. • To properly follow the effect of the time of commencement on the lactational-test curves for the Jerseys, all the Jersey data would need to be utilized, as has been done in the case of the Friesians, in order that curves could be shown for each month of commencement.

The difference in range of variation of the lactational' tests for, cows commencing at different periods of the year is of some interest. In Table 10 figures for the chief breeds are given according to the season of the year of commencement. In considering range of variation the year seems to be divided into two rather than into four, since on the one hand figures for spring and winter, and on the other figures for summer and autumn, agree fairly closely. Range-of-variation figures were run

out for several breeds for different months of commencement, and it was found that the greatest ranges occurred for July, August, September, and October. The next highest were May, June, and November, while December, January, February, March, and April ranked lowest. We arrive at the interesting conclusion that cows commencing during the wet season of the year show a greater range of variation in test than do cows calving in the dry season, although in each case cows were milked through all periods of the year.

LACTATIONAL VARIATIONS DUE TO LENGTH OF PERIOD OF GESTATION. The length of time a cow is in calf during test has considerable influence on the lactational-test variations, inasmuch as the general conformation of the monthly-test curve is appreciably affected. This is clearly shown in Graph 9. In compiling this graph records for the Friesians were taken. On the left of the graph the first test for each curve is indicated, and along the curves the limits of the number of days between calving at commencement of test and subsequent calving are given. The numbers on the right indicate the number of records represented in the respective curves. Only three curves are shown, but these serve to indicate the trend of the variation very well. More curves were not shown for the reason that some would necessarily have been somewhat irregular owing to the smallness of the number of records included. The top curve is for all first-class certificates, being

for cows calving within 455 days after start of test; The middle one represents all second-class ■ certificates, and also includes other records equivalent to second class but which were made before the. introduction of the second-class certificate, and are for cows calving between 455 and 485 days after start of test. The bottom curve represents some of the records for which calvings were available and for which the period was 485 days and more — i.e., all these cows failed on subsequent calving for both first and second-class certificates. From the ninth test onwards the difference of trend in these curves is readily apparent. No difference in general conformation was found in the curves for the limitssls-545, and 545 and more days. The change in trend of the test curves evidently ceased after a period of about 515 days. From this it may be accepted that the effect of gestation on the test curve of a cow calving 515 days after commencement is negligible. In other words, a cow may be pregnant for a period up to about four and a half months during a 365-day lactation without this influencing the test in any way. A. C. Ragsdale, C. W. Turner, and S. Brody stated as a conclusion to their investigation on the “ Effect of Gestation upon Lactation in the Dairy Cow ” that when during lactation the period of pregnancy exceeds about five months the effect, of pregnancy becomes apparent in a reduced rate of milk-secretion Journal of Dairy Science). When it is remembered that quality and quantity of milk depend on one another to some extent, the fact that the two conclusions quoted agree so closely is not at all surprising. LACTATIONAL VARIATIONS DUE TO NATURE OF SEASON. The Friesian data have been utilized to test this factor, and the results are given in Graph 10. The dotted curve in each case denotes the average lactational-test curve for all Friesians, while the wholelines represent the average monthly tests for different seasons. On the left the seasons are. given, and on the right the number of records for each season is supplied. By showing the curve for all Friesians each time comparison is considerably facilitated. For reference and comparison, tests for the dotted curve are given at the foot of the graph. In all cases it will be noticeable that the curves do not vary much from the average, and that their general conformation remains fairly uniform. In the case of the 1918-19 season the yearly curve approaches the average right throughout better than for any other season. The Milking Shorthorns gave a similar result.* In general it is found that where bad seasons were experienced the yearly curves lie above, while for good years they lie below, the curve for all years. Good seasons were experienced for 1913-14, 1915-16, 1921-22, and 1922-23, while for 1912-13, 1914-15, 1919-20, and 1920-21 the seasons were considered to be poor ones, and in each case the foregoing statement holds good. ' - ■ LACTATIONAL VARIATIONS DUE TO CONDITION OF THE COW. . This factor has been already mentioned in the second article of this series, but, as there stated, it has no influence on daily variations in test, since the latter are variations considered for much shorter periods

than a month. The condition of a cow at time of calving is believed to somewhat influence the monthly test curve. Unfortunately, our C.O.R. data cannot be used in this case, since no record is obtained in regard to the condition of a cow at commencement of test.

There are two kinds of condition, known respectively as “ soft ” and “ hard,” and either is obtained by difference in ration. As the names indicate, a soft condition is soon reduced, and is suitable for butterfat trials at shows, while a hard condition is not reduced so quickly by flush of milk-production, and “ milks oft ” much slower. The latter is therefore the better condition to obtain in fitting a cow for C.O.R. test. The test curves for poor-, soft-, and hard-conditioned cows should show a difference in conformation one from the other, and all would no doubt be influenced by feeding during test. It is patent, therefore, that the whole question could be properly treated only from the results of a carefully planned experiment. LACTATIONAL VARIATIONS DUE TO AGE. A trial with the 1918-19 season Jersey data showed that the lactational-test variations were similar for different ages, all the curves being alike in general conformation. This factor may therefore be dismissed as one which does not. materially affect the trend of the lactational-test curves. LACTATIONAL VARIATIONS DUE TO QUANTITY OF MILK-PRODUCTION. The 1918-19 two-year-old Jersey figures, when grouped according to quantity of milk-production, produced curves which were all uniform in general conformation. Difference in annual milk-production does not produce difference in the respective lactational-test curves. As a matter of fact, all curves for this factor as well as those for that of age were inclined at practically the same angle. GENERAL EXAMPLES. Before passing on to conclusions a few individual examples may prove of interest. In the Tables 11 and 12 the monthly tests are supplied for the highest- and lowest-testing cows of the principal breeds. Table 13 has been compiled to illustrate how high and how low cows may test during a lactation. The tests quoted were taken from first-class C.O.R. records, and were allowed to stand, the cows at the time of test being normal as far as was known.

For the final Table 14 examples have been specially selected from C.O.R. data, and serve to show how remarkably some cows test. CONCLUSIONS. To sum up, there appear to be only five factors which materially influence monthly-test variations. . They are (1) the breed ; (2) time of commencement of lactation during year (3) length of period of gestation during test; (4) nature of season ; and (5) condition of cow, feeding, &c. Note.—Unless otherwise stated the C.O.R. data used in the case of. each breed include all first-class records from the year 1913 up to 31st December, 1923. The figures quoted for. the Jerseys in each case are for the 1918-19 season, with the exception of Tables 11, 12, and 13, which are for all Jerseys up to 31st December, 1923.

(To be continued.)

* Much laborious work is entailed in obtaining results for lactational-test-variations, and it is regretted that time would not permit all the Jerseys being included. ' The year selected was 1918-19, as this was considered to be a typical average year, and should therefore give a good idea of what the complete figures would reveal. •f The equation for this line is : Number of days in the dip of the lactationaltest curve of a particular breed (722 .163) times the average test of that breed.

- *By “ annual test” is meant the average test for the lactation, this being equivalent to the test of the whole of the milk produced taken in one lot.

breeds here is divided by all is average quoted. In of tests is Jersey Table (see

* The curves shown in these two graphs are of the “ smoothed ” type.

* It was for this reason that the 1918-19 data of the Jerseys were taken in every instance as being likely to supply the probable average for all Jerseys.

* As given in the first article of this series (Journal, September, 1924)

A and B give tests for the same cow for two consecutive seasons, and provide a good example of consistent high testing C is a remarkable case of extreme range in variation. D is an exceptional case of small range in variation. E, F, G, and H are all examples where a number of consecutive monthly tests have all come out the same.

: — — Breed. Number of Records. Lactational Order. Tests given Average Test. in Order. Average Test. 1st. 2nd. 3rd. 4th. th. 5 6th. 7th. th. th. 9 10th. nth. th. 13th. Jersey 163 5 4’ 5-ii 5’37 5'52 5’70 5-83 6-03 6-22 6-45 6-6o 6-59 6’63 6-86 5'66 Red Poll 34 4-24 4-n 4-04 4-22 4-25 4-48 4’59 4-82 5-oo 5-02 5’12 5-°6 4-37 Ayrshire . 113 4-00 3-95 3’99 4-02 4-10 4'io 4-20 4-33 4'47 4-60 4-62 473 4-12 Milking 263 3-85 3’73 3-80 3-85 3’92 4-02 4-17 4'32 4’44 4A7 4-67 4'75 3'99 Friesian 1 , 3’53 3'39 3-40 3’44 3'47-3-53 3-60 3-7 3-82 3-93 4-°3 4-12 4'29 3'55

Table 7. Lactational Tests of the Chief Breeds.

Breed. Number of Records. Lactational Tests given in Order. 1st. 2nd. 3rd. 4th. 5th. 6th. 7 th. 8th. 9th. roth. nth. 12th. 13th. ... . . Lactational Tests given in Order. 1st.' 2nd. 3rd. 4 th. 5th. 6th. 7th. 8 th-. 9 th. 10th. nth. 12th. 13th. Jersey 163 86 90 95 98 IOI 103 107 IO 114 117 116 117 121 Red Poll 34 97 94 92 97 97 103 105 no 114 115 117 Il6 Ayrshire 113 97 96 97 98 100 IOO 102 105 109 112 112 115 Milking . 263 97 94 95 97 98 IOI 105 109 112 15 117 119 Friesian ... 1,042 99 96 96 97 98 99 IOI 105 108 III 113 Il6 121

Table 8. Lactational Tests of the Chief Breeds expressed as Percentges of the respective Average Annual Test.

Lactational Tests. Highest Test Lowest Test Number Total Percentage expressed expressed Breed. Average Number of Test. Breed. of Records. Average Test. Average Lactational Tests. Total Range of Variation. Percentage Range of Variation. Highest Test expressed as Highest. Percentage Test. of Average Average Lowest Test. : Range of •Variation. ■ Range of Variation; as Percentage of Average Test. Lowest Test expressed as .. - Percentage of Average . Test. Average Highest Test. Average Lowest Test. Jersey 163 5-66 7'01 4-68 2-33. 41 124 83 j Red Poll . . - 34 4'37 5-51 3-82. 1-69 39 126 87 Milking 263 3’99 5-oi 3'47 i-54 ■ 39 ■ ' 126 87 • Shorthorn J Friesian . . 1,042 3’55 4’37 .3-12 .1-25 35 123 88: Ayrshire .. 113 4-12 4'95 .3-64 f’3i 32 120 88 ■

Table 9. Range of Variation in Lactational Tests of the Chief Breeds.

Period. Number of Records. Average Test. Lactational Tests. . Total Range of Variation. Percentage Range of Variation. Highest Testexpressed as ’ Percentage of Average Test. Lowest Test expressed as Percentage of Average Test. Average Highest Test. ' Average Lowest Test. JERSEYS. Spring 134 5’69 . 7-09 JERSEYS 4-70 2-39 42 125 83 Summer . . 18 5-50 6’52 4'57 i-95 36 119 83 Autumn . . 4 5'79 6'75 4-63 2-12 37 117 80 Winter . . . 7 5-56 • 6-8i . 4-67 2-14 38 122, 84 MILKING SHORTHORNS. . - ■ Spring . . . 218 .3-98 MILKIN 5'02 G SHORT 3'45 HORNS. i-57 39 126 87 Summer . . 20 4-09 4-97 3-56 1-41 34 121 87 Autumn . . 5 - 3-96 4-65 3-58 1-07 27 . 117 . 90 Winter . . 20 4'07 5-oo 3-51 1’49 37 123 86 FRIESIAN Spring . . 615 3’54 F 4'41 RIESIAN 3-II 1-30 37 125 88 Summer . . 255 3-56 4-26 3-15 111 3i 120 120 89 80 Autumn . . •57 ' 3'55 4-21 3-14 1-07 30 119 ' 89 Winter . . 115 3-56 4-46 3-13 i-33 37 125 88 AYRSHIRES. Spring . . ' 75 ■ 4-12 A' 4'97 iTRSHIRES 3-62 i-35 33 121 ■ 88 Summer . . 27 ■ 4 -I 4 4-86 3-72 1-14 28 117 89 Autumn . . 3 ■ 3'74 4-5° 3’43 1-07 29 120 9i Winter . . 8 . 4-18 5-16 3-66 1-50 36 123 87

Table 10. Range of Variation in Lactational Tests of Chief Breeds-for Cows commencing during Different Periods of the Year.

Breed. Monthly Tests given Average Test. Test made ISt. 2nd. 3rd. 4th. 5th. 6th. 7th. 8th. th. 10th. nth. th. 13 Jersey 5-70 7'30 6-50 y 40 8-90 9-60 8-io 8'3° 8-50 8-30 8-55 8-34 7-65 November Red Poll 4'47 4’90 5’34 5-13 5'27 5-44 5-io 5-61 6-05 5’44 5’88 5’37 5’29 October. Ayrshire 4'50 4-65 5-48 5’58 5’20 5'30 5-64 5’29 5'49 6-21 5'25 September Milking 5-80 5-80 5'io 4-60 5-50 5-io 5-60 6-io 6-20 6-50 7-10 6-30 5-60 August. Friesian . 4’°5 4’34 4-80 4-58 4’37 4’91 5’27 5’2.5 5-32 5-70 5-9i 5-64 4-89 July.

Table II. Lactational Tests of the Hightest-testing Indivìduals of the Chief Breeds.

Breed. Monthly Tests given in Order. Average Test. First Test made in ISt. 2nd. 3rd. 4th. 5 th. 6th. 7lh - 8 th. 9 th. 10th. nth. 12 th. 13th. j Monthly Tests given in Order. ■ Average Test. First Test made in rst. . 2nd. 3rd. 4th. 5 th. 6th. 7th. 8 th. 9th. ■ 10th. nth. 12 th. 13th. Jersey '2-60 ; 3 3'14 3’95 3-85 4-09 478 4-10 4’24 4'44 4-81 4'64 3-85 November Poll 3-5° 3-80 3'43 3'33 379 3'9° 3’99 3'88 374 3’91 3’99 3-86 372 October. •r 3’72 October. Ayrshire 3-60 3'50 3’50 3-io 3-4° 3-00 3-io 3-oo 3’20 3-20 3-29 December. Milking 3'90 3'34 3’°4 2-80 2-90 3-04 7 2-95 3'34 3-41 3’29 3-48 3-16 August. Friesian 2-29 3-08 2-39 2-6i 2-54 2-80 3'45 3-io 2-94 2-83 274 August.

Table 12. Lactational Tests of the Lowest-testing Individuals of the Chief Breeds.

Breed. Highest Test. Lowest Test. \. Average Test of Breed.* Jersey n-6o 1 2-60 5'55 Red Poll 6-50 3-24 4-39 Ayrshire * ■ 9'53 2-91 4-II Milking Shorthorn . .. 7-80 2-60 3-97 Friesian ' . . 8-oo 2-13 3'54

Table 13. Highest and Lowest Monthly. Tests, of the Chief Breeds.

£ o o w Order of Tests during Lactation. Average Test. Breed. ist. 2nd. j 3rd. 4th. ' 5th. 6 th. 7th. 8th. 9th. 10th. nth. ■ 12th. . 13th. M A Order of Tests during Lactation.. Average Test. Breed; ist. 6-oo 2nd. 5’7° 3rd. 5-So 4th. 80 5’ th. 5 6-70 6th. 6-6o 7th. . 7-00 th.. 7’4° 9th. 8-io th. 10 9-20 nth. 9’3° 9’20 6'65 Jersey. . •' ■ 12th. - 13th. A B '■ 6-oo 5'20 5’7° 4’90 5-80 6-6o 5-80 7-10 6-70 6-8o 6-6-70 7-00 7-30 7’4° 8-20 8-io 9-00 9-20 9'4° 9-3°' 8-8o 6-67 Jersey. 9-20 • 6-65 Jersey. B C 5-20 3-64 4’90 4’14 6-6o . 3'32 7-10 3-78 6-8o 4-58 6-70- 5’14 7’30 6-oo 8-20 6-27 9-00 8-20 9-40 9'53 4'4 8 Ayrshire. 8-8o 6:67 Jersey. C D 3'64 , 3’33 -I 43-06 3-32 3’23 8 3’73'22 8 4-53-15 5-14 3-oi 6-oo 3’°2 6-27 3-17 8-20 3’04 9’53 3-18 4A8 ' Ayrshire. 3-10 3-08 3-!3 Friesian. D E 3’33 3’4° 3-°6 4-00 3-23 4-00 3’22 °° 4; 3-i5 4-00 3-oi 3-60 3-°2 3’7° 3-17 4-00 3’04 3-9° 3-18 3-80 3-io 3'90 3-84 Milking Shorthorn 3-08 3’13 Friesian. E F 3-40 4-14 4-00 3’8o 4-00 3’99 4-00 3’95 4-00 3’95 3-60 3’95 3’70 3’95 4-00 4-01 3-9° 4-3i 3-80 6 5 43’90 4’35 3-84 Milking Shorthorn 4-40 4’°9 Jersey. F G 4-14 3-5° 3-80 3'50 3’99 3-5° 3’95 3‘6o 3:95 3-60 3’95 3’6o 3’95 3-60 4-01 3’6° 4-3i 3’4° 6 5 4-3-3° 4'35 3’90 4’4° 4-50 4-09 3’54 Jersey. Friesian. G H 3'50 3’4° : • 3/50 3’4P 3’50 3’4° 3-60 3’50 3-60 3’8o 3-60 3-60 3-60 3'7° 3-6o 3-80 3’4° 3-80 3-3° 3-80 3’90 4-00 3’59 Ayrshire. 4-5° 3’54 Friesian. Key. 3’40 3’49 3-40 3’50 3-8° 3-60 3'7° 3’8o 3‘8o 3-80 4-00 3-59 Ayrshire. ' . Key. ■■ H

Table 14. Exceptional Examples.