SF UNITED STATES DEPARTMENT OF AGRICULTURE jyJ^'^Lru BULLETIN No. 996 Contribution from the Bureau of Animal Industry JOHN R. MOHLER, Chief Washington, D. C. October 12, 1921 FLUSHING AND OTHER MEANS OF INCREASING LAMB YIELDS. 1 By FT Rf Marshall and C. G. Potts, Animal Husbandry Division. CONTENTS. Page. Extent to -which flushing is practiced in the United States 1 Factors influencing size of lamb crop 2 Results of experiments in flushing ewes :i General plan of the experiments 3 Number of lambs dropped ! Relation of weight gains to number of twins. 6 Feed for flushing t; Earlinass of lambing 6 Uniformity of lambs' ages 7 Twin production as affected by age of ewe. . . V Twin production as affected by breed of ewe. 8 Twin production as affected by sire 8 Breeding for twin lambs 9 Value of twin lambs in comparison with singles i o Comparative weights of single and twin lambs 11 Summary 13 EXTENT TO WHICH FLUSHING IS PRACTICED IN THE UNITED STATES. THE BELIEF that extra feeding of ewes at the time they were bred would result in larger yields of lambs has long been held by students of sheep husbandry. Feeding for this purpose is commonly called "flushing." Though flushing has been practiced by a few breeders of purebred slice]), it has not been generally or seriously considered by Hock owners producing market lambs. Practically no figures or results of experiments have been obtainable as to the extent of increase in the lamb crop that could be obtained by flushing the ewes. This bulletin presents the results of live years' experiments in Hush- ing ewes, covering a total of 302 matings. These experiments were conducted on that portion of the Bureau of Animal Industry farm at Beltsvillc, Md., known as "Sheep Acres," and at the United States Morgan Horse Farm, Middleburv. Vt. There is added a discussion of other matters to be considered in endeavoring to obtain maximum yields of lambs. !R. B. Millin, now of the Montana Agricultural College, assisted in the early development of the experi- ments reported in this bulletin. 56405°— 21— Bull. 996 2 BULLETIN 996, U. S. DEPARTMENT OF AGRICULTURE. FACTORS INFLUENCING SIZE OF LAMB CROP. The size of the lamb crop is dependent upon two things — the num- ber of dry ewes (those not having lambs) and the number of twins and triplets. Under ordinary farm-flock conditions the proportion of dry ewes is insignificant. In range flocks, however, it is a principal cause of lower lamb yields, and it is often impossible to furnish the' feed necessary to put the ewes into condition to make sure of their getting in lamb. The advantages of flushing are to be obtained principally through an increased number of twins. It has not been proved that the sire influences the number of twins occurring among his offspring. The production of twins or triplets is determined chiefly, if not entirely, by the ewe. Twins may result in either of two ways. First, two developed ova (eggs) may be discharged from the ovaries during the period of heat. Second, a single fertilized ovum may become divided at an early stage and each part develop a fetus. The first is believed to be the more common cause of twins. The production of a second or third ovum is thought to be largely influenced by the condition of the ewe and on this basis the connection between flushing and twin births is rendered very clear. To show the connection between production of twin lambs and maturing of extra ova, Marshall 2 slaughtered 55 Black-faced High- land sheep shortly after breeding and examined the ovaries to learn the number of ova that were produced. His findings were as follows: 1 ruptured follicle in one ovary — 1 ovum produced 42 cases. 1 ruptured follicle in each ovary — 2 ova produced 7 cases. 2 ruptured follicles in one ovary — 2 ova produced 5 cases. 2 ruptured follicles in one ovary and one in the other — 3 ova produced .... 1 case. In this case if the ewes had been kept and if all the ova had been fertilized and all developed normally, the result would have been 42 single lambs, 12 pairs of twins, and one set of triplets, a total lamb crop equal to 125.4 per cent of the number of ewes bred. The report of this experiment states that this is higher than the ordinary returns from flocks of the breed and that apparently under ordinary condi- tions some of the ova do not produce lambs. It is a common observation that the twin lambs in a flock are pro- duced chiefly in the early part of the lambing season. In 302 cases of lambing in purebred Southdown ewes used in experiments con- ducted by the Bureau of Animal Industry and extending over five years, 78 per cent of the ewes dropping twins lambed during the first half of the lambing period. The explanation of these facts must be found in one of two things, either of which has an important relation to management for maxi- 2 The (Eslrous Cycle and the Formation of Corpus Luteum in Sheep. In Philosophical Transactions of the Royal Society, Series B, No. 196, LIBRARY OF cONQKflt •'•.FKcovto ooouMENTs ammo* w FLUSHING TO INCREASE LAMB YIELDS. mum lamb yields: (1) The ewes that are in the best nourished condi- tion and therefore more likely to produce two ova are also those first to come in heat and get in lamb, or, (2) the feed and pasturage are more nutritious early in the season and cause production of more ova by the ewes bred at that time. In the Bureau of Animal Industry experiments the use of rams began about September 7 to September 10, but in most cases only one or two ewes were bred during the first 10 days. It therefore seems that the ewes ordinarily bred first are those that owing to their better physical condition first come in heat at the beginning of the breeding season, and that their condition is at once the cause of their showing heat earlier and their producing twins. An overfat condition may derange the normal action of the ova- ries even more seriously than a thin condition. It seldom occurs except in stock fitted for show or in ewes that have missed getting in lamb and grown fat while running dry. In our experiments there have been some rather fat dry ewes which got in lamb as readily as other ewes. They were not allowed grain, however, and had plenty of exercise. RESULTS OF EXPERIMENTS IN FLUSHING EWES. GENERAL PLAN OF THE EXPERIMENTS. The experiments were made with purebred Southdown ewes, all of which had been bred in the Bureau of Animal Industry flocks. The number of animals was 302, divided into 17 lots, of which Lots 1, 2, 5, and 6 were at the Morgan Horse Farm, Middlebury, Vt., and Lots 3, 4, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, and 17 at the bureau farm, Beltsville, Md. The ewes were usually in good condition com- pared with general farm flocks, which it is thought would tend to lessen the effect of flushing. In all the tests the lots of ewes were divided with careful regard to age, number of lambs dropped in previous years, gain in weight prior to experiment, and the rams to which bred. As it was necessary to use a number of different rams in each flock, hand breeding 3 was followed, which with the use of teasers 4 made it possible to record the date of each service and at the same time mate each ewe to the ram desired. Approximately the same number of ewes in each lot were bred to each of the rams used that season. In one case the proportion was disturbed because one of the rams was not a sure breeder and some of the ewes booked to him did not get in lamb. Such cases were eliminated from the experiment. The data given are in all cases for ewes having lambs. The matter of dry ewes and the influence of rams upon the lamb yield are discussed separately. 3 " Hand breeding " is a term applied to the individual mating of a ram and ewe outside the flock. « ' ' Teaser " applies to a ram used in a flock for t he purpose of indicating the ewes which are in heat. Copu- lation is prevented by an apron tied around the ram's body. 4 BULLETIN 996, U. S. DEPARTMENT OF AGRICULTURE. The amount of grain fed was never less than one-half pound per ewe daily; in a few cases it was as much as three-fourths of a pound. The grain mixture used is given in Table 1. The gains shown are the averages for each lot, each ewe's gain covering the period from one to two weeks prior to the dates the rams were first used up to the date of the service to which she lambed. It was planned to hold the unflushed lots without making any gains in weight during the breeding season. Except in the case of Lots 1 1 and 14, the pasturage was the same for the flushed and unflushed lots. There was considerable irregularity in the gains of individual ewes. Some of those in the unflushed lots made quite large gains, while some in the flushed lots made little or no gain. However, the fact of having grain feed may be considered to result in a more highly nourished body condition, even if not reflected in gains in weight . In all the Vermont flock tests, all lots were run on blue-grass pas- tures. All the lots in the Maryland flock were grazed on forage crops, soy beans being chiefly used during the breeding season. NUMBER OF LAMBS DROPPED. An average increase of 18.1 per cent in the number of lambs dropped was obtained as a result of flushing in the experiments here reported. This figure applies to 302 lambings of 143 different ewes used in the fall breeding seasons of 1916 to 1920, inclusive, and includes the total number of lambs dropped, living or dead, the percentage being based on the number of ewes having lambs. As shown in Table 1, the smallest result from flushing obtained in any one of the separate trials was 3.2 per cent. This was in the case of Lots 5 and 6, bred at the Morgan Horse Farm, Middlebury, Vt., in the fall of 1917. Both lots of ewes were in very good condition and ran on similar grass pastures. Although fed grain (14 parts corn, 5 parts bran, 1 part linseed-oil meal) at the rate of one-half pound each daily, the flushed lot made an average gain of only 6.3 pounds per head from September 10 until they were in lamb, while in the corre- sponding time the ewes not fed grain gained 4.6 pounds per head. The most pronounced effects of flushing are seen by comparing Lot 3 with Lot 4 and Lot 9 with Lots 10 and 11. In the case of Lots 3 and 4, both of which were in quite low condition when breeding commenced, there was an increased gain of 12.38 pounds in weight in the flushed over the unflushed lot and an increase of 30 per cent in lambs dropped. In the second instance the 15 unflushed ewes (Lot 9) gained 3.03 pounds each from September 10 to the time of getting in lamb, the average date of which was October 5. This lot produced 120 per cent of lambs. During the same time 150 per cent of lambs was yielded from a lot of 14 ewes (Lot 10) receiving grain, and 145.5 per cent for 11 ewes (Lot 11) kept on sufficiently good pasture to cause them to gain 10.96 pounds each from September 10 until in lamb., the average date being October 14. FLUSHING TO INCREASE LAMB YIELDS. Lambs dropped. I't CI II 1 . 140 140 140 161. 5 L50 145.5 147.6 152. 1 150 1 17. 05 ■z Average gain in weight. 11 Pounds. 4. 32 17. 13 2. 85 12.31 111. i"i 111. S 3. 76 5.80 10. 00 00 OS a "- 03 ~Z m — tlr f- - IB C > - Oct. 17,1916 Oct. 24,1916 Oct. 10,1917 Oct. 8, 1917 Oct. 6,1918 Oct. 14,1918 Oct. 20,1919 Oct. 7,1919 Del. 3,1920 Oct. 9, 1920 Oct. T2 a , coo l~ t- X X OS OSO o 5c — Ol CM OS OS 0: OS os z 3 33 O. O I ~ ". I- ro X) Ttj i-t C4 G 1003 oo r- n ■{■ o CN 01 0) CN ro OSOl Ol OS £ • • _;•.._■ _: _; _; c >•- p. a ^^— -r ft — z. ft ft as asirir as OS OS OS OS — < ° /. X. DC DO O O GO X X. DO X- as CSO •c -.o r^r^ oo Os o OS ^ OS OS i. "?T "t,~l rt _ II OS — o> .-. o o o — ' -* o rt c ^ © - _; «j _:-^-:~^ •O^j t: S to D.C — — z- \ z. 33 as OS as as as .as cs Q DO X &o co go • x X ~-, ■TOI~ -t" Ol — 53? : .- o M C Its ~"° «8 "** — n -' - § 1 os ■- P *: a) ;; *- 1 S > g-g r 5 : : IP! — - V i- ■ - '— - ~ z _ OS E o osS — -7 ^ — —"7 ™ "C ft i-* — i ; ; ol 3 - r^ 2 -,S — 1 r _J Q. — :/ :; - - — :,- • • - c H as as S rt p. ~ feu- £ '• : « stra orn, 4 parti parts meal xtra orn, i parts parts meal U O : :--'- WO p£ Num- ber iii lot. mio C ^O -* -H-H HO ^ Ol iH OS « — 1 — i Ol OS OS — " o - Ol -r (B.OOOHM -f o t- jz •-l-<-H fH W I- m'd ='■: oc ■* SO 00 ~ c - ft | - Sajc '»' OS o tc 00 ^cm — 00 OS OS OO OS OS h ^ *"* ^H ^H ^H M Ol ^ ^■S a. 60 H ^ 03— ' r- so coin 00 CO oo o CO CO ! ■a 'o. t~ fjoi-r ooo ■^ i-3 cc t- -V t- > 03 O o 1 ^^ ft. c. "_• ISO t^- t^ 00 os o 1 Ol 03 O a r. o. OS — — z. — oo"of os"-c in oT oT o3 a '~ p 5- o _j > > - u c es - ■/. <-c OZ ooo O o o z c , cc -.o I- [~ X OS o s - Ol o OS OS OS OS Z « ". 3o\h I ■x"-h'-1-" r^." ,-- CC sj: — is Ol OS Ol OS q A vera of firs ic ftft-^ ^ ^_ & x C X X - X 02 OS ■J. M CO co l^t^OO OS o 01 = _: OS OS Os OS "I Os"-T cTi-To* s = ~ fH — < ~"* ^^ £a CQ ftft a^^ ft ft 99 - as as as OS - Q X V. /. X X -f. 00 \ um- ber in lot. mo r-. t- i~ in >* 00 oi 01 • ph « '0 r- os OS „, o is. . Z 03 OS I a < ^ • 03 = ;. ~ — 'S r ~ be s- ^ c ~ ^ ^1 - / °ca c — as m S ?'" as * e5.S c ® S s_ i£ ' /J S-r 1 c3 = 5-0 ZZ\ _T) 03 go r x — '1-- ° ftgg-2 i;- — — o o =- o ^ .£ ^ ■32g2 = cc:- as m ® 5 3 « ® os "5 03 jj"0 ~~~ £ .-3 as as q 6 BULLETIN 906, U. S. DEPARTMENT OF AGRICULTURE. Breeds having a larger proportion of twin births than the South- downs can be expected to give larger returns from flushing. Also, most farm ewes should show larger increases in lamb yields as a result of flushing than were obtained in these experiments because of the fact that ordinarily they are thinner at the time of breeding. RELATION OF WEIGHT GAINS TO NUMBER OF TWINS. Individual weights of the ewes were kept, and these permit an analysis of the relation between actual gains and numbers of twins prpduced. Of 30 cases in which ewes lost 1 to 7 pounds during the breeding season, twins were produced by 11, or 37 per cent. Of 133 cases in which ewes gained up to 7 pounds during the breed- ing season, twins were produced by 45, or 34 per cent. Of 74 cases in which ewes gained 7 to 30 pounds during the breed- ing season, twins were produced by 33, or 44 per cent. These results suggest that in order to produce the largest number of twins, ewes should gain 7 pounds or more. FEED FOR FLUSHING. With regard to kinds of feed most effective and most economical for use in flushing, no data have been obtained except in the cases of Lots 11 and 14. These lots had first access to good growths of soy beans, while the grain-fed lots and the unflushed lots followed on the same grazing after the pasture-flushed lots were moved forward. The difference in the grazing so obtained caused Lot 11 to gain 3 pounds more than Lot 9 in 1918 and Lot 14 to gain 4 pounds more than Lot 12 in 1919. In 1918 the pasture-flushed Lot No. 11 pro- duced 4.5 per cent fewer lambs than the grain-flushed Lot No. 10, while in 1919 the pasture-flushed Lot No. 14 produced 4.8 per cent more lambs than the grain-flushed Lot No. 13. This would indicate that there is no decided advantage in the kind of feed used in flushing other than the saving in labor and more expensive grain feed when pasture flushing is practiced. There are times, however, owing to unfavorable weather, when good pas- ture is not available and pastures are too short even to hold ewes at their initial weight. In such cases grain could no doubt be economically used. EARLINESS OF LAMBING. In addition to the increase in the number of lambs from flushed ewes, it has been believed that the extra nourishment brought the ewes in heat earlier and thereby resulted in earlier lambs. This is a reasonable expectation when the rams are in service at the beginning of the breeding; season. FLUSHING TO INCREASE LAMB YIELDS. In all but two cases flushed ewes came to the first service earlier than those unflushed, the time ranging from one-half day to 8 days. UNIFORMITY OF LAMBS' AGES. Having all the lambs of about the same age is a great advantage in marketing and flock management. Since flushing brings ewes to service earlier, it should be a great aid in preventing late lambs. This expectation was not realized, however, in the experiments. More of the cases of not getting in lamb until the second or third service occurred among the flushed ewes and was of course followed by a larger proportion of late lambs. The number of ewes lambing to each successive service in each year is shown in the following table; the ewes which were bred after being interchanged between lots are not included : Table 2. — Number and per cent of ewes getting in lamb at each servici . Unflushed lots. Flushed lots. Year. Lot No. Ewes in lot. Service. Lot No. Ewes in lot. Service. 1st. 1 2d. 3d. 4th. 5th. 1st. 10 7 i:, 6 11 7 6 12 13 9 2d. 3d. Hh. 5th. 1916 1 3 5 7 9 15 10 19 17 15 8 15 7 1 t 5 4 3 4 1 2 5 1 .5 1 1 •0 2 4 6 8 10 11 13 14 16 17 25 15 20 13 14 11 21 21 20 16 id 7 3 3 3 2 7 3 4 3 1 2 3 1 6 6 2 1 1 1 2 (1 1 1 1 1917 1918 1919 12 - 25 13 9 2 1 1920 15 20 U 8 1 Total . . . 121 6S 34 56 28 15 13 4 3 176 96 40 25 5:. -v, ii 7 4 2 Percent. TWIN PRODUCTION AS AFFECTED BY AGE OF EWE. Flock records of the Bureau of Animal Industry show a gradual rise in the proportion of twins born until the ewes are 5 and 6 years old. There is a possibility that this is due in part to elimination of ewes not dropping twins. In our experimental flocks, however, ewes have never been discarded on that account, although some ev\ ts have been kept to advanced age that might have been disposed of one or two seasons earlier if it had not been for their marked prolifi- cacy. The figures given in Table 3 for ewes over 6 years old may, therefore, have been to a slight extent affected by selection. The data given include nine years' records. The ages are those at time of lambing and not at time bred. 8 BULLETIN 996, U. S. DEPARTMENT OF AGRICULTURE. Table 3.- -Effect of age of ewes on per cent of lambs dropped in experimental flocks. Age of ewes. ('uses Of lambing. Lambs dropped. Age of ewes. Cases of lambing. Lambs dropped. Years. 2 79 63 67 62 Per a nt. in. i 123.8 143. 3 143.5 Years. 6 49 35 22 8 Pir cent. 161.2 3 7 112. 8 4 8 113.6 ;, 9 162. 5 TWIN PRODUCTION AS AFFECTED BY BREED OF EWE. Evidence as to inheritance of fertility makes it appear that vari- ous breeds or strains have each an inherited limit of fertility and that the obtaining of the full possibility in any one season will be deter- mined by the extent to which the conditions and management favor the full utilization of the inherited capacity. Records of actual returns from 189 flocks representing 9 breeds for the seasons of 1919 and 1920 were obtained by the Animal Husbandry Division through the kindness of breeders who reported. The aver- age per cent of lambs in proportion to ewes, lambing in the spring, is shown below, also the separate record of the 2-year-old ewes (drop- ping lambs the first time) and the highest flock average reported for each breed. The table gives the average of the two seasons' reports. The high return for the breed is for a single season. Table 4. — Record of lamb crops, bi/ breeds, from 189 flocks. [Average of seasons 1919 and 1920.] Flocks. 2-year-old ewes. Aged ewes. Total ewes. Highest flock. Breed. Ewes. Lambs dropped. Ewes. Lambs dropped. Ewes. Lambs dropped. Ewes. Lambs dropped. Number. 26 11 IS 27 26 16 25 16 24 Number. 215 40 96 138 549 91 167 84 186 Per cent. 146 145 144 143 139 135 134 123 111 Number. 566 106 214 378 857 190 402 184 667 Per cent. 163 161 156 153 148 148 154 149 125 Number. 7sl 146 310 516 1,406 281 566 268 853 Per cent. 158 157 152 151 144 141 119 141 122 Number. 6 11 6 6 6 23 6 8 9 Per cent. 200 191 Oxford.... 200 Southdown 200 200 200 183 200 Rambouillet 177 TWIN PRODUCTION AS AFFECTED BY SIRE. A study of 334 cases of lambing from the services of 5 rams, none of which were used less than three seasons or on less than 20 ewes, does not indicate any important variation in proportions of twins that can be attributed to the sire. Such differences as were shown in the average number of lambs per service for each sire appeared to be due to differences in the ewes to which they were bred. Records of 380 lambings from services of 8 different sires were studied in relation to FLUSHING TO INCREASE LAMB YIELDS. 9 the ages of the rams at the time of service. The results did not show any connection between age of sire and number of twins among his "get." Overworked or run-down condition of a ram may result in a smaller lamb crop through a lack of numbers or of vitality of the sperm cells in the seminal fluid. Such a lack is most likely to occur in the case of ram lambs running in a flock with the ewes, and it would be more likely to result in ewes not getting in lamb than in a decreased number of twins. Though improbable, it is possible that a ewe might pro- duce two" ova and have one of them fail to become fertilized because of a lack in the number or vitality of the sperm cells. BREEDING FOR TWIN LAMBS. It appears quite logical to suppose that the proportion of twin births in a flock can be increased by selecting, for breeders, rams and ewes themselves born as twins. However, the facts do not bear out such a supposition. This does not preclude the possibility of increasing lamb yields by breeding, but selection for this purpose should be based on average yields of different strains rather than on records of individuals. The fact that a ram was born as a single or as a twin can not reason- ably be expected to have any relation to the number of twins among his offspring. The function of the sperm cells of the male is to fertilize the ova produced by the female, and under ordinary condi- tions the number and strength of the sperm cells is many times greater than actually needed. It is reasonable, however, to expect a son of a ewe that is a regular producer of twins to transmit some or all of his dam's capacity to his daughters. Any ewe's inheritance of capacity for bearing twins must therefore be traced through the prolificacy records of her female ancestors. In the breeding of the bureau's purebred Southdown flock, records have been obtained of 458 cases of lambing which include only ewes that have dropped lambs at least three times. A few of the ewes had eight or nine lambing records. The relation of the production of these ewes, the fact of their having been born as single or twin lambs, and also whether their sires and dams were born as singles or twins are shown in Table 5. It must be remembered that the fact of a ram or a ewe having been born as a single or twin is in itself an incomplete record of the dam's productive rapacity. As shown in the table twin-born ewes were found to be 4.7 per cent more prolific than those born singles. The highest record, however, is from ewes born as singles with both parents twins, and the second- highest record is for single-born ewes by single sires from twin dams. 10 BULLETIN 996, U. S. DEPARTMENT OF AGRICULTURE. There does not appear to be any connection between lamb produc- tion and the fact of sires and dams having been singles or twins. Table 5. — Effect of breeding on twin production of Southdown ewes in Government flock. Ewes' breeding. Number of ewes. Number of lambings. Lambs dropped Born twins: Sires twins, dams twins. . . Sires singles, dams twins . . Sires twins, dams singles. . Sires singles, dams singles. Average for twin ewes . . . Born singles: Sires twins, dams twins . . Sires singles, dams twins. . Sires twins, dams singles . . Sires singles, dams singles . Average for single ewes . 12 84 12 L34 Per cent. 133 142. g 116.6 142. 5 140.9 14 70 21 111 157.1 145.7 109.5 132.7 136.2 A safer way of appraising the possibility of increasing twin pro- duction by selection is to take into account the full records of female ancestors rather than a single birth in which the particular sire or particular dam was produced. VALUE OF TWIN LAMBS IN COMPARISON WITH SINGLES. Sheep raisers differ in their ideas of the desirable size of the lamb crop. At one extreme are ranchmen chiefly interested in wool pro- duction who consider twins as undesirable because feed conditions are unfavorable to a ewe's furnishing more than sufficient milk for one lamb. Even in such cases, however, it is always likely that a number of ewes will lose their lambs and a corresponding number of pairs of twins would allow transferring one from each pair of twins to a ewe in milk and without a lamb. This would render possible the rearing of 100 per cent of lambs. There are, also, a few breeders of registered sheep who believe that there is no gain in obtaining twin lambs. Their position is based upon the fact that some twin lambs do not develop so fully as singles. Since a good individual animal sold for breeding purposes may bring as much as or more than two inferior ones, single lambs might be an advantage, provided they always proved more valuable at selling age. At the other extreme are raisers of market lambs in whose hands a pair of twins, even though comparatively underdeveloped and sold perhaps at a lower price per pound, still will bring a much larger amount than the single lamb. With ewes lambing for the first time, it is less desirable to have twin lambs than with older ewes. Young ewes do not ordinarily milk so well nor look after their lambs so faithfully, and thus they FLUSHING TO INCREASE LAMB YIELDS. 11 have a greater rate of loss in twin lambs than older ewes. In most flocks, if not all, containing the ordinary proportion of ewes of vary- ing ages, the mark can well be set at 150 per cent of lambs in working for the greatest net returns. Possible disadvantages in twin lambs must come from one or all of three causes: (1) Greater rate of loss among twins; (2) slower rate of growth as lambs; or (3) inability to reach the same size, weight, and breeding value as single lambs. As regards the rate of loss, the experience of the Bureau of Animal Industry shows no greater losses among twins. In the lambing seasons of 1916 to 1920, inclusive, in the two flocks of Southdowns used in the experiments a total of 224 single lambs and 290 twin lambs was born. Of these 14.3 per cent of the single-born lambs died before reaching the age of 2 weeks, and 13.4 per cent of the twins. COMPARATIVE WEIGHTS OF SINGLE AND TWIN LAMBS. Comparative weights of twins and single lambs at six months old show that the milk received by the lambs is more important in in- fluencing growth than is birth as a single or twin. Records of 184 lambs dropped through three different years are grouped to show weights attained by both sexes and by single lambs, twin lambs, and lambs born as twins but having all of one ewe's milk (twins raised as singles) . The weights of the ram lambs include 3 or 4 wethers. Table 6. — Weights of 6- months-old twin and single lambs. Singles. Twin-. Twins raised as singles. Kind. Number. Average weight. Number. Average weight. Number. Average weight . 32 46 Pounds. 85.4 73.8 45 37 Pounds. 81.6 07.6 9 15 Pounds. 82 7 78 3 78 78.6 82 75.3 24 79.9 The twin-born ewe lambs averaged 6 pounds lighter at six months than those born singles, while in the case of ram lambs the difference was 4 pounds. In the smaller groups of twin lambs raised as sin- gles the ewes made an especially good growth, averaging more than the single lambs. 12 BULLETIN 996, U. S. DEPARTMENT OF AGRICULTURE. Three experiments were conducted to determine whether twin ewe lambs would catch up in weight with the singles if given an oppor- tunity. Of the 1915 crop, 19 head of singles and 22 twin-ewe lambs were fed separately for 112 days (December 8, 1915, to March 28, 1916). At the outset the singles were 8.6 pounds heavier and at the close of the test they were 1 1 pounds heavier. They received similar feed, but that eaten by the twins contained about 6 per cent more total energy. After running in the same lot on pasture until August 30, 1916, the single-born lambs were still 8 pounds heavier. In November, 1916, 8 head of single ewe lambs and 8 head of twins that had been raised as twins were placed in a similar experiment. The average daily ration fed the twins contained 0.28 pound of pro- tein and 1.84 therms of energy as compared with 0.24 pound of protein and 1.57 therms of energy for the singles. At the start the singles were 7 pounds heavier and at the close of the special feeding the weights were identical. The twins were somewhat fatter, how- ever, and after running with the others on pasture until August 25, 1917, were 2.4 pounds lighter. In the summer of 1917 an attempt was made to furnish lambs raised as twins sufficient extra grain to permit them to catch up with those born singles. From birth (about March 1) until July 1, there were 11 single ewe lambs and 3 born twins but raised as singles in one lot, and 8 twin-raised ewe lambs in the other lot. During that time the former ate 50 pounds of grain each and the twins 63 pounds, with the result that the twin lambs averaged 7 pounds lighter than the others. Subsequently the lambs were fed and pastured as one lot until December 29, 1917, at which time the singles were 3 pounds heavier. From December 29, 1917, to April 6, 1918, the single lambs ate an average daily ration of 1 pound of the following grain mix- ture: Cracked corn 100 parts, bran 30 parts, in connection with 2 pounds timothy hay and 2 pounds turnips per head. That eaten by the twin lambs consisted of If pounds of a mixture consisting of cracked corn 80 parts, oats 60 parts, bran 35 parts, in connection with 2 pounds of timothy hay and 2 pounds of turnips. On April 6 the singles weighed 107 pounds and the twins 1 10. That their extra gain was not fat is shown by the fact that after running on pasture with no feed until August 10, the twins were 3.1 pounds heavier per head than the singles. A comparison of the gains made by singles, twins raised as twins, and twins raised as singles is shown in Table 7. The fact that lambs born as twins but receiving all of one ewe's milk often equal and sometimes outweigh single lambs makes the matter appear to be one of nourishment. The slight lack of growth and development of twin-ewe lambs below that of singles at market age is not serious in comparison with the advantage of larger numbers. FLUSHING TO INCREASE LAMB YIELDS. 13 Table 7. — Gains made by single and twin lambs in bureau flocks at Middlebury, 17., and Beltsville, Md. Single lambs. Twin lambs raised as (wins. Twin lambs raised as singles. Averag e weight. Average weight. Average weight. 03 <3 C3 £) ,d o o o e & s, c z ■d ,g g o o 6 L ft a 6 £ = n a o a fl R a S3 pa M 101.0 Average: Rams 1.8 8.5 64.6 91.1 123.1 25 8.1 5S.0 83.1 121.0 6 7.7 58. 1 86.0 116.6 Ewes 32 8.7 59.4 77.9 110.5 29 7.3 49.1 68.9 ION. 4 13 7.4 62.1 79.9 111.7 Beltsvilb flock: 1917— Rams . 2 8.4 40.5 69.0 96.0 3 6.9 45.2 73. 5 94.7 2 6.2 38.8 67. 8 91.8 Ewes.. 8 7.7 42. 8 63.6 82.0 3 6.4 39.8 60.8 86.7 1 6.6 55.0 74.0 96.0 1918 — Rams . 5 3 8.3 8.2 59.9 56.5 82. 7 69.0 109. 4 104. 7 6 3 7.1 5.8 49.1 42.5 77.3 6S.2 119.8 111.7 Ewes. . 1 5.7 39.6 61.0 83.0 1919— Rams . 9 8.0 54.4 65.1 100.2 5 7.3 45.6 60.6 97. 8 4 6.6 55. 1 66.3 97.0 Ewes.. 13 8.1 53.2 66. 5 91.2 8 6.6 50.7 64.9 90.3 3 6.7 53.5 68. 7 92.7 1920— Rams . 7 8.4 60.3 82.0 124.5 14 7.2 51.0 78.9 117.5 1 6.8 40.0 67.0 106.0 Ewes.. 17 8.3 47.8 64.5 98.8 17 6.5 44.8 64.8 L02.3 2 6.9 55. 5 70.0 92.0 Average: Rams 23 8.2 56.2 74.4 109.2 28 7.2 49.0 74.7 112.0 7' 6.5 is. 3 66.8 96.8 Ewes 41 8.1 49.2 65.3 93.6 31 6.5 45.6 64. 8 9S.6 7 6.6 52. 2 68.7 91.6 Average: All rams... 41 8.4 59.9 81.7 115.3 53 7.6 53. 2 78.7 116.3 13 7.0 52. 8 75.7 105.9 All ewes... 73 8.4 53.6 70.9 99.8 60 6.9 47.3 66.7 102.3 20 7.1 58. 6 75.9 104. 7 All lambs.. 114 S.4 55.9 74.8 105.8 113 7.2 50. 2 72.3 L09.5 33 7.1 56.9 75.9 105.2 1 Placed in special experiment and weight not comparable. SUMMARY. 1. Feeding at breeding time to increase the number of twins pro- duced by ewes is called "flushing." 2. The percentage of lambs produced by a flock depends upon the number of dry ewes and the proportion of ewes producing twins and triplets. 3. The practical advantage of flushing lies in the production of twins, which in turn depends upon the number of ova produced by the ewe. 4. Experiments reported herein indicate that ewes getting in lamb first produce the largest percentage of twins. 5. Data from experimental work indicate that ewes should gain at least 7 pounds a head during the breeding season to obtain largest percentage of twins. 6. There seems to be a natural tendency toward twin production, which varies in different breeds. 7. It is only in extreme cases that the ram has shown any influence on the number of twin lambs produced by the flock. 14 BULLETIN 996, U. S. DEPARTMENT OF AGRICULTURE. 8. Records do not indicate that ewes born twins of twin parentage are more prolific than single ewes. 9. There is no material difference in the size of twins and singles when fully developed. 10. Although at market age twin lambs would not weigh so much as singles, the difference in weight would be small compared to the total weight of the lambs for sale, thus making twins far more profit- able. ADDITIONAL COPIES OF THIS PUBLICATION MAY BE PROCURED FROM THE SUPERINTENDENT OF DOCUMENTS GOVERNMENT PRINTING OFFICE WASHINGTON, D. C. AT 5 CENTS PER COPY V LIBRARY OF CONGRESS 002 847 759 4