1. In the course of a 2-year field trial (1959–60) conducted in mid-Wales, 525 cattle were entered by 317 co-operating farmers for treatment with p.m.s. in the induction of multiple-pregnancy. Most animals were commercial, non-pedigree cattle maintained in accord with local custom. The cattle were of mixed ages and included some milking cows as well as animals used in calf rearing.

2. The method of hormone treatment tested consists of a single subcutaneous injection of a freeze-dried preparation of p.m.s. administered in the follicular phase of the oestrous cycle. Most cows were injected 16 or 17 days after a previous oestrus and bred by a.i. at the heat period following injection.

3. Of the 525 cattle injected 416 provide comprehensive data on factors affecting ovulatory response; seventy cattle, which had a ‘silent heat’ after injection, provide additional but partial information. In thirty-nine cattle, it was not possible to obtain reliable evidence of response. Data on ovulation and pregnancy were obtained by examinations per rectum conducted during the early weeks of pregnancy.

4. Ovulation data relate to 416 cows which were injected, came in oestrus shortly afterwards and inseminated, then examined for evidence of fresh corporea lutea at the following mid-cycle stage. Five dosage levels of p.m.s. were employed (800, 1000, 1200, 1600 and 2000 i.u.). The normal process of ovulation did not appear adversely affected by any of the doses employed. 173 of the 416 cows (41·6%) produced more than one egg.

5. There is clear evidence of a dose-response relationship over the dosage range employed. The mean number of ovulations varies from 1·43 at 800 i.u. to 3·97 at 2000 i.u. The percentage of cattle shedding additional eggs rises from 33 at 800 i.u. to 56–6 at 2,000 i.u., but increasing dosage level leads to considerable variability in ovulatory response: range of ovulations at 800 i.u. is 1–5, and at 2000 i.u. is 1–25.

6. Little affect was observed in ovulatory response according to the stage of p.m.s. administration. At low dosage levels, an interval of at least 2 days between injection and oestrus was necessary to allow response to occur. Among factors affecting response to a standard dose of p.m.s., lactation and state of normal ovarian activity appeared to have some importance. There was no evidence of an effect on cycle length. Treatment did not give rise to abnormal mating behaviour and the incidence of adverse side-effects following p.m.s. injection was negligible. The occurrence of large unovulated follicles occurred more frequently following treatment with 2000 i.u. than with lesser dosage levels. 59·1% of all ovulations occurred in the right ovary and 40·9% in the left. In examinations made prior to injection, the incidence of spontaneous double ovulations was 4·1%.

7. With the p.m.s. approach, it is found that one problem in cattle is that of silent heat, the condition where ovulation is not accompanied by the usual symptoms of oestrus. In 486 animals for which information was obtained, silent heat occurred in 70 (14·4%). There is some tendency for the condition to occur more frequently where low dosage levels are used or where p.m.s. is administered very late in the oestrous cycle. Lactation appears to affect the incidence of the defect. The lowest incidence (4·8%) occurs in ‘dry’ cows and the highest (22·7%) in cattle rearing three to four calves. Superovulation occurred less frequently at silent heat than at normal oestrus. There is no evidence that p.m.s. injection is the direct cause of the condition. It appears that time of ovulation in animals showing silent heat is comparable with that for cattle showing the normal symptoms.

8. Of the 416 cattle put to a.i. shortly after p.m.s. administration, 317 (76·2%) were pregnant when examined some 6 weeks after first insemination. This compares closely with the normal 30–60 day non-return figure for the mid-Wales area (78·5%). Cattle conceived as readily after injection with 2000 i.u. as after treatment with 800 i.u. Even where animals shed many eggs (e.g. more than six) the conception percentage (84·0) was high. It is clear that this form of therapy does not adversely affect the processes of ovulation, fertilization and implantation. There is some suggestion that p.m.s. aided conception in repeat-breeder cattle. Examinations in the mid-cycle stage following a.i. did not adversely affect conception.

9. Data on the number of eggs surviving (as foetuses) in the 6th week after mating, in relation to the initial number of ovulations, clearly show a heavy loss in eggs in early pregnancy. 41·6% of cattle shed additional eggs but only 23·1% possessed multiples in early pregnancy.

10. Reduction to a single foetus seemed particularly liable to occur in cattle producing two eggs-only. Of sixty-seven double-ovulating animals that conceived, thirty-two possessed twins at the 6th week (47·7%). The survival of both eggs in such cows was better when one egg was shed by each ovary (61·5%) than where both ovulations occurred in the one ovary (28·6%). Analysis of data on egg survival according to factors such as age, breed, time of treatment did not reveal differences.

11. Trans-uterine migration of eggs was rarely (1 case in 197 examinations) noted. The ability of a single uterine horn to sustain more than two foetuses was limited. In cattle where three to six eggs were shed by an ovary, the survival of twins in the associated uterine horn was observed in a majority (57%) of cases. Where many eggs (ten to fifteen) were shed by an ovary, twins were rarely sustained by the associated uterine horn.

12. A steady increase in the percentage of cows carrying multiples (at 6 weeks) was noted over the dosage range employed. 20·8% of cows injected with 800 i.u. carried multiples, the figures for doses of 1000, 1200 1600, and 2000 i.u. being 25·2, 31·3, 41·7 and 42·7, respectively. Multiples in the form of triplets and quadruplets occurred more frequently at the higher dosage levels than at 800–1000 i.u. of all cows with multiples considered, 36·5% possessed more than two foetuses at examination in the 6th week.

13. Pregnancy diagnosis examinations were conducted over a period ranging from 35 to 101 days after mating, with most falling in the period 41–44 days. It appeared that the most appropriate time for diagnosing multiples was at the end of the 6th week of pregnancy. The need to conduct diagnosis at a specific time after mating, and the fact that the examination may lead to total litter loss in many animals, makes the procedure unsuitable for any widespread practice. Partial litter loss (loss of one foetus) was observed in 12·5% of cows carrying twins at 6 weeks and in 20·0% of those with triplets.

14. In nineteen cows, each possessing either three or four foetuses, an attempt was made in the 6th week of pregnancy to control litter size to two. This was by rupture of excess amniotic vesicles by manipulation per rectum. Two of those so treated eventually produced calves, the others either aborting or returning in oestrus soon after the control procedure. The method is unsuitable for farm application for reason of the delay in returning to normal breeding after such cases of failure of pregnancy.

15. A substantial proportion (32·5%) of those cows which were subjected to pregnancy diagnosis lost the conceptus a short time afterwards. Foetal loss in cows with twins was greater (43–46%) where foetuses were both in the one horn than where distributed between both horns (31%). The various factors which were possibly involved in foetal loss are discussed. Evidence was obtained that total foetal loss occurred more frequently in animals shedding many eggs than in those shedding few.

16. Information is recorded for 147 injected cattle which produced single calves following conception at the oestrus immediately following hormone treatment. Additional data on single calvings are provided by 200 untreated cattle which were studied in the area of the trial. These data show birth weight of single calves to be 78·6 lb., and the incidence of calf mortality, retention of foetal membranes and difficult calvings to be 4·9, and 6·0%, respectively. The mean duration of the gestation period was 283·4 days.

17. The outcome of multiple calving was observed in forty-four cattle (thirty-five sets of twins, eight sets of triplets and one set of quintuplets). As a result of earlier examinations, farmers were usually alert to the fact that cows were carrying twins. Herd owners were encouraged to provide such cows with an additional 2 cwt. of a suitable concentrate feed during the final 2 months of pregnancy. Although a number of herd owners followed the recommended feeding programme, in a number of instances the most appropriate pre-calving preparation was not given. The outcome of calvings in relation to the level of additional feeding is considered under three headings, high, moderate and low plane.

18. Given good feeding (high plane) the outcome of twin calvings was satisfactory. Results suggest that the problems often associated with twins (low-weight calves, high rate of mortality, retained foetal membranes, loss of conditon and poor subsequent breeding of the dam) may largely be overcome by an adequate ‘steaming-up’ policy. In cows adequately prepared, birth weight ranging from 59 to 89 lb. were recorded (mean 73·9 lb.) Mean gestation was a little less (280·6 days) than for singles. The incidence of retained foetal membranes (18·2%) was greater than that for single-bearing cows, but did not constitute a serious problem in any animal.

19. Less satisfactory results were noted in cows not given adequate preparation for twins. Where triplets were carried, the outcome for cow and calves, regardless of the feeding attention given, was unsatisfactory. Triplet-bearing cows calved earlier, and the incidence of retained foetal membranes, difficult calvings and calf deaths was much above that noted for the twin-bearing animals. The single set of quintuplets was born prematurely, all calves being dead.

20. Problems in the field application of the p.m.s. technique are considered in the light of the results obtained in the trial and from discussion with farmers in the area of work.

21. Information was obtained on the incidence of natural twins in 3826 calvings in the mid-Wales area. The figure of 2·82% compares closely with that noted in other surveys in the United Kingdom. Literature on the incidence is reviewed.

22. It is clear, from the results obtained in this trial, that the technical problems facing commercial usage of p.m.s. are too numerous to allow any practical recommendation to be made. While none seems to be of such a nature as to make the eventual development of a twinning technique unlikely, there is little hope of immediate solution to many of them. The lines of work which may lead to more satisfactory results are discussed in the final section of the paper.