In dealing with fertility in sheep there are a number of initial difficulties. On the one hand, there is generally a lack of any uniform definition of such terms as ‘fertility’, ‘fecundity’ and ‘prolificacy’ (cf. Marshall & Hammond, 1947; Lush, 1938; Rice, 1942; Asdell, 1946; and others); on the other hand, investigators have been confronted with a real difficulty in procuring suitable information which would conform to the requirements of any single and adequate definition of fertility. Because of this, fertility figures have been calculated in different ways* and may differ by as much as 30%. The situation is at present so confused that reference to similar work, especially when the original papers are not available, or the terms not clearly denned, is often of dubious value if not altogether misleading. In the past the data analysed were mostly based on farmers' answers to questionnaires or on flock records. The limitations of such methods are only too obvious; yet it must be clearly realized that in commercial flocks there is no possibility of any basic departure from them, even though their accuracy could in many cases be improved. In other words, only such information is collected as the circumstances allow. This, for instance, may be the number of lambs docked per ewes put to ram in one flock and number of lambs docked per ewes lambed in another.

In the course of two breeding seasons 137 mature to aged ewes of mixed breeds but predominantly Border Leicester × Cheviots, Dorset × Cheviots, Suffolks and Hampshires and their crosses, have been injected with PMS or PU, and at different levels of dosage, and slaughtered at intervals after service to determine the sequence of events following treatment. In addition ten ewe lambs received similar treatment and eighty-one flock Suffolk and Romney Marsh ewes, of which forty-nine were injected, have been studied. The results may be summarized as follows:

1. The injection of 500–2000 i.u. PMS on the 12th day of the oestrous cycle will regularly induce multiple ovulations at the ensuing heat.

2. There is a significant dose-response relationship between 500 and 2000 i.u. PMS. Mean ovulation rates observed were: 500 i.u., 4·1; 1000 i.u., 10·6; 2000 i.u., 15·8. There is, however, considerable variation, the ranges being 2–9, 4–33 and 8–29 respectively.

3. The injection of 1000 i.u. PU subcutaneously on the 12th day of the cycle causes marked ovarian and cyclic abnormalities, and heat may be entirely suppressed. When injected intravenously at heat following a priming injection with PMS on the 12th day there is no evidence that it increases the rate of ovulation; in other words, the ewe can ovulate up to at least thirty ova without difficulty, provided the follicles are matured.

4. PMS levels of 500 and 1000 i.u. do not cause any apparent ovarian abnormalities apart from superovulation. A very high proportion of all follicles developed rupture and form apparently normal corpora lutea. PMS at 2000 i.u. causes lutein cysts and ovulation may be inhibited.

5. PMS at 500 and 1000 i.u. causes no cyclic abnormalities, although cycle length is slightly shortened. In the event of the ewe failing to conceive, the subsequent oestrus is normal, and ovulation and fertilization of the ova occur.

6. Commercial PMS is probably as effective as fresh PMS provided it is correctly standardized. There seems no reason to suspect differences in multiple ovulating efficiency between different batches of fresh PMS.

7. While breed differences in response possibly exist they were not observed in these experiments. Nor were differences apparently related to the relative time of injection within the breeding season.

8. There is an ovarian weight-PMS level, doseresponse relationship which is almost entirely accounted for by the numbers of corpora lutea and hence of luteal tissue.

9. Multiple-ovulated ova are highly fertilizable. However, when fifteen or more are shed the rate of tubal transport is considerably accelerated and the proportion fertilized appears to fall. One- and two- cell ova have been recovered from the uterus within 48 hr. of service. Nine fertilized ova have been recovered from one ewe.

10. Considerable embryonic mortality occurs before attachment of the blastocysts. None the less, multiple impregnation does occur but is followed by further early death. Up to thirteen attachments have been observed in one ewe, but all but three were showing signs of regression by the 19th day.

11. Post-attachment mortality takes the form of an initial retardation commencing about the 15th day, embryos apparently dying a day or so later. The peak of mortality occurs between the 17th and 19th day, by which time definite signs of resorption are apparent.

12. By 21 days equilibrium has been reached, the mean numbers of survivals of those ewes pregnant being some 260%. This level is maintained without further loss until the 68th day. Several cases of four normal foetuses were noted up to the 41st day, and one uterus containing six perfectly normal foetuses was recovered at 61 days.

13. The overall fertility indicated by these slaughterhouse ewes varies with the level of PMS administered. While equilibrium is achieved by animals conceiving, at about 250–260% viable foetuses, regardless of the dose injected, there is increasing proportion of returns to service with increasing dosage. For the most part this appears due to hormonal imbalance at the time of ovulation and fertilization when an excessive number (> 15) ova are shed. This results in accelerated ovum transport through the tubes and lowered rate of fertilization. Between ovulation rates of 4 and 12, produced by 500 i.u., rate of conception is extremely high. Of twenty-five ewes receiving 500 i.u. PMS, twenty-three (92%) conceived to the first fertile service. The conception rates were reduced to 80 and 60% by 1000 and 2000 i.u. respectively.

14. Although after the end of the 3rd week there is no direct evidence of further foetal mortality, the percentage of lambs born in the flock ewes injected with 500 i.u. PMS was only 192% of those lambing or 167% of all ewes mated (147% for controls). This indicates a loss late in pregnancy. This can in part accounted for by abortion; one ewe aborted, due, is believed, to her carrying more lambs than she was physically capable of retaining.

15. OffifteenflockSuffolksinjectedinthe 2nd year with 500 i.u. PMS, thirteen lambed, one aborted and one did not conceive. The thirteen ewes lambing presented twenty-five lambs, including one set triplets and one of quadruplets. All the latter were viable, but two sets of twins were bom dead. One triplet was lost on fostering and one other lamb was lost, so nineteen were tailed. Of fifteen controls, all lambed, giving twenty-one lambs bom of which twenty survived. Conception to first fertile service was higher in the injected than in the control ewes. In the preceding year when 1000 i.u. was given, both conception and lambing rates were appreciably lowered as compared with untreated controls. The significance of this in respect to the level of PMS administered is stressed.

16. It is concluded that the major part of the embryonic loss is due to a uterine environment which is incapable of supporting more than a limited number of embryos. It is considered possible that there are breed differences in this, and that each breed, and individual within the breed, has ‘Maximum Potential Fertility’.

17. This ‘Maximum Potential Fertility’ is not attained in normal breeding practice, since the number of ova shed by the ewe tends to constitute a limiting factor. Use of a level of PMS which will induce between three and ten ovulations—in the case of this experiment, 500 i.u.—removes this limiting factor, giving the individual full opportunity to express its full potential breeding capabilities.

18. There is some indication that the use of PMS in lambs may not give good results, and its use for the while should be restricted to mature ewes.

19. Since 500 i.u. PMS has resulted in a higher conception rate to first service and to a greater number of lambs born, it is apparent that this technique of injecting on the 12th day of the cycle a level of PMS which will result in a moderate number of ovulations is worthy of exhaustive field trials, notwithstanding the high post-natal mortality observed in the few cases which were allowed to go to term.

20. In view of the higher indicated fertility of the slaughtered ewes which were mainly Border Leicester cross and Dorset × Cheviots—normally very highly fertile ewes as compared with the flock Suffolks—it is considered possible that this technique will yield best results with more highly fertile breeds.

21. To have any hope for successful increase of fertility in mammals following gonadotrophin treatment the hormone must be administered at a time and at such a concentration that it will augment, but not upset, normal hypophysial and ovarian function.

1. The experiment previously described (series 4, Holmes, 1949) on the effect of massive applications of nitrogenous fertilizers on the productivity of a ryegrass dominant pasture was continued for 3 years (1946, 1947, 1948). The manurial treatments ranged from none to the application of 312 lb. nitrogen per acre and this was applied with and without 135 lb. P2O5 and 168 lb. K2O per acre. Farmyard manure was applied to one block in 1948.

2. With the heaviest nitrogen treatment plus phosphate and potash the average yield for 3 years was 8000 lb. dry matter and 1640 lb. crude protein (similar to the yield in 1946) compared with a control yield of 4720 lb. dry matter and 590 lb. crude protein. The yields declined from year to year when phosphate and potash were not applied, the decline being greatest with the heaviest application of nitrogen.

3. The seasonal distribution of the yield of herbage was very considerably modified by the time when fertilizers were applied.

4. The botanical composition of the swards was related to the yield, 70% of the grasses in the highest yielding sward being perennial ryegrass and cocksfoot while the lowest yielding sward contained only 35% of these grasses.

5. The manurial treatments had no effect on the pH, loss on ignition or the content of readily soluble P2O5 in the soil, but in the first year there was some evidence of a general reduction of readily soluble K2O to a low level.

1. A small-scale plot experiment was carried out from 1947 till 1949 to study the effect of massive dressings of nitrogen with and without phosphate and potash on the yield of a ryegrass-dominant sward. A 4 × 4 Graeco-Latin square was used.

The nitrogen treatments were: (1) no nitrogenous fertilizer, (2) 260 lb., (3) 520 lb., (4) 416 lb. nitrogen per acre per annum. These applications were divided into four or five equal dressings, the first being applied in March and the others after successive cuts.

In 1948 and 1949 the following mineral treatments were also applied: (A) no mineral fertilizer, (B) 180 lb. K2O per acre, (C) 120 lb. P2O5 per acre, and (D) treatments B and C combined. These treatments were applied in two parts, half in spring and half in midsummer.

2. Applications of phosphate did not affect the yields of herbage but yields were severely restricted in the absence of potash.

3. Where potash was applied there was no significant change in the annual yield of herbage from 1947 to 1949 for any one of the treatments. The nitrogen treatments increased the yields from 4300 lb. dry matter and 600 lb. crude protein per acre with treatment 1 to 9000 lb. dry matter and 1900 lb. crude protein with treatment 3. The yield response of dry matter to nitrogen applications was curvilinear, but that for crude protein was rectilinear.

4. Increasing the application of nitrogen per cut from 52 to 104 lb. per acre raised the protein content in the dry matter of the herbage from 15·20 to 18·25%.

5. The plots receiving the nitrogen treatments gave earlier production and also produced a greater proportion of the total yield in July-September than the untreated control plots.

6. Where potash was present the average net efficiency of recovery of nitrogen for 3 years was 47, 39 and 37% for treatments 2, 3 and 4 respectively.

7. Where potash was present the content of vigorous grasses in the sward (perennial ryegrass and timothy) was increased in relation to the quantity of nitrogen applied, from approximately 40% in treatment 1 to 75% in treatment 3. The content of vigorous grasses in the sward was reduced in the absence of potash.

8. The results and their practical implications are discussed.

1. A statistical study of the causes of variation in milk yields has been made on the basis of 5000 lactation records from twelve leading herds of Ayrshire cattle in south-west Scotland for the period 1930–9.

2. The unit of measurement of milk production employed was the yield during the first 180 days of the lactation period. This eliminated the effect of variations in length of current calving interval on milk yield.

3. The effect of month of calving on milk yield varied significantly between herds, and it was shown that correction factors for month of calving should be calculated on a within-herd basis. The average difference in 180-day yield between the summer and winter calvers of all herds was about 10% in favour of winter calvers.

4. The milk yield of a cow was found to be influenced both by the number of her previous lactations and also by her age at calving. The types of corrections for age employed by previous investigators were discussed, and it was shown that percentage corrections are the most satisfactory.

5. There was a positive correlation between milk yield and length of preceding calving interval. From an economic point of view, the optimum length of calving interval was about 400 days for the first lactation, and 1 year for subsequent ones. Corrections for preceding calving interval, like those for age, were most satisfactory when they were proportionate and not additive.

6. No significant differences were found among the first three records of a cow in their ability to indicate her actual production capacity. The probable performance of a cow in any lactation was predicted as accurately from the lactation immediately preceding it as from the average of a number of previous lactations.

7. The average repeatability of milk yield was 0.46, and heritability was of the order of 0.25–0.30.

8. The probable effect on herd improvement of selecting breeding females was found to be very small, extremely little genetic progress being attained by this method in the twelve herds.

1. A statistical study of factors affecting variation in persistency of lactation has been made using 5000 lactation records from twelve leading herds of Ayrshire cattle in south-west Scotland for the period 1930–9.

2. It is suggested that a satisfactory numerical expression for the shape of the lactation curve could be obtained from the formula, Persistency = where A is the milk yield during the first 180 days and B is the initial milk yield, namely, the milk yield during the first ten weeks of lactation.

3. The interrelationships of persistency, initial milk yield and 180-day milk yield showed that these three characteristics are positively correlated with one another on a between cows within herd basis. This meant that it should be possible to combine high initial milk yield with high persistency to obtain increased total production.

4. The effect of month of calving on persistency of lactation varied significantly between herds. On the average, the highest persistency was attained by cows calving in the winter, and the lowest by summer calvers.

5. The variation of persistency with age showed that it is necessary to correct for the high persistency of first calvers when comparing cows with different lactations.

6. The average repeatability of persistency was 0.242 in the present material, while heritability was of the order of 0.10–0.15.

7. It has been concluded that improved feeding and management would bring about the greatest returns in the direction of improved persistency.

1. On the basis of 5000 records from twelve leading herds of Ayrshire cattle in south-west Scotland for the period 1930–9, a statistical study has been made of some of the causes of variations which occur in butterfat percentage.

2. Butterfat percentage was found to be affected only very slightly by month of calving, age and length of calving intervals.

3. Selection for milk yield would not affect butterfat percentage adversely to any appreciable extent.

4. The average repeatability of butterfat percentage was 0·691 in the present material, and heritability was of the order of 0·50–0·60.

5. Mass selection would be an effective tool in bringing about improvement in butterfat percentage.

The cost of conducting digestibility trials with cows is high, due to the need for either elaborate equipment for separating faeces and urine or a considerable number of assistants to remain constantly behind the cows catching the excreta as they are voided. Manual collection of excreta was used extensively by earlier workers, and, in a recent report of the most suitable arrangements for experiments using this method, Eheart, Holdaway & Pratt (1945) found that one attendant was necessary for every cow on test with a trained chemist also present in the shed at all times.

In the carrying out of comparative determinations of the supplemental values in pig feeding of the proteins in different protein concentrates, the writers consider the following conditions to be fundamental, and must be observed if reliable results are to be obtained:

(1) The comparison must be conducted over the period from weaning to about 90 lb. live weight. Measurements made in the later stages of growth can have no meaning, since supplementation of the basal mixture (cereal plus middlings or fine bran) with the protein-rich concentrates under comparison has no significant effect at tins stage on the rate of growth or nitrogen retention. Examples of the misleading results that are obtained by prolonging the period of comparison to 150 or 200 lb. live weight, instead of terminating the trial before the attainment of 90 lb. live weight, were encountered during the course of the present investigation, and are cited in the text.

(2) The diets to be compared must be so designed as to contain, as nearly as possible, equal percentages of ‘total digestible nutrients’, as based on computations from digestion coefficients derived from digestion trials with pigs.

(3) The daily allowances of meal must be scaled to live weight, and the diet must be sufficiently palatable to ensure clean food consumption.

(4) The diets must make adequate provision for the mineral and vitamin requirements of young growing pigs.

(5) The comparison in the case of growth experiments must be made by the method of individual feeding, and in accordance with the requirements of the modern statistical lay-out, so that the records of live-weight gain and food consumption may be submitted to statistical analysis.

(6) The diet forming the standard for comparison should contain the minimum amount of protein needed for the maximum rate of growth compatible with the available supply of net energy. In the case of white-fish meal, which was the standard protein supplement in the present comparisons, this condition has been found in earlier work to be realized when a diet composed substantially of barley meal and middlings contains 7% of this animal-protein food.

(7) In the initial tests, the experimental diet should contain, as nearly as possible, the same percentage of digestible crude protein as the standard diet, and the percentage of the protein supplement under test should be such as to provide as much digestible crude protein as is contained in the 7% of white-fish meal in the standard diet. Under such conditions of feeding, differences of protein quality in the two feeding treatments will be the factor responsible for any observed differences in the rate of live-weight increase or of nitrogen retention.

Should the pigs on the experimental treatment display a poorer rate of growth, or of nitrogen retention, than those on the standard treatment, then further trials should be undertaken in which the percentage of protein supplement under test is progressively increased, whilst still maintaining equality between the experimental and standard diets in respect of ‘total digestible nutrients’, until a level is reached at which the experimental pigs show a rate of live-weight increase, or of nitrogen retention, equal to that of the pigs on the standard treatment. The final stage of the comparison, therefore, is based on the results of feeding tests in which the maximum possible rate of growth, or of nitrogen retention, is obtained, both on the standard and experimental treatments, with diets supplying, respectively, the minimum amounts of the two types of protein supply.

The object of the present investigation has been to compare the supplemental values of a typical animal-protein concentrate (white-fish meal) and vegetable-protein concentrate (ex. dec. ground-nut meal). The aim has been to determine what percentage of ex. dec. ground-nut meal must be incorporated with a basal diet, composed approximately of 2 parts by weight of barley meal, 1 part of middlings (or fine bran) and a small allowance of lucerne meal and minerals, to promote the same rate of growth, or of nitrogen retention, in young pigs as is obtained with a standard diet containing roughly the same proportions of the basal foods in conjunction with 7% white-fish meal, the diets under comparison being made up so as to contain equal percentages of ‘total digestible nutrients’.

On the basis of pig digestion trials, 8 parts by weight of ex. dec. ground-nut meal contains as much digestible crude protein as 7 parts of whitefish meal. The diets under comparison in the initial growth and balance trials were made up, therefore, to contain, respectively, 7% of white-fish meal and 8% of ex. dec. ground-nut meal. Their contents of crude protein, digestible crude protein and ‘total digestible nutrients’ were approximately equal. Over a range of live weight from 36 to 90 lb., the pigs receiving the white-fish meal averaged 0·92 lb. of live-weight increase per day, with a mean efficiency of food conversion of 3·08 lb. per lb. of live-weight gain. The corresponding figures for the pigs receiving the ex. dec. ground-nut meal were 0·73 and 3·87 lb. These differences were strongly significant and pointed clearly to the poorer supplementa. value of the vegetable protein supplement. Confirmation of this finding was given by the results of nitrogen-balance trials on young pigs. At all stages, pig (1) receiving the standard diet containing 7% of white-fish meal was retaining nitrogen at a distinctly higher rate than pig (2) subsisting on the diet containing 8% of ex. dec. ground-nut meal. Over the 43 days of continuous measurement, pig (1) consumed 1719·5 g. of nitrogen, and retained 720·5 g., or 41·9% of the total nitrogen intake; whereas pig (2), which had steadily fallen behind in live weight during this period, consumed a total of 1591·9 g. of nitrogen, and retained 566·6 g., or only 35·6% of the ingested nitrogen. It was clear from the results that the protein assimilated from the diet containing ground-nut meal is not so suitably balanced in respect of amino-acids as the digestible protein in the standard diet, and is subject on this account to a higher degree of de-amination, with a correspondingly greater wastage in terms of urinary nitrogen.

This conclusion was substantiated by re-calculating the results for the same range of live weight, namely, 62·92 lb., a range which falls within the limits that have been shown to be necessary for a sensitive comparison. For this live-weight gain, pig (1), receiving white-fish meal, required 22 days, over which period the total consumption of nitrogen amounted to 771·5 g., of which 327·6 g., representing 42·5% of the ingested nitrogen, was retained, and 226·9 g. (29·4% of the nitrogen intake) eliminated in the urine. The corresponding results for pig (2) receiving groundnut meal were distinctly inferior. The time needed for the same gain in live weight averaged 25·3 days; the necessary nitrogen consumption amounted to 859·3g., and of this, 301·8g.(35·1% of the intake) was voided in the urine and 309·5 g. (36·0%) was retained in the body of the animal.

A succession of growth trials was next carried out in which the effect of increasing the level of ex. dec. ground-nut meal to 12, 14, 15, 16, 17 and 20%on the rate of live-weight increase between weaning and 90 lb. live weight was investigated. Although the successive increases led to progressive improvement of the results in relation to those obtained on the standard diet containing 7% of white-fish meal, was not until the 20% level of feeding was reached that the diet containing ex. dec. ground-nut meal gave as good results in respect of rate of growth and efficiency of food conversion as were obtained by the use of the standard diet.

The same result was obtained in the investigation of the problem from the standpoint of nitrogen retention, the inclusion in the experimental diet of as much as 20% of ex. dec. ground-nut meal being found necessary to so rectify the ill-balance of the amino-acids in the cereal part of the diet as to make possible, between weaning and 90 lb. live weight, the maximum rate of nitrogen retention such as was obtained in the case of the pigs receiving the control diet containing 7% of white-fish meal.

In view of the variation from consignment to consignment of the protein content of such proteinrich concentrates as white-fish meal and ex. dec. ground-nut meal, it is necessary to define the different diets under test in terms of their protein content rather than their content of protein supplement. Examined from this standpoint, the results of the growth trial and the balance trial, in which ex. dec. ground-nut meal formed 20% of the diet under test, showed a substantial measure of agreement. Averaging the results of these trials, it is found that the standard diet supplying 7% of white-fish meal contained 14·2% of crude protein on the air-dry basis (16·4% on the basis of dry matter), and of this, 32%, or roughly one-third, came from the white-fish meal. In the case of the diet supplying 20% of ex. dec. ground-nut meal, the crude protein content on the air-dry basis rose to 18·5% (21·2% on the basis of dry matter), and as much as 54%, or rather more than half, of the total protein was derived from the vegetable-protein supplement. The two diets contained, respectively, 11·6 and 15·6% of digestible crude protein, and of ‘total digestible nutrients’, 62·4 and 62·7% respectively (air-dry basis). On the two diets as so made up, equality in respect of rate of growth and nitrogen retention in young pigs was made possible, only, however, at the expense of a less efficient and more wasteful utilization, as evidenced by the greater degree of de-amination and a correspondingly higher loss of nitrogen in the urine, of the protein in the ration containing ex. dec. ground-nut meal.

The feeding of as much as 20% of ex. dec. groundnut meal in the rations of young bacon pigs for the purpose of ensuring the maximum rate of growth compatible with the available supply of net energy may seem an extravagant usage of the protein concentrate, but this is really not so if it is kept in mind that, in accordance with the writers’ earlier findings, the protein supplement may be omitted altogether at about 90 lb. live weight without occasioning any depression of the rate of live-weight gain or the efficiency of food conversion. Thus, in the growth trial in which the diet containing 20% of ground-nut meal was compared against the standard diet, no more than 35·5 lb. of the vegetable-protein concentrate was required, on an average per pig, between the live weights of 36 and 90 lb., and following the attainment of 90 lb., the use of the ground-nut meal could, if desired, have been discontinued.

Perhaps the best way of economizing in the use of ground-nut meal, however, is to feed it in conjunction with a small allowance of an animal-protein food, such as white-fish meal. In growth trial no. 4, for example, the diet containing 6% of ex. dec. groundnut meal and 2% of white-fish meal gave as good results as the diet containing 15% of ex. dec. ground-nut meal. Thus, the introduction of as little as 2% of white-fish meal into the diet enabled the percentage of ground-nut meal to be reduced from 15 to 6%, with a consequential reduction of the total digestible crude protein from 13·9 to 11·2%, without affecting the rate of live weight gain or the efficiency of food conversion.

1. A method of extraction and isolation of the alkaloid of blue lupine was adopted for small samples of the various plant parts: seeds, leaves and pods.

2. From the evidence presented, it is shown that the seeds of blue lupine contain a higher percentage of crude alkaloid, than do the pods and leaves. On the basis of the extractions and isolation, seeds contained as much as 2·25%, leaves were found to contain 0·69%, and pods contained 1·46%.

3. The alkaloid isolated was identified as D-lupanine by its characteristic physical and chemical properties. The anhydrous monohydrochloride salt was prepared in pure form and from it were obtained samples of the pure free base, d-lupanine.

4. Toxicity determinations, involving the injection of water solutions of the crude free base, hydrochloride salt and the pure free base, administered intraperitoneally in guinea-pigs, rats and mice were carried out. Mice were found to be most susceptible, rats less susceptible, and guineapigs least susceptible to the effect of the various solutions injected. The minimal lethal dose (m.l.d.) is given for all three species of animals, and in addition the 50% lethal doses l.d. 50 are given for mice. The purest samples were the most toxic.

5. The symptoms shown by guinea-pigs, rats and mice poisoned by the alkaloid are similar. Gross pathology of the animals killed indicated death by asphyxiation.

A study has been made of the bias in error which occurs when Latin-square change-over trials conducted on dairy cattle are analysed by the usual method for Latin-square experiments, with modification for carry-over effects. Bias is present for adjusted direct effects and for permanent effects (direct plus carry-over), but does not exist for unadjusted direct effects.

Analyses of fifteen sets of experimental data showed that the bias is of no importance in 3 × 3 designs but might be serious for some practical situations in the 4 × 4 designs. A tentative factor to correct for bias was given for the latter case.

1. The methods of investigation of the effect of ground water-level on crop growth, together with tho field installations in use, are discussed.

2. Direct field experiments are handicapped by the difficulties of achieving close control on a sufficiently large scale, due to considerable variations of surface level and depth of peat within individual fields and to rapid fluctuations in rainfall and evaporation. Many recorded experiments are associated with climatic conditions of substantial precipitation during the growing season.

3. Seasonal fluctuations of ground water-level in Fen peat soils in England, in natural and agricultural conditions, are described.

4. The local soil conditions are outlined and the implications of profile variations are discussed.

5. The effective control of ground water-level on a field scale requires deep and commodious ditches and frequent large underdrains to ensure the movement of water underground with sufficient freedom to give rapid compensatory adjustment for marked disturbances of ground water-level following the incidence of heavy rain or excessive evaporation.

6. A working installation for a field experiment in ordinary farming conditions is described and the measure of control attained is indicated.

1. Two groups, each of six Ayrshire dairy cows, have been watched in the field for a total of twentytwo 24 hr. periods during the months May to September. No supplementary foods were fed at any time. One group grazed under a system of close-folding and the other in small rotational paddocks.

2. The effect of the different systems of management was small. It was clear, however, that closefolding increased the time spent in grazing and decreased the weight of herbage eaten. The milk yields from both groups were very similar.

3. A very constant pattern of behaviour over the 24 hr. period, similar for both groups, was observed. The most prolonged grazing period always occurred immediately after the cows returned from evening milking.

4. The amount of grazing during the hours of darkness was small at all times of the season but tended to increase as the evenings shortened and the nights lengthened. The cessation of evening grazing was closely linked with the fall of darkness at all times of the year.

5. The activities during the ‘day’ period were less regular but, nevertheless, showed a marked division into three main spells of grazing, of which one occurred immediately after morning milking, separated by two main spells of rest.

6. Stage of lactation, individual milk yield and live weight appeared to have no direct relation to the amount of time spent in grazing. Cows watched in both 1949 and 1950 grazed for very similar times despite differences in their stage of lactation.

7. Increasing defoliation of the paddock resulted in an increased grazing time by the rotationally grazed cows.

8. Close-folded cows grazed for a shorter time on herbage of low moisture content than on herbage of high moisture content, whilst eating virtually the same weight of dry matter.

9. Climatic changes to the extent met with in south-west Scotland had only a limited effect on the general pattern of grazing.

10. The physiological activities of dunging, urinating and drinking showed only small day to day variations.

11. The distance walked during grazing in the 24 hr. by one individual cow from each group showed only a slightly longer track for the rotationally grazed animal, the average distance being about 2000 yards.

Ten experiments on sugar beet in 1949 tested a phosphate-potash fertilizer applied in different ways. There was no damage to germination or plant establishment by fertilizer placed in bands 2 in. to the side and 2 in. below the level of the seed.

There were no significant differences between the yields of sugar given by placed and broadcast fertilizer. Similar yields were given by broadcast applications applied early and worked deeply into the seed-bed and by dressings on the seed-bed which were worked in shallowly. Split dressings where half the fertilizer was broadcast and half was drilled beside the seed were not markedly superior to placing or broadcasting all the fertilizer. In most of the experiments placing gave more vigorous growth than broadcasting the fertilizer during late spring and early summer; at harvest this superiority had vanished.

The results of these experiments confirm those carried out in 1947 and 1948. There is no advantage from placing the full dressing of fertilizer beside the seed in districts where the crop is normally grown, except that labour is saved in applying the fertilizer.

The digestibility of carotene in dried grass was measured by the Cr2O3 method in both cows and goats. In the goats the direct method was simultaneously used for comparison with the Cr2O3 method.

The main conclusions were:

1. When mixed with the concentrate portion of the ration of lactating goats, Cr2O3 was recovered in the faeces in four experiments in amounts equal to 100·0, 99·1, 99·1 and 96·0% of the amount ingested.

2. The apparent digestibility of carotene in the goat was measured both directly by the use of metabolism crates and indirectly by using Cr2O3 as an indicator of equivalence of faeces and food. By the former method digestibilities of carotene in four goats were 68·7, 62·7, 61·8 and 59·6%. The corresponding digestibilities obtained from the same faeces samples by the Cr2O3 method were 67·4, 62·9, 61·8 and 58·9%.

3. In six cows, on the diet similar to that of the goats, the Cr2O3 method showed that 59·3, 54·0, 54·4, 54·4, 57·1 and 55·5% of carotene was apparently digested.

4. Reasons for preferring the use of added Cr2O3 to natural constituents of the food as a marker are discussed.

5. The individual results from cows and goats have been statistically analysed to partition the various sources of variance, and it has been shown that fewer animals for a longer time give a more reliable result than more animals for a shorter time.

1. Eight Cheviot ewes, bearing twins, housed in individual pens, and given the choice of a protein concentrate, a carbohydrate concentrate, hay, minerals and water, failed to select a ration that allowed them to bear and rear strong healthy lambs.

2. Three ewes maintained under these conditions developed a condition resembling pregnancy toxaemia.

3. The amount of food eaten by each ewe decreased during the last few weeks of pregnancy, This is probably the reason for the disappointing performance of the ewes at lambing.

4. The possible influence on the behaviour of the ewes of the constant presence of food is discussed.