1. Peck order position was assessed by aggressiveness scores for birds housed intensively on deep litter on the floor and in cages.

2. Relationships were shown between scores and egg weight, egg number and egg mass for the birds on the floor but not for the caged birds. The nature of the relationships found was discussed.

3. The correlation between egg weight and egg number was higher in the flock on the floor than in the caged flock. This appeared to be due to a significant positive correlation among the lowest scoring birds on the floor.

Behaviour observations and measurements were carried out on three classes of beef steers kept in yards under drought feeding conditions. The classes were (i) 2-year-old Brahman × Hereford crossbreds, (ii) 2-year-old Herefords and (iii) 3-year-old Herefords.

Animals of each class were placed in two groups and fed on low quality bush hay, one group ad lib. and the other on a restricted basis. An intermingled group comprised two animals from each class and was fed ad lib.

Fairly stable linear social orders were observed in each yard, social success depending mainly upon height. In the intermingled group, the crossbreds dominated the 3-year-old Herefords, which in turn dominated the 2-year-old Herefords.

Crossbreds bunted more actively and were more often disturbed at feeding during the day. They consequently consumed proportionately more feed at night than the Herefords.

Although high social ranking animals had priority at the feed trough there was no relationship between social rank and growth.

The orders in which the 2-year-old Herefords and crossbreds entered a crush were non-random. These orders appeared not to be related to social rank.

1. Nitrogen balance studies have been conducted on eight early-weaned calves receiving four diets containing, respectively, groundnut meal, groundnut+lysine, groundnut+lysine+methionine and white-fish meal as the major protein source.

2. There were no significant differences between groundnut meal diets i n any of the measurements examined. On the fish meal diet all measures of nitrogen retention tended to be higher than the corresponding values for groundnut meals.

3. Daily live-weight gains were significantly higher on the fish meal diet than on any of the groundnut diets (P<0·05).

4. It is suggested that the supplementary amino acids, being highly soluble, were rapidly deaminated by rumen micro-organisms.

While it is generally agreed that fish meal protein is of high biological value for non-ruminants, variations in quality are known to occur as a result of differences in the nature of the parent material (Duckworth, Woodham and McDonald, 1961) or in the method of processing (Grau et al., 1959; Ousterhout and Snyder, 1961). Duckworth, Woodham and McDonald (1961) have shown that the Gross Protein Value offish meals can range from 92 to 119; within this range white-fish meals had a mean value of 102 while the mean value for herring meals was 113. Differences of a similar order have been reported by other workers (Record and Bethke, 1933; Heiman, Carver and Cook, 1939; Robertson, Carver and Cook, 1940; March, Stupich and Biely, 1949; Carpenter, 1954; Carpenter, Ellinger and Shrimpton, 1954; Duckworth, 1955).

It has been estimated (Mason, 1963) that about 70% of beef stores bred in Britain are out of dairy cows. Some 30–40% of them are by beef bulls, so that about 40–50% of animals slaughtered for beef are of pure dairy breeding.

The possibility of using hormones to increase the productivity of meat animals has been considered for a number of years (Hammond, 1957 and Scarisbrick, 1960). Lamming (1957) has pointed out that most of the work with hormones and beef cattle has been done in the U.S.A. under feedlot conditions. In this country Aitken and Crichton (1956), Dodsworth (1957) and Lamming (1957 and 1958) all obtained increases in rates of gain with implantation. Lamming found that responses in steers varied from 0 to 15% increase in gain per day in spring, about 20% in summer, 40% in autumn, i n each case on pasture and up to 60% under winter feeding conditions. He also used several levels of implantation, from which it appeared that the optimum level was about 45 mg. hexoestrol. Smith (1958) described trials conducted at eight farms in the North of Scotland in which 189 steers fattened solely on grass were implanted with hexoestrol. Implantation resulted in an extra daily live-weight gain of about 0·5 lb. Increases in dose levels above 60 mg. produced no significant response in terms of live-weight gain. Eighteen experiments at Experimental Husbandry Farms and the Norfolk Agricultural Station, described by Jones (1961), also showed that live-weight gain was increased by implanting; that the correct dose to give maximum response appeared to be 45 to 60 mg. hexoestrol; and that the implant should be made from 100 to 130 days before slaughter. Second implantations were made in several of these experiments and it appeared doubtful if this practice was desirable. In New Zealand, Lucas (1958) and Everitt (1959) also found that hexoestrol implantation increased weight gains of grazing steers.

The first paper of this series (Alder, Tayler and Rudman, 1964) described experiments carried out over 4 years on the effects of hexoestrol on live-weight gain and herbage intake in grazing steers. The 52 steers used weighed between 750 and 1,100 lb. initially and included four pairs of twins. The experiments lasted between 55 and 124 days and involved both summer and winter grazing. This paper is concerned with observations made on the animals after slaughter.

Donald (1962) has recently pointed out the paucity of available data on production from British sheep, though some information is available for hill sheep (see for example Donald, 1958; Purser and Roberts, 1959; Purser and Young, 1959; Dalton, 1962). Most of these refer mainly to the Scottish Blackface and Welsh Mountain breeds. Hill sheep form approximately 40% of the national flock of over 11 million breeding ewes. Another 40% of ewes are found in crossbred flocks mainly kept on the lowlands; these produce fat lambs and hoggets with wool as a secondary product. Very little information has so far been published on these crossbred ewes, exceptions being work reported by Bywater (1945) and Rennie (1957). This is the first of three papers to be presented with the object of providing such information, and is based mainly on the work of Yalçin (1963). It is hoped that these papers will be of use on two counts:

1. To provide a documentation of one specific cross, the Border Leicester × Cheviot ewe and her Suffolk-cross lambs under North of England grassland conditions.

2. To analyse the usefulness of keeping production records within such flocks as aids to good commercial management, to culling and to selective breeding.

In this paper the traits studied are body weight, fleece weight and litter size of the ewes, and the weights of the lambs between birth and slaughter or weaning. Estimates were obtained for the effects upon some of these traits of age and live-weight of ewe, and of sex, year and type of birth and rearing of the lambs. The usefulness of such correction factors is considered within recording schemes. In subsequent papers the effects on production of differences between crossbred ewes and between the rams used as fat lamb sires will be considered.

There are many flocks in Britain where first-cross ewes are mated with a Down ram in order to produce offspring of both sexes to be fattened for slaughter between 3 and 12 months of age. The ewes are normally bred out of hill ewes by rams of the special crossing breeds, the Border Leicester being the most popular example. Commonly the crossbred flock is maintained by the annual buying in of young sheep at 6 or 18 months old. These replace animals lost through normal wastage and those sold after a set number of lamb crops. In such a situation the flock owner relies on others to produce his crossbred ewes and can himself do nothing to improve their performance through selective breeding. He can only attempt to cull ewes with a poor performance. Any gains from such culling are, of course, temporary and must be maintained by a regular system of culling within each separate intake of females. The success of a culling scheme depends partly upon the extent to which future production is predictable from past performances. This is normally assessed by the correlation between an individual's successive records: the intra-class correlation or repeatability (t). The work reported here estimates the repeatability of various productive traits within two crossbred ewe flocks and considers the scope for culling within such flocks.

1. Total concentrations of volatile fatty acids and molar proportions of acetic, nrooionic and butyric acids were determined in the rumen of Blackhead Persian sheep given sixteen low-protein diets.

2. Total concentrations of acids ranged from 2·89 to 9·33 millimoles/ 100 ml. of rumen liquor. Both digestibility and protein content of the ration had a significant effect on the concentration of volatile fatty acids in the rumen liquor. Protein content of the diet probably influenced the latter through controlling the rate of ruminal fermentation.

3. Molar proportions of acetic acid varied from 59·7% to 69·3%, of propionic acid from 21·2% to 29·6%, and of butyric plus higher acids from 7·6% to 16·3%. Protein content of the diet had no significant effect on the composition of the acid mixture, but organic matter digestibility was inversely related to the proportion of acetic acid.

4. The fermentation in the rumen of Blackhead Persian sheep, as judged by the volatile fatty acids produced, appears to be similar to that found in European breeds of sheep.

1. Carcass dissection data and specific gravity determinations from 76 pigs slaughtered at either 120 lb., 200 lb., or 260 lb. live-weight have been analysed to study the value of specific gravity and its reciprocal in predicting fat and lean in the carcass.

2. Possible effects that the bone content of the carcass might have on the accuracy of prediction have been investigated.

3. Specific gravity was highly correlated with percentage fat and lean in the half carcass. Correlation coefficients for the reciprocal were of similar magnitude but opposite in sign.

4. The residual standard errors in percentage lean and fat at constant specific gravity were 1·74% and 1·72% respectively.

5. The replacement of specific gravity by its reciprocal did not increase the accuracy of prediction for either lean or fat tissue expressed as a percentage of the trimmed half carcass.

6. The muscle/fat ratio exerted a markedly greater influence on specific gravity than did the muscle/bone ratio. However, the inclusion of the latter ratio in equations relating specific gravity and percentage lean in the carcass, reduced the residual variation.

7. Application of the pooled regression equations to independent data, obtained by the use of identical techniques, has given promising results.

1. The transferrins in two cattle herds were studied by means of starch gel electrophoresis. Three phenotypes were observed in a Hereford herd and six in a Hybrid herd comprised of Galloway, Aberdeen Angus and Charolais breeds.

2. Gene frequencies were calculated for each herd. Great variation was found in the frequency of the transferrin alleles between breed and source within the Hybrid herd and between the Hybrid and Hereford herds. The frequencies of the transferrin alleles found in the cows in the Hereford herd were TfA = 0·467 and TfD = 0·533; those found in the cows in the Hybrid herd were TfA = 0·382, TfD = 0·506 and TfE = 0·112.

3. Progeny data were in general agreement with the three-allele theory of inheritance, although the data suggested that offspring in the Hybrid herd carrying the TfE allele occurred at a lower frequency than expected.

4. The possible relation of transferrin type to embryonic mortality was studied. The results were in general agreement with those of other workers, indicating a possible interaction between the genotypes of the foetus and dam.

5. The characteristics of the transferrin types and the postulated three-allele genetic mechanism for the control of the observed variation are discussed.

In determining the milk yield of the sow by weighing the litter before and after suckling, loss of weight due to metabolic processes must be considered.

An experiment is described to investigate the loss of weight of piglets caused by metabolic processes between two consecutive weighings, and to investigate the relationship between this loss on the one hand and the body weight and the age of the piglets on the other. Weight changes of 15 litters of average 9·8 piglets were studied during 66 weighing periods of 16 minutes, by putting the pigs in a cage, fixed on a differential balance. Each weighing period consisted of 8 periods of 2 minutes, so that in all, loss of weight was studied during 528 2-minute periods.

In poultry and dairy cattle heterosis is occasionally found in crosses of different strains of the same breed (Hutt and Cole, 1952; Johansson, 1939; Nordskog and Ghostley, 1954). Limited data on the milk production of reciprocal crosses between Dutch and American strains of Friesian dairy cattle being available in Israel, an investigation was carried out to determine whether crossing these two strains resulted in a heterotic effect on milk production.