This paper will in the main deal with the possibilities of the use of progeny testing for the improvement of livestock, in particular under British conditions. Before dealing with individual species, it will perhaps be of value to consider the problem generally and create a framework in which the practical discussion may more fruitfully be carried on.

Let us consider first then an ideal case in which we are dealing with a large population of animals all kept under the same general conditions of management. In such a situation, it is possible to predict, using recent advances in population genetics, the probable genetic improvement that could be achieved by different selection programmes. The basic factor that needs to be known for such a calculation is the heritability of the character for which selection is practised—the accuracy with which the breeding value can be predicted from the performance. From the heritability, the accuracy of a progeny test in measuring the breeding value of a sire can be calculated and by this means the probable rate of advance in a progeny testing programme can be evaluated.

In 1945 the Milk Marketing Board (M.M.B.) of England and Wales undertook responsibility for the development of an A.I. service to cover large areas of the two countries.

By 1951, a total of 24 main Centres had been set up, housing some 600 dairy and beef bulls and supplying semen to 77 Sub-Centres. During the year ending 31st March, 1952, approximately 707,000 first inseminations were carried out, representing 25% of the available cattle population. Further progress continues to be made and it is estimated that this proportion will be raised to 30% for the corresponding year ending in 1953.

From the outset it has been agreed that the rate of cattle improvement through A.I. will be largely dependent upon the use made of the older progeny-recorded bulls available to the scheme. Here it should be noted that the M.M.B. took over responsibility for milk recording under ‘ National Milk Records ‘ in 1943. By adopting a system based upon lactation record cards, it was possible to establish in 1947 a central clearing house for milk records of animals registered with Breed Societies, viz. the Bureau of Records. One of the main services operated by the Bureau is that of progeny recording for sires and summaries of bulls fulfilling certain requirements are now published annually.

I want to correct one or two misconceptions about the way the Danes have sought to improve their native Landrace breed, to try and decide the extent of the improvements which can be fairly attributed to progeny testing, and to describe a few of the features which are characteristic of Danish methods. Owing to the shortage of time I shall refrain from giving any account of the mechanics of testing in Denmark. As this has often been reported you should have sufficient knowledge of the daily procedure at testing stations to provide an adequate background to this paper.

The progeny test is generally regarded as an important aid to selection in livestock improvement. It is by no means a new method, but it seems to be in the limelight today. The usual methods of selection have brought us to a certain standard of performance, and apparently any further imporvement can only result through the application of the progeny test.

It is worth pointing out that cattle, especially dairy cattle, pigs and poultry have received more attention by the scientist than sheep. This may be due to the fact that these animals are more under the direct control of the breeder than sheep. They require daily attention whereas a hill flock of sheep may not be seen for weeks or months by the breeder. Again one has definite objectives in view when breeding dairy cows or hens; one can weigh the milk and count the eggs but the task of the hill sheep worker is considerably more complicated.

I propose taking a pretty wide interpretation of the term progeny testing. When selecting poultry for improved economic characters, most of these characters such as egg numbers, viability and disease resistance in which we are interested have a low heritability and consequently progeny testing is an essential poultry breeding technique. During the last 15 years in England we have mainly been trying to improve the economic characteristics of egg production, but now the Americans are showing us that if we follow closely their broiler industry we can develop a valuable industry for the production of poultry meat in Britain which makes better feed utilisation of scarce concentrated feeds than any other form of meat production. The heritability of meat conformation in poultry is high but I believe progeny testing will come into the picture of selection for better broiler strains, because such strains must be capable of growing fast and evenly with the most efficient feed conversion ratio whilst being housed together in large mobs of manv thousand of chicks.

Since 1948 the Animal Breeding Research Organisation has been collecting twin cattle in the West Midlands of England. The dairy cattle population here is dense—an important factor in reducing cost of collection. Propaganda was issued in various forms all directed at getting in touch with farmers owning one-egg twins, i.e. twins originating from the splitting of a single fertilised egg as distinct from two-egg twins coming from two fertilised eggs. As the twinning rate in cattle is only 1-2% of all births, and probably only about 1 in 20 of these twins are one-egg, their purchase and collection under British conditions is rather costly. Up to the end of 1952, 103 pairs of one-egg heifers twins had been collected.

There has been a rapid expansion of pig research at the Rowett Institute during the last two years, and the object of this paper is to give some idea of our facilities, the kind of work we are doing, and a few of our recent findings.

We are now able to feed individually up to 206 growing pigs at any one time, but only rarely will more than 150 be fed individually whilst about the same number will be fed in groups.

Our permanent fattening house has four farrowing pens which are at the moment being used for baby pig experiments. It has three pens in which 36 pigs are individually fed and four pens somewhat similar to those used by the late Mr. McGuckian in Ireland. These house eight groups of seven pigs, which can, if we wish, be individually fed in a ‘ dining room ’ containing sixteen individual feeding compartments.

During the early stages of investigations concerned with the dietary requirement of dairy calves for certain amino-acids (Blaxter & Wood, 1952a), we had occasion to devise rations simulating milk to meet the calves’ basic requirements. These consisted of reconstituted dried skim-milk powder, glucose, trace elements, lard as a source of fat and a concentrated solution of vitamins A and D in arachis oil. The diet was comparatively cheap and the growth of the calves which were given it appeared quite satisfactory, at least for the first month. Then, quite suddenly, some animals developed alarming symptoms of muscular and cardiac derangement and some died. Examination of the calves at post mortem showed the presence of muscular degeneration which appeared comparable to a similar type of degeneration which had been observed in guinea-pigs and rabbits given rations low in vitamin E content (Goettsch & Pappenheimer, 1931). The ration which we had been giving to the calves contained only a few milligrams of vitamin E, since experiments in America had suggested that E deficiency in cattle took several years to produce (Gullickson & Calverley, 1946).