The theme of the meeting was ‘Dairying’. After an introduction and talk on the Shropshire Landscape on the first evening, the first day's programme consisted of three invited papers in the morning, followed by visits to two dairy farms in the afternoon. The second day was spent visiting contrasting types of dairy farm. On the last day of the meeting the morning session consisted of three invited papers. In the afternoon a local farmhouse-cheese dairy was visited, and finally members had a tour of the College dairy herd.

One of Bertrand Russell's contributions to philosophy was to show that there are certain questions, quite simple to state, to which no sensible answer can be given. While writing this paper I began to feel that the question ‘what is the optimum size of the dairy cow ?’ is perhaps among these. Let me first ask what we mean by the question, what assumptions we have to make to answer it, and finally what use is the answer.

A survey of the accumulated experimental evidence suggests that the general effect of genetic differences in body size on input-output relationships in livestock might be largely accounted for by the following two genetic sizescaling rules:

1. treat all age and time variables for the ith genotype as directly proportional to Mi0·27, where Mi is the mature body weight of the ith genotype;

2. at every age standardized as in (1), treat all cumulated inputs and outputs for the zth genotype as directly proportional to Mi.

With our increased knowledge of genetics and our ability to control breeding populations of farm animals we are now in a position to select for optimum size if we can define it. A consideration of size in relation to nutritional efficiency therefore becomes relevant. (I confine myself to nutritional efficiency since this was my remit, but of course there are many non-nutritional aspects of productivity which also may be related to size, e.g. larger animals need less labour or housing per unit of production; smaller animals may yield more suitable carcasses for some markets, and these may indeed be more important economically than nutritional efficiency.) Nutritional efficiency will be considered in relation to energy use, since energy is normally the limiting nutrient, and the concentration of the other nutrients, proteins, minerals and vitamins can be varied by adding supplements to the diet.

The economic significance of size in animal production processes manifests itself mainly through an influence on the biological efficiency of the animal population or the individual animal. Within limits, those factors affecting biological efficiency are also ones which have a dominant effect on financial efficiency. These effects relate almost solely to total financial returns and the cost of feeding livestock per unit of product. However, size of animal may have a major impact on the need for resources other than feed, particularly labour and capital, and these aspects also merit some consideration.

Discussion of this topic requires clear definition of terms. For instance, because size can be such an ambiguous word, it is essential at all times to state the units of measurement of size of the animal. The units may referto height, weight, skeletal size, metabolic body weight or to some component of body composition. Most frequently the subject is discussed in terms of body weight, or some function of it representing metabolic body size, and little cognisance is taken of possible differences in body composition. I consider this to be a serious omission, because it seems highly unlikely that weight of the animal will have the same relationship to productivity irrespective of the composition of the weight.

The request to deliver this paper called for a presentation on the methodology of research and development (R and D) resource allocation in a form suitable for a lay audience. As a pragmatist who believes that tools only have a value if they can be used to improve the way in which the job is done, I would very much prefer to be speaking on practical case studies. Nevertheless, I will attempt to stick to the given remit and describe some of the approaches to R and D resource allocation that are available, and to comment on their practical usefulness based on the experiences of myself and my colleagues at the Programmes Analysis Unit.

THE Management of Research and Development (R and D) is not a precise science and it would be foolish to pretend that it is so. For this reason Management Science is somewhat of a misnomer, but this does not mean that the art of management is not susceptible to analysis and study. On the contrary, it is possible to analyse how good managers work and to identify their strong points. When we do this we are struck by the systematic way in which good managers approach problems. This systematic approach is frequently not appreciated in a conscious way by the managers themselves, but it seems to be the key to their success. Since no manager can operate effectively in isolation, but must communicate with others, it is important for a successful manager to be able to communicate not only his decisions but also the reasoning behind his decisions. A systematic approach to decision making offers a framework for communications, and the larger the organizational unit involved the greater the necessity for information to be written down and passed on in a formal way.

The discussion which followed the papers also contained contributions which were related to a paper entitled ‘Focus on Animal Production Research and Development’ and given earlier by Mr W. F. Raymond (Deputy Chief Scientist, Ministry of Agriculture, Fisheries and Food); these contributions are included in this report.