Operational Research Quarterly Vol. 21 No. 1 Chapter 7 deals with Markov chains and queues. The approach to Markov chains is by means of Z-transforms, not the more usual canonical decomposition of matrices. This approach has certain advantages, but also the obvious disadvantage that most scientists are not aware of Z-transforms; even though there is an appendix devoted to Z-transforms, some difficulty may be found reading this chapter. Chapter 8 deals with decisions and games. This is an introductory treatment, with a rather brief description of the link between linear programming and game theory that could perhaps have been further explored. The final chapter is concerned with graphs and networks and includes a section on special algorithms for transportation-type problems. The reviewer found this an interesting book to read and would recommend it as an elementary introduction to mathematical methods in operational research for those who like to proceed slowly up the mathematical ladder. But there is still a long way to go after this. JOHN REYNOLDS REFERENCE 1

T. L. Saaty (1959) Mathematical Methods of Operations Research. McGraw-Hill, New York.

Quantitative Methods in Marketing.

RoNALD E. FRANK and PAUL E. GREEN. Prentice Hall, New Jersey, 1967. 118 pp. 33s.

This paperback is one of the Foundations of Marketing series, which has been prepared to serve the need for up-to-date teaching materials for marketing courses. This volume will doubtless contribute usefully to the aims of the series. From the point of view of the operational research practitioner the text is never more than elementary. It begins with a clear worked example in Bayesian decision theory which pays particular attention to evaluating alternative investments in market research on a cost vs. value basis. The interrelated roles of researcher and marketer are emphasized and the section is well endowed with bibliographical references, as indeed is the whole book. The potential value of experimental work and the analysis of observational data, including correlation and regression, are clearly described by means of further examples. Due attention is rightly drawn to the pitfalls and practical difficulties that are associated with such work. The book closes with a rather short section on simulation studies. I should have liked sections on games theory and perhaps distribution, though it seems likely that these subjects are covered fully in other volumes in the series. To my eyes, this book could represent good value to the operational research worker who experiences difficulty in convincing his marketing management of the potential of operational research. It is the sort of book that might usefully be left lying around an unenlightened marketing director's office. ALAN KITCHENER

142

Operational Research Society is collaborating with JSTOR to digitize, preserve, and extend access to Operational Research Quarterly (1970-1977). ® www.jstor.org