Certain noises, many aspects of turbulence, and almost all aspects of finance exhibit a level of temporal and spatial variability whose "wildness" impressed itself vividly upon the author, Benoit Mandelbrot, in the early 1960's. He soon realized that those phenomena cannot be described by simply adapting the statistical techniques of earlier physics, or even extending those techniques slightly. It appeared that the study of finance and turbulence could not move forward without the recognition that those phenomena represented a new second stage of indeterminism. Altogether new mathematical tools were needed. The papers in this Selecta volume reflect that realization and the work that Dr. Mandelbrot did toward the development of those new tools.

This book presents the papers arising from the ICMI study seminar on the popularization of mathematics held at the University of Leeds, England, 17-22 September 1989. The event was organized in conjunction with a highly successful touring exhibition known as the 'Pop Maths Roadshow'. Inspired by the discussion document prepared by Howson, Kahane and Pollak, the symposium consisted of three plenary sessions discussing the problems faced in the popularization through particular media. Members were present from a variety of backgrounds and discussion groups were devoted to specific themes, such as the image of mathematicians, TV and films, and mathematics in different cultures. This volume will be valuable reading for all teachers and mathematics educators and general readers who are interested in the popularization of mathematics.

Based on the International Commission on Mathematical Instruction conference held in early 1987, this volume consists of a number of key papers presented by international authorities on the role of mathematics in applied subjects, such as engineering, computer science, and mathematical modelling. The importance of certain mathematical ideas, such as geometry and discrete mathematics is stressed, as well as the more classical methods. The book includes a long article by the editor synthesising some of the main themes and trends debated at the conference.

First published in 1986, the first ICMI study is concerned with the influence of computers and computer science on mathematics and its teaching in the last years of school and at tertiary level. In particular, it explores the way the computer has influenced mathematics itself and the way in which mathematicians work, likely influences on the curriculum of high-school and undergraduate students, and the way in which the computer can be used to improve mathematics teaching and learning. The book comprises a report of the meeting held in Strasbourg in March 1985, plus several papers contributed to that meeting.

The three major aims of the present study were to (a) test some of the major game-theoretic solutions for n-person games in characteristic function form with data obtained from "rational" players; (b) locate, assess, and explain differences between sophisticated and naive subjects in coalition frequencies and payoff disbursements; and (c) provide a common data base for bargaining process analyses and testing of both present and future models. To this end, five quartets of subjects each partici pated in 16 different coalition formation tasks presented as computer-controlled, four-person, characteristic function games with sidepayments. All 20 subjects (a) were relatively mathematically sophisticated, (b) were familiar with the major solution concepts for characteristic function games, (c) could safely assume that their co-players were equally sophisticated and knowledgeable, and (d) possessed a high motivation to maximize reward. The results showed that (a) sophisticated players are less restricted than naive players in coalition choice, (b) their payoffs are less variable and less galitarian, (c) these differences between the two populations of sub jects increase with the complexity of the game, and (d) for III the four-person, mDstly non-superadditive games played in the present experiment, the bargaining set M i) accounts more accurately for payoff disbursements than do the competitive bargaining set H i), the kernel K, and Selten's equal share analysis."