It has been widely recognized recently that in order to make scientific progress on large and important problems (eg, carbon dioxide effects on climate, viability of various sites for nuclear waste disposal etc.), it is necessary to integrate knowledge from wide ranging sets of disciplines. This is certainly true in the climate sciences, for progress in understanding the cause of the ice ages or the effects of industrial pollution on the future climate or even the likelihood of severe climatic consequences in the aftermath of nuclear war. All require state-of -the -art input from many...