time, in social life and physical environment. If scientific imagination, such as Professor Tyndall once explained and illustrated, is strictly bound by the laws of hypothesis, the magic of the literary artist which looks so free is as strictly bound within the range of ideas already marked out by the language of his group. Unlike the man of science, the man of literature cannot coin words for a currency of new ideas; for his verse or prose, unlike the discoveries of the man of science, must reach average, not specialised, intelligence. Words must pass from special into general use before they can be used by him; and, just in proportion as special kinds of knowledge (legal, commercial, mechanical, and the like) are developed, the more striking is the difference between the language of literature and that of science the language and ideas of the community contrasted with those of its specialised parts. If we trace the rise of any civilised community out of isolated clans or tribes, we may observe a twofold development closely connected with the language and ideas of literature—expansion of the group outwards, a process attended by expansions of thought and sentiment; and specialisation of activities within, a process upon which depends the rise of a leisure-enjoying literary class, priestly or secular. The latter is the process familiar to economists as division of labour, the former that familiar to antiquaries as the fusion of smaller into larger social groups. While the range of comparison widens from clan to national and even world-wide associations and sympathies, the specialising process separates ideas, words, and forms of writing from the proper domain of literature. Thus, in the Homeric age the speech in the Agora has nothing professional or specialised about it, and is a proper subject of poetry; but in the days of professional Athenian