Many cities owe their origin and their fame to some event, often some religious association which draws great numbers of pilgrims and others. It is often doubtful, however, to what extent the supposed cause is the real cause of the city's importance. Not every saint founds a city, not every holy city keeps its fame, and in the struggle for existence those cities will persist whose natural advantages are greatest. Another very important cause of cities is a junction of routes, for this means that many different types of merchandise will pass this way, and will give abundant raw material for many minor industries. London is a good example of a town upon which many routes converge, these being both land routes and water routes. Even these few examples may serve to suggest the point of view from which the modern geographer regards towns, and to illustrate the fact that in this as in other branches of his subject his interest is in the study of causes and of interrelations. NOTES ON BOOKS Most of the subjects which have been treated in this book fall under the heading of physical geography, and therefore we may begin these notes by recommending two large works of reference on this subject. Salisbury's _Physiography_ (New York: Henry Holt & Co., 1907, $3.50 net) is the most comprehensive work on the subject in English. In French there is an admirable book by E. de Martonne, _Traité de Géographie Physique_ (Paris: Armand Colin, 1909, price 22 fr.), to be especially recommended for its beautiful plates and diagrams, and for its copious references.