A and in the writings that were labeled "geography." The meaning given to the term "physical geography" had, it is true, been modified in the preceding fifteen years, alrnast entirely through the efforts of the founder of the Association, William Morris Davis. A decade earlier Davis had imposed on all but one of his fellowmembers of the sub-committee on geography of the Committee of Ten on Secondary School Studies his then novel conception of its c0ntent.l Before about 1890 there was no evident disagreement about what physical geography was. The most succinct definition given in the textbooks is Russell Hinman's of 1888: "Physical geography seeks to trace the operation of the laws of nature upon the earth; upon the air, the water, and the land; upon plants, animals, and even upon man."* This definition, except for the reference to man, was one that Humboldt would have subscribed to. Its derivation was clear: it could be traced back through Humboldt, Bergman, and Lulofs to Varenius; and from Varenius to the encyclopedic writers of the Middle Ages, who built on Aristotle.The notion of physical geography embodied in the recommendations made to the Committee of Ten in 1893 differed from Hinman's only in giving primacy to the surface of the lands, in the spirit of Davis's concepts, over the atmosphere and the oceans. This "curious and persistent insistence on the peculiar claims of physiography" was one of the points at which the dissident Houston directed his opposition. But the spirit of the study of the earth that Davis recommended for secondary schools was the same as the one that animated the instruction in physical geography at Harvard in which he had participated for twenty years, the same that Hinman had expressed. That is to say, the earth was approached from the viewpoint of the physical and natural sciences ; it was studied in its own terms and for its own sake. 1 National Education Association, Report of the Committee of Ten on Secondary School Studies (New York, 1894), pp. 204-240. The lone dissenter was Edwin J. Houston.

No abstract

Most of the work done in climatology in the past 150 years can be assigned to one of the following classes of investigation or presentation: empirical formulation of climatic data, descriptive climatology, climatologic cartography, organization of observational data by synoptic categories, investigation of the physical bases of climate, definition of climatic types and the delineation of climatic regions, and reconstruction of past climates. Although their relative importance has varied, through much of the period under consideration all these fields have been cultivated simultaneously. Especially in the latter part of the Nineteenth and the early part of the Twentieth Century, climatology has been too strongly dominated by a descriptive approach. Satisfactory physical interpretations of the phenomena recorded have often been impossible, but even the physical insights available at any given time have not always been applied. It would appear that the rational treatment of climatologic data can be attained more readily through synoptic concepts than through a frontal attack with the general formulations provided by physical theory. “… das arithmetische Mittel, in welchem de allerverschiedensten Zustände zusammen vergraben werden, ist Nichts Wirkliches, sonders eine abstrakte Grösse.” W. Köppen, 1874.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. American Geographical Society is collaborating with JSTOR to digitize, preserve and extend access to Geographical Review. HIS recent article on the Wolf Creek glaciers of the St. Elias Range,Robert P. Sharp describes and illustrates a type of surface of melting glacier ice that he calls "scalloped."' His term was evidently suggested by the profile of a section normal to the surface. A more suitable name for a three-dimensional surface such as he describes might be "cuspate" or "negatively mammillate." According to Sharp, "scalloped" surfaces are associated with fluted ones, and always face downward. "Scallops" appear on a horizontal or nearly horizontal surface and pass into flutings as the inclination of the surface from the horizontal increases. Sharp is unable to suggest a "satisfactory explanation ... for the form and pattern of flutes and scallops." I propose here a hypothesis of the origin of these surface forms of melting ice. As a supplement to the illustrations in Sharp's article, Figure i is included, reproduced from a photograph of part of the roof of a stream tunnel under a firn bank. I took the photograph on August 7, I940, in the canyon of Lowell Creek, which flows into Resurrection Bay at Seward, Alaska. In winter, snow accumulates to great depths in parts of this canyon, as a result of the piling up of avalanches that have slid down its steep tributary valleys. As late in the summer as the date when the photograph was taken, three such accumulations had not yet melted. All the exposed snow had been changed into firn, and the stream flowed through tunnels under the remaining banks. Figure i was taken from the upper end of one of these tunnels, looking downstream. The avalanche snow is rather dirty, so that many of the ridges in the melting surface of the firn are lined with bits of rock and plant material. Most of this foreign matter has been carried to the ridges and cusps by the water that flows as a film along the melting surface toward the cusps and there drips off. In the upper right of the picture, melting has exposed an ice crust containing debris; it has an appearance different from that of the true cuspate or scalloped surface. MELTING ON OVERHANGING SURFACESA part of the heat that melts downward-facing surfaces of ice and firn is probably derived from the absorption of reflected solar radiation and of I R. P. Sharp: The Wolf Creek Glaciers, St. Elias Range, Yukon Territory, Geogr. Rev., Vol. 37, I947, pp. 26-52; reference on pp. 46 and 47 (Figs. 27 and 28).

No abstract