E.— GEOGRAPHY. 103 



map-distributions are momentary cross-sections, needing to be recombined, 

 like the microtome-layers of the anatomist or the successive snapshots of 

 a film, if their significance is to be recovered as phases of an event. 

 Thus we speak of the historical duration of a glacier as an obstacle to 

 traffic over a mountain pass ; and of the geographical distribution of 

 Greek city states or Parliamentary institutions. 



It was indeed this coalescence of geographical and historical outlook 

 and method, late in the eighteenth century, which made possible to 

 von Humboldt and Ritter our modern geography, the study of the distribu- 

 tion and interrelation of terrestrial processes ; and reacted, through 

 Lyell, Darwin, Lubbock and Pitt-Rivers — to give only British names — 

 on the humanities, by supplying a method of geographical analysis for 

 what are popularly called historical situations. 



No one, I hope, will have been led by any part of this argument to 

 suppose any intention to ignore those other aspects of science — of intelli- 

 gence exercised on the intelligible around us — which are concerned neither 

 with relations in space nor with relations in time, but ultimately and 

 sometimes quite obviously with quantities and qualities ; all those 

 observations which go to make up the Physical Sciences ; and all con- 

 clusions and results of the kind which Aristotle was illustrating when he 

 said that ' fire burns here as in Persia '—and he might well have added 

 that ' fire burns now as it burned Persepolis or Troy.' In respect to all 

 those expressions of hoiv things happen, or Jiow they are composed, the 

 historical and distributional sciences stand in the relation of applied 

 sciences to the ' pure sciences ' of physics, chemistry and physiology : 

 accepting and employing their conceptions and interpretations, like their 

 vocabulary and notation, as a gunner employs range-finder and explosive 

 to solve his regional problem of making this projectile here hit that target 

 over there. This intellectual outlook is quite consistent with the possi- 

 bility that any occasion of gunnery may suggest fresh problems to the 

 physicist or the chemist, or offer them significant data ; and may even 

 do so by reason of local and temporal conditions. It was a sound instinct, 

 as well as wholesome criticism of somebody's educational technique, that 

 made the schoolboy bring into class a lump of wayside chalk and beg that 

 by the method demonstrated yesterday carbon dioxide might now be 

 made out of this. 



Similarly, those aspects of science which are concerned with the 

 estimation and interpretation of values — with relations, that is, as 

 irreducible to quantitative expression as they are to conjunctions of 

 region or period, and wherein the notion even of quality parts company 

 almost at the outset from anything that has significance for a chemist — 

 have nevertheless ultimately this point of contact with geographical and 

 historical science, that all the values with which they are concerned are 

 values-to-man, and consequently are, as phenomena, characteristic of — 

 perhaps even peculiar to — terrestrial life, and to a relatively recent phase 

 of it. Indeed, when we speak of these sciences as the Humanities, we 

 mark their distributional and historical limitations, even while we 

 recognise their high rank among aspects of knowledge and their supreme 

 significance to ourselves. 



Now of these three main groups of studies : the Human Sciences and 



