PHYSICAL GEOGRAPHY IN THE UNIVERSITY. 8 1 



ful an analysis as is given to the forms of the land ; but they 

 cannot be traced here for lack of space. 



Illustrations of the deductive scheme. — However much the advance 

 of a deductive scheme of study may be aided by reference to con- 

 crete illustrations during its progress, its statement should be 

 abstract, in order to emphasize the essentially deductive side of the 

 study. It is difficult to follow such a method without artificial aids. 

 Hence, in discussing the theory of the tides, a model of certain 

 theoretical tidal circles was introduced for the convenience of 

 definition and argument. It was found to be an effective aid in 

 reaching certain geometrical consequences that follow from the 

 rotation of the earth on an axis that is not coincident with the 

 axis of the tidal circles. This model was an illustration of the 

 same order as the diagrams employed in text-books on geometry. 

 In the same way, a series of some thirty rough paper reliefs, con- 

 structed several years ago to illustrate a course of lectures to 

 teachers under the auspices of the Boston Society of Natural 

 History, are introduced to aid in giving clearness to the concep- 

 tion of the geographical scheme. They are roughly made ; hardly 

 better than blackboard diagrams, except in having three dimen- 

 sions ; yet they certainly serve a good purpose as aids in follow- 

 ing deductive statements. Being two or three feet in length and 

 yet light enough to handle easily, they are frequently brought 

 into the lecture room, although they are used chiefly in the labo- 

 ratory, where they can be examined and described deliberately. 

 Nearly all the points thus far mentioned are illustrated in one way 

 or another by these models ; but I can here give account of only 

 a few of them. 



While occupied with the first considerations of the cycle and 

 its systematic variations of relief, both in intensity and variety, 

 use is made of three simple models, which are found to be of 

 particular value in fixing the fundamental ideas. The first shows 

 a broad upland, traversed by a main river with a few branching 

 streams, all in valleys of the canyon type. The form of the 

 second is well diversified, there being about as much of lowland 

 in its wide open valleys as there is of upland on its well separated 



