242 SCIENTIFIC APPARATUS. 



Society to reduce their area. This great aim has been steadily 

 persevered in for forty-six years, and the interval has seen a very 

 extensive amount of work done. But a still greater extent 

 remains untouched round the north and south Poles, in Asia, 

 Africa, South America, and in the eastern islands. For many 

 years to come the Society must continue its efforts and offer its 

 rewards ob terras redusas. 



The instruction to be derived from the study of a large col- 

 lection of maps, especially of a complete series representing 

 the same region at certain intervals of time, is varied in its 

 character, and extends to several branches of knowledge. To 

 the historian such study is indispensable for the due appreciation 

 and efficient handling of his subject. The historian who has not 

 got the geographical instinct is without one of the most necessary 

 qualifications for his task ; and there is no more useful aid to 

 the historical student than maps showing the territorial divisions 

 .of a region at various periods ; such, for instance, as those given 

 by Mr. Freeman in his History of the Norman Conquest. 



The study of a series of maps of the same region may often 

 throw light on the physical changes that have taken place, and 

 thus convey knowledge respecting the tendencies of changes in 

 the courses of rivers or the condition of harbours, which may be 

 of the utmost value to administrators, and especially to engineers. 

 It is thus that we learn the marvellous changes that have taken 

 place in the valley of the Indus. Multan, as we know through 

 these comparisons, was once on an island in the Ravi. It is 

 now thirty miles from that river. The classic Saraswati once 

 watered a now arid desert, and the history of the change is 

 taught to us by comparative geography. The gradual changes 

 in the deltas of the Ganges and Brahmaputra are of the utmost 

 importance, and their character is revealed to us by a comparison 

 of Major RennelFs first maps of Bengal, and of the still earlier 

 Dutch maps, with the results of the latest surveys. The same 

 causes are still at work, and by learning their tendency in former 



