GENERAL CONSIDERATION. 23 



in Burma and in Baluchistan which still further increase the vast- 

 ness of the task. Consequently, the parts of the map that have 

 been submitted to detailed examination are very few, as the time 

 necessary for work of that kind can seldom be spared except for 

 some special object like the detailed survey of a small area occupied 

 by a coal-field. Under the circumstances, the minute examination 

 necessary to decide these questions is quite out of the question. 

 The delineation on the map of the areas of metamorphic rocks at 

 once excludes certain portions from the range of research since 

 artesian water can scarcely be expected from them. But if a geolo- 

 gical map of India be examined, it will be noticed that an enormous 

 area of the Peninsula is occupied by a volcanic formation known 

 as the " Deccan trap." Practically the whole of that area is un- 

 surveyed : we know next to nothing of that immense formation, but 

 the mere boundary of the tract which it occupies. As to its capacity 

 as a water-bearing formation, little is known and great differences 

 of opinion exist. The map also shows a large area occupied by the 

 " Vindhyan " rocks. These have received more attention than the 

 "Deccan trap," but still in a very general manner, and very little 

 is known regarding their behaviour as water-bearing strata. Lastly, 

 we have scarcely any information respecting the underground 

 structure of that immense spread of recent deposits constituting 

 the Indo-Gangetic alluvial plain, whose strata are still in the posi- 

 tion in which they were deposited, and no exact knowledge of which 

 can be arrived at except from actual borings. In the opening 

 paragraph of these notes, I mentioned that a review of the whole 

 question as it then stood in India had been written by Mr. Medlicott 

 in 1881. It cannot be said that the knowledge gained in the two 

 decades that have since elapsed can allow much to be added to 

 what he so ably discussed. Artesian problems are always so full 

 of uncertainty that the results of actual experiment play an import- 

 ant part in arriving at any definite conclusions respecting subsequent 

 trials. The experiments carried out since the publication of 



( 23 ) 



