l899-'00 TRANSACTIONS L,43'i 



will also be a fixed one, and if once known, it will be as easy to 

 detect any sewage contamination in this kind of well as in the; 

 other. 



It is also to be noted that the normal chlorine number for a 

 given area, will vary from month to month, and will be especially 

 affected by unusually heavy rainfall, or by prolonged drought. 

 But experience shows that variations, due to these causes, are 

 insignificant in comparison with those resulting from sewage con- 

 tamination. 



For some years past I have endeavored to put this method 

 to the test of experiment, and for that purpose I have collected 

 personally and by deputy, over 730 samples of well water, chiefly 

 within the drainage area of the Ottawa valley. The difficulties 

 which lie in the way of any single individual's accomplishment of 

 so gigantic a task as this, are almost insurmountable ; and I can 

 only hope to illustrate the subject in a very imperfect way from 

 the data in mj^ possession. The first difficulty is to obtain samples 

 which truly represent the normal ground- water, and the normal 

 deep water supply or supplies. Very few country wells are pro- 

 tected against surface soakage by the method indicated in Fig. 4, 

 and I may say here, that I think the Provincial governments 

 might profitably entrust to certain of their officials whose work 

 takes them to different parts of the country, as in the case of the 

 Road Inspectors, the Board of Health officers, etc. , the additional 

 duty of seeing that new wells are properly protected from surface 

 drainage. It costs very little more to properly protect the well 

 by tamping clay behind the stones when the well is being made, 

 than to finish it in the unscientific way in which we find this im- 

 portant matter usually performed, and I am sure that it is ignor- 

 ance rather than any wish to save a few dollars at the risk of 

 health, which explains the unsatisfactory condition of nearly all 



the wells which I have* visited. 



If, in each topographical area, we could find a few thoroughly 

 protected wells, of known depths, and of whose history a full 

 record had been kept, we should possess the data which we re- 

 quire, and which we cannot now obtain with any such certainty 

 as would g:ive a sure basis for the illustration of the scheme I 

 have suggested. 



It is by so much the more important that new wells should 

 be constructed in such a way '^s to fulfil these conditions. 



