438 COEEESPONDEJfCE, 



METHOD OF OBTAINING THE AMOUNT OF WATER IN 



EIVEES. 



Information in regard to the amount of water wliich passes through 

 rivers at different times is an important element in the improvement 

 of internal navigation in the application of hydraulic power, and in the 

 solution of various questions relative to meteorology. The following 

 letter from Captain A. A. Humphreys, of the Topographical Engi- 

 neers, who has had much experience in gaging rivers, will be of in- 

 terest to all who are disposed to co-operate in investigations of this 

 character : 



Dear Sir : In reply to your request to give suggestions as to the 

 hest method of ascertaining the velocity and depth of rivers adapted 

 to amateur observers, I must first observe that any such plan must be 

 rough, as accurate measurements are very expensive. 



The most feasible plan for ascertaining the cross section of the river 

 ■will be to employ a civil engineer to fix the position of a boat by tri- 

 angulation, while soundings are made from it directly across the river 

 from bank to bank, -v^ith a common chain and lead. A time of low 

 water and gentle current should be selected in order to insure accu- 

 racy. The contour of the bed between the water surfaceand high water 

 level must be ascertained by a spirit level. The area of cross section on 

 any day corresponding to any stage can then be readily computed from 

 the reading of a gage-rod when reference to high water is known. 



Accurate velocity observations are too delicate and expensive for 

 amateur observers, but fair results may be obtained by the following 

 plan: A base line, say 200 feet in length, is measured parallel to the 

 current. Two range posts are placed at each end. Their lines of 

 direction being perpendicular to the base, and of course parallel to 

 each other, and perpendicular to the current, an observer is placed at 

 each range and a man with a skiff" directed to throw over a block of 

 wood in the swiftest part of the current^ entirely above the upper range. 

 As this floats past the upper range, the observer shouts to his assistant 

 at the lower, who notes the exact time of its passing the lower range, 

 which he can see for himself. The difference of these times gives the 

 means of computing the velocity, which is equal to 200 divided by it. 

 This velocity is the maximum velocity of the river. The mean velocity 

 is about 0.8 of it. The discharge is, therefore, equal to the area of 

 cross section corresponding to the reading of the gage-rod, multi- 

 plied by the observed velocity, multiplied by 0.8. 



This method, although rough, will give fair results, if properly 

 carried out. JSfo instruments are required, except a laatch and the 

 transit and level used by the civil engineers to assist at the soundings. 

 Your obedient servant, 



A. A. HUMPHREYS, 

 Captain Ibpographical Engineers. 



Prof. Jos. Henry, 



Smithsonian Institution, Washington, D. 0. 



