CHAPTER XXIV. 181 



which this talies place depends upon the texture of the soil and the 

 viscocity of the fluid. 



In any case the drops that first appear are not part of the water 

 just added. This Is one of the many difficulties in the way of accu- 

 rately measuring the speed of filtration. The quantity that passes in 

 a given time can, of course, he easily measured, ubt the speed iu 

 inches per minute, or in any other unit of distance and time, is 

 quite another matter. That is the difficult point. If we knew the 

 area of the passage ways between the grains we could divide the 

 quantity discharged by this area, combine with the time of dis- 

 charge and arrive at the speed. Unfortunately we cannot measure 

 these passageways — we can measure the cubical voids but not the area 

 of the passageways between the grains. 



An instructive experiment, one easily made, is to fill a glass tube 

 or glass sided box with small marbles and pour oil through them. 

 The use of the slow-moving oil in place of water gives time to note 

 the changes in the shapes of the films and in the filling of the inter- 

 stitial spaces. Coloring the subsequent additions of oil will aid in 

 tracing the movements of the fluid. 



The viscosity, or internal friction of water, has been previously 

 alluded to. Experiment shows that it increases with decrease of 

 temperature. If the viscosity at 32 deg. Fahrenheit be assumed as 

 100, it will be 50 at 77 deg., 45 at 86 deg., and 31 at 112 deg. This vari- 

 ation of viscosity due to temperature has a marked effect upon the rate 

 of percolation. This effect is very important. The following quo- 

 tation from experiments of the United States Department of Agricul- 

 ture will serve to illustrate and emphasize the point: 



Flow at 48 deg. Fah., 6.15 grams per minute. Flow at 90 deg. 

 Fah., 10.54 grams per minute. The ratio of the two rates of flow being 

 as 100 to 171. Now the viscosity of water at 48 deg. Fah. is to the 

 viscosity at 90 deg. Fah. as 75.6 is to 32.5, or as 177 to 100. The flow 

 should be in the inverse ratio, and this agrees very closely with the 

 ratios of the actually measured flows. 



The viscosity of gases, contrary to that of fluids, increases with 

 increase of temperature. Air is frequently used in making "perme- 

 ability" tests of soils, and unless corrections are properly made for this 

 different characteristic very erroneous and misleading results will 

 ensue. 



Speaking of air as a means of testing permeability of soils re- 

 calls an instance in connection with coal mining. Water broke into 



