104 



Table II. 



It will be seen that the amount tlowing in at the first division of A B 

 is al out tAYo-thircIs the total amount floAving- into the entire area, and that 

 this snpplits the f.owage for the first nine divisions of A D while the 

 tenth division of A D gives out the water from j\ the distance A B. If 

 the rock in the area be soluble it is easily seen that the water flowing 

 from this lowest division of A D will be very highly charged with mineral 

 matter, while the remaining two-thirds that flows out above will be very 

 slightly charged. This is more especially evident when the long sweeping 

 paths of the water are considered compared with the very short paths of 

 the waters of the first division of A B. We have this represented graph- 

 ically in Fig. 5, where the lines of flow are drawn for the case where 

 A B = 100 and A D = 10, or a typical artesian area. If A D be a crevice 

 in the rock it is evident that this place will be favorable for the deposition 

 of the mineral salt dissolved in the Avater since the pressure is released at 

 this point and there is apt to exist some reagent that aaMII cause a precip- 

 itate of the ore. This reagent may exist in the crevice itself or in the 

 opposite wall. 



In Fig 6 the curve has been plotted for the flowage from A D for 

 the case A D = 10 and A B = 100. This does not differ much from the 

 case where A D = 10 and A B = 10, except that the convexity dowuAvard 

 is somewhat more pronounced, making the curve less like the tractrix. 



Ten equal divisions were taken along A D and the A'alues of y taken 

 from Table II corresponding to different A'alues of f. 



The absorption curve for the case A D = 10 and A B = 100 is given in 

 Fig. 7. Here the scale has been somewhat changed due to the large 

 A-alue of A B. The distance A B was divided into one hundred equal 

 divisions, while the same vertical scale was used for y as in the preceding 

 cases. The values of y were taken from Taljle II, being the different 

 Aalues of a in that table. 



The rapid fall of the curve at first and then more gradual fall corre- 

 sponds to the values of a found in Table II and also emphasizes the 

 relatlA'e slowness of the motion of the Avater in the right-hand half of the 

 area A B C D, Fig. 5, as compared Avith that of the left-htuid half. 



The method used in the preceding cases might be extended to areas of 

 different dimensions, but the results Avoiild not differ much from those 

 already stated. 



