166 JV. H. Barton — Geoihermal Data from 



which is at the rate of 1° for each 38 feet, somewhat less than 

 in the deeper well recorded in the table. I prefer, however, 

 to take the figures of Professor Shepard for this deeper well, 

 for here the conditions appear to be more definitely deter- 

 mined. 



It will of course be obvious that where underground water 

 is circulating rapidly in horizontal beds, it would retain the 

 same temperature over quite a wide area under high land and 

 low land alike, consequently, if a well were sunk on high land 

 the rate of in crease in feet to every degree would be a very 

 much larger figure than in the adjoining low lands, where the 

 water would be reached at much less depth. I have weighed 

 this suggestion in the region in question, and excepting, possi- 

 bly, about Tyndall, Yankton and Springfield, believe that its 

 influence is practically nothing. It must be borne in mind in 

 this connection that underground waters move at an extremely 

 low rate of speed, which is probably not over a mile or two a 

 year in the sands of the Dakota formation, and under these 

 conditions the water of a flowing well represents very nearly 

 the local underground temperature at the bottom of the casing 

 of the well. The altitudes of the wells have been found to 

 have practically nothing to do with the differences of tempera- 

 tures, excepting probably where wells are quite near together. 

 It is not fair to assume that the isogeothermal^>Z<2w<?<s, even for 

 moderate depths, are closely parallel to minor undulations of 

 the surface of the ground. Consequently in a deep well on a 

 high hill the rate of temperature-increase would be lower than 

 in a correspondingly less deep well at the foot of the hill. 

 The ratios in a case of this sort have not as yet been ascer- 

 tained and they probably vary locally. 



The difference of rate in the increase of temperature in the 

 wells at Chamberlin is thought to be due to the relative posi- 

 tions of these wells. The city well with the smaller rate is 

 some distance east of the mill well, which may account for the 

 difference, as it is in line with the diminution eastward to 1° 

 in 41 feet at Kimball. In the wells at Yankton the difference 

 of actual temperatures of the waters is not great, but as the 

 depths of the wells vary considerably the ratio of temperature- 

 increase varies from 3 k 3 to 38 feet for each degree. It is prob- 

 able that in this case the water may all come from one stratum, 

 and if the minimum depth of this be taken, the rate of 1° for 

 each 32 feet is obtained. In the cement works 4 miles west of 

 Yankton an increased rate is indicated, The areal relations of 

 the data are shown in fig. 1. As the observations are distrib- 

 uted at short intervals over a wide district, and the geothermal 

 ratios vary gradually and uniformly, it has been possible to 

 introduce isogeothermal lines in this figure. Without a knowl- 



