HISTORICAL 705 



township 56, range 33, gave 43.3 degrees, one of the very best observa- 

 tions — that is, not less than 42.6 degrees Fahrenheit surface temperature. 

 These variations are but such as might be expected in the mean soil tem- 

 perature from streaks of warm years and cold years, as we see from the 

 table of temperature below, from the State Weather Service. In addition 

 to these variations, we have inaccuracies in thermometers, and in my 

 case at least the thermometer was a pocket one, too small to read fractions 

 of a degree accurately. There is also allowance to be made for mine 

 ventilation, tending to increase readings below 60 degrees and decrease 

 those above 70 degrees. There is also the fact that in no case was the 

 real mean depth below the topographic surface computed nor the effect 

 of underground circulation allowed for. All these factors must be al- 

 lowed for if closer results are to be obtained, though they hardly will 

 change the results 5 per cent. 



While State Geologist, I made various observations in 1901, 1906, and 

 1910, assembled in my report on the Keweenaw series, and a very im- 

 portant set was made by S. Smillie, engineer of Quincy Mine. All ob- 

 servations were assembled in my report in 1911, and all the more impor- 

 tant were reprinted by N. H. Darton in United States Geological Survey 

 Bulletin 701. C. E. Van Orstrand is continuing that work. 



There is also a method of attacking the problem of mean soil tempera- 

 ture from the Weather Bureau side, thus : Find the mean temperature for 

 a term of years, preferably just preceding the observation. Then assume 

 that for the months when the mean temperature is below freezing, the 

 ground temperature is freezing, for it is known that frost does not go 

 down into the ground more than for a couple of feet. If we also assume 

 that for the first month in spring, when the temperature averages above 

 freezing (April), the ground temperature averages about freezing by the 

 melting snow, we come to 43.7 degrees — a temperature within a degree of 

 the temperatures directly measured. 



SuMMAEY or Data 



The data have been so summarized in my report in 1911, and especially 

 in Barton's valuable Bulletin on the Geothermal data of the United 

 States,* that it seems hardly needful to do more than give a few important 

 figures to save the bother of reference and add a few observations I re- 

 cently made, especially as I understand C. E. Van Orstrand has a yet 

 more elaborate report in hand, which may precede this, and it is to be 

 hoped will include in full the valuable records of the Calumet and Hecla 

 Mine for Alexander Agassiz. 



U. S. Geol. Survey Bulletin 701. 



