EXPERIMENT STATION BULLETINS. 731 



34°F and by April 15 when the 12-inch depth began to thaw it had risen 

 to 36.43 °F. 



These instances therefore, prove conclusively that an appreciable 

 amount of heat traverses to the surface from the lower depths and tends 

 to maintain a higher temperature there than would otherwise exist. It 

 is perhaps partly on this account that after the soils freeze at the upper 

 depths their temperature tends to remain at about the freezing point. 

 At least a portion of the heat they lose at these upper depths is re- 

 plenished by the heat coming from the lower depths. In the spring, how- 

 ever, when the air temperature begins to rise and the loss of heat from 

 the soil is thereby diminished if not entirely stopped, the heat coming 

 from the interior of the earth accumulates near the upper depths and 

 causes a rise in temperature, just as the experimental data show. The 

 amount, however, coming to the surface during a single day is un- 

 doubtedly small, partly because of the low heat conductivity of the soil, 

 and partly on account of the small gradient of temperature existing at 

 the adjacent depths. 



Immediately upon complete thawing the temperature of the sand and 

 gravel rose at once a few degrees above that of the clay and loam and 

 several degrees above that of peat. In about three or four days, how- 

 ever, from the day that the actual thawing occurred, March 31, and 

 freezing did not reoccur, the temperature of the sand, gravel, loam and 

 clay reached about the same degree; that of peat, however, did not rise 

 to the same magnitude until April 26 or about 27 days later. The reason 

 for this is that its 12-inch depth did not begin to thaw until April 15 or 

 almost one month after the corresponding depth of the other soils com- 

 menced to thaw. As soon as its entire frozen layer, however, had thawed 

 then its temperature rose rapidly and approached that of the other soils 

 in a short time. 



The slow rate of thawing of peat and the rapid ascend of its tempera- 

 ture thereafter is extremely interesting and suggests very strongly that 

 it is the latent heat of fusion of ice, which amounts to 80 calories per 

 gram of water, which delays the warming of those soils possessing a large 

 amount of water and not necessarily the high specific heat of the water in 

 the liquid state. This inference finds strong confirmation in the fact 

 that during the fall the peat which possesses the .greatest total specific 

 heat cools at the same rate and has practically identical temperature as 

 the other soils. 



The rapidity with which the clay and loam thawed and the rise of 

 their temperature to the same magnitude as that of the sand and gravel 

 in such a short time contradicts the exaggerated general belief that 

 these soils possess a greatly lower temperature during the spring and 

 therefore should be classed as cold soils. 



Although by April 26 the average temperature of all the diflFerent types 

 of soil approached the same magnitude it did not come any nearer be- 

 tween them, as it continued to rise in the subsequent months, until next 

 September. Thus during the months of May, June, July and August it 

 varied appreciably in the different types, being always highest in sand 

 and gravel followed by loam, clay and peat, respectively. At some days 

 the sand and gravel had as much as 10°F higher temperature than the 

 peat at the 6-inch depth. This large difference, however, occurred only 

 whenever the air temperature rose very high suddenly. As a monthly 



