EXPERIMENT STATION BULLETINS. 499 



HEAT CONDUCTIVITY 01-' NATURAL SOILS UNDER LABORATORY CONDITIONS. 



The next best way to study the heat conductivity which would yield 

 data of practical as well as of theoretical value, would be to cut cones 

 of soils at different seasons of the year, dependini,^ upon the meteoro- 

 logical elements which Avould bring the soils under the different condi- 

 tions of moisture, texture and structure, and thus measure the heat 

 conductivity of these cores of the different kinds of soils under controlled 

 laboratory conditions. In the present investigation, that is just what 

 was attempted. The method of operation was as follows: Four plates 

 of iron 8 inches long, 7 inches wide, and Yg inch thick, with their front 

 and opposite edges sharpened, were pressed or bored into the different 

 types of soil — which types were placed one year before in adjacent 

 wooden boxes 3x3x3 ft., without top or bottom, as will be explained 

 later — so as to form a square. The soil around the outside of the plates 

 was then removed to its full depth. The plates came out and left 

 the core of the soil standing. A wooden box 8 inches deep, 6 % inches 

 wide, and 1 inch thick, was placed over this core until the top soil was 

 even with the upper side of the box. Next a copper plate was fitted on 

 this side of the box which came in intimate contact with the top of the 

 soil. The bottom of the core was next severed from the ground with 

 a sharp knife, then smoothed and kept in place by a wooden cover. This 

 whole operation, i. e., driving the plates into the ground, placing the 

 box over the core, severing the core from the ground, was done with the 

 greatest possible care so that the structure of the soil core may not be 

 disturbed. Precautions were also taken to secure the core 6% inches in 

 diameter so that any slight friction on the sides would not affect its 

 structure. 



The box which contained the core of soil consisted, as already said, 

 of four wooden sides, one wooden bottom, and one copper top. Before 

 it was placed over the soil, it was boiled in paraffin. Along the center 

 of the lengthwise side, 7 holes were bored one inch apart to admit 

 thermometers. 



Four cores of soil were taken, one from sand, loamy, clay, and peat 

 respectively. No core was taken of the gravelly soil since its texture 

 would not permit the plates being driven into it without materially dis- 

 turbing its structure. 



Mention has been made that the top of the core of each soil was smooth 

 and came in contact with the copper plate. To be more certain of this 

 a layer of moist quartz sand was placed over each core before covering 

 it finally with the copper plate. There could now be no doubt of an 

 absolutely perfect contact. Each box was then made air tight by paraf- 

 fin so that there would be no evaporation. The soils were allowed to 

 remain in the room for several hours to attain the same temperature, 

 before the study of their heat transmission was commenced. 



The copper plate end of the box was placed up against the copper 

 side of a tank the water in which was kept at the constant temperature 

 of 33.5° C by means of an electrical arrangement shown in the diagram 

 I. A is a tank filled with water, C is a 32 candle power electric light 



