Generally the snow laying on the ground near the earth thermometer is cleared away to make the observation easily. The author made a special observation in the last February at Sapporo, to find its _??_ff_??_ct on the re_??_ding. A c_??_r_??_fully standardised thermograph was put in a tin box and baried at a depth of 30 cms. under the ground covered with snow. The observation was compared with the readings of two earth thermometers-one is suspen ed in an iron tube and the other directly ins_??_rted in the soil. The thermograph reading was always above the freezing point and higher than those of the thermometers. The mean differences of the thermograph reading from those of the thermometers in tube and directely inserted in soil were 1.°5C. and 2.°OC respectively. The soil near the iron tube of the earth thermometer was frozen to a depth of 20cm. or more, while that under snow was warmer and soft.