Whether it used in the climic in comparison with the prediction accuracy of electronic clinical thermometer was examined under the cooperation of the volunteer. There was the high correlation on predictive value and measured value with A c202, c220 types. The reproducibility by the iterative measurement by the equal clinical thermometer is high.
 Though the sensor is being set in the temperature sensing element metal cap of the clinical thermometer, the thermister has been minimized in order to improve the thermal reaction. It becomes an element in which this fact stimulates the short time measurement.
 The prediction accuracy lowers, and on the other hand the sensor is small, and it is difficult to be fixed at best warm division on arteria auxillaries. The artifact occurs, when the sensor is not rightly fixed, and it concludes easiness.

Experiments to determine the relationship between the exposure time following death from cold and the amount of residual NPN in blood and organs.
Sixty rabbits weighing about 2500 grams were used in these experiments. Their whole bodies except the head and face were immersed in water at a temperature of 2 to 10°C and fastened by tight binding. Their rectal temperatures were measured five minutes apart using a thermocouple thermometer. The measured amount of residual NPN in tissues and blood with each experiment group by the Kjeldahl method after the death are as follows:
1) The amount of residual NPN in organs and blood differed depnding on aggressiveness and the length of time until death.
2) When the length of time until death was relatively short, the amount of NPN in organs decreased and that in blood increased.
3) When the length of time until death was relatively long, the amount of NPN inorgans increased extremely and that in blood also increased.
4) The amount of residual NPN in organs and blood starts to change at the initial stage of aggression.

Recently, many deaths while bathing in the home bath have been reported. During the winter season, hot water at a high temperature is commonly used for bathing. The circulatory system may thus be seriously affected by the hyperthermia load produced by undressing in a cold environment and subsequent bathing in a bathtub at a high temperature followed by a rapid temperature change due to exposure to cold air after bathing. However, death while bathing also occurs in the summer season.
From the cases of death while bathing in the summer season, we found commonality in terms of sex, age, water temperature, bathtub size, and causes of death.
The average age in all cases was 69.7. There were 47 male and 46 female deaths indicating an almost equal ratio. The average water temperature at the time of deaths while bathing was 40.7°C. The typical bathtub size was small, 750 to 900mm. In a small Japanese style bathtub, one must compress the body and therefore become more vulnerable to water pressure.
From these results, warming, drawing in of limbs, and effects of water pressure on the body may contribute to deaths while bathing in the summer season and also are factors produced by bathing throughout the year.

We recently conducted 10-minute axillary temperature measurements on 699 healthy individuals, ranging in age from 5 to 83 years (mean: 36.5±15.3 years).
Axillary temperature readings are sometimes inaccurate because the thermometer is not inserted in the correct place (i.e. the point of the highest temperature) or due to incomplete closure of the axilla during measurement. As the result of analysis of temperature rises during the 10-minute axillary temperature measurement, we found that the results are not always accurate because of incorrect conditions of measurement such as when the thermometer reading does not reflect the surrounding temperature.
In this study, the following temperature readings were regarded to be inaccurate when: (1) the temperature temporarily fell during the rising phase; and (2) the rise in temperature was accelerated during measurement and the temperature had not become stable after 10 minutes. As a result of analyzing 10-minute axillary temperature measurements, we believe the temperature rise during the measurement must be considered to obtain more accurate readings.
When analyzing the readings of short-term temperature measurements using a predictive algorithm, awareness required of possible errors which may be caused by the measurement method employed. Error factors other than the algorithm used for prediction become large when the thermometer is temporarily withdrawn from the axillary pit or its direction is changed during measurement to check an interim reading. After error factors associated with the measurement method have been eliminated, high accuracy is obtained and the difference between predicted reading and 10-minute becomes 0.01±0.13°C. A difference smaller than±0.2°C has been achieved in 98.0% of all measurements. Such a difference causes no problem from a clinical point of view.