=69. Importance of Self-Registering Thermometers.=--Heat being apparently the most effective agent in producing meteorological phenomena, the determination of the highest temperature of the day, and the lowest during the night, is a prime essential to enable an estimate of the climate of any place to be formed. To observe these extremes by means of the ordinary thermometer would be impracticable, from the constant watchfulness which would be necessary. Hence, the utility and importance of self-recording thermometers are evident. A thermometer constructed to _register_ the highest temperature is usually called a _maximum thermometer_; one to show the lowest temperature is termed a _minimum thermometer_; and if made to record both extremes of temperature, it is designated a _maximum-and-minimum thermometer_. We will, for the sake of method, describe the instruments in use in this order. It would carry us beyond our scope to explain in detail the methods of dealing with temperature observations; but we may remark that half the sum of the maximum and minimum temperature of each day of twenty-four hours, is not what meteorologists designate the _mean daily temperature_, although it very frequently approximates to it. The mean temperature of the day is understood to be the average of twenty-four consecutive hourly readings of a thermometer; and meteorology now supplies formulae whereby this result can be deduced from two or three observations only in a day. But we would observe that the actual mean temperature of any place has not such an important influence upon life, either animal or vegetable, as the abruptness and magnitude of the variations of temperature. Climate, therefore, should be estimated more by the range of the thermometer than by the average of its indications. The Registrar General's returns prove that with a wide range of the thermometer, the mortality greatly increases; and it is now becoming apparent to meteorologists that the daily range of the thermometer marks the effects of temperature on the health of men, and the success of crops, better than any other meteorological fact of which we take cognizance. Now that self-registering thermometers are constructed with mercury, the most appropriate of all thermometric substances, not only for maxima, but likewise for minima temperatures, the determination of the diurnal range of temperature is rendered more certain, and observations at different places are more strictly comparable. MAXIMA THERMOMETERS.