EECEEATIVE SCIENCE. 



331 



a globe 1000 feet in diameter would in one 

 second of time ; for how many ages, there- 

 fore, has the cooling process gone on, to 

 bring it to its present condition of greenness 

 and fertility — its once confused and cloudy 

 vapours condensed into seas and lakes, and 

 its axes capped with everlasting snows ! 



It must be confessed that we have no de- 

 finite data by which to determine how many 

 eons must elapse for the cooling of the entire 

 globe to the present general temperature of 

 the surface, nor can we say at what rate it 

 parts with heat. The earth may be said to 

 weigh about three thousand six hundred and 

 fifty-two millions seven thousand and ninety- 

 two billions of tons.* How long would a 

 globe of such a bulk, if heated to the tempe- 

 rature of melted iron, and exposed in a 

 medium at 0" of Fahrenheit, require to cool 

 down to that same point on our thermometric 

 scale? 



The late M. Arago, in one of his brilliant 

 scientific contributions to the Annuaire, fur- 

 nishes a proof, deduced from the moon's 

 motion, that the general temperature of the 

 mass of the globe has not varied the tenth 

 part of a degree during the last 2000 years. 

 Reducing his ample explanations into a 

 small compass, the argument takes shape as 

 follows : — 



Suppose a heavy mass were adapted to 

 each radius of a wheel, so that we could slide 

 the mass from the extreme end of the radius 

 up to the axle. If the masses are set near 

 the axle, a certain force would be required to 

 give the wheel a rotation of (say) one revolu- 

 tion per second. Move the masses to the 

 outer extremities of the radii, and though 

 the total weight of the apparatus wiU not be 

 changed, yet a greater force will be required 

 to give, as before, one revolution per second. 

 It foUows, therefore, that, as to turn a mass 



* This is the result of a rough calculation, made 

 on the hasis of Cavendish's pendulum experiment. I 

 have lost my original memorandum of the calculation, 

 •tad have- now grave doubts of its accuracy. 



of a given weight with a given velocity, the 

 force must be increased as the elements of 

 weight are removed from the axis of rotation ; 

 so, with a constant force, the rotary velocity 

 of the mass wiU be retarded in proportion as 

 its different particles are more distant from 

 the axis. That which is true of a flat wheel 

 is, in this case, also true of a spheroid, such 

 as the earth, which rotates on its axis by 

 virtue of an original impulse. 



The materials of which the earth is com- 

 posed expand by heat and contract by cold, 

 and if contraction take place, the rotation on 

 its axis must of necessity be increased in 

 velocity. Hence the rate of the earth's rota- 

 tion is at all times a measure of contraction 

 or expansion of its bulk. The duration of 

 the revolution is an unit of time, used by 

 astronomers in all ages as the basis of the 

 boldest inquiries and the most sober predic- 

 tions, and may be used here to determine if 

 any change has taken place in the heat of the 

 globe during 2000 years. The ancients knew 

 nothing of the thermometer, and hence have 

 bequeathed no data for determining changes 

 of heat, but they were accurate in their mea- 

 surements of time. 



Amongst the uses made by the ancient 

 astronomers of the unit of time afforded by 

 the rotation of the earth, was that of measur- 

 ing the revolutions of the moon. The school 

 of Alexandria has left materials by which 

 to deduce the velocity of the moon in past 

 times with the greatest exactitude. The 

 Arabian astronomers have furnished similar 

 data for the time of the caliphs, and there 

 not, says Arago, a single catalogue of modern 

 observations in which the moon's mean mo- 

 tion during a sidereal day is not given. Now 

 the arc passed over by the satellite in that 

 unit of time is found to be precisely the 

 same, whether deduced from the Grecian, 

 Arabian, or from recent observations ; the 

 phenomena of perturbation proper being, of 

 course, left out of the deduction. Hence the 

 sidereal day has been precisely the same at all 



