November 16, 1900.] 



SCmNCE. 



755 



tare ; such an average is merely a compro- 

 mise, and a nearer approach to a correct re- 

 sult will possibly be attained by selecting 

 some low rate of increase, provided it is 

 based on accurate observations. 



It is extremely doubtful whether an area 

 such as the British Isles, which has so fre- 

 quently been the theater of volcanic activity 

 and other subterranean disturbance, is the 

 best fitted to afford trustworthy results ; the 

 Archfean nucleus of a continent might be 

 expected to alibrd surer indications. Un- 

 fortunately the hidden treasures of the 

 earth are seldom buried in these regions, 

 and bore- holes in consequence have rarely 

 been made in them. One exception is af- 

 forded by the copper-bearing district of 

 Lake Superior, and in one case, that of the 

 Calumet and Hecla mine, which is 4,580 

 feet in depth, the rate of increase, as de- 

 termined by Professor A. Agassiz, was 1° F. 

 for every 223.7 feet. The Bohemian ' horst ' 

 is a somewhat ancient part of Europe, and 

 in the Przibian mines, which are sunk in it, 

 the rate was 1° F. for every 126 feet of de- 

 scent. In the light of these facts it would 

 seem that geologists are by no means com- 

 pelled to accept the supposed mean rate of 

 increase of temperature with descent into 

 the crust as affording a safe guide to the 

 rate of cooling of a solid globe ; and if the 

 much slower rate of increase observed in 

 the more ancient and more stable regions of 

 the earth has the importance which is sug- 

 gested for it, then Lord Kelvin's estimate 

 of the date of the ' consistentior status ' may 

 be pushed backwards into a remoter past. 



If, as we have reason to hope. Lord Kel- 

 vin's somewhat contracted period will yield 

 to a little stretching. Professor Joly's on 

 the other hand, may take some paring. 

 His argument, broadly stated, is as follows: 

 The ocean consisted at first of fresh water ; 

 it is now salt, and its saltness is due to the 

 dissolved matter that is constantly being 

 carried into it by rivers. If, then, we know 



the quantity of salt which the rivers bring 

 down each year into the sea, it is easy to 

 calculate how many years they have taken 

 to supply the sea with all the salt it at 

 present contains. For several reasons it is 

 found necessary to restrict attention to one 

 only of the elements contained in sea salt : 

 this is sodium. The quantity of sodium 

 delivered to the sea every year by the riv- 

 ers is about 160,000,000 tons ; but the 

 quantity of sodium which the sea contains 

 is at least ninety millions of times greater 

 than this. The periods during which riv- 

 ers have been carrying sodium into the sea 

 must, therefore, be about ninety millions of 

 years. Nothing could be simpler ; there is 

 no serious flaw in the method, and Profes- 

 sor Joly's treatment of the subject is ad- 

 mirable in every way ; but of course in cal- 

 culations such as this everything depends 

 on the accuracy of the data, which we may, 

 therefore, proceed to discuss. Professor 

 Joly's estimate of the amount of sodium in 

 the ocean may be accepted as sufficiently 

 near the truth for all practical purposes. 

 We may, therefore, pass on to the other 

 factor, the annual contribution of sodium 

 by river water. Here there is more room 

 for error. Two quantities must be ascer- 

 tained : one the quantity of water which 

 the rivers of the world carry into the sea, 

 the other the quantity or proportion of sod- 

 ium present in this water. The total vol- 

 ume of water discharged by rivers into the 

 ocean is estimated by Sir John Murray as 

 6,624 cubic miles. The estimate being based 

 on observations of thirty- three great rivers 

 although only approximate, it is no doubt 

 sufSciently exact ; at all events such alterna- 

 tions as it is likely to undergo will not greatly 

 affect the final I'esult. When, however, 

 we pass to the last quantity to be deter- 

 mined, the chemical composition of average 

 river water, we find that only a very rough 

 estimate is possible, and this is the more 

 unfortunate because changes in this may 



