6o CARNEGIE INSTITUTION 



We have all been very sorry that circumstances did not allow the 

 continuation and extension of that laboratory, and I am sure it 

 will be a good thing if it can be followed up in the manner you pro- 

 pose. I cannot but think that Mr. Carnegie himself will be inter- 

 ested in the subject. 



lu answer to your letter of May 15, I am sorry I do not know 

 enough to allow me to recommend anyone as specially well quali" 

 fied for conducting the work. 



Yours sincerely, 



Kelvin. 



IE. Sui^ss to Mr. Walcot.t. Jime 7, /po^.] 



Vy£,^'^.^, June y, igo2. 

 Dear Sir : 



Your very kind letter, dated May 5, has occupied me very much 

 during the last days. There is danger in proposing experiments 

 which must be conducted under conditions too different from those 

 of nature. I hope I have not gone beyond your intentions in invit- 

 ing two distinguished friends, one our ex -professor of physics, Mr. 

 Mach, and the otlier our professor of petrography, Professor Becke, 

 to a confidential discussion of your question. Experiments are now 

 being made on melting points at Gratz (Doelter) and at Geneva 

 (Brin), and are in preparation here. You know the brilliant experi- 

 ments on cooling by Sir Robert Austen. But it seemed to my 

 friends as well as to m3-self that our knowledge is singularly deficient 

 in regard to the influence of pressure on the melting point, as well 

 as on increase or decrease of solubility. For the sake of attaining 

 exactitude of expression, I asked my friends to give me their 

 opinions in a few written lines, and these I beg to append to this 

 letter. 



But I must confess that quite a flood of different pieces of work 

 and questions of the most varied character arise before my mind. 

 I beg permission to speak of one of these tasks. 



It is a very curious fact, that any brick or piece of pottery during 

 baking attains, and then retains for all time, the magnetic orienta- 

 tion of the place of baking. Every fragment of brick or of old pot- 

 tery gives the magnetic meridian of the place and time of baking, 

 not the declination, because the situation during baking is not 

 known. The variation of the magnetic meridian during the last 

 thirty or forty years can easily be read from our bricks, and if the 



