22 SECTION PHYSICS. 



and a-half atmospheres. This is very feeble to that given by 

 the ammonia. 



3. This liquid cannot be decomposed, and has no action on metals 



or grease. 



4. This liquid, in a gaseous form, is so good a lubricator that all 



grease is unnecessary. 



5. By using this liquid all danger of fire or explosion is avoided. 



6. This liquid can be produced at a lower price than that of ethei 



or ammonia. 



7. The price of a ton of ice, when sulphurous acid is used, is about 



seven shillings, all told. 



These are the principal advantages gained by the use of anhydrous 

 sulphurous acid in the manufacture of cold. I thank the Committee of 

 the Exhibition for having allowed me to explain my process. 



The PRESIDENT : It is perhaps already known to most present 

 that the machine, which has been so well described by M. Pictet, is 

 exhibited downstairs, and is frequently in operation, so that you can 

 all see not only the results, but the actual modus operandi. I believe 

 I may say M. Pictet will have much pleasure in explaining it again on 

 the spot to any who may be desirous of understanding it. 



I will now call on Sir William Thomson to speak on 



THE PRINCIPLES OF COMPASS CORRECTION IN IRON SHIPS. 



Sir W. THOMSON, LL.D., F.R.S. : Mr, President, Ladies, and 

 Gentlemen, the principles by which a compass in an iron ship may be 

 corrected so that it shall point to the true magnetic north in every 

 position of the ship, were pointed out about forty years ago by the 

 Astronomer Royal. He then shewed that by the application of masses 

 of soft iron, and also of permanent magnets, in the neighbourhood of 

 the compass, the whole disturbance produced by the iron of the ship 

 could be annulled and the compass brought to indicate the true north 

 in whatever position the ship's head might lie. In the Astronomer 

 Royal's investigation, however, there was a certain assumption made 

 regarding the magnetic induction, which could only be approximately 

 true for substances of exceedingly small inductive capacity, and which 

 was very far from being true for iron. The nature of this assumption 



