TRANSACTIONS OF THE SECTIONS. 2a 



of tliat paper printed in the 'Transactions' for last year at page SO; and, in 

 partii-ular, to the diagram of three curves shown slvetched in fig. 1 of that 

 abstract. 



In respect to these curves several essential features had been, at tlie time of last 

 year's Meeting, clearl}' discerned, and were pointed out and reasoned on by the autlior 

 in his paper then read. His attempt to sketch out the curves, however, in such 

 a -way as tliat tliey sliould be in agreement wntii the Icuown conditions then taken 

 into consideration, soon forced on his attention the question whether tlie two 

 curves, one of which is that between gas and liquid, and the other is that between 

 gas and solid, ought to be drawn crossing as represented here in fig. 1 a, or as in 

 fig. 1 b ; and his object at present is to give a demonstration, subsequently deve- 



Fig. 1 a. 



Fig. 16. 



loped, showing that they must cross as in fig. 1 a ; or, in other words, as in the 

 diagram which he gave in the abstract of his last ^-ear's paper. 



It is to be understood that A X and A Y are the axes of coordinates for pressures 

 and temperatures respectively ; A, the origin, being taken as the zero for pressures, 

 and as the zero for temperatures on the Centigrade scale ; and, for simplicit}^ in 

 expression and in thought, the diagram may be taken as relating to the particular 

 substance of water, steam, and ice, rather than to substances in general. The 

 curve E T P is the hoiUng-Une, or the line which has its successive points such that 

 for any one of them the two coordinates represent a pressure and temperature 

 for a boiling-point, or a pressure and temperature which the water and steam can 

 have when in mutual contact. It may also he called, for brevity, the steam- 

 u-ith-wuter line. In like manner the curve N T Q is the steam-iuith-ice line ; and 

 the curve ^I T li is the wafer-witJi-ice line. The full meaning of these diagrams 

 may become more distinctly intelligible to the reader if he will advert to the ex- 

 planations given in the paper already referred to in last year's 'Transactions,' as to 

 fig. 1 in that paper, — explanations which, though now useful, need not be wholly 

 repeated here, as the present paper is meant to be read in connexion with that 

 previous one. 



If we now look to fig. 1 a and suppose that we have water and steam in mutual 

 contact, the pressure and temperature must be represented by the coordinates of 

 some point of the steam-with-water curve L T P. Let us now suppose that we 

 lower the temperature gradually while keeping water and steam in mutual contact : 

 the point whose coordinates show the successively coexistent temperatures and 

 pressures will pass downwards along the steam-with-water curve L T P. Let us 

 suppose this operation continued so far as to bring this point into that part of the 



