294 Dr. H. Herwig's Investigations on the Conformity 



Table I. b. 



Tempera- "1 

 ture t. j 



23° 



30°-5 



36°-4 



41°-9 



47°-8 



57°-8 



62°-9 



69°-9 



Limits of v 1 < 



Correspond- J 

 ing c \ 



197-8 



201-6 



005957 



005613 



1259 



133 

 006137 

 005809 



89-6 

 95 

 006247 



005898 



69-9 



72 



0-06036 



005860 



525 



54-4 



005974 



005765 



33 25-2 

 342 275 



006004 0-06304 

 005793 005777 



19 8 



20-5 

 005977 

 05773 



A consideration of these figures gives great probability to the 

 assumption that in c we have a genuine constant ; and compa- 

 ring the v } calculated with c=*0595, as above, with the parti- 

 culars of the tension in the neighbourhood of these volumes, as 

 they may be seen in Table I., it would appear with the highest 

 probability that these values are correct. Hence it appears to 



PV _ 



me that the relation — — = -0595 va + 1 holds actually, at least 



P\ v \ 

 for such temperatures of alcohol-vapour as have been ex- 

 amined. Taking this relation as universally correct for alcohol, 

 it follows hence that, for the particular temperature at which 

 •0595 */a+t = l, the product P\ v \ — PV ; i. e. that at this tempe- 

 rature the vapour of alcohol, so soon as it is separated from the 

 fluids already follows Mariotte's law. As to the temperature at 

 which this happens, it is calculated from the value c = '059487 

 as / = 9 0, 589 Celsius. The investigation^ unfortunately, could 

 not be carried on as far as this temperature in the warm weather 

 of the season ; ice thrown into the bath would not have given, 

 sufficiently steady temperatures. Moreover the deviation of the 

 vapour from Mariotte's law which exists at 23° is already so small, 

 that it only slightly oversteps the possible errors of observation 

 in the slight tension belonging to that temperature. Now, 

 whether the relation f(t) ='0595 \/a + t holds good quite to 

 the temperature of 9 0, 5 for vapour of alcohol, and whether at 

 that and lower temperatures the vapour follows Mariotte's law 

 when free from fluidity, or whether a slight departure from 

 Mariotte's law takes place in the opposite direction (perhaps 

 even according to the law f(t) =c V a 4- 1) , just as Regnault 

 found for hydrogen under a high pressure*, is a question which 

 must be decided by further investigations, attended, of course, 

 by greater difficulties ; and these I intend to execute. 



With respect to the particulars of the products pv which lie 

 between p x v x and PV, after many trials I have not been able to 

 find any formula to which these products would conform as to an 

 actual law. It is probable that the relation actually existing for 

 these products is complicated, like the tension-curve of saturated 

 vapours, the theoretical expression for which has hitherto been 



sought in vain, 



* Mem. de VAcad. des Sciences, vol. xxi.p. 395. 



