344 
DK, A. DUPEE ON THE SPECIFIC HEAT AND OTIIEE 
10, 20, 30, 40, 50, 60, 70, 80, and 90 per cent, alcoholic strength respectively. Excluding 
the last temperature, as in this case a slight experimental error would greatly influence 
the result, the agreement between the rest is close enough to warrant the conclusion 
that the actual variations observed are due only to slight experimental error ; and we 
may therefore look upon the mean of all, namely 8G°T, as being approximately the tem- 
perature at which no heat at all would be evolved, whatever be the proportion in which 
the alcohol and water are mixed. 
The idea at once suggests itself that the whole of the phenomena under consideration 
are due to dissociation. For every given temperature an equilibrium may exist between 
the free alcohol and water on the one hand, and the compound formed between the two 
on the other. A rise in temperature might be accompanied by dissociation of some of the 
compound present, and consequent absorption of heat ; a fall of temperature, on the 
other hand, would then be accompanied by reunion and consequent liberation of heat, 
whereby the apparent specific heat of the mixture would be augmented. The same idea 
has already been advanced by Pfaundler, and also by Marignac, to account for the 
observed deviations of the specific heat of many mixtures from their calculated mean value. 
A closer examination of some of the other properties of these alcoholic mixtures does 
not, however, tend to confirm this supposition. Thus, as is well known, considerable 
contraction in volume follows the mixing of alcohol and water ; but this contraction is 
not, as might be supposed, in any degree proportional to the amount of heat evolved. 
Thus the maximum contraction takes place in a mixture containing about 45 per cent, 
of alcohol, whereas the maximum amount of heat is evolved at an alcoholic strength of 
about 30 per cent. The maximum elevation of the specific heat above its calculated 
mean value is also observed at this strength of 30 per cent. ; and hence, in order to recon- 
cile the above theory with fact, we must assume on the one hand that the amount of 
combination taking place at any given temperature is greatest when 30 parts of alcohol 
are mixed with 70 parts of water, and on the other that a given rise of temperature will 
produce in this mixture of alcohol and water a greater amount of decomposition or 
dissociation than in any other. We should thus have the maximum amount of chemical 
action and the feeblest union both occurring with one and the same proportion of alcohol 
and water. 
The rate of expansion of the various mixtures seems also opposed to the supposition 
that dissociation is the cause of the high specific heat observed. It is to be supposed 
that the greater the amount of dissociation the more nearly would the specific gravity 
observed correspond to its calculated value. In other words, the rate of expansion of 
all the mixtures should be above its mean value, and this excess should stand in some 
relation to the observed elevation of the specific heat. The following Table will show 
that this is by no means the case. It gives for spirits of 30, 20, and 10 per cent, strength 
by weight, the specific gravities as found and as calculated for the temperatures 0°, 10°, 
20°, 30 ', and 70 C. The specific gravities for the temperatures 0, 10, 20, and 30 are 
taken from the Tables of Mendelejeff. 
