BOILING POINTS OF HOMOLOGOUS LIQUIDS. 977 example, in the alcohols themselves, as the proportion of multiples of EH, increases the boiling point falls. 2 The ethers of acids homologous with oxalic acid, ЄHan-2 are isomeric with certain ethers of the diatomic alcohols, which affords an opportunity of contrasting these bodies; but it must be admitted that the accuracy of the numbers given for some of the glycolic ethers is questionable. A comparison of metameric bodies of analogous composition shows that the boiling points of such compounds coincide very nearly, whereas if they belong to series which differ in their composition, the boiling points may vary widely. The differences between the hydrated acids and the compound ethers metameric 978 DETERMINATION OF BOILING POINTS. with them, given in the table p. 974, show the latter point very clearly. Many of the differences observed between the boiling points obtained by experiment, and the theoretical numbers (calculated according to Kopp's law of a constant difference) may possibly disappear when the boiling points shall have been more accurately determined but Kopp observes, that it must not be assumed that the boiling points of liquids, when compared at any given pressure, will always exhibit the same differences as they do when compared at the average pressure of 30 inches (760 m.m.) of mercury. It does not follow, for instance, though the boiling point of ether is 79°5 below that of alcohol, under a pressure of 30 inches, that under a pressure of 15 inches the difference between these boiling points should still be 79°5: since it is well known that what is called Dalton's law (190), viz., 'that all liquids at equal distances from their boiling point emit vapours of equal tension,' is not correct. The accurate determination of the boiling point of a liquid is by no means easy, as might be inferred from an inspection of the foregoing tables, where many experimentalists of great skill and experience give for the same well-known substances, numbers which differ by several degrees from each other. Indeed, it has been found, independently of the irregular manner in which many liquids boil in glass vessels (180), that in several cases impurities so slight as not materially to affect the result of ultimate analysis of the substance, may yet produce a considerable effect upon its boiling point. Numerous attempts have been made to determine the effect which the addition or the removal of a certain number of atoms of each of the elements, carbon, hydrogen, and oxygen, exerts upon the boiling point of the resulting compound. These attempts, however, have not led to any satisfactory result, since the compounds thus compared belong generally to different series. The aldehyds, for example, cannot satisfactorily be compared with the alcohols, and it is only in the same homologous series that an equal rise in the boiling point can be expected for equal increments in the number of the constituent elements. In the case of the alcohols, the volatile acids, and the compound ethers, Kopp, assuming a constant difference of 34°2 F. (19° C.) for EH,, has calculated a table of boiling points which approximate pretty closely to the observed numbers, the mean of the best observations of which are contrasted with Kopp's calculated numbers. CALCULATION OF BOILING POINTS. 979 If the calculated results given in the ensuing table be assumed to be accurate, it follows, if one of the bodies contained in this table be compared with others analogous to it in composition, that the boiling point will be raised by about 52°2 (29° C.) for each additional atom of carbon; and that it will be reduced by about 52°2 for a diminution of each atom of carbon: but the effect produced by hydrogen is the reverse of that produced by carbon; since for every additional atom of hydrogen, the boiling point is lowered 9° (5° C.); and for every atom of hydrogen which is removed, it is raised 9° :— The following compounds are adduced by Kopp as affording an illustration of this alteration of the boiling point by the addition of carbon and hydrogen in isologous compounds; but the experimental numbers often differ very considerably from the calculated values: |