i 4 2 SCIENCE PROGRESS 



in which thirteen breaks have been detected at concentrations 

 ranging from 2 to 98 per cent. H 2 S0 4 . It is noteworthy that, in 

 spite of the well-recognised instability of complex hydrates, the 

 cumulative evidence for breaks at 3*99 and 9*69 per cent. H 2 S0 4 

 (recognised in five properties) is greater than in the case of the 

 seven breaks between 10 per cent, and 80 per cent. H 2 S0 4 (recog- 

 nised in three properties), and is actually greater than in the case 

 of the hydrate H 2 S0 4 H 2 0, the presence of which was recognised 

 in four properties of the solution. 



As the modern theory of solutions does not admit of the 

 recognition of hydrates of sufficient complexity to correspond 

 in composition with the breaks that have been recorded for 

 dilute solutions, it has been customary to pass them over as if 

 they were inadequately established. This involves, however, a 

 refusal to recognise the whole of the breaks between o and 

 80 per cent. H 2 S0 4 , even if indeed it is permissible to select for 

 recognition the break at 84/5 per cent, (corresponding with the 

 hydrate H 2 S0 4 . H 2 0), which is hardly more pronounced than 

 those recorded at the lower concentrations. This refusal might 

 be justified in the case of breaks deduced from the study of 

 any one curve or property, but it is less easy to override the 

 cumulative evidence which Pickering has put forward in this 

 particular case. It appears, therefore, to be worth while to 

 consider whether an alternative explanation cannot be devised 

 which will admit of the recognition of at least some of the 

 breaks without postulating the existence of an improbable series 

 of unknown hydrates. 



A possible clue to an alternative explanation is afforded by 

 Riicker's analysis of the sulphuric acid density curves {Phil. 

 Mag. 1 89 1, 32, 304). He found that, between 46*9 per cent, and 

 8o'5 per cent. H 2 S0 4 , the relationship between density and con- 

 centration might be represented approximately by the equation 



j = a + bx — cd x , 



but that there was a small deviation of the calculated from the 

 experimental values, which reached a maximum at about 60 per 



from the study of independent properties was necessary before such changes could 

 be regarded as established ; and it was only after obtaining such evidence from the 

 study of three or four other properties that I ventured to call this evidence proof, 

 and then only with the oft-repeated caution ' that many of these changes were 

 admittedly of a very doubtful nature'" (Pickering, Phil. Mag. 1892,33, p. 143). 



