102 Mr. E. H. Hayes on Objections to Mr. Pickering's 



personal element is introduced," nor that " considerable 

 care and practice are requisite before it can be safely used/' 

 Uniformity of results can be no real argument in favour 

 of a method, which, as I have above endeavoured to show, 

 would seem by its very nature incapable of being applied 

 in practice so as to deal effectively with the question at 

 issue. 



Mr. Lupton attempts to clinch his argument by applying 

 an arithmetical process to some ten or a dozen density deter- 

 minations forming only a small part of one of Mr. Pickering's 

 series of experiments. In order to exhibit the value of the 

 method, he is, of course, compelled to take them so that 

 they include an assumed break, but minimises this disad- 

 vantage by choosing them in a region where, according to 

 Mr. Pickering, the curve " is of a very doubtful character." * 

 Although under the circumstances the chances in favour of a 

 very fair agreement between the observed and calculated 

 values were considerable, the result is most unsatisfactory 

 from Mr. Lupton's point of view. He seems to have been 

 misled by a statement of Mr. Pickering as to 0*0002 gram 

 being a safe limit of error; but, since the piknometer held 

 25 c.c, the corresponding error in the density is, as Mr. Picker- 

 ing says, only 0*000008, which is far exceeded by every one 

 of the differences between observed and calculated values 

 tabulated by Mr. Lupton. Even if the limit of error were 

 ten times as great as the above, it would only mend the 

 matter in two cases, and this occurs in spite of the fact that 

 his equation contains three arbitrary constants. Comment is 

 needless. 



There are several minor points in Mr. Lupton's paper 

 which call for some remark. After making an assertion 

 concerning the equation of the second degree which is untrue 

 unless it happens to represent a parabola whose axis is 

 parallel to the axis of y, he proceeds to say, " Hence the 

 hydrates of sulphuric acid are apparently made to depend 

 upon the flexibility of the steel lath used." f It is extremely 

 difficult to fathom the precise meaning of this statement, still 

 more so to see how it depends upon the preceding one. It may, 

 however, be as well to point out that any two uniform laths 

 may be bent so as to form identical curves (provided of course 

 that the bending does not exceed a certain limit), although if 

 the flexural rigidity (stiffness) were too great the operation 

 might be trying to the fingers. 



Mr. Lupton very rightly remarks that " the assumption 



* " Nature of Solutions," p. 76. 

 t P. 421. 



