THE CHALLENGE OF STAS. 179 



points determined by the data of Stas in error lie away above 

 our curves. Our true values lie exactly on the curves. We 

 have supposed that Stas, as a matter of fact, was not work- 

 ing in a pit 4,000 yards deep, nor in a room the temperature 

 of which reaches within 70 degrees of the absolute zero; but 

 that temperature and pressure at No. 6 were about the 

 average of what they were for Nos. i, 2, 3, 4, 5 and 7. 



VI. Of course, we would not say that Stas intentionally 

 falsified his record for No. 6 and No. 8; but all the same it 

 is a remarkable fact, that by getting No. 6 " high " and No. 

 8 " low " his values got very much nearer into line than they 

 are in fact; that is, by the curious "mistake" committed in 

 the reduction to vacuum, giving in No. 8 a value 40% low 

 and in No. 6 a value iSofo high, the false results given by 

 Stas approach more nearly a common mean, than they do 

 without those two " mistakes." 



That none of the eminent and honest Re-Calculators 

 detected these mistakes of Stas is perfectly natural. " Blind 

 followers of a blind guide " are not expected to see much. 



VII. Taking now the points both for "dried" and 

 " fused " nitrate, marked i mean 2, 3 8, raised mean 

 4, 5 true 6 we can draw a curve through each of these 

 two sets, namely (always see Plate II) : 



Curve A B C D for dried silver nitrate; and 

 Curve E F G H for fused silver nitrate. 



VIII. The total range of actually determined values 

 runs from 6, fused, true, 14.041 or 14.04, to 2, 3, dried, 14.097 

 or 14.10, showing a total range of 0.06. 



IX. But the range actually observed is still greater, 

 because No. 2, dried, is 14.11, making the entire actual 

 range 0.07. 



X. Since now the pretended value of Stas is 14.04, the 

 range of his own determinations (say only 0.06) is greater 

 by 50% at least, than the value he has pretended to deter- 

 mine, namely, the exact excess 0.04 above 14. 



XI. When the determinations are of such varying kind, 

 science holds that no determination has been affected. 



