4.VJ 



PROFESSOR W. DITTMAR ON THE BEHAVIOUR OF THE 



Experiment. 



Pressure. 



. 



31-7 



A . 



31-8 



B . 



31-8 



C . 



31-8 



D . 



30-2 



G . 



30-2 



N . 



44-7 



P . 



59-7 



M . 



60-0 



H . 



601 



J . 



59-9 



K . 



60-0 



Percentage of C0 2 

 59-601 

 59-645 

 59-529 

 59-655 



Li 2 = 



29-824 

 29-770 



(29 - 913) suspected exp. 

 29-757 



Not determined, but substance proved neutral. 



59-683 29-723 



59-604 29-821 



59-517 29-928 



59-663 29-748 



60-143 (29159) 



59-794 

 59-584 



29-586 

 29-845 



In addition to B, I exclude Experiment H, because the percentage of carbonic acid 

 found is far higher than that obtained in any other of the high-pressure experiments, and, 

 besides, the analysis was made with an exceptionally small quantity of substance. The 

 mean of the other (nine) numbers is 29 "7 78 ; from the deviations of the individual 

 results from the mean, I calculate that the "probable" error of the mean is = ± 0*020. 

 Hence Li = 6*889±0'010. If we allow only the low-pressure experiments (O, A, C, G) to 

 vote, we have Li 2 = 29769 and Li = 6*8842±0*007 ; but we have no excuse for excluding 

 the high-pressure experiments. Another question is whether we may not neglect Experi- 

 ments J and P as being in all probability infected with abnormal errors. In their case, 

 indeed, the deviations from the mean amount to about three times the probable error of a 

 single determination, which is ± 0*06. If we do exclude these two results, the mean of 

 the remaining seven becomes equal to 29 '7 8 4 for Li 2 and 6 '892 for Li with a probable 

 error of ± 0*0136 for the former. I adopt this number 6*892 =b 0*007, or rather 6*89 = 

 Li, as the net result of my work, the more unhesitatingly as it, after all, does not differ 

 much from the general mean of the nine experiments. But, how does it agree with the 

 results of previous observations by others ? Of these the following chiefly come into 

 consideration : — * 



I. Mallet, in 1857, analysed chloride of lithium in two ways, namely — 

 A. By determining the chlorine gravimetrically as chloride of silver ; results of two 

 analyses — 



Li - 



Minimum. 

 6-947 



Maximum. 

 6-950 



Mean. 

 6-949 



B. By titrating the chlorine with silver ; one analysis gave Li = 6*934. 



II. Troost, in 1857,t analysed carbonate of lithia dried, "sometimes in vacuo, 



* I quote from Lothar Meyer and Seubert, Die Atomgewichte der Elemente, Leipzig, 1883. With Lothar 

 Meyer and Seubert, "Mean" means not the arithmetical mean of the several determinations, but the result as it, 

 comes out if all the analyses are calculated as one. Thus, for instance, in the case of four analyses of chloride 

 of lithium by nitrate of silver, the four quantities of chloride are added together, and compared with the sum 

 of the four precipitates of chloride of silver. 



t Lothar Meyer and Seubert refer to a memoir published in 1857. I find, in the Zeitschrift fur Chemie for 

 1862, an abstract of a memoir of Troost's on the same subject, in which the numbers quoted as immediate data of the 

 analyses agree with those given by Lothar Meyer and Seubert. 



