
TRANSACTIONS OF THE SECTIONS. 33 
CHEMISTRY. 
On the Discovery of Minute Quantities of Soda by the Action of Polarized 
Light. By Professor Tuomas Anprews, M.D., MRA, FLRS., 
V.P. Queen’s College, Belfast. 
Tue double chloride of potassium and platinum crystallizing in regular octahedrons, 
exercises, when placed in the dark field of the polariscope, no depolarizing action; 
and the same remark applies to the bichloride of platinum in consequence of its im- 
perfect crystallization. On the other hand, the chloride of sodium and platinum in 
thin crystalline plates is remarkable for its depolarizing power, and a trace of this salt, 
which is invisible to the naked eye, may be at oncé detected by the brilliant display 
of prismatic colours which it exhibits under the action of polarized light. The author 
applies this property to the detection of soda in the following way. The other bases 
having been removed by the ordinary methods, and the alkalies converted into chlo- 
rides, a drop of the solution is placed on a glass slide, and a very small quantity of 
a dilute sclution of the bichloride of platinum added, avoiding as far as possible an 
excess of that reagent. The drop is then evaporated by a gentle heat till it begins to 
crystallize, and afterwards placed in the field of a microscope furnished with a good 
polarizing apparatus. On turning the analyser till the field becomes perfectly dark, 
and excluding carefully the entrance of light laterally, the crystals remain quite invi- 
sible if either potash alone or no alkali whatever be present; while the presence of 
the slightest trace of soda is at once indicated by the depolarizing action of its pla- 
tinum compound. With a drop of solution of chloride of sodium, weighing 0-0015 
gramme, and containing TT of its weight of chloride of sodium, a very distinct 
effect was obtained. The quantity of soda thus detected was only ;. of a gramme, 
1 
or about Prd 
of a grain. 
On the Atomic Weights of Platinum and Barium. 
By Professor T. ANDREws, M.D., M_R.ILA., F.R.S. 
No determination of the atomic weight of platinum having been made since the 
recent revision of atomic weights, and the number adopted by chemists for that metal 
resting on the authority of a single experiment of Berzelius, the author considered it 
of importance, on practical as well as theoretical grounds, to institute some new expe- 
riments on the subject. The salt of platinum selected was the double chloride of 
potassium and platinum, which, after being dried in vacuo at a temperature of 105° C., 
was decomposed by digestion with metallic zinc and a small quantity of water, the 
action being assisted by the application of heat towards the end of the process. After 
the complete precipitation of the platinum and the formation of chloride of zinc from 
the decomposition of the double salt, the excess of zinc was removed by the addition, 
first of acetic and subsequently of nitric acid. The precipitated platinum was then 
removed by means of a small and carefully washed filter, and the amount of chlorine in 
the solution of chloride of zinc ascertained by Gay-Lussac’s process, which has been 
of late so successfully applied by Pelouze to the determination of several other 
atomic weights. The double chloride of potassium and platinum was found to retain 
varpdths of its weight of moisture, even when dried at a temperature considerably supe- 
rior to the boiling-point of water. In three experiments performed by this process, 
the numbers obtained were 98°93, 98-84, and 99-06; the mean number 98°94 ex- 
presses therefore the atomic weight of platinum. 
For the atomic weight of barium, the author obtained from two closely-accordant 
experiments the number 68°789, and concluded with some general observations as to 
the importance of a systematic series of experiments to settle, if possible, definitively, 
whether the law of Prout, that the atomic weights of all bodies are multiples of that 
of hydrogen, be universally true. He concluded by reading an interesting extract 
from a letter which he received from Baron Liebig :—‘ It is not certain that Prout’s 
law may not be true for oxygen, nitrogen and carbon, without it being necessary to 
assume, as a consequence, that other bodies behave similarly ; that is, their atomic 
weight must be exactly multiples by whole numbers of the atomic weight of hydrogen. 
The law is certainly not true of all bodies, but it may be true of certain groups, whose 
members, in respect to atomic weight, stand in a simple numerical relation to each 
1852. 3 

