So Intelligence and Miscellaneous Articles, 
cumstances usually observed would be accounted for by supposing the 
rain-drops of the diameter of 1-72 of an inch. ‘The considerations 
account for the supernumerary bows being seen near the summit of 
the bow only, since the drops, as they fall lower, coalesce, and be- 
come larger; and the white fog-bows, which are often seen, would 
result from very minute drops, of the diameter perhaps of one thou- 
sandth of an inch, 
Afterwards a communication was also read from C. Darwin, Esq., 
containing a notice that animals (lizards, &c.) which are oviparous in 
certain districts of South America, as they are in this country, are 
viviparous in the province of Mendoza, which he visited. Mr. Dar- 
win also gave an account of red snow observed by him in the road 
from St. Jago de Chili to Mendoza, by the Portello pass, and of a 
microscopical examination of the substance. 
XV. Intelligence and Miscellaneous Articles. 
ON THE ANALYSIS OF GERMAN SILVER, AND THE SEPARATION 
OF ZINC FROM NICKEL, &c. BY MR. JOSEPH D. SMITH. 
ISHING to ascertain in what proportions the copper, zine, and 
nickel were combined in a specimen of German silver, or pak- 
fong, an alloy which has within a few years been extensively employed 
as a substitute for silver plate, I referred to Rose's Analytical Chemis- 
try, in which I find it stated, that when a solution of potash or soda is 
added to one containing the oxides of nickel and zinc, their separa- 
tion is by no means completely effected, a great excess of alkali not 
redissolving the whole of the oxide of zinc, even when boiled; he 
therefore directs the oxides to be converted into chlorides, and then 
to volatize the chloride of zinc ; but the manipulation in this process 
is of so complicated a nature and requires such numerous precau- 
tions to ensure accuracy, that it is quite unsuited to any but the most 
experienced and skilful analysts. 
Among many of the experiments performed I tried to discover a sim- 
pler method than that recommended by Rose: there was one which I 
made to ascertain to what extent the oxide of zinc was soluble in soda 
or potash after precipitation in mixture with oxide of nickel ; for if the 
method would give results correct within 3 or 4 per cent., it would 
have answered my purpose sufficiently well: therefore I took 20 grains 
of zinc and 30 grains of oxide of nickel, dissolved them in muriatic 
acid, and having largely diluted the solution, boiled it for half an 
hour with four times as much soda as it took to precipitate the oxides. 
The precipitate when dried and ignited weighed 46°4 grains, show- 
ing that 16-4 grains of oxide of zinc were precipitated with the 36 
grains of oxide of nickel. Thus the results obtained by the use of 
caustic alkali are as incorrect in this instance, or even more so, than 
when this process is employed to separate copper from zinc, the un- 
fitness of which method was pointed out by Mr. Keates in a paper on 
the Analysis of Brass (Ann. of Philosophy, vol. xix.), and he at the 
same time gave directions for the performance of its analysis. Although 
the results obtained by the employment of his process are sufficiently 
