Nov. 11, 1869] 
CHEMISTRY 
New Test for Alcohol 
LIEREN has discovered a new and very delicate test for the 
presence of alcohol, depending upon its conversion into iodoform. 
The liquid under examination is heated in a test-tube, into 
which are then introduced a few grains of iodine, and a few 
drops of potash-solution ; whereupon, if alcohol is present, a 
yellow crystalline precipitate of iodoform is produced im- 
mediately or after sometime, according to the degree of 
dilution of the liquid. This test is said to be capable of 
detecting 1 part of alcohol in 12,000 parts of water. For 
greater certainty, it is best to examine the precipitate with the 
microscope, iodoform exhibiting the appearance of hexagonal 
plates or six-rayed stars. 
The test just described is capable of an important physiological 
application. It is generally supposed that alcohol introduced 
into the animal organism in the form of wine or other spirituous 
liquors becomes completely oxidised, and does not pass into the 
urine as alcohol, but in the form of some product of transforma- 
tion. Lieben, however, by applying the new test to the urine of 
a man who had drunk a bottle of wine half an hour before, was 
able to detect the presence of alcohol in it. A second portion of 
urine voided by the same individual, an hour later, and a third, 
after another half-hour, still exhibited the peculiar reaction under 
consideration, ‘The urine was of course distilled before applying 
the test, and it had been previously ascertained that none of the 
other volatile matters contained in it would produce a similar 
reaction. —[Ann, di Chim, app. alla Med., Sept. 1869, p. 136.] 
Preparation of Silver Nitrate 
P, ScIVOLErTo proposes the following modification of the 
process of preparing silver nitrate for use in medicine, photography, 
&c, This salt is usually prepared from old silver containing 
copper, by dissolving the alloy in nitric acid, evaporating to 
dryness, and calcining the residue as long as nitrous fumes 
continue to escape, ‘The product is a mixture of silver nitrate 
and cupric oxide, from which the former may be dissolved out by 
water. ‘The inconyeniences of this process are the time it takes, 
and the difficulty of ascertaining when the cupric nitrate is 
completely decomposed. To obviate these inconveniences, the 
author, after evaporating the solution of the mixed nitrates to 
dryness, redissolyes them in water, and precipitates the silver 
from the neutral solution by means of a clean spiral of copper 
foil. The precipitated silver is then redissolved in nitric acid, 
and the resulting nitrate is either crystallised, fused, or left in 
solution, according to the use to which it is to be applied.— 
[Ann, di Chim, app. alla Med., August 1869, p, 70.] 
A, SAy?zeEFF has discovered a new method of converting fatty 
acids into the corresponding alcohols, namely, by the action of 
dry sodium amalgam on a mixture of a fatty acid with the 
corresponding chloride; e.g. acetic acid and acetyl chloride 
yield ethyl alcohol. In this manner he has prepared ethyl, 
propyl, and butyl alcohol.—[Zeitschr, f. Chem. (2), v. 551.] 
Grune has found that the photographic image, as ordinarily 
produced, is on the surface, and not in the substance of the 
collodion film, By transferring the film to wood, and then dis- 
solving out the collodion by means of ether, a purely metallic 
image is left, admirably suited for the purposes of the engraver. 
PHYSICS 
Thalen’s New Map of the Spectra of the Metals 
M. Rosert THALEN has contributed to the Royal Society of 
Upsala an important memoir on the determination of the wave- 
lengths of the metallic lines of the spectrum. Dissatisfied with 
the pure results of refraction, as not being sufficiently refined to 
meet the requirements even of ordinary analytical accuracy, 
the author resolved to construct a new chart, based on the 
principle of wave-lengths. For the systematic examination of 
spectra, an electric source of light should always be employed, 
and entire groups of characteristic lines ought to be observed in 
all cases. The ordinary spectroscope, with a fine micrometer 
scale, gives readings which vary sensibly with the temperature 
and material of the refractive medium; and two such instruments 
cannot be compared with each other unless by specific tables, or 
graphically. Accordingly, the highest accuracy can only be 
attained by direct comparison with the dark lines of the solar 
spectrum, which themselves furnish an excellent micrometric 
scale. M, Thalén has therefore founded his experiments on the 
NATURE 
61 
laborious achievement of Angstrom, with whose “normal solar 
spectrum” he was early associated. 
The actual course of operations was as follows. Each bright 
metallic ray, whose spectrum it was desired to study, was 
laid down on the plates given by Kirchhoff and Hoffmann (a to G) 
° . ia 
or by Angstrom and Thalén (G to H); these rays were next 
fo} 
referred to Angstrém’s plates of the normal spectrum of the sun, 
unless a direct comparison with the solar lines could be made; 
and, lastly, the rays were drawn in the order of their wave- 
lengths as thus obtained, and sometimes with the assistance of a 
graphic method, on a map which accompanies the memoir. 
The instruments employed in this research consisted of a large 
Ruhmkorff induction coil, aided by a sufficiently powerful 
condenser ; and a voltaic battery of fifty pairs furnished the 
light for certain determinations. The spectroscope consisted of 
two tolerably large telescopes (one being used as a collimator) 
and a carbon disulphide prism of 60°. In favourable cases, 
two such prisms or six flint-glass prisms of 60° were employed ; 
but when the intensity was very feeble, only one (of the latter 
kind) could be used. 
The registration in the solar spectrum of the lines of in- 
candescent bodies may be effected by different methods. When 
the voltaic are is operated with, or even the induction spark 
(provided, in this case, that the electrode is made of the metal 
submitted to experiment), it is convenient to bring the rays from 
the two sources of light into the slit of the collimator in such a 
manner that the solar and metallic spectrum are one above the 
other. If the lines of the latter have sufficient intensity, the 
reference is effected without difficulty. Onthe other hand, when 
the intensity of the electric spectrum is feeble—which is generally 
the case when the spark is taken between electrodes moistened 
with saline solutions—it is better that the two pencils should 
enter the slit in the same direction, so as to be mutually super- 
posed. As the bright lines are now scarcely visible on the 
illuminated background of the solar spectrum, the latter must be 
temporarily excluded by a screen ; the vertical wire in the eye- 
piece of the telescope is made to coincide exactly with a bright 
metallic line ; and then, on re-introducing sunlight, its position 
among the dark lines is seen with precision. It is not unworthy 
of notice that the exactness of this observation is impaired by a 
somewhat singular circumstance. If the wire and the Fraunhofer 
lines are seen simultaneously in the focus of the eye-piece, the 
wire being placed among the weaker and narrower lines, it 
commonly happens that these entirely disappear, or can only 
be made out with difficulty. The great difference between the 
intensities of the two objects, and the diffraction fringes produced 
by the two sides of the wire, are, no doubt, the causes of this 
curious phenomenon. 
M. Thalén gives a table in which the normal spectrum of the 
sun is recorded in wave-lengths, and compared with the refraction 
spectrum of Kirchhoff. By its aid, the metallic lines on the 
chart accompanying the author’s paper may be identified with 
those of the refraction spectra alluded to, and an approximate 
value can be obtained of the wave-length corresponding to any 
line. The chart itself gives, in millimetres, the wave-lengths of 
metallic lines within about 0‘oooooor of their true value. It was 
drawn by hand on paper upon which the scale had already been 
printed without the usual damping process; in this manner all 
shrinking was avoided. It is rendered still more valuable by a 
long appendix of tables, in which all its numerical elements are 
appropriately distributed among the respective metals. Only 
the most intense lines, such as are obtained by the induction 
apparatus, have as a rule, been submitted to measurement. 
The following are the names of the metals whose lines coincide 
with those of the solar spectrum: sodium, calcium, magnesium, 
iron, manganese, chromium, nickel, cobalt, and #tanium. The 
chart contains lines belonging to forty-five metals. Iridium, 
rhodium, ruthemium, tantalum, and niobium were examined, but 
without any definite result. The spectrum of air is given at the 
bottom of the chart, for the sake of reference, and some integers, 
roughly representing the intensity of the lines. 
Some of the lines which show very strongly with metallic 
electrodes become very weak, when a saline solution is taken, 
and the more so as this is diluted. Two large and well-marked 
groups belonging to zinc and cadmium appear only when the 
metal itself forms the electrode, not the slightest trace of them 
appearing with a saline solution. 
In a concluding note, M, Thalén points out the probable 
existence of titanium in the sun, Titanic oxide only gave feeble 
