configurations of the two gases are very similar. 
i= all _~) 
: June 23, 1923] 
NATURE 
859 

Journal (vol. 17, No. 1, 1923). She shows, first, that 
ormation of vitamin-A, 
ht is necessary for the 
though neither chlorophyll, carbon dioxide nor 
oxygen need be present. It can also be formed in the 
almost complete absence of calcium. A further clue 
was given by the apparent close association of the 
lipochrome pigments (carotene, etc.) with vitamin-A in | 
various articles of diet. 
that the two substances might be identical. 
It had indeed been suggested 
pigment, carotene, is well known as giving the colour 
to carrots. 
or parts of flowers, exposed to light, if they contained 
carotene, also contained vitamin-A, and that absence 
of the pigment meant absence of the vitamin. But 
both may be present in tissues not exposed to light, 
as in the root of the carrot. Evidence is given that 
the vitamin has been transported to the root from 
the leaves. Although the investigator ventures no 
hypothesis on the matter, it seems highly probable, — 
from the necessity of both light and carotene for 
the production of vitamin-A, that the pigment acts 
as an optical sensitiser, similar to chlorophyll for 
the formation of formaldehyde. It would be of 
interest to know whether the rays absorbed by 
carotene are the most effective. 
STRUCTURE OF CARBON MoNnoxIDE AND NITROGEN. 
—In an interesting paper in the Proceedings of the 
Physico-Mathematical Society of Japan for April, 
H. Nagaoka discusses the band spectra of nitrogen 
and carbon monoxide. He starts with the assumption 
put forward by Langmuir, that the external de 
band spectra, which are presumably due to the exter- 
nal electrons of the molecule, should therefore be in 
close agreement. This is shown to be the case, with 
small differences indicating slight peculiarities of 
structure. The author then remarks that the ratio 
of the specific heats of the two gases cannot be 
accounted for on the assumption of Langmuir that 
the two nuclei are in the same cube (a difficulty 
pointed out by Partington in 1921), and he therefore 
proposes another model for the gases, in which two 
cubes are joined at an edge. This would give the 
correct value of 1-40. The two connecting electrons 
in the edge are pulled together by the nuclei, so that 
the resulting external electronic arrangement is that 
of two tapering six-faced figures connected by a 
narrow neck. ‘The author points out that the values 
of the ratio of the specific heats can serve as a useful 
criterion in differentiating between possible and 
impossible electron configurations. 
PHOTO-ELECTRIC CONDUCTIVITY OF CRYSTALS.— 
A number of contributions to our knowledge of 
this subject have been made during the past three 
years by Drs. B. Gudden and B. Pohl, of the Uni- 
versity of Géttingen, in communications to the 
Zeitschrift fiir Physik and the Physikalische Zeit- 
schrift, and a short summary of these is given in the 
issue of Die Naturwissenschaften for May 11. They 
find that all crystals with high refractive indices 
possess this conductivity, and that if, when with- 
drawn from the influence of light, they are insulators, 
when exposed to it they show an initial conductivity 
which is relatively large and proportional to the 
energy of the incident light. When the wave-le 
of the light is altered, the quotient of the quantity 
of electricity transmitted divided by the energy of the 
light incident shows the usual maximum at the wave- 
length of greatest absorption, but when it is calcu- 
lated for the energy of the light absorbed, it continues 
to increase towards the longer waves and eventually 
NO. 2799, VOL. IIT] 
But this — 
was disproved by Drummond. The orange-yellow 
Dr. Coward found, however, that flowers, — 

becomes a linear function of the wave-length. Over 
this region the authors consider that the observations 
justify the conclusion that one quantum of light energy 
absorbed gives rise to one electron in the crystal. 
AMMONIUM SULPHIDES.—Although a solution of 
ammonium sulphide has been in use in the laboratory 
for many years, the anhydrous substances are not 
well known. The solid compounds NH,HS and 
(NH,).S were said to have been obtained by Bineau 
in 1838-39 by the interaction of gaseous ammonia 
and hydrogen sulphide in the required proportions 
by volume, but doubt was thrown on the formation 
of the second compound by experiments of Bloxam 
in 1895. The matter has been reinvestigated by 
Thomas and Riding, whose experiments are described 
in the May issue of the Journal of the Chemical 
Society. Anhydrous NH,HS is best prepared by 
alternately passing ammonia and hydrogen sulphide 
into dry ether. Attempts to prepare (NH,).S were 
not very successful. The prolonged action of 
ammonia on the hydrosulphide in ether produced no 
sulphide, but on the addition of alcohol a yellow oil 
separated, which on standing gave some transparent 
cubic crystals, believed to be (NH,).S. The research 
throws very little light on the formation of the 
latter substance, but the method of preparation of 
NH,HS is a convenient one. 
ABSORPTION SPECTRA AND ATOMIC STRUCTURE.— 
In the Comptes rendus of the Paris Academy of 
Sciences for April 23, M. Victor Henri derives from 
the study of the absorption spectra of a large number 
of substances, both in solution and in the state of 
vapour, some important conclusions bearing on 
Bohr’s theory of atomic structure. He shows that 
the absorption spectrum of a solution may be either 
one composed of narrow bands (10-30 A) disposed 
in regular series, or one of broad bands (200-500 A); 
in a few cases both types of bands are present, but 
the narrow ones then occur only in the less refrangible 
regions. When the vapour of the substance is examined, 
the narrow bands of the solution are replaced by fine 
lines, while the broad bands of the solution appear 
also in the vapour as unresolved bands. He explains 
the narrow bands by the theory of quanta, the 
molecule being supposed to possess a_ series of 
stationary states, of which the energy is determined 
by the movements of the electrons, atoms, and the 
molecule. He distinguishes four cases. When the 
molecule contains only a single double bond, such 
as C=C, C=O, C=N, N=O, the other atomic 
groups in the molecule being all saturated, only 
broad bands are afforded either by the solution or 
the vapour. When the molecule is as simple as 
possible, but contains two or more groups with 
double bonds, narrow bands are given by the solution 
and fine lines by the vapour, distributed in series 
conformably to the theory of quanta. When the 
two double bonds are removed by the introduction 
of CH,, the narrow bands run together and form 
wide bands both in the solution and the vapour. 
When the molecule is rendered more complex by the 
substitution of more and more complicated groups 
of atoms, the narrow bands of the solution enlarge 
and the fine lines of the vapour fuse together, so 
that eventually a complicated molecule shows only 
broad continuous bands. He therefore finally con- 
cludes that for molecules containing only one double 
bond, the first postulate of Bohr is inapplicable, 
only the second postulate being valid; whereas for 
molecules with two adjacent double bonds both 
postulates apply, the first being determined by the 
existence of an electric polarity in the molecule. 
