684 
DESTRUCTIVE DISTILLATION OF BonES.—Mr. E. V. 
Aleksejevski, in the Journal of the Russian Physical 
and Chemical Society, 1921, vol. 53, describes a 
research he has carried out, at the request of the 
Russian Government, on the dry distillation of large 
quantities of bones which have accumulated in the 
towns of the Tersk district since 1914. He finds that 
the quality of the bone charcoal obtained is better if 
horizontal retorts are used, instead of vertical ones. 
The ammoniacal liquor produced by distillation from 
such retorts contains more than twice as much 
ammonia as was usually obtained by the old method. 
The bone charcoal left in the retorts has a medium 
carbon content, and possesses a high degree of 
efficiency as a decolourising agent, for which purpose 
it is used in the beet sugar industry. It may with 
advantage be used as a contact catalyst, as, for 
example, in the direct synthesis of phosgene from 
carbon monoxide and chlorine, or in any other 
reaction of gaseous combination. Its catalytic power 
is found in a number of cases to compare very favour- 
ably with that of cocoanut-shell charcoal, which is 
considered to be the most efficient carbon containing 
contact catalyst. 
CORRELATION OF UPPER AIR WVARIABLES.—Mr. 
P. C. Mahalanobis contributes two Memoirs to the 
Indian Meteorological Department (Volume xxiv. 
Part ii.) entitled ‘‘ The Errors of Observation of Upper 
Air Relationships’’ and ‘“‘The Seat of Activity in 
the Upper Air.” He comes to the conclusion that 
Chapman’s corrections to W. H. Dines’s correlation 
coefficients are open to doubt. But he has fallen into 
error in stating that Douglas's coefficients are based on 
true heights. In a footnote in the Professional Notes 
of the Meteorological Office, No. 8, Douglas explains 
how he obtained his heights. He (Douglas), in the 
quotation given, merely meant that he did not use 
altimeter heights based on the erroneous supposition 
of a uniform temperature of 50° F. In the second 
Memoir Mr. Mahalanobis discusses the height at 
which the correlation coefficients between the five 
variables are numerically greatest and obtains a much 
lower value than 9 kilometres. However, he seems 
to have confused the T,, used by Dines, namely, the 
mean temperature between 1 and 9 kilometres, with 
the mean temperature between o to 9 kilometres, and 
this fully explains the discrepancies he finds. Leaving 
out the temperature of the first kilometre in forming 
the mean prevents the relationship between Py, Py, 
and T,, being a fixed one, whereas the relationships 
between the partial correlation coefficients given by 
Mr. Mahalanobis depend upon P,, P-, T, being ¢on- 
nected by a definite equation. If these three quantities 
be rigidly connected, the connexion is equivalent to 
reducing the independent variables from five to four, 
and as a matter of course the partial correlation 
coefficients involving the three related quantities 
must be 1 or-—1, and the second and third order 
partials must take the form found by Mr. Mahalanobis. 
DEVELOPMENT CENTRES IN THE PHOTOGRAPHIC 
PratE.—It is well established that photographic 
development starts at definite points or ‘‘ reduction 
centres ’’ in the individual grains of silver bromide. 
Silberstein favours the view that the corpuscular 
nature of light is the cause of this, while others regard 
these centres as pre-existing in the grains. The 
practical importance of the matter is that, if the 
latter is true, the emulsion maker may eventually 
be able to control the production and sensitiveness of 
these centres, and perhaps even to isolate them. Mr. 
Walter Clark, of the British Photographic Research 
NO. 2794, VOL. 111] 
NATURE 

[May 19, 1923 pte 
Association, gives some important results of his 
investigation of this question in the May number of 
the Journal of the Royal Photographic Society. 
He finds that a solution of sodium arsenite~has no 
measurable reducing action on silver bromide pro- 
duced by precipitation, and confirms the fact that 
a dilute solution of it applied t6 a plate renders the 
plate developable. This is evidence that there is in — 
the plate something besides simple pure silver — 
bromide, which is affected by sodium arsenite (as~ 
well as by light) to form development centres. By 
giving a plate a suitable exposure to light to render 
the centres developable and then treating the plate 
with chromic acid, the sensitiveness of the plate is’ 
reduced to a very low figure but always of the same — 
order of magnitude if the action is thorough (the 
preliminary exposure is necessary). It appears 
probable that the chromic acid dissolves the “ centres ”” 
produced by the exposure and that the low remaining 
sensitiveness is the sensitiveness of pure silver — 
bromide. 
Mass SPEcTRA.—In a communication which appears 
in the May issue of the Philosophical Magazine, Dr. 
F. W. Aston gives an account of his work with the 
mass spectrograph to the end of 1922. The general 
technique has been in the main unchanged, but softer 
rays from the discharge tube are being used, and the 
photographic plates have had some of the emulsion 
dissolved from them to concentrate the sensitive 
grains more highly. Helium, nickel, lead, zinc, xenon, 
tin, iron, cadmium, thallium, selenium, tellurium, 
beryllium, aluminium, and antimony have been 
tested, and the constitution of nickel, tin, iron, 
selenium, aluminium, and antimony determined for 
the first time. Two new isotopes of xenon have also 
been discovered. Tin and probably iron show devia- 
tions from the whole number rule on the oxygen 
scale, and with hydrogen give three exceptions to 
that rule. A complete table of elements and isotopes 
determined by any of the positive ray methods up 
to the present time is given. 
A Frencu O1L-WeEtL.—In the Comptes vendus of 
the Paris Academy of Sciences of March 19, M. Ph. 
Glangeaud gives a note of the oil-well of Crouelle, 
near Clermont-Ferrand, Puy-de-Déme, about which 
some paragraphs have recently appeared in the Press, — 
The well-log is an interesting one, particularly from 
the geological point of view, and much information 
has been obtained regarding the Oligocene facies of 
the district between the Puy de Crouelle and the 
better-known Puy de la Poix. The beds traversed 
seem to belong to the Upper Sannoisien and Lower 
and Middle Stampien stages of the system, and, 
according to M. Glangeaud, recall in many respects 
similar Oligocene beds at Pechelbronn ; further, the 
prevalence of abundant organic material and the 
conditions of sedimentation are cited as being dis- 
tinctly favourable circumstances to the formation and 
accumulation of petroleum. The well was carried 
to a depth of about 856 metres, but operations were 
subsequently interrupted by casing breaking at 787 
metres, which, with consequent water trouble, 
curtailed developments. Notwithstanding this, M. 
Glangeaud regards the results as being among the 
most important and encouraging yet achieved in this 
district, still an unknown factor as regards oil poten- 
tialities. Certainly the oil obtained from the well, 
both in quality and quantity, seems to augur well for- 
future developments in the area, though on general 
geological grounds one can scarcely be optimistic as 
to the possibilities of a large field being discovered in 
this region of France. 
~ 
