bi 
SEPTEMBER 7, 1916] 
Griffith Taylor has, however, given some account of 
the initial experiments (Bulletin No. 13, Common- 
wealth Bureau of Meteorology). He points out that 
Melbourne is not suited for these experiments, since 
the prevailing northerly winds cause a large proportion 
of the balloons to be carried out to sea and lost, while 
those carried by the easterly upper-air movement to the 
Victorian mountains are seldom recovered. No con- 
clusive results can be deduced from the meagre data 
at present available, but after the war experiments 
are to be conducted under more favourable conditions 
at the meteorological observatory at Mount Stromlo, 
in Federal territory. 
' Dr. W. van BEMMELEN has published in English a 
very interesting account of the ** Results of Registering 
Balloon Ascents at Batavia’ (Batavia Javasche Bock- 
handel en Drukkerij, 1916). Batavia lies a few de- 
grees south of the equator, and these observations are 
valuable on account of the obvious care that has 
been taken to ensure accuracy and also on account of 
the equatorial situation. The ascents were 103 in all, 
spread over the six years 1910-15, and sixty-six avail- 
able records were obtained. Dr. van Bemmelen gives 
tables showing the temperature and humidity, and also 
discusses the annual and daily variations. The most 
striking result is the low temperature that is found at 
yreat heights in these equatorial regions. Eighteen 
ascents reached the stratosphere, the mean height of 
which is shown as just under 17 km. At sea-level the 
mean temperature is 26° C., the freezing point is reached 
at 4-7 km.; at 10 km. the temperature is —34° C., 
compared with —51° C. in England, but at 17 km. 
over Batavia the low value of —84° C. is found, 
against —54° C. over England. On one occasion a 
temperature of —go-2° C. (183° A. or —130° F.) was 
reached at 16-7 km. If any doubt remained about the 
existence of these low temperatures over the equator 
it has been removed by the publication of these results. 
The value given at 17 km., viz. —84° C., is based on 
twenty observations, and the standard deviation is 
small, so that there is no room for serious error in 
this value. 
Tue Journal of the Society of Siberian Engineers 
(Tomsk, January, 1916) devotes an article, illustrated 
by climatological charts, to the possibility of extending 
and developing the beet-sugar industry in Siberia. 
The desirability of establishing this industry in Siberia 
Was pointed out by the Russian Government thirty 
years ago, and substantial fiscal relief was offered to 
pioneers, with the result that the first factory was set 
up in 1889 in the Minusinsk district of the Govern- 
ment of Yenisei. The seat of the beet-sugar industry, 
the western provinces of Russia, being now in enemy 
occupation, there is among the refugees from those 
regions a large amount of highly skilled labour avail- 
able for employment elsewhere. The present moment 
is therefore opportune for directing attention to the 
subject and taking practical steps to foster a growing 
industry the develonment of which is of the greatest 
importance for the future of Siberia. 
Messrs. B. Artis and H. L. Maxwell have esti- 
mated the amounts of barium present in the leaves of 
certain tobaccos and trees, and publish the results in 
the Chemical News for August 11. The barium 
seems to be present in the ash of the leaf, partly as 
sulphate and partly in a form soluble in hydrochloric 
acid. The tobaccos examined were grown in Cuba, 
Pennsylvania, Connecticut, Sumatra, Wisconsin, and 
Mexico. The amount of barium (calculated as 
sulphate) found varied from 0-0132 per cent. to 0-0980 
per cent. in the leaf, the lowest amount being found 
in the tobacco from Mexico and the highest in that 
NO. 2445, VOL. 981 
NATURE 21 
from Pennsylvania. The stems invariably contained 
a larger amount., In the leaves of the trees examined 
the amount of barium sulphate found varied from 
0-0071 per cent. (Sumac) to o-0g41 per cent. (wild grape) 
in immature leaves. In the mature leaves the amount 
is generally greater than in the immature where com- 
parison is possible. 
AccorDING to a note by Mr. A. W. Knapp in the 
Chemical News for August 18, the pink colour fre- 
quently noticed by analysts to develop on the surface 
of margarine fat which has been exposed in the 
laboratory is not due to bacterial growth or to the 
action of light or oxygen. It is caused by the action 
of the vapour of mineral acids on a dye (probably 
dimethylamidoazobenzene) frequently present in mar- 
garine. A method of detecting the dye is described. 
A NEw system of signalling which dispenses with 
semaphores has been in use on a section of the Penn- 
sylvania Railroad for nearly eighteen months, and is 
described in the Engineer for September 1. Daylight 
lamp signals having a range of visibility of about 
2500 ft. in broad daylight had been obtained, using 
a lamp of not less than 20 watts and a lens 8 in. to 
10 in. diameter. In 1914 Dr. Churchill, of the Corn- 
ing Glass Company, discovered the possibility of secur- 
ing very long range from a small lamp arranged in the 
exact focal centre of a small wide-angle lens. Each 
fight unit consists of a box painted dull black on the 
inside, and containing a 12-volt 6-watt lamp with 
tungsten horizontal helical filament. The lamp is 
placed in the focus of a lens 5-5 in. diameter, having 
a focal length of 225 in. In front of the lens is a 
convex glass cover of the same diameter, and is so 
constructed as to avoid the difficulty of sun glare, 
which was at first experienced when a flatter cover- 
glass was used. A 4-in. spherical mirror is placed 
over the lamp, and is so arranged as to give the 
signal indication at extremely close range. A hood 
11 in. long is placed over the cover-glass in order to 
concentrate the lamp rays and to exclude the sun’s 
rays. The article contains clear illustrations of the 
arrangement. 
We are informed that Messrs. Macmillan and Co., 
Ltd., have become the sole agents for the sale at home 
and abroad of the publications of Messrs. W. and 
A. K. Johnston, Ltd., of Edinburgh. 
OUR ASTRONOMICAL COLUMN. 
MaxtMa oF Mirra CETI, 1915-—From observations made 
between November 21, 1914, and March 8, 1916, Mr. 
Felix de Roy has concluded that the dates of maxima of 
Mira Ceti were January 25, 1915 (mag. 3-8), and Decem- 
ber 20, 1915 (mag. 3-0), while the intervening minimum 
(mag. 8-7) occurred on August 22 (Mem. della Soc. 
degli Spett. Ital., vol. v., series 2, July). The first 
maximum of 1915 was the feeblest observed since 
1896, and its abnormal character has already been 
discussed by A. Bemporad (NaturE, vol. xcv., Pp. 405). 
The brightness at the minimum was also exceptional, 
having been equalled or exceeded only at_six of the 
forty-three minima which have been sufficiently re- 
corded. The second maximum of 1915 showed a 
normal amplitude of variation, and a normal interval 
from minimum to maximum, and a return of the star 
to normal conditions is suggested. A general dis- 
cussion of the ‘‘perturbations” indicates the prob- 
ability that the diminution and re-establishment of the 
brightness at maximum, of the amplitude, and of the 
interval from maximum ‘to minimum, are progressive, 
and pass through a minimum when the maximum is 
abnormal. The disturbances usually extend through 
