100 

Tue HarvarD COLLEGE OBSERVATORY REPORT.—The 
annual report of the director of the Astronomical Ob- 
servatory of Harvard College for the year ending 
September 30, 1914, indicates the completion of a large 
amount of work in both observation and publication. 
Prof. E. C. Pickering refers in the first place to the 
principal work of the observatory, namely, its publica- 
tions and the importance of issuing these as soon as 
possible to prevent loss by fire. Thus observations 
from 1892-1912 with the 15-in. equatorial, from 1888-98 
with the §-in. transit circle, and from 1898-1912 with 
the 12-in. meridian photometer are now printed, and 
the discussions are in progress. The director directs 
attention to the improvement in photographic processes 
resulting in the replacement of practically all visual 
work. The report then describes in more detail the 
work of the Henry Draper Memorial, the principal 
research of which is the New Draper Catalogue; last 
year Miss Cannon classified 60,386 spectra, making a 
total of 160,541. The activities of the Boyden depart- 
ment, the Bruce photographic telescope, and the Blue 
Hill Meteorology Observatory are briefly summarised, 
while among the many items mentioned under the 
heading ‘‘ Miscellaneous” the work of Prof. W. H. 
Pickering at the Mandeville Station, in Jamaica, is 
described, much time having been devoted to the study 
of the nlanet Mars during its recent opposition. 
Star CHARTS FOR METEOR OpBSERVERS.—One of the 
contributions to the January number of the Journal 
of the Royal Astronomical Society of Canada (vol. ix., 
No. 1, p. 7) is entitled ‘A Gnomonic Star Atlas,”” and 
contains a set of thirteen maps, prepared by Mr. Rey- 
nold K. Young, intended to facilitate the observa- 
tion of meteors and the plotting of their paths. The 
method of the projection of the map is such that 
great circles in the sky are equivalent to straight 
lines on the map, thus making the plotting of the 
meteor trails more easy. The maps are devoid of 
unnecessary detail. The positions of the stars are 
given for the epoch 1900 correct to within one-tenth of 
a degree, and all stars down to 5th magnitude and 
the brighter variables are included. A good margin 
of overlap has been allowed in each map, which 
should prove very useful. 

FLUCTUATIONS OF TEMPERATURE IN 
EUROPE AND AMERICA. 
R. H. ARCTOWSKI, in a paper published in 
vol. xxiv. of the Annals of the New York 
Academy of Sciences, considers the problem of varia- 
tion of temperature over the whole earth. After a 
brief statement of the general problem and_ the 
methods by which it may be attacked, he explains 
that he could not deal single-handed with the arrange- 
ment and discussion of the actual values even over 
the whole of the northern hemisphere, and confines 
himself to the detailed survey of the variations over 
North America and Europe. He also compares the 
results with one or two representative equatorial and 
southern hemisphere stations. From a study of the 
values at one of these, Arequipa, in Peru, he deduces 
that the temperature changes are partly of a short 
period of about fifty-five days, brachypleionian waves; 
partly of a long period of twenty years or so, macro- 
pleionian waves; and partly of an intermediate period 
of between one and two years, pleionian waves. 
In dealing with the longer periods the normal 
annual variation is eliminated by taking a series of 
means for twelve months beginning with each con- | 
secutive month of time. For the European stations 
he finds that the continental ones resemble Arequipa 
in having marked pleionian waves, while those sta- 
NO. 2369, VOL. 95| 
NATURE 

[MarcH 25, 1915 

tions near the Atlantic are characterised mainly by 
brachypleionian variations. 
In an earlier paper Mr. Arctowski dealt with the 
period 1891-1900, and he takes the mean values for 
this period as normals, and plots on maps the differ- 
ence from normal of the values during each year of 
the decade 1900-09. The areas where the differences 
are positive he cails thermopleions, and the areas of 
negative differences antipleions. He finds that certain 
years, in particular 1900, 1908, are characterised by 
thermopleionian areas, while others, such as 1904, 
1907, are years of antipleions. The most important 
cause of these differences is the variation of solar 
radiation, but there are also supplementary causes 
such as the presence of volcanic dust in large quan- 
tities, or exceptional ice conditions in the polar 
regions. 
Many of the maps which illustrate the results of the 
investigation are on a very small scale; the course of 
the thermopleions and thermomeions is obscured by 
the attempt to show relatively microscopical geograph- 
ical details. 
Mr. Arctowski finds it astonishing that after all the 
efforts which are made to organise and maintain 
meteorological stations all over the world, the actual 
results of the work are so inaccessible. Even for the 
area with which he dealt he could only get much of 
the data by writing personally to the directors of the 
different meteorological institutes. This is’ a défect 
which will be remedied when meteorologists of 
different countries undertake to contribute to a central 
bureau representative regional values based on a selec- 
tion of stations which can only be chosen satisfactorily 
by the local organisation. 
There is another defect which is almost ‘more 
serious, viz., the lack of continuity in ‘the records for 
individual stations due to changes of situation or 
instruments. For example, Mr. Arctowski finds: that 
the difference of temperature between Chicago and 
Milwaukee was nearly 4° F. in the decade 1873-82, 
while in the decade 1896-1909 the difference was only 
2° F. The change is almost certainly due to change 
of instrument or site, and as it is of the same order 
of magnitude as the changes with which he deals, it 
indicates the need for great caution. 
The difficulty of securing comparable continuous 
records is indeed one of the most serious problems 
with which organised meteorology has to deal. 
E. G. 

REFINING GOLD BY ELECTROLYSIS. 
pale problem presented by the necessity of refining 
gold was one for which a solution was sought 
at least as early as the time, about B.c. 700, when 
coins were first manufactured in the Western world. 
Apart from toughening or the removal of base metals, 
which was sufficiently cared for by the ancient pro- 
cess of cupellation, it is clear that some measure of 
success attended the efforts made to part gold and 
silver. Thus, some of the ancient Greek coins con- 
taining 997 or 998 per tooo of gold. The earliest 
parting process used was one of cementation, which 
was succeeded by the nitric acid process. At the 
present day chlorine is the predominant agent for 
parting gold from silver in Australia, electrolysis in 
America, and sulphuric acid in Europe. 
The electrolytic process was brought forward by 
Charles Watt, at Sydney, in 1863, and was first put 
into operation by Wohlwill at Hamburg in 1878 and 
by Tuttle at the Philadelphia Mint in 1902. In the 
gold chloride process the solution used in the bath 
1 Abstract of the presidential address delivered before the Institution of 
Mining and Metallurgy on March 18, by Sir T. K. Rose. . 
