NovEMBER 10, 1899.] 
with great ease with the formation of calcium 
alloys, and, in a few cases, of the free metal. 
As this is the case with chromium, it may be 
possible to devise a commercial method for its 
manufacture by the use of calcium carbid. 
SoME years ago the sterilization of water by 
chlorid of lime (bleaching powder) was sug- 
gested by Traube, and the subject was further 
studied by Bassenge. In arecent number of the 
Hygienische Rundschau, A. Lode, of Innsbruck, 
describes further experiments along the same 
line. The process, as practically carried out, 
demands 0.15 gram of commercial, dry chlorid 
of lime per liter of water to be purified. This 
is rubbed with an equal weight of water in a 
porcelain dish, or on a large scale in a suitable 
wooden or stone vessel, to a thin paste, and 
added with constant agitation to the water. 
The corresponding amount of hydrochloric acid, 
for which the author gives a table, is then 
added. Inthe course of half an hour the water 
has cleared and 0.3 grams of sodium sulfite per 
liter is added. The cost of the process is found 
to be about 8 cents per cubic meter of water. 
It is claimed that by this process the water is 
completely sterilized, and even very bad waters 
rendered potable. 
Jo) Mig 1b 
THE NOVEMBER METEORS OF 1899. 
PROFESSOR EH. C. PICKERING has sent from 
the Harvard College Observatory the following 
account of the approaching meteoric shower: 
The predicted time of maximum of the No- 
vember meteors is November 15, 1899, at 18 h. 
Greenwich mean time. As a similar shower 
may not occur again for thirty years, no pains 
should be spared to secure the best possible ob- 
servations. The most useful observations that 
can be made by amateurs are those which will 
serve to determine the number of meteors vis- 
ible per hour throughout the entire duration of 
the shower. They should be made on Novem- 
ber 15th, and also on the two preceding and 
following evenings. The most important time 
for observation is from midnight until dawn, as 
comparatively few meteors are expected earlier. 
Observations are particularly needed at hours 
when they cannot be made at the observatories 
of Europe and America. In general, the time 
SCIENCE. 
697 
required for ten or more meteors to appear in 
the region covered by the accompanying map, 
should be recorded. This observation should 
be repeated every hour or half hour. If the 
meteors are too numerous to count all those 
appearing upon the map, the observer should 
confine his attention exclusively to some small 
region, such as that included between the stars 
yw Ursae Majoris, 40 Lyncis, J and a Leonis. If 
the meteors occur but seldom, one every five 
minutes, for instance, the time and class of each 
meteor should be recorded. Also note the time 
during which the sky was watched and no 
meteors seen, and the time during which that 
portion of the sky was obscured by clouds. 
Passing clouds or haze, during the time of ob- 
servation should also be recorded. The date 
should be the astronomical day, beginning at 
noon, that is, the date of early morning obser- 
vations should be that of the preceding even- 
ing. Specify what time is used, as Greenwich, 
standard, or local time. When a meteor bursts, 
make a second observation of its light and color, 
and when it leaves a trail, record the motion of 
the latter by charting the neighboring stars, 
and sketching its position among them at short 
intervals until it disappears, noting the time of 
each observation. If the path of a meteor is 
surely curved, record it carefully upon the map. 
On November 14, 1898, thirty-four photo- 
graphs were obtained of eleven different me- 
teors. Their discussion has led to results of 
unexpected value. The greatest number of 
meteors photographed by one instrument was 
five. Only two meteors were photographed 
which passed outside of the region covered by 
the map, although the total region covered was 
three or four times as great. No meteors 
fainter than the second magnitude were photo- 
graphed. 
Photographs may be taken, first, by leaving 
the camera at rest, when the image of the stars 
will trail over the plate and appear as lines, or 
secondly, attaching the camera to an equatorial 
telescope moved by clockwork, when a chart of 
the sky will be formed, in which the stars will 
appear as points. A rapid-rectilinear lens is to 
be preferred in the first case, a wide-angle lens 
in the second. The full aperture should be 
used and as large a plate as can be covered. 
