June 13, 1912] 
of git. 
Houlae phenomenon of this summer than to direct the 
attention of meteorologists to a new point of view. 
In so far I succeeded, for a series of letters in NATURE 
devoted attention to this point. The fullest treat- 
ment was contained in a letter from Mr. L. G. Schultz 
in the issue of March 14. 
I should like to reply briefly to this letter, which, 
owing to university holidays, I have only lately seen. 
According to my views, his interesting observation, 
that for both the middle and end of the year 1911 the 
state of the weather in South America was diametric- 
ally opposite to that in Europe (extremely dry summer 
in the north with rainy winter in the south, and 
extremely dry summer in the south with rainy winter 
in the north) does not contradict my attempt at ex- 
planation, but rather proves its correctness. 
With normal ultra-violet radiation from the sun, 
i.e. with normal production of condensation nuclei, 
the water vapour formed in the north or south hemi- | 
sphere will condense again on the same hemisphere 
if the necessary conditions are brought about by cool- 
ing and alterations of pressure. With abnormally 
small production of nuclei rain will probably not cease | 
all over the earth, as Mr. Schultz seems to conclude, 
for the evaporated water must come down somewhere | 
or other, but the occurrence of condensation will be 
rendered more difficult. Consequently it is possible 
that the water evaporated on the summer half of the 
earth will first find the required preliminary condi- 
tions for condensation on the colder winter half, and 
so come down there. 
In other words, if the abnormal weather of 1911 
was conditioned by the decrease of ultra-violet radia- 
tion from the sun, then the abnormal dryness on the 
summer hemisphere had to be accompanied by 
abnormal rainfall on the winter hemisphere. This is 
exactly what Mr. Schultz has shown beyond doubt 
occurred not only for the period of the northern 
summer, but also for the period of the southern 
summer. Accordingly, the period of abnormally low 
ultra-violet radiation of the sun extended over the | 
whole of the year torr. Cart RAMSAUER. 
Radiologisches Institut, Heidelberg, May 30. 
Alleged Ultra-violet Rays from Filament Lamps. 
In the note referring to the proposed electric light- 
ing in the House of Commons by metal filament 
lamps, in Nature of June 6 (p. 352), it is stated that 
“The present proposal is to use metallic filament 
lamps enclosed in holophane globes behind amber- 
coloured glass to cut off completely all ultra-violet 
rays.’ Investigations made in America and in Ger- 
many show that the ultra-violet rays from such lamps 
are insignificant, and are far less than in daylight 
giving the same illumination, and probably less than 
with some kinds of incandescent gas mantles. 
Both physicists and electrical engineers would be 
interested to learn if there is any foundation for the 
allegation that metal filament lamps emit any appre- 
ciable ultra-violet rays or any rays which are injurious 
to eyesight. A. P. TRrorter. 
June 8, 1912. 
Earthquake of May 23. 
REFERRING to the second paragraph of Fr. Sid- 
greaves’s letter (NaTURE, June 6), I think that in the 
reading of seismograms it is often very difficult to 
My intention was less to explain the par- | 
NATURE 377 
of the epicentre by means of the time elapsing 
between the arrival of the first and of the second 
phase. On May 23, by this method, both horizontal 
| booms here gave the origin at 73.8° (8200 km., 
determine which is the first long wave from a distant | 
earthquake. 
ful enough to give a definite impetus to the seismo- 
graph at the inception of each of the two preliminary 
In the case of shocks which are power- | 
roughly), which would not be far from Burmah. 
F. Epwarp Norris. 
Woodbridge Hill, Guildford, June 8. 
SOLAR HALOS AND MOCK SUNS. 
Gj eeee have been recently many observations 
of optical phenomena in the atmosphere 
which can usually be identified with the halos of 
22° and 46° radius or with the allied and com- 
plementary arcs and mock suns. A brief descrip- 
tion of the principal phenomena which can be 
attributed to reflection and refraction of the sun- 
light by ice-crystals may therefore be of interest. 
Full accounts of such phenomena and of the 
theoretical explanation of their production are 
given in the classical memoir on halos by Bravais, 
in the third volume of Mascart’s “Optics ” and in 
the third part of Pernter’s ‘Meteorological 
Optics.” 
Ice-crystals are mainly hexagonal, and may be 
divided into two main classes, plates or stars with 
short axes and needles or prisms with long axes. 
The resulting optical prisms have angles of 60° 
or go° for the most part. The 22° halo is formed 
| by light which has passed through those prisms of 
60°, the right cross secticns of which pass 
through the sun. The prisms must be in 
the position in which the rays to the sun and 
to the observer make equal angles with the faces, 
and this is possible, for yellow light, only for 
prisms on a cone of 21° 50’ angular radius. The 
46° halo is produced in a similar way by prisms 
of 90°. 
If there is a preponderance of crystals floating 
with faces vertical, the reflection of light from 
these faces will give rise to a horizontal circle of 
light passing through the sun; and at points 
on this circle where the light is reinforced by re- 
fracted light, there will be unusual brilliance or 
mock suns. Hence the name mock sun ring. 
Two of the mock suns are formed by light re- 
fracted through prisms of 60°. They are at 22° 
from the sun when it is on the horizon, and their 
distance increases with the altitude of the sun. 
The mock suns produced by prisms of 90° are 
similarly at 46° or more from the sun, according 
te its altitude. 
The arcs of contact or tangent-arcs of the 22° 
halo are produced by refraction through prisms 
of 60° floating with their axes horizontal. If the 
sun’s altitude is less than 29°, the upper and 
lower arcs are distinct, but for greater altitudes 
they are joined to form the elliptic halo circum- 
scribing the ordinary 22° halo. 
The arcs of contact of the 46° halo are formed, 
according to Galle and Pernter, by refraction 
through crystals with vertical axes oscillating 
about the equilibrium position; according to 
Bravais and Mascart by simple refraction through 
crystals with one face horizontal. In the former 
ease the arc is not part of a circle, but always 
phases, it would seem easier to determine the distance | touches the 46° halo; in the latter it forms part of 
NO. 2224, VOL. 89] 
