DECEMBER 13, 1906] 
INRA OLE: 
1oB 
hardly fail to discover some. important fact about the 
interior structure of a molecule.’’ I think this statement 
remains as true now as it was thirty-two years ago. There 
can be no doubt, I think, that spectrum analysis, and 
especially the magnetisation of the spectral lines, will give 
us a clue to-the inner structure of the atom. I hope that 
I have succeeded in imparting to you this, my conviction. 
THE ERUPTION OF VESUVIUS IN APRIL, 
1906. 
HE most complete published account of the eruption 
of Vesuvius in April last is due to the enlightened 
liberality of the French Government, which commissioned 
Prof. Lacroix to study and report upon the eruption, and 
it is gratifying to find that this, as all other detailed 
accounts by qualified scientific observers of the eruption of 
Vesuvius, confirms in every respect the description which 
we were able to disentangle from contemporary newspaper 
reports and publish in our issues of April 12 and 109. 
As a result of Prof. Lacroix’s researches he has, in addi- 
tion to more detailed memoirs published or to come, com- 
municated to the Revue générale des Sciences of October 30 
and November 15 an interesting account of the result of 
his observations and deductions, some of which are suffici- 
ently interesting to deserve notice, in extension of what we 
have already published. 
The earlier stage of the eruption was of the Strombolian 
type, that is to say, the material ejected from the crater 
was formed by the breaking up of molten lava; it was 
consequently red hot, and Prof. Mercalli, who was watch- 
ing the eruption from Torre Annunziata, noticed that the 
mountain became covered, for from 200 metres to 300 
metres from its summit, with a continuous sheet of glowing 
material, from which blocks incessantly rolled down to 
lower levels. At oh. 31m. and again at 2h. 4om. a.m. on 
April 8 violent earthquakes were felt, corresponding 
to the most violent paroxysms of the eruption, accom- 
panied by a lowering of the height of the cone and a 
change from the Strombolian to the Vulcanian type of 
eruption. From this time onward the ejected material was 
less and less composed of fresh lava, and less and less 
incandescent, being composed, in increasing degree, of the 
old solidified lavas and tuffs of the cone. 
For several days after April 8 the summit was hidden 
by a thick cloud of ashes, and when this cleared away the 
mountain was found to have changed its form, from a 
pointed to a truncated cone, like that left after the erup- 
tion of 1822, though not so low or with so large a crater. 
When it became possible to ascend the cone it was found 
that the new crater was a true caldera, almost circular, of 
640 metres to 650 metres in diameter, surrounded by walls 
almost vertical, except at the top, where a steep talus 
reached up to the crest, and at the bottom, where a funnel- 
shaped talus sloped down into a cloud of vapour escaping 
from the fumeroles. The rim was irregular in height and 
generally sharp-crested, but cut by a deep gap on the 
north-east, where, for some 80 metres, the crest was not 
only lower, but comparatively flat-topped; this gap faces 
the crest of Somma in the direction of Ottajano, where 
scoriz and ashes fell in quantity sufficient to crush in 
the roofs of houses, while the observatory, less than half 
-as far from the crater in the opposite direction, received 
but a very small quantity of these same ejections. Prof. 
Lacroix rejects the explanation that this difference was 
solely due to wind, and considers that he has established 
a case of oblique eruption, the average direction of pro- 
jection being, not vertical, but inclined at a considerable 
angle towards the north-east. 
The greater part of the material blown out from the 
crater fell on the slopes of the cone, which was covered 
many yards deep with a loose deposit of fine dust, ashes, 
and blocks of all sizes. Even before the eruption ceased 
the surface of this deposit began to be broken by dry 
avalanches, which crashed down on every side, leaving 
the cone deeply scored by a series of radiating valleys, 
separated by steep-sided, sharp-crested ridges. Later on, 
rain-water sinking into and saturating these loose deposits 
set them in motion as the well-known mud lavas, the 
NO. 1937 VOL. 75 
mode of flow of which resembles closely that of the molten 
lava, and still later the rain-water, flowing off the surface, 
formed torrents of more liquid mud, which cut through the 
earlier accumulations of the dry avalanches and mud 
lavas. 
The eruption was accompanied by a change in level of 
the land, but this was confined to the immediate neighbour- 
hood of the volcano, for the tide-gauge shows that there 
was no alteration in the relative level of land and sea at 
Naples, while Prof. Mercalli found an elevation of from 
30 cm. to 48 cm. between Portici and Vico Equense. 
Of mineralogical interest is the new mineral, of which 
the first published description appeared in Nature of 
May 31, and the discovery of galena as an addition to the 
list of Vesuvian minerals. 
RUSSIAN OBSERVATIONS OF THE SOLAR 
ECLIPSE, AUGUST 30, 1905. 
(CONSIDERING “the unfavourable weather conditions 
experienced by many of the eclipse parties last year, 
the members of the Russian expedition, in charge of 
M. A. Hansky, are to be congratulated on the results they 
obtained, which have been recently circulated as a publi- 
cation of the Pulkowa Observatory. The observers were 
stationed at Alcocebre, on the Mediterranean coast near 
Valencia. The central line of totality passed almost exactly 
through the station, and various local conveniences com- 
bined to render the choice of site favourable to efficient 
observation. On August 15 all the instruments were re- 
ceived in good condition, and after observations had been 
made for determining the azimuth of the sun’s rising 
point, the various pillars and stands for the apparatus 
were erected. 
Photographs of the corona were taken on two scales :— 
small pictures with the Bredikhine double photographic 
telescope, furnished with a Zeiss objective of 170 mm. 
aperture and 800 mm. focal length, giving a field of 12°-4 
in R.A. and 8°-8 in declination; large pictures, for the 
delineation of fine detail in the coronal streamers, with 
an objective of 5 imches aperture and 13-28 m. focal 
length, the light being supplied from a ccelostat 30 cm. in 
diameter. Spectroscopic observations of the corona and 
prominences were made with a direct-vision spectroscope 
without slit, and the polariscopic phenomena examined by 
the aid of a Savart polariscope. Measurements of the solar 
radiation were taken with an actinometer and actinograph 
of Crova’s pattern. 
Near the time of eclipse the sky became clouded 
over, but about a minute before totality the sun was seen 
in clear sky. The corona was seen five or six seconds 
before totality, and the last ray of sunlight was visible for 
some two seconds, probably through a deep valley in the 
moon’s limb. This feature is also shown very clearly in the 
photograph of the chromosphere accompanying the report, 
which is divided up into a series of bead-like masses at 
that particular place. Visually the corona was seen of a 
brilliant, silver-white colour, its brightness increasing 
rapidly towards the moon’s limb. The longest rays seen 
extended about one and a half lunar diameters, and were 
situated near the poles of the sun, one at the north and 
two very fine ones at the south pole. The sky had a green 
colour, similar to that often seen about half an hour 
before sunrise. Careful examination of the spectrum of 
the corona during one of the forty-seconds’ exposures 
showed that the continuous spectrum was especially strong 
in the green, yellow, and red, the latter region being so 
brilliant that it suggested the possibility of photographing 
the corona in ordinary daylight by means of suitably pre- 
pared colour screens. 
With the polariscope the coronal light was seen to be 
strongly polarised, and the conditions were such that the 
dark bands were not visible on the sky surrounding the 
corona. The bands were much stronger when tangential 
to the sun’s limb than when radial. There appeared to 
be a rotation of about 23° of the plane of polarisation, 
which may possibly be ascribed to the action of a magnetic 
field round the sun. 
Eight photographs of the corona were obtained with 
the long-focus telescope, the exposures varying from 40-45 
