May 15, 1902] 
NATURE 
53 
LEILERS TC) Siemon. 
(The Editor does not hold himself responsible for opinions ex- 
pressed by his correspondents. Neither can he undertake 
to return, or to correspond with the writers of, rejecied 
manuscripts intended for this or any other part of NATURE. 
No notice ts taken of anonymous communications. | 
Mont Pelée Eruption and Dust Falls. 
FALLs of dust are caused in two ways ; either the dust, as for 
instance Sahara sand, is transported by means of the lower air- 
currents over wide areas, or matter is ejected from volcanoes, 
thrown high up into the air and carried by the upper currents, 
falling eventually in places at great distances from the seat of dis- 
turbance. The eruption of Krakatdo is a good example of the 
latter case, while the dust fall that occurred last year in March 
and was recorded in northern Africa, southern and northern 
Europe is a good representation of the former kind of dust fall. 
The occurrence of such falls of dust is very interesting meteor- 
ologically, because they afford us means of increasing our know- 
ledge of the actual movements of the air-currents, both low 
down and high up in our atmosphere. Falls of dust originating 
from volcanic eruptions are perhaps of greater interest, because 
the dust in such cases is thrown to very great heights, and we 
are able to deduce the directions of the currents at this eleva- 
tion, there being no other method of doing so available. 
To those interested in the movements of these upper currents, 
the recent disastrous eruption of Mont Pelée in Martinique 
may afford us some valuable information on this subject. 
From the accounts of the eruption already published, it is found 
that it occurred on May 8, and we gather from the information 
supplied by the British schooner Ocean Traveller, and printed in 
the Zzes, that when the ship was about a mile off St. Pierre, 
the volcano on Mont Pelée exploded. That the eruption 
was on a stupendous scale is undoubted from the numerous 
descriptions already made public, and the report from the 
British steamer Zsé, which passed St. Pierre, that ‘‘ she was 
covered with ashes although five miles distant from the land, 
and from on board nothing could be seen owing to the im- 
penetrable darkness,” gives some idea of the result of the 
disturbance. Later reports have indicated a further spreading 
of the dust, the island of Dominica recording a fall of sand on 
its southern boundary. 
It is interesting for a moment to make a brief survey of the 
atmospheric circulation in the lower and upper reaches of the 
atmosphere in the region of the West Indies and to the north 
and south, and see whether we can trace out the probable path 
of the fine dust thrown into the air. 
An examination of the fine pressure charts published in Bar- 
tholomew’s ‘‘ Atlas of Meteorology” tells us that during the 
month of May the West Indies lie between, but a little to the 
west of, two high-pressure regions, the more northern one being 
situated in the Atlantic Ocean and that to the south over the centre 
of South America, the intervening belt being one of low pres- 
sure. We also know that the sun has a declination of about 17° 
north at this time, and as the island of Martinique is situated 
in latitude about 15° north, the sun therefore passes daily near 
the zenith of that place, or, in other words, the sun is exerting 
its greatest heating power. In consequence of this fact, the low- 
pressure belt has a maximum in this region, and a low-pressure 
area means that the air is rising from the earth’s surface into the 
higher regions, The two high-pressure areas already mentioned 
correspond to downcast shafts of air, ze. air moving from 
higher to lower regions. With our present knowledge of 
atmospheric circulation it seems most probable that the heated 
air, rising into the upper reaches of the atmosphere from the 
low-pressure region (which includes t eWest Indies), bifurcatesin 
a north-easterly direction in the northern hemisphere and in a 
south-easterly direction in the southern hemisphere. Since 
these currents of air must again reach the earth’s surface, where 
they fall they will give indications of high pressure, 7.e. indi- 
cations of descent of air. As the two high-pressure areas 
already mentioned lie in the correct positions and directions in 
relation to the West Indies, it seems very probable that these 
are the downcast shafts corresponding to the upcast shaft or 
low-pressure area. 
If the circulation above mentioned be correct, then, as the 
region of the volcanic eruption of Mont Pelée lies in this low- 
pressure area, some of the finest particles ejected to the upper 
reaches of the atmosphere might possibly be carried in these 
NO. 1698, VOL. 66] 
currents and begin to fall in these high-pressure areas. They 
may also, if the dust be thrown sufficiently high, reach that 
elevated current of air travelling from east to west and make a 
circuit of the earth, as was the case in the eruption of Krakatdo. 
The most favourable position in the northern hemisphere to 
observe this fall of dust, should there be such, would be probably 
in the middle of the Atlantic Ocean, and this could only be 
recorded by passing ships. Since, however, the descending 
air moves in a spiral manner and in the direction of the hands 
of a watch, some of this current reaches Britain as a south-west 
wind, and it will be interesting to see whether any fall be 
recorded. There seems little doubt, however, that, just as in 
the case of Krakatdo, a great fall of dust fell to the westward 
of the volcano, so we shall probably soon hear in this case of 
such records from Mexico and Central America. 
Further, the eruption of Krakatdo was responsible for the 
magnificent coloured sunsets that were observed nearly all over 
the world, and as these were due to the fine dust particles 
ejected from the voleano—particles at very great altitudes—so 
it is quite probable that similar effects will ensue from the 
eruption of Mont Pelee. 
It seems desirable, therefore, that information relating to the 
present eruption should be collected while facts are still in the 
memory of those who have observed them, and that a complete 
account be recorded similar to that published on the Krakatdo 
eruption. It is satisfactory to learn that already expeditions 
are about to be sent from the United States of America to 
investigate the scene of the eruption. 
WILLIAM J. S. LockYER. 
Symbol for Partial Differentiation. 
*) or (a) at 
dx) ax 
there was only one other independent variable y) as the 
differential coefficient when y was constant. I still keep to 
this symbol. Thus, if £ is a certain kind of thermal capacity, 
(3) or (2) or (*) are in my thermodynamic work 
v p dt) 
In my college days we used the symbol 
dt dt 
perfectly definite. The mathematicians have introduced the 
convenient symbol for a partial differential coefficient Ou 
ie 
and in much work there is no doubt about the meaning. But 
even in hydrodynamics there is trouble. In thermodynamics 
there is so much trouble wi th this symbol that I venture to ask 
for help. 
The German translator of one of my books uses the same 
"symbol = for each of the above quite different things. Baynes 
in his thermodynamics does the same, and so do all other 
writers ; it seems to me that everybody is doing this without 
thought. Are they writing for the average examination man 
who does not need to think, or for the real student? If the 
letter 0 is to be retained, would it not be possible to use 
oe or OF oy Oe in the above three cases? 
ee Gis Ge 
own students to use 0, and I speak in the interest of such men. 
For myself it does not much matter, as I mean to continue 
using the symbolism of my youth. JOHN PERRY. 
May 6. 
I encourage my 
The Pines of Western Asia. 
ON p. 15 of NATURE of May 1, it is stated that Herr Hugo 
Bretzl, in a thesis for his Doctorate at Strassburg, has shown 
that ‘‘the Greeks realised such facts as the absence of the pine 
in all the countries which intervene between Macedonia and 
India.” That this statement is erroneous is proved by the 
following facts relating to the distribution of Macedonian and 
other species of Pzzws in the countries alluded to. PP. Pinea, 
Anatolia and Syria; P. sydvestr¢s, Asia Minor, the Caucasus 
and Tauria; P. halepensis, Anatolia, Transcaucasia, Syria ; 
P. Brutia, Asia Minor, the Lebanon, N. Persia; P. Laricio, 
Asia Minor. To these should be added various species of Pzcea 
and Adzes, which the Greeks may have included under Penus. 
J. D. Hooker. 
THE pine referred to is Pzmus excetsa Wall., which forms a 
feature of the Macedonian Mountains and also of the Himalayas, 
