608 
the sun-spot wave with lowest number (1879-89) with the 
time of least warmth. If the relative position of all five 
points does not exactly correspond in the solar and terrestrial 
curves, we should remember the large uncertainty 
necessarily attaching to sun-spot measurements. 
We might smooth these curves with averages of three, 
getting the thick line curves, which indicate general de- 
cline (in a wide'sense). Thus in the case of temperature, we 
find the grouped sun-spot waves I, 2, 3, associated with 
more warmth than 2, 3, 4, and the latter with more than 
3, 4) 5- 
It is known that the recent researches of Nordmann, ex- 
tending and confirming the work of Képpen thirty years 
ago, indicate a state of things in the tropics which is 
Sunspots Numbers. 
Mean Temperature Winter. 
Frost Days Winter (inverted). 
Days with Maximum Tempera- 
ture 70° or more. 
Mean Temperature Summer. 
Mean Temperature Year. 
essentially opposite to that in our region, that is, minimum 
of sun-spots is associated with much heat (relatively) and 
maximum with little heat. 
I may here be allowed to submit for criticism a specula- 
tion regarding northern regions. I know little of tempera- 
ture conditions in the Arctic regions and of ice in the 
Atlantic, and I suppose very little is known of these in 
their relation to the sun-spot cycle. Let us suppose, how- 
ever (and the supposition does not seem a wild one), that 
it is in the Arctic regions as in the tropics, that is, more 
heat about sun-spot minima than about maxima, and that 
the Arctic régime is further opposite to ours in showing 
a general rise of temperature (in a wide sense) since the 
‘forties. (I may remark that Nordmann’s data for the 
tropics seem to point to a gradual rise in the last thirty 
years.) What should we expect from this state of things? 
Would not the greater heat with minima cause more melt- 
ing and loosening of ice, a more open season, and more ice 
to be carried down into the Atlantic? This would have a 
cooling effect on the Gulf Stream, and our temperatures 
would correspondingly decline. Thus heat in the Arctic 
would mean cold to us, and a gradual rise of temperature 
in the far north would mean a gradual fall in western 
Europe. 
I have been told that a certain shrinking of the northern 
ice covering has been noticed in places in recent times, and 
I read lately, in connection with the voyage of the Discovery, 
that since Ross’s time the Antarctic ice pack has broken 
back some thirty miles. 
There may, however, be facts adverse to the above theory. 
Perhaps some of your readers may be able to throw light 
on the subject. ALEx. B. MacDowatt. 
NO. 1800, VOL. 69] 
NATURE 
[APRIL 28, 1904 
A NEW EPOCH IN SOLAR PHYSICS. 
] P to the year 1868 those rose coloured appendages 
round the solar limb, the prominences, could 
only be observed at the times of total eclipses of the 
sun. The ingenious method for watching these pheno- 
mena any time when the sun shines we owe to the 
labours of Lockyer and Janssen, and the striking of 
a medal by the French Government in honour of this 
important solar physical advance properly noted an 
epoch in this branch of astronomy. 
By this new device, which was spectroscopic, the 
positions, forms, structure and movements of the 
prominences that encircle the solar disc could be 
accurately watched and determined, and we owe 
a debt of gratitude to such men as Respighi, 
Tacchini, Ricco, Mascari and others for the great work 
they have accomplished in taking advantage of this 
new line of research by recording daily the state of 
the solar limb in respect to these appendages. It must 
not be forgotten that all this work has been accom- 
plished by eye observations alone. Sweeping round 
the solar limb and noting accurately the position, form, 
&c., of each prominence is not the work of a moment, 
even if the sky is clear, and it is astonishing what a 
great amount of valuable information has been 
gathered by this apparently sluggish method. When 
it is considered that one sweep of the spectroscopic slit 
round the solar limb only makes us acquainted with 
the prominences that exist in a very small section of the 
solar atmosphere and at only one particular moment of 
time, it was natural that early attempts were made not 
only to employ photography as a means of quickly 
recording these, but of devising, if possible, some 
method by which prominences on the solar disc itself 
could be also photographed. 
It is not the object of the present article to trace 
the history of the development of the instrument, the 
photospectroheliograph, which now affords a means of 
satisfying these and other unlooked for requirements, 
but to give an account of the latest form adopted and 
results obtained by Prof. George E. Hale, of the Yerkes 
Observatory, to whom belongs a large part of the 
credit of designing and constructing an instrument 
capable of giving most successful results. i 
It may nevertheless be mentioned that Janssen 
in 1869 conceived the first idea of the method; 
he was followed by Braun, of Kalocsa, in 1872, and by 
Lohse, of Potsdam, in 1880. In 1889 Hale commenced 
work in this direction, and after him came Evershed 
in England and Deslandres in Paris, who both de- 
signed and used instruments which gave excellent 
results. 
From time to time Prof. Hale has published accounts 
of the design of, and work accomplished by, his former 
instruments, but in a recent publication! he gives us 
a very full and detailed description not only of the 
latest form he has adopted, but of the magnificent 
photographs which he has secured with it. 
To pass then at once to the modern photospectro- 
heliograph, reference may first be made to the principle 
involved. The feature of the instrument is that it is 
capable of giving us pictures of the sun in light of one 
wave-length, or in monochromatic light. The instru- 
ment itself differs little in principle from an ordinary 
spectroscope if the eye-piece be replaced by a (second) 
slit. If the solar image be thrown by means of a lens 
on the first slit, then after the solar light has passed 
through the lenses and prisms of the spectroscope it 
will fall on the second slit, which will only allow a 
narrow portion of the spectrum to pass through it 
1 Publications of the Yerkes Observatory, vol. iii. part i; also the 
Astrophysical Journal, vol. xix., No. 1, p. 41. 
