
June 15, 1871 | 
NATURE 
I2I 

The Eclipse Photographs 
Ir would have given me much pleasure to have shown Mr. 
Winstanley the original negatives of the photographs of the late 
eclipse of the sun if he had called on me to see them, and by so 
doing he would have avoided falling into the mistakes which his 
letter contains. 
At the time when the last photograph was taken the sky was 
perfectly clear, and unless Mr. Winstanley is in possession of ex- 
clusive information he has no right to assume that the American 
photograph was not taken under equally favourable conditions. 
Some of my photographs (which Mr. Winstanley cannot have 
seen) were taken through the edges of a cloud, the whole of which 
could be covered by the hand when held with the arm extended ; 
and there was a perfectly cloudless sky near the sun excepting 
towards the east. 
The imperfection in my No. 5 picture, which Mr. Winstanley’s 
experienced eye detects, arose from the shaking of the telescope, 
caused by the high wind blowing at the time. Probably a single 
gust during the eight seconds while the plate was exposed caused 
the mischief, and this defect would never have been seen but for 
the extremely actinic power of the red prominences which leave 
their impression on the sensitive plate instantaneously. The 
moon’s limb is perfectly sharp, excepting where the red promi- 
nences appear. 
Let it be clearly understood that this ‘‘ indifferent definition” 
refers to the moon’s limb only; the details of the corona 
do not appear to have suffered; after the gusts of wind the 
telescope has returned to its proper position, and Mr. Win- 
stanley must know from experience that the image of an object 
giving off feeble light would not be materially injured by a slight 
blow given to a firmly mounted camera. 
Mr. Winstanley says that ‘‘ the identity of the coronal rifts in 
the Cadiz and Syracuse photographs is not satisfactorily conclu- 
sive.” Assertion is not proof. In NATURE of March 9 I 
gave evidence which appeared to me to be conclusive (I need not 
here refer to the opinions of others who are equally satisfied), 
and up to the present time no counter-evidence has been pro- 
duced. 
It is not for me to defend the American photograph. In due 
time we shall know all about how that was produced. But has 
Mr. Winstanley failed to notice that the light on the moon’s disc 
does not extend all round and all over it as it would do if caused 
by our atmosphere? It is chiefly on the east and west sides. 
We may expect the explanation of this defect when we hear how 
it happens that the corona in this photograph is cut off instead of 
extending as in all the other photographs. 
I fail altogether to see the connection between the solar corona 
and a lunar halo—the phenomena bear no resemblance to each 
other. The solar corona comes close up to the perfectly black 
disc of the moon. I never sawa lunar halo close up to the 
moon’s limb. When seen through a mist or in a ‘* sky burdened 
with innumerable clouds,” there can be no doubt that the lunar 
surface is obscured by the moisture in our atmosphere. 
A, BROTHERS 
Ocean Currents 
Havine had occasion in the spring of 1868 to consider the 
subject of Ocean Currents as discussed by Captain Maury and Sir 
John Herschel, I was led to certain views respecting the origin 
of the oceanic circulation, which are briefly touched upon in a 
paper which appeared in the Stadent for July, 1868. At that 
time an experimental test of my theory (or rather of that portion 
of the theory I advocated, which was, as I judged, novel) occurred 
tome. The experiment might, I conceive, be very readily tried. 
It somewhat resembled that by which Dr. Carpenter illustrated 
lately at the Royal Institution his views respecting the influences 
of evaporation and polar cold ; but as I wished specially to show 
how the westwardly equatorial current came about, the experi- 
ment was somewhat more complex. Let the circumference of a 
large and shallow cylindrical basin represent the equator and the 
central part the north polar regions. Within this cylinder let 
solid matter be so placed as to represent the northern halves of 
the continent, in such sort that the resulting configuration would 
correspond to that of a map of the northern hemisphere (say on 
the equidistant projection). Let sea-water be poured in to repre- 
sent the northern portions of the terrestrial oceans. Now to re- 
present the Arctic ice-fields, let lumps of ice be placed at the 
centre of the cylindrical vessel (they should be circled round by 
a wire-guard) ; and to represent the effects of equatorial heat, let 


a stout iron ring round and above the rim of the cylindrical 
vessel be heated. In this state of things the process of circula- 
tion, which actually took place in Dr. Carpenter’s experiment, 
would take place after such modified sort as the contour of the 
continent masses permitted. Now suppose that the cylindrical 
vessel is set in steady and somewhat slow rotation about its axis. 
It is clear that on the currents flowing from the pole and pole- 
wards, effects will be produced which precisely resemble those 
due to our earth’s rotation. If I am right in regarding these 
effects as the true cause of the direction in which the equatorial 
currents, the Gulf Stream, and in fact all the currents in open 
ocean are observed to flow, abundant evidence to that effect will 
be obtained. If no such evidence be obtained, the westwardly 
direction of the equatorial currents must, I imagine, be ascribed 
to the trade winds, as Franklin and Sir J. Herschel have main- 
tained. 
In the summer of 1868 I suggested to Prof. Pepper that such 
a contrivance as the above, if it worked as I judged (and still 
judge) that it would, would form an interesting and instructive 
addition to the models exhibited at the Polytechnic Institution. 
Dr. Carpenter has already proved that the vertical circulation 
takes place in an experiment of this sort. If the eastwardly and 
westwardly circulation takes place as I expect, the expe:imental 
illustration of oceanic circulation would be singularly complete. 
The circulation in the southern hemisphere could be illustrated 
in like manner. 
I may note here that the vast distance separating the polar 
from the equatorial regions must not be overlooked in theories 
respecting oceanic circulation, The influence of arctic cold may 
be paramount in very high latitudes ; but equatorial evaporation 
must, it should seem, be the prime moving cause in tropical and 
sub-tropical regions. RIcHARD A, PROCTOR 
Brighton, June 6 
Day Auroras 
I HAVE read attentively the numerous letters which have ap- 
peared in your columns on this subject ; but so far as I can discern 
it seems not to have occurred to any of your correspondents that 
the auroral force, whatever it may be, affects every kind of cloud 
as well as the cirrus. On June 15, 1870, at 9 A.M., I witnessed 
here as complete a display of auroral motions in the cirrus cloud 
as ever I beheld in a midnight sky ; and from that date I dis- 
missed in my own mind all doubt as to the identity of auroral 
force, whether seen by day affecting the cirrus cloud or appearing 
as streams and rays of light at night. On Thursday and Friday 
last I witnessed a configuration of cirro-stratus cloud, evidently 
the result of magnetic polarisation, which I have no hesitation in 
characterising as auroral. There were on Thursday two poles, 
both in the line of the magnetic meridian; but on Friday 
night, at nine o’clock, only one pole in the direction 
of the true meridian. The phenomenon to which I 
refer is of very frequent occurrence, especially before a 
track of fine warm weather ; and without at present offering a 
theory on the subject of auroras, I venture to class polarisations 
of clouds, whether cirrus or not, as arising from the same cause 
as luminous aurora. The transverse or dia-magnetic lines are 
generally as well marked in cloud auroras, and it is an interesting 
task to watch the transformation of cirrus cloud from the meri- 
dional to the equatorial direction. I have also noted that when 
these auroral lines converge towards the magnetic pole, a steady 
barometer and fine weather ensue ; but that when the transverse 
or equatorial lines predominate and continue long visible, rain 
soon follows. These transverse lines of cloud are always lower 
and seem to be dia-magnetic, D. Low 
Burntisland, May 22 
Peruars you will allow me to add my mite to the dis- 
cussion which has been going on in your pages on the question of 
the visibility of the aurora in daylight. 
On the 3rd of September in last year, when at Nairn in the 
north of Scotland, I witnessed an aurora, such as I never heard 
or read of, or saw before; and strangely enough it was not 
noticed, as far as I am aware, in any of the newspapers. I had 
gone down to the beach at about 10 15 P.M., and immediately 
noticed what appeared at first to bea kind of haze over the whole 
sky, which slightly dimmed the light of the stars. 
For a few minutes I thought no more about it, but, happening 
to turn my eyes towards the zenith, there was a sight I never 
shall forget. A number of sheets of whitish light were con- 
