enter my eye. And what I saw is shown, side by side with 
Respighi’s observations, in this diagram, in which I have 
arated the rings somewhat, so that there should be less con- 
fusion that in the actual observation. Here is Prof. Respighi’s 
first observation. He gets indications of C, D*, F, and the 
hydrogen line near G. He was observing the very lowest, 
brightest region of all, and therefore 1474 was obliterated by the 
brightness of the continuous spectrum ; but as the eclipse went 
‘on D3 was entirely obliterated, and afterwards he got C and F 
building up rings together with 1474, which was not represented 
n the lower regions of the prominence—not because it was not 
there, but because, as I have already insisted, of the extreme 
brilliancy of the background. Now my observation was made 
intermediately as it were between the two observations of Prof. 
 Respighi’s. Let me show the observations together, 
Respighi ... CD? F G Prominences at beginning 
of eclipse. 
Lockyer .... C 1474 F G Corona at 80 seconds from 
commencement. 
1474. F Corona at mid eclipse. 
a Respighi ... C 
Note that I had no object-glass to collect light, but that I had 
more prisms to disperse it ; so that with me the rings were not 
so high as those observed by Respighi, because I had not so 
much light to work with: but such as they were I saw them 
' better because the continuous spectrum was more dispersed, and 
because, with my dispersion, the rings—the images of the corona 
' —therefore did not so much overlap. Hence doubtless Respighi 
missed the violet ring which I saw, so faint, however, that both 
that and 1474 were almost invisible, while C shot out with mar- 
vellous brilliancy, and D* was absent. 
_ These observations thus tend to show, therefore, that instead 
f the element—the line of which corresponds with 1474—exist- 
ig alone just above the prominences, the hydrogen accompanies 
‘it to what may be termed a great height above the more 
“intensely heated lower levels of the chromosphere, including tae 
‘prominences in which the lower vapours are thrown a greater 
height. With a spectroscope of small dispersion attached to the 
Targest mirror of smallest focus which I could obtain in Eng- 
land, the gaseous nature of the spectrum, as indicated by its 
clure, that is, bands of light and darker intervals as distin- 
guished from a continuous spectrum properly so called, was also 
rendered evident. 
These are results of the highest importance, which alone 
are worth all the anxiety and labour connected with the expedi- 
tion. 
But there is more behind. 
_ The photographic operations (part of the expense of which 
was borne by Lord Lindsay) were most satisfactory, and the 
solar corona was photographed to a greater height than it was 
observed by the spectroscope, and with details which were not 
observed in the spectroscope. 
Mr. Davis was fortunate enough to take an admirable series 
‘of five photographs at Bekul, and Captain Hogg also obtained 
feme alae ; but I am sorry to say the latter lack somewhat 
in detail 
I have prepared two lamps, because I am anxious to exhibit 
the photographs two at a time, that you may compare one with 
the other. [This was done.] You see that so far as the camera 
goes—and mark this well—the corona was almost changeless 
during the whole period of totality; this is true, not only 
for ce. place, but for all the places at which it was photo- 
Z raphed. 
Bos exhibit two other photographs—one taken at Jaffna and 
the other at Ootacamund. Actinically the corona was the same 
d practically changeless at all the stations. You see that, 
ough not so obvious as in the other case, there is the same 
sin ilarity. 
Before I leave the actinic corona, I am anxious to show you 
an image of it, taken during the American eclipse of 1869 in a 
‘camera exposed to the sun during the whole of the totality ; to 
a certain extent in our recent photographs we have reproduced 
what was photographed in 1869. 
The solar nature of most, if not all, of the corona recorded 
on the plates is established by the fact that the plates, taken in 
different places, and both at the beginning and end of totality, 
closely resemble each other, and much of the exterior detailed 
“structure is a continuation of that observed in the inner portion 
independently determined by the spectroscope to belong to the 
sun 
7 
NATURE 3 An 
While both in the prism and the 6 inch equatorial the corona 
seemed to form pretty regular rings round the dark moon, of 
different heights, according to the amount of light utilised by 
the instrument, on the photographic plates, the corona, which, 
as I have before stated, exceeds the limits actually seen ia the 
instrument I have named, has a very irregular, somewhat stel- 
late outline, most marked breaks or rifts ({gnored by the spectro- 
Scope), occurring near the sun’s poles, a fact perhaps connected 
with the other fact that the most active and most brilliant pro- 
minences rarely occur there. 
From the photographs in which the corona is depicted 
actinically we pass to the drawings in which it is depicted 
visually. I would first call attention to two drawings made by 
Mr. Holiday, who formed part of the expedition, and in whose 
eye every one who knows him will have every confidence. 
First there is a drawing made at the commencement of the 
totality, and then a drawing made at the end. There is a won- 
derful difference between the drawings ; the corona is in them 
ba much more extensive than is represented actinically on our 
plates. 
Here is another drawing, made by Capt. Tupman, in which 
again we have something absolutely different from the photo- 
graphs and from Mr. Holiday’s sketches, inasmuch as we get an 
infinite number of dark lines extending down to the moon, and a 
greater extension than in the photographs, though in radial 
places the shape of the actinic corona and some of its details are 
shown, 
Now the corona, as it appeared to me with the naked eye, 
was nothing but an assemblage of bright and dark lines, it 
lacked all the structure of the photographs, and apppeared 
larger ; and [ have asked myself whether these lines do not in 
some way depend on the size of the telescope, or the absence of a 
telescope. It seems as if observations of the corona with the 
naked eye, or with a telescope of small power, may yive us such 
lines ; but that when we use a telescope of large power, it will 
give, close to the moon, the structure to which I have referred, 
and abolish the exterior structure altogether, leaving a ring round 
the dark body of the moon such as Prof. Respighi and myself 
saw in our prisms, and in the 6-inch telescope, in which the light 
was reduced by high magnification so as to bring the corona to a 
definite ring some five minutes high, while Prof. Respighi, using 
a 4-inch telescope and less magnifying power, brought the 
corona down to a ring something like 7 minutes high. 
And here we have an important connection between spectro- 
scopic and telescopic work. If we employ a telescope in which 
the light is small or is reduced by high magnification, we bring 
the corona to a definite ring, and perhaps here we have the 
origin of the ‘‘ring-formed” coronas. 
Many instances of changing rays, like those seen by Planta- 
mour in 1860, were recorded by observers in whom I have every 
confidence. One observer noted that the rays revolved and dis- 
appeared over the rifts. 
We have next to deal with the poiariscopic observations. 
Mr. Lewis, in sweeping round the corona at a distance of 6’ 
or 7’ from the sun’s limb, using a pair of compensating quartz 
wedges as an analyser, which remained parallel to itself while 
the telescope swept round, observed the bands gradually change 
in intensity, then disappear, bands of a complementary character 
afterwards appearing, thereby indicating radial polarisation. 
Dr. Thomson at Bekul saw strong traces of atmospheric, but 
none of radial polarisation, with a Savart. With the same class 
of instrument the result obtained by myself was precisely similar ; 
while on turning in the Biquartz, at the top and bottom of the 
image of the corona, zc., near the sun’s equator, faint traces of 
radial polarisation were perceptible for a short distance from the 
moon’s limb, Captain Tupman, who observed with the polari- 
scope after totality, announces strong radial polarisation extend- 
ing toa very considerable distance from the dark moon, 
Leaving the extreme outside of the corona as a question to be 
determined at some future time—and it can well wait—let us 
come to the base of the corona, and deal with the region to 
which I have already referred, close to the sun. 
What was the general conclusion at which we arrived on this 
important point? Before I state it, let me tell you the instru- 
mental conditions of the inquiry. We can use such a spectro- 
scope as the one with which you are all familiar, and so arrange 
matters that the slit shall be carried by a clock, so that it may 
follow accurately the edge of the moon ; but if the least varia- 
tion in the rate of motion takes place, the observation is ren- 
dered almost valueless, But if we employ a spectroscope, in 
