188 
NATURE 
[ Dec. 21, 1882 
hypotheses of the greatest stringency. The stratum B, being 
further removed from the photosphere than the stratum A, will 
be cooler, its lines therefore will be dimmer, and the lines of C 
will be dimmer than the lines of B, andso on. So if we could 
really observe the strata, the longer a line ts, 7.e., the greater the 
height at which the stratum which gives rise to it lies, ‘he dimmer 
the line will be. 
12. Now our best chance of making such an observation as 
this is during a total eclipse. We do not see the lines ordinarily 
in consequence of the illumination of our air. As during an 
eclipse before totality the intensity of this illumination is rapidly 
diminishing, the lines first visible should be short and brighr, 
and should remain short while the new lines which become 
visible as the darkness increases should be of gradually increas- 
ing length, so that the spectrum should become richer in the way 
indicated in Fig. 5. 
13. Further, the lines in 1 should be lines seen in prominences, 
and not in spots, and relatively brighter in the spark than in the 
arc, while the longer lines added in 2 and 3 should be lines 
affected in spots, and of in prominences. 
14. All these phenomena were predicted for the Egyptian 
eclipse a year before its occurrence, and were verified to the 
letter for the lines of iron over a purposely limited region. 
15. The actual observations of the iron lines made at Sohag 
are shown in the accompanying map, and these actual observa- 
tions are contrasted with the lines thickened in spots, the lines 
observed in the prominences by Tacchini, those intensified on 
passing from the are to the spark. The Fraunhofer lines are also 
given according to Angstrém and Vogel, and the iron spectrum 
of the are and spark according to Angstrom and Thalén. The 
observations during the eclipse were made 7 minutes, 3 minutes, 
and 2 minutes before totality as the air was gradually darkened, 
by which darkening, succe sive veils, as it were, were lified so 
that the more delicate phenomena could be successively seen. 
16. We begin with one short and brilliant line constantly seen 
in prominences, never seen in spots. Next, another line appears, 
also short and brilliant, constantly seen in prominences, and 
now, for the first time, a longer and thinner line appears, occa- 
sionally noted as widened in spots, while last of all we get very 
long, very delicate relatively, two lines constantly seen widened 
in spots, and another line not seen in the spark and neyer yet 
recorded as widened in the spots. 
17. The procession from the hot to the colder is apparent, and 
the simplicity of the spectrum as opposed to the Fraunhofer 
spectrum even yet, is eloquent of the gradual approximation 
which would be still possible if the darkness could be greater 
and our attack more complete. 
18. It will be noted over what an excessively small range the 
observations extend. We want similar observations over a wider 
range during future eclipses, and to do this work properly many 
observers armed with similar instruments must divide the whole 
or part of the solar spectrum amongst them, preferably that part 
between F and D which has been most closely watched in pro- 
minences and spots by Tacchini and myself. 
19. I next pass to another point on which an observation was 
made in Egypt. 
20. In Fig. 4 we considered the sun’s atmosphere, taking the 
sipplest case, that of one element; but evolution and the 
chemistry of our earth teach us that when the sun cools it will 
a very complex ma-s chemically. If the laws of evolution hold 
we need not expect that this will largely increave the complexity 
of the hottest layers A and B, but higher up, say at H—I, t 
complexity of chemical forms produced by evolution along t 
fittest lines will be very considerable. 
21. These strata H-—L may be taken to represent the corona. 
Its spectrum, therefore, should not be a continuous one, but 
should consist of an integration of all the radiations and atsorp- 
tions of these excessively complex layers. 
The spectrum of the corona, as I saw it in Egypt exactly 
answered to this description. Instead of the gradual smooth 
toning seen, say in the spectrum of the limelight, there were 
maxima and minima producing an appearance of ribbed structure, 
the lines of hydrogen and 1474 being, of course, over all. his 
observation, however, requires confirmation, for the look I had 
at the corona spectrum was instantaneous only. 
23. This observation should certainly be repeated during 
future eclipses with the proper ins:rumental conditions, +e. 
small intensely bright image on narrow slit and spectroscope of 
small disper-ion. I believe that, under these conditions, photo- 
graphs could readily be obtained with the new plates. 
ye 
ne 
1€ 
22 
| exhaustive. I 
24. Now an eclipse occurs next May at a critical time of the 
sun’s activity, for, so far as we can see, we shall be nearly at 
sun-spot maximum, and I hold that it will be a disgrace to our 
nineteenth century science, if efficient steps are not taken by 
those who are regarded as the leaders of science in this and 
other civilised countries to secure adequate observations, 
oc 
25. So far I have only referred to those special observations 
undertaken this year to discriminate between two rival hypo- 
theses, but both hypotheses may be wrong in many points, so 
that we must not limit ourselves to such observations, but collect 
facts over the whole field, as has always been the custom in 
eclipse expeditions. 
26. In my opinion the following scheme shows the observa- 
tions which, in the present state of our knowledge, it is most 
desirable to secure. The scheme, I am aware, is by no means 
give the observations in the order of im- 
portance I attach to them, having regard to the present posi- 
tion ef solar theory and the conditions of eclipse observations. 
(1.) 6-inch equatorial of long focus, perfect clockwork, spectro- 
scope with dispersion of at least five prisms of 60°. 
Clamp point of disappesrance of sun at base of normal slit, 
and record phenomena observed from ten minutes before totality 
to actual totality. 
a. Order in which lines appear. 
é, Brightness and length when first visible. 
The spectrum from A 4800 to A 5900 should be distributed 
among at least nine observers. 
Repeat observations after totality on point of reappearance. 
(2.) 6-inch photographic lens of 4-feet focus, perfect clock, 
same dispersion as above, 
Clamp point of disappearance of sun on centre of tangential 
slit and record phenomena observed from ten minutes before 
totality to actual totality. 
a, Order in which lines appear. 
6. Brightness and length when first visible. 
Repeat observations after totality on point of reappearance. 
Same part of spectrum, same distribution as in (1). 
(3.) 6-inch photographic lens as in (2). 
Photographic phenomena before and after totality on slowly 
ascending or descending or rotating plate, taking care to expose 
only narrow strip of plate. 
(4.) Ditto. Spectroscope of small dispersion, long slit. 
