470 
MR. J. EVERSHED ON THE SOLAR ECLIPSE OF 1900, MAA’ 28. 
are enhanced lines or not. If these are omitted, all the titanium lines abnormally 
strong in the flash, and all the iron lines excepting the three at XX 3856-5, 4325-9, and 
4404-8, are enhanced lines. 
If all the enhanced lines in the above-mentioned lists are considered, it is found 
that all the more strongly enhanced lines of iron and titanium coincide vith strong 
lines in the fla.sh (11 Fe lines and 21 Ti lines). But since many of these lines are of 
compound origin in the flash, it is not possible to say whether they are all of ahnormal 
intensity, ej/., 4351-9, 4549-6, 4556-1, 4629-6, and other.s. The (piartette of enhanced 
iron lines at 4508-5, 4515-5, 4520-4 and 4522-7 are all abnormally strong in the flash 
considered as Fe lines ordy, hut according to Bowland three f)f these are of 
com|)onnd origin, one including Ti. Ho^vever, it seems prohalfle that the ahnormal 
inten.sity of tliis gi-on}) is chiefly due to the fact that the lines are enhanced lines. 
'Ihere can he little doul)t from this impiiry that the enhanced lines do play a 
significant part in the flash spectrum, and the ahnormal intensities of these lines are 
not due to errors in the assignment of origin in Howland’s tables or to over-estimates 
of intensity in the flash. 
Of tlie abnormally weak lines a considerable number are probably the re.sult of 
under-estimates due to the close proximity of very strong lines of other elements. 
There remain a few, however, which cannot 1)6 thus explained ; among these 
particular attention may be called to the titaninm lines at XX 3753-00, 3924-67, 
4171-21, and 4306-08, all of intensity 4 in the solar spectrum, and the chromium line 
at X 4626-36. No satisfactory reason can at present he given for the weakness or 
absence of tliese lines in the flash s})ectrum. 
Notwithstanding these instances of disagreement between the intensities of the 
Fraunhofer lines of an element and their flash spectrum equivalents, the general 
agreement between the two spectra rs so sti'iking that it can scarcely he maintained 
that there is a fundamental difference in the conditions under which they are 
produced. The abnormally strong lines in the flash, whicli in so many cases are also 
lines which are enhanced in the spark, would, it is true, indicate that some of the 
radiatim'- cas at all events must he in a condition differing from that in the absorbing 
layer, and this, it must be acknowledged, is of great interest and importance, 
particularly in view of the flrct pointed ont by Fowler, that under some stellar 
conditions, e.g., in a Cygni, these particular lines constitute a separate and much 
simpler spectrum cjuite free from admixture with the ordinary arc lines.* 
But, as I hope to show in what follows, the prominence of these enhanced lines in 
the flash can he simply explained without abandoning the view that the flash region 
is really identical with the ahsorljing layer, and in the great majority of cases the 
flash lines are true reversals of the dark lines. 
In all photographs hitherto obtained at stations near the central line of eclipse, the 
flash spectrum must represent tlie more elevated region of the radiating gases, since 
* ‘ Observator}',’June, 1902. 
