472 
^IR. J. EVERSIIED ON THE SOLAR ECLIPSE OF 1900, MAY 28. 
Tills structure suggests that the chromosjihere is in reality a region of innumerahle 
small eruptions of the same nature as the jets of highly luminous gas which are 
constantly to he seen with the spectroscope in all regions of the sun’s limb. It is 
})i’ohahle, indeed, that these jets, and the larger eruptive prominences, are in reality 
only the more pronounced manifestations of a phenomenon occurring on a smaller 
scale everywhere over the solar surflice. 
The higlily-heated gases composing these eruptions, which may be supposed to 
originate below the photospheric level, would lose heat as they ascended by adiabatic 
exjiansion and by radiation, and at a certain elevation would 2 )recl})itate the more 
refractory substances as highly luminous clouds, forming, in fact, the photosjiheric 
granules and the columnar filaments observed in sunspots. But the gaseous streams, 
dejuived of their condensable materials, would continue to ascend above the photo¬ 
sphere, finally becoming diffused in the region of the chromosphere. The expanded 
gases, subsequently subsiding in a relatively cooled condition, would form a strongly 
absorbing atmosphere settling down uniformly and slowly upon the photosphere and 
throimh which the ascending streams would be forced. 
If this really represents roughly the actual state of things, it is clear that the 
temperature conditions repinsented by the electric spark and by the arc may both 
exist at the same altitude above the pliotosphere, the spark condition in the highly- 
heated ascending gases and the arc condition in the cooler descending gases. 
Seen at the sun’s limb, as under the conditions of a total eclipse, the more intense 
spectrum of the ascending gases would be neutralised to a considerable extent by the 
absorption of the cooler gases in which the jets would be immersed, and through 
which for immense distances the line of sight must pass. But just those particular 
rays which are characteristic of the high temperature spectrum would not suffer 
absorption to nearly the same extent, consequently these rays (the enhanced spark 
lines) would stand out conspicuously in a spectrum which in its main features would 
he the emission spectrum of the cooler descending gases, i.e., the reversed Fraunhofer 
spectrum. 
The relatively cool gases would obviously determine the character of the absorption 
spectrum of the disk, and the only effect of the hotter eruptions, supposing them to 
be too small to be individually distinguishable in the spectroscope, would he to 
produce a faint emission line of about the same intensity as the background of 
continuous spectrum, and tending to diminish the intensity and width of all the dark 
lines, particularly the enhanced spark lines. 
In this way, by assuming the presence of innumerable eruptions of hot gas and 
cooler but quietly descending absorbing gases, the abnormal intensity of the 
enhanced lines in the flash can be simply explained without abandoning the view that 
the flash spectrum is really the reversed Fraunhofer spectrum, and that the entire 
depth of the flash region, and, indeed, of the chromosphere itself, is effective in 
producing the absorption lines. 
