•28 



TILL 5lKL <- lURE OF THE NUCLEUS. 



with table 4, the latter therefore advance faster than those in the first part. After 

 exhaustion (reheating) the coronas follow each other in reversed order, but they 

 soon vanish. Wlu-n the nuclei are very dense (as when obtained from burning 

 sulphur) it may take ten or more e.vhaustions to clear the fogs befoi-e the colors 

 .•ii)pear proliably as a result in part of unevenness of nucleation. It is then difficult 

 to place the coronas in the second and third columns. The nuclei are not removed 

 in tliirtv exliaustions, though to determine the limit greater care must be given to 

 the tiltration than was liere thought necessary. The occurrence of new coronal se- 

 (piences (change of the gr, rd, etc., into the rd, gr, series), after so many exhaustions 

 is none the less striking. 



9. Axial colors. — The final exhibit to Vie preliminarily made is the contrast 

 between the color of the central field of the corona and the axial or flame color. 

 It has been stated that the interior ciicular field of the coronas is at first colored 

 and that this color is more and more invaded by diffuse white light growing from 

 the center outward. Only after many exhaustions does the normal coiona with a 

 permanently white central field and the regular distribution of colored annuli, 

 appeal'. The color of the central field always differs strongly from the axial color, 

 and this is one of the important observations of this chapter. Indeed, the follow- 

 ing tables show that the axial color is nearly complementary to the color of the 

 inner field of the corona. The two ai'e thus distinct as to the nature of their 

 ori"-in, and one is tempted to conclude that the colors ab.sorbed in the axial beam 

 are the ones which illuminate the central coronal patch. 



Obsei'vation is naturally very difficult. Both colors are fleeting. The axial 

 color is always strongly admixed with white light, which soon overpowers the 

 effect of absorption altogethei'. A long column is thus needed for strong axial 

 colors, even under favorable conditions (colors of the fii'st order). The globe, 

 therefore, which shows the coronas very well (admirably when piojected against a 

 black background), will not show the axial color favorably as they are necessarily 

 projected against the flame. I endeavored to overcome this difficulty by a variety 

 of devices, using polarized light, but without substantial results. 



TABl-E 3— PREI.IMIN.ARY COMPARISON OF CORONAL CENTER AND AXIAL 



COLOR. EXHAUSTION, 76-58 cm. 



From Exhaust 

 Table 4.! No. 



I 



2 

 3 

 4 



5-6 

 6-7 

 7-8 



9 

 9-10 



10 



3 

 4 



5 



6 



7 

 8 



9 

 10 



Coronal 

 Center. 



Fog 

 Fog 



Olive 



Yl-gr 



' Exhaustion, 76-53 cm. 



