916 



SCIENGK 



[N. S. Vol. XII. No. 311. 



discovered and they were classed as un- 

 known lines. Comparison of these two 

 periods, with the sunspot curve as con- 

 structed from the measurements of the 

 mean spotted area for each year, indicated 

 that when that area was greatest the wid- 

 ened lines belonged to the unknown class, 

 but when it was least to the known class. 

 Now, in the laboratory it was possible to 

 differentiate between three stages of tem- 

 perature — that of the flame, of the electric 

 arc and of the electric spark of the highest 

 tension. At the lowest temperature — that 

 of the flame — a certain set of lines was ob- 

 tained ; as the temperature of the arc was 

 reached a new set was seen ; and at the 

 temperature of the high tension spark other 

 lines, called enhanced lines, were added, 

 while many of the arc lines waned in in- 

 tensity, At sunspot minimum, when the 

 known lines were most numerous, the lines 

 were almost invariably those seen most 

 prominently in the arc. But as the miiii- 

 mum passed towards the maximum the 

 unknown lines gradually obtained the pre- 

 dominance, and these were possibly ' en- 

 hanced ' lines, indicating the action of a 

 much higher temperature on known sub- 

 stances. It was therefore justifiable to 

 assume a great increase of temperature at 

 the sunspot maximum where these unknown 

 lines appeared. The curves of the known 

 and the unknown lines had been obtained 

 by determining for each quarter of a year 

 the percentage numbers of the two kinds, 

 and plotting them as ordinates with the 

 time elements as abscissse. But for the 

 purposes of the investigation, instead of 

 using the mean curves of all the known ele- 

 ments involved, that for iron alone was 

 emploj'ed, since that metal was a good rep- 

 resentative of the known elements, and had 

 been most fully studied. Such curves when 

 drawn crossed each other at points where 

 the percentage of unknown lines was in- 

 creasing and that of iron or known lines 



decreasing, and vice versa. There seemed) 

 therefore, to be three well-marked stages 

 of solar temperature. At the crossing 

 points, where the numbers of known and 

 unknown lines were about equal, a mean 

 condition might be assumed, a plus pulse or 

 condition of temperature being indicated 

 when the unknown lines reached their 

 maximum, and a minus pulse when the 

 known ones reached their maximum. The 

 curves obtained during the last 20 years 

 endorsed the conclusion that the unknown 

 lines curve varied directly and the iron- 

 lines curve inversely with the sunspot area. 

 The widened-line curves were quite differ- 

 ent from those furnished by the sunspots. 

 The crossings were sharply marked, and 

 since 1879 three of them had occurred, in- 

 dicating the presence of mean solar tem- 

 perature conditions in 1881, 1886-87, and 

 1892 ; another such crossing was antici- 

 pated in 1897, but has not as yet taken place. 

 Sunspots were indications of excess, not of 

 defect, of heat, and it was now known that 

 the spots at maximum were really full of 

 highly heated vapors produced by the 

 prominences, which were most numerous 

 when the solar atmosphere was most dis- 

 turbed. The Indian meteorologists had 

 abundantly proved that the increased radia- 

 tion from the sun on the upper air currents 

 at maximum was accompanied by a lower 

 temperature in the lower strata, and that 

 with this disturbance of the normal tem- 

 perature pressure changes must be expected. 

 Chambers was the first to show that large 

 spotted area was accompanied by low pres- 

 sures over the land surface of India. To 

 pass from the consideration of individual 

 spots to the zones of prominences with 

 which they were probably associated, it 

 was of the highest interest to note the solar 

 latitudes occupied when the crossings 

 already referred to took place, as in this 

 way were discovered the belts of promi- 

 nences which were really effective in pro- 



