THE STKUCTUIIK OF THE NUCLEUS. 



81 



TABLE 7.— CORONAS IN PETROLEUM VAl'OR. PUNK NUCLEI. EXHAUSTIONS, 

 76-59 cm. DIAMETER OF DROPLETS, d IN cm. NEW FILTER. 



' Shaken. 



' Irregular. 



TABLE 8.— CORONAS IN PETROLEUM VAPOR. SULPHUR NUCLEI. EXHAUS- 

 TIONS, 76-59 cm. DIAMETER OF DROPLETS, d IN cm. NEW FILTER. 



' Successive coronas too faint and fleeting for measurement. 



A feature of these experiments is the result that mere exhaustion is ahnost 

 powerless to clear the vessel of nuclei. Thus, in table 6, 40 exhaustions do not 

 suffice (though the effect is doubtless counteracted by the shaking applied), so that 

 any imperfection iu the filter is at once fatal. Even air nuclei, in table 6, are apt 

 to persist indefinitely, unless long lapses of time intervene between the exhaustions. 

 Consequently, in tables 6, 7, and 8, the observations were repeated with a new 

 conical filter 40 cm. long and of compressed cotton. The coronas in this case are 

 somewhat smaller in diameter, showing fewer nuclei and larger particles, but their 

 persistence nevertheless remains a striking feature. 



16. Spontaneous mtclei, efc.— The nuclei vanish completely on the lapse of 



