343 
Dust from the Krakatoa Eruption of 1883. 
By Joseph Wharton, Philadelphia. 
The splendid roseate glows which in the winter of 1883-4 were 
visible in the western sky after sunset and in the eastern sky 
before sunrise, gave rise to many conjectures, but apparently to 
almost no experiments. A few persons believed those glows to be 
sunlight reflected from the under surface of a stratum of fine solid 
particles suspended at a great height in the atmosphere; some 
thought with me that those particles might be volcanic dust which 
had floated to us from the eruption at Krakatoa, but, as no one 
offered any proof of this, I attempted on the morning of January 
20, 1884, to demonstrate it. Six miles northward from the centre 
of Philadelphia, where I reside, a light and fine snow was then 
gently falling in an almost calm atmosphere, presumably from a 
high altitude. Of that snow, while it was yet falling, I collected 
about a gallon by skimming it carefully with my hands from a con- 
siderable surface in a field a hundred yards to windward of my 
house and a quarter-mile from the nearest windward building. 
This very clean new-fallen snow I melted under cover in the 
porcelain bowl it was gathered in, and was at first unable to detect 
any sediment; after maintaining for several minutes a gentle 
rotatory movement of the bowl in order to bring into its deepest 
part any solid matter which might be present, I poured off most of 
the water and evaporated the remainder. A minute quantity of 
fine dust was then discerned by the tiny vitreous reflections which 
it gave in the sunlight. My practice in chemical analysis, and 
therefore in weighing small quantities, affords some justification for 
the estimate that the total weight of this dust was less than one- 
hundredth of a grain. 
Under the microscope, where it was immediately placed, this 
dust showed the characteristics of volcanic glass ; it consisted in part 
of irregular, flattish, blobby fragments, mostly transparent and 
showing no trace of crystalline structure, in part of transparent fila 
ments more or less contorted, sometimes attached together in wisps, 
and mostly sprinkled with minute glass particles. The filaments of 
glass had about the same diameter as single filaments of silk placed 
on the microscope slide for comparison with them. 
