226 PHENOMENA (CONNECTED WITH CLOUDY CONDENSATION. 



appears that a tine form of condeiisatiou had taken phice, as Mr. W. R. 

 Manley states* that there was at the time a sort of haze all over the 

 sky, and from the letters of different observers this haze seems always 

 to have accompanied the great sun. 



One almost wonders that a blue or green sun is not oftener seen, as 

 there are often present all the materials necessary for producing these 

 colors in the atmosphere. On a few occasions I have observed the sun 

 to be of a silvery whiteness, when the vapor in the upper air was begin- 

 ning to condense, and the sky was covered with a thin filmy cloud. It 

 IS liowever possible that this slightly bluish tint may have been due 

 to the sun being seen more in its natural color than usualj thatis, made 

 much less yellow by our atmosphere than it generally is. There seems 

 to be something preventing the dust and the vapor in our atmosphere 

 acting under ordinary conditions in such a way as to color the sun blue 

 or green. Perhai)S it may be the tendency the particles have to dif- 

 ferentiation. This tendency, we have seen from the experiments, 

 rapidly destroys all color effects, and from this we might suppose it 

 would be impossible that the colors if produced by Avater drops could 

 remain in nature visible for so long a time as they did. But it must be 

 remembered that the particles in the experimental vessels are extremely 

 close together, and the vapor exchanges can therefore take place (juickly. 

 If however the drops Avere widely separated, the exchanges would 

 take place slowly. For instance, if the drops in 1 centimeter were 

 separated so as to form a column 1 mile long, with a section of 1 Sipiare 

 centimeter, we should have the same amount of color in 17 miles that 

 we had in the 17 centimeters of air in the flask, and the particles would 

 be so far apart that dififerentiation Avould then take place extremely 

 slowly. lUit further, if the supply of dust and vapor Avere constantly 

 kept up by the volcano, the color phenomena would continue for the 

 same reason that they continue in a steam jet, namely, by the tlrops 

 being constantly renewed. 



A neic instruinent for tcsthuj the amount of dust in the air. — As 

 this investigation progressed it became evident that these color phe- 

 nomena placed in our hands an easy and simple way of estimating, 

 in a rough way, the number of dust i)articles in the atmosphere of our 

 rooms, Avliich might be useful for sanitary i)urposes. An instrument 

 Avas therefore constructed to see how far the idea could be practically 

 carried out. This new instrument Ave intend to call a Koniscope. In 

 its present form this instrument consists of an air-pump and a metal 

 tube with glass ends, which we shall call the test-tube. The capacity 

 of the pump should be from half to three-quarters the capacity of the 

 test-tube. Kear one end of the test-tube is a passage by which it com- 

 municates with the air-pump, and near the other end is attached a 

 stop-cock for admitting the air to be tested. The test-tube and air- 

 pump may be attached parallel to each other, and held vertically Avhen 



"Nature, Oct. 11, 1883; vol. xxviii, p. 576. 



I 



