ATMOSPHERIC NUCLEI. 125 



fully described by Professor Barus (loc. cit., chap. 8), and the same 

 apparatus (fig. 57 a) which had proved efficient in the earlier investi- 

 gations was used throughout the present experiments. F is the revolu- 

 ble wood fog chamber with plate-glass windows, V the vacuum chamber, 

 i the influx, e the exhaustion pipe, g the gage, G the goniometer. Source 

 of light beyond F is not shown. Two or more observations were usually 

 made daily, together with the meteorological elements of wind, weather, 

 and similar data. The ionizations referred to below, section 91, were 

 taken in the same place. 



87. Data for nucleation. In table 48 the first column shows the day 

 and month, the second the time in hours and tenths of an hour. The 

 third gives the current weather, F denoting fair, F' partly fair, C' partly 

 cloudy, C cloudy, R rain, Sn snow, etc. The temperature of the fog 

 chamber. (C.) and the temperature of the atmosphere (F.) follow. 

 The remaining columns show the data referring to the coronas, the 

 sixth giving the diameters (s) of the coronas at the end of a radius 

 of 30 cm. Hence s/30 is nearly their angular diameter when the 

 eye and the source of light are at distances 85 cm. and 250 cm. on oppo- 

 site sides of the fog chamber. The seventh column indicates the colors 

 of the successive annuli of the coronas, reckoned from within outward, 

 w denoting white, r red, o orange, y yellow, g green, b blue, p purple, etc. 

 A vertical line ( | ) shows an indeterminable color, a prime (') an ap- 

 proach to a color, etc. The last column gives the number of nuclei in 

 thousands per cubic centimeter, deduced from the amount of water 

 precipitated per cubic centimeter, and the sizes of particles in succes- 

 sive coronas, as listed by Professor Barus (loc. cit.). 



88. Remarks on the table of nucleation. With regard to the individ- 

 ual observations very little can be adduced that has not already ap- 

 peared in the earlier work. Conformably with the mild winter, the 

 nucleation as a whole is relatively low, an unfortunate occurrence in its 

 bearing on the purposes of the present work; but deductions of this 

 character will be brought forward with advantage in connection with 

 the daily and monthly mean nucleations below. It is rather curious 

 that the forest fires on Cape Cod in the early May and the powder com- 

 bustion on July 4 produce so little impression; on the other hand, the 

 cold weather in August is at once marked by large coronas. 



89. Mean daily nucleation. The data of table 48 have been averaged 

 for the successive days in table 49, with other data at once intelligible. 

 The results, if given graphically with the current days as abscissas, the 

 corresponding mean nucleations in thousands per cubic centimeter 

 as ordinates, show no new points of view. One may note the rare 

 occurrence of the large g-b-p coronas so frequently met in the high 

 nucleations of the preceding winter. 



