124 NUCLEATION OF THE UNCONTAMINATED ATMOSPHERE. 



with divergences on February 2, 3, etc. The exceptionally high 

 results for Block Island on February n, 14, 16, 18, penetrate, as it 

 were, into corresponding elevations in the upper curve. In fact, 

 between February 10 and 20 the wave of both places is of the same 

 general type, with the Providence data somewhat behind the other in 

 phase. This is one of the most interesting parts of the results. It 

 may be seen more or less clearly on February 3, 4, 7, u, 14, 16, 18. 



9 10 Jl IZ 13 14. 15 16 H 18 



FIG. 99. Showing daily record for February 21 to March 21. 



20 21 



Late in March (8-20) there is a final attempt at a sustained maxi- 

 mum at Block Island, while the Providence data gradually decrease 

 to the rain on March 20. From March 20 to March 31 both curves 

 are relatively low. March 31 to April 4, April 6 to April n, April 

 14 to April 1 8, show a distinct tendency of both curves to pass through 

 moderate maxima. The remaining data to March 4 are too low for 

 further comparison. As a whole, therefore, little more than a general 

 correspondence can be made out, in which the correspondence of excep- 

 tionally high winter uucleation correspond in the data of both stations. 



