NO. 3 PROVISIONAL SOLAR-CONSTANT VALUES ABBOT 3 



the completion of them will not change much the relative daily values 

 we are about to give. It is possible that there may be found a neces- 

 sity to alter the general scale of the results slightly to make them 

 comparable to our previous publications. 



We would have preferred not to make public any provisional values. 

 But the demands for recent results have become so insistent and 

 numerous that we believe it will be better to publish briefly at this 

 time the best knowledge we now have. The reader must understand 

 distinctly tjiat small modifications will probably be made in the final 

 publication which we hope to make sometime in Volume V of the 

 Annals of the Astrophysical Observatory. 



Table i gives a summary of the results from Mt. Harqua Hala and 

 Montezuma. A part of these results was published in the Monthly 

 Weather Review for February, 1923. There were certain of them 

 which needed correction, and so it seemed best to gather in one place 

 all of the values determined since the publication of Volume IV of 

 the Annals. Column i gives consecutive -dates. Columns 2 and 5 give 

 the numbers of observations at Harqua Hala and Montezuma respec- 

 tively. Columns 3 and 6 give the weighted mean values ; columns 4 

 and 7, their grades ; and columns 8 and 9, the general weighted mean 

 and its grade. 



All of the individual determinations entering into the mean values 

 have been slightly corrected by general formulae based on extensive 

 statistical treatments suitable to eliminate vestiges of atmospheric 

 influences not entirely removed by the preliminary reductions. We 

 shall omit here the description of these statistical investigations, as 

 they are too extensive for brief presentation. They will appear in 

 Volume V of the Annals of the Astrophysical Observatory. We may 

 remark, however, that except for slight horizontal corrections to 

 scale, the methods employed are such as to leave the two stations 

 essentially independent up to the last combination into the general 

 mean given in column 8. 



In choosing the best value for the general mean, we have been 

 inclined to give greater weight to the better station, Montezuma. We 

 have also been influenced by the view that sporadic values, quite out 

 of the line of march indicated by values on either side of them, are 

 apt to be erroneous. Finally, we have availed ourselves of the notes 

 as to sky and instrumental conditions in forming our views as to the 

 best value and its grade. Four grades are given, S, S — , U + , and U, 

 meaning satisfactory, nearly satisfactory, rather unsatisfactory, and 

 unsatisfactory. In what follows we omit all results marked U, and 

 merely include them in table i for the sake of completeness. 



