102 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 



The wavelengths assigned by M. Becquerel to the baud at the limit of his researches, at 

 1460 to 1480, appear to me too great, for this limit corresponds to the band whose wavelength is 

 given at l,a.3C to 1/^.37 on my chart published in the Comptes Rendus of the previous year (Sep- 

 tember 11, 1882), and on a larger scale in the American Journal of Science for March, 1883, and 

 in the Annales de Chimie for August of this year. I regret that M. Beoqtjerel has not read the 

 article in the Comptes Rendus. Had he done so he would have seen that the wave lengths there 

 given were not conjectural, but directly determined by the only practical method — from the use of 

 a grating. They were the result, in fact, of the measurements I have just described, and were 

 specially intended to give information about the unknown region extending beyond the limit of 

 M. Becquerel's researches, such as the great newly discovered band 12, for instance, which 

 stretches from wave-lengths 1/^.80 to 1//.90, while M. Becquerel's furthest baud, as I have said, 

 is at l/a.48, according to him, but really nearly at 1/^.38. The present memoir will show what degree 

 of reliance may be placed on these measurements. 



It is understood that a photographic map of the spectrum to 1,^.6, and therefore covering the 

 ground of M. Becquerel's paper, but not extending as tar as my il, will shortly be published 

 from the joint labors of Professor Rowland and Captain Abney, and as their results will prob- 

 ably be accepted on all hands as more exact than the preliminary explorations in which M. Bec- 

 querel and myself have been engaged, we may await its appearance for the determination of a 

 part at least of the points in question. 



I would call attention to the fact that M. Becquerel has stated that the furthest band 

 known to him in September, 1883 (except from my owu researches), had a wave-length of not over 

 1/2.50, according to his own estimate. 



