E-Book Content
782 PHYSICS: KEMBLE AND G UILLEMIN PROC. N. A. S. an increase- in crystal dimensions causes a decrease in wave-length of the phosphorescence. The key to the solution of this discrepancy lies no doubt in the different electron configurations of the activators themselves. The rare earth group, of which samarium is a member, is generally considered to have its valence electrons on an inside shell, and not on the outside, as is the case with the other elements. The approach of neighboring atoms would naturally be expected to produce a different, perhaps even opposite, effect on these sheltered electrons, than when the electrons responsible for the emission are on the outside and more directly approached. Further experiments with different activators might show whether or not all of the rare earths behave in a like manner in this regard. 1 This investigation was made under a grant from the Hecksher Fund of Cornell University. It was inspired by previous work with Professor E. L. Nichols on uranium as an activator. The author wishes to acknowledge his continued interest and helpful suggestions during the course of the present work. 2 E. L. Nichols and M. K. Slattery, J. 0. S. A., 12, 1925 (449-466). 3 Papish and Hoag, Proc. Nat. Acad. Sci., 13, 1927 (726-728). 4 T. Tanaka, J. 0. S. A., 8, No. 2, Feb., 1924 (287-318). 3J. Ewles, Leeds Phil. Lit. Soc. Proc., 1, 1925 (6-19). 6 Davey, Phys. Rev., 21, 1925 (143-161). 7 Travnicek, Ann. Physik, 84, 1927 (823-839). NOTE ON THE LYMAN BANDS OF HYDROGEN By E. C. KEMBLII AND V. GUILLIMIN, JR.* JsFFpRSON PHYSIcALLLABORATORY, HARvARD UNIVURSITY Communicated September 13, 1928 In a very admirable paper, Horil has carried through a complete and thorough analysis of the rotational terms for the Werner and Lyman bands of hydrogen, using his own new data for the former and the data of Witmer2 for the latter. Starting out from an analysis of the vibrational levels given by Dieke and Hopfield,3 he obtains a complete assignment of rotational quantum numbers to all the observed lines of the thirty-two bands and also derives values for the constants in the formulas for the band frequencies. His value of the moment of inertia of the normal hydrogen molecule has received striking confirmation through Dennison's4 solution of the problem of the specific heat of hydrogen at low temperatures. There remains however some uncertainty regarding the structure of the Lyman bands (B3-An)' inasmuch as the identity of the two branches actually observed is not definitely fixed. Without being categorical, Hori favors the interpretation that these bands consist of P, Q and R branches, like VOL. 14, 1928 VPHYSICS: KEMBLE AND G UILLEMIN 783 the Werner bands but with the corresponding lines of the P and Q branches happening to lie so close together as to form unresolved doublets throughout the entire system, and has shown that theoretical formulas for the frequencies based on this hypothesis are indeed capable of describing the bands within the limits of experimental error. If the bands are to be interpreted in this way it is necessary, however, to assume a quite unheard of value for the constant a which fixes the magnitude of the "sigma-type" doubling. This constant displaces each Q line relative to the corresponding P line so that it lies practically on top of the other instead of in its normal position approximately midway between corresponding P and R lines. To the improbability of such a large doubling constant is added the improbability that even with a large 6 the Q and P lines would lie so close together throughout the system of bands as to have completely escaped resolution. The doubt thus cast on Hori's interpretation of the bands is increased by a theoretical analysis of the possible types of molecular states belonging to the B-A system of which the Lyman bands fo