1946ApJ. . .104 . . 430C ON THE CONTINUOUS ABSORPTION COEFFICIENT OF THE NEGATIVE HYDROGEN ION. Ill S. Chandrasekhar and Frances Herman Breen Yerkes Observatory Received Äugtest 14, 1946 ABSTRACT In this paper the contribution to the continuous absorption coefficient of the negative hydrogen ion by the free-free transitions is evaluated in terms of the matrix elements of the acceleration in the Hartree field of a hydrogen atom. The new coefficients are larger than the earlier determinations by factors exceeding 10 over the entire range of wave lengths of astrophysical interest. Tables, of the continuous absorption coefficient of H~, including both the bound-free and the free-free transitions for various temperatures (2520° ^ T ^ 10,080°) and wave lengths (X > 4000 A) are also provided. It is further shown that the new coefficients are sufficient to account for the solar continuous spectrum from X 4000 A to X 25,000 A. 1. Introdmtion.—In two earlier papers1 the continuous absorption coefficient of the negative hydrogen ion has been considered, and its cross-sections for the radiative processes leading to its ionization have been determined with some degree of definitiveness. And recent astrophysical discussions2 relating to the continuous spectrum of the sun have shown that the cross-sections which were derived for these “bound-free” transitions of H~ are adequate to account for the continuous absorption in the solar atmosphere between X 4000 A and X 10,000 A. Beyond X 10,000 A, however, there appeafrs to be an additional source of absorption, which it would be natural to suppose is due to the radiative transitions of free electrons in the field of neutral hydrogen atoms.3 But the existing evaluations4 of these “free-free” transitions make them insufficient to account for. the observed amount of absorption beyond X 10,000 A by factors exceeding 10. Indeed, on the strength of this discrepancy, the existence of a hitherto “unknown source of absorption” has been concluded.6 However, as similar conclusions relating to the boundfree transitions have proved premature in the past, we have examined the earlier evaluations of the free-free transitions of H~ and have found, as we shall presently explain, that there are ample grounds for mistrusting them, even as to giving the correct orders of magnitude. We have, accordingly, made some further calculations to obtain estimates of the free-free transitions, on which at least some reliance could be placed. It is the object of this paper to present the results of such calculations and to show that these newly determined cross-sections for the free-free transitions, together with the cross-sections for the bound-free transitions given in Paper II, are sufficient to account for the continuous absorption in the solar and in the stellar atmospheres of neighboring spectral types in a manner which dispels any remaining belief in an “unknown source of absorption.” 2. The inadequacy of the Born approximation for the evaluation of the free-free transitions of H_.—The essential approximation which underlies all the earlier evaluations4 is that of Born’s for describing the motion of an electron in the field of a hydrogen atom. 1 S. Chandrasekhar, Ap. J., 102, 223, 395, 1945. These papers will be referred to as Papers I and II, respectively. 2 G. Münch, Ap. J., 102, 385, 1945; D. Chalonge and V. Kourganoff, Ann. d’ ap. (in press). 3 The possible astrophysical importance of this process was first suggested by A. Pannekoek, M.N., 91, 162, 1931. 4L. Nedelsky, Phys. Rev., 42, 641, 1932; D. H. Menzel and C. L. Pekeris, M.N., 96, 77, 1935; J. A. Wheeler and R. Wildt, Ap- J-} 95, 281, 1942. 6 See particularly the discussion of Chalonge and Kourganoff (op. cit.). 430 © American Astronomical Society • Provided by the NASA Astrophysics Data System 1946ApJ. . .104 . . 430C NEGATIVE HYDROGEN ION 431 This is apparent, for example, from the agreement of the calculations of Menzel and Pekeris with those of Wheeler and Wildt, in which the Born approximation is explicitly made. However, on consideration it appears that for the range of energies which occur in stellar atmospheres the Born approximation for incident ^-electrons must be a very bad one; for, as the mean energy of the electrons in a Maxwellian distribution at temperature Tis Ji<-0.0243 (5^) (1) when expressed in atomic units, it is evident that it is only electrons with k2 < 0.05 that will principally contribute to the continuous absorption. And for ^-electrons with as small energies as these, the Born approximation must fail. This is indeed well known from the work of P. M. Morse and W. P. Allis.6 But it may be useful to illustrate this failure in a manner which will emphasize the magnitude of the errors to which the Born approximation may lead in the cross-sections for radiative transitions. For this purpose, we have compared in Figure 1 the radial j-waves Xo (Bom) = sin kr (2) on the Born approximation with the properly normalized ^-spherical waves in the Hartree field of a hydrogen atom, which the authors have recently tabulated.7 It is seen that for r < 2 the Hartree waves have amplitudes which are larger than xo (Born) by factors which, on the average, exceed 3. However, similar comparisons between the ^-waves Xi (Born) = k rk.r — cos ^ r (3) on the Born approximation and the corresponding Hartree waves show that equation (3) provides a satisfactory approximation for the energies in which we are interested. Remembering that, in the evaluation of the matrix elements of the acceleration, the wave functions for r < 2 are all that matters (cf. Fig. 2, in which we have plotted the acceleration r in the static field of a hydrogen atom) and, further, that the contribution to the absorption coefficient arising from the s—*p and p —+ s transitions must far outweigh all the others, it is apparent that in the Hartree approximation we shall obtain crosssections which will be larger by ten or more times the values obtained with the Born approximation. In other words, we may expect a correct evaluation of the free-free transitions of Hr to bring about an agreement between physical theory and the demands of astrophysical data without the need of postulating a still unknown source of continuous absorption. 3. Formula for the evaluation of the cross-sections for the free-free transitions of Fir on the Hartree approximation.—Our remarks in the preceding section have shown the inadequacy of the Born approximation for the evaluation of the free-free transitions of H~. The use of the Hartree approximation suggests itself as the next best, though the effects of exchange and polarization may very well be appreciable for the very slow ^-electrons in which we are primarily interested. But it may be hoped that these latter effects will not, at any rate, affect the orders of magnitude of the derived quantities! In any case, to improve on the Hartree approximation would require an amount of numerical work which will be several fold; and the task is immense even as it is.8 These considerations, to- ñ Phys. Rev., 44, 269, 1933. 7 S. Chandrasekhar and F. H. Breen, Ap. /., 103, 41; 1946. 8 For example, the present work has required the numerical integration of 63 radial functions and the evaluation of 523 infinite integrals, not to mention the computation of numerous auxiliary functions and tables. (All this work was done with a Marchant.) < © American Astronomical Society • Provided by the NASA Astrophysics Data System 1946ApJ. . .104 . . 430C Fig. 1.—A comparison of the radial wave functions of 5-electrons in the field of a hydrogen atom on the Born (B) and the Hartree (H) approximations for various energies of astrophysical interest. (The abscissa measures the distance from the center in units of the Bofir radius). Fig. 2.—The acceleration r = [r-2 + 2(1 + r-l)]e~2r of an electron in the Hartree field of a hydrogen atom. © American Astronomical Society • Provided by the NASA Astrophysics Data System 1946ApJ. . .104 . . 430C NEGATIVE HYDROGEN ION 433 gether with the urgency of the astrophysical needs, have prompted us to undertake in some detail the evaluation of the cross-sections for the free-free transitions of Hr, using the s- and the ^-spherical waves in the Hartree field of the hydrogen atom which we have tabulated in an earlier paper.7 Now, in the Hartree approximation a wave function representing a plane wave at infinity is given by *« ¿z=o^44 k r ^.0.) (4) where the radial functions xi{r] k) are solutions of the equation d2Xi d r ■4 k* 2(1+7)e"rSx' = 0’ (5) which have unit amplitude at infinity. (In the foregoing equations we have adopted the atomic system of units). For wave functions of the form (4) the standard formula in the quantum theory, which gives the cross-section for a transition in which a free electron with an initial momentum (in atomic units) becomes an electron with a momentum k\ by the absorption of a quantum of appropriate energy from an incident beam of unit specific intensity, can be written in the form9 a Mi*) = 256x2 3 K/2xhe2eN)v(44ir2&w2 e2)V ¿2M1A¿V 1 oo (6) X I (Z, ¿02| r\l- 1, ¿2) |2+ I (/- 1, k¡\ r\l, k¡) I2}cm6, where the various matrix elements (/, k\\r\l — 1, k^) have to be evaluated in atomic units and Ak* = kl~kl. (7) Equation (6) gives the cross-section for a single hydrogen atom in the ground state for the transition in question when there is one electron with momentum ko in unit volume. Moreover, the wave length, X, of the radiation for which formula (6) gives the atomic absorption coefficient is 01 1 3 X (Aß2) (8) In the Hartree field of a hydrogen atom the acceleration is given by *=[72+2 0 "*"7)]e-2r’ (9> and the matrix elements which have to be evaluated are of the type =jr“Xí(r;^)[ii+2(l+Í)]e-2^¡_1(r;^)¿r. (10) To obtain the absorption coefficients appropriate for an assembly in which we have a Maxwellian distribution of electrons corresponding to a temperature Ty we must average a(kl\ Ak2) over all initial kl (and for a fixed Ak2) with the weight function 197.8 03/2&oe~31*3mo (11) 9 Cf. J. A. Gaunt, Phil. Trans. R. Soc., London, A, 229, 163, 1930; see particularly formula (5.24) on p. 194; also Wheeler and Wildt, op. cit., eq. (6) on p. 284. © American Astronomical Society • Provided by the NASA Astrophysics Data System 1946ApJ. . .104 . . 430C 434 S. CHANDRASEKHAR AND FRANCES HERMAN BREEN where 5040 (12) In astrophysical applications it is convenient to express the free-free atomic absorption coefficients as per unit electron pressure. If we denote by k(A&2) the corresponding atomic absorption coefficient after averaging over a Maxwell distribution of initial veloc- ities, we find that we can write our basic formula in the form K ,(A¿2) 7.251 X IO-29#6/2 r,f^î{kl) (ÄFp 1 d^ oo cm4 X £/{ 1 = 1 I (l, k>0\ r\l-\, kl) |^+ I (/- 1, k¡\ f\l, k¡) I’} dyne’ > (13) where f(kl) =100k0e-sl'*2<. (14) 4. Details of the evaluation: tables of the necessary matrix elements.—The problem of numerically evaluating absorption coefficients for free-free transitions is a specially troublesome one, since for each wave length the coefficients must be computed for a range of initial velocities sufficient to allow for the averaging over Maxwell distributions for various temperatures. If the matrix elements for all the necessary values of k2 and AÆ2 must be individually evaluated, then the number of wave functions which would be needed will be so many as to make the problem an impracticable one on this score alone. It may, therefore, be useful to place on record the details of our procedure in this instance. First, we may observe that the infinite series in equation (13) is so rapidly convergent that we may ignore all terms except the first one. This may be verified in the following manner: We have already indicated in § 2 that for / ^ 1 we can use the Born approximation without any serious error. We may therefore write Xii*', k2) Ji+\{kr) . (15) Accordingly, for 1^2, *'0 r 1 / 1\“| r (16) Writing equation (16) in the form {],, k\ \ r \l — l, k§ = —lirkikif rJi+i (¡hr) Ji~i (k2r) (17) •i ~ and integrating by parts, we End after some further reductions that G, flz-l, = It (M2) I/2 n (1 + r) ] ; — k^Ji+i(kir) Ji+i (k2r)]dr . ) © American Astronomical Society • Provided by the NASA Astrophysics Data System NEGATIVE HYDROGEN ION 435 The integrals over the Bessel functions which occur in the foregoing equation can be expressed in terms of the Legendre functions of the second kind, Qi(x)y and their derivatives, Qiipc) ™ We find {l,k\\r\l-l,k*) =\hlQl-l{x) +¿e'z(*), (19) where the argument for the Legendre functions is _k\+k\+4: X 2jM2 ' (20) Using equation (19), we can readily verify the fact that all transitions in which the inci- dent electron is characterized by an l greater than 1 contribute less than a fraction of 1 per cent to the sum in equation (13). It is therefore sufficient to consider only the ma- trix elements (0, ^î) and ^D* In evaluating these matrix elements it was found convenient to distinguish three cases, discussed below. i) ko < 0.015 and \\ > 0.015.—It has been shown7 that, when k2 < 0.015, the radial functions xo(^; k2) and xi(^; k2) can be expressed with sufficient accuracy in the form X»{r-W)=1[^-[X,(r)-WYO0v-0OoO'0O00'0OOov00Or\0OoOOt0^'C0OOo'O0AT0Oov^00Of00OoO''0OOQ0Oo'm0OO0OoOO'0^ ONt^esiOí^OOOO^Q-^t^OO 1^0 CT) ©receHsOsr©e^OisOHT©rt^qí^^t'mi>.—O^.li^i©>’0>-OTt'tTÍO-O-'^iOOO>'0OOO0^t‘'OOO'^ Q 1'Oí^í^f^í^í^'O'OiO^eo ©eN©C©Sc©se©scs©es©cs©es©cseosceosc©se©sc©se©ses©es©cs©es©es ©«©(i■^M-Oic©©©0Nt\©^o©©0O©•0^a\T©0©-o0lH■vC©^O,H-^H©''T^t-r©'^Os©•aí^O^\-©(tf»~so-^>^©O'(O^^c^\©©irsO~t\ t©©e0^s\'©0ef©#sL5TeOi©s-Oie'*ot©s^l|ef<'0Os©e\esT'<>O«-T0i)Ouir©^©©^0^r50o3 f^©0T'00t^00©o0©oO©'^©\HO©^‘\0O<©''-OOi©e\es0Sen©'o^CT-W©^ittCl'^w©»Á'>íiO*©-L-t4'oeO©ioO\iOoLW©'\O0Ot©'LHO\OO^T©L\tOOiv©-H\i*0Oy-©n-Li*CtO^©K.'^Tc^^©HSíiOeti©-ooH\eoOvoNoH\ d © f©r©©■^^O'©'O^-©eHL©st^OOo■©«ofOO©•s\^\'oL^'i©>TO-tiHOee©ooiO-©'©H^eON'To©O©-oL-e'©O^COiSeot©O^CoeOSo©See\'0-SS©^'P^LO-O©OiNifOOee©jO©evOO©0ir0'Ní^1©Cl'^^OSi-©oiiOO©e©N''LOOCOte©^K's0O'©(r0NO-0i j'©'o-<^io-e^©©os»oe©c-soiscooT©soOeti<©sNocte^s©©se.ao©r©»)\efT'^©'o©-'e»<©ooer-afo0^«-oT©^eiiooL-‘c©voO^Co'h'©Oi\t-ohHt0ot^-he'©o0so'eiTOH©—0'>i0e-eOio<'©N^iiierOo0vs©eeOs'c'e©OePsOetO©N^'^'oOí©c's^vs'f©L0Lto©Oo©r'Ot^'«^o©'TéOoefht©eo0e\oOeeoseosoees»s'-0( ©©©©©0©0000©©©©00©00 © © L•©^\0r'00H0©©0‘^0©-©L'i0eeON©'s0e©Tts©^ÍLc<.sOt©vc^0s©0©0e0©s0e0e©0DSs\'0©íeO'vo~i©-©TlLt'©<^O©-<©l■e\?^0vs©i’eeo-0s©\ro©o©©^oet©O^o©Ore^To©o©-t©oi—i ©©©ee'^osTee©©OOsOí©©O0—>0i©OírOil~e©o'NO'l0oO©v0ío^s\ oeTSOf TTTff+riofio'Ooot^^\OoTosÍ©L^\©OL©©OO'©©^^OeOt©sL^'©OO©©xoL^Ooo©t^'e\O0eo©0orrO^ie<©QNse't©O^o^ L©©'f^OOiO^O©©'L©H>O0©©^i^0HrH50©©^O0e-O0©\©iL0^\O©H0©©©“©ç0©©rv©t^,jO-ç©'o^iOvi-©Oj©líf'-^sTi©'Oie>^çr-'^>©^©lcTT0s©-0}©ele©O's-OMe©©\Os'i-oHt©'-©^slíiet-|^©e©slt^se^O's©©r'0e-\l'se©O-©eS^Ls-'©©Oo©4ioO©^'\~©©-©^ ©©©©©©©©©©©©©©©©©©©© l©©©©eOsTtt©i^©^HoOeve©©'sONtt^^©f^©-'LeOOs©o©'t^te^^iO©Otoe^S''oO©^©’OHO-©H©'OOOOs'o'rO*C--hO©iOrO'Or©i-▼OiO—c©f.O0—'is©eíe©©-o©©oo^©\©^©es0^'^©T'O"-©< o©oe©s1ee0ss ©©l0O-^©©iT©L-iO©cS©©ClS©OrO©©'^©0L©©O'0©O©t©-00©0^0©’\-©i©'C^©S-i-©T^s-©Oi^©O-Ot©c©LseeOSs © American Astronomical Society • Provided by the NASA Astrophysics Data System o0o0 !o/)oTojov0O©o©0\^'©\OT-cH-t©s^CHvtS©-or|.0Coj©^»C\0reV~©rs^Oo'©'0ov<©Oo—©o©Toi©’'v-tOiiOf©'r©©3\©©i'r^O~l©0’if--0©'-Hm©-^-t’H^-'-<-Clil'oOCrv-OSHTtOO^-l' \©-r^)•\©r^)■\©n^\•r©^)iirT©)i^r©T>irn©-v^m©-^r-©)tvti~©ri^\©ro)om©r\of©r~r>i©^ino\©rjo~©r)ior^>©oin©o\on©i\rc)s©\críNv©--ir©i-r)i O»©0T0OcÿCN©ivNot0rt©)o0»\rTo0©5-H\©'0O0'OeJ>0s0v-0'0^OOC'©Nt0S^'^Q>t©^O--tlif^'tf^©0)O|v00Oo©0\ofOTs0-©l'O>0c-©s©0Hcf0O<©^\©rf'}0Ci^i'-r^-)iOfOOPOCO\<\CO;fOCCffS,D5tr—Cr!SmCCOfCS»CrHr)S©Ci-S0tC0©UNO0CO^S-tCl^vSlOOCt^fSCvCOrSH-'«>t-^^l ©©©©©©©©©©©©©©©©©©©©©© r©rTH^jC^<©<-T5HT©tt'Hr^S5--l'^^if'-^0Hl 00C0\NO0OC0 vS0O0Ci0VO0T)0Ci0-0CH0N^0O-0l\’0'O-0^«liií<0^0f*00 .lllOO^T-jOslHTOO-OOIO^OOHOOOO^Hl©©^^vf»^SO0HT\'OO^lHHvTOcfOrs\O©lOCT©CSH©tS^O'*CO-vS(OOiC-v'C^S©i-H»'^<©-0O©L©0fOsO©O0ifn\NT©©»--ll iPo0©T'O*i^—iOii-i~O-©'Oc'oOr''-O©l->>--tt-^«t-^iO'O'v'tO'~'-O'v»o-ot©to0r0^ ©©©©©©©©©©©©©©©©©©©©© ©lfr'^fOO>VO©rTOO*O©^0VfOOÖM©-©'rOHC©0O0-N^s©©-iCr^}S©-Tt^-U^IL©"TOO^tO«Sjt©©Ci'SO0-©ICVvNO0 ©©©©©©©©©©©©©©©©©©©©© ©irC>oi<©^!iHes^e»soe'os^eSs'íeos' es es oes'íeos' oes^eíos' eses oes'íeos' 0es^e0s oes'íeos' te^siens íe^s'eíos' íe^síeos' ! I ! I I I I I I I I II II II II II II II II II M COOn t''» On On © eeos eTotf Oeo TO-nI ceoo \000 vo t— eo iO Oes O lo On OO CS rOon On '*—( NOO es lOon teos O i t''» oo lo eo es th oo vo r+i rf © colcos totoo esetos CcoO »-HcOo »-H 0c0s On LeOs ecss NO eOs OcsOcs^c-^scloslcoslcosloelsocnsOesncOselos I I I I II II I II II II II !O>O.OCOSt^Ol>n.N^HOOOOl»o-OHnl>O-t\O^»N-Oh N»“OH LOO NCOS 0_0 NO T-H TfH es CS T-H IO CS Th On coco 0c0scThscThscLOscLOscLOscioseioseNsOeNsOcLsO NcOs »eOs NcOs lcos NcOs lcos NcOs LcOs NcOs LcOs NcOs LeOs cs LcOs Jc>s- IcOs es LeOs I II I I II II I I M II II II II II II II II M nTOtflooooTqhONncOooOonTnhOi>C-S©Ot^n*O^ncOo LoOo *e>s• tO"N- eTsh LeOs lO>» 0N0O ^Q CTOh L©O ©es onoo »©-H ©^ »N-OJ c-o »—ILO»—INOCOCS-<-H»-HNOTh CO Th CSt^. CS 00 »-HON »-HON »-H 00 00 00 NOL>» LO NO Th LO OeOse^hsTehsLeOsLeOsLeOsLeOsNeOsNeOsNeOsLeOs NcOs ieos NeOs LeOs NeOs LeOs NCOS LcOs NcOs LeOs ecs- lcos Cc's- LeOs ÍC"S- LcOs I I II I II I I II II II II I I II II II II I CesS O»-Hn CTSh O»-H »C-HS O©n *T>h• LTOh NcOs OOnn NO ©©ClOo C^ONcOo OOn0 C0O0 lnoo l>- »C-HS0©0 l©oOTnh LC>S> C©-» Th NO LNOO C©O CSO©n LO CO CS »H 00 LO co © CO CS CS CO Th »-H Th »-H Th 00 Th NO Th LOCO Th CO CO CS OenseThseseLOseLOseLOseLOseNOsenOseNsOensO I I I I I I I I I I I II II II II II II II II II II 0O0N»©-H©»L-HOVNOO»N-HOOCNSO©NThLNOOeSlO»-eHS TThh© ©ON »-H LCOO 00 CNOS NO N©O © 0NO0 LO ©Th NO COOn NO LNOOON On© On Th »-h Th es NO Th co co cs th ©CS »-H CCSS On CCOS NO »C-HS io On The^ COLO CS Th ^e's^e'Ísoe'sÍeos'Íeos'leos'e©s'eos'©es'eos'neos' enos'enso' eos'Oes' eos'Oes' íe^s'eos' íe^s'eos' íe^s'eos' íe^se'os' íe^se'os' te^s'eos' ees- eos iiiiiiiiiii m m m m m m m m m m ■lo'—o CO LLOO©eoCeOsCOSNCOOOLO^CLOSCCOS O00n LT'hCO CS On LO CO CS »H »-H Th On ON On © CO O lo On ■»h oo ocoo ©On co t'- ©00 »e-Hs cs e- es es eso OesNes»ehsieosLeOseLsOesLesoeosoeosoeos Ocs eos oes ocs cos cos te's- cos cs eos Xe—s cNOs es cos te—s oes *c-s* oes II I I I M I I I I II II II II II I Il II II II W O »-H C0O0 © xL>O. »T-Hh »©-H TNOh L©O L»—-H- LoCOoS cOo jO>>n XC>S-0O0 c©oxT>h- o Tc—oi Ooelso tL"O- »O-H lOon COO ©On O©n T©h O© lOon l©o NOesNOco»-í»-HOOTheocs© »-íes »-h Th »-hlo oono o no Th o coo eso es »h Th m eOsNes'^e'Lseos'Leose'LsreTsoe'soe'os'eos'oe'os' cos eos eos ocs rc-s cos es cos xe>s. eos ocs- eos Le's- NeOs ic-s- ncos tc"s cos Jc>s- ocs I I I I I I I I I I II II II II II II II II II I xT^h©coOtN^O©0TÇh»o-Ht^f-O.»N-HL*^oCoOoOTcNho 0co0 tO—n ©00 ocoo OO L©O T»-Hh O00 »t—HI c©s 0O0 0c0s c©o L»“OH 3 ©es x0>0- N^O LcOo LOeSLOCS»-HONNOCOCS»—i© »—l»-H »-HCS oo co t^-co lo Th coco coco es co cs es »—i es OeNseThsLeOsLeOsLeOsNeOsNeOsNeOsNeOsNeOsNeOs eNOs cNOs te"s. eNOs ecs- enos íe-s eNOs es enos ees- cnos ec-s enos re-s- enos te'"s eNOs es eNOs I I I I I I I I I I I I I I I II II II I I. I I I I II II tN'-OON»O-HO©»-LHOCCSOSCNOLOt^CCSST»--HHL©OteO^Sn. c»-oH ©CS 0T0h NcoO Tc“sI c©o Te-sI co t0h0 NeOs OThn O©n t©h Oon nnoo Xc>s- cToh lcos ThCS»-Ht^LOCOCS»-H© »-H 00 es toes Th cs co es eses cs »-h »-h »-h S OesNeTshLeOseLOseNsOeNsOeNsOe'Ose'OsNeOsNeOs xe>s. enos ee—s enos OCS- CNOS ees- enos es cNOs rc—s cnos Oes- eNOs tC"S- NcOs tC'S- CNOS es eNsO I I I I I I I I I I I I I Il II II I I I I I I I I o»-oH lcoo oToh es ee-- ceCosS TThh LNOO 0N0O TOHn ^1-H On lLToOh ©cs eO0N0- LNOO »©-H nNoO eCOs CTOh c»-oH LCOS cNOs 0ON0 L»-H'- N©O e©- ^eCOs^íO—hn- ^©nLo é0nT0o CO »-H CO »-H On NO Th CS »-H »-H © 00© L-. © LO co »-H CS OeNseThsLeOsLeOsNeOseNsOeNsOeNsOeNsOets^NeOs Le—s• cnOs cs NeOs ees- ncos hcs- neos es enos tc—s NeOs te's- enos íe—s eNOs 0e0s cNOs oeos ncos I I I I II I I I I I M M II II II II II II I Oesn »-H COON »C-Oi LTOh »0-0t 0CO0 TOhn CS C©S t©'- Xt—>-CNOS COSn CTOh es lo LeOs »Ne-Osh leos lOnoNo Th CO »-H »-H On © NO© LO© Th© CO © es© L©O ©NCOO ct-Hs LLNOOO oOon LO©On L©O LTTOhh © T—i © ON © 00 © ©©co' c»©-oh tL©'——O llTooh L»nC©OOho c©0©CoO0 e0cC©—0sS oCe©TOo-st L»-OH xLO>On- LLNTOOOh * The arrangement of this table is the same as Table 5. L8O »©©—i ©L■*—Oi©©cs L©eOs c©©o 443 S S es American Astronomical Society • Provided by the NASA Astrophysics Data System o0o0 H 5 P ¡z; § CQ w H w !í ^ M w H Offee s 0CQS,C^S'C,ts,»ic^5SlCOSllOMVCOSC»SOClOSClOSCVSOC»SOCtSOCtOSClOSClCSlC/5S»C/5SvCrSj>COS ¿i¿¿JJ.¿¿¿¿¿¿¿vLLLL¿¿¿ OCCCSC0irÍ'jO^f-O-t«i^-rI\O-00.0tO\M»UC-0'IN^Cc-Tot0rtefOUs<'OO3■ OfO0s0OO0OO^CCNNOO'\O0'0OrTHfCtNOCOM’O--ToiHOot-C^^'heOsOrcts0t<<-o0OO'0\ 'Oes>r5eST-ta\vocoeseseoe^^Tí0c-Ocs»si-O-(rtr-^i(oo-oeo--iOooOto^v'iOoo »oeoes>-'eseseseseseseses'--i^-i eOsnc^sieOsicOse'Ose'Osc'Ose'Ose'Osc'Os'cOsOcs'Oes'Ocs'OcsvcOsc'Ose'Ose'Ose'Os ie'OOO0v~'TOOe^'sOtule^-O'S3O'''Ov-í^4re-"sO'-^eÍe-ÍOsirOsírOO^\'eO0o\TO- -slOtV'OJlOOt^Í^'0O'OPueo) >r-(cS'i-tr~^er)i-i'rHOOOOOOOOOOO OO0fC<0rSOOO5^'NiOOrO-'iOi\OOOt~'t^Oi|rOvOOOoOOe-O>f'vt0OOO0r'iO»^OToiOOoji iOO0»OO-»-O0iCOOSiOC0SOOeO0OriOO ag tíO ^s |^-a3 «Ö Î3 -d w £!t^safíljmIotuí -dS -5Cg?? fH*e 68« ,H—d W) © American Astronomical Society Provided by the NASA Astrophysics Data System 1946ApJ. . .104 . . 430C NEGATIVE HYDROGEN ION 445 6. Concluding remarks.—It is not our intention here to make detailed applications of our new absorption coefficients to various astrophysical problems. But one comparison is of interest. In Figure 4 we have compared the theoretical variation of the continuous absorption coefficient of E~ with wave length for T = 6300° with that derived by Chalonge and Kourganoff in their recent discussion of the continuous spectrum of the sun for the same temperature. It is seen that the agreement is very satisfactory. In any event, there is no basis for the conclusion that the free-free transitions of E~ are not X In Fig. 4.—A comparison of the theoretical variation of the continuous absorption coefficient with wave length of for T = 6300°, with the corresponding variation derived empirically from the solar continuous spectrum by Chalonge and Kourganoff. adequate to account for the observed amount of absorption in the red beyond X 10,000 A. Indeed, it would appear that the negative ion of hydrogen by itself is able to account quantitatively for the entire continuous spectrum of the sun over the range of wave lengths XX 4000 A to 25,000 A. On the physical side we should, however, emphasize that, while our new coefficients for the free-free transitions of E~ are probably to be trusted generally, the importance of exchange and polarization for the slow ^-electrons may lead to further changes. It is likely that these effects will have a tendency to reduce our coefficients somewhat. We hope to return to these questions in the near future. © American Astronomical Society • Provided by the NASA Astrophysics Data System