( 66B ) 



clescripiroii of I lic slroiiülv sliadtMl l^'i-aiiiiliolcr lines. Close lo the 

 central al)S()f|)lioii line tliei-e was also an nnmislakaMe increase of 

 liiniinositv (resenihlinij,' llie supposed emission liii(\s in lli(> s(»lar s]>ec- 

 Iriini); hnl lliis inci-ease onüiit. williont donln, lo Ite allrihnled lo 

 llie niosl stronuiy cnr\ed i'ays heinu' J<epl louellier hy the Inhnlar 

 sti-nctnre (»f the llanie, and not lo dii-ect radiation IVoin the llanie. 

 For, the eliM'tric liuht Ixmiiu' intercepted, the emission-lines wei'e 

 scarcely ^■isil)je in the dark tield. Aud besides, as soon as the llanie 

 was disturbed by blowing- upon it, or Avhen it was j)arlially c(tvered 

 by a diaphra,ü,in, the l)right bajid, as well as the shadinu', became 

 nnsymmetrical with respect to the absoi'ption line. Neither Doim'I-Kk's 

 [»rinciple, nor the iidlnence of jjressure on \\a\e-lenuth can here 

 have played an apj>reciable pai-t. 



I moreo\'er obser\'(Ml fringe-like maxima and minima in the shadin.üs, 

 but they showed irreLtnlar and so nnsleady. that I could not thiidc 

 of measuring" their distances. Nor can ihei-e be any tpu'stioii 

 of photographiuu' this peculiarity before means ha\ c Ixhmi dexised 

 to Jv(H'|» a sti'ucture of sodium \a|»oui'. as desci'ibcMl above, steady 

 foi' a reasonable linu'. Such means are however being prepared. 



Imperfect as our present experiuKUit must be, it still ^er\es lo 

 bear out the assertion, that numerous pecidiarities of the solai' spectrum 

 may be explained from anomalous dispersion. 



Physics. — "(hi tli(' I'lii/'ss/'o// (inil (ihsorjition hij nn'fdls of /■ai/s of 

 hi'dt iif (Ji-t'ill irni-t'-li'lKltlis." \\\ II. A. LOHKNTZ. 



§ J. H v(fKN and IvrnKNs have recently shown by their measure- 

 ments of the i-ellecting powei- of metals') that the beluuiour of these 

 bodies toxvards rays of great wave-lengths (larger than IS »/; may be 

 accoimted for, if one applies lo the pi-o|»agation of eleclric \ibralious 

 the eipiatious that hold for slowly vai'ving currents, and which con- 

 tain iu> othei' physical constant of the metal but its condiu'tivity. It 

 follows from this lesull that a theory \vhich can gi\e an ade(|uale idea 

 of the mechanism of a current of conducti(Ui \\ ill also suflice for the 

 explanation of the absoi-[)tion of the rays that have been used by these 

 ex[)erimenters. A theory of this kind has been developed by Rikckk -) 

 and Drudk "). Accordijig to their \iews a nu'tal cojitains an immense 

 number of free electrons juo\ing to and fro in much the same way 

 as the molecules of a uas or as ih(> ions in an electi'ohtic solution, 



\) H.VGEN and PiVBEXs, nerliiici- Sitzungshriiclite. I'.I0:5, \\. -Jt')'»: llciichh' il. ili'iil- 

 sclioii phys. flcscUscli., \'M)?k p. IT). 



~) RiKcicK, \Vi(Ml. .\nn., bd. (id. |). Wo?,, 1S!)S. 

 •'j DuiDE, Di'uclo's Ar>n., Bd. 1, p. 0(1(1, 1900, 



