240 



PROCKEDINGS OF THK AMI-^IIICAN ACADEMY 



111 the list which follows, of the plates existing in the po-scssioii of 

 ;\Ii-s. Diaper, the remarks against each plate give all neces>ary details. 

 The photographs were all taken with the llnggins star spectroscope 

 with its two prisms, attached sometimes to the reflectoi', sometimes to 

 the refractor, as indicated. 



The first column of the tahle gives a cuirent reference uuinliii'; the 

 second, the dale; the third, the name of the object; the foiiiih. tii ■ local 

 mean time of l)eginning of the cxposiii-e, when the note-bonks liunisii 

 it; the fifth, the duration of the expo.-ure ; the sixth, the width of the 

 slit in thousandths of an inch ; and the seventh, the apei'ture of the 

 instrument used. Remarks follow cputed fom the note-book. An 

 asterisk denotes that the plate was one of those measured l)y Professor 



PickeriuiT. 



TABLE I. 



No. 



18,9. 

 1 Aug. 6 



2 Aug. 9 



3 Aug. 9 



Aug. 11 

 Aug. 12 



D:ite. Oliject. 



G Auj. 19 



a Lyroe 



Sun 

 Dajliglit 



Jupiter 

 and 



a LjTsc 



a Lyras 

 and 



Jupiter 



a BootiA 



Time. 



8h. 30ui. 



to 



12h. 



Expos. 



45m. 



5s. 

 30s. 



48m. 



27m. 

 40m. 



30m. 



45m. 



Slit. 



.006 



.007 



.007 

 .007 



.007 



Inst. 



12-in 



12-in. 



12-in. 

 12-in 



liemurks. 



12-in. 



The spectrum is about i inch long from 

 O to U. Put a finger in front of tli.- 

 slit .<o as to be able to .shut olT eituer 

 the upper or lower part of it, and tliu.-' 

 take a .spectrum of a -star with the 

 moon or a planet juxtaposed uiKjn the 

 .same plate for .a rcl'i'ieiicu .'spectruni 



With the stellar spettro-cope and a he:iMi 

 of sunlij;ht from a lie.io.^tiit, took some 

 photojiraph.s of the sun"s spectrum. 

 Altlioiigli the sun was :>oinewhat ob- 

 scured by clouds, 5 s. w-.s too long an 

 exposure, .md the picture was burned 

 out below U. The exjio.sure of 30 s. to 

 d.iyli^lit answered better, and showed 

 that these photographs extend nearly 

 to K, instead of ending at (i. On coni- 

 parinj; tlic-se with the |>lii>t(ignipli of 

 tile spectrum of a L> ra^ of August 6, 

 it w:is found fo extend .ahovo N (S'lSO), 

 and looks as if the great solar groups 

 about Ij (.5815) were represented in a 

 LyriB by bands. 



The evening was quite misty. — too uiutdi 

 so to photograpli the spectrum of a 

 Bootis. 



The evening was not clear, the sky 

 being misty orcovered by heavy clouds, 

 which formed and dissolved away at 

 interv:ils. The li'nglh of exposure 

 specified is therefore uncertain, beiiii 

 tile estimated clear periods during ot 

 hours the pl.'ite wa.»i iu tlie spei^tro-cope. 

 The mist cut off the more ri^frangible 

 part of the spectrum. The two spectr.i 

 sliirti'd p:ist each other a little in the 

 photograph, tuit the adjacent coinci- 

 dences are pl;iiii. To try to prevent, 

 on another occ.i.'^ion, the spectra moving 

 past each other when the telescope i-i 

 changed from one star to another, as 

 in the preceding photograph, the spec- 

 troscope was fastened to u piece of 

 board. 



Faint picture. 



