CIHM ICMH Microfiche Collection de Series microfiches (Monographs) (monographles) Canadian Institute for Historical Microreproductions / Institut canadien de microreproductions historiques ©2000 Technical and Bibliographic Notes / Notes techniques et bibliographiques The Institute has attempted to obtain the best original copy available tor filming. Features of this copy which may be oibliographically unique, which may alter any of the images in the reproduction, or which may significantly change the usual method of filming are checked below. Z D D D D Coloured covers / Couverture de couleur r~7] Covers damaged / Couverture endommag^e □ Covers restored and/or laminated / Couverture restaur^e et/ou pellicul^e Cover title missing / Le titre de couverture manque I I Coloured maps / Cartes g6ographiques en couleur □ Coloured ink (i.e. other than blue or black) / Encre de couleur (i.e. autre que bleue ou noire) □ Coloured plates and/or illustrations / Planches et/ou illustrations en couleur Bound with other material / Reli6 avec d'autres documents Only edition available / Seule Edition disponible Tight binding may cause shadows or distortion along interior margin / La reliure serree peut causer de I'ombre ou de la distorsion le long de la marge int6rieure. Blank leaves added during restorations may appear within the text. Whenever possible, these have been omitted from filming / II se peut que certaines pages blanches ajout^es lors d'une restauration apparaissent dans le texte, mais, lorsque cela ^tait possible, ces pages n'ont pas 6t6 film^es. L'lnstitut a microfilme le meilleur exemplaire qu'il lui a ete possible de se procurer. Les details de cet exem- plaire qui sont peut-etre uniques du point de vue bibh- ographique. qui peuvent modifier une image reproduite. ou qui peuvent exiger une modification dans la metho- de normale de filmage sont mdiqu^s ci-dessous. Coloured pages / Pages de couleur \l/\ Pages damaged / Pages endommagees D D Pages restored and/or laminated / Pages restaur^es et/ou pelliculees yl Pages discoloured, stained or foxed / k_l Pages decolor^es. tachetees uu piqu6es Pages detached / Pages d^tach^es 1^ Showthrough / Transparence I I Quality of prim varies / D D Quality in^gale de I'impression Includes supplementary material / Comprend du materiel suppl^mentaire Pages wholly or partially obscured by errata slips, tissues, etc., have been refilmed to ensure the best possible image / Les pages totalement ou partiellement obscurcies par un feuillet d'errata, une peiure, etc., ont ^t^ film^es d nouveau de fa?on a obteni.- la meilleure image possible. Opposing pages with varying colouration or discolourations are filmed twice to ensure the best possible image / Les pages s'opposant ayant des colorations variables ou des decolorations sont film^es deux fois afin d'obtenir la meilleure image possible. Additional comments / Commentaires suppl6mentaires: Pagination is as follows: p. 373-379. La pagination est comme suit: p. 373-379. This item is filmed at the reduction ratio checked below / Ce document est filmi au taux de reduction indirut ci*dc«tou>. lOx 14x IBx 22x 26x 30x V^''^ 12x 16x 20x 24x 28x 32x The copy filmed h«r« hu b««n raproduead thanks to tha ganarotity of: Gerstein Science Information Centre University of Toronto Tha imagas appaaring hara ara tha bast quality poaaibia considaring tha condition and lagibility of tha original copy and in kaaping with tha filming contract spiaeificationa. L'axamplaira film* fut raproduit grica k la g*n*rositi da: Gerstein Science Information Centre University of Toronto Las imagas suivantas ont M raproduitas avec Is plus grand soin, compts tanu da la condition st da la nattst* da l'axamplaira film*, at an conformit* avac laa conditions du contrst da filmaga. Original copias in printad papar eovars ara fllmad baginning with tha front covar and anding on tha laat paga with a printad or illuatratad impraa- sion, or tha back covar whan appropriata. All othar original copiaa ara filmad baginning on tha first paga with a printad or illuatratad impras- sion, and anding on tha laat paga with a printad or illuatratad impraaaion. Laa axamplairas originaux dont la couvartura tn papiar ast imprim*a sont filmte an commancant par la pramiar plat at an tarminant soit par la darni*ra paga qui comporta una amprainta d'impraasion ou d'illustration, soit par la sacond plat, salon la cas. Tous las autras axamplairas originaux sont film*s an commandant par la prami*ra paga qui eomporta una amprainta d'impraasion ou d'illustration at an tarminant par la darniira paga qui eomporta una talle amprainta. Tha laat racordad frama on aach microficha shall contain tha symbol —^ (moaning "CON- TINUED"), or tha symbol V (moaning "END"), whiehavar applias. Un das symbolas suivants spparattra sur la darni*ra imaga da chaqua microficha. salon la cas: la symbols — »■ signifia "A SUIVRE". la symbols ▼ signifia "FIN". Mapa. platas, charts, ate, may ba filmad at diffarant reduction ratios. Thosa too larga to ba antiraly included in ona axposura ara filmad baginning in tha uppar laft hand eornar, laft to right and top to bonom. as many frames as required. The following diegrams illustrate the method: Lea cartes, plonches, tableaux, etc.. peuvent *tra filmis i des taux da reduction differents. Lorsque le document est trop grand pour *trs raproduit en un seul cliche, il est film* * partir da Tangle sup*rieur gauche, de gauche A droits. et de haut en bes, en prenant le nombre d'imeges n*cessaire. Lea diegrammas suivsnts illustrent le methode. 1 2 3 1 2 3 4 5 6 Miotocopv tuourrioN tbt chabt (ANSI and ISO TEST CHART No 2) ^ /APPLIED \MA3E: I ^K '653 East Uatn Stpwt S T' J! Roch«5tef. Htm York T4b0i* USA ■^S5 (716) 482 - 0300 - Phone "^^f^ THE PLANE GRATING FOR STELLAR SPECTROSCOPY ■ ro •CM J S. IM.ASKK'n \^ ^*^- \ ( .jUN ^' ' l''i'* -J KqirinUil for private tiriulaiinn fr'un __,,'"'' The Asirophysical JtnnNAi . \o|. XXXVII. X<'. s- Ju'h- mi? -T/'^a ^^^^^^ J -■<-., '. '."'-' ', -r- 1 •.■•> I [.* ;>■ .1.- ' j' -r " -if. * 1 : THK PLANK (;K.\ri\(; FOR STKl.I.AK SPI.( TKOSCOPV ii\ J s i'i.\ski rr At the nu't'linnol the inn.niittif of thf A-lroimntiial ami A>tr(i physical Soiiety of Aiiu-rira on Co o|'tr,ition in Kailial \«'lix ily Determinations at Mt. Wilson in luio. the iiuf-tion of in< reasinj; the eftic'ienty of the nuMJern >tcllar spec tro^;rai)h wa>
  • ni-i t a plane j;ratin>; as the dispersion piece of a >tellar >] e* tro^'raph. Some earlier experi.iients on the u>e tif );ratinj;s in stellar spei tro>- copy were mad • hy I'oor and Mitihell.' l>y \Vad>worth;' and l>y others hut without very em oura^iti^; re>ult>. The ditliiulty was probably due to the loss of li>;hl entailed by Its distribution over several s|)ectra and ean be o\ere<>nie by obtaining a ;;r.itin^' frivinj; a strong conientration in one order. (Iratings have been ruled by Rowland giving a very strong tirst-order s|)eetruni. ami both Michclson anhould be possible to rule gratings giving a large proportion, as nnieli as 75 jier lent of the incident light, into «)ne t)rder. K\en if this etfuitncy was not reached, grating sj)ectra should ioni|>are fa\ ^'i\rn to the optiial ilc^inn of llu" sp<'(tro>;ra|>h ami three nulhods of [irndiu iiig a >piitriini wtTf trunieiit. in tlie l'ir>t of the>e, the usual method, in whiih a |iarailel |>ei)(il Iron) a doulilet (ulliinator ol>jei - tive of 57 mm (.'.JS in.) a|)erture is ilitTraded liack. makinj; an annle of alwiut ,^o with the initial direi tion. is emi>loyed. th'.' spei- trum luini; formed hy a triplet i aniera (shjeitive of ri^ 4 mm h 5 in. I aju'rlureand 507 mm (join.) foi us. In the second methiKJ the autoeolliiritiiij; or I.ittrow prim iple is umiI. the (onihined eollimator and (amera ohjei ti\e heinfja triplet of 6,^.4 mm ( .' . 5 in.) aperture and 051 mm (,,7 s in ) foeus. I'he sjie'trum is formed in the plane of. and as » lose as possible to. one end of the slit. The same prineiple is used in the third method, the gratinn Ininn replaeecl \>y a hahprism of o. 10.? k'I-'ss haxinj; anjiles of 58?5, and yo°, and >ilvereil on the side opposite lli' 58?5 ijjle. The light is refracted into the prism on the hy|)otenu>e ane of the half-prism for radial velocity work was jjrojiosed l»y Profess' r Caniphell at the meeting above referred to. and I understand a stellar siiectio- graph has been eonstrueted at the Liik Observatory from his designs whii h gives very i>romising results. Neverthele.ss. as no ihange was nee])e'-sions of tlie three forms are ^vo A. 17.5 A, and 17.5 A per millimeter at //,. These values are almost identical with those given by >i)ectrograi)h I (,^3.4 A) and spectrograph III S (17.5 A), one- an//// ia- s/Vi i Ho-^i nrv ^75 mailf of lira>-' plait-. tirniU ^inwiil tuyttlur .il llic .m^'lc^ ainl tln>r<>u(ihl\ l)ran'il iiitfriially. ( >\vinu In il> nlaliM- i Minpat liu»> a^ < onipari'il with llic ( )llawa uiu- |>ri~iii. it \\a> not iImiiiiIiI iit«i>- sarv to ii'.truilim- a lliinl i ouiiliilMlaiu ini; >tippoit a^ in llir iailtr iii>tnmifnt. l>ul thr l»ox i> lulil tlc\il>i\ ami \il tirnil\ in llii- hrainl T iron ir.mic l>\ two support- -o platcil a> to riilinf llrMirc to a niininuini. I lie ohjti li\i- wliiili \ the J. A. Hia~luar Co Iml all oilur parts of tlu- in-trnimiil were i nn-trui tctl in ilu' ol)-tr\alor\ work shop. I'rfliniiiiary lots a'nl ailjn-lnniit- ni.,.ic in the ial>oralor\ usin^ till' iarl)oii an a- tlu li^ht -ouric. sliow pr.ntiially ilal o\it the 4 (iti;rti - iiu hultil on the plate and that tin- lU imition of all thrte forms ua- 1 vicllinl. I 111' nii'thod adoi)tiro- portional to the ' ndxrs 1 . 1.5..'. ^.4. Inr ixanipie, on /Vixyc;/, 5 1 xpo-ures wi made with III S. ^ with the I.ittrow irratin^; ^ with the I.ittn .lalf |)ri-ni. and then ;; with III S. all three m-tru- ment- having; the -ame dispersion. .\ similar le-t wa- arried throuf;h with the lirst foriii of the ,L'ratin>: spei ito>;rai)h and the one-prism instrument, both of the same di-|itrsion. ami reiieated ! 1! ther stars and lor the -un. .\s the linear disper-ions were the sanie the spettra of the same width, the tihservin^ tonditions f irl\ ■instant, tlu ■ \po-ure titiies the -ame. • nd the plates dexel- I tomther for tli. same lime, the intensity of the re-ulting e- v;i\es a direi I 1 om|>arison of the intensitx ol the spet tra. '\ equal iniensities on the plates tan he tomjiareii. liltk ■)hi( errc - likely to otiur. '! he re-ull- of these toin- re ^i\eii •!! the -uhjoined tahle 1 -I striUns,' feature in the aiii>earann' of the ;;ratin!; as ,..-np..i: h tlu- pr -m.ru speitr.i and this is al-o e\ ideiit Irom he talii' reni.irl- d)le uniformity in inten-it\ hetwieii llg .iiid / ,;^ I I'.i- unifo' iity is i)erhaps the mo-t u-eful property 370 ./. v /•/. i.v. „ ^ „.,,,., _ o O O I . -, ^ o ? J ^ ^ -. o x. .' -^ o « -r - •' " " " " " ' -.-..-_ o o o o o '^ "■ r, \e •*_ ^ t, '^. -m '■ t ~r '• ooo OOOO-OOOOJi-^-. ■r ^ ^ .' .r ^ .^ - „ -. no/ ■ o o o r. r. -^ -r o < c t -. o vc t •> .^ .■ .r .,..,,,, „ ooo o o -■ o ? • - a: -<- - -ooo oouoooooojv: c*^, """------OOOO O ", .^ O O 3 -«■ O ", -. O I - -^ O •< C •• """•--"■'""'"■■■' OOO OOO"'. "-. ' o I - -r o :r »-- - -• ooo o o "-- o >: .J M a: CO-^--. -. -ooo- 0003CN ° 000 O O ■...'. o .<-. ox C OXC -'O .--. •"^ -^ -^ -^ - - o o : --. >r. A, ,ri Q -^ o -■' 1-^ *r, '^ -r -f ./-, -r "-,'•-.- o o o 3 -^ »^' "■, - O O o O O ^ X »*: - — ooo •^t^^?5: 5^:;^:; ^%=-x 1,^.1.- T 'K ri.ASI CUM ISC FOK KIF.U.AR SfKl IKiiSu >"] ,77 cif the Kralini; |ieitrof;ra|>l). '\hv mrtrast lutwcen Kratin*: in.l pri^m in this ri'.|Hi t i> v«ry striking;, as pri^n.atii -\>v, tra an- hmt ten tinifs (.'rating -h-i Ira only .nif an'i a thhil tini;>. a-< intfiM ,it A 4700 as at X ,<()00. Thi-. «. TiTituc i> Im- in the nviii to two causfs: lirst, the im reast-d tlisiR-rsion towanl thr vii.lti .nil liimin- ishfil toward the red of pri>niatii >|ri tra (dis|Hr.ion at K • nc and a half tinus. at Hg two thirds, that at //,>. whiliMiilTrat ti.in -.jK-itra are nearly normal; ami semml. the strong absorption oi the pri>in glass for the shorter wave lengths. Caliulation- from Vo^eis am- stants for o loj ^lass sho». that {hrou;;h 5 mi 01 thi- i;las», alwiut the mean lenjjth of path through the >.ini;le and lh<' hall pri>m \\ |)er tent is transmitted at the K line as tompared with 71 per lenl at //^ and S^ per »ent at //». The elTei t of thi> alivirption is strikingly shown in the table where the relative inUn>ilv u' three prism as compared with sinjjie- and half pri>m ?peitra toward the violet is gi\en. Disnissinn lirst the relative inter. Nity of thrte pti m and uraliiiK' soettra. we find the former has the advanta ;e from lla to aliout A4;ratinn s|H'etra .ire deiidediv iperior. t • ,)rism speitra (.is;. arinj; helow //j. If.ispettrum .ii which tn. region from A 4^00 to the ultra violet 1- rei|iiiretl onlv the grating could be used, while from //« to //, the |)risms would have the advantage. In other words for early t\pe >t.ir> u-i the grating, for solar t>-j)e. three-prism dispersion. Comparing next grating with single and half prism spectra we tind the advantage lies decidedly with the latter ahove the H am! K lines, but if the K line is required, as is the ca>e in many c irly type stars, it can be obtained with the gratin'i in the .ime time as with the one-prism si)ectrograph with(>ut making the lines l;et"iin Ha and I/y immeasurable by overexiM)sure. If now the jmsmatic spectra are intercon*. •"•■d the superiority of one- over three-prism sjiectra i> markedly show •, The former gives nearly three times the ini isity above / 4.'oo and a nun h greater ratio below, a difference of mo-c than a m:'gnltutle in the stars within reach of the telesco[)e. It must n.it Ik forgotten, however, that this dilTerence is partially olTse« 'v the threefold greater resolving j)ower of three prisms, although -.a i)hotographit .,.„., Th. c..n„an,. .ktvr.nin.n, .he roMt,..,. of ,lu ,..„.,,.. i„,„ps M..UUnh to :•..•,■.'.-.- when the rUw-ly .lonhk.l ...,^s ap,.ar. i..a\cn,an,s,,acticalU o...,antM.h,n«a. th. ah..r,u....u. is present ( )r. Jnne 1" no l.s.than n-nel.en ...n^.-.nt.v. // n, s. be.,nn,n. at //.. ^^eK• n.eaM.r..l an.l thu. n.u.c u. r. s ..hU- Th. -.vera^c .kv.at.on of the ocn,,.nte.l ,...Mt,ous, Haln.er . on- Htant ■;(•.' 1 '.'■• -t the niuWlc th.rtc.n of thc^c lines fum ih. ob :.,,v.t Jas ..nW ..O: A., ana of nine onlv n04 A , .how.n, that U,.v are nnaonhtc-aiv a.te to lnaH.,.n. d.splacea hv .,on>e nn , ,u,wn ana iaent.oal cause f.on, thcr normal ,>osU.ons. The a,M.IaocnK-nt .,f nu.nv euhancea n-.etall:c hnes uh.ch wasd,scover.ai.v Messrs. Aaan.s atui Jov, .,f Mt. Wilson a.>a ,,„.onncea as •-':; A. at U^^m has been ftdly contunKa ,n the ,,,ecira obtains! here between June 10 and June 1... I he pos,- „ons of sotne .„-, nulalUc line, iaentthea .hKliv a. aue to /,. /v, (V S-, S /-Have bten n.easurea on these plates, ana U h:.s Ueeti show',, b.-vona aoubt that the a.splacen.ents are .Iso p, ■ ,,ort,onal to the wave lcn,th, a,e exactlv the same as those o, Ihe hv.i. .,en absorpt.on ana merea^e fron, spect.un, to spectrum n.exa.th the same way. These a.spl rcen.ents m.rst evu entlv ,11 he ane to the san.e cause but what the can^e .s cannot be ae- ,,„„,nea. Ahhon.h the a.splacetnen.s follow the -"-'=»-- ,,U,evwereveloc,tvsh.lts, it seetns to me very unhkelv that M.ch enonnon. vclocit.es as 1 inO kn. per ^cc on Jnne 10 tncreas- „„. U, lT:.Oon June :;4 conUl he present. A tu.ther evulence • ha k on not be due to velocUy is guen bv the secona nanow d,~orKion .erus of seven Inu., also nnnustakabh ane to h>..ro „,„ who^e shuts wonia conespo.ul to a vcUhUv ot -:;<() k,u rn 1„„L i:, tb.t .^ to sav the absotb.ns; hyaiov:... wonKl have two a.lTerentvelocme^ lOT'^ uui :i::. .. kn.. per -tc at the same tnue. When these a.sp'. .cements .o evactl « follow the Hahner law ,or hvarouen, and an of the same amount for the enhanced n.s an.l also tor cakinn, as wdl be seen below, it seems more hkely tliat till \ iif dm- to Mitiif i>!i\*km1 i aii-i- .iitii'n iii t'.u- «..irt:«' niaiitifi ')ii Itif iiiohviiU-" .it .ill th«»«r -.Iftnnit- wli.H the utMii is cat! onh W mwe-iilatfl on tli.iii to vrloiHv rirli.it'* tlir iiio»l r.ni.irkaltlf 1k.1i.m ■! ot tlu .^asrs i-. »lio«ii bv thf //aii'l A'liitt"^ ot .••ti-uim winch t.ikf thu-r tot in* I ! I IvMw^ion, not viiv »tton>;!v ..uvti I'tit (m.|oiitii..llv ptrsi-nl ill -oiiif ^Kfctr.j aii.l of miihI ii wi.lth .iiul t>"-itioii to tlic livilro>;en eitii»suin ( 'i ) iMspUifd i»h-.ori'rioii liiu'-. that of A' -liowiiiv; aUviV- when the (Iwphu-ed hy^ .r|>tioii, l.v whu-h it in w ivf '.■ ii^th I's a\w\\\ ■1 A (I'O Verv narrow an. I >harp iuu- // iii.l A' ii«- ii Iv m tin ir normal p«'-ition atitl jirtsriit in nll■^t ot llu- s,„h tra, iv. n w li. n tliL- hvilro.;t;n aiwl •)tlur c iltiniii hI'v.i ptioti .Ik! not sli< *inali .m lu- tlnii-i- in \ilocltv lint tins is so close to \\\v prol.al.l.- etr..i ot iiik-asiiieniciit th.nt it van not Ix- r»u if>if'l -'^ ''^ "'V in<-.iti* . -tal. lislu.l '1"1h- n ■ an valiif of tli-- veloiitv ot>ti»iiU'il from tlu-se // iiui A'lim-s .,,1 17 platts n-lnc..! to thi v„n is t;( 7 U 4 km. ]>fr SCO Consulerin- tli<- v. l.Hitv a. constant, the proh.ihli- .iror of a plate as Iclrrininf.l l>v thv tnea«iire «' tlu-sc two lines i, 1 7 km. per sec- Tilt- i-arK platfs i;avc a vrlocitv ot ahoiit 17 km , which incrcas»r.l to -J.S m about K' .iivs, (IroitH-.l to !'.• ami in creas« km per sec. .; 378 ./. .V. ri.xsKErr spectra thi'ir lii<;h ri'solutioii taniiot he effectively employed and it is (if IK) value in early Ixjie stars. A final interesting; mniparison is that between the one-prism and till' half |)risni siieitra. ihe former jjives spectra about 20 l)er(ent nuiri' intense than the latter, a result to be expected when the los> at the silvered retlec tinj; surface is taken into account and when the larj^er aperture of the half-|)rism (0^.4 mm as comi)ared with 51 mm) and the (()nse(|uent f^reater length of o|)tical path are considered. It is evident that more intense spectra would be secured by usinj; the regular one-prism instrument rather than the half |)rism, making' the camera the same length as collimator. The further advantage of narrower lines for the same slit-width would be obtained by increasing the length of lollimalor. camera remain- ing the sanu'. The advantages of the half-prism instrument arc it> nun h simpler and more comjiact and self-contained mechanical form, avoiding >()inc of the tle.xure and temj)erature dilViculties occurring with the extended one-prism spec trogniph. In cone lu^ion. it ma\' be said thai although the spectra obtained from the grating are disajipointingly weak and show that liie pro- |)oriion of the incident light diffracted into the spectrum used is nearer ^o than the 50 per cent estimated by Anderson" or the 75 per cent considered po>sible. yet even under this handicap it can be used to ad\antage when the K line is re(|uire(l and if spectra of iMiilorni intensity or of uniform disi)ersion are needed. It would also be Useful in the red ('\\i\ where |)rismatic sjiectra are so unduly compressed. If a grating u'iving twice the intensity could be obtained it would be sii|)erior even to single-])rism dis])ersion for n\ii>t work. The relati\e lle.xure of prism and grating spec trogra])h should aUo be considered. It is well known that a small change in the position of a pri^m when at minimum deviation does not displace the sjiectrum. while the angular movement of a diffracted pencil will be twice that of the grating. Tests of the single-prism spectro- 'I li.u<' Ir.iriuil siiin- lln- .iliu\f wm- wriUi-ii thai Dr. .XiidiTsuiiV r>tinialf rt-- Icrri-d n.il in tin- iru iilinl linlil IjiiI In llu- luTn-nlaK'f ritiiriii'd \>y llii- sliui uliim niftal Mni.in' .\> till' ritli'i li\ ily ol >]ifi Ilium iiu'tal i-- .ilmiU (15 perifiit llii- IvMn'sliriiaU-s an III i;(Hi,| .mrcciiKnt. THE PLASE GKATIXa FOK SIEIJ.AK SI'ECl KOSiOl'l graph showed very small tiexures of the onier of : or ^ km jier second. A similar te>t of the h.df-prjsm sjh'c trojjraph showed Maximum tlexure of about 7 km and of the ^ratinj; sjiei troj;rai)h of al)oiit 20 km per second. In most exposures the disjihu fnieiit caused l)y the change of position in hour anjile will not nun h exceed one-tenth of these amounts, and it is evident that, willi proper distribution of the lomparison exposure^, there >hould be no appreciable elTei t on the \elocity mea>ure>. Ilenie the (|ues tion of tlexure need introduce no serious ditVuultie> in the u>e of a gratinj; spectrofirajjh. I have much pleasure in acknnwlely the apparatus re