32S 



.VA TURE 



[January 31, 1895 



.. .ho« of mies Qi-9-5. This correlation .ppears to indicate 



*K-.i thr fainiest «:tars and those of the mn or lom nia^- 

 p'ot^bfy Kart of one sy.en,, and are »' "-^^ ':/,"- 

 His-ance from us At .he same t.me, some . '»" ^''I ?f^^\v *" 

 oth i.-ae -=e<:m to be intimately associated w.th '^e M Iky " ay^ 

 '^The hypothesis of an annular sy.tem, relatively -obted rom 

 theceninS part of the Rreat galactic system is regarded as not 



n^mpL™ble »ith the di-lribmion of stars «h.ch he has founC 

 F^" at nearly the same distance from us, stars vary so much 

 in sz"e or int-insic brightness as to give •""g"''"'^" ""^'"e 

 from 9 to 15. there seems no reason why some should not be of 

 creitcr bnuhiness than 9th mag. , 



* There is nothing to prove, however that the various par so 

 the Mdky Way .rl at an equal distance from us, nor even ha 



.may be'^an enclosed ring. It does not ^PPear improbable .ha 

 sub«quent researches may show the ex.>t.nce of one or several 

 tu.raN emanating from a cer...al accumuUition, and recurving o 

 a^ .0 form a nearly annular sys.cm. or ore consisting of nearly 

 Toncentrk rings. Mowever it may he, Mr. Easu,n's r.sulls seem 

 r . S c" .e that the portion of the Milky Way accessible o 



.mr ».eans of obser%a.ion. has but little thickness in relation .0 



"In LTa^r en the same subject (AV-.W.*.. February^ Mr^ 

 Maunder fiiidsit difficult to rc-ist the conclusion that the daik 

 Un. s- of the M.lky Way a.e really region- of barrenness, and 

 regards these features ..s indications of a process of condensa- 

 lion toing on in the stellar as well as in the . ebulous matter. 



The Snstem of ALr.ot.—tJai.e recemly (Naturk, vol. 

 xlv p. 446) Mr. Chandler credited Algol with an oljscure com- 

 uanion in addition to that which was recognised l,y Coodricke, 

 r,!d .he existence of which has been fully confirmed .y the in 

 vstigaUons of Pickering and Vogel. Mr. Chandlers concl,.. 

 lion^weie based on a discussion of the s)S,ematic 1, regularities 



of the epochs of minima, and were ''Pl'J^'"!'^/!;" '''",'' E 

 , la.er discussion of Ibe proper motion nf Algol itseK (Nature, 

 vol xlix P -,49\ The evidence of irregular proper n.olion, 

 howe^er, il not legarded as conclusive by some aulhon.ies. and 

 M Tisserand, the Director of the Pans (Observatory, is ap- 

 parently one of the unconvinced, lie has therefore a<<e"ip ed 

 ^Tlind fome other explanation of the phase variations (C.",//..- 

 AW«., January 21, 1895), and the result is to =^how .ha. 

 ,hey can be simply and sufficiently explained by supposing a 

 single dark companion moving in =">,«"'p".\ '"^'"'^ °i ^,' '' 

 cular. orbit; and, in actdilion. that the bright star exh.bi.sa 

 sensible polar compression. The resul. of this depa.....e (rom 

 the spherical form would be a movement of the periastion point, 

 and ihi- would explain .he apparent iriegularines 



Assuming that the plane of the orbit is coincidenl with he 

 equa.or ol Algol, its tccen.riciiy is found to be 0-132, and the 

 polar diameter is shorier ihan the equatorial by .U. 



The consequences of these condiiions would be a very slight 

 variation of minimum brightness in the long period of 140 

 years, and an entirely negligible d.f.erence in the t.me "f Pas;'»H': 

 10 minimum and recovery of normal bngh.ness. I he to ai 

 duration of the eclipse, however, will vary very considerably 

 Taking .he mean epochs 1800 a..d 1SS4 for the observations of 

 Wurm and Schonfeld respectively, M. Tisserand finds that .he 

 duration would be increased in this lime by 1 63 hours, tiince 

 the lime^ given b. these observers are 6 5h. and QXih. respec- 

 tively, M. Tisserand is entitled to regard this as confirmation ol 

 hi%hypithesis. lie points out the importance of spectroscopic 

 obsctvalif.n* at short intervals from miminum, in connection 

 with his explanation. .. „ , . 



The irregularities in U Ophiuchi and U Cephei are pro 

 biblyto be explained in the same way. 



THE EXPLOSIVE NATURE OF THE SODIUM 

 AND POTASSIUM DERIVATIVES OF 

 NITIiOM ETHANE. 

 COMK additional inloni.alion cf an inlereMing char.icler con- 

 '^ ccining the extiemely explo»ivc sodium and polas-iuin 

 compounds of nilromctliam-, is contributed to the current 

 Jlnirhltoi ihe German Chemical Society by Prof. Zclinsky ol 

 Mo-cow. A thoil lime ago Prof. Vict ir Meyer dt«cribed 

 (lUricliU, 27, 1601) a mode of preparing the sodium compound 

 CH,NaNO, in a «l.-iie of purity. The process consists in dilut- 

 ing a quan ily of nilromelhane. CHjN< >,. wi.h clhcr and treat- 

 ing the liquid willi a solution ol sodium in alcohol, wh.n the 

 .odium comiK.und i» precipitated. The piccipiiatc requires lo 

 be washed wi.h c.her, and is then dried over oil of vitriol ; the 

 NO. 13 iH, VOL. 51] 



dry compound thus obtained affords numbers on analysis agree- 

 ing with the anhydrous formula above given In a former 

 mf.hod of prepar,tlian described by PiofMeyer, alcoholic 

 To^a was employed as precipitant, but .he sodium n'tromelhane 

 ob.ained invariably contained e.ther water or alcohol ; the use 

 of so.Uum ethvlalc affords it anhydrous. Kven the hjdratei 

 compound fir^t isolated proved .0 be explosive ; but upon 

 placfng a small quan.i.y of it upon a w.-chglass, ami warming 

 over a wa.crba.h, in a short time it '"^"'"^ ^".'''^'="'>. ',°"- 

 verted in.o the anhydro.s compound which mimed uitely 

 exploded with great violence. Wnen a small qt.antily of the 

 anhydrous compound prepared by use of sodium ethylate was 

 placed in a test tube, gently compressed, and .hen warmed, an 

 explosion of so violent a nature occurred that the test-tube was 



I Prof z'linsky has recently had occ.ision to prepare consider- 

 able ouantities of -odium and potas-ium nilromelhane, and has 

 1 had the oppor.unity o( lesting and observing their explosive 

 power upon a larger scale. He appears to have adopted 

 essentially the same process for the preparation of sodium ni.ro- 

 methane as that described by Prof. Meyer, employing an 

 alcoholic solution of sodium eihylate as precipitating reagent. 

 Being desirous of obtaining the sodium compound peviectly 

 anhydrous, an attempt was made .0 achieve this object by 

 use of the «a.er b.aih, but lor the sike precaution only 

 about a gram of .he .subs.ance was employed as a preliminary 

 ?est of the efficacy of this method of dehydration. It was 

 fortunate .hat such was the case, lor «ithin hve minutes an ex- 

 plosion of so violent a nature occurred, that the w^'ch-glas-- 

 ipon which the compound was supported was redviced to 

 powder, and the water bath cons.derably .njured. I""' " 

 demonstrate the explosive nature "f ','^'%'^°'"P°""i;^''!'°,"' 

 danger upon the lecture .able, Pro', /elinsky recommends the 

 fonowing^xperimem :-A th.ck clock-glass or l.etler a s^ul 

 metal plate, is sp.i.ikled with small drops of water, and a very 

 small i>iece of sodium nilromelhane dropped upon 1.. Alter a 

 f..w seconds, provided .he amount of water has not been 

 rxcessive, a deafening deiona.ion occurs, with production of 

 flame and projection of a thick cloud of smoke. The expert 

 ment may be varied by placing the substance upon the P^lectly 

 dry plaic, and invoking its explosion by means of a sm.arl blow 

 with a hard object. . „ \ a -u j 



M Nef has previously (Ann. der Clumic, 280, 2n) described 

 seve.al of the metallic deriva.es of the nitroparaff.ns, and has 

 referred to the instabiU.y of the sodium compound, .and the 

 possibility of occasion.al explo-ions. Prof. Zel.nsUy "o-v siip- 

 plemenis this statement by lemarking that an explosion always 

 results from .he contact of the dry sod.um compound with a 

 minute quan.i.y of wa.er. One of his assistan.s upon one 

 occasion incautiously placed abou. five grams of sodium ni.ro- 

 methane in a glass vessel whose su.face happened to be mo st 

 with the result that a teirilic explosion instantly occurred, wh ch 

 shattered every piece of apparatus upon '^^ '"'•'•■■. »"f"^^=f 

 atmospheric wave pioduced occasioned ,he sudden ex.iuction of 

 lhewhdeoflheg.as flames in .he laboratory. The assistant 

 fortunately escaped more than .rilling injury, but » ^ec"n-i *u.h 

 occurrence might have a very dilleren. result. This incd.nl 

 will doub.less se.ve .0 emphasise the great precaution which is 

 necessary in handling these compounds. 



The potassium compound, ClUKNO,, has been P'cP"ed in 

 a similar manner, and found to I.e even ..yue "";'••>''!;;''•; 

 1 ihe sodium cuipound. explo<l,ng at the ordinary "^"'P' *'""=, 

 ' shortiv alter its isolation. It separates upon the addition of 

 hepoussium ethyl.ite in well defined crjslal-. .Jhe crystalline 

 form^. however, sJon disappears and upon rapidly t--n;^femng 

 10 a filler, an explosion invari.ibly occurs as soon as "'«."'" 

 pound becomes dra.ned Iree ol mos. of .he "-'her ..pior^ 

 The in-tabili.y of the polnssium compound at the orciinary 

 tenrp'a.ure may also bc"readily demons, i.a.ed upon the lec.u^o 

 table It is. ol course, necessary to prepure it freshly on the 

 po., because of .he impossil.ili.y of preserving it for any length 

 of time. An ethereal -oluiion ol nilromelhane is mixed with a 

 Tolu-rn of po.as.siuin e.hyla.e in alcohol, .lie -P"n.a.an liqu d 

 rapidly decanted from the precipitate V'"'^^^'^'^- '^'^ }^\l^\^^^ 

 as quickly as possible bc.ween filler paper, and left quietly 

 "stmg upon "•"= paP". After a lew minutes the substance 

 explodes wi.h a loud de.onaiion. 



These expcrimen.s will serve .0 indicate the «"■«""=" 

 slabili.y of .he alkali-metal derivatives of nilromethane. and 

 the violence of the explosions produced by their ^^^l^^"^; 



