158 



KNOWLEDGE 



[August 1, 1891. 



one class to another, may be expected to prove instructive 

 as to the essential characters of each. The very fact that 

 such migratory instances are to be met with is in itself 

 significant. It assures us, at any rate, that red and white 

 stars are not the products of widely separated epochs of 

 cosmical history. But besides this piece of negative 

 information, much positive intelligence should be derivable 

 fi-om them. The co-ordination of spectral and colour- 

 change has never yet been satisfactorily accomplished. 

 The two kinds of variation are so far united that although 

 the former might conceivably take place without the 

 latter, the latter inevitably involves the former. Thus a 

 star spectrum might, although it is unlikely that it would, 

 vary fundamentally in character without any attendant 

 variation in tint. This would be the case, for example, if 

 emergent bands or lines were so situated as to be comple- 

 mentary one to the other. And in point of fact, relatively 

 trifling modifications of colour appear to have been 

 concomitant with the striking spectral changes detected 

 by Mr. Espin in R Corona; and E Scuti. But coloiu-- 

 variation munt be explicable, so to speak, by spectral 

 variation. Enquiries on this head are superfluous, their 

 upshot being self-evident. What needs to be investigated 

 is the form and manner of a correspondence which unques- 

 tionably exists. In this direction next to nothing has been 

 done ; to the questions that suggest themselves no answers . 

 are forthcoming ; yet by their means, if at all, the enigma 

 of star-colours ought to prove soluble. A thorough 

 examination of a single colour-variable in its slowly 

 alternating red and pale phases could hardly fail to 

 disclose the essential condition of the peculiarity of its 

 light. What supervenes, one desires to know, in the 

 atmosphere of that star, now to blanch, and again to 

 flush its rays ? Are they subtracted from by additional 

 absorption, or reinforced by special but transient 

 emissions ? Is continuous absorption included among 

 the elements of change '? That is to say, does the dusky 

 veil thrown over the upper part of the spectrum appreciably 

 lift or lighten with the pahng of colour '? The providing 

 of definite replies to these definite queries would in itself 

 make a solid beginning of knowledge as regards the cause 

 of redness in stars. But they can only be provided by the 

 persevering exertions of some one competent observer. 

 Such disjointed notes of colour as have been hitherto 

 visually recorded are of little use except in the way of 

 suggestion. They have furnished the means of con- 

 structing a working hst of objects more or less vehemently 

 suspected of change, and thus served their main purpose 

 of prescribing the aim, and limiting the scope, of a fresh 

 series of more concentrated operations. Half-a-dozen red 

 stars assiduously watched would probably be foimd more 

 genuinely commimicative than hundreds passed in review, 

 and then abandoned until perhaps recovered after a decade 

 or two by some other collector of celestial curiosities, 

 whose perplexities at certain incongruous results of his 

 search would remain as unprofitable as those of his 

 predecessor. But the half-dozen stars chosen for detailed 

 scrutiny should be tested spectroscopically and spectro- 

 graphicaUy, no less than visually. The photographic 

 delineation of the spectra of small red stars would no 

 doubt demand large instruments ; yet without it the 

 enquiry would be lamentably incomplete. Indeed, the 

 tell-tale modifications looked for would be more likely to 

 present themselves in the higher than in the lower reaches 

 of the spectrum. From its exceptional brightness among 

 colour-variables, ;■ Velorum might be singled out as a 

 particularly tempting subject for investigations of the 

 kind described, were not astro-physical observers and 

 apparatus equally scarce m the southern hemisphere. 



^mwt Column. 



By W. Montagu Gattie, B.A.Oxon. 



c 



Eefusino to Overtrump. 



ASES frequently arise in which it is not advisable 

 to overtrump. The following hand furnishes a 

 simple illustration. 



Hand No. 22. 



Score — Love all. 



Z turns up the three of hearts. 



Note. — A and B are partners against Y and Z. A has 

 the first lead ; Z is the dealer. The card of the leader to 

 each trick is indicated by an arrow. 



V^ 



Tricks— AB, 1 ; YZ, 0. 



Trick 3. 

 B 



9~~^ 



TrtrLs—AB, 3 ; YZ, 1. 



TricKs—AB, 4 ; YZ, 1. 

 Trick 7. 



Tricks— AB, 6 i YZ, I. 



Tricks— AB, 7 ; YZ, J. 



