December 23, 1915] 



NATURE 



455 



in the fourth diagram I have brought the observa- 

 tions to the same scale by dealing- with per- 

 centages instead of the actual number of stars 

 as a base line. In both cases it will be seen that 

 the shape of the curve is in harmony with the 

 analogy to which I referred, and I think I am 

 justified in adding that since the curves are so 

 similar, although they have been obtained from 

 such different sources, not only is the method of 

 inquiry amply justified, but the principles on 

 which the stars have been classified and separated 

 into the various ascending and descending groups 

 are shown to be fundamentally sound. This, of 

 course, goes to strengthen my main contention 

 that the life-history of a star is a rise, followed 

 by a fall, of temperature. 



Further inquiries are necessary before all 

 the conclusions to be drawn from a com- 

 parison of the two curves can be arrived at, 

 but some are already suggested 



One point on which information might be ex- 



ARGONIAN 

 ALNITAMIAN 



inclusion of solid meteorites in stars classified in 

 the Ibwer part of the ascending curve give rise, 

 in some cases, to large masses, while on the oppo- 

 site side of the curve we should expect densities 

 above the normal. 



Since my classification is based upon absorption 

 phenomena, the curves do not include the stars 

 in which bright lines alone have been studied, but 

 I am continuing my inquiries concerning them, 

 and it looks as if they will ultimately be found to 

 represent special classes of nebulae — that is, 

 swarms in which something besides an ordinary 

 condensation is taking place. 



I have already shown (Nature, February 4, 

 1915, and 2nd Bulletin of the 'Hill Observatory) 

 that the stars with constant bright lines added to 

 absorption phenomena occur near the top of the 

 temperature curve. 



I have to thank Mr. H. E. Goodson, A.R.C.Sc, 

 assistant at the Hill Observatory, for the prepara- 

 tion of the curves. Norman Lockyer. 



Ptrtenuge of Stars m Respective Groups. 



Fig. 4.— Temperature Curves based on the percentages of stars in the separate 

 ' and " 470 Catalogue." 



Stars in " Catalogue of 470 Brighter Stars. 

 Stars in list in *' Bulletin III." 



[pected from the method under notice is an indica- 



jtion of the various maximum temperatures 



attained by the individual stars. If all the stars 



[reached the highest temperature we should expect 



broader apex in the average result than 



lif the number were restricted. A first glance 



it the composite curve shows that the number 



the highest temperature is very much restricted. 



1^ are justified, therefore, in suggesting that 



Jose groups containing a relatively larger number 



stars represent the highest temperature condi- 



)n of a larger number. On the other hand, the 



iarly perpendicular part of the curve in the middle 



the ascending side as contrasted with the slowly 



jscending curve on the descending side indicates 



le action of a rapidly increasing temperature in 



»e former. The kinks in the ascending curve, 



lerefore, at the Aldebarian and Crucian stages 



both groups of stars possess great interest from 



iiese points of view. 



When a larger number of stars have been 



issified, and further evidence is afforded that 



ire are absolutely dealing with homogeneous 



roups, this average temperature curve may help 



in the study of masses and densities. We 



lould expect, for instance, that the probable 



NO. 2408, VOL. 96] 



A PLIOCENE FLOTSAM.^ 



BOTANISTS expect from 

 the authors of this in- 

 ALGOLiAN stalment of the results of their 

 MARKABiAN sustained investigation of the 

 siRiAN Pliocene vegetable deposits of 



PROCYONiAN *^^ delta of the Rhine, then 

 much more extensive than 

 now, work of a high standard. 

 Those who may be led to study 

 *" •*•' the work before us will find 



that this standard has been 

 groups in " Bulletin III. fully Sustained, and will be left 

 , in doubt whether to admire 



most the care and patience in 

 the examination and arrange- 

 ment of their material to which the text and the 

 plates which illustrate it bear testimony, the 

 clearness with which their evidence is tabulated, 

 or the caution and restraint with which their con- 

 clusions are formulated. 



We are already indebted to our authors for 

 various contributions to our knowledge of similar 

 deposits associated with the brick-earth of Tegelen 

 on the Meuse, and in that case they have been 

 able to show from other evidence that the deposits 

 in question are, at least approximately, contem- 

 porary with the Norwich Crag, and perhaps in 

 part with the somewhat more recent forest-bed 

 of Cromer. The present instalment deals with 

 material from an older horizon, belonging to the 

 middle, not the upper Pliocene, obtained from 

 brick-clay pits near Reuver and Swalmen, and 

 some distance further south at Brunssum, and 

 the circumstance that, owing to the lack of col- 

 lateral evidence, it is impossible yet to suggest a 

 corresponding formation west of the North Sea, 

 in no way detracts from the interest which their 

 results arouse. 



1 " Mededellingen van He Rijksopsporlnd van delfstoffen." No. 6. " The 

 Pliocene Floras^ of the Dutch-Prussian Border." By Clement Reid and 

 Eleanor M. Reid. _ Pp. im+xx plates. (The Hague : Institute for the 

 Geological Exploration of the Netherlands, 1915.) 



