192 



SCIENCE. 



[N. S. Vol. XXIII. No. 579. 



rediscuss the results. This Professor Poor 

 has done. His conclusion is that these ob- 

 servations, when properly interpreted, also 

 confirm his theory of the variable form of the 

 sun, and curves are given which confirm his 

 views. 



It should be stated, however, that the dif- 

 ferences derived from the Gottingen observa- 

 tions are extremely small, and that Ambronn 

 was of the opinion that the differences between 

 his determinations of the polar and equatorial 

 diameters were due to accidental errors. 



RELATION BETWEEN THE MOTION IN THE LINE OF 



SIGHT, AND THE VARIATION IN BRIGHTNESS 



OF VARIABLE STARS. 



The discovery of new variable stars has 

 gone on with increasing rapidity in recent 

 years, until now about three thousand va- 

 riables are known. The majority of these 

 have been found by photographic means at 

 the Harvard Observatory. Fair progress has 

 been made also in the study of the light- 

 curves of these stars, and definitive elements 

 have been found for several hundred of them. 

 It can not be said, however, that great progress 

 has been made in the determination of the 

 underlying causes which produce the varia- 

 tions. In the case of the Algol stars it is 

 sure, both from theoretical considerations, 

 and from spectroscopic determinations of 

 their motion, that the variation is caused by 

 a relatively dark, eclipsing body. The bril- 

 liant lines of incandescent hydrogen, which 

 appear near maximum in the spectra of many 

 stars of long period indicate with considerable 

 certainty that the variations in their light 

 are associated with eruptive disturbances of 

 some sort. It was long ago pointed out that 

 our sun is probably a variable star of long 

 period and small range, and many variable 

 stars may exist, whose changes are caused by 

 spots of greater or less size. Nevertheless, it 

 still remains true that for the great majority 

 of variable stars no sure key has been found 

 to the secret of their changes. 



Two recent bulletins of the Lick Observa- 

 tory (Nos. 62 and 83) have dealt with a new 

 and extremely important phase of this ques- 

 tion. In these. Dr. Ralph H. Curtiss, of that 



observatory, shows that for the well-known 

 variable star W Sagittarii there is a most 

 intimate relation between the velocity of 

 motion in the line of sight and the changes 

 in brightness. This was well indicated by a 

 comparison of the Lick determinations of 

 motion with the early observations of the 

 light-curve by Schmidt. The similarity is 

 most strikingly shown, however, when the 

 velocity-curve is compared directly with the 

 light-curve derived from recent photometric 

 measurements by Professor Pickering, given 

 in the Harvard Annals, Vol. 46, Part 2. A 

 comparison of these curves shows a close re- 

 semblance even in the details, and proves 

 conclusively that both phenomena are due to 

 the same underlying causes. Incidentally a 

 striking proof is furnished of the accuracy 

 of the two widely separated investigations 

 thus critically compared. Dr. Curtiss's work 

 marks a distinct step in advance in pie study 

 of variable stars, and it is to be hoped that 

 the research may be extended to as many and 

 as faint variables as possible. 



S. I. Bailey. 



CURRE^'T ^'0TE8 ON METEOROLOGY. 



BRIEF COMMENT ON RECENT ARTICLES. 



In the recent numbers of the Meteorologische 

 Zeitschrift (Nos. 7-10, Vol. XXII., 1905) 

 there have been many contributions of gen- 

 eral scientific interest, on which the following 

 brief comments are made : 



The exploration of the free air at great 

 heights has been giving records of very low 

 temperatures. On March 2, last, at a height 

 of 9,717 meters, the temperature was — 85.4° 

 C, and on April 4, at 11,010 meters, it was 

 — 79.6° C. These records were obtained by 

 means of hallons-sondes sent up from Vienna. 

 So rapid has been the accumulation of data 

 from the free air that the mean annual tem- 

 perature and the vertical temperature gradi- 

 ents at each interval of one kilometer up to a 

 height of eleven kilometers have been deter- 

 mined, using the results obtained on nearly 

 600 balloon ascents. 



Antarctic meteorology is making rapid 

 progress. On the Swedish expedition, under 



