March 23, 1906.] 



SCIENCE. 



457 



during totality was placed in a nearly hori- 

 zontal position above the mirror, and not 

 between the lens and mirror, as is cus- 

 tomary. The sunshade over the building 

 and tube, and the wall of the tube, which 

 was on the opposite side from the sunshade, 

 were all whitewashed with ordinary lime. 

 The sunshade was of opaque black cloth, 

 and all the inner surfaces were black. 



At the time of the eclipse eight exposures 

 were made as follows: 2^, 5^.2, 13^.2, 

 l'°28=.l, 38^3-, 6«.8, 5« and an exposure of 

 one fourth second about ten seconds after 

 third contact. All the negatives are sharp 

 and clean. The unusual completeness of 

 the series gives every part of the corona in 

 fine detail on one or more of the plates. 

 The first exposure shows a very large group 

 of prominences. The short exposure made 

 after the reappearance of the sun shows 

 the entire ring of the corona. On this 

 plate the prominences on the bright limb 

 are shown in fine detail, and from it good 

 measures may be made of the height of the 

 chromosphere. 



The first and fourth contacts were ob- 

 served visually from the image on the face 

 of the focal plane shutter, and recorded by 

 a stop-watch. I think that this method of 

 observing the outer contacts has advantages 

 over the ordinary method of observing the 

 limb with a small telescope. The times of 

 the exposures were marked on the chrono- 

 graph by a key operated by an assistant 

 who looked after the coelostat clock, set the 

 mirror, etc. 



The plates were thoroughly backed with 

 water-color lampblack, which was carefully 

 wiped off before development. Distilled 

 water was used in developing and fixing, 

 and after washing, the plates were rinsed 

 in distilled water. 



Polarized Coronal Light, August 30, 1905: 

 C. D. Perrine. 

 Observations of polarized light in the 



corona were made by the Lick Observatory- 

 Crocker Eclipse expedition to Alhama, 

 Spain. A double-image prism was used 

 in connection with a camera of twenty and 

 three fourths inches focal length, in con- 

 tinuation of the work at Sumatra in 1901. 

 More efficient apparatus was designed for 

 the recent eclipse, and plane glass mirrors 

 were used as analyzers. These cameras 

 were of fifty' inches focal length and yielded 

 sharp images of good scale for photometric 

 measurement. Strong radial polarization 

 is shown on all the negatives. 



Anomalous Refraction: Frank Schles- 



iNGER and C. B. Blair. 



Under ordinary circumstances the atmos- 

 phere is disposed in horizontal strata of 

 uniform density; but these strata may be 

 inclined occasionally to the earth's surface, 

 giving rise to anomalous refraction. 



In this case the usual expression for re- 

 fraction should be replaced by a somewhat 

 different one, deduced by the writer. The 

 new formula has been applied to observa- 

 tions made in 1900 and 1901 at the six 

 international latitude stations, with the ob- 

 ject of discovering whether the effect of 

 anomalous refraction is large enough to 

 justify the frequent, but vague, appeals 

 which have been made to this source of 

 error in order to account for inconsistencies 

 in meridian work. From this discussion it 

 appears that observers have little to fear 

 from this cause. 



An important corollary to this conclusion 

 relates to Kimura's recently discovered 

 term in the latitude variation, independent 

 of longitude. The present paper makes it 

 highly probable that this phenomenon has 

 a real existence and is not due, as has been 

 suggested, to anomalous refraction. 



Neiv Variable Stars in the Small Magel- 

 lanic Clouds: Henrietta S. Lbavitt. 

 At the meeting of the Astronomical and 



Astrophysical Society held in Philadelphia 



