Jaxuary 3, 1913] 



SCIENCE 



35 



from plate to plate and with direction on the 

 same plate. The erratic variations seem now to 

 be eliminated. At present, the point of greatest 

 interest in photographic photometry relates to the 

 determination of an absolute scale. The first 

 determination of magnitudes has been for stars of 

 the polar sequence. Fifteen plates with the wire 

 gauze screen and diaphragms of 32 and 14 inches, 

 affording 27 separate determinations of the scale, 

 have been used. The exposures range from 1 to 

 11 minutes, the magnitudes from 8.8 to 17.6, 

 although the limits for reliable results are 10.5 

 to 15.5. The mean deviations of groups of stars 

 from results obtained at the Harvard Observatory 

 are as follows: 



The zero point of the Mt. Wilson scale is such 

 that the sum of the deviations from the Harvard 

 scale between 10.5 and 15.5 is zero. The cause 

 for the divergence beyond mag. 15.5 is not yet 

 established, but may be settled by comparing the 

 scale here given with that derived from plates of 

 longer exposure. 



Variable Asteroids: S. I. Bailey. 



In connection with the photometric measure- 

 ments of Eros, made at Arequipa in the years 

 1902 and 1903, observations of Ceres, Parthenope, 

 Massalia, Kalliope, Urania, Harmonia, Melete, 

 Hecuba, Sirona, Baucis, Celuta, Chryseis, Kal- 

 listo, Kleopatra, Germania, Bamberga, Tercidina, 

 Melusina, Ilmator, Aquitania, Aurelia, Eros and 

 Tergeste, were undertaken especially for the detec- 

 tion of new cases of variability in light. The 

 Eumford photometer was used with the 13 -inch 

 Boyden telescope. Five eases of variability, in 

 addition to Eros, were detected: Urania (30); 

 indications of variability, no satisfactory period 

 found. Hecuba (108) ; indications of a small 

 range of variation, having a period which appears 

 to be a sub-multiple of 0''.99, perhaps 0''.330. 

 Sirona (116) ; Indication of range of half a mag- 

 nitude with period 0''.403. Celuta (186); varia- 

 tion small, if real, period 0''.364 satisfies all ob- 



servations. Tercidina (345) ; small variation, 

 satisfied with period 0''.366. 



The light curves of Eros independently deter- 

 mined by the Eumford photometer and by photo- 

 graphs made with the Bruce telescope agree 

 closely, but those obtained with the photographs 

 have the smaller accidental errors. The photo- 

 graphic method offers many advantages. In order 

 that photographs shall serve for the accurate 

 determination of magnitudes, it is very desirable 

 that the images of the asteroids shall be com- 

 parable with those of the stars. This can be ac- 

 complished by giving to the plate, while the tele- 

 scope is moving at sidereal rate, a motion equal 

 to one half the apparent motion of the asteroid. 



Besults of Latitude Oiservaiions at the Flower 

 Observatory from December, 190^, to July, 



1911 : C. L. DOOLITTLE. 



The series here considered consists of 13,852 

 determinations of latitude made with the zenith 

 telescope, and 11,591 with the Wharton reflex 

 zenith tube. All of the observations were made 

 by myself and the two series are practically simul- 

 taneous. A comparison of the results given by 

 the two instruments furnishes data, to be found 

 nowhere else, so far as I know, for examining the 

 daily fluctuations to be found in all extended 

 series of this character. 



For this purpose we have 1,540 comparisons. 

 We find 149 cases where the residuals from both 

 series are at least twice the probable error. Of 

 these, 79 have like signs and 50 unlike. There are 

 12 cases where both residuals are at least three 

 times the probable error; 10 of these have like 

 signs. In these cases the residuals are four times 

 the probable error and in one case six times the 

 probable error. All of these have like signs. 



It is obvious that we have to do with a class of 

 errors other than those due to observations, and 

 that before much greater progress can be hoped 

 for in this direction, means must be found for 

 this elimination.^ 



The Constant of Aberration: C. L. Doolittle. 



Prom the observations carried on at the Sayre 

 and Flower Observatories from December 1, 1889, 

 to June 8, 1911, with the zenith telescope and the 

 reflex zenith tube there have been derived 22 

 values of this constant. Although these range in 

 value from 20". 448 to 20". 605, the computed value 

 of the probable error in no case differs very much 

 from 0".01. If we give all equal weight we have 



= See A. J., 641. 



