January 3, 1913] 



SCIENCE 



27 



quartz spectrograph of the Yerkes Observatory. 

 As the spectrograph was to be used with the 2-foot 

 reflector, of which the angular aperture is //4, 

 the collimator should also have such ratio. The 

 form proposed is in all essentials a small reflector 

 of the eassegrain form. The light coming through 

 the slit falls on the perforated concave mirror, 

 which throws the light upon a small perforated 

 convex mirror, mounted just back of the slit. This 

 second mirror is of such curvature that it throws 

 a parallel beam through the perforation of the 

 first mirror upon the prism. The ratio of perfora- 

 tion to aperture in both cases is equal to the ratio 

 of aperture of the objective of the telescope to 

 the diagonal mirror, and the aperture of the con- 

 vex mirror is equal to the perforation of the con- 

 cave, so there is no loss of light. The perforation 

 of the convex mirror centered behind the slit is of 

 suflScient size to permit the use of a slit one eighth 

 of an inch long. 



Systematic Motions of the Stars: Benjamin Boss. 

 Diagrams representing the distribution of stars 

 of the different spectral types and down to the 

 6.5 magnitude were exhibited. The striking pref- 

 erence of the B type stars for the plane of the 

 galaxy, the similar, though somewhat less pro- 

 nounced, preference of the A type for the same 

 plane, and the more random distribution of the 

 more advanced types was clearly shown. We must 

 either conclude that eventually a state of approxi- 

 mately random motion must prevail, or that the 

 later type stars originated uniformly over the 

 entire sky, rather than selecting the plane of the 

 galaxy. The facts revealed by the study of the 

 motions of the stars indicate that the first view 

 is the correct one (A. J., 635-636). The early 

 type stars show a decided crowding toward certain 

 portions of the galaxy, toward the vertices of 

 preferential motion, as discovered by Kapteyn, 

 and toward the apex and antapex of a group of 

 stars called Group IV., or the antapex group, 

 discovered by the writer (A. J., 635). That the 

 motion of this group is in the direction of the 

 antapex of solar motion, and that there seems to 

 be a tendency of motion in the opposite direction, 

 may indicate the existence of a solar group. The 

 analysis of the motions within these groups is now 

 in process of development by the writer. 



Oiservations of Variable Stars at the Vassar Col- 

 lege Observatory : Caroline E. Fdrness. 

 Observations of variable stars were begrm at 

 this observatory in 1901. The telescopes in use 

 had apertures of twelve, five and three inches. 



A photographic wedge photometer which could be 

 attached to the twelve-inch was in use for part 

 of the time, but for the most part the comparisons 

 were made by the Argelander method. The ob- 

 serving list included chiefly long-period variables 

 which were selected so that they might pass 

 through a maximum or minimum during the time 

 of observation. The predictions were usually 

 taken from Hartwig 's ephemeris. The standard 

 of magnitude was the Harvard photometry, other 

 sources being reduced to this scale. 



The resulting observations, about five thousand 

 in number, with a full discussion of the curves 

 and periods of the stars so far as they may be 

 determined, are to be published as soon as their 

 revision is~cDmplete.' 



On the Cause of the Earth's Magnetic Field: 



L. A. Bauer. 



The communication is confined to a considera- 

 tion of the portion of the field which is symmet- 

 rical about both the axis of rotation and the 

 equator. If the intensity of the magnetization be 

 computed for each parallel of latitude over the 

 region best covered by magnetic observations, 

 60° N. to 60° S., on the basis that the average 

 values of the observed magnetic components along 

 a parallel correspond to those of a uniform mag- 

 netization parallel to the axis of rotation, then a 

 systematic and regular increase in the intensity 

 is observed in both hemispheres with approach to 

 the equator. This may be stated mathematically 

 thus : If X , be the magnetic component along a 

 true meridian, positive toward the north, and Z 

 be the vertical component, positive downward, then 

 for the field, symmetrical as stated, the com- 

 ponents, for any co-latitude u, will be 



X^fxiu) sin « and Z = f~{u) cos w. 

 These "characteristic functions" show an in- 

 crease of 13 to 17 per cent, from the parallel of 

 60° to the equator. 



None of the theories thus far advaiced to 

 account for the origin of the earth's magnetic 

 field gives the law of the observed increase in the 

 characteristic functions. With but few exceptions, 

 the attempts have been restricted to ascertaining 

 a cause for a uniform magnetization. Since this 

 course has not led to generally accepted results, it 

 was decided to begin with the geographical varia- 

 tions, hoping that, if their cause be determined, 

 valuable clues might be found as to the origin of 

 the primary field. Thus a closed physical expres- 



■ See Popular Astronomy, 20, 645, 1912. 



