644 POPULAR SCIENCE MONTHLY. 



make up a world even one-hundredth as large as the earth would take 

 hundreds of thousands of such worlds. 



On the night of August 13, 1898, Herr Witt made a photograph of 

 the region near /? Aquarii, with an exposure of two hours. He 

 wished to obtain an observation of a known asteroid which had not 

 been observed for nine years, and which his calculations assigned to 

 that region. When developed and examined on the following day, 

 the plate not only showed the object desired, and also a second known 

 asteroid, but a faint and long trail of some unknown object. From 

 its rapid motion it was at first thought to be a comet, but an exami- 

 nation on the following night with a visual telescope revealed its 

 true nature. As soon as the well-known computer of minor planet 

 orbits, Herr Berberich, had computed its approximate orbit, the as- 

 tonishing nature of the new planet became apparent. Of all the pre- 

 viously known members of the solar system, with the obvious exception 

 of our moon, Venus and Mars approach nearest to the earth. Venus 

 is distant from us at the most favorable times about twenty-five million 

 miles, and Mars thirty-five million miles. Eros, however, approaches 

 the earth at the most favorable oppositions within less than fourteen 

 million miles, so that he is our nearest celestial neighbor. This leads 

 to a solution, under better conditions perhaps than ever before 

 granted, of that fundamental problem in astronomy, the distance of 

 the sun, or, in other words, the determination of the solar parallax. 

 In order to determine the orbit and position of a planet, certain quan- 

 tities must be found, based upon at least three observations of the 

 planet's place in the sky. It is, however, highly desirable to have 

 more than three observations of the planet's position and to have them 

 widely separated in time. 



The following elements for Eros were computed by Dr. S. C. 

 Chandler, and were based on the observations of 1898, combined with 

 those of the Harvard photographs made in the years 1893, 1894 

 and 1896: 



EPOCH 1898, AUGUST 31.5, GREENWICH MEAN TIME. 



Mean Anomaly 221° 35' 45. "6 



Perihelion Distance of Ascending Node 177 37 56. CM 



Longitude of Ascending Node 303 31 57. 1 > 1898.0 



Inclination of Orbit to Ecliptic 10 50 11. 8) 



Angle whose Sin is the Eccentricity 12 52 9. 8 



Mean Daily Motion 2015."2326 



Logarithm of Semi-major Axis 0.1637876 



Period of Revolution around Sun 643 d . 10 



Later observations will doubtless slightly modify these results, but 

 they are sufficiently precise for our purpose. These elements were 

 published in December, 1898, and well illustrate the enormous pho- 

 tographic resources which at the present time are in the possession 

 of the Harvard Observatory. Twenty years ago, the present Director, 



