Todd and Baker — Predictions for the Total Eclipse. 245 



Art. XV III — Local Predictions for the Total Eclipse of the 



Sun, 1907, January 13-14-, in Turkestan and Mongolia • 



by David Todd and Robert H. Baker. 



[Contributions from Amherst College Observatory — Ixxiii.] 



Six total eclipses of the sun happen during the next six 

 years, and it might be expected that contributions to knowledge 

 of the corona would be correspondingly ample. 



Of these eclipses, however, the tracks of those of 1908 and 1911 

 are wholly confined to the Pacific Ocean, with the possibility of 

 observing stations on only two or three difficult islands ; the 

 eclipse of 1909 is too near the north pole and that of 1910 too 

 near the south pole for ready observation ; totality of 1912 can 

 perhaps be excellently obtained in Brazil ;* but of them all, 

 that of 1907 seems least uncertain to yield significant results. 



Following the last eclipse (1905, August 30) by an interval 

 of seventeen months, the figure and type of the corona will 

 doubtless have changed completely ; so that it is in the highest 

 degree important to photograph this totality. 



Fortunately, the track of the eclipse of 1907, January 13- 

 14, is wholly on land. But a good part of the region visited is 

 so remote and difficult of access, in Mongolia and the Gobi 

 desert, that it can be occupied only by equipping tedious and 

 expensive expeditions. Only one station in Mongolia, Tsair- 

 osu ( see table below given), seems likely to be considered. 



But the western half of the track crosses Turkestan, a trans- 

 'Caspian region penetrated by the imperial railways of Russia. 

 For travelers from the United States, it can readily be reached 

 by Naples, Constantinople, the Black Sea, Tiflis, the Caspian 

 Sea, Bokhara and Samarkand. 



For Europeans a convenient route would be Berlin, Warsaw, 

 Moscow, Samara, Orenburg and Tashkent. On this railway 

 and about two-thirds of the way from Tashkent to Samarkand 

 lies Jizak, only a few miles from the exact line of central 

 eclipse. Other easily accessible places near Jizak, and well 

 within the belt of totality, are Zaamin, Nau and Ura-tiube. 

 The last is practically central. If we go farther east, the track 

 of the eclipse leads into a region more and more difficult in 

 every way, although the totality is a few seconds longer there, 

 and the eclipsed sun a few degrees higher. 



In order to exhibit the exact circumstances of the eclipse, 

 throughout the entire length of its track, the indicated data 

 have been calculated for eleven stations, as exhibited in the 

 following table. Also the results of the calculated example in 

 the British Nautical Almanac, for a station between Yarkand 

 and Cherchen, are included. f 



The computations are based on the Besselian elements of the 



American Ephemeris, and the geographical positions of the 



towns have, for the most part, been obtained from a map 



recently published by the Carnegie Institution of Washington.;}: 



* Todd, Total Eclipses of the Sun (Boston, 1900), p 249. 



f The Nautical Almanac for the Meridian of Greenwich, 1907, p. 594-5. 



\ Pumpelly and Davis, Explorations in Turkestan (Washington, 1905), p. 157. 



