460 



SCIENCE 



[N. S. Vol. XLVII. No. 1219 



The reasons for these new suggestions are: 

 (1) That the sharpest focus is normally pro- 

 duced by the central portion of the lens and 

 lateral aberrations depend on differences of 

 focus that may result from the passage of 

 light rays through a marginal region of the 

 lens; (2) that the best measure of lateral 

 aberrations are the extreme deviations, and 

 these are those of a point at the edge of the 

 object field through the nearest and through 

 the most distant marginal point of the lens; 

 (3) that the greatest difference of focus of a 

 lateral object between the central image and 

 that produced through an edge point is the 

 one produced by the most distant point on the 

 lens surface and therefore this may most ap- 

 propriately be designated astigmatism; (4) 

 that the focus through the nearest marginal 

 point of the lens may lie on either side of the 

 median focus and if on the same side as that 

 of the distant marginal point there is pro- 

 duced the characteristic optical effect called 

 coma, and finally (5) that these two measure- 

 ments are strictly comparable with the meas- 

 urement always made to determine the longi- 

 tudinal aberration of the axial rays and are 

 therefore the only consistent methods of de- 

 termining the two lateral aberrations. 



c. w. woodwoeth 

 University op California 



observations on the auroral con- 

 vergent, april 5. 1918 



An auroral display of more than usual in- 

 terest occurred on Friday evening, April 5, 

 1918, and was observed by the writer from a 

 point about one and a half miles southeast of 

 the Dominion Observatory, Ottawa. 



At about 10.30 p.m. the rays seemed to be 

 converging at a more or less well-defimed point 

 approximately half way between Saturn and 

 the "Big Dipper." For all that the writer 

 knew the position or path of the point of 

 auroral convergence and its height above the 

 earth's surface had been subject to such fre- 

 quent observation that any measurements he 

 might make on this particular evening would 

 be superfluous, but they seemed to him more 

 worthy of record for a scientific magazine than 



the random descriptions of color, play of light 

 and duration which have recently apjieared 

 and he decided to see whether or not the posi- 

 tion of the point of convergence could be de- 

 termined with any degree of accuracy. 



Exact Western Union time was obtained 

 from " central," but the rough nature of the 

 observations makes the times recorded below 

 approximate only, say within one minute, the 

 fact that they are recorded as 11.20 and 

 11.40 being due not to rough estimation but to 

 choice. The writer used a clothes reel with 

 taut wires, revolving it so that one of the 

 wires intersected both Saturn and the point 

 of convergence. Three small markers (J inch 

 wide) were hung on the wire and moved about 

 until they covered Saturn, the point of con- 

 vergence (convergent), and another known 

 point or star, all in line. 



The following observations were made: 



10.55 P.M. Saturn, convergent, and Mizar in 

 line. 

 Saturn to convergent : convergent to Mizar 

 ::llf:10i. 



11.20 P.M. Saturn, convergent, and star at 

 end of handle of " Big Dipj^r " in line. 



Saturn to convergent : convergent to star 

 ::11|:8| 



11.40 P.M. Saturn, convergent, and point in 

 sky on line from Mizar through end of " Big 

 Dipper" handle and the barest fraction (say 

 one sixth) farther from the end of the handle 

 than that is from Mizar, all in line. 



Saturn to convergent : convergent to point 



::13J:7 



11.55 P.M. Saturn, convergent, and Gamma 



of Bootes in line. 



Saturn to convergent : convergent to Gamma 



: : 15J : 7J 



For the last observation (11.55 p.m.) the 

 rays of light had become faint enough to make 

 the exact position of the convergent somewhat 

 doubtful and measurements were discontinued. 

 In fact the latter observation was taken at 

 11.55 instead of at midnight, which would have 

 preserved the 20-minute interval, because of a 

 fear that the position of the convergent would 

 become too indistinct for observation. 



There was a perceptible tendency for the 



