390 ANNALS NEW YORK ACADEMY OF SCIENCES 



ber 1813 were of the horizontal and 1826 of the vertical diameter. Of 

 this large number, however, only 584 were made on corresponding days 

 and were available for the test. These observations were freed from per- 

 sonal equation, and the resulting series of residuals found. Each Wash- 

 ington residual was multiplied by the corresponding Greenwich residual, 

 and the sum of the products taken with the following results: 



HoR. DiAM. Vert. Diam. 



Sum of products — OM077 — 23''.14 



Number of observations 313 271 



This preponderance of negative products would shov,- that positive 

 residuals at Greenwich are followed five hours later by negative residuals 

 at Washington and vice versa. This would apparently indicate vibrations 

 in the sun of short period, somewheres around ten hours' duration. Again, 

 the result seems, according to Newcomb, to be conclusive against any 

 variability whose period is a multiple of a day. 



The apparent short-period vibration is, however, attributed by the 

 authors to chance. 



AuwERS. — Auwers * enlarged the scope of his investigation, and dis- 

 cussed all available transit material. This included the following long series : 



Greenwich Observations . , , 1851-83 Radcliffe Observations . 1862-83 

 Washington Observations . . 1866-82 Neuchatel Observations . 1862-83 



and a number of shorter series. 



Practically every series showed fluctuations, more or less periodic, in 

 the observed diameters ; and an investigation as to the reality or non-reality 

 of these apparent fluctuations formed the main part of Auwers' papers. 

 These apparent variations fell into two classes; 1st, Irregular or long period 

 variations, and, 2d Annual variations. 



In discussing the variations of the first class, Auwers carefully investi- 

 gated the personal equations of the ninety-two observers who took part in 

 the four main series of observations. These personal errors were deter- 

 mined in the usual manner. In each series the observations (mean of each 

 year) of each observer were compared with the corresponding mean of all 

 the observers, or with that of a standard observer or observers. In the 

 case of the Greenwich series, the mean of four observers, — Dunkin, Ellis, 

 Criswick and J. Carpenter [the observers who had the largest consecutive 



1 Neue Untersuchungen Qber den Durchmesser der Sonne, I, II (Sitzungsberichte of the 

 Berlin Academy, December, 1886 and June 9, 1887). 



