XXII .
On the Periods of the Changes of Light in the Star Algol .
In a Letter from John Goodricke , Esq. to the Rev. Anthony Shepherd , D. D. F. R. S. Professor of Astronomy at Cambridge .
Read April 1 , 1784 .
SIR , York , Dec. 8 , 1783 .
ASI am now able , by collating some of my late observations on Algol with those I sent you last May , to determine with greater precision the periodical return of tis changes , I wish to add this as a kind of supplement to that account .
The method I have here pursued is by taking the intervals between accurate observations of Algol 's least brightness or greatest diminution of light made at long distances of time from each other , and dividing those intervals by a certain number of revolutions , as will be best understood by the table below .
The reason of my chusing long intervals is , that the number of revolutions being greater , the errors of observation are thereby diminished : all error cannot , however , as yet be excluded , but I think the period is now , by the following calculation , ascertained within ten or fifteen seconds .
Mean times of Algol 's least brightness .
1783 h. ' d. h. ' ' ' Jan. 14 9 25 an interval of 99 revolutions , each of 2 20 49 14 Oct. 25 6 39 Jan. 14 9 25 Ditto 106 Ditto 2 20 49 10 Nov. 14 8 17 Jan. 14 9 25 Ditto 107 Ditto 2 20 49 2 Nov. 17 4 52 Feb. 68 15 Ditto 91 Ditto 2 20 49 3 Oct. 25 6 39 Feb. 68 15 Ditto 98 Ditto 2 20 48 59 Nov. 14 8 17 Feb. 68 15 Ditto 99 Ditto 2 20 48 51 Nov. 17 4 52 Feb. 26 9 43 Ditto 84 Ditto 2 20 49 14 Oct. 25 6 39 Feb. 26 9 43 Ditto 91 Ditto 2 20 49 9 Nov. 14 8 17 Feb. 26 9 43 Ditto 92 Ditto 2 20 49 0 Nov. 17 4 52 Jan. 31 14 29 Ditto 100 Ditto 2 20 49 4 Nov. 14 8 17 Mar. 21 8 36 Ditto 84 Ditto 2 20 48 46 Nov. 17 4 52 Hence the period of Algol 's variation is , on a mean , 2 20 49 3 I could have added several more comparisons of the like kind ; but these are , I think , sufficient .
It is to be remembered , that all the observations contained in the above table are reduced to mean time .
It appears to me now , that the duration of the variation is about eight hours ; but , as it is difficult to hit exactly the beginning and end of the variation , this may occasion different observers to differ in this respect .
Before I conclude , I beg leave to mention a circumstance deserving of notice ; which is , that FLAMSTEAD has also amongst other stars observed Algol , and in two places has marked it of less magnitude than at other times , viz. of the third magnitude , 1696 , January 16 .
6 h. 24~ , and 1711 , December 5 .
9 h. 13~ , both mean time and old stile * .
Suspecting these might probably be days of Algol 's variation , I computed the interval between them , but could not find a period answerable to that which I have above determined .
Upon examining more closely the observations , I find , in that of 1696 , he marked at the same time the magnitude of ~ Persei ; which , considering especially the nearness of ~ Persei to Algol , makes this observation to be relied on for its justness , and less liable to any mistake of judgement ; whereas the other observation of December 5 , 1711 , is more liable to error or doubtfulness , because he did not then mark the magnitude of ~ Persei , or of any star of the same magnitude near enough to Algol .
Presuming , therefore , on the justness of FLAMSTEAD'S observation of 1696 , to think that it probably was made at a time when Algol varied , I compared it with one of mine , viz. October 25 .
6 h. 39~ , 1783 , and I find there is , in the interval between those observations , either 11,176 periods , each of 2 d. 20 h. 49~ 18~ ; or 11,177 , each of 2 d. 20 h. 48~ 56~ .
The last , as it approaches nearest to the results of my best observations , I think , is the exactest determination of the period .
This , however , all proceeds upon the supposition that Algol varied at the time of FLAMSTEAD'S observation , and also that the period is regular .
* Historia Caelestis , vol. II , edit 1725 , p. 284 .
and 534 .
The following is a short abstract of my late observations on Algol , when its least magnitude was accurately determined .
August 17 , 1783 .
App. time .
h. ' 10 52 About equal to ~ Persei , though Algol seemed to be rather brighter .
11 7 Evidently less than ~ Persei .
11 22 Ditto ; but rather difficult to distinguish them from each other .
11 30 Rather brighter than ~ , and not so bright as ~ Persei .
12 0 About the brightness of ~ Persei , and rather less than ~ Trianguli .
12 30 Brighter than ~ Persei , and rather not so bright as ~ Trianguli .
From those observations , by taking a mean between 11 h. 7~ and 11 h. 22~ , it appears , that its least brightness happened at 11 h. 14~ ; true , I think , to 5~ .
October 25 .
6 40 It was considerably less than ~ Persei .
75 Ditto .
7 20 Equal to ~ Persei , though Algol seemed rather less .
7 35 About equal to ~ Persei .
7 50 Brighter than ~ , and also than ~ Persei .
8 25 About the third magnitude , and equal to ~ Trianguli .
9 35 Between the second and third magnitude ; brighter than ~ Arietis , and rather less than ~ Pegasi .
10 10 About the second magnitude ; rather brighter than ~ Pegasi , rather less than ~ Cassiopeae , and not so bright as ~ and ~ Cassiopeae .
App. time .
h. ' 10 40 Rather brighter than ~ Cassiopeae , but less than ~ and ~ .
11 0 Nearly equal to , if not rather brighter than , ~ Cassiopeae , and less than ~ Cassiopeae * .
In 20~ afterwards it was of the same brightness ; hence we may conclude , that the variation has ended at 11 h. 0~ .
Its least brightness from the observations appears to have happened at 6 h. 55~ ; true , I think , to 10~ .
November 11 .
10 5 Third magnitude ; not much different from ~ Persei and ~ Trianguli .
10 45 Between the third and fourth magnitude ; believe equal to ~ Persei .
11 14 Less than ~ Persei .
11 48 Ditto ; but think it rather increased .
Its least brightness from those observations appears to have happened at 11 h. 31~ ; true , I believe , to a quarter of an hour .
The weather was rather hazy .
November 14 .
50 Between the second and third magnitude , and less than ~ Cassiopeae .
5 45 A little brighter than ~ Arietis .
6 50 Not so bright as ~ Arietis , and rather brighter than ~ Trianguli .
8 10 A little brighter than ~ Persei , and believe equal to ~ Persei .
* Algol 's usual and greatest brightness , by my later and more accurate observations , is thus : a little less than ~ Cassiopeae , brighter than ~ Cassiopeae and ~ Pegasi , and rather a little brighter than ~ Cassiopeae .
App. time .
h. ' 8 25 Less than ~ Persei .
8 40 Ditto .
90 Equal to ~ , though Algol appeared rather brighter .
9 15 A little brighter than ~ and ~ Persei .
By taking a mean between 8 h. 25~ and 8h 40~ , it appears , its least brightness happened at 8 h. 32~ ; true to 10 minutes .
The weather was rather hazy during some part of this observation .
November 17 .
4 58 A little less than ~ Persei .
5 15 Ditto .
5 35 Rather brighter than ~ Persei .
5 50 A little brighter than ~ Persei , but less than ~ Persei .
65 Rather brighter than ~ Persei .
6 40 Equal to ~ Trianguli , and brighter than ~ and ~ Persei .
7 20 A little brighter than ~ Arietis .
8 30 Between the second and third magnitude , and equal to ~ Cassiopeae , but less than ~ and ~ .
8 50 Second magnitude , and equal to ~ Cassiopeae .
9 25 Nearly the same , if not rather brighter .
The variation has therefore ended at 9 h. 0~ nearly , and its least brightness by taking a mean between 4 h. 58~ and 5 h. 15~ , happened at 5 h. 7~ ; true , I believe , to 10 minutes .
The weather was fine .
I have several more observations on Algol , where I have not been able to ascertain its least brightness , which all happened agreeable to the period as above determined ; viz. May 20 .
July 5 .
and 22 .
August 14 .
September 6 .
9 .
12 .
and 26 .
October 2 .
5 .
19 .
and 22 .
and December 7 .

