500 



NA TURE 



{April 5, 1877 



degree of precision, many varied arrangements as well as expe- 

 rienced assistants. That idea has secured me the protection of 

 His Imperial Highness the Grand Duke Constantine and the 

 support of the Impenal Russian Technical Society. 



I thus conclude this communication which I have the honour 

 to send you. The researches on the co-efficients of dilatation of 

 gases in their general features confirm the accuracy of the deduc- 

 tions drawn from the observations on compressibility. But that 

 matter is in process of elaboration, and only a part of the 

 researches has been completed. Consequently I abstain from 

 expounding all the details of the subject. I shall only remark 

 that the true coefficient of dilatation of air under constant pres- 

 sure and with variable volume is found to be greater for pressures 

 near to atmospheric pressure than the number generally accepted, 

 notably from the researches which I have made with M. Kaian- 

 der, it is equal to o 003683 if we take 100° C, as the temperature 

 of boiling water under an atmospheric pressure of 760 mm. in 

 latitude 45°. 



If you do not find devoid of interest more ample details on this 

 subject, as also on the determination of the weight of a litre of 

 air, I shall have the greatest pleasure in explaining to the 

 English public through your interesting journal the essential 

 points of the researches made on this subject in my laboratory. 



St. Petersburg, January I De Mendeleef 



OUR ASTRONOMICAL COLUMN 

 Binary Stars. — From one of the very careful and complete 

 investigations on the orbits of the revolving double stars, by 

 which Dr. Doberck is so greatly contributing to our knowledge 

 of the motions of these interesting objects, we have an orbit of 

 f Scorpii which is probably a very near approximation to the 

 true one. 



We have called the star by what appears to be its correct 

 designation, ^ Scorpii, but few stars have been subjected to 

 more varied nomenclature than the one in question. In Dr. 

 Doberck's paper, published in No. 2,121 of the Astronomische 

 Nachrichten, in all probability through one of the typographical 

 errors which have of late so much disfigured this periodical, the 

 star is styled ^ Libras. It has been previously very commonly 

 lettered | Libras, and it is 51 Librae of Flamsteed ; Secchi, in 

 Astron. Nach., No. 1,614, calls it p' Librae, though we are 

 ignorant upon what precedent. 



Dr. Doberck's elements are those of a very nearly circular 

 orbit, and it may be remembered that some thirty years since 

 Madler gave elements for circular motion with a period of revo- 

 lution of about 104 years. In Astron. Nach., No. 1,683, ^^• 

 Thiele gave the results of a very complete discussion of the 

 measures up to 1856, in which he has assumed that Sir W. 

 Herschel's angle on the night of his discovery of the duplicity of 

 the star has been registered correctly, though a doubt has been 

 entertained upon this point. He thus arrives at a highly excen- 

 trical elliptic orbit with a period of a little over forty-nine years. 

 It should be remarked that from the near approach of the mag- 

 nitudes of the components forming the double star | Scorpii (it 

 is more correctly a triple star), an error of 180° in the measured 

 angle of position is by no means an improbable one. 



Dr. Doberck assumes that Sir W, Herschel's angle of 1782 

 requires this correction, and deduces an orbit in which the period 

 of revolution is nearly twice that of Thiele, and which therefore 

 approaches the period originally assigned by Madler. His ele- 

 ments are as follow : — 



Peri-astron passage 1859 62. 



Node ... 12° 15' Inclination 



Node to peri-astron on orbit 



Eccentricity 



Semi-axis major ... 



68» 42' 



89° 16' 



0-0768 



I "•26 



Revolution 95 '90 years. 



A full comparison with the measures up to the present time, 

 appears in the Astronomische Nachrichten. 



In the same number Dr. Doberck gives first elements of that 

 exceedingly difficult object 7 Coronae Borealis, in which the 



period of revolution is 95I years, and the peri-astron passage 

 18437. The distance calculated from this orbit is still under 

 two-tenths of a second, but it will increase, until towards the 

 end of the first decade of the next century the components, 

 according to Dr. Doberck's calculation, will be separated by 

 more than o" "8. 



The Annular Eclipse of the Sun, 1737, March i. — 

 Prof. Grant, in his " History of Physical Astronomy," mentions 

 this eclipse as the first annular one of which we have any detailed 

 account. This phase passed over Edinburgh, where it was ob- 

 served by Maclaurin, the well-known mathematician, by Short, 

 the optician, Lord Aberdour, and others. The times were 

 determined by Maclaurin by a pendulum clock of Graham's, 

 and he was also furnished with a meridian instrument by the 

 same maker, with the aid of which the clock was rated by Short 

 for " a long time before and after the eclipse." The clock used 

 by Lord Aberdour, who was located in Edinburgh Castle, was 

 compared with Maclaurin's at noon on the day of the eclipse, 

 and in addition signals were exchanged between the Castle and 

 Maclaurin's station at the college. Both observers determined 

 the duration of the annular phase to have been 5m. 48s. 



The following elements of this eclipse have been deduced from 

 a similar system of computation as regards the moon's place to 

 that adopted for other eclipses to which reference has been made 

 from time to time in this column, a system which gives results 

 for the total solar eclipse of 17 15 agreeing very closely with the 

 observations of Flamsteed and Halley. 



G.M.T. of conjunction in R. A., March i, at 3h. im. 31s. 



R.A 



Moon's hourly motion in R. A. 

 Sun's „ „ „ 



M oon's declination 



Sun's „ 



Moon's hourly motion in decl. 

 Sun's „ „ „ 



Moon's horizontal parallax . . . 

 Sun's „ „ 



Moon's true semi-diameter ... 

 Sun's ,, ,, 



The equation of time was 12m. 40s. subtractive from mean time. 

 The following were points upon the central line by the above 

 elements : — 



Long. 10 10 W. Lat. 54 53 N. 

 7 26 W. „ 55 42 N. 



Long. 4 1 1 W. Lat. 56 35 N. 

 „ o 4W. „ 57 36N. 



If reduction equations are founded upon a direct calculation i 

 for Edinburgh, there result for any place not far distant : — 



CosTO = 27'6369 — [i 59611] sin / + [i'22oi8] cos/, cos (£. - 51° i''4) 

 t — 2h. 13m. 3S'5S. ^ [2-29350] sin in + [3-49325] sin / 



- [3-89289] cos /, cos (L + 147° 35"S)- 



where / is the geocentric latitude of the place, L its longitude 

 from Greenwich + if E, — if W, and t is expressed in Green- 

 wich mean time ; the upper sign is to be used for beginning of 

 annular phase, and the lower for the ending. The quantities 

 within square brackets are logarithms. 



The calculated duration of annularity at Edinburgh 

 5m. 46-63., differing only r'4?. from the observations of Mac- 

 laurin and Lord Aberdour, but the middle of this phase is givenl 

 later by im. 51s. At other places mentioned in Maclaurin's 

 memoir on this eclipse, published in the Philosophical TransaC'\ 

 tions, the duration of the annulus was as follows : — At Alnwick,| 

 2m. OS. ; at Crosby, near Ayr, 5m. 54s. ; at Montrose, 6m. 27s. ; 

 and at St. Andrew's, 6m. 12s. At Aberdeen, which was verj 

 near the central line, the annulus was formed at 3h. 43m. 

 local mean time, and continued 6m. 30s, On the east coast 

 Scotland, where the duration of annulus was longest, it did no 

 exceed 6m. 35. The eclipse is not given annular at Morpet 

 therein agreeing with the observation. 



