fication of star spectra.* He there suggested a rough classification of the spectra into three groups: "First, those having many lines and resembling the Sun, all reddish or golden stars; secondly, white stars, like Sirius; the third group comprising Spica, Rigel, etc., also white stars, but showing no lines.'

In the course of his observations with the spectroscope Mr. Rutherfurd discovered its use for making a photographic objectglass, and in 1864 he succeeded in devising and constructing an 1-inch glass (Am. J. of Sc., 1865 May) of fifteen feet focal length, corrected for the photographic rays, with which many very successful photographs were taken of the Sun, Moon, and stars. In 1868 he finished a 13-inch-an ordinary achromatic, to which a third flint lens could be attached in a few minutes when it was desired to use the telescope photographically. The beautiful photographs taken with these instruments were in 1890 presented to the Columbia College Observatory, together with twenty folio volumes, of about 200 pages each, containing the measures of many of the plates. A complete catalogue of them, given by Mr. J. K. Rees (Annals of the N.Y. Acad. of Sc. vol. vi. 1891 June) shows 175 solar, 435 lunar, and 664 stellar plates, and 174 plates of the solar spectrum. The measures were made by Rutherfurd with micrometers of his own devising. In the first he used a long screw, and that the measures even with this instrument were excellent was shown by Dr. B. A. Gould in two papers read to the Nat. Acad. of Sc. in 1866 and 1870. The merest summary of these papers was published at the time, and the details were not printed till many years after. In 1870 Dr. Gould was, as is well known, called to Cordoba for (as he then thought) three years, and he left the publication of the star measures till his return. He took with him the 11-inch telescope, but the object-glass was unfortunately cracked in transit. He put together the pieces and made some photographs of the southern stars, but afterwards obtained another lens. It was fifteen years before Dr. Gould returned to the States, and could arrange for the printing of the two memoirs by the National Academy. The real investigator, Mr. Rutherfurd, meanwhile made no effort towards this consummation, partly

* It should not be forgotten that Fraunhofer, using a prism before the object-glass, observed the spectra of Capella, Betelgeux, Procyon, and Pollux, in which he recognised the solar D; and also of Castor and Sirius, which were different. These results in 1823 were really the first to show that the spectra of stars differ from each other and from that of the Sun. Immediately after Kirchhoff's discovery of the true meaning of the Fraunhofer lines, Donati in Florence, Rutherfurd in the United States, and Huggins and Miller in England, all worked independently (and unknown to each other) at the spectra of stars and planets. Donati published a paper on stellar spectra in the Annali del Muso Fiorentino for 1862, but his method and results were very imperfect. When Dr. Huggins and Dr. Miller arrived at the Royal Society, on 1863 February 26, to read their first paper (dated 1863 February 17), they were surprised to find the copy of Silliman's Journal containing Rutherfurd's paper mentioned in the text.

from ill-health and partly from that dislike to the act of publication only too prevalent among eminent scientific men. But the long-screw micrometer was not satisfactory, and Rutherfurd was led to devise one with a glass scale, the principle of which is now too well known to need description. The measures made with the glass-scale micrometer are at last being reduced and published by the Columbia College, and Mr. H. Jacoby has shown in two memoirs (on the Pleiades and the ẞ Cygni group) how excellent are photographs and measures alike. Thus the chief part of Rutherfurd's astronomical work is only now being made generally accessible.

During 1870 Mr. Rutherfurd constructed a ruling-engine, described and figured in Appleton's Cyclopædia. With this beautiful apparatus he produced superb interference gratings on glass and on speculum metal. Some of the ruled plates had 17,000 lines to the inch; they were superior to all others down to the time when Professor Rowland perfected his machine.

Among his other work may be specially mentioned his photograph of the solar spectrum by using a bisulphide of carbon prism (1864); his method of adjusting a battery of prisms to minimum deviation (1865); his demonstration of the stability of a collodion film (1872); and his project for using a glass divided circle instead of one of metal (1876).

Mr. Rutherfurd was one of the founders of the National Academy of Sciences. He took a most important part in the International Meridian Conference at Washington in 1885. For more than twenty years he was a member of the Board of Trustees of Columbia College, and received the degree of LL.D. at the Centenary in 1887. Weak health prevented his accepting many scientific engagements in which he was invited to take a prominent position. For twenty years past he was compelled to spend his winters either in Florida or in the south of France. He contracted a severe cold on his journey south in 1891, and never fully recovered from its effects. He died peacefully, and without apparent pain, on 1892 May 30, in his seventy-sixth year. He was elected an Associate of this Society on 1872 November 8.

PROCEEDINGS OF OBSERVATORIES.

The following reports of the proceedings of observatories during the past year have been received from the directors of the several observatories, who are alone responsible for the

same:

Royal Observatory, Greenwich.

The past year has been an exceptionally heavy one in various ways, and has somewhat severely taxed the energies of the staff. Two new assistants were appointed in February to fill vacancies which had existed for some time; and the observatory was for some months without the services of a clerk. Important changes in the buildings have occupied much of the attention of the Astronomer Royal; and, finally, two important longitude operations have required nearly the whole time of Mr. Hollis during three quarters of the year, and a considerable portion of that of the Chief Assistant.

It might therefore be expected that the regular work would necessarily suffer to some extent; but it is reassuring to find that the arrears (chiefly in the printing are not serious, and will probably be effaced early in 1893. The total number of transits observed in the year 1892 is 6,212, and of zenith-distances 5,607, which numbers compare favourably with those for any previous The total number of stars observed in 1892 was 1,722. year. The reductions for the Five-year Catalogue of Fundamental Stars from observations in the years 1887-1891 are completed, and the Catalogue will be published as an appendix to the volume of Greenwich Observations for 1891. Meanwhile the resulting star-places have been communicated to the Superintendent of the Nautical Almanac and used by him in the Nautical Almanac for 1896 (just published).

The N.P.D.'s of the Ten-year Catalogue have been analysed by comparing the results for the separate years with the mean. The comparison suggested periodic fluctuations of latitude of the kind found by Mr. Chandler in his recent investigations; and on a more direct reduction of the Greenwich observations from 1851 to the present time being undertaken by Mr. Thackeray. Mr. Chandler's doubly periodic variation of the latitude (in about 365 and 427 days respectively) was very satisfactorily confirmed. The results have been given in two papers communicated to the Society by Mr. Thackeray (Monthly Notices, liii. pp. 2, 120).

The mean error in R.A. of Hansen's Lunar Tables, with Newcomb's corrections, as deduced from III observations with the Transit-Circle in 1892 is + o's069, agreeing well with the results for previous years since 1883.

The

Recent observations of horizontal flexure by means of the collimators show that the coefficient is still very small. apparent correction to nadir observations deduced from reflection observations of stars is o" 25, a rather large quantity. No correction for flexure, or for the above nadir discordance, has been applied to the observations apart from the R-D correction.

Observations with the altazimuth were intermitted during the months May to October, owing to the great pressure of other work. In the remaining months the Moon was observed regularly during the first and last quarters. The proposition to replace this instrument by a Universal Transit-Circle, as described by the Astronomer Royal (Monthly Notices, lii. p. 480), has been sanctioned by the Admiralty.

Five comets have been observed during 1892 with the Sheepshanks Equatoreal, viz.: Comet (a) 1892 (Swift) on fourteen nights; (c) 1892 (Winnecke) on one night; (d) 1892 (Brooks) on nine nights; (ƒ) 1892 (Holmes) on nine nights, and (g) 1892 (Brooks) on three nights.

Twenty-five occultations of stars (including seven disappearances and three reappearances of stars during the lunar eclipse of 1892 May 23); the occultation of Uranus by the Moon on July 3; the occultation of 73 Piscium by Jupiter on May 23; and 49 phenomena of Jupiter's satellites, have been observed with the Equatoreals, by one or more observers.

With the Astro-photographic Equatoreal 489 plates, involving 1,222 exposures, have been taken during the past year. The total time of exposure, exclusive of trails for orientation, was 127h 8m. Of the plates, 88 were Chart and 220 Catalogue plates, 67 were for instrumental adjustment, 75 of Nova Auriga, 13 of the Pleiades, 16 of various regions, and 10 were attempts at Chart plates cut short by clouds. The plates of Nova Auriga have been measured for determination of the magnitude of this star at different epochs, and the results have been communicated to the Society. Four of the miscellaneous plates were exposed on Jupiter, a concave enlarging lens (specially made by Dallmeyer) being used to magnify the image about 15 times. The photograph micrometer used for the measurement of Transit of Venus (1874) photographs has been modified for use with the stellar plates. The measures can be made quite independently of the réseau, which will be an advantage if this auxiliary has ultimately to be dispensed with, owing to the decomposition of the film, which it seems difficult to prevent. As regards the selection of guiding stars for the plates of the chart:-Since the date of the last report the remainder of the catalogue for zone +60° to +65° (Rome) has been sent to Father Denza; the

U

second 12 hours of R.A. of the zone +25° to +29° (Oxford) has been sent to Professor Pritchard; the catalogue for 12h-18h R.A. for the zone -3° to -5° (San Fernando) has been completed, and the approximate star-places for R.A. oh - 12h and R.A. 18h -24h have been communicated to Dr. Becker, of Strassburg, who is determining accurate places by meridian observation.

The 12-inch Merz refractor, dismounted in 1891 November to make room for the new 28-inch refractor, was mounted in place of the Lassell reflector on 1892 April 13, the Thompson photo-heliograph being attached to the tube. In its present position, however, it cannot be effectively used owing to the interference of the new building of the proposed Physical Observatory, on the central tower of which it will, as is hoped, soon be mounted, together with the Lassell dome. The old south-east dome was dismounted in November, and operations for the erection of the new 36-foot dome which is to cover the 28-inch refractor, were commenced by Messrs. T. Cooke & Sons in the middle of December. No spectroscopic observations have

been made during the year.

Photographs of the Sun have been taken with the Dallmeyer 4-inch photo-heliograph on 210 days, and of these 518 have been selected for preservation, including 29 photographs with a double image of the Sun, taken to determine the position of the wires with reference to the parallel of declination.

For the first three months of the year the Dallmeyer photoheliograph was mounted on the first floor of the new museum, from which it was removed on April 6 to its old position in the wooden dome south of the Photographic offices. Here it remained during the summer months until the Sun was no longer sufficiently high to be observed over the roof of the new museum building, and on September 9 the photo-heliograph was once again transferred to the new museum.

Photographs of the Sun have also been taken with the Thompson photo-heliograph on 35 days during the year. new enlarging lens has lately been supplied for this instrument by Messrs. Ross & Co., with very satisfactory results as regards flatness of field and freedom from distortion. Photographs have also been received from India and Mauritius up to December 6, leaving seven days in the year ending on that day for which no photograph is yet available for measurement.

The increase in the solar activity which was so marked in 1891 has continued during 1892. The Sun has not been free from spots on a single day; as many as ten or eleven distinct groups have often been seen on the disc at the same time. Several of the groups observed have been of very great size and complexity, and one (that of Feb. 5-18) was by far the largest ever photographed at the observatory. It is premature yet to decide whether the actual maximum has been attained, but the great increase in the number of the spots appears to point to its being near at hand.