stars, including a Centauri, being chosen for historical associations or for conspicuous brightness, and five others for large proper motion. These stars were judiciously divided between the two observers, so that each had six stars on his list. The comparison stars were chosen with great care, and in the case of the three stars common to the two lists each observer had his special comparison stars, so that we must regard the determinations as entirely independent. Each observer discussed his own observations, and the results were combined in the final account of the research published in the Memoirs of our Society, vol. xlviii. (1884).

The results were valuable, both in their bearing on the exact solution of the problem undertaken and by the effects produced by their attainment, for immediately after the completion of the work Sir David Gill represented to the Admiralty that continued research on stellar parallax and a new determination of the solar parallax by observations of minor planets were much to be desired, if only they could be carried out with a larger instrument provided with certain structural improvements which experience with the small heliometer had suggested. The Lords Commissioners promptly sanctioned the addition of a 7-inch heliometer to the equipment of the Cape Observatory.

Regular observing was begun with the new heliometer in January 1888, and in the next four years an astonishing programme of observations was successfully completed. It included

(1) The complete determination of the constants and errors of the heliometer (scale value, errors of the micrometer screw and of the scale divisions, etc., involving nine months' labour).

(2) Observations for parallax of twenty-two stars.

(3) Observations of the three minor planets, Iris, Victoria, and Sappho, for the determination of solar parallax.

(4) Observations of Jupiter's satellites.

Gill's expectations of the new heliometer were fully realised; he satisfied himself that one observation made with it was of the same weight as three observations with the old instrument, and that one set of observations could be made in half the time that had been previously required.

The tables which summarise (in vol. vii. of the Annals of the Cape Observatory) the results for stellar parallax contain details relating to 12 bright stars and 10 stars of large proper motion; and among the 32 distinct determinations the name of our Medallist is attached to 18, Elkin's to 7, Finlay's to 3, and de Sitter's to 4. In several instances, by way of test, the parallax of the same star was investigated with both instruments, with substantially the same results. Gill completes his exposition of the observations by a discussion of the conclusions to be drawn from the results obtained. So far as so small a number of parallaxes as 22 can be utilised for such a generalisation, the Cape results seem to support the view which has gradually formed in the last twenty years, namely, that large proper motion is a safer criterion of proximity of a star to our system than brightness. We have learnt from the work of

Stumpe, Kapteyn, and Newcomb and others to realise how much emphasis is to be laid on the probability that stars differ enormously in actual luminosity.

The point is beautifully illustrated by the tables which Gill has drawn up. It must be remembered, as he points out, that the derived parallaxes are differential or apparent, not absolute; for we are not at liberty to assume that the comparison stars are infinitely remote. The only legitimate procedure is to deal with limiting values, which can be obtained by adopting different hypotheses as to the absolute parallax of the comparison stars. Gill accordingly gives tables exhibiting the variation of the total light radiated by each of his parallax stars (in terms of the Sun's light), with different hypothetical parallaxes of the comparison stars, ranging from o" oo to o"05. It is thus made clear that the southern stars whose apparent parallaxes have been determined would differ in brightness by about 12-15 magnitudes if they were all removed to the same distance from the Sun.

The further tables which Gill gives to convert the observed proper motions of his parallax stars into absolute velocities in miles per second afford strong confirmation that the deductions about actual luminosity are legitimate. For the velocities deduced are nearly all within the limits which spectroscopic determinations have led us to expect among stars.

The beautiful photographic reproductions, given in the volume of results, of the spectra of a, Centauri and the Sun show what may be expected of the astrophysical department which Sir David Gill has laboured to develope at the Cape. I will not attempt to refer in detail to this spectroscopic work; for though the papers published by the Cape Observatory hitherto have dealt with important points, we are able to gather from references in the annual reports that the main forces are concentrated on yet another determination of the solar parallax by the spectroscopic method. We can well imagine the interest with which Sir David Gill is looking forward to the results of the work.

Cape Catalogues.-Before passing on to refer to Gill's work on solar parallax, I would advert to another of his large contributions to the astronomy of the southern hemisphere, namely, his faithful fostering of the Cape Catalogue tradition. Meridian work must necessarily form an important part of the work of a large national observatory. At the Cape much of the energy of the staff has always been devoted to this branch, and the result is a series of catalogues of star places which have a special value.

When Gill took up his duties at the Cape he found that, in spite of the indefatigable zeal of Mr. Stone, there were two periods (1849-52 and 1861-70) of valuable observations of Sir Thomas Maclear which still remained to be reduced and published. Stone had already begun the reduction of the observations of the first period, and Gill made it one of his first duties to complete and revise the reduction, and in 1884 the Cape Catalogue of 4810 stars for 1850 was published. The last remaining period (1861-70) of

Maclear's observations was dealt with by degrees, and in 1900 the Cape Catalogue of 1905 stars for 18650 was published. We can well understand with what satisfaction the completion of this work was hailed. The preface contains the following remark :-"Thé publication of this catalogue marks an epoch in the history of the observatory. For the first time in that history the Director can feel that the accumulated labours of his predecessors are available for the use of astronomers, and that the work being done under his own direction is in a healthy and forward state of reduction and publication."

The first catalogue for which Sir David Gill is solely responsible is a catalogue of 1713 stars for 18850 published in 1894; the next is a catalogue of 3007 stars for 1890'0 published in 1898. In 1906 two more catalogues were published, an achievement of which the meridian circle department might well be proud. One is a catalogue of 8560 astrographic standard stars for the equinox 1900, to serve for the Cape section of the Astrographic Catalogue. The other is a catalogue of 4464 stars for the equinox 1900, based on observations made between 1900 and 1904, and containing, inter alia, the results of Cape observations of 2798 zodiacal stars which were entered on the list circulated by Sir David Gill, and adopted at the Paris Conference in 1896 as a list of stars suitable for reduction of heliometer observations of planets, etc. In this connection I may refer to another large undertaking. Since 1897 all the oppositions of major planets have been observed with the heliometer at the Cape.

The transit-circle, which has been in use since 1856, was carefully renovated in 1885. The instrument is similar to that at Greenwich. I wish I could have included in these remarks reference to a fifth catalogue, which will deal with other observations, already on the way to completion at the time of Sir David Gill's leaving the Cape. We shall look forward to it with great interest, for it will contain the results of comparison of the old and the new transit-circles at the Cape.

Gill has laboured to embody in the new transit-circle, which was erected in 1901, many new features which were designed with the object of making it possible to detect or eliminate all ascertainable sources of error in observations of fundamental stars.

In connection with his efforts to probe into these refinements of observational work, I would recall to your memory his careful discussion of the meridian observations made for the reduction of his heliometer measures of Mars at Ascension. He then definitely discovered that curious variation of personal equation known as the magnitude equation. In virtue of this troublesome peculiarity, a transit observer watching the passage of bright and faint stars over a wire records the transit a little later for a faint star than for a bright star. The result is that the right ascension interval between a faint and a bright star is a little too large when the bright star leads, and is a little too small when the faint star leads. The effect is a small one, but is very troublesome in cases where

great refinement is needed in star places; the trouble is all the greater because the magnitude of the effect varies appreciably among different observers.

Sir David Gill has introduced in the Cape Catalogues, e.g. that of Astrographic Standard Stars for 19000, corrections for magnitude equation applied separately for each observer. The latest contribution to the experimental study of the effect is one made at the Cape Observatory, and communicated to our Society last April. The authors, Sir David Gill and Mr. Hough, discuss therein the results of a comparison of right ascension differences measured in three independent ways-(1) with the old transit method, (2) with the new transit and travelling wire, and (3) with the heliometer. The results show that within the range of magnitudes (31-8) of the stars observed, the magnitude equation is barely appreciable with the new transit-circle fitted with the travelling wire.

Cape Photographic Durchmusterung.-I turn now to remind you of another great undertaking which we owe to the initiative of our Medallist, viz. the Cape Photographic Durchmusterung. There is scarcely need for me to refer to it in any detail, for the history of it must be vivid in your memory from the account which Dr. Glaisher gave of it in presenting the Medal to Professor Kapteyn in 1902. The Cape Photographic Durchmusterung arose out of the success which Gill achieved in photographing the comet of 1882; he saw the beauty of the many star-images that appeared on the photograph down to the 9th magnitude, and soon afterwards he proved that large photographic cameras could be effectively employed for constructing photographic star maps. He proceeded forthwith to secure the systematic photographical survey of the southern skies, and gave us what may be briefly described as an extension of the Bonn Durchmusterung to the South Pole. Kapteyn's measurement and reduction of the plates remains a splendid instance of scientific devotion. It gives us, under the title of the Cape Photographic Durchmusterung, in three large volumes of the Annals of the Cape Observatory, the approximate places and magnitudes of 454,875 stars,-in fact, a catalogue of all the stars down to the 94 magnitude, and most of those to the 10th magnitude between declination 19° S. and the South Pole. This survey has not only served admirably for a study of stellar distribution in the sky, but by the industrious co-operation of Professor Kapteyn at Groningen and Mr. Innes at the Cape it has also led to the discovery and study of many variable stars, a fact which is attested by observations recorded in the fourth volume, containing the details of the work of revision of the C.P.D.

Astrographic Chart and Catalogue.-The present year is the twenty-first anniversary of the inception of the Astrographic Chart and Catalogue. In the original initiation of this scheme of international co-operation Sir David Gill took a deep interest and also a large share. Admiral Mouchez, by whose tact and energy the initial steps of promoting the harmonious co-operation of

eighteen observatories in different parts of the world were successfully conducted, has testified to the fact that Gill's success in photographing the comet of 1882 opened the way to the Chart, for it proved that we had reached the point of being able to construct the chart of the heavens by photography. The skill of the Brothers Henry, not only in developing the construction of objectglasses large enough and suitably achromatised for stellar photography, but also in producing their superb photographic star charts, was probably the decisive factor in the acceptance of the plan agreed upon in 1887.

The Cape Observatory at once became one of the eighteen contributories to the scheme; the zone -40° to - 52° was allotted as its portion of the sky. The astrographic telescope made by Sir H. Grubb arrived at the Cape in 1890, and after some alterations and experimental researches the work was begun in 1893 and has gone steadily forward. By the end of 1897 the required 1512 catalogue plates had all been obtained and nearly half of the chart plates. In the following year it was decided to re-photograph the whole zone for the catalogue, in order to bring the epoch at which the plates were taken nearer to that at which the comparison stars were observed on the meridian. This plan also rendered it possible to determine with sufficient accuracy the proper motions of the reference stars in the numerous instances where no meridian observations existed previous to the epoch of the latest Cape observations.

In 1900 Sir David Gill undertook, in addition to the zone, to photograph the area within 2° of the South Pole, originally allotted to the Melbourne Observatory; it will thereby be possible to combine the photographic work with the Cape heliometer triangulation of this region, and with the results of the Cape discussions of all the meridian observations of southern close circumpolar stars.

The measurement of the plates was well started in 1898, with the aid of a new type of measuring apparatus designed by Gill (M.N., lix. 61); a second measuring instrument of the same type was afterwards added; and at the time of Sir David Gill's retirement from the directorate in February 1907 he was able to announce that 1086 out of the 1512 catalogue plates had been measured. It is thus seen that the zone undertaken by the Cape Observatory is well on the road to completion.

Solar Parallax.-In these days, since the discovery of minor planets with peculiar orbits like that of Eros, it is not easy to speak of the older methods of determining the solar parallax in such a way as not to appear antiquated. Yet, in a retrospect over the gradual development of the methods, there are points of great interest.

It is just half a century since Airy summarised his views of the relative values of available methods of determining the solar parallax from observations of Venus and Mars. His summary

was given by way of calling attention to approaching favourable oppositions of Mars in 1860, 1862, and 1877, and the two transits of