MR. E. R. PRATT, of Ryston Hall, Norfolk, writes an interesting article on the East Anglian timber willow in the recently published Journal of the Royal Agricultural Society. The supply of timber suitable for their trade has in recent years caused manufacturers of cricket-bats some anxiety. In East Anglia, and apparently in other parts of England, all large willows have been felled, except those kept by landowners for ornamental purposes, and the price of timber good enough for bat-making has risen to 5s. per foot. Two varieties of willow are purchased, the "closebark," which is considered much the best, and the " openbark." Growers of willows have found it a difficult matter to ascertain what variety the bat-maker wants, as his descriptions of the tree have been very vague. Mr. Pratt has gone carefully into the question of variety, and has examined a great many willows in the eastern counties. On the authority of a botanist who has given special attention to the genus (Rev. E. F. Linton, of Edmondsham, Salisbury), he states that the "close-bark willow" is not Salix alba, but S. viridis, a variable hybrid between S. alba and S. fragilis. Many of Mr. Pratt's specimens closely approached the former species, but could always be distinguished by the bronze-red winter shoots. He believes that the genuine S. alba, of which he has cultivated specimens obtained from Kew, is very rare in the eastern counties. Mr. Pratt further states that the " open-bark willow of the bat-maker is S. fragilis, the crack willow, or its variety, S. Russelliana, the Bedford willow. In his experience S. viridis is much more common in East Anglia than S. fragilis.

IN the Engineering Review (vol. xiv., No. 5) illustrations are given of the works at Notodden where the synthesis of nitrates from the air has been found commercially practicable.

SOME very useful hints for horseback travel and transport are given in an article by Mr. F. L. Waldo on outfitting for the prospecting trail in northern Mexico in the Engineering Magazine (vol. xxxi., No. 1). During a ten years' residence in Mexico the author's attention has many times been directed to the incongruity of the outfits prepared and brought into that country by those whose business or pleasure call them into the Sierras. While the suggestions given refer specifically to a certain region, many of them will be of value elsewhere.

A BULLETIN (vol. iii., No. 54) has been issued by the Department of Agriculture, Madras, describing experiments on well irrigation made by Prof. A. Chatterton at Metrosapuram in 1902-5. The results show that with adequate pumping power it is possible to improve the water supply and to cultivate a very considerable tract of land from a single well. If in the future oil engines and pumps are extensively used for well irrigation in India, it appears certain that the 3-inch centrifugal pump will be most largely employed, and that such a pump will water six acres per day of twelve hours, and will be suitable for areas ranging from thirty to fifty acres.

THE annual progress report of the Geological Survey of Western Australia for 1905 (Perth, 1906) shows that much valuable work has been done. The Wodgina tinfield has been carefully examined, and it is believed that it will prove an important tin and tantalite producer. About I cwt. of tantalite specimens have been presented to the Survey museum. During the year there was a sudden demand for tantalum ores which had hitherto been considered useless. Considerable interest has been taken locally in deposits of graphite, and those on the Donnelly

River have been worked to a slight extent. A sample of laterite iron ore from Comet Vale, North Coolgardie, proved of great interest on account of the occurrence in it of a notable proportion of chromium, mostly in the form of a hydrate. A large portion of the report is devoted to the results of examinations of the various goldfields, full reports of which will be published in due course.

THE weather report of the Meteorological Office for the week ending Saturday, May 19, shows that the recent rains were excessive in places, while in other parts of the United Kingdom the rainfall was below the average. In the east of Scotland and in the north-east of England the aggregate fall was at least four times the average. The measurements due to the exceptionally heavy rains of Saturday. May 19, were -2.53 inches at North Shields, 2.40 inches at Alnwick Castle, and 2-23 inches at Seaham. Both France and the Spanish Peninsula participated in the heavy rains of Saturday, the measurement for the twenty-four hours at Lyons being 2.25 inches, and at Corunna 2:40 inches.

WE have received copies of the Boletim mensal of the Observatory of Rio de Janeiro, issued under the auspices of the Ministry of Industry. Anyone wishing to study the climate of that part of the Atlantic shore lying between the mouth of the Amazons and Rio de Janeiro will find trustworthy statistics for several of the coastal stations. The data are chiefly for ten-day means, with monthly means and extremes, but for Rio de Janeiro the actual observations for three-hourly periods are given in addition, and furnish most valuable details for all meteorological elements.

THE Republic of Uruguay has recently established a National Institute for Weather Prediction, with its central observatory at Monte Video; the meteorological observatory at that place was founded by the municipal authorities in 1895. Observations have been made at several stations for some years, and the new institution has commenced its operations by the collation and discussion of the means and extremes already available, and by the investigation of the characteristics of the severe storms which affect the navigation of the estuary of the Rio de La Plata. The most dangerous storms are those from the south-east, as they usually occur with a rising barometer, in connection with anticyclonic conditions over the Atlantic, and are frequently accompanied by thick fog on the coast. The first number of the bulletin of the institute contains an exposition of the hydrography of the estuary, and tables showing, inter alia, the effect of the various winds upon the

tides of the river.

Two reports have recently been issued on rates of deck watches and of box and pocket chronometers on trial for purchase by the Board of Admiralty at the Royal Obseryatory, Greenwich, in the latter half of last year. The number of deck watches on trial from August 5 to November 25, 1905, was 125, and the makers of the first five in the list, in which the watches are arranged in order of merit, are:-(1) W. Potts and Son, Leeds; (2) L. Hall, Louth, Lines; (3) S. D. Neill, Belfast; (4) and (5) J. Player and Son, Coventry. The makers of the first five box chronometers of those on trial from June 17, 1905, to January 6, 1906, are:-(1) Kullberg, London; (2) and (3) Johannsen, London; (4) Lilley and Son, London; (5) M. F. Dent, London. In the same period the makers of the leading five pocket chronometers are:-(1) Newsome and Co., Coventry; (2) and (3) Kullberg, London; (4) Lindqvist, London; (5) Newsome and Co., Coventry,

PROF. B. WALTER states in a brief note published in the Annalen der Physik (vol. xix., p. 874) that the ultra-violet portion of the spectrum of a high-tension arc in air shows a series of bands identical with those observed by Eder in 1892 as characterising the combustion of ammonia, and considered by him as ammonia bands. It would appear probable that these bands are to be attributed rather to an oxide of nitrogen, produced in both cases, than to the cause suggested by Eder.

SOME SUCCESSful attempts, made in the geophysical laboratory of the Carnegie Institution, to prepare small plates of quartz glass suitable for the construction of lenses, mirrors, or other optical apparatus, are described by Messrs. Arthur L. Day and E. S. Shepherd in Science (vol. xxiii., No. 591). The glass obtained was nearly free from air bubbles, and was only slightly discoloured by the presence of silicon. The conditions for obtaining such a material by the fusion, in a small graphite box, of pure crystallised quartz or tridymite, are summarised follows: an initial temperature of 2000° or more without pressure, so as to allow of the production of sufficient vapour to drive out the air between the grains, followed by pressure (at least 500 lb.) and a reduced temperature of about 1800°, with time for the quartz to flow compactly together without being attacked by the graphite.

as

As interesting contribution to the study of fluorescence is contained in a paper published by Mr. Harry W. Morse in the Proceedings of the American Academy of Arts and Sciences (vol. xli., No. 27) under the title "Studies of Fluorite." The fluorescence and thermoluminescence of fluorite and the nature of the gaseous and liquid inclusions in fluorspar are dealt with under different headings. The fluorescence spectra shown by fluorite when excited by the light of the condensed electric spark between electrodes of certain metals contain sharp lines and narrow bands; the lines of these fluorescence spectra do not appear to belong to any known substance, and are remarkable inasmuch as different lines are obtained with different exciting sources. The spectrum also varies sharply from crystal to crystal with the same means of excitement. The cause of fluorescence, whatever be its nature, is removed or destroyed by heating at a temperature of about 300° C. At the same temperature the colouring matter of the different varieties of fluorite is destroyed; the nature of this colouring matter is discussed by reference to the gaseous products liberated at higher temperatures. As these consist of hydrogen, carbon monoxide, and carbon dioxide, the colouring matter would appear to be organic in its origin; the gases are probably produced by its undergoing a process of destructive distillation

No. 3. vol. xxiii., of the Astrophysical Journal contains an important paper by Mr. Theodore Lyman on the extreme ultra-violet spectrum of hydrogen. Part of this spectrum was previously photographed and investigated by Dr. Schumann, whose work was briefly described in vol. Ixix. (p. 262) of NATURE. Unfortunately Dr. Schumann, although he photographed the spectrum down to λ 1270, was unable to give the wave-lengths beyond A 1850, but this omission has now been rectified by Mr. Lyman, who | has not only determined the missing wave-lengths, but has also extended the known spectrum down to A 1030 (see NATURE, p. 45, vol. Ixix., and p. 110, vol. lxx.). In the present paper the author describes the apparatus and methods employed in great detail, in the hope that an exact knowledge of the conditions necessary to success may prove of value to investigators who work in this

field "; he also gives reproductions of his spectrograms, with wave-length scales, for the region between A 1670 and A 1270.

THE Watkins Meter Co., Hereford, has published a third edition of "The Watkins Manual of Exposure and Development, by Mr. Alfred Watkins.

THE report for the year 1905 of the council of the Hampstead Scientific Society has been received. The Christmas juvenile lectures, and those on nature-study, intended to encourage the teaching of this subject to children, proved very successful. Among lectures delivered at the general meetings of the society may be mentioned those by Prof. Marcus Hartog, on the end and beginning of individuality as shown in the living cell; Dr. R. S. Clay, on the peculiarities and paradoxes of fluid pressure; Sir Samuel Wilks, F.R.S., on spirals; Dr. C. W. Andrews, F.R.S., on fossil hunting in the Libyan Desert; and Mr. F. W. Rudler, on the geology and scenery of the British Isles.

THE Seventy-second annual report of Bootham School (York) Natural History, Literary, and Polytechnic Society shows that the pupils of this school continue to receive every encouragement to devote their leisure hours to the outdoor study of natural phenomena. During 1905 the boys were particularly successful in discovering rare plants, and though we have been unable to find a specific caution in the report, we trust that all observers are urged not to uproot plants or in any other way to assist the disappearance of rare species. It is satisfactory to find that attention is given to many branches of natural science so that the predilections of as many boys as possible may be satisfied.

OUR ASTRONOMICAL COLUMN.

SPECTRUM OF NOVA AQUILE No. 2.-A visual observation of the spectrum of Nova Aquila No. 2, made at the Lick Observatory on September 5, 1905, showed a number of bands, the brightest of which was recognised as HB. Hy and a band near A 4600 were distinguished with difficulty owing to their extreme faintness.

Photographs obtained with the one-prism spectrograph on September 6 and 10 (exposures three and four hours respectively) confirmed the visual observation, the intensities of the bands at A 4600 and Hy being respectively onefifth and one-tenth that of HB. H8 was also seen, but was very faint.

A faint continuous spectrum was seen to extend from about A 4500 to the region of the Hy band. He and the so-called nebular lines were not visible.

Visual and photographic observations made on October 11, 1905, agreed in showing a marked diminution in the brightness of HB, which was then no brighter than Hy (Astrophysical Journal, vol. xxiii., No. 3).

A number of magnitude observations of this Nova, made on various dates between September 20 and November 24 at the Utrecht Observatory, are recorded in No. 4089 of the Astronomische Nachrichten by Dr. A. A. Nijland.

STEREO-COMPARATOR DISCOVERIES OF PROPER MOTIONS.At a meeting of the Paris Academy of Sciences, held on May 7, Prof. Loewy announced that Prof. Max Wolf had met with considerable success in discovering and measuring stellar proper motions by means of his stereo

comparator.

In one instance a star of known proper motion was seen to be obviously displaced after the very short interval of four years. When the two photographs, taken at this interval, were placed in the stereoscope, the star in question was seen to stand out in a plane considerably different from that in which the neighbouring stars appeared to be

set.

Prof. Wolf has also been able to show that a ninthmagnitude star in the constellation Leo has a proper

motion hitherto unsuspected, and he has obtained a value for the motion which he believes to be more correct than could be determined by ordinary micrometric measures. In this case a period of fourteen years separated the times of taking the photographs (Comptes rendus, No. 19, 1906). MEASURES OF DOUBLE AND MULTIPLE STARS.-The

a

measures of 1066 double and multiple stars are published in vol. ii., part iii. (astronomical series), of the Publications of the University of Pennsylvania by Prof. Doolittle. The measures were made with wire micrometer attached to the 18-inch refractor of the Flower Observatory, and include, among others, 733 Burnham stars, 109 O2, and 102 Σ stars. Four hundred and ninety-two stars from Prof. Hough's catalogue have also been remeasured but are not included; it is Prof. Doolittle's intention to re-measure all the stars discovered by this observer.

The micrometer, the corrections of the instrument, and the method of observing are all fully discussed in the present publication, which is a continuation of part iii., vol. i.

Part ., vol. ii., of the same publications gives the results of the observations made with the zenith telescope of the Flower Observatory from October 1, 1901, to December 28, 1903, and also contains a re-discussion of the 1896-1898 series, of which the details appeared in part ii., vol. i., in 1899.

OBSERVATIONS OF COMET 1905c.-Numerous observations of comet 1905c are recorded in No. 4090 of the Astronomische Nachrichten.

This object was observed at Vienna from December 17, 1905, to January 14, 1906, and during that time its apparent diameter increased from 2' to 4'-5', the length of its tail from 5' to 40', and its total magnitude from 9.5 to 4.0. On December 30 a nucleus of magnitude 6.0 was observed.

Heliometer observations at the Cape Observatory showed the comet as a faint nebulous mass with no visible nucleus. Observations of position were recorded from February 5 to February 20, 1906.

The observations at Strassburg Observatory extended over the period December 10 to March 21, and the apparent position, the total magnitude, and the diameter were recorded on eleven different dates.

A LUNAR TIDE ON LAKE HURON.-Whilst examining the curves showing the periodical oscillations of the seiches on Lake Huron, Prof. W. J. Loudon, of Toronto University, was struck by the regularity of their general outline, which seemed more marked in calm weather. Further investigation of the matter showed a well-marked and regular rise and fall twice a day, and also showed that no oscillation of the lake could have a period of more than four hours.

From these facts Prof. Loudon concluded that a true lunar tide occurs on Lake Huron, a conclusion which his further experiments seemed to verify.

THE HAUNTS OF THE OKAPI.

ACCORDING to a report in Monday's Times (May 21)

the expedition to the Congo Free State under the charge of Captains Boyd Alexander and C. B. Gosling has been successful, not only in procuring a fine skin (and it may be hoped a skeleton) of the okapi, but likewise in obtaining some important particulars with regard to the habits of this animal. The specimen, which it is stated will ultimately find a home in the Natural History Branch of the British Museum, was obtained at Bima, on the River Welle, in the northern territory of the Congo State. It is mentioned in the letter that the animal was seen alive by the expedition, but further particulars on this point are desirable, as it is not stated whether anyone but the Portuguese collector by whom it was trapped had this good fortune. The animal was caught in a pit according to native fashion, previous attempts to shoot it having proved ineffectual.

Hitherto the only definite account of the kind of country inhabited by the okapi and the probable nature of its food is one given by Mr. J. David under the title of "Weitere Mitteilungen über das Okapi," and published in vol. lxxxvi. of Globus (1904). Captain Alexander's notes, which differ

in some respects from the former, are therefore of great value and interest, and may be quoted in full.

"The okapi here is generally found singly or in pairs, but Mobatti hunters state that sometimes three may be found together. An essential to the life of the okapi is a small stream of water with some muddy and swamp、 In this grows a certain large leaf ground on either side. that on its single stalk attains a height of 10 feet. It is the young leaf of this plant that is the favourite food of the okapi, and I venture to say that where the plant is not to be found the animal will not exist. During the night he will wander along in the mud and water in search of it. Here he may be found feeding as late as 8 a.m. in the morning, after which he retires to the seclusion of the forest, where he remains until nearly dusk. On the three occasions that I was at close quarters with the beast, he was perfectly concealed in this swamp leaf. Near the River Welle I found his spoor on ground frequented by buffalo and waterbuck, but this is unusual, and his companions in the forest are the elephant, the greater bushbuck, the yellow-backed and small red duikers. The okapi is very quick of hearing, and in that respect is classed by the Mobatti with the bushbuck (local name bungana ). In the forest here I consider this latter beast to be more difficult to obtain than the former. On the hunting ground of the first village that I visited I estimated the number of okapi as five or six, at the second and third nil; and twenty miles south in the forest, on very likely ground where my guide said they were formerly numerous, there was one only, probably owing to rubber-collectors who had been there."

Several specimens had been speared, shot, or trapped by natives shortly before the date of Captain Alexander's visit, but time did not admit of further investigation. The sex of the new specimen is not stated, but it is to be hoped that it will prove to be a male, as Sir Harry Johnston's example, now exhibited in the Natural History Museum, is a female. A pair of okapis are exhibited in the Congo Free State Museum at Tervueren, near Brussels, which also possesses other skins; and there are likewise a few other examples in Europe, notably one in Italy and another in Mr. Rothschild's museum at Tring. It is a great pity that the Belgian Government does not take immediate steps to publish coloured figures of its specimens in order to aid in solving the question as to whether there is more than one species (or race) of okapi. Important information on this point will, however, doubtless be afforded by the Alexander-Gosling specimen, which, it may be hoped, will also indicate (if a male) whether the tips of the horns always protrude through the skin, and thus foreshadow the antlers of deer.

THE TARAWERA VOLCANIC RIFT, NEW ZEALAND.

MR. JAMES MACKINTOSH BELL, director of the New Zealand Geological Survey, contributes a paper to the April number of the Geographical Journal describing the present topography of the great volcanic rift of Tarawera, in the north island of New Zealand, and the changes which have taken place in the configuration of the region since the great eruption of Mount Tarawera on June 10, 1886, which is memorable for the destruction of the famous pink and white terraces, and their submergence in Lake Rotomahana.

Mount Tarawera lies near the centre of the Taupo volcanic zone, and about 135 miles south-east of Auckland. This zone, which has a breadth of some twenty-five miles, extends from near the great volcanic cones of Ruapehu, Tongariro, and Ngaurahoe north-eastwards to White Island, on the Bay of Plenty, a distance of nearly 160 miles. A great rift, which was the scene of greatest intensity of the 1886 eruption, stretches from near Lake Okaro along the Tarawera range to Mount Wahanga, in the most northeasterly part. This rift, really a line of craters, forms a huge fissure about nine miles in length, cutting the summit of the range, and appearing on its south-western slope. It is divided into several somewhat distinct craters by low partitions, and on the south-west side a long narrow rift extends to the base of the hill, so far as the edge of Lake

habitation as possible. The charts show that the quantity of chlorine near the coast amounts to 6 parts in a million, at 4 miles away to 5 parts, at 20 miles to 3 parts, at 40 miles to 1 part, and at 100 miles to 0.4 part.

The fact that chlorine exists in rain water to a large extent near the sea coast was stated in the report on domestic water supply of the Rivers Pollution Commission in 1874. It was there shown that on the coast of Devonshire, where with south-west winds sea spray is blown over the land, the amount of chlorine varies from 1-20 to 2.10 parts in 100,000, and at the Land's End, with a strong south-west wind blowing, it amounts to as much as 21.8 parts. Inland the average quantity of chlorine diminishes to 0.39 part; increases to 0.99 part at Liverpool and 0.79 part at Newcastle.

The

Paper No. 151, by Mr. Marshall O. Leighton, deals with the field assay of water, and describes the methods which have for some time been used in connection with the investigations into the quality of water in various parts of the United States. The methods described relate, not to laboratory experiments, but to simple tests to ascertain the general character of the water by methods which can be carried out on the spot. These field determinations give the turbidity and colour of the water, the presence of chlorine, carbonates, calcium, and iron, and the amount of hardness; also the amount of suspended matter. former are more particularly required in water for domestic supply, and the latter for that used for irrigation purposes. The amount of gradient to be given to a canal for conveying water for irrigation is governed to a great extent by the solid matter in suspension, and this also affects the capacity of the storage reservoirs. The method for determining turbidity, accompanied by an illustration of the gauge used for this purpose, was given in NATURE of January 7, 1904. A description and illustration of the Geological Survey field case is given in the paper. Paper No. 143, by Mr. J. H. Quinton, details the experiments made under the direction of the Reclamation Department on steel concrete pipes for the purpose of determining the durability and permanence of these structures in connection with the supply of water for irrigation purposes. The pipes experimented on were 5 feet in diameter, 20 feet long, and 6 inches thick, of concrete, enclosing an armour of steel rods sufficient to resist a head of 150 feet of water with a factor of safety of 4. The experiments showed the difficulty, even with the closest attention to the construction, of making pipes of, this kind that would stand a head of 100 feet.

Paper No. 150, by Mr. Robert E. Horton, gives the results of an investigation of the theory of weir measurements, and the discharge over different forms of weirs. The various coefficients of Bazin, Fteley, Stearns, and Hamilton Smith are analysed. A further description is given of the experiments performed at the Cornell University laboratory, where a closely regulated volume of water was passed over weirs of different forms placed across an experimental canal, and the results obtained compared with the different formulæ for obtaining the discharge. Tables are also given for calculating the discharge over weirs.

GREENWICH OBSERVATIONS.1

IN the introduction to the first work mentioned below, an opinion is expressed that the revision of an old catalogue must always be a source of anxiety to those who advise and undertake the revision, and that only the final result can justify the expenditure of the time and labour. Those who are responsible for this work need be under no apprehension that their efforts have been misspent. It 1 "New Reduction of Groombridge's Circumpolar Catalogue for the Epoch 1810-0." By F. W. Dyson, F. R.S., and W. G. Thackeray. Under the direction of Sir William H. M. Christie, K. C. B., F.R.S., AstronomerRoyal. (Published by order of the Board of Admiralty in obedience to His Majesty's command. Edinburgh: Neill and Co., Ltd., 1905.) Price

245.

"Telegraphic Determinations of Longitude made in the Years 1888-1902, under the direction of Sir W. H. M. Christie, K. C. B.. F.R.S, Astronomer. Royal. (Published by the Board of Admiralty in obedience to His Majesty's command. Edinburgh: Neill'and Co., Ltd., 1906.) Price 158.

would rather seem that in this case they have fulfilled a necessary duty, and discharged an honoured trust. It has always seemed to the writer that the ancient authorities at Greenwich were a little wanting in patriotism and enterprise in entrusting to a foreigner, however eminent, the reduction and discussion of Bradley's observations. Groombridge's observations, in a sense, may not be so completely a national possession as those of Bradley, but certainly it is not unfitting that at the Royal Observatory, almost within the shadow of which Groombridge erected his transit circle, his observations should be examined and discussed.

There are several circumstances which tend to give distinction to Groombridge's work. At the beginning of the last century his instrumental equipment was equal to, if not more powerful than, that of any other observer in Europe. The fact that, as an amateur, he gave his time and leisure to the repetition of the same mechanical performance shows that he was a lover of order and accuracy. Pond, the Astronomer Royal, whatever his failings may have been, appreciated the necessity for certainty and accuracy, and he must have impressed these qualities upon Groombridge. Further, the lapse of time, that factor which has increased the value of so much astronomical work and enhanced the reputation of so many worthies, has fought on the side of the retired West Indian merchant. The method to be pursued in the reductions, how far the observations are to be treated as independent, how far they are to be regarded as differential, are points which must be left to the decision of the computers. They must accept the entire responsibility, since the knowledge and experience is theirs. In this case it is not impossible but that they have had the assistance of tradition. The interesting remarks of Colonel Colby and Dr. Firminger quoted by the revisers, probably do not exhaust the information at their disposal. It would be an impertinence for anyone who has not even seen the originals to offer any criticism on the methods employed by those who have gained familiarity and experience by long contact with Groombridge's figures. These methods are described with clearness and in sufficient detail, but the revisers must know so much more than they can set down.

The result is to obtain a catalogue for the equinox of 1810 of 4239 stars. The number in the original Groombridge catalogue was 4243, but of these nine have been rejected on various grounds, and five have been added as separate stars. The places of a few more stars have been considered discordant, and have not been used in the subsequent discussion of proper motion. The accuracy of the catalogue and the care of the observer can both be estimated in some measure from the fact that a discrepancy of four seconds of arc in either right ascension or polar distance has been considered a proper limit to warrant the exclusion of the observation. The number excluded is 75 in right ascension and 214 in polar distance, slightly more than I per cent. of the total number of observations.

The peculiar value of this catalogue lies in the fact that its epoch is 1810. Therefore, by comparison with modern observations, it offers the means for a new determination of the precessional constant, while the new proper motions which it makes available should give greater certainty to researches into the amount and direction of the solar motion. The length of time elapsed since Groombridge's day is not much less than that available in the case of Auwers-Bradley, and the accuracy of the observations would seem to be of the same order; but Bradley's optical means were smaller, and the average of his stars considerably brighter. Groombridge's stars include many of the ninth magnitude, and fill a gap between those to which Bradley's observations refer and the results that will be derived from photography. On the other hand, Bradley's stars were better distributed over the whole sky. Groombridge limited his observations to the circumpolar regions. Against this drawback, as against many others, the Greenwich authorities have struggled with apparent success, and a few of their final results may be given.

We have, in the first place, the proper motions of more than four thousand stars determined by comparison of places at intervals of approximately ninety years. These proper motions have been derived for the most part by a