"And Hony soit qui mal y pense write In emerald tufts, flowers purple, blue, and white, Like sapphire, pearl, and rich embroidery."-Ibid., v. sc. 5. Here there is no confusion. The comparisons are exact and beautiful. Again we have "When wheat is green, when hawthorn buds appear." Midsummer Night's Dream, i. sc. 1. The season indicated shows there was no confusion between green and brown. We must not forget the well-known song- Love's Labour Lost, v. sc. 2. I think the above quotations afford good proof of the poet's correctness of colouring with regard to green and blue. It is true that he occasionally uses a small degree of licence with purple and blue, in the case of violets; but clearly not from unconsciousness of the difference. I cannot remember any instance where he confuses green with blue except purposely and humorously. In the use of other colours Shakespeare is in most instances I am acquainted with equally true to nature. To give examples would occupy too much space; but if there are exceptions I have no doubt that your correspondents-now that the matter is broached-will be able to furnish them. THE following incident may interest some of the readers of NATURE, as affording evidence of the possession and exercise of reasoning power by a brute. During the present frost the window-sills of my drawing-room are supplied with bread for the benefit of the birds, who, finding food there, are constantly fluttering about the windows. One day a large water-rat was seen on the window-sill, helping himself to the bread. In order to reach the window he had to climb to a height of about thirteen feet: this he did by the help of a shrub trained against the wall. Neither instinct nor experience will easily account for his conduct : since he never found food there before. If neither experience nor instinct, what save reason led him? His action seems to have been the result of no small observation and reasoning. He seems to have said to himself-I observe the birds are thronging that window all day; they would not be there for nought; it may be they find there something to eat if so, perhaps I too might find there something which I should like. I shall try. Bardsea EDWARD GEOGHEGAN OUR ASTRONOMICAL COLUMN OLBERS' COMET OF 1815.-On March 6, 1815, Olbers discovered a small comet at Bremen, in about 49° right ascension, and 32° north declination, or between Perseus and Musca; it had an ill-defined nucleus and was not visible without telescopic aid. The first parabolic elements were calculated by Olbers himself, and he was followed by Bessel, Gauss, Triesnecker and others in the determination of similar orbits. Ephemerides founded upon them showed that the comet would be observable for a considerable period, and as the result proved observers were not negligent of this circumstance. Gauss, writing to Bode on April 24, alludes to the long visibility of the comet, and the probability that elliptical elements would be found, but this remark apparently was merely intended to imply that the grasp which a long course of observation would afford upon the orbit, might lead to an ellipse, not that Gauss had remarked any sensible deviation from parabolic motion; indeed he mentions that he had not then reduced his April observations. The first detection of the inadequacy of the parabola to represent accurately the comet's course, is due to Bessel: he had calculated parabolic elements from observations on March 11, April 11, and May 20, which, while agreeing well with the positions employed, gave the right ascensions sensibly too small from March 11 to April 11, and between April 11 and May 20, as decidedly too great, even to as much as 4', and on May 26, the calculation was again many minutes in defect; these differences naturally induced Bessel to relinquish the parabolic hypothesis, and after some disappointment from the failure of the first method he employed, he communicated to Olbers on June 23 the elements of an elliptical orbit, in which the period of revolution was a little over 73 years. At the end of June Gauss deduced an ellipse with a period of 77 years, and soon afterwards Nicolai, then assistant to von Lindenau at Gotha, added a further confirmation of the elliptical character of the orbit, assigning a revolution of 72 years. On July 22, being in possession of observations to the middle of the month, Bessel improved upon his first calculation, and now found an ellipse with a period of 73.8968 years, which was made the foundation for his subsequent investigations, of which we have presently to speak. Thus was the periodicity of the comet established, and Bessel, after remarking upon the importance of the addition to the system (at that time Halley's comet was the only one that could be considered certainly periodical) he proposed that it should bear the name of its discovererOlbers. Besides a long series of observations taken by Olbers himself, the comet was observed by Gauss at Göttingen, Bessel at Königsberg, Triesnecker at Vienna, Struve at Dorpat, Oriani at Milan, Lindenau at Gotha, Maskelyne at Greenwich, and Bouvard at Paris. Its distance from the earth continued pretty nearly constant (about 1'45) during the greater portion of the time it was visible, and at no period was it a conspicuous object; its nucleus was pretty bright at the beginning of May, and it then had a tail about 1o in length. On the disappearance of the comet Bessel collected the observations which extended to August 25, the last having been made by Gauss at Göttingen; indeed, he was the only observer after July 25. He then commenced the work which is incorporated in his great memoir upon this comet, published in "Abhandlungen der königlichen Akademie der Wissenschaften in Berlin, 1812-13," a volume which was not published until 1816. He formed ten normal positions, in which all the observations appear to be brought to bear, excepting those at Greenwich and Paris, which were doubtless unknown to him. He corrects these normals for the effect of perturbations from the action of Venus, the Earth, Mars, Jupiter, and Saturn, during the comet's visibility, and by a fine series of observations of the sun at Königsberg between March 8 and August 29, 1815, he applies corrections to the sun's places obtained from Carlini's first tables. Equations of condition were then formed and solved on the method of least squares, and thus the following definitive elements of the comet's orbit in 1815 were obtained: These elements represent the normals upon which they are founded very closely, considering that observations of comets in 1815 did not pretend to the degree of precision which is now sought to be attained, and, moreover, were subject in the reductions to errors in the places of the comparison stars. But Bessel's labours did not stop here. With a special interest in the comet of 1815, not, it may be presumed, alone due to its exceptional character, but in no small degree to the circumstance of its having been detected by his most intimate and revered friend, Olbers, Bessel undertook, and in the year of its appearance accomplished, the laborious task of computing the perturbations of the planets Jupiter, Saturn, and Uranus upon the motion of the comet during the present revolution, and so determining the epoch of the next perihelion passage. The principal details of this work are comprised in the memoir to which we have already referred. The masses of Jupiter and Uranus were Laplace's, while the mass of Saturn was taken from Bouvard's tables. The whole period is divided into three sections, the first extending from August 4, 1815, to July 30, 1833; the second from the latter date, with new values of the semi-axis and excentricity to July 21, 1869, and the second from July 21, 1869, to the next perihelion passage. The action of each of the three planets tends to accelerate the comet's return, that of Jupiter by upwards of two years; the final result indicating an acceleration of 824'51 days, with reference to the period belonging to Bessel's definitive ellipse for 1815; it was thus found that the duration of the actual revolution would extend to 26222'4 days, and consequently the next perihelion passage is fixed to February 9'4, 1887. This conclusion will be affected not only by the imperfect values of the planetary masses which were available when Bessel undertook the investigation, but in a greater degree by the uncertainty which still remained as to the precise length of the revolution at the last appearance; this Bessel found to extend to 027657 of a year, or IOI days. With such an amount of probable error attaching to Bessel's result it must soon be a matter for the consideration of the astronomer, whether a nearer approximation may not be yet attained. We have much more accurate values of the masses of Jupiter, Saturn, and Uranus than Bessel possessed, and are able to take into account the influence of Neptune, though this is not likely to be very material. Fortunately, in several series of observations, the observed differences of right ascension and declination between the comet and the comparison stars are preserved to us, and thus we can reduce the observations anew, with much improved positions of many of the stars and with modern elements of reduction. The series of observations thus available include the long one of Olbers (Berliner astronomisches Jahrbuch, 1818), and those of Greenwich, Paris, and Dorpat. It is a work which, together with the recalculation of the perturbations to the next perihelion passage, may perhaps be made the subject of a prize by one or other of our scientific academies; on the last return of Halley's comet, the first approximation to the epoch of arrival at perihelion was due to action of this kind on the part of the Academy of Turin, and though a much higher degree of interest attached to the reappearance of that famous body, we do not despair to see Olbers' comet deemed worthy of a new and more refined calculation. If these cometary bodies wandering into the confines of the solar system from the stellar spaces are fixed therein by the action of one or other of the planets, it will have been owing to a very close approach to the planet Mars that Olbers' comet presented itself in 1815, moving in an ellipse of moderate dimensions. The nearest approach of the two orbits in that year was 0.07 in 86°.4 heliocentric longitude, but this distance must have varied in successive revolutions through the perturbations of the other planets, and at some past time there may have been an intersection of the orbits and a close encounter of the two bodies. METEOROLOGICAL NOTES BEFORE the commencement of the summer rains this year Mr. Eliot, the officiating meteorological reporter to the Government of India was called upon for a report on the prospects of the season. His reply, to which we have already referred in the "Notes," consisted of a short résumé of the most important characteristics of the southwest monsoons of recent years, from which the following conclusions were deduced:-" 1. The persistent excessive pressure over Northern India at the present time (June, 1878), tends to diminish the baric gradient between Southern Asia and the Mid-Indian Ocean, and if this is not compensated by increased pressure over the sea area to the South of India, the monsoon current will be below its average strength. 2. There appear to be no stronglymarked abnormal variations of pressure over Northern India. It is therefore probable that the rainfall will be much more equally distributed than last year. 3. Comparing the present year with 1865, it is probable that the heavy rainfall during the cold weather, and more especially in May, will slightly retard the advent of the monsoon in Upper India. 4. The probable effect of the low pressure along the Bombay coast cannot be determined except by comparison with last year. It appears to promise fairly abundant rain over that portion of the country." These conclusions have now been subjected to the test of experience and are found to have been verified in almost every particular. The southerly current from the Indian Ocean has been decidedly below its normal strength; the rains set in from a fortnight to a month after the usual time; every district in the country has received a moderate supply of rain, though the average rainfall for the whole country has been less than usual, and over the Bombay Presidency, from Belgarum to Kurrachee, the rainfall has been in excess of the average for previous years. The only peculiarity of the monsoon of 1878, that was not predicted, was the frequent recurrence of heavy falls of rain over a few small and well-defined areas; but this would seem to be the character of the rainfall of every year in which the monsoon current is of less than the usual strength. The percentage of verifications reached by Mr. Eliot has thus been as great as that attained by the American observers, and the predictions in his case were made months, not days or hours, in advance. The same meteorologist has recently made a discovery which promises to be of the greatest possible value in connection with the system of storm-warnings to the ports round the Bay of Bengal. It is that a cyclonic vortex, when generated in the middle of the Bay, always travels towards that part of the coast where the wind velocity for the time being is least in comparison with the average velocity for the same place and time of year. This law has been verified by almost all the cyclonic disturbances that have occurred in the Bay since a chain of meteorological observatories was established round it, and it lends a great deal of support to the theory that a cyclonic vortex is developed through the accumulation, concentration, and condensation of aqueous vapour over a region of comparative calm. All that appears now wanted to render cyclone prognostications for the Bay of Bengal almost absolutely certain is a submarine cable to the Andaman and Nicobar Islands, by which the meteorological stations on these islands, near the place of origin of all the great cyclones of the Bay, would be brought into telegraphic communication with the rest of the empire. In his "Tenth Contribution to Meteorology," which appears in the American Journal of Science and Arts for the present month, Prof. Loomis gives the results of an examination he has made as to the course of seventyseven storms after leaving the eastern coast of the United States, these storms having occurred from March, 1874, to November, 1875. Of these seventy-seven storms he was able to follow thirty-six of them entirely across the Atlantic Ocean, eight of them, however, becoming merged in other storms before reaching Europe. The annual average of storms which are found to cross the Atlantic from the United States to Europe is eighteen, and nearly all of these storms pursued a course north of east, passing in their eastward course considerably to the north of Scotland; indeed, in only four of the storms did the centre pass as far south as the north of England. Prof. Loomis concludes that, when a storm with a centre depression at least below 29.5 inches leaves the coast of the United States, the probability that it will pass over any part of England is only one in nine; that it will occasion a gale anywhere near the English coast, one in six; and that it will give rise to a fresh breeze, one in two. A characteristic feature of these storms is the slow rate of their onward progress in crossing the ocean, as compared with their rate over the United States-a feature of the utmost possible importance in attempting to predict the time of their descent on the shores of Europe of those American storms which cross the Atlantic. About half of the whole number of the storms originated in the neighbourhood of the Rocky Mountains, five in or near Texas, and four were distinctly traced to the Pacific coast. Of six West India cyclones which occurred in the same time only two could be traced across the Atlantic, and even one of these became blended with another storm. The rest of the paper is taken up with a discussion of the fluctuations of the barometer on Mount Washington, 6,285 feet, and Pike's Peak, 13,960 feet, as compared with what takes place on the level ground at the base of these mountains. As regards Mount Washington, the valuable result is arrived at that the diurnal maxima and minima of the barometer occur more than three hours later at the summit than at the base, showing an average retardation of one hour for each 900 feet of elevation. In the case of Pike's Peak, the rate of retardation is one hour for an elevation of 1,380 feet. It is evident from these figures that the law of the rate of retardation is yet to be sought, one of the most important factors, in all probability, being the absence or presence of high plateaux and their extent near the high station, to which must be added the latitude of the place. Observations of the wind at these high levels show, just as at places near sea-level, a circulation about a low centre, the movement of the wind being approximately at right angles to the direction of the low centre; and further, that at the height of Mount Washington, the low centre of storms sometimes lags behind the low centre at the surface of the earth as much as 200 miles. This last result is so vital in the theory of storms as to demand a much more extended examination, the most special care being taken that the retardation of the time of occurrence of the diurnal barometric minima be allowed for in the discussion. IT is with extreme satisfaction we learn that at a recent meeting of the Council of the Scientific Association of France, M. Mascart, Director of the Meteorological Department, submitted a proposal from the Departmental Commission of Vaucluse, for the establishment of an This observatory on the top of Ventoux, situated to the northeast of Carpentras, and rising above all the surrounding summits to a height of 6,300 feet above the sea. observatory in the south of France, along with the observatories of Puy de Dôme in the north, and of Pic du Midi in the south-west, may be regarded as furnishing France with an enviable system of elevated observatories for meteorological observations such as no other country possesses, thus putting French physicists in possession of the essential data whence the more difficult meteorological problems may be attacked, and the systems of weatherwarnings for navigation and agriculture more rapidly developed and improved. It is estimated that 150,000 francs will be required to establish the station, of which sum there are already subscribed by M. R. Bischoffsheim 10,000 francs, by the Commune of Bédoin, situated at the foot of Mt. Ventoux, 10,000 francs, the Council of the Scientific Association 500; and as the Meteorological Commission of Vaucluse has opened a subscription-list, the General Council of the Department has promised to aid in forming the roadway up the mountain, and a subsidy is looked for from the Minister of Public Instruction, the establishment of this important observatory will doubtless soon become an accomplished fact. GEOGRAPHICAL NOTES WITH reference to the reports that Prof Nordenskjöld's vessel had got shut in by the ice near East Cape, in Behring Strait, the Committee for Promoting Russian Trade and Industry have resolved to apply to the Governor-General of Eastern Siberia, requesting him to assist in instituting a search for Prof. Nordenskjöld, and in obtaining more certain information as to the situation of the expedition. Mr. W. H. Dall, the well known U.S. Alaska explorer, has written a letter to an acquaintance in Stockholm, mentioning the previously-reported statement of whalers, from which it is supposed that the Vega, has been stopped by ice east of Cape East. Should this be the case, Mr. Dall entertains no fears for the fate of the expedition. If these suppositions be correct, he says, "the breaking up of the ice next July will leave open water for the Vega to proceed to Behring Strait. Vessels pass to westward of East Cape every year. There is a creek there. (The letter here gives a sketch map describing a bay, with a small island in the middle of it, and an anchorage inside.) A river with fresh water runs into the bay, and on the coast is a native village. This is not marked in the ordinary maps and charts, and it is just here that the vessel, according to the repots of the natives, must be lying. She can safely winter there. There is a large village, inhabited by Tchuktchees, who would be able to supply fresh meat. This place is situate not more than 200 English miles from the white men's trading station at Plover Bay. If the Vega is lying there, the success of the operation is practically achieved, because, as I said, the bay is open every year, and does not get closed by ice until October. Vessels sail there, and carry on trade every summer." THE last number of the Isvestia of the Russian Geo graphical Society contains an interesting paper by M. Grigorieff, on the temperature and density of water in the Arctic Ocean, along the coast of Russian Lapland, and in the White Sea, being the result of observations carefully made on board the schooner Samoyede, by means of good instruments. As to the Arctic Ocean, M. Grigorieff confirms the existence of a warm branch of the Gulf Stream which flows along the coast as far as Gavrilovskiye Islands, and thence turns due east to the Kanin Peninsula and Kolgueff Island, and further, to the Moller Bay on Novaya Zemlya. Beneath this warm current there is a cold one flowing in an opposite direction at some depth. When it meets with a rising bottom, and especially with the deep bank of less than 100 fathoms under 71° N. lat., this denser and cold current is compelled to change its direction, and makes its way between the Gulf Stream and the shore; hence the low temperatures and great density of water at the Lapland coast, in the space between Svyatoy Nos and the Seven Islands. The density of the eastern (North Cape) branch of the Gulf Stream (1'025 to 1'026, figures which correspond to a percentage of salt of from 3.28 to 3.41), seems to be smaller than that of the Spitzbergen branch, where Nordenskjöld has found a percentage of salt as highas 3.625. As to the White Sea, M. Grigorieff denies the entrance of a branch of the Gulf Stream into that sea, as was supposed some years ago by Prof. Middendorff; the Gulf Stream does not penetrate further than the Gulf of Mezen, and the warm temperatures observed by Middendorff are due to purely local causes. On the contrary, a cold polar current enters the White Sea along the Tersky coast, whilst the current which flows out of the sea into the ocean, follows the Winter and Kanin Coasts. The water of the White Sea on the whole has a very low temperature, especially in the deeper parts; on depths more than 100 fathoms the temperature is always below 32° Fahr., and this, because of the great loss of heat during the long winter. Altogether, the observations having been made and computed very carefully, and published in extenso in the Isvestia, are a real acquisition to science. Two new expeditions to Central Asia are planned in Russia for the next spring. The first, by Col. Prjvalsky, to Hlassa in Thibet, and thence to Afghanistan; and the other, by M. Blumenfeld, a German savant who has studied in Russia, for botanical and geological explora. tions; M. Blumenfeld will follow nearly the same route as that proposed by M. Prjvalsky. • UNDER the title of "D'Orenbourg à Samarkand " Madame de Ujfalvy has commenced in the Tour du Monde an illustrated account of her travels in Ferghanah and Western Siberia. Leroux, of Paris, has just brought out the first volume of M. de Ujfalvy's account of the results obtained during his mission. These results are mainly ethnological, and contain many observations and careful and detailed measurements of a large number of individuals representing the various races of that part of Central Asia visited by the traveller and his wife. DURING the year 1878 the following accounts of Russian exploration were published in Russia: that of Col. Prjvalsky to Lob-Nor, now translated into English; of M. Wojeikoff in India and Japan; rather literary than scientific is that of M. Minayeff on his journey to India, which contains very interesting observations on Buddhism; of M. Ogorodnikoff to Persia, giving among other data an account of the trade-routes to Persia and Afghanistan: and of M. Skalkofsky to Eastern Asia and California. As Sir H. Rawlinson has announced his intention of delivering an address at the next meeting of the Geographical Society, on the road to Merv from the Caspian, it will not be without interest to note some particulars respecting the earlier part of the route, as far as the Tekké fort of Kizil Arvad, from an account lately furnished to the Moscow Gazette by a writer who appears to have been attached to General Llamakin's staff. The party were obliged to strike eastwards from the Chikishliar littoral by a road which has never yet been described, but which is the most practicable route to the Attrek, the bank of that river, from its mouth at Hassan-Kuli Bay almost to Balt Adji, being, bordered by inapproachable morasses. The ground traversed was at first covered with shells, but soon presented the appearance of a salt marsh petrified by the sun; then, after a stretch of sand, firmer soil was met with. No water was found until the wells of Karadji-Batyr were reached. About twelve versts from the wells the party arrived at the gates, as it were, of an enormous wall, which bore a greater resemblance to an artificial structure than to a natural conformation of the soil. Three versts further on the valley of the Attrek appeared in sight, with the river itself winding between high and verdant banks. Here is Bayat-Adji, a name which is also applied to the whole of the surrounding country. From this spot the party proceeded up the Attrek to Chat or Chad, following an excellent road. About ten versts before reaching Chat the road turned to the left, leaving the Attrek at a point where there are large auls, or settlements of the Atabai tribe. At length Chat was reached, and it is described as the most repulsive place along the whole Attrek, although from a strategical point of view the most important, because it is here that the River Sumbar (which Capt. Napier calls the Sunt) flows into the Attrek, and the delta might be made an impregnable position. Fifty versts above Chat two enormous rocks rise out of the Attrek, forming a sharp delimitation of the geological structure of the country. This place is called Su-Sium; after this point the road is impassable for camels, and 10 versts further on is difficult even for horses; 100 versts beyond Chat the course of the Attrek can only be followed on foot, and it would take three months to make the road practicable. In consequence of the difficulties mentioned, the party was obliged to abandon the course of the Attrek at SuSium, and to strike a new road. After making the necessary surveys they turned to the left at a place called Alun-Yak, and proceeded over the high Sugundag chain. The ascent and descent of the Sugundag extends over a distance of 16 versts, the descent terminating at the small River Chandyr, which falls into the Sumbar. Twenty-five versts from Chat the party crossed the Sumbar, and marching between that river and the Chandyr, reached an elevated mountain called Bek-Tépé, belonging to the spurs of the Kurindag. Leaving the Sumbar they proceeded through the waterless defiles of the Ters Akon, and through the Morgo defile (belonging to the Kaplandag range), and reached the ruins of HadjanKala, near the Tekké fort of Kizil Arvad. The road through the defiles presents many difficulties, and only two horses can proceed along it abreast, but it is thought that a good road could be made without much trouble or expense. THE Society for Promoting Christian Knowledge publishes a very excellent small wall-map of Africa, by Stanford, containing all the most recent discoveries and useful both for teaching and general purposes. ON January 25 the Geographical Society of Paris will hold a public reception in the large hall of the Sorbonne, in honour of MM. de Brazza and Ballay, the two French Ogowé explorers. The great medal for 1879 will be delivered on this occasion by Admiral La Roncière le Nourry, the president of the Society. No. 78 of the Zeitschrift of the Berlin Geographical Society contains a careful geographical and statistical study on the Brazilian province of Rio Grande do Sul, by M. Bescharn. Botanical students will be interested in Dr. Klunzinger's elaborate paper on "The Vegetation of the Arabian Desert near Koseir." This number contains a carefully arranged, most complete, and valuable bibliography of geographical literature and cartography for the year from November, 1877, to November, 1878. No. 3 of Globus of this year contains a fine illustration of the wonderful reclining statuary figure of Chac-Mool, unearthed in Yucatan some time ago by M. le Plongeon. Plo The same number contains the sixth contribution of Herr Zehme to a résumé of recent exploration in Arabia. THE GEOLOGICAL HISTORY OF THE COLORADO RIVER AND PLATEAUS II. IN N the Pliocene period the climate of the region gradually experienced a great change. Miocene times were characterised by a moist and ordinary sub-tropical climate; the Pliocene by developing an arid one, like that which now prevails there. Let us look at the causes which make this climate what it is. In whatever rectilinear direction we may undertake to pass from the Pliocene Country to the ocean we shall be compelled to cross some of the loftiest barriers of the Continent. It is hemmed in by range after range of high mountains. The winds laden with moisture are wrung dry long before they reach the plateaux in the heart of the province. The prevailing wind throughout the year is from the westward, and must cross the Sierra Nevada. Sweeping across the great basin it blows over many ranges, and at last strikes the Wasatch and the chain of high Plateaux which form the western wall of the Plateau Province. Here it is suddenly projected upward more than a mile and flings down moderately copious rains. Descending into the Cliff and Cañon Country, its humidity is so much exhausted that it can yield but the scantiest pittance of snow and showers. Thus the country is a desert. Now the strange forms impressed upon this land-its cliffs and cañons, By Cart. C. E. Dutton, U.S. Army, Ass start-Geologist U.S. Strvey of the Rocky Mountain Region, under fro. J. W. Powell, in charge. Continued from p. 252. with their myriads of wonderful shapes and their astounding architecture-are due, as we shall presently find, in great part to the aridity. The aridity is due to the great barriers which surround it, and above all to that great barrier of high plateaux which lies upon its western verge. Here, then, we may look for another key which may unlock another door within the vestibule. The search will not be fruitless. The district of the high plateaux has been during the last four years a field of special study by myself, and has |