thing sensible, which a persevering study of Nature in every direction and down to our own times, has brought to light, is the material from which our delineation is to be drawn; it includes its inherent testimony of truth and fidelity.'-Kosmos, p. 80. And farther on, after referring to a future section of the work for the history of science, he adds My duty is to depict generally the state of knowledge, according to its measure and limits, at the present time. Mean results are the ultimate aim, nay, the expression of physical laws, as regards what is subject to motion and change. They exhibit to us Constancy in the midst of Change and the ceaseless course of events. So, for example, the progress of the modern measuring and weighing science of physics is eminently indicated by the attainment or the correction of the mean values of certain magnitudes; so numerical cyphers present themselves again, but with an enlarged meaning, as they formerly did in the schools of Italy, the last and only remain of hieroglyphics in our writings, but all-powerful in Cosmical science.'-p. 82. He thus proceeds in a more lively strain : The zealous philosopher is delighted by the simplicity of the numerical relations by which the dimensions of space, the magnitudes of the planets, and their periodical disturbances, are denoted; or the threefold elements of the earth's magnetism, the mean pressure of the atmosphere, or the quantity of heat which the sun sheds daily or yearly on any spot of the fixed or fluid surface of our globe. But unsatisfied is the poet, unsatisfied the ever-curious multitude. To both of these, Science seems as if desolate, many questions being rejected as dubious or insoluble which formerly were entertained. In her more rigid form and stiffer drapery she loses the more seductive charm with which she was invested by a philosophy of forms and symbols calculated to deceive the judgment and amuse the fancy. Long before the discovery of the New World it had been supposed that land was visible from the Canaries and Azores. But these were phantoms, not caused by extraordinary refraction, but due only to the conjectures of the spectators, whose longing eyes strove to penetrate the distant haze. The natural philosophy of the Greeks, and the physics of the middle ages, and even of a later period, abundantly offered similar airy visions. At the limits of exact knowledge (as from a lofty island-shore) we cast a sanguine gaze towards unknown regions. The belief of the unusual and the marvellous lends a distinct outline to every creation of fancy; and the realms of imagination, with their cosmological, geognostical, and magnetic dreams, are immediately confounded with the domain of reality.'— Kosmos, p. 82-3. In the astronomical part of Cosmos, our author invariably treats the so-called nebular hypothesis as an ascertained physical fact, and in so far appears for once to abandon the cautious limits of descriptive writing and simple classification which he has imposed upon himself. Not only does he maintain Herschel's doctrine doctrine of the progressive consolidation of nebulous matter (which, however, he ascribes (p. 87) to Anaximenes and the Ionic school); not only does he affirm this process to be going on under our eyes,' and to be in all respects similar to the development' of organic beings-thus assimilating the universe to a garden or a forest. He also accepts as established, and apparently not admitting of a doubt, the theory peculiar to Laplace of the genesis of nebulous rings by centrifugal force, and the subsequent still more incomprehensible agglomeration of these rings into solitary rotating planets and satellites: and he even assumes it as established (p. 89, 95), that the zodiacal light arises (as Cassini imagined) from a still uncondensed ring of world-vapour (welt-dunst) between the orbits of Venus and Mars. On all this doctrine we retain the most energetic doubts.* The progress of discovery at the present time is decidedly unfavourable to it, as every one conversant with the scientific literature of the day is aware of; as a physical description of what exists, it is inaccurate, because it is uncertain; as a physical account of what has been and what will be, it can rank at best amongst the numerous list of bold but unestablished inductions. Nor can we think more favourably of an idea of Humboldt's own, that there exists an analogy between the distribution of plants and that of satellites in groups round their primary and planets round the sun. A still more palpable similarity would, we imagine, permit us to compare the individuals of celestial groups to the stamens and pistils of flowers; to call our earth and moon of the order Monandria Monogynia, Jupiter's system Monandria Tetragynia, and the like. This shows how mere analogies from collocation, without reference to the end or design of the whole, may retard science. What is barely tolerable in the poetry of Darwin, cannot come well from the matter-of-fact pen of the astronomer.‡ It * The sole phenomenon of our system which might lend countenance to Laplace's notion (and which perhaps suggested it) is the unique and imposing one of Saturn's ring. We observe a very good remark on this subject in Mr. Monck Mason's Creation by the immediate Agency of God,' p. 50, which is undoubtedly correct; to wit, that the excessively small and uniform thickness of this vast expansion of matter (estimated at only 100 miles, with an extreme diameter of nearly 200,000) indicates a degree of oblateness quite inconceivable under the circumstances, the planet whose centrifugal force is supposed to have generated it being almost spherical, or flattened at the poles only by one-eleventh part (Laplace, Système du Monde, 1, 79). . † So Milton and other suns perhaps, With their attendant moons, wilt thou descry, Which two great sexes animate the world,'- Par. Lost, viii. 148. Botanic Garden., iv. 359, commencing 'So, late descry'd by Herschel's piercing sight,' A noble passage, though in Darwin's inflated style. His cosmogony seems to have some It is not to be supposed that much of novelty should be elicited in the purely astronomical part of the subject. But starting with the Nebular Hypothesis, our author manages with much ingenuity to consider in succession a series of phenomena which lead into one another, and which convey us, by easy steps, from the celestial to the terrestrial part of the science of Cosmos. Surveying in succession the heavenly bodies with whose density we are tolerably acquainted, the sun and planets, he next passes to comets, whose rarer texture forms a step to that inconceivable attenuation of gravitating matter which constitutes, according to Laplace and Humboldt, the Zodiacal Light; and to shooting stars and aërolites, celestial in their origin, terrestrial in their component parts (iron, nickel, cobalt, manganese, chromium, copper, arsenic, tin, soda, potash, sulphur, phosphorus, and carbon), which bring us down to the vulgar chemistry and geology of our own Earth. Of cometary astronomy we have (p. 105, &c.) an interesting synopsis, which we should willingly have transferred to our pages were it not too long; besides, our readers will be more interested in parts of the subject more akin to Humboldt's own pursuits. It may be mentioned in passing, as a curious fact, that the earliest valuable observations of comets are due to the Chinese, and extend as far back as the years A.D. 240 (under Gordian III.), 539 (under Justinian), and 565. Our author does not fail to draw a contrast between the terror with which these bodies were then regarded throughout Europe, and the scientific composure of the Chinese. In 837, when a comet of alarming magnitude approached the earth within twice the moon's distance, whilst Louis I. of France was trying to avert the impending danger by vowing to found a monastery, the countrymen of Confucius were coolly measuring the length of its tail and determining its course amongst the stars. The comets of 1402, 1532, 1577, 1744, and 1843, were so bright that the nucleus was visible in broad daylight; but their well-defined disks are commonly excessively small, and indicate a diameter of but a few hundred miles, or even less. The cometary light is ascertained by Arago to be polarised, and therefore he concludes it to be reflected from the sun; whereas it seems to us that were the matter the same as that of the Sidereal Nebulæ, as has been supposed, it ought to be, like them, selfluminous. The tail is sometimes double (1807, 1843), and in 1744 was divided into six. The apparent length was, in 1618, soire analogy with that in the work before us (Kosmos, p. 86), which appears to ascribe to matter generally a power of indefinite 'development and regeneration, such as is usually admitted only to exist in living plants and animals, and that to a limited degree. 104°. or 14° greater than the distance from the horizon to the zenith. The comet of 1680 had an absolute extent of tail as great as from the sun to the earth (95,000,000 miles). A star of the 10th magnitude lost no sensible part of its brilliancy in being eclipsed by Halley's comet in 1835 at a distance of only 2" 2 from the comet's centre (Struve), nor do stars appear refracted out of their course by the intervention of the nebulous matter, which is therefore conjectured to be dusty, not fluid. The mass of comets is conjectured not to exceed 1-5000th of the earth's at a maximum, and perhaps not 1-100,000th at an average. The periods and eccentricities of comets have, as is well known, an enormous range. Three orbits are considered to lie wholly within the recognised limits of our solar system:-1. Encke's comet, which revolves in 3 years, and whose aphelion or most distant point lies within Jupiter's orbit; 2. Biela's comet of 63 years extends its path beyond Jupiter's orbit, but far within Saturn's; 3. Faye's comet (discovered in 1843, and of which the return has yet to be observed) is supposed to have a smaller eccentricity than any other known comet, and a period of 7, years, with an orbit lying wholly between those of Mars and Saturn. On the other hand, the comet of 1680 is supposed to reach its aphelion at a distance of 80,000 millions of miles from the sun, forty-four times further than Uranus. Yet the nearest fixed star, whose distance has been approximately estimated (a Centauri), is distant no less than 11,000 radii of Uranus's orbit, and the star 61 Cygni 31,000 radii. Yet this same comet of 1680 approached the sun's surface within 1-6th of the sun's diameter, or 7-10ths of the moon's distance from the earth. It was then moving with the velocity of about 250 English miles in a second, whilst at the other extremity of its eccentric orbit it must toil along at the rate of but 10 feet in a second, a speed comparable to that of many large rivers. Nor does it return to the sun until the lapse of 8800 years from the time of its departure. It is singular, that in enumerating (p. 118) instances of the near approach of comets to the body of the sun, our author has omitted that of 1843, of which the orbit was first calculated by an accomplished young astronomer, M. Plantamour of Geneva, and shown to have a perihelion distance less than that of any previously known, even that of 1680. The next topic is one of general interest, and is treated of with great fulness and originality: the phenomena and origin of meteors, including aërolites and common shooting stars. This part of the work (pp. 120-137) will be studied with interest by men of science as well as by popular readers. It begins by recalling the general phenomena which are probably due to a common cause. cause. The appearance of luminous fire-balls, sometimes so large and bright as to shed a visible gleam in broad daylight, is unequivocally connected by experience with the fall of aërolites or meteoric stones-as was the case (to cite only recent instances) in 1790 at Barbotan in the south of France; in 1794, at Siena in Italy; in 1804, at Weston in Connecticut; and in 1821, in the department of the Ardèche in France. Sometimes a small dark cloud appears to originate the meteoric shower, whose descent is accompanied by a noise like thunder. The fire-balls, which occasionally appear to exceed the diameter of the moon, have every intermediate magnitude down to that of common shooting stars and this is the strongest, perhaps the sole evidence, for their identity of nature; both one and the other leave phosphoric trains behind them, a real phenomenon, and not due to an optical deception, as has been sometimes imagined (p. 394, note 30). The important consideration which has recently recalled particular attention to these curious and beautiful appearances of luminous meteors, is their alleged periodicity. On this subject Humboldt says: Shooting stars fall either singly and rarely (sporadically), or in groups of many thousands. In the latter case they are periodical, and generally move in parallel directions. Of periodic groups the best known are the November-phenomenon (12th-14th of November), and that of the Feast of St. Lawrence (10th of August), whose " fiery tears" have long since been suspected by tradition, and in an old monkish Calendar, to be a recurring meteorological phenomenon. Although a mixed shower of falling stars and fireballs was seen in the night of the 12th-13th of November at Klöden near Potsdam, and in 1832 throughout all Europe, from Portsmouth to Orenburg on the Ural river, and even in the Isle of France in the Southern Hemisphere, still the idea that great meteoric showers are connected with certain days was first occasioned by the observations of Olmsted and Palmer in North America, on the 12th-13th of November, 1833, when the falling stars appeared compressed like snow-flakes about one spot in the sky, so that in nine hours not less than 240,000 must have fallen. Palmer in Newhaven (Massachusetts) recollected the meteors of 1799 (also on the 12th-13th of November), which were first described by Ellicott and myself, and which it is proved, by the observations which I have cited, were simultaneously seen in the New Continent from the Equator to Herrnhut in Greenland (lat. 64° 14′), and between 46° and 82° of longitude. The identity of the periods was observed with astonishment. The meteoric stream which filled the whole sky on the 12th-13th of November, 1833, from Jamaica to Boston, was repeated on the night of the 13th-14th of November, 1834, in the United States of North America, but with somewhat less brilliancy. In Europe the periodicity has been since more regularly established. *Said to exist in Corpus Christi College, Cambridge. 'A second |