i think that every thoughtful student of the stars must have wondered how the figures of the various objects now pictured in our star-maps came to be imagined in the heavens themselves. it is a convenient answer to inquiries of the sort to say that it became necessary at an early stage in the progress of astronomy to have some means of identifying and naming star-groups, and that the arrangement into constellations was as suitable as any other that could have been desired. but it seems to me altogether unlikely that, in the infancy of a science, a mere arbitrary arrangement, such as this explanation supposes, should have been adopted. if we try to imagine the position of the first observers of the stars, what they wanted, and what they were likely to do,—and this a priori method of dealing with such questions is, i believe, the only safe one,—we perceive that the division of the stars into sets named after animals and other objects, without any real resemblance to suggest such nomenclature, is as unlikely a course as could possibly be conceived. beyond all question, i think, the first watchers of the skies (they can scarcely be called astronomers) would have taken advantage of imagined similarity, more or less close, between each remarkable group of stars and some known object, to identify the group, and to obtain a name by which to speak of it.

yet it must be admitted that, as the constellations are at present arranged and figured, it is very difficult, in the great majority of cases, to imagine the least resemblance between a constellation and the object from which it derives its name. this is not only true of the modern constellations, the preposterous pneumatic machines, printing presses, microscopes, and so forth, with which hevelius and his successors foolishly crowded the heavens. even the oldest of the old constellations of ptolemy, nay, some even of those which are found among all nations, present, according to their present configuration, scarce any resemblance to their antitypes. for instance, it is well known that the great bear was recognised by many nations besides the greeks and those, whoever they may have been, from whom the greeks derived the constellation. we learn that when america was discovered the iroquois indians called this constellation okouari, or the bear. so the inhabitants of northern asia, the ph?nicians, the persians, and others, called this constellation the bear. the egyptians, not knowing the bear, called the constellation the hippopotamus, an animal resembling the bear in several respects, as in its heavy body, short inconspicuous tail, small head, and short ears. yet the constellation, as at present figured, is certainly not in the remotest degree like a bear. apart from the enormous tail given in the pictures to the bear (almost tailless in reality), it is impossible for the liveliest imagination to recognise a bear as the constellation is at present formed. flammarion says that, "even if we take in the smaller stars that stand in the feet and head, no ingenuity can make it in this or any other way resemble a bear," adding the absurd explanation given by aristotle, "that the name is derived from the fact that of all human animals the bear was thought to be the only one that dared to venture into the frozen regions of the north, and tempt their solitude and cold." as though the shepherds and tillers of the soil, who first gave names to the stars, were likely to consider such far-fetched reasons, even if they had known either the habits of the polar bears or had considered the relation of the northern star-groups to the polar regions of the earth.

now the question whether any real resemblance attracted the attention of the earlier observers in such cases as this is by no means without interest. if such a resemblance formerly existed, and does not now exist, it would follow that quite a considerable proportion of the stars have changed in brightness. considering that each star is a sun, the centre, most probably, of a system like that which circles around our own sun, such a conclusion would be very startling indeed. it would have a special interest for ourselves, somewhat in the same way that the news that many railway accidents occur has an interest for those who travel much by rail. if accidents frequently happen to those other suns, in such sort that they either lose or gain greatly in brightness, an accident of one or other kind might well happen to our own sun, in which case the inhabitants of this earth would perish. for many of the stars, by our supposition, would have changed so much as either to lose their character as the defining stars of a constellation or by accession of brightness to acquire that character which in old times they had not possessed. now, assuredly, a change of brightness competent to affect our sun's character (as viewed from any remote star system) in equal degree, would be destructive to the inhabitants of the earth. none at least of the higher races of animals or plants could bear the intense cold resulting from a change of the former kind, or the intense heat resulting from a change of the latter kind. yet, if the constellations were once named because of their imagined resemblance to various objects, and if no such resemblance can now be even imagined, a change of one or other kind in the condition of our sun must be regarded as probable,—much in the same way that a regular traveller by train on any line must be regarded as exposed to danger, if accidents are known to be continually happening on that line.

what i now propose to do is to inquire whether we may not find the true figures and proportions of the ancient constellations in another way—viz., not by looking for them among the constellations as at present bounded and figured in our star-maps, but by searching the heavens themselves for them. this general method of search occurred to me very long ago while i was preparing various star-atlases, but the special mode of illustration here adopted occurred to me lately, while preparing for young astronomers in the united states a series of monthly maps showing the skies towards the north, south, east, and west, at different times of the night all the year round, and in various latitudes within the limits of the states. when i was in america i noticed, as i travelled about over a tolerably wide range of latitude, that the varying attitudes assumed by several of the constellations suggested features of resemblance to different objects. in constructing maps, simple in appearance, but based in reality on careful calculations, this characteristic came out more clearly. adopting a particular way of presenting the connection between the various stars of a constellation, i often found the figure suggested which had actually been associated with the group of stars thus connected. lastly, the idea of extending this method to other cases naturally occurred to me, and some of the results are presented in the present essay.

the method of delineation referred to is simply that of connecting the stars of a group by lines, ad libitum, that is, not merely introducing so many lines as will connect all the stars into a single set, but where necessary to complete the delineation of the imagined figure, adding other lines connecting pairs of stars belonging to the group, yet not so many that every pair of stars is connected by a line. the lines, again, need not be straight. on the contrary, where a group of stars forms a stream, the natural way of joining them is by lines so curved as to follow the serpentine course thus suggested. and in other cases a slight curvature of the lines joining pairs of stars will seem permissible, because corresponding to a configuration suggested by the stars themselves.

it is easily seen that in some of the simplest cases, the figure associated with a constellation is at once suggested by this method of delineation. for instance, take the case of the northern crown.

fig. 32.—the northern crown.

fig. 33.—the dolphin.

in this constellation we have a group which, while consisting of only a few stars, yet suggests very naturally the idea of a coronet of gems, as shown in fig. 32. the same is true also, though perhaps in less degree, of the dolphin, as shown in fig. 33. it is noteworthy, by the way, that this constellation can hardly have been invented by landsmen. for though in our own time when the pictures of sea-creatures are accurately drawn, so that persons who have never been to sea may have a correct idea of the figure of such creatures, in old times it was exceedingly unlikely that any but sailors would have such familiar knowledge of the dolphin as to be reminded of that creature by a group of stars.

fig. 34.—the scorpion.

a much more complex constellation than either of those just mentioned—the scorpion,—is even better represented by lineation, as shown in fig. 34. it is not, however, with cases so remarkable as these that the difficulty suggested at the outset is really connected. the instances of really remarkable resemblance are so few that they must be regarded as altogether exceptional. the best proof that the scorpion is unmistakably pictured by the stars is to be found in the fact that the modern map-makers have not in this case departed much from the older delineations. no one, in fact, who knows what a scorpion is like, could have any doubt as to the configuration of the body, at least, of the celestial scorpion. so that though such a case illustrates well the way in which the method of delineation i have suggested may be made to picture the object seen by the ancient observers in the heavens, it does not afford any answer to the difficulty indicated by those who assert that the great bear, the lion, the ship, and other of the old constellation figures, have no real existence among the stars.

before leaving the scorpion, however, i must call attention to one or two points which this remarkable constellation seems to establish. first, it is clear that in its case real resemblance suggested the association of a group of stars with a familiar object. since this resemblance remains, we infer that the group of stars presents now an appearance closely resembling that which it presented four or five thousand years ago. and as there is no special reason why the stars of the scorpion more than those of other constellations should retain their lustre unchanged, we gain a certain probability for the belief that all the constellations are now very much as they were when first named. indeed, it so happens that the region occupied by the scorpion is perhaps that part of the heavens where changes would on the whole most probably occur, the region of the milky way crossed by the scorpion being exceptionally irregular. we may note also that the part of the earth where the observers lived who called this constellation the scorpion must have been one where the reptile is well known, a conclusion which seems to dispose of the belief that the first astronomers lived in high latitudes.

let us, now, however, take some of the more difficult cases. we cannot do better, perhaps, than take at the very outset the great bear, a constellation of which many astronomers have asserted that it no longer presents and probably never did present the slightest resemblance to a bear.

i would lay down, in the first place, the hypothesis that the stars in the region of the heavens now occupied by the great bear must have reminded the earliest observers of a large, heavily-bodied, small-headed, short-eared, and short-tailed creature, such as either a bear or a hippopotamus. next, it may be taken for granted that the creature of which they were thus reminded was one with which they were familiar; and as we have already seen that the inventors of the oldest constellations cannot have lived in very high latitudes, we may conclude with great probability that the bear imagined in the heavens was not the polar bear, but the bear from which the first shepherd astronomers had to defend their herds and flocks,—the syrian bear, as it is commonly called, though the species inhabited also the greater part of asia minor in former times. the indians may be supposed to have seen the grizzly bear, not the smaller black bear, in the heavens. the features to be looked for, then, among the stars, are those common to the bears of comparatively low latitudes—not those of the polar bear.

so much premised we may proceed to inquire whether the region of the heavens occupied by the great bear presents such a creature with sufficient distinctness to suggest the idea of the animal to persons familiar with its aspect.

it is perhaps hardly necessary to remark that we must not expect to find a complete far less a perfect picture of a bear, or lion, or ship, in a large region of the heavens such as is occupied by these constellations. if some characteristic feature of a bear could be recognised in a group of stars, the ancient observer would be content to recognise the region of the heavens which would be occupied by the entire figure of the animal, as belonging to a great bear, unless some marked peculiarity in the stars of that region absolutely prevented the most lively imagination from conceiving a bear's body there. as an instance of the latter kind may be mentioned the bull and the ship, both of which constellation figures are seen only in part. the bull's head is exceedingly well marked, as is the stern of the ship argo, but the liveliest imagination cannot recognise the body and tail of a bull, or the fore-part of a ship, where these should be. consequently the ancients always regarded the bull as a half bull,

and (as aratus is careful to mention) they recognised only the stern of the good ship argo. but in general, where only some marked feature of an object could be imagined, or perhaps two or three, they yet conceived the whole object to be shown in the heavens, though it may have been altogether impossible to distribute the other stars over the remaining portion of the object in such a way as to show any natural association.

fig. 35.—the great bear.

the great bear seems to have been a constellation of this sort. one can recognise the head of an animal like the bear or the hippopotamus, and also the feet of such a creature, but the proper disposal of the stars forming the animal's body is not so easy. this would not interfere, however, with the choice of the bear to represent the region of stars occupied by the constellation. every one who has seen faces and figures in the fire—and who has not?—knows that one or two features will suffice to suggest a resemblance; either the imagination does all the rest, or else the idea is suggested that some other object partially conceals that portion of the imagined figure which is wanting.

fig. 35 shows how, as i conceive, a bear was figured in the heavens by those who, in various nations, gave to the stars of this part of the sky the name of the great bear.

fig. 36.—the bear's head.

it will be noticed in the first place that the famous septentriones (the seven stars of the plough, as in england the set is called, the dipper as it is called in america, the corn-measurer as it was called by the ancient chinese) has little or nothing to do with the configuration of the bear, though forming a part of the constellation. it is the set of small stars forming the head which seems to have suggested the idea of a bear, though two of the paws are also well defined by the stars. but the outlining of the head of a bear or hippopotamus is really sufficiently close to require no very lively imagination to fill it in. fig. 36, giving these stars only, serves to show this, i think. that the entire figure of a bear or hippopotamus was not recognised seems further shown by the figure assigned to the constellation in the zodiac of tentyra, or denderah, where it appears as in fig. 37. the smaller figure is supposed to represent the little bear.

fig. 37.—the constellations of the bears, represented as a hippopotamus (?) and wolf (?) in the denderah zodiac.

in the second place, the reader familiar with the constellations will perceive that several stars not at present appertaining to the great bear are included within the configuration itself of the animal in fig. 35. thus the third magnitude star behind the right ear belongs to the constellation of the dragon; the third magnitude star near the hind quarters is cor caroli, the chief star of the modern constellation canes venatici, or the hunting dogs. it appears to me that we ought not to expect that the first observers of the heavens, in recognising imaginary features of resemblance between a group of stars and some known object, would be careful to inquire whether some among those stars were included in a group which they had compared or might afterwards compare with another object. it is very necessary for the astronomer of our time, nay, it may have even been very necessary for the astronomers of the times of hipparchus, ptolemy, etc., to have the limits of the constellations clearly defined, and to let no conspicuous star be common to different constellations. but as regards the figures fancied in the heavens by the first observers of the stars, considerations of that sort would be of no importance whatever. indeed, it is worthy of notice that even so late as the time of bayer, who gave to the stars their greek letters, the constellations were not separated from each other. he called the star now known as beta tauri only, gamma auriga also, so that now auriga has stars alpha, beta, delta, and so forth, but no gamma. similarly, we look in vain for any star delta in the constellation pegasus, simply because bayer called one and the same star alpha andromed? and delta pegasi, the astronomers of our own time retaining only the former name for this star,—the bright one adorning the head of andromeda. even in our time it has been found impossible properly to separate the older constellations from each other, so that to this day the scorpion remains entangled with the legs of ophiuchus, who is further inextricably mixed up with the serpent. in fact, the serpent is divided into two separate parts by the body of ophiuchus, map-makers having no choice but either to allow ophiuchus to divide the serpent, or the serpent to divide ophiuchus.

fig. 38.—the original constellation of the lion.

in the next case, that of the great lion, we have still further to depart from the modern configuration of the constellation. no one can imagine the remotest resemblance between any part of a lion and the grouping of stars falling on the corresponding portion of leo in the modern constellation. the nose of the lion now falls near λ (fig. 38); μ and ρ forming the outline of the mane, β the end of the tail, ε the nearer fore-paw, τ the nearer hind-paw. the original lion, i cannot doubt, was imagined somewhat as pictured in fig. 38. the head and mane are unmistakably pictured among the stars, the paws fairly, the relatively small quarters and the tufted tail exceedingly well—always remembering that anything like very close resemblance is not to be looked for between a widely extended group of stars and the figure of an animal or other large object. if we remember also that uncultured nations, like children, are much quicker in imagining resemblances than those carefully trained to recognise the artistic delineation of objects, we cannot be surprised to find that nearly all those nations who were acquainted with the lion imagined a large leonine figure in the part of the heavens now centrally occupied by our modern and most puny lion, but including portions of cancer, the whole of leo minor (one of hevelius's absurd inventions), the hair of berenice, and a star or two belonging to virgo.

fig. 39.—the original ship "argo."

we have to treat in a similar way the constellation argo of our present maps, to get the good ship argo, as the ancients must have conceived the constellation. fig. 39 shows the ship as i imagine she was originally pictured. the stars which mark her curved poop belong in part at present (as doubtless they have long belonged) to the larger dog, while those which mark the steering-oar belong to the modern constellation columba noachi, or noah's dove. it must be observed that the bright star canopus, shown in the water, was not visible in the time of the first observers in the latitude where they probably dwelt. the mighty gyrating motion of the earth has caused these stars to be brought five or six degrees further from the southern pole of the heavens. but canopus and a few of the small stars near it are the only stars which have thus been added to the constellation as seen from the regions inhabited by the first observers. (canopus was known to the arabian and egyptian astronomers.)

this introduces another point which seems worth noticing. at present the ship argo is never seen from any part of the earth's surface as pictured in fig. 39. when due south, the position whence in all northern latitudes the constellation is most favourably seen, the ship is always tilted up at the stern: one would say, in more nautical phrase, she is down by the head, if the ship had any fore-part; but from time immemorial she has been a half-ship only. some 4,000 years ago, however, argo stood nearly on an even keel when due south. again, it is to the mighty gyrational motion of the earth that we have to look for the cause of the great change in the apparent position of the ship. the sphere of the fixed stars has remained all the time unchanged, or very nearly so, but the direction in which the earth's axis of rotation points has swayed round (much as the axis of a reeling top sways round) through about one-sixth part of a complete gyration.

in the regions where astronomy first began as a science, argo not only stood on an even keel but almost on the horizon when due south; and the features of resemblance to a ship, which i have endeavoured to portray in fig. 39, must have seemed much more striking there (and then) than now.

the fore-part of the ship, or rather that region of the heavens where the fore-part should be, is occupied by great masses of the milky way in one of its brightest and most remarkable portions. i have sometimes fancied that in some of the old zodiac temples of star-worshippers the constellation argo was depicted as a mighty ship, gemmed with stars, and heavily laden in its fore-part with great masses of gilded cloud to represent the milky way, and that from such representations of the constellation came the tradition of the ship argo and its cargo of golden fleece. many parts of the story of jason and his companions seem to relate to objects depicted in the old constellation-domes,—as those relating to the dragon, to hercules, castor and pollux, the centaur, etc. there is also a curious reference, in the tradition, to the stern of the ship, which is much like what we can imagine as resulting from an attempt to explain the appearance of this part only, in the set of constellation figures. we read that the entrance to the euxine sea was fabled to be closed up by certain rocks called symplegades (the clashers), which floated on the water, and when anything attempted to pass through came together with such velocity that not even birds could escape. phineas advised them to let a bird fly through, and if the bird passed safely, to venture the passage. it passed with only the loss of its tail; and the argo, favoured by juno, and impelled by the utmost efforts of its heroic crew, passed also, though so narrowly that the meeting rocks carried away part of her stern-works, which remained fixed there thenceforward.

for my own part, i think we may not only regard the story of the ship argo as in reality a version, though much modified, of the account of noah's deluge, but consider the series of constellations, aquarius, cetus, eridanus, argo, corvus, centaurus, ara, and sagittarius, as typifying the same narrative. it is somewhat curious that if we place these constellations in their original position,—that is, as they were before the changes which the earth's great gyration has introduced during the last four thousand years or so,—we find the following coincidences with the account of the deluge. first comes aquarius (whose beginning would correspond with the sun's position on or about the seventeenth day of the second month of the old pleiades year) pouring water. his range on the ecliptic (or the space he occupies in the annual range represented in the zodiac temple) is about forty days. then came the watery constellations eridanus, the river, and cetus, the sea monster, having, with the ship argo, a range of about 150 days of the annual circuit. about forty days later in the circuit we find corvus, the raven, whose feet rest on hydra, the great celestial sea-serpent, as though no dry land could be found by the bird. a dove also, if we accept the interpretation above given of the argo narrative, may have been represented in this part of the star temple. next we have the centaur, originally we know represented as a man only, offering an animal as sacrifice on the altar ara. there is a cloud of stars rising from the altar: we may recall manilius's account of the constellation,—

"ara, ferens thuris, stellis imitantibus, ignem."[16]

in this cloud is the bow of sagittarius, the bow being originally alone shown, as it is indeed the only figure which can be imagined among the stars of this region. so that these constellation figures seem to typify noah offering sacrifice on the altar, and the bow of promise set in the cloud above the altar. it is curious, too, that while the time of noah's leaving the ark was a year and ten days from the beginning of the rains, the constellation sagittarius overlaps the conjoined watery signs capricornus and aquarius (running south of them) by about so much as would correspond to ten days of the annual circuit of the heavens.

the objections to the view of matters above indicated are, first, that the constellations referred to seem to have been formed because of real resemblance between the star-groups and the figures associated with them; and, secondly, that the zodiac temples were probably erected by star-worshippers, and would scarcely have been employed to typify such a narrative as that of the deluge. the theory that the narrative itself was an attempt to interpret pictures represented on a zodiac temple will, of course, be objectionable to many readers; though they may not be unwilling to believe that the fable of the argonautic expedition had its origin in some such way.

it will have been noticed that in the figures which i have given of the great bear, lion, and ship, i have not altogether adhered to my idea of simply connecting the stars of a group by lines. to say the truth, although a rough notion of a bear, lion, or ship may thus be given, the figure so presented is not altogether satisfactory to the mind. in any case, as for instance even in the scorpion (of all these figures the best marked), the line-figure is very imperfect. but in some cases it does suggest the idea of an animal or figure, or a part of either, much in the same way that the idea of a human figure can be suggested by a few lines forming a skeleton figure, such as our old friend tommy traddles used to draw. now the lion, bear, and ship are not well suited for this sort of delineation, as anyone will find who tries to suggest the idea of a bear, lion, or ship (of the old-fashioned heavily-sterned sort) by means of a few lines.

in order, however, to show that in some cases a skeleton figure can be formed by joining the stars of a constellation, and that the figure thus formed represents (of course in an utterly inartistic sort of way) the object associated with those stars, i will now take one or two instances in which such resemblance suggested itself to me without being specially sought for. i might add to the crown, dolphin, and scorpion, the chair of cassiopeia, the figure of orion, and the constellation of the cup; i omit these, however, not because they are unfit for my purpose, but because they so obviously illustrate my argument. no one, with the least power of imagination, can fail to see how a chair, a belted giant, and a cup, are pictured, as it were, in these constellations. i will take others where the resemblance is less obvious.

thus, i think scarcely anyone who is acquainted with the constellation andromeda can have failed to be perplexed by the association of the figure of a chained lady with this group of stars. in the arrangement of the stars themselves, without lines drawn to connect them, no such figure can be imagined; at least i fail utterly for my own part when i attempt to picture such a figure, even now that i recognise how the figure is formed, skeleton-wise, by connecting lines. i cannot but think this figure must have been imagined from pictures of the groups of stars with lines connecting them, and not from the stars themselves. there is this reason, among others, for so thinking. the lady's head is represented by a single star, alpherat. now a single star in the sky, however bright, is not large enough to represent the head of a human figure like andromeda's. but the representation of a bright star like alpherat in a chart or sculpture has sufficient size to serve for a head, because size is the only way in which brightness can be indicated.

in fig. 40 the stars forming the constellation andromeda are shown; also the chair of cassiopeia; and, on the right, one of the fishes and the triangle. a group of stars in the upper left-hand corner marks the place of the rock to which the chains are fastened which bind andromeda's right hand.

fig. 40.—andromeda.

it cannot be said that the skeleton picture shewn in fig. 40 is very graceful or artistic; but, on the other hand, it cannot, i think, be doubted that there is enough in it to suggest the idea of a chained person. the fish naturally suggests the idea that the place is by the sea-shore. and the chair suggests the idea of some one on the shore waiting and watching. in our own time, probably, the idea suggested would be that of a person taking a bath, while some one sat in a chair on the sands and waited for their turn. but to the old observers of the heavens, unfamiliar as they were with sea-side diversions, the notion would more naturally occur of a woman chained to a rock,

lifting her long white arms, widespread, to the walls of the basalt;

while not far off was imagined among the stars the monster cetus coming onward,

bulky and black as a galley,

lazily coasting along, as the fish fled leaping before it.

one of these fish is seen close by the figure of the chained andromeda. near at hand they imagined the father and mother of the lady; cassiopeia sitting close to the shore; but

cepheus far in the palace

sat in the midst of his hall, on his throne, like a shepherd of people,

choking his woe dry-eyed, while the slaves wailed loudly around him.

the story of andromeda, as the reader doubtless knows, is not of greek origin. its real origin is lost in a far antiquity. the indians have the same story in their astronomical mythology, and almost the same names. thus wilford, in his asiatic researches, relating his conversation with an indian astronomer, says, "i asked him to show me in the heavens the constellation of antarmada, and he immediately pointed to andromeda, though i had not given him any information about it beforehand. he afterwards brought me a very rare and curious work in sanscrit, which contained a chapter devoted to upanachatras, or extra-zodiacal constellations, with drawings of capuja (cepheus), and of casyapi (cassiopeia), seated and holding a lotus flower in her hand; of antarmada, chained, with the fish beside her; and last, of parasiea (perseus), who, according to the explanation of the book, held the head of a monster which he had slain in combat; blood was dropping from it, and for hair it had snakes."

as another illustration of the method i have described, i give the constellation pegasus, or, as it was sometimes called, the half-horse. i do not assert that fig. 41 presents a very well shaped steed, any more than that in fig. 40 a lady of exquisite proportions is pictured. but one can perceive how the stars suggest the idea of a horse in one case, and of a human figure with upraised fastened arms in the other. it is commonly stated that pegasus is one of the constellations showing no resemblance at all to the figure associated with it. i think fig. 41 suffices to show that there is some slight resemblance at least.

fig. 41.—pegasus.

it may be mentioned, in passing, that all the nations of antiquity would not be likely to form equally clear conceptions of figures in the heavens. there are marked differences between the various races of the human family in this respect, just as there are marked differences between various persons in the power of imagining figures under different conditions. some persons see figures at once in a cloud, in the outline of a tree, in a fire, in a group of accidental markings, and so forth; while others not only do not see such figures, but cannot imagine them even when their outlines are indicated. so it is with different races of men. there have been some which, even when only just emerging from the utterly savage state, possessed so much of the imaginative power as to be able to picture for themselves, by lines cut with rude flint instruments on pieces of bone, horn, or ivory, the animals with which they were familiar. we have even among such pictures some belonging to an age so remote that the mammoth (or hairy elephant) had not yet entirely disappeared from europe; for, in the cave of la madeleine, at dordogne, among other relics of the stone age, there has actually been found a drawing of the mammoth scratched on a piece of mammoth tusk. on the other hand, there are some races in existence at the present day, in a more advanced stage of civilization, who cannot perceive even in well-executed coloured drawings any resemblance to the objects pictured. an aboriginal new hollander, says oldfield, "being shown a coloured engraving" of a member of his own tribe, "declared it to be a ship, another a kangaroo, and so on; not one of a dozen identifying the portrait as having any connection with himself." a rude drawing, with all the lesser parts much exaggerated, they can realise. thus, to give them an idea of a man, the head must be drawn disproportionately large. dr. collingwood tells us that when he showed a copy of the illustrated london news to the kibalaus of formosa, he found it impossible to interest them by pointing out the most striking illustrations, "which they did not appear to comprehend." denham (i quote throughout from lubbock's most valuable and interesting work on the origin of civilization) says that bookhaloum, a man otherwise of considerable intelligence, though he readily recognised figures, could not understand a landscape. "i could not," he says, "make him understand the print of the sand-wind in the desert, which is really so well described by captain lyons' drawing. he would look at it upside down; and when i twice reversed it for him he exclaimed, 'why! why! it's all the same.' a camel or a human figure was all i could make him understand, and at these he was all agitation and delight. 'gieb! gieb!—wonderful! wonderful!' the eyes first took his attention, then the other features; at the sight of the sword, he exclaimed, 'allah! allah!' and, on discovering the guns, instantly exclaimed, 'where is the powder?'"

we have in the consideration of this diversity of character between different races and nations, as respects the power as well of imagining as of delineating figures (the two are closely connected), one means of judging to what race we owe the original constellations. for although some figures in the heavens are manifest enough, others require a considerable power of imagination. and it should be noted that this must have been true even if we suppose (which i think i have succeeded in showing we need not do) that many of the stars have changed in brightness, and that thus resemblances have disappeared which formerly existed. for, in any case, the heavens four, ten, or twenty thousand years ago, or at whatever remote period we set the original invention of the constellations, must have presented the same characteristics as at present. it can never have been the case that all the star-groups could be compared at once, obviously, with the figures of men and animals. so that only a race of lively imagination could have found figures for all the star-groups, as was certainly done in very remote times by some race.

the race, then, to whom we owe the general system of constellations, was probably one with so much talent for artistic delineation that in later ages this people would have become distinguished for skill in painting and sculpture. i think the sculptures found in babylon, and the traditions left of the artistic skill of the babylonians, correspond well with the belief that the constellations had their origin, and astronomy its first development, among that people or a kindred race.

but the chief lesson to be derived (and i think it may fairly be derived) from the study of the constellation-groups is, that enough resemblance still remains, if only the arbitrary boundaries invented for the constellation figures in recent times are overlooked, to assure us that no very great changes have taken place in the aspect of the heavens for thousands of years. a few stars here and there have certainly changed greatly in brightness, and some few have changed considerably even in position; while a considerable number have probably changed slightly in brightness, and all, or very nearly all, have changed somewhat in position. but on the whole the aspect of the stellar heavens now is the same as it was when the constellation figures were first imagined.

this thought not only assures us of the permanence of our own sun (seeing that among the thousands of his fellow-suns which spangle the heavens so few have changed in lustre), but seems to me to give to the study of the stars a singular charm. our antiquaries and arch?ologists present for our study the relics of long past ages, and we may often rest assured that the objects thus gathered for us were really used in old times, though probably in a manner not understood by us, and when in a condition very unlike that in which they have reached our times. in nearly all such instances, however, doubt exists as to the antiquity of the relic, as to the race to whom it really belonged, and as to its real use and purport. but as regards the stellar heavens we have no doubt. of all the objects on which the eyes of remote races have rested, the celestial bodies are undoubtedly the most ancient, while at the same time they and they alone were most certainly contemplated by all mankind. from the very earliest ages, from the time when the child-man first turned his thoughts from mere animal wants to the wonders of nature, the stars, and the sun and moon and planets must have drawn to themselves the attention of all who had eyes to see even though they had no power to understand the glories of the star-depths. men pictured among the stars the objects most familiar to them, the herds and flocks which they tended, the herdsman himself, the waggoner, the huntsman, the birds of the air, the beasts of the field, the fishes of the sea, the ship, the altar, the bow, the arrow, and, one may say, all that according to their knowledge existed in the heavens above, in the earth beneath, and in the waters under the earth. imperfect and anomalous as these meanings are, in relation to modern astronomy, with its exact methods, elaborate instruments, and profound investigations into the meaning of all the phenomena of the heavens, they nevertheless retain their place, and are likely long to do so, in virtue of the hold which they took, in remote ages, on the imagination of mankind in general.