In the last paper we saw how badly off for water Mars, to all appearance, is; so badly off that any inhabitants of that other world would have to irrigate to live. As to the actual presence there of such folk, the broad physical characteristics of the planet have nothing to say beyond a general expression of acquiescence, but they do have something very vital to say about the conditions under which alone their life could be led. They show that in these Martian minds there would be one question paramount to all the local labor, women's suffrage, and Eastern questions put together, the water question. How to procure water enough to support life would be the great communal problem of the day.

If Mars were the earth, we might well despair of detecting signs of any Martians for some time yet. Across the gulf of space that separates us from Mars, an area thirty miles wide would just be perceptible as a dot. It would, in such case, be hopeless to look for evidence of folk. Anything like London or New York, or even Chicago in anticipation, would be too small to be seen. So sorry a figure does man cut upon the earth he thinks to own. From the standpoint of forty millions of miles distance, probably the only sign of his presence here would be such semi-artificialities as the great grain-fields of the West when their geometric patches turned with the changing seasons from ochre to green, and then from green to gold. By his crops we should know him, a telltale fact of importance because probably the more so on Mars.

For Mars is not the earth. Conditions hold there which would necessitate a different state of things, inorganic and organic, apparently a much more artificial one. If cultivation there be, it must be cultivation upon a much more systematic scale, due in large part to a system of irrigation; just as any Martians must be quite different physically from men. Now, at this point in our investigation, when the broad features of Mars disclose conditions which imply irrigation as their organic corollary, we are suddenly confronted on the planets face with phenomena so startlingly suggestive of this very thing as to seem the uncanny realization of the deduction. Indeed, so amazingly lifelike is their appearance that, had we possessed our present knowledge of the planets physical condition before, we might almost have predicted what we see as criterion of the presence of living beings. What confronts us is this: —

When the great continental areas, the reddish-ochre portions of the disc, are attentively examined in sufficiently steady air, their desert-like ground is seen to be traversed by a network of fine, straight dark lines. The lines start from points on the coast, of the blue-green regions, commonly well marked bays, and proceed direct to other equally well-marked points in the middle of the continent. At these latter termini the lines meet, very surprisingly, other lines that have come there from different starting-points in a similarly definite manner. And this state of things exists all over the reddish ochre regions.

All the lines, with the exception of a few that are curved in a regular manner, are absolutely straight from one end to the other. They are arcs of great circles, taking the shortest distance between their termini. The lines are as fine as they are straight. As a rule, they are of scarcely any perceptible breadth, seeming on the average to be less than a Martian degree, or between twenty and thirty miles, wide. Some are broader; some even finer, possibly not above fifteen miles across. Their length, not their breadth, renders them visible; for though at such a distance we could not distinguish a dot less than about thirty miles in diameter, we could see a line of much less breadth, because of its length. Speaking generally, however, the lines are all of comparable width.

Still greater uniformity is observable in the different parts of the same line; for each line maintains its individual width throughout. Although at and near the point where it leaves the dark regions, or the Solis Lacus, for the same phenomenon appears there, some slight enlargement seems to take place, after it has fairly started on its course it remains substantially of the same size from one end to the other. As to whether the lines are even on their edges or not, I should not like to say, but the better they are seen, the more even they look. It is not possible to affirm positively on the point, as they are practically nearer one dimension than two.

On the other hand, their length is usually great, and in some cases enormous. A thousand or fifteen hundred miles may be considered about the average. The Ganges, for example, which is not a long one as Martian canals go, is about 1450 miles in length. The Brontes, one of the newly discovered, radiating from the Gulf of the Titans, extends over 2400 miles. Among really long ones, the Eumenides, with its continuation the Orcus, the two being in truth one line, runs 3540 miles from the point where it leaves the Phoenix Lake to the point where it enters the Trivium Charontis; throughout this great distance, nearly equal to a diameter of the planet, deviating neither to the right nor to the left from the great circle upon which it set out. On the other hand, the shortest line is the Nectar, which is only about 250 miles in length; sweetness being, according to Schiaparelli its christener, as short-lived on Mars as elsewhere.

That with very few exceptions the lines all follow arcs of great circles is proved: first, by the fact that when near the centre of the disc they show as straight lines; second, that when seen toward its edges they appear curved, in keeping with the curvature of a spherical surface viewed obliquely; third, that when the several parts of some of the longer lines are plotted upon a globe they turn out to lie in one great circle. Apparent straightness throughout is only possible in short lines. For a very long arc upon the surface of a revolving globe tilted toward the observer to appear straight in its entirety it must lie in certain positions. It so chances that these conditions are fulfilled by the canal called the Titan. The Titan starts from the Gulf of the Titans, in south latitude 20~, and runs due north almost exactly upon the 169th meridian for an immense distance. I have followed it over 2300 miles down the disc to about 430 north, as far as the tilt of the planets axis would permit. As the rotation of the planet swings it round, it passes the central meridian of the disc simultaneously throughout its length, and at that moment comes out strikingly straight, a substantialized meridian itself.

Although each line is the arc of a great circle, the direction taken by this great circle may be any whatsoever. The Titan, as we have seen, runs nearly due north and south. Certain canals crossing this run, on the contrary, almost due east and west. There are others, again, belting the disc at well-nigh every angle between the two. Nor is there any preponderance, apparently, for one direction as against any other. This indifference to direction is important as showing that the rotation of the planet has no direct effect upon the inclination of the canals.

But, singular as each line looks to be by itself, it is the systematic network of the whole that is most amazing. Each line not only goes with wonderful directness from one point to another, but at this latter spot it contrives to meet, exactly, another line which has come with like directness from quite a different starting-point. Nor do two only manage thus to rendezvous. Three, four, five, and even seven will similarly fall in on the same spot, a sociability which, to a greater or less extent, takes place all over the surface of the planet. The disc is simply a network of such intersections. Sometimes a canal goes only from one intersection to another; more commonly it starts with right of continuation, and, after reaching the first rendezvous, goes on in unchanged course to several more.

The result is that the whole of the great reddish-ochre portions of the planet is cut up into a series of spherical triangles of all possible sizes and shapes. What their number may be lies quite beyond the possibility of count at present; for the better our own air, the more of them are visible. About four times as many as are down on Schiaparelli's chart of the same regions have been seen at Flagstaff. But before proceeding further with a description of these Martian phenomena, the history of their discovery deserves to be sketched, since it is as strange as the canals themselves.

The first hint the world had of their existence was when Schiaparelli saw some of the lines in 1877, now eighteen years ago. The world, however, was anything but prepared for the revelation, and, when he announced what he had seen, promptly proceeded to disbelieve him. Schiaparelli had the misfortune to be ahead of his times, and the yet greater misfortune to remain so; for not only did no one else see the lines at that opposition, but no one else succeeded in doing so at subsequent ones. For many years fate allowed Schiaparelli to have them all to himself, a confidence he amply repaid. While others doubted, he ~vent from discovery to discovery. What he had seen in 1877 was not so very startling in view of what he afterward saw. His first observations might well have been of simple estuaries, long natural creeks running up into the continents, and so cutting them in two. His later observations were too peculiar to be explained even by so improbable a configuration of the Martian surface. In 1879, the canali, as he called them (channels, or canals, the word may be translated, and it is in the latter sense that he now regards them), showed straighter and narrower than they had in 1877: this not in consequence of any change in them, but from his own improved faculty of detection; for what the eye has once seen it can always see better a second time. As he gazed they appeared straighter, and he made out more.

Lastly, toward the end of the year, he observed, one evening, what struck even him as a most startling phenomenon, the twinning of one of the canals: two parallel canals suddenly showed where but a single one had showed before. The paralleling was so perfect that he suspected optical illusion. He could, however, discover none by changing his telescopes or eyepieces. The phenomenon, apparently, was real. At the next opposition lie looked to see if by chance he should mark a repetition of this strange event, and went, as he tells us, from surprise to surprise; for one after the other of his canals proceeded startlingly to become two, until some twenty of them had thus doubled. This capped the climax to his own wonderment, and, it is needless to add, to other peoples incredulity; for nobody else had yet succeeded in seeing the canals at all, let alone seeing them double. Undeterred by the general skepticism, he confirmed, at each fresh opposition, his previous discoveries; which, in view of the fact that no one else did, rather tended in astronomical circles to the opposite result.

For nine years he labored thus alone, having his visions all to himself. It was not till 1886 that any one but lie saw the canals. In April of that year Perrotin at Nice first did so. The occasion was the setting-up of the great Nice glass of twenty-nine inches aperture. In spite of the great size of the glass, however, a first attempt resulted in nothing but failure. So did a second, and Perrotin was on the point of abandoning the search altogether when, on the 15th of the month, he suddenly detected one of the canals, the Phison. His assistant, M. Thollon, saw it immediately afterward. After this they managed to make out several others, some single, some double, substantially as Schiaparelli had drawn them; the slight discrepancies between their observations and his being, in point of fact, the best of confirmations.

Since then other observers have contrived to detect the canals, the list of the successful increasing at each opposition, although even now their number might almost be told on ones hands and feet. The fact that so few men have yet seen these lines is due to poor air. That in ordinary atmosphere the canals are not easy objects is certain; while for the detection of their peculiar fineness and straightness a steady air is essential. So also is attentive perception on the part of the observer, size of aperture being distinctly a secondary matter. That Schiaparelli discovered the canals with an 8~ object-glass, and that the 26-inch at Washington has refused to show them to this day, are facts that speak with emphasis on the point.

Although skepticism as to the existence of the so-balled canals seems now pretty well dispelled, disbelief still makes a desperate stand against their peculiar appearance, dubbing accounts of their straightness and duplication as sensational, whatever that may mean in such connection; for that they are both straight and double, as described, is certain, a statement I make after having seen them instead of before doing so, as is the case with the gifted objectors. Doubt, however, will not wholly cease till more people have seen them, which will not happen till the importance of atmosphere in the study of planetary detail is more generally appreciated than it is to-day. To look for the canals with a large instrument in poor air is like trying to read a page of fine print kept dancing before ones eyes, and increase of magnification increases the motion. Advance in our study of other worlds depends upon choosing the very best atmospheric sites for our observatories.

As we shall now have to call these Martian things by our names, that is, it may be well to consider cursorily the nomenclature which has been evolved on the subject. Unfortunately, the planet has been too much benamed, — benamed, indeed out of all recognition. There are no less than five or six systems current for its general topographical features. The result is that it has become something of a specialty just to know the names. The Syrtis Major, for example, appears under the following aliases: the Syrtis Major, the Mer du Sablier, the Kaiser Sea, the Northern Sea, to say nothing of translations of these, such as the Hour-Glass Sea. After which ample baptism it is a trifle disconcerting to have the sea turn out, apparently, not to be a sea at all. Everybody has tried his hand at naming the planet, first and last; naming a thing being mans nearest approach to creating it. Proctor made a chart of the planet, and named it thoroughly; Flammarion drew another chart, and also named it thoroughly, but differently; Green made a third map, and gave it a third set of names; Schiaparelli followed with a fourth, and furnished it with a brand-new set of his own; and finally W. H. Pickering found it necessary to give a few new names, just for particularization. To know, therefore, what part of the planet anybody means when he mentions it, one has to keep in his head enough names for five worlds. To cap which, it is to be remarked that not one of them is the things real that is, its Martian name, after all!

Fortunately, with the canals matters are not so desperate, because so few people have seen them. Schiaparelli's monopoly of the sight pleasingly prevented, in their case, christening competition. What is more, lie named them very judiciously and most picturesquely after mythologic river names. Where he got his names is another matter. Whether he started by being as learned in such lore as lie afterward became may well be doubted. Certainly, one of the greatest discoveries made at Flagstaff has been the discovery of the meaning of Schiaparelli's names; some of them still defying the penetrating power of the ordinary encyclopedia. Among them are classical mythologic ones of the class known only to that himself mythical character, Macaulay's every schoolboy, which speaks conclusively for their reconditeness. Others, I firmly believe, even that omniscient schoolboy can never have heard of. Want of space here precludes instances; but as a simple example I may say that the translation to Mars of the Phison and the Gehon, the two lost rivers of Mesopotamia, satisfactorily accounts for their not being found on earth by modern explorers.

With due mental reservation as to their meaning, I have adopted Schiaparelli's names, and where it has been necessary to name newly discovered canals have conformed as closely as possible to his general scheme. If even in an instance or two I have hit upon names that are incomprehensible, I shall feel that I have not disgraced my illustrious predecessor. For a brand new thing no name is so good as one whose meaning nobody knows, except one that has no meaning at all.

Schiaparelli's scheme embraces all the other Martian features as well as the canals, and the same poetic imagination pervades the whole. For example, the central promontory of what used to be known as Dawes Forked Bay, a prominent point, since it has for some time been used as the zero meridian for Martian longitudes, he calls the Fastigium Aryn. The Fastigium Aryn was, it appears, the cupola of the world, a mythic spot supposed to be the absolute centre of the earth regarded as a plane in mid-heaven, a point midway between the north and south, the east and west, the zenith and nadir; an eminently suitable name, indeed, for the origin of longitudes and the beginning of time.

To return now to the objects of so much human incredulity. The first point worth noting about them is that their actual existence is quite beyond question; the second, that the better they are seen, the odder they look. Observations at Arequipa in 1892 not only confirmed Schiaparelli's, but extended the canal system considerably both in quantity and in character; observations last year at Flagstaff extended it still further, so that now we know of about half as many more canals as are down on Schiaparelli's chart, and of certain phenomena connected with them no less peculiar, to say the least, than themselves. What these strange dependencies are we will note after we have considered the canals.

So far we have regarded the canals only statically, so to speak; that is, we have sketched them as they would ap. pear to any one who observed them in sufficiently steady air, once, and once only. But this is far from all that a systematic study of the lines will disclose. Before, however, entering upon this second phase of their description, we may pause to note how, even statically regarded, the aspect of the lines is enough to put to rest all the theories of purely natural causation that have so far been advanced to account for them. This negation is to be found in the supernaturally regular appearance of the system, upon three distinct counts: first, the straightness of the lines; second, their individually uniform width; and third, their systematic radiation from special points.

On the first two counts we observe that the lines exceed in regularity any purely natural regularity of which we commonly have cognizance. Physical processes never, so far as we know, produce perfectly regular results; that is, results in which irregularity is not also plainly discernible. Disagreement amid conformity is the inevitable outcome of the many factors simultaneously at work.

From the orbits of the heavenly bodies to philotaxis and human features, this diversity in uniformity is apparent. As a rule, the divergences, though small, are quite perceptible; that is, the lack of absolute uniformity is comparable to the uniformity itself, and not of the negligible second order of unimportance. In fact, it is by the very presence of uniformity and precision that we suspect things of artificiality. It was the mathematical shape of the Ohio mounds that suggested mound-builders; and so with the thousand objects of every-day life. Too great regularity is in itself the most suspicious of circumstances that some finite intelligence has been at work. If it be asked how, in the case of a body so far off as Mars, we can assert sufficient precision to imply artificiality, the answer is twofold: first, that the better we see these lines, the more regular they look; and second, that the eye is quicker to perceive irregularity than we commonly note. It is indeed surprising to find what small irregularities will shock the eye.

The third count is, if possible, yet more conclusive. That the lines form a system; that, instead of running anywhither, they join certain points to certain others, making thus, not a simple network, but one whose meshes connect centres directly with one another, is striking at first sight, and loses none of its peculiarity on second thought. For the intrinsic improbability of such a state of things arising from purely natural causes becomes evident on consideration.

Were lines drawn haphazard over the surface of a globe, the chances are ever so many to one against more than two lines crossing each other at any point. Simple crossings of two lines would of course be common in something like factorial proportion to the number of lines, but that any other line should contrive to cross at the same point would be a coincidence whose improbability only a mathematician can properly appreciate, so very great is it. If the lines were true lines, without breadth, the chances against such a coincidence would be infinite; and even had the lines some breadth, the chances would be enormous against a rendezvous. In other words, we might search in vain for a single instance of such encounter. On the surface of Mars, however, instead of searching in vain, we find the thing occurring passim; this ~ priori most improbable rendezvousing proving the rule, not the exception. Of the crossings that are best seen, almost all are meeting-places for more than two canals.

To any one who had not seen the canals, it would at once occur that something of the same improbability might be fulfilled by cracks radiating from centres of explosion or fissure. But such a supposition is at once negated by the uniform breadth of the lines, a uniformity impossible in cracks, whose very mode of production necessitates their being bigger at one end than the other. We see examples of what might result from such action in the cracks that radiate from Tycho, in the moon, or, as we now know from Professor W. H. Pickering's observations, from the craterlets about it. These cracks bear no resemblance whatever to the lines on Mars. They look like cracks; the lines on Mars do not. Indeed, it is safe to say that the Martian lines would never so much as suggest cracks to any one. Lastly, the different radiations fit into one another absolutely, an utter impossibility were they radiating rifts from different centres.

In the same way, we may, while we are about it, show that the lines cannot be several other things which they have, more or less gratuitously, been taken to be. They cannot, for example, be rivers; for rivers could not be so obligingly of the same size at source and mouth, nor would they run from preference on arcs of great circles. To do so, practically invariably, would imply a devotion to pure mathematics not common in rivers. They may, in some few instances, be rectified rivers, which is quite another matter. Glaciation cracks are equally out of the question: first, for the causes above mentioned touching cracks in general; and second, because there is, unfortunately, no ice where they occur. Nor can the hues be furrows ploughed by meteorites, another ingenious suggestion, since in order to plough, invariably, a furrow from one centre to another, without either swerving from the course or overshooting the mark, the visitant meteorite would have to be carefully trained to the business.

Such are the chief purely natural theories of the lines, excluding the idea of canals, theories advanced by persons who have not seen them. No one who has seen the lines well has or could advance them, inasmuch as they are not only disproved by consideration of the character of the lines, but instantly confuted by the mere look of them.

Schiaparelli supposes the canals to be canals, but of geologic construction. He suggests, however, no explanation of how this is possible; so that the suggestion is not, properly speaking, a theory. That eminent astronomer further says of the idea that they are the work of intelligent beings, "Io mi quadero bene dal combattere questa supposizione la quale nulla include d'impossible". (I should carefully refrain from combating this supposition, which involves no impossibility.) In truth, no natural theory has yet been advanced which will explain these lines, while recent observations furnish material that seems to render artificial construction probable.

After so much necessary digression on what the canals are not, we will resume our inquiry as to what they are.

So far we have considered their aspect at any one time, and we have seen that it is such as to defy natural explanation, and to hint that in these lines we are regarding something other than the outcome of purely natural causes. Indeed, such is the first impression upon getting a good view of them. How instant this inference is becomes patent from the way in which drawings of the canals are received by incredulously disposed persons. The straightness of the lines is unhesitatingly attributed to the draughtsman. Now it is to be remembered that accusation of design, if it prove inapplicable to the draughtsman, devolves ipso facto upon the canals.

We come next to a consideration of their successive appearances night after night, and month after month. After the fundamental fact that such curious phenomena as the canals are visible is the scarcely less important one that they are not always so. At times the canals are invisible, and this invisibility is real, not apparent; that is, it is not an invisibility due to distance or obscuration of any kind between us and them, but an actual invisibility due to the condition of the canal itself. With our present optical means, at certain seasons they cease to exist. For aught we can see, they simply are not there.

That distance is not responsible for the disappearance of the canals is shown by their relative conspicuousness at different times. It is not always when Mars is nearest to us that the canals are best seen. On the contrary, they show a sublime disregard for mere proximity. This is evidenced both by the changes in appearance of any one canal and by the changes in relative conspicuousness of different canals. Some instances of the metamorphosis will reveal this conclusively. For example, during the end of August and the beginning of September, at this last opposition, the canals about the Lake of the Sun were conspicuous, while the canals to the north of them were almost invisible. In November the relative intensities of the two sets had distinctly changed: the southern canals were much as before, but the northern ones had most perceptibly darkened.

Another instance of the same thing was shown in time case of the canals to the north of the Sinus Titanum when compared with those about the Solis Lacus. In August the former were but faintly visible; in November they had become evident; and yet, during this interval, little change in conspicuousness had taken place in the canals in the Solis Lacus region.

With like disregard of the effect due to distance, the canals to the east of the Ganges showed better at the November presentation of that region than they had at the October one, although the planet was actually farther off at the later date, in the proportion of 21 to 18.

A more striking instance of the irrelevancy of distance in the matter was observed in the same region by Schiaparelli in 1877. It is additionally interesting as practically dating his discovery of the canals. In early October of that year, on the evenings of the 2d and the 4th, he tells us, under excellent definition, and with the diameter of the planets disc 21 of arc, the continental region between the Pearl-Bearing Gulf and the Bay of the Dawn was quite uniformly, nakedly bright, and destitute of suspicion of markings of any sort. A like state of things was the case with the same region at its next presentation, on the 7th of November. Four months later, when the diameter of the disc had been reduced by distance to 5Tf~7, or, in other words, when the planet had receded to four times its previous distance from the earth, the canal called the Indus appeared, perfectly visible, in the region mentioned. At the next opposition, in 1881, similar effects occurred; the canals in this region remaining obstinately invisible while the planet was near the earth, and then coming out conspicuously when it had gone farther away. Distance, therefore, is not, with the canals, the great obliterator.

As to their veiling by Martian cloud or mist, there is no evidence of any such obscuration. The coast line of the dark areas appears as clear-cut when the canals are invisible as when they become conspicuous.

A canal, then, alters in visibility for some reason connected with itself. It grows into recognition from intrinsic cause. But during all its metamorphoses, I A presentation of any part of the planet is the occasion when that part of the disc is turned toward the observer. Man~+ causes combine to make the face presented each night vary, hut the chief one is that the earth rotates about in one thing, and in one thing only, it remains fixed, in position. Temporary in appearance, the canals are apparently permanent in place. Not only do they not change in position during one opposition; they seem not to do so from one opposition to another. The canals I have observed this year agree quite within the errors of observation with those figured on Schiaparelli's chart. In general they conform to their representations, and failure to do so is explicable not only by errors of observation, but by certain other facts. First, by seasonal variation in the canals themselves; the visibility or invisibility of a canal combined with the visibility or invisibility of a neighbor being capable of producing strange permutations in the region observed.

The Araxes is a case in point. On Schiaparelli's chart there is but one original Araxes and one great and only Phasis. But it turns out that these do not possess the land all to themselves. No less than five canals traversing the region, including the Phasis itself, were visible this year at Flagstaff, and I have no doubt there are plenty of others waiting to be discovered. These cross one another at all sorts of angles. Unconscious combination of them is quite competent to give a turn to the Araxes one way or the other, and make it curved or straight at pleasure.

Unchangeable, apparently, in position, the canals are otherwise among the most changeable features of the Martian disc. From being invisible, they emerge gradually, for some reason inherent in themselves, into conspicuousness. In short, phenomenally at least, they grow. The order of their coming carries with it a presumption of cause, for it synchronizes with the change in the Martian seasons. forty-one minutes faster than Mars, and consequently gains a little less than ten degrees on him daily. After about thirty-seven days, therefore, the two planets again present the same face to each other at the same hour. Their first appearance is a matter of the Martian time of year.

To start with, the visible development of the canal system follows the melting of the polar snows. Not until such melting has progressed pretty far do any of the canals, it would seem, become perceptible.

Secondly, when they do appear, it is, in the case of the southern hemisphere, the most southern ones that become visible first. Last June, when the canals were first seen, those about the Lake of the Sun and the Phoenix Lake were easier to make out than any of the others. Now, this region is the part of the reddish-ochre continent, as we may call it, that lies nearest the south pole. It extends into the blue-green regions as far south as 4O~ of south latitude. Nor do any so-called islands that is, smaller reddish-ochre areas stand between it and the pole. It lies first exposed, therefore, to any water descending toward the equator from the melting of the polar cap.

Having once become visible, these canals remained so, becoming more and more conspicuous as the season advanced. By August they had darkened very perceptibly. As yet those in other parts of the planet were scarcely more visible than they had been two months before. Gradually, however, others became evident, farther and farther north, till by October all the canals bordering the north coast of the dark regions were recognizable; after which the latter, in their turn, proceeded to darken, a state of things which continued up to the close of my observations toward the end of November.

The order in which the canals came out hinted that two factors were operative to the result, latitude and proximity to the dark regions. Other things equal, the most southern ones showed first; beginning with the Solis Lacus region, and continuing with those about the Sea of the Sirens and the Titan Gulf, and so northward down the disc. Other things were not, however, always equal in the way of topographical position. Notably was this the case with the areas to the west of the Syrtis Major, which developed canals earlier than their latitudes would warrant. Now, to the Syrtis Major descend from the pole the great straits spoken of before, which, although not in their entirety water, are probably lands fertilized by a thread of water running through them. They connect the polar sea with the Syrtis Major in a tolerably straight line.

The direction of the canal also affects its time of appearance, though to a less extent. Canals running north and south, such as the Gorgon, the Titan, the Brontes, and the like, became visible, as a rule, before those running east and west. Especially was this noticeable in the more northern portions of the disc. Time of appearance was evidently a question of latitude tempered by ease of communication.

After the canals had appeared, their relative intensities changed with time, and the change followed the same order in which the initial change from invisibility to visibility had taken place. A like metamorphosis happened to each in turn from south to north, in accordance with, and continuance of, the seasonal change that affected all the blue-green areas.

To account for these phenomena, the explanation that at once suggests itself is, that a direct transference of water takes place over the face of the planet, and that the canals are so many waterways. This explanation has the difficulty of involving enormously wide canals. There is another objection to it: the time taken would appear to be too long, for some months elapsed between the apparent departure of the water from the pole and its apparent advent in the equatorial regions; furthermore, each canal did not darken all at once, but gradually. We must therefore seek some explanation which accounts for this delay. Now, when we do so, we find that the explanation advanced above for the blue-green areas explains also the canals, namely, that what we see in both is, not water, but vegetation; for if the darkening be due to vegetation, time must elapse between the advent of the water and its perceptible effects, time sufficient for the flora to sprout. If, therefore, we suppose what we call a canal to be, not the canal proper, but the vegetation along its banks, the observed phenomena stand accounted for. This suggestion was first made some years ago by Professor W. H. Pickering.

That what we see is not the canal proper, but the line of land it irrigates, disposes incidentally of the difficulty of conceiving a canal several miles wide. On the other hand, a narrow, fertilized strip of country is what we should expect to find; for, as we have seen, the general physical condition of the planet leads us to the conception, not of canals constructed for waterways, like our Suez Canal, but of canals dug for irrigation purposes. We cannot, of course, be sure that such is their character, appearances being often highly deceitful; we can only say that, so far, the supposition best explains what we see. Further details of their development point to this same conclusion.

In emerging from invisibility into evidence, the canals first make themselves suspected, rather than seen, as broad, faint streaks smooching the disc. Such effect, however, seems to be an optical illusion, due to poor air and the difficulty inherent in detecting fine detail; for on improvement in the seeing I have observed these broad streaks contract to fine lines, not sensibly different in width from what they eventually become.

The parts of the canals which are nearest the dark areas show first, the line extending sometimes for a few hundred miles into the continent, sometimes for a thousand or more; then, in course of time, the canal becomes evident in its entirety. Complete visibility takes place soon after the canal has once begun to show, although it show but faint throughout.

This tendency to being seen in toto is more strikingly displayed after a canal has attained its development. It is then not commonly seen in part. Either it is not seen at all, owing to the seeing not being good enough, or it is visible throughout its length from one junction to another.

Apart from their extension, the growth of the canals consists chiefly in depth of tint. They darken rather than broaden, a fact which tends to corroborate their vegetal character; for that long tracts of country should be thus simultaneously flooded all over to a gradually deepening extent is highly unlikely, while a growth of vegetation would deepen in appearance in precisely the way that the darkening takes place.

As for color

, the lines would seem to be of the same tint as the blue-green areas. But, owing to their narrowness, this is only an inference. I have never chanced to see them of distinctive color. At this point it is probable that a certain obstacle to such wholesale construction of canals, however, will arise in the mind of the reader, namely, the thought of mountains; for mountains are by nature antagonistic to canals. Only the Czar of all the Russias if we are to credit the account of the building of the Moscow railway would be capable of running a canal regardless of topography. Nor will the doings at our own antipodes help us to conceive such construction; for though the Japanese irrigate hillsides, the water in the case conies from slopes higher yet, whereas on Mars it does not.

Indeed, for the lines to contain canals we must suppose either that mountains prove no obstacles to Martians, or else that there are practically no mountains on Mars. For the system seems sublimely superior to possible obstructions in the way; the lines running, apparently, not where they may, but where they choose. The Eumenides-Orcus, for example, pursues the even tenor of its unswerving course for nearly 3500 miles. Now, it might be possible so to select ones country that one canal should be able to do this; but that every canal should be straight, and many of them fairly comparable with the Eumenides Orcus in length, seems to be beyond the possibility of contrivance.

In this dilemma between mountains on the one hand and canals on the other, a certain class of observations most opportunely comes to our aid; for, from observations which have nothing to do with the lines, it turns out that the surface of the planet is, in truth, most surprisingly flat. How this is known will most easily be understood from a word or two upon the manner in which astronomers have learnt the heights of the mountains in the moon.

The heights of the lunar mountains are found from measuring the lengths of the shadows they cast. As the moon makes her circuit of the earth, a varying amount of her illuminated surface is presented to our view. From a slender sickle she grows to a full moon, and then diminishes again to a crescent. The illuminated portion is bounded by a semicircle on the outside, and by a semi ellipse on the inner. The semicircle is called her limb, the semi-ellipse her terminator. The former is the edge we see because we can see no further; the latter, the line upon her surface where the sun is just rising or setting. Now, as we know, the shadows cast at sunrise or sunset are very long, much longer than the objects that cast them are high. This is due to the obliquity at which the light strikes them; the same effect being produced by any sufficiently oblique light, such as an electric light at a distance. Imperceptible in themselves, the heights become perceptible by their shadows. A road illuminated by a distant arc light gives us a startling instance of this; the smooth surface taking on from its shadows the look of a ploughed field.

It is this indirect kind of magnification that enables astronomers to measure the lunar mountains, and even renders such vicariously visible to the naked eye. Every one has noticed how ragged and irregular the inner edge of the moon looks, while her outer edge seems perfectly smooth. In one place it will appear to project beyond the perfect ellipse, in another to recede from it. The first effect is due to mountain tops catching the suns rays before the plains about them; the other, to mountain tops further advanced into the lunar day, whose shadows still shroud the valleys at their feet. Yet the elevations and depressions thus rendered so noticeable vanish in profile on the limb.

Much as we see the moon with the naked eye do we see Mars with the telescope. Mars being outside of us with regard to the sun, we never see him less than half illumined, but we do see him with a disc that lacks of being round, about what the moon shows us when two days off from full. It is when he is in quadrature that is, a quarter way round the celestial circle from the sun that he shows thus, and we see him then with the telescope at closer range than we ever see the moon. When we so observe him, we notice at once that his terminator, or inner edge, presents a very different appearance from the lunar one. Instead of looking like a saw, it looks comparatively smooth, like a knife. From this we know that, relatively to his size, he has no elevations or depressions upon his surface comparable to the lunar peaks and craters.

His terminator, however, is not absolutely perfect. Irregularities are to be detected in it, although much less pronounced than those of the moon. His irregularities are of two kinds. The first, and by all odds the commonest phenomenon consists in showing himself on occasions surprisingly flat; not in this case an inferable flatness, but a perfectly apparent one. In other words, his terminator does not show as a semi-ellipse, but as an irregular polygon. It looks as if in places the rind had been pared off. The peel thus taken from him, so to speak, is from twenty to forty degrees wide, according to the particular part of his surface that shows upon the terminator at the time.

Now it is a significant fact that this paring of his disc appears usually where the dark regions are coming into view or passing out of sight, according as it is the sunrise or the sunset terminator that is presented to observation. And even in the few cases where it is not coincident with them, it is never far removed from their position. Two causes undoubtedly combine to produce the effect. One of them is irradiation. It is a well-known fact that bright bodies look larger than they are, probably because of the sympathetic vibration of the rods in the retina adjoining those directly affected. A familiar instance of the effect is the seemingly wizened look of the old moon seen in the new moons arms. The lusty young moon seems a sixth the broader of the two. The same thing would appear in the case of the Martian terminator; a bright area would seem to project beyond a dark one. This accounts for a part of the loss. The other part is doubtless due to an actual depression iii the Martian surface. Thus from the appearance of the terminator comes corroboration of the lower level at which we found reason (in the last paper) to suppose the dark markings upon the planet to lie.

That these long parings do not always coincide with the dark areas may help confirm, paradoxical as it sounds, their real depression; for it is only the relative, not the actual height that is projected on the terminator, and a more elevated area, if sloping at the proper angle, would be projected as a depression beside a lower one, in spite of being the higher surface of the two. It may also, however, not be due to this cause, but to the presence of an actually elevated district; verdure, such as a forest, standing on high land.

Such long, low depressions are characteristic of the Martian terminator, which is thus in kind quite unlike the lunar one. In addition to them there are elevations, some long and low, some short and sharp. Both are relatively rare. Of the former variety Professor W. H. Pickering discovered two striking specimens. Each looked to be, and probably was, a plateau, very level on top, and sloping more or less equally on both sides. Of the short and sharp variety Mr. Douglass has detected some noteworthy instances; but whether they mean high dust cloud or mountains is not yet predicable. Mr. Douglass has very systematically observed the Martian terrain at every longitude, and is now busy upon a contour map of the planet. His map may enable us to say something more definite as to whether the canals traverse low regions from preference or not. But certain it is that Mars is a flat world; devoid, as we may note incidentally, of summer resorts, since it possesses, apparently, neither seas nor hills. To canals w~ will now return.

The canals so far described all lie in the bright reddish-ochre portions of the disc, those parts which bear every appearance of being desert. But Mr. Douglass has made the discovery that they are not the only part of the planet thus privileged. He finds, in the very midst of the dark regions themselves, straight, dark streaks not unlike in look to the canals, and still more resembling them in the systematic manner in which they run. For they reproduce the same rectilinear arrangement that is so striking a characteristic of their bright-area fellows. He has succeeded, indeed, in thins triangulating all the more important dark areas. What is more, lie finds that these canals in the dark regions end at the very points at which the others begin, so that they make continuations of them.

This fact is another telltale circumstance as to the true character of the so-called seas; for that the seas should be traversed by permanent dark lines is incompatible with a fluid constitution. But the lines are even more suggestive from a positive than they are from a negative standpoint. That they make continuations of the lines in the bright regions shows that the two sets are causally connected, and affords strong presumption that this causal relation is the very one demanded by the theory of irrigation. For if the canals in the bright regions be strips of vegetation irrigated by a canal (too narrow to be itself visible at our distance), and there he a scarcity of water upon the surface of the planet, the necessary water would have to be conducted to the mouths of the canals across the more permanent areas of vegetation, thus causing bands of denser verdure athwart them, which we should see as dark lines upon the less dark background.

Before passing on to certain other phenomena connected with the, canals of like significance, we may note here an obiter dictum of the irrigation theory of some slight corroborative worth; for if a theory be correct, it will not only fit all the facts, but at times go out of its way to answer questions. Such the present one seems to do. If the seas be seas, and the canals canals, we stand confronted by the problem how to make fresh-water canals flow out of salt-water seas. General considerations warrant us in believing that the Martian seas, like our own, would contain salts in solution, while irrigation ditches, there as here, should flow fresh water to be most effective, and we seem committed to the erection of distilleries upon a gigantic scale. But if, on the contrary, the seas be not seas, but areas of vegetation, the difficulty vanishes at once; for if the planet be dependent upon the melting of its polar snows for its spring freshet, the water thus produced must necessarily be fresh, and the canals be directly provided with the water they want. The polar sea is a temporary body of water, formed anew each year, not a permanent ocean; consequently there is no chance for saline matter to collect in it. From it, therefore, fresh water flows, and, like our rivers, gathers nothing to speak of in the way of salt before it is drawn off into the canals.

We now come to some phenomena connected with the canals, of the utmost suggestiveness. I have said that the junctions held in a twofold way the key to the unlocking of the mystery of the canals; in the first place, in the fact that such junctions exist. The second and more important reason remains to be given, for it consists in what we find at those junctions. These phenomena will form the subject of the next paper.