MAP OF THE TRAVELLER’S ROUTE.
GENERAL OBSERVATION.
Before entering, in somewhat minute detail upon the examination of M. Caillié’s routes, I should make an observation to warn the reader against the apparent differences which he may observe between the journal of the travels and the map of the route. To render this journal fit for perusal, and not to interrupt perpetually the description of places and people, it has been found necessary to retrench the number of distances and bearings, to such an extent that any one who should attempt to form a plan of the route by the assistance of the narrative alone would fall into serious errors: respecting this circumstance however the reader has been already cautioned in a note in the first volume. The construction of the lines of route upon the map which I have designed is the result of all the indications given by the traveller, and which I have collected in the form of a table. It is also necessary to consult the following article, respecting the use which I have made of these materials, to render them accordant with anterior information and the positions already admitted as incontestable.
The entire journey of M. Caillié, from Kakondy to Tangier occupied five hundred and eight days: from which, in order to obtain the days of actual travel, three hundred and one days of stationary residence in eighteen different places, must be deducted, which will leave two hundred and seven days’ travel. I shall here subjoin a list of these places of residence, but it must not be inferred that the traveller had no convenient means of making observations in other stations, because it will be observed that when he arrived early at any place of consequence, the remainder of the day was spent there.
When it is proposed to offer a new map to the public it is an indispensable duty to submit to inspection the elements which form its basis, and not to ascribe to the work which results from them a greater value than appertains to the materials; these again have no other than what they derive, whether from the instruments and methods which the traveller has employed, or from the manner in which he has collected the observations and information which form the basis of his description. M. Caillié was not provided with any astronomical instruments; he had no watch, and measured the hour only by the height of the sun; but he possessed two compasses which were of great assistance to him. All his bearings were carefully noted by this instrument during the day, and by the stars at night. With regard to distances they were estimated according to several experiments made by himself at Sierra-Leone while preparing for his enterprise. He made a practice of walking over a certain space measured exactly in English miles and of observing the time he spent upon the road. It is thus that he calculated the number of miles in each of his marches from Kakondy to Djenné at three English miles an hour, or two geographical miles and six-tenths: as far as Timbo, however, that is to say during the first days of the journey, this number must be a little increased; this results from the situation of Timbo, as determined by Major Laing; anden passantwe may observe that this part of M. Caillié’s journey proves that the ancient position of Timbo, according to Watt and Winterbottom, is totally inadmissible. This rate of two geographical miles six-tenths an hour, or more exactly from two miles four-tenths to two miles six-tenths, can in general only be ascribed to isolated marches, and to caravans lightly laden. An hour’s march of a lightly laden caravan seldom exceeds two geographical miles three-tenths, and that of a numerous caravan heavily laden does not exceed a mile and a half, or a mile and three quarters[50].
On quitting Djenné, M. Caillié travelled by water. The current of the river was slow, the obstacles arising from the faulty construction of the vessel, the ignorance and unskilfulness of the pilot, and the islands and sand-banks, reduce an hour’s direct navigation to two English miles. Different causes producing, however, similar results determine the same value for an hour’s march in the deserts between Timbuctoo and Fez. This number of two miles taken as an average for a long journey[51]admits of almost all possible differences in the degree of expedition with which a caravan will move at different times: while it forms also the intermediate length between the progress of a caravan heavily and of one moderately laden, which was the case with the caravans which M. Caillié accompanied from Timbuctoo, as may be seen by his narrative.
This same deficiency of space with a variation in the bearings, has caused the frequent breaks which I have been obliged to make in the route on the itinerary map. The scale of a millionth part, adopted for this map, is perhaps too small to describe the details and all the angles of the route; I had, accordingly, at first constructed it upon a scale of double dimensions, but this development would have required too much space.
The line of route in the first part of the journey depends essentially upon Timbo. This place is well worthy of attention both in respect to its physical geography and to the European relations with Africa. The English have always considered it as an important station for commerce, but it is not less deserving of study as the culminating point of this part of the continent. It is fortunate that several travellers have made efforts to determine its geographical situation. Major Laing has definitively fixed it in latitude 10° 25’ north, and longitude 12° 54’ west from Paris; its actual elevation above the sea remains to be correctly ascertained. I could not establish M. Caillié’s route upon a better basis; and this I have done, by means of a direction which the traveller obtained from the village of Dité. There the inhabitants informed him that Timbo was at two days’ distance in a certain direction, towards which he immediately applied his compass and found it to bear S. E. ¼ S. Now, we may observe (and it is a remark common to the whole interior of Africa), that the inhabitants are scarcely ever mistaken with respect to the bearings of places; their accuracy in this respect is surprising; it is the effect of habit and of a particular tact. Secondly, I subjected this line of route to the position of the town of Labé, which is tolerably well known though not so certainly as Timbo. Our traveller, when near Teléouel learnt that Labé lay N. E. ¼ E. at two days’ distance. Another confirmation is drawn from the position of Cambaya, in respect to Timbo, which latter, according to the natives, bears W. S. W. at two days’ journey from Cambaya. Here, then, are five points well connected together, Teléouel, Labé, Dité, Timbo, and Cambaya, forming a whole, the main support of which rests upon Timbo[52]. From the nature of these data I have had no hesitation in subjecting the route from Kakondy to them; and the result has been that the hour’s march proves to be a trifle more than three English miles, which, for the commencement of the journey, is not extraordinary.
But, be this as it may, the construction of this route upon the itinerary map has undergone no reduction or modification, and I have confined myself to designing as accurately as the scale would permit, all the angles of this line, the rivers which the traveller crossed or had sight of, the different elevations of the soil with his estimate of them, and in fact all the variations of the ground which he has carefully noted.
At thirty-two miles from Kakondy is a fine river called Tankilita, a name which much resembles that of Tingalinta, marked in several maps at the same distance;[53]and which word is perhaps pronounced Tenkalita; it is the same, according to the inhabitants, with the Rio-Nuñez; this stream, which flows to the north, would, in this case, be only the origin or principal tributary of the Rio-Nuñez.
Other rivulets or rivers, as the Bangala, the Doulinca, and the Kakiriman, a large river in the neighbourhood of the high mountains of Antegua, on the contrary flow towards the south. The soil is chiefly composed of granite in this district and beyond it: the mountains are very high, steep and pointed; chains still increasing in elevation succeed each other; and there are among them mountains two thousand feet in height. All this country, bordering on the Fouta-Dhialon, presents an accumulation of obstacles, which are with difficulty surmounted; and yet men and even women, with heavy loads upon their heads, are continually traversing it. One of these mountains, the Touma, forms the limit between the Tenanka and the Fouta, and beyond it are the cataracts of the Cocoulo, a large and very rapid river. Thence we proceed to the Bâfing, the chief arm of the Senegal: it appears that its source is to the west of Timbo, and that its course is to the west and north, (and not to the east); this, at least, is the inference I draw from the new itinerary; and the north-easterly course of the Tankisso, which belonged to the bason of the Dhioliba, confirms this opinion.[54]
We afterwards descend into vast and very fertile plains, watered by the Tankisso, a large river, with steep banks, the Bandiegue and others less considerable, till we reach the principal river of this part of Africa, the Dhioliba.
On leaving the Soulimana the river takes a northerly direction, as I have said above, on the authority of Major Laing, but afterwards turns to the east. This bend is precisely given by a distance in the new itinerary. It may be observed that the river passes at one day’s journey to the south of Saraya. This bearing is in perfect agreement with the map of Laing’s journey to the Soulimana,[55]which will confirm if necessary the position of Saraya, and my plan of the route. It is known that the actual height of its source above the level of the sea, is, according to Major Laing, about sixteen hundred feet, or a hundred and thirty more than the source of the Rokelle.
It was at Couroussa, in the little country of Amana, that our traveller crossed it. Though so near to its source, it had already attained a breadth of nine hundred feet, and the moderate velocity of two miles and a half. At a short distance from this point, the Yendan, a large river, falls into it; farther on it receives the Milo, coming from the town of Kankan, and the Sarano, which waters the rich plains of the Wassoulo. Proceeding towards the east, and leaving the Dhioliba to the left, the traveller reached Timé.
The situation of Timé is determined by two data: the one, that the march, throughout its whole length from Kakondy, was the same; the caravan was equally numerous, and if the country is less mountainous, the travellers were more fatigued, which forms a compensation: the other is furnished by the meridian altitude of the sun, twice taken at Timé, by means of the length of the shadow. Imperfect as this observation probably was, we ought to have some respect for it, when it coincides with the data of the itinerary; I shall soon resume this subject.
To conclude the examination of this first part of the itinerary, I shall make some remarks on the situation of the country, which appears to be the spot where the waters of this portion of the African continent separate. The first great line of partition is that which divides the waters of the Senegambia from those of the Soudan. The mountains which form it have their nucleus at Timbo, where their distinction into separate ranges is totally lost. Thence one range diverges to the north, and another to the east, and it is remarkable that the divisions of the countries or states correspond with the physical regions. Thus in the country called Fouta-Dhialon, Timbo and its mountains are the sources of the Rio-Grande, the Gambia, the Falemé, the Senegal, &c. The Soulimana and its fountains are the source of the Dhioliba on the one side, and the Rokelle and the Mungo take their rise on the other. From the backs of the mountains of the Fouta-Dhialon spring the tributaries of the Dhioliba, as the Tankisso (which cannot be an arm of the Senegal, as M. Caillié was informed): the spot where it rises is not far from Timbo, and the true point of division between the basons of the north and the east. Thus we have a proximate acquaintance with the relief of the country, which presents a satisfactory whole, where all appears to bear a natural connection. To this result M. Caillié’s observations have materially contributed.
Besides the line of division of which I have just spoken, there is another which separates the waters of the Senegambia from those of the Timannie; it is the result of an attentive examination of M. Caillié’s route. He crossed in fact two considerable rivers: the one, the Kakiriman, which is from seventy to eighty paces in breadth; the other, the Cocoulo, which is forty-five; and both very rapid. He saw them flowing to his right, that is southward, towards the bason of the Timannie.[56]Further on, that is after Popoco, all the rivers which he crossed flow to the left, towards the north, and into the Senegambia. Thus, between the route of M. Mollien and that of M. Caillié, there must be a very elevated mountain crest, running from N. W. to S. E. and passing near the point of intersection of the two routes.
The names of the countries traversed by M. Caillié, in the first part of his travels, deserve a particular examination. According to the circumstantial description of the Baleya, it appears to me, that Major Laing has placed on his map a village of Beilia, where he ought to have marked and where in fact is, the country of Baleya. The Firia of the maps is the same with the Fryia or the Firya of M. Caillié; the Sangaran is nearly where it has been placed upon the recent maps, but on both sides of the river. With respect to the Couranco, if M. Caillié was rightly informed, it extends considerably towards the N. W., whereas Major Laing confines it to the interval between the rivers Rokelle and Camaranca, in the south. This country borders it should seem on the Baleya and the Soulimana. The true source of the Dhioliba is in the Kissi, to the south of Couranco, according to M. Caillié (or of the Soulimana, according to Major Laing). Thus the whole discrepancy between the two travellers is reduced to the lengthening of the country of Couranco; but, are the limits of these petty kingdoms well defined, and are the natives themselves fully agreed as to frontiers which violence is perpetually changing? On many maps the names of Sangala and Couronia are to be found not far from Couranco and Sangara. I suspect they are doing double duty here, as in so many other instances in geography, either through the difference of the orthography or the negligence of the transcriptions. Is not Couronia altered from Couronca (Couranco) and Sangala, by the frequent mutation of therintol, from Sangara? The Wasselon, or Wassoulo, should be placed much nearer to the ocean, the incontestable consequence of M. Caillié’s march; and already this example affords a presentiment of the fact as relates to the course of the Dhioliba and all the districts which it traverses. This river having also been supposed too much to the east by three or four degrees, the result has been the necessity of lengthening all the distances in order to cover the space between it and Timbo, and the constructors of maps have scattered over this space the names of countries and places, made all the positions too remote from each other, and transformed hamlets into towns and towns into kingdoms. They have not sufficiently remarked the custom common to almost all travellers of lengthening distances by an exaggerated estimate, and have neglected to make allowance for the frequent curves and inflexions of the line of route: a double cause for throwing into the interior many countries much nearer to the sea than they are believed to be.
This want of exactness in the compiling of itineraries, or in laying down those itineraries upon maps, is no novelty in geography; the Ancients have furnished many examples of it, and the maps designed after the positions of Ptolemy are nearly all tainted with this defect; I shall mention Arabia alone as an example. But I must here terminate this short digression, the subject of which would almost furnish matter for a book equally useful and instructive. I also pass over many names of countries with which M. Caillié has made us acquainted, and which will figure for the first time upon the map of Africa, and many more which are already known, and the existence of which is now confirmed by authentic testimony.
The largest portion of the space between Kakondy and Timé may be regarded as an entirely new acquisition for geography; the same remark applies to that which we are about to run over with our traveller. We must not regret that he abandoned the banks of the Dhioliba, and that this circumstance deprived us of information concerning the banks of that river from Couroussa to Djenné; we are indemnified for this loss by his discoveries respecting countries of which we have hitherto been utterly ignorant. Moreover, had he passed through Bamacou, Sego, and Sansanding, he would not have had time to sojourn in them so long as Mungo Park did: he might, perhaps, even have been recognised at Sego, and like Dochard have been detained. Leaving the river far to the west, he became acquainted with the tributary streams which traverse the triangular space comprised between Couroussa, Timé, and Djenné, and with all the positions of this vast tract. He also acquired information concerning the positions situated near the rivers, and of all the towns of any importance, by the attention which he paid to inquire their distance and direction at different points of his route.
I have already observed that, the route from Timé to Djenné having been performed under the same circumstances as the preceding, the estimate of three English miles an hour applies to this line of road; I have therefore admitted this with M. Caillié, and I have had the satisfaction to see that, in forming this line, without any alteration, upon the distances and bearings as he has given them, the latitude of Sego would remain very nearly the same as that which results from the observations of Mungo Park, made in this vicinity.[57]But this point remains for future examination, when I shall discuss the situation of Timbuctoo. Thus the delineation of this portion of the itinerary has not presented any great difficulties. Some topographical circumstances relating to this extent of country merit observation.
Beyond Timé the traveller proceeds for two days more towards the east; he then turns to the (magnetic) north, and continues afterwards in nearly the same direction, which ought to be borne in mind. There are high mountains of granite at Timé, and again at four days’ distance, after which the ground subsides and becomes level. The sand is succeeded by a fertile soil, furrowed by the frequent streams running westward towards the Dhioliba, amongst others the Bagoe, a navigable river, and the Couara-ba. Tangrera, near the commencement of this line, appears to be a large and very commercial town; this point is placed, according to the journal to the E. N. E. of Timé, (east by the compass), but not ten days to the east as appears by the journal.[58]At Tiara a part of the caravan directed their course upon Sansanding, and to the N. W., according to M. Caillié (the line traced on the map gives the true N. N. W.); agreeably to the information which he obtained at Badiarana. Caya is nine days to the north, and Sego nine days farther on; now Sego (as it is placed upon the map, as I have elsewhere said)[59]is situated due north by the compass, with respect to Badiarana; but the eighteen days’ journey, if they really exist, must be very short, occasioned probably by the stagnant waters, and other obstacles, which render a winding course necessary. The situation of Cayaye is determined by its bearing upon Couara, five days to the N. N. W., which places it, as it should be, half way upon the road between Badiarana and Sego. This spot, Couara, doubly merits our present attention. I have had occasion, in another work, to remark that this is a generic word, the sense of which is analogous to that of river. Here we see beside a village of this name a pretty considerable river calledCouara-ba, that is to sayriver-river; we have some examples of a like denomination inBa-ba, and other names of the same kind; an additional reason for not identifying rivers and currents on account of the similitude of their names, because it is above all things necessary to know whether these names are generic terms, or individual appellations.
In the second place, the Couaraba runs across a country through which, according to the map of Mungo Park, a river calledBanimmapasses, parallel with the Dhioliba; this would appear impossible since the Couaraba falls into the latter stream, if we may believe the report of the inhabitants.
This portion of the itinerary furnishes data for the site of a country and even of a town calledKong, placed at a considerable distance in all the maps. The bearing of a line directed thither from Douasso is between S. S. E. and E. ¼ S. E. by the compass, which is in fact due south; so much for its longitude: but a distance of forty-five days’ journey would carry it much too far to the south; doubtless the soil is very mountainous upon nearly the whole road, and the journeys very short. If we stop at the seventh parallel north, we shall make each day’s march but seven or eight geographical miles, including the windings of the route.
Still proceeding northward, the traveller crosses large open plains and rich countries, bounded on the left by swamps, lakes, or ponds, which indicate the neighbourhood of a large river. At length he reaches its banks at Galia, opposite to Djenné, at ten miles distance. Before we enter this capital, let us examine some important geographical positions. Nothing could be more obscure than the situation of Bouré; we were ignorant whether there existed a town of this name, and even what was the site of the country of Bouré: in the maps we shall find differences of several degrees. M. Caillié did not visit Bouré; but, in throwing together all the particulars which he collected, I find sufficient data to determine the situation of this town; for there is a town, and a very important one, on account of the vicinity of its rich gold mines, and the commerce of which it is the centre. Bouré stands upon the left bank of the Tankisso, a large tributary, as I have before said, of the Dhioliba, and at three quarters of a day’s journey from the latter in a direct line, or one day’s above the confluence. These particulars which agree well together, were furnished by various persons in various places. Moreover the position of Bouré is given by its distance from Kankan; and its bearing, namely, four or five days to the north, ¼ N. E., in descending Milo. Finally, as a fourth testimony, we learn that Bouré was five days’ march from Couroussa, descending the Dhioliba in a canoe, andafterwards ascending the Tankisso.
Bamakou is known to us from the second journey of Mungo Park; but its actual situation should be considerably further to the west: the information obtained by M. Caillié, and the construction of the map, place this town at about the tenth degree of longitude west of Paris, and in latitude eleven degrees forty-five minutes. The idea of forming an establishment there, is founded upon a real knowledge of the country; it was long ago suggested in memorials submitted to government, pointing out its advantages. The documents of M. Caillié confirm the possibility of the scheme at the present time. He thinks that from this point it is eight or ten days’ journey to the nearest point of the Senegal. Now, it appears, from the map, that, by ascending the Bâfing, as far as thirty leagues above the first cataract of Felou, we should be seventy leagues in a direct line from Bamakou, which agrees very well with what has gone before; this, however, is not the place for examining that question.
The course of the river, from Couroussa to Sego, as I have traced it, deviates from the hitherto received opinions;[60]and I am bound to justify so material a change, when submitting it to the judgment of geographers. In laying down the routes of M. Caillié, I was not prepared for such a result, or rather I was apprehensive of finding former conjectures relative to the much more westerly position of the river, and of the towns through which it flows destroyed by actual observations, as an illusion is dissipated by the clear light of truth. It has however proved quite otherwise, and this opinion was far from rash. In fact, beyond Couroussa the river runs due N. E., then eastward, and afterwards for a great distance to the north, instead of pursuing invariably an eastern direction as marked upon the maps. But on what is this latter direction supported? On what foundations does it rest? It has no other cause than the position arbitrarily assigned to Timbuctoo, much too far east, much too far from the mouth of the Senegal. On the more modern maps attempts have, it is true, been made to remove this position farther west; the necessity of bringing it nearer to the ocean has been felt,[61]but it has not been advanced far enough, and has still been retained in nearly the same parallel, instead of being brought at the same time farther north.
If the objection be admitted that M. Caillié has carried his route by water too far west, the result would then be that his route from Timbuctoo to Fez must have been carried too far east: now these two results are opposed to each other, since it is the same line of route, and the same bearing, which lead from Couroussa to Sego, from Sego to Timbuctoo, and from Timbuctoo to Fez.
One powerful consideration is that M. Beaufort has fixed and ascertained the position of the town of Elimané, eastward of Bakel, by a great number of observations; this is the present capital of Kaarta. From that place travellers are daily going to Sego; our unfortunate countryman was himself on the point of proceeding thither, and would have done so but for the destitution caused by the pillage of which he was a victim, and which compelled him to turn back. He was then told that Ségo was only at ten days’ distance; that the bearing of Elimané with regard to Sego was E. S. E.; and that these days’ journeys were journeys on foot. Let us work on these data, and see whether they confirm the course of the river drawn from the itinerary of M. Caillié.
Can these days’ journeys be computed at more than twenty-one geographical miles? This can scarcely be admitted; but should they even be extended to twenty-two miles and in a direct line, it would be on the whole two hundred and twenty miles, which in the above direction would not reach the ninth meridian west of Paris.
While M. Caillié was at Kiebala, he was told the distances and bearings of Sego, which differ very little from the position resulting from that of Elimané in the opposite direction; and if the river was here carried eastward, as it is found in all the maps calculated from Park’s travels, there would no longer be any agreement between these two different sources of information. If, on the contrary, these respective data are preserved, they naturally harmonize, and the northerly course of the stream is confirmed.
Besides, the latitude of Sego, although not yet observed, cannot greatly differ from that of Sami, which is near it: the observation there made by Mungo Park gives thirteen degrees seventeen minutes, and this is perhaps, a little too northerly. The approximative position of Sego, resulting from various data, and which I have adopted as a medium, would thus be, latitude thirteen degrees; longitude, nine degrees west.
Sego, according to the itinerary, lies due north by the needle of Badiarana, that is north seventeen degrees west. This direction exactly passes through the position which I have assigned to Sego, from the observation of Sami, by Mungo Park.
The bearing of Sego is likewise precisely indicated by the N. N. W. line of the compass, drawn from Saraclé.[62]
The course of the Dhioliba, from Sego to Jenné, west and east, and the position of Sego, are still more strongly confirmed by data with which the traveller was furnished at Bamba. Sego is three days’ journey N. W. of that place; and at noon on the fourth, the city of Sego is reached: these are long days’ journeys. The proportion of three to four, or rather of three to three and two thirds, will be actually found as I have traced it.
Lastly, it is elsewhere expressly said, that Sego is five days west of Jenné; the new map gives W. ¼ S. W. All these data, it is plain, perfectly coincide; and although they are not authenticated by a single good celestial observation, yet they agree too well together to allow one to venture upon deranging the whole itinerary of our traveller, and carrying the city of Sego and with it the whole course of the river, two or three degrees farther east, when besides there exists no observation whatever, either geographical or astronomical, for carrying the river and its towns farther into the interior of the continent.
The position of Djenné, the arms of the river by which it is surrounded, its situation in a large island apart from the Dhioliba, the stream which branches off in the neighbourhood of Sego, and rejoins the river at Isaca, four days further on, are so many circumstances all equally new, and which tend greatly to modify our previous notions. Looking at this complication of details, the obscurity and contradictions contained in all the accounts of the negroes on the subject of Djenné may be easily conceived; but, upon an attentive study and careful comparison of the different data, we arrive at a tolerably just idea of these localities: the reader will judge whether I have attained this object in the course of these observations and in the drawing here submitted to him. Unquestionably, there yet remain many circumstances to be learnt relating to this double branch of the Dhioliba, the true form and actual extent of the island of Djenné, and its double or perhaps triple communication with the branches of the river; for every step yet gained in our acquaintance with this mysterious river seems to throw us to a greater distance from the ultimate object: it is, therefore, to confess the truth, nothing more than a slight sketch that I have presumed to delineate; time will rectify our yet imperfect notions, and complete the observations of our indefatigable traveller, when the use that I have made of them may be judged of. M. Caillié presumes the island of Djenné to be eighteen miles in circumference; but the construction of the route and the general appearance of the map induce me to believe that it is of greater extent: I also suspect, that the junction of the island with the western branch of the river cannot be seen from Djenné, but only its junction with the eastern branch; this would be a natural consequence of the westerly situation of Sego, with regard to Djenné. For the rest, I refer to the observations in the following article, that this discussion may not be prolonged here.
I shall not stay to inquire how it happened that Mungo Park knew nothing of a second branch of the river, when he proceeded from Sego to Silla. I shall likewise abstain from entering into further developments respecting the course of the Dhioliba beyond Djenné. The narrative furnishes ample details, not less instructive than new, respecting the course of the river, its depth, its immense width in some parts, to the extent of half or even a whole mile. One of the most interesting points of this route by water is the great lake Debo, or Dhiebou, which M. Caillié met with half-way between Djenné and Timbuctoo. I cannot help thinking that this lake is the same which figures on the maps under the name of Dibbie, but with a widely different position, form, and extent.
The enthusiasm of our traveller at the sight of this sea of fresh water must appear excusable to every one, and his eagerness to take possession, in a manner, of the spot, on behalf of his country, by giving names to three islets situated in it, cannot but be generally approved. Twenty three years earlier, Mungo Park had accomplished this same navigation, perhaps, had also given names to these little mediterranean isles. Who, either in France or in England would have blamed him for this national feeling, had the account of his discovery reached Europe?.
The lake has two divisions, one eastern, the limits of which may be seen, the other western, extending beyond the reach of sight. M. Caillié is ignorant whence this mass of water proceeds; is it from a tributary, or is it, on the contrary an efflux from the inundations of the Dhioliba? This geographical question, on many accounts an important problem, remains then in obscurity; further on I shall offer a rather probable opinion[63].
To the east of this lake are sands and sterile hills; to the west, it is lost in vast marshes; the navigation of the latter division must be extremely difficult. Passing from Djenné to Timbuctoo, the traveller navigates the eastern division of the lake, keeping close to the right bank.
Below the lake (on the map) a winding of the river will be observed, which perhaps appears forced and unnatural; it results from the lines of route as noted down and the difficulty of substituting any thing else has alone determined me to retain it, although doubtful of its accuracy: an error, however, in this tracing will have little influence on the general result.
In approaching Cabra, the port of Timbuctoo, M. Caillié perceived to the right a large arm of the river, running E. N. E. and he continued to navigate the other, holding a north-west course. He was informed that the latter rejoined the first at some distance; but this important fact has not been verified by any European eye; the traveller could not have ascertained it, without repairing himself, at considerable risk, to the point of junction. As for the inhabitants, perfectly indifferent as to obtaining or transmitting intelligence of this sort, they spoke very vaguely to him on the subject. The problem, therefore, of the ulterior direction, and of the final termination of these branches of the river, is yet to be resolved; I defer this discussion to a subsequent article[64], as well as the particular question of the situation of Timbuctoo[65]; and shall conclude this with repeating, that the itinerary from Timé to Galia, and from Galia to Timbuctoo, has been framed on the calculation of two miles an hour.
The progress of M. Caillié through the desert has also been estimated at two miles an hour, at least as far as Tafilet, and that for reasons explained at the beginning. The principal interest presented by this part of the route consists in the exact information it contains of the wells and stations to be met with amidst this ocean of sand. Science is indebted to M. Caillié for numerous and correct notions of these vast solitudes, which travellers cannot confront without consternation. Thus, we knew the spot called el-Araouan only by the wells found there, as a place at which the caravans usually fill their water-skins; but our traveller informs us that it is an important town: seeing it thus surrounded on all sides by the deserts, we are the less surprized at the situation of Timbuctoo in the midst of the sands.
The wells of Telig are remarkable for the vicinity of granite mountains, and for the neighbourhood of Toudeyni, which, in all the maps, is carried far to the west of the line between Timbuctoo and Tafilet. M. Caillié’s description leaves no room for supposing that he is treating of some other place of the same name, since this city is well known to be a great mart for salt. I pass over in silence the immense banks of moving sand, and the rare accidents of the soil, which scarcely vary throughout this long tract of road, the dreary uniformity of which is interrupted only by the wells. Beyond the wells of Mayara appear the last branches of the chain of Mount Atlas. The granite shews itself at first in fragments and hillocks, afterwards in high hills and steep mountains. Twelve days’ journey beyond el-Harib, we enter the district of Tafilet. Here the report of M. Caillié is widely different from the received opinions: 1st. He heard nothing of anytown of Tafilet, it is merely, he says, the name of a country. It is nevertheless possible that a town may have heretofore existed there, and have disappeared like so many other towns of central Africa. I observe in the travels of Ebn-Hassan from Fez to Tafilet, quoted in theRecherches sur l’Afrique septentrionale[66], &c. that the territory only of Tafilet is mentioned and not the town; which would support the account of M. Caillié. 2ndly. This country is much nearer to the meridian of Fez than it is marked on all the maps[67]. 3rdly. It is farther north[68], Ghourland, M. Caillié informs us, is the principal place in this territory; near it is a place called Afilé, and another of the name of Boheim[69], besides Ressant, the residence of a governor under the Emperor of Morocco. M. Caillié mentions another position unknown upon the maps; a large town called Rauguerute or Rogrut, S. S. E. from Morocco.
The narrative presents but few details concerning the crossing of Mount Atlas; it is not a matter of surprize that, after so many fatigues and perils, our adventurer should be impatient to reach the end of so long a journey. However, I observe in this part the course of a little river or brook, called the Guigo, flowing from Soforo (or perhaps from beyond) towards M-Dayara, and probably as far as Tafilet. Upon carefully examining the traveller’s route, we perceive that he first saw the Guigo, or had some knowledge of it, between Rahaba, and L-Eyarac, afterwards at Tamaroc, and Kars; that he crossed it at L-Eksebi; and that the river fertilizes the environs of N-Zeland, L-Guim, Guigo, and Soforo. The water-mills which our traveller saw at Soforo again announce the presence of the same river. Thus the part of its course with which we are acquainted will be at least fifty leagues in length and probably much more. I do not know whether it passes to el-Fez, where however several different rivulets are seen.
Neither was the journey across the Sahara altogether unprofitable with regard to our knowledge of the oases and of the stations which it contains. The distances of these places from the several points of his route were reported to our traveller; and these distances are very different from those previously admitted: but are there two accordant testimonies concerning these positions? Each caravan and each tribe makes a different estimate of the intervals; and if, to the actual mistakes which the Moors and natives may commit, the errors in which they are interested be added, what a source of uncertainty will these present! Therefore, though I have fixed upon the general map the points of Tychyt, Oualet, Ouadan or Hoden, Ouad-Noun, &c., according to the indications of M. Caillié, combined with other documents, I consider them as very uncertain. Oualet, says our traveller is at ten days’ distance from el-Araouan, to the west, fifteen north from Sego, and eighteen from Ouadan: this point falls near nineteen degrees of latitude and nine degrees forty minutes longitude west from Paris: the three lines meeting at nearly the same point. But if it agrees with the three data given above, it is at variance with the situation of Oualet, as hitherto admitted, either from Mungo Park, or the testimony of the natives. It is perhaps another place named Oualet, not that intended by Mungo Park. Can it be the Gualata of Leo Africanus? but his description suits this no better.[70]In this case Gualata would be much farther to the south than is generally supposed. It is however certain that geographers cannot confound the Walet of Mungo Park with Gualata.
Although M. Caillié has not mentioned Agably which is believed to be the chief place of the oasis of the Touats, I have been obliged to insert it in the map, because the traveller gives a position called Touat much nearer to the route from Timbuctoo to the Tafilet. It is not surprising that several places should be so called, because the wandering tribe bearing this name occupy a considerable portion of the great desert, from the south of Morocco to beyond Agably. Besides, the situation of this place, which could not have been brought within the limits of the map, if we were to adopt the mere reports of the Arabs, to which geographers have had recourse, falls there, on the contrary, according to the astronomical observation of Major Laing, who fixes it much farther to the west. This observation is published, for the first time; I owe it to my learned friend Captain Sabine, himself united in the most affectionate intimacy with the unfortunate traveller, who transmitted it to him after sustaining a first attack on leaving this oasis. Nor was it at Agably that Major Laing made the observation, but at Ain-Salah, belonging to the same canton and at two days’ journey from it; this place is cited in all the itineraries between Tripoli and the western Soudan. According to the Major its situation is twenty nine minutes west of Paris, and twenty-seven degrees eleven minutes thirty seconds north latitude. Notwithstanding the difference between this and the situation allotted to this town upon all the maps, I have thought it right to enter it in the general map of the travels.
The other places of the Sahara, such as Akka, Tatta, el-Kabla,[71]for want of new data, have been fixed from a comparison of those upon which the map of M. Walckenaer, or those of Messrs. Lapie, Brué, Berghaus, &c., were formed. With respect to the names of tribes and colonies, I have endeavoured to place them with all possible exactness, and have been obliged to differ from several recent maps in other respects very valuable.
To the right of his route, and in front of el-A’raouan, M. Caillié heard of an important situation, that of Sala; but the accounts were too vague to be compared with the descriptions of the Arabs.
I shall finish the examination of this third part of the travels by the general list of the wells and stations of the desert on the line from Timbuctoo to Fez; which line appears to be the most advantageous and the shortest, because it is the most frequented. It seemed to me that the collection of all these names into a single table might be of some utility. It will be remarked that Toudeyni does not appear in this list: may not this place be without water fit for drinking? and has Telig succeeded it as a more convenient station for caravans? I submit this doubt to the reader.