Chapter I · (continued 2)

I will add one consideration which will not have escaped those geographers who have studied the calculation of probabilities. In a series of observations made under the same circumstances, and especially by the same observers, the greater the number, the more probable is it that their amount will approach to the total quantity required. When there is no reason that the errors committed should be on one side rather than another, they mutually destroy each other, and the more so, the more the observations are multiplied. There is even a rule which teaches us how much the sum found differs from the truth; its discovery belongs to the learned geometrician, who is at present the organ of the French Institute for mathematical science. Knowing the error which may have occurred in an observation, it must be multiplied by the square root of the total number of observations. Thus, instead of growing with this number the total possible error decreases proportionably. For example, for four observations it would be represented by two, and for a hundred observations, by ten only. The proportion of total errors is therefore as ¹⁰⁄₂ whereas the proportion of the number of observations is as ¹⁰⁰⁄₄: thus the error is but the fifth part of what it would be proportionally in four observations.[83] Hence it follows that the more observations are multiplied the more any imperfection in the processes by which they have been made will be corrected.

Are we not authorised to apply this principle to the length of M. Caillié’s stages, since the number of the lines of route is not less than six hundred and thirty three? I may add that the same remark applies to the angular deviation. Indeed, considering the total distance between the meridian of Timé and that of Fez, or the difference in longitude (which is equal to one degree forty four minutes) as being the real sum of all the angles of the route to the east or west of the first, it will follow that the forming of the lines of bearings noted by the traveller (or the calculation of the angles, which is more exact) gives a total variation, so much nearer to the above difference of longitude as the number of bearings is greater; and this number is also six hundred and thirty three. As it has been seen above, I have found the deviation equal to less than six degrees, and it has been easy to divide it over the whole line.

OBSERVATIONS ON THE SITUATION OF TIMBUCTOO, AND PARTICULARLY ON ITS LONGITUDE.

From the determination of the lines which join Timé, Djenné, Timbuctoo, and Fez, and from the latitude of Timbuctoo thus fixed according to the itinerary at seventeen degrees fifty or fifty one minutes,[84] necessarily results a more westerly longitude of this city, than has been hitherto supposed, and even than that which I had formerly admitted, and which brings its position much nearer to the ocean than any geographer has placed it.

Such is the importance of the situation of Timbuctoo that I deem it necessary to insist again on the discussion of the elements; in order, not to establish it with certainty (I am far from believing that I have accomplished this) but to furnish the reader with new means of approaching the truth, as nearly as the state of our knowledge will permit. It is true that a single astronomical observation made by a skilful observer, furnished with good instruments, and worthy of confidence, might overthrow these inductions; but it is equally true that till we are possessed of such an observation, no geographer can fix this main position on a map, without deducing all the reasons which determine him, especially if he dissents from the opinions of his predecessors.

Timbuctoo is placed, on the general map of the travels, in the latitude resulting from the observation of the shadow and the length of the route from Timé to Timbuctoo. A great part of this route, it may be observed, is in a northerly direction, which was far from being conjectured from the course of the river in this quarter. If Mungo Park could have acquainted us with the particulars of his journey beyond Sansanding, we should not have been so long uncertain of this direction, which has uniformly been carried easterly, and that because the situation of Timbuctoo was imagined (as it is still) to be very central in the continent. Park’s map (travels in 1805) places it under the meridian of Paris; Clapperton at 0° 5° west. Rennell at 2° 3°: but the data of which we have been for ten years in possession oblige us to bring it nearer to the ocean; M. Walckenaer has done so, in adopting a longitude from nine to twelve minutes more westerly. I have always proposed to carry this position much farther west, and even placed it four years ago four degrees west.[85] M. Brué has since adopted 3° 34’: perhaps it should have been advanced as far as the sixth degree. For the reasons I have elsewhere given, the route from Timbuctoo to Fez cannot be fixed at more than two English miles an hour: to this rate I have reduced the computation of M. Caillié, who at first estimated the whole route equally at three miles. Now the line of route thus constructed and resting upon Arbate, a well known position, carries Timbuctoo very near the eighteenth degree north; that is to say, from 17° 50’ to 17° 55’; had not this been shortened a little, the city would have been carried to 19°, or 20°, a position far too northerly, and totally inadmissible. Moreover, the line from Timbuctoo to Arbate would be ten degrees too far west; it would also be increasing the declination to 27° instead of 17°, which seems to be its amount for the mean meridian.

But, if the bearing of the line from Timbuctoo to Fez resulting from the itinerary, and that of the line from Timé to Timbuctoo, are preserved, these two intersect each other about the parallel of 17° 50’. This latitude agrees so well with that resulting from the height of the shadow, 17° 51’ that we derive from these data a very satisfactory confirmation. Now the line of the parallel 17° 51’ and the two lines of direction meet, all three, on nearly the same meridian, the sixth west from Paris.

No positive data are yet known which can better fix the longitude: not only are the itineraries of the Moors too vague, but they support themselves upon the Mediterranean, while M. Caillié’s routes, though they are continued to the Mediterranean, support themselves upon the ocean; which is much nearer than the former, and also upon the known positions of Timbo and Fez.

Suppose, however, it were determined, notwithstanding all these reasons to withdraw Timbuctoo farther to the east by a degree or more, it would then become necessary greatly to increase the journeys from Kakondy to Timé. In making them three English miles and three fifths an hour, the true measure has probably been somewhat exceeded, but the situation of Timbo compelled this; while, to place Timbuctoo on the third degree of west longitude, would be to suppose a march of more than four miles and a half an hour.

Another circumstance here comes to our assistance: it is that by carrying it farther towards the east, we must suppose the traveller greatly and constantly deceived in a western direction on his journey from Timé to Timbuctoo, and, on the contrary, in going from Timbuctoo to Fez, he must have made a similar and not less considerable error in all his bearings towards the east. How should we account for this singularity?

From all these considerations, could I risk changing the result of M. Caillié’s itinerary and altering the construction of his route? could I, in short, without sufficient motive, give the preference to any one map over another? Some judgment may be formed upon this question from the diversity of situations assigned to Timbuctoo by different authorities:

In 1720, the longitude according to Delile 0° 0’ Wt.of Paris

1749, d’Anville 2 15

1790, Major Rennell 2 30

1796, Park’s first map 1 0

1805, Park’s second map 0 0

1821, M. Walckenaer 2 42

1821, M. Lapie 2 44

1822, Clapperton 0 50

1828, M. Brué 3 34

It is on the English maps, even those of most recent date, that this town has always been placed the most to the east. The advantage of a more eastern position relatively to the proximity of a great river flowing into the Gulf of Guinea would be evident: but this proximity is nothing less than certain. By a coincidence worthy of notice the English maps have also brought Timbuctoo farther to the south than any others except Delile, as will appear from this table of different latitudes:

North latitude according to

Delile 15° 0’ Park’s second map 16° 27’

Clapperton 15 0 M. Walckenaer 17 38

M. Lapie 17 38

M. Brué (after Clapperton) 15 0 D’Anville 19 15

Park’s first map 15 14 Major Rennel 19 38

I return to the latitude of Timbuctoo. Ain-Salah, in the oasis of Touat, is usually placed about 1° east of Paris and in 24° 30’ north latitude; but Major Laing’s observation, as communicated to me by Captain Sabine, is very different, namely: 0° 29’ west of Paris, and 27° 11’ 30". M. Walckenaer has calculated the distance from this oasis to Timbuctoo at six hundred and seventy five geographical miles: it is evident that this measure, if admitted, would advance Timbuctoo towards the north to between the 17th and 18th degree; now, it happens that this distance of six hundred and seventy five miles is found exactly between the two points as placed upon the present map. This last coincidence, which I remarked after my labour was concluded, contributed still farther to dispel my doubts. I have estimated the day’s journeys at 18’ four tenths and not with M. Walckenaer at 15’; but it was the medium journeys, not those of the great caravan, which formed the object of his researches. It appears then that the approximative situation assigned to Timbuctoo, 6° west and 17° 50’ north, satisfies the different data, and the best itineraries, that it agrees with the positions of Timbo and of Fez, and that it is not contradicted by Mungo Park’s observations on the latitude, taken at Sami and Yamina; finally, it is confirmed by the particulars brought from Elimané by Captain Beaufort.

Fifty four days’ journeys will be found on my map from Timbuctoo to Tatta; an itinerary cited by M. Walckenaer (p. 297) marks fifty. The fifty four days from Fez to Timbuctoo according to Mr. Jackson[86] also agree: it is the same with the sixty-four days from Timbuctoo to Morzouk. Fewer than twenty eight journeys are reckoned from Houssa to Timbuctoo, according to the itinerary of Mohammed Ebn-Foul; this distance is too short on our map, and on all the others, even those in Clapperton’s travels. There are, besides, other reasons for believing that there may be two towns or countries called Houssa[87].

I do not compare with the map the distance reckoned between Timbuctoo and the town of Tafilet, because M. Caillié heard no mention of a town so called: he affirms that none such exists, in which case it is not possible to make use of this distance.

I shall add in conclusion of this discussion, that nothing can authorise us to depart from the observations of latitude taken by Mungo Park, at Yamina and Sami, namely, 13° 15’ and 13° 17’, and to remove these points much further to the south, as M. Brué has done upon his map, otherwise so rich in details and nomenclature. The determination which I propose for the latitude of Timbuctoo agrees better with these observations, the only ones we possess in this direction as far as Timbuctoo.

From the whole discussion, it results, first, that the different lines of route forming the itinerary map have been subjected to divers conditions resulting from the journal of the traveller, from anterior observations and from good geographical data; secondly, that the hour’s journeys, estimated at three English miles, mean measure, between Kakondy and Galia or Djenné, are a little increased between Kakondy and Timé, in consequence of the position of Timbo; thirdly, that, from Djenné to Timbuctoo and thence to Fez, the average of two miles an hour has been a little modified by the real difference in latitude of the two extremities. I have now only to speak of the bearings and measure of the lines of march.

BEARINGS OF M. CAILLIÉ’S ROUTES.

The reduction of the true north of our traveller’s lines of route was one of the most important points; and to arrive at their actual bearing, I was assisted by a single new datum only; namely, the observation of the angle of the meridian shadow with the magnetic north. The travels furnish two of these observations: the one, of the 30th of October 1827, gave the N. N. E.; the other of the 1st of November, the N. ¼ N. E., that is to say, the compass had on these days a declination of 22° 30’ and 11° 15’ to the east. I was then obliged to have recourse to other researches to determine between these two measures, and to avail myself of several observations taken by Mungo Park and other travellers.

The following are the names of the places and observations; I omit their geographical situations:

At Badoo, Mungo Park[88] 14° 6’ E. declin.

On the Bâfing, idem 16 30

On the Dhioliba, idem 16 36

At Yamina, idem 17 11

At Sansanding, idem 17 40

At Bathurst, Captain Owen[89] 17 56

At St. Louis, Maritime Annals[90] 17 32

At St. Mary, M. le Prédour[91] 17 25

At Cape Roxo, idem 17 20

In Mungo Park’s journal of his second expedition, as printed in London, is a tracing of the Gambia,[92] from which a smaller declination would result: it appears evident to me, that some error has crept in here, either in the copying or in the engraving, when the coincidence of the preceding observations from five different travellers is considered, and particularly that of Park’s own observations.[93] To these might be added the declination observed by Major Laing in his first journey, and that which has been observed at Sierra-Leone and other places; but this inquiry would be superfluous. It will be remarked, that the mean term of 16° 55’ declination east, is also the mean between the two observations of M. Caillié; but, as the observation of Badoo seems hither small compared with all the others, I have thought myself entitled to adopt 17° as the mean declination and to subject the whole route to it.

I shall here make another remark respecting the situation of places marked upon the two maps, upon information communicated by others, and not direct observations; it is that the natives are extremely clever at indicating the direction of places at a greater or less distance: they are seldom mistaken in this indication; and they point out with the finger with great accuracy, the direction which must be taken to go in a straight line to a given spot. This observation has been made before. When a certain direction was thus pointed out to M. Caillié, he remarked some particular object on the line, and applied his compass to it at the first favourable opportunity. These bearings have been extremely useful to me for the points situated beyond the route; without this assistance I should have found it impossible to trace, even tolerably, the course of the Dhioliba above Djenné.

COMPUTATION OF THE DAYS’ JOURNEYS.

The geography of countries which have not been explored by observers furnished with instruments is usually reduced to the calculations of days’ journeys. What can be more vague or doubtful than such documents? The most learned discussion (as remarked above), can only elicit feeble scintillations from them. How are contradictory accounts to be reconciled? How are common days’ journeys to be distinguished from double days, or even longer still? It is evident, that itineraries must be examined and compiled from the number of the hours journeys, and not by the days, and there would then be a less degree of uncertainty. Should European travellers themselves compute their lines of march by the days’ journeys? And how happens it that it is not an established rule, in exploring distant and unknown countries, to keep an exact account of the hours and every fraction of time? The journal of M. Caillié, although he has not rigorously complied with this condition, at least presents an uninterrupted continuity of marches measured by time, generally by hours, sometimes by half-hours and even quarters. But for this persevering (and amidst so many fatigues truly laudable) attention, positive geography would have gained very little by these long and toilsome peregrinations.

In a memoir inserted in the eighty-first volume of the Philosophical Transactions, Major Rennell has fixed the day’s journey of a caravan heavily laden at sixteen geographical miles (or minutes of a degree) and one sixth; and that of a light caravan, at seventeen miles and one third. It appears to me, that the second of these results is too small, and especially that an intermediate term should have been established, between the light and the heavy caravans, for there is a vast difference between the two extremes. The former certainly advance more than twenty geographical miles a day, especially, as they are able to continue their journey longer, that is to say, for a greater number of hours each day. Then, again, it would be convenient to fix an intermediate valuation between sixteen and twenty miles, that is to say, the day’s journey of the medium caravans, if they may be so called. I find Major Rennell’s appreciation of the first denomination of days’ journeys confirmed by the experience of the engineers attached to the French expedition, in crossing the deserts near Egypt. We estimated the hour’s march at nineteen hundred toises; eight hours would amount to fifteen miles two hundred toises, or very nearly sixteen geographical miles; the hour’s march would thus be established at two miles, or thirty hours to the degree. In consequence of several calculations much too long to be reported here, I consider the day’s journey of the medium caravan to be eighteen geographical miles and four tenths, the hour being between two miles and two miles and one tenth. I shall not here take the light caravans into account; these travel longer, and the progress of each hour is at the same time greater; but I think it may in many cases be estimated at twenty-two miles, or ten hours of two miles and two tenths. Persons travelling in small parties, without reckoning men mounted on horses or dromedaries, move still quicker.

According to Captain Lyon, the days’ journeys of the caravans are less than twenty English miles, and above seventeen; that is to say, more than the short day, and less than the medium. It is not by the pace either of a pedestrian or of a lightly loaded camel that the progress of a caravan must be measured, but, on the contrary, by that of the man or camel bearing the heaviest burden; for the latter must be waited for by the former, and is perpetually retarding the march; otherwise the usual pace would be much greater than that just fixed. Besides, this reduction of the average value is independent of that which must be allowed for the deviations and turnings which are often unknown: another source of hesitation and of error to geographers.

The earlier journeys of M. Caillié on leaving Kakondy afford an example of the real amount of the day’s journey: the first day, he advanced twenty-three English miles, the following sixteen miles and a half; the third eighteen miles; the average rate nineteen miles and one sixth.[94] The hour’s journey was estimated by him at three miles only, but the construction of the itinerary proves that this computation was more than one tenth too small. The result is twenty one English miles,[95] which is very near eighteen geographical miles and four tenths; all, by my estimation, medium days, (or days of a medium caravan). Indeed, the small caravan to which M. Caillié then belonged may be balanced against these considerations—first, that though small in number it included a woman; secondly, that all were on foot; thirdly, that all were loaded. In traversing the desert, the day’s journeys were of twelve hours and sometimes more;[96] but exhaustion, thirst, and fatigue, prevented both the travellers and camels from keeping up the same pace as on their first departure, and a mile and a half an hour is a large allowance for the march.

I ought here to report the opinion of M. Walckenaer, who has discussed with much sagacity this important geographical question:[97] he fixes at fifteen minutes the amount of a day’s journey of a caravan heavily loaded; this amount falls short by only one sixth of that which I have adopted, from our own experience in Egypt, and which further confirms the opinion of Major Rennell.

But M. Walckenaer does not estimate the other two denominations of days’ journeys, those of the medium and of the light caravans. I shall only remark that the result of his arguments seems to be, that the real and effective progress is considerably more than 15’, and about equal to 18’ or 19’: the average of a day’s journey reduced to measures taken in a straight line upon the map. But will the diversity of elbows and inflexions, resulting from that of the obstacles which produce them, allow us to work a uniform reduction, and to apply a uniform factor to every distance travelled over? Let us, for instance admit that part of the route amounts to sixty miles, and that the straight line between these extremities is forty-eight miles; can the same rule be adopted with regard to the whole route, which may be three hundred miles or five times as much? would the total direct line amount to two hundred and forty miles? The longer the route, the greater should be the variation in the rate of the reduction. The following table contains a summary of these observations upon the amount of the hours’ and days’ marches of caravans, and the different computations which have been made of them.

AUTHORITIES. MARCH OF THE CARAVAN

HEAVILY laden. MODERATELY laden. LIGHTLY laden.

per hour. per day of 8 hours. per hour. per day of 8 hours. per hour. per day of 10 hours.

Geograph. miles. Geograph. miles. Geograph. miles.

Egyptian expedition. 2 „ „ „ „ „

Major Rennel. „ 16⅙ „ „ „ 17⅛

Captain Lyon. „ „ „ above 17 „ „

M. Walckenaer. „ 15[98] „ „ „ above 22

M. Caillié’s travels. 1⁹⁄₂₀ „ 2³⁄₁₀ „ „ „

Computation from miscellaneous data. 1¾ 15 to 16 2³⁄₂₀ 18⁴⁄₁₀ 2⁴⁄₁₀ to 2⁶⁄₁₀ 24[99]

N.B. To reduce these numbers into English miles, they must be worked by the proportion of 60:69.[100]

I had at first intended to construct separately each of M. Caillié’s routes, taking into consideration in each particular case the difficulties of the ground, the composition of the caravans, and the physical condition of the travellers; but, as it was impossible to perform this work correctly, it would also have been useless. Besides, the reflections which I have already made tend to shew that, by following a uniform plan of reduction from one end of the route to the other, one must arrive at a result nearly approaching to the truth.

All that I have hitherto advanced relates solely to that part of the general map, which represents the space travelled over by M. Caillié; I have now to speak of the remainder, namely, the extremity of the map towards the north, and that towards the S. W. The former portion, or the country of Morocco, has been chiefly extracted from the fine map of M. Brué,[101] but reduced to much fewer details; we know that he sought his data in the best Spanish and English maps, and in the works of Badia, Jackson, Walckenaer, Ritter, &c.; authorities which furnish excellent materials, but which, the former especially, must not always be employed without the assistance of criticism.

The second part, to the south of the Rio-Nuñez, has been drawn from the Senegambie of M. Dufour, who has skilfully combined the materials of Major Laing with those of preceding English travellers: it will be seen by comparing our two maps, that I have been obliged to differ from him with respect to the situations and names of several countries. I have also been necessitated, as well for the sake of perspicuity on a map upon a very small scale, as on account of uncertainty, to retrench many details. I shall conclude with the table of geographical positions, forming the foundation of the general map, independently of the coast of Africa, which is pretty accurately known.[102]

N. Latitude. Longitude W. from Paris. AUTHORITIES.

Timé 9° 3’ 0’’ 9° 2’ 0’’ Construction of the map.

Sources of the Dhioliba at Mount Loma 9 25 0 12 5 0 Major Laing.

Timbo 10 25 0 12 54 0 Major Laing.

Kakondy 11 10 0 16 17 0 Construction of the map. M. Brué’s map, long. 16° 24’

Djenné 13 0 0 9 0 0 Park.

Yamina 13 15 0 "" "" "" Idem.

Sami 13 17 0 "" "" "" Idem.

Bakel 14 13 30 14 41 40 Dussault.

Elimané 15 2 50 12 36 0 De Beaufort.

Timbuctoo 17 50 0 6 0 0 Construction of the map.

Ain-Salah 27 12 30 0 29 0 Major Laing.

Morocco 31 37 30 9 56 0 Badia.

Fez 34 6 0 7 18 0 Idem.

§ III

OF THE NOMENCLATURE.

I shall not here offer to the reader a general list of the names of towns and villages, or of the different nations, states and districts visited by M. Caillié during his travels; this duplicate labour is rendered quite unnecessary by the minute list of the Itinerary[103] contained in this volume, and by the journal itself, both of which I recommend it to the reader to consult upon this subject. My object is merely to make some remarks upon the method of retracing these names and upon a few other relative points. The nomenclature is so much the more important in the compilation of African maps, as inattentive travellers frequently confuse generic appellations with proper names and vice versa; or they spell the same name in several different ways, or in writing various names they lose sight of the trifling differences by which they are distinguished. Hence the faulty multiplication upon the maps of places which do not exist, and on the contrary the suppression of many which do. The difficulty is great, particularly with respect to countries recently explored and the language of which is unknown. I have chiefly confined myself to obtaining the names from the mouth of the traveller as his memory furnished them, and comparing them in his presence with his journal.

Among other generic terms which have been considered as proper names, I shall particularize two, on account of the confusion which they have introduced, and which involves in obscurity some important geographical questions, namely the situation of a considerable chain of mountains to the south of the 8th degree of latitude, and the still unknown outlet of the great central river. Kong, is the name given, particularly since Mungo Park, to a great chain of transverse mountains which he reports to have seen at a distance to his right, when travelling from the Gambia to the Dhioliba. Now, M. Caillié learnt from the natives that Kong is a generic word, and that in the Mandingo language it signifies a mountain; the mountain or chain of mountains in question is accordingly far from being the only one of its name. I remark also, that the English traveller in his Mandingo vocabulary explains the word Kong by head; whence perhaps the signification of Kong; and himself translates Konko by hill.[104]

When the later English travellers had gained intelligence of a river called Couara, to the west of Saccatou, and of the river which is near Funda, it was remembered that this name is also borne by the upper Dhioliba, and these three rivers have been at once identified; but it appears that Couara is a general term signifying a river. The inhabitants who in three different places have been asked the name of the river, not understanding the question, have answered by the word river. Already had this confusion taken place a hundred times with the words ba, bahr, and nil, which also mean river, running water, great water. There is on the road from Timé to Djenné a village named Couara, and near it a river of moderate breadth, called Coraba (or according to my idea Couara-ba,[105].) It is easy to perceive from the face of the country, that it is a tributary of the Dhioliba, which was also reported to M. Caillié; here then is another river of the same name, or rather another general denomination which confirms the import of the word Couara, already observed by travellers.

I have noticed that the Arabic, kha غ, is used every where, even in the countries where the Arabic language and Islamism do not prevail: the traveller had expressed it by a blank; I have written it kh according to the general custom. The slightly lisping sound ﻎ is used in many central districts, as is a liquid sound, common also in Senegambia, and which may be written ghi or dhi. The name of the town of Jenné has been written Djenné, because the Arabs of the present day write جنع or جنا; but M. Caillié remarked that the natives pronounce dhi I am therefore inclined to prefer Dhienné.[106]

M. Caillié knows nothing of the Island of Jinbala marked upon Park’s map, but he was not wholly a stranger to the name; he mentions a tribe of Jinbalas to the north of Timbuctoo.

I entertain doubts, which time alone can remove; respecting the names of many places, and I have therefore thought it better to preserve those names without alteration. I have only omitted, upon consultation with the traveller, letters which appeared useless to the pronunciation, or which might create difficulties. The names of such places as Brahihima no doubt require rectification; the name of Abrahima, or Ibrahima, &c., are met with in some travels.

I have suppressed the letter q and almost always the k, except before e or i, confining myself to the use of c. The double ss merely expresses the sound of an initial S. The w and oo of the English, which M. Caillié acquired the habit of using at Sierra-Leone, I have expressed upon the maps by ou; the words in the text have been generally subjected to the same rules.

On the route to the Tafilet such names as Tamaroc and M-dayara, &c., occur on account of the use of the initial letters T and M which announce the vicinity of the Berbers. The wells of Trasas or Trazah should perhaps be pronounced T-ghazah or T-ghazzah, which will correspond with the Tegasa or Tegazzah of Leo Africanus[107]. In several words beginning with L as L-Eksebi, L-Guedea, L-Eyarac, L-Guim, &c., the Arabic article al appears to me to be joined to the name by contraction, as in the vulgar pronunciation throughout Northern Africa. The words beginning with the letter n should perhaps be pronounced ain, which signifies a source or fountain. I hazard this conjecture from the presence of wells at such places, and from the example of Ain-Salah, in the oasis of Touat, which is frequently written Ensalah, or Nsalah, in a single word. It is difficult for Europeans to pronounce the guttural ع, and they frequently leave a blank for it, as well as for the ق and the خ. I presume, therefore that the names Nzeland, Nyela, &c., stand for Ain-Zeland, Ain-Yela; but this supposition may perhaps be refuted by the orthography of the name of Hanalak or (Hen-Alak) هنالﻜ a place situated upon the route from the country of Galam to Morocco.

I shall take this opportunity of giving the Arabian orthography of the names of several places belonging to the space between St. Louis, Timbuctoo, and Morocco; I am obliged for them, to the Baron Roger, formerly governor of the Senegal, and who has lately enhanced, by important publications, the title he had long since acquired to general esteem, by the improvements of all kinds which he introduced into that country.

LIST OF THE NAMES OF SEVERAL PLACES IN NORTHERN AFRICA.

Ouad-Noun وادنون

Sous سوس

Marakch, or Maroc مراكش

El-Rebat, or Arbate الرباط

Fes, or Fez فاس

Kalam, or Galam كّلم

Takant, or Tajacant[108] تاكّانت

El-Zaouat الزاوات

Oualata, or Walet وَلات

Ouadan, or Hoden وأدَان

Hanalak هنالﻜ

El-A’rousyoun, or el-Ross العروسيون

Tychyt تيشيت

Oualata, or Oualyata[109] ولياتَ

A’raouan عَروَان

Bouzbeyah بوزبيه

EI-Mabrouk المبروك

Tymboctou[110] تيمبُكتُ

Touat توَات

This list contains the names of several Moorish tribes and colonies of different nations, the orthography of which is uncertain. It would require very laborious research, and assistance which will be long unsupplied, to write these names more correctly: I have contented myself with discarding from the names occurring in the journal all useless letters, and those which do not accord with the sound as remembered by our traveller.

Science would be much benefited by the careful examination of the names engraved on the maps of Northern Africa, without excepting those recently published, and the indication of the duplicates and false names, which are real blemishes in works otherwise of great value; but this labour, even as regards the countries which belong to my subject, would occupy too much time. The nomenclature is of more importance than is usually ascribed to it by the compilers of maps; I consider it one of the fundamental principles of geography: there are maps which would merit the appellation of classical if they answered in this respect to their precision and merit in all others. Two things are wanting to enable us accurately to fix this nomenclature: the names written by a native, and the pronunciation collected with great attention[111].

§ IV.

OF SOME RESULTS OF THE TRAVELS OF M. CAILLIÉ.

Of all the results, for which we are indebted to the new travels, that which most excites curiosity is unquestionably the knowledge of the city of Timbuctoo; but, perhaps that which is most important to geography is the course of the great central river. Although M. Caillié was unable to give information of it beyond Timbuctoo, he has conferred a real service on science, since he describes its banks with circumstantial detail from Djenné to that city, and gives us an idea of its course above Djenné. By crossing it at Couroussa, and after advancing more than two hundred English miles farther east, then travelling as far as Djenné on the right bank, he has furnished reason for concluding that no river parallel with the Dhioliba exists there, as traced upon the maps. The river, on the contrary, receives in this quarter pretty numerous tributaries, which, from their importance, seem to denote a distant source.[112] We perceive from the description that both banks of the river, a little beyond Bamakou, are very open, and their declivity gentle, which explains the existence of vast accumulations of water, of which the Lake Debo or Dhiebou is the most considerable. Whatever opinion may have been previously formed of the existence of a lake in this part of Africa, we were utterly ignorant of its situation, its true name, its form, and extent. The islets, on which our countryman has thought himself entitled to bestow names, will perhaps one day prove remarkable points. Their occupation would suffice to intercept the navigation upwards and downwards, and by thus rendering their possessor master of the river to influence the commerce of interior Africa; a commerce, comprehending on one hand, the gold extracted from the rich mines of Bouré, and the productions of Kankan, Ouassoulo, and Fouta-Dhialon, the merchandise of the Mandingoes, and all the articles of Djenné; on the other, the productions which Timbuctoo receives from the northern coast, and every thing transported thither by the Moors from the north and north east.

The observation of the river near Djenné affords a still more remarkable modification of received opinions; it must now be acknowledged that Djenné stands on an island, and that this island is double; which, let me remark, by the way, may serve to explain many contradictory accounts both of travellers and natives. The new narrative itself is, I confess, not free from difficulties, I have consequently only traced with dotted lines the branches of the river round Djenné. The following results may, at any rate, be deduced from M. Caillié’s routes by land and water, along the eastern part, and from the intelligence which he obtained respecting the western portion; first, that a large branch diverges near Sego, about forty leagues, (in a direct line), above Djenné; and that it rejoins the main stream at Isaca, twenty-seven leagues below, (forming the first or largest island,. including Djenné); secondly, that near Galia, (or Cou Galia,) is another communication, having itself two branches, and forming a second or smaller island, of which Djenné occupies one extremity; another channel also joins this island with the eastern branch.

This description appears at first sight rather complicated, but its perplexity will be removed if the tracing dotted on the map be admitted. The eastern branch could not have been crossed by Mungo Park, and he knew nothing of it; the same with Dochard. And, as for the second journey of Park, its narrative breaks off abruptly: it is probable that the branch which he followed in the first instance, on which Silla and six other villages beyond Sansanding are situated, is the same branch, which, according to M. Caillié, runs from Sego to Isaca; but Djenné must be carried to the right bank, whereas in the map of Park’s travels it is placed on the left. This circumstance was already learned from Baron Roger[113]; but the new discoveries perfectly explain this contradiction, which is merely apparent; for if Djenné is on the right of the western branch, it is at the same time on the left of the eastern.

This discussion shews that the immense volume of the waters of the Dhioliba is still greater than has been supposed. Park, indeed, who saw only one branch, stood in silent admiration of this majestic stream; but that which M. Caillié crossed before Djenné and followed as far as Isaca is not inferior to the other in importance[114].

The junction of the island with the river, visible from the tops of the houses in Djenné, was pointed out to M. Caillié. Is it with the western arm, or with a branch of the eastern? I am disposed to believe that it is with the latter, the other being at too great a distance to be within sight. I can say nothing positive of the island of Jimbala, and which the maps represent between Lake Dibbi and Cabra, the port of Timbuctoo[115]. There is, however, no reason for doubting its existence, since M. Caillié could not examine all the windings of Lake Debo or Dhiébou, from which an arm may diverge to the right, and rejoin the branch which it runs off to the east near Cabra.

From the travels of M. Caillié we further learn, on the subject of commercial communications, that the navigation of the Dhioliba is almost every where practicable; it is navigable as early as Couroussa, and no doubt yet nearer to its source; there is no reason to suppose any serious obstacle at Bamakou, although there appear to be at this place three principal rapids, but not cataracts. Park navigated it here; the current, by his account (but on the 22d of August, the season when the river is full,) ran at the rate of about five knots an hour. Its bed was a mile in width, and at the point of the rapids twice as much[116].

The observations which I have made above upon the position of Timbuctoo[117] apply equally to the course of the river. Ever since the year 1720, the tracing of this course on the maps has been progressively moving from east to west, drawing nearer and nearer to the Senegambia and to the west coast of Africa. It becomes, also, more and more probable that the mountainous space, which separates these two basons, will be found shorter and more practicable for the passage both of men and animals. Who knows but some large tributary to the Dhioliba may run in the close vicinity of a similar tributary to the Bâfing, or even to the Senegal below Galam, such as the Red River, or the Baoulima, or the Kokoro, so that the communication between these two rivers might be established without much difficulty? Who knows, in short, whether the progress of civilization may not one day effect the construction of a canal between such tributaries, suitable for commercial purposes; and whether we may not then find a navigable communication opened between the mouths of the Gambia and Senegal and Sego, Djenné, Timbuctoo, Houssa, and all the large towns by which the Dhioliba flows?

Even though Europeans should attempt only the land passage between the two rivers, this would be an enterprise founded on a more accurate knowledge of the situation and distance of places, and, if not an immediate consequence, would, at least, be an indirect result of M. Caillié’s travels on the two banks of the great river. It is unnecessary for me to add, that if, for want of positive documents, I should have erred in tracing the itinerary, the fault will be all my own, and will detract nothing from the merit of our traveller, or from the gratitude due to him from all friends and patrons of discoveries.

To continue our examination of the geographical results of M. Caillié’s travels, I ought not to omit the attention he has paid to make himself acquainted with the situation and depth of the wells; a circumstance from which useful inferences may be drawn relative to the course and distance of the waters: he has not neglected remarks upon the climate, the periodical rains,[118] and the state of the atmosphere. The learned will, no doubt, deeply regret that he was unprovided with instruments for observing and measuring meteorological phenomena: but it is not on a man’s first travels that these lights can be obtained; and, besides, none of our readers have forgotten the perils which attended our countryman in the execution of his enterprize.

The situations of several known places experience extraordinary changes in consequence of M. Caillié’s peregrinations, without mentioning the towns washed by the Dhioliba. Toudeyni, which was supposed to be 3½ degrees west of the meridian of Timbuctoo, proves by M. Caillié’s route to be very near the wells of Telig, only 40’ west of that meridian. Is this another place of the same name? I doubt it: its importance, proved by the description given of it by our traveller, repels the supposition. A’raouan[119] is inscribed in the maps as a mere station, with a well of brackish water; but M. Caillié found this a considerable place, a commercial entrepôt, in a word, an important town, notwithstanding the partial decay of its prosperity.

M. Caillié makes us acquainted in the north with a place called Oualet, at a great distance from the Walet of Mungo Park, and another of the name of Sala, which confirms the testimony of Arabic writers; in the south the towns of Teuté, Cagny, or Canny, and Koung, fifteen days and more south of Timé, that is to say, near the seventh degree. This account rectifies our notions of the kingdoms or states east of Fouta-Dhialon, and distinctly points out the mountainous tracts, the sandy plains, and the fertile territories enriched by numerous rivers.

We were before uncertain about the district of Bouré; the new notes enable us to give it on the maps very nearly its true site.

The large towns in which he resided, such as Timbuctoo, Fez,[120] Djenné, Kankan, are very minutely described, and were I not unwilling to lengthen this paper, it would be easy to shew how much he has added in this respect to our information, and also how many false and exaggerated ideas he has corrected: this is a merit for which we ought to give him double credit; since he is, perhaps, of all travellers the one who has dissipated the greatest number of illusions. Sound minds will feel the more grateful to him in proportion to the currency formerly gained by these exaggerations. The lapse of time may, indeed, have effected actual changes in the importance and population of towns; we must not, for instance, be surprized at the difference between the description of Leo and that given at the present day.[121] Who was there but believed, only a short time since, that Timbuctoo was equal in extent to one of our large cities, and contained within its walls a population of a hundred thousand inhabitants, or even from a hundred and fifty to two hundred thousand? The most moderate computation, says M. Walckenaer, gives it a hundred thousand inhabitants.[122] The exaggerations of the Arabs have constantly held the Europeans under the same error, and in the itinerary of Mohamed-Ebn-Aly Ebn-Foul especially it is said: “This is the largest city which God has created.”[123]

With regard to the account of the mountains of Fouta-Dhialon, and the means afforded by the descriptions of M. Caillié for forming an idea of the configuration of the country, or the relations which exist between the situations of the different basons, I must refer to what I have said in § II. (articles 2 and 3); I shall also refer to the same for the positions of various countries, pompously called kingdoms by travellers or writers.

Although he did not receive either from government, or from any learned societies, those questions or instructions which might have guided him in his course, M. Caillié has observed much; if he has not treated his subject very profoundly, if indeed he has but glanced at it, he has at least opened the road to his successors.

It is thus that during his travels he has lost no opportunity of speaking of the manners and customs, of the costume and food, of the people; of their religious practices and superstitions; of their commerce and navigation; of their industry, agriculture, and habitations; of the population of the countries; of the character, the physiognomy, and the language of the inhabitants; of their warlike or peaceful habits; in a word, of the whole state of society amongst these still half barbarous people. The picture of a flourishing agriculture, a peaceful and industrious population in the countries of Kankan, Ouassoulo, Baléya, &c., will be read with peculiar interest. Could it be expected that he should penetrate deeply into these subjects, or even that he could direct his attention to them with benefit?

It would be superfluous here to repeat all the new results for which we are indebted to him, and which make ample amends for the frequently minute simplicity of his accounts; they well depict the difficulties of the journey and the precautions which it requires. Future travellers will not complain of the multiplicity of his details, monotonous, it is true, but convenient in pointing out to them the measures necessary to avoid miscarrying in their enterprize. On this ground M. Caillié will have contributed usefully to the progress of discovery.

The physical conformation of the various races, and the colour of their complexion, are most frequently noticed by our traveller. This is one means of clearing up some questions still enveloped in great obscurity, for example, the origin of the Fellatas, as they are called by the latest English travellers. What relation do they bear to the great nation of Foulahs? We learn from M. Caillié that the Touariks are established much farther towards the south than has hitherto been supposed; their camps are seated upon the Dhioliba, considerably above Timbuctoo. He has also taught us that they bear a second name, that of Sorgous; above all, he has given us some very valuable information respecting the tyranny exercised by this wandering and predatory tribe against the peaceful natives; the portrait he has drawn of them bears visible marks of truth.

With regard to idioms, it is vexatious that M. Caillié, although he visited so many tribes, has been able to collect only two vocabularies. That of Timbuctoo consists of only a hundred and twenty words. We must regret that it is not more extensive. I have already said that the words in Denham’s relation agree with him, but not those of Adams and of Bowdich.

The particulars respecting commerce have been collected with care. M. Caillié has scarcely ever omitted to point out the native or foreign commodities which he saw at every market, their price, and the kind of coin. He confirms the fact that European merchandise reaches central Africa; goods of English manufacture were to be seen at Djenné as well as at Saccatou. Upon the commerce of Bouré in gold the traveller gives us some information which appears to be equally new and certain, and fit, probably, to direct the calculations of speculators, or the efforts of the European governments. We have yet but a vague knowledge of the actual degree of the riches of the mines of Bouré, and the quantity of gold now circulating in commerce; but there can be no doubt that it abounds in this part of Africa. We know for certain that this rich country is a hundred and twenty or a hundred and forty leagues in a straight line from the establishments of the Gambia and the Senegal. If we could, at some future day, open a direct communication, we should avoid the journey from Bouré to Ségo, from Ségo to Djenné, thence to Timbuctoo, and afterwards to Morocco, across the great desert; not only should we thus shorten the road by at least four hundred leagues in a direct line[124], but we should escape the cupidity of the Moors and the Jews, who appropriate the largest part of the profits, and also the ferocity of the predatory Arabs of the desert. At a trifling sacrifice, and in a short time this result might be obtained, if but a small part of those efforts were consecrated to it, which are absolutely wasted upon less useful objects. But, whatever the consequence may be, geography and the genius of discovery will have had the merit of pointing out a source of wealth to ancient Europe, overcharged with debts and population, and ready to sink under this double burden, if some new outlets are not speedily opened to her industry.

If it were possible to doubt the veracity of the traveller; if it could be supposed that all which has been hitherto said has still left any uncertainty in the minds of those who, in the first instance, manifested some incredulity, the results which I have just recapitulated would remain equally doubtful: it is their importance which induces me to neglect nothing here which may dispel doubts, if they still exist. I shall first cite, as a sure testimony of this veracity, an Arabic manuscript which Lander received two years ago from the hands of his master Captain Clapperton, and which M. Salamé translated in London, at the very time that M. Caillié was completing his enterprise: it is a description of part of the Soudan. One may read a portion of the African text, with the new map in one’s hand, for the countries common to this description and to the itinerary, that is to say, as far as Timbuctoo. I will even add that it would have been completely unintelligible to me, without the assistance of this map[125]. The alteration in the names of places does not prevent their being recognized; for many of the differences, it appears evident to me, arise only from carelessness in copying. I have made an extract from this curious document, which was accompanied by a map of the Kouara, traced by the hand of the secretary or amanuensis of the Sultan Bello. I place, according to geographical order, the different parts of the description, extracted as I have said above.

“The route from Sakkatou to Masera crosses first a tributary to the Kouara, and afterwards that river itself.

“The Touara flows from south to north, thence to the east, and afterwards returns to the south.

“It is formed of two arms, the one, called Balio, the black river, coming from the Fouta-Djalo; the other, called Raniou, the white river, coming from Ségo[126].

“At their junction it takes the name of Couarra, according to the secretary.

“Upon the first is a large island containing the town of Djeri.

“Many streams fall into the river on the left side.

“Beyond is the lake Djebou.

“Timbuctoo is at the farthest turn of the river: Kabara is half a day’s journey from that city.

“Masera is beyond the branch which comes from Ségo.

“The Touariks occupy the country adjoining Timbuctoo to the north, &c.

I pass over in silence the other places unconnected with the route of M. Caillié, or which he has not mentioned.

This account appeared at first very obscure; but by reading, with me, Massina instead of Masera, and Djené (or Djenné, instead of Djeri), every thing is explained.

It is easy to confound a noun without a point ﻥ with a ﺭ re. A similar error, I think, has been committed in the name of Ba-niou which has been read Raniou taking a be without a point ﺑ for a ﺭ re.

The word Dombari, a mountain, represented upon the map of the Fellata writer, ought, upon the principle already laid down, to be read in two words Oun-bari.

M. Caillié having informed us that the situation of Djenné is in an island, it will be immediately recognised in that of Djéri, actually situated at seven days’ journey from Ségo. In like manner, the arm, which separates in the environs of Ségo, says M. Caillié, and rejoins the western branch at Isaca, explains the Balio and the Banio, which meet, (according to the Fellata), below Djéri (or Djenné). Our traveller, not having inquired the names of these branches of the river, could not be acquainted with them. Masera is here, like Massina, to the west of Djenné: I remarked that, in travelling from Timé to Djenné, M. Caillié neither saw nor heard of the mountain or the town of Ounbari, nor of the road leading to Saccatou.

The lake Debo or Debou is here placed as it was seen by M. Caillié, between Timbuctoo and the confluence of the two branches (at Isaca); it is called in the Arabic description Djebou. The name of this great lake is doubtless written جبو, and I suppose that in this word, the ﺟ is pronounced dhi, as at the Senegal. On this subject, I remark that, according to M. Caillié, the name of the town of Djenné is pronounced, in that country, in a peculiar manner, expressed here by Dhienné.

The same is the case with the name of Fouta-Dhialon which the English translator, after the Fellata, writes Fouta-Djalo[127]. In studying the Arabic nomenclature of the places bordering on the Senegal, this observation has appeared to me to be more and more confirmed; and it must by analogy be applicable to the neighbouring countries. The Marabouts had only the letter djim ﺝ at their disposal for expressing this liquid sound, which is of a peculiar nature, intermediate between the English th and the Arabic dj.

The marigots observed by M. Caillié on the left bank of the stream, correspond with the four rivers or canals belonging to Masera or Masina.

Kabera appears here to be placed beyond Timbuctoo (coming from Djenné): the place in question is perhaps distinct from Cabra; or Timbuctoo may be a little farther west of Cabra than has been supposed. M. Caillié in going to this city went due north by the compass, which places it only 17° west of Cabra.

With regard to Baniou, White River, it should be remembered that M. Caillié also translates in the same manner the word Bagoé, the name of a considerable tributary of the river running from Teuté, very far south, but falling in on the right bank, whereas the Baniou is on the left: there is therefore a difference both in the name and situation. Goé (and koué, according to Mungo Park) certainly signify white in Mandingo. But the words niou and lio, I cannot find in any vocabulary of Northern Africa, with the meaning of white and black. In Wolof, moul signifies black.

The Banimma of the maps cannot flow parallel with the great river, as I have explained above[128].

It would not be uninteresting to compare with this draught the pretended map of Bello himself, which Clapperton has given in his first travels[129]. Five or six sites only are connected with my subject; Djenné is here, as in the other, placed between two branches of the river, (a fresh point of conformity with our traveller) and the Massina is separated from it by the western branch. There again the r is substituted for the n; (and I presume for the same reason) for it is there spelt Jesni or Jenri, and Mashira[130].

For want of room, Bello has placed Sego and Masina, much too near together, as well as Fouta and Djenné. The city of Timbuctoo (written Tonbaktou) is not less misplaced by the august geographer; but he has marked between it and Mashira (Massina) a large tributary or off-branch: this is probably one of the four rivers which his secretary has indicated in the same space, and one of the marigots, the outlet of which was seen by M. Caillié[131]. Thus the only two native maps (if they may be so called) which we possess, and the recent descriptions of two Africans, all confirm the discoveries of M. Caillié, who was entirely unacquainted with them.

A corroborative testimony is that of Park himself. Similar names will be found in the two narratives for certain indigenous productions and for several instruments employed in the arts. The Mandingo words and names are either the same or analogous in the two narratives.

In a preceding article, I think I have sufficiently shewn the agreement of M. Caillié’s observations with those of other distinguished travellers, Mollien, Watt, and Winterbottom, and Major Laing. The particulars of Major Laing’s death collected by our traveller at Timbuctoo, and afterwards on the very theatre of the catastrophe, so far from contradicting those obtained either by the English Consul at Tripoli, or by the governor of the Senegal, confirm all the important circumstances of both[132].

We did not learn from the first travels of Clapperton the name of the prince reigning at Timbuctoo; it was even imagined that the supreme authority resided at that period in a female; and when M. Caillié informed us that the supreme chief of Timbuctoo was named Osman, we could not reconcile this report with those which had preceded it: but we have now letters from Major Laing himself, received since the return of our countryman, one of which dated from Timbuctoo, the 21st of September, 1826, reveals the name of the prince then reigning, and this name is also Osman[133].

In the same letter, Major Laing states the circumference of the city at four miles; this account confirms the small population assigned to it by M. Caillié. He adds that he has collected documents in abundance relating to Timbuctoo. If we may rely on the account of a Moor from Saint Louis, the books of the Major are at the disposal of a certain Saleh, son of the Iman of Timbuctoo; and according to another they are in the hands of the Touariks. Time will in all probability produce some of the documents, the only consolation for so deplorable a loss.

The same accuracy appears in M. Caillié’s observations respecting the journeys, if we admit as a fair specimen those which concern countries whose geography is known; thus the distance from Fez to Arbate by Mequinaz agrees with the itinerary of M. Caillié, at the rate of three miles an hour.

So many motives of confidence and interest sufficiently recommend the accuracy of the travels, and consequently authenticate the results I have above deduced; yet the history of the enterprize itself will, if possible, increase this effect. The inclination of Réné Caillié for travels of discovery early announced his vocation. From the moment of his second landing at Senegal, he employed himself in acquiring a familiarity with the language of the Moors.[134] He talked of nothing but penetrating into the interior of Africa, the object of all his thoughts; his resources became exhausted, yet he refused every other occupation, every other mission. This fixed impression was regarded as a mania; nothing could shake his purpose, not even the insults which his Moorish costume drew upon him from the negroes; he was content to be considered by them as an idiot, and almost an object of derision. The want of sufficient support having obliged him to take another course, he set out for Sierra-Leone; there he remained the period necessary for collecting some resources, and soon quitted it for Rio-Nuñez: thence he announced to a friend at St. Louis (in April 1827) his departure for the interior. He was supposed to be lost, and nearly forgotten like so many other victims, when, at the expiration of eighteen months, he suddenly appeared at the further extremity of Africa, triumphant over every obstacle; like an expert swimmer, who, having plunged into the bosom of a broad stream, after a long interval unexpectedly appears on the opposite bank, while his friends are already lamenting his loss as certain.

Scarcely had he quitted the banks of the Rio-Nuñez and entered upon this new career of difficulty and danger, when he at once displayed a consummate prudence, and far above his years: as skilful in appreciating obstacles as he was firm in encountering them, his embarrassments increased at every step; but his sagacity constantly suggested the means of extricating himself from every new perplexity. Had he not possessed this just estimate of his difficulties and resources, united with a fortitude not to be shaken, he could not have concluded, perhaps not have commenced, his enterprize. How ingenious was the fiction he invented to obtain the confidence of these numerous tribes! This thread, slender as it was, served every where for his guidance and protection. He rightly supposed that the fame of the French expedition to Egypt had spread over Northern Africa: it was natural that a child carried off from its parents at the age of three years, and transported into the heart of France, should be but ill versed in his mother-tongue; equally so that good Musulmans should congratulate him on his return to his country by the most direct line, although he was supposed to be destitute of resources. Now this line must inevitably conduct him beyond the great river: afterwards manifesting a desire of repairing to Alexandria by sea, it was necessary to return to the river and embark on it, and thus his arrival at Timbuctoo was accomplished with certainty; having attained this object, he must naturally seek the readiest and surest means of reaching some place occupied by Europeans, and the caravan of Tafilet offered an opportunity that he could not hesitate to seize.

I shall not advert to all his misfortunes at Timé, and during the journey, nor to his perils at Tangier, when already within sight of his haven; but with what intelligence and courage must he have been endowed to resist and triumph over so many enemies. It was necessary to avoid the most trifling error; a single one would have infallibly proved his destruction.

Few are acquainted with the history of the unfortunate Antonio Piloti, a Spaniard, who took refuge in Morocco in consequence of the political events of 1811. After having assumed the Moorish dress, and habituated himself to the language, he succeeded in enrolling himself amongst the Emperor of Morocco’s guards. Incessantly occupied with the project of going by this means to Timbuctoo, he secretly offered his services to the French consul, M. Sourdeau; he solicited the protection of the French government: the Consul had ascertained that he combined all the qualifications essential to success, yet his offers were rejected. Neverthless Piloti daily prosecuted his preparations for the journey: nothing it should seem could have prevented his success, since he would have set out under the disguise of a Moor, and returned with a caravan of Moors. In the absence of direct assistance from France, he was furnished with instructions by M. Delaporte, our vice-consul, and from a member of the Geographical Society, who was on the point of sending to him some instruments, when he was suddenly implicated in the political movements of the court of Morocco. The Moors and Jews, always intriguing, suspected some secret designs on his part, and denounced him as belonging to the party opposed to the prince. His trial was summary, and Piloti was beheaded. Such is the jealous distrust manifested by the mercenary race, against any stranger, who, making himself acquainted with the localities, should attempt to deprive them of any portion of their commerce with the interior. M. Caillié, ignorant of this adventure, was more fortunate; he did not continue long enough exposed to the suspicions of the Moors, or by prudence and sacrifices he contrived to escape the effects of them.

On the first arrival of the letters which I received from M. Delaporte and M. Caillié himself, I entertained some doubts of the authenticity of his narrative, and I immediately arranged some questions by way of trial, on the language spoken at Timbuctoo, the customs of the country, its natural productions, the nomenclature and distance of places, &c.: but meanwhile I attentively compared the two letters, and found the result so conformable to the most established notions of science, that I determined to publish on that very day the news of the journey to Timbuctoo. Many were incredulous; I expected it. I requested the traveller, who in the interval had landed at Toulon, to commit to paper without delay his recollections respecting the questions which I held in readiness for his arrival, but to which he had beforehand in a great measure replied. The very day of his arrival he submitted to my inspection a journal of his travels complete, and continued from the 19th of April 1827 to the 21st of September 1828, modestly observing: “I do not know whether I can answer all your questions, but here are my notes.” He then shewed me part of the original notes, written in pencil, on the spot, and the narrative written and completed during his residence at Tangier and in quarantine, though suffering under a high fever. He also shewed me the pieces of cord with which he had measured the meridian shadow at Timbuctoo and in other places, some fragments of plants brought from the interior, the vocabularies, and some simple sketches of the town of Timbuctoo. If after such testimony I could retain any doubts, the construction of all the routes of the traveller, which I drew up on the following day, would have effectually dispelled them, for I discovered that the observations were continued without any interruption, and that the whole produced a result agreeing with the data already acquired.

It remained that the learned Geographical Society should share my conviction, an additional success for which M. Caillié had not long to wait;[135] and he obtained a brilliant recompense which he had amply merited. This example will restore confidence to minds discouraged by so many fatal catastrophes; and will inflame the zeal of those who are actuated by a desire for glory and the advancement of geographical science.

§ V.

OF THE COURSE OF THE DHIOLIBA ABOVE AND BELOW TIMBUCTOO.

The attentive reader who may have patiently followed me through the preceding pages, will, no doubt, have remarked the new and principal fact which results from M. Caillié’s observations; the division of the Dhioliba in the environs of Sego into two branches equally broad and deep, and the existence of a large island. It elucidates the description of Mungo Park, and reconciles him with our traveller; it explains the contradictions between the situations assigned to the same towns by different travellers, sometimes on the right, and sometimes on the left of the river, and finally it enlarges our ideas of the advantages of navigation in the interior of the Soudan. This fact also accounts for the great collection of waters which forms the lake Dhiebou or Debo, because many considerable branches which separate from the western arm unite again with the main stream beyond the tributary which falls into it at Isaca; and the want of declivity in the direction of this junction is the cause of the stagnation of the waters.

It would appear that the river has different names which change with its course. Called at its source Tombia, Ba, Dhioliba, &c., it retains the latter name as far as Sego, where or in the environs it divides; if we may trust the description of the amanuensis of Bello, the left arm is called Baniou, and the right Balio, and after the junction it is called only Couara. But M. Caillié never once heard that the river, which he reached at Galia, and upon which he navigated thirty days, had different names; perhaps because he did not inquire. He only saw a river, Couaraba, which falls into the right branch, but very far to the south. I think then that if the stream is called Couara below Isaca, it is only because the term is generic and signifies a river.

I might stop here and leave it to the reader to draw other consequences from the new observations. But the question of the outlet of this stream is so closely connected with my subject, that the reader would have a right to complain if in this work he found neither information nor opinion upon it. It is universally inquired what becomes of this immense collection of water below Timbuctoo; it is at least necessary to exhibit in a few words the different opinions at present current upon this subject.

The most ancient identifies this river with the Nile of Egypt. It does not appear that the partisans of this opinion had any other foundation than the pretended unanimous reports of blacks, Arabs, and natives. Thus without considering the physical conditions, or taking account of insurmountable obstacles, they maintained as a fact, that the waters which had their rise in the heights of the Soulimana, that is to say at an elevation of from fourteen to fifteen hundred feet, reached the Mediterranean after a course of two thousand leagues. But, what is perhaps still more strange, this notion rested wholly upon the equivocal interpretation of a word, or as we may express it in plain English upon a pun; the word Nile or Nil is generic. In saying that the Dhioliba joins the Nile, the Africans mean no more than it communicates with some other great water, whether it falls into it, or whether it receives it, (for this distinction of arm or tributary is very important). When therefore the Arabs say that the Dhioliba communicates with the Nile or the Bahi, they understand thereby either a great river, or a sea, and this may be an inland sea as well as the Ocean. This opinion that the Dhioliba empties itself into the Nile of Egypt, though it was supported only a few years since by a learned writer, appears to be now altogether abandoned.

But this is not the case with the opinion of those who, like Major Rennell, consider the central lake as the outlet of the river. Before the discovery of the lake Tchâd by the English travellers, the existence of this inland sea might have been doubted, the evidence of it was so vaguely attested. This opinion, however supported by probability, is nevertheless liable to two objections: first, that upon the whole western coast of the lake is found the mouth of only one inconsiderable river, the source of which is at no great distance in the E. S. E.; secondly, that the town of Boussa, to which Park navigated upon the Dhioliba, is now known by the second journey of Clapperton, and that it is very far to the S. E. of Timbuctoo.

With regard to the first objection, it may not prove a serious difficulty, because recent travellers have not followed the river Yéou, which falls into lake Tchad; they have left it at a certain distance from the lake, and it is very possible, that that which they have seen farther on may have been only a tributary to the former. As to the second objection, it might be more important if it were certain that the Dhioliba runs in a single bed from Timbuctoo to Saccatou and to Boussa; but there is nothing to prove this. Continuing eastwards, towards the central lake, it may send out a branch to Boussa; and this division would account for the Yéou consisting of but an inconsiderable body of water.[136]

Reichard was one of the first who imagined that the Dhioliba may run into the Gulf of Guinea. This hypothesis has for some time past assumed a certain degree of probability, to which the opinions of the later English travellers, Clapperton and Major Laing,[137] have added much weight. They differ, however, respecting the outlet of the river: the one preferring the river Benin (or Formosa), with Reichard[138]; the other, but with much less probability, the Rio-Volta. The objection always raised to this hypothesis is the great height of the mountains called Kong. To reach the sea, the river must cross them; but it may not be absolutely impossible that there should exist an opening in them deep enough to admit of its passage. Another difficulty arises from the small declivity of these waters: but I will here make an observation on this subject. The actually known course of the Dhioliba, from its source as far as Timbuctoo, is about three hundred and sixty leagues: it issues from Mount Loma, at a height of nearly sixteen hundred English feet above the sea, or less than five hundred metres. The velocity observed by M. Caillié leads to the belief, that the average inclination from Djenné and also from Bamakou to Timbuctoo is two thirds of a metre to a league: Timbuctoo would stand, according to this datum only, at a height of two hundred and sixty metres; but it is very probable, that the inclination is much greater from Mount Loma to Bamakou than it is below this latter point, which would lower the position of Timbuctoo at least to two hundred and thirty metres, taking the fall of the first part at only a metre for a league. But this quantity would greatly exceed that supposed by Capt. Beaufort, who, after having observed the elevation of Elimané, conceived Timbuctoo to be upon the same level, that is to say, eighty-four metres above the sea.[139]

Now, from Timbuctoo to the mouth of the river of Benin, following the course of the waters (as it is traced by the partisans of this opinion), the distance is not less than four hundred and sixty leagues. Thus, in the second part of its course, the river would have a total declivity of 230 metres, or 0,51 metre to a league. It is known that the Seine has an inclination of 0,72 metre to a league; the Mississipi, 0,84 metre; the Rio-Apure, 0,92 metre, &c. but others have a much less fall, such as the Wolga, the Missouri, the Senegal, &c., which have one of 0,50 only;[140] so that, strictly speaking, the above inclination is sufficient.

According to a fourth opinion, the river, on reaching the Kong mountains, makes an elbow to the left and runs eastward, by Djacoba and Adamowa to Chary, and thence to lake Tchad into which it discharges itself. It is here that the objection of the want of sufficient inclination applies: how can it be admitted, that the river, after passing Funda, where it would scarcely have an elevation of fifty metres[141] above the sea, (supposing it to be the river of Timbuctoo which flows to Funda), can run on to lake Tchad, three hundred and fifty leagues farther, through a country represented by all accounts as mountainous? But this, even, is not the greatest difficulty.

It is hardly conceivable that any geographer should have admitted an hypothesis, the absurdity of which must have been manifest upon the slightest reflection. The height of lake Tchad has been ascertained: it is nine hundred and twenty French feet above the level of the sea, or something less than three hundred metres; it cannot, therefore, receive the waters which flow to Funda.

The course of this river east of Funda must be reversed, and the supposed elbow converted into a tributary: we shall then probably approach the truth. Major Denham was the first to conceive this easterly turn of the river, running north of the great chain of mountains, and falling at a great distance into the central lake; he had been assured that a communication existed between this river and lake Tchad by the Chary. How is it that the physical impossibility of this course did not occur to him?

A very simple consideration appears to afford a solution of this difficulty, namely, the existence of a lake in an elevated point of the Mandara chain, giving rise both to the Chary and the river which flows by Adamowa and Djacoba. The reports made to Major Denham demonstrate the importance of this stream, which is sufficiently proved by its extent; but, why, without ocular testimony, did he imagine its course to be easterly? If the negroes did not inform him that it ran to the west, neither did they state the contrary. Let us admit the westerly direction: a certain communication will then exist also between the Couara and the Chary; only, after having descended a current southwards, we must ascend another eastward, and thence redescend northwards, into the central lake. This is the most plausible theory I can devise upon Major Denham’s opinion; this is nearly the case with the White Nile and the Misselad, both taking their rise in one of the lakes of Gebel-Koumri. Upon this system, the Couara will continue, after the confluence, to flow southwards, and fall into the sea near the coast of Benin.

A fifth opinion has been recently broached by the English General Sir Rufane Donkin;[142] the summary of which is that the Niger crosses the Wangarah, enters the valley of Ouadi-el-Ghazel, formed by the continuation of the Misselad, and thence runs into the Mediterranean (in the great Syrtis) by a subterranean channel under the sands of Bilmah; and moreover that the Niger rises near the Gulph of Guinea, instead of running towards it. This rather extraordinary opinion has met with adversaries, at which we need not be surprised, even after having read the arguments on which the dissertation is founded. I do not therefore think it necessary to discuss it here; neither shall I enter into the complicated notions, upon the courses of the central river, hazarded by the English traveller Bowdich on very vague information; and I shall be cautious not to offer an additional hypothesis of my own respecting this problem, still full of obscurity. On what basis can an entire and complete system be founded while even the names of the central regions are unknown to us, and our researches into the physical geography of these vast tracts are yet in their infancy; when, in short, the papers of Major Laing, should they be recovered, may at once throw strong light on these chasms in science?[143] It may however be affirmed, and I think with certainty, that the rivers called Dhioliba and Couara neither join the Egyptian Nile, nor contribute one drop to its waters; I think besides that, if the Couara of Funda actually is the continuation of the Dhioliba, flowing to Sego and Timbuctoo, and falls into the Gulph of Guinea, there is nothing to hinder it from throwing off a branch to the east, which may have its outlet in the Yeou and the central lake: this would be the branch seen by M. Caillié before he reached Timbuctoo, and which our traveller followed, the other being on his right flowing east-south-east; and there is no proof that the whole of the former rejoins the latter. The great lake Tchad, or central sea, would then not be the general receptacle of the Dhioliba, but only one of its outlets.