Some two years since, I had the honor to read before the Naval Institute, a paper in which I attempted to present a comprehensive view of the Interoceanic Canal question, as it then appeared in the light of the latest reliable information.
Since that time important additions to our knowledge concerning two proposed routes have been made, and my duty as a member of the Institute requires that I should acquaint it with the facts. It will not, however, be necessary that I should speak of the Panama route, since you will doubtless receive information concerning the results of the survey of that line by Commander Lull, from other sources.
I will therefore confine myself to a description of the route by way of the Atrato and Napipi rivers, giving you the information acquired by an expedition under my command that made a careful survey of that route during the winter just past.
For a general description of the country in which this proposed route lies, I would refer to my previous paper. I wish now merely to describe its physical features, and to discuss briefly the question of its adaptability to the construction of a ship canal.
At the time of the reading of that paper, our knowledge of the country in the vicinity of the Napipi, was confined to observations made from the river beds. All our surveys had been made in the beds of streams, and in projecting a canal line, it was assumed that the country for half a mile or so back from the river, was the same as it appeared on the banks. More intimate knowledge shows such a supposition to be erroneous. The country that, from the river, had appeared to be flat, was found to be covered with a network of hills.
This fact is of great importance as showing the difficulty of acquiring information concerning the character of those regions. It also shows how easily experienced men, anxious only to learn the truth, may be deceived, and it enables us to enunciate the general rule that favorable reports concerning any route, not substantiated by the results of actual survey over the ground itself, are worthy of little consideration.
In saying this, I do not by any means wish to be understood as depreciating the value of surveys made in the beds of streams. On the contrary I accord to such surveys great value, and believe that all preliminary reconnoissances, in such a country as Darien at least, should always be so made.
But a most important distinction must be made in estimating the value of the results so obtained. And for this reason; a survey in the bed of a stream, naturally shows the lowest possible profile within the region that the stream and its tributaries drain. If, therefore, such a survey shows an unfavorable profile, the question of the possibility of finding a favorable one anywhere within those limits must be regarded as conclusively settled in the negative.
But— and mark the difference—if such a survey shows a favorable profile, the question of the adaptability of the line is by no means to be regarded as settled in the affirmative. No ship canal can follow the windings of a mountain stream, and before any accurate estimate can be formed of the character of the line of the canal, that line itself must be carefully surveyed from end to end.
The principles then to be borne in mind when called upon to judge of the value of explorations are these—unfavorable reports founded upon reliable surveys in the beds of streams may be accepted without hesitation; favorable reports with such foundation can be regarded only as indicating the desirability of a more careful examination before judgment can be formed as to the merits of the route.
Furthermore, let me say that, reports of favorable routes on the authority of "old Spanish maps and documents," or from "conversations with the Natives," or of passes in the Cordilleras seen from either shore; or of low land seen from a height, or while passing up or down rivers; or of passes determined by the temperature of boiling water or the velocity of streams—all such reports, I say, need no longer agitate the geographical and commercial worlds—they may be set down at once as a snare and a delusion.
Applying these principles to the results of the surveys of the past six years, we find that from the surveys in the beds of streams the various routes that have been from time to time proposed between the Panama railroad and the Napipi river (with the exception of the Truando route which has not been examined since Nichler's survey) have been pronounced unfavorable; and such unfavorable verdicts may be accepted without hesitation.
From a similar examination the Napipi route was pronounced favorable. This result should have been accepted simply as indicating the desirability of further surveys; it was altogether wrong to suppose that sufficient data had been acquired to justify the preparation of estimates of cost. The additional examination that has since been made has shown the character of the country to be essentially different from what we had been led by the previous surveys to suppose.
This route still remains, however, even with all its disadvantages, the most favorable one south of Panama, at least. The fact that it is so poorly suited to the purpose, is a sufficient index to the character of the others.
But let me proceed with a description of the country in that vicinity, which was my avowed purpose when I set out, but to which I have been long in coming.
Our survey followed the left or northern side of the valley of the Napipi, from the Atrato to the junction of the Doguado. At this point the line crossed the Napipi and followed the valley of the Doguado till opposite Chiri-Chiri Bay, when it struck across the "divide" and descended to the Pacific, in the valley of the Chiri-Chiri River.
In reference to the general topographical character of this region, I may say that our observations on the main line of survey, and in extended reconnoissances, showed its physical features to be wonderfully systematic. From the main divide, which skirts closely the Pacific coast, come down, to the eastward, long spurs or ridges that form the divides between the various western tributaries to the Atrato. These ridges send off to the northward and southward smaller spurs and these divide again and again, till the whole country is overspread with ranges of hills, running the one into the other, like the veins on a leaf. A detached hill is rarely to be found.
The crests of these hills, are usually very narrow and their sides descend abruptly—often precipitously. These crests rising and falling with gentle slopes, always afford good ground for walking. On this account they are used for roads or trails almost exclusively by the natives, they having long ago learned that the longer way round, with a good road, is a surer way home than the shorter, which, cutting across the ridges, presents a succession of steep and slippery hillsides. They thus practically exemplify our proverb, even if they do not put it into words.
Immediately along the line of survey, the country naturally divides itself into four divisions, as regards its topographical features. First, there exists from the banks of the Atrato, for some five miles to the westward, a flat, swampy region of a lower average level than the banks of the adjacent streams. During the wet seasons, this region is frequently inundated to a considerable depth. During the dry seasons its more elevated portions become sufficiently dry to be passable, but those of a lower level always remain open water swamps or miry morasses.
This portion of the route is in fact the delta of the Napipi, since it is bounded to the westward and northward by a second mouth of that river called the Braso Muriel, while a third mouth, the Palmerito, flows through its central part. It includes, in the portion near the Atrato, two large cienegas or lagoons, which during the wet months, are shallow lakes, but which become more or less dried up as the rain lessens.
The second topographical section extends from the Braso Muriel some six miles to the westward. It is characterized by the extension of the spurs of the divide between the Napipi and Opogado rivers to the very banks of the former, rendering it necessary to cross them continually with the survey. Extended reconnoissances were made to determine the practicability of flanking these hills by a detour to the northward, but in almost every case they were found to increase in altitude as they receded from the river, proving to be parts of the system just described, rather than detached hills that might have been flanked.
When, therefore, these hills butted on the river, there was no course left but to cross them, and our line through this section shows a succession of steep hills, the highest being 253 feet in elevation.
The third topographical section extends from the western limit of the second, to the point at which the projected canal crosses the Napipi, near the junction of the Doguado and Meriudo rivers. In this section, the spurs or ridges generally terminate at a thousand or fifteen hundred feet back from the river, and it was accordingly found possible to avoid them by keeping the line well down to the southward. Our profile, therefore, of this section shows level or gently undulating ground, with no elevations of any considerable magnitude.
This section, in common with all except the first, is well covered with heavy timber, which appears wherever the ground loses its swampy character.
The fourth topographical section extends from the point of crossing the Napipi, to the Pacific ocean. It is characterized by being extremely broken, and by the great height of its ridges as compared with those of other sections.
For the greater part of this section the line lies in the valley of the Doguado. The distance between that river and the Meriudo is so small, and the divide between them so high, that the spurs extend almost invariably directly to the river banks. A line of survey carried up the valley anywhere except in the river bed, must cross these spurs continually.
In this section the main dividing ridge between the valley of the Atrato and the Pacific slope is found. It was crossed at an altitude of 778 feet, and so steeply does its western slope descend that the crest of the divide is only 7,000 feet in direct horizontal distance from the beach. The tunnel required for passing under this ridge is three and one half miles in length.
The direct distance from the Pacific at Chiri-Chiri Bay to the Atrato, is but about 28 miles, but the introduction of the curves, necessary to follow the best ground, increases the actual length of the canal to thirty and one quarter miles.
From the data acquired by the last expedition, I have calculated the probable cost of a canal by this route, according to the general plan proposed by Commander Thos. O. Selfridge. This plan, in its essential features, is to follow the left bank of the Napipi for about 30 miles, to its junction with the Doguado. At that point the canal is to cross the Napipi by a dam of sufficient height, and then to follow the valley of the Doguado, till the cutting becomes too deep to continue an open cut. A tunnel will then be resorted to, to carry the canal under the dividing ridge, and bring it out on the Pacific, at Chiri-Chiri Bay.
In the plan on which the following estimates have been founded the summit level has been placed at 143 feet above mean tide, and 12 locks with a lift of 10.3 feet each on the Atlantic side, and 10 locks
with a lift of 14.9 feet each on the Pacific side will be required.
The Canal is to be fed at the summit from the Napipi river with its tributaries, the Doguado and Meriudo, as well as from the next river to the southward, the Cuia, for which purpose an aqueduct three and two-tenths miles long is, necessary.
The estimates include all the works supposed to be necessary to the successful operation of the Canal, and for its preservation and protection from accident, by floods or otherwise.
The following dimensions have been assumed as best suited to meet the requirements of the case:
Width at bottom | 72 feet |
Width at water surface in earth | 150 feet |
Width at water surface in rock | 98 feet |
Slope of sides in earth cuttings | 1 ½ to 1 |
Slope of sides in rock cuttings below water | ½ to 1 |
Slope of sides in rock cuttings above water | ¼ to 1 |
Width of top embankment | 9 feet |
Slope of embankment, exterior | 2 to 1 |
Slope of embankment, interior | 1 ½ to 1 |
Width of “bench” (at ten feet above water) | 9 feet |
Width of locks inside | 60 feet |
Length of locks between mitre-sills | 400 feet |
For the tunnel calculations have been made with several different forms and dimensions. In the estimates given below, a tunnel 60 feet wide at water surface, with 30 feet depth of water and 86 feet height .above water, has been allowed for. If a tunnel 70 feet wide should be considered necessary, the cost would be increased by about three and a half millions of dollars. In order to be on the safe side it has been assumed that the tunnel would require lining throughout with an arch of masonry.
The following prices have been allowed in computing the cost of the various portions of the work.
Excavation in earth | 33 cts. per cubic yard |
Excavation in rock | 1.25 to 1.50 per cubic yard |
Excavation of tunnel | 5.35 per cubic yard |
Arched lining of tunnel | 20.00 per cubic yard |
Dredging | 50 cts. per cubic yard |
Hydraulic concrete for Locks, Culverts, Dams, &c., $7.00 to $8.00 per cubic yard.
As close a calculation of the amount of work of every kind, as is possible with the data at hand, has been made, and the cost calculated at the above prices with the following result:
Estimated Cost of Canal.
Excavation and Embankment | $28,697,398 |
Tunnel | $33,341,923 |
Locks | $5,049,214 |
Culverts | $3,031,405 |
Side Drains | $2,449,953 |
Dam for crossing the Napipi | $616,057 |
Aqueduct for feeding Canal from Rio Cuia | $548,726 |
Diversion of rivers | $1,670,159 |
Grubbing and Clearing | $191,900 |
Breakwaters at Chiri-Chiri Bay | $2,613,000 |
Improvements at mouth of Atrato | $817,780 |
Light house at each terminus | $60,000 |
| $78,557,515 |
Add 25 per cent, for contingencies | $19,639,379 |
Total estimated cost | $98,196,895 |
Without entering into any comparison of this canal route, with others that have been proposed, I present the following as the chief advantages and disadvantages of the Napipi route, according to the best of my judgment.
Advantages.
- Shortness of artificial channel required.
- Good Harbors. That on the Atlantic side is all that could be desired, while at the Pacific terminus there is deep water with good holding ground, and the region is seldom visited by violent gales.
- The cutting mainly in rock or stiff, tenacious clays. In such materials the amount of excavation can be reduced to a minimum; the clay will form stable embankments, and its impervious character will greatly reduce losses from leakage and filtration.
- Proximity of the heaviest work to the Pacific, rendering transportation of labor, "plant" and supplies inexpensive.
- The greater part of the work to be performed lies in a healthy region (for the tropics).
- Abundance of good timber for construction.
- Absence of high winds along the line of the canal. Transit would be greatly impeded in a canal lying through a region of violent winds.
- Freedom from liability to terrestrial convulsions of a nature likely to interfere with the permanency of the canal works.
- Absence of large streams, or of deep valleys to be crossed at a high elevation.
- Friendly attitude of the inhabitants.
- Fertility of the soil. With proper management, the country in the vicinity of the line could be made to produce the greater part of the supplies required by the laborers.
Disadvantages.
- The necessity of resorting to a tunnel. This, while it is no doubt practicable, involves great expense in construction, uncertainty in estimate of cost, and a probable increase in the difficulties attending transit, especially for large ships,
- The steep descent of the Pacific slope, requiring the grouping of a large number of locks, and consequently increasing the liability to accident to the works.
- Very heavy cuttings required in the valleys of the Doguado and Chiri-Chiri.
- Limited water supply during dry seasons.
- Liability to damage to the works from sudden floods. It is believed that this contingency is well guarded against, yet the liability to sudden and violent floods in a hilly country, subject to excessive rains, cannot be overlooked.
- Excessive rains likely to wash away embankments, while in course of construction, and to interfere generally with the progress of the work.
- Shortness and uncertainty of the yearly periods well suited to the work of construction.
- Undeveloped state of the country and scarcity of native labor.
- Remoteness from the great commercial centers of the world.
The above are all that have occurred to me, with what attention I have been able to give the subject. Those accustomed to the contemplation and execution of great engineering schemes will doubtless see many more on both sides, while it is quite certain that, in the actual execution of the work, many complications and contingencies will arise that the best minds will now be unable to foresee.
In concluding this paper, it may not be inappropriate to present my views as to what general conclusions may be drawn from the results of the long series of explorations that have been perseveringly carried on during the past six years.
The main fact to be deduced is that the construction of a ship canal between the Atlantic and Pacific Oceans is to be a work of truly Herculean proportions; a work involving the expenditure of much time and treasure in its execution, and demanding the exhibition of as great engineering skill as has been put forth in any work as yet accomplished by man.
The dreamy hope, that has existed since the days when Columbus searched in vain for a natural strait, that somewhere among the gloomy fastnesses of the isthmus there might be found a spot so exactly suited to the purpose that the construction of an artificial channel would be an easy task, must be regarded as forever dispelled.
Nature has not been so kind as to leave the gateway open, and when man shall essay the task, he must be prepared to find his highest faculties and greatest energies taxed to their uttermost.
In fact, in view of the difficulty of constructing a canal, it does not appear to me improbable that the question of transit will be finally solved by resort to some form of marine railway, by means of which ships with cargoes intact shall be hauled overland from one sea and launched to continue their voyages upon the other. Such a project is by no means new, but the inherent difficulties which it presents have caused it to be kept in the background, so long as any chance of finding a place well suited to the purposes of a canal existed.
But with the demonstration of the magnitude of the Canal project, and the recent improvements in the raising of ships, and in mechanical appliances generally, it is likely to assume greater importance. I have recently been informed that certain parties in New York are taking the matter into serious consideration, and, if they will be able to insure the safety of ships while in transit, their project may be as likely to meet with favor as that of a canal involving locks, dams or viaducts and tunnels, or other difficult engineering works.
The question as to which route presents the most favorable conditions to the construction, maintenance and successful operation of a ship canal involves so many considerations as to make its solution a matter of extreme difficulty. Without entering into any discussion of the relative merits of each, I will say that, taking everything into consideration, and after careful deliberation, I am perfectly well satisfied that the Nicaragua route presents more favorable, and fewer unfavorable conditions than any of the others. The fact that this route is open to serious objection only goes to prove the remarks just made concerning the inherent magnitude of the projected task.
I believe that the construction of a successful ship canal through the isthmus of Nicaragua will be one of the most difficult enterprises that man has ever yet undertaken. Still I am confident that it will be less difficult there than elsewhere, and, with time, money, and the highest order of engineering talent, it cannot be regarded otherwise than as perfectly practicable.
In regard to the possibility that a better route than any that has been examined may have been overlooked I would say that the explorations have been so conducted as to preclude such a possibility.
In a country as densely wooded as is the one in question it is practically impossible to cover every square mile of it with lines of survey; hence there will always be left a chance for interested parties to declare that, had the last explorers gone a little farther to the north, or a little farther to the south, they would have found the exact spot desired. But the fact that the survey of the bed of a stream is fully competent to decide the unfavorable character of all the region drained by that stream and its tributaries, has already been dwelt upon. It is therefore easy to see that, when all the principal water courses have thus been followed up with unfavorable results, further search is not only unnecessary, but positively absurd.
I am aware of the fact that certain persons are loudly proclaiming that the United States are politically opposed to the opening of a canal, and that all these explorations have been instituted for the purpose, and carried out with the idea, of demonstrating the impracticability of such an undertaking.
I do not propose to enter into any controversy with persons capable of conceiving and publicly proclaiming such idiotic notions. The good faith of the government and the honor of its officers cannot need to be vindicated to its own citizens. And if people can be found elsewhere sufficiently simple to pin their faith to visionary enthusiasts or unprincipled adventurers, in preference to accepting the statements of responsible officers whose very position guarantees absolute truthfulness, we can well afford to let them do so.
Further surveys will greatly add to our knowledge of the topography of the various regions, and in the interest of geographical science it is to be hoped they will soon be made. But they can only result in demonstrating the general correctness of the conclusions already drawn, and those who embark in them with any other hope will find, in the end, that they have parted with their money to no other purpose than the demonstration of their own foolishness.
In a word, it may be said that no reasonable doubt can now exist that the data necessary to a determination of the most favorable place for the connection of the oceans by a ship canal have been secured.
Commander Farquhar stated that he was very sorry that Lieut. Collins was not present himself, in order to answer some very interesting questions which would most probably have been asked by some of the members. He himself, he said, was of the opinion that the great difficulty in regard to the building of the canal, was the estimating of the cost of the work. It being proposed that the lowest depth of water in the canal should be twenty-six feet, he did not see how any reliable estimate of the cost of the simple excavation could be made. Should rock in large quantities be encountered during the process, the expense would be, of course, greatly increased. So little was known of the nature of the country, that no guarantee could be given that such a contingency might not arise.
He was attached to surveying expeditions across the Isthmus of Tehuantepec, some years ago, and it was while on that duty that he had come to the conclusion that it was next to an impossibility to make, with any degree of certainty, an estimate of the cost of the work. Lieut. Collins, himself, he continued, in an article read before this Institute, about a year ago, estimated the cost of this Canal to be fifty millions of dollars; now, in his last paper, which has just been read before the Institute, he says that it will cost over ninety millions of dollars. Another great difficulty, would be to get from the tunnel, which is twenty-five hundred feet above the level, to the Pacific. The idea of having twelve locks, in such a short distance, with a fall of over two hundred feet each, was an almost impracticable undertaking. The only way that he could see of arriving at any conclusion in regard to the cost of the canal was to compare it with other works of a like character. The Suez Canal cost about eight hundred thousand dollars a mile. It was to be remembered however that it was built almost entirely by forced labor, and that the country through which it runs is perfectly level, and the soil entirely free from
rock. The country across the Isthmus of Darien offered no such advantages. Labor would be very high. Taking these facts into consideration, and acting upon them as a basis for an estimate, the members of the Institute might be able to form some little idea of how vast was the undertaking, and what tremendous expense it involved.
Rear Admiral Rodgers moved that the thanks of the Institute be tendered to Lieut. Collins for the great service he had rendered to science in his thorough description of the topographical and geographical nature of the country across the Isthmus of Darien.