Papers Relating to the Foreign Relations of the United States, Transmitted to Congress, With the Annual Message of the President, December 3, 1889

[Inclosure 1.]

Major Ernst to Mr. Bayard.

United States Engineer Office,
Galveston, Texas, December 12, 1888.

The Secretary of State,
(Through the honorable Secretary of War, and Chief of Engineers, U. S. Army).
Washington, D. C.:

Sir: I have the honor to report that, in compliance with the instructions contained in your letter of the 19th ultimo, I have visited El Paso, Texas, where I arrived on the 24th ultimo and remained until the 30th ultimo, and have examined the works under construction by the Mexican Government for the protection of the right bank of the Rio Grande at that place. The following maps, photographs, and documents are herewith transmitted:

(1): Comparative map, showing approximately the change which occurred in the location of the river between the years 1855 and 1885. Scale 1/15000.
(2): Map which accompanied the original project of the Mexican engineer, showing the banks of the river in 1885, the proposed location of the works, some details of their construction, and the water-ways at a medium stage. Scale 1/5000.
(3): Sketch showing approximately the water-ways as they appeared November 26, 1888, and the location of the works as actually constructed. Scale 1/5000.
(4): Six photographic views of the works taken November 27 and 28, 1888.*
(5): Translation of the original project for the protection of the bank submitted by the Mexican engineer to his Government in 1886, under which the work has been done.
(6): Minutes of the proceedings at the conference between the engineers of Mexico and the United States in the Spanish† and English languages.

Leaving a rocky gorge just above El Paso the Rio Grande issues into an alluvial plain some 5 or 6 miles in width, through which it flows in a bed of its own formation. The banks and bottom are of light earth and sand, easily moved by the currents. The river is a silt-bearing stream carrying at times an excessively large proportion of solid matter. At the lowest stage the water ceases to flow and except in detached pools the bed becomes entirely dry. During the spring freshets the water [Page 623] sometimes rises to a height of from 9 to 10 feet above the lowest, and flows with a rapid current heavily charged with sediment. At such times it possesses great building as well as destructive power. Between these two extremes there are at different seasons all degrees of volume and velocity. The size and character of the stream are ever varying, and its requirements as to form and dimensions of bed vary equally. The river’s work of altering its bed to suit the necessities of the moment is never ending. The bed as it happens to exist now is the final resultant of all efforts of the stream, some of which have neutralized and some have aided each other, but the bed is not fixed. It is shifting from one position to another, altering its course, eroding one bank and building up the opposite one, forming islands and bars and then destroying them. The result of the natural changes is most noticeable in a bend, where the erosion of the concave shore is sometimes continuous for many years, as appears to have been the case at El Paso. The location of the river at this place as it was found in 1855 is shown in black upon Plate I, which is an enlargement of a portion of sheet 29, United States Mexican Boundary Survey, furnished me with your letter of instructions. Upon the same plate is shown in red ink the location of the stream as it was found by Mr. Garfias, the Mexican engineer, in 1885. It will be observed that between those dates there has been a very considerable encroachment upon Mexican territory, with a corresponding advance of the American shore. The maximum distance between the shore of 1855 and that of 1885 is about five-eighths of a mile, and the total area added to American territory is about 490 acres. I have been unable to learn whether this movement has been continuous throughout the thirty years or whether it has been intermittent. Of late years there has been a tendency to erode the American side. Beginning a short distance below Dike 1, for half a mile or more the American shore bears evidence of recent erosion. The vertical face of the bank indicating recent caving is partially shown upon photographic view No. II.

The evidence as to when the erosion began or how far it has extended was contradictory. These points must be determined, if at all, in a court of law.

In 1886 Mr. Ignacio Garfias, civil engineer, submitted to the Mexican Government a project for the protection of the right bank. A translation of this document is herewith transmitted. A tracing of the map which accompanies it constitutes the inclosed Plate II. Mr. Garfias took the ground that the left bank, as well as the right bank, belonged to Mexico, and that any works which might be required to push the river back to the position which it occupied in 1855 were legal and proper, provided they did not interfere with navigation. (There is no navigation of any kind upon this portion of the river, except a small skiff used as a ferry, and there never can be any.) He furthermore held that in this case it would not be necessary to encroach upon the left bank, and that while the use of dikes projecting into the stream to the injury of left bank was justifiable, some persons might object to it, and he could and would locate his works so that the current would not be thrown upon the left bank. He proposed a series of deflecting dikes, eight in number, which are shown upon Plate II. The total length of shore which they are designed to protect is about 2½ miles. They are numbered consecutively according to location, and not according to the order of their construction. He rejected the system of continuous revetment of the bank, on account of its supposed greater cost, and he announced that at present in the United States and in Holland, no works for the rectification of a river’s channel, the defense of its banks, or the improvement of a bar are constructed upon any other system than the one adopted by him. It may be remarked, in passing, that he was misinformed as to the United States.

The work was begun in August, 1886, under the direction of Mr. Felipe Zavalza, civil engineer, with the construction of Dike 4, which was completed in January, 1887. It was followed by Dike 3, which was finished in July, 1887, also under the direction of Mr. Zavalza. Mr. Garfias then took personal charge of the works, relieving Mr. Zavalza, and has remained in charge up to this time. Between August, 1887, and January, 1888, he built No. 1; between January, 1888, and March 1888, No. 2, and subsequently No. 5, which was completed early in November, 1888. He then began No. 6, but soon suspended operations there to await the action of higher authorities upon the complaint made against his works by the citizens of El Paso, Tex. At the time of my visit he was engaged in preparing material for the construction of No. 8.

The general method of construction is the same for all the works.

Brush is first made into closely-bound bundles, called fascines, about 6 inches in diameter and of varying length. A course of fascines is laid at right angles to the general direction of the dike and about 3 feet apart. Over these a second course is laid, parallel with the general direction of the dike and also about 3 feet apart. Above these, again, a third course is laid parallel with the first, and also about 3 feet apart, and so on until six courses have been placed, the fascines in each course being laid in a direction perpendicular to that of those immediately below. Each course is secured to those below by withes and pickets which pass through three courses. The [Page 624] fascines intended for the longitudinal courses are made long enough to extend from one end of the work to the other. Those intended for the enter courses are made of a length necessary to extend through from one side of the dike to the other. In this state the structure constitutes a sort of crib-work or grillage, having very large interstices. In the voids short brush is packed vertically as closely as possible by hand. These six courses, after being filled in with brush, constitute what Mr. Garfias calls a mattress. It is made in place, or, when the water is too deep for that, it is made upon inclined ways and then floated into position and sunk. In either case it is loaded with stone sufficient to hold it in place when submerged. One mattress being completed and placed, another is constructed in a similar manner and superposed on the first.

The width of the top mattress is 20 feet, that being the thickness of the dike on top. The width of the others increases with their depth below the top as shown in the cross-section, Plate II. The top of the dike when completed is on a level with the highest floods. It is finished with a top-dressing of loose gravel, as shown in photograhic view No. 6.

When first placed the dike is permeable and allows the water to pass freely through it. It is a silt-catching device similar in its action to the hurdles used for building up new banks upon the Mississippi River. Checking the velocity of the water and smoothing out its boils and whirls, it causes the water to drop its load of solid matter and to make deposits above and below and within the work itself. It is most active at the higher stages of the river, when the water is most fully charged with sediment. It is a thoroughly efficient device for building up a new bank in advance of the old one, but, like all of its kind, must be kept in repair.

Some changes in the river bank subsequent to the original project involved some slight alterations in the location of the works. As actually constructed they are shown upon Plate III, upon which also are sketched the water-ways approximately as they appeared at the time of my first inspection, November 26, the stage of the river being a few inches above low water. The works not yet built are shown upon this plate in broken lines. In plan all the dikes have one main branch, connected with the shore at its upstream end and running obliquely into the river until its down-stream end is at a distance from the shore which differs from the different works. Those first connected, Nos. 4, 3, and 1, have a second or return branch which connects the downstream end of the first branch with the shore below. This return branch being found unnecessary, has been omitted in the later works.

The following lengths and distances have been measured from the map furnished by the Mexican engineer, Plate III: The main branch of No. 1 is about 325 feet long, and its greatest projection from the Mexican bank is about 90 feet, the total distance between the two banks here being about 325 feet. A view of this dike is given in photographic view No. 1. No. 2 is about 450 feet long, and its greatest projection is about 110 feet, the total distance between the banks here being about 300 feet. This dike appears in photographic view No. 1 and also in view No. 2. The main branch of No. 3 is about 250 feet long, and its greatest projection is about 60 feet, the total distance between the banks here being about 180 feet. This dike appears in photographic view No. 3. The main branch of No. 4 is about 325 feet long, and its greatest projection about 250 feet, the total distance between the banks here being about 430 feet. This dike appears in photographic view No. 4. No. 5 is about 500 feet long, and its greatest projection is about 170 feet, the total distance between the banks here being about 600 feet. The down-stream end of this dike connects with a pier of the railway bridge, as shown in photographic views Nos. 5 and 6.

Just above the railway bridge there begins a decided increase in the distance between the banks. Near the wagon bridge, where dike No. 6 was begun, the distance is about 700 feet. The river is here divided into two arms (see Plate III), which are separated by a high bar or low island. Incipient vegetation is growing upon this bar or island, and it is submergable only at the higher stages of the river. It was not entirely submerged during the year 1888 at all, the highest freshets of this year having reached a level less than 6 feet above the lowest and between 3 and 4 feet below the highest. A few days before my visit the left arm was entirely dry, and all the water which was passing flowed through the right arm in a stream which, I am informed, was not over 30 feet wide and but a few inches deep. This fact demonstrates that the right arm is the deeper arm, and that the middle of it is therefore the boundry of the two countries.

Dike No. 6, as projected upon Plate III and as its construction was begun, is about 350 feet long and extends entirely across the right arm at the lower stages. At the higher stages of course the right arm becomes wider and the obstruction is not total. As already stated, work upon this dike had been suspended. The first courses of fascines had been placed for the entire length of the dike, and at the outer end they rested upon the dry ground of the bar or island. They had been ballasted with stone, for only a short distance from their up-stream end—say about one-quarter of their entire length. I was unable to obtain a photographic view of this work, for the reason [Page 625] that between the date of my first inspection, November 26, and that of my second, November 27, when I was accompanied by a photographer, there was a rise of over 4 feet in the river, which submerged a part of the work and disturbed much of it that had not been ballasted.

Dike No. 7, as projected upon Plate III, also extends entirely across the right arm. It had not been begun, for the same reasons that had caused the suspension of work upon No. 6.

The main branch of dike No. 8, as planned, is about 325 feet long, and its greatest projection from the Mexican bank is about 160 feet, the total distance between the banks here being about 780 feet. No work had been actually placed in the river here, but the fabrication of fascines had been begun, and it was the intention of Mr. Garfias to proceed with the construction of the dike.

It thus appears that the five dikes constructed project from 60 to 250 feet from the Mexican bank, and occupy at the points where they are located from about one-quarter to more than one-half the entire width of the stream between high-water banks. They are of a class which are designed to utilize the building power of the river itself, and their effect will be to push out the Mexican shore between them as well as behind them. The river will not endure such material reduction of its bed. The area of cross-section will be restored by excavations elsewhere. This will be done partially at the expense of the opposite bank and partially at the expense of the bottom. There will be a deepening of the channel at the outer extremities of the dikes which will tend to undermine them. But the material of the bottom being heavier than that of the bank, the restoration of the area of cross-section will, in all straight or gently curving portions of the river, be made principally at the expense of the opposite shore. In the sharp bends, where much the strongest current is found upon the concave side, the restoration of the area will be more at the expense of the bottom. The excavation caused by this sudden contraction of the width will at first load the stream with an amount of solid matter which it can carry only while it retains the increased velocity due to the contracted section, and a portion of which it will drop at the first wide” place below, where in finding its normal area of cross-section its velocity is slackened. This may cause a temporary bar to make its appearance below the dike and on the opposite side of the river. Mr. Garfias claimed that such bars had actually formed below his works, and he held that they proved that his works were not injuring the opposite bank. The fact may be admitted, but not the conclusion. Subsequent freshets, when not overloaded with sediment, approaching such bars, will sweep them away, and may then attack the bank behind them. He claimed also that no erosion of the opposite bank had thus far been caused by his work, and that such erosion as is now going on began before the works were constructed. If it be admitted that the erosion began before the works were constructed—and I am unable to say that it did not—it does not by any means follow that it has not been increased or hastened by them. Or, if it be admitted that even this has not as yet occurred, it does not follow that it will not occur in the future. Dikes 1 and 2 have been in existence through only one high-water season, that of 1888, in which the river was less high than usual. Their bank-building effect has thus far been moderate, and their maximum effect upon the opposite shore has therefore not been reached. They are permeable now in their upper courses, but will hereafter become solid. The deposits above and below them are now comparatively low, but they will increase in height with every flood. The proposition of the Mexican engineer is, in substance, that the width of a river flowing in a straight reach between alluvial banks can be contracted one-fourth by works constructed upon one side of it only. Hydraulic engineers in general will not assent to this proposition, In the case of Dikes 1, 2, and 3, I should expect that if the works be maintained in full efficiency as they have been constructed they would push out the Mexican shore to approximately the same distance that they themselves extend into the stream, and that as the result of this there would be a slight deepening of the channel, but that the American shore would recede a distance only a little less than that by which the other shore advanced. If allowed to stand the wear and tear to which they will be subjected without repair, they will cause a less advance and less recession of the opposite shore. In the case of Dikes 4 and 5 I should expect a similar advance of the Mexican shore, but owing to the sharp bend here not much, if any, recession of the opposite one, except what might be caused by the works above. The probable effect of Dike 6, if completed, is more serious. In connection with No. 5 above and No. 7 below” it would probably close the right arm of the river and throw the entire volume into the left arm, thus making the boundary difficult of identification, and transferring all the destructive power of the American territory. It should be remarked that when these works were planned the condition of the river was different to what it is now. The main stream then crossed the bed in a single channel over ground which is now occupied by the high bar or island, as may be seen by superposing Plates II and III and looking through the tracing. No. 8 would cause an advance of the Mexican shore, and possibly some recession of the opposite one, but owing to the fact that the opposite shore here [Page 626] diverges with a sharp bend, the latter result may not occur. These are given as the probable effects of the works after two or three high-water seasons. Their ultimate effects are somewhat uncertain. A new force is injected among a multitude of natural forces which are already in a state of unstable equilibrium. A well-defined erosion of the left bank having been inaugurated may possibly continue after the original inciting cause has ceased to exert a direct influence.

The ground upon the American side which will probably be destroyed by these works is of recent formation, subject to overflow, and but sparsely occupied where occupied at all. Its money value is not great.

I am required to give my opinion as to whether these works are in conformity with those permitted by the third article of the convention of November 12, 1884. That article reads as follows:

“No artificial change in the navigable course of the river by building jetties, piers, or obstructions which may tend to deflect the current or produce deposits of alluvium, or by dredging to deepen another than the original channel under the treaty when there is more than one channel, or by cutting water-ways to shorten the navigable distance, shall be permitted to affect or alter the dividing line as determined by the aforesaid commissions in 1852, or as determined by Article I hereof and under the reservation therein contained; but the protection of the banks on either side from erosion by revetments of stone or other material not unduly projecting into the current of the river shall not be deemed an artificial change.”

The article does not appear to prescribe what character of works shall or shall not be constructed in the river, but it defines what the effect of different classes of works shall be upon the location of the boundary. If the course of the river be changed by jetties, piers, etc., or by revetments which project unduly into the river, the boundary between the two countries does not move with the channel, but remains where it was before.

There are various methods in common use among engineers for the protection of river banks. One of them is the continuous revetment, which consists in covering the face of the bank with a material, such as brush and stone, which can not be washed away by running water. It conforms to the natural outline of the shore, projecting uniformly and slightly, and is strictly defensive in its character. It is regarded by American engineers as the surest, and therefore in the end the cheapest, means of defence, and is in general use upon the largest American rivers. Another method is in the use of the ordinary spur, in which at intervals along the shore to be protected, short dikes, called spurs, are run out at right angles to the shore or slightly inclined up or down stream. They are designed to hold the bank where they find it, but not to advance it. They also are defensive, though less strictly so than the former, but they are uncertain in their action. Another method is in the use of deflecting dikes, designed to turn the current entirely away from the shore, and, in silt-bearing streams, to push the latter forward. They are aggressive in character. The works under consideration belong to the latter class. They have been adopted upon the ground of supposed economy. They do, in my judgment, project unduly into the current of the river.

I was unable to carry out the spirit of the third paragraph of your instructions, which contemplated an exchange of views between myself and a Mexican engineer not heretofore associated with the works, for the reason that the engineer appointed by the Mexican Government to confer with me was Mr. Garfias himself, who projected the works and is now in charge of their construction. This gentleman was thoroughly committed in his official reports and public utterances to the ideas that his works would not injure the left bank, and that if they did, it was still his right to build them. I did not consider it necessary or desirable to enter into any discussions with him, further than were required to elucidate his views and to procure the information I was seeking. He was very courteous and obliging, going with me through mud and rain to examine his works, showing me his drawings, freely making explanations, answering all questions, and allowing me to take into my service his draughtsman for making copies of his maps.

Viewed from a Mexican standpoint, his work is a credit to him, showing much ingenuity and fertility of resource. There was associated with him as his subordinate Mr. Felipe Zavalza, who had been appointed by the Mexican Government to aid in the conference. This gentleman is not at present connected with the works, but he formerly was, having built the first two dikes.

Mr. Beckford Mackey, United States consul at Juarez, was present at all the conferences between the Mexican engineers and myself, and, with his knowledge of the Spanish language, was of much assistance in the mutual understanding of what was said. He made the translation of Mr. Garfias’ project, which is hereto appended, and he was of material service in bringing me into communication with the citizens of El Paso, who might be able to furnish information. He cheerfully rendered me all the assistance within his power.

I was unable to procure from American sources any recent maps which I considered [Page 627] as valuable for the present purpose as the Mexican map of 1885. I held interviews with Mr. F. Ashton, the present city engineer of El Paso, and with Mr. G. N. Marshall, his predecessor, and with other citizens; but they did not know of any maps except those which had been made to exhibit the boundaries of land owners. These maps indicated the location of the river in a general way, but they showed only the American side, and that, as it appeared to me, only approximately, as would be natural with a land map in the case of a boundary which is changeable.

The general result of my investigation is that the Mexican Government has constructed and is constructing works which project unduly into the current of the river, to the injury of American citizens owning the left bank; an injury not as yet great, but which may become serious. The injury to be expected from those already constructed may be largely diminished by leaving them without repairs and allowing them to suffer the deterioration to which all such works are subject. Even thus they will protect a large part, if not the whole, of the right bank. Should portions of the hank remain unsheltered by them, the money which would otherwise be expended in repairs might be directly applied to such unsheltered parts. The works planned but not yet constructed, viz, Dikes 6 and 7, promise to do more harm than the others, and in the present condition of affairs, which is so different from that when they were planned, they should be abandoned, and recourse had to a direct revetment for this portion of the bank.

It may be pertinent to add that the changes which have occurred in the Rio Grande at El Paso are perfectly natural to an alluvial stream, and that similar changes have doubtless occurred throughout its course below. It is highly probable that at the date of the convention of 1884 there were very few points in common between the location of the river at that time and its location at the time of the survey of 1855. To hold that the convention of 1884 was a new agreement, not retroactive, and that the boundary of that time was where it was in 1855, without reference to changes in the river which might have occurred prior to 1884, is to make the convention meaningless. The boundary of 1855 was in 1884 upon dry land, except at isolated points where it crossed the river. To make it the subject of stipulations which could refer only to a river would not mean anything.

My account of disbursements and mileage will be forwarded with a separate communication hereafter.

Very respectfully, your obedient servant,

O. H. Ernst,
Major of Engineers.

Office of the Chief of Engineers,
United States Army,
Washington, D. C., December 17, 1888.

Sir: I have the honor to transmit herewith a report, just received, from Major O. H. Ernst, Corps of Engineers, U. S. Army, addressed to the honorable the Secretary of State, in the matter of the works of construction by the Mexican Government for the protection of the right bank of the Rio Grande at El Paso, Texas.

As the report is made in compliance with instructions from the Department of State (with the sanction of the War Department), it is forwarded at once.

Very respectfully, your obedient servant,

Thos. Lincoln Casey,
Brig. Gen., Chief of Engineers.
Hon. William C. Endicott,
Secretary of War.

[Inclosure 2.]

Project of works of defense in the Bio Bravo in front of the town of Paso Del Norte. Dated March 25, 1886.

[Translation.]

Galveston, Texas, December 12, 1888.

Respectfully forwarded to the honorable Secretary of State, with letter of this date.

O. H. Ernst,
Major of Engineers, U. S. Army.

state of the river and form of its bed in 1848.

At that point in the west at which the river commences to be the line of boundary its bed is well defined, flowing as it does through a narrow valley and encased for a considerable distance between the mountains which form the great range known as the Sierra Madre Oriental.

[Page 628]

When the river reaches the point at which is placed the dam for the irrigation of the right bank the valley widens, the flat country begins, and the bed becomes variable. The course of the river for the whole of the space described, and especially between the dam mentioned and the town of Paso del Norte, is from northwest to southeast, but precisely in front of the town it changes to the east, forming a large curve, the concavity of which extends toward the south.

The maps surveyed (?) by the joint commission which marked the line of boundary, and which, according to Article V of the treaty, “form part of the same treaty and have the same force as though they were inserted therein,” exhibit the condition of the river-bed at that epoch, the form of its curve, and above all its location, which, as will be seen further on, is most important in order to establish the right of Mexico to recover the land which has been lost by the aggressions of current employing such means as should be adequate for the purpose.

The accompanying tracing has been taken from the original on file at the department of public works.

damage suffered by the mexican side.

The natural form of rivers presents always the same phenomena. The conditions of the regions which rivers traverse make these phenomena more or less characteristic.

The modifications of a river’s bed depend upon the class of land through which it flows and upon the velocity of the current. Near the source of a river the stream is rocky, narrow, abrupt, irregular, and generally precipitate as a torrent; its bed is fixed and invariable because the obstructions which the current encounters are of a permanent character, formed by the inequalities of the soil.

Lower down the bed of the stream becomes more tortuous and truly serpentine, as it crosses broad valleys and forms great curves, which seem to recognize no law, as the slightest obstruction, or at times no known cause, will make the river leave its channel and force a new bed entirely different in its location and development.

In such places bars and shores are formed, the apparition of which phenomena follows laws perfectly understood.

When a river leaves the higher regions from which it takes its source the increase in volume caused by freshets may deepen the channel and leave its banks unchanged. The matter held in suspension by the waters and that torn away from its bed or banks will not be deposited except in those places where the force of the current has so diminished as to be unable to bear them away.

At the beginning of its course not only does the river bear along the material dislodged from its bed and sides, but by the irresistible operation of the water upon the fragments of rock and the rotation and shocks to which these fragments are subjected they become round and smooth and what are known as “cantos rodadas” (bowlders).

Passing from the mountains to the level regions accessions to the volume of the water enlarge the channel, not in depth but in width; because, as the force of the upper and middle part of the current is greater than at the bottom, and the banks of the river in the valleys and plains more spacious and yielding, they give way more easily than the soil which forms the bottom of the river’s bed.

In support of this theory, which is perfectly corroborated by experience, Professor Molesworth makes use of the following expressions:

“Representing V as the velocity of the surface, that of the middle will be equal to a (n+05)—VV, and at the bottom equal to a(V+1)—2 VV; from which it results that 4, for example, being the velocity of the surface, that of the middle will be 2.5 and the velocity of the bottom of the stream 1.”

So, now, if the dams must yield to every increase in the volume of water, let us see which of them will be most likely to suffer.

Every body in movement unless meeting resistance follows a straight line. The body in movement which does not follow the straight line must have met some resistance which causes it to pursue a different direction. This rule holds with rivers, which tend always to follow a straight line, but when encountering resistance change their course and wear away portions of their banks. In the curves of rivers is produced this effect, which science calls “avulsion.”

On the side opposite, that is to say, in the convex part of the curve, from the same tendency of the river to follow a straight line it will follow the tangent of the curve for a distance more or less great, dependent upon the breadth of the river and its velocity; and consequently the space embraced between the tangent and the curve from the point of tangency until where this change of course occurs is occupied by still water, where is deposited the matter held in suspension and where is formed what is technically called “alluvion.” These two phenomena just described are directly contrary, and as the convex side of the curve gains what is lost by the concave, and this evolution is constant and natural, unless the hand of man should interpose a limit the river would continue its advance, adding land each day to the convex side and each [Page 629] day carrying away territory on the other. This is exactly what has occurred in the case now under consideration.

The curve of the river in front of Paso del Norte presents its convex side towards the Mexican bank, and from a comparison of the surveys of the commission of 1852 with one made by me, it results that the river has changed its course towards the south 1,735 meters. Without the necessity for the inspection of maps the aspect of the soil on both sides of the river proves the invasion of the right bank; for this side is covered with vegetable earth, which waters never deposit, and the left bank displays only sand and that character of soil produced only by alluvion. I think, therefore, that I have sufficiently demonstrated the loss of territory suffered by the Mexican side.

legal part.

Let it now be seen whether the Mexican Government possesses the right to construct works which may prevent the continuance of this damage and re-establish the bed of the river in the state it held when the united commissioners surveyed the line of boundary.

The question is perfectly clear and simple. No necessity exists for recurring to an examination of doctrines in regard to alluvion and avulsion, for the existence of a formal compact nullifies all else.

The treaty of peace, friendship, limits, and definitive settlement between the Mexican Republic and the United States of America, signed February 2, 1848, at Guadalupe Hidalgo, and ratified at Queretaro, May 26 of the same year, says in Article V, part 3: “In order to designate the boundary line with due precision upon authoritative maps, and to establish upon the ground landmarks which shall show the limits of both Republics as described in the present article the two Governments shall each appoint a commissioner and a surveyor, who, before the expiration of one year from the date of the exchange of ratifications of this treaty, shall meet at the port of San Diego, and proceed to run and mark the said boundary in its whole course to the mouth of the Rio Bravo del Norte.

“They shall keep journals and make out plans of their operations; and the result agreed upon by them shall be deemed a part of this treaty, and shall have the same force as if it were inserted therein. The two Governments shall amicably agree regarding what may be necessary for these persons, and also as to their respective escorts, should such be necessary.” And in part 4 of the same, Article V: “The boundary line established by this article shall be religiously respected by each of the two Republics, and no change shall ever be made therein except by the express and free consent of both nations, lawfully given by the General Government of each, in conformity with its own constitution.” And in Article VII of said treaty it is agreed that “The River Gila and the part of the Rio Bravo del Norte lying below the southern boundary of New Mexico, being, agreeably to the fifth article, divided in the middle between the two Republics, the navigation of the Gila and the Bravo below said boundary shall be free and common to the vessels and citizens of both countries, and neither shall construct any work that may impede or interrupt, in whole or in part, the exercise of this right; not even for the purpose of favoring new methods of navigation.”

From these conventions it is seen that the course of the river as shown on maps of that epoch shall, be the line of boundary and that “no change shall ever be made therein;” so that the boundary line is not the course of the river with the changes which it has had or may have, but the line which the river occupied as shown by the surveys of the joint commission on limits, and consequently the land which is to-day on the left bank of the river, but to the south of the line marked by the maps of the commission, belongs to Mexico, and are subject to the jurisdiction of the Mexican Government.

With respect to works on the river the Mexican Government can establish such as may be necessary to restore the line of boundary to its legal position, with the sole limitation that they “may not interrupt or impede in whole or in part the exercise of the right of navigation.”

class of works of defense proposed.

Established the necessity and right to construct works to defend the right bank of the river, it remains to select the class of works which, in consideration of the circumstances, will be most expedient.

In view of the works constructed by me under similar circumstances at the port of Matamoros, I have formed the opinion that in this case it will only be necessary to modify my plans in one detail, on account of the force of the current, which is greater here than at Matamoros.

All works of defense for river banks, with the exception of these spurs, are based on the system of facings, and in consequence must be of considerable extent and [Page 630] therefore greater in cost than spurs which occupy only a sixth part of the surf acre which they are expected to protect, it being known that a spur of this kind protects an extent of territory five times longer than its face.

The only inconvenience sometimes presented by these works is that it is not always possible to so select their location and direction that the opposite bank of the river may not be injured.

In such a case it is necessary to abandon the system.

In the case under consideration no such difficulty exists, for such a location may be chosen and such directions and dimensions given that the deflection of the current will not reach the opposite bank.

If, then, the only objection to the construction of these works in the present instance disappears and it is beyond doubt that spurs of this description defend the same extent of territory at a smaller cost and with less labor, there is no reason for hesitating to adopt this system.

Let us see now the class of material which must be used in their construction.

The piles system gives good results when the force of the current does not pass a certain limit, as at Matamoros, where the works constructed by me resulted satisfactorily; but if the current is sufficiently strong to cause the piles to oscillate, this, aided by the floating tendency and shocks received from matter carried down the stream, loosen their fastenings, and the connection between the stakes once destroyed the works go to ruin.

I do not know the velocity of the current at Paso del Norte, but it is sufficient to know that the pile bridges of the railways here have not been able to resist the current during high water, and that Paso del Norte is much higher up than Matamoros, to presume that works constructed on the pile system will not give good results and should not be adopted while there is another system exempt from these inconveniences. The mattress system has been known from time immemorial and was used and is still used in several of the countries of Europe as a simple facing of the banks, usually in one layer only.

The genius of Captain Eads has worked a genuine revolution in this respect, and at present in the United States and in Holland no works for the rectification of a river’s channel, the defense of its banks, or the improvement of a bar are constructed on any other system.

The new system consists in forming a dam of proper dimensions by means of superposed layers of willow or other branches; each mattress so constructed being sufficiently ballasted to secure immersion. The mattresses consist of bundles of branches, and their length varies according to the importance of the work, but in general is never less than 100 feet.

The diameter of the bundles from which the mattresses are formed can not be fixed, but is usually from 4 to 6 inches. The bundles are so united as to form the mattresses by means of flexible reeds or vines or with tarred ship-cord.

The mattress is laid on an inclined plane of beams placed at such distance as permits the tying or weaving of the bundles, and at the bank of the river in order that the mattress may be easily floated. The mattress is then drawn to the place where it is to be situated, and once, in place is ballasted and sunk.

The mattresses are placed one upon the other, and diminish in breadth to give the necessary slope.

The dike is built to the highest point which the waters of the river may attain during the freshets.

This is the system which Captain Eads employed and which he called “mattresses of fascines.”

In Holland these mattresses are differently constructed, and two kinds of work of this description are used, one for simple revetments, and one for the formation of dams. The first are called “kizzen-bedden,” the second “zink-stukken.” To the first is given greater longitudinal strength, to the second greater transverse resistance.

A frame of crib-work is made, which commences with a transverse series of branches tied or twisted in bundles and placed at a distance from each other of 1 meter or nine-tenths of a meter. Over this is laid a longitudinal series at a distance which is never less than nine-tenths of a meter for the “zink-stukken.” In order that the longitudinal strength may be greater and may better resist the force of the current the bundles of branches are made of the whole length of the work, if this should not be greater than 150 or 200 meters, and never less than these dimensions.

The diameter of the bundles is from 4 to 6 inches. Where the bundles cross each other they are fastened together by means of flexible reeds or vines, except at the points of the perimeter, in which places use is made of unraveled ship-cable, by which the mattresses are fastened to stakes, and the upper layers united with the lower after these are filled.

The openings in the crib-work are filled with twigs tied together and placed vertically, the height of the two bundles which form the layer.

[Page 631]

Another layer is then superposed so that the openings coincide, and these are filled in the same manner, the layers being united with ship-rope as described.

Three layers form the mattress, which will be from forty-hundredths of a meter to a meter in thickness. A greater diameter is not desirable, for the reason that the mattress then assumes too great rigidity and does not adapt itself to the form of the river’s bed.

These mattresses are submerged by ballasting of stones of the proper weight and dimensions. In order that the ballast may be kept in position, small stakes are placed in all the openings of intersection, and at each fourth stake is placed one which is called the “stake of anchorage” (estaca de anclage), which penetrates the river’s bed and prevents the movement of the system.

These stakes must be strong, but thin, so that the floating tendency will be as slight as possible. The ballast should be in the proportion of eighteen-hundredths of a cubic meter of stone or gravel to a square meter for the center, and twenty-six hundredths of a ton of stone to the square meter for the perimeter.

This is the system which I believe advisable to adopt for the works now under consideration.

survey.

Having adopted the system and class of works to be constructed, let us pass to the location and dimensions of said works, which require more exact data.

This data I began to collect immediately upon my arrival at Paso del Norte, and with more or less success. I encountered no obstacles in the topography.

Provided with an excellent theodolite, I quickly made a detailed plan of the river and its relative location.

The soundings also offered but little difficulty, for, as the river was found almost dry, the transverse sections which I believed would be necessary became rather profiles taken on terra firma.

The examination of the river’s bed and subsoil gave a like result.

Unfortunately I can not say the same in regard to the other data relative to the regimen, the lowest point of the water during the dry season, the freshets and middle stage of the waters, the velocity of the current, its direction, the points at which the matter carried along or held in suspension strikes the banks, the discharge, and other particulars which result from the above.

My stay in Paso del Norte has been during the months of October and November last, when the rains had ceased more than a month. During this season the rain-fall had been slight and freshets produced had disappeared; and the season of the breaking up of the snow and ice not having arrived, I found the river in what may be termed at low-water state; that is to say, almost entirely dry. This will be shown by the transverse sections which appear on the plans herewith annexed.

This was the first occasion of my visiting this locality on a scientific mission, and consequently I lacked the data proper for determining the elements necessary to base my calculations. 1 inquired on both the American and Mexican sides whether any observations had been made of the low and high stages of the waters, its volume, the force of the current, etc., and I found that no such data had been obtained.

On the Mexican side, which is the side injured, and consequently where a very natural interest should exist in collecting the information referred to, no person has been found of sufficient knowledge, united with the zeal and foresight, necessary to make the observations required, which demand time, constancy, and expense not inconsiderable from its duration and continuance.

As on the American side, far from having experienced any damage, the land has been added to day by day by alluvion; no attention has been given to changes which could not affect them injuriously; and no such observations had been made, which under adverse circumstances would have been carefully registered, as occurred at the military post near Matamoros.

In view of these circumstances, and reserving the completion of my observations for a favorable epoch, I have resolved, time and circumstances permitting, to proceed by deductions founded on the theories of the most reputable authors and the experience acquired for many years of study of the phenomena presented by rivers and the various works intended to regulate their regimen, protect their sides, and prevent the formation of banks and bars.

To deduce the regimen of this river in the locality referred to, I availed myself of the following theories and facts:

As is known, the regimen of a river is the relation which exists between the transverse section, the longitudinal slope, the nature of the bed, and the volume of water It is said that this regimen is fixed when these relations are constant, or change insensibly at long intervals of time. This regimen is, on the contrary, variable when the relations expressed are subject to changes which recognize various causes. It is sufficient to glance at the accompanying map to be convinced of the irregularity and inequality of the transverse sections, and if this map is compared with those of anterior [Page 632] epochs, it will be seen how much the course of the river has changed, invading a great extent of land on the south, and widening in all that part which remains in front of the town of Paso del Norte. Now, if the transverse sections are so variable, and if the course of the river changes so easily, augmenting the development of the line, and consequently modifying the longitudinal slope, the regimen of the river can not be fixed at this place. The volume of water also is very variable.

Not only is this manifested by facts in a passage of time sufficiently great, but this may also be deduced from the relation which exists between the volume, the velocity of the current, and the section; for these last being variable, the volume will also vary.

The velocity may be deduced approximately from the nature of the matter deposited in the river’s bed by the following table:

A current of one-quarter foot per second carries fine clay.
A current of one-half foot per second carries fine sand.
A current of three-quarters foot per second carries coarse sand.
A current of 1 foot per second carries fine gravel.
A current of 2 feet per second carries coarse gravel.
A current of 3 feet per second carries a stone as large as an egg.
A current of 5 feet per second carries a stone as large as 2 feet square.

In the bed of this river are found all these materials, which shows that the lighter matter has been deposited in the dry season when the velocity of the current is reduced to the minimum; but as large stones are also found in the bed of the river it may be deduced that during the freshets the velocity of the current is sufficiently great to carry this matter, and consequently is at least that of the maximum in the, above table. This datum is what has made me reject the pile system and adopt the system of fascines, which offers greater resistance and at less cost, at least in the locality now treated of.

location.

Having selected the system of works and the materials to be used let us proceed to fix upon the location and dimensions to be given to each of the dams in order that the desired effect may be obtained.

Assuming, according to the established theory, that when water encounters a convex curve it follows the tangent of said surface, this natural effect should be availed of in such a manner that it may be produced in the whole extent which it is sought to defend.

It is known that a spur protects an extent of land five times greater than the length of its face. Now, in order that the opposite bank may not be injured, in spite of the justice of the title which I have shown to exist in Mexico, I have fixed upon the most pronounced convex curves and given to each spur such length and direction that they will not carry the current to the left bank, but will protect our own side, which has been worn away and must continue to be worn away unless defended.

Between the irrigating dam and the first work I did not think it necessary to do anything, because there the course of the river is almost straight, possessing only a slight curvature which presents its concave side rather to the left bank.

The first work, as marked on the plan, will have 100 meters face of deviation and 50 of counter-force (return face?), with a breadth of 10 meters, and a height which according to the depth of the bed but attains the altitude of the river’s bank which supports it.

The second work will be 75 meters in length, and have 50 meters of counter-force (return face?), it will also have 10 meters of breadth and a corresponding height. The third will have respective dimensions of 80, 20, and 10 meters.

The fourth, 75, 75, and 10 meters; the fifth, 95, 85, and 10; the sixth, 95, 55, and 10; the seventh, 200 meters, 70, and 10; the eighth, 150, 130, and 10.

The ninth must be a simple dike, without counter-force (return face?), supported in its lowest part by the bank, or natural island, marked on the map and commencing above at half the distance defended by the last spur.

This work might be left isolated because the current will not strike it except from the third part formed.

The construction of this dike appears to me of the greatest importance, although I must remark that it will remain in the river beyond the middle of the bed at present; and although the part treated of is no determined channel, and the bed of the river where it now flows is not the line of boundary, I believe it my duty to submit to the high consideration of the Department whether I may adopt the location indicated.

The dimensions of this work must be 600 meters by 10. Its greatest height will be 3 meters and its medium 2.

The order in which the works should be constructed is that which I have mentioned. But, above all, the last should not be constructed until the next has produced its [Page 633] effects, forming a bank on the right side and causing to disappear the movable sand on which nothing can be built without danger of the work being undermined and losing its stability.

estimate.

From the anterior data it results that there are required 2,000 longitudinal meters of mattresses, with a breadth of 10 meters, and a height of 1 meter 80 centimeters; or a total of 60 mattresses of a 100 meters by 10 and of 60 centimeters.

For greater clearness, I will calculate the cost of one mattress of these dimensions:

The branches for the fascines are purchased in Paso del Norte, in cart-loads which average a ton and a half, and cost $2 delivered at the works. The stone costs $1 a ton; the gravel 75 cents per cubic meter; the wood costs $2 per thousand square feet of 1 inch in thickness.

The wages of the laborers will be from 75 cents to $1 per diem.

The average weight of branches necessary for a mattress is 240 tons, or 160 cartloads, at 2$.

160 cart-loads, at $2	$320.00
144.90 cubic meters of gravel, at 75 cents	108.67
44.5 tons of stone, at $1	44.50
164 feet of wood for 109 frames, at 3 by 3 by 2, $3	3.28
82 feet of wood for stakes	64.00
5 pounds nails, at 10 cents	.50
8,100 yards tarred hemp cord	4.00
50 spikes, at 12½ cents	6.25
50 yards thick cord	6.25
1,250 yards medium	12.50
Labor on frames	7.50
Labor on 222 fascines or bundles	55.50
Labor on 11 fascines, at $2.50	27.50
20 beams, 1,312 square feet	26.24
Labor on inclined plane	20.00
Placing, binding, and formation of a mattress, at 50 cents square meter	500.00
Floating into position of each mattress	40.00
Ballasting mattresses, 25 cents a ton	38.25
Total	1,222.58
For 60 mattresses	73,354.80
Contingent expenses	7,335.48
Total	80,690.28

The time required for the construction of the work will depend upon the sum assigned each week for the purchase of materials and payment of labor, and may be reduced by beginning at the same time construction on all except the last, which should not be erected until a year later than the others.

March 25, 1886.

Y. Garfias.

Consulate of the United States,
Paso del Norte, Mexico, December 4, 1888.

I hereby certify that the accompanying translation was made by me, and after being submitted to Mr. Ygnacio Garfias declared by him to be true and correct.

Given under my hand and seal of office the day and year above written.

[seal.]

Beckford Mackey,
United Stales Consul.

[Inclosure 3.]

minutes of proceedings.

[Seal.]

I.

In the city of Juarez, on the 26th day of November, 1888, in the office of the director of the hydraulic constructions for the Rio Grande at (Paso del Norte) said city, a meeting was held by and between O. H. Ernst, major of the United States Engineer [Page 634] Corps, and special envoy of the American Government; Beckford Mackey, consul of the United States at said city; Felipe Zavalza, engineer, and the directing engineer of the works, Ygnacio Garfias, as a result of the protest made by the city authorities of El Paso, Tex., against the construction of said works, and these persons proceeded to hold a conference to elucidate the probable effects of said works; and Engineer Garfias having been called upon to do so, as the author of the project, and as builder of the works, began his explanation by exhibiting the plans of the river as made in 1852 by Commissioners Emory and Salazar Harregui, and a map made by himself in 1885 for the purpose of formulating the project. He called attention to the difference in the lines occupied by the channel in the two maps, a difference which is wholly against Mexico in front of the city in its central portion and most populated part, making the comparison by a reference to the church, which appears in both plans. He called attention to the fact that not only had Mexico lost a very considerable part of cultivated and irrigable lands and some dwelling houses, but also one of the irrigating canals, known as the Chamizal ditch, the loss of which constituted a greater damage because it ruined and converted into arid lands a considerable amount of ground formerly used for viticulture and the cultivation of choice fruits; that these damages were caused in the beginning by the natural effect of the water, which in this part of its channel attacks the right bank as it makes a big curve, to the detriment of the concave part, which is on the right-hand side, and partly caused by small wing-dams constructed for defense on the left side, which helped powerfully to increase the destruction which already without them had been considerable. Afterwards the embankments of the Mexican Central and street-car railroads were built in an oblique direction to the channel, forming an obtuse angle with the axis of the current, whereby the water in meeting these dikes deviated almost normally from its true current, producing decretion on the opposite bank every day of greater importance.

These facts, which are notorious on both sides of the river, caused the Mexican Government to decide to defend its territory, where already the only remaining canal for irrigation was threatened; wherefore Engineer Garfias was appointed to make plans for defensive works. During the months of October and November of 1–85 he made his preliminary topographical and hydrometric studies, and presented the project in March, 1886.

The project includes eight wing-dams, whose location and angle of deviation are so calculated that the current shall occupy the center of the channel, so that the parabolic curve described by flowing fluids shall in no case reach the opposite bank.

He explained the theory of wing-dams as consisting in the claim that they defend a distance equal to five times the length of their face of deviation when the departing curve (whose tangent defines the location of the wing-dam) follows the channel for that distance, increasing or diminishing the distance defended according to the curve of the channel to the right or left.

According to this theory the projected wing-dams can not cause the current to reach the opposite side, and if this happens it may be considered one of the many caprices of the Rio Grande, but in no way caused directly by the wing-dams.

After this those present went to examine the works already constructed, beginning with No. 5, which was the nearest.

Engineer Garfias described the method of construction, gave the date of the completion of this work, which was in the early part of the current month, and explained that as there had been no rise in the river during its construction, no effect had been produced.

They then proceeded to work No. 4, and the same engineer, after giving the date of its completion, explained that this work, in spite of having faced the rises in the river for the past two years since its completion, had not caused any damage to the opposite side. The same was shown of work No. 3, and the examining party proceeded to work No. 1. This work was concluded before the rises of the present year, and had therefore produced its defensive effects. The low-water season permitted the view of the formation of the special circumstances of the form of the channel and justified the exactness of the theory, as the said bar preserved almost exactly the form of a parabolic curve, which, commencing at the extremity of this wing-dam, continued until it intercepted nearly the middle of the next work. He explained here also that no effect detrimental to the opposite side had been produced, though this work had faced all the rises of the present year. Thence they proceeded to work No. 2. This was also completed before the rise of the current year, and has not caused any damage to the other side, its effects being defensive in a greater degree, because the channel has a cnrvation towards the right.

Hence they visited work No. 6, the cause of all these proceedings. Arise of 3 inches of water two days previous had filled a channel which had been dry on account of low waters during the days of the protest. Engineer Garfias explained this circumstance, which would completely set aside any ground for the protest of the authorities of El Paso.

[Page 635]

He called the attention of Engineer Ernst to the inexactness of the telegram written by Attorney Caldwell, there being nothing true therein except the fact that the work crossed the small current, which in those days was the only one; but he added that even in such a case the work could not be considered as a dam, because the water passed freely through the fascine cribs that formed it, as was manifest at the examination. He explained other data which are not stated herein, because they appear in the plans, tracings of which Engineer Ernst will receive, showing the extent of the channel occupied by the current, the location of the work, the breadth of the true and proper channel, etc.

To conclude, the party passed on to visit the place designed for the location of work No. 8, of which nothing appears in the river but where they are constructing various fascines, which are accumulating, to be put in place when an opportunity is offered.

Thus terminated the first conference, with the understanding that the second should be held on the 28th, as Engineer Ernst intended to devote the 27th to private interviews with the engineers of El Paso just as he had previously given his time to Engineers Garfias and Zavalza.

On the 28th of the same month the conference continued. The minutes of the 26th were read, and having been translated into English by Consul Mackey, Mr. Ernst desired to have it understood that he manifested neither his agreement nor disagreement with the propositions of Mr. Garfias, some of which were expressed in these minutes in a manner to imply his indorsement, and that making this reservation and extending the minutes in both languages he would sign them upon the termination of the conference. He added that in his opinion the works would necessarily cause the erosion of the left bank, because they were an obstacle to the free course of the water and they confined the channel, although theoretically and under the hypothesis that the left bank would resist it might not happen, because the section would be equalized by additional depth; but he believed the bank would not resist. Furthermore, that as the flow in the river was so variable, having a dry channel at times, then again its waters overflowing its banks, there was no fixed regimen, and it was difficult to calculate the effects of the work with such uncertain data.

Engineer Garfias answered that the theory was a technical reason as against a simple hypothesis; that, besides, the facts justified till now the exactness of the theory, as two of the existing works had produced their effects during two years and two others had done so for one year without causing damage to the left side, and that the effect of the works was calculated for the highest water, for the reason that the current was so variable. That, furthermore, it was not to be supposed that he had projected works that would throw the water against the left bank, because by the theory of the angles of incidence and reflexion the water would return to attack the right side and cause destruction.

Mr. Ernst asked why they had not adopted the system of continuous revetment, to which Engineer Garfias answered, because it was very expensive and the wing-dam system produced the same result for one-sixth of the expense.

Engineer Zavalza said that at the time of a rise both banks were attacked, and that if one was defended and the other not, the latter would suffer somewhat, but not as a direct effect of the works.

Mr. Juan S. Hart was proposed as translator. Mr. Mackey observed that the gentleman might act as such as a special favor, as he did not usually act in that capacity.

The four members of the conference then called upon Mr. Hart, who willingly offered his services, keeping in his possession the minutes already written in order to begin their translation.

Engineer Ernst said he had ordered photographs taken of the works completed, and that before leaving the city he would present a collection of them to Engineer Garfias.

On the 29th of the same month the conference continued. The minutes of the preceding day were read, and after Mr. Ernst received the copies of the plans and the project translated by Mr. Mackey, the conference was declared closed, during the whole of which the English language was spoken, as being the only one understood by all who signed.

Before signing Mr. Garfias added that he desired to state that the proposed second visit to the works had not taken place, not only on account of the bad weather but also because Major Ernst admitted that it was probable that among the earlier effects of the work would be the formation of a bar on the opposite side of the river, but well below the work, and, Major Ernst added, immediately opposite the work there would be erosion.

He thought that later on this opposite erosion might work down stream, but he believed the ultimate results of such works to be uncertain. To this Mr. Garfias replied that, even admitting that this erosion might occur, compensation would exist between the territory lost and that gained and even supposing the loss of the left bank caused by said erosion greater than would be gained by the resulting, accretion, this would be grounds on which the owners of this property might found a claim for indemnity, [Page 636] but not for the suspension of these works, which have for their object the restraining of the destructive effects of the current on the Mexican side, where, from the form of curve, not only cultivated and habitable lands are threatened with destruction, but also the only remaining canal of irrigation, which once destroyed would completely kill the agriculture of the place, on which its existence depends.

He also said that it should be held in mind that the land on the left bank which might be exposed to injury had been separated from the Mexican territory by the action of the waters, and that should the line of jurisdiction or even boundary be declared changed, this would not affect the right of property of individuals, and that the title to these lands remained in their ancient proprietors, who are Mexicans now living on the right bank of the river, and who have neither claimed or asked anything for the damage which might be inflicted upon their properties left by the river on its left bank.

With this the conference terminated, and the minutes were signed in quintuplicate, two copies in Spanish and one in English remaining in the possession of Mr. Garfias. One in each language to be sent to the Department of Public Works, and one to be retained by Mr. Garfias. A copy in each language was delivered to Major Ernst.

During the last moments of the conference Major Ernst expressed a desire to modify one paragraph of the minutes, but as Major Ernst desired to leave for Galveston in a few moments no time remained for making the desired correction; it was therefore agreed that the quintuplicate copies of the minutes should be forwarded to Galveston for his signature, those intended for the Mexican commissioners to be returned by mail.

O. H. Ernst,
Major of Engineers, U. S. Army.
Beckford Mackey,
U. S. Consul.
Y. Garfias,
Felipe Zavalza.