SCIENCE. 



[Vol. XI. No. 264 



the work recommended, and still more as to the results to be ex- 

 pected of them. In such a state of aflfairs, the work could properly 

 be continued ; since the concurrent opinion of all as to the work to 

 be done would probably be right, though of the discordant reasons 

 and diverse expectations of individuals, some, of course, must be 

 wrong. 



At the beginning of the session above referred to, a strong and 

 determined opposition developed to one of the principal features of 

 the work proposed. It was stated that the revetment of caving 

 banks was unnecessary and wasteful, since the object of that work 

 the cessation of caving, would result from the contraction-works, 

 or those designed to concentrate the water over the shoals. 



Some of the most active champions of the improvement and of 

 the commission by which it was being carried on, announced the 

 discovery that this commission had effected an insidious and 

 dangerous change in its original plan, by virtue of which the ob- 

 jectionable feature of bank-revetment had been introduced or made 

 more prominent. It is without the present line of argument to 

 contest this statement, several refutations of which have appeared 

 in print. The issue is now squarely made between revetment and 

 anti-revetment theories, and must be met on its merits. It makes 

 no difference, in the decision of the question, whether the Missis- 

 sippi River Commission have changed front on it or not ; except, 

 perhaps, that if they have done so, it has been in the light of four 

 years' experience, which in itself would be a strong argument in 

 favor of their later views. 



The two theories are so antagonistic that compromise is scarcely 

 possible. The revetment or protection theory makes prevention of 

 caving by means of this class of work one of the prime causes of 

 the improved condition of the river : the anti-revetment theory pre- 

 dicts the cessation of caving as a result of the contraction-works. 

 The one would stop the deterioration of the channel, and cut off 

 one of the principal causes of impaired navigation, preliminary to or 

 concurrent with the effort to improve the channel : the other would 

 attempt to remove the effect, while leaving the cause in full opera- 

 tion. 



The hypothesis on which the anti-revetment theory is based is 

 very simple. It is assumed that the ability of flowing water to 

 carry suspended sediment is directly proportional to its velocity ; 

 that at any given velocity it can carry a certain normal quantity, 

 refuses more, and is not content with less ; that if undercharged it 

 takes up the deficit from the adjacent bed, producing scour and 

 caving ; if overcharged it drops the surplus on the spot, causing a 

 fill. A corollary of this is, that if the channel can be so regulated 

 that the velocity will be uniform throughout its whole extent, and 

 always bear the proper ratio to the supply of sediment, there will 

 be no scour or fill in the bed or on the banks of the stream ; the 

 sediment supplied by tributaries will be carried without loss or gain 

 to the sea ; the deterioration of the channel will cease ; the bars, 

 having been removed, cannot re-form ; and the problem is solved. 

 As evidenced by their practice, engineers are overwhelmingly in 

 favor of ideas the reverse of these. Many civil engineers have ad- 

 dressed themselves to the problem in the interest of individuals or 

 corporations. No case is known where any of them have proposed 

 any remedy for a caving bank, except a direct protection of some 

 kind. When called upon to induce a scour along a bank, as in 

 some cases of important landings, they have, strangely enough, 

 successfully employed for that purpose the very means now pro- 

 posed to prevent the same thing. 



Such proponderance of professional opinion would be accepted 

 in any question of law or medicine. In matters of engineering, 

 however, the public demand not only that we shall be agreed, but 

 that they shall be convinced. It is therefore necessary to further 

 argue the question, and for the same reason the argument must be 

 addressed, not to the profession alone, but to the public as well. 



It is not proposed to test the hypothesis by any of the well-as- 

 certained facts of the river's regimen. Thus no argument will be 

 founded on the facts that by hypothesis the proportion of sediment 

 should increase from the bottom to the surface, as does the 

 velocity, whereas the increase is from surface to bottom ; that the 

 sediment should be greatest at the thread of swiftest water, and 

 diminish toward either bank, as does the velocity, while, on the con- 

 trary, it is sensibly equal all the way across, and as often in excess 



on the slow as on the swift side ; that a greater proportion of sedi- 

 ment should always be carried at a higher than at a lower stage, 

 the reverse having been again and again observed ; or that the rate 

 of caving should be least when that proportion is greatest, which 

 rarely or never happens. Nor will any attempt be made to weaken 

 the force of this doctrine by pointing out any of the well-established 

 causes of caving, such as outflow of ground-water, eddies, or 

 whirls, and wash of wind or steamboat waves, which, being inde- 

 pendent of the velocity, will survive any regulation of it which may 

 be effected. 



Inquiry will be made as to the applicability of this theory to the 

 problem in consideration, or, in other words, whether the conditions 

 under which the hypothesis is claimed to be effective can be pro- 

 duced in the Mississippi River. Should it be found that these con- 

 ditions can be produced, the truth or falsity of the hypothesis could 

 be quickly decided by trial. On the other hand, should it appear 

 that the conditions precedent cannot be realized, the truth or falsity 

 of the theory is immaterial. 



The actual velocities or rates of current are, in the river's present 

 state, any thing but uniform. Their distribution may be illustrated 

 by the motion of the wheels of a cart driven over a crooked road. 

 On a straight stretch, the wheels revolve with equal velocity. If a 

 turn to the left be made, the right wheel is accelerated and the left 

 retarded, and the reverse in case of a turn to the other side. If the 

 curve be sufficiently short, the inside wheel stops ; while, if still 

 more abrupt, it must turn backward. Add to this that the top of 

 each wheel moves faster than the bottom, and the motion of the 

 water of the Mississippi and like streams is completely illustrated. 



The channel of the Mississippi River is just such a road, and the 

 relative velocity of its current at any point of its course may be 

 readily predicated from the above analogy. The depth is always 

 roughly proportional to the velocity. The highest velocity and 

 greatest depths coincide on the concave sides of the bends, corre- 

 sponding to the outside wheel on a curve, and it is here that the 

 caving banks are found. On the convex side, deposits of sediment 

 from the retarded currents are constantly being made, the accre- 

 tions nearly, though not quite, keeping pace with the recession of 

 the caving line opposite. 



A word now as to the location and operation of the contraction- 

 works, which are the means to be employed to bring about the re- 

 quired conditions. At every flood the river builds up its principal 

 shoals, so that the bottom is as high as the surface of the water at 

 the lowest stage. Low waters, such as now occur, would be im- 

 possible but for the fact that the river, in falling, cuts a channel for 

 itself through these barriers. Were these natural channels suitable 

 and sufficient for navigation, river improvement would be without 

 its strongest claim to public support. They are unsuitable by 

 reason of the uncertainty when and where they will form, and their 

 frequent tortuousness. When they are deficient, it is usually by 

 reason of a division of effort whereby two or more small channels 

 are formed by an expenditure of the work which would suffice for 

 a single one of sufficient size. 



Above and on these shoals the contraction-works are to be 

 placed. Their effect will be to localize and accelerate the natural 

 channel cutting, but not to increase the amount of energy so ex- 

 pended. The amount of material scoured from the shoal nearly or 

 quite equals each year the amount deposited on it. Otherwise the 

 river would shut itself up. If so little as one per cent of the ma- 

 terial deposited in a year on any shoal remained there permanently, 

 the shoal would be raised perceptibly each year, and, within the re- 

 corded history of the river, should have become a dam as high as 

 the banks, to turn the river out over the country. The regulated 

 river, flowing through the contraction-works, can remove from the 

 shoal each year but a small excess of material over what is deposited 

 on it, and this for a limited period only : ultimately it can not, and 

 by our hypothesis must not, carry away any excess. 



It appears from the foregoing that the aggregate amount of ma- 

 terial scoured from any shoal will not be sensibly changed by the 

 contraction-works. The amount passing through in suspension can- 

 not be affected at all: hence the total amount in suspension in the 

 bend below will not differ. That the volume of water discharged 

 will not be affected, it is scarcely necessary to argue. These two 

 quantities unchanged, their quotient, which is the degree of satura- 



