Moreover, since the jetties are about half constructed, a maximum depth between them might be expected of the half of 75 feet, or 37 feet, instead of 13, if the assertion in 4 and 5 be correct.

As a matter of fact, a contract was made for dredging between the jetties, for an amount not to exceed 200,000 cubic yards, at 30 cents per cubic yard. This contract was terminated on the 31st of December, 1885, after 85,549 cubic yards of material had been removed.

In the report of 1888, ten years later, it is stated that the jetties have been partially completed, so that the north one extends 14,327 feet seaward from Sullivan's Island, and the other 16,440 feet from Morris Island; total 30,767 feet, or nearly 6 miles.

The effects produced may be briefly stated in the language of the officer in charge :-"There has been a decline of about 25 per cent. in the rates of freight and insurance, and in marine insurance of 50 per cent. between 1878, when the jetties were begun, and the present; but these gratifying results cannot be ascribed to any improvement thus far made by the Government" (p. 974).

"The 12-foot curve has throughout made a marked seaward movement, and has preserved its form of former years remarkably. The outer 15-foot curve has been pushed seaward, so that the distance between the 15-foot water inside and outside of the bar is the same as last year. The outer slope of the bar is steeper than in 1887, and still more so than in former years. All these changes point to a general movement seaward of the material between the jetties, and a very considerable activity of the ebb-tide scour" (p. 973).

In fact, the results as shown by both the report and the chart show a decided progression of the bar without the desired lowering of the plane of tidal scour. The "thalweg" of the channel will be found to be reflected from the convex northern side of the south jetty, where spurs have been needed for protection, across the mouth or gap diagonally to the outer end of the north jetty, and having a least depth of only 10 4 feet. The least distance across the bar between the 12-foot curves within the jetties is 1,100 feet. Under the most liberal construction, therefore, the Author fails to perceive wherein this experiment has proved beneficial.

In this connection, however, it would seem important to place upon record the unexpected improvement effected by the sinking of the "stone fleet" in the channel over the Charleston bar, with the intention of blockading the harbour during the Civil War. It serves to impress the importance of controlling the sand movements, and protecting the ebb currents in their path to the sea.

This phenomenal experience is described in a Paper by Professor Julius E. Hilgard, late Superintendent of the United States Coast and Geodetic Survey.1

Professor Hilgard says:-"On the accompanying diagram is seen the 'stone fleet' sunk in the main channel, which at that time had 12 feet of water at low tide, where the figure 7 indicates the present depth. There was, moreover, another channel, making out more to the southward, with 9 feet of water, where the figure 3 indicates the present depth. The vessels were placed checker-wise, in such a manner as to impede navigation, while interfering least with the discharge of water. The effect, nevertheless, was the formation of a shoal in a short time, and the scouring out of two channels, one on each side of the obstructions, through which 12 and 14 feet can now (January 27, 1871) be carried at low-water. The increased water-way thus given to the ebb-tide, caused it to abandon the old 9-foot channel on the less direct course to deep water. We have here the total obstruction of a channel which was of considerable importance to the southward trade, by new conditions introduced at a point 4 miles distant from where the effect was produced, and we are warned how carefully all the conditions of the hydraulic system of a harbour must be investigated before undertaking to make any change in its natural conditions, lest totally unlooked-for results be produced at points not taken into consideration."

So that, instead of obstructing the entrance, this accidental barrier to the flood actually deepened the water on the bar 2 feet, and induced the ebb-currents to effect an escape in its lee, closing a channel several miles to the westward by furnishing a line of less resistance, and withdrawing the water from the former distant channel. Moreover, it cut a second channel quite as deep as the first on the opposite side of the fleet, thus creating two channels as good as or better than before, instead of the one formerly existing on the site of the fleet.

This and other instances, where single jetties have been correctly located, go far to establish the soundness of the requirements to be observed in all constructions for improving the crossing of the outer bar, whenever it is the result of littoral drift.

These requirements briefly stated are:

(1) To keep out the material carried up on the bar by the flood component.

1 Annual Report of the Board of Regents of the Smithsonian Institution, 1874, p. 221.

[THE INST. C.E. VOL. XCIX.]

U

(2) To admit as large a portion of the tidal-prism as possible. (3) To prevent the advance of the foreshore, resulting from the construction of solid jetties springing from the shore and forming groynes.

(4) To change the conditions of equilibrium of the flood and ebb currents over the same part of the bar in favour of the ebb.

(5) To prevent the dispersion of the ebb over a large sector of the outer bar, and to concentrate it along the line of least resistance.

(6) To construct a barrier of such form and position as not to generate injurious or destructive cross currents and eddies.

(7) To leave an ample waterway, and provide additional aids to navigation.

(8) To reduce the length of the breakwater, and consequently its cost, to a minimum, and yet furnish a safe roadstead and harbour entrance.

To meet these requirements, the Author has submitted a form of structure of a crescent or concave enceinte, and having a straight or slightly curved flank at its inner or shore end, which is intended to be placed upon the outer bar to seaward of the main channel, and to be raised to or above high-water.

By this exterior concavity it is expected that the wave action will be decomposed and tranquilized, while its burden of sand will be partially deposited in the cove thus formed, and tend to reinforce the outer slope of the work.

By leaving the beach channel open, and directing the inshore flank towards the gorge, the littoral movement of the flood will not be materially interrupted, and the foreshore will be prevented from advancing, the beach channel will be improved, while the ingress of the flood through this opening will exceed the egress of the ebb, thus throwing a larger volume at ebb over the main crossing of the bar.

Thus, by a single line of defensive works, it is expected to secure much more beneficial results at about one-half the cost.

A knowledge of the causes of the prevailing direction of the littoral drift enables the engineer to determine the proper position for those works. This subject was investigated by the Author in 1887, in a Paper entitled "The Physical Phenomena of Harbour Entrances," which was published in the Proceedings of the American Philosophical Society, Philadelphia, 1888.

In the discussion which followed, the general impressions appeared to be that there was no prevalent angle at which the breakers approached the shore, or if there were, it was mainly influenced by

the direction of the wind, and a standard authority in a work just issued states, that while on the open ocean the direction of the waves is the same as that of the wind, along shore they move normally to the beach. To show that these impressions do not accord with the facts, the Author made some instantaneous photographs of the breakers at Atlantic City, N.J., when the wind was blowing parallel to the crest of the waves, and during the second quarter of ebb-tide, which exhibit a very decided angle of approach, the direction of the movement being south-west, which accords with that of the inlets along this part of the coast.

The conclusion reached by the Author was, that the dynamic action of the breakers, especially during flood, being guided by a receding shore line, controlled the direction of the beach movement, and indicated the remedy for bar improvements.

A remarkable confirmation of the good effects secured by locating a single jetty upon the proper side of the channel, namely, that from whence the drift approaches, is to be found in the experience furnished at Aransas Pass on the coast of Texas.

Here, in 1869, a private company built a jetty only 600 feet in length, at a cost of $10,000. It was placed on the north side of the pass, and almost immediately resulted in an improvement of the channel, which was maintained until this temporary structure was destroyed by the waves and teredo several years thereafter.

(Paper No. 2411.)

'Compound Locomotives." 1

By ERNEST POLONCEAU, Chief Locomotive Engineer,
Orleans Railway.

THE results of experience show that stationary compound condensing engines and marine engines are economical; but it does not follow from that that the compound system is advantageous for locomotives. The duty of locomotives is very different from that of other engines, in that their work constantly varies. The compound system is particularly economical where the work to be done is constant, but not otherwise; and the system suitable for the one kind of engine is unsuitable for the other. For fair comparison reliance must be placed only upon trials made with locomotive engines of different systems, and under exactly similar conditions.

If an engine is modified by increasing its adhesive weight, which was previously very slight in regard to the heating surface, economical results will evidently follow the adoption of the compound system, which necessarily increases the adhesive weight, but these results prove nothing. New apparatus, when first tried and carefully attended to by its inventors or agents with a picked staff, always shows a considerable economy; but when left to the care of more or less capable ordinary attendants, it often does not give half the economy found on the trials.

The Author is convinced that there is an economy of fuel with the compound system; but, all other things being equal, he does not think it can exceed 5 to 8 per cent. ; now, between a good and an average engine-driver the differences far exceed 5 to 8 per cent., reaching even 50 per cent. with a mean of 15 to 20 per cent. In the same way, between a locomotive having its working parts and tires in perfect order, and another on the point of return to the shops after prolonged service, there is a difference of consumption, which often reaches 15 to 20 per cent.; comparing the cost of fuel per 100 kilometric tons gross, there was for the entire network of the Orleans Company in 1887, 5.86; in 1888, 5.76; or a difference of 2 per cent.

If the premiums obtained by engine-drivers be taken into con

This communication was intended for the Correspondence on Mr. E. Worthington's Paper, vol. XCVI., p. 2, but was not received until that article had gone to press.-SEC. INST. C.E.