shall render the improvement of the entrance a more urgent matter than at present.

For the deepening of the navigation in the canal and the removal of the obstructions at Biddle's Landing and the minor points above the expenditure would vary from $4,000 to $6,000, depending upon the prices of dredging. It is impossible to forecast these for the smaller works under the contract system, as the proposals received in response to advertisement are regulated not by the cost of the work, but by the greater or less amount of other work and the presence or absence of competition. The commercial statistics for "Sharptown" or "Upper Salem" Creek are given in the accompanying schedule. Respectfully submitted.

The CHIEF OF ENGINEERS, U. S. A.

WILLIAM LUDLOW,

Captain of Engineers,
Bvt. Lieut. Col., U. S. A.

COMMERCIAL STATISTICS OF SALEM, N. J.

There are fifty vessels of all classes that belong to Salem and trade in and out of Salem Creek, the tonnage aggregating about 4,000. The larger part of merchandise received and shipped at Salem is in vessels belonging to other ports.

Two large side-wheel steamers ply between Salem and Philadelphia, carrying annually 50,000 passengers, and their freight receipts each year amount to $25,000. There is also a propeller of 250 tons, carrying freight exclusively, which makes two trips a week between Salem and Philadelphia.

The receipts at Salem during the past year were as follows:

COMMERCIAL STATISTICS OF SHARPTOWN, N. J., AND VICINITY. The population of Sharptown is 600; that of Woodstown, 14 miles further up the creek, 3,000. The annual amounts received and shipped by the creek are about as follows:

Much produce is also taken from the lower landings by open boats.

F 29.

EXAMINATION OF MURDERKILL CREEK, DELAWARE.

UNITED STATES ENGINEER OFFICE,
Philadelphia, Pa., March 18, 1882.

SIR: I have the honor to submit the following report on the examination of Murderkill Creek, Delaware, required by river and harbor act of March 3, 1881.

The accompanying report of Mr. Gieseler, assistant engineer, fully explains the methods of the examination and the existing physical conditions of the stream.

The actual present head of navigation is at Frederica, 71⁄2 miles from Delaware Bay, the distance having been reduced from 11 miles by the successive construction of canals across the greater bends.

The lowest mile of the creek is rather wide and shallow, with lowwater channel depths of less than 4 feet, but from Webb's Landing upward the 6-foot low-water curve is nearly continuous, and the interrupting shoals aggregate about 2,000 cubic yards only in volume.

The main obstruction to the present navigation of the creek proper is the inadequate dimensions of the "canals" or "cut-offs," which were originally designed to have a width of 30 feet and a depth of 6 feet. This cross-section is about one-third only of the normal area of the creek, and not only is the width too narrow for convenient navigation, but, as Mr. Gieseler suggests, the tidal prism, which is the main source of power of maintenance, is gradually diminishing in proportion as the bends in the old creek bed become filled and obliterated.

The obvious remedy is that the improvement company, which, under charter granted by the State, has made the canals, should widen and deepen them to such extent as may be necessary to facilitate navigation and conserve the tidal volume of the stream.

The obstructions at the mouth of the Murderkill are different in character, and require for their amelioration much larger expenditures and less simple methods of treatment.

As in the case of nearly all the streams entering Delaware Bay from the west, the entrance is across mud and sand shoals of considerable width, with low-water least depths of from 1 to 13 feet, the rise of tide varying from 4 to 5 feet. The ebb current from the creek spreads across the shoal without any defined channel, which its unconfined volume, in conflict with the tides of the bay and the disturbing effect of wind and waves, is unable to maintain.

The conditions are entirely similar to those of Saint Jones Creek, which enters the bay only one-half mile to the northward, and the remedies to be applied, if the two be considered independently, are the same. The examination of the Saint Jones was made in 1880, and the report thereon is printed in the Annual Report of the Chief of Engineers for 1881, pages 799 to 804, from which the following extracts are made:

Delaware Bay is an arm of the sea, varying in width from 12 to 30 miles, with the ship-channel traversing its axis. Its entire western border is shallow, and the entrance to nearly every stream is obstructed by broad flats of sand, mud, and clay.

The proper improvement of these entrances is the most important and costly feature of any valuable increase to commercial facilities, while, at the same time, needed harbors of refuge for the numerous small craft plying in the bay would be provided.

There is no probability of maintaining a dredged channel through the flats without the aid of a jetty. The daily tide movements, aided by the action of the waves in the shallow waters, would speedily fill it up. A channel suited for navigation, with the tides flowing across it, would need to be 200 or 250 feet wide, while the protection afforded by the jetty would justify a reduction in the width by one-half and effect a corresponding reduction in the dredging estimates.

The proposed jetty was to start from the north side of the Saint Jones entrance, with a direction nearly east, and after crossing the 2-foot lowwater contour, to curve to south and east, with a total length to the 3-foot low-water contour in the bay of about 3,300 feet.

The cost of the jetty and of the necessary dredging to make a 3-foot channel 100 feet in width was $35,000. To extend the jetty to the 4-foot low-water curve, with corresponding additional dredging, would cost $42,500.

Were the improvement of the Murderkill entrance only in question, the project and estimate would be almost exactly the same as those given for Saint Jones.

Since, however, the mouths of the two streams are at a less distance apart than that to which the projected works for either would extend into the bay, and since the improvement of the Saint Jones has already been appropriated for, and is under consideration by Congress, it becomes necessary to combine the two projects in order both to avoid injurious interference, and, if practicable, to secure more advantageous results.

Upon the accompanying chart of the entrances to the two creeks two projects for their combined improvement are shown, marked, respectively, A in red and B in blue lines.

The leading feature of project B (which is an ingenious suggestion by Mr. Gieseler) is to unite, by means of concave half dikes of suitable radius, the effluent waters of the two streams, in the anticipation that their combined volumes may suffice to erode a channel across the flats on the prolongation of their common tangent, which has a direction such as not to interfere with the free admission of the flood-tide into both.

Good results may reasonably be expected from this project, which furthermore admits of future extension and attainment of greater depths by the construction of dikes beyond the point of tangency and parallel to the sides of the exterior channel.

The large amount of excavation required, however, makes the cost considerable, and the project is liable to the further objection that no provision is made for the securing of harbor space.

Project A, shown in red lines, includes the dike hitherto projected for the Saint Jones entrance, the position and development of which, owing to the circumstances of the case, admit of no material variation. The position of the Murderkill dike, however, may be varied in the manner to be hereafter determined as most favorable regarding results and

costs..

Three positions are shown on the chart, marked A', A2, and A3.

The first maintains a concave bend throughout its length and gives the most harbor space; the second affords a better entrance for the flood-tide; while the third more nearly unites the two streams, but largely diminishes the area of the harbor. By reason of the greater amount of dredging necessary, it is, besides, much the most expensive of the three.

Estimating on the basis of a low-water entrance depth of 4 feet, the cost of the several projects, including dikes and dredging, is computed as follows:

To the above amounts should be added the 12,000 estimated as necessary for the dredging in the bed of the Murderkill between Webb's Landing and the mouth.

The estimates above given are for a 4-foot low-water entrance.

It is not necessary that the entire amounts should be given at any one time, nor need the depth estimated for be necessarily adhered to.

An increase of the present depth to 3 feet at low-water, or even a reliable entrance, fixed in position, of 2 feet would be of great value to the navigation and trade of both streams. To effect these, a nearly proportional reduction in the estimate may be made of 25 per cent. for each foot of draught surrendered. Thus, taking project A as a basis, to obtain 3 feet would cost about $55,000, and for 2 feet about $40,000.

The argument as to the propriety of the expenditure of these amounts depends less upon the present trade of the streams in question than upon its probable development, when by the opening of the mouths reliable access may be had to the interior.

The construction of harbors of refuge for the numerous small vessels navigating the bay is likewise a matter worthy of consideration. Respectfully submitted.

The CHIEF OF ENGINEERS, U. S. A.

WILLIAM LUDLOW,

Captain of Engineers,

Bvt. Lieut. Col., U. S. A.

REPORT OF MR. E. A. GIESELER, ASSISTANT ENGINEER.

UNITED STATES ENGINEER OFFICE,
Philadelphia, Pa., March 4, 1882.

COLONEL: I have the honor to report as follows upon the examination of Murderkill Creek:

The examination extended over that part of the creek lying between Quillen's Landing, 5 miles above Frederica, and the mouth, a total distance of about 124 miles, measured on the water.

The field-work was executed between the 19th and 27th of October, 1881.

The turnpike bridge above Frederica not being permanent, but a draw allowing the passage of vessels, Frederica, strictly speaking, cannot be called the head of navigation. Yet that term is applicable to the place mentioned, since the trade above it is very small. In view of this fact and of the limited means at disposal for the examination, and considering the sufficient depths from Quillen's Landing to Frederica (54 feet at high-water), obstructed slightly in two places only, it was thought sufficient merely to examine these shoals and to omit that detailed examination of the upper creek which would have been necessary for subsequent mapping. The chart, therefore, does not show the upper 5 miles of the creek above Frederica.

This course appeared the more desirable since the original intention of confining the field-work to the collection of hydrographical data only had to be abandoned, the very first exploration developing the unreliability of the creek lines as shown in the atlas used as a basis for the topography. It was found impossible to locate the lines of soundings by reaches and points as given on this map, because hardly any of them could be identified in the field, and the range of the necessary field-work was thus unexpectedly widened at the outset.

A rapid polygon was run from Frederica down to the mouth, the angles of which were taken by sextant, and the distances measured by means of a telameter rod, while cross-lines were located and sounded at suitable distances and the existing shoals examined as to their extent and material. From work like this, that lies between "field-sketching" and "surveying," no great degree of exactness can be expected, but merely that reasonable accuracy which is sufficient for the practical purposes in view.

Two tide-gauges were established, viz, one at the mouth and one at Frederica. The one at the mouth was established on the same level as the one that had been used

during the preceding survey of the mouth of Saint Jones, thus allowing the use of previous observations in computing the tidal element. Even including those former observations, the entire amount of data available is small, and as simultaneous observations on a standard gauge were not practicable, the results arrived at in regard to the tidal conditions of the creek are necessarily only approximate. According to these it appears that mean rise and fall at the mouth is 4.2 feet, while at Frederica it is about 1.2 feet, or 3 feet less.

The distance from the mouth to Frederica, 7 miles, seems to be traversed by the high-water in somewhat more, and by the low-water in somewhat less, than three hours.

The greater part of the course of the creek is through marsh land at about the highwater level; here and there the ground rises a little more, and areas of farm land or high meadow stretch down to the banks. At the mouth a belt of sand, running along the beach, separates the meadow and marsh from the Delaware Bay.

A number of landings that are marked on the map are convenient shipping points for the products of the farms bordering on the creek, and, combined with the mercantile and manufacturing interest of Frederica, indicate a considerable trade in timber, lumber, phosphate materials, canned fruits, farm truck, &c.

Ten vessels, trading to Philadelphia and New York principally, are owned on the creek. Further details concerning these matters are given in a subjoined table, based on information received from Governor Hall, of Delaware, and Mr. Anderson, of Frederica.

The general direction of the creek from Frederica to the mouth is northeasterly. The original course was an extremely sinuous one, the distance from Frederica to the mouth, as measured along the creek, being about 11 miles, while the air-line distance is only about 5 miles. During the last sixty years, however, the parties interested in the trade of the creek have straightened its course by the construction of six canals at various places, and the water distance has been reduced to about 74 miles. The entire length of these cuts amounts to nearly 14 miles, cutting off three times that length of creek.

According to information received from Mr. Anderson, of Frederica, the cost of the Bradley Cut, 30 feet wide by 6 feet deep and 2,000 feet long, was $1,100, which would represent a rate of less than 10 cents per cubic yard of material actually removed. The cuts, however, are only 30 feet wide, and the excavated material could be thrown on the shore directly from the dredge without further handling. Basing an estimate on the above price, the entire amount expended on the cuts would have been about $5,000, besides an unknown amount paid to property-holders who were damaged by the excavation of the cuts. In addition to this, $2,000 were expended in an unsuccessful attempt to improve the mouth of the creek by dredging.

The canals or cuts are a peculiar feature of the creek. Having been constructed with the single idea of reducing the distance at the least cost, their effect, however satisfactory for the time being, can scarcely prove entirely beneficial in the course of years. Their width is pretty uniformly about 30 feet, and depth about 6 feet, giving a mean low-water cross-sectional area of about 180 square feet. Now, the mean crosssectional area of the creek proper at Frederica may be put at about 450 square feet, gradually increasing to about 700 square feet at the mouth. Manifestly, therefore, the capacity of the cuts is entirely insufficient for the accommodation of the body of water moving in the creek. It is perhaps in appreciation of this fact that the old course has everywhere been left open, but naturally it is undergoing a gradual process of shoaling and narrowing, which in some cases has proceeded so far as to almost close the old creek bed. This is notably the case at Walker's and Warren's cuts, which are the two oldest ones.

It may be assumed that, as this shoaling of the old creek advances and the entire in-and out flow is more and more restricted to the narrow and insufficiently capacious ents, the tidal volume of the creek is correspondingly diminished. Although no reliable historical data on this point could be obtained, it is quite probable that the mean rise and fall in the upper sections has already been reduced through the action of the causes named. On the other hand, it appears safe to say that a suitable increase in the area of the cuts would be followed by an increase in the tidal volume; assertions which, if they are correct, would have an important bearing not only on the future of the creek itself, but also on any contemplated improvement of the mouth.

The following list, besides giving other data concerning the cuts, gives the crosssectional area which, according to the mean profiles of the creek, they shouli possess: