362 



THE CIVIL ENGINEER AND ARCHITECT'S JOURNAL. 



[November, 



the floor plank of the tow-path, on one side, and of the footway for 

 ordinary purposes, on the other. 

 The roof is boarded and shingled. 



Where the tops of the inner arches get below the line of the tow- 

 path floor, a piece of plank on edge resting on the floor is nailed 

 against the queens, to prevent horses from slipping into the canal 

 trunk. 



Fig. 5 shows an excellent device for counteracting, to some extent, 

 the tendency which trusses always have to settle in the centre. The 

 poles or caps of the adjoining arches are here connected by fishing 

 splices, each composed of two pieces of 4 x 12 in. well bolted, and 

 trenailed together, through the caps ; and as a further security, two 

 blocks, 5, «■, of hard wood, about 4 inches square, are driven through 

 from side to side. By the introduction of these splices, it is obvious 

 that the inward draw of two adjoining trusses is mutually counter- 

 acted to a considerable extent. This arrangement, together with the 

 extra braces, serves very much to stiffen the trusses, and something of 

 the kind should always be resorted to in large spans. 



Fig. 8 shows the mode of scarfing the pieces composing the curved 

 ribs. These pieces break joint where the ribs pass the queens, so 

 that the bolts of tlie scarfs pass entirely through the queens, and the 

 rib pieces on each side of them. The scarf of the poles is shown in 

 Fig. 3 ; that of the chords is the same as that of the poles, only that 

 •which constitutes the side view of the pole scarf, forms the top, or 

 plan, of that of the chord. 



Fig. 11. As it is always a matter of importance in a bridge erected 

 over a stream liable to freshets, to secure each span as soon as pos- 

 sible, so as to stand in case the scaffold should happen to be carried 

 away, the introduction of the secondary timbers is generally deferred 

 until those essential to the support of the bridge, in such an event, 

 are put together. Therefore, the diagonal braces are not inserted 

 imtil after the girders are put into their places. 



As these braces are tenoned into the girders at both ends, they 

 could not be inserted into the mortises in the girders, unless some play 

 were allowed at one end ; this play is afterwards filled up by a pair of 

 double wedges, as shown in Fig. 11. 



Where the counterbraces of the trusses intersect the main braces, 

 the former are merely tenoned into the latter, as shown in Fig. 3, at Y. 

 Where the chords and queens intersect, they are notched equally into 

 each other, so as to bring the two pieces composing the chords within 

 about an inch of each other. Figs. 2 and 4. 



The planking of the canal trunk is single, and well caulked. The 

 courses of plank are from six to fifteen inches wide. 



Remarks. — This aqueduct evinces, more strikingly than any other 

 structure I know of, the capability of timber for the purpose of bridge 

 building. The weight of water in the canal trunk on a single span, 

 when 4 ft. 3 in. deep, amounts to 275 tons, of 2240 lb. ; and we may 

 safely say that that weight is frequently increased to at least 300 

 tons, during the passage of boats; for although a boat, of course, dis- 

 places a bulk of water of equal weight with herself, yet it may rea- 

 dily be conceived that the water so displaced does not instantaneously 

 leave the span, on her entrance ; and 1 think we may assume that at 

 least 25 tons of it are frequently on a span at the same moment with 

 the boat. Yet on a most critical examination, made with that view, I 

 could not detect in any part of the timbers the slightest symptom of 

 what might with propriety be called crushing. Slight compressions, 

 (if I may be allowed to draw such a distinction,) were visible at the 

 heads of the queen-posts, but not to a greater extent, apparently, than 

 invariably attends all trusses of this kind in common bridges of large 

 sjtans, after having been some time in use. In all bridges there is a 

 tendency to settle, or sag, in the centre ; and this tendency, of course, 

 brings a heavy compressing strain upon the pole plates ; but beside 

 this compression, incident to the truss considered as a whole, there is 

 another, acting at the several points at which the heads of the posts 

 tenon into the poles. This compound compression explains a fact for 



which I was for some time at a loss to assign a satisfactory reason. I 

 have already stated that in the inner trusses of this aqueduct, a 

 strainiug-piece, like that shown at T, fig. 3, was inserted between the 

 heads of the posts, in preference to the short butting piece, P, figs. 

 3, 6, employed in the outer trusses. This was evidently done under 

 the impression that it opposed a more perfect resistance to the com- 

 pressions alluded to, than the shorter pieces; and, at first sight, it 

 will probably strike most of my readers in the same manner. But it 

 is of great importance to know, that although the long piece is almost 

 invariably introduced, both by engineers and bridge builders, whenever 

 extraordinary compression of the pole is anticipated, it is in fact en- 

 tirely ineffective ; whereas the short butting pieces perform the duty 

 assigned them perfectly, 



I shall endeavour to point out the cause of this. 



The compression of the poles evidently increases from the piers 

 towards the centre of the span, in the same manner as in a single long 

 piece of timber, supported at two ends, when it sags in the middle: 

 consequently, when the bridge settles, as it always will, more or less, 

 the head of any one post is moved a greater distance towards the 

 centre of the span than the post behind it, that is, between it and a 

 pier. Therefore, the opening,/'', behind the post, Q', must be a little 

 wider than the opening/;, behind the post Q ; and, consequently, the 

 inner end of the straining- piece, T, cannot be forced up into contact 

 with the head of the post Q, but must remain distant from it an amount 

 equal to the difference of the compression which takes place in that 

 part of the pole between Q' and the centre of the span, and that part 

 which extends fromy top. This difference in the amount of com- 

 pression between any two consecutive posts, is very perceptible in all 

 large bridges, being generally about i of an inch, that is if there be 

 7 spaces in the truss, between a pier and a king post, the opening at 

 the inner one will generally be about | of an inch, at the next one 

 f, at the next I, and so on to the queen post near the pier, where it 

 will diminish to nothing. In some bridges, and those excellent ones, 

 I have seen the openings behind the queen posts much greater than 

 this, at least double ; but, I believe, only in such bridges as have no 

 chords to confine the feet of the ribs. Of course some portion of 

 these openings, in every case, is due to the compression which takes 

 place in the heads of the posts themselves. This is frequently very 

 perceptible. I could just detect it in a few of the queen posts of the 

 aqueduct. 



But it may be objected that if the explanation I have given be cor- 

 rect, then even the short butting piece, P, should also be ineffective ; 

 because, if the compression of the pole increases so perceptibly to- 

 wards the centre, then supposing the length of the butting piece to be 

 i of the distance between two queens, the inner end of the butting 

 piece should not come into contact with the pole, by an amount equal 

 to i of the opening which occurs between these two queens. Plausi- 

 ble as this deduction would seem, it is, nevertheless, incorrect, for as 

 I have before remarked, the short butting pieces act admirably, and, 

 as I conceive, for this reason, that although the entire length of ^that 

 portion of a pole, or cap, between two adjacent posts, is in a state of 

 compression, which, considering the whole truss as one great beam, 

 gradually increases towards the centre, still the action of the main 

 braces against the back part of the head of each post, tends to bring 

 an additional strain upon the portion of the pole next adjoining the 

 inside of the post head. This additional strain produces a compression 

 of its own, which, unlike that operating on the truss considered as a 

 whole, decreases towards the centre. Therefore, that part of the 

 pole into which the head of any post tenons, is more compressed than 

 the part at the end of the butting piece, and, consequently, the latter 

 is brought into full action. 



This matter is a very important one, and my remarks on it were 

 suggested by seeing that in tiiis aqueduct the long straining'piece had 

 superseded the short butting piece, evidently in expectation of its 

 greater efficiency. In the Market street bridge, at Philadelphia, the 



