906 TRANSACTIONS OF THE AMERICAN INSTITUTE. 



Here the water at the pivot span is 40 feet deep, the current riin- 

 ninfj some four miles an hour. The cylinder dcsicrned was to be 

 40 feet in diameter. 



C0NTR.VCT10N OF Water-way. 

 Let ns see now in what way the velocity of the river is affected 

 or, in other words, the contraction of the water-way. Assuming 

 the average depth of 60 feet, our pier Avould present an area 

 against the stream of say 40 X 60=2,400 square feet. The original 

 area of the channel being, say, 60x1,350=81,000 square feet, the 

 ratio being nearly as 1 to 34. The mean velocity of the natural 

 water Avay l)eing 2 feet .per second, that of the contracted way 

 would be m^V=1.097X:I'^!"I X 2=2.24 feet per second (m is a con- 

 stant according to du Buat; s=original area; B=climiuished area; 

 and V=original velocity.) The increase is, you see, .24 per second 

 or 864 feet per hour, an increase of speed that could not be 

 noticed except by careful mcasui'ements. The fall corresponding 



to this velocity is | inch, (fall= ; — c' so that you see the 



apprehended injurious effect upon the river current amounts 

 almost to nothino; at all when examined in the li2;ht of figures. 

 The office of masonry being to distribute pressure over a great 

 area, after the superincumbent weight has been provided foj-, the 

 piers and alnitments should be built cellular, in well bounded 

 courses tied together. 



CiiAXGES BY Variation of Temperature. 



The effect of change of temperature in the superstructure is 

 provided for, in mounting the end cast iron saddles to v.'hich the 

 chains are attached, upon steel rollers, secured in a frame, rolling 

 upon a plate imbedded in the masonry. This would be all the 

 provision needed, since the material employed being all of the 

 same kind and character, the contraction and expansion would be 

 the same. The expansion of steel would be about .0000067 of 

 its length for ever}- degree Fidi., which for 100 degrees (taken as 

 extremes of temperature in this latitude) would cause a vai-iation 

 in length for 650 feet, of .435 feet or 5| inches or 2-^^^ inches at 

 either end of each span. This in the horizontal member. The 

 chains being 658 feet long they w^ould have an additional incre- 

 ment of expansion or contraction, due to the additional eigl)t feet, 

 amounting to -j\ of an inch at the center, which, of course, would 

 be inapprecial)le in effecting a change of versed sine. The matter 



