576 ANNUAL REPORT SMITHSONIAN INSTITUTION, 19 31 



from temporary into final position in the cable and finally the cable 

 is compacted into cylindrical form and wrapped with a layer of 

 finer wire (pi. 3, fig. 1). While the principle of this method is an 

 old one, having been used already 50 years ago in the Brooklj^n 

 Bridge, the machinery and equipment necessary for the spinning 

 have undergone radical improvements which have resulted in greater 

 accuracy and speed of erection. 



THEORY AND RESEARCH 



Advances in theory and extensive research work have aided 

 materially the construction of large bridges, in fact, refinements in 

 theory and experiments are called for and justified mainly in con- 

 nection with structures of unusual size. The refinements in theory 

 include the elaborate calculation of secondary stresses of all kinds, 

 stresses which are usually neglected in ordinary structures, but 

 whose magnitude it is well to determine, and where necessary pro- 

 vide for, in larger ones. Such elaborate calculations were carried 

 out in connection with the Hell Gate Bridge, the George Washington 

 Bridge, and more extensively in the Bayonne Bridge. 



Stress determinations by calculation are now being supplemented 

 by stress measurements on models. In order to check the highly 

 statically indeterminate stresses in the towers of the George Washing- 

 ton Bridge a celluloid model 6 feet high of one of the tower bents 

 was constructed and the stresses were measured by means of very 

 sensitive extensometers. For the Bayonne Bridge a complete model 

 of the main arch was built of brass, loaded in various manners ver- 

 tically as well as horizontally and the stresses measured by exten- 

 someter. A complete model of the Mount Hope Suspension Bridge 

 was recently built by Professor Beggs of Princeton and the stresses 

 in it measured with excellent results. 



As a further means to check the theory, stress measurements by 

 extensometers on the actual structure have been undertaken. Such 

 measurements are now in progress in certain parts of the George 

 Washington Bridge and more extensively in the Bayonne Bridge. 



Finally, in order to gain further knowledge of the actual be- 

 havior of large-sized members and connections when loaded to de- 

 struction, a series of such strength tests have been undertaken. In 

 connection with the George Washington and Baj^onne Bridges, for 

 instance, compression tests were made of a number of columns of 

 various materials and of the largest sizes ever tested, taxing the 10,- 

 000,000-pound testing machine of the Bureau of Standards in Wash- 

 ington to its capacity. Many tests of large-size riveted connections 

 have also been made in recent years. 



