576 PROCEEDINGS OF SECTION H. 



existence, many in very open and exposed situations, which, apart 

 from the adhesion of the mortar, would nifallibly overturn with a 

 pressure of not more than 15lbs. per square foot. How much the 

 adhesion of the mortar helps them it is hard to say, but it 

 certainly cannot double their resistance. These chimneys are 

 regarded by architects as perfectly safe, although placed in posi- 

 tions where their fall would almost certainly cause loss of life. 



The most valuable contribution to the subject of wind pressure 

 for many years has been made by Sir H. Baker, in conne-'tion with 

 the Forth Bridge. He established three Osier anemometers, one 

 being of the gigantic area of 300 sq. ft., the others small, and 

 arrived at the most important and valuable result that the average 

 pressure on an object as lai-ge as a railway carriage or small house 

 was not more than about two- thirds of that indicated by the small 

 instruments previously used. He also showed that the effect of 

 inertia of moving parts, unless carefully guarded against, might 

 enormously exaggerate the apparent piessure, and thus probalaly 

 accounted' for the incredibly high results sometimes given. He 

 also made very valuable experiments on the effect of wind on 

 plates and lattice work, which are to be fovmd described in 

 " Stoney on Stresses," p. 524, and, therefore, need not be repeated 

 here. 



In "Engineering" of May 30th, June 6th and 13th, 1890, is to 

 be found a complete account of some most interesting researches 

 by Mr. O. T. Crosby on wind pressure. The experiments were 

 made by attaching the object to be experimented upon at the end 

 of an arm projecting from a vertical axis which could be rotated 

 at a known velocity. The result of the experiments which took 

 place at linear velocities of from ten to 100 miles per hour was 

 entirely to negative the earlier theoretical deductions and ex- 

 perimental determinations as to the law connecting velocity and 

 pressure. These earlier formulae were all of the form P=cV-, 

 and when P was expressed in pounds per square foot, and V in 

 miles per hour, c was given as -005 by Smeaton and Rouse, and 

 •0035 by Dines ("Engineering," 14th March, 1890, p. 333). 

 Crosby, on the contrary, makes P vary directly as V, and the 



V 

 formula P=— represents closely his results for a plane surface. 



He himself expresses surprise at this result, but states that both 

 he and his assistant verified it repeatedly^ with the whirling 

 apparatus, and also Avith an Osier anemometer placed^ upon an 

 electric locomotive travelling at various speeds up to fifty miles 

 per hour. 



The experiments carried out by the author were intended, first, 

 to corroborate or negative Crosby's remarkable conclusion that 

 the pressure varied directly as the velocity, and, next, to determine 

 the relation existing between the pressure on solid bodies of 

 rectangular, pyramidal, conical, or other form, and that on thin flat 



