The Sani> Blast, 639 



Tilghman u long time to perfect liis inventiou, to make it a 

 success. Tlie manner of introducing the stream of sand 

 required patient study and experiment ; the flow must be 

 reguhited by tlie quantity required, and only as fast as 

 needed. This he accomplishes by carr}ing tlie steam into 

 the gun above tlie sand entrance, by which the pressure acts 

 as a syphon in drawing the sand down, and ceasing the 

 moment the steam is shut off. 



The proper distance between tlie end of the blast tube 

 and the face of the stone was a matter of study and costly 

 experience on the part of Mr. Bridges, (who first demon- 

 strated the availability of this apparatus for lettering on 

 marble to any extent). If the blast is an inch too close the 

 steam blows off the letters, and if an inch too far away the 

 blast loses its effect. Experience shows that nine inches is 

 just right for our work. 



The force of steam required has only been determined 

 by long trial ; we are using from one hundred to one hun- 

 dred and twenty-five pounds ; less than ninety-five pounds 

 will not accomplish any satisfactory results. 



The kind of sand is important. At Keokuk, Iowa, where 

 this work was done to a small extent last season, we found 

 the island shoals oi- bars of the Mississippi Kiver afforded 

 the best cutting sand. These are made up of the broken 

 down particles from the trap and other hard rocks pi the 

 upper Mississippi, thoroughly washed by the river cm-rent 

 from earthy impurities. 



At West Rutland we tried the sand from the adjacent 

 sand pits, Ijut it is altogether too crumbly and impure. 

 Tlien we experimented on sand brought by Messrs. Sheldons 



