730 NEW YOKK STATE MUSEUM 



water level is a rack on which the cement can be placed. The 

 pat is made and then pnt on the rack for 6 to 8 hours,, after 

 which it is pnt in warm water from 16 to 18 honrs. If at the 

 end of that time it is firm and adheres to the glass, it can be con- 

 sidered safe. -If it does disintegrate, it may simply indicate that 

 the cement is too fresh. Cement is said to blow very often if 

 tested 24 hours after making. 



In making the pats it is necessary, specially in the case of slow 

 setting cements, to protect them from the sunlight and drafts. 

 For this reason they are covered with a moist cloth. 



Gary claims^ that, according to German experience, all tests 

 to determine blowing with the exception of the German cake 

 method are misleading, and that a swelling of cements (Portland) 

 is really a rarity. 



Cements of changeable volume, he maintains, differ in other 

 properties, specially their tensile strength, from Portland cement, 

 so that they are easily recognized. Some cements, however, such 

 ■as highly magnesium ones, will, when burnt to a clinkered con- 

 dition like Portland, refuse to swell when first mixed, and some- 

 times do not show an increase in volume when kept under water 

 till nearly a year later, but they then show the property to a 

 marked degree. 



The apparatus used by the Germans for determining change 

 in volume is knoAvn as Bauschinger's caliper apparatus, and can 

 be made to show the change in volume that takes place in a speci- 

 men over an extended period of time (pi. 26). 



It enables one to determine by direct measurement the changes 

 in length of small parallelopipeds of about 100 mm (4 inches) 

 long and 5 square cm (.78 square inch) area with an accuracy 

 of -glo" ^"^ (Woo" of an inch). The apparatus consists princi- 

 pally of a stirrup-shaped caliper, having a fine micrometer screw 

 on its right arm, the left being the support of a sensitive lever. 

 The shorter arm of the lever terminates in a blunt caliper point, 

 and is pressed against the measuring screw by a spring attached 



1 Trans. Am. soc. civ. eng. 30: 15. 



