In a message dated 5/11/2014 2:24:09 P.M. Eastern Daylight Time,  
[log in to unmask] writes:

In  many, if not most, instances, the visible tile on the lower surface of 
vaults  (what I recall RGCo calling “soffit tile” on their drawings) are 
built  integrally with the subsequent layers of tile. So, in a way, they are “
wedged”  in place like voussoirs. There are exceptions. Tile below the gross 
domical  form in RGCo’s vaults -- think of the vaults in the Oyster Bar in 
Grand  Central, as one example -- were applied to the “structural” tile in 
the manner  you describe. It is likely, though I have no documentation to 
support this  statement, that the elaborately decorated surfaces of vaults 
(the Nebraska  State Capitol comes to mind) were also applied to the surface of 
the  structural tile above. 
Derek, as always, thank you.  I posted this  query after seeing the 
Guastavino show at MCNY.  There they have a  demonstration dome, perhaps 4' square, 
which indicated to me that the bottom  tiles were held in place only by 
some adhesive.  Having just seen a  sofa-sized piece of cast stone fall from 
the Zeckendorf apartment house (c.  1985) at 86th and Lexington (apparently an 
attachment failure - no one injured)  it occurred to me that I have never 
heard of such a failure with Guastavino  tiles.   And there are so many of 
them.
 
So ... what is that glue?
 
Christopher 
 
(PS If you're interested in G as an architect, you  gotta see these photos: 
http://www.nytimes.com/2014/03/09/realestate/off-with-the-face-paint.html )
 

--
**Please remember to trim posts, as requested in the Terms of Service**

To terminate puerile preservation prattling among pals and the uncoffee-ed, or to change your settings, go to:
<http://listserv.icors.org/archives/bullamanka-pinheads.html>