ANALYSES OF ROMAN GLASS FROM SILCHESTER. 163 
ANALYSES oF ROMAN GLASS From SILCHESTER, 
WITH SPECIAL REFERENCE TO THE AMOUNT OF MANGANESE 
AND IRON PRESENT. 
By C. J. WHITE, Caird Scholar, University of Sydney. 
(Communicated by Professor LIVERSIDGE, F.R.S.) 
[Read before the Royal Society of N. S. Wales, December 5, 1906.] 
[Introduction.—The following analyses of Roman glass 
have been made under my direction by Mr. O. J. White, 
Caird Scholar, in the Ohemicai Laboratory of the University 
of Sydney, mainly with the object of ascertaining if possible 
whether manganese peroxide had been purposely used in its 
preparation, and partly because the amount of manganese © 
in Roman glass is not given in the analyses contained in 
the principal books of reference. 
The material consisted of about 8 ounces of fragments of 
Roman glass found in 1896, during the excavations on the 
site of the Roman city at Silchester) between Basingstoke 
and Reading, Hampshire, England), which I had obtained 
for the purpose of examination from Mr. W. H. St. John 
Hope, the Assistant Secretary of the Society of Antiquaries. 
The fragments consisted mainly, if not entirely, of portions 
of bottles. The glass for the most part is of a dull greenish 
colour, somewhat blebby, and most of it shows the usual 
iridescent scale due to superficial decomposition. The 
hardness of undecomposed surfaces was found to be slightly 
greater than that of ordinary Hnglish window glass, that is, 
some of this Roman glass will just scratch window glass, 
but the difference in hardness is not very great, nor is it 
a matter of much importance. 
The presence of manganese is of interest, because pyro- 
lusite may have been added to counteract the green colour 
