448 L. Waldo— Papers on Thermometry. 
8. The Tonnelot standards have a maximum correction of 
about —0°-25 C., which occurs at some point beyond 50°. 
There has not yet been proposed a thermometer which can 
be used to supplant the mercurial. A slight experience with 
the air or non-mercurial liquid thermometers will satisfy the 
SSaghie that they are too difficult to use for ordinary scien- 
work outside of researches in thermometry. e non- 
aerate liquid thermometer wets the glass and absorbs air to 
such an extent as to always make it unreliable, though the pe- 
culiar errors depending on the glass itself are reduced to an 
exceedingly small amount owing to the large linear expansion 
of any liquid used for a thermometric substance as compared 
ih mercury. 
t becomes therefore of importance that we should have 
a Sis of known constitution and methods of manufac- 
ture, for mercurial standards. It is not sufficient that we 
should know only the commercial name of the glass, because 
this seldom gives any good idea of its constitution. Presuma- 
bly the chemical constitution of the glass is of more importance 
than its manipulation in manufacture, and it hardly seems just 
to science for the glass makers to decline to furnish such data 
regarding thermometer tubing as will best enable students to 
investigate the peculiarities of particular standards. The 
manufacturers could hardly suffer financially from such a 
revelation, since the manufacture of thermometer tubing is so 
largely a question of individual skill in drawing the tubing 
Ss. 
Out of a very large number from many makers which we 
have examined in this connection, it may be safely said that 
the glass used in the Kew thermometers (and in general in the 
about the same amou 
or our own convenience in the preparation of standards we 
desire to pursue our inquiries far enough to determine if possi- 
ble in how far the behavior of glass as a thermometric sub- 
stance is affected by its constitution and its methods of 
manufacture. The present state of the matter leads to a con- 
fusion in the making and examination of standards which is 
perplexing alike to the maker and the student of science who 
has not the refined ee necessary to refer every ther- 
mometer to the air standa 
After pee the deectripiten of the Fuess standards at p. 
eceived a note from Dr. Foerster asking me to give 
Cablacity to the following letter from Dr. Thiesen in wee to 
—— 
