580 
SECOND REPORT-I Su£. 
pared every twelve hours, and thus a complete check is ob¬ 
tained on any occasional error. The registering of this instru¬ 
ment is entrusted to the warders at the gate, all of whom are 
men of character; and there is every reason to hope that a se¬ 
ries of unexceptionable observations will be thus obtained. A 
copy of the register for May and part of June was presented to 
the Meeting. 
Description of a new self-registering Maximum Thermometer . 
By John Phillips, F.G.S. &c. 
The advantage of this invention is stated to be, the acquisi¬ 
tion of an instrument, capable of exactly the same delicacy and 
exactness as the best mercurial thermometer, possessing the 
same durability as that instrument, and applicable to measure 
the extremes of heat in a variety of positions,-—objects to which 
the ordinary maximum thermometer is, from the principles of 
its construction, entirely inadequate. In the first part of the 
paper certain facts are stated, as observed in the process of 
making mercurial thermometers, from -which the construction 
of the instrument flows as a simple inference. They relate to 
the extrication of air by boiling, to the position of the residual 
air-speck, and to the size and form of sections of the tubes. 
The small residual air-speck, which is supposed to be, with¬ 
out any exception, left in every mercurial thermometer, is em¬ 
ployed, in Mr. Phillips’s construction, to separate a small portion 
of the column, and thus to permit that portion to be acted on 
exactly in the same manner as the iron, or other cylinder, above 
the mercury of the common instrument. The air-speck is for 
this purpose brought to a certain place in the tube, and the 
bore of the tube is chosen so slender, as to render it extremely 
difficult, except by artificial refrigeration, to change the place 
of the air-speck ;—consequently instruments of very great deli¬ 
cacy may thus be made, and appear no more liable to injury or 
deterioration than the finest common thermometer. 
In applying this principle to practice, various methods have 
been tried by Mr. Phillips; and, finally, he has for some time 
preferred the following plan. 1. The bulb and tube are filled 
in the common way, carefully boiled, allowed to retain the pro¬ 
per quantity of mercury, and sealed. 2. The end of the tube 
is melted, and instantly afterwards the bulb is plunged into the 
flame of a spirit-lamp. The consequence is, that the end of 
the tube is blown out into a spherule, in which, when cooled, 
the elastic fluids are so highly attenuated as to offer no sensible 
resistance to the movement of the mercury. This operation 
