alone that by such shocking the weight of the 

 mercury thread which presses on the narrow 

 point (at A) will cause the small fissures in the 

 narrow capillary part of the dead arm and en- 

 tirely ruin the thermometer. 



b. The bulb is completely filled with mercury 



(fig. A-2e) 



With a Bunsen or alcohol flame one heats 

 very carefully the glass jacket surrounding the 

 bulb. The mercury in the bulb should expand 

 and enter the capillary (fig. A-2e). Of course 

 the heating has to be done very carefully, pref- 

 erably with the thermometer in a horizontal 

 position, continuously rotated, and withdrawn 

 frequently from the flame in order to prevent 

 a too sudden and one-sided heating which might 

 cause the jacket to break. 



The auxiliary thermometer naturally rises 

 also, and one must take care lest some mercury 

 in the bulb of the auxiliary thermometer be- 

 comes separated or that the heat becomes too 

 great. Later, placing the thermometer reversed 

 in ice will indicate the position of the zero point. 



During the heating process one raises the 

 thermometer frequently with the reservoir 

 downward and examines whether the mercury 

 has already expanded sufficiently and whether it 

 runs down into the capillary. Naturally the 

 air bubble wiU thus be pushed downward, and 

 it wiU be necessary to force it back to the bulb 

 and above the mercury. In many cases (where 

 air bubbles are present) a check of the breaking 

 point will indicate that the mercury breaks cor- 

 rectly, since the air bubble is perhaps too small 

 and consequently has no effect on the breaking 

 point; but it is advisable to follow the method 

 outlined for removing the air bubbles. It is 

 very possible that the mercury will break at 

 another point, perhaps below point A. In that 

 case an adjustment is obviously necessary. 



c. The thermometer breaks incorrectly, i. e., the 



mercury separates at a place other than 

 point A (fig. A-2g) . 

 In the case of careless handling of the instru- 

 ments, especially during a long transportation, 

 it is quite possible that air bubbles will travel 

 from the bulb downward (see also 6). They 

 must again be brought above the mercury in 

 the bulb. 



The following manipulation is necessary: 

 The thermometer is placed in an upright 

 position and cooled, preferably in ice water, 

 during which the mercury must be continuous. 

 If one now reverses the thermometer, it will 

 be of some advantage to take hold of the 

 instrument at the point where the loop is 

 situated and with a little jerk, to turn it around 



its axis in order to break up the mercury at the 

 point where the air bubble is located (look up 

 fig. A-2g), because it should then be possible 

 to push it forward during the subsequent heat- 

 ing of the reservoir. It is immaterial whether 

 the mercury also breaks at the regular breaking 

 point. After the mercury has run into the 

 bulb, one tilts the thermometer into a hori- 

 zontal position, and heats very slowly and 

 carefully the reservoir of the thermometer by 

 a very small flame (up to about 60° C). 



The rising mercury now pushes the air 

 bubble forward, the greater the previous cool- 

 ing, the farther the air bubble will be pushed 

 on; it will be of advantage if now it be located 

 at the end of the loop or at the beginning of 

 the capillar3^ One now places the thermometer 

 quickly in an upright position in order to allow 

 the mercury to run out of the bulb, joining the 

 mercury below, thus all of the mercury in an 

 upright position forms now again a coherpnt 

 quantity. 



Now another cooling in ice will take place, 

 the thermometer will be caught again at the 

 loop, then will be turned around its axis with 

 a jerk in order to cause the mercury to break 

 at the point to which the air bubble has been 

 pushed. 



The subsequent heating need only be slight; 

 if the mercury rises up to the lower end of the 

 bulb, the air bubble has been pushed to that 

 point. One now combines the mercury again 

 in the bulb and finally cools down the ther- 

 mometer, while in an upright position, prefer- 

 ably to below 0° C. in order to make sure that 

 all the air bubbles are located above the mercury. 



The indicated manipulations are to be per- 

 formed with several stages of interruptions, the 

 principal point, however, is that the air bubble 

 is always to be pushed forward successively. 



III. The keeping and handling of the instruments 

 Of course a very careful handling of the 

 instruments is imperative in order to avoid 

 disturbances in their perfect functioning. The 

 thermometers must be shipped in an upright 

 position or at least in a half horizontal position 

 and should also be stored in this manner, and 

 should be especially protected from jolts or 

 shocks. An accurate test must consider the 

 following points. 



1. Observe whether the mercury at an up- 

 right position forms a continuous thread, i. e., 

 that the mercury is not separated and that 

 there are no globules of mercury in the bulb. 



2. Check the breaking point with a mag- 

 nifying glass. This check is especially im- 

 portant after transportation or long storage. 



H. O. 607 



169 



