Particular care must be taken to ensure the correctness of the standard temperature, for a mistake of 0·1° causes the weight of 10 c.c. of water to be estimated either too high or too low by 0·14 milligram, giving rise to an error in the fifth decimal, or making 100,000 parts
100001·4 parts. These determinations have been made in Dupré’s apparatus, which, when furnished with a sensitive thermometer, allows the fluctuations of temperature to be fixed within the limits of 0·01°. If many determinations had to be made, I should avail myself of Scheibler’s (‘Zeitschrift für Analytische Chemie,’ vol. vii, p. 88, 1868) electro-magnetic regulator for maintaining a constant temperature.
A peculiar feature of my instrument is the ease and precision with which the measurement of the liquid can be adjusted at the moment it has taken the standard temperature; for it will be found that the liquid expands and contracts only in the wider capillary tube, viz. in the direction of the least resistance. The narrow capillary tube remains always completely filled. Supposing the liquid reaches beyond the mark b, it may be reduced through capillary force by touching the point a with a little roll of filtering paper. Supposing, however, that in so doing too much liquid is abstracted, capillary force will redress the fault if point a be touched with a drop of the liquid under examination; for this gentle force acts instantly through the whole mass of the liquid, causing it to move forward again to or beyond the mark.
As the instrument itself possesses the properties of a delicate thermometer, the time when it has reached the standard temperature of the bath may be learned from the stability of the thread of liquid inside the wider tube. The length of this thread remains constant after the lapse of about five minutes.
In wiping the instrument (after its removal from the bath) care should be taken not to touch point a, as capillarity might extract some of the liquid; otherwise the handling of the liquid requires no especial precaution.
The capillary tubes need not be closed for the purpose of arresting evaporation, at least that of water. I have learned from the mean of several determinations that the error arising from this source amounts in one hour to 1⁄20 of a milligram.
In cases where the temperature of the balance-room is high, and the expansion coefficient of the liquid to be examined is considerable, it may be found necessary to put a small cap (bead-shaped and open at both ends) over the extremity of the wider capillary tube, for the purpose of retaining the liquid, which during the time of weighing might otherwise be lost, owing to its expansion. When a cap is used the wider capillary tube need not be longer than the narrow one.[180]
[180] This instrument is manufactured by E. Cetti & Co., 11, Brooke Street, Holborn, London.
The ‘Compte Rendus’[181] describes a new specific gravity apparatus, the invention of M. Pisani. The apparatus in question consists of a glass vessel about 5 c.c. capacity, closed with
a perforated stopper like an ordinary specific gravity bottle. To the side of the vessel is joined a tube, coming off at an angle of about 45°, about 25 cm. long, and 4 mm. internal diameter, and graduated at 50ths of a c.c. The vessel is filled with water, the level of which is read off in the tube held vertically, the finger being held over the hole in the stopper; 2 or 3 grams of a mineral are then placed in the flask, the stopper is replaced, care being taken to lose no water, and the level is again read off in the graduated tube, held vertically as before. The difference in the two readings gives the volume of the mineral taken.