8 
THE VOYAGE OF H.M.S. CHALLENGER. 
with the table and weight No. iv (the combination OOiv) weighing together 164*2633 
<;nns., is immersed as far as No. 74 on the stem, we find from Table I. that had the tem- 
perature been 0° C. the immersed volume would have been 160*502 c.c. From Table II. 
we find that at 20° C. the volume of the body of the instrument is greater by 0*0912 c.c. 
than it is at 0° C., neglecting as insignificant the variation in volume of the immersed 
portion of the stem, and adding the above correction 0*091 to the volume 160*502 at 0° C. 
found in Table I., we have for the correct immersed volume at 20° C. 160*593. This 
volume of the liquid is equal in weight to that of the displacing instrument (OOiv), which 
is, its above, 164*2633 grms. 
Dividing the weight by the volume we have the density of the liquid 
= 164*2633-^ 160*593 = 1*02285. The specific gravity of a substance is the ratio of 
its density to some standard density. If we choose as our standard density that of 
distilled water at 4° C., at which temperature, by definition, a cubic centimetre weighs 
a gramme, our densities are identical with specific gravities, “water at 4° C. being 
unity.” 
Fig. 2 shows the hydrometer C, loaded with the table D, and weight E, floating in 
the surface of water contained in the cylinder B, of about one litre capacity. The 
cylinder stands on the tray A, suspended by a hook from the beams above. As the ship 
rolls the cylinder preserves a sensibly vertical position, and in all ordinary weather the 
observations could be made easily and accurately. 
The volume of the body of the instrument may be taken to vary between 160*27 
and 160*42 c.c. If we add the volume of the stem (0*865 c.c.) to the last, we have the 
extreme variations of the volume of displaced liquid between 160*27 and 161*285 c.c. 
from these volumes and the weights of the combinations, namely, 00iv= 164*2633 and 
00v= 165*1 198 grms., Table V. is constructed, in which under Y. we have the volume 
of liquid displaced, and under D |y , D v the corresponding densities, with the combinations 
OOiv and OOv respectively. 
By the use of Tables III., IV., and V. we find without calculation the density 
corresponding to the observed readings of hydrometer and thermometer. 
lor purposes of future reduction it is necessary that the temperature of the water 
at the time of the hydrometer observation should be accurately ascertained. For this 
purpose one of Geissler’s “ normal ” thermometers, divided into tenths of a degree 
centigrade, was used. Its zero was frequently checked in melting ice, and the correction 
applied. At low temperatures (below 10 or 12° C.) a tenth of a degree makes no 
.“•■nsible difference in the resulting density; but at the high temperatures (25° to 30°C.) 
common in tropical and equatorial waters, a difference of even 0°*1 C. in the temperature 
a difference of three to four in the fifth place of decimals in the density. In 
’ - warm latitudes, and with so delicate a hydrometer, it was absolutely essential that 
• : • a. iter under observation should have sensibly the temperature of the atmosphere. 
