SUEFACE WATERS OF THE NORTH ATLANTIC OCEAN. 
09 
to the work than the smaller Fahrenheit decree ; the simpler numbers are also an 
advantage in charting large numbers of observations, and I found the occasional 
negative sign no trouble whatever. Taking into consideration the excellent quality 
of the instruments used, and the general nature of the observations, no attempt was 
made to correct the observations for either instrumental error or diurnal range. 
With the view of obtaining material for constructing salinity charts the following 
arrangements were made. At my request the Meteorological Council authorised 
their secretary, Mr. Ik H. Scott, F.R.S., to write to a number of the commanders 
and officers taking temperature observations, asking if they woidd be willing to 
undertake the additional trouble of preserving and sending to me samples of the 
v/ater collected for their temperature and hydrometer observations, in bottles supplied, 
and to fill up a form with the necessary details as to ship’s position, temperature, &c. 
The Ijottles, when full, were to he sent to me as soon as the ship reached its nearest 
point, and were to be replaced by a fresh supply as required. 
Favourable replies were received from a considerable number of observers; and this 
number was added to by observers working for the Danish Hydrographical Depart¬ 
ment, the officials of which, under Commodore AAaxdel, took a great deal of trouble 
in the matter. A number of volunteei'S were obtained in response to an appeal 
pu])lished in the ‘ Field ’ newspaper, amongst whom was the late Sir CfEOiiC4E Baden 
Powell. Mr. C. M. Mundahl, of Grimsby, added largely to the material obtained 
from the regions between Faeroe and Iceland, during 1896, by observations made on 
board his vessels, and I am indebted to friends for assistance from other observers. 
The necessity for restricting tlie expense connected with the investigation, an 
important item of which was the cost of carriage of boxes to and from the vessels, 
made it necessary to kee}) the number of oljservers consideraljly Ijelow what could 
easily have been oljtained, and even to diminish the number during the second year 
of the work. The detailed list of observers is given in Appendix I., and I have to 
express my most cordial thanks to them for the enthusiasm, skill, and accuracy, witli 
which the work was carried out. In addition to the work of collecting and forwarding 
the samples, done with one exception altogether gratuitously, I received a number ot 
letters containing valuable hints and suggestions bearing on the interpretation of tlie 
results. 
Tlie boxes used were each capalde of holding thirty G-ounce bottles, the liottom of 
the box being lined with felt, and the bottles kept in position by perforated frames, 
also lined with felt. The tops of the corks in the Ijottles were exactly level with the 
top of the box, so as to be in contact with the lid when closed, preventing all 
possibility of movement, and yet avoiding all need foi' special packing. The efhcacy 
of the arrangement is apparent from the fact that not one of tlie 4100 samiDles was 
lost throuo-h a breakage in transit. 
The bottles were made of ordinary blue glass, and v/ere supijlied by Messrs. Baird 
and Tatlock. 
