PHYSICAL OCEANOGRAPHY 



215 



De Bilt 3J millions, the Hydrographical Bureau in Washington 

 51^ millions, and so on. Add to this the surface observations 

 made by scientific and other expeditions, and it will be evident 

 that in the course of the last sixty years a good deal of know- 

 ledge regarding the surface of the sea has been gained. 



Making surface -temperature observations is very simple Temperature 

 work. All that is necessary is to haul up a bucket of sea-water observations 



dn • /* 1 * 1 ''■ ^LlG suri3.cc 



measure the temperature by means 01 an ordmary thermo- of the sea. 



meter. It is a far more difficult thing to record the actual 



temperature of the deeper layers. In 1749 Captain Ellis Temperature 



brought up water from 11 90 metres and from 1645 metres to beneath \h? 



the south of the Canaries, and, on measuring the temperature surface. 



of the water inside the water-bottle after it had been hauled 



up, found it to be 17.2° C. lower than the temperature at the 



surface. Some investigators coated their water-bottles with an insulating 



insulating substance, so that the temperature might remain water-botties. 



unaltered during the process of hauling up. This principle has 



recently been developed to a high degree of perfection in one 



of the water-bottles now most used, viz. the Pettersson-Nansen 



water-bottle, which will be described later. 



Attempts were also made to insulate the thermometer itself insulating 

 by surrounding the bulb with a stout sheath of caoutchouc or thermometers. 

 wax. This insulated thermometer was lowered to the depth 

 desired, where it was left for hours to assume the temperature 

 of the water ; it was then hauled up quickly and the temperature 

 read off. In this manner de Saussure was able, in 1780, to 

 determine correctly the temperature in the Mediterranean at 

 585 metres, finding it to be 13° C. Thermometers made on 

 this principle have been much used until our own times, but they 

 have one serious drawback, for the operation takes a very long 

 time, and this makes them unsuitable for use in expeditions, 

 where the work must be done as quickly as possible ; they may, 

 however, do good service in cases where the very greatest 

 accuracy is not required, and where there is unlimited time at 

 disposal, as on light-ships. 



Nearly a hundred years ago some one thought of employing Maximum 

 Six's maximum and minimum thermometer for temperature S'-^ermom"""^^^ 

 observations in the sea, various modifications being introduced, 

 until finally in 1868 it became quite serviceable as made by 

 Casella under the direction of Dr. Miller. The Miller-Casella Miiier- 

 thermometer (see Fig. 154) was the one principally used on board Caseiia. 

 the "Challenger" and during other great expeditions. At the 



