74 DR W. S. BRUCE, MR A. KING, AND MR D. W. WILTON ON THE 



3. The question of its reliability when used during low air temperatures, and the fact 

 that the whole instrument may be solidly frozen up immediately it comes out of the 

 water (this actually occurred on board the Scotia). 



4. The fine scale of the Richter thermometer, which is difficult to read on the deck 

 of a ship in bad weather and poor light. 



5. The fact that it is not possible to use it with reliable results at greater depths 

 than 1000 fathoms. 



The Buchanan-Richard water-bottle is more desirable than the Pettersson-Nansen — 



1. Because it is inexpensive. 



2. Because it is easily manipulated. 



3. Because it does not jam by freezing up during cold weather. 



4. Because the Negretti-Zambra reversing thermometer scale is easily read under 

 conditions of weather and light when the Richter thermometer cannot be read at all, 

 or with difficulty. 



5. Because it can be used at any depth desired. 



It has been contended that the point at which the mercury breaks off in the 

 reversing thermometer is not constant. Errors due to this will probably be found 

 to be less frequent than those which occur with an instrument with such drawbacks 

 as have been mentioned in the case of the Pettersson-Nansen bottle with the Richter 

 thermometer. 



From what has been already said, it is unnecessary to emphasise that the hydrometer 

 observations taken by Mr Wilton were those of a trained physicist, and such methods* as 

 using " towels or wash-leather ' : that " were not generally quite clean," or " the surface 

 of the water samples " being " more or less contaminated during the determinations with 

 the hydrometer," were not practised on the Scotia. Mr Wilton had all his instruments, 

 silk cleaning-cloths, vessels, and fingers, and surroundings generally, scrupulously clean 

 before commencing observations. Besides this, he was most careful in maintaining the 

 laboratory and the water samples as nearly as possible at the same temperature. In 

 this he was in the highest degree successful, even in weather conditions that were on 

 many occasions very difficult to deal with. 



The laboratory was fitted with a specialserving-table for this work ; and there was 

 also constructed a special swinging platform, which on some occasions, but not as 

 a rule, was more convenient than the table. The table proved most useful for this 

 and other delicate operations. 



The form of hydrometer used by Mr Wilton in determining the density of the 

 sea-water was the Buchanan absolute-weight hydrometer. One of these hydrometers 

 (No. 25) was used for Samples 1 to 199, but unfortunately it was broken on 24th 

 February 1903. Thereafter hydrometer No. 14 was used. This form of hydrometer 

 is fully described in Mr J. Y. Buchanan's elaborate monograph on " Experimental 



° The Norwegian North Polar Expedition, 1893-90, vol. iii, pt. ix, pp. 147, 148, "The Oceanography of the 

 North Polar Basin," by Fridtjof Xansen. 



