104 INTERNATIONAL CONFERLNCE, 



have a space at least 10 mm. between the marks of oue degree, and the 

 degree should be divided in ten parts. 



For greater depths of the ocean Negretti-Zambra's or other thermo- 

 meters of a similar type should be used. 



The glass to be used for the thermometers as well as the thermo- 

 meters should be tested and approved by the central bureau (see C, a). 



For the determination of salinity and density, either chemical or 

 physical methods may be adopted, provided that the salinity can be 

 determined with an accuracy of 0,o5 in a thousand parts (and the 

 density up to 0,oooo4). 



The determination of these constants can be founded either upon 

 chemical analysis of the halogen by weighing or titration, or upon 

 physical determination of the specific gravity by means of hydrostatical 

 balance pycnometers and hydrometers, provided that measures be taken 

 to exclude disturbances arising from thermal effects, capillarity, vis- 

 cosity, etc. 



The chemical analysis shall be controlled by physical methods, and 

 the physical determinations by chemical analysis in the following 

 manner : — 



From every collection of samples examined at least three shall be 

 selected and sent to the central bureau. Standard samples shall be 

 sent in return.* 



The specific gravity is to be represented in the tables by the 

 formula S (jo). 



V. 



Samples for gas analysis are to be collected each time in a pair 

 of sterilised vacuum tubes. 



It is desirable that the existing tables of absorption of nitrogen and 

 oxygen should be revised. 



VI. 



Qualitative plankton-observations should be made every six hours by 

 pumping through a silk net (Nr. 18) for the space of fifteen minutes, 

 and at the same time a sample of water (HI. c) should be taken. 



At the points mentioned in ii. samples for quantitative analysis are 

 to be collected according to the method of Prof. Hensen at different 

 depths depending on the hydrographical circumstances. 



Petersen's modification of Hensen's net is recommended. 



* By standard water shall l)e understood samples of filtered sea-water, the physical 

 and chemical projjerties of which are known with all possible accuracy by analysis, 

 and statements of which are sent to the different laboratories, together with samples. 



In respect to halogen the ordinary water-samples have to be compared with the 

 standard water by analytical methods. 



