Chapter III 



— 31 



Collecting Samples at Sea 



In practice, several dozen glass or collapsible rubber bottles are as- 

 sembled and sterilized in the autoclave for use at sea. After collecting the 

 sample, the water is analyzed or transferred aseptically to other con- 

 tainers. The sample bottles are then sterilized in a pressure cooker in the 

 ship's galley and refitted with new capillary tubes. In an emergency they 

 may be sterilized with 70 per cent alcohol or by boiling for an hour, but 

 neither procedure is recommended for critical work. 



Metal bottles are bactericidal: — It has been noted that in many of 

 the early observations on the occurrence of bacteria in the sea, water 

 samples were collected in metal containers. The very low counts ob- 

 tained with unsterilized metal cylinders led such workers as Fischer 

 (1894a), Otto and Neumann (1904), Bertel (191 i), Gazert (i9o6Z>), 

 and others to believe that adventitious organisms were effectively flushed 

 out of the open-end cylinders as the latter dropped through the water. 

 Actually some of the low or anomalous counts reported by these workers 

 may be attributable to the bacteriostatic or bactericidal effect of the metal 

 (mostly brass) containers or to contamination. 



After finding a reduction of 97 to 100 per cent in the bacterial popula- 

 tion in 100 ml. of sea water exposed for six hours to 2 square inches of 

 bright bronze, nickel, brass, silver coins, or copper foil. Drew (1914) con- 

 cluded that platinum is the only metal suitable for the interior of bac- 

 teriological sampling apparatus. Bedford (1931) noted little bacteri- 

 cidal effect of the Prince Rupert brass and phosphor-bronze sampling 

 bottle of Young et al. (1931) in sixty minutes, but in ninety minutes there 

 was an appreciable diminution in the bacterial count of sea water stored 

 in the metal bottle. Bacteriologists working in the Ocean ographic Labo- 

 ratories of the University of Washington sought to obviate the bacteri- 

 cidal effect of the Prince Rupert water bottle by having its interior plated 

 with platinum for collecting samples of sea water for bacteriological 

 analysis. 



ZoBell (1941c) noted a marked decrease in the bacterial population 

 in sea water stored in brass Nansen bottles as compared with that stored 

 in glass bottles. This is shown by the data in Table VI which gives the 

 average of four experiments. Characteristic of stored sea water, there 

 was a slight decrease in the bacterial population in the glass bottles during 

 the first few hours, followed by an increase later. 



Table VI. — Number of viable bacteria in sea water after dijjerent periods of storage in brass 

 Nansen bottles and in glass bottles for different periods of time at 22° C. The number of bacteria 

 is expressed as per cent of the number present at the beginning of the experiment: — 



Some of the Nansen bottles were found to have a greater bacterio- 

 static effect than others, probably due to differences in the metallic sur- 



