On the Precipitation of Calcium Carbonate. 17 



the leakage, if any, would be outward, due to the expansion of the sample 

 through the regularly increasing temperature and decreasing pressure, so 

 that the sample would not be contaminated by any of the surface layers 

 through which it was hauled. There was, however, no reason to suppose 

 that any leakage occurred, and it appears that the expansibility of the 

 rubber washers and tubing was sufficient to allow for the small changes in 

 bulk of the fluids within the cylinder after the first sterilization by steaming. 

 The action of the alcohol was relied on for sterilization between successive 

 samples, and both experimentally and in practice this method was found to 

 be absolutely safe, as all the marine bacteria are very readily killed by 

 alcohol, and they do not form resistant spores. 



After the collection of a sample it was siphoned off into a sterilized glass 

 bottle by means of a sterilized length of rubber tubing; this method was 

 considered preferable to any arrangement of taps leading from the collecting 

 cylinder, owing to the difficulties of cleaning and sterilization which would 

 be involved. Part of the sample was also siphoned off into bottles which 

 were returned to Plymouth for analysis for salinity ; these bottles had previ- 

 ously been thoroughly washed, rinsed with several changes of distilled water, 

 and then dried in an oven; they were corked with rubber stoppers. 



It was found in practice that this design of water-bottle worked extremely 

 well and gave very little trouble. It is to be noted that the sample of water 

 collected is only kept in contact with rubber and glass throughout, so that 

 the bactericidal action of metals is avoided. 



Surface samples of water were taken in wide-mouthed stoppered bottles, 

 holding about 12 ounces; the samples were always taken from the bow of 

 the boat when moving ahead, in order to avoid any possible contamination 

 from the sides of the boat. 



Some samples from depths up to 80 fathoms were collected off the 

 Tortugas in 191 1 in retort-shaped glass flasks of about 300 c.c. capacity, 

 with narrow, recurved, and long-drawn-out necks. These were sterilized, 

 exhausted, and sealed; they were then lowered in an apparatus in which 

 the extremity of the neck could be broken off at any desired depth by 

 sending a messenger down the sounding-wire, when the flasks became 

 completely filled with water. After hauling up, a little water was shaken 

 from the neck, and it was then sealed with the blowpipe. This method 

 avoids risk of contamination from more superficial layers of water as the 

 apparatus is drawn up, since the changes in pressure and temperature as 

 it ascends tend to cause a continuous outflow through the narrow neck until 

 the surface is reached. 



A somewhat similar apparatus was used for obtaining deep samples 

 from the station 70 miles west of Ushant, but the glass bulbs were smaller 

 and the tube leading from them was bent at right angles to itself. Con- 

 siderable difficulty was caused by the breaking of the tube owing to the force 

 of the inrushing stream of water impinging on the wall where it was bent at 

 right angles. 



