Appendix 2 

 Data on origin oi 12-gailon polyethylene drum samples 



For tlio most pai-t tlio coluinii lieadings are soil" explanatory. The 

 surface salinit_\" at the time of collection of tlie lnv^o sample was that 

 of a sample collected by Xansen bottle as measui'ed by the Coast 

 Guard Oceano^i'raphic Unit's Weiir.cr l)rid^i-e whereas the large sample 

 was taken from the ship's fire main after thorouo-h flushing of the 

 system. While most of the (hfferences b(>tweeii the surface salinity 

 at the tim(> of collection and the subsecpient sahiiity by ('1 are small, 

 drums 11 and 8 showed large differences. In the case of drum 11. 

 there was a considerable salinity gradient near the surface so that the 

 fact that the fii'e main intake was below the surfac(> may explain 

 the higher dnnn salinity. .\o such ex])lanation can be offered for the 

 large dilFerenee in the case of (h'um S which leaxcs the possibility of 

 inadequate Hushing of the fire main system (this was one of the 

 first drums collected in 19")'.) and |)eisonnel had little previous experi- 

 ence in this method of collection), contamination fi-om other than sea 

 water soui'ces, and excessive evaporation. Since each of the last two 

 possibilities (le|)en(l on a leaky seal, they seem im|)robable because in 

 the case of a leaky seal the contamination from rain water would 

 probably gr'eatly reduce t he salinity. One other drum which did have 

 ii leaky seal changed from ap|)roxiniately .■^6.()°/oo u|)on collection on 

 10 April 1959 to ;^3.22%o in Kel)iua!y I'.MiO. Conlidence in the seal 

 is justified by the fact that at the end of the (iist survey, before the 

 di'um was put aslioi'e for storage, a small sample was withdrawn and 

 measured a|)proximatel\ ;!().89°/oo by conduct i\ii\. Drums 1 and '.) 

 wei'e composites, each being mixtures of about e(|iial parts of di'uins 

 which wei-e about /4°/oo on either side of the desii-ed salinities. 



80 



