and Naidu and Hood, 1972). However, there also 

 seems to be some sorting of calcium carbonate 

 based on size in the different environments. It is 

 apparent (Table 6) that calcium carbonate in the 

 deltaic sediments is concentrated in the clay frac- 

 tion, whereas in the nondeltaic marine sediments 

 most of it occurs in the sand fraction (Naidu and 

 Hood, 1972). 



Interelement correlations (Table 6) have been 

 used by the authors in attempting to understand 

 element partitions in these regions. The strong 

 covariance between all alkali metals (Table 6) is 

 to be expected because of their similar geochemi- 

 cal behavior. However, on the basis of the exist- 

 ing correlations (Table 6) it would seem that ex- 

 cept for Na all the alkali metals are predominantly 

 tied up with the clay fraction (plausibly in 

 adsorbed/exchangeable sites of clay minerals), 

 and with the organic matter. The association of 

 alkali metals with organic matter is related to 

 primary fixation of the metals by living organisms 

 through metabolic activities; such a fixation is 

 now well-known and appears to need no further 

 elaboration here. It is apparent that the bulk of 

 the Na has been distributed in some other sedi- 

 ment phase than the argillaceous or the biogenic 

 fraction. The strong covariance of all alkali ele- 

 ments, particularly Na, with both Fe and Mn is 

 difficult to explain unless the premise is made 

 that the alkalies, Fe, and Mn have, at least in 

 part, a common derivation in interstitial water. 

 Sodium being a thallasophile element, a large 

 amount of it can be accounted for in salts sol- 

 idified from interstitial water, and to a smaller 

 extent this is also applicable to other alkali met- 

 als. 



The strong covariance of Fe and Mn suggests 

 that a significant part of this element has either 

 coprecipitated as fenimanganic hydrate or is as- 

 sociated in a common sediment phase. Assuming 

 that this is true, it is suggested that a part of this 

 precipitate has originated from interstitial water. 

 This does not seem untenable, in view of the fact 

 that a few inches of the surface sediment have 

 been analyzed in this study, and that post- 

 depositional upward migration of soluble Fe and 

 Mn, with oxidative precipitation of these ele- 

 ments at the sediment surface, would not be an 

 unusual occurrence in this region. Naidu (1973) 



has shown that in fact such a precipitation of Fe 

 and Mn is taking place on the adjoining shelf and 

 extrashelf sediment surfaces of the Beaufort Sea. 

 However, it is strongly suspected, from the data 

 in Table 6, that the predominant part of the total 

 Fe and Mn is tied up with organic matter, and in 

 the silt size fraction of sediments — in heavy min- 

 erals or discrete ferrimanganic particles. 



There are strong positive correlations between 

 smectite, Fe and Mn (Table 6), but we are not sure 

 of the significance of these correlations, because 

 no significant covariance has been observed be- 

 tween Fe, Mn and the clay size fraction (of which 

 latter smectite is of course a part). Assuming that 

 the Fe-Mn-Smectite correlations are true, it fol- 

 lows that some of the Fe and Mn is associated with 

 either adsorbed or exchangeable sites of smectite, 

 and/or in basic lattice (octahedral, presumably) 

 positions in smectites. The work of Anderson and 

 Reynolds (1966) with the Umiat Bentonite does 

 not suggest any appreciable nontronitic character 

 of that material. Of course, this does not preclude 

 the occurrence of nontronite clays elsewhere in 

 the Colville River drainage area. However, we 

 have observed in our detailed work (refer to Ap- 

 pendix for details) that the sediments at various 

 sample locations along the Colville River contain 

 varying amounts of ferruginous materials, much 

 of it of an X-ray amorphous "limonitic" character, I 

 intimately associated with the clay minerals. 



It seems quite probable that Mg, along with Ca, ■ 

 is related to the argillaceous and/or carbonate ^ 

 fraction of the deltaic sediments. Preliminary at- 

 tempts made to understand the partition pattern of 

 Co (Naidu, 1972) suggest that Co is being 

 scavenged by ferric hydroxide. 



The geochemistry of Cu in the deltaic sedi- 

 ments is not well understood, especially because 

 of its positive correlations restricted to Ca, Mg, K, 

 Li, Rb, illite, and carbonate fraction of sedi- I 

 ments. A natural conclusion of such an associa- 

 tion would be that Cu is chiefly bound with the 

 carbonate phase, and possibly to a limited extent 

 with the argillaceous fraction. Limestone and 

 dolomite grains of terrigenous origin probably 

 constitute the major portion of the carbonate ma- 

 terial in these sediments. Copper is likely to be 

 present in these grains as fine disseminations, 

 probably of discrete sulphide phases. Prelimi- 



250 



