of nondetrital manganese varies between 0.01 

 and 2.62 percent, and nondetrital Mn/Fe ratio 

 varies between 0.02 and 2.15 (table 14). The 

 mean values for iron, manganese and Mn/Fe 

 ratio are highest in the East Novaya Zemlya 

 Trough and lowest in the Voronin Trough, 

 with the Svyataya Anna Trough having values 

 more similar to the Voronin than to the East 

 Novaya Zemlya Trough (table 3). All values 

 show a high degree of variability within a 

 trough but iron shows the least variability 

 (tables). 



The data for the East Novaya Zemlya 

 Trough are subdivided because the E-series 

 samples represent the top centimeter of each 

 sediment core and the N-series represent stra- 

 tigraphic channel samples encompassing as 

 much as 15 cm. of the top of a core. The signif- 

 icance of the difference in these two sets of 

 samples is related to the distribution of iron 

 and manganese with depth in the cores (fig. 8). 

 Surface samples which include more than the 

 upper centimeter or two of sediment may in- 

 clude iron and manganese from layers just 

 below the surface which are greatly enriched 

 in these elements. 



The data from Gorshkova (1957) for the 

 southwestern Kara Sea (tables 3 and 14) 

 which include several samples from the East 

 Novaya Zemlya Trough and which represent 

 samples from a wide range of water depths, 

 tend to support the earlier suggestion that the 

 reagents used in the present study extract only 

 a small part of the total iron but almost all of 

 the total manganese. 



The regression analyses of the geochemical 

 data revealed no significant (at the 95 percent 

 confidence level) surfaces higher than linear. 

 Thus it is possible to express a trend as a com- 

 pass direction with the dependent variable in- 



creasing in that direction. 



The content of nondetrital iron in surface 

 sediments from the East Novaya Zemlya and 

 Svyataya Anna Troughs increases towards the 

 southeast (128° to 131°). In the Voronin 

 Trough, nondetrital iron increases towards the 

 south (192°). The residuals do not appear to 

 have any geological significance (fig. 9). 



The similar trends observed for nondetrital 

 iron in each trough are not observed for nonde- 

 trital manganese (fig. 10). Each trough dis- 

 plays a distinctly different trend direction. 

 Manganese increases towards the coast of No- 

 vaya Zemlya (287°) in the East Novaya Zem- 

 lya Trough, towards the south (186°) in the 

 Voronin Trough and towards the Central Kara 

 Plateau (95°) in the St. Anna Trough. The re- 

 siduals reveal a large area in the central St. 

 Anna Trough which has a higher content of 

 nondetrital manganese than the surface would 

 predict. The positive residuals which appear in 

 the East Novaya Zemlya Trough probably re- 

 flect the sampling procedure used for the seven 

 Northwind cores from this trough, i.e., strati- 

 graphic channel sampling instead of the "spot" 

 sampling used for all other cores, rather than 

 any natural phenomenon. 



The nondetrital Mn/Fe ratio (fig. 11) in- 

 creases northward (18°) and northeastward 

 (45°) towards the Arctic Ocean in the Svy- 

 ataya Anna and Voronin Troughs respectively, 

 but increases westwards (288°) towards No- 

 vaya Zemlya in the East Novaya Zemlya 

 Trough. The residuals suggest that the Mn/Fe 

 ratio is higher in the central parts of all three 

 troughs than the linear surface would predict. 

 No geological significance can be attached to 

 the latter observation for the East Novaya 

 Zemlya Trough because of the sampling dis- 

 crepancy outlined previously. 



Discussion 



Although color is one of the most readily ob- 

 served mass characteristics of a sediment it is 

 also one of the more difficult to evaluate. The 

 color of naturally moist sediments under simi- 

 lar lighting conditions depends upon (1) the 

 intrinsic color of detrital minerals of which 

 each sediment is composed, (2) the size and 

 packing of these minerals, (3) the amount and 



kind of organic matter, including skeletal ma- 

 terial, and (4) the color and amount of any 

 nondetrital constituents such as oxide coatings, 

 cement and concretions (Weller, 1960; Krum- 

 bein and Sloss, 1963). Any interpretation of 

 sedimentary color must therefore consider the 

 contribution of each of these factors to the 

 total mass effect of color. 



8 



