488 



Fishery Bulletin 91(3). 1993 



40 67 22 97 74 124 75 98 29 80 25 88 125 51 56 110 34 



Station at Middle of Window 

 Figure 1 1 



Example output from gradient analysis performed by GRADSECT, with the 1986-87 

 AMLR survey data sorted by depth. 



the populations of these species 

 than will those of the summer. 

 To better understand this ecosys- 

 tem, information on the seasonal 

 dynamics of the community and 

 its constituent populations (in- 

 cluding krill and benthic inver- 

 tebrates) is needed. 



Acknowledgments 



My thanks go to S. Saila, J. 

 Heltshe, and K. Sherman for 

 their technical and editorial as- 

 sistance. Supporting research for 

 this study was funded by grants 

 from the National Oceanic and 

 Atmospheric Administration 

 through the Antarctic Marine 

 Living Resources Project of the 

 National Marine Fisheries Ser- 

 vice. 



resource (e.g., krill) without experiencing competition 

 (McKenna, 1991). However, during the long winter, 

 resources may be limiting, causing strict resource par- 

 titioning, and more highly defined community struc- 

 ture may develop. If this is the case, the events of the 

 austral winter will be more important to regulation of 



Literature cited 



Belyaev, G. M., and P. V. Ushakov. 



1957. Certain regularities in the quantitative distri- 

 bution of bottom fauna in Antarctic waters. Doklady 

 Akad. Nauk, SSSR. 112(1): 137. 



53.08 



53.88 



DEGREES 



SOUTH 



LATITUDE 



54.67 



55.58 



'88 n 



'89 Tl»b 



'88 b 



-r-r 



'8b n«b 



43.88 



42.08 



41.88 



40.88 39.80 38.88 



DEGREES UEST LONGITUDE 



37.88 



36.08 



35.00 



Figure 12 



Summary of major ecotones identified by GRADSECT analysis. The horizontal lines represent the location of major breaks in the 

 latitudinal gradient. The vertical lines represent the location of major breaks in the longitudinal gradient. The dashed lines represent 

 the major breaks identified along the depth gradient. Abbreviations: '86 = based on 1986-87 survey data: '88 = based on 1988-89 survey 

 data; n = based on numerical abundance; b = based on biomass. 



