Walsh et a\ : Juvenile fish assemblages on the southeast United States continental shelf 



265 



blages and the three overlapping groups were 

 not as distinct as they were in spring or fall 

 (Fig. 3, E and F). The species composition and 

 abundance at station 7 from January 2001 was 

 dissimilar from the rest of the winter samples 

 (Fig. 3E) and generated a high eigenvalue for 

 the second dimension (Table 5). The cross- 

 shelf gradient along the first dimension had 

 the highest eigenvalue (Table 5) in the winter; 

 the first dimension showed three cross-shelf 

 groups (Fig. 3F): an inner-shelf group (stations 

 1-2), a mid-shelf group (stations 2-5), and an 

 outer-shelf group (stations 4-7). 



Species composition of juvenile fish 

 assemblages 



Five seasonal juvenile fish assemblages were 

 associated with the cross-shelf station groups 

 (Fig. 4, Table 2). The cross-shelf distribution of 

 juveniles was related to adult distribution pat- 

 terns and habitat use (Appendix). During the 

 spring, the inner- and mid-shelf assemblage 

 comprised species that are found in estuarine, 

 coastal, and open-shelf habitats as adults (Fig. 

 4A, Appendix). Syiwdus foetens fell between 

 the two station groups on the ordination, and 

 was found in the both assemblages (Fig 4A). 

 The outer-shelf assemblage was predominately 

 coastal and open-shelf species (Fig. 4A, Appen- 

 dix). Three reef-associated species were pres- 

 ent in both assemblages (Table 2). 



Eighteen juvenile reef-associated species 

 were collected during the summer (Table 2). 

 Six species (Centropristis striata, Apogon 

 pseudomaculatiis, Haemulon auroUneatum, 

 Stenotomus sp., Halichoeres bivittatus, and 

 Parablennius marmoreus) were collected in 

 June 2001 at station 2.3, when sampling was 

 conducted over live-bottom. Sixteen of the 56 

 species from the summer sampling were col- 

 lected solely during this season, and overall, 

 the entire shelf was considered a single as- 

 semblage (Table 2). 



Only estuarine and coastal species were ob- 

 served in the inner-shelf assemblage during 

 the fall (Fig. 4B, Table 2). The greatest num- 

 ber of species was observed in the mid-shelf 

 assemblage (Fig. 4B, Table 2), which consisted 

 of a mixture of estuarine, coastal, and open- 

 shelf species (Fig. 4B, Appendix). Diplectrum 

 formosum was collected from stations 2 to 7 

 and was in both the mid- and outer-shelf as- 

 semblages (Fig 4B). The outer-shelf assem- 

 blage consisted of only coastal and open-shelf 

 species (Fig. 4B, Appendix). Seven reef-asso- 

 ciated species were present in mid- to outer- 

 shelf assemblages during the fall (Table 2). 



There was a cross-shelf gradient in the 

 winter juvenile fish assemblages that was in- 



CM 



< 



Cvj 

 < 



o 



CM 



< 



O 



c\j 



< 

 O 







gi ,2.1 



6 -'24 .23 

 2 1 .,. 



C/) 



c 



CD 



""225 



2 1 



CA 1 



CA1 



Figure 3 



Correspondence analysis (CA) ordinations (portraying the first and 

 second dimension scores) of the 10% cross-shelf data set showing 

 station groups in each season; spring (A), summer (B). summer 

 without station 2.3 (C), fall iD). winter (E), and winter without 

 station 7(F). The cross-shelf station groups were identified within 

 each season. Solid lines enclose the boundary of each station group 

 with three or more stations. Station groups comprising one or 

 two stations are not enclosed by a solid line. Each station group 

 is labeled and identified by a different symbol. The dashed lines 

 intersect at the origin of the plot. Analyses were conducted by 

 using juvenile fish abundance data only. Data from each cruise 

 within a season are shown together. 



