170 



Fishery Bulletin 104(2) 



in each group, identified to the lowest taxonomic level, 

 by the area of their associated habitat patches. For the 

 larger invertebrates we recorded the geographic position 

 and estimated maximum height of each solitary sponge, 

 gorgonian. and black coral. We also noted the color of 

 black corals. We made observations on the occurrence 

 of animals (i.e., epizoids) found directly on sponges, gor- 

 gonians, and black corals, and also noted any damaged 

 or dead individuals. Voucher specimens were collected 

 to assist in taxonomic identification. 



To quantify fish-invertebrate associations we used 

 ArcGIS - to estimate the distance between each sponge, 

 gorgonian, and black coral invertebrate and the nearest 

 fish. These data were compared to the total number 

 of fishes counted in habitats that contained sponges, 

 gorgonians, and black corals by using a chi-square 

 test to look for significant differences in the frequency 

 of fish observed near corals in relation to overall fish 

 abundance. 



Results 



We completed 112 dives and surveyed 3189 habitat 

 patches (Table 1), covering 26.1 hectares at 32-320 m 

 depths (median depth of 110 m). The distributions of 

 number of patches and of surface area of habitats were 

 similar except for sand (SS), sand-gravel (SG). and 

 mud (MM) habitats, all of which had greater surface 

 areas in relation to number of patches (Fig. 2, A and B). 

 Overall, cobble-sand (CS), sand, mud, and sand-gravel 



TT RR RM BB BC BS CB MB SB CS SC SP MG SG MM MS SS 



CD 2 - 



26 1 hectares 



-l?l 



m 



ra 



kl 



FU 



I 



a 



■4 



TT RR RM BB BC BS CB MB SB CS SC SP MG SG MM MS SS 



^, 2 



n n 



TT RR RM BB BC BS CB MB SB CS SC SP MG SG MM MS SS 



Habitat type 



Figure 2 



Characteristics of habitat patches surveyed on southern 

 California rocky banks. (A) Number of patches in each 

 substratum type. (B) Total area of each substratum 

 type. (C) Mean relief (±1 standard error). See text for 

 description of method and habitat codes. See page 169 

 for definitions for the substrate abbreviations along 

 the .V axis. 



constituted the largest habitat areas, but cobble-sand 

 and sand were the most frequent habitat types. Verti- 

 cal physical structure varied among habitat types; the 

 highest structure was found in high-relief rock areas 

 (TT to BS) and lower structure was found in low-relief 

 mixed rock (CB to SP) and mixed sediment areas (MG 

 to SS; Fig. 2C). The frequency of patches of each sub- 

 stratum type varied by depth (Fig. 3). The incidence of 

 most high-relief rock categories (TT, RR, BB, and BS) 

 decreased with depth, and the occurrence of mud-domi- 

 nated habitat patches (MB, MG, MS, MM) increased 

 with depth. 



Overall, 12,360 observations were made on 521,898 

 individuals from 15 taxa of megafaunal, structure-form- 

 ing invertebrates; during these observations, estimates 

 of size and incidence of epizoic animals on 9105 sponges, 

 black corals, and gorgonians were made (Table 2). The 

 most common structure-forming invertebrates (98% 

 of total) included the crinoid Florometra serratissima 

 (40%), the brittle star iOphiacantha spp. [33%]), bra- 

 chiopods (order Terebratulida [11%]), the white sea ur- 

 chin {Lytechinus anamesus [9%]), the fragile sea urchin 



