FISHERY BULLETIN: VOL. 83, NO. 2 



(as abundances m^^), excluding those taxa which 

 significantly changed in total abundance in this 

 layer from day to night. The weighted-pair den- 

 drogram of Spearman's rank difference correla- 

 tion coefficients (Fig. 3A) shows an imperfect sep- 

 aration into profiles taken before and after the 

 storm, the first poststorm profile (#7) being more 

 like those before the storm. This is evidence 

 against the hypothesis that physical advection of 

 new populations caused all the poststorm differ- 

 ences, though it is also possible that advection 

 caused by the storm affected our site only after a 

 delay. The storm's apparent effect on the thermal 

 gradient (Fig. 1) was also delayed for some time. 

 Even with some of the migrating taxa excluded, 

 there is a partial separation in the dendrogram of 

 nocturnal from diurnal profiles. 



o 



UJ 



a: 

 o 

 (J 



ui 

 o 



z 

 < 

 a: 



1.0 

 .95 

 .9 

 .85 

 .80 

 .75 



1.0 

 .9 5 

 ,90 

 8 5 

 .8 

 .75 

 .7 

 .65 

 .60 

 .55 



DDNDD DNN NN DD 

 BBBBA AAA BB AA 

 15 6 3 7 9 10 12 2 4 1113 



N Doy or Night 



A Before or After storrr 



8 Profile number 



A 



DNDND NDNDD DM 

 8B8BB AAAAA AA 

 12 3 4 5 8 9 12 II 13 7 10 



N Day or Night 



B Before or After storm 



6 Profile number 



B. 



FIGURE 3. — Dendrograms of faunal (A) and floral (B) 

 similarities of the upper 50 m of water off Dana Point, Calif 

 Faunal assemblages are based on 39 taxa, floral assemblages on 

 126 taxa (not just those listed in Appendix). "Floral" includes 

 protozoans. All coefficients are significant at P ■: 0.001. 



B. Chlorophyll, Phytoplankton, and 

 Protozoa 



Because of the mechanisms of feeding used to 

 separate small particles of food from water, there 

 are probably no strict herbivores among the zoo- 

 plankton we studied, i.e., no animals which ingest 

 living phytoplankton without also ingesting other 



particulate organic matter Nevertheless, we used 

 the distribution of chlorophyll (see Appendix) as 

 the measure of the distribution of food for 

 particle-grazing species; in the euphotic zone of 

 the Southern California Bight, the concentration 

 of chlorophyll is closely correlated with that of 

 particulate organic carbon, with particulate ATP, 

 and (within any one season) with the chlorophyll 

 in particles >5 ^tm (Mullin and Brooks 1976; 

 Eppley et al. 1977; Mullin 1979). 



We had adequate data to answer Questions 1-4 

 from Table 2 for chlorophyll ( = "taxon"). We used 

 the phytoplanktonic and protozoan abundances 

 from the physically integrated samples for all 13 

 profiles (see Methods) to perform Tests 1, 3, and 5 

 concerning the whole 50 m water column. We re- 

 stricted Questions 2 and 4 to the upper 40 m (since 

 these taxa were rare below this depth) and used 

 data from five diurnal and three nocturnal profiles 

 in answering these questions, since only those 

 profiles were suitable for counting (see Methods). 

 Only one of the nocturnal profiles was poststorm. 

 In order to obtain estimates of "within classifica- 

 tion" variability and still maintain a balanced de- 

 sign, we reduced the ANOVA to a two-way design, 

 retaining "before vs. after storm" and "depth" as 

 classifications. Thus, diurnal and nocturnal sam- 

 ples were considered replicates (there was no evi- 

 dence of diel migration in the phytoplanktonic 

 taxa). We again restricted the analysis to the 

 upper 40 m. Variances of log-transformed data for 

 these taxa were all homogeneous in the four pro- 

 file data set (profiles 5, 6, 9, and 10). Thus we 

 applied the ANOVA to a subset of those profiles 

 suitable for nonparametric tests. 



The concentration of chlorophyll per m^ did not 

 change from day to night (Ho 1 accepted), nor did 

 the vertical distribution of chlorophyll within the 

 upper 50 m change from day to night (Ho 2 ac- 

 cepted). The median chlorophyll concentration 

 (m~^) was greater after the storm, but not sig- 

 nificantly so by Test 3. Vertical profiles of in vivo 

 fluorescence of chlorophyll and samples of phyto- 

 plankton from the fluorescence maximum layer 

 (cf. Kiefer and Lasker 1975; Cullen et al. 1982) 

 were taken from the second ship working concur- 

 rently at Dana Point. Comparison of the inte- 

 grated fluorescence profiles indicated that this 

 measure of chlorophyll increased significantly 

 after the storm (P 0.01 by a variant of Test 3). 



Inspection of the data ( see Appendix ) indicated a 

 shoaling of the chlorophyll maximum layer after 

 the storm, and this was significant by a Mann- 

 Whitney U test for differences in depth of occur- 



158 



