LOEB and RO.I.XS ICHTHYOPLANKTON COMPOSITION AND ABUNDANCE 



notus. Bathylagus nigrigenya , and Vinciguerna 

 lucetia, have a concordance of higher abundances 

 in 1972, 1973, and 1983 vs. other years (W = 0.81, 

 P<0.01). The abundance rankings of B. ni- 

 grigenys are strongly correlated with those of D. 

 laternatus and V. lucetia (p = +0.82, P < 0.01 in 

 both cases); the correlation between D. laternatus 

 and V . lucetia is not significant. 



The abundance rankings of sardine and Mycto- 

 phum nitidulum (Pair I species) are significantly 

 correlated (p = +0.96, P<0.01); both species 

 were rare or absent in 1970 and most abundant in 

 1972, 1973, and 1983. 



With adjustment to 1973 and 1983 abundances 

 of T. mexicanus, its abundance rankings are 

 strongly correlated with those of Lampanyctus 

 parvicauda (Pair II species; p = +0.84, P < 0.01). 

 Highest relative abundances of both species were 

 in 1964, 1966, 1967, and 1969. 



Three relatively frequent species iTrachurus 

 murphyi, Diogenichthys atlanticus , and Diaphus 



sp.) do not conform to any of the above patterns. 

 If the 1973 Trachurus abundance is adjusted to 

 accommodate for possible undersampling, its 

 abundance pattern is similar to that of anchoveta 

 (p = +0.68, P = 0.05) but not to any of the other 

 Group I species (p = +0.03 to +0.46). 



VARIATIONS IN ABUNDANCE AND 



COMPOSITION RELATIVE TO 



HYDROGRAPHIC CONDITIONS 



Ichthyoplankton abundances in the 10 years 

 sampled show no consistent patterns relative to 

 warm water-cold water events (Table 8). High PL 

 and OL abundances occurred during strong El 

 Nino events (1972 and 1983) and during cold or 

 transition years immediately following El Nihos 

 (1966 and 1973). Lowest abundances of both frac- 

 tions were associated with the 1965 El Nino and 

 warm 1968. Neither the PL nor OL have signifi- 

 cant correlations with ranked (high to low) July- 



Table 7. — Larval fish taxa grouped according to similar across-year (1964-83) ranked abundance patterns. 

 Significant agreement of group rankings indicated by Kendall's concordance {W) values. Correlations between 

 species pair rankings indicated by Spearman's rhio (p) values. Significant values at P £ 0.05 are indicated, but 

 note use of multiple testing. " = abundance ranks adjusted to accommodate for possible sampling deptti 

 related catch differences (Table 2). 



13 



