included larvae in the families Gerreidae, Haemulidae, and 

 Nomeidae. 



The task of producing a reliable and equitable 

 ichthyoplankton data base required extensive procedures to 

 verify, correct, and edit the original identifications. The 

 primary data source was the original identification sheets (see 

 Kramer et al., 1972, for examples); however, a critical resource 

 used in all phases of this process was the CalCOFI 

 ichthyoplankton collection in which the samples are archived. 

 Throughout the course of CalCOFI ichthyoplankton studies, samples 

 have been identified to the lowest taxon possible. In reviewing 

 these identifications for the data base, our approach has been 

 conservative and we have preserved those identifications and 

 counts which we could confirm, while correcting as many of the 

 errors as possible. During the coding of the identification 

 sheets, the "descriptive types" were assigned scientific names 

 and reexamined, if necessary. After computer entry, taxonomic 

 errors and inconsistencies in the data base were corrected and 

 the most obvious identification errors were corrected. Our 

 current knowledge of ichthyoplankton techniques coupled with a 

 precise understanding of the development of identification 

 competency in the program over the years allowed us to critically 

 judge the historical records. Identifications were changed to 

 different taxa, lumped to a higher taxonomic category, or given a 

 more precise taxonomic name. In many cases, identifications of a 

 taxon were inconsistent among cruises in a year, because of 

 varying competency of identifiers. These records were made 

 equitable by lumping to the higher taxonomic category to avoid 

 biases that could result in quantitative misinterpretations. 



Next, statistical, seasonal, and geographic outliers were 

 identified, employing a series of graphic summaries and listings. 

 Examination of geographic outliers proved to be especially 

 effective because of our accumulated knowledge of species 

 distributions. In the course of examining samples for these 

 outliers, other identification errors were discovered and 

 eventually all taxa were scrutinized to some extent. Lastly, 

 certain taxa were reexamined in all samples for the entire 

 CalCOFI time series. These taxa were selected because of their 

 commercial, ecological, phylogenetic, or zoogeographic importance 

 or because taxonomic confusion was at the ordinal level. The 

 following is a list of the taxa for 1957 which received special 

 attention, with explanations and caveats intended to aid in 

 quantitative interpretations: 



Anguilliformes - tentative and sporadic identifications to family 

 or lower taxon lumped to order. 



Sardinops sagax - all specimens south of line 120 checked for 

 misidentif ication of Opisthonema spp. 



Engraulidae - includes nearshore taxa (mostly Anchoa spp.) large 

 enough to separate from Engraulis mordax . Some nearshore 



