2 
PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES 
Series 4, Volume 65, 28 Sept. 2018, No. 1 
Only species of the subgenera Serranus and Paracentropristis are known from the eastern 
Atlantic: one species is endemic to the central Atlantic islands of Ascension and St. Helena and 
nine valid species are known from the eastern Atlantic. Heemstra and Anderson (2016) recorded 
eight species of Serranus in their chapter on the Serranidae in the revised edition of the FAO series 
The Living Marine Resources of the Eastern Central Atlantic (Carpenter and Angelis, 2016), 
including among them Serranus africanus (Cadenat, 1960), which we here treat as a member of the 
genus Chelidoperca. Williams and Carpenter (2015:287), in describing Chelidoperca santosi from 
the Philippines, did not include C. africana in their list of the seven nominal species of the genus, 
noting that “an eighth nominal species from the Atlantic has been reassigned to the genus 
Serranus .” That eighth species is presumably C. africana. Heemstra and Anderson (2016) did not 
include S. pulcher Wirtz and Iwamoto, 2016, because it was described after their FAO publication. 
We describe herein one new species from the Gulf of Guinea island of Sao Tome and a second from 
Angola and Gabon. Thus, a total of ten species of Serranus is now known from the eastern and cen¬ 
tral Atlantic. Four of the ten have distributions extending into the Mediterranean Sea, and 
S. cabrilla, with the widest range, is also recorded from the Black Sea, as well as in the Red Sea 
where it apparently arrived as an invasive species (Norman 1927; Tortonese 1954). 
Heemstra and Anderson’s (2016) contribution to the FAO Species Identification Guide to the 
Marine Resources of the Eastern Central Atlantic includes descriptions and illustrations of seven 
Serranus species plus Chelidoperca africana. We have incorporated information from that work 
and added additional information based on our findings. For some species, new collections have 
extended their geographical ranges; extensive diving observations by the second author have 
increased our knowledge of the life history and habits of the species. Color photographs of living 
and fresh specimens, of which we provide many, are of particular value in showing the wide vari¬ 
ation in color pattern found in some species. Chelidoperca africana Cadenat, 1960, is included in 
the key because it is frequently treated as a member of the genus Serranus. 
Material and Methods 
Specimens examined are deposited in the California Academy of Sciences (CAS and 
CAS-SU), the National Museum of Natural History (USNM), the Natural History Museum of Los 
Angeles County (LACM), Royal Belgian Institute of Natural Sciences (RBINS), South African 
Institute of Aquatic Biodiversity (SAIAB), Stuttgart Natural History Museum (SMNS), the Zoolo- 
gische Staatssammlung, Miinchen (ZSM). 
Anatomical terms and abbreviations, and methods for making measurements and counts 
follow those described in numerous ichthyological texts, including Hubbs and Lagler (1964), 
Smith and Heemstra (1986), and the various FAO species identification guides for fishery purpos¬ 
es (e.g., Carpenter 2002:603-610; 2016:1513-1518). The last ray of the dorsal and anal fins is usu¬ 
ally split to the base and is counted as one ray; almost all eastern Atlantic species have seven soft 
rays in the anal fin, although S. atricauda is an exception with a normal of eight, S. hepatus has six 
or seven, and S. drewesi has six. Pectoral-fin ray counts include the small splintlike uppermost ray. 
Pored lateral-line scales are taken to the caudal-fin base and do not include those on the caudal fin 
itself; the value of lateral-line and circumpeduncular scale counts as applied to Serranus was 
discussed by Robins and Starck (1961:260). Gill-raker counts are from the first arch and include 
all rudiments; counts of the upper arm are separated by a plus sign (+) from those of the lower arm, 
the raker whose root spans both arms is included in the count of the lower arm. Meristic characters 
provided in the text but not in the tables are ours and do not include counts from the literature that 
are extralimital to the ranges we recorded. The descriptions of color patterns are based on numer- 
