Notes on the of Tahiti, with Description of a New Serranid Genus'

JOHN E. RANDALL

DURING 1956 and part of 1957 the author car­ U.S. National Museum (USNM) , with the as­ ried out research on the biology of groupers sistance of Leonard P. Schultz and others of the (Epinephelinae; ) and snappers (Lut­ Division of . Dorothea B. Schultz made janidae) in the Society Islands, with the support the drawing of Epinephelus truncattts Katayama. of a fellowship from Yale University and the Bernice P. Bishop Museum. The following 12 STATUS OF albomarginatus species of groupers were discussed in a paper by Randall and Brock (1 960) which dealt pri­ After considerable effort a serranid fish un­ marily with food habits of Tahitian fishes: known to the author, which was sighted und er­ water on several occasions in the Tuamotu Ar­ Epinepbelus metra Bloch chip elago, was finally collected at Tetiaroa atoll Epineph elus bexagonatas ( Bloch and in the Society Islands, near Tahiti. It was ulti ­ Schneider ) mately identified as Cepbalopb olis albomargi­ Epinephelus fuscoguttatus ( Forskal ) natus Fowler and Bean (1 930 ). Th e specimen Epinephelus elongatus Schultz was deposited in the collection of the George Epinephelus fasciatus (Forskal) Vanderbilt Foundation at Stanford University. Cephalopholis argus ( Bloch and Schneider ) Th e species has been recorded on only two Cephalopholis urodelus (Bloch and occasions since the original description of East Schneider ) Ind ian and Philippine specimens , once from Cephalopholis miniatus ( Forskal ) East Africa (Smith, 1954 ) and once from AI­ Cephalopholis leopardus ( Lacepede) dabra in the Indian Ocean (Smith, 1955 ) ; thu s Variola louti ( Forskal ) a record from the Society Islands represents a Plecsropomus leopardus (Lacepede ) noteworthy range extension. Plectropomus maculatus ( Bloch) Smith elevated the subgenus A etbaloperca In addition to the above, a few specimens of Fowler ( 1904) , the type species of which is four rare groupers were collected which were Perea rogaa Forskal, to generic rank and in­ not positively identified in the field. Consider­ cluded albomarginata. Although not properly able museum research was needed to identi fy belonging in Cepbalopb olis, albomarginata fits these fishes, one of which does not fit into either no better in A ethaloperca; therefore a new genus of rwo genera in which it has been previously is proposed for this fish. classified. Further study has necessitated the al­ Graci/a, n. gen. teration of the names of two of the more com­ mon groupers in the list of 12, Epinepbelas DIAGNOSIS. One dorsal fin with nin e spines; fuscoguttatus and Epinephelus elongatus. Dis­ spines of fins slender; supramaxillary bone pres­ cussions of the four rare species and the two ent; depressible teeth in jaws, those at side of name changes are presented under separate head­ lower jaw in a single row medial to a row of ings below. fixed teeth; a single pair of enlarged canine Most museum work was carried Out at the teeth at front of each jaw; head small, the length about 3.1 in standard length; depth about 2.6 in standard length; caudal fin emarginate. 1 Cont ribution from the Institute of Marine Biology, University of Puerto Rico, Mayaguez, Puerto Rico. DESCRIPTION. Dorsal rays IX, 15; anal rays Manuscript received August 25, 1963. 1II, 9 (rarely 10 ) ; dorsal and anal spines mod-

281 282 PACIFIC SCIENCE, Vol. XVIII, July 1964 erately slender; dorsal fin unnorched, the spinous The latter is distinctive in the steep dorsal pro­ portion of fin equal in basal length to soft por ­ file of the head ( the snout forms an angle of tion; body compressed, the width about 2 in nearly 60 degrees to the horizontal; the snout head length and 6.5-7 in standard length; head angle of Gracila is about 40 degrees) and its small, about 3.1 in standard length; body depth deep body (d epth about 2.2 in standard length ) . moderate, about 2.6 in standard length; supra ­ Also, the teeth at the side of the lower jaw of maxillary (supplemental) bone present ; mouth A ethaloperca (and most species of Cepbalopb­ moderately large, the m axillary extending olis ) occur in more than two rows (except slightly posterior to eye; a pair of enlarged posteriorly on the jaw ) ; in Graeila there are canine teeth anteriorly in each jaw separated only two rows of teeth, the outer fixed and the by a broad gap ( two adjacent canines may be inner depressible. Also, Gracile has much smaller present in place of a single canine ) , and a single anal spines. row of smaller canines on sides of jaws; depress­ ible canine teeth medial to anterior canines and Gracila albomarginata in one medial row on side of lower jaw; upper Figs. 1, 2 jaw with a band of villiform teeth medial to Cephalopholis albomarginatus Fowler and fixed canin es at side of jaw; a band of villiform Bean, 1930. Bull. U.S. Natl. Mus. 100, p. teeth on palatines and on vomer (in V-shape on 235, fig. 11 ( type locality: Danawan Island, latter ) ; tongue smooth ; opercle with three flat vicinity of Sibuko Bay, Borneo ). spines, middl e one closest to lower; edge of A ethaloperca albomarginata: Sm ith, 1954. preopercle rounded with only a slight indenta­ Ann. Mag. N at. Hi st., ser. 12, vol. 7, p. tion above angle; upper preopercular margin 925, pl. 33 A (first record from Indi an finely serrate, lower margin smooth.margin of Ocean ). interopercle and subopercle finely serrate; head, including maxillary, scaled; scales moderately One specimen: 257 mm standard length, 318 small (about 110 vertical rows between upper mm total length, Tetiaroa, Society Islands,Feb. end of gill opening and end of hypur al ); scales 16, 1957, J. E. Rand all. Natural History Mu­ seum, Stanford University, uncatalogued. ctenoid except on head, thorax, abdomen, and The following counts, measurements, and ob­ anterodorsally on body; lateral line single, con­ servations were made from the fresh specimen : tinuous to base of caudal; posterior nostrils dorsal rays IX, 15; anal rays III, 9; pectoral rays spherical; pectoral fins obtusely pointed; caudal 19; lateral-line scales about 115; head length 82 fin emargi nate; gill rakers moderately long; the mm; body depth 87 mm; body width 48 mm ; one at angle slightl y longer than gill filaments; least depth of caudal peduncle 33 mm; diameter 24 vertebrae. of eye 14.5 mm ; interorbital space 18.5 mm; Monotypic. Type species, Cephalopholis al­ length of pectoral fin 52 mm ; lower jaw proj ects bomarginatus Fowler and Bean. 5 mm beyond upper jaw when mouth is closed; DISCUSSION. Gracila shows affinities to Cepb­ caudal concavity 13 mm ; three spines on opercle, alopholis but differs principally in external the middle located about one-third distance morphology in its smaller head ( head length of from lower to upp er spine ; preopercle entire, Cephalopholis 2.4-2.7 in standard length ) and rounded, the upper limb finely serrate; four ernarginate caudal fin (caudal always rounded canine teeth in lower jaw in two close-set pairs, in Cephalopholis). It also displays a different separa ted by a 5 mm gap at symphysis; teeth in mode of life from Cephalopholis; it charac­ upp er jaw similar (o nly one canine on one teristi cally swims well above the bottom like side ), separated by 11 mm; remaini ng teeth species of Plectropomus. Cephalopholis dwells nearly as long, more slender, depressible, and in more upon the bottom and is more retiring in a double row; narrow band of villiform teeth on its habits. Gracila differs notably from Plectro­ vomer and palatines. pomus in having 9 instead of 6-8 dorsal spines Color in life reddish brown with a suffusion and in lacking enlarged canines of the side of of orange on head, especially around mouth; the lower jaw. numerous light grayish-blue bars on sides which Graeila is also closely related to A etbaloperca. nearly disapp ear in preservative; four diagonal ?o .., Group ers of Tahiti-RAND/.LL _ 0 )

FIG. 1. Gracila albomarginata. ( Fowler and Bean ) . Ph otograph by Charles E. Cutress of a color painring from the "Albatross" Philipp ine Expedition, through the courtesy of Leonard P: Schult z of the U.S. N ational Museum .

deep blue lines ( dark brown in pres ervative ) on Th e largest specimen examined is the holo­ head as in Figure 1 of an Indo-Malayan speci ­ type (USNM 89985), 295 mm in standard length. men, except thr ee upper lines not continuous (uppermost line broken into two long sections TAHITI RECORD OF Epinephelus truncatus and a spo t; second line, from beneath eye, into a long section and · two spots; and the third Two groupers bearing some resemblance to from maxillary groove into three spots). In Epinephelus fasciatus were purchased in the preservative the edges of the fins were noted to market in Papeete, Tahiti, in 1957. They differ have become white. They may have been reddi sh from fasciatus notably in having a shorter max­ in life, as was described from Afr ican specimens illary (maxillary of fasciatus extends to beneath by Smith ( 1954) . posterior edge of eye), broader lips, and an Th e single specimen was collected with a opercular flap with an angularly rounded upper spear a t a depth of 60 fr on the seaward side of margin ( nearly straight on fasciatus ) . The two the atoll of Tetiaroa, When seen und erwater, G. fish were caught by Tahitian fishermen with albomarginata has two large white saddle-like hook and line at an unknown depth, but prob­ areas on the back which are its most striking ably in excess of 100 ft, judging from the de­ color markings. Before the speared specimen gree of distens ion of the stomach into the mouth from an expan ded air bladder. The speci­ was brought to the surface, however, the white mens were believed to represent an undescribed areas had disappeared and did not reapp ear. species. In the same year that they were collected Th e species was observed but not collected at a paper by Katayama app eared that described the arolls of Takaroa, Takapoto, and Tikahau the species from islands near Japan as Epi­ in the Tuamotu Archipelago. It was seen outside nephelus trun catus. reefs and in passes, usually at depths of about 40 to at least 120 ft, but was never sighted in Epinephelus truncatus lagoons. Only once was it observed in water less Fig. 3 than 30 ft deep, and when this individual was pursued it retreated into deeper water. Epinephelus truncatus Katayama, 1957. Jap. 284 PACIFIC SCIENCE, Vol. XVIII, July 1964

Jour. Ichth., vol. 6, p. 158, fig. 4 (type ranee from pelvic tips (when fins applied to locality: Izu Islands and Bonin Islands). abdomen ) to origin of anal fin; first dorsal spine Epinephelus truncatus Katayama , 1960. Fauna nearly one-third length of second, and second Japonica Serranidae (Pisces) , pp. 66, 77, about three-fourths of third; third, fourth, and pI. 47. fifth spines the longest and nearly equal in length; remaining spines progressively shorter, Two specimens : 222 and 238 mm standard the last only slightly longer than second; front length, Papeete market, Tahiti, May 28, 1957, of upper jaw with twO large adjacent canines on J. E. Randall. USNM 75400 . each side, each of these pairs separated by a gap Dorsal rays IX, 16 (injury to rear base of fin about equal to half eye diameter; a single row of of one fish); anal rays III, 8; pectoral rays 19; 15 small canine teeth along side of upper jaw, pored scales in lateral line 68 (70); vertical posterior to larger canines; lower jaw with one scale rows from upper end of gill opening to 'or a pair of large fixed canines on each side at base of caudal fin 123 (125 ); scale rows above front separated by a gap about equal to one­ lateral line 14 or 15; scales below lateral line to third eye diameter; large canines followed by a origin of anal fin 28 or 29; gill rakers 7 + 1 +15(8+1+15). series of about 40 smaller ones extending the Caudal fin trun~ate; greatest depth of body length of the jaw; bands of slender depressible at origin of dorsal fin, 'the depth contained 3 canines medial to fixed canine teeth in both times in standard length; maxillary reaches .upper and lower jaws; teeth on vomer and slightly posterior to a vertical at center of eye; palatines; tongue smooth. middle opercular spine much larger than other Color when fresh dull orange, the centers of two, posterior in position, and slightly closer to scales light 'greenish gray (centers now darker the lower than the upper spine; opercular mem­ than edges in preservative) ; five faint, broad brane moderately pointed, reaching posterior vertical bars on body which faded shortly after to a vertical at origin of pelvic fins; preopercular preservation; head mottled with light red and margin rounded, with a slight indentation above orange-brown; edge of orbit, except anrero­ angle; 42 (48) serrations on vertical margin of ventrally, bright red (now pale in alcohol); a preopercle, the last four enlarged; dermal flap pale blue line adjacent to red rim of orbit and on maxillary (continuous with upper lip) nearly surrounding entire eye (blue line now dark in reaching upper edge of maxillary; base of alcohol); caudal fin orange-red except olive dorsal and anal fins moderately fleshy; origin of green on upper fifth; dorsal fin olive green, the dorsal fin slightly posterior to a vertical at upper outer third of spinous portion light red basally base of pectoral fin; anus about one-third dis- and dark red distally (now pale with a blackish

FIG. 2. H olotype of Gracila albomarginata, 295 mm standard length, Borneo (USNM 89985). X ray by Victor G. Springer of the U.S. National Museum. Groupers of Tah iti-RANDALL 285

diagonal streak ) ; a few diagonal orange lines in part of the membranes of the spinous porti on of olive green in basal part of fin; rays orange in the dorsal fin of the holotype are much better de­ soft portion of dorsal; extreme base of fin with veloped ( there are only traces of pigment dis­ a red band; remaining fins light orange-red. tally on the spinous dorsal of the Tahitian Katayama described the color of his speci­ specimens ). There is no white caudal margin mens as dark red, each scale with a basal spot of and dark submarginal band on the specimens greenish brown; membranes between dorsal from Tahiti (possibly overlooked when speci­ spines tipped with black; other parts of dorsal mens were fresh ). The border caudal markings olive; upp er several rays of caudal olive, the are now very faint on the holotype. Th e pale rest of the fin red, with a dark submarginal edge and narrow dark submarginal line around area and a white edge; other fins red. the eye are identical on all thr ee fish. Th e slight Because some differences seemed to exist differences in morphology and color are probably between the Tahiti specim ens and the original due to the disparity in size of the specimens description given for truncatus, a request was under comparison or are within the range of made of Katayama for comparative material. variability of the species. Some differences might He sent the holotype, a well-preserved specimen be expected of fish collected at such distant 319 mm in standard length. There is now little localities as Japan and Tahiti. doubt that his specimen and the tWO from Tahiti are conspecific. Meristic data are essentially the ON THE VALIDI1Y OF Cephalopholis obtusaurus same. The only morphological differences noted are the slightly more convex interorbital space The same fisherm en who caught the two . and more rounded pectoral fins of the specimens specimens of Epinephelus truncatus from rela­ from Tahiti. The black areas in the pale distal tively deep water off Tahiti brought another

FIG. 3. Epinephelus truncatus Katayama, 238 rom standard length, Tahiti (USNM 75400 ) . Drawing by Dorothea B. Schultz. 286 PACIFIC SCIENCE, Vol. XVIII, July 1964

FIG. 4. Holorype of Cepbalopbolis aurantius (Cuvier and Valenciennes) , 165 mm standard length , Sey­ chelles ( MNHN 765 ).

.small to market which proved difficult scales 53 ( includes 3 beyond hypural ); vertical to identify: At first it was believed -to be Cepb­ scale rows from upper end of gill opening to alopholis aurantius ( Cuvier and Valenciennes ) end of hypural about 113; gill rakers 9 + 1 + ( 1828 ). Although this species had not been 15; depth of body 2.83 in standard length; head recorded previously from the Society Islands, length 2.5 in standard length; width of body another, Serranus roseus Cuvier and Valenci­ 2,3 in head length; greatest diameter of eye 6 ennes with a type locality of Tahiti, has been in head length; interorbital space moderately placed in the synonymy of aurantius by authors convex, the fleshy width equal to vertical diame­ such as Boulenger ( 1895). Cuvier and Valen­ ter of eye; maxillary reaches slightly posterior cienn es described roseus from a painting by to eye; middle opercular spine slightly nearer Parkinson in the library of Banks. A. C. Wheeler lower than upper spine; opercular flap pointed; of the Briti sh Museum (N atural History) kindly preopercular margin broadly rounded with a sent a photograph of Parkinson's painting. The slight indentation at angle; upp er limb and in­ fish is obviously variola louti ( Forskal} ; thus dentation of preopercular margin finely denticu­ it has been improperly placed in the synonymy late; small scales on head, including maxillary of aurantius. and mandible; scales on head and anteriorly on M. 1. Bauchot of the Museum National body cycloid; a pair of adjacent canine teeth in d'Hisroire Narurelle in Paris sent the' holorype upper jaw (one missing on one side) separated of Serranus aurantius ( Fig. 4) so that it might by a space equal to .8 eye diameter; comparable be compared with the specimen from Tahiti. It canines in lower jaw separated by about .3 eye was soon evident that the tWO are not the same diameter; bands of villiform teeth in jaws broad species. anteriorly ; narr ow V-shaped band of villiform Th e following observations were made of teeth on vomer, and narrow band on palatine in the holorype of aurantius, which is 165 mm in length equal to.5 eye diameter; longest gill standard length and 201 mm in total length: raker at angle, about 1.5 times longer than gill dorsal rays IX, 15 (last ray composite, first filaments ; pectoral fins reach a vertical at origin unbranched); anal rays III, 9 (last ray compos­ of anal fin; tips of pelvic fins reach slightly ite); pectoral rays 18 (all branched except upper­ beyond anus; color in alcohol uniform pale most ; 15 branched caudal rays; pored lateral-line yellowish; a black submarginal band at posterior Groupers of Tahiti-RANDALL 287 border of caudal fin, its width about one-third Cephalopholis obtusaurus Everm ann an d diamet er of pupil; narrow blackish margin on Seale, 1907. Bull. Bur. Fisheries , vol. 26, p. rear half of soft portion of dorsal and all of 77, fig. 12 ( type locality: Bacon, Sorsogon, soft portion of anal fins; a trace of dark pigment Luzon, Philippine Islands). marginally on tips of pelvic fins. Cuvier and Valenciennes described the color One specimen : 185.5 mm standard length , of the body and fins of aurantius as orange-red; 228 mm total length, Papeete market, Tahit i,. without spots or bands. Bleeker (1873-1876:37, May 28, 1957, J. E. Randall. USNM 175409. pI. 248, fig. 3) portrayed the species in color. It Dorsal rays IX, 15; anal rays III, 9; pectoral is rose-orange with small round blue spots on rays 17; pored lateral-line scales 52 (includes: the head and the back beneath the spinous 3 beyond hypural ) ; vertical scale rows from portion of dorsal fin; there is a broad yellow upper end of gill opening to end of hypural103; posterior border on the medial fins; the caudal gill rakers 7 + 1 + 13. fin has a black line within this yellow border. ' Depth of body 2.77 in standard length ; head The Tahiti specimen was finally identified as length 2.50 in standard length ; width of body Cephalopholis obtusaurus Evermann and Seale 2.3 in head length; eye 5.1 in head length; ( 1907 ) , after examination of the holotype in bony interorbital nearly flat, the bony width 7.8 the U.S. National Museum . This species has not in head length; least depth of caudal peduncle been recognized by authors after Evermann and 2.74 in head length; pectoral fins 3.83 in stand­ Seale. Fowler and Bean (1930), Herre ( 1953 ), ard length; pelvic fins 4.52 in standard length, and Katayama (1960) have all placed it in the the tips reaching anus; third dorsal spine the synonymy of aurantius; however, obtusaurus longest; its length 3.36 in head length; first appears to be valid. dorsal spine half as long as third; tenth dorsal soft ray the longest, its length 2.24 in head Cephalopholis obtusaurus length; second anal spine the longest, its length Figs. 5,6. 3.04 in head length ; fifth anal soft ray the long-

FIG. 5. H olotype of Cepbalopbolis obtusaum s Evermann and Seale, 191 mm standard length , Philippine Islands ( USN M 559'10 ) . . 288 PAOFIC SOENCE, Vol. XVIII, July 1964

FIG. 6. Cepbalopbolis obtusaurus Evermann and Seale, 185.5 mm standard lengrh, Tahiti (USN M 17540 9 ) . est, its length 1.95 in head length; dorsal fin ( the tips reach slightly beyond anus in holo­ very fleshy at base; maxillary reaches posterior type ) , and longer gill rakers. In view of the edge of eye; maxillary scaled, but scales mostly many similarities, however, such as the meristic imbedded; opercular spines flattened, not sharp, .data (all counts the same except for fewer scale the middle one closer to lower spine (which is rows from gill opening to end of hypural­ somewhat anterior) than to upper; hind part of about 95 in holotype ), obtuse opercular flap, opercular membrane truncate with rounded position of opercular spines, fleshy dorsal fin, corners; preopercular margin very finely serrate. ' and color, the two specimens are' regarded as Color when fresh: body and fins bright belonging to the same species. orange-red, the edges of scales faintly brownish, The stomach of the holotype of obtusaurus is with widely scattered small pale blue blotches; protruding into the mouth cavity, thus suggest­ head, nape, and fins with numerous small red ing that the specimen was taken from moderately spots, these more evident in faint blue areas on deep water. head and caudal fin. In alcohol the specimen is Cephalopholis obtusaurus and C. aurantius pale yellowish brown with no trace of spots. are evidently closely related species. They may Evermann and Seale described the color of the be distinguished principally by the more pointed holotype as uniform yellowish, with a slight opercular flap, more convex interorbital, smaller wash of dull brown, and stated that the fish was eye, larger mouth, and higher gill raker counts evidently uniform red in life. However, the pre­ of aurantius. Also, aurantius has the charac­ served specimen shows faint, close-set pale spors teristic black submarginal line at the posterior (about 2 mm in diameter ) on the nape. edge of the caudal fin, which is lacking in Some differences were noted between the obtusaurus. specim en from Tahiti and the holotype (which Although other specimens of obtusaurus may is nearly the same size; it measures 191 mm in be present in museums, possibly labelled as standard length and approximately 238 mm in aurantitts, the author knows of only the holotype total length; the latter measurement difficult to from the Philippines and the one specimen from obtain because the mouth is fully open ) . The Tah iti. holotype has a slightly smaller eye ( 5.64 in head length ), slightly deeper caudal peduncle ADDITION AL SPECIMEN OF Epineph eltts (3 .03 in head length ), maxillary not as broad socialis FROM TAHITI (greatest width 5.5 in head length of Tahitian specimen and 6.3 in head length of holorype) ; Epinephelus socialis longer pectoral fins, slightly longer pelvic fins Fig. 7 Groupers of Tahiti-RANDALL 289

FIG. 7. Lectotype of Epinephelus socialis (Gunther ) , 190 mm standard length , Tahiti (B M 1873.4.3.1) .

Serranus socialis Giinther, 1873. Fische der men of this grouper. A second specimen was Siidsee (J. Mus. Godeffroy) ; vol. 1, pt. 1, speared by the author in the lagoon of Takaroa p. 7, pI. 8, fig. B ( type locality: Tahiti , as in the northern Tuamotu Archipelago, at a restricted by lectotype designation below ) . depth of 15 ft. This was deposited in the col­ lection of the George Vanderbilt Foundation, One specimen: 276 mm standard length, Pa­ Stanford Uni versity. peete market, Tahiti, May 25, 1957, J. E. Ran­ E. socialis has been recorded from Mangareva dall. USNM 175407-. " ', in the southern Tuarnotus (Kendall and Rad­ Color when fresh : head and body with nu­ cliffe, 1912 ). Recently it has been listed among merous small dark brown spots, these becoming the fishes of the Marshall Islands (Schultz et aI., confluent on posterior part of body to form 1953 ). Poll (1942 :6, fig. 1) recorded two juve­ irregular horizontal lines; caudal and soft por- niles from the lagoon of Punaauia, Tahiti. ' -rions of dorsal and anal fins dark brown with white SpOtS, the anal and posterior part of THE IDENTITY OF Epinephelus fuscoguttatus caudal with a white border ; spinous porti on of ( FORSKAL) dorsal fin light brown with irregular dark brown spots; pectorals light brown with brown spots Forskal (1 775:42) described fuscoguttatus basally, dark brown distally with a few pale from the Red Sea as a variety of Perea sum ­ spots in middle of fin; pelvic fins dark brown, mana. The latter is a white-spotted grouper the rays spotted basally, the lateral edge of the now classified in the genus Epinephelus. The fin white near tip. description of fuscoguttatus is brief, but the Three syntypes of Epinephelus socialis were name has long been recognized as a species of examined in the British Museum in London. Epinephelus distinct from summana. The most Th e largest of the three specimens ( 190 mm important characteristics given by Forskal' are a standard length, 236 mm total length )(8M black spot on the dorsal part of the caudal 1873.4.3.1 ) from Tahiti is here designated as peduncle, circular reddish-brown spots, and 18 lectotype. Meristic data from this specimen are pectoral rays. as follows: dorsal rays XI, 15; anal rays III, 8; Morgans (1958) utilized the name fuscogut ta­ pectoral rays 19; vertical scale rows from upper tus for a species which Randall (1955 ) identified end of gill opening to end of hypural 103; gill as Epin ephelus horridus (Cuvier and Valenci­ rakers 9 + 1 + 16. ennes). H e applied the name Epin ephelus dispar A year and a half of collecting fishes in Tahiti (Playfair) (in Playfair and Giinther, 1866) to resulted in the taking of only the single speci- the grouper identified as fuscoguttatus by Schultz 290 PACIFIC SCIENCE, VoL XVIII, July 1964

(1953) and Randall (1955 ). The confusion in Data on the holorype of microdon were ob­ deciding to which species the name fuscoguttatus tained from M. Boeseman of the Rijksmuseum should be applied is by no means confined to the van Natuurlijke Historie in Leiden, and later authors just mentioned. the specimen was examined by the auth or. This Information was requested of J~rgen Nielsen grouper ( RMNH 5510 ) is now 400 mm in of the Uni versiterets Zoologiske Museum in standard length and 490 mm in total length. Copenhagen on the type of fuscoguttatus. He Although in good condition, it was preserved replied that the type of summana, a dried skin with the body curved; therefore, according to about 160 mm in standard length and still show­ Boeseman, the original length was probably a ing the characteristic white spots of the species, few centimeters longer. The specimen was pur­ is extant, but no type material of variety fus­ chased at auction in 1879, and since Bleeker coguttatus is available. (1 873- 1876) indicated that he had only a The two species so long confused under the single 510 mm example of the species, there is one name fuscoguttatus have numerous small little doubt that the Leiden specimen is the true reddish-brown spots and a prominent black halotype. Because of the curvature of the body, saddle-like marking on the caudal peduncle. The the photograph supplied by Boeseman (Fig. 8 ) one character given by Forskal which suggests shows the head foreshortened. The two broad one of the two species and not the other is 18 dark bars on the body are not pigmented regions pectoral rays. The name fuscoguttatus is there ­ but shadows from large wrinkles resultin g from fore restricted to the species with 18 or 19 an attempt to straighten the body before the pectoral rays, which is in agreement with the photograph was taken. A second illustration of use of the names by Morgans, the most recent microdon (Fig. 9) is provided from a photo­ revisor (he did not, however, record pectoral graph of a smaller specimen from the Phoenix ray counts) . Islands. Th e oldest available name for the other spe­ E. fuscoguttatus and E. microdon may be cies app ears to be Serranus microdon Bleeker distinguished as follows: fuscogutta tus, pectoral ( 1856) ( type locality, Java); thus Epinephelus rays 18 or 19; gill rakers on lower limb of first microdon replaces the name Epinephelus dispar arch 17-20 (i ncluding rudiments but not raker ( Playfair) used by Morgans. at angle ); dorsal profile of head with an inden-

FIG. 8. Holorype of Epinephelus microdon (Bleeker), 400 mm in standard length , Java ( RMNH 5510). Photo supplied by M. Boeseman of the Rijksmuseum van Naruurlijke Hisrorie at Leiden. Groupers of Tahiti-RANDALL 291

ration above hind edge of eye; maxillary extends which can be confused with fuscoguttatus and from % to 1 eye diameter posterior to eye microdon. It is distinguished by the lack of small (seven specimens: 120-334 mm standard length , spots on the body, many rows of teeth at the Red Sea, Zanzibar, and Gilb ert Islands); micro ­ front of both jaws, subequal nostrils, convex in­ don, pectoral rays 16 or 17 (usually 17); gill terorb ital, and large size ( it is believed to attain rakers on lower limb of first arch 15 or 16; a weight of at least 240 lb ) . In the author's dorsal profile of head smoothly convex; maxillary opinion tukula is the same as Playfair 's Serranus ends beneath hind edge of eye or extends up to dispar variety a. Since Morgans restricted dispar half an eye diameter posterior to eye ( nine to variety b, his new species tukula is valid specimens: 153-350 mm standard length; from whether variety a is the same or not (unless an Red Sea, Zanzibar, Gilbert Islands, Marshall earlier name is found). Islands, and Phoenix Islands ). In view of the absence of a holotype and the There is a slight difference in the depth of the instability of the classification of this complex body, microdon being the more slender form on of serranid fishes, a neotype is herein described the average (depth 2.7-3 in standard length; of fuscoguttatus, based on a specimen from the that of fuscoguttatus is 2.6-2.9) . The dorsal soft Red Sea, the type locality. rays of fuscoguttatus are 14 (rarely 15) and those of microdon 14 or 15 (usually 15) . NEOTYPE OF Epin ephelus fuscoguttatus ( FORS­ Morgans (1 958:656) has separated the spe­ KAL) : USNM 147594, 216 mm in standard cies in a key on color. He noted that the spots length and 267 mm in total length, collected by of microdon (at least of adults) are more regular Donald S. Erdman at the S.A.M. pier at Jidd a, in outline than those of fuscoguttatus. An addi­ Red Sea, on July 2, 1948 ( Fig. 10) . tional color difference is the natur e of the spot Dorsal rays XI, 14; anal rays III, 8; pectoral beneath the spinous portion of the dorsal fin. rays 19; gill rakers 12 + 1 + 18 ( two addi­ Adults of E. microdon have a single roundish tional small rakers between larger rakers of dark blotch center ed at the base of the fifth upper limb not counted ) ; vertical scale rows dorsal spine; fuscoguttatus has two close-set from upper end of gill opening to end of hypural blotches or a single bilobed one. plate about 127; 25 scales in diagonal row above Morgans described tukula, a third species of lateral line to origin of dorsal fin. Epinephelus from the western Ind ian Ocean The following measurements are expressed as

FIG. 9. Epinepbelas microdon (Bleeker ) , 153 mm standard length , Canton, Phoenix Islands ( USN M 115367) . 292 PAOFIC SOENCE, Vol. XVIII, July 1964

FIG. 10. Neotype of Epinephelus fuscoguttatus ( Forskal ) , 216 mm srandard Iengrh, Jidda, Red Sea (USNM 147594) . a percentage of 'the standard length: greatest at orbits; lower jaw projecting 2.5 mm anterior depth of body 38.2; width of body just behind to upper; middle opercular spine nearer lower gill opening 18.5; head length 41.4; snout length than upper spine (7 mm separa te tip s of lower 6.1; eye diameter 6.8; postorbital length of head and middle spines and 12.2 mm separate tips of 27.8; horizontal distance of maxillary posterior upper and middle spines ); upper margin of to eye 5.4; bony interorbital space 4.9; least opercular membrane nearly horizontal anteriorly, depth of caudal peduncle 13.0; length of caudal becoming rounded as it meets vertical hind peduncle (to rear base of dorsal fin ) 9.0; snout margin (which has a protrusion slightly below to origin of dorsal fin 42.4; snout to origin of level of middl e opercular spine) ; preopercular anal fin 71.1; snout to origin of pelvic fin 39.8; margin very finely serrate, with no marked in­ length of dorsal fin base 57.2; length of anal dentatio n; scales cycloid, those on head very fin base 17.3; length of pectoral fin 22.1; length small; maxillary finely scaled; posterior nostril of pelvic spine 11.1; length of pelvic fin 20.3; subrriangular in shape, 3.8 mm in greatest length of first dorsal spine 5.7; length of second measurement ( height), with no rim; anterior dorsal spine 8.6; length of third dorsal spine nostril less than 1 mm in diameter, with a mern­ 11.3; length of fourth (longest) dorsal spi ne branous rim which is higher posteriorly; teeth 12.2; length of eleventh dorsal spine 11.4; length typical of genus, those on vomer small; teeth on of first dorsal soft ray 15.0; length of longest palatine about twice as large as vomerine teeth dorsal soft ray 18.0; length of last dorsal soft but still not large; enlarged canines of upper ray 10.2; length of first anal spine 4.3; length jaw (a close-set pair on one side, a single tooth of second anal spine 9.6; length of third anal on the other) separated by a gap of 10 mm ; spine 9.3; length of first anal soft ray 14.7; single row of small, evenly spaced, fixed canine length of longest anal soft ray 18.5; length of teeth along side of upper jaw; smaller depress­ last anal soft ray 12.2; length of caudal fin 23.6. ible teeth in upper jaw in a broad patch an­ Profile of head consisting of two convexities teriorl y on each side (up to a maximum of which meet above center of eye; interorbital flat about six irregular rows); depressible teeth of over most of its width (in center), rising slightly lower jaw in a band of more uniform width Groupers of Tahiti-RANDALL 293

(about thr ee or four irregular rows ) ; hind into fin, more obviously into spinous portion; borders of all fins well-rounded; tips of pelvic soft portion of dorsal and other fins with inter­ fins reach within 8 mm of anus. . connected round brown spots with small black Color in alcohol brown with irregular dark centers (dark centers not very evident on pec­ brown blotches ( those on upper third of body torals ); a dark brown marginal spot on each nearly black) and numerous small dark spots interspinous membrane of dorsal fin (except which are more evident on head and anteriorly cirrus of each which is pale ); other fins with on body. The irregular dark blotches are located very narrow pale margins. as follows: an elongate one begins from lower Two syntypes of Serranus horridus Cuvier half of hind edge of eye and angles upward to and Valenciennes from Java are also located at nape, meeting its fellow anterior to :origin of Leiden ( RMNH 19 and 2160) . Both are stuffed dorsal fin; five other blotches or pairs of blotches specimens, one 188 mm in standard length and may be seen middorsally on head anterior to the other 600 mm. The larger is in better condi­ one just mentioned; three irregular dark bands tion and is here designated as the lectotype. cross maxillary, lower jaw, and chin; all dark Th ese fish have 18 or 19 pectoral rays, a dark blotches and bands on head superimposed with saddle on the caudal peduncle ( not mentioned the small round darker brown spots; a bilobed by Valenciennes in the original description) , dark brown spot on back centered at base of fifth and 14 dorsal soft rays. There are two dark dorsal spine; a small dark brown spot at base saddles on the back between the bases of the of ninth and tenth dorsal spines; two irregular fourth and sixth dorsal spines, another at the dark brown spots at base of soft portion of dor­ end of the spinous portion of the dorsal fin, and sal fin and a black saddle dorsally on caudal one or two below the soft portion of the fin, peduncle; an elongate irregular dark brown S. horridus therefore seems to be a synonym of blotch below and slightly anterior to bilobed Epinephelus fuscoguttatus. spot at base of fifth dorsal spin e (lateral line runs Another fish which seems to belong in the through upper part of this blotch); a dark brown synonymy of fuscoguttatus is Serranus lutr« blotch in line with but posterior to the one just Cuvier and Valenciennes from Mauritius. The mentioned, also touching lateral line ; five irregu­ holorype (MNHN 7278) , 315 mm in standard lar brown bars on lower half of body in line length and 382 mm in total length, was exam­ with the more dorsal darker markings; dark ined by M. Blanc and R. Roux at the Museum spots along base of dorsal fin extend irregularly N ational d'Histoire Naturelle in Paris. They

FIG. 11. H olotype of Serrenus lutra Cuvier and Valenciennes ( = !uscoguttat us), 315 mm standard length ; Mau ritius (MNHN 7278) . Photo by]. Abel , through the courtesy of M. Blanc of the Museum N ational d' Hi stoire Naturelle in Paris. 294 PACIFIC SCIENCE, Vol. XVIII, July 1964 report that it has 19 pectoral rays and 14 dorsal upper third of the body than on the lower two­ soft rays. A photograph was kindly supplied by thirds. Blanc, herein reproduced as Figure 11. N o individuals of fuseoguttatus were seen or Blanc and Raux also examined the type of collected in Tahiti or other islands of French Serranus taenioehirus Cuvier and Valenciennes, Oceania. E. microdon, on the other hand, is a species which has been placed by some authors common in the Tuamotu Archipelago, although in the synonymy of fuseoguttatus; however this rare in the Society Islands. does not seem to be correct. S. taenioehirus has 16 dorsal soft rays and lacks a dark spot dorsally THE IDENTIlY OF Epinephelus tauvina on the caudal peduncle; therefore it is probably (FORSKAL) neither fuseoguttatus nor mierodon. At the request of the author, Eugenie Oark Figure 12 represents a photograph of the procured a specimen of fuseoguttatus from the halotype of Perea tauvina Forskal. a dried skin, Red Sea. The specimen ( USN M 197323) meas­ from the Red Sea. It was provided by J~rgen ures 574 mm in standard length, 700 mm in Nielsen of the Uni versitetets Zoologiske Mu­ total length, and weighs (in preservative ) 9 Y:2 seum in Copenhagen. He also supplied the fol­ pounds. It was caught by trolling in 12 ft of lowing fin-ray counts for the specimen: dorsal water off Entedebin, Dahlak Archipelago, on rays XI, 15 or 16; anal rays III, 8; pectoral rays April 5, 1962. The spots on the body were 18. orange-brown in life, more orange ventrally, and Epinephelus elongates ,Schultz was :described there is a prominent black saddle .on the caudal from specimens collected inthe Marshall Islands, peduncle. There are 19 pectoral rays, 15 dorsal Mariana Islands, Samoa -Islands, and Phoenix soft rays (the last two closely spaced), and 9 + 1 Islands. Two specimens in the N ational Mu­ + 17 gill rakers. The rakers on the upp er limb seum, 110 and 218 mm in standard length, are sessile and difficult to count. Previous counts (USNM 166985- 6 ) collected at Ghardaqa, Red of the rakers on the upp er limb of smaller speci­ Sea, by Eugenie O ark, were compared to the mens of fu seoguttatus were all 11-13. Possibly Pacific material of elongatus and proved to be there is a loss of gill rakers in larger fish because the same. Specimens of elongatus of the same of fusion. size as the . halotype of tauvina (larger speci­ This is the largest specimen examin ed by the mens have more spots than smaller ones) were author. Morgans (1958) reported fuseoguttatus comp ared with the photograph of the holotype, from East Africa to a standard length of 760 and they appear identical ; thus the decision by mm , a total length of 885 mm , and a weight of Katayama ( 1960) to refer elongatus to the 24 lb. Boulenger ( 1895) reported the largest as synonymy of tauvina seems correct. 900 mm total length. It seems obvious that it Epin ephelus tauvina is not commo n in the attains a greater size than does m ierodon, the Society Islands. The largest of 12 specimens that largest of which examined by Morgans is 465 mm in standard length, 565 mm in total length, were collected is 498 mm in standard length. In and 8 Y:2 lb in weight. the smaller sizes it can be confused with two A 334-mm specim en from Onotoa, Gilbert small dark-spotted groupers, Epin ephelus m erra Islands, collected by the author in 1951, was and E. hexagonatus. It may be differentiated colored in life as follows : light brown ish yellow from these in having 15 instead of 16 dorsal with numerous small orange-brown spots (more soft rays, 27-30 gill rakers ( total counr) on the evident on head than on body ) and large irregu­ first arch (gill rakers of metra range from 20 lar dark brown blotches (smaller spots super­ to 23 and those of bexagonatus from 23 to 27 ), imposed on the large dark blotches ) ; oneo f the and more elongate body ( depth 3.3-3.7 in irregular blotches begins behind the eye and standard length in contrast to 3.2-3.3 for the extends to the nape ; those on the body occur in other two species) . E. tauvina was observed at an irregular series of five bars, the last on the depths of 10-150 ft (few observations were caudal peduncle beginning with the dark dorsal made in deeper water) in both lagoon and Outer blotch . Th e large blotches are darker on the reef environments. Groupers of Tahiti-RANDALL 295

FIG. 12. Holorype of Epinephelus tauvina (Forskal ), 171 mm standard length , Red Sea (Z. M. Cph . 18 ) . Photo supplied by Jprgen Ni elsen of the Uni versirerets Zoologiske Museum in Copenh agen.

Smith and Smith ( 1963:15, pI. 141) de­ heretofore regarded as a synonym of Cepha­ scribed Epinephelus salonotus from the western lopbolis aurantius (Cuvier and Valenciennes) , Indian Ocean. The type, a lO-inch specimen, is synonymous with Variola louti ( Forskal ) was taken at Delgado, East Africa. The illustra­ (1775). tion depicts a 2-ft specimen from the Seychelles. 4. Cephalopholis obtusaurus Evermann and Although the description is only 21;2 lines in Seale, repr esented previously only by the type length and the figure small, there seems littl e from the Philippines, ·is resurrected from the doubt that the species is tauvina. synonymy of C. aurantius and recorded from Tahiti . 5. A second specimen of the rare grouper SUMMARY Epinephelus socialis (Gunther) is recorded from Tahiti . The first, the largest of three syntypes in 1. Cepbalopbolis.albomarginates Fowler and the British Museum, is designated lectotype, Bean, known previously from the East Indies thus restricting the type locality to Tahiti. and Philippines (original description) and the 6. The name Epinephelus fuscoguttat us (For­ western Indian Ocean, is here reported from skal) has been applied to either of two common Tetiaroa, Society Islands. It is reclassified in the groupers with reddish-brown spots and a black new genus Gracila, distinct from Cephalopholis saddle on the caudal peduncle. It is here re­ principally in having a shorter head length ( 3.1 stricted to the species with 18 or 19 pectoral in standard length) and an emarginare caudal rays, 17-20 gill rakers on the lower limb of the fin. The related A ethaloperca may be separated first arch (including rudiments but not raker at from Gracila by its steeper head profile, deeper angle ), a more marked indentation in the profile body, more than two rows of teeth at the side of the head, and a larger mouth. The oldest of the lower jaw, and larger anal spines. available name for the other species, which has 2. The range of Epinephelus trun catus Kata­ 16 or 17 pectoral rays and 15 or 16 gill rakers yama is extended from the Izu and Bonin islands . on the lower limb, is Epin ephelus microdon (type locality) to Tahiti. (Bleeker) (1856). A neotype of fuscoguttatus 3.Serranus roseus Cuvier and Valenciennes, is described. 296 PACIFIC SCIENCE, Vol. XVIII, July 1964 ·

7. Serrsnes horridus Cuvier and Valenciennes KATAYAMA, M. 1957. Four new species of ser­ and S. lutra Cuvier and Valenciennes are syno­ ranid fishes from Japan. Jap. J. Ichth. 6(4-6): nyms of fuscoguttatus, and S. dispar Playfair is 153-159,4 figs. the same as microdon. S. taeniochirus Cuvier and Valenciennes, considered a synonym of fus co~ut­ - -- 1960. Fauna Japonica Serranidae (Pis­ tatus by some authors, is neither fuscoguttatus ces). Tokyo News Service, Ltd., Tokyo. viii + nor microdon. 189 pp., 86 pIs. 8. Epinephelus elongatus Schultz and Epi­ KENDALL, W . c., and 1. RADCLIFFE. 1912. Re­ nephelus salonotus Smith and Smith appear to ports on the scientific results of the expedition be synonyms of Epinephelus tauvina Forskal, to the eastern tropical Pacific, . .. "Albatross" . .. XXV. The shore fishes. Mem. Mus. Comp. REFERENCES Zool. 35(3 ):77- 171, 8 pIs. MORGANS, J. F. C. 1958. Three confusing species BLEEKER, P. 1856. Verslag am trent eenige of serranid fish, one described as new, from vischsoorten gevangen san de zuidkust van East Africa. Ann . Mag. Nat. Hist., ser. 13, Malang in Oost-java. Nat. Tijdschr. Neder.­ 1:642-656, 3 pIs. Indie 11:81-89. PLAYFAIR, R. 1., and A. C. GUNTHER. 1866. 1873-1876. Atlas Ichthyologique des Indes Orientales Neerlandaises . . . Percoides The Fishes of Zanzibar. J. Van Voorst, Lon­ I. Vol. 7. De Breuk and Smits, Leiden. 126 don. xiv + 153 pp., 21 pIs. pp., 41 pIs. POLL, M. 1942. Les paissons de Tahiti recueillis BOULENGER, G. A. 1895. Catalogu e of the par G. A. de Witte. Bull. Mus. Roy. Hist. Nat. Fishes of the British Museum. Vol. 1. 2nd Belgique 18 (61 ) :1-20,4 figs. ed. London. xix + 394 pp., 15 pIs. RANDALL, J. E. 1955. Fishes of the Gilbert Is­ CUVIER, G., and A. VALENCIENNES. 1828. His­ lands. Aroll Res. Bull. 47:xi + 243 pp., 2 toire Naturelle des Poissons. Vol. 2. F. G. figs. Levrault, Paris. xxi + 490 pp. - - -and V. E. BROCK. 1960. Observations on EVERMANN, B. W. , and A. SEALE. 1907. Fishes the ecology of epinepheline and lutjanid fishes of the Philippine Islands. Bull. Bur. Fisheries of the Society Islands, with emphasis on food 26: 51-110,22 figs. habits. Trans. Amer. Fisheries Soc. 89 (1): FORSKAL, P. 1775. Descriptiones Animalium 9- 16. Avium , Amphibiorum, Piscium, .. . . Molleri, Hauniae. 164 pp., 43 pIs. SCHULTZ, 1. P. et al. 1953. Fishes of the Mar ­ shall and Marianas Islands. Bull. U.S. Natl. FOWLER, H . W . 1904. A collection of fishes Mus. 202, 1:xxxii + 685 pp., 74 pIs., 90 from Sumatra. J. Acad. Nat. Sci. Phila., ser. 2, text figs. 12 (4 ) :495-560, 21 pIs. SMITH, J. 1. B. 1954. Four rare serraniform - --and B. A. BEAN. 1930. Contributions to fishes from East Africa. Ann . Mag. Nat. Hist. , the Biology of the Philippine Archipelago ser. 12, 7:925-933,1 pI. and Adjacent Regions. The Fishes of the Families Amiidae, .. . . Bull. U.S. N atl. Mus. - -- 1955. The fishes of Aldabra. Part I. Ann. 100, lO:ix + 334 pp., 27 figs. Mag. Nat. Hist., sec. 12,8:304-312, 1 pI. HERRE, A. W . 1953. Checklist of Philippine - - - and M. M. SMITH. 1963. The Fishes of Fishes. U.S. Fish Wildl. Servo Res. Rep . 20. the Seychelles. Dept. of Ichthyology, Rhodes 977 pp. University, Grahamstown. 215 pp., 98 pIs.