428

o

^ °'c%

J&

<b

^ rEs o* *

NOAA Technical Report NMFS Circular 428

Morphological Comparisons of North American Sea Bass Larvae (Pisces: Serranidae)

Arthur W. Kendall, Jr.

August 1979

U.S. DEPARTMENT OF COMMERCE

National Oceanic and Atmospheric Administration

National Marine Fisheries Service

NOAA TECHNICAL REPORTS National Marine Fisheries Service, Circulars

The major responsibilities of the National Marine Fisheries Service (NMFS) are to monitor and assess the abundance and geographic distribution of fishery resources, to understand and predict fluctuations in the quantity and distribution of these resources, and to establish levels for optimum use of the resources. NMFS is also charged with the development and implementation of policies for managing national fishing grounds, development and enforcement of domestic fisheries regulations, surveillance of foreign fishing off United States coastal waters, and the development and enforcement of international fishery agreements and policies. NMFS also assists the fishing industry through marketing service and economic analysis programs, and mortgage insurance and vessel construction subsidies. It collects, analyzes, and publishes statistics on various phases of the industry.

The NOAA Technical Report NMFS Circular series continues a series that has been in existence since 1941. The Circulars are technical publications of general interest intended to aid conservation and management. Publications that review in considerable detail and at a high technical level certain broad areas of research appear in this series. Technical papers originating in economics studies and from management in- vestigations appear in the Circular series.

NOAA Technical Report NMFS Circulars are available free in limited numbers to governmental agencies, both Federal and State. They art also available in exchange for other scientific and technical publications in the marine sciences. Individual copies may be obtained (unless otherwise noted) from D822. User Services Branch, Environmental Science Information Center, NOAA, Rockville, MD 20852. Recent Cir- culars are:

396. Whales, dolphins, and porpoises of the western North Atlantic. A guide to their identification. By Stephen Leatherwood, David K. Cald- well, and Howard E. Winn. August 1976, iv + 176 p., 168 figs., 1 table, 14 app. figs. For sale by the Superintendent of Documents, U.S. Gov- ernment Printing Office, Washington, DC 20402; Stock No. 003-020- 00119-0.

397. Marine flora and fauna of the northeastern United States. Cnidaria: Scyphozoa. By Ronald J. Larson. August 1976, iii + 18 p., 28 figs. For sale by the Superintendent of Documents, U.S. Govern- ment Printing Office, Washington. DC 20402; No. 003-017- 00387.

398. Marine flora and fauna of the northeastern United States. Higher fungi: Ascomycetes, Deuteromycetes, and Basidiomycetes. By A. R. Cavaliere. March 1977, iii + 49 p., 24 figs. For sale by the Super- intendent of Documents, U.S. Government Printing Office, Washington, DC 20402; Stock No. 003-020-00126-2.

399. Marine flora and fauna of the northeastern United States. Copepoda: Harpacticoida. By Bruce C. Coull. March 1977, iii + 48 p., 100 figs. For sale by the Superintendent of Documents, U.S. Govern- ment Printing Office, Washington, DC 20402; Stock No. 003-020-00125 - 4.

400. Fishery publication index, 1965-74. By Mary Ellen Engett and Lee C. Thorson. March 1977, iii + 220 p. For sale by the Super- intendent of Documents, U.S. Government Printing Office, Washington, DC 20402; Stock No. 003-020-00127-1.

401. Fisheries and fishery resources of New York Bight. By J. L. McHugh. March 1977, v + 50 p., 43 figs., 49 tables. For sale by the Superintendent of Documents, U.S. Government Printing Office, Wash- ington, DC 20402; Stock No. 003-020-00129-7.

402. Guide to the identification of scorpionfish larvae (Family Scor- paenidae) in the eastern Pacific with comparative notes on species of Sebastes and Helicolenus from other oceans. By H. Geoffrey Moser, Elbert H. Ahlstrom, and Elaine M. Sandknop. April 1977, v + 71 p., 40 figs., 38 tables. For sale by the Superintendent of Documents, U.S. Gov- ernment Printing Office, Washington, DC 20402; Stock No. 003-020- 00128-9.

ernment Printing Office, Washington, DC 20402; Stock No. 003-017- 0040-4.

406. Marine flora and fauna of the northeastern United States. Copepoda: Lernaeopodidae and Sphyriidae. By Ju-Shey Ho. De- cember 1977, iii + 14 p., 16 figs. For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, DC 20402; Stock No. 003-017-00412-1.

407. Distribution of decapod Crustacea off northeastern United States based on specimens at the Northeast Fisheries Center, Woods Hole, Massachusetts. By Austin B. Williams and Roland L. Wigley. De- cember 1977, iii + 44 p., 2 figs., 1 table, 57 charts.

408. Collection of tuna baitfish papers. (20 papers.) By Richard S. Shomura (editor). December 1977, iii + 167 p.

409. Marine flora and fauna of the northeastern United States. Copepoda: Cyclopoids parasitic on fishes. By Ju-Shey Ho. February 1978, iii + 12 p., 17 figs.

410. The 1976 Ceratium tripos bloom in the New York Bight: Causes and consequences. By Thomas C. Malone. May 1978. iv + 14 p., 17 figs., 1 table.

411. Systematics and biology of the tilefishes (Perciformes: Branchio- stegidae and Malacanthidae), with descriptions of two new species. By James K. Dooley. April 1978, v + 78 p., 44 figs., 26 tables.

412. Synopsis of biological data on the red porgy, Pagrus pagrus (Lin- naeus). By Charles S. Manooch III and William W. Hassler. May 1978, iii + 19 p.. 12 figs., 7 tables. For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, DC 20402; Stock No. 003-017-00418-0.

413. Marine flora and fauna of the northeastern United States. Crustacea: Branchiura. By Roger F. Cressey. May 1978, iii + 10 p., 15 figs. For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, DC 20402; Stock No. 003-017-00419-8.

414. Synopsis of biological data for the winter flounder, Pseudopleuro- nectes americanus (Walbaum). By Grace Klein-MacPhee. November 1978, iii + 43 p., 21 figs., 28 tables.

403. Marine flora and fauna of the northeastern United States. Sipun- cula. By Edward B. Cutler. July 1977, iii + 7 p., 6 figs. For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, DC 20402; Stock No. 003-012-00404-0.

404. Revision of the sea basses of the genus Diplectrum (Pisces: Ser- ranidae). By Stephen A. Bortone. September 1977, v + 49 p., 15 figs., 9 tables.

405. Marine flora and fauna of the northeastern United States. Echino- dermata: Holothuroidea. By David L. Pawson. September 1977, iii + 15 p., 3 figs. For sale by the Superintendent of Documents, U.S. Gov-

415. A basis for classifying western Atlantic Sciaenidae (Teleostei: Perciformes). By Labbish Ning Chao. September 1978, v + 64 p., 41 figs., 1 table.

416. Ocean variability: Effects on U.S. marine fishery resources - 1975. (20 papers.) By Julien R. Goulet, Jr. and Elizabeth D. Haynes, Editors. December 1978, iii + 350 p.

417. Guide to the identification of genera of the fish Order Ophidiiformes with a tentative classification of the order. By Daniel M. Cohen and J^rgen G. Nielsen. December 1978, vii + 72 p., 103 figs., 2 tables.

NOAA Technical Report NMFS Circular 428

j«Sv.

^fNT Of C

Morphological Comparisons of North American Sea Bass Larvae (Pisces: Serranidae)

Arthur W.Kendall, Jr.

August 1979

■a

■ a

U.S. DEPARTMENT OF COMMERCE

Juanita M. Kreps, Secretary

National Oceanic and Atmospheric Administration

Richard A. Frank, Administrator

National Marine Fisheries Service

Terry L. Leitzell, Assistant Administrator for Fisheries

The National Marine Fisheries Service (NMFS) does not approve, rec- ommend or endorse any proprietary product or proprietary material mentioned in this publication. No reference shall be made to NMFS, or to this publication furnished by NMFS, in any advertising or sales pro- motion which would indicate or imply that NMFS approves, recommends or endorses any proprietary product or proprietary material mentioned herein, or which has as its purpose an intent to cause directly or indirectly the advertised product to be used or purchased because of this NMFS publication.

CONTENTS

Page

Introduction 1

Methods and Materials 1

Results 2

Subfamily Serraninae 2

Centropristis 2

Paralabrax 4

Diplectrum 5

Serranus 6

Hypoplectrus 10

Schultzea 12

Serraniculus 12

Serraniculus pumilio 12

Discussion 15

Subfamily Anthinae 16

Plectranthias 16

Anthias 16

Pronotogrammus 20

Hemanthias 22

Hemanthias uivanus 22

Discussion 26

Subfamily Epinephelinae 26

Paranthias 30

Epinephelus 31

Mycteroperca 34

Mycteroperca microlepis 34

Discussion 35

Subfamily Grammistinae 38

Liopropoma 38

Rypticus 39

Pseudogramma 40

Pseudogramma gregoryi 41

Discussion 43

Discussion 45

Summary 45

Acknowledgments 47

Literature Cited 47

Figures

1. Larvae of Centropristis striata from the northwestern Atlantic Ocean 4

2. Spine-bearing bones of the opercular and posttemporal regions on a 10.6-mm Centropristis striata

larva 4

3. Larvae of Paralabrax sp. from the northeastern Pacific Ocean 5

4. Spine-bearing bones of the opercular and posttemporal regions on a 9.5-mm Paralabrax sp. larva ... 5

5. Larvae of Diplectrum sp. Type 1 from the northwestern Atlantic Ocean 6

6. Larvae of Diplectrum sp. Type 1 from the northeastern Pacific Ocean 7

7. Spine-bearing bones of the opercular and posttemporal regions on a 10.3-mm Diplectrum sp. Type 1

larva 8

8. Larvae of Diplectrum sp. Type 2 from the northeastern Pacific Ocean 8

9. Spine-bearing bones of the opercular and posttemporal regions on a 9.9-mm Diplectrum sp. Type 2

larva 8

10. Larvae of Serranus sp. from the northwestern Atlantic Ocean 9

11. Spine-bearing bones of the opercular and posttemporal regions on a 9.7-mm Serranus sp. larva 10

12. Laboratory-reared larvae of Hypoplectrus sp. from eggs collected near Miami, Fla 11

13. Spine-bearing bones of the opercular and posttemporal regions on a 12.5-mm Hypoplectrus sp. larva . 12

14. Young stages of Serraniculus pumilio from the northwestern Atlantic Ocean 13

iii

15. Larvae of Plectranthias garupellus from the northwestern Atlantic Ocean

16. Larvae oiAnthias sp. Type 1 from the northwestern Atlantic Ocean

17. Larvae oiAnthias sp. Type 2 from the northwestern Atlantic Ocean

18. Larva oiAnthias tenuis from the northwestern Atlantic Ocean

19. Larva oiAnthias sp. Type 3 from the northwestern Atlantic Ocean

20. Larvae of Anthias gordensis from the eastern Pacific Ocean

21. Spine-bearing bones of the opercular and posttemporal regions on a 10.2-mm Anthias sp. Type 1 larva

22. Larvae of Pronotogrammus aureorubens from the northwestern Atlantic Ocean

23. Larvae of Pronotogrammus eos from the eastern Pacific Ocean

24. Spine-bearing bones of the opercular and posttemporal regions on a 9.5-mm Pronotogrammus aureorubens larva

25. Larvae of Hemanthias vivanus from the northwestern Atlantic Ocean

26. Larva of Hemanthias peruanus from the eastern Pacific Ocean

27. Spine-bearing bones of the opercular and posttemporal regions of a 4.3-mm Hemanthias vivanus larva

28. Spine-bearing bones of the opercular and posttemporal regions on a 10.3-mm Hemanthias vivanus larva

29. Development of the supraoccipital crest of Hemanthias vivanus

30. Details of the head spination on a Hemanthias vivanus larva

31. Larva of Paranthias furcifer from the eastern Pacific Ocean

32. Larva of a member of the Epinephelus striatus species group from the northwestern Atlantic Ocean

33. Larva of an Epinephelus sp. (subgenus Dermatolepis or Alphestes) from the eastern Pacific Ocean . .

34. Spine-bearing bones of the opercular and posttemporal regions on a 10.2-mm Epinephelus niveatus larva

35. Young stages of My cteroperca microlepis from the northwestern Atlantic Ocean

36. Spine-bearing bones of the opercular and posttemporal regions on a 9.8-mm Mycteroperca microlepis larva

37. Larvae of Liopropoma sp. from the northwestern Atlantic Ocean

38. Spine-bearing bones of the opercular and posttemporal regions of a 12.0-mm Liopropoma sp. larva

39. Larva of Rypticus sp. from the northwestern Atlantic Ocean

40. Larvae of Pseudogramma gregoryi from the northwestern Atlantic Ocean

41. Spine-bearing bones of the opercular and posttemporal regions on a 10.3-mm Pseudo- gramma gregoryi larva

42. Larvae representative of four groups of American Serranidae

43. Interoperculars of larvae of representatives of four groups of American Serranidae

Tables

1. Some adult and larval characters of genera of North American serranids

2. Meristic character development of larvae of Serraniculus pumilio

3. Body proportions of larvae of Serraniculus pumilio

4. Meristic character development of larvae of Hemanthias vivanus

5. Body proportions of larvae of Hemanthias vivanus

6. Meristic character development of larvae of Mycteroperca microlepis

7. Body proportions of larvae of Mycteroperca microlepis

8. Meristic character development of larvae of Pseudogramma gregoryi

9. Body proportions of larvae of Pseudogramma gregoryi

17 18 19 20 20 21

22 23

24

24 25 26

26

26 28 29 31

32 33

34 36

38 40

41 41 42

44 46 47

3

14 15 27 28 37 38 43 44

IV

Morphological Comparisons of North American Sea Bass Larvae (Pisces: Serranidae)

ARTHUR W. KENDALL, JR.

ABSTRACT

Larvae of 17 of the 23 nominal genera of American serranid fishes are described. Representatives of only two of these genera have been described previously from American waters. The genera fall into four groups which closely follow subfamilial groupings based on adult characters. Larvae of the Serraninae, representing seven genera, appear to be the most generalized and are most similar to Morone-like percichthyid larvae. They have some larval specializations but exhibit rather direct de- velopment of adult morphology. The serranine genera can be ranked in order of increasing larval spe- cialization as follows: Serraniculus, Paralabrax, Centropristis, Diplectrum, and Serranus. Hypoplectrus and one type of Diplectrum, although considered serranines on the basis of adult char- acters, are quite different from the other serranine larvae observed. One line of divergence from the serranines is the Anthiinae, with larvae of four genera being represented in the present collections. These larvae have variously developed strong spines on the head and in the opercular region, also the pelvic fin spine and some dorsal fin spines are strong and develop precociously. These spines are ser- rated in more specialized genera. Among the anthiine genera described there is a progression of larval specialization as follows: Plectranthias, Pronotogrammus, Anthias, and Hemanthias. A third general type of serranid larvae is represented by members of the three American genera of the Epine- phelinae and Gonioplectrus. These larvae all are similar in general appearance and specialized in having elongate, strong serrate spines — primarily the second spine of the dorsal fin and the pelvic spine. The fourth larval type is comprised of three genera whose affinities have been unclear. Liopropoma is generally considered by others to be a serranid of the subfamily Liopropominae, Pseu- dogramma has been placed by others in the family Pseudogrammidae, or considered to be a member of a subfamily of the Grammistidae, the family in which Rypticus, the third genus, is placed. Because of the similarity of their larvae and evidence from adult characters, I consider these genera to be members of the serranid subfamily Grammistinae. Larvae of these genera share some larval charac- ters with the serranines. Their most outstanding larval feature is the development of one or two great- ly elongated flexible dorsal fin spines.

INTRODUCTION

Jordan and Eigenmann (1890) reviewed the American and European serranids known at that time. The limits of and relationships within the family have continued to be sources of study and, as presently understood, in American waters it consists of three of Jordan and Eigen- mann's (1890) six subfamilies (Serraninae, Anthiinae, and Epinephelinae) plus the Liopropominae (Gosline 1966). Their Grammistinae has been elevated to familial status (Gosline 1966).

Serranid larvae are poorly known, although adults of most species are common and important commercially. Early descriptions of Mediterranean larvae indicated a diverse morphology (Fage 1918). The larvae have vari- ously developed head and fin spination that seem related to the group within the family to which the species are assigned. Larvae of only 2 of the 100 or so species of American serranids have been described — Epinephelus niveatus by Presley (1970) and Centropristis striata by Kendall (1972). Descriptions of these and other serranid

'Northeast Fisheries Center Sandy Hook Laboratory, National Marine Fisheries Service, NOAA, Highlands, N.J.; present address: Northwest & Alaska Fisheries Center, National Marine Fisheries Service, NOAA, 2725 Montlake Boulevard East, Seattle, WA 98112.

larvae indicate that the family may be subdivided based on specializations of the larvae. In this paper, sub- familial and generic characteristics of American ser- ranid larvae are defined from material from western Atlantic and eastern Pacific waters and one representa- tive of each of the four subfamilies is described in detail.

METHODS AND MATERIALS

Since only Hypoplectrus and Centropristis (Hoff 1970) among American serranids have been reared beyond yolk exhaustion, the indirect method of assem- bling series of similar appearing larvae from plankton samples was used.

Representative larvae were drawn using a camera lucida attachment on a dissecting microscope. Pigment on the formaldehyde-preserved specimens was limited to melanophores which is variously referred to as it ap- peared on the larvae, e. g., as spots, patches, or blotches. Body proportion measurements were made with an ocular micrometer. The larval period was divided into three stages: preflexion (before notochord flexion), flex-

2W. J. Richards, National Marine Fisheries Service, Southeast Fish- eries Center, Miami, pers. commun. January 1974.

ion (during notochord bending), and postflexion (after the notochord tip had completed flexion) following rec- ommendations of Ahlstrom et al. (1976). Body length (BL), the reference for proportion measurements, was measured from the tip of the snout to the tip of the noto- chord (notochord length (NL) in preflexion and flexion larvae and in postflexion larvae it was measured from the tip of the snout to the margin of the hypural bones (stan- dard length (SL) as in Ahlstrom et al. (1976). Some specimens were cleared and stained following Hollister (1934) and Clothier (1950). The ventral portion of the interopercular bone is not shown in the figures of the opercular area because its outline was hidden by over- lying tissue in the cleared and stained specimens.

Specimens included materials from collections at the National Marine Fisheries Service (NMFS) Labora- tories in La Jolla, Calif.; Miami, Fla.; Narragansett, R.I.; and Sandy Hook, N.J.

Most of the material was collected as part of one of three large-scale ichthyoplankton surveys — one in the Atlantic Ocean and two in the Pacific Ocean. The survey in the Atlantic was that of NMFS, Sandy Hook, aboard the RV Dolphin in 1965-68 when ichthyoplankton was sampled from Cape Cod, Mass., to Palm Beach, Fla. Most of the Pacific material resulted from the CalCOFI program, a long-continuing survey of waters off Cali- fornia and Baja California. The other major source of material from the Pacific Ocean was the EASTROPAC program, which consisted of three multivessel cruises in the eastern tropical Pacific Ocean. Appendix Table 1 lists collection data associated with specimens used for the illustrations.

RESULTS

Subfamily Serraninae

The serranines have rather uniform characteristics which help distinguish them from other serranids (Table 1). They are represented by seven genera in North Amer- ican waters with 52 species and are moderate to small- sized serranids apparently representing the base from which other subfamilies arose (Smith and Atz 1969; Ken- dall 1976). Most serranines are synchronous hermaphro- dites; however, protogyny and secondary gonochorism are present in some species (Smith and Young 1966) . The lateral line is complete and not highly arched and the scales are moderate to large and strongly ctenoid (Jordan and Eigenmann 1890). The larvae of three Euro- pean species of Serranus (Fage 1918; Bertolini 1933; Aboussouan 1972) as well as those of Centropristis striata (Kendall 1972) are the only serranines described.

'Serranids studied to date are hermaphroditic (Smith 1965; Kendall 1977), i.e., male and female gametes are produced in each individual. Some genera are secondary gonochorists, i.e., the gonads show histological traces of both sexes but only male or female gametes are pro- duced in each individual. Hermaphroditism takes on two forms in serran- ids: synchronous hermaphroditism, where both types of gametes are produced simultaneously in one individual, and protogyny, where in- dividuals mature first as females and later in life the gonad transforms into testes and produces sperm.

Larvae of all but one (Schultzea) of the North Ameri- can genera of serranines are herein described. Serranine larvae, except those of Hypoplectrus (see below), can be distinguished from larvae of other fishes on the basis of several characters. Development in serranines is direct and they do not possess many of the larval specializa- tions of other serranids. They do not have elongate pre- opercular spines, rather a series of blunt points. The dorsal and pelvic fin spines are thin and only slightly elongate in some genera. Larval pigment generally con- sists of melanophores in characteristic positions mostly along the ventral midline. Generally, there are spots at the angle of the jaw, at the junction of the cleithra, be- tween the bases of the pelvic fins, near the anus, at the bases of the anal rays and some on the caudal peduncle. Some of these spots are intensified in some species. Pig- ment on other parts of the body is variable within the subfamily but some is consistent within genera. The body shape of the larva gradually changes to that of the adult, without abrupt changes in proportions.

Centropristis. — Centropristis occurs only in the northwest Atlantic where four species are recognized (Kendall4). These are moderately large protogynous her- maphrodites. Centropristis striata occurs farther north than any other serranid on the east coast. A closely re- lated species, C. melana, is widespread in the Gulf of Mexico."' The two other species, C. philadelphica and C. ocyurus, cooccur with C. striata in the South Atlantic Bight. The species of Centropristis are distinguished by meristic and other characters (Table 1).

The eggs of C. striata were the object of one of the early detailed studies of fish embryology (Wilson 1891) and Hoff (1970) illustrated eggs and prolarvae of C. melana. Larvae of C. striata were not described until recently (Kendall 1972), although Sette (see Merriman and Sclar 1952) and Pearson (1941) had apparently recognized them in plankton samples.

I found only one type of larva assignable to Centropristis in the present material. The description of Kendall (1972) has been condensed and supplemented with recent observations in the following account.

Larval morphology of Centropristis (Fig. 1) is typical of other serranines. There is no armature development associated with the fin spines, the preopercular bone has only small, unserrate protruding spines (Fig. 2) and body shape, once the caudal fin has formed, approximates that of small adults. Fin spines are thin and not pro- duced. Dorsal fin spines develop at about the same stage as the soft rays. Pelvic fins develop after the other fins have several rays ossified and grow isometrically throughout development.

In early larvae (<5 mm) there is some dorsal body pig- ment but afterwards nearly all larval pigment is asso-

JA. W. Kendall. 1977. Biological and fisheries data on black sea bass. Centropristis striata (Linnaeus). Sandy Hook Lab. Tech. Ser. Rep. No. 7, 29 p.

''Some (e.g., Miller 1959) consider C. melana a subspecies of C. striata.

os ~

01 S

01 c

■3 b

& M

2 S

S «

i- —

1 «

.Si oi

5 s

c 0) !- cd

-n	CD
	c
eg	a
Tl
	T3
	to
1,	OJ
	_c

O T3

S B

OJ CO

MUBIJ

jBSJOQ

apunpad I'epneo

aseq

utj ]euy

snuy

uoijjasui oiAjaj

uirumiaiQ

MB[>

JB|nojado -jajuj

jB|najado

-ajj

|B}idpao

-Bjdng

IBuy aiAjaj

[BSJllfJ

oiA|a,-[

[BSJOQ

<
-1
OS

Cfi	CD
	6
O	3
X	2

A"3o|OL[d.ioui ajBJg

ujaj^Bd IBSJopajj

uiajjBd viH

3TJI0BJ

3I}UB|)V

X X

X

X X XX

k^ ^ *v* *v*

**N rS ^N **S

XK*- *-y k> "^ *v* ^rf-

rS rS r*S r*S rS rN

X X X X XX

X X X X X X X

X X r> X

X X

c c c c a c c

0; OJ O) Oi CD o* o>

55 3 55 53 55 55 co

CM CM CM CM

^T c5 jd c5

CD Q £3 ^ O O O O

cj} <j} oo ao

■3 o >> •-

C O C C

2 c c*> 2

o "> 2 o

c ci o a

Tj* Tf Tf *3*

CM CM

o c5

O CD

o o

s s

£ x x'£!

^H -t CC

X 4 ~*

*x'x'

X X	X X
X	XX
X	X
X X	X X
X	X
X

X X

XX X X

X 5 rS i^S

X J J J

X XX

Q Q Q Q

a; os os os

o o o

XXX

X X

*>S ?N k^N rS

?S ?S ^N ?N

r*N f^i ?S <*^

m ■* ^ ■*

f) f) o o

0 o c5 o

01 0J CD OJ

c c c c

o o o o

Z Z Z Z

c£> iO CO iO CO

+ o	+	+	+ 3	f 3	+	+	+	+ 3'
urs	35	O	X	X	-f	[ -	CO	D : i
-t	to	iO	'C	CO		iO		x.
r~	'JO 'i>	c~	CO	op	co I -	CO CO	c^	CO

o o o o

O O CO

ai co 6

CO CO iO -t (35

•— ' C>3

H CO i-H <N

	m	Tf CO
en

	or>	M IO
f-H	^	X -1
>	X	X X
	-	-H IO
^	-	IO 00

XXX

g 1?

a .3	\|
a	C
3	3
0)
en	11
>	t

-J	J

X X X X
CO CO CO CO	CO	CO
O) CM CM CM	CM	CM
T3 T3 T3
a> © dj a a a ca S3 ca t^ -c x j=: 3 ai c/i co	J3	CO -a CO	CD u c: a
-Q : ; ; o qj <d a>	CU	0,	X! D
os .-g s s	=	CD	6.

o o o o

Oi 01 0* 01

C C c c

o o o o

Z Z Z Z

C)D -f- O

+ + -H + + O O + O CT3

cj	c k e	s	CD 'iJ
c	a ? u	c	Q. ii
CO J.		a	° 3

X	Q, O Q	"X	It) 09

J I

?o.t

t: «! o t. o

a; tq to u cj

01 ? ^ ci

CO C CO _2 CD

— -~Z j* -C O » CliS Q, J.

& -3 a Co

■^ 5 ^ ?; >^ c ° a Q. jT'

■s co eg tg^

3 H- ■* u" B

CM S CM X

3- > g 3 =n

> > K > >

r-l ^H Ol

iO ^ CO CO H

k o -^ Q Cy

Q. 3 K CD O

O CD IS- ^ -o

3 £, oe: a; ^

co H

s =3

■& 2

Q Oh

Figure 1.— Larvae of Centropristis striata from the northwestern Atlantic Ocean: a) 4.7 mm; b) 8.3 mm.

posttemporal

preopercular

supracleithrum

Figure 2. — Spine-bearing bones of the opercular and posttemporal regions on a I0.(i-mm Centropristis striata larva.

ciated with the ventral midline. Melanophores (spots) in characteristic positions are larger than others. Large spots are seen on the posterior margin of the angular, at the junction of the cleithra, between the bases of the pelvic fins, near the anus, and near the insertion of the anal fin. Several slightly smaller spots develop on the caudal peduncle, and still smaller spots are seen at the bases of the anal fin rays, between the larger caudal pe- duncle spots, and at the bases of some caudal fin rays. Several melanophores also occur on the posterior portion

of the optic lobes and on the dorsoposterior lining of the coelomic cavity.

Paralabrax. — Paralabrax with eight species is the only North American serranine limited to the eastern Pa- cific (Table 1). It is ecologically similar to Centropristis in the Atlantic in that it is larger than most other serran- ines and occurs at higher latitudes than other serranines on the Pacific coast. Smith and Young (1966) indicated that Centropristis and Paralabrax are also closely re- lated. The only species that has been studied is a sec- ondary gonochorist, while most other serranines are syn- chronous hermaphrodites (Smith and Young 1966).

Paralabrax larval development is typical of serran- ines. Paralabrax larvae approach adult body pro- portions soon after tail flexion is complete (Fig. 3). The spinous and soft portions of the dorsal fin form at about the same rate, as they do in Centropristis. The fins of Paralabrax larvae are more heavily pigmented than those of Centropristis but the ventral melanophores are smaller and more uniform in size. Pelvic fins form with pigment on the membrane as the dorsal fin starts to form. Pigment develops ventrally at the base of the pelvic fin, near the anus, along the base and at the inser- tion of the anal fin, and on the caudal peduncle. Pigment also forms on the angular, on the dorsal surface of the brain, and at the base of the caudal fin. Some pigment develops dorsally at the base of the spinous dorsal fin and a large blotch of pigment develops on its membrane. The pectoral fin rays are pigmented on their distal third and the pelvic fin is heavily pigmented on and between the rays. Opercular spination is as in other serranines (Fig. 4): the preopercular has about three spines dorsal and

Figure 3. — Larvae of Paralabrax sp. from the northeastern Pacific Ocean: a) 6.(1 mm; b) 9.0 mm.

two ventral to the one at the angle, on the preopercular ridge there are a few blunt spines, and the subopercular has a smooth posterior margin.

Diplectrum. — In contrast to reef-dwelling habits of many other serranids, Diplectrum occurs mainly over sand or mud bottoms on the continental shelves (Bortone 1977). These moderate-sized synchronous hermaphro- dites have one or two clusters of strong and divergent spines on the preopercular which is produced into a bony flap. The Pacific species were reviewed by Rosenblatt and Johnson (1974) and the entire genus by Bortone (1977).

Diplectrum is the only serranine with larvae from both Atlantic and Pacific waters. Since species of Diplectrum have overlapping meristic features and geographic ranges, the larvae cannot be assigned to species. Two dis- tinct larval types with the unique meristic characters of Diplectrum were found from both oceans. They will be referred to as Type 1 and Type 2 and discussed sepa- rately, since they share few larval characters in common.

Type 1 — Larvae of Diplectrum called Type 1 from the Atlantic and Pacific Oceans are similar in development. They exhibit most of the features of other serranine lar- vae. The body shape approaches that of the adult and in later larval stages, after the fin rays form, they are among the most slender serranines (Figs. 5, 6). Fin develop- ment is modified from the serranine pattern (see page 2) in that the first dorsal fin spines and pelvic fin elements

posttemporal

supracleithrum

preopercular

bopercular

interopercular

Figure 4. — Spine-bearing bones of the opercular and posttemporal regions on a 9.5-mm Paralobrax sp. larva.

form early. None of the early forming dorsal fin spines are very elongate, as they are in some Serranus larvae. Pigment on larvae from both oceans is present along the ventral midline in positions characteristic of serranine larvae (Table 1). Individual melanophores are more uni- form in size than those of Centropristis. A series of about

Figure 5. — Larvae of Diplectrum sp. Type 1 from the northwestern Atlantic Ocean: a) 5.8 mm; b) 10.0 mm.

five spots occur ventrally on the caudal peduncle of Atlantic larvae while there is only a spot at the insertion of the anal fin and one two-thirds of the way back on the caudal peduncle in Pacific larvae. At least one spot is present at the base of the caudal fin on larvae from both oceans. The membranes of the spinous dorsal and pelvic fins of Atlantic larvae are variably pigmented with scat- tered spots. Pacific larvae have intense pigment on the membrane of the spinous dorsal fin centered around the fourth spine. The pectoral and pelvic fins of Pacific lar- vae have pigment on and between the fin rays. The pre- opercular and subopercular have more armature than is seen in other serranine genera (Fig. 7).

Type 2 — Larvae assigned here to Diplectrum Type 2 are distinct from Diplectrum Type 1. I have not found such intrageneric differences in larvae of any other ser- ranid genus. This would indicate that either these larvae are not of Diplectrum but are of an undescribed genus, possibly even in a closely related family with representa- tives in both oceans, or that there are two quite distinct types of Diplectrum, possibly worthy of subgeneric rank.

While Type 1 larvae are similar to other serranine lar- vae, Type 2 larvae show specializations not seen in other genera. The general body shape is similar to that of Type 1 but the snout is more elongate and is concave in dorsal profile in smaller specimens (Fig. 8). The sequence and extent of fin ray formation is different from that in other serranines. The pectoral fin develops early, before most other fins, as a fan-shaped structure and extends to the anus. Only in grammistines is similar larval pectoral fin development seen. The pelvic fins form before the rest of

the fins and are elongate, reaching beyond the anus. The distal third of the rays of the paired fins are covered with contracted melanophores. Elements of the dorsal, anal, and caudal fins all develop at about the same time. Pig- ment pattern differs from that seen in other serranines in that the body anterior to the anus is unpigmented before the fin rays are complete and later in development a large diffuse internal pigmented area forms between the lateral line and the anal fin, dorsal to the anterior half of that fin. The bones of the opercular region are similar to those of Diplectrum Type 1 except the subopercular margin is smooth and does not have the small spines seen in Diplectrum Type 1 (Fig. 9).

Serranus. — In the western Atlantic there are 13 species and in the eastern Pacific there are 5 nominal species assigned to Serranus (Table 1). These are rather small fishes ( <200 mm) sharing 17 characters in common which, taken together, exclude all other genera (Robins and Stark 1961). Their median fin ray counts fall out- side the ranges of other American serranids except Schultzea beta, which has only six branchiostegal rays (Robins and Stark 1961) and Diplectrum, which general- ly has more pectoral rays (Table 1).

Robins and Stark (1961) examined east coast Serranus species in some detail. Since Serranus in the eastern Pa- cific is poorly known with few specimens of some species taken, some species may prove invalid.

While no American Serranus larvae have been de- scribed, eggs and early larvae of Mediterranean species of Serranus were among the first serranids described

Figure 6. — Larvae of Diplectrum sp. Type 1 from the northeastern Pacific Ocean: a) 4.3 mm; b) 5.0 mm; c) 6.1 mm; d) 13.0 mm.

posftemporal

opercular

preopercular

supracleithrum

Figure 7. — Spine-bearing bones of the opercular and posttemporal regions on a 10.3-mm Diplectrum sp. Type 1 larva.

opercular

preopercular

supracleithrum

Figure 9.— Spine-bearing bones of the opercular and posttemporal regions on a 9.9-mm Diplectrum sp. Type 2 larva.

Figure H. — Larvae of Diplectrum sp. Type 2 from the northeastern Pacific Ocean: a) 4.5 mm; b) 8.4 mm.

(Raffaele 1888). Larger larvae of Mediterranean species have since been illustrated and described (Bertolini 1933) and there are brief descriptions of Serranus larvae collected off Africa (Aboussouan 1972).

Serranus larvae were found in samples from the Atlan- tic; however, none were found in Pacific collections.

Although there are differences among Serranus larvae in extent of fin ray elongation, pigment, and body shape, all found so far are separable from other larvae by the be- ginning of the flexion stage. More detailed study may al- low descriptions of the several types and their assign- ment to species

Figure 10. — Larvae of Serranus sp. from the northwestern Atlantic Ocean: a) 3.7 mm; b) 5.0 mm; c) 5.5 mm; d) 9.4 mm.

Serranus development shows several modifications of the basic serranine plan (page 2). The body is consider- ably deeper from the head back to the anus from before notochord flexion until after all the fin rays have formed (Fig. 10). The pelvic fin forms before the other fins and the elements are elongate in some larvae. The first three to five spines of the dorsal fin form before the rest of the fin and in some the third spine is elongate. Among Ser- ranus larvae described so far these fin spines are longest in 5. cabrilla (Fage 1918).

In Serranus larvae some of the characteristic ventral pigment is intensified and some is eliminated. The more contracted melanophores of other serranines are absent in Serranus. Intensified melanophores are on the angular, at the junction of the cleithra, and near the anus. A spot at the insertion of the anal fin and one midway back on the caudal peduncle are the largest spots ventrally and are more pronounced in some larvae than in others. Dorsally, on at least some larvae, there is a spot near the base of the last dorsal spine. Another spot ventral to the posterior third of the second dorsal fin is variable in position from near the base of the fin to near the lateral line. This spot opposes the one at the insertion of the anal fin and in S. cabrilla these two spots are large and distinctive (Fage 1918). Pigment spots are scattered on the membranes of the spinous dorsal, anal, and pec- toral fins in some larger larvae.

The opercular region is armed with blunt spines, which are generally more pronounced than those of other ser- ranines (Fig. 11). There seems to be some variation in spine length among the species of Serranus since S. cabrilla has longer spines (Fage 1918) than the larvae I examined. The preopercular has two spines ventral to and three spines dorsal to the one at the angle. There are spines anterior to these on the ridge of the preopercular bone in some specimens. The subopercular has two spines dorsal to the one seen at its angle in other serran-

posttemporal

supracleithrum

preopercular

bopercular

interopercuiar

Figure II. — Spine-bearing bones of the opercular and posttemporal regions on a 9.7-mm Serranus sp. larva.

ines. Among larval serranines, only Diplectrum has a more ornately armed opercular region.

Hypoplectrus. — This genus of brightly colored small reef fishes has long perplexed scientists. In the western Atlantic there are several nominal species with different color patterns but with identical meristic features and body proportions. At present, these are generally consid- ered valid species; however, other species of serranids such as Mycteroperca rosacea and M. olfax have more than one color phase (Rosenblatt and Zahuranec 1967). Barlow (1975) observed spawning between two nominal species and so considered them synonymous. Thresher (1978) concluded that the color patterns observed in Hypoplectrus demonstrated mimicry of nonpredatory reef fishes. Apparently, coloration in Hypoplectrus is a plastic character; and with the possibility of clones in a synchronous hermaphrodite, different color patterns can become established locally in a relatively short time. The poorly known eastern Pacific Serranus lamprurus Jordan and Gilbert was placed in Hypoplectrus by Meek and Hildebrand (1925), but Robins and Stark (1961) stated it is indeed a Serranus and is close to S. flaviventrus of the Atlantic, although they gave no reason for their opinion. Hypoplectrus lamprurus is deeper bodied than Serranus and has a dorsal fin ray count of X, 14, whereas Serranus species generally have X, 12 (Table 1). Although a de- tailed study is needed, it seems that H. lamprurus should still be considered a Hypoplectrus.

No larvae assignable to Hypoplectrus were found in plankton collections. Their unique dorsal fin ray count (X, 14) among east coast serranines assured that if larvae had been present they would have been recognized. It is possible that Hypoplectrus larvae are planktonic for only a short time before they become demersal.

A series of Hypoplectrus sp. was reared from wild planktonic eggs at the NMFS Laboratory in Miami (Richards see footnote 2). I was able to examine the specimens briefly and copy drawings of them. Also, three cleared and stained specimens were examined in detail.

These specimens are different from other serranine larvae. It is unlikely that rearing produced these differ- ences and it is difficult to consider them as modifi- cations of serranine larval characters. These Hypo- plectrus larvae are not similar to any known serranid larva and provide reason to doubt that Hypoplectrus be- longs in the Serranidae.

The body shape indicates rather direct development from larval to adult proportions (Fig. 12). The first few spines of the dorsal fin develop early, during notochord flexure, but the associated membrane is more fleshy than in serranines. The pelvic fins form at about the same stage as the first dorsal spines but the rays are shorter than those of serranines with early developing pelvic fins. The membranes of these early forming fins are pigmented. The rest of the dorsal spines are pro- gressively shorter than the third and the posterior few form late after the rayed portion of the fin is developed. Some of the rays of the second dorsal fin become longer

10

a

b

d

Figure 12. — Laboratory-reared larvae of Hypoplectrus sp. from eggs collected near Miami, Fla.: a) 3.4 mm; b) 4.7 mm; c) 5.7 mm; d)

8.5 mm.

11

than the longest spines. The rest of the fins form over a short length interval of 4.7-5.7 mm.

The head and particularly tissues around the mouth are fleshier than in serranines. The opercular region has spines developed on several bones; however, they do not appear on the larval drawings since they are covered by flesh. The preopercular is edged with about 12 subequal spines (Fig. 13). The subopercular has about three blunt spines on its margin. Besides the three definitive spines on the opercular, there are about four spines ventral to them, a condition not seen in serranines.

Pigmentation consists of irregular shaped and sized melanophores in characteristic positions. More pigment is present on Hypoplectrus than on most other serran- ines. A series of spots along the ventral midline consists of a spot near the base of the pelvic fins, one below the middle of the gut, one just anterior to the anus, one near the insertion of the anal fin, and another on the caudal peduncle. Spots ventral near the tip of the notochord come to lie at the base of the caudal fin rays as they form. A few spots near the base of the dorsal fin migrate to form the dorsal fin membrane pigment. A few spots are scat- tered on the dorsal surface of the brain posterior to the eyes. All the fin membranes have pigment in later larval stages. By 8.5 mm, small melanophores become scat- tered over the dorsal portion of the anterior two-thirds of the fish.

opercular

preoperc

supracleithrum

interopercular

Figure 13. — Spine-bearing bones of the opercular and posttemporal regions on a 12.5-mm Hypoplectrus sp. larva.

Schultzea. — Schultzea is monotypic represented by S. beta of the western Atlantic (Table 1). It lives more pe- lagically than other serranines and is adapted to feed on small, free-swimming prey. The mouth is modified to be highly protrusible through extensive changes in the shape of the premaxillary (Robins and Stark 1961). The fin counts are similar to those of Serranus but there are only six branchiostegal rays instead of seven (Robins and Stark 1961).

Two juvenile specimens (24.0 and 25.4 mm) collected by John McCosker (Steinhart Aquarium, San Francisco, CA 94118) provided for study did not retain any larval characters, so they cannot be compared with other serranines and their relationships within the group can- not be determined. However, from adult adaptations, it appears that Schultzea is the most highly modified American serranine.

Serraniculus. — Serraniculus contains one diminu- tive species (<80 mm), S. pumilio, in the northwest Atlantic (Table 1). Its life history has been studied and it is a synchronous hermaphrodite (Hastings 1973).

A serranine larval type from the Atlantic had meristic characteristics of S. pumilio, notably six branchiostegal rays as opposed to seven for all other serranines except Dules auriga and Schultzea beta, which occur south of the area where these larvae were collected.

Serraniculus pumilio (Fig. 14). — Meristic element development (Table 2) — A total of 41 specimens from 3.1 to 7.3 mm was studied to trace meristic character de- velopment. Since the smallest larvae have a straight no- tochord and vertebrae are the only ossified meristic ele- ments, and the largest have adult complements of all meristic characters except gill rakers and scales, the pat- tern of meristic character development can be seen in this series. Vertebrae ossify from anterior to posterior ex- cept for the urostyle which ossifies before the two or three vertebrae anterior to it. The smallest larva examined (3.1 mm) had 1 ossified vertebra. Numbers of vertebrae gradually increase until by 4.6 mm some larvae have all 24 ossified. However, some larvae as large as 6.2 mm have only 23 vertebrae. Throughout the series, there is considerable variation in the number of ossified verte- brae at a given length. Branchiostegal rays are ossified in some larvae as small as 3.8 mm and the adult comple- ment of six is seen in most specimens by 4.0 mm, when the notochord has completed flexion. Other meristic ele- ments ossify when the notochord completes flexion at 4.0 mm and larvae as small as 4.6 mm have adult fin ray complements. Pelvic and spinous dorsal fin rays do not develop precociously as they do in other serranid sub- families and even in Diplectrum and Serranus. The dor- sal fin spines and rays each develop at about the same rate in S. pumilio.

Body proportions (Table 3) — Several body propor- tions of 21 larvae from 2.9 to 8.3 mm were measured and expressed as a percentage of body length (BL) to docu- ment changes in body shape during larval development. Total length increases rapidly from about 105% of BL at 2.9-3.9 mm, while the notochord is straight, to 115% of BL at 4.0 mm after notochord flexion is complete; there- after, it increases gradually to about 125% of BL as the caudal fin forms. Preanal length increases from about 60% of BL in larvae <4.0 mm to about 65% of BL in 7.0- to 8.0-mm larvae. Head length remains fairly constant at about 39% of BL throughout larval development. Eye length and snout length decrease slightly during de- velopment from about 12% to slightly less than 10% of

12

a

b

?-- T*s?3r

rk&^$>.

'MM

v'^-vSW ^ '-% ra$*

Figure 14. — Young stages of Serraniculus pumilio from the northwestern Atlantic Ocean: a) 3.8 mm; b) 5.8 mm; c) 55.3 mm.

BL. The greatest body depth decreases from 32% to about 28% of BL. Depth at anus increases during early development up to 4.0 mm and then holds constant at about 25% of BL. Caudal peduncle depth increases from <6% in BL in the 2.9-mm larvae to about 14% of BL in 4.4-mm larvae; thereafter, it changes little. A compari- son of the proportions of the larvae after notochord flex- ion with those of the adult shows that the adult body shape of S. pumilio is established early in development, in larvae as small as 4.2 mm.

Pigmentation — Throughout larval development, S. pumilio is more heavily pigmented than any other

serranine. The characteristic serranine ventral spots are present, but they are supplemented by many other spots. Pigmentation is composed of small, individual melano- phores making up a pattern, rather than the pattern be- ing comprised of large melanophores in characteristic positions as seen in other serranids.

The beginning of the pattern is already established in 2.6-mm larvae with three rows of melanophores on the trunk: one along the dorsal midline, one along the lateral line, and the third along the ventral midline. Dorsally, there is a series of about four spots, which are larger than those in the other two rows. The anteriormost of these

L3

Table 2. — Meristic character development of larvae of Serraniculus pumilio. Specimens between dashed lines are undergoing notochord flexion. For the caudal fin,"P" and "S" indicate primary and secondary fin rays.

Body	Branchio- stegal rays	Verte- brae	Pec- toral fin	Pel- vic fin	Caudal fin	Anal fin	Dorsal fin Spines Rays
length (mm)	Dorsal Ventral S P P S	Gill rakers

3.1 NL

3.4

3.5

3.6

3.7

3.8

:■; S

3.8

4.0

4.3

4.4

1 3

4

8

5

11

10

15

in

8

3

16

14

12

10

16

13

14

3.8 4.1 4.2 4.3 3.9

16

17 17 17

is

1 12

1,2

VI

3.8 SL

4.4

4.6

4.7

5.0

5.2

5.7

5.8

6.0

6.1

6.2

7.0

7.2

7.3

19

22 24 22 24 24 24 21 24 24 23 24 24 24 24 24

13

15

14

9

14

15

6

9

15

8

13

14

15

14

1,3

[,5

I

1,5

1.4

1,5

1,4

1,5

	H,7	VIII	9	1
1	111,7	IX	9	2
2	111,7	X	11	4 2
3	111,7	X	in	3
2	111,7	X	11	5
4	111,7	X	11	5
	II. 1	IV		4
	11,1	VIII		5
5	111,7	X	10	5
	11,1	VII		5
	11,1	VII		4
	III	IV		4
	111,7	X	11	5
6	IH,7	X	11	6
6	111,7	X	11	6

spots is just anterior to the origin of the dorsal fin and re- mains pronounced throughout development. Addi- tionally, a series of smaller spots forms along each side of the dorsal fin. When all of the dorsal rays are formed, one or two of the original large spots can be seen internal to the base of the fin and the smaller spots have developed into a series of nearly contiguous dashes on each side of the fin base.

The lateral line series begins on 2.6-mm larvae as a series of dashes from the head about two-thirds of the length of the trunk. These dashes seem to develop in as- sociation with the vertebrae and lie internally along the lateral septa. The spots above the gut become internal while those farther posterior remain on the surface. From the anus back, as the original dashes become internal, another series develops distal to them on the body sur- face. Superficial spots develop from the anus to the ter- mination of the dorsal fin between the dorsal and lateral

series of dashes. In larger larvae similar spots also extend ventrally to a lesser extent from the lateral dashes. These superficial spots are small but variable in size and posi- tion and appear as a single pigmented blotch. Superfi- cial pigment also develops as a line on the head passing from the snout posteriorly at the level of the eye and another line on the opercular region in line with the lateral series of dashes on the trunk. A patch of small spots also develops laterally in the area of the pectoral fin. Ventrally, spots extend the length of the larva. A few spots are present on the tip of the lower jaw, along the isthmus, on the angular, and at the junction of the clei- thra. A larger spot is present on the ventral midline be- tween the junction of the cleithra and the base of the pelvic fins. Along the gut the ventral spots diverge slight- ly to form a series on each side of the midline. A few spots are also present on the midline ventral to the posterior third of the gut. The anteriormost of these is the largest

14

Table .'!. — Body proportions of larvae of Serraniculus pumilio. Specimens between dashed lines are undergoing notochord flexion.

			Percent of standard length
Body					Pre-	Caudal	Depth
length	Total	Body	Eye	Head	Snout anal	peduncle	at
(mm)	length	depth	length	length	length length	depth	anus
2.9 NL	103	32	13.0	39	10.0 58	5.7	19
3.7	104	31	11.0	38	12.0 59	8.0	25
3.9	107	33	12.0	38	10.0 63	10.0	27
4.0	107	34	11.0	39	11.0 61	9.8	27
4.0 SL	115	33	11.0	36	12.0 61	11.0	28
4.2	118	34	12.0	39	10.0 63	13.0	28
4.2	115	36	12.0	42	11.0 62	12.0	30
4.4	119	33	11.0	39	10.0 63	14.0	29
5.5	117	31	10.0	38	10.0 66	14.0	27
5.7	122	31	11.0	38	11.0 59	13.0	25
6.0	123	31	11.0	38	11.0 62	15.0	26
6.1	123	29	10.0	36	9.3 65	14.0	29
6.7	117	29	9.4	37	9.4 62	14.0	26
6.8	120	31	11.0	38	9.0 67	14.0	J 7
7.0	120	29	9.8	36	9.8 65	14.0	25
7.7	125	23	9.8	32	7.4 61	13.0	28
7.8	124	29	10.0	39	10.0 64	15.0	26
7.9	117	29	11.0	38	9.5 64	14.0	25
7.9	124	28	10.0	39	9.5 62	13.0	25
8.3	125	27	9.9	39	9.9 66	14.0	26
Adult1		27-31	8.5-10	34.5-36	8.5-10.5	12-13.5
'Gins	burg (1952	).

and is ventral to the deepest point of the gut. Posterior to the gut, a series of spots extends along the ventral mid- line to the caudal peduncle area in small larvae. As the anal fin develops, these spots are positioned on either side of its base. A series of spots also develops at the base of the anal fin rays and extends posteriorly onto the caudal peduncle. A spot develops ventrally near the tip of the notochord. When the caudal fin develops, this spot is situated ventrally at its base and several other spots develop along the base of the fin. The posterior third of the caudal peduncle is unpigmented throughout de- velopment. Other small spots present during develop- ment are on the membranes associated with the bran- chiostegal rays, pectoral and pelvic fins, and internally on the dorsolateral surface of the gut cavity.

Late in larval development the superficial spots inten- sify and coalesce giving the larva, particularly the trunk, a gray appearance and making S. pumilio the most heavily pigmented serranid larva found to date.

Discussion. — Of the seven genera of North American serranines, larvae of all but Schultzea have been ex- amined. Larval material for five of these genera has been gleaned from plankton collections, and their develop- ment shows a similar pattern. Laboratory-reared mate- rial of the other genus, Hypoplectrus, showed a pattern of development dissimilar to that of other serranines. This suggests that its adult characters, upon which is pres- ently understood affinities are based, need to be re- examined.

Two distinct types of larvae were found that had me- ristic characters of Diplectrum. Both types of larvae oc-

curred in both oceans and more than two species of Di- plectrum are known from each ocean. Diplectrum Type 2 larvae are not similar to those of other known serranines. They do not have the characteristic serranine ventral pigment, but have a large internal blotch dorsal to the anal fin. The fin rays of the pectoral, pelvic, and caudal fins are pigmented with small melanophores. The fin rays are thin, but longer than in other serranines. On the basis of larval characters, the affinities of Diplectrum Type 2 cannot be determined, but it does not appear closely related to any other serranine.

Among the other serranine genera, the extent of modi- fications of the basic pattern of development may indi- cate relationships. Among American serranids, the serranine genera Serraniculus, Centropristis, and Para- labrax show larval characters most similar to those of Morone-type percichthyids (Mansueti 1958, 1964). In these serranines the dorsal spines develop at the same stage of development as the dorsal soft rays and are not elongate. The pelvic fin rays are not precocious or elon- gate during larval development. Spines of the opercular region are less developed in these genera than in other serranine genera. Pigmentation of Serraniculus larvae is heavier than in other serranines and therefore, most closely resembles that of Morone-like percichthyids. Within the serranines it seems that there has been a trend from a uniformly heavily pigmented larva to one in which there are a few large spots in characteristic posi- tions primarily along the ventral midline. Diplectrum Type 1 and Serranus show modifications of the basic serranine pattern in that the spinous dorsal fin develops before the rest of the fin, and in some species of Serranus

L5

some of the dorsal spines are elongate and tipped with pigmented "flags." The pelvic fins are also early de- veloping and become elongate, more so in Serranus than in Diplectrum Type 1. Serranus is deeper bodied than other serranines and Morone-type percichthyids. Also some large pigment spots develop dorsally in Serranus. Thus on the basis of larval characters, there seems to be a progression of specialization in the order: Serraniculus, Centropristis-Paralabrax, Diplectrum Type 1, Serranus.

Subfamily Anthiinae

This subfamily is presently under study for revision by Anderson and Heemstra. This is a cohesive group of small, brightly colored fishes that occur over hard bot- toms near continental slopes. They are distinguished by several morphological modifications from serranine char- acteristics (Table 1). They have large scales (29-60 lateral line scales), an arched lateral line, deep bodies with rather large heads, and at least 14 gill rakers, but mostly more than 25 (Jordan and Eigenmann 1890, Heemstra see footnote 6). There are about 15 species of anthiines in American waters grouped in seven genera. Four genera have representatives in both oceans but no species are common to both oceans.

Ten types of larvae assignable to four genera {Anthias, Pronotogrammus, Plectranthias, and Hemanthias) are present in the material I examined. No American anthiine larvae have been described previously although larvae of two species of Anthias from the Mediterranean Sea have been described (Bertolini 1933; Sparta 1932) and several anthiines from the southeast Pacific have been illustrated (Fourmanoir 1976).

Anthiine larvae have deep bodies and stout, produced preopercular spines. They also have strong spines in their dorsal, anal, and pelvic fins; but the spines are not as produced as they are in epinephelines. There seems to be a trend in anthiines toward development of armature in the form of serrations on the spinous fin rays and on spines of various head bones such as the opercular series, frontals, parietals, and supraoccipitals. In some species spiny scales develop at a small size, while the fish are still planktonic. Pigment consists of a few blotches in characteristic places on the trunk and head.

Plectranthias. — This genus is represented in the western Atlantic by one species (Robins and Stark 1961). The genus contains another species found in the Celebes and Arafura Seas. Plectranthias garupellus, the Atlantic species, is distinguished from other anthiines in the west- ern Atlantic by the high number of dorsal soft rays (16) and the low number of pectoral rays (13) (Table 1). The spinous dorsal fin is quite angular with the third spine being the longest. There are two strong antrose spines on the lower limb of the preopercular. It is known from Florida and the Bahamas, but is apparently not readily available to normal collecting procedures.

6P. C. Heemstra, CSIRO, Cronulla, N.S.W., Australia, pers. commun. April 1973.

Only a few P. garupellus larvae were found, so a de- tailed description is not possible. However, the larval characters indicate relationships of the species within the subfamily and will be discussed briefly. Plectran- thias garupellus shows little development of character- istic anthiine larval features (Fig. 15). None of the fin spines are serrate. The third dorsal spine in small larvae is elongate but not stout. The head has spines in areas characteristic of other anthiine larvae but they are weak- ly developed and not serrate. A simple spine protrudes from the frontal ridge above the eye. Too few specimens were available for staining and illustrating the opercular series but the preopercular has two ridges of blunt spines with an elongate, thin spine at the angle of the posterior ridge. The interopercular has an elongate thin spine lying proximal to the long preopercular spine and the other opercular bones have some spines protruding. The post- temporal and supracleithral spines protrude bluntly. There is no protruding supraoccipital crest.

The pigment pattern of larval P. garupellus is dis- tinctive. There are three characteristic blotches of pig- ment on the trunk. Two blotches are dorsal, one below the sixth and seventh dorsal fin spines, and the other below the four posteriormost dorsal soft rays. The third blotch is on the ventral midline just posterior to the in- sertion of the anal fin. Other spots occur on some speci- mens, such as one ventrally on the caudal peduncle.

Anthias. — Anthias is circumtropical with several species in the western Atlantic and one in the eastern Pa- cific Ocean (Table 3). They are rather small, brightly colored, deep-bodied fishes. Anthias anthias larvae have been described (Fage 1918; Roule and Angel 1930; Ber- tolini 1933) and they closely resemble Anthias larvae I have examined. I found four types of anthiine larvae in western Atlantic material and one type in Pacific material, which were quite similar and larger specimens had meristic features of Anthias species (Figs. 16-20). Only one western Atlantic type could be identified to species {Anthias tenuis) on the basis of meristic charac- ters of the larvae. Anthias nicholsi seems to be the most common adult of this genus in the western Atlantic and Anthias Type 1 is the most common larval type col- lected; therefore, Anthias Type 1 may be Anthias nicholsi. Since there is only one Anthias species in the eastern Pacific (Anthias gordensis), the larvae are pre- sumed to be that species (Fig. 20).

Anthias larvae are deep bodied and have large heads, as is characteristic of all anthiines. The second dorsal spine is elongate but thin in at least some species. The first three dorsal spines form while the rest of the dorsal fin is still an undifferentiated finfold. When all the dorsal rays have formed, the third spine is the longest and the spines approach the proportions of adults. The dorsal fin spines are rather stout during later larval development, but are never serrate. The elongate spine has pigment on its associated membrane, giving it a flaglike appear- ance. Pigment is retained on the membrane between the second and third dorsal spines into the late larval period. There is a deep notch in the dorsal fin, the last four

L6

Figure 15. — Larvae of Plectranthias garupellus from the northwestern Atlantic Ocean: a) 5.5 mm; b) K.O mm.

spines being quite short. The pelvic fin forms early in Anthias larvae and the first soft ray is produced. The pelvic fin and first three spines of the dorsal fin develop before other fin rays. The membrane of the pelvic fin is pigmented, like the dorsal fin. The pelvic fin spine is stout but not elongate or serrate.

The dorsal and anal fins form concurrently except for the precocious anterior dorsal spines. The caudal fin completes formation after the dorsal and anal fins have formed. The pectoral fin is last to form rays, the dorsal

rays of the pectoral fin forming first, as in all serranids studied to date.

Anthias larvae have characteristic head spines. In some species there is a simple supraoccipital spine. There is a serrate ridge above the eye. The degree of ser- ration varies among the species, some having 3 to 6 serrae while others have up to 12. The most conspicuous spines are in the opercular region (Fig. 21). There are two spinescent ridges on the preopercular. The anterior ridge has 3 to 5 broad points. The posterior ridge has 4 to 8

17

a

b

Figure 16. — Larvae of Anthias sp. Type I from the northwestern Atlantic Ocean: a) 3.K mm; b) 5.3 mm.

strong points, some of them serrate. This ridge is marked by a large serrate spine at the angle of the preopercular. Lying directly proximal to this preopercular spine is a similar spine on the interopercular, serrate and just as large. This spine is obscured by the one on the pre- opercular and is easily overlooked. The posttemporal region has a number of protruding spines, some of which are serrate.

Pigment on Anthias larvae varies among the species; however, there seems to be a generic pattern. Usually a small melanophore is at the posterior end of the anal fin, one is on the ventral edge of the caudal peduncle, one or a few spots are at the base of the central rays of the caudal

fin, usually a few spots are on the skull above the optic lobes, and in some species there is a spot at the symphy- sis of the lower jaw. Dorsal trunk pigment varies among the species. In the Atlantic there are four types of Anthias larvae based on dorsal trunk pigment and other characters. Anthias Type 1 has a spot on the trunk mus- culature ventral to the middle of the dorsal fin (Fig. 16). In Anthias Type 2 there is a line of pigment ventral to the posterior part of the dorsal fin (Fig. 17). Anthias tenuis has some pigment between the dorsal body margin and the lateral line ventral to the last few rays of the dorsal fin (Fig. 18). Anthias Type 3 has no dorsal trunk pigment (Fig. 19).

18

b

Figure 17. — Larvae of Ant hias sp. Type 2 from the northwestern Atlantic Ocean: a) 4.7 mm; b) 5.7 mm; c) .H.4 mm.

19

Figure 18.— Larva of Anthias tenuis from the northwestern Atlantic Ocean, 6.7 mr

Figure 19. — Larva of Anthias sp. Type 3 from the northwestern Atlantic Ocean, 5.1 mm.

These four types of Anthias larvae are distinguished by other characters also. Of the types, Type 2 and Anthias tenuis have a protruding supraoccipital crest. Also Type 2 is scaled while it is still pelagic and has larval pre- opercular spines, pigment, and body shape. Although there are not enough specimens for a thorough analysis, there appear to be differences in the number of spines above the eye with Type 1 and Anthias tenuis having the fewest (about 4), Type 3 having more, and Type 2 having the most.

Anthias gordensis has no dorsal trunk pigment, no pro- truding supraoccipital crest, and a moderate number of spines in the ridge above the eye. Thus, Anthias gordensis larvae correspond most closely to Anthias Type

3 larvae among the Atlantic types. As more becomes known about the relationships within this genus, clues about the identity of Anthias Type 3 may be sought in the Atlantic species most closely resembling the Pacific Anthias gordensis.

Pronotogrammus . — There are three American species of Pronotogrammus: P. aureorubens in the Atlantic and P. eos and P. multifasciatus in the Pacific. These fish are more slender and have smaller heads than other Ameri- can anthiines.

In plankton samples, I found one larval type in each ocean which can be ascribed to Pronotogrammus on the basis of meristic features (Figs. 22, 23). The Atlantic type

20

a

b

Figure 20. — Larvae of Anthias gordensis from the eastern Pacific Ocean: a) 5.2 mm; b) fi.O mm.

has meristic features of P. aureorubens and the Pacific type those of P. eos. The two larvae share common char- acters that differentiate them from larvae of other genera. None of the fin rays are serrate or much elongate but the second and third dorsal spines form before the other elements and the pelvic fin is formed early. The opercular region is heavily armed with several serrate spines (Fig. 24). The preopercular has a long serrate spine at its angle and the interopercular has a similar

spine proximal to the one on the preopercular. The post- temporal and supracleithrum have protruding serrate spines. The supraoccipital crest protrudes as a simple ridge until it is "grown over" in late-stage larvae. There are several spines above the eye on a ridge of the frontal bone. Midlateral trunk pigment is characteristic of both types of Pronotogrammus larvae. In P. aureorubens this pigment consists of 5 to 12 dashes along the midlateral septum, roughly the length of the anal fin base. In P. eos

21

posttemporal

opercular

preopercular

'i \^ V supracleithrum

nteropercular

Figure 21. — Spine-bearing bones of the opercular and posttemporal regions on a 10.2-mm Anthias sp. Type I larva.

the pigment consists of a group of melanophores that form a blotch of pigment laterally on the trunk over the last few rays of the anal fin. Other pigment consists of several melanophores on the surface of the optic lobes of the brain. A few spots of pigment are at the base of the caudal fin on both species, and on P. eos some pigment is on the pelvic fin membrane. Lower jaw pigment is pres- ent on P. eos and two ventral midline spots are on P. aureorubens — one at the insertion of the anal fin and one midway on the caudal peduncle.

Hemanthias. — Hemanthias is represented by three American species: H. vivanus and H. leptus in the Atlan- tic and H. peruanus in the Pacific.

I found one type of Hemanthias larvae in each ocean, the Atlantic species with 47 lateral line scales (Fig. 25) is presumably H. vivanus, the most northerly species. From larval catches, H. vivanus is the most abundant anthiine off the east coast of the United States. The lar- vae from both coasts share several characters which sep- arate them from larvae of other genera. They are the most ornately spined anthiine larvae. They have well- developed spines in the opercular region, marked by an elongate, serrate spine at the angle of the preopercular and one underlying it on the interopercular. The inter- opercular also has a series of heavy spines ventral to the larger one. Both species have a protruding supra- occipital crest which is serrate, resembling a "cocks- comb." There are numerous ridges of complex spines on the frontals and parietals. The frontals also have a ridge of spines on their ventral margins. The posttemporal and supracleithral bones have protruding, serrate spines. The second through fourth dorsal spines develop early and the third spine is quite elongate early in development. The first dorsal spine develops after these three. The pelvic fin also develops early. In both species the pelvic fin spine is serrate and stout. In H. vivanus the second and third dorsal spines as well as the first anal spine are

also serrate. Pigment on H. vivanus consists of opposing spots on the trunk medial to the posterior extent of the second dorsal and anal fins. There is also a spot ventrally on the caudal peduncle and one at the caudal fin base. Spots also occur on the dorsal surface of the optic lobes. Hemanthias peruanus has pigment on the membranes of the spinous dorsal and pelvic fins (Fig. 26). Both species become scaled with ornate ctenoid scales while they still possess larval characters.

Hemanthias vivanus. — Meristic element develop- ment (Table 4) — In the smallest recognizable larvae on hand (3.1-4.0 mm), some meristic elements are already forming: the third dorsal fin spine and the pelvic spine are present. The third dorsal spine remains prominent throughout development with pronounced serrations. By 4.1-4.5 mm the dorsal fin has added one or two spines (the second and fourth) and some specimens are forming gill rakers. Between 4.6 and 5.0 mm the vertebral column has two to six vertebrae ossified, the first dorsal fin has three to five spines, the anal fin had one spine on about half of the specimens, and the branchiostegal rays range from two to six. By 5.1-5.5 mm, six branchiostegal rays, one spine, and one ray in the pelvic fin, and four to nine first dorsal spines are present. Between 5.6 and 6.0 mm the vertebral column is adding vertebrae and 14 to 20 are seen, the dorsal fin has 8 or 9 spines, there are 7 to 9 gill rakers, and branchiostegal ray complement is complete at 7. Between 6.1 and 6.5 mm there are 18 to 23 verte- brae, 9 or 10 dorsal spines, 1 to 4 pelvic rays, and 8 to 12 gill rakers. The second dorsal, anal, pectoral, and caudal fins are developing rapidly during this period showing wide ranges of numbers of elements formed. Between 6.6 and 7.5 mm, the same pattern of wide ranges of formed elements is seen, indicating rapid formation. Some spec- imens have adult complements of vertebrae, second dor- sal, anal, caudal, and pelvic fin rays. By 7.6-8.0 mm, the adult complements of all characters examined except gill rakers are present. Notochord flexion takes place at 5.5 mm and specimens >6.0 mm are developing scales. That is, scales form when the only meristic character with its adult complement is the branchiostegal rays.

Serrae are well developed on the third dorsal spine and the pelvic spine. The second and fourth dorsal and the first and second anal spines also develop serrae.

Body proportions (Table 5) — Larvae from 3.1 to 9.2 mm were measured to trace the proportional develop- ment of several body dimensions. The only proportion that shows noticeable difference with size is total length, which increases from 110' V of BL to 120% of BL as the larvae grow. This difference reflects the formation and growth of the caudal fin. Greatest body depth remains fairly constant ranging from 29 to 44' c of BL; the depth at the anus is 30* c of BL; and caudal peduncle depth is 12' b of BL. Head length is about equal to greatest depth ranging from 34 to 46' b of BL, snout length is 12c? of BL as is eye diameter. The preanal length ranges from 53 to 65'V of BL. Thus, H. vivanus larvae are large headed and deep bodied with a trunk that tapers quickly from the anus to the caudal peduncle.

22

a

b

c

Figure 22. — Larvae of Pronotogrammus aureorubens from the northwestern Atlantic Ocean: a) 4.6 mm; b) 6.0 mm; c)

9.(1 mm.

Head spine development (Figs. 27, 28) — The serrate "cockscomb" spine protruding from the supraoccipital is pronounced in the smallest larvae (3.1 mm) observed. Serrate ridges above the eye and transverse ridges of spines on the surface of the cranium as well as beginnings of spines in the posttemporal and opercular region are also present. During larval growth, these spines increase in complexity and size, those in the opercular area show- ing the greatest change (Figs. 29, 30). The supra- occipital crest remains the largest cranial spine and is strongly serrate. The ridges on the cranium form a com- plex pattern covering the area between the eyes and the supraoccipital region (Fig. 28). The serrate supraorbital

ridges develop about eight strong, blunt points. In the opercular region, the preopercular has a long serrate spine at its angle and several spines along the rest of the posterior margin of the bone (Fig. 30). The inter- opercular has a long serrate spine which lies directly proximal to the long spine on the preopercular and several smaller spines ventral to the long one. The sub- opercular has a few small spines on its posterior margin and some are serrate. The opercular bone develops the three spines characteristic of serranids on its posterior margin during the larval period. The posttemporal has several serrate spines and the supracleithrum has a pos- teriorly directed, elongate serrate spine. Apparently

23

b

Posttemporal

supracleithrum

preopercular

Figure 23. — Larvae of Pronotogrammus eos from the eastern Pacific Ocean: a) 3.6 mm; b) 7.8 mm.

these spines are resorbed in the late larval or early juvenile period but specimens to describe these events are lacking.

Pigmentation — Pigmentation is present on the small- est larvae examined (3.1 mm) and remains essentially unchanged during larval development. Two spots are present on the ventral midline of the caudal peduncle, one just posterior to the insertion of the anal fin and the other, slightly smaller, at the anterior end of the caudal fin base. Another spot develops at the base of the ventral lobe of the caudal fin and extends along the bases of several rays. Dorsally a spot is present on some larvae at the insertion of the second dorsal fin. This spot seems to be quite superficial, seen on only one side of the larva in some cases, and apparently abraded off of other larvae. A few melanophores are present at the symphysis of the lower jaw, on the dorsal surface of the optic lobes of the brain, on the fin membrane between the third and fourth dorsal spines, and on the membranes between the rays of

Figure 24. — Spine-bearing bones of the opercular and posttem- poral regions on a 9.5-mm Prontogrammus aureorubens larva.

24

a

b

Figure 25.— Larvae of Hemanthias vivanus from the northwestern Atlantic Ocean: a) 4.2 mm; b) 5.3 mm; c) 6.8 mm.

25

Figure 2(>.— Larva of Hemanthias peruanus from the eastern Pacific Ocean, 9.3 mm.

post temporal

posttemporal

opercular

preopercular

supracleithrum subopercular

ropercuiar

Figure 27.— Spine-bearing bones of the opercular and posttemporal regions of a 4.3-mm Hemanthias vivanus larva.

the pelvic fins. Internal pigment on the dorsolateral sur- face of the gut shows through the body wall.

Discussion. — Larvae of four genera of American anthiines have been found. These larvae are separated from other serranids by their deep bodies, large heads, produced preopercular and interopercular spines, a tendency for development of armature on the head, and serrations of fin spines. Within the anthiines, the genera can be ranked according to the development of these characters. Plectranthias has little development of these characters. The preopercular spine is elongate and ser- rate, but the interopercular spine is not serrate. There is no protruding supraoccipital crest, neither the frontal

supracleithrum

preopercular

interopercular

Figure 28. — Spine-bearing bones of the opercular and posttemporal regions on a 10.3-mm Hemanthias vivanus larva.

ridge over the eye nor the fin spines is serrate. Anthias develops more extreme larval characters. The pre- opercular and interopercular spines are stronger and both are serrate. The ridge above the eye has several ser- rae. In some species there is a nonserrated protruding

26

Table 4. — Meristic character development of larvae of Hemanthias vivanus. Specimens between dashed lines are undergoing notochord flexion. For the caudal fin, "P" and "S" indicate primary and secondary fin rays.

Body

Branchio- stegal rays

Verte- brae

Pec- toral fin

Pel- vic fin

Caudal fin

Anal fin

Dorsal fin Spines Rays

length (mm)

Dorsal Ventral S P P S

Gill

rakers

3.2 NL

3.3

3.5

3.6

3.7

3.8

3.9

4.0

4.1

4.2

4.3

4.4

4.5

4.7

4.8

4.9

5.0

5.2

5.3

5.4

3 3 3 2 3 3 4 4

3 5 3 6 6 5 5 5 I J

III

111 111 III III III III III III 111 IV 1\ IV IV

III

V V V

VI IV V

2 2

3

damage

5.5

15

11

IX

5.6 SL

5.6

5.7

5.8

5.9

6 I

6.3

6.4

6.5

6.8

7.0

20 14 18 16 16 15 14 18 18 23 18 20 22 23 20 26 18

II

14

II

II. 1

11,1

II

11,1

II

11.9

II. 1

11,1

[1,9

11,1 11,1

11,9 11,1

IX VIII

IX

IX VIII

IX VIII X X X X

IX X X X X X

14

12

8 9 10

11

13

9

27

Table 4.— Continued.

Bodv	Branchio- stegal	Verte-	Pec- toral	Pel- vic		Cau	dal fin		Anal	Dorsal fin
length	Dorsal	Ventral	Gill
(mm)	rays	brae	fin	fin	S	P	P	S	fin	Spines	Rays	rakers
'7.1	7	26	14	1,5		8	7		11,9	X	14	13
'7.3	7	26	16	1,5		9	8		11,9	X	14	14
'7.3	7	23	2	1,2		6	4		n,i	X		12
'7.7	7	26	17	1,5		9	8		11,9	X	14	15
'7.7	7	26	17	1,5		9	8		11,9	X	14	15

'Scales present.

Table

-Body proportions of larvae of Hemanthias vivanus. Specimens between dashed lines are undergoing notochord flexion.

			Percent of standard length
Body						Pre-	Caudal	Depth
length	Total	Body	Eye	Head	Snout	anal	peduncle	at
(mm)	length	depth	length	length	length	length	depth	anus
3.1 NL	106	36	14	34	12	60	8	28
4.0	103	29	10	35	11	57	5	:\2
4.4	114	39	11	37	14	53	11	31
4.6	108	36	11	41	15	58	10	27
4.6	112	38	12	42	15	55	10	30
4.6	109	36	12	39	14	57	11	27
4.6	109	38	12	42	14	59	11	27
4.7 SL	116	41	15	43	16	61	12	35
5.0	116	38	13	39	13	58	11	30
5.0	113	41	15	48	15	60	14	38
5.2	116	37	12	40	12	55	12	31
5.2	118	39	13	41	1(1	60	13	31
5.4	123	40	15	43	15	63	14	30
5.5	117	44	16	43	15	62	14	36
5.7	121	38	13	40	15	58	L3	31
6.0	118	43	16	46	13	65	16	35
6.0	122	42	14	44	15	60	14	36
6.3	122	39	13	44	13	60	14	31
6.5	124	40	14	42	14	61	14	32
6.8	117	33	13	38	14	(11	13	29
7.0	124	42	14	43	11	61	14	32
7.8	124	37	14	42	14	59	13	29
8.0	113	40	11	38	11	54	11	30
8.1	122	40	15	41	14	60	11	32
9.2	121	35	12	37	14	55	12	28

supraoccipital crest. The fin spines are strong and the third dorsal spine and the pelvic spine are elongate. Some species develop ctenoid scales while retaining lar- val characters. Pronotogrammus develops more extreme larval characters including a strong protruding supra- occipital crest and heavy opercular armature. These lar- vae, however, are not scaled. Among American anthiines, Hemanthias exhibits the most extreme development of larval characters. The head is armored with a serrate supraoccipital crest, numerous serrate ridges on the fron- tals and parietals, and heavy serrate spines on bones in the opercular series. Ctenoid scales develop during the larval period. In Hemanthias vivanus larval anthiine characters reach their apex with the development of ser- rations on several dorsal and anal fin spines.

Figure 29. — Development of the supraoccipital crest of Hemanthias vivanus: a) 4.3 mm; b) 5.8 mm; c) 6.6 mm.

28

Figure 30.— Details of the head spination on a Hemanthias vivanus larva, 7.2 mm.

29

Subfamily Epinephelinae

The Epinephelinae is represented by three genera in American waters from both oceans. These are Paranthias with 1 species, Mycteroperca with 13 species, and Epi- nephelus with 21 species grouped in 5 subgenera (Table 1). Epinephelus is worldwide in warm water while the other two genera are indigenous to the Americas. The monotypic genus Gonioplectrus has close affinities to the epinephelines based on larval and adult characters (Ken- dall and Fahay in press). Members of this subfamily are quite disparate in size as adults, some reaching only 30 cm, while others reach 180 cm and a weight of 320 kg. Until recently the subgroupings of Epinephelus were given generic status but Smith (1971) regarded them as subgenera.

Larvae of several species of Epinephelus have been il- lustrated and described. The species are outstanding in having the second spine of the dorsal fin and the pelvic spine strong, serrated, and produced to nearly the length of the larva. These larvae are frequently associated with the entire family and have been cited as representatives of the family (e.g., Rosenblatt and Zahuranec 1967; Smith 1971). Among the earliest published illustrations of Epinephelus larvae are those of Fage (1918) who called them Serranus scriba. Bertolini (1933) reviewed Fage's work and indicated that the fish was Epinephelus guaza. Sparta (1935) figured the eggs of Epinephelus guaza and the larvae of Epinephelus alexandrianus. Fage's (1918) figures and description were reviewed by Vodjanitzki and Kazanova (1954). Ukawa et al. (1966) and Mito et al. (1967) described larval development of reared Epine- phelus akaara from Japan. Guitart Manday and Juarez Fernandez (1966) described eggs and early stage larvae of Epinephelus striatus obtained from aquarium held adults.

Aboussouan (1972) illustrated and described three types of larvae from the west coast of Africa that he ascribed to the genus Epinephelus. However, only one of these, E. aeneus, has the larval characteristics of an epi- nepheline. The other two do not have elongated fin spines; in fact, the spinous dorsal and pelvic fins develop after the soft dorsal and anal. The preopercular margin is comprised of several spines with the one at the angle only slightly longer than the rest and not serrate. Pigment is made up of small melanophores as opposed to the large caudal peduncle blotch of epinephelinesm. These larvae are almost certainly not epinephelines, rather one is pos- sibly a sciaenid and the other a haemulid, judging from the number of fin elements and general larval morphology. Presley (1970) described Epinephelus niveatus larvae from the Florida Straits. Smith (1971) il- lustrated a larval Epinephelus from off the Carolinas. Fowler (1944) described a new genus and species (Serri- hastaperca exul) from a larval fish from the eastern Pa- cific. As Heemstra (1974) has shown, this is probably a larval Epinephelus, possibly Epinephelus panamensis. Fourmanoir (1976) illustrated and briefly described three epinepheline larvae from the southeastern Pacific. Ken- dall and Fahay (in press) described a larval Gonio-

plectrus hispanus and concluded that although it had some characters of the anthiines, it was more closely related to the epinephelines.

Although the figures are not as detailed as might be desired, and the descriptions are generally brief, most Epinephelus larvae described thus far appear quite similar. They all possess stout, elongate, serrate pelvic fin spines and second dorsal spines. The preopercular is armed with an elongate spine at its angle and several smaller spines. The first spine of the dorsal fin is short and serrate, the second long and serrate, and the third somewhat elongate and serrate. The succeeding spines become progressively shorter and less robust until it is difficult to distinguish the last dorsal spine from the first dorsal soft ray. Pigmentation generally consists of a few spots on the surface of the skull, a spot at the junc- tion of the cleithra, internal pigment lateral to the gut, and a large spot on the caudal peduncle. The elongate fin spines also have some pigmentation on their membranes.

The larval material I have examined contains many species of epinepheline larvae all agreeing closely with the above descriptions. There are no forms which possess only some of this suite of characters, rather they are all uniform and distinguishable from other types of serranid larvae. Besides specimens having meristic characters of Epinephelus sp., there are larvae representing the two other American genera, Mycteroperca and Paranthias. Larvae of these genera can be separated at present on the basis of meristic characters but not larval morphology.

Paranthias. — The genus Paranthias contains one species (P. furcifer) which is found on both sides of the Americas. It is separated from other epinephelines by morphology and habits. It converges with anthiines in several of its adaptations, although on skeletal and other characters its affinities clearly lie with the epinephelines (Table 1). As opposed to the demersal, piscivorous habits of other members of the subfamily, P. furcifer occurs in aggregations somewhat above the bottom where it feeds on plankton (Randall 1968). Morphological character- istics related to these habits include a deeply forked caudal fin, dorsal and ventral profiles nearly equally curved, a smaller head with a small upturned mouth, and more gill rakers (35-40) than other epinephelines (Smith 1971).

As larvae, the three genera of American epinephelines can only be separated using meristic characters. These are sufficiently formed only late in development. Sepa- ration of Paranthias from the other two genera depends on the number of dorsal fin rays. The posteriormost spines in the dorsal fin of epinephelines are short, late forming, and first form as rays. The total count of dorsal spines and soft rays in Paranthias (27-28) is commonly seen in members of the other two epinepheline genera also. At present, only large, well-developed larvae of Paranthias furcifer have been identified.

As mentioned, Paranthias furcifer larvae share larval characters with all other epinephelines found so far (Fig. 31). The pelvic spine and second spine of the dorsal fin are long and serrate. They bear pigmented "flags" near

30

Figure 31. — Larva of Paranthias furcifer from the eastern Pacific Ocean, X.(i mm.

their tips, which are often abraded off in netted speci- mens. The first and third dorsal spines and second anal spine are also serrate. The third dorsal fin spine is mod- erately elongate. The spine at the angle of the pre- opercular is long and serrate and there are usually two spines on the preopercular bone dorsal and ventral to the spine at the angle. There is a serrate ridge of spines dor- sal to the eye and the supracleithrum has a protruding blunt serrate spine. Pigment spots are present over the optic lobes of the brain and laterally over the gut. There is a large, intense blotch midlaterally on the caudal pe- duncle.

A 34-mm specimen collected at the surface as part of a large school showed loss of larval characters. The body was fully scaled and the supraocular ridge and the pro- truding supracleithral spines were unpronounced. The second dorsal spine was only slightly larger than the third and all were well formed and nearly equal in length. The second spine retained minute serrations. The pelvic spine also retained serrations and was about as long as the second dorsal spine, not reaching the posterior end of

the pectoral fin. The first ray of the pelvic fin was longer than the spine. The spine at the angle of the pre- opercular was still slightly elongate and serrate; but rather than two spines on each side of the spine at the angle, there were several (6 ventral and 20 dorsal). The specimen I examined was cleared and stained so no as- sessment of pigment pattern could be made.

Epinephelus. — This is the only American epi- nephrine genus with representatives in other parts of the world. Epinephelus is the most speciose genus of ser- ranids, containing 21 American species grouped in 5 sub- genera. Fifteen species occur in the Atlantic and 11 in the Pacific, with 5 species common to both oceans. They are separated from other American epinephelines by a com- bination of characters (Table 1). Those species with IX dorsal spines have fewer than 18 rays, whereas Paranthias furcifer has a dorsal fin count of IX, 18-19. Epinephelus species differ from Mycteroperca by having 8 or 9 anal rays, rather than 10 to 12. Other diagnostic

31

characters include characteristic skull crests and the presence of a postocular process (Smith 1971).

As discussed earlier, larvae of several species of Epi- nephelus from around the world have been described (e.g., Bertolini 1933; Mito et al. 1967; Fourmanoir 1976). Presley (1970) described larvae he assigned to Epi- nephelus niveatus from Florida. His description is rather brief, and considering the similar appearance of all epi- nephrine larvae that have been described and all those I have found, it is probably inadequate to separate that species from others in the area.

Larval epinephelines are not separable into genera un- til medial fin rays have formed. Only on the basis of these counts could Epinephelus larvae be distinguished from those of Mycteroperca (by a lower anal fin ray count) and from Paranthias (by a different dorsal fin count). Two larval specimens of Epinephelus are illustrated (Figs. 32, 33). They are both about the same length and appear similar. One specimen (Fig. 32) from the Atlantic has XI, 16 dorsal fin elements and III, 9 anal fin elements making it most likely a member of the Epinephelus striatus species group and probably either E. morio or E. gut- tatus. The other specimen (Fig. 33) is from the eastern

Pacific with XI, 19 dorsal fin elements and III, 9 anal fin elements making it most likely either E. dermatolepis dermatolepis or E. alphestes multiguttatus . Although these specimens are in different subgenera, no charac- ters other than meristics could be found to distinguish them from other genera within the subfamily.

The following account describes features of all larval Epinephelus specimens seen so far, since no characters have yet been found to identify the species. Most of the characters mentioned with the account of the subfamily apply to the specimens identified as Epinephelus sp. The second spine of the dorsal fin is elongate and serrate as is the pelvic spine. The first and third dorsal fin spines are also serrate and the fourth may be as well. The anal fin forms with only the first two spines as such until late in development. The third anal fin spine forms as a ray and later becomes a spine. The first two anal spines may be serrate. The preopercular margin is armed with an elon- gate, serrate spine at its margin and one or two blunt spines dorsal and ventral to the one at the angle. Later in development more spines form on this margin and it becomes serrate. There is a serrate ridge above the eye and the posttemporal and supracleithral bones have pro-

Figure 32. — Larva of a member of the Epinephelus striatus species group from the northwestern Atlantic Ocean,

7.6 mm.

32

Figure 33. — Larva of an Epinephelus sp. (subgenus Dermatolepis or Alphestes) from the eastern Pacific Ocean,

8.4 mm.

trading, serrate spines. The pigment pattern consists of a on the lateral surface of the gut, and a large blotch on the

few melanophores on the surface of the optic lobes, some caudal peduncle. The caudal peduncle blotch is var-

33

iously placed, from along the ventral midline to laterally on the lateral line. It tends to become internal on larger specimens. The elongate dorsal and pelvic spines are var- iously pigmented but the pigment is absent on many net- ted specimens. The pigment is seen on the membranes associated with the extended portion of the spines.

Besides the transformation of the third anal spine from a ray to a spine during development, as many as two pos- terior spines of the dorsal fin undergo similar transfor- mation. The second dorsal fin spine is elongate and the third through seventh are progressively smaller. The 7th through the 11th dorsal spines are quite small during lar- val development. It seems that the first nine spines develop as such, but the last two develop as soft rays and later transform into spines. However, in specimens which have only 9 or 10 spines as adults, no such transforma- tion may occur.

The bones in the opercular region of a 10-mm Epi- nephelus sp. (possibly E. niveatus) are characteristic of the genus (Fig. 34). The preopercular bone has the most produced spines of any of the bones in the region. As mentioned above, it has a long, serrate spine at its angle and two spines dorsal and ventral to the one at the angle. These spines may also be serrate. The opercular has a broad fan shape with three blunt spines. The area ventral to the lower spine is broad. The subopercular has a spine on its posteroventral margin. The interopercular has a spine on its posterodorsal edge. The posttemporal and supracleithrum each have a protruding serrate spine.

posttemporal

opercular

supracleithrum

preopercular

Figure 34. — Spine-bearing bones of the opercular and posttemporal regions on a 10.2-mm Fpinephelus niveatus larva.

Mycteroperca. — This is the second most speciose epi- nepheline genus with eight species in the Atlantic and five in the Pacific Ocean (Table 1). They are distin- guished from the other two epinepheline genera by hav- ing 10 to 13 anal rays and straight, parallel lateral skull crests that join the rim of the orbit anterior to the mid- orhital area (Smith 1971).

Larvae with developed III, 10-13 anal fin elements and epinepheline larval characteristics were found in samples from both oceans. They had other diagnostic characters of Mycteroperca. These larvae had the full suite of char- acters of epinepheline larvae and, on the basis of larval characters, could not be separated from other epi- nepheline larvae. They had elongate, serrate second dor- sal and pelvic spines and a long serrate spine at the angle of the preopercular. There was a serrate, bony ridge above the eye and protruding, serrate spines in the post- temporal region. They had a few pigment spots on the dorsal surface of the optic lobes of the brain, some on the lateral surface of the gut, and a large blotch on the caudal peduncle. Some pigment was present on the membranes of the elongate fin spines.

Mycteroperca microlepis (Figs. 35, 36). — The mer- istic characteristics of Mycteroperca microlepis fall within the ranges for several species of Mycteroperca in the western Atlantic (Table 1). However, among these species, M. microlepis is the most northerly occurring. Late larvae and early juveniles with characteristics of M. microlepis have been caught in the Middle Atlantic Bight.

In plankton samples from as far north as Cape Hat- teras, N.C., large epinepheline larvae (>7 mm) with meristic characters of M. microlepis are frequently found. In these samples smaller larvae with similar larval characters but incomplete meristic characters are also found. The following description is based on these epi- nepheline larvae taken off North Carolina. Larger larvae and juveniles (12-35 mm) from nearshore and estuarine collections along the Atlantic coast from Florida to New Jersey were examined to note the development of juvenile pigment and the regression of the elongate spines of early larvae. No other species of Mycteroperca could be identified in the samples with certainty.

Meristic element development (Table 6) — Develop- ment of most meristic elements occurs over a length range of 4.0-9.8 mm. In the smallest larvae available (4 mm), the dorsal fin already has the first three spines ossi- fied, the pelvic fin has one spine and one ray, and there are three branchiostegal rays and two gill rakers. Be- tween 5 and 6 mm, the notochord is flexing and most meristic features are forming ossified elements. The ossi- fied vertebrae increase from 4 to 18, forming from ante- rior to posterior except for the urostyle. Several verte- brae are partially developed in most specimens. The branchiostegal rays reach their adult complement of 7 and there are 6 to 10 gill rakers. The pelvic fin has three rays ossified and the pectoral fin starts to form rays. The dorsal fin increases from three to five spines and the anal starts to form. The caudal fin has its adult complement of nine dorsal and eight ventral primary rays as the no- tochord becomes fully flexed at 6 mm. When notochord flexion is complete, by 7 mm, most meristic features are well formed, approaching adult counts. Vertebrae in- crease to 22 with the urostyle forming prior to the 2 or 3 vertebrae anterior to it. The pectoral fin increases to 12 rays and the pelvic fin attains its adult complement of 1

34

spine and 5 rays. The anal fin increases rapidly having 2 spines and 11 rays. The first soft ray of the anal fin becomes the third spine later in development, as is char- acteristic of most adult serranids with three anal spines. The dorsal fin is difficult to count separately because the posterior spines are quite small at formation and, like the third anal spine, originate as soft rays. It appears that there are 10 spines and 15 rays in the dorsal fin. In addi- tion to the 9 or 10 gill rakers on the lower limb, 1 forms on the upper end of the first gill arch. Procurrent caudal rays start to ossify. Between 7 and 10 mm, the meristic elements, except gill rakers and scales, reach their adult complements. Not until about 20 mm do the 3rd anal spine and 11th dorsal spine change from soft rays to spines.

Body proportions (Table 7) — Proportions of various measurements of larvae and juveniles of Mycteroperca microlepis to body length (BL) were examined. Most proportions show little ontogenetic change; however, total length increases sharply with the formation of the caudal fin and the depth at the anus and caudal pedun- cle depth also increase between 4 and 6 mm. The greatest body depth remains between 29 and 35% of BL through- out the size range examined. The head length stays about 31 to 43% of BL. The preanal length is 51 to 66% of BL. Eye and snout length are about 9 to 14% of BL. The elongated larval spines were measured to trace their relative growth. The second dorsal spine was broken on several smaller specimens but it appeared to increase from 40% of BL at 4.6 mm to 60% from 5 to 10 mm and then decrease to 11% by 35 mm. The pelvic spine showed a similar pattern of increase and decrease, reaching a maximum relative size of 68% of BL at 8.3 mm and de- creasing to 12% of BL by 35 mm. The preopercular spine increases from about 0.5% of BL at 5 mm to 1.1% at 8 mm and decreases to 0.2% by 35 mm.

Pigmentation — The pigment pattern seen on the smallest larvae examined remains basically unchanged throughout larval development. On the heads of 4-mm larvae there is a spot on the dorsoposterior surface of each optic lobe, and there are several spots in the same area on 6-mm specimens. A spot develops internally on the nape near the dorsal fin origin which later occurs in the area of the neural spines of the first two vertebrae. The dorsolateral surface of the gut cavity is covered with rather dense pigment throughout larval development. Ventrally, on most specimens there is a spot at the cleithral junction. A large spot occurs on the ventral region of the caudal peduncle in small larvae. In 7.5-mm specimens this spot has moved dorsally to lie over the lateral line in the midlateral caudal peduncle area. This spot extends internally to the area of vertebrae 20 to 21. The size and vertical position of this spot varies some- what among the specimens observed from along the ventral midline to the lateral line. A few small spots that develop on the ventral tip of the notochord lie near the bases of the caudal rays as the caudal fin is formed.

Opercular series spines (Fig. 36) — Spine-bearing bones of the opercular region of a 9.8-mm specimen of Mycter-

operca microlepis appear similar to those of other epi- nephelines. The preopercular bone has an elongate, serrate spine at its angle which reaches a maximum of 1.3% of BL in 8-mm specimens. Smaller spines form along the vertical and ventral edges of the preopercular. During larval development, there are two spines in each area with all but the uppermost spine being serrated. Later in development, as the elongate spines regress, additional small spines appear along the edge of the pre- opercular to form the serrate margin of the adult. The interopercular bears two blunt spines along its posterior margin. The subopercular has one blunt spine at its ventralposterior angle. The opercular is fan-shaped with three blunt spines on its posterior margin. The supra - cleithrum has a posteriorly directed, serrate spine and the lateral posttemporal ridge is serrate.

Discussion. — The 35 presently recognized species of epinephelines in American waters constitute a cohesive group (Smith 1971). Recently, five genera have been given subgeneric rank, leaving the species grouped in only three genera. Epinephelus, the most speciose Amer- ican genus with 21 species, is a speciose genus in other parts of the world also, e.g., there are 34 species in Japan (Katayama 1960). Among American genera, Paranthias has evolved obvious ecological and morphological dis- tinctions from the general "grouper" characteristics. A clearer understanding of the relationships within this subfamily must await its comprehensive revision on a worldwide basis.

Epinepheline larvae representing all three American genera were found. Their identification was based solely on meristic characters. No larvae of epinephelines could be separated on the basis of any larval characters. They shared all recognized larval characters. Thus it appears that evolutionary differences within this group are not reflected in larval morphology.

Epinepheline larvae are quite distinct from larvae of other serranids. The other subfamily with serrated spines, Anthiinae, does not have the extremely elongate, serrate fin spines seen in epinephelines. In anthiines the interopercular has a produced spine which lies under the spine at the angle of the preopercular, both of which are usually serrate. Epinephelines have only a blunt spine on the interopercular. Anthiines also are heavier bodied than epinephelines. Gonioplectrus shares trenchant lar- val characters with epinephelines but has less variation in dorsal spine length and a stouter body, both reminis- cent of anthiine larvae. These resemblances to anthiine larvae may represent convergence. However, Gonioplectrus may represent an evolutionary plateau between anthiines and epinephelines; this remains for further study to elucidate. Serranines have simple ar- mature on the preopercular and no fin spine serrations. Grammistines have the second spine of the dorsal fin produced into a long, filamentlike structure, in contrast to its stout-serrated form in epinephelines. Apparently, in American waters at least, intermediates between epinephelines and other groups of serranids have been lost.

35

a

b

36

d

Figure 35.

-Young stages of Mycteroperca microlepis from the northwestern Atlantic Ocean:

mm; d) 22. (i mm.

a) 4.0 mm; b) 7.4 mm; c) 14.2

Table 6. — Meristic character development of larvae of Mycteroperca microlepis. Specimens between dashed lines are undergoing notochord flexion. For the caudal fin, "P" and "S" indicate primary and secondary fin rays.

Body	Branchio- stegal	Verte-	Pec- toral	Pel- vic		Caudal fin		Anal	Dorsal fin
length	Dorsal	Ventral	Gill
(mm)	rays	brae	fin	fin	S	P	P	S	fin	Spines	Rays	rakers
4.0 NL	3			1,1						UI		2
4.6	4			1,2						III		3
5.1	5	4		1,2		1	2			III		6
5.6	6	6		t,3		5	5			III		7
5.3	5	4		1,3		4	4			III		6
5.3	6	6	2	1,3		6	6			IV		8
5.3	6	8	2	1,3		7	7			IV		H)
5.5	7	18	6	damage	9	8		1,6	V		9
6.0 SL	7	20	9	1,4		9	8	1	11,9	X	11	1+9
6.1	7	19	6	1,4		9	8		1,6	Yl		9
7.0	7	22	12	1,5	1	9	8	2	11,11	X	15	1 + 10
7.4	7	24	12	1,5	1	9	8	2	11,11	X	15	2+ 9
7.9	7	24	14	1,5	3	9	8	3	rj.,13	X	19	3 + 10
8.1	7	24	15	1,5	3	9	8	2	n,i2	X	18	3+10
9.2	7	24	17	1,5	4	9	8	5	11,11	X	18	3+12
9.8	7	24	16	1,5	5	9	8	5	11,12	X	18	4 + 11

■M

Table 7. — Body proportions of larvae of Myctoperca mierolepis. Specimens between dashed lines are undergoing

notochord flexion.

Percent of standard length

Body

length (mm)

Total

length

Body depth

Eye

length

Head length

Caudal Snout Preanal peduncle length length depth

Depth

at anus

Second spine of dorsal fin

Pelvic

fin spine

Preoper- cular spine

4.6 NL

104

30.4

10.9

32.6

9.6 52.2 7.6

15.2

39.1

-

0.43

5.3 5.3 5.3 5.5 5.6 5.9

106 109 L13

110 107 112

32.1 34.0 34.0 36.4 32.1 32.2

11.3

11.9 12.1 12.7 11.1 10.8

32.1 35.8 35.8 40.0 35.7 42.4

9.4 10.6 12.1 13.6 10.9

50.9 56.6

54.7 56.6 55.4 64.4

8.5

8.3 9.4

10.2 7.9

10.5

18.9 20.8 26.4 25.5 19.6 22.0

69.2

49.1

54.7 58.5

0.58 0.75 0.94 0.91 0.55 1.02

6.0 SL	120	36.7	13.0	38.3	11.5	60.0	11.5	23.3	61.7	61.7	0.93
6.1	120	34.4	12.3	41.0	13.0	67.2	10.2	26.2	—	59.0	0.92
7.0	120	34.3	12.1	37.1	11.6	61.4	11.6	24.3	54.3	61.4	0.99
7.4	122	33.8	11.9	39.2	11.5	60.8	11.1	21.6	—	62.2	1.15
7.9	120	32.9	11.9	38.0	10.8	60.8	11.9	24.1	—	60.8	1.04
s 1	121	34.6	9.9	37.0	11.6	59.3	13.6	24.7	59.3	58.0	0.99
8.3	119	33.7	12.0	38.6	13.3	57.8	12.0	25.3	65.1	67.5	1.33
9.2	123	33.7	12.0	38.0	12.0	57.6	12.0	23.9	59.8	64.1	1.02
9.4	121	35.1	12.8	38.3	10.6	59.6	12.8	25.5	58.5	63.8	1.06
9.8	123	33.7	11.2	39.8	14.3	66.2	11.2	26.5	60.2	65.3	1.02
12.0	123	35.0	11.7	38.3	10.8	57.5	13.3	28.3	49.2	55.0	1.00
13.6	125	33.1	9.6	37.5	11.8	58.1	11.8	27.9	39.7	42.6	0.92
14.0	131	33.6	10.7	37.9	11.4	62.9	10.7	25.7	39.3	47.9	—
16.8	123	32.7	10.1	39.3	9.5	61.9	11.9	25.6	40.5	39.3	0.83
17.6	122	31.8	11.4	35.3	8.5	58.0	10.2	24.4	40.3	46.0	0.91
19.3	115	31.1	10.4	36.3	9.3	58.5	10.9	24.9	21.3	20.2	0.62
21.1	122	31.8	10.9	31.3	13.7	60.7	10.9	27.0	20.4	29.4	0.52
21.5	120	30.2	10.7	37.2	10.7	52.1	11.2	25.6	20.5	17.2	0.65
24.4	125	32.8	11.1	42.6	11.1	63.5	11.1	32.8	13.5	14.8	0.33
35.4	120	28.5	9.0	36.2	9.0-	61.6	10.7	27.1	11.0	12.1	0.14

opercular

preopercular

supracleithrum

subopercular

interopercular

Figure 36. — Spine-bearing bones of the opercular and posttemporal regions on a 9.8-mm Mycteroperca mierolepis larva.

Subfamily Grammistinae

Larvae of three genera, Liopropoma (larvae of Pikea may be included here), Rypticus, and Pseudogramma,

have been recognized and will here be considered members of a single group, the subfamily Grammistinae. These larvae are similar in appearance and readily separable from larvae of other fishes. Considerable work has been done on the taxonomic affinities of these fishes but their relationships remain perplexing. Liopropoma has been considered a member of the serranid subfamily Liopropominae (Katayama 1960). Rypticus has been considered a member of the family Grammistidae, a closely related offshoot of the serranids (Gosline 1966). Pseudogramma has been considered a pseudogrammid, a group which has been considered a subfamily of the grammistids or a separate family (see Randall et al. 1971). Reasons for considering these fishes members of a single group are detailed in Kendall (1976, 1977).

Liopropoma. — Pikea has been included with Liopropompa in the subfamily Liopropominae. They have identical predorsal bone patterns and similar meristic formulae (Table 1). Chorististium has been syn- onymized with Liopropoma. Two genera, Joboehlkia and Flagelloserranus, described from preadult material, apparently belong with this group. Pikea has three north- western Atlantic species and Liopropoma includes five species in the northwest Atlantic, one in the eastern Pacific, and several from the central and western Pacific. Within this group, larvae of Liopropoma and/or Pikea

38

have been found in material from the northwest Atlantic. These larvae appear similar to those described by Kott- haus (1970) as the genus Flagelloserranus. He described two species of Flagelloserranus, one from the Indo- Pacific Ocean and one from the western Atlantic Ocean. Fourmanoir (1971, 1976) indicated that similar speci- mens from near New Caledonia were young of Liopropoma. All the characters Kotthaus (1970) used to define this genus and separate it from Liopropoma ap- pear to be transient larval or juvenile characters. The similarity of meristic and other characters, the unique predorsal bone pattern, and the lack of adults of Flagel- loserranus indicate that these are larval Liopropoma or Pikea. Thus, Kotthaus (1970) provides descriptive material for larval development of Liopropoma-Pikea so it will be considered only briefly here.

I considered larval types resembling Flagelloserannus, one with pigment on the ventral caudal peduncle surface and one without, members of the genera Liopropoma and/or Pikea. However, since the meristic characters and areas of occurrence of the several species of these two genera overlap, I did not attempt to allocate the larvae to species.

Their general body shape is similar to that of the ser- ranines, although the gut is shorter and there is a space between the anus and the origin of the anal fin (Fig. 37). The caudal peduncle is both longer and deeper than it is in serranines.

The most outstanding developmental feature is the presence, even in small larvae, of two elongate, thin dor- sal spines. These develop before other fin rays, reach a length of up to three times the fish length and become the second and third dorsal spines. These spines are deli- cate and are broken in many specimens. Kotthaus (1970) described the presence of thick tissue surrounding these spines; the tissue around the second spine having two vanelike swellings on its distal third, the tissue around the third spine being tubular for its entire length. The distal portion of both spines is pigmented with several large melanophores. The remaining fin rays develop their adult proportions without any pronounced elongations. The ventral fins develop more slowly than those of most other serranids. The pectoral fins do not develop precociously, as they do in Rypticus and Pseudogramma.

Except for the pigment on the elongate dorsal fin spines, most larvae examined were unpigmented. Prior to notochord flexion some larvae bear a melanophore on the ventral midline anterior to the insertion of the anal fin. Whether this spot disappears in larger larvae or larger specimens with this pigment were not collected cannot be determined presently. Some spots develop on the hindbrain surface in larger larvae, probably representing the onset of juvenile pigment.

The bones of the preopercular region are armed with characteristic spines (Fig. 38). The preopercular has four blunt spines above and three below the one at the angle. Flagelloserranus meteori from the Indo-Pacific has a similar preopercular with two spines above and three below the one at the angle (Kotthaus 1970). Flagelloser- ranus danae from the western Atlantic has a similar pre-

opercular except it has a hooked dorsalmost spine and the vertical limb is longer than in F. meteori (Kotthaus 1970). The opercular of the larvae examined in this study as well as those of Kotthaus (1970) is fan-shaped with three spines on its margin. Both the subopercular and interopercular have a blunt spine on the posterior edge in the larvae I examined; however, Kotthaus (1970) reported a smooth margin in both species he recognized. A low posttemporal ridge that penetrates the epidermis of the larvae I examined was not mentioned by Kotthaus (1970).

Rypticus. — Occurring in the Atlantic and eastern Pacific Oceans, this genus has meristic characters that are unusual among serranids (Table 1) and small, em- bedded cycloid scales. The unique gonadal morphology (Smith and Atz 1969) and predorsal bone patterns (Ken- dall 1976) also indicate that Rypticus is highly special- ized and no other known fish can be derived from it. These are secretive, nocturnal reef fishes of small to moderate size (50-260 mm). Altogether, there are about 11 species, with 8 of them occurring in the western Atlan- tic (Randall et al. 1971). Atlantic members of the genus have been reviewed by Courtenay (1967).

Aboussouan (1972) illustrated and briefly described a single 9.0-mm larva as Rypticus saponaceus and a 4.5- mm specimen as "Rypticus (?)." Courtenay (1967) il- lustrated a 15.5-mm juvenile of Rypticus saponaceus.

Only one larval specimen with meristic characters of Rypticus was found in collections from the Pacific, while the genus was represented by many specimens in collec- tions from the Atlantic. Small larvae were so similar in body shape and lack of pigment to Liopropoma and Pseudogramma that it was difficult to separate them un- til some fin elements were well formed. The only pigment seen on the larvae was on the fleshy material enveloping the single elongated first dorsal spine (Fig. 39). This pigmented "flag" is more than half the body length; how- ever, it is fragile and often broken in plankton-collected material. Aboussouan (1972) seems to show rather heavy pigment on the large fanlike pectoral fin of the 9.0-mm larva he illustrated; however, I saw none on the material I examined. Besides the elongate dorsal spine and the large pectoral fin, the rays of the second dorsal and anal fins are also uniformly long and the caudal fin appears rounded. The pelvic fins are quite short in contrast to their large size in most other serranids. In most fish lar- vae with an elongate dorsal spine, the pelvic spines are comparably elongate (Ahlstrom'); Rypticus, however, has an elongate dorsal spine but a relatively undevel- oped pelvic fin. Aboussouan (1972) reported an element which resembled a spine in the anal fin; I found none in the specimens I examined. Most serranids have three anal spines, the third of which originates during larval development as a soft ray. Adult Rypticus have no anal

7E. H. Ahlstrom, National Marine Fisheries Service, Southwest Fish- eries Center, La Jolla, CA 92038, class notes, August 1971.

39

b

Figure 37. — Larvae of Liopropoma sp. from the northwestern Atlantic Ocean: a) fi.3 mm; b) 7.0 mm.

spines. If larvae of some Rypticus have an anal spine, it is apparently resorbed as the fish develops.

The body shape is similar to that of the serranines; but the nape is deeper and the bases of the second dorsal and anal fins are fleshier.

The bones of the preopercular region are similar in lar- val spination to those of Liopropoma, although speci- mens of the same size as those observed of Liopropoma

were not available for study, so comparisons may be in- valid.

Pseudogramma. — Rhegma has been synonymized with Pseudogramma (Gosline 1960) and at times con- sidered a serranid, a grammistid, or a pseudogrammid (Kendall 1977). Larval and other characters indicate that this genus is closely related to Rypticus and

40

posttemporal

opercular

supracleithrum

preopercular

Figure 38.— Spine-bearing bones of the opercular and posttemporal regions of a 12.0-mm Liopropoma sp. larva.

Liopropoma and they all were derived from epi- nephrine serranids. Pseudogramma contains one species from each of the following areas: the central Pacific, the eastern Pacific, and the western Atlantic. They have in- complete lateral lines and moderate-sized scales (Schultz 1966).

Larvae assignable to this genus by their meristic features (Table 1) have been found in samples from the western Atlantic; and since this genus is represented in

this area by a single species, it is assumed to be Pseudo- gramma gregoryi. Although there is only a small series in the present collections, it will be described in detail, since both of the other genera in this group contain sev- eral species, none of which could be identified using lar- val characters. Also Liopropoma has been dealt with previously (Kotthaus 1970).

Pseudogramma gregoryi (Fig. 40). — The proper generic name and number of species assigned to the Pseudogrammidae in the northwestern Atlantic have been sources of confusion. Gosline (1960) and Schultz (1966) listed three species, one in Pseudogramma and two in Rhegma. At present, Pseudogramma gregoryi is the only recognized pseudogrammid species in this area, all others being synonymized with it. However, the reported range of meristic characters among species syn- onymized with P. gregoryi indicates that there may be more than one valid species.

Meristic element development (Table 8) — A series of 18 larvae from 3.3 to 10.3 mm was cleared and stained. At 3.3 mm, the notochord is straight but 12 vertebrae (10 precaudal and 2 caudal), 4 branchiostegal rays, 15 pec- toral rays, 1 dorsal spine, and 2 gill rakers are already ossified. The notochord undergoes flexion from 3.6 to 4.2 mm and during this time the vertebrae increase to 19, the branchiostegal rays to 6, the pectoral rays to 16, the dor- sal spines to 2, and the gill rakers to 4. The caudal fin begins formation and has six dorsal lobe and six ventral lobe primary rays. The dorsal fin spines which become elongate during larval development are the second and third. Most of the meristic complements are complete in

Figure 39.— Larva of Rypticus sp. from the northwestern Atlantic Ocean, (i.H mm.

41

a

b

d

Figure 40.— Larvae of Pseudogramma gregoryi from the northwestern Atlantic Ocean: a) 4.7 mm; b) 6.1 mm; c) 6.1 mm; d) 10.2

mm.

42

Table 8. — Meristic character development of larvae of Pseudogramma gregoryi. Specimens between dashed lines are undergoing notochord flexion. For the caudal fin,"P" and "S" indicate primary and secondary fin rays.

Body Branchio- length stegal (mm) rays	Verte- brae	Pec- toral fin	Pel- vic fin	Caudal fin Dorsal Ventral S P P S	Anal fin	Dorsal fin Spines Rays	Gill rakers
3.3 NL 4	10+2	15				I	2

3.6

4.1 4.1 4.2

10+8 15

10+8 16

10+9 16

4 6 6 6 6 6

4.2 SL 4.3 4.4 4.5 4.6 4.7 5.3 5.6 5.9 6.7 7.8 9.6 10.3

10 + 14 10 + 11 10 + 16 10 + 12 10+14 10 + 16 10 + 16 10+16 10 + 16 10 + 16 10+16 10+16 10+16

16 15 15 15 15 15 15 15 15 15 15 15 15

1,3 3

1,4 1,4 1,5 1,5 [,5 1,5

1,13 9

1,16 10

1,16 111,15 111,16 111,15 111,15 111,15 III, 15 111,15 111,15

IV

II

VIII

II

VIII VIII VIII VIII VIII VIII VIII VIII VIII

17 18

18

19 19

18

18 18 19 19

1 + 7 1 + 7 1 + 7 1 + 5

8

7

8 3 + 7

3 + 7 3+8

4 + 8 3+9 3+10

5-mm specimens (branchiostegal rays, vertebrae, pec- toral rays, anal fin rays, dorsal fin spines and rays, and primary caudal fin rays). The pelvic fin is notably late in formation, relative to the other fins, and does not attain its adult complement of rays until 6.7 mm, after all other elements have reached their adult numbers. Thus, the sequence of meristic character development in Pseudo- gramma gregoryi is quite different from that of other serranids in that the pectoral fin forms before other fins and the pelvic fin forms last. The third anal spine develops as a spine rather than as a ray that transforms into a spine as it does in other serranids.

Body proportions (Table 9) — A series of larvae from 3.9 to 11.4 mm was measured to trace changes in body pro- portions during larval development. Due to the limited number of specimens available, some partly damaged larvae had to be used for these measurements; therefore, some measurements may not represent actual body shape. As the notochord flexes, total length increases from about 110% to about 125% of BL, caudal peduncle depth increases from 6-10% to about 14% of BL, and the preanal length increases from about 45% to about 55% of BL. The rest of the proportions remain rather constant over the size range at hand. The body depth is about 25% of BL (23.2-31.7%) and the depth at the anus is only slightly less (22.5-29.5%). The measurements associated with the head are: eye, 8.3 to 11.6% of BL; snout, 4.7 to 10.0% of BL; and head length, 26.6 to 37.8% of BL. Pseu- dogramma gregoryi then, is shallower bodied and has a shorter head and preanal length than other serranids.

Pigmentation — Larvae of Pseudogramma gregoryi are notable, as are those of Liopropoma and Rypticus, for their lack of pigment. The only body pigment of P. gregoryi during larval development occurs on the dorso- lateral surface of the gut cavity. Some pigment develops on the membrane associated with the elongated second spine of the dorsal fin and also on the tips of the pectoral

fin rays. Since many specimens were bleached, had broken dorsal spines or were in otherwise poor condition, it was not possible to determine exactly when this pig- ment first appeared. It was only present in larger larvae among those examined.

Spines in the opercular region (Fig. 41) — Several bones in the opercular region develop unserrated spines during the larval period. Their character was observed in a 10.3- mm larva. The spines are more pointed in P. gregoryi than in Liopropoma. The preopercular has five subequal spines on its posterior margin. In contrast to other ser- ranids where the spine at the preopercular angle is longest, the spine dorsal to the one at the angle is longest in P. gregoryi. The interopercular has a sharp spine on its posterior edge, as does the subopercular. The suboper- cular also has a strengthening ridge that lies under the opercular and extends posterodorsally near the opercu- lar margin. The opercular has three posteriorly directed spines that are pointed and closer together than those of other genera examined. The posttemporal and supra- cleithrum lack noticeable spines protruding from the lar- val surface.

Discussion. — Larvae of the three genera described here (Pseudogramma gregoryi, Liopropoma, and Ryp- ticus) share characters which unite them with each other and separate them from other groups of serranid larvae.

1. They are practically devoid of pigment at all sizes. Some Liopropoma larvae have one to three melano- phores on the caudal peduncle and the fleshy parts of the elongated dorsal spines are pigmented. Nevertheless, these genera have markedly less pigment than other serranid genera I examined.

2. They all possess one or two elongate, thin, dorsal spines with fleshy sheaths. These are broken in many

43

Table 9. — Body proportions of larvae of Pseudogramma gregoryi. Specimens between dashed lines are undergoing notochord flexion.

			Percent of standard length
Body						Pre-	Caudal	Depth
length	Total	Body	Eye	Head	Snout	anal	peduncle	at
(mm)	length	depth	length	length	length	length	depth	anus
3.9NL	111	27.4	9.7	30.6	6.5	37.1	6.5	24.2
4.1	106	29.2	9.2	29.2	7.7	46.2	9.2	26.2
4.3	110	23.2	10.1	30.4	8.7	44.9	10.1	23.2
4.3	110	25.0	10.2	30.9	8.8	44.1	8.8	25.0
4.3	112	26.1	11.6	30.4	8.7	46.4	10.1	26.1
4.6	112	25.7	9.5	31.1	—	50.0	9.5	23.0
4.5 SL	120	26.8	11.3	32.4	7.0	52.2	14.1	26.8
4.5	118	23.9	9.9	29.6	9.9	46.5	11.3	22.5
4.8	113	27.6	10.5	27.5	6.6	42.1	13.2	26.3
4.8	116	27.6	9.2	27.6	—	46.0	13.2	26.3
4.8	118	26.3	9.2	30.3	9.2	50.0	14.5	23.7
5.0	118	26.3	10.0	31.3	10.0	47.5	12.5	23.8
5.2	117	31.7	9.8	37.8	9.8	59.8	11.0	24.4
5.3	112	27.4	10.7	31.0	7.1	50.0	16.7	26.2
5.7	122	26.4	9.9	29.7	7.7	50.5	15.4	25.3
5.9	119	26.6	8.5	26.6	5.3	46.8	13.8	23.4
6.0	121	26.0	10.4	34.4	8.3	45.8	14.6	25.0
6.1	120	26.8	9.3	29.9	8.2	48.5	15.5	25.8
6.0	119	25.0	8.3	28.1	6.3	44.8	12.5	24.0
7.2	121	26.7	9.5	29.5	6.7	50.5	16.2	27.6
8.0	118	25.2	9.4	29.9	4.7	50.4	14.2	27.6
9.7	125	26.0	8.4	31.8	9.1	53.2	13.0	27.3
9.8	125	6.3	9.0	28.8	7.1	53.8	13.5	28.2
10.4	122	28.3	9.0	30.7	6.6	56.0	14.5	29.2
11.4	124	25.4	8.8	32.0	7.7	53.6	14.4	26.5

posttemporal

supracleithrum

opercula

preopercula

subopercular

interopercular

Figure 41. — Spine-bearing bones of the opercular and posttemporal regions on a lll..'1-nim Pseudogramma gregoryi larva.

field-collected specimens; but the bases of these elon- gate spines are larger than those of other spines and this is useful in recognizing these larvae.

3. The pectoral fins develop rays earlier and are larger than in other serranids. In other serranid genera the pec-

toral fins form last; however, in these genera (except Lio- propoma) they form while only a few dorsal spines and caudal fin rays are developing.

4. The body shape is little elevated and the caudal pe- duncle is deep, producing a tubular-shaped body similar to that seen in some labrid and scarid larvae. Serranid larvae in the subfamilies Anthiinae and Epinephelinae are deeper bodied than grammistines. Only larvae of serranines have the depressed condition of grammistines.

These three genera have been considered members of separate families or subfamilies; but on the basis of lar- val morphology and other characters (Kendall 1976), they appear more closely related to each other. Several other genera have been aligned with those described here but their larvae are unknown, so an accurate assessment of the relationships within this group based on larval de- velopment is not possible at present. However, Hubbs and Chu (1934) illustrated late larvae of Diploprion bi- fasciatum, a genus considered a grammistid by Randall et al. (1971). Their specimens had greatly elongated flex- ible second and third dorsal spines. Jeboehlkia gladifer, described from a small specimen by Robins (1967), has an elongated but stiff first dorsal fin spine and is prob- ably a small specimen of a species closely related to those described here. Fourmanoir (1976) illustrated and briefly described larvae of Grammistes sexilineatus and Aporops bilinearis, both with an elongate dorsal spine and preopercular spines similar to the grammistines de- scribed here. The unique first pterygiophore of gram- mistine and related fishes (Kendall 1976) may help sup- port the elongate larval dorsal spines, as the enlarged

41

predorsal bone supports the vexillum of larval carapids (Strasburg 1965). Possibly all fishes belonging in the Grammistinae have one or two elongated dorsal spines as larvae or have lost them secondarily. As larvae of more members of this group are described, relationships be- tween them may become clarified.

DISCUSSION

This study of larvae of most American serranid genera permits an examination of the phylogeny of the group on the basis of larval characters. Among the percichthyids, only the larvae of Morone have been adequately de- scribed (Mansueti 1958, 1964). Since they develop in es- tuarine areas whereas serranids develop at sea, their morphology may show adaptations not present in ser- ranids. Nevertheless, the larvae of Morone appear generalized and it appears that unspecialized serranines could have been derived from them. Some of the larval characters of Morone considered trenchant in this respect are their late-developing spinous dorsal and pelvic fins, pigment consisting of evenly sized melano- phores scattered over the body, and the lack of heavy ar- mature in the opercular area.

Serraninae larvae demonstrate various degrees of specialization from the Morone pattern. There seems to be a trend among the genera toward a deeper body, earlier development of the spinous dorsal and pelvic fins, larger spines on bones in the opercular region, and pig- ment concentrated in blotches mainly along the ventral midline. Hypoplectrus and a type here designated Diplectrum Type 2 larvae do not show these features, so their taxonomic position based on other characters should be reevaluated.

Among the other serranids there seems two major lines of divergence from the serranines. These are the anthiines and the epinepheline-grammistines (Kendall 1976).

Four genera of anthiine larvae (Plectranthias, Anthias, Pronotogrammus, and Hemanthias) were recognized and among them there appears to be a trend toward in- creased development of strong spines in the opercular region and on the head. Also fin spines in the dorsal and pelvic fins become strong. The strong ctenoid scales develop precociously. Serrations develop on the spines in the opercular region and on the thick fin spines. Among the anthiine genera and among the species within the genera these features are variously developed. These features probably afford protection against predation and thus make possible a longer planktonic larval period to increase the dispersion of these fish.

The American epinephelines contain three genera. One of these genera, Epinephelus, has several subgenera. Although there are considerable differences in size and body shape of adults of members of this subfamily, indi- cating different ecological requirements, the larvae are similar. Several species of epinepheline larvae have been described from other parts of the world and these are similar to American larvae. In fact, no characters were found to separate larval epinephelines to genera. Epi-

nepheline larvae possess a suite of specialized characters that apparently enable them to have a long pelagic ex- istence. They have extremely elongate and serrate sec- ond dorsal and pelvic spines. The preopercular also has a long serrate spine at its angle and there is a serrate supra- orbital crest. Pigment is generally confined to the ventral area of the caudal peduncle, the hindbrain, the lateral surface of the gut cavity, and the membranes of the elon- gate spines. Gonioplectrus larvae resemble epinepheline larvae in most respects, differences being in "degree" rather than "kind" (Kendall and Fahay in press). Whether Gonioplectrus should be considered an epi- nepheline awaits further analysis of adult morphology. Since there are no other larvae with only some of the epi- nepheline suite of characters or degrees of development of the characters, it is not possible to relate these fish to any others on the basis of their larvae. However, other evidence indicates that the epinephelines arose from serranine-type fishes (Smith 1971).

Larvae representing three genera in the grammistine line were found. No consensus on the relationship of these three genera is available; but their predorsal bone patterns (Kendall 1976) as well as larval similarity indi- cate that they are part of a single lineage. The most out- standing feature of these larvae is the development of one or two elongate, flexible, dorsal spines. Each spine has a variously pigmented membranous sheath around it. Otherwise, the larvae are practically devoid of pigment. In several features of development, these fish are similar to serranines. Their body shape is little elevated and there are no strong spines associated with the head or fins. The sequence of fin development is similar to that in serranines, except the pectoral fin develops early in Pseu- dogramma. In summary, it appears that these fishes could have developed from fishes with serranine-like lar- vae and that epinepheline larval specializations occurred in that lineage after the grammistine line diverged from it. Based on predorsal bone patterns, Pogonoperca is intermediate between serranines and epinepheline- grammistines (Kendall 1976) so its larvae may show par- tial development of epinepheline-grammistine characters.

SUMMARY

There are four distinct groups among the American serranids based on the morphology of their larvae. These groups represent the subfamilies Serraninae, Anthiinae, and Epinephelinae and the genera Liopropoma, Ryp- ticus, and Pseudogramma, here grouped as the subfami- ly Grammistinae. Within the serranines and anthiines, larvae of several genera were found and a progression of morphological characters which seem to make the larvae better fit for a longer planktonic existence was seen.

The four groups of American serranid larvae can be partially characterized as follows (Figs. 42, 43):

Serraninae — Body proportions show rather direct de- velopment. There are no elongate spines in the opercu- lar region, rather a series of blunt points. The fin spines are thin and only slightly elongated in some. Most larval

45

c

Figure 42. — Larvae representative of four groups of American Serranidae: a) Serraninae (Paralabrax sp., 6.0 mm); b) Anthiinae (Anthias tenuis, 6.7 mm); c) Epinephelinae (Mycteroperca mierolepis, 7.4 mm); d) Grammistinae (Rypticus sp., 6.6 mm).

46

a

Figure 43. — Interoperculars of larvae of representatives of four groups of American Serranidae: a) Serraninae (Centropristis striata, 10.6 mm); b) Anthiinae (Hemanthias vivanus, 10.3 mm); c) Epinephelinae (Epinephelus niveatus, 10.2 mm); d) Grammistinae (Pseudogramma gregoryi, 10.3 mm).

pigment consists of melanophores in characteristic posi- tions along the ventral midline.

Anthiinae — These are deep-bodied larvae with pro- duced spines on several bones in the opercular region, some of which are serrated. The pelvic and some dorsal fin spines are strong, and serrate in some, but not very elongate. Pigment consists mainly of large blotches and dashes in characteristic positions on the trunk.

Epinephelinae — The body is roughly "kite" shaped, deepest at insertion of second spine of the dorsal fin and pelvic fin spine. Elongate serrate spines are present on the preopercular, but there are no serrations on other bones in the opercular series. The second dorsal fin spine and pelvic fin spine are greatly elongate and heavily ser- rate. Pigment mainly consists of a large blotch on the caudal peduncle, heavy pigment over the body cavity, and some pigment on the membranes associated with the produced fin spines.

Grammistinae — The body is roughly tubular with a deep caudal peduncle. Bones in the opercular series are armed with variously elongated, simple spines. One or two dorsal fin spines become quite elongate, but are thin and flexible with pigmented membranous sheaths around them. Bodies of the larvae are practically devoid of pigment throughout development.

ACKNOWLEDGMENTS

This work was a thesis project at Scripps Institution of Oceanography. Elbert H. Ahlstrom and Richard H. Rosenblatt provided much invaluable advice and guidance during this study.

I thank the following people for their help in obtaining specimens in collections under their control and for

engaging in useful discussions about this work: Phillip C. Heemstra, CSIRO, Cronulla, N.S.W., Australia; C. Lavett Smith, American Museum of Natural History; Leslie Knapp, Smithsonian Oceanographic Sorting Center; and William J. Richards, National Marine Fisheries Service, Miami, Fla., who also reviewed the manuscript. L. A. Walford, New Jersey Marine Science Consortium, and H. Geoffrey Moser, National Marine Fisheries Service, La Jolla, gave much encouragement and helpful advice.

LITERATURE CITED

larves de teleosteens de l'ouest african. XI. Larves Bull. Inst. Fondam. Afr. Noire, Ser. A Sci. Nat.

the genus Rypticus (Gram- Sci. Phila. 119:241-293.

Oceanogr. Exped. 1908-1910.

ABOUSSOUAN, A.

1972. Oeufs et serraniformes 34:485-502.

AHLSTROM, E. H., J. L. BUTLER, and B. Y. SUMIDA.

1976. Pelagic stromateoid fishes (Pisces, Perciformes) of the East- ern Pacific: kinds, distributions, and early life histories and observations on five of these from the Northwest Atlantic. Bull. Mar. Sci. 26:285-402.

BARLOW, G. W.

1975. On the sociobiology of some hermaphroditic serranid fishes, the hamlets, in Puerto Rico. Mar. Biol. (Berl.) 33:295-300.

BERTOLINI, F.

1933. Uova, larvae e stadi giovanili di teleostei: familia 2: Serran- idae. Fauna e flora del Golfo di Napoli, Monogr. 38:310-331. BORTONE, S. A.

1977. Revision of the sea basses of the genus Diplectrum (Pisces: Serranidae). U.S. Dep. Commer., NOAA Tech. Rep. NMFS Circ. 404, 49 p.

CLOTHIER, C. R.

1950. A key to some southern California fishes based on vertebral characters. Calif. Dep. Fish Game, Fish Bull. 79, 83 p. COURTENAY, W. R., JR.

1967. Atlantic fishes of mistidae). Proc. Acad. Nat. FACE, L.

1918. Shore-fishes. Rep. Dan. Medit. Adj. Seas 11(3), 154 p. FOURMANOIR, P.

1971. Notes ichthyologiques (III). Cah. O.R.S.T.O.M., Ser. Oceanogr. 9:267-278.

1976. Formes post-larvaires et juveniles de poissons cotiers pris au chalut pelagique dans le sudouest Pacifique. Cah. Pac. 19:47-88.

FOWLER, H. W.

1944. The fishes. In Results of the fifth George Vanderbilt expe- dition (1941). Monogr. Acad. Nat. Sci. Phila. 6:57-583. FREIHOFER, W. C.

1963. Patterns of the ramus lateralis accessorius and their systematic significance in teleostean fishes. Stanford Ichthyol. Bull. 8:80-189. GINSBURG. I.

1952. Eight new fishes from the Gulf coast of the United States, with two new genera and notes on geographic distribution. J. Wash. Acad. Sci. 42:84-101. GOSLINE, W. A.

1960. A new Hawaiian percoid fish, Suttoma lineata, with a discus- sion of its relationships and a definition of the family Gram- mistidae. Pac. Sci. 14:28-38. 1966. The limits of the fish family Serranidae, with notes on other lower percoids. Proc. Calif. Acad. Sci. 33:91-111. GUITART MANDAY, D., and D. JUAREZ FERNANDEZ.

1966. Desarrollo embriouario y primeros estadios larvales de cherna criolla, Epinephelus striatus (Bloch) (Perciformes: Serran- idae). Acad. Ciencias de Cuba. Estudios-Inst. de Occanol. 1:35- 45. HASTINGS, R. W.

1973. Biology of the pygmy sea bass, Serraniculus pumilw (Pisces: Serranidae). Fish. Bull., U.S. 71:235-242.

47

HEEMSTRA. P. C.

1974. On the identity of certain eastern Pacific and Caribbean post-larval fishes (Perciformes) described by Henry Fowler. Proc. Acad. Nat. Sci. Phila. 126:21-26. HOFF. F. H.. JR.

1970. Artificial spawning of black sea bass, Centropristes striatus melanus Ginsburg, aided by chorionic gonadotrophic hor- mones. Fla. Dep. Nat. Resour. Mar. Res. Lab., Spec. Sci. Rep. 25, 17 p. HOLLISTER, G.

1934. Clearing and dyeing fish for bone study. Zoologica (N.Y.) 12:89-101. HUBBS, C. L., and Y. T. CHU.

1934. Asiatic fishes (Diplopnon and Laeops) having a greatly elon- gated dorsal ray in very large post-larvae. Occas. Pap. Mus. Zool., Univ. Mich. 299:1-7. JORDAN. D. S., and C. H. EIGENMANN.

1890. A review of the genera and species of Serranidae found in the water of America and Europe. Bull. U.S. Fish. Comm. 8:329-441. RATA YAM A, M.

1960. Fauna Japonica Serranidae. (Pisces). Tokyo News Serv. Ltd.. Tokyo, 189 p.

KENDALL, A. W., JR.

1972. Description of black sea bass, Centropristis striata (Lin- naeus), larvae and their occurrences north of Cape Lookout, North Carolina, in 1966. Fish. Bull., U.S. 70:1243-1260.

1976. Predorsal and associated bones in serranid and grammistid fishes. Bull. Mar. Sci. 26:585-592.

1977. Relationships among American serranid fishes based on the morphology of their larvae. Ph.D. Thesis, Univ. Calif. San Diego, 208 p.

KENDALL, A. W., JR., and M. P. FAHAY.

In press. Larva of the serranid fish Gonioplectrus hispanus with comments on its relationships. Bull. Mar. Sci. KOTTHAUS, A.

1970. Flagelloserranus a new genus of serranid fishes with the de- scription of two new species (Pisces, Percomorphi). Dana Rep. 78, 31 p. MANSUETI. R. J.

1958. Eggs, larvae, and young of the striped bass, Roccus sax- atilus. Md. Dep. Res. Educ. Contrib. 112:1-36.

1964. Eggs, larvae, and young of the white perch, Roccus americanus, with comments on its ecology in the estu- ary. Chesapeake Sci. 5:3-45. McCULLY, H. H.

1961. The comparative anatomy of the scales of the serranid fishes. Ph.D. Thesis, Stanford Univ., Palo Alto, Calif., 258 p.

MEEK, S. E., and S. F. HILDEBRAND.

1925. The marine fishes of Panama, Part II. Field Mus. Nat. Hist., Publ. No. 226, Zool. Ser. 15:331-707. MERRIMAN, D., and R. L. SCLAR.

1952. The pelagic fish eggs and larvae of Block Island Sound. In Hydrographic and biological studies of Block Island Sound, 13(31:165-219. Bull. Bingham Oceanogr. Collect., Yale Univ. MILLER, R. J.

1959. A review of the seabasses of the genus Centropristes (Serran- idae). Tulane Stud. Zool. 7:35-68.

MITO, S., M. UKAWA, and M. HIGUCHI.

1967. On the larval and young stages of a serranid fish, Epi- nephelus akaara (Temminck et Schlegel). Bull. Naiki Reg. Fish. Res. Lab. 25:337-347. PEARSON, J. C.

1941. The young of some marine fishes taken in lower Chesapeake Bay, Virginia, with special reference to the gray sea trout Cynos- cwn regalis (Bloch). U.S. Fish Wildl. Serv., Fish. Bull. 50:79- 102. PRESLEY, R. F.

1970. Larval snowy grouper, Epinephelus niveatus (Valenciennes, 1828), from the Florida Straits. Fla. Dep. Nat. Resour., Leafl. Ser. 4(18), 6 p.

RAFFAELE, F.

1888. Le uove galleggianti e le larve dei Teleostei nel golfo di Napole. Mill. Zool. St. Neapel. 8, 84 p. RANDALL, J. E.

1968. Caribbean reef fishes. Trop. Fish Hobbyist Publ., N.J., 318 P-

RANDALL, J. E., K. AIDA, T. HIBIYA, N. MITSUURA, H. KAMIYA, and Y. HASHIMOTO. 1971. Grammistin, the skin toxin of soapfishes, and its sig- nificance in the classification of the Grammistidae. Publ. Seto Mar. Biol. Lab. 19:157-190. ROBINS, C. R.

1967. The status of the serranid fish Liopropoma aberrans with the description of a new, apparently related genus, copeia 1967;591- 595. ROBINS, C. R„ and W. A. STARCK, II.

1961. Materials for a revision of Serranus and related fish genera. Proc. Acad. Nat. Sci. Phila. 113:259-314. ROSENBLATT, R. H„ and B. J. ZAHURANEC.

1967. The Eastern Pacific groupers of the genus Mycteroperca, in- cluding a new species. Calif. Fish Game 53:228-245. ROULE, L., and F. ANGEL.

1930. Larves et alevins de poissons provenant des croisieres du Prince Albert 1" de Monaco. Resultats des Campagnes Scientif- iques Accomplies sur son Yacht, Fasc. 79, 148 p.

SCHULTZ, L. P.

1966. Pseudorhegma diagramma, a new genus and species of grammistid fish, with a key to genera of the family and to the species of the subfamily Pseudogramminae. Ichthyol. Aquarium J. 37:185-194. SMITH, C. L.

1965. The patterns of sexuality and the classification of serranid fishes. Am. Mus. Novit. 2207, 20 p.

1971. A revision of the American groupers: Epinephelus and allied genera. Bull. Am. Mus. Nat. Hist. 146:69-241. SMITH. C. L., and E. H. ATZ.

1969. The sexual mechanism of the reef bass Pseudogramma ber- mudensis and its implications in the classification of the Pseudo- grammidae (Pisces: Perciformes). Z. Morphol. Tiere 65:315-326.

SMITH, C. L., and P. H. YOUNG.

1966. Gonad structure and the reproductive cycle of the kelp bass. Paralabrax clathratus (Girard), with comments on the relation- ships of the serranid genus Paralabrax. Calif. Fish Game 52:283- 292.

SPARTA, A.

1932. Contributo alia conoscenza dello sviluppao post-embrionale

in Athias sacer Bloch. Rev. Comit. Talassogr. 195, 11 p. 1935. Contributo alia conoscenze dello sviluppo nei Percidi. Uova ovariche mature di Epinephelus guaza L. e stadi post-embrionali e larvali di Epinephelus alexandrinus Cuv. e Val. Rev. Comit. Talassogr. 224, 12 p. STRASBURG, D. W.

1965. Description of the larva and familial relationships of the fish Snyderidia canina. Copeia 1965:20-24.

THRESHER, R. E.

1978. Polymorphism, mimicry, and the evolution of the hamlets (Hypoplectrus. Serranidae). Bull. Mar. Sci. 28:345-353. UKAWA, M., M. HIGUCHI, and S. MITO.

1966. Spawning habits and early life history of a serranid fish, Epi- nephelus akaara (Temminck et Schlegel). Jpn. J. Ichthyol. 13:156-161.

VODJANITZKI, W. A., and I. Z. KAZANOVA.

1954. Opredelitel pelagicheskikh ikrinok i likinok rub Chernogo Morya [Key for identification of the pelagic eggs and fish larvae in the Black SeaJ. Tr. Vses. Nauchno-Issled. Inst. Morsk. Rybn. Khoz. Okeanogr. 28:240-323. WILSON, H. V.

1891. The embryology of the sea bass (Serranus atrarius). Bull. U.S. Fish. Comm. 9:209-277.

48

Appendix Table 1. — Data associated with larval fish used for illustrations. Institution abbreviations: NE: NMFS, Northeast Fisheries Center; SW: NMFS, Southwest Fisheries Center; SE: NMFS, Southeast Fisheries Center; UNC: University of North Carolina.

Figure

Species

Length (mm)

Insti- tution

Cruise

Station

Lat. (N)

Long. Date

(W) Day/Mo/Yr

la b

Centropristis striata Centropristis striata

Centropristis striata

4.7	NE	D-67-4	FF-5(D	31°38'	79°56'	12	5 67
8.3	NE	D-71-5	N-2p	34°56'	75°55'	29	4 71
0.6	NE	D-67-4	CC-5U)	32°20'	78°25'	14	5 67

3a b

Faralabrax sp. Paralabrax sp.

6.0

9.0

SW

6509B S-5509

118.39 28°19' 115°24' 17 10 65

(CalCOFI sample — no precise data)

4 Paralabrax sp. 9.5 SW C-51

5a Diplectrum sp. Type 1 (Atl.)

b Diplectrum sp. Type 1 (Atl.)

6a Diplectrum sp. Type 1 (Pac.)

b Deplectrum sp. Type 1 (Pac.)

c Diplectrum sp. Type 1 (Pac.)

d Diplectrum sp. Type 1 (Pac.)

7 Diplectrum sp. Type 1 (Atl.) 10.3 NE D-67-4

8a Diplectrum sp. Type 2

b Diplectrum sp. Type 2

9 Diplectrum sp. Type 2 9.9 SW ETP

5.8	N E	D-67-4
10.0	NE	D-67-16
4.3	SW	5706S
5.0	SW	5706S
6.1	SW	5706S
13.0	SW	5706S

130.30	26°21'	113°49'	10	8	53
GG-4U)	31<W	80°51'	U	5	67
HH-5(2)	30°22'	80°48'	24	10	67
107G02	29°42'	114°07'	20	6	57
107G02	29°42'	114°07'	20	6	57
107G02	29°42'	114°07'	20	6	57
143G100	29°19'	109°14'	11	6	57

CC-5(1)

33°20'

78°25' 14 5 67

4.5	SW	P-5412	157.10	22°35'	109°19'	4	7 54
8.4	SW	5612-H	160G30	22°39'	107°49'	3	7 56

47.244

6°59'

1°54' 27 8 67

10a b

11

Serranus sp. Serranus sp. Serranus sp. Serranus sp.

Serranus sp.

3.7	NE	D-71-5	M-6	35°07'	74°55'	5	5	71
.VI)	NE	1	4	35°03'	75°10'	11	4	72
5.5	NE	D-66-3	M-5(2)	35°11'	75°07'	20	4	66
9.4	NE	D-68-1	NN-2(2)	27°11'	80°03'	4	2	68
9.7	NE	D-66-12	L-4(l)	35°46'	75°05'	30	9	66

12a b

Hypoplectrus sp. Hypoplectrus sp. Hypoplectrus sp. Hypoplectrus sp.

3.4	SE
4.7	SE
5.7	SE
8.5	SE

Reared at NMFS, Miami, Fla. from eggs collected nearby Reared at NMFS, Miami, Fla. from eggs collected nearby Reared at NMFS, Miami, Fla. from eggs collected nearby Reared at NMFS, Miami, Fla. from eggs collected nearby

13 Hypoplectrus sp.

14a Serraniculus pumilio

b Serraniculus pumilio

c Serraniculus pumilio

15a Plectranthias garupellus

b Plectranthias garupellus

16a Anthias sp. Type 1

b Anthias sp. Type 1

17a Anthias sp. Type 2

b Anthias sp. Type 2

c Anthias sp. Type 2

18 Anthias tenuis

19 Anthias sp. Type 3

20a Anthias gordensis

b Anthias gordensis

21 Anthias sp. Type 1

12.5

SE

Reared at NMFS, Miami, Fla. from eggs collected nearby

3.8	NE	D-67-16	DD-4U)	32°55'	78°51'	21	Hi	67
5.8	NE	D-67-16	DD-5(2)	32°47'	78°44'	21	10	67
55.3	NE	D-71-3	SA-2	29°51'	81°10'	3	9	71
5.5	NE	D-67-16	NN-4(1)	27°12'	79°51'	29	10	67
7.0	NE	D-68-1	PP-2U)	26°47'	79°56'	4	2	HH
3.8	NE	D-66-5	N-4(2)	34°42'	75°48'	24	5	66
5.3	NE	D-71-5	BB-7	32°53'	77°26'	1	5	71
4.7	NE	D-68-1	DD-6(1)	32°36'	78°34'	29	1	68
5.7	NE	D-71-5	M-4	35°13'	75°12'	5	5	71
8.4	NE	D-68-1	EE-6	32°08'	79°10'	30	1	68
6.7	SE	BLM	IV-2	26°10'	96°39'	30	5	76
5.1	NE	D-71-5	BB-8	32°39'	77°17'	1	5	71
5.2	SW	5706S	109G55	29°52'	113°04'	18	6	57
6.0	SW	5706S	115G40	28°52'	112°44'	17	6	57
10.2	NE	D-68-1	EE-6	32°08'	79°10'	30	1	68

49

Appendix Table 1.— Continued.

Figure

Species

Length (mm)

Insti- tution

Cruise

Station

Lat.

(N)

Long.

(W)

Date

Day/Mo/Yr

22a Pronotogrammus aureorubens

b Pronotogrammus aureorubens

c Pronotogrammus aureorubens

23a Pronotogrammus eos

b Pronotogrammus eos

24 Pronotogrammus aureorubens

4.6	NE	D-68-1	PP-3(1)	26°47'	79°50'	4	2	68
6.0	NE	D-68-1	CC-7(2)	32°54'	78°07'	29	1	68
9.0	NE	D-68-1	AA-7	33°37'	76°47'	27	1	68
3.6	SW	5706S	133G40	26°16'	110°50'	12	6	57
7.8	SW	OP	052	00°04'	84°58'	19	11	67

9.5

NE

D-68-1

HH-7

30°19'

313'

1 2

25a b	Hemanthias viuanus Hemanthias uiuanus
c	Hemanthias uiuanus
26	Hemanthias peruanus
27	Hemanthias uiuanus
28	Hemanthias uiuanus
29a b c	Hemanthias vwanus Hemanthias uiuanus Hemanthias uiuanus
30	Hemanthias uiuanus
31	Paranthias furcifer
32	Epinephelus striatus
33	Epinephelus sp.
34	Epinephelus niueatus
35a b c d	Mycteroperca microlepis Mycteroperca microlepis Mycteroperca microlepis Mycteroperca microlepis

Mi

Mycteroperca microlepis

37a b	Liopropoma sp. Liopropoma sp.
38	Liopropoma sp.
39	Rypticus sp.
40a b c d	Pseudogramma gregoryi Pseudogramma gregoryi Pseudogramma gregoryi Pseudogramma gregoryi

41 Pseudogramma gregoryi

42a Paralabrax sp.

b Anthias tenuis

c Mycteroperca microlepis

d Rypticus sp.

43a Centropristis striata

b Hemanthias uiuanus

c Epinephelus niueatus

d Pseudogramma gregoryi

4.2	NE	D-71-5	LK-5	35°59'	74°33'	5	5	71
5.3	NE	D-71-5	BB-8	32°39'	77°17'	1	5	71
6.8	NE	D-71-5	BB-8	32°39'	77°17'	1	5	71
9.3	SW	6611J	147.30	23°36'	111°42'	13	11	66
9.3	NE	D-71-5	N-7	34°13'	75°35'	29	4	71
10.3	NE	D-68-1	HH-7	30°19'	80°13'	1	2	68
4.3	NE	D-71-5	N-7	34°13'	75°35'	29	4	71
5.8	NE	D-71-5	N-6	34°23'	75°39'	29	4	71
6.6	NE	D-71-5	N-7	34°13'	75°35'	29	4	71
7.2	SE	BLM-2	1-3 Day	27°34'	96°07'	10	4	75
8.6	SW	5612	157G10	22°30'	10915'	3	12	56
7.6	NE	D-67-4	GG-7(2)	30°54'	80°00'	12	5	67
8.4	SW	OP	162	11°10'	92°00'	26	11	67
10.2	NE	D-67-16	GG-7	30°54'	80°00'	23	10	67
4.0	NE	1	4	35°03'	75°10'	11	4	72
7.4	NE	3	2	35°03'	75°08'	27	4	72
14.2	UNC		T-9	New Rivei	,N.C.	26	4	72
22.6	UNC	(UNC 7934)		Bogue Sd	, N.C.	16	6	73
9.8	NE	D-71-5	AA-7	33°37'	76°47'	30	4	71
6.3	NE	D-67-4	NN-4(2)	27°12'	79°51'	7	5	67
7.0	NE	3	2	35°03'	75°08'	27	4	72
12.0	SE	OR II 702	0-2-13	26°47'	84°34'	13	9	70
6.6	NE	D-67-8	AA-4U)	34°11'	77°11'	1	8	67
4.7	NE	D-67-4	MM-4(1)	27°46'	79°56'	8	5	67
6.1	SE	OR II 7739	56	20°00'	80°13'	30	7	72
6.1	NE	D-66-3	N-5U)	34°33'	75°44'	20	4	66
10.2	NE	D-67-4	JJ-6U)	29°49'	80°14'	10	5	67
10.3	SE	OR II 7239	56	20°00'	80°13'	30	7	72
6.0	s\\	6509B	118.39	28°19'	115°24'	17	9	65
6.7	SE	BLM	IV-2	26°10'	96°39'	30	5	76
7.4	NE	3	2	35°03'	75°08'	27	4	72
6.6	NE	D-67-8	AA-4(1)	34°11'	77°11'	1	8	67
10.6	NE	D-67-4	CC-5U)	33°20'	78°25'	14	5	67
10.3	NE	D-68-1	HH-7	30°19'	80° 13'	1	2	68
10.2	NE	D-67-16	GG-7	30°54'	80°00'	23	10	67
10.3	SE	OR U 7239	56	20°00'	80°13'	30	7	72

50

NOAA TECHNICAL REPORTS NMFS CIRCULAR AND SPECIAL SCIENTIFIC REPORT GUIDELINES FOR CONTRIBUTORS

FISHERIES

CONTENTS OF MANUSCRIPT

First page. Give the title (as concise as possible) of the paper and the author's name, and footnote the author's affiliation, mailing address, and ZIP code.

Contents. Contains the text headings and abbreviated figure legends and table headings. Dots should follow each entry and page numbers should be omitted.

Abstract. Not to exceed one double-spaced page. Footnotes and literature citations do not belong in the abstract.

Text. See also Form of the Manuscript below. Follow the U.S. Government Printing Office Style Manual, 1973 edition. Fish names, follow the American Fisheries Society Special Publica- tion No. 6, A List of Common and Scientific Names of Fishes from the United States and Canada, third edition, 1970. Use short, brief, informative headings in place of "Materials and Methods."

Text footnotes. Type on a separate sheet from the text. For unpublished or some processed material, give author, year, title of manuscript, number of pages, and where it is filed — agency and its location.

Personal communications. Cite name in text and footnote. Cite in footnote: John J. Jones, Fishery Biologist, Scripps Insti- tution of Oceanography, La Jolla, CA 92037, pers. commun., 21 May 1977.

Figures. Should be self-explanatory, not requiring reference to the text. All figures should be cited consecutively in the text and their placement indicated in the left-hand margin of the manuscript. Photographs and line drawings should be of "professional" quality — clear and balanced, and can be re- duced to 6'/< inches (40 picas) for page width or to 3 '/s inches (19 picas) for single-column width, but no more than 9 inches (54 picas) high. Photos should be printed on glossy paper — sharply focussed, good contrast. Label each figure. List, and typed dou- ble spaced, each figure legend. DO NOT SEND original figures to the Scientific Editor; NMFS Scientific Publications Office will request these if they are needed.

Tables. Each table should start on a separate page and should be self-explanatory, not requiring reference to the text. Headings should be short but amply descriptive. Use only horizontal rules. Number table footnotes consecutively across the page from left to right in Arabic numerals; and to avoid con- fusion with powers, place them to the left of the numerals. If the original tables are typed in our format and are clean and leg- ible, these tables will be reproduced as they are. In the text all tables should be cited consecutively and their placement indi- cated in the left-hand margin of the manuscript.

Acknowledgments. Place at the end of text. Give credit only to those who gave exceptional contributions and not to those whose contributions are part of their normal duties.

Literature cited. In text as: Smith and Jones (1977) or (Smith and Jones 1977); if more than one author, list according to years (e.g., Smith 1936; Jones et al. 1975; Doe 1977). All papers re- ferred to in the text should be listed alphabetically by the senior author's surname under the heading "Literature Cited"; only the author's surname and initials are required in the author line. The author is responsible for the accuracy of the literature cita- tions. Abbreviations of names of periodicals and serials should conform to Biological Abstracts List of Serials with Title Abbre- viations. Format, see recent SSRF or Circular.

Abbreviations and symbols. Common ones, such as mm, m, g, ml, mg, °C (for Celsius), %, %„, etc., should be used. Abbrevi- ate units of measures only when used with numerals; periods are rarely used in these abbreviations. But periods are used in et al., vs., e.g., i.e., Wash. (WA is used only with ZIP code), etc. Abbreviations are acceptable in tables and figures where there is lack of space.

Measurements. Should be given in metric units. Other equivalent units may be given in parentheses.

FORM OF THE MANUSCRIPT

Original of the manuscript should be typed double-spaced on white bond paper. Triple space above headings. Send good duplicated copies of manuscript rather than carbon copies. The sequence of the material should be:

FIRST PAGE

CONTENTS

ABSTRACT

TEXT

LITERATURE CITED

TEXT FOOTNOTES

APPENDIX

TABLES (each table should be numbered with an Arabic numeral and heading provided)

LIST OF FIGURE LEGENDS (Entire figure legends, includ- ing "Figure" before each number)

FIGURES

ADDITIONAL INFORMATION

Send ribbon copy and two duplicated copies of the manuscript

to:

Dr. Jay C. Quast, Scientific Editor

Northwest and Alaska Fisheries Center

Auke Bay Laboratory

National Marine Fisheries Service, NOAA

P.O. Box 155

Auke Bay, AK 99821

Copies. Fifty copies will be supplied to the senior author and 100 to his organization free of charge.

PENN STATE UNIVERSITY LIBRARIES

UNITED STATES DEPARTMENT OF COMMERCE

NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION

NATIONAL MARINE FISHERIES SERVICE

SCIENTIFIC PUBLICATIONS STAFF

ROOM 450

I 107 N E 45TH ST

SEATTLE. WA 98105

A0QDD72Dn0SM

OFFICIAL BUSINESS

NOAA SCIENTIFIC AND TECHNICAL PUBLICATIONS

NOAA, the National Oceanic and Atmospheric Administration, was established as part of the Department of Commerce on October 3, 1970. The mission responsibilities of NOAA are to monitor and predict the state of the solid Earth, the oceans and their living resources, the atmosphere, and the space environment of the Earth, and to assess the socioeconomic impact of natural and technological changes in the environment.

The six Major Line Components of NOAA regularly produce various types of scientific and technical infor- mation in the following kinds of publications:

PROFESSIONAL PAPERS— Important definitive research results, major techniques, and special in- vestigations.

TECHNICAL REPORTS— Journal quality with extensive details, mathematical developments, or data listings

TECHNICAL MEMORANDUMS— Reports of preliminary, partial, or negative research or tech- nology results, interim instructions, and the like.

CONTRACT AND GRANT REPORTS— Reports

prepared by contractors or grantees under NOAA sponsorship.

TECHNICAL SERVICE PUBLICATIONS—

These are publications containing data, observations, instructions, etc. A partial listing: Data serials; Pre- diction and outlook periodicals; Technical manuals, training papers, planning reports, and information serials: and Miscellaneous technical publications.

ATLAS — Analysed data generally presented in the form of maps showing distribution of rainfall, chemical and physical conditions of oceans and at- mosphere, distribution of fishes and marine mam- mals, ionospheric conditions, etc.

Information on availability of NOAA publications can be obtained from:

ENVIRONMENTAL SCIENCE INFORMATION CENTER

ENVIRONMENTAL DATA AND INFORMATION SERVICE

NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION

U.S. DEPARTMENT OF COMMERCE

6009 Executive Boulevard Rockville, MD 20852