The effects of temperature and photoperiod on ovarian development in captive grey mullet (Mugil cephalus L)

Aquaculture, 3 (1974) 25-43 @ Elsevier Scientific Publishing

THE

EFFECTS

Company,

OF TEMPERATURE

RIAN DEVELOPMENT

CHING-MING

KUO,**

** Oceanic Institute,

Amsterdam

COLIN

IN CAPTIVE

E. NASH**

- Printed

AND

in The Netherlands

PHOTOPERIOD

GREY MULLET

ON OVA-

(Mugil cephalus L)*

and ZIAD H. SHEHADEH***

Waimanalo, Hawaii (U.S.A.)

***Department of Fisheries FAO, Rome (Italy) * Contribution (Received

No. 111 of the Oceanic

9 October,

1973;revised

Institute

29 October,

1973)

A series of sixteen environmental experiments, conducted under controlled laboratory conditions during the refractory period in the reproductive cycle of the grey mullet, determined the effect of photoperiod and temperature on the vitellogenesis of intraovarian oocytes. Fish subjected to the natural light cycle and ambient water temperatures (2426°C) served as controls. A classification of stages of vitellogenesis (I-V) is used to determine the percentage composition of oocytes for each fish at intervals throughout the experiment following sampling in vivo. Onset of vitellogenesis is timed by the environmental conditions. A retarded photoperiod, irrespective of preconditioning photoperiod, plays a dominant role in stimulating oocyte growth. Temperature regulates vitellogenesis towards functional maturity. The combination of retarded photoperiod (6L/ 18D) and constant temperature of 21°C is the most effective for the completion of vitellogenesis of oocytes to functional maturity. Regular injections of pregnant mare’s serum gonadotropin (PMSG) at 1 IU/g body weight are effective in initiating vitellogenesis.

INTRODUCTION The endocrine system of vertebrates forms the main link between the reproductive organs and environmental regulators. Fluctuating regulators such as photoperiod and temperature, mediated through the central nervous system, trigger neurosecretions which regulate, in turn, the activities

26 of the pituitary gland. As one of many target organs, the gonads are influenced accordingly, and the reproductive cycles are thus regulated intimately by the trophic hormones of the pituitary. Many experiments with fishes have observed gonad development in order to interpret the effects of certain environmental regulators on the reproductive cycle. Among the factors concerned, temperature and photoperiod are the two most important which initiate pituitary activity for fish in temperate and subtemperate regions (Hoar, 1959). However, the relative importance of each factor varies with different species of teleosts. Photoperiod has been reported as the dominant factor influencing the reproductive cycle of Enneacanthus obesus, Notropis bifrenatus and Fundulus confluentus (Harrington 1956, 1957, 1959a, 1959b); Gasterosteus aculeatus (Baggerman, 1957); Salvelinus fontinalis (Henderson, 1963) and Uryzias Zatipes (Yoshioka, 1962). Temperature has been shown by other workers to be the dominant entity for Fundulus (Matthews, 1939); Phoxinus Zaevis (Bullough, 1940); Apeltes quadracus (Merriman and Schedl, 1941); Gambusia affinis (Medlen, 195 1) and Couesius plumbeus (Ahsan, 1966). It was concluded for Salvelinus fontinalis (Henderson, 1963) that the influence of an environmental regulator varied with the stage of gonad maturation. The most responsive period to a. regulator could in fact differ between males and females of the same species, and the gametogenetic process may be independent of environmental regulators at certain stages of maturity. The most appropriate time for testing effects of environmental regulators on successive phases of the annual reproductive cycle is the refractory or post-spawning period, when the gonads undergo their maximum annual regression. The refractory period can be short, particularly for fish in temperate and subtemperate latitudes. The refractory period for Mugil cephalus in semitropical Hawaii is long, lasting from the end of March through to November. The effects of certain regulators on the reproductive cycle therefore can be interpreted accurately if the regulators are manipulated in the refractory period and spawning is subsequently induced. Experiments in fish reproduction on the interaction of environmental regulators and the endocrine system indicate that influences on the reproductive cycle are reflected more readily by changes in the ovaries rather than in the testes. Most of the work has therefore been confined to females of the species. A reliable source of female breeding stock is vitally important to aquaculture operations, particularly when those operations are linked to mass propagation of juveniles in hatcheries. Female grey mullet have been induced to spawn by hypophysation techniques (Tang, 1964; Liao, 1969;

27 Yashouv, 1969; Shehadeh and Ellis, 1970). Captive broodstocks have been maintained for long periods satisfactorily and spawning later has been induced (Shehadeh et al., 1973c). In an attempt to prolong the breeding season of the grey mullet, and eventually to supply fertile eggs to a hatchery throughout the year, experiments have been conducted on the effects of two environmental regulators on ovarian development. This paper reports the effects, of photoperiod and temperature on development, indicated by the ovarian stage composition of oocytes during vitellogenesis following controlled photoperiod and temperature manipulation. A comparison is made with fish exposed to ambient seasonal and diurnal variations as a control. In addition, observations are recorded on the use of pregnant mare’s serum gonadotropin (PMSG) to stimulate vitellogenesis.

MATERIALS

AND METHODS

All experiments and controls were performed in an environmental control laboratory (Bacon, MS 1973). Eight circular tanks (42 x 48 in. i.d.) were available in each of three rooms of the laboratory; the controlled environmental conditions included light intensity, photoperiod, water temperature and air temperature. An incandescent lamp (300 watt), suspended above the center of each holding tank, was adjusted vertically to provide an intensity of 75 ft-c at the water surface throughout all experimental periods. The lamps in each room were controlled by separate timers and dimmer units, operated automatically. Water temperature was maintained by adjusting the flow rate of either prechilled or preheated seawater. The salinity of the water throughout all the experiments was 32%0. Air temperature in each room was maintained by an air conditioner regulated by a separate thermostat. This minimized any sudden fluctuation of water temperate in the tanks. Female adults were selected from an established broodstock originally collected in Kaneohe Bay, Hawaii. These fish were then held captive for more than two years in earthen ponds lined with butyl rubber. Selection of individuals was based on a minimum fork length of 31 cm, which is known to be the size of sexually mature fish in local waters. The majority of females used were those which had been induced to spawn in the natural breeding season preceding the experimental period. No sexual dimorphism has been noted in the grey mullet. Sex was therefore determined by examining gametes sampled in vivo by the technique described by Shehadeh et al. (1973b). The natural breeding season

28 of grey mullet in Hawaii is generally between December and March. Spawning in captivity without inducement has never been recorded, although well developed oocytes can be identified in captive females during the seasonal active breeding period. It was essential to ensure that the effects of environmental regulators on intraovarian oocyte development were not reflecting the natural reproductive cycle, Therefore, all experiments were conducted in the refractory period during which.only recruitment crops of oocytes were observed. The number of fish maintained in each tank and used for each experiment was small. For identification the fish were at first tagged, either in the operculum with metal tags, or by ‘hot marking’. Neither method proved satisfactory and subsequently individuals were recognized either by size or fin incision. The fish were fed daily a prepared artificial diet (Table I). Uneaten food was removed regularly and the tanks kept clean. TABLE Formula

I for the feed presented

Component (Dry ration)

or adult fish

Component B (Liquid ration)

A

Fish meal Soybean meal Chicken starter mash Fish bone meal Dairy whey Wheat germ Wheat middlings Components

to juvenile

400 g 250 g 1oog 50 g

Water Choline chloride Urea Propylene glycol

900 0.5 0.5 5.0

ml g g g

50 g 50 g 50 g

A and J3 are mixed separately

and combined

before

use.

The oocytes were sampled in vivo - monthly at first, and more frequently as they matured during the latter part of the experimental period. Individual records of oocyte development were obtained and classified by defined stages of vitellogenesis. The sampled oocytes were preserved in Bouin’s fixative, dehydrated in alcohol and cleared in xylol. They were embedded in paraffin and later cut into sections (10 /..L)which were stained with Heidenhain’s iron-hematoxylin, counter-stained in eosin, and mounted. The study was conducted between March and November 1971, and repeated the following year. Three general procedures for the experimental work were established. These were:

29 (1) To subject fish to the natural photoperiod and ambient temperatures throughout, thus establishing the control. (2) To precondition fish to the natural photoperiod and ambient temperatures first, followed by manipulation of regulators in the laboratory. (3) To subject the fish to manipulated regulators in the laboratory throughout the test period. A photoperiod of 6 h light and 18 h darkness (6L/ 1SD) for simulating extreme conditions was used in the experiments. The operating range for water temperature was selected as 17-26°C a range including both surface and deeper coastal water temperatures in Hawaii. VITELLOGENIC

DEVELOPMENT

AND CLASSIFICATION

The histological details of the intraovarian oocytes during vitellogenesis have been described and illustrated by Kuo et al. (1974). Five development stages were named - the primary oocyte stage, yolk vesicle stage, yolk globule stage, the ripe stage, and atresia. The development of ovarian oocytes from each experimental female was recorded at intervals. The oocytes obtained at each sampling period were examined microscopically and classified into one or more of the five stages. A total of 350-500 oocytes from each sample was classified to avoid bias in assessing the stage composition. Percentage of oocytes at various stages of development was therefore calculated for each fish at frequent intervals throughout the experiment. For descriptive purposes the proliferation of primary oocytes from oogonia will be considered as the primary growth phase (stage I), and growth of primary oocytes toward yolk-laden oocytes as a secondary growth phase (stages II and III). Natural oocyte development has been observed in captive females only to the tertiary yolk globule stage. Functional maturity, therefore, refers to the late tertiary yolk globule stage (stage III) when the mean oocyte diameter is greater than 600 p. This is regarded as the proper stage at which to induce spawning artificially by the technique of hypophysation (Shehadeh et al., 1973a). EXPE-RIMENTAL

DESIGN

Effect of photoperiod and temperature Eight individual experiments were designed to determine photoperiod and temperature on ovarian development.

the effects

of

30 Trial I Accelerated photoperiod; constant temperature.of 21 +_1°C. March 15-April 5: initial 12L/12D photoperiod increased by equal increments to 14L/ 1OD April 5-April 19 : 14L/ 1OD photoperiod maintained April 19-May 27: 14L/ 10D photoperiod decreased by equal increments to 9L/15D May 27-June 3 : 9L/ 15D photoperiod maintained June 3-Nov 16: 6L/ 18D photoperiod maintained Trial 2 Accelerated photoperiod; constant temperature of 26 + 1°C. March 15-Nov 16: same photoperiod regime as in Trial 1 Trial 3 Retarded March 15-May May 27-June 3: June 3-Nov 16:

photoperiod; ambient temperature to 21 * 1°C. 27: natural photoperiod and ambient water temperature 9L/15D photoperiod maintained at 21°C 6L/18D photoperiod maintained at 21°C

Trial 4 Retarded photoperiod; ambient temperature to 26 * 1°C. March 15-Nov 16: same photoperiod regime as in Trial 3 Trial 5 Natural photoperiod; March 1.5-Nov 16: control

ambient temperature for Trials l-4

Trial 6 Retarded photoperiod; constant temperature March 15-Nov 16: 6L/ 18D photoperiod maintained

(24-26”(I). of 17 + 1°C.

Trial 7 Retarded photoperiod; ambient temperature to 17 + 1°C. March 15-April 18: natural photoperiod; ambient temperature 26°C) April 1&May 16: natural photoperiod at 21°C May 16-Nov 1: 6L/18D photoperiod maintained at 17’C Trial 8 Natural photoperiod; ambient temperature March 15-Nov 1: control for Trials 6 and 7.

Effect of temperature

(24-

(24-26’(Z).

alone

Four individual experiments were designed to provide data on the effect of temperature on ovarian development. Under a constant photoperiod of 6L/ 18D between June 3 and November 16, fish with ovaries still in the refractory period were maintained under the following Conditions: Trial 9 Constant photoperiod, 6L/ 18D; constant temperature, 17 + 1°C. Trial 10 Constant photoperiod, 6L/18D; constant temperature, 21 + 1°C. TriaE II Constant photoperiod, 6L/18D; constant temperature, 26 + 1°C. Trial 12 Natural photoperiod; ambient temperature (24-26”(Z). Control for Trials 9- 11.

31 Effect of exogenous gonadotropin

injection

Four experiments were conducted on the effect of injecting pregnant mare’s serum gonadotropin (PMSG) on fish under controlled and natural conditions. Briefly, the experimental conditions were as follows:

Trial 13 Retarded photoperiod; constant temperature of 17 + 1°C. March 15-Nov 1: 6 L/IS? photoperiod maintained. Injections of PMSG at the rate of 1 II-l/g body weight three times per week commencing four weeks after the start of the experiment. Trial 14 Retarded March ment

15-Nov

photoperiod, constant temperature of 17 + 1°C. 1: same conditions as Trial 13 but without hormone

treat-

Trial 15 Natural photoperiod; ambient temperature (24-26°C). March 15-Nov 1: natural conditions. Injections of PMSG at the rate of 1 IU/g body weight three times per week commencing four weeks after the start of the experiment. Trial 16 Natural photoperiod; ambient temperature (24-26”(Z). March 15-Nov 1: natural conditions as in Trial 15 but without treatment.

hormone

RESULTS Photopheriod

and temperature

effects on ovarian development

The mean percentage compositions of the oocytes during the stages of vitellogenesis, as sampled throughout the experimental period, are given in Tables IIa (Trials l-5) and IIb (Trials 6-8). Fish subjected to the accelerated photoperiod regime at both 21 (Trial 1) and 26°C (Trial 2) showed onset of stage II development by day 129 (July), 49 days following the regulated conditions of 6L/ 18D. Similarly fish subjected to a retarded regime of natural conditions in March and April, followed by constant 6L/18D conditions (Trials 3 and 4), also showed identical development within the same time following the change in test conditions. The photoperiod patterns, employed prior to establishing the constant regime of 6L/18D, differed significantly in Trials 1 and 2 compared with Trials 3 and 4. In the former trials the fish were preconditioned to a cyclic change involving equal increments of increasing and decreasing photoperiod. In the latter (Trials 3 and 4), the fish were being exposed to an increasing daylength regime when the photoperiod was suddenly adjusted and shortened. The changing photoperiod cycle in Trials 1 and 2 was an

0

I II III V n

I II III V n

I II n

4 (26°C)

5 Control (24-26°C)

I II III V n

5

5

5

5

100

100

100

5

5

5

5

100

100

100

100

6

6

6

6

100

100

100

100

3

3

100

100

3

3

3

3

100

100

20 (3)

100 (15)

100

74 (11)

100

34 (5)

100

3

100

Stage I 100 II III V n 3

3 (21°C)

F26”C)

Trial 1 (21°C)

Days (weeks) at 6~/18~

Days (weeks) of experiment

5

100

4

93.1 (1.8) 6.9 (1.8)

4

88.0 (6.5) 8.5 (3.3) 3.5 (3.5)

2

93.6 (2.3) 6.4 (2.3)

2

80.3 (2.9) 15.9 (1.0) 3.8 (3.8)

49 (7)

129 (19)

5

100

4

87.4 (2.7) 11.5 (2.0) 1.1 (0.7)

6

66.8 (8.4) 10.7 (1.3) 22.5 (9.1)

2

93.5 (2.1) 6.5 (2.1)

1

69.2 12.2 18.6

74 (11)

154 (22)

5

100

4

81.3 (4.9) 16.5 (4.2) 2.1 (0.8)

6

51.7 (6.6) 4.9 (1.0) 43.4 (6.3)

2

83.0 (9.1) 16.7 (8.8) 0.3 (0.3)

3

61.0 (5.8) 7.0 (0.4) 31.9 (5.9)

105 (15)

185 (27)

Mean percentage composition (* SD) of oocytes by stages of vitellogenesis. Number of samples-n. -

TABLE Ha

(8.5) (1.3) (9.9) (4.6)

5

100.

47.7 7.4 42.2 2.6 5

2 (7.7) (0.9) (6.2) (1.7)

88.9 (1.3) 11.1 (1.3)

49.4 10.3 34.9 5.4 3

125 (18)

205 (30)

(10.9) (1.7) (14.9) (2.3)

5

100

93.5 5.5 0.8 0.2 4

59.6 5.9 32.4 2.1 2

(2.8) (2.7) (0.4) (0.2)

(3.6) (2.6) (1.2) (2.1)

80.9 (15.7) 6.3 (3.5) 11.7 (11.7) 1.1 (0.5) 2

64.8 5.9 27.0 2.3 2

137 (20)

217 (31)

(7.9) (0.9) (7.9) (0.3)

(4.4) (1.8) (2.7) (0.9)

98.7 (1.3) 1.3 (1.3) 5

92.0 5.8 1.9 0.3 1

87.5 5.4 5.8 1.4 3

2

92.9 (6.8) 7.1 (6.8)

65.6 5.0 29.1 0.3 2

155 (23)

235 (34)

(2.1) (1.5) (0.5) (0.2)

(3.0) (1.1) (1.9) (1.9)

(3.7) (1.9) (3.6) (1.7)

96.9 (1.0) 3.1 (1.0) 5

94.6 4.2 1 .o 0.2 3

89.0 3.1 5.9 1.9 3

95.0 2.8 1.4 0.9 1

75.4 5.8 16.6 2.2 2

166 (24)

246 (36)

33 TABLE IIb Mean percentage composition Days (weeks) at 6L/18D Trial 6 (17OC)

0 Stage 100 I II III V 3 n

33 (5)

100

3

62 (9)

93 (14)

96.7 (3.3) 3.1 (3.1) 0.2 (0.2)

87.4 (8.2) 3.7 (1.9) 8.9 (7.2)

3 94.7 (4.2) 2.1 (1.1) 3.2 (3.2)

3 91.1 (1.8) 3.4 (2.0) 5.5 (5.5)

125 (18)

64.7 (12.9) 7.6 (3.1) 27.7 (15.6)

72.5 (2.4) 9.3 (2.9) 18.2{0.7) 2

2 81.0 7.3 11.6 0.1 6

154 (22)

(9.2) (3.8) (9.1) (0.1)

95.5 2.3 0.3 1.8 5

(1.4) (1.1) (0.3) (0.8)

169 (25)

67.4 7.7 23.6 1.3 2

(1.5) (0.1) (3.0) (1.3)

97.7 (1.9) 0.5 (0.4) 0.0 1.8 (1.5) 5

I II III V n

100

100

6

6

6

6

I

100

100

100

100

100

100

100

n

2

2

2

2

2

2

2

:17”c) 8 Control (24-26°C)

(? SD) of oocytes by stages of vitellogenesis

attempt to stimulate and condense a natural preseasonal cycle. In nature, vitellogenesis in the ovaries of the grey mullet begins shortly before daylength reaches its annual minimum. Neither set of preconditions stimulated development. The control fish (Trial 5) subjected to natural photoperiod regimes at no time attained the same stage of development as those under constant photoperiod of 6L/18D, although some had begun stage II vitellogenesis by day 235. The accumulated photoperiod for the entire experimental group in Table IIa was 947 L-h for Trials I and 2, compared with 1009 L-h for Trials 3 and 4. The minor difference in the accumulated total L-h was not judged to be significant or to have influenced the results before the light’ regime of 6L/18D was established. The responses of individual fish subjected to the constant photoperiod and differing experimental temperatures varied in the time of appearance and proportion of stage III oocytes. The latter were observed four to eight weeks earlier in the ovaries of fish maintained at 21°C (35-43%) than in those maintained at 26°C (2-12%). Atresia (stage V) of vitellogenic oocytes commenced more rapidly at 26°C before the fish reached the stage of functional maturity. During the first year the constant photoperiod regime of 6L/ 18D (Trials 1-5) proved to be effective in stimulating development of vitellogenesis irrespective of preconditioning adjustments. In the following

34

year the most successful procedure was repeated (Trials 6-8) but at 17°C and with controls. As in the first year, the time of onset of vitellogenesis was observed within 62 days for both Trials 6 and 7 following exposure to a constant photoperiod of 6L/18D, and irrespective of preconditioning. Fish in Trial 7 had been subjected to an advanced photoperiod at 21°C before the experiment began, with an accumulated exposure of 784 L-h. The difference in percentage composition of stage III oocytes (28% compared with 12%) was pronounced. Atresia (stage V) was observed earlier in Trial 7 although the reason was not apparent.

natural photoperiod

first year: experiments #l through 5 second year: experiments # 6,7 and 8

--

I I I I I I, 3115

w12

I III lllll

I I I I I I I I,

5116 6/36/13

7n8

I!,

my15

I I I I I, sn2

I I I,

ml7

II

lln5

Time Fig. 1. Experimental light regimes and the natural light cycle in Hawaii. (*trial)

The photoperiod regimes and their duration for eight individual trials are illustrated in Fig. 1. The conclusions from the experiments are readily discerned from the figure: vitellogenesis of the oocytes was stimulated effectively by a short and constant photoperiod regime of 6L/18D and it was not related to any previous photoperiod condition. Temperature appeared to regulate maturation. femp.erature

alone and ovarian development

Data on mean percentage composition of oocytes by stages of vitellogenesis in relation to temperature (with photoperiod constant at 6L/18D) are given in Table III and summarized in Fig.2. Experimental temperatures were 17’, 21”, and 26°C for Trials 9-l 1, respectively. Trial

001

001

9

001

9

(SZ) 9LI

(6’0) 8’; tI.8)L’SI (0.I) L’9 (S-8) 8’SL

(1'2)I.: (E'9)0'8 (VZ) I’L (P'S)0’08

(EZ) z91

9’IP S’9 6’IS

(TZ) IPI (Z1) 88

I’SI 6’2 O’Z8

9

6’l7Z S'II 6’E9

9

8’9 Z’E6

(8) 9s

(9.Z) L’Z (9'1)Z'Z (‘FE) Z’S6

9

(VZ) 9’E (9.z)O'II (0) P’S8

&‘I) &‘I)

9

001

S

.,

(P) 82

6

001

9

9

OOI

001

9

001 U

a%paxad

A III II I a8w

u

III II I

A

6 I”!‘1

(3,LI)

(3,IZ) 01

0,9z) II

ZI

IO-QUO3

t3,9z-!Jzz)

wan

asm9 $8 (syaafi) s&a

Ll

9 A III II I

I

u

001

S

001

9

001

9

001

S

s!saua8ollalra 30 sa8els dq sabboo 30 (as T) uogsodwo3

(LI) STT

(I’L) t0.I) (9.L)

(8.6) tw (I.61

S

9 (Z’L) (9’0) (2-L)

(S’O) L’O CL.0 6’6 (Z’Z) 8’68

9

001

S

CL.01 S-1 (S’E) 9’91 (6’0 6’I8

9

001

S

P (1.0)1'0 (01) 8’SZ (S’OI) O’OE (9.Z) Z’S (9’Z) O’S (8X1) 6'89 (9'6)O'S9

e.: L’6Z 6’S Z’Z9

(E’I) (E’9) (E’I) (6.P)

CL.01 (2.6) (S-0) (56)

CL’01 Z.i (I’$) O’L CL.01 9’Z (S'P)Z’L8

I.! 6’LI 6’E I’LL

(8‘E) P’P (6’0 8’S (Z’S) E’68

(8'9)I’L (8’9) 6’26

(P’O)T’T

(1'1)8'E WI) L’P6

(E-0) S’O

(Z’O) p’o

9

S

S

8,

s

,.,I

36 Primary

oocyte

(I)

stage

(II

)

Yolk

stage

(III

)

globule

Functionat

maturity

26’C

I

#12 fcontrolf

E”“““““““’ ,,/,,/,,,/,,,,/,,::/::/::: I

I

stage

Yolk vesicle

50

Fig. 2. Summary Induced spawning

I

100 Days of ovarian of individual

’

’

I

150

4

I

200

development fish indicated

in trials 9-12. Photoperiod, by arrows. (#=trial)

6L/18D;

time

in days.

12 acted as a control and employed ambient temperature conditions (24-26”(Z) and a natural light cycle. At 17°C the onset of stage II occurred before day 56. Oocytes at the yolk globule stage (III) were also observed at that time. By day 115 the percentage of primary oocytes (I) had decreased substantially to 65%, whilst oocytes at the yolk globule stage (III) had increased rapidly (30%). Atretic ova (V) were first observed by day 141, and the percentage of stage I oocytes had increased once more. At 21 and 26°C the onset of stage II oocytes also occurred before day 56, although yolk-laden oocytes (III) were not apparent until day 84 in the fish maintained at the higher temperature (26°C). In general, the percentage composition and time of appearance of the vitellogenic oocytes of fish held at 21°C were similar to those of the fish at 17°C. The percentage of stage III oocytes was high, 41.6% by day 115. Fish held at 26°C showed less-advanced development. Only 4.2% yolk-laden oocytes were observed at day 141 when atresia was again noted. Towards the end of the experimental period, several fish held at 17 and 21°C had developed through to the stage of functional maturity (III, and over 600 p mean oocyte diameter). A total of eight fish were selected and induced to spawn successfully by injections of partially purified salmon gonadotropin (1 mg equivalent to 2 150 IU HCG). The injection dose and sequence are presented in Table IV. None of the fish held at 26°C com-

37 TABLE IV Injection sequence and dose of salmon pituitary with fish from trials 9 and 10. Fish No.

Date

Injection dose.(mg)

gonadotropin

Cumulative dose

Mean egg diameter

(Irdg

C/J)

body wt)

(1 mg equivalent

to 2150 IU HCG) and spawning results

Fertiliz&ion rate (%)

Experimental

conditions

Temperature

Photoperiod

natural

‘75

17°C

retarded

12 h 30 min

natural

53

17°C

retarded

Time to spawn from last injection

Spawning

12 h

643

2.5 2.5 2.5 2.5 5.0 10.0

2.2 4.4 6.6 8.8 11.0 13.2 18.6 24.1 29.6 35.1 46.1 68.0

S/O9 S/l0 8/l 1 S/12 8113 S/14 S/15 8116 8111 S/18

1 1 1 1 2.5 2.5 5.0 10.0

1.8 3.6 5.4 1.2 9.1 13.6 18.1 27.2 45.3

614 631

l/l8 7119 7120

2.5 5.0 10.0

6.0 18.0 42.0

673 921

10 h 30 min

artificial

0

17°C

retarded

3.0 10.0

5.0 21.5

617 617 931

15 h 10 min

artificial

0

21°C

retarded

IO/12 10/13 10114

3.0 5.0 10.0

5.4 14.4 32.4

643 650 667

19 h 30 min

natural

87

21°C

retarded

lo/l8 IO/19 10120

5.0 10.0

9.1 27.4

670 681 966

16 h 10 min

natural

94

21°C

retarded

10/31 11/01 11102 11/03

5.0 10.0 5.0 10.0

4.2 12.5 16.7 25.0

654 663 798 996

19 h 50 min

natural

90

21°C

retarded

12/13 12/14 12/15 12116 12111

10.0 5.0 10.0 10.0

11.9 11.8 30.0 41.5

671 684

9h

natural

90

21°C

retarded

1

S/16 S/l I 8118 S/19 8121 S/22 S/23 S/24 8125 S/26 S/28 8129 S/30

9/20 9121 9122

1

;

654

682

699 930

664 926

)

689 933

38

pleted vitellogenesis and no attempt was made to induce spawning. For induced spawning of fish with oocytes in the later period of stage III (mean diameter below 650 FL), Kuo et al. (1974) describe procedure options. Although the oocytes in fish maintained at 17°C were in the range 614-643 p in diameter, the fish received either small increasing doses of hormone with an accumulated total of between 8.12-35.09 pg/g body weight from seven to ten injections, or between 27.8-32.9 pg/g body weight from two injections, the second twice the dose of the first. Spawning of fish held at 21°C was induced with total doses of between 2 1.5-41.5 pg/g body weight from two to four injections (Table IV). Some differences were observed in the morphology of stage III oocytes from fish maintained at 17°C compared with those found in fish during the natural breeding season or in other experimental conditions (Trial 10). The oocytes were lighter in appearance and probably the yolk was less dense. The oocytes darkened as the accumulated hormone dose reached 20 pug/g body weight, and more yolk was deposited. Examination of the oocytes of fish in the control (Trial 12) showed that no stage II oocytes developed within the experimental period.

Exogenous gonadotropin

injections on ovarian development

The possible use of exogenous mammalian gonadotropin, pregnant mare’s serum gonadotropin (PMSG), was tested for initiating ovarian development in association with photoperiod and temperature. Data on the effects on vitellogenesis of a series. of injections (three per week) is presented in Table V. Injections were begun 28 days after the start of the experiment. In these trials fish were held under a constant 6L/18D regime photoperiod and maintained at 17°C. Stage II and stage III oocytes were observed both in injected individuals (Trial 13, after 15 injections) and in uninjected individuals (Trial 14) on day 62 of the experiment. Thereafter, development was more rapid in the fish receiving the injections (Trial 13), although the difference in percentage composition of vitellogenic oocytes between the two groups was not significant. Vitellogenesis reached a peak by day 120 (33.5%) in Trial 13 and by day 90 (27.7%) in Trial 14. Atresia was more advanced in the injected fish. The duration of ovarian development was 138 days in Trial 13, shorter than that in Trial 14 (over 168 days). The PMSG injections therefore accelerated ovarian development. Data from the control experiments (Trials 15 and 16) indicated that under natural photoperiod and ambient temperature conditions vitellogenesis was initiated after 29 injections (day 94). The peak of oocyte

100

2

I

n

16 Control Ambient (24-26”(Z) without

PMSG

100

I II III V

2

3

100

I II III V

n

3

100

n

I II III V

0

n

15 Ambient (24-26°C) plus PMSG

14 (17°C) without PMSG

13 (17°C) plus PMSG

Number of injections

Days (weeks) of experiment

100

2

2

100

2

100

2

2

2

100

2

2

104

2

5.9 (5.6)

0.7 (0.7)

2

94.1 (5.6)

99.3 (0.7)

2

72.4 (2.4) 9.3 (2.9) 18.2 (0.7)

0.3 (0.3)

3

64.7 (12.9) 7.6 (3.1) 27.7 (15.5)

2

100

69.2 3.1 20.0 7.8 2

(28.4) (0.7) (20.0) (7.8)

1.3 (1.3) 2

2

100

(6.9) (28.5) (0.1)

(21.5)

2

100

10.5 (6.5) 19.9 (19.9) 0.2 (0.2) 2

69.4 (13.3)

4.8 (3.0) 6.3 (6.3) 2.0 (1.7) 2 61.5 9.4 28.9 0.1 2

87.0 (5.0)

9.3 (1.9) 11.9 (8.5) 3.0 (3.0) 2

1

100

73

75.8 (3.6)

1

90.8 2.9 0.0 6.3

67

200 (29)

of PMSG for ambient

186 (27)

injections

67.4 (1.5) 7.7 (0.1) 13.6 (2.2)

2.7 1

92.7 4.6 0.0

65.3 (33.4) 1.2 (0.1) 33.5 (33.5) 2

60

169 (25)

53

154 (22)

with and without

99.7 (0.3)

100

100

3

3.7 (1.9) 8.9 (7.2)

87.4 (8.2)

3

2.3 (0.4) 32.4 (17.5)

3

65.3 (17.2)

2.1 (1.0) 21.4 (12.7)

41

124 (18)

for females

76.6 (13.5)

29

94 (14)

3

96.7 (3.3) 3.1 (3.1) 0.2 (0.2)

3

93.6 (6.4) 1.2 (1.2) 5.2 (5.2)

15

62 (9)

3

100

3

100

35 (5)

Mean percentage composition (+ SD) of oocytes by stages of vitellogenesis and 17°C temperatures with controls. (Trials 13-16)

TABLE V

2

100

2

7.3 (7.3) 0.4 (0.4)

92.3 (7.7)

9.7 (4.0) 0.3 (0.3) 1.2 (0.8) 2

88.9 (5.1)

1

100

79

214 (31.)

2

LOO

2

92.3 (4.4) 4.4 (2.7) 0.0 3.3 (1.7) 2

1

100

86

230 (33)

40 development (28.9% stage III) was reached after 67 injections, thereafter followed by some atresia (stage V). No vitellogenic oocytes were observed in the control experiment without injection (Trial 16). It was apparent that ovarian development was advanced by the injections of PMSG, by which vitellogenesis was initiated. No fish was observed to reach the stage of functional maturity at the ambient temperatures of 24-26°C with injections (Trial 15) or without (Trials 5, 12, and 16).

DISCUSSION

Many experiments with a variety of fishes have indicated that photoperiod and temperature are the two most significant external regulators of the reproductive cycle. However, the relative importance of these factors varies with species. Burger (1939) hypothesized that the time of avian gonad maturity could be predicted on the basis of a summation of photoperiod and thermal effects. The results of this study demonstrated that the onset of vitellogenesis in the grey mullet was not determined by a cumulative effect of day length but directly by the effective photoperiod of the time. Similar results have been reported for the brook trout (Henderson, 1963). For the grey mullet, the time of onset of vitellogenesis was determined to be about eight weeks after exposure to a short photoperiod regime (6L/ 18D) at temperatures ranging from 17 to 26°C. The response of oocyte development to such a retarded photoperiod was consistent; it was unrelated to any preconditioning photoperiod regime, including a simulated natural photoperiod cycle. It is certain that retarded photoperiod was more effective than temperature in initiating development. The fact that vitellogenesis is induced naturally prior to the breeding season demonstrates that endogenous rhythm of reproductive activity is initiated by photoperiod and temperature. However, that rhythm is not totally controlled by either factor. The controlling mechanism of vitellogenesis in the grey mullet was the marked endogenous activity of the pituitary gland, which was itself initiated by the external environmental factor of daylight. Once the secondary growth phase of oocyte development had begun, rate of development was directly influenced by and inversely proportional to temperature for the experimental temperatures used, namely ambient, 17, 21 and 26°C. A long daily photoperiod appeared to reinforce the effect of the higher temperatures (24-26°C) inhibiting vitellogenesis. A retarded photoperiod reinforced development at 17 and 21°C. A constant temperature of 21°C and a photoperiod of 6L/18D produced the most significant result in advancing the development of vitellogenesis. In sum-

41 mar-y, it was demonstrated that the short photoperiod of 6L/ 18D had a stimulatory effect initiating oocyte development, whilst temperature appeared to regulate vitellogenesis toward functional maturity. The data indicate that development also was accelerated by constant exposure to a temperature of 17°C and a 6L/18D photoperiod. However compared with development at 21°C and above, oocyte maturation at 17’C was incomplete. At *17”C, only limited yolk deposition occurred in stage III oocytes. Subsequent induced spawning and egg incubation showed that embryonic development failed to pass the blastula stage. The spawning of grey mullet can be induced by injections of partially purified salmon pituitary gonadotropin SG-G 100 (Shehadeh et al., 1973a) if hypophysation is begun when the fish are functionally mature. The injection schedule described by them proved to be unsuitable for the experimental fish maintained at 17°C as the oocytes were functionally mature but the mean oocyte diameter was between 600-650 p. Spawning of these fish was successfully induced by a series of injections using procedures described by Kuo et al. (1974). .Their schedule had been developed to induce spawning of fish that were still undergoing prolonged and slow final development at the end of the natural breeding season. For the fish maintained at 17”C, microscopic examination of stage III oocytes revealed that they were lighter in appearance than normal. This was probably due to reduced yolk deposition, as the oocytes became normal in appearance following an accumulation of SG-GlOO. The reproductive cycle of many if not most vertebrates is under the dual control of an internal physiological rhythm and an external seasonal rhythm. The refractory period in the reproductive cycle is considered to be the time during which these two rhythms coincide and reinforce each other. As fish are exposed to changing environmental conditions, such as photoperiod and temperature, the external rhythm begins to dominate and its influences on the reproductive processes are transmitted by changes in the quantity of gonadotropin released from the pituitary gland. The results from this study demonstrated that vitellogenesis was initiated eight weeks after fish were exposed to a retarded photoperiod. This delay in the response to external stimuli implied that the changeover from the primary to secondary growth phase was a result of cumulative hormone influence (Barr, 1968). Many experiments have demonstrated the effect of gonadotropins on stimulation of vitellogenesis. Ahsan and Hoar (1963) reported that luteinizing hormone (LH), alone or in combination with follicle stimulating hormone (FSH), stimulated vitellogenesis and growth of oocytes in Gasterosteus aculeutus. Neither human chorionic gonadotropin (HCG) nor PMSG was as effective as LH, although both produce some stimulation of

42 oocyte growth. A negative effect on vitellogenesis of PMSG, alone or combined with other mammalian gonadotropins, was reported in hypophysectomized Carassius auratus (Yamasaki, 1965). By contrast, Tchechovitch (1952) found appreciable stimulation of the ovary of three teleost species injected with mammalian hormone. The response of fish ovaries to HCG and PMSG is inconsistent. In this study, injections of PMSG were found to be effective in initiating vitellogenesis, and the completion of oocyte maturation was then dependent upon water temperature. Functional maturity was noted only for those fish maintained at a low temperature. Atresia (stage V) of oocytes was observed in the injected fish earlier than in untreated controls held under identical environmental conditions. ’ The reason for this early appearance of atretic oocytes after injections of PMSG is not cle;lr. It was probably due to the low injection dose (1 III/g body weight) used throughout, which was insufficient to meet the requirement for successful completion of vitellogenesis. However, the result indicates the stimulatory effect of PMSG on vitellogenesis and encourages further investigation. In summary the experimental work with manipulation of photoperiod and temperature, alone or in conjunction with the use of exogenous hormones, demonstrates that two primary objectives for aquaculture could be achieved, namely : ( 1) Individual spawning frequency could be increased. (2) The spawning season could be prolonged throughout the year. The benefits to aquaculture are that the mass propagation of fingerlings can be increased annually and regulated through a hatchery system. The efficiency of subsequent farm operations and the utilization of facilities can therefore be substantially improved. ACKNOWLEDGEMENTS This study was funded by the National Oceanic and Atmospheric Administration, Office of Sea Grant, Grant No. 04-3-l 58-14. The authors wish to thank Diane Henderson and Holly Maxson for assistance with the manuscript. REFERENCES Ahsan, S.N. (1966). The control of cyclical changes in the testicular activity Couesius plumbeus (Agassiz). Can. J. 2001. 44, 149-159. Ahsan, S.N. and Hoar, W.S. (1963). Some effects of gonadotropic hormones stickleback, Gasterosteus aculeatus. Can. J. Zool. 41, 1045-1053.

of the lake chub on the three-spine

43 Bacon, N. (MS1973). Induced Breeding

Design of an environmental control laboratory for fish. In: The Grey Mu&t: and Larval Rearing Research 1972-1973, Vol. II. Oceanic Institute Report

01-73-128. Baggerman, B. (1957). An experimental study on the timing of breeding and migration in the three-spined stickleback (Gasferosteus aculeatus L.). Arch. nkerl. Zool. 12, 105-317. Barr, W.A. (1968). Patterns of ovarian activity. In: E.J.W. Barrington and C.B. Jqrgensen (Editors), Perspectives in Endocrinology. Academic Press, London and New York, N.Y., pp. 163-237. Bullough, W.S. (1940). The effect of the reduction of light in spring on the breeding season of the minnow (Phoxinus laevis Linn.). Proc. zool. Sot. Lond. A, 100, 149- 157. Burger, J.W. (1939). Some aspects of the roles of light intensity and daily length of exposure to light in the sexual photoperiodic activation of the male starling. J exp. zool. 81, 333-341. Harrington Jr., R.W. (1956). An experiment on the effects of contrasting daily photoperiod on gametogenesis and reproduction in the centrarchid fish, Enneacanthus obesus (Girard). J. exp. Zool. 131,203-223. Harrington Jr., R.W. (1957). Sexual photoperiodicity of the cyprinid fish, Notropis bifreuatus (Cope), in relation to the phases of its annual reproductive cyc1e.J. exp. Zool. 135, 529-556. Harrington Jr., R.W. (1959a). Effects of four combinations of temperature and daylength on the ovogenetic cycle of a low-latitude fish, Fundulus confluentus (Goode and Beau). Zoologica 44, 149-168. Harrington Jr., R.W. (1959b). Photoperiodism in fishes in relation to the annual sexual cycle. In: R.B. Withrow, (Editor), Photoperiodism and Related Phenomena in Plants and Animals, Vol. 55. Washington, D.C. Am. Assoc. Adv. Sci. Publ. 651-667. Henderson, N.E. (1963). Influence of light and temperature on the reproductive cycle of the eastern brook trout, Salvelinus fonfinalis (Mitchill). J. Fish. Res. Bd Can. 20, 859-897. Hoar, W.S. (1959). Endocrine factors in the ecological adaptation of fishes. In: A. Gorbman, (Editor), Comparative Endocrinology. Wiley, New York, N.Y., pp. l-23. Kuo, C.-M., Nash, C.E. and Shehadeh, Z.H. (1974). A procedural guide to induce spawning in grey mullet (Mugil cephalus L.). Aquaculture 3, l-14. Liao, I.C. (1969). Artificial propagation of grey mullet, MugiZ cephafus Linnaeus. Fish. Ser. Chin.Am. jt Comm. rur. Reconstr. 8, 10-20. (In Chinese, English summary). Matthews, S.W. (1939). The effects of light and temperature on the male sexual cycle in Fund&us. Biol. Bull. 77, 92-95. Medlen, A.B. (1951). Preliminary observation on the effects of temperature and light upon the reproduction in Gambusia affinis. Copeia 148-153. Merriman, D. and Schedl, H.P. (1941). The effects of light and temperature on gametogenesis in the four-spined stickleback, Ape&es quadracus (Mitchill). J. exp. 2001. 88,413-449. Shehadeh, Z.H. and Ellis, J.N. (1970). Induced spawning of the striped mullet Mugil cephaZus L. J. Fish BioL 2, 355-860. Shehadeh, Z.H., Kuo, C.-M., and Milisen, K.K. (1973a). Induced spawning of grey mullet Mugil cephalus L. with fractionated salmon pituitary extract. J. Fish Biol. 5,471-478. Shehadeh, Z.H., Kuo, C.-M., and Milisen, K.K. (1973b). Validation of an in vivo method for monitoring ovarian development in the grey mullet (MugiZ cephalus L.) J. Fish Biol. 5, 489-496. Shehadeh, Z.H., Kuo, C.-M., and Nash, C.E. (1973c). Establishing brood stock of grey mullet (Mugi cephalus L.) in small ponds. Aquaculture 2, 379-384. Tang, Y.A. (1964). Induced spawning of striped mullet by hormone injection. Jap. J. Ichthyol. 12, 23-28. Tchechovitch, G. (1952). La ponte artificielle chez quelques Poisson d’eau deuce provoque par les hormones hypophysaires. Arh. biol. Nauka, Belgrade 4, 27-85. Yamasaki, F. (1965). Endocrinological studies on the reproduction of the female goldfish, Carassius aura&s L., with special reference to the function of the pituitary. Mem. Fat. Fish. Hokkaido Univ. 13, l-64. Yashouv, A. (1969). Preliminary report on induced spawning of M. cephalus (L.) reared in captivity in freshwater ponds. Bamidgeh 21, 19-24. Yoshioka, H. (1962). On the effects of environmental factors upon the reproduction of fishes. 1. The effects of day-length on the reproduction of the Japanese killifish Oryzias latipas. Bull. Fat. Fish. Hokkaido Univ. 13, 123-136.