Marine Biology (1996) 124:637-649

9 Springer-Verlag 1996

J. C. Bonardelli 9 J. H . H i m m e l m a n 9 K. D r i n k w a t e r

Relation of spawning of the giant scallop, Placopectenmagellanicu to temperature fluctuations during downweUing events

Received: 8 March 1995 / Accepted: 28 March 1995

We examined the relation of spawning to biological and physical factors for the scallop Placopecten magellanicus (Gmelin, 1791) over 8 yr (1984 to 1991) in the Bale des Chaleurs, southwestern Gulf of St. Lawrence, Canada. Spawning was always abrupt and occurred between July and mid-September. It did not appear to be related to the abundance of phytoplankton or particulate organic carbon and nitrogen in the water. It further showed no relationship to lunar or tidal phases or to current velocity. Spawning consistently occurred during the summer temperature maximum, but did not coincide with any critical temperature or cumulative temperature threshold. All but one of the 33 spawning events, for which temperature data were recorded, were associated with temperature changes; 25 of these were sharp temperature increases and 7 were during strong temperature fluctuations when the mean temperature was 9 to 14 ~ Both types of temperature changes were caused by downwelling of warm surface water. The delay by about 1 d in time of spawning between sites coincided with the rate at which downwelling events propagated into the bay. Virtually all of the spawning events resulted in gametes being ejected into warm water masses where conditions are likely to favour larval development. Abstract

Communicated by J. P. Grassle, New Brunswick J. C. Bonardelli ( [ ] ) . J. H. Himmelman D~partement de biologie et GIROQ (Groupe interuniversitaire de recherches oc6anographiques du Quebec), Universit6 Laval, Qu6bec, Qc, G1K 7P4 Canada K. Drinkwater Bedford Institute of Oceanography, Department of Fisheries and Oceans, Physical and Chemical Sciences Branch, Dartmouth, N.S. B2Y 4A2 Canada

Introduction

Many environmental factors have been suggested as proximate spawning cues for marine invertebrates. One is phytoplankton. Numerous species, particularly spring spawners, release their gametes during phytoplankton increases (Thorson 1950), and for several species laboratory studies demonstrate that phytoplankton can stimulate spawning when other factors are maintained constant (Himmelman 1975; Starr et al. 1990). Coupling spawning with plankton blooms ensures that larvae have an abundant supply of food. Numerous studies, especially in tropical regions, indicate that spawning is related to monthly (generally moonlight) or semi-monthly (tidal) rhythms (Korringa 1947; Hauenschild 1960; Skreslet and Brun 1969; Kennedy and Pearse 1975). However, other studies report contradictory evidence (Skreslet 1973; Pearse 1975). The relation between spawning and a tidally related exogenous signal is difficult to evaluate, particularly in regions where tidal cycles directly influence circulation patterns and water column stability (and ultimately temperature and food availability). Temperature is often suggested as a signal, based on the observation that spawning occurs at about the same temperature in different locations or years (Loosanoff and Davies 1950, 1952; Galtsoff 1961; Loosanoff 1969; Barber and Blake 1991). Most hypotheses relating temperature to spawning originate from Orton's (1920) suggestion that a critical temperature, a "physiological constant" for each species, determines the start and end of the breeding season. However, subsequent workers have demonstrated that latitudinally separated populations breed at different temperatures (Korringa 1957; Sastry 1970; Newell et al. 1982; MacDonald and Thompson 1988). Other studies suggest that spawning is triggered by increases in temperature (Bricelj et al. 1987; Minchin 1987, 1992a; Barber and Blake 1991). Sharp temperature changes often

638 cause s p a w n i n g in the l a b o r a t o r y a n d are used to o b t a i n g a m e t e s in hatcheries ( L o o s a n o f f a n d Davies 1963; Sastry 1963; Culliney 1974), however, the fluctuations required to p r o v o k e s p a w n i n g in the l a b o r a t o r y rarely o c c u r d u r i n g s p a w n i n g in the field ( H i m m e l m a n 1981). M o s t analyses of the relation of s p a w n i n g to enviro n m e n t a l factors are limited b y (1) infrequent s a m p l i n g o f animals a n d (2) insufficient e n v i r o n m e n t a l data. Studies are often confined to d o c u m e n t i n g a n n u a l rep r o d u c t i v e cycles by s a m p l i n g at m o n t h l y intervals. S p a w n i n g usually occurs within days or weeks so t h a t the invertebrates need to be s a m p l e d m o r e frequently (preferably every 2 to 3 d). Biological a n d physical variables (food, t e m p e r a t u r e , current) s h o u l d be m e a s ured at even m o r e frequent intervals so t h a t events coinciding with s p a w n i n g can be identified. T h e present s t u d y investigated the relation of s p a w n i n g to e n v i r o n m e n t a l factors for the giant scallop, Placopecten magellanicus, in the Baie des Chaleurs, the largest b a y in the G u l f of St. Lawrence. N u m e r o u s studies h a v e described r e p r o d u c t i v e events in the giant scallop. G a m e t o g e n e s i s is usually relatively sync h r o n o u s a m o n g s t individuals in the p o p u l a t i o n ( N a i d u 1970). A l t h o u g h the r e p r o d u c t i v e cycle is usually annual, with g o n a d a l d e v e l o p m e n t d u r i n g the spring a n d s p a w n i n g in the s u m m e r or a u t u m n ( N a i d u 1970; R o b i n s o n et al. 1981; P a r s o n s et al. 1992), instances of s e m i - a n n u a l cycles have been r e p o r t e d ( D u p a u l et al. 1989; D i B a c c o 1993). W e l c h (1950) a n d N a i d u (1970) state t h a t t e m p e r a t u r e fluctuations, p e r h a p s c a u s e d by w i n d events or s t o r m - i n d u c e d mixing, m a y initiate spawning, a n d P a r s o n s et al. (1992) indicate t h a t s p a w n i n g in B a y of F u n d y is associated with tides. O u r s t u d y of s p a w n i n g is u n i q u e because m a n y s p a w n ing events are d o c u m e n t e d , using s a m p l i n g at frequent intervals. F u r t h e r , we collected an extensive set of biological a n d physical d a t a in the vicinity o f the scallop beds p e r m i t t i n g the e v a l u a t i o n o f n u m e r o u s p o t e n t i a l s p a w n i n g factors.

Methods Determination of spawning of Placopecten magellanicus Spawning events were identified as decreases in gonadal mass. Adult scallops were collected periodically by SCUBA divers from July to mid-September from 1984 to 1991. Each sample consisted of ~ 30 scallops, measuring 95 to 161 mm in shell height. Relative gonadal size is maximal and independent of size within this range (Bonardelli 1994). In any given year, spawning was followed in one to four different scallop beds. Beds at Gascons (25 m in depth), Bonaventure (16 m) and Carleton (13 m), spaced at ~ 50-kin intervals along the north shore of Baie des Chaleurs (Fig. 1), were studied from 1984 through to 1988. In 1989 sampling was limited to Gascons, and in 1990 to Gascons and Bonaventure. Finally, in 1991 we sampled at Pabos (23 m, 30 km east of Gascons) in addition to the three regular sites to further examine geographic variation in spawning. The sampling intervals were generally every 2 to 4 wk in 1984, every

Fig. 1 Map of the Baie des chaleurs showing locations where giant scallops were collected (star) and where current meters and thermographs were moored (open circle) during 1984 to 1991 week in 1985, 1988, 1989, 1991 and every 3 to 7 d in 1986, 1987 and 1990. (A final sample was sometimes collected in October to obtain the minimal gonadal mass after spawning). Scallops were transported in an insulated box and dissected within 3 h of collection. No spawning occurred during transport. We measured shell height (to the nearest 0.1 ram) and then dissected the scallops to determine the sex and wet mass of the gonad (to the nearest 0.01 g; foot removed) after draining for ~ 5 min on absorbant paper. Changes in gonadal mass between sampling dates were quantified using the gonadal mass index (Bonardelli and Himmelman 1995). Briefly, we determined the index (a) for each scallop by directly scaling gonadal mass (Y') to shell height, using the following allometric equation: Gonadal mass index (a) = measured gonadal mass (7) x shell height b. For each site and year, the coefficient "b" was the common slope determined from the relation of gonadal mass to shell height (log transformed) after pooling the samples collected during July, the month prior to the first spawning. Slopes of the regressions of gonadal mass to shell height were similar before and after spawning (Bonardelli and Himmelman 1995). During spawning, gonadal mass decreased significantly, indicating the release of gametes. Both males and females were treated together, because no differences in allometric relationships between the sexes were detected in any of the locations or years (Table 4.1, Bonardelli 1994). The gonadal mass index (a) for each scallop was then transformed to the mass for a standard 125 mm scallop using the equation, / ~ ' = a x 125b, b being the pooled slope. This size was near the mean shell height of scallops collected in this study (mean = 127mm, SD = 12, n =7122). To illustrate changes in gonadal mass over time, we plotted the mean scaled gonadal mass and its standard error from the individual values. During spawning, scaled gonadal mass and gonosomatic indices violated assumptions of normality and homogeneity of variance. Thus, we tested for differences in gonadal mass between successive dates by applying a non-parametric Jonckheere test to ranked scaled gonadal mass values (Cap6ra/t and Van Cutsen 1988). We made a modification of the statistic to specifically compare observations

639 taken at successive time points where evidence is sought of a downward trend in gonadal mass as would occur during spawning. The modified statistic accounts for certain selected elements of the sum; in this case only successive dates are compared (date1 with date> date~ with datei + 1; Bonardelli and Himmelman 1995). When significant differences were detected in the time series for gonadal mass, a post-hoc modified Jonckheere test was performed to identify the intervals in which significant decreases occurred.

Pabos 1991 ] Gascons 1984-

Results F r e q u e n c y a n d t i m i n g of s p a w n i n g O v e r the 8 yr s t u d y p e r i o d , we o b s e r v e d 37 s p a w n i n g events, b a s e d o n decreases i n g o n a d a l m a s s (Figs. 2, 3).

'I

i ,

'1

i

m

1987

m m m

1988"

Abundances of phytoplankton (chlorophyll a determinations corrected for phaeopigment content) and of suspended organic particles [particulate organic carbon (POC) and nitrogen(PON)] were determined from duplicate samples taken twice daily (06:00 and 18:00 hrs; weather conditions permitting) collected at 1 m above the bottom (using 2-liter Niskin bottles) at the three major sites in 1985 and from weekly samples collected at 3 m above the bottom at Gascons in 1989 (the latter from C6t~ et al. 1993). In both years, samples were filtered using Whatman GF/F filters, and chlorophyll a concentration was determined with a Turner fluorometer (Model III) using the method of Strickland and Parsons (1972). We used the 90% acetone extraction method of Yentsch and Menzel (1963), as modified by Holm-Hansen et al. (1965). Filters for POC and PON determinations were frozen at - 2 0 ~ and later analyzed using a Perkin Elmer CHN elemental analyzer (Model 240B). To examine the relation of spawning to the lunar cycle, we determined the number of spawning events that occurred during different phases of the moon, based on spawning events at Pabos, Gascons, Bonaventure and Carleton from 1985 to 1991. In these analyses the lunar cycle was divided into eight equal periods (arcs). Since gonadal decreases often did not occur within the defined arcs, we used the midpoint of the interval of the decrease in gonadal mass to assign spawning events to arcs. When several significant decreases in gonadal mass occurred sequentially, only the first decrease was considered, because the subsequent decreases could have been initiated by gametes released during the first decrease. To remove imprecisely documented spawnings from the analysis, gonadal decreases indicated from sampling at > 7-d intervals were not considered. The statistical methods of Batschelet (1981) were used to calculate a mean vector and axis (corrected for grouping). We applied a Rayleigh's test for randomness to the circular distribution of the spawning events (Ro, lunar phase) and to the axial distribution obtained by doubling the angles (R2o, tidal phase). For each scallop bed, we obtained continuous records of temperature, salinity, and current velocity and direction between July and December, for the years 1985 to 1991, using Aanderaa RCM-4 current meters that were moored at 1 m above the bottom. These data were edited to remove spikes, filtered using a three or five point equally weighted running mean (for 30- and 15-min sampling periods, respectively). We previously used the 1985 to 1987 data to describe temperature and current variability in relation to wind forcing and the Gasp~ Current (Bonardelli et al. 1993). Additional temperature data were obtained with Peabody-Ryan and Hiigriin thermographs and a Beckman CTD profiler. Finally, current meters and thermographs were also employed to record bottom temperatures in the winters of 1986 and 1991. All data were reduced to hourly values.

i ,

19851986-

Biological and physical conditions

'I

m

1989

m

1990 1991" Bonaventure

1984" 1985

1986-

mmm

1987

m m

1988"

m m ~

m

m

m m m m m

1989" no samples 1990" 1991 Carleton

1984 1985 1986-

19877

~ m m

1988 1989" no samples 1990 no samples

1991

"" ..... ijul i'" .... ,,11,,,,,'l,,,,,,i,,,,,,i,,,,,,i,,,,,,I,Aug J

..... I'"'"l'"'Hl'"i"lSep

Fig. 2 Placopecten maqellanicus. Summary of spawning events for scallops collected at Pabos, Gascons, Bonaventure and Carleton in the Baie des Chaleurs from 1984 to 1991. Bars indicate the intervals over which a significant decrease (p < 0.05) in gonadal mass was recorded. Successive sampling intervals are considered as a single event. Longer length of some bars due to reduced frequency of sampling when gonads appeared spent

All o c c u r r e d b e t w e e n m i d - J u l y a n d e a r l y S e p t e m b e r . S p a w n i n g u s u a l l y o c c u r r e d later i n the i n n e r p o r t i o n of the b a y (Fig. 2). T h e f r e q u e n c y of s p a w n i n g v a r i e d a m o n g sites a n d s p a w n i n g events were g e n e r a l l y m o s t a b r u p t at B o n a v e n t u r e a n d C a r l e t o n (Fig. 3). M u l t i p l e s p a w n i n g events, s e p a r a t e d b y p e r i o d s of s t a b l e or

640

iZ1

Pabos

0 / b=1.4, 27-

Gascons

J'

/ b:l ~

I b:2

b=1.7

~

J

Bonaventure

2

27-

E 18-

"O

g 9-

b=2.0

b=2.1 t

~

J

J

i

A

%

36-

--

~

b=1.8

i

b=2.4 t

i

u

\

i

b=1,9

h

i

J

Carleton

36-

b=2.6 L

i

I

A

<_

2718-

b=2.0

b=2.0

b=2.3 i

J

A

1984

S

J

i

A

1985

b=2.1 i

i

S

J

< i

1986

S

b=1.8

b=1.7 i

i

A

I

S

J

A

1987

Fig. 3 Placopecten magellanicus. Change in mean scaled gonadal mass (9) and standard errors for scallops collected at Pabos, Gascons, Bonaventure and Carleton from July to October in the years 1984 to 1991. Horizontal bars show significant decreases in gonadal mass (spawning events), as indicated from Jonckheere tests (p < 0.05). Arrows indicate intervals when spawning was initiated. Significant decreases when gonadal mass fell to < 10 g were not considered spawning events. The b value for each year and site indicates slope used in calculating gonadal mass for a standard scallop measuring 125 mm in shell height (gonadal mass index x 125 b)

S

J

i

J

A

1988

1989

1990

i

A

S

1991

much as 45%. This was because additional gonadal production often occurred between spawnings. At all sites, reproductive output was greater in years with multiple spawnings than in years with a single spawning. Lastly, when the same years were compared, the total output at Bonaventure and Carleton was nearly 50% greater than at Gascons. Relation of spawning to abundance of organic matter

increasing gonadal mass, occurred more frequently at Gascons, in the outer bay (1985, 1986, 1987, 1988, 1989, 1990), whereas single annual spawnings were most common at Carleton, in the inner bay (1984, 1985, 1987 and 1988). The mean number of spawnings decreased from 2.0 yr- 1 at Gascons to 1.6 yr- 1 at Bonaventure and 1.3 yr-1 at Carleton. In years when the three principal sites were sampled, maximum gonadal mass was less at Gascons than at Bonaventure and Carleton (Table 1). Total reproductive output, as indicated by the sum of the decreases in gonadal mass in any given year, varied from 11 to 35 g. If reproductive output was calculated from the difference between gonadal mass before and after the spawning season, it would be underestimated by as

In 1985, chlorophyll a levels at the three sites were generally low and stable throughout the spawning period (Fig. 4). The seasonal development of the phytoplankton community was best illustrated by weekly measurements made at Gascons in 1989, when sampling began on 15 May (Fig. 5). Low levels characteristic of summer conditions began in late May. An increase to 1.7 gg 1- 1 was recorded for one date in June (26), but no spawning occurred (the scallops were possibly not ready to spawn at this time). Chlorophyll a levels may have increased somewhat during the second gonadal decrease but were low ( < 0.6 ggl-1) during the first and third. We only began measuring the POC and P O N concentration just prior to spawning in 1985 and levels

641 Table 1 Placopecten magellanicus. Maximum prespawning wet gonadal mass; number of spawnings (decreases in gonadal mass, Fig. 3); reproductive output as indicated by (1) difference between pre-spawning and post-spawning gonadal mass and (2) sum of successive decreases in gonadal mass (sum of spawnings); and increase in gonadal output obtained using latter calculation. All gonadal mass calculations based on estimates for a 125 mm scallop. Bold values indicate means for all years sampled

Site Year

Maximum gonadal mass (g)

Number of spawnings yr - 1

Reproductive output (g) Pre-spawn -post-spawn

Sum of spawnings

(%)

18.4

18.4

Pabos (23 m in depth) 1991 25.72

1

Gascons (25 m) 1984 20.38 1985 23.99 1986 22.29 1987 21.22 1988 25.39 1989 26.74 1990 21.71 1991 17.49

1 1 4 2 2* 3 2 1

12.7 18.4 13.9 14.7 18.8 20.2 16.0 t0.9

12.7 18.4 16.4 15.1 19.2 23.8 16.3 10.9

22.40

2.0

15.69

16.60

(16 m) 25.07 34.3 29.98 30.02 37.44 36.04 31.08

1 1 2 2 2* 2 1

19.5 27.6 20.2 22.6 29.6 27.4 23.8

19.5 27.6 25.2 28.9 35.5 30.4 23.8

32.04

1.6

24.39

27.26

Carleton (13 m) 1984 22.67 1985 34.19 1986 28.72 1987 30.82 1988 34.60 1991 30.43

1 1 3 1 1" 2

17.6 26.6 19.7 23.2 27.2 22.2

17.6 26.6 28.5 23.2 27.2 26.1

30.24

1.3

22.76

24.94

Bonaventure 1984 1985 1986 1987 1988 1990 1991

Increase in output

18.0 2.7 2.1 17.8 1.9

24.8 27.9 19.9 10.9

44.7

17.6

* Post-spawning gonadal mass (g) and output estimated assuming a minimum mass equal to the average values in other years at the same site (6.7 g at Gascons, 7.8 g at Bonaventure, and 7.4 g at Carleton)

were similar to those during the remainder of the summer (Fig. 4). Levels of P O N were similar at all three sites. In contrast, P O C concentrations were 3 to 4 times higher at Gascons than at the other sites. In the weekly P O C and P O N measurements at Gascons in 1989, increases were suggested during spawning (Fig. 5), however, these were probably not significant since they were within the range of daily variations recorded in 1985 (Fig. 4). Thus, there was little evidence suggesting that phytoplankton or organic substances stimulated spawning.

dom distributions were also obtained when the lunar phases were divided into 6 or 12 arcs). The lack of a relation of spawning to lunar or tidal cycles was also indicated by the variations in spawning times at different sites in any given year (Fig. 3), since lunar and tidal phases were the same throughout the bay. Synchronous spawning at different sites rarely occurred. A more frequent pattern was a progressive delay in spawning towards the inner bay.

Critical spawning temperature Relation to lunar and tidal phases Rayleigh tests showed no association of spawning events with either lunar or tidal phases. Tests applied to the circular distribution (Re = 0.278, ~b = 25 ~ p = 0.08) and to the axial model (R2o = 0.257, q5 = 135 ~ p = 0.12) both showed random distributions (Fig. 6). (Ran-

Although the spawning period, late July to early September, corresponded to the annual temperature maximum, temperatures above the scallop beds during spawning events were highly variable. The mean temperatures during spawnings (the mean calculated over the period when the first drop in gonadal mass occurred) varied considerably, from 4 to 16 ~ (Table 2,

642 ~.s-

:.~::

~!~t.,:

~i~

~!~

g~,..,,.~#.

!~:.,,~

~g,;N

~/.~

o~.

Bonaventure

G~.~con5 :~]

i~]~]

0.0

I

18 -

^

i

i

I

I

I

~!

~ "~ -:y. ::

.,'y ~.<,

N~

/~

I

I

I

,~)~

I

~

I

~

i

i

Ni

t

Aug

i

I

I

I

I

I

I

I

I

I

I

i

i

i

i

I

|

i

i

i

l

i

i

I

i

I

I

I

[

i

I

N

!

I

I

,

I

.:. ~&,.:,:

~ ! 1

1--

I

"~"

I

I

J

.

t

s~

.... ~

'

N~

I

,

i

;'*rlt'~

|

/ I

I

i

t

,

i

:~:~:~ !4

"

r,.a

i

I

.,,~

o

o

I

" i

I

.%:g~'.:;:

~o.

i

I

|

;.%:~

I

|

al/v

8-

:'..%':~1~,~ 9~ <

.....

-

i;~ "

~

Carleton

4.~;:~:i';;~.'~;.'4:;" :~

~;.~

&

:;" ,r

N@)N::N

I

I

I

I

I

l,.a

Aug

Fig. 4 Change in chlorophyll a concentration, bottom temperature, particulate organic carbon (POC) and nitrogen (PON) in the summer of 1985 at Gascons (25 m), Bonaventure (16 m) and Carleton (13 m). Shaded areas indicate periods when giant scallop spawning occurred as indicated by decreases in gonadal mass (from Fig. 3)

Fig. 7). Spawning was infrequent at the extremes in temperature, only once below 6 ~ (Gascons, 1986) and once above 14 ~ (Carleton, 1988). Our data suggested that spawning was not associated with temperatures attaining either the summer maximum or any particular fixed temperature. In only 4 of the 33 spawning events, for which temperature data were available (Table 2 and data for Pabos), the temperature at the time of spawning was the first time that level was attained. Thus, our data did not support the hypothesis that spawning was stimulated by a critical temperature. Cumulative spawning temperature Cumulative temperature at the time of spawning (sum of ~ d) varied among sites and with depth, because of the horizontal and vertical temperature gradients in the bay. Our most complete data were for the 1986 and

i

I

I

I

s~p

I

I

i

I

!

r~

i

I

I

I

^ug

I

I

S~p

i

I

1991 spawnings at Gascons and the 1986 spawning at Carleton, where we had continuous records from the gonadal minimum in October of the previous year. In these instances, the cumulative temperatures at the time of the first spawnings were 800, 773 and 1150 ~ respectively. At Carleton, the warmest and shallowest site, cumulative temperatures always attained higher levels earlier than at the other sites. That spawning was generally latest at Carleton contradicted the hypothesis that the time of spawning was determined by a cumulative temperature. Relation of spawning to temperature fluctuations Most spawning events (25 out of 33, for which temperature data were available) coincided with a rapid increase in temperature, caused by the downwelling of warm surface water (Fig. 7). This was documented for three spawnings in 1985 (one at each of three sites), four in 1986, four in 1987, four in 1988, three at Gascons in 1989 (although the downwelling was only weak during the second spawning), one in 1990 (the second at Bonaventure) and for five in 1991. During all of these spawnings, temperature increased markedly compared

643

~

2.0-

~~

1.0- ~ .

A

Gascons

2.5-

moon

@

~i~V:::,~1!:~ :~*-~ }~".:i~

I ~

::-"-'~:.':~ ~ N

0.50.0

New

ThUd ( ~ qua~er

9)

First quarter

,

O

Full m o o n

~ 2

B

~{!~

~::i! i~:,~!~!. . . . . . . . . .

-2

135o

~

Neap 45~

Spring tides

1200-

2250

315~

Placopec~en magellanicus. Relative frequency of spawning events for the period July to September in the years 1984 to !,991 (n = 34) in relation to A lunar and B tidal phases. Rayleigh tests were used to determine randomness of spawning events for (A) circular and (B) axial distributions Fig. 6

g

f,'...i.~.

['""P

I ~1

~0

I

I

I

I

!

I

I

I

I

I

I

200-

:~::

150-

~o-

r

~"i~,~

i::N

/ 21m I

May

t

[

I

I lun

l

I

I

~

~

Jul

I

'l'

!

I

1

t

I

I

Aug I s~v

I

l

I

I

t

o~

1

l

l

I

Fig, 5 C h a n g e in chlorophyll a concentration, b o t t o m temperature,

particulate organic carbon (POC) and nitrogen (PON) in the summer of 1989 at Gascons (24 m). Shaded areas indicate periods when spawning occurred as indicated by decreasesin gonadal mass (from Fig. 3)

to the mean 2 d earlier. We suspect that the 8 ~ rise in temperature at Carleton on 5 and 6 September 1984 initiated spawning. Although we did not obtain a temperature record for Bonaventure in 1986, a sharp temperature increase likely occurred during the abrupt 5 to 8 September spawning (Fig. 3). This was indicated by the rate of propagation (westward at ~ 1 d between sites, Bonardelli et al. 1993) of the downwelling event observed at Gascons in late August and at Carleton in early September. Conditions during seven additional spawnings, three in 1986 (the last at Gascons and the two August spawnings at Carleton), one in 1988 (the first at Gascons) and three in 1990 were different. There were strong (4 to 6 ~ temperature fluctuations around a high mean temperature (9 to 14 ~ The only spawning that occurred in conditions which were different

from the above two categories was at Carleton in 1987 (24 to 27 August). During spawning, temperatures were below 8 ~ and falling, with minor fluctuations. This gonadal decrease continued until the following sampling date (27 to 31 August), and this part of gamete release was during a downwelling event (this was not counted in the 25 spawning events above, because it was a successive spawning). Thus, out of 33 spawnings for which we had temperature records, 25 coincided with downwellings, 7 with fluctuations at high mean temperatures and 1 spawning began while temperatures were decreasing (Carleton, 1987). To examine whether temperature changes occurring during spawning differed from those found throughout the reproductive period (mid-July to early September), we made the following analysis based on spawning events that had been documented by sampling at frequent intervals (7 d or less). We first calculated the difference between the hourly temperature and the mean for the previous 48 h. This procedure selected for departures from the temperature to which the scallops were physiologically adapted. Then, we used a Kolmogorov-Smirnov test to compare the frequency distribution for the maximum temperature increases during all 48-h periods between dates when spawnings were recorded with the distribution for the maximum temperature increases for all 48-h periods throughout mid-July to early September. The distributions were different (Z2 = 26.7, p < 0.0001): temperature increases were greater while spawning occurred (Fig. 8). Although our data indicated a strong association of spawning with sharp temperature increases and,

644

Placopecten mageUanicus. M a x i m u m Table 2

t e m p e r a t u r e s (~ attained (1) before s p a w n i n g (period w h e n a drop in g o n a d a l m a s s was first detected a n d (2) d u r i n g spawning; a n d (3) m e a n t e m p e r a t u r e s d u r i n g spawning, for s p a w n i n g events at G a s c o n s , B o n a v e n t u r e a n d C a r l e t o n in the years I984 to I99i. E a c h line represents a s p a w n i n g event (decrease in g o n a d a l mass). *Indicates t e m p e r a t u r e at the time of s p a w n i n g was the first time it attained that level. M a x i m u m t e m p e r a t u r e s before a n d during, a n d m e a n t e m p e r a t u r e d u r i n g s p a w n i n g at P a b o s in 1991 were 12, 1 5 ' a n d 10 ~ respectively (Fig. 7)

Year

Gascons

Bonaventure

Max. before

Max. during .

Mean during .

Max. before

Carleton

Max. during

Mean during

Max. before

Max. during

Mean during

21

15

10

12

15

16"

10

.

12 I6 15 .

12 t5 t7"

11 t4 13

1984

.

1985

15

15

9

16

15

1986

11 i4 9 13

9 8 12 11

6 4 9 9

-

-

1987

14 14

13 12

9 10

16 18

15 11

13 7

18

8

7

1988

17 17

12 14

7 10

14 18

16" 17

9 13

21 -

16 -

16 -

1989

13 11 11

12 9 11

7 6 7

1990

16 15

14 14

10 8

1991

16 .

16 .

secondarily, with fluctuations around high mean temperatures, spawnings were not always recorded when downwelling conditions were first encountered during the summer. For example, spawning did not occur during the sharp downwelling event that took place in late July at Bonaventure and Carleton in 1985, during two downwellings in Juiy in 1986, during several downwellings at Carleton in 1988 and during an early August downwelling at Bonaventure in 1990. Downwellings were generally more frequent than the number of spawnings recorded (e.g. Bonaventure in 1988, Fig. 7). The lack of spawning during certain downwelling events may indicate that the scallops were incapable of spawning. That the ability to spawn varied was indicated by the spawning response of male scatlops subjected to 250 ppm hydrogen peroxide (Bonardelli 1994). This response was high at the gonadal maximum. During partial spawnings it fell markedly and was regained somewhat later. Salinity and current velocity Although salinity was strongly negatively correlated with temperature (r = - 0.9, p < 0_05; Bonardelli et al. 1993), it is unlikely that salinity triggered spawning. This is suggested because the salinity variations were low (27 to 33~o). During the recorded spawnings, the most abrupt change in salinity was only by 3Yooo.Current velocity and direction were strongly influenced by local topography. During spawning, tidal current velocity varied by a factor of five between Gascons and

.

. . -

11 .

.

. .

.

. .

.

.

.

.

.

-

-

19 15

17 16

14 12

-

-

-

18

17

13

18 16

16 12

12 9

.

Bonaventure, and mean residual currents varied tenfold (from 0.01 to 0.1 m s - l ; Bonardelli et al. 1993).

Discussion Giant scallop spawning controlled by external factors A number of observations in our present study indicated that spawning in Placopecten magellanicus was triggered by environmental cues. The first was its abruptness. We documented numerous complete spawnings within 1 wk and twowithin 2 to 3 d. Abrupt spawnings have also been documented in other locations (Dickie 1995; Posgay and Norman 1958; MacDonald and Thompson 1988; Parsons et al. 1992; DiBacco 1993). Our sampling at frequent intervals ( < 1 wk) revealed instances of repeated abrupt spawnings separated by periods of gonadal growth or stability. Histological analysis of female gonads after the first spawning on 3 August at Bonaventure (Fig. 3, 1987) revealed that mature oocytes were absent (Bonardelli unpublished). Several days were required for non-vitellogenic oocytes to reach maturity and during this process gonadal mass increased. Repeated spawnings have also been documented by Parsons et al. (1992). The variability of spawning in space and time (present study; Barber and Blake 1991; Parsons et al. 1992) further indicates control of spawning by external factors. The final evidence of external control of spawning was that, when scallops were collected just prior to

645 a spawning event and maintained in a closed sea-water system at various temperatures (6 to 7, 10 to 12 and 15 to 16 ~ massive spawnings did not occur when they were occurring in the field (Bonardelli 1994). Some scallops spawned sporadically at the higher temperatures when manipulated. Relation of spawning to environmental factors Spawning in Placopecten magellanicus did not coincide with increases in p h y t o p l a n k t o n or organic substances,

as reported for some marine invertebrates (Himmelm a n 1975; Starr et al. 1992, 1993). Spawning took place \

\ \, \

Fig. 7 Hou\rly temperatures as recorded by near-bottom currentmeters and thermographs during 1984 to 1991 at Pabos (23 m, current-meter at Grande Rivi+re), Gascons (25 m), Bonaventure (16 m) and Carleton (13 m) in the Baie des Chaleurs. Shaded areas indicate periods when giant scallop spawning occurred as indicated by decreases in gonadal mass (Fig. 3). Arrows indicate when spawning is believed to have been triggered

Ga~oons

iiiiEii!ii Bonaventure (..,1

2o

c,

:~iiii:ii:i:i:i:i:i:i)!i:iiii

[.-,

C~lelxm 20

=================================

~5

::i~:~i::i!i::ili::i~iNii::i::i::i::i':i~:i!i::i::i::i::iiiiii::iiii::i':::i

10

i ...... i::.......... iN~iiiiii)il

:::::::::::::::::::::::::::::::::::::::::::::::::::: 1984

,5

i,~ii~,i'~i~,iil

1985

i,,~ii i

1986

iI'~i!'~i'~i'~ iitliii',i'~i',i',

Bonaventure

~'

I

C~lelxln

0|'"'"l"'"

................... I

1987

iiii!~i. . . . . . . . . . . . . . . . . . . . . . I

I

1988

1989

Aug

I

646

Pabos 20-

60-

~ 50-

15-

2

5-

~50-

0 20~

iiii':iiiiiiiiii

10

ou

Gascons

~ " " ":

5 0 r

~20 ..................

10-

0-

":'i?:':':'~

0

......iii r

r

r

1~20-

r

r Bonaventure

fiiiiiiiiii',iiiii

15-

40.

10-

r

iii:::: iiiiii',iiii

105-

2

4 6 Temperature increases (~

8

10

Fig. 8 Placopecten magellanicus. Frequency distribution for temperature increases in 48-h periods (n = 118) during spawning events (filled bars) compared to temperature increases in 48-h periods throughout the summer reproductive period (mid-July to early September, n = 783) (shaded bars). For each hour within each 48-h period, the difference between hourly value and mean for the preceding 48-h period was calculated. Maximum of these values was temperature increase for that period

0 [ ~ Cafleton

Aug

20-1 is 10

"i"

Jul [ 1990

......~::: :::i!::!i~ "

i:~:i:i~i!:!:i:!:i:~ i:i:i~!~i~i ":ili

Aug 1991

Fig. 7 (Continued)

2 to 3 mo after the spring bloom, when the water column was strongly stratified. Associated with this were low chlorophyll a levels and a dominance of picoplankton ( < 5 gm cells) (Claereboudt et al. 1995). Further evidence that phytoplankton increases did not induce spawning was that scallops, collected on 10 August 1985 at Bonaventure, did not spawn when exposed to high diatom densities (40 x 107 cells of Phaeodactylum tricornutum 1-1) in the laboratory for a 5-d period (data of M. Starr reported by Bonardelli 1994). The fact that the natural population at Bonaventure began spawning while the phytoplankton tests were being performed in 1985 indicated that these scallops were ready to spawn. Our conclusion that spawning is not related to lunar or tidal phases contradicts that of Parsons et al. (1992). They indicate that spawning in Passamaquoddy Bay (45~ occurs just before spring tides and is associated with new and full moons. However, their data do show instances in which spawning does not occur at the same time at different sites in the same year, even though lunar and tidal phases were identical. For example, the gonadal decrease began later at Western Passage than at Navy Island in 1987 and later at Navy Island than at Hill's Point in 1988 (Figs. 4 and 7 in

Parsons et al. 1992). Tides are the dominant oceanographic force in Passamaquoddy Bay, and an alternative hypothesis is that spawning there is triggered by a factor associated with tides. Environmental data were not collected during their study. Whereas giant scallop spawning was a seasonal event and only occurred during the later part of summer when temperatures were between 4 and 16 ~ there was little evidence that spawning was related to either a critical or a cumulative temperature. The temperature during spawning varied among sites, among years and among spawnings when there were several spawnings in the same year. Spawning rarely occurred when the mean daily temperature was below 6 ~ or above 14 ~ The lack of spawning at low temperatures probably reflects physiological inactivity because of the temperature (Dickie 1958). In spite of the prevalence in the literature of the hypothesis that spawning is controlled by a critical temperature, the lack of a clear demonstration of this idea is surprising. All but one of the 33 spawnings, for which temperature records were available, coincided with marked changes in temperature. Twenty-five of these occurred during temperature increases associated with downwelling. Bonardelli et al. (1993) describe the mechanisms involved. During the summer the water column is strongly stratified with temperatures on the beds being cooler and undergoing greater fluctuations than at the surface. Low temperatures and high salinities on the beds correspond to periods of upwelling when deeper cold water is moving shoreward. At the same time the warmer surface layer moves offshore. The increase in bottom temperature (downwelling), which coincided with most spawnings, corresponded to a readjustment of the pycnocline caused by a decrease in wind stress or a change in wind direction.

647 Downwelling events (n = 90 from 1984 to 1991) were randomly distributed in relation to the lunar cycle (Rayleigh test, R2o = 0.102, p = 0.42). They were principally generated from strong winds towards the east at periods of 3 to 10 d, associated with the passage of cyclonic systems through the region. The strength of the local vertical temperature gradient largely determined the amplitude of temperature increases. Downwelling is initiated at the mouth of Baie des Chaleurs and propagated westward at a speed of ~0.44 ms -1 along the coast (Bonardelli et al. 1993). Given the 50-km distance between scallop beds, ~ 1 d would be required for a downwelling event to travel from one site to the next. This, we believe, is why many spawning events occurred progressively later toward the inner portion of the bay. In 1985 and 1987, spawning began first at Gascons and successively later into the bay (Fig. 7). A progressive delay was observed for the second spawning in 1988 (Fig. 3) and for the first spawning at the various locations in 1991. About 20% of spawnings did not coincide with strong temperature increases, but rather with strong temperature fluctuations (at tidal intervals) around a mean temperature of 9 to 14 ~ The sustained high temperature on the beds was maintained by persistent downwelling of warm surface water. The temperature changes during these spawnings were generally less than those during sharp rises from a low temperature. Possibly scallops spawned under these conditions because their sensitivity to temperature increases with increasing temperature. In Passamaquoddy Bay, where tidal amplitudes are an order of magnitude greater than in Baie des Chaleurs, giant scallops spawn when mean temperatures are ~ 12 to 14 ~ At this temperature level, smaller changes may be sufficient to provoke spawning. Changes of 1 to 3 ~ have been recorded near-surface (Gregory et al. 1988) and changes nearbottom are probably greater. Our study corroborated previous suggestions by Welch (1950) and Naidu (1970) that temperature changes or wind events are associated with giant scallop spawning. Skreslet (1973) studied the scallop Chlamys islandica and also suggested that spawning may coincide with temperature increases. Yamamoto (1951) stated that Patinopecten yessoensis spawned when warm water flowed into Mutsu Bay, Japan, and Minchin (1992b) noted that the mass spawnings of various invertebrates were associated with periods of onshore winds and higher temperatures. Reproductive strategy For a marine invertebrate which releases its gametes into the water, one criterion which is likely important in the evolutionary selection of an environmental spawning cue is its "distinctiveness". To ensure coordination of gamete release within a population, the individuals must be able to distinguish the signal from other

environmental variations. Fertilization success is only likely to be achieved if there is a high degree of spawning synchrony (Pennington 1985; Olive 1992; Sewell and Levitan 1992). During the summer in Baie des Chaleurs, the only environmental factor showing sudden changes is temperature (abundances of phytoplankton and organic particles were stable) and our data showed that virtually all spawning events coincided with periods when temperature changes were accentuated. Evidence of the importance of spawning synchrony is provided by an analysis of the population structure of scallops in Baie des Chaleurs from sampling in 1992 (Claereboudt and Himmelman 1996). The only years with strong cohorts were 1988 and 1990. These were years when spawnings in Baie des Chaleurs were abrupt, synchronous among sites, occurred in late August, and coincided with mean gonadal mass attaining near-maximum levels (Table 1). The years with the least recruitment were 1986 and 1989, when multiple spawnings occurred. Dickie (1955) and Langton et al. (1987) also proposed that year class success may correlate with spawning synchrony. Whereas one might expect that multiple spawnings, by increasing total output of gametes (Table 1) and by decreasing the risk that gametes be subjected to unfavourable conditions, would ultimately increase recruitment success, this does not appear to be the case. The year classes from the years when gamete release was spread over a prolonged period were poorly represented in the population. This is possibly explained by a decrease in fertilization success. Since spawning was associated with downwelling events, one would expect the east-west geographic variation in spawning frequency to be related to the frequency of downwelling events. Our data show that both the frequency of spawning and the frequency of downwelling events decreased toward the inner bay. Only about half of the downwelling events reached Carleton. When downwellings did occur at Carleton, the corresponding temperature change was generally more abrupt than at Gascons. This corroborates with more abrupt spawnings at Carleton. That giant scallops spawn during downwelling events means that their gametes are likely ejected into the warm water mass that is temporarily advected from the surface to the bottom. This should enhance fertilization success and larval survival, since fertilization success and early embryonic development are optimal at temperatures characteristic of the warm water mass (12 to 16 ~ Bonardelli unpublished). Dickie (1955) and Culliney (1974) similarly indicate that larval success depends on their being ejected into a narrow range of temperatures appropriate for larval growth and survival. Low temperatures retard development (Dickie 1955) and high temperatures are lethal ( > 19 ~ Culliney 1974). Tremblay and Sinclair (1988) found larvae at temperatures of 11 to 13 ~ in the Gulf of Maine, with peak concentrations at the thermocline. The

648

infrequency of spawning at temperatures above 14 ~ in the present study suggests that scallops inhibit gamete release when temperatures are unfavourably high. Although spawning occurred during the summer period of low phytoplankton abundance and domination by small cells (picoplankton), larval development was probably not limited by food availability. Raby et al. (1994) sampled bivalve larvae in a stratified and mixed water column for several weeks at Caplan (between Bonaventure and Carleton) and Grande Rivi6re (Fig. 1) following spawning and found that most larvae, including those of scallops, aggregated just below the thermocline, where temperatures were ~ 10 ~ and salinities ~ 29%o. The stomachs of scallop larvae were fuller than those of other bivalves and contained cyanobacteria and other picoplankton (1 to 15 gm in size). Scallop and other bivalve larvae were observed making rapid vertical migrations to richer food sources. In starvation experiments on scallop larvae, Raby (personal communication) found that 80% of the gut contents were digested within 2 to 3 h, yet scallop larvae collected in field samples were never observed with empty guts. Our values for phytoplankton concentration following spawning in 1985 and 1989 ( < 1 ~tg 1-1 Chl a), were similar to the values recorded by Raby et al. (1994). The timing of spawning of giant scallops is variable. It usually occurs in August to September in Newfoundland (Naidu 1970; MacDonald and Thompson 1988), in July to August in Baie des Chaleurs, Bay of Fundy (Beninger 1987) and Passamaquoddy Bay (Parsons et al. 1992) and in September to October on Georges Bank (MacKenzie et al. 1978; DiBacco 1993). Spawning is occasionally observed in May or June on Georges Bank (DiBacco 1993), in the mid-Atlantic Bight (Dupaul et al. 1989) and off New Jersey (MacDonald and Thompson 1988). The hypothesis that spawning is triggered by temperature changes (associated with downwelling events or tides) needs to be studied in other regions. Numerous factors (temperature, salinity, food availability, chemical substances, etc.) vary during downwelling events, and any factor associated with the change in water mass is a potential spawning cue. The hypothesis that temperature, rather than the other factors, triggers spawning is suggested by the spawnings which occurred during fluctuations around a high mean temperature. These spawnings occurred after downwelling conditions had been present for several days. There was no change in water mass. These conditions resembled those observed during spawning in tidally wellmixed areas such as the Bay of Fundy. Laboratory studies are required to determine the response of scallops acclimated to various temperatures and to different levels of temperature changes. In either spawning situation (a rapid increase or fluctuations in temperature), reproductive success is probably coupled with properties associated with the warm surface water mass.

Acknowledgements We are indebted to C. Cayouette and S. Naidu for support and encouragement at the onset of this project, to I. Lamontagne for advice on phytoplankton analysis, to J.-Y. Anctil of GIROQ for assistance with the mooring arrays, to M. Fr6chette for providing logistical assistance, to N. Hamel for analyses on the Elner Burner and to W. Petrie and R. Pettipas at Bedford Institute of Oceanography and technicians at McGill University for processing the current meter data. We further appreciated the aid of J.-S. Duguay, S. Bernier, D. Dorion, J. C6t6, M. Moisan, L. Michaud, R. No~l, M.-F. Dalcourt, J. Poirier and J. Gaudreault in diving to collect the scallops and of H. Gaudreault, P. Hasty, A. St-Jean and S. T6trault in making the gonadal mass determinations. Finally, the study greatly benefited from discussions and critical comments on the manuscript by J. Parsons, J. McNeil, J. Dodson, J. Bovet, M. Starr, R. Rochette, D. Arsenault, M. Claereboudt and K. Stokesbury. The project was supported by Canada Employment and Provincial grants to JCB at Aquatek Mariculture Inc. and through CRSAQ, NSERC and OPEN grants to JHH.

References Barber BJ, Blake NJ (1991) Reproductive physiology. In: Shumway SE (ed) Scallops: biology, ecology, and aquaculture. Developments in aquaculture and fisheries science. Vol. 21. Elsevier, Amsterdam, pp 377 428 Batschelet E (1981) Circular statistics in biology. Academic Press, London Beninger PG (1987) A qualitative and quantitative study of the reproductive cycle of the giant scallop, Placopecten magellanicus, in the Bay of Fundy (New Brunswick, Canada). Can J Zool 65:495-498 Bonardelli JC (1994) Relation entre la ponte chez le p6toncle g+ant, Placopecten magellanicus,et les facteurs environnementaux.PhD thesis, Universit~ Laval, Qu6bec Bonardelli JC, Drinkwater K, Himmelman JH (1993) Current variability and upwelhng along the north shore of Baie des Chaleurs. Atmosphere-Ocean 31:541-565 BonardeUi JC, Himmelman JH (1995) Examination of assumptions critical to body component indices: application to the giant scallop, Placopecten magellanicus. Can J Fish aquat Sciences 52(i1): (in press) Bricelj VM, Epp J, Malouf RE (1987) Intraspecific variation in reproductive and somatic growth cycles of bay scallops Argopecten irradians. Mar Ecot Prog Ser 36:123-137 Cap6ra/t P, Van Cutsen B (1988) M6thodes et mod6les en statistique non param6trique-expos6 fondamental. Presses de l'Universit6 Laval, Qu6bec Claereboudt MR, C6t6 J, Bonardelli JC, Himmelman JH (1995) Seasonal variation in abundance and size structure of phytoplankton in Baie des Chaleurs, southwestern Gulf of St. Lawrence, in relation to physical oceanographic conditions. Hydrobiologia (in press) Claereboudt MR, Himmelman JH (1996) Recruitment, growth and production of giant scallops (Placopecten magellanicus) along an environmental gradient in Bale des Chaleurs, eastern Canada. Mar Biol 124:661-670 C6t6, J, Himmelman JH, Claereboudt MR, Bonardelli JC (1993) Influence of density and depth on the growth of juvenile giant scallops (Placopecten magellanicus) in suspended culture. Can J Fish aquat Sciences 50:1857-1869 Culliney JL (1974) Larval development of the giant scallop Placopecten magellanicus (Gmelin). Biol Bull mar biol Lab, Woods Hole 147:321-332 DiBacco C (1993) Reproductive cycle of the giant scallop, Placopecten magellanicus (Gmelin), on northeastern Georges Bank. MSc thesis, Dalhousie University, Halifax

649 Dickie LM (1955) Fluctuations in abundance of the giant scallop, Placopecten magelIanicus (Gmelin), in the Digby area of the bay of Fun@. J Fish Res Bd Can 12:797-802 Dickie LM (1958) Effects of high temperature on survival of the giant scallop. J Fish Res Bd Can 15:1189-1211 Dupaul WD, Kirkley JE, Schmitzer AC (1989) Evidence of a semiannual reproductive cycle for the sea scallop, Placopecten magelIanicus (Gruelin, 1791), in the mid-Atlantic region J ShellfishRes 8:173-178 GaltsoffPS (1961) Physiology of reproduction in molluscs. Am Zool 1:273 289 Gregory DN, Verge E, Dobson D, Smith C (1988) Long-term temperature monitoring program 1987 Scotia-Fundy, Gulf of St. Lawrence and Newfoundland. Can Data Rep Hydrogr Ocean Sciences No. 65 Hauenschild C (1960) Lunar periodicity. Cold Spring Harb Symp quant Biol 25:491-497 Himmelman JH (1975) Plankton as a stimulus for spawning in three marine invertebrates. J exp mar Biol Ecol 20:199-214 Himmelman JH (1981) Synchronization of spawning in marine invertebrates by phytoplankton. In: Clark WH, Adams TS (eds) Advances in invertebrate reproduction. Vol. II. Elsevier NorthHolland, New York NY, pp 3-19 Holm-Hansen O, Lorenzen CJ, Holmes RN, Strickland JS (1965) Fluorometric determination of chlorophyll. J Cons int Explor Mer 30:3-15 Kennedy B, Pearse JS (1975) Lunar synchronization of the monthly reproductive rhythm in the sea urchin Centrostephanus coronatus Verril. J exp mar Biol Ecol 17:323-331 Korringa P (1947) Relation between the moon and periodicity in the breeding of marine animals. Ecol Monogr 17:347-351 Korringa P (1957) Water temperature and breeding throughout the geographical range of Ostrea edulis. Colloque international de biologie marine, Station biologique de Roscoff, Ann6e biologique 33:1-17 Langton RW, Robinson WR, Schick D (1987) Fecundity and reproductive effort of sea scallops Placopecten rnagellanicus from the Gulf of Maine. Mar Ecol Prog Ser 37:19-25 Loosanoff VL (1969) Maturation of gonads of oysters, Crassostrea virginica, of different geographic areas subjected to relatively low temperature. Veliger 11:153-163 Loosanoff VL, Davies HC (1950) Conditioning V. mercenaria for spawning in winter and breeding its larvae in the laboratory. Biol Bull mar biol Lab, Woods Hole 98:60-65 Loosanoff VL, Davies HC (1952) Temperature requirements for maturation of gonads of northern oysters. Biol Bull mar biol Lab, Woods Hole 103:80-96 Loosanoff VL, Davies HC (1963) Rearing of bivalve mollusks. In: Russel FS (ed) Advances in marine biology 1. Academic Press, London, pp 1 136 MacDonald BA, Thompson RJ (1988) Intraspecific variation in growth and reproduction in latitudinally differentiated populations of the giant scallop Placopecten magellanicus (Gmelin). Biol Bull mar biol Lab, Woods Hole 175:361-371 MacKenzie CLJ, Merril AS, Serchuk FM (1978) Sea scallop resources off the northwestern United States coast, 1975. Mar Fish Rev 40:19 23 Minchin D (1987) Sea-water temperature and spawning behaviour in the seastar Marthasterias glacialis. Mar Biol 95:139 143 Minchin D (1992a) Induced spawning of the scallop, Pecten maximus, in the sea. Aquaculture 101:187-190 Minchin D (1992b) Multiple species, mass spawning events in an Irish sea lough: the effect of temperatures on spawning and recruitment of invertebrates. Invert Reprod Dev 22:229-238 Naidu KS (1970) Reproduction and breeding cycle of the giant scallop Placopecten magellanicus (Gmelin) in Port au Port Bay, Newfoundland. Can J Zool 48:1003-1012 Newell RIE, Hilbish TJ, Koehn RK, Newell CJ (1982) Temporal variation in the reproductive cycle of Mytitus eduIis L. (Bivalvia, Mytilidae) from localities on the east coast of the United States. Biol Bull mar biol Lab, Woods Hole 162:299-310

Olive PJW (1992) The adaptive significance of seasonal reproduction in marine invertebrates: the importance of distinguishing between models. Economic and applied aspects of invertebrate reproduction. Proc 6th int Congr Invert Repr, Trinity College, Dublin, Ireland, Balaban 22 (1-3): 165-174 Orton JH (1920) Sea-temperature, breeding and distribution in marine animals. J mar biol Ass UK 12:339-366 Parsons G J, Robinson SMC, Chandler RA, Davidson LA, Lanteigne M, Dadswell MJ (1992) Intra-annual and long-term patterns in the reproductive cycle of giant scallops Placopecten magettanicus (Bivalvia: Pectinidae) from Passamaquoddy Bay, New Brunswick, Canada. Mar Ecol Prog Set 80:203-214 Pearse JS (1975) Lunar reproductive rhythms in sea urchins. A review. J interdiscip Cycle Res 6:47 52 Pennington JT (1985) The ecology of fertilization of echinoid eggs: the consequences of sperm dilution, adult aggregation, and synchronous spawning. Biol Bull mar biol Lab, Woods Hole 69:417-430 Posgay JA, Norman KD (1958) An observation on the spawning of the sea scallop, Placopecten magellanicus (Gmelin), on Georges Bank. Limnol Oceanogr 3:p 478 Raby D, Lagadeuc Y, Dodson JJ, Mingelbier M (1994) Relationship between feeding and vertical distribution of bivalve larvae in stratified and mixed waters. Mar Ecol Prog Ser 103:275-284 Robinson WE, Wehling WE, Morse PM, McLeod GC (1981) Seasonal changes in soft body component indices and energy reserves in the Atlantic deep-sea scallop, Placopecten magellanicus. Fish Bull US 79:449-458 Sastry AN (1963) Reproduction of the bay scallop, Aequipecten irradians Lamarck, influence of temperature on maturation and spawning. Biol Bull mar biol Lab, Woods Hole 125:146 153 Sastry AN (1970) Reproductive physiological variation in latitudinally separated populations of the bay scallop, Aequipecten irradians Lamarck. Biol Bull mar biol Lab, Woods Hole 138:56-65 Sewell MA, Levitan DR (1992) Fertilization success during a natural spawning of the dendrochirote sea cucumber Cucumaria miniata. Bull mar Sci 51:161-166 Skreslet S (1973) Spawning in Chlamys islandica (O.F. Miiller) in relation to temperature variations caused by vernal meltwater discharge. Astarte 6:9-14 Skreslet S, Brun E (1969) On the reproduction of Chlamys islandica (O.F. Miiller) and its relation to depth and temperature. Astarte 2:1-6 Starr M, Himmelman JH, Therriault J-C (1990) Direct coupling of marine invertebrate spawning with phytoplankton blooms. Science, NY 247:1071-1074 Starr M, Himmelman JH, Therriault J-C (1992) Isolation properties of a substance from the diatom Phaeodactylum tricornutum which induces spawning in the sea urchin Strongylocentrotus droebachiensis. Mar Ecol Prog Ser 79:275-287 Starr M, Himmelman JH, Therriault J-C (1993) Environmental control of green sea urchin, Strongylocentrotus droebachiensis, spawning in the St. Lawrence Estuary. Can J Fish aquat Sciences 50:894-901 Strickland JDH, Parsons TR (1972) A practical handbook of seawater analysis. Bull Fish Res Bd Can 167:1-311 Thorson G (1950) Reproductive and larval ecology of marine bottom invertebrates. Biol Rev 25:1-45 Tremblay MJ, Sinclair MM (1988) The vertical and horizontal distribution of sea scallop (Placopecten magelIanicus) larvae in the Bay of Fundy in 1984 and 1985. J northw Attant fish Sci 8:43-53 Welch WR (1950) Growth and spawning characteristics of the sea scallop, Placopecten magellanicus (Gmelin), in Maine coastal waters. MSc thesis, University of Maine, Orono Yamamoto G (1951) Ecological study on the spawning of the scallop Pecten (Patinopecten) yessoensis in Mutsu Bay. Bull Jap Soc scient Fish 17:53-56 Yentsch SS, Menzel DW (1963) A method for determination of phytoplankton chlorophyll and phaeophytin by fluorescence. Deep Sea Res 10:221-231