,1 , 5' - i

1

¥

: i" \*' % :

■ : ' .

h

■

i . ■ . ■ ! ■ " ■ ■ ■ I ,: i

■■■.■•■'.

!

I

i

.

i

!

■

| h K 6, If

, «. s. %: l- . I I I i i

» *• I i » • » > s

M i, i(j i't|

a

■' ' ' : I | . ' . I

■ ■'■ i | :

. . p ■ •

. - ft :.

iiii

1 tf 4 4 <*

■ -•4 i it

i it & '* v

■ •:

- - . -■ ■

i

' '

. ■

. • v ••

.'/

• III I

:•

■ * ft :

•

Xv!

■

CONTENTS

List of Abstracts by Author of Technical 1 Presented at 1976 NSA Annual Meet

Michael Castagna and John N. Kraeuter Mercenaria Culture Using Stone Agg Predator Selection

Robert E. Malouf and Wilbur P. Breese Food Consumption and Growth of L Pacific Oyster, Crassostrea gigas (Th Constant Flow Rearing System

Kwang H. Im and Don Langmo

Economic Analysis of Producing Pac Seed in Hatcheries

Joseph G. Loesch and John W. Ropes

Assessment of Surf Clam Stocks in N along the Delmarva Peninsula and in Virginia Fishery South of Cape Henr

Victor G.Burrell, Jr.

Mortalities of Oysters and Hard Clar with Heavy Runoff in the Santee Rivi South Carolina in the Spring of 1975

Raymond J. Rhodes, Willis J. Keith, Peter Burrell, Jr.

An Empirical Evaluation of the Leslie Applied to Estimating Hard Clam, M Abundance in the Santee River, Sout

George E. Krantz and Donald W. Meritt An Analysis of Trends in Oyster Spa Maryland Portion of the Chesapeake

Gary H. Cole, Ronald L. Copp, and Davi Estimation of Lobster Population Siz Point, Connecticut, By Mark-Recapt 1975-1976

■

■ ». »

	.'
								.

PROCEEDINGS

NATIONAL SHELLFISHERIES ASSOCIATION

OCT 1 7 1978

VOL. 68

PROCEEDINGS

OF THE

NATIONALSHELLFISHERIES ASSOCIATION

OFFICIAL PUBLICATION OF THE NATIONAL

SHELLFISHERIES ASSOCIATION;

AN ANNUAL JOURNAL DEVOTED TO

SHELLFISHERY BIOLOGY

VOLUME 68

Published for the National Shellfisheries Association, Inc. by The Memorial Press Group, Plyniouth, Massachusetts

JUNE 1978

PROCEEDINGS

OF THE

NATIONAL

SHELLFISHERIES

ASSOCIATION

CONTENTS Volume 68 - June 1978

List of Abstracts by Author of Technical Papers Presented at 1977 NSA Annual Meeting, Hunt Valley, Maryland, and NSA Coast Section, Tumwater, Washington v

Gordon Gunter and W. David Burke

Further Notes on How Oysters Land When Planted 1

Frederic Lipschultz and George Krantz

An Analysis of Oyster Hatchery Production of Cultched and Cultchless

Oysters Utilizing Linear Programming Optimation Techniques 5

Herbert Hidu, William G. Valleau and Fletcher P. Veitch

Gregarious Setting in European Oysters - Response to Surface

Chemistry vs. Waterbone Pheromones 11

Richard A. Lutz and Herbert Hidu

Some Observations on the Occurrence of Pearls in the Blue Mussel,

Mytilus edulis L 17

George E. Krantz and John F. Chamberlin

Blue Crab Predation on Cultchless Oyster Spat 38

Jack M. Whetstone and Arnold G. Eversole

Predation on Hard Clams, Mercenaries mercenaries by Mud Crabs,

Panopeus herbstii 42

Paul A. Haefner, Jr.

Seasonal Aspects of the Biology, Distribution and Relative Abundance of the Deep-Sea Red Crab, Geryon quinquedens Smith, in the Vicinity of the Norfolk Canyon, Western North Atlantic 49

Hubert J. Squires and G. Riveros

Fishery Biology of Spiny Lobster (Panulirus argus) of the Guajira

Peninsula of Colombia, South America, 1969-1970 63

Abstracts

NSA Annual Meeting

NSA Pacific Coast Section 75

HI

LIST OF ABSTRACTS BY AUTHOR OF TECHNICAL PAPERS

PRESENTED AT THE 1977 NSA ANNUAL MEETING

HUNT VALLEY, MARYLAND

Erik Baqueiro and Craig B. Kensler

Mexican Molluscan Fisheries of the Gulf of Mexico and the

Caribbean: 1970-1975 75

Robert S. Brown

A Disease Survey of New England Soft-Shell Clams, Mya arenaria 75

Robert S. Brown and Carole J. OToole

Histochemical Analyses of Pigment Accumulations in Mercenaries

mercenaria L. and Mya arenaria L 76

H. Arnold Carr

Culture of Hatchery-Spawned Mercenaria mercenaria in Massachusetts 76

Melbourne R. Carriker

Ultrastructural Evidence that Gastropods Swallow Shell Rasped

During Hole Boring 76

Keith Cooper, Sammy Ray and Jerry Neff

The Interaction of Water Soluble Fractions (WSF) of South Louisiana

Crude Oil and Dermocystidium (Labyrinthomyxa) marina at Varying

Temperatures in the American Oyster, Crassostrea virginica Gmelin 77

N. Dean Dey and Ellis T. Bolton

Tetracycline as a Bivalve Shell Marker 77

William J. Eckmayer

Oyster Reef Cultivation for Cultch Material 77

Arnold G. Eversole

Marking Clams with Rubidium 78

RegGillmor

Suspension Culture of European Oysters (Ostrea edulis L.) 78

Reg Gillmor

Growth Responses of European and American Oysters (Ostrea edulis L

and Crassostrea virginica G.) to Intertidal Exposure 79

VI

Harold H. Haskin

The Onshore Surf Clam Resource along the Southern New Jersey Coast 79

Harold H. Haskin and Susan E. Ford

Mortality Patterns and Disease Resistance in Delaware Bay Oysters 80

Dexter S. Haven, William J. Hargis, Jr. and Paul C. Kendall

The Oyster Industry of Virginia 1931 to 1975 80

Mary T. Hickey

Age, Growth, Reproduction and Distribution of the Bay Scallop,

Aequipecten irradians irradians (Lamarck), in Three Embayments

of Eastern Long Island, New York as Related to the Fishery 80

Herbert Hidu, Richard D. Clime and Samuel R. Chapman

Oyster Setting - Evolution of Commercial Hatchery Technique 81

Sewell H. Hopkins

An Oyster Family Tree: Ancestry of Crassostrea virginica 81

Louis Leibovitz

Bateriologic Studies of Long Island Shellfish Hatcheries 81

Louis Leibovitz and John Hamlin Gordon II

Water Quality Studies of Long Island Shellfish Hatcheries 81

Richard A. Lutz

A Comparison of Hinge-Line Morphogenesis in Larval Shells of Mytilus edulis L. and Modiolus modiolus (L.) 83

K. S.Naidu

Culture of the Sea Scallop, Placopecten magellanicus in Newfoundland 83

Bruce J. Neilson

Engineering Considerations in the Design of Oyster Depuration Plants 84

George S. Noyes, Harold H. Haskin and Cindy Van Dover

Oil and the Oyster in Delaware Bay 84

Linda Plunket

The role of Uronema marinum (Protozoa) in Oyster Hatchery Production 85

Edwin E. Rhodes and Ronald Goldberg

The Use of Pumped Raceway Systems for the Intermediate Grow-Out

of Hatchery Reared Bivalves 85

R. J. Rhodes, M. Wolff and J.L. Music

Status Report on the Commercial Blue Crab Fishery of

the Carolinas and Georgia 85

VII

Scott E.Siddal

The Development of the Hinge Line in Tropical Mussel Larvae

of the Genus Perna .. . . . 86

Robert S. Steneck, Richard A. Lutz and Robert M. Cerrato

Age and Morphometric Variation in Subtidal Populations of Mussels 86

ABSTRACTS OF TECHNICAL PAPERS PRESENTED AT NSA 1977 WEST COAST SECTION MEETING, TUMWATER, WASHINGTON

John H. Beattie, William K. Hershberger, Kenneth K. Chew and Conrad Mahnken

Breeding Disease Resistance into the Pacific Oyster 88

Rick D. Cardwell

Oyster Larvae Mortality in South Puget Sound 88

Darrell Demory

Helicopter Crabbing 89

Richard A. Eissinger

Disease Control in a Molluscan Shellfish Hatchery 89

Jim Glock

Squaxin Island Manila Clam Reseeding Studies 90

William K. Hershberger

Oyster Breeding: Where Can It Go 90

Chris R. Jones

A Comparison of Survival, Growth, and Yield of Pacific Oysters

(Crassostrea gigas) from Different Sources 90

Mark Miller, Kenneth Chew, Charles D. Magoon, Lynn Goodwin and Chris Jones Preliminary Report on Manila Clam Reseeding Program at Five Puget Sound Beaches 91

Thomas L. Richards

Seed Oyster Production in the Salton Sea, California 91

M. A. Toner

Preliminary Studies on the Development of a Synthesized Diet for

Juvenile Oysters, Crassostrea gigas 91

VI11

J. R. Vanderhorst and P. Wilkinson

Clam Resource Measurement for Estimation of Pollution Damage 91

Paul Waterstrat

Musseling in on a New Market 91

Anthony Weaver

Pigeon Point Shellfish Hatchery: Past, Present and Future 93

Ronald Thurber Zebal

Vibrio anguillarum and Larval Mortality in a California Coastal

Shellfish Hatchery 93

Cover

Pearls in the blue mussel, Mytilus edulis.

Photo by R. A. Lutz (See page 17).

Proceedings of the National Shellfisheries Association Volume 68-1978

FURTHER NOTES ON HOW OYSTERS LAND WHEN PLANTED1

Gordon Gunterand W. David Burke

GULF COAST RESEARCH LABORATORY OCEAN SPRINGS, MISSISSIPPI 39564

ABSTRACT

Woodward, Huxley and Moebius saw that oysters always set on the left valve. Later Stafford, Prytherch and Ranson observed cementation by the left prodissoconch (lar- val shell). However Cunningham and Ortou stated that Ostrea edulis often come to lie on the right valve. Apparently this is of little consequence on hard bottom, but in mud the situation is considered to be harmful for reasons summarized by Needier. Gunter and McGraw found that oysters culled from clumps landed upside down when planted in water 59% of the time. This is attributed to poor shape of the oysters.

Natural single oysters numbering 1,423, which had been dredged from wild reefs, were selected or culled to singles and shovelled overboard into fresh water 1.5 to 3 meters deep. Oysters were seen to stabilize in about the first half meter when thrown into the water and then fall unchanged, upside down or not, to the bottom.

Out of 4,269 shovel-tosses of the 1,423 oysters (3 each) 2,447 (57%) landed on the left valve and 1,822 (43% ) landed on the right valve. In three replications of the experi- ment the preponderance of the total sample came to rest normally on the left or lower valve, and of the total 488 individuals (34%) consistently settled normally whereas on- ly 108 individuals regularly came to rest on the dextral valve.

INTRODUCTION AND REVIEW

In an attempt to analyze some of the aspects of oyster culture a little closer, the writers have con- tinued work on how single oysters land on the bottom when thrown overboard. This appears to be a simple and overly pedestrian matter but it has an interesting history in the literature, with some examples of how disagreements and misunder- standings can occur over simple matters.

The early conchologists showed that the oysters come to rest on the left or convex valve by attach- ment (Cf. Woodward 1878), and generally lie that way, contrary to Anomia and Pecten which rest on the right valve. Nevertheless, Cunningham (1885) of the Scottish Marine Station disputed this view and said that most oysters in the Firth of

Presented at 1976 Annual Meeting of the National Shell- fisheries Association.

Forth rest on the right valve and are heavily fouled only on the upper (left) valve. This was quickly disputed by Hunt (1885), who said that all oysters attach by their convex (left) valves as stated by Woodward, Jeffrey and Huxley. Moebius (1885) quickly affirmed that oysters set with the nucleus of the left valve attached to the substrate. It is now well known through the observations of Huxley (1883), Stafford (1910) and Prytherch (1934) that the larvae always cement the left valve down. The "nucleus" of Moebius is the left valve of the pro- dissoconch or late larval shell stage, which Ran- son, (1960) found attached even in fossil oysters. Thus, how oysters set and how they come to lie on the bottom became confused, at least in the minds of some people.

And so it would appear that the natural position of oysters was well settled years ago. Even so, modern workers have caused further discussion

G. GUNTER AND W.D. BURKE

and according to Orton (1937, p. 14), "In life the flat or right valve usually rests on the sea-bottom and is often referred to as the lower one." This caused Galtsoff (1964, p. 16) to say that "Orton's (1937) statement... is an obvious oversight." But Galtsoff overlooked Orton's deferred explanation on p. 27 which says, "When oysters are thrown in the water they usually settle with the flat valve on top. In this position the shell is liable to be washed about by any water-current and eventually to top- ple over to rest on its flat side." In short Orton ex- plains the observations of Cunningham (1885) while verifying the general ideas and observations of zoologists concerning setting on the left valve.

With regard to the turning over of the European oyster, Ostrea edulis L., by currents as observed by Cunningham (1885) and Orton (1937), we call attention again to the observations of Emery (1°68) on the same phenomenon regarding empty pelecypod shells on the continental shelf.

How Oysters Come to Lie When Planted in the Water

Orton (1^37, p. 27) said merely in passing that when oysters are thrown into the water they usually settle with the flat (right) valve on top.

Needier (1938) applied himself to this question and came to several conclusions, some of which confirmed our observations independently, be- cause we never acquired his paper until the last moment. They may be summarized as follows: The normal position of an oyster on firm bot- tom is with the curved (left valve) down. On soft bottom they sink by the hinge end with the "lip" up. In upside down oysters the "lip" grows upward, producing distortion. For the best shape oysters should lie in the normal position. Oysters smother more readily when upside

down. About 95% of well-shaped oysters landed right side up in water over three feet deep. The same percentage righted themselves within a foot or two when let in the water upside down. Crooked oysters more often land upside down. Loose shells or other oysters caused oysters to turn over. Water helps in positioning oysters and it is better to plant with a tide than on bare ground.

Galtsoff (1964, p. 16) touched upon this ques- tion with his statement, "In C. virginica the left valve is almost always thicker and heavier than the right one. When oysters of this species are dumped from the deck of a boat and fall through water they come to rest on their left valves. I observed this many times while planting either small oysters not greater than 2 inches in height, or marketable adults of 5 to 6 inches."

Other authors seem to have used estimates. Gunter and McGraw (1974) seem to be the only workers who have presented quantitative data on this question. They found in Mississippi that wild reef oysters separated from clusters and culled into singles landed right side up only 41% of the time. However, the oysters used in these trials were essentially misshapen from having been grown together and lingering doubts over these results caused us to re-examine the situation. We might say here briefly and quickly that we have come to the conclusion that these results are an example of what Needier was talking about when he said, "Crooked oysters more often land upside down."

PRESENT WORK AND RESULTS

The writers carried on a further series of simple experiments using 1,423 oysters of which 937 were seed oysters under 50 mm in length. These were collected from natural reefs in Mississippi Sound by dredge and most of them were selected as singles, but some were culled. They were shovel- led overboard into clear fresh water in swimming pools 1.5 to 3 meters in depth. Observation was then made by scuba diving and picking up all the oysters that landed right side up. Then the upside down oysters were collected and both sets counted. This whole process was repeated three times. The results are given in Table 1 .

The departure of these results from a predic- table norm based on mere chance is given in Table 2. Actually when this is calculated as the prob- ability function of a binomial distribution, the chances that the type of distribution shown here would occur naturally intrinsically in a 50-50 distribution is less than one in a billion. Translated from statistical terms to the vernacular, this means that there is practically no chance that these results are matters of chance.

HOW OYSTERS LAND

TABLE 1. Resti)ig positions of 1423 "planted" oysters during each of three consecutive trials.

▼ = At rest on right valve A= At rest on left valve

Total Oysters

Run I.

Run. II.

Run III.

1423

108V	16t>A	101T	155A	126Y	488A	125V	154A
39%	61%	39%	61%	20%	80%	44%	56%

TABLE 2. Comparison of calculated and observed orientation of 1423 "planted" oysters.

Dextral Sinistral

(Chi Right Valve) (On Left Valve)

Calculated Observed Calculated Observed

712 530 712 893

712 553 712 870

712 460 712 963

Run I. Run II. Run III.

Totals

2136

1543

2136

2726

DISCUSSION AND CONCLUSIONS This all means, as several writers have stated before, that well-shaped natural oysters do have a very strong tendency to land right side up when planted single in the water. One might say that well-shaped oysters are fitted to fall through the water in a way that is most likely to lead to sur- vival after the oyster strikes the bottom. We might wonder what circumstances throughout the ages would cause a natural oyster to fall through the water and what would cause these helpless in- vertebrates to be so attuned to the environment in this respect. But whatever the situation, they are obviously among the most successful marine animals on Earth being present in all continents except Antartica, at depths from the surface to

two kilometers, and in such abundance that they leave organic remains such as the Point au Fer Reef in Atchafalaya Bay in Louisiana which was 30 miles long in its heyday. As relicts of one single animal these are probably the largest conglomera- tions on Earth, for the sometimes larger coral reefs are formed from the remains of many species of animals and plants.

Observations made by the scuba divers (W. D. Burke, Ron Lukens and lohn DeMond) showed that oysters thrown overboard with a shovel stabilized themselves within two feet and then continued to fall in the same way to the bottom. This confirms the previous statement of Needier who implies that oysters were held in various posi- tions at the surface and then released. Looked at another way, our results indicate that well-shaped oysters have a 36% better chance of survival on soft mud where they would die less quickly than misshapen oysters which might fall wrong 59% of the time as indicated by Gunter and McGraw (1974).

Examination of Table 1 shows that oysters fall- ing through the water in the proper stance, so to speak, have a tendency to do so on repeated trials, whereas oysters that fall upside down have a similar tendency to continue to fall upside down

G. GUNTER AND W. D. BURKE

with repeated trials. This is the conclusion we would expect if the shape of the oysters modifies its manner of falling through the water.

The matter of how well oysters survive on various bottoms when they fall upside down or right side up is another question and although at first blush it might seem to be simple, easy and clearcut, it is a bit more complicated than it first appears. Be that as it may, we hope to treat that subject in a subsequent study.

ACKNOWLEDGEMENTS

We thank Mr. Ronald Lukens and Mr. John De- Mond for assisting us with the diving and Mrs. C. E. Rasor and Mr. Anthony Becker for the use of their swimming pools.

LITERATURE CITED

Cunningham, T. J. 1885. The resting position of

oysters. Nature, London 32:597. Galtsoff, P. S. 1964. The American oyster

Crassostrea virginica Gmelin. U. S. Fish Wildl.

Service. Fish. Bull. 64:1-480. Gunter, G. and K. A. McGraw. 1974. Basic

studies on oyster culture 1. How do single

oysters land on the bottom when planted? Proc. Nat'l Shellfish. Assoc. 64:122-123.

Hunt, A. R. 1885. The resting position of oysters. Nature, London 33:8.

Huxley, T. H. 1883. Oysters and oyster question. Eng. Illustrated Mag. London 1:47-55, 112-121.

Moebius, Karl. 1885. The resting position of oysters. Nature, London 33:52.

Needier, A. W. H. 1938. How do oysters land on the bottom under water? Fish. Res. Bd. Can., Oyster Farming Circ. No. 1.

Orton, 1. H. 1937. Oyster biology and oyster- culture being the Buckland Lectures for 1935. 211 pp., 57 fig. Edward Arnold and Co. Lon- don.

Prytherch, H. F. 1934. The role of copper in the setting, metamorphosis and distribution of the American oyster, Ostrca virginica. Ecological Monographs 4(1): 47-107.

Ranson, G. 1960. Les prodissoconques (coquilles larvaires) des Ostreides vivants. Bull. Inst. Oceanogr. 1(183):1-41.

Stafford, J. 1910. The larva and spat of the Cana- dian oyster. II. The spat. Amer. Nat. 44:343-366.

Woodward, S. P. 1878. Manual of Mollusca. 3rd ed. London.

Proceedings of the National Shellfisheries Association Volume 68-1978

AN ANALYSIS OF OYSTER HATCHERY PRODUCTION

OF CULTCHED AND CULTCHLESS OYSTERS

UTILIZING LINEAR PROGRAMMING TECHNIQUES

Fredric Lipschnltz * and George Krantz

UNIVERSITY OF MARYLAND

CENTER FOR ESTUARINE AND ENVIRONMENTAL STUDIES

CAMBRIDGE, MARYLAND 21613

ABSTRACT

Manpower and operational requirements for cultched and cultchless oyster pro- duction schedules in a large-scale hatchery were compared using a system of linear equations and a computer optimization program. The matrix of equations defined the operational sequence within the oyster hatchery, the resource requirements, and restrictions, if any. The optimization program minimized a cost objective function within the constraints defined by the system matrix.

Data for the calculations of the matrix coefficients were taken from records of the University of Maryland small-scale hatchery and from Dupuy (1973). The hatchery data provided estimates of manpower requirements for each activity, oyster mor- tality rates, and equipment costs. Dupuy's paper provided space and density re- quirements. The temporal sequence of oyster development stages was based on well- documented literature and observations at the model hatchery.

Results show that labor was the major cost component in all types of hatchery schedules. The optimal solution involved purchase of large amounts of equipment which remained idle most of the year, being fully utilized in two pulses during the year. Constant maximal use of equipment required less equipment but more labor and therefore increased production costs. Use of the cultched mode of hatchery operation as opposed to the cultchless resulted in approximately 45% savings in pro- duction costs.

This study represents the first phase of a long- term project to optimize production scale oyster hatchery operations. Several problem areas indicated by this model will be investigated and changes incoporated into a revised formulation.

INTRODUCTION cultched spat to cultchless spat; and (3) test the This paper describes a mathematical model of design of a production scale oyster hatchery just the operation of a commercial oyster hatchery. completed at the UMCEES Horn Point En- Through linear programming (LP) techniques the vironmental Laboratories Cambridge, Maryland, model has been used to: (1) determine an optimal The goal of this paper is to demonstrate the utility production schedule and equipment mix for oyster of LP as a tool for economic analyses in produc- hatcheries, (2) compare hatchery production of tion aquaculture.

•Present Address: Ecosystems Center. Marine Biological LineaI" Programming is a mathematical tech-

Laboratory. Woods Hole, Mass. 02543. nique which determines optimal levels of activity

F. LIPSCHULTZ AND G. KRANTZ

within given resource restrictions, using well- documented theory (Gass, 1969) and readily available computer programs. As an analytical tool, linear programming has been used in a varie- ty of biological applications (Arroyo, 1965; Davis, 1967; Swartzman, 1973.) The model form is a matrix of linear equations describing the pro- duction or use of restricted resources. The use of linear equations assumes all functions modelled to be linear, thereby precluding economies of scale and exponential cost curves. Since matrices of this form generally have a large number of possible solutions, an additional equation, the objective function, can be defined which is maximized within the given set of possible solutions to the matrix. For example, an objective function could be a series of terms relating the profit in all of the activity levels, thereby determining the most pro- fitable solution consistent with the imposed con- straints.

METHODOLOGY The temporal sequence used in the model con- sisted of a 360-day year divided into a 300-day production period, a 30-day maintenance period, and 30 days of vacation. The oyster life-cycle was divided into 15-day periods to consolidate the model activity into units closely approximating life stages of oysters in the hatchery culture pro- cedure. The life-cycle activities and their durations (Table 1) for cultchless hatchery operation are based on observations at the University of Maryland Horn Point hatchery. Each of the life- cycle activities could be initiated in any of 20

TABLE 1. Definitions and Durations of Life-cycle Activities Available for Initiation during the Pro- duction Year.

Duration

No. Activity (Periods)

1	Grow 1 carboy of algae	1
2	Condition 1 oyster	3
3	Fill 1 Larval Cone
	(600,000 larvae/cone)	1
4	Set and harden 1 Mylar sheet
	of oysters	2
5	Fill 1 trough with oysters for
	final hardening	4
6	Heat water to condition oyster for 1 period	1

periods, although the developmental biology in the oyster life-cycle obviously precluded certain choices such as initiating spat set prior to the con- ditioning and larval stages.

Listed in Table 2 is a set of activities through which the model could purchase resources, such as labor and equipment, necessary to conduct the hatchery operations. Initiation of these activities (i.e. purchasing of resources) took place prior to the production year and permitted use of the pur- chased items throughout the year. This restriction precluded purchase of additional equipment or temporary labor during the production year. Labor to install purchased items is included in the cost.

TABLE 2. Purchasing Activities and Associated Costs Initiated Prior to the Production Year.

Cost U.S. Activity Dollars

Hire labor to work 8 hrs/day
for 15 days	10,000
Buy 1 carboy	15
Buy 1 Larval cone	250
Buy 1 spat tank and set of
associated Mylar trays	1,180
Buy 1 sq. ft. of hardening trough	5
Buy 1 conditioning tank
(9 sq.ft.)	100

The resource equations for a period in mid-year are shown in Table 3. The matrix coefficients are the levels of the resource (manpower, equipment or energy) required per unit of hatchery activity. For example, growing a carboy of algae (activity 1) would require 1 carboy, 30 hours of labor and produce 55.8 liters of algae with a mean cell con- centration of 107 cells/ml. The negative sign in- dicates production as is illustrated in the following equation:

-55.8X! +84X2 +240X3 + 75X4 < 0

Addition of 55.8 to both sides produces an equation which limits utilization of algal food to less than or equal to the algal production. Produc- tion in successive periods is linked by use of this convention. Thus, larvae initiated in period 5 and grown for 15 days would add to the supply of spat for use in the period 6 activity of hardening the spat on Mylar sheets.

ANALYSIS OF HATCHERY PRODUCTION

TABLE 3. Matrix Coefficients for Cultchless Hatchery Production Activity Initiated During a Mid-Year Period.

Resource

No. of carboys

Sq. ft. conditioning space

No. of larval cones

No. of spat tanks

Sq. ft. of trough

Labor (hours)

Liters of food supply (Algal)

Larval supply (Individuals)

1st hardening spat supply

(Individuals) Final hardening spat supply

(Individuals) Hot water (gpm) Oysters (Individuals)

X,

x2

Activity X, X.

.25

.004

				1
30		65.25	39	.625
•55.8	84 —60,000	240 600,000 —240,000	75 120,000 —60,000	1

-30

— 1

X, = Grow 1 carboy of algae X2 = Condition 1 oyster X3 = Grow larvae

X4 = 1st hardening X5 = Final hardening X6 = Heat water

The coefficients in Table 3 were calculated from the records of a small scale oyster hatchery that was operated for one year at Horn Point En- vironmental Laboratories. Labor coefficients were developed on the assumptions of U.S. union labor (i.e. 40 hr work week and 30 day annual leave). The design of a cultchless hatchery and its re- quisite space, oyster density requirements, and flow rates for water were taken from Dupuy (1973). The costs for purchase of equipment sup- plies and expendible materials are actual figures for (1976) construction of a pilot production hat- chery currently in operation at Horn Point.

A cost objective function was chosen for this model utilizing the costs for purchasing the resources listed in Table 2. The life-cycle activities also had associated costs, though they were small compared to the costs of materials, labor, and energy.

RESULTS

The initial computer run of the model was made to optimize the production schedule and minimize the amount of labor, space, and equipment re- quired to produce 50 million oysters in the 10- month cultchless production period. The resource

requirements and production schedule to meet this constraint are shown in Figure 1. Since the initial high level of conditioning activity (labelled A), propagates through subsequent life stages, fully utilizing the purchased resources, new condition- ing activity (labelled B,C,D) during the first 10 periods is either low or non-existent. After the pulse of activity passes out of the system, (e.g. oyster spat are planted on natural bottom), a new high level of activity (labelled E) is initiated.

A typical pattern of equipment utilization shows an initial maximal use followed by several periods where the equipment is largely idle. This seemingly non-optimal solution on closer ex- amination was found to be soundly based. Since labor to use the equipment was by far the greatest cost element in the objective function, the model utilized a mobile labor force shifting from equip- ment types as it followed the oyster life-cycle.

It is possible within the LP framework to restrict the level of an activity within a given range, thereby obtaining a solution in which the hatchery equipment was more fully utilized in all periods. The oyster conditioning was therefore constrained to a level of activity currently used in the HPEL pilot production hatchery. This restriction limited

F. LIPSCHULTZ AND G. KRANTZ

1 —

2 ....

3 -- 4

NUMBER OF CONDITIONED OYSTERS NUMBER OF LARVAL CONES NUMBER OF SPAT TANKS SQUARE FEET OF TROUGH

4 3 2 1

2 5,000 | 20pOO	500 400	250 • 200;
1 5,00 0	300	150 :
10,0 00	200	ioo :
5,000	100	so :

1000

500

2 I- -+ I

C

B

2

H I-

0 2 4 6 8 10 11 14

FIFTEEN DAY PERIODS

FIGURE 1. Time sequence and activity levels for 50 million oysters/year.

20

1500	L
V X X
1/1
> § 1200 CC <	\	v-
u		A

o

1/1 a:

LU CD

Z)

900

600

300o

0

UNCONST RA I NED CONSTRAINED

1

15 16

3 5 7 9 11 13

FIFTEEN DAY PERIODS FIGURE 2. Carboy utilization for algal culture.

all subsequent life stages and activities. Figure 2 ed is less for the constrained mode and the number

compares the time sequence of a particular of periods in full use is longer than the un-

resource for the constrained and unconstrained constrained mode. Table 4 lists all of the equip-

modes of operation. The number of carboys utiliz- ment and labor needs for these two possible hat-

ANALYSIS OF HATCHERY PRODUCTION

TABLE 4. Resource Requirements for Constrained and Unconstrained Brood Stock in a Theoretical Hatchery Design to Produce 50 Million Oysters Per Year by the Cultchless Hatchery Process.

Resource

Not Constrained

Constrained Brood Stock

Cost

($1000) Carboys

(#) Condition

(sq.ft.) Cones

(#)

Spat Tanks

(#) Troughs

(#) Manpower

2,688

1,469

27

195

407

1,017 291

2,968

1,531

23

80

191

995 376

chery production schedules. The constrained solu- tion costs more because of the increased labor costs associated with conducting several opera- tions simultaneously as opposed to shifting the labor force from activity to activity as the produc- tion pulse passes through the system.

The cultched mode of hatchery operation was modeled by removing the first hardening activity (X4) and putting all newly set spat directly into the hardening troughs. The number of periods re- quired for maturation was increased to 90 days (6 periods). Table 5 compares the results of this run with the requirements for production of 50 million oysters by the cultchless production method. The total cost for production of spat on cultch was 44% less than by the cultchless method partially because of space economies and greater survival after setting, but mostly as a result of the lower labor requirements.

Finally, the cultched and cultchless modes were compared using the actual equipment and space available for the operation of a full scale produc- tion hatchery at Horn Point. An unrealistic assumption allowed the hatchery program to hire unlimited labor, though the cost of the labor to conduct the various specific procedures remained constant. An oyster selling activity ($100/1000 oysters) (10c each) was added to replace the pro-

duction quota as a motivating force for the model. The results in Table 6 indicate a non-optimal pur- chase of equipment when the hatchery was de- signed. The non-starred equipment remained idle, limited by the starred resource. The model output permits calculation of the percent of idle equip-

TABLE 5. Resource Requirements for Comparison of Cultchless and Cultched Hatchery Procedures. Production of 50 Million Oysters Per Year.

Resource

Cultchless

Cultched

Cost

($1000) Carboys

(#)

Condition (sq. ft.) Cones

(#)

Spat Tanks

(#) Troughs

(#) Manpower

2,688

1,469

27

195

407

1,017 291

1,503

753

14

100

1,042 195

TABLE 6. Comparison of Resources in HPEL Hatchery Under Cultched and Cultchless Opera- tion Using a Profit Objective Function.

HPEL Hatchery Design Resource Cultchless Cultched

Profit
($1000)	175.4	366.9
Carboys
(#)	103	101
Condition
(sq.ft.)	15	17
Cones
(#)	6	13
Spat Tanks
(#)	12*
Troughs
(#)	60	140*
Manpower	37	45
Oysters
(X1000)	4,320	6,720

"Limiting Constraints

10

F. LIPSCHULTZ AND G. KRANTZ

ment thereby providing guidance for the optimal hatchery design. Within the given resource restric- tions the cultched mode again was more efficient than the clutchless mode producing more oysters (48%) than the cultchless mode.

CRITIQUE OF THE MODEL

This model represents the first phase of a long term project to determine the economic and biological feasibility of using hatchery-raised oysters to sustain public and private oyster fisheries. The modeling effort, in conjunction with existing conventional biological research pro- grams, has demonstrated some uses as a tool in determining the future of aquaculture in the Maryland portion of Chesapeake Bay. These uses include satisfying the initial objectives of optimiz- ing a production schedule and equipment mix for oyster hatchery production, pointing out con- trasts between cultched and cultchless hatchery operation and identifying potential production bottlenecks due to design errors. Beyond these im- portant products, modeling efforts can benefit research by clearly and formally organizing data on hatchery operation, and indicating where carefully planned new research or technical im- provements could be of greatest benefit.

Analysis of the output from this initial modeling effort suggests several areas which need more refined equations and/or more data from the ac- tual operation of a production scale shellfish hatchery which produces both cultched and cultchless oyster spat. Operational manpower and production calculations were based on figures from a laboratory scale hatchery which set only 10 million spat in 1976. A great economy of scale which the linear program framework cannot model, is expected in the operation of a larger facility in 1977 which should produce 100 million spat during its first year of operation. Therefore,

during 1977, the HPEL facility is being devoted to collection of the types of data necessary for the reformulation of the model. The reformulated model in addition to performing tasks illustrated in this report can then be used as a predictive tool, allowing production modifications to be tested without disrupting hatchery operation.

ACKNOWLEDGEMENTS

This project was jointly supported by the Uni- versity of Maryland and by the Fisheries Admini- stration of the Maryland Department of Natural Resources.

Current studies at HPEL are also being support- ed by the NOAA Office of Sea Grant.

Computer time and facilities were provided by the University of Maryland Computer Service Center.

We wish to thank Dr. Filmore Bender, Depart- ment of Agricultural and Resource Economics for his constructive advice, technical assistance, and review of the manuscript.

BIBLIOGRAPHY

Arroyo, G.J. 1962. Fish Pond Development Plan- ning with the help of Linear Programming. FAO Fisheries Technical Paper 21p.

Davis, L.S. 1967. Dynamic Programming for Deer Management Planning. J. Wildlife Mgmt. 31(4):667-679.

Dupuy, J.L. 1973. Translation of mariculture research into a commercial oyster seed hatch- ery. Proc. World Mariculture Soc. p677-685.

Gass, S.I. 1969. Linear Programming Methods and Applications. 3rd ed. McGraw-Hill Book Co.

Swartzman, G.L. 1973. Dynamic Programming Approach to Optimal Grazing Strategies using a succession Model for a Tropical Grassland. Appl. Ecol. ll(2):537-548.

Proceedings of the National Shellfisheries Association Volume 68-1978

GREGARIOUS SETTING IN EUROPEAN AND

AMERICAN OYSTERS — RESPONSE TO

SURFACE CHEMISTRY vs. WATERBORNE PHEROMONES1

Herbert Hidu, William G. Valleau, and Fletcher P. Veitcli

HERBERT HIDU DEPARTMENT OF OCEANOGRAPHY

IRA C. DARLING CENTER

UNIVERSITY OF MAINE AT ORONO

WALPOLE, MAINE 04573

WILLIAM G. VALLEAU

DEPARTMENT OF ZOOLOGY

UNIVERSITY OF MAINE AT ORONO

ORONO, MAINE 04473

FLETCHER P. VEITCH

DEPARTMENT OF CHEMISTRY

UNIVERSITY OF MARYLAND

COLLEGE PARK, MARYLAND 20742

ABSTRACT Action of a waterborne pheromone in gregarious setting was demonstrated by ex- posing European oyster larvae to extrapallial fluid (EPF) prior to and during exposure to cultch shell. Both treatments resulted in significant increase in larval setting over controls, indicating that the setting response can be released without contact with a treated cultch surface. Cross reactivity of EPF between American and European oysters was also indicated. Additional experiments should investigate the role of the pheromone in planktonic movements of larvae.

INTRODUCTION 1965, 1974; Bayne, 1969). Various test materials, In 1949, Cole and Knight-Jones discovered Le" fractionated oyster tissue extracts and ex- "gregarious setting" in the European oyster trapalhal fluid (oyster shell liquor), they state, Ostrea edulis L. larvae. The presence of were effective only when applied to surfaces, postmetamorphic oysters stimulated, in some When test chemicals were administered in suspen- manner, the setting of larvae near the site The s.on or solution there was no observable modifica- authors surmised that a waterborne material tion of the behavior of the larvae and a change in released by the juveniles stimulated the mature Settmg rateS dld not result- They sPeculate that it larvae to set. Since that time British workers have WOuld be imProbable that a larva could respond held the strong opinion that gregarious setting is t0 a waterborne material because of very low con- mediated by the direct contact of larvae with com- centratloris or the material in all but the immediate pounds adhered to the setting surface (Crisp, SOurce area (boundry laVer theory). Further, they

hypothesize that a small animal, such as an oyster

1 IraC. Darling Center Contribution No. 119 larva, could not sense a concentration gradient

11

12

H. HIDU, W.G. VALLEAU AND F.P. VEITCH

and that their swimming speeds are so low that they could not effectively approach the source (Crisp, 1965).

Our developing information with the American oyster (Crassostrea virginica), however, indicates the action of a waterborne pheromone (Hidu, 1969; Keck et al., 1970; Veitch and Hidu, 1971). Spat enclosed in bags of plankton mesh too small for passage of larvae stimulated setting on cultch outside the bags. Addition of extrapallial fluid (EPF) or seawater which had contained adult oysters significantly increased setting rates within one to two hours. Of course, there is the possibili- ty that the administered materials became adhered to the cultch surfaces and the larvae responded to surface contact after all, as suggested by Crisp (1974).

To really understand and manipulate recruit- ment in oysters it is important that we determine the mechanisms of action in gregarious setting, i.e., response to surface attached molecules and/or response to waterborne materials. A set- ting response released only after surface contact would mean that mature larvae are benthic and undergo extensive random "searching" of the bot- tom until the "proper" substrate is encountered. With the relatively low current velocities just at the bottom and feeble swimming speeds of larvae, it is difficult to conceive how a surface contact response alone could be of significant benefit in in- creasing the efficiency of setting of larvae. Rather, the response would be highly adaptive if pelagic mature larvae could sense the presence of a large adult population and then become benthic and set in the immediate area. Indeed Welch (1930) and Crisp (1965) suggested such a mechanism; the response to a dissolved chemical being coupled to a second sense, i.e., a light response. Photoposi- tive larvae would turn photonegative, thus caus- ing the larvae to reach a desirable setting area. This behavior, possibly coupled with the response to specific surface chemistry, would be more ad- vantageous to the setting larvae. Thorson (1964) suggested that oyster larvae were photopositive throughout their larval life, but then turned photonegative and set in response to environmen- tal stimuli such as increased temperatures en- countered in the intertidal zone. The sudden ap- pearance of the larval eyespot just prior to setting

is highly suggestive that this receptor is functional in setting in some manner.

In the present experiments, European oyster lar- vae were exposed to extrapallial fluid applied in suspension prior to and during exposure to cultch shells in order to demonstrate the release of a set- ting response through the waterborne action of the setting pheromone. Further, the cross reactivity of EPF of American and European oysters has been determined for these species.

This study was supported in part by National Science Foundation Grant No. GA-28741 and by NOAA, Office of Sea Grant, Project No. NG-40-72.

MATERIALS AND METHODS

All tests determining effects of EPF on setting were conducted in a similar manner. European and American oyster larvae were first reared by methods described by Walne (1966) and Loosanoff and Davis (1963).

Trials were run with eyed larvae which had in- itiated setting in large 400-liter culture vessels. Larvae were concentrated from the larger vessels and then aliquoted into a series of 1-liter beackers which each contained five bay scallop, Ae- quipecten irradians, shells as cultch. Each beaker contained approximately 5000 larvae with five to ten replicates per treatment. Twenty-five mis of EPF was added to 1-liter of larval suspension in the appropriate series. The experiment was then held for one to two hours to allow significant numbers of larvae to attach to the shell cultch. All vessels were then emptied at the same time and the cultch shells were gently dipped in clear seawater to remove larvae from the shells that had not set. This process, no doubt, removed some newly set larvae; however, all repetitions were carried out in a uniform fashion to permit a valid measurement of the effect of the treatment on setting. Cultch shells were dried and the total number of larvae that set were counted under a binocular micro- scope. The results were analyzed by calculating the 95% confidence limits of the mean, ±2 SE„,, or by subjecting results to a single classification ANOVA where appropriate (Steel and Torrie, 1960).

Extrapallial fluid, intra-and interspecies trials. The effects on setting of EPF administered in suspen-

GREGARIOUS SETTING OF OYSTERS

13

sion, was determined first at the intraspecies level and then at the interspecies level. With American oysters, an initial series of 12 experiments tested the effects of American oyster EPF by comparing setting rates with untreated controls. Later, a single experiment determined the effects of Euro- pean oyster EPF on American oyster setting by comparing its performance with American oyster EPF and untreated controls. Similarly, with Euro- pean oysters, 7 experiments were performed to determine the effects of intraspecies EPF and two with interspecies EPF. Significant differences in oyster set between control and EPF treated cultures within each experiment were determined by a "t" test (Steel and Torrie, 1960).

American oyster EPF was procurred by first holding in-season scrubbed oysters out of water for two to three hours. The shells were then gently pried open with an oyster knife and the EPF removed. This procedure generally produced several mis. of a rather viscous yellow fluid with a protein concentration of 90 to 100 ug/ml. This procedure was not possible with European oysters because this species retains less EPF and prying of the fragile shells resulted in damage to our valuable brood stock animals. An alternate method was devised, that of allowing 36 air-dried oysters to pump in a single 3-liter volume of sea water for two hours. Oysters readily transferred EPF to this medium; however, the protein concen- trations were much lower, approximating 20 ug/ml.

Extrapallial fluid — action of waterborne vs. at- tached chemicals.

We felt that waterborne action could be further demonstrated if prior exposure of oyster larvae to EPF in suspension resulted in increased setting rates after transfer to experimental cultures con- taining only seawater and cultch, but no EPF. A series of four experiments was thus set up to deter- mine setting rates for European oyster larvae which were pretreated in two ways (Figure 4). The first group was held in a 60-liter vessel with the ad- dition of 6 liters of EPF from European oysters (procurred as above). Larvae were stirred to pre- vent prolonged contact with container surfaces. After 10 minutes, larvae were decanted onto a sieve, thoroughly washed with filtered seawater and placed in a second 60-liter vessel made up to

the same volume with EPF-free seawater. A se- cond 60-liter vessel with larvae served as a con- trol. Concentrations of larvae in the vessels ap- proximated 5000 per liter.

Three experimental treatments were initiated, i.e.,

a. Control larvae with no EPF (10 reps)

b. Control larvae with 25 mis EPF added to experimental cultures

c. Experimental larvae pretreated with EPF but with no addition of EPF to experimental cultures. (EPF released).

Results from each of the four repetitive ex- periments were analyzed with a single classifica- tion ANOVA. Significant differences among treatments were determined with an lsd test (Steel and Torrie, 1960).

RESULTS

Intraspecies response to EPF. Extensive trials with American and European oysters indicate that there is a very rapid response to EPF administered in suspension (Figure 1). One to two hours of ex- European oyster larvae

H r-

©O© ©

©

©

European oyster ex <r opolhal fluid

0

0 30 60 90 120 150

Total Set per Experiment American oyster larvae

H h

H h

©@® @© © © I

American Oy

r a pa 1 1 ia • i u - J

■12 ' i ' I

© © ®

©

-no odd.t m

40 60 80 100

Total Set per Experiment

FIGURE 1. Setting of European and American oyster larvae in response to intraspecies extrapal-

14

H. HIDU, W.G. VALLEAU AND F.P. VEITCH

Hal fluid administered in suspension. Seven ex- periments were run with European oyster larvae and 12 with American oysters, with ten replicate cultures in each experiment. "T" tests revealed significant differences in all experiments at least at the 95% level.

posure resulted in over twice the setting in ex- perimental cultures. Gross examination revealed that setting behavior is activated within 10 minutes of application of EPF. Shells extracted from experimental cultures at this time had greater numbers of larvae loosely attached in a swim- crawl phase preparatory to setting.

Additional trials with both species indicated a possible interspecific response, although the results were variable for unknown reasons. With

European oyster larvae Expt.

1 1 Cor	trol -	1	ropol	trapol ioi f I	■ al f id	Uld
1
i American oysler ei no additive
0 I 2 1		5 6 7 8 9 10 i Set per Culture	12	13

European oyster larvae Expt. 2

1	1 1 *			, 1 _, 1 Europeon oyster extrapal Amencon oyster ex trapaiiiai fluid I - no additive	ial	luid
1
0			■:.	6 10 12 14 16 16 20 22 24 Set per Culture	ZG	28

FIGURE 2. Setting of European oyster larvae in response to extrapallial fluid of American and European oysters. Points indicate replicate cul- tures within each experiment. Mean values of rep- licates and 95% confidence limits (±2 SEm) are indicated.

European oyster larvae, for example, a first trial with American oyster EPF (Figure 2) resulted in a very significant increase in setting rates. A second trial, however, resulted in a very little increase over untreated controls. A preliminary trial with American oyster larvae indicated increased setting in response to European oyster EPF (Figure 3) .

American oyster larvae

				1 ,	ial r	Uld
4).	Cont			1 Europeon oyster enlropol 1
ro		1 American oytler eitrapoliial tluid no additive
0 2	4		6	B 10 12 14 16 IB 20 22 24 Set per Culture	26	26	30

FIGURE 3. Setting of American oyster larvae in response to extrapallial fluid of American and European oysters. Points indicate replicate cul- tures with mean value and 95% confidence limits indicated.

Exposure of European oyster larvae to EPF prior to exposure to cultch shells resulted in significant increases in setting, demonstrating a waterborne action of the pheromone (Figure 4). Two of four repetitions of the experiment resulted in significant differences between treatments, with controls be- ing significantly lower than the groups exposed to EPF and rinsed, which were in turn lower than the groups exposed to EPF during the exposure to cultch shell. Two other experiments, although they showed similar numerical trends showed no overall significant differences as revealed by the ANOVA.

Thus, oysters of both species are immediately stimulated to set in response to EPF administered in suspension. The response appears to be in- terspecific. The action of a waterborne material is indicated with European oysters because the larval setting response was released by exposure to suspended EPF prior to exposure to cultch shells.

GREGARIOUS SETTING OF OYSTERS

15

	EUROPEAN	OYSTER SET
Expt.	Control		EPF Released		EPF
1	51	<	83	<	121 **
2	(Total of 10 cultural 100	)	141	=	130 NS
3	60	=	99	=	124 NS
4	7	<	114	<	** 414
Total	218		437		789

OtSTER EXTRA PA LLIAL FLUID

HOLD LARVAE 10 MIN

CONTROL WATER WITH TREATED LARV*

E XTRAPAL LI A L FLUID

,,.0 b",ebb LJ»"> Q»'o

^ ' COUNT SET

FOUR EXPTS

FIGURE 4. Setting of European oyster larvae in response to intraspecies extrapalhal fluid adminis- tered prior to and during exposure to cultch shells. An appropriate ANOVA analyzed significance of differences in each of four experiments.

DISCUSSION

These experiments add evidence that oyster set- ting is initiated through action of waterborne chemicals. Thus there appears to be an additional component to the gregarious setting response not generally recognized by the British workers. However, the mechanism of action of waterborne materials is obscure, again by reasons stated previously. The attractive option is that water- borne pheromones present in the vicinity of very large concentrations of oysters act to draw larvae from the plankton by modifying their photo- or geotaxis. American oyster populations, in the American mid-Atlantic region at least, in their un- disturbed state are, or were, concentrated on huge bars of many acres surrounded by vast expanses

of mud bottom. It would be quite expedient for larvae to "recognize" these areas, drop from the plankton, and begin exploring the bottom, with then at least some chance of encountering a favorable attachment surface.

We have tried to measure change in larval geo- and phototaxis by administering EPF under condi- tions of light and dark in small-scale water col- umns. Results were interesting but inconclusive because of inadequate experimental apparatus. Needless to say, it would be very instructive to demonstrate positive geotaxis or negative phototaxis which is released by the presence of waterborne chemicals from oysters. Realistic-sized experimental water columns with precise control of environmental factors would be a necessity.

These experiments indicated considerable cross reactivity of EPF between Crassostrea and Ostrea, and this suggests that chemicals emitted from oyster populations may modify recruitment rates in other species; that is, act as kairomones as described by Kittredge et al (1974). It is not hard to imagine the adaptive value of such evolution; predators may locate prey (Welch, 1930; Crisp, 1965); larvae of epifaunal species may detect a substrate favorable for epifauna; finally the most intriguing possibility, interspecies setting re- sponses may be the biological basis for the establishment of many marine benthic com- munities. The British (Meadows and Campbell, 1972) appear to have considerable evidence for in- terspecies response for many groups but tend to dismiss its importance in favor of the dominant in- traspecies response. Much remains to be ac- complished in determining the ecological significance of the chemical senses in marine meroplankton, but their potential diversity of form and function cannot be disputed at this point.

LITERATURE CITED

Bayne, B. L. 1969. The gregarious behavior of the larvae of Ostrea edulis L. at settlement. J. Mar. Biol. Assoc. U.K. 49:327-356.

Cole, H. A. and E. W. Knight-Jones. 1949. The setting behavior of larvae of the European oyster, Ostrea edulis L., and its influence on methods of cultivation and spat collection. Fish. Invest., Lond. Ser. II 17(3):39 pp.

16

H. HIDU, W.G. VALLEAU AND F.P. VEITCH

Crisp, D. J. 1965. Surface chemistry, a factor in the settlement of marine invertebrate larvae. Botanica Gothoburgensia III, Proc. of the Fifth Marine Biological Symposium, Goteburg, 1965, pp. 51-65.

Crisp, D.J. 1974. Factors influencing the settle- ment of marine invertebrate larvae. In: P. T. Grant and A. M. Mackie (eds.), Chemorecep- tion in Marine Organisms, Academic Press, London, pp. 177-265.

Hidu, H. 1969. Gregarious setting in the American oyster, Crassostrea virginica Gmelin. Chesa- peake Sci. 10(2):85-92.

Keck, R., D. Maurer, C. Kauer, and W.A. Shep- pard. 1970. Chemical stimulants affecting larval settlement in the American oyster. Proc. Nat. Shellfish. Assoc. 61:24-28

Kittredge, J.S., F.T. Takahashi, J. Lindsey and R. Lasker. 1974. Chemical signals in the sea: marine allelochemics and evolution. Fishery Bull. 72(1):1-11.

Loosanoff, V. L. and H. C. Davis. 1963. Rearing of bivalve mollusks. In: F. S. Russell (ed.), Ad-

vances in Marine Biology, Academic Press, London, 1:1-136.

Meadows, P.S. and J. I. Campbell. 1972. Habitat selection by aquatic invertebrates. In: F.S. Russel (ed.), Advances in Marine Biology, Academic Press, London, 10:271-382.

Steel, R. G. D. and J. H. Torrie. 1960. Principles and Procedures of Statistics. McGraw-Hill Book Co., Inc. New York. 481 pp.

Thorson, G. 1964. Light as an ecological factor in the dispersal and settlement of larvae of marine bottom invertebrates. Ophelia 1:167-208.

Veitch, F. P. and H. Hidu. 1971. Gregarious set- ting in the American oyster, Crassostrea virginica Gmelin. I. Properties of a partially purified "setting factor". Chesapeake Sci. 12(3):173-178.

Walne, P. R. 1966. Experiments in the large scale culture of the larvae of Ostrea edulis L. Fish. In- vest. Lond., Ser. II 25(4):l-53.

Welch, J. H. 1930. Reversal of phototropism in a parasitic water mite. Biol. Bull. Woods Hole 59:165-169.

Proceedings of the National Shellfisheries Association Volume 68-1978

SOME OBSERVATIONS ON THE OCCURRENCE OF PEARLS IN THE BLUE MUSSEL,

MYTILUS EDULIS L.'

Richard A. Lutz2 and Herbert Hidu

DEPARTMENT OF OCEANOGRAPHY UNIVERSITY OF MAINE WALPOLE, MAINE 04573

ABSTRACT

Pearl incidence in raft and shore populations of mussels (Mytilus edulis L.) of similar lengths was quantitatively compared in four separate experiments in the Damariscotta River and Boothbay Harbor, Maine. Highly significant differences be- tween the two population types were found in each of the experiments, with fewer and smaller pearls occurring in raft-based individuals. A positive correlation was observed between the number of pearls in mussels of similar lengths sampled from each of the rafted populations and the amount of time the rafted substrate had been in the water. The relationship between number of pearls per mussel and age was qua)itified for one rafted and one shore population of mussels in South Bristol, Maine. No significant age- independent differences were found between the two populations with regard to the quantity of pearls present. Quantification of differences in pearl incidence between geographically isolated Mytilus edulis populations should facilitate correlation of the presence and quantity of pearls with the presence and abundance of the definitive and/or intermediate host(s) of the digenetic trematode parasite reputedly responsible for the initiation of pearl formation.

INTRODUCTION isolated pearls from specimens of M. edulis

According to Giard (1907), the first documented sampled from certain areas along the Gaspe' coast

account of the presence of pearls in mussels of Canada. More recently, Scattergood and

(Mytilus edulis L.) was given by Olaus Worm in Taylor (1949) have reported pearls in various

1655 upon examination of mussels taken near mussel populations along the northeast coast of

Copenhagen. Since that time, various workers the United States from Eastport, Maine, to Cape

have commented on the presence of pearls in Cod, Massachusetts. In all of the studies above,

isolated M. edulis populations. European in- considerable variation in both number and size of

vestigators have reported pearl-infested mussels in pearls with geographical location has been

a number of localities along the coasts of Den- reported. Such variable incidence has resulted in

mark, France, and England (Garner, 1857; d'Ham- the designation of distinct pearl-producing areas

onville, 1894; Dubois, 1901, 1909; Jameson, 1902; and is of considerable biological interest. Factors

Herdman, 1904; Giard, 1907). Stafford (1912) responsible for the observed differences have been

the research subject of several workers (Garner,

' Contribution number 110 of the Ira C. Darling Center. 1872; Dubois 1901, 1909; Jameson, 1902; Scat-

Umversity ot Maine at Orono, Walpole, Maine 04573. , , .

■ Present address: Department of Geology and Geophysics, Yale tergood and 1 aylor, 1949).

University, New Haven, Connecticut 06520 The independent studies of Garner (1872) and

17

18

R.A. LUTZANDH.HIDU

Dubois (1901) showed pearl incidence in M. ednlis to result from infection by a parasitic trematode. Jameson (1902) described the organism which, ac- cording to Odhner (1905), is probably Gytn- nophallus bursicola. Since Jameson's description, considerable work has been done to ascertain the life cycle of the distome parasite (Odhner, 1905: Nicoll, 1906; Giard, 1907; Dubois, 1907a, b, 1909). Thus far, the suggested life cycle involves either the common eider duck, Somateria mollissima, and/or the black scoter, Oidemia nigra, as the definitive host(s) and the blue mussel, M. edulis, as the intermediate host (Dubois, 1907b; Jameson and Nicoll, 1913; Stunkard and Uzmann, 1958). Whether a second intermediate host is required remains uncertain (Nicoll, 1906; Dubois, 1909; Stunkard and Uzmann, 1958). In general, therefore, the life cycle of the trematode is not well established and attempts at further unravelling the cycle have been frustrated largely because of taxonomic chaos. This is emphasized by the statement of Stunkard and Uzmann (1958, p. 298), "specific identification (of gymnophallids) is so uncertain that we prefer to list the worms by host and location rather than propose names that might further confuse the taxonomic situation.''

During the past century, various workers have speculated on factors responsible for the presence or absence of pearls in various populations of mussels (and, hence, factors controlling the distribution patterns of the trematode initiating pearl formation). Of these, salinity has been allud- ed to by two workers. Jameson (1902, p. 143) stated that the "most favorable places (for pearl formation) seem to be estuaries or landlocked channels", while Dubois (1909) observed that pearl-bearing populations were frequently located at the mouths of rivers. d'Hamonville (1894), stu- dying a bed of "rnoulieres perlieres" in Billiers (a borough in Brittany, France near the mouth of the Vilaine), could detect no differences between various environmental parameters (sediment, cur- rent velocity, seaweed distribution, and plankton) of the pearl-infested area and those of an adjacent area harboring "rnoulieres sans perles". Lack of contact of the mussel with the sediment has been suggested as contributing to reduced pearl in- cidence by both Jameson (1902) and Nicoll (1906) based on studies of the creeping motion of the sup-

posed cercaria of the gymnophallid responsible for initiation of pearl formation. Furthermore, Jameson (1902) observed reduced pearl incidence in mussels taken from stakes or floating objects, although Dubois (1909) found pearls within Mytilus galloprovincialis on ropes in the ex- perimental culture operations at the University of Lyon. A final factor, and one which has received only limited attention, is that of age. d'Hamon- ville (1894) commented that pearls were only found in the largest and less regularly formed mussels {"celles qui sont le mains regulierement formees") at Billiers, while Jameson (1902) ob- served that pearls were seldom found in mussels less than 40 mm in length. These stunted and larger individuals may well have been the older members of the population. Jameson (1902) at- tributed the relatively small size of pearls in the Billiers population to the active mussel fishery in the area which prevented individual mussels from reaching "a great age". The effect of age is sum- marized by Jameson's (1902, p. 162) statement that the "general experience of everybody ac- quainted with pearl-fisheries is that the largest pearls were found in the oldest and thickest shells, which proves how intimately the growth of pearl and shell are associated." Despite this realization, no workers have attempted to quantify the rela- tionship between the presence of pearls and age. Furthermore, it has never been demonstrated that age-independent differences exist between various populations with regard to the size and quantities of pearls present. The lack of such quantitative data in these studies renders conclusions concern- ing factors responsible for the presence or absence of pearls highly speculative. It is the purpose of this study to compare quantitatively the size and number of pearls in rafted and shore-based M. edulis populations and to establish if age- independent differences actually exist between these two population types.

MATERIALS AND METHODS

Preliminary Studies

1. Pearl Removal and Sorting

Throughout the following studies, pearls were removed from individual mussels using the following modification of the potassium hydrox- ide technique employed by Scattergood and

PEARLS IN BLUE MUSSEL

19

Taylor (1949). The soft tissues of each mussel were carefully removed from the shell and placed in a 5% solution of boiling KOH for a period of 10 min. This resulted in complete maceration of the tissues. The solution was diluted with water and carefully decanted, leaving the pearl-containing residue at the bottom. The residue was then poured through a specially constructed set of sieves which was rotated and gently shaken under running water for a period of 3 min. This resulted in the separation of pearls into one of the follow- ing six size categories (in microns):

(A) > 1050 (D) 308-471

(B) 602-1050 (E) 153-308

(C) 471-602 (F) 102-153 The total number of pearls per size category was determined by examination of individual sieves under a dissecting scope. The selection of the smallest mesh (102 ^m) was based on two con- siderations. First, the smallest diameter recorded for 486 pearls from 10 mussels was 120 pm. Secondly, no pearls were found on a 50 ^m screen attached to the bottom of the sieve set during analyses of 165 mussels which contained a total of 1574 pearls.

2. Homogeneity of Shore Population A preliminary study was undertaken in Clark Cove on the Damariscotta River in Lincoln Coun- ty, Maine, to determine the variability in the number of pearls per mussel and to test the homogeneity of a shore population with regard to pearl incidence. Five stations with dimensions of 1 m2 were established at the mean low water level (MLW) over a distance of 100 m, with 25 m be- tween stations. Eleven mussels were gathered from each station, sampling only those with lengths be- tween 63.5 and 76.2 mm (2V2 -3"). These were considered representative of the size mussels nor- mally gathered commercially. Numbers of pearls were determined for individual mussels and the results subjected to a one-way classification ANOVA.

Shore vs Raft

Pearl incidence in raft and shore populations of mussels of similar lengths was quantitatively com- pared in four experiments in the Damariscotta River and Boothbay Harbor, Maine. Shell lengths of specimens sampled from the two population types in Experiments #1 - #3 were restricted to the

size range 63.5 - 76.2 mm. Lack of a sufficient quantity of rafted mussels with lengths greater than 63.5 mm in Experiment #4 necessitated the use of a smaller size range (50.1 - 63.5 mm) for adequate comparison of the two populations. 1. Experiment #1: Clark Cove Raft (31 2 years)

Specimens of M. edulis obtained from a raft left in the water approximately 31 -2 years (June, 1968 - December 17, 1971) were quantitatively compared with shore specimens of similar lengths with regard to the size and number of pearls present in the soft tissues. The raft, which was located in the Darmariscotta River (Clark Cove), is depicted in Figure 1. A site with dimensions corresponding to those of the raft was chosen on the shore at MLW within 50 m of the raft. The shore site and the underside of the raft were divided into five sta- tions (Figure 1). The vertical distance between the intertidal and subtidal shore stations was less than 0.2 m. On December 7, 1971, 11 mussels were gathered from each station. The total number of pearls in each size category was determined for in- dividual mussels and these numbers combined to arrive at station and location (shore and raft) totals. Prior to statistical analysis, a logarithmic transformation [log (x + 1)] (Steele and Torrie, 1960) was applied to all data since the variance was found to be approximately proportional to the square of the mean. Nested analyses of variance (mussels within stations, within loca- tions) were run on the transformed values. 2. Experiment #2: Boothbay Harbor Raft (5 years)

Pearl incidence in mussels obtained from a raft left in the water for approximately 5 years was quantitatively compared with that in mussels of similar lengths sampled from an adjacent shore population. As in Experiment #1, a one-way hierarchial classification (mussels within stations, within locations) served as the experimental design for the following study. The raft, which had been in the water of Boothbay Harbor since July, 1967, consisted of a floating wooden bumper approximately 13 m in length. Five stations with dimensions of 1 x 0.75 m were selected on the underside of the raft. Each station was separated from the adjacent one by approximately 2 m and numbered consecutively R4A - R4E. The depth of water under the raft at mean low water varied from approximately 1 m at the inshore station

20

R.A. LUTZANDH.HIDU

3 m

EXPERIMENTAL RAFT

MLW

INTERTIDAL

SHORE SAMPLE PLOT

FIGURE 1. Raft and shore stations (Clark Cove, Damariscotta River, Maine) for Experiment §1. MLW mean low water level.

(R4E) to 7 m at the offshore station (R4A). Similarly, five stations with dimensions of 1 x 0.75 were located on a mussel bed at MLW on the adja- cent shore. Again, the stations were spaced at in- tervals of approximately 2 m. On August 4, 1972, 11 mussels were gathered from each of the 10 sta- tions. The total number of pearls in each size category was determined for individual mussels and the numbers combined to arrive at station and location (shore and raft) totals. The data were transformed [log (x + 1)] and the two populations compared using nested analyses of variance. 3. Experiment #3: Manila Rope (21 months)

A 5/8" diameter Manila rope (2 m in length) was hung from a floating dock in Clark Cove in May, 1971. On January 4, 1973, a 0.5 m section of the rope, suspended 1 m beneath the dock, was

stripped of all attached mussels. Of these, 55 specimens, varying in length from 63.5 - 76.2 mm, were sampled and subjected to the potassium hydroxide treatment. The total number of pearls in each size category was determined for the in- dividual mussels. Through calculation of Stu- dent's t values, the transformed [log (x + 1)] data were compared with the transformed values for the 55 mussels from the adjacent shore population described in Experiment #1. 4. Experiment #4: Asbestos Panel (7 months)

An asbestos panel measuring approximately 0.3 m: was placed at a depth of 1 m beneath a floating dock on the Damariscotta River (Wentworth Point) in May, 1972. On December 2, 1972, 55 mussels (from natural larval settlement) were removed from the panel and an additional 55

PEARLS IN BLUE MUSSEL

21

specimens sampled from an adjacent shore population at MLW. The total number of pearls in each size category was determined for each specimen and the transformed (log (x+1)] values for both populations compared through calcula- tion of Student's t values. Comparison of Rafted Environments

Transformed [log (x+1)] pearl counts from mussels sampled from each of the rafted en- vironments assessed in the preceding four ex- periments were quantitatively compared. Seven separate analyses of variance were run comparing totals per mussel and totals for each of the six size categories. For analyses indicating a difference among a set of population means, Tukey's w- procedure was used to judge the significance of differences between the individual population means within the set. This test is more conser- vative than Duncan's multiple range test as the en- tire experiment is the unit used in stating the number of errors of the type where an observed difference is falsely declared to be significant. The test criterion is

w =q.05(p.n2)(sr) where q.os is a tabulated value (Steele and Torrie, 1960, p. 445) for p treatments and n2 degrees of freedom and sr is estimated from the error mean square, w is used to judge the significance at the 0.05 level of each of the observed differences (Steel and Torrie, 1960).

Shore vs Raft (Age-Independent Studies) 1. Transplantation

Pearl incidence in mussels left in a rafted en- vironment for 14 months was quantitatively com- pared with that in mussels left on an adjacent shore for the same period of time. On October 19, 1972, 250 mussels with lengths varying from 13.0 - 17.0 mm were gathered from the underside of the

1 Engle and Loosanoff (1944) found the season of setting of mussels in Milford, Connecticut, to be of comparatively short duration, extending from the early part or middle of June to the middle or end of August. Plankton samples of McAlice et al. (unpublished data) suggest that the vast majority of M. edulis larvae within the Damariscotta River are present between the months of May and September. It follows that the greatest pro- portion of larval settlement in this estuary should be restricted to these warmer months. In light of these considerations, the majority of mussels from the two sampled populations are ex- pected to be restricted to summer year classes. Age estimates in this study are. therefore, generally considered accurate to within 1 year.

raft described earlier in Experiment #1. These were individually notched for subsequent identifica- tion. Twenty-five individuals were placed in each of the 10 sections (20 cm long) of cylindrical 5 mm VEXAR plastic netting with a layflat diameter of 10 cm. Five sections were hung from a raft off Wentworth Point at a depth of 0.5 m below the surface. The remaining five stations were placed in contact with sediment at the mean low water level on the shore within 50 m of the raft. On December 19, 1973, 65 of the original 125 mussels were recovered from the shore station and all 125 mussels from the raft. The pearls were removed from 65 of the 125 rafted specimens and all of the shore mussels. These were separated into size categories as previously described. Pearl counts from the two populations were quantitatively compared.

2. Regression of Pearls on Age

The regression of the number of pearls per mussel on age for a rafted M. edulis population was quantitatively compared with a similar regression for an adjacent shore population. On July 26, 1973, 85 mussels of 25 - 96 mm length were sampled from a raft placed in the water at South Bristol, Maine in June, 1964. Similarly, 102 mussels varying in length from approximately 19- 92 mm were gathered on the same date from an adjacent shore population. The pearls were removed as described earlier and the total number per individual recorded. One shell valve of each specimen was embedded in epoxy and longitudinally sectioned along the antero- posterior axis. The exposed shell surfaces were polished and etched and acetate peels prepared for examination with both compound and dissecting microscopes. Age estimates were based on counts of annual growth patterns within the inner shell layer [See Lutz (1976) for a detailed description of annual growth patterns within the shell of both shore and raft-based specimens].1 Occasional sub- jectivity associated with the counting of lines may somewhat reduce the accuracy of age estimates for certain specimens. Such subjectivity, however, is expected to be equal for both populations and should have little effect on the interpretation of differences between the two. As the variance of pearl counts for individual year classes was found positively correlated with the mean, actual counts

22

R.A. LUTZANDH.HIDU

were logarithmically transformed [log (x + 1)1 (Steel and Torrie, 1960). A regression of trans- formed pearl counts on age was run for each population, using a total of 78 rafted and 82 shore specimens. Five of the original 85 rafted mussels and seven of the original 102 shore specimens were discarded because of poorly preserved peels or suspected subjectivity associated with the count- ing of lines. Two 1-year-old rafted and 13 shore specimens ranging in age from 10 - 16 years were also excluded because of the lack of corresponding age classes in the other population. Differences in slope and elevation between the two regression lines were assessed using an analysis of covariance (Steel and Torrie, 1960; Sokal and Rohlf, 1969).

RESULTS

Preliminary Studies

The Clark Cove shore population (at MLW) was found to be relatively homogeneous with regard to the quantity of pearls present in the soft tissues. Table 1 shows the results of the ANOVA, with the null hypothesis that no difference exists between station means other than that due to chance. The null hypothesis was accepted, sug- gesting that a sampling site at MLW could be located randomly at any point along the 100 m stretch with confidence of obtaining a represen- tative sample of the population. Shore vs Raft

Highly significant differences in pearl incidence were found between rafted and shore-based populations of mussels. In each of the four ex- periments, rafted individuals contained fewer and smaller pearls than mussels of similar length ob- tained at MLW from an adjacent shore popula- tion.

1. Experiment #1: Clark Cove Raft (31 2 years)

Specimens of M. edulis obtained from the raft left in the water for approximately 31 2 years con- tained significantly fewer pearls than specimens of similar length obtained from an adjacent shore population (Table 2). The nested ANOVA in- dicates a highly significant difference (F =188.83) between the raft population mean (2.2 pearls per mussel) and that of the adjacent shore population (14.3 pearls per mussel). The null hypothesis of no significant difference between stations means was accepted.

TABLE 1. Total pearl counts and analysis of vari- ance for mussels sampled from each of five sta- tions at the mean low water level in Clark Cove, Damariscotta River, Maine.

		Station
Specimen	A	B	C	D	E
1	3	14	2	2	0
2	22	2	4	12	29
3	35	8	9	0	6
4	7	61	8	10	2
5	5	18	6	3	68
6	2	11	2	1	18
7	2	0	6	2	6
8	11	9	8	16	29
9	4	10	14	4	1
10	86	6	29	9	5
11	13	6	0	13	5
Total	190	145	88	72	169
Mean	17.3	13.2	8.0	6.6	15.4
Source of
Variation	df	SS	MS	F
Among stations	4	948. 9C	) 237.20 0.84 n.s.
Within stations	50	14957.11	282.06
Total	54	15906.01

In addition, pearls found in the rafted mussels were significantly smaller than those obtained from the examined shore specimens. Results of the six separate analyses of variance (Table 3) il- lustrate that the shore population mean number of pearls in any size category is significantly greater at the 0.01 level than the mean for the raft popula- tion in any corresponding size category. Further- more, no pearl was found in this study with a diameter greater than 1 mm in any raft-based in- dividual. No significant differences between sta- tion means (experimental error) were shown in any of the size category analyses. 2. Experiment §2: Boothbay Harbor Raft (5 years)

Pearl incidence in mussels obtained from the raft left in the water for approximately 5 years was significantly lower than that in specimens of similar length from the adjacent shore population (Table 4). The null hypothesis (for the nested ANOVA) of no significant difference between locations (shore and raft) was rejected, with the shore population mean (27.7 pearls per mussel) considerably greater than the mean of the rafted

PEARLS IN BLUE MUSSEL

23

TABLE 2. Total pearl counts and analysis of variance for raft and shore mussels in the Damariscotta River (Clark Cove)."

			Raft"	Station			Shorec
Specimen	A	B	C	D	E	A	B	C	D	E
1	1	3	0	0	0	28	2	26	15	13
2	0	1	0	0	1	11	70	7	12	9
3	12	0	1	3	0	46	13	11	3	2
4	1	0	2	18	0	8	39	5	8	2
5	8	5	0	2	10	14	1	10	38	12
6	2	0	9	2	1	5	8	0	2	1
7	1	0	0	1	5	7	3	4	17	42
8	1	7	1	0	7	15	0	14	6	10
9	0	0	3	0	5	19	3	67	5	16
10	3	1	3	0	0	14	5	5	0	59
11	0	2	0	0	2	3	7	6	16	4
Total	29	19	10	26	31	170	L69	155	122	170
		Raft	mean =	2.25		Shore mean =	14.29
Source of Variation				df	SS		MS		F
Among stations					9	11.78		1.31
Locations					1	11.33		11.33		L88.83**
Among stations w	ithin li	jcations			8	0.45		0.0b		0.38 n.s.
(experimental	error
Among mussels w	thin stations			100	16.33		0.16
(sampling error)
Total					109	28.11

"The results of the analysis of variance are based on transformed [log(x + 1)] pearl counts. 'Rafted specimens were obtained from a raft left in the water approximately 31 2 years (June, 1°68 December 17, 1971).

'Shore specimens were gathered at the mean low water level within 50 m of the raft. * indicates significance at the 0.01 level.

TABLE 3. Analyses of variance for individual size category pearl counts from raff and shoreb mussels in the Damariscotta River (Clark Cove).'

Size		Transformed	F	F
category	Population	mean	Locations	Experimental error
>1050 Mm	Raft Shore	0 0.0199	28.9043*'	1.1605 n.s.
602-1050 nm	Raft Shore	0.0748 0.3184	25.8653**	1.1090 n.s.
471-602 um	Raft Shore	0.0455 0.2686	21.8546**	1.2108 n.s.
308-471 um	Raft Shore	0.1122 0.4439	53.9982**	0.6080 n.s.
153-308 Mm	Raft Shore	0.1680 0.6293	239.8483**	0.2161 n.s.

24

R.A. LUTZANDH.HIDU

102-153 Mm

Raft Shore

0.0917 0.2692

13.9485**

0.9227 n.s.

"Rafted specimens were obtained from a raft left in the water approximately 3V2 years (June, 1968- Decemberl7, 1971).

fcShore specimens were gathered at the mean low water level within 50 m of the raft. 'The results of the analyses of variance are based on transformed [log(x + 1)] pearl counts. "Indicates significance at 0.01 level

TABLE 4. Total pearl counts and analysis of variance for raff and shoreb mussels in Boothbay Harbor, Maine'.

Station

Specimen

Raft" C

D

Shore" C

D

1

2

3

4

5

6

7

8

9

10

11

Total

1 2 4 1 2 1 1 0 0 0 3 15

0 0 0

12 0 0

49 0 0 0 1

62

2

0

0

0

144

0

12

2

1

20

1_

182

1 97 1 0 5 1 0 0 0

1 !_

107

0 2 0 0 0 4 4 0 0 1 0 11

1	75	47	16	24
12	61	4	30	0
98	5	75	5	108
1	43	15	7	9
6	18	27	13	10
25	26	35	12	8
6	13	7	4	61
4	63	7	18	1
13	20	24	7	10
64	q	5	7	27
95	57	45	75	67

325

390

291

194

325

Source of Variation

Raft mean = 6.85

df

Shore mean = 27.72 SS MS F

Among stations

Locations

Among stations within locations

(experimental error) Among mussels within stations

(sampling error) Total

9	22.9170	2.5463
1	21.3044	21.3044	105.6765**
8	1.6126	0.2016	0.8096 n.s

100

109

24.8980

47.8150

0.2490

"Rafted specimens were obtained from a raft left in the water approximately 5 years (July, 1967 - August 4, 1972).

"Shore specimens were gathered at the mean low water level within 50 m of the raft. 'The results of the analysis of variance are based on transformed [log(x + 1)] pearl counts. "Indicates significance at the 0.01 level.

population (6.8 pearls per mussel). The raft population, however, was extremely variable and, occasionally, individuals were found with numerous pearls (as many as 144). No significant

difference between station means (experimental error) was shown in this experiment.

In addition, the average pearl diameter en- countered in rafted specimens was considerably

PEARLS IN BLUE MUSSEL

25

smaller than that encountered in shore-based in- dividuals. As in Experiment #1, highly significant differences were found in each of the six separate analyses of variance (Table 5), indicating that the shore population mean number of pearls in any size category is significantly greater than the mean number for the raft population in any correspond- ing size category. No significant difference be- tween station means (experimental error) was show in any of the size category analyses. Only one pearl was found with a diameter greater than 1 mm. 3. Experiment #3: Manila Rope (21 months)

The total numbers of pearls in individual mussels from the rope left in Clark Cove for ap- proximately 21 months are shown in Table 6. The calculated Student's t indicates a highly significant difference between the two population (rope and adjacent shore) means, with the shore population mean number of pearls (14.3) per individual con- siderably greater than that of the rope population (mean = 0.3 pearls per mussel).

Highly significant differences in pearl size were also found between the two populations. The ac- tual and transformed population means, standard

deviations of the transformed means, and Stu- dent's t values calculated from the transformed data are summarized in Table 7 for each of the six categories. The shore population mean number of pearls in any size category was found to be significantly greater (at the 0.01 level) than the raft population mean for any corresponding size category. No pearls were found with a diameter greater than 1 mm in any of the mussels examined from the rope. 4. Experiment #4: Asbestos Panel (7 months)

Mussels obtained from the asbestos panel left off Wentworth Point for 7 months were found to be relatively pearl-free in comparison with specimens of similar length sampled from the adja- cent shore population (Table 8). The actual and transformed means, standard deviations of the transformed means, and Student's t comparing the two population means (based on transformed data) are given in Table 8. The shore population mean (7.5 pearls per mussel) was found to be significantly greater (at the 0.01 level) than the mean of the asbestos panel population (0.1 pearl per mussel).

As in Experiment #3, highly significant dif-

TABLE 5. Analyses of variance for individual size category pearl counts from raff and shoreb mussels in Boothbay Harbor. '

Size category

Population

Transformed mean

Locations Experimental error

>1050um 602-1050 um 471-602 Mm 308-471 nm 153-308 um 102-153 Mm

"Rafted specimens were obtained from a raft left in the water approximately 5 years (July, 1967 - August 4, 1972).

'Shore specimens were gathered at the mean low water level within 50 m of the raft. cThe results of the analysis of variance are based on transformed [log(x + 1)] pearl counts. '"Indicates significance at the 0.01 level.

Raft	0.0055
Shore	0.1578	19.2810**	0.7291 n.s.
Raft	0.0274
Shore	0.3401	34.2510**	1.1684 n.s.
Raft	0.0930
Shore	0.3665	26.8486**	0.9087 n.s.
Raft	0.1340
Shore	0.5989	59.5426**	0.5977 n.s.
Raft	0.2064
Shore	0.8966	44.9931**	1.4177n.s.
Raft	0.1212
Shore	0.4801	31.1176**	0.8762 n.s.

26

R.A.LUTZANDH.HIDU

TABLE b. Total pair! counts front rafted mussels and results of Student's t test comparing these counts with counts from shore specimens."

Specimen	Pearl	s Specimen	Pearls	Specimen Pearls	Specimen	Pearls	Specimen	Pearls
1	0	12	0	23	1	34	0		45	1
2	0	13	4	24	0	35	0		46	0
3	0	14	0	25	0	36	0		47	0
4	1	15	0	26	0	37	0		48	0
5	2	16	0	27	0	38	0		49	1
6	0	17	0	28	2	39	0		50	0
7	0	18	0	29	0	40	0		51	0
8	0	19	2	30	0	41	0		52	0
9	1	20	0	31	0	42	0		53	0
10	0	21	0	32	1	43	0		54	0
11	0	22	0	33	0	44	1		55	0
			Actual	Transformed	Std.	Computed
		Population	mean	mean		dev.	t
		Raff	0.31	0.0770		0.1655
		Shore'	14.29	0.9824		0.4238	14.7700*	*

"Student's t values were calculated using transformed [log(x + 1)] pearl counts.

''Rafted specimens were obtained from a 5/8" Manila rope suspended from a floating dock in the Damar-

iscotta River (Clark Cove) for approximately 21 months (May, 1971 - January 4, 1973) . cShore specimens were gathered at the mean low water level within 50 m of the raft. * indicates significance at the 0.01 level

TABLE 7. Student's	t test results for	individual size	catego	ry pearl counts	from raft and	shore mussels in
the Damariscotta River (Clark Cove)	,°
Size		Actual	Transformed	Std.	Computed
category	Population	mean		mean	dev.	t
>1050 urn	Raft*	0		0	0
	Shore-	0.93		0.1988	0.2560	5.7620**
602-1050 ^m	Raft	0.04		0.0109	0.0568
	Shore	1.65		0.3184	0.2935	7.6303**
471-602 ^m	Raft	0		0	0
	Shore	1.45		0.2686	0.2966	6.7150**
308-471 um	Raft	0.07		0.0164	0.0901
	Shore	3.13		0.4438	0.3689	8.3477**
153-308 um	Raft	0.18		0.0524	0.1218
	Shore	5.44		0.6293	0.3868	10.5466**
102-153 Mm	Raft	0.02		0.0055	0.0406
	Shore	1.49		0.2692	0.3016	6.4317**

"Student's t values were calculated using transformed [log(x + 1)] pearl counts.

''Rafted specimens were obtained from a 5/8" Manila rope suspended from a floating dock in the Damari- scotta River (Clark Cove) for approximately 21 months (May, 1971 - January 4, 1973). 'Shore specimens were gathered at the mean low water level within 50 m of the raft. ** Indicates significance at the 0.01 level.

PEARLS IN BLUE MUSSEL 27

TABLE 8. Total pearl counts and analysis of variance for raft " and shoreb mussels in the Damariscotta River (Wentworth Point).'

Asbestos Panel Shore

Specimen Pearls Specimen Pearls Specimen Pearls Specimen Pearls

1	0	28	0
2	0	29	0
3	0	30	0
4	1	31	0
5	0	32	0
6	0	33	0
7	0	34	1
8	0	35	1
9	0	36	0
10	0	37	0
11	0	38	0
12	0	39	0
13	0	40	0
14	0	41	0
15	0	42	0
16	0	43	0
17	0	44	0
18	0	45	0
19	1	46	0
20	0	47	0
21	0	48	0
22	0	49	0
23	0	50	1
24	0	51	0
25	0	52	0
26	0	53	0
27	0	54	0
		55	0
		Actual
	Population	mean
	Raft	0.09
	Shore	7.51

1	3	28	1
2	0	29	0
3	3	30	85
4	1	31	9
5	2	32	1
6	3	33	5
7	10	34	5
8	9	35	7
9	4	36	11
10	2	37	6
11	1	38	12
12	2	30	23
13	2	40	3
14	1	41	13
15	3	42	4
16	2	43	7
17	3	44	2
18	1	45	4
19	1	46	0
20	o	47	6
21	14	48	4
22	7	49	1
23	2	50	4
24	27	51	2
25	8	52	9
26	23	53	7
27	10	54	16
		55	7
	Std.	Computed
	dev.	t
	0.087
	0.374	13.822**

Transformed mean 0.027 0.743

° Rafted specimens were obtained from an asbestos panel suspended at a depth of 1 m beneath a floating dock for approximately 7 months (May- December 2, 1972).

6 Shore specimens were gathered at the mean low water level within 50 m of the raft.

' The results of the analysis of variance are based on transformed [log(x + 1)] pearl counts.

* 'Indicates significance at the 0.01 level.

ferences in pearl size were also found between the calculated from the transformed data are sum- two population types (rafted and shore). The ac- marized in Table 9. The shore population mean tual and transformed means, standard deviations number of pearls in any size category was found of the transformed means, and Student's t values to be significantly greater (at the 0.01 level) than

28

R.A. LUTZANDH.HIDU

the asbestos panel population mean. No pearls were found with a diameter greater than 471 fim in any of the mussels sampled from the asbestos panel.

Comparison of Rafted Environments

A positive correlation was observed between the number of pearls in mussels sampled from each of the rafted populations in Experiments #1 -

TABLE 9. Student's t test results for individual size category pearl counts from raft and shore specimens in the Damariscotta River (Wentworth Point)."

Size		Actual	Transformed	Std.	Computed
category	Population	mean	mean	dev.	t
>1050fjm	Raft"	0	0	0
	Shoref	0.66	0.1426	0.2231	4.7375**
602-1050 Mm	Raft	0	0	0
	Shore	0.71	0.1765	0.2130	6.1498**
471-602 Mm	Raft	0	0	0
	Shore	0.64	0.1560	0.2140	5.3979**
308-471 Mm	Raft	0.02	0.0055	0.0406
	Shore	1.38	0.2746	0.2892	6.9000**
153-308 Mm	Raft	0.07	0.0219	0.0789
	Shore	3.09	0.3848	0.3648	7.3760**
102-153 Mm	Raft	0	0	0
	Shore	1.05	0.1861	0.2818	4.8974**

"Student's t values were calculated using transformed [log(x + 1 )] pearl counts.

"Rafted specimens were obtained from an asbestos panel suspended at a depth of 1 m beneath a float- ing dock for approximately 7 months (May-December 2, 1972). cShore specimens were gathered at the mean low water level within 50 m of the raft. * 'Indicates significance at the 0.01 level.

TABLE 10. Analyses of variance and results of Tukey's w-procedures for total and individual size cate- gory pearl counts."

Source of Variation

Total Number of Pearls df

SS

MS

Among Rafts Within Rafts

Total

3 216

4.7677 22.8338

1.5892 0.1057

219 27.6015

w = q 05(4,216)sr = 3.63(0.0438) = 0.0438

Raft Rl" R2< R3" R4<

Mean

0.0274

0.0770

0.3402

0.3487

15.0353*

Source of variation

Size category >1050 Mm df

SS

MS

Among Rafts Within Rafts

3 216

0.0012 0.0890

0.0004 0.0004

.0000 n.s.

Total

219

0.0902

PEARLS IN BLUE MUSSEL

29

Source of variation

Size category: 602-1050 f^rn

df SS

MS

Among Rafts Within Rafts

Total

3 216

219

0.1795 2.3123

2.4918

W = q 05(4,216)5* = 3.63(0.0140) = 0.0506

Raft

Mean

Rl

0

R2 0.0109

R4 0.0274

R3

0.0748

0.0598 0.0107

5.5888**

Source of variation

Size category: 471-602 ^m

df SS

MS

Among Rafts Within Rafts

Total

Source of variation

3 216

219

0.3259 3.4878

3.8137

0.1086 0.0162

w = q 05(4,216)sr + 3.63(0.0172) = 0.0623

Size category: 153-308 /^m

df SS

MS

6.7058'

	Raft Mean	Rl R2 R3 0 0 0.0455	R4 0.0930
		Size category: 308-471 /jm
Source of variation		df	SS	MS	F
Among Rafts Within Rafts		3 216	0.7538 9.8919	0.2513 0.0458	5.4869**
Total		219	10.6457
	w	= q 05(4,216)sj = 3.63(0.0289)	= 0.1048
	Raft Mean	Rl R2 R3 0.0055 0.0164 0.1122	R4 0.1400

Among Rafts Within Rafts

Total

	3 1.3038	0.4346
	216 14.2086	0.0658
	219 15.5124
w	= q 05(4,216)sF = 3.63(0.0346) = 0.1255
Raft	Rl R2 R3 R4
Mean	0.0219 0.0524 0.1680 0.2064

6.6069*

30

R.A. LUTZANDH.HIDU

Source of variation

Size category: 102-153 ^m

df SS

MS

Among Rafts		3 0.6160
Within Rafts		216 7.8973
Total		219 8.5133
	w	= q 05(4,216)sr= 3.63(0.0258) = 0.0936
	Raft	Rl R2 R3 R4
	Mean	0 0.0055 0.0917 0.1212

0.2053 0.0366

5.6093*

"Mussels were sampled from substrates left in the water for varying lengths of time.

"Rl = Asbestos Panel (May, 1972 - December 2, 1972)

•R2 = Clark Cove Rope (May, 1971 - January 4, 1973)

JR3 = Clark Cove Raft (June, 1968 - December 17, 1971)

<R4 = Boothbay Harbor Raft (July, 1967 - August 4, 1972)

"Indicates significance at the 0.01 level.

#4 and the amount of time for which the rafted substrate had been in the water (Table 10). The null hypothesis of no significant difference be- tween pearl counts in mussels from the four rafted substrates was rejected for six of the seven analyses of variance (Table 10), the only non- significant difference occurring for pearls with diameters greater than 1050 ^m. Although Tukey's w-procedure failed to show a significant difference (at the 0.05 level) between each of the individual population means, examination of the transformed means for the totals indicates a general trend for an increasing number of pearls per mussel with increasing length of submergence of the substrate from which the specimens were collected. With the exception of minor discrepan- cies, this same general trend holds true for the population means calculated for each individual size category.

Shore vs Raft (Age-Independent Studies) 1. Transplantation

No significant difference in pearl incidence was found between mussels left in a rafted environ- ment for 14 months and specimens left on an adja- cent shore for the same period of time. The 65 in- dividuals which had been on the shore contained a total of two pearls. These were in two individuals and the two size categories, 102-153 and 309-471 |im. The 65 individuals examined from the raft population also contained a total of two pearls in two individuals. These were found in the two size categories, 154-308 and >1050 ptm. The pearl

found in the latter size category had a diameter of approximately 1100 ^m. The identical number of pearls found in each population rendered statistical analysis of the data superfluous. 2. Regression of Pearls on Age

No significant age-independent difference in the number of pearls within the soft tissues was found between rafted mussels in South Bristol and specimens obtained from the adjacent shore population. The total numbers of pearls within in- dividual mussels in each year class are shown in Tables 11 and 12 for the raft and shore popula- tion, respectively. The regression lines and 95% confidence belts for the regression of the transformed [log (x + 1)] counts on age are depicted in Figure 2 for each population. The regression equations for the raft and shore popula- tions are, respectively,

ind

Y = 0.1540X — 0.3765

Y = 0.1217X — 0.3266,

where Y is the estimated mean of the transformed [log (x+1)] pearl counts for a mussel of age X (in years). An estimate (Y) of the mean number of pearls per mussel of age X is given in the following formula from Elliot (1971):

Y = antilog[1.15sr + Y] — 1,

where s,. is the standard error of Y for a given value of X. Following the procedure outlined by

PEARLS IN BLUE MUSSEL

31

TABLE 11. The total number of pearls within individual mussels in each year class (raft)."

Raft (Number of pearls per individual)

Specimen	0				Year Class
		2 0	3 0	4 1	5 6 0 3	7 7	8 113	9
1	5
2		0	1	5	1 9	2	14
3		0	0	1	0 1	1	3
4		0	0	0	0	1
5		0	0	0	1	2
6		0	0	0	o
7		0	0	1	7
8		0	0	0	4
9		0	1	0	1
10		0	0	5
11			1	0
12			0	1
13			2	1
14			0	0
15			0	0
16			0	0
17			0	0
18			1	0
19			0	4
20			0
21			0
22			0
23			1
24			0
25			0
26			1
27			0
28		0	0 8
Total	19	23 13	13	130	5
Mean		0	0.3	1.0	2.6 4.3	2.6	43.3	5.0
"Mussels were samples from a	South B	ristol rafted	population.

Sokal and Rohlf (1969), an estimate of s, for the raft and shore populations are provided by the equations

s,. = JO. 1521 [0.0128 + (X — 3.9872)- 1

1447

and

s, = /0.0277 [0.0122 + (X — 5.4268)2]

2739 respectively. Once again, X is the age in years of

the specimen. Insertion of these estimates into Elliot's formula and substitution of the ap- propriate values of Y for each population provides an estimate (Y) of the mean number of pearls per mussel for both the raft and shore population. The equations are, respectively,

Y = antilog (1.15 s2 + 0.1540X — 0.3765) — 1 and

Y = antilog (1.15 s,2 + 0.1217X — 0.3266) — 1

32

R.A.LUTZANDH.HIDU

TABLE 12. The total number of pearls within individual mussels in each year class (shore).

Shore (Number of pearls per individual)

Specimen	#				Year	Class
		2	3	4	5	6	7	8	9
1		0	0	0	0	5	6	4	2
2		0	0	1	0	8	0	9	3
3		0	0	1	3	3	0	9	9
4		0	0	0	1	1	0	9	4
5		0	0	0	0	1	2	13	30
6			1	0	0	1	6	0	6
7			0	1	2	0	22	2	3
8			0	0	0	30		1	1
9			0	0	0	1
10				0	0	3
11				0	0	1
12				3	4	0
13				1	3	1
14				4	0	0
15				0	2
16				0 11				47
Total	0	1	15	55	36	58
Mean		0	0.1	0.7	1.0	3.9	5.1	5.9	7.2
"Mussels were sampled from a South Bristol shore	population at the mean	low water	level.

However, in consideration of the relatively small magnitude of sj, the comparatively wide 95% con- fidence intervals, and the subjectivity associated with the counting of the annual lines on acetate peels, a relatively accurate estimation of Y for each population is provided by substitution of ap- propriate values in the equation

Y = antilogY— 1.

For both populations the linear regression on age was found to remove a highly significant por- tion of the variation of the transformed pearl counts (Table 13). The deviations from regression were not significant. Table 14 shows the results of the analysis of covariance comparing the two regression lines. No significant difference between the slopes of the regression lines nor between the adjusted transformed means of the two popula- tions (elevation of lines) was detected in this study.

Finally, no pearls with diameters greater than 1 mm were found in mussels less than 5 years old. Two pearls (one from each population) with diameters greater than 1 mm were found in 5-year- old specimens.

DISCUSSION

The results of this study suggest that, with regard to the size and quantity of pearls in the mantle tissue, no difference exists between raft- based mussels and those of comparable age ob- tained from adjacent shore populations. The dif- ferences found between mussels of similar lengths obtained from rafts left for varying periods of time are probably a reflection of age differences be- tween the individuals sampled. Similarly, the dif- ferences observed between raft and shore popula- tions in this study are probably a reflection of a difference in the age distribution of specimens from the two environments. Lutz and Porter (1977) have shown that a 50 mm mussel is ob-

PEARLS IN BLUE MUSSEL

33

RAFT

Y= I540X - .3765

•3 '

SHORE

Y= 1217 X - 3266

-l 1 1 1 1— rlXH — S

2 345678902

AGE (years)

FIGURE 2. Regressions of logarithmically transformed pearl counts on age for raft and shore based mussels. Dashed lines represent 95% confidence belts.

tained in at least one Maine environment in a period of 12 months from settlement using rafting techniques. Reported estimates of the average growth rate of mussels (using length measure- ments) in natural populations (submerged and in- tertidal - the North Atlantic region) vary from 1.9 - 16.0 mm per year (Mossop, 1922a, b; Field, 1922). Thus, the rafting of mussels may increase linear growth rate as much as 3-25 fold over that of natural benthic populations. Therefore, although the age structure of the two population types (raft and shore) in most of the experiments conducted is not known, the average age of the shore mussels is suspected to be considerably greater than that of the rafted specimens.

If the rate of pearl formation and pearl growth

remains relatively constant throughout the life of a mussel, one might expect little or no difference between old and young individuals with regard to the number of pearls in the smallest size category. The results of Experiments #1 - #4 and the com- parison of rafts left for varying lengths of time in- dicate that this is generally not the case. The greater number of small pearls in shore-based in- dividuals and in mussels sampled from the rafts left longest in the water suggest that there is a general increase in the number of pearls being formed and/or a decrease in the growth rate of pearls as the individual ages. Both of these are probably occurring to some extent. As the in- dividual ages, there is generally an increase in the surface area of the mantle epithelium. If, as has

34

R.A. LUTZANDH.HIDU

TABLE 13. Analyses of variance for regressions of transformed pearl counts on age.

Source of variation

Among year classes Linear regression Deviations from regression Within year classes Total

Raft

df	SS	MS	F
7	5.8213	0.8316	12.3934**
1	4.9086	4.9086	32.2722**
6	0.9127	0.1521	2.2335 n.s
70	4.7667	0.0681

77

10.5880

Shore

Source of variation

df

SS

MS

Among year classes Linear regression Deviations from regression Within year classes Total

7	4.9670	0.7096	6.5945**
1	4.8010	4.8010	173.3213**
6	0.1660	0.0277	2.2574 n.s
75	8.0720	0.1076

82

13.0590

"An analysis is shown for each of two populations (raft and shore) in South Bristol, Maine. * 'Indicates significance at the 0.01 level.

TABLE 14. Analysis of covariance comparing the regressions of transformed pearl counts on age."

Analysis of Covariance

Transformed adjusted raft mean = 0.2375 Transformed adjusted shore mean = 0.3338

Within Raft Shore

Pooled, W

Between, B

df x: xy y2 b

77 206.9872 31.8750 10.5880 0.1540 81 324.0610 39.4440 13.0390 0.1217

Total 158 531.0482 71.3190 23.6270 0.1343 Difference between slopes 1 82.8518 5.5404 0.3705

613.9000 76.8594 23.9976 0.1252 Between adjusted means

df SS MS F

7b 5.6794 0.0747

80 8.2379 0.1030 1.3788 n.s.

156 13.9173 0.0892

157 14.0490 0.0895

1 0.1317 0.1317 1.4765 n.s.

158 14.3749

1 0.3259 0.3259 3.6413 n.s.

The analysis compares the regression lines for two populations (raft and shore) in South Bristol, Maine.

been suggested by numerous workers (Garner, 1972; Dubois, 1901, 1909; Jameson, 1902; Jameson and Nicoll, 1913), pearl formation is a result of trematode infection, this increased sur- face area may offer a greater number of sites for penetration by the parasite, and, hence, would result in an increased number of pearls. As the ma- jority of pearls in M. edulis are nacreous, the same

processes controlling thickening of the inner shell layer may control the growth of pearls. This was recognized by Jameson (1902, p. 161) who stated in reference to pearls, "their growth is, in fact, regulated by the causes which control the thicken- ing of the shell." Examination of the polished shell sections and corresponding acetate peels presented by Lutz (1976) reveals that there is a general

PEARLS IN BLUE MUSSEL

35

decrease in the width of annual increments with increasing age of the specimen. A similar decrease in the growth rate of pearls probably occurs.

Jameson (1902) and Nicoll (1906) claim that the gymnophallid trematode responsible for pearl for- mation in M. edulis from French and English waters enters the mussel as a tailless cercaria. This, in turn, led Jameson (1902, p. 160) to con- clude that "in order to be abundantly infected, Mytilus must be on the bottom." Dubois (1909) disagrees, claiming to have found all larval stages of the trematode within the mantle of M. edulis. If pearls in mussels from Maine waters are also a result of infection by the trematode, Gym- nophallus, the results of the present study appear to discredit the claims of Jameson (1902) and Nicoll (1906). If a larval stage of Gymnophallus enters M. edulis, it is a stage that is not restricted to a bottom habitat.

Dubois (1907b) conducted an experiment to test the effect of increased temperatures on the metamorphosis of metacercariae larvae of the gymnophallid trematode responsible for the initia- tion of pearl formation. Based on the acquisition of adult characteristics at higher temperatures (35 - 40°C), he concluded that larval metamorphosis to the adult form must take place in a warm- blooded animal. Jameson and Nicoll (1913) sug- gest that the common eider duck, Somateria mollisima, and/or the black scoter, Oidemia nigra, serve(s) as the definitive host(s). Stunkard and Uzmann (1958) fed metacercaria-infected mussels from Long Island to newly hatched eider chicks and recovered adult gymnophallids, pro- bably G, bursicola (Odhner, 1905), from the in- testines of the chicks. Both Oidemia nigra and Somateria mollissima are abundant in certain areas throughout the Gulf of Maine and either one or both may serve as the adult host(s) for this trematode. If pearl formation in these waters is a result of gymnophallid infection, it is reasonable to expect the presence and quantity of pearls in mussels to be somewhat correlated with the presence and abundance of eiders and/or scoters in the surrounding waters. Before such a correla- tion can be made, however, age-independent dif- ferences (if existent) in pearl incidence between geographically isolated populations of M. edulis must be quantified. The results of this study sug-

gest a method of obtaining such quantification.

Several workers (Jameson, 1902; Nicoll, 1906; Lebour, 1906) have suggested that a second in- termediate host may be required for completion of the life cycle of the trematode responsible for pearl formation. Jameson (1902) found sporocysts con- taining larvae, almost identical with those which occur in M. edulis, within the mantle margins of the bivalves, Tapes decussatus and Cardium edule. Nicoll (1906), in his examination of numerous specimens of Cardium edule, failed to find such sporocysts in the mantle margin, but noted (P. 149) their frequent occurrence in "one well-defined, somewhat oval-shaped mass situated in the middle line dorsally just over the posterior border of the liver." Neither Tapes decussatus nor Cardium edule are found within the Gulf of Maine. Thus, if the trematode (probably Gymnophallus bursicola) responsible for the initiation of pearl formation in Maine waters requires a second intermediate host for the completion of its life cycle, such a host remains to be found. Furthermore, if this trematode is responsible for the initiation of pearl formation in Maine waters and if one and only one additional intermediate host is required, the presence and quantity of pearls in mussels should be correlated with the distribution patterns of this host. Again, the results of this study suggest a method of quan- tifying differences (independent of age) between geographically isolated M. edulis populations with regard to pearl incidence, facilitating correla- tion of the presence and quantity of pearls with the presence and abundance of such a host.

ACKNOWLEDGEMENTS

This work was supported by NOAA Grants 1-36099, NG-40-72, 04-3-158-63, 04-3-158-38, 04-5-158-50, 04-6-158-44056, SGI-77-17, and 04-7-158-44034. We thank Drs. David Dean, Franklin Roberts, and Michael Mazurkiewicz for their guidance and critical reviews of initial drafts of the manuscript; Dr. Bernard J. McAlice and Mr. William Soule, Jr. for their advice and direc- tion with the statistical aspects of this work; Mr. Barnaby Porter for his helpful suggestions and assistance, particularly in the design and construc- tion of numerous experimental rafts used

36

R.A. LUTZANDH. HIDU

throughout this study; Mr. Keith Leeman and Mr. Kenneth Fossett for assistance in setting up lapidary equipment; Miss Phyllis Coggins and Mr. Richard D. Smith for their assistance with graphic illustrations; Miss Patricia Higgins for typing the manuscript; and especially Miss Sarah E. Hurlburt for constantly reminding the senior author that there are more important things in life than Mvtilus edulis L.

LITERATURE CITED

d'Hamonville, B. 1894. Les moules perlieres de Billiers. Bull. Soc. Zool. France 19:140-142.

Dubois, R. 1901. Sur la mecanisme de la forma- tion des perles fine dans le Mytilus edulis. Com- pt. Rend. Hebd. Seances Acad. Sci., Paris 133:603-605.

Dubois, R. 1907a. Sur les metamorphoses du Distome parasite des Mytilus perliers. Compt. Rend. Seances Soc. Biol. 63:334-336.

Dubois, R. 1907b. Action de la chaleur sur le distome immature de Cymnophallus margar- itarum. Compt. Rend. Seances Soc. Biol., 63:502-504.

Dubois, R. 1909. Contribution a l'etude des perles fines de la nacre et des animaux que les produi- sent. Ann. Univ. Lyon n. s., 29:1-126.

Elliot, J. M. 1971. Some methods of the statistical analysis of samples of benthic invertebrates. Freshwater Biol. Assoc. Sci. Publ. No. 25, 144

P- Engle, J. B. and V. L. Loosanoff. 1944. On the

season of attachment of larvae of M. edulis

Linn. Ecology 25:433-440. Field, I. A. 1922. Biology and economic value of

the sea mussel, Mytilus edulis. Bull. U. S. Bur.

Fish. 38:127-259. Garner, R. 1857. On the pearls of the Conway

River, North Wales, with some observations on

the natural productions of the neighboring

coast. Brit. Ass. Adv. Sci., Part 2, p. 92-93. Garner, R. 1872. On the formation of British

pearls and their possible improvement. J. Linn.

Soc. 11:426-428. Giard, A. 1907. Sur les Tre'matodes margar-

itigenes du Pas-de-Calais (Gymnophallus

somateriae Levinsen et G. bursicola. Odhner).

Compt. Rend. Seances Soc. Biol. 63:416-420.

Herdman, W. A. 1904. Recent investigations on

pearls in shellfish. Proc. Trans. Liverpool Biol.

Soc. 17:88-97. Jameson, H. L. 1902. On the origin of pearls.

Proc. Zool. Soc, London 1:140-165. Jameson, H. L. and W. Nicoll. 1913. On some

parasites of the scoter duck (Oedemia nigra)

and their relation to the pearl-inducing

trematode in the edible mussel (Mytilus edulis).

Proc. Zool. Soc, London, p. 56-63. Lebour, M. V. 1906. Notes on Northumbrian

trematodes. Rep. Scient. Invest. Northumb. Sea

Fish. Comm. 1905, p. 100-105. Lutz, R. A. 1976. Annual growth patterns in the

inner shell layer of Mytilus edulis L. J. Mar.

Biol. Ass. U. K. 56:723-731. Lutz, R. A. and B. Porter. 1977. Experimental

culture of blue mussels (Mytilus edulis L.) in

heated effluent waters of a nuclear power plant.

Proceedings of the Eighth Annual Meeting,

World Mariculture Society, January 9-13, San

Jose, Costa Rica, Louisiana State University

Press (in press). Mossop, B. K. E. 1922a. A study of the sea mussel

Mytilus edulis Linn. Contr. Canad. Biol.

2:17-48. Mossop, B. K. E. 1922b. The rate of growth of the

sea-mussel (Mytilus edulis L.) at St. Andrews,

N. B., Digby, N. S. and in Hudson Bay. Trans.

Roy. Canad. Inst. 14(3):3-22. Nicoll, W. 1906. Notes on trematode parasites of

the cockle (Cardium edule)and mussel (Mytilus

edulis). Ann. Mag. Nat. Hist. 17:148-155. Odhner, T. 1905. Die Trematoden des arktischen

Gebietes. In: Romer u. Schaudinn, Fauna Artica

4(2):291-372. Scattergood, L.W. and C. C. Taylor. 1949. The

mussel resources of the North Atlantic region.

Part II. Observations on the biology and the

methods of collecting and processing the

mussel. Comm. Fish. Rev. ll(10):ll-23. Sokal, R. R. and F. J. Rohlf. 1969. Biometry. W.

H. Freeman and Company, San Francisco,

776p. Stafford, J. 1912. On the fauna of the Atlantic

coast of Canada, third report-Gaspe,

1905-1906. Contrib. Canad. Biol. 1906-1912, p.

45-67. Steel, R. G. D. and J. H. Torrie. 1960. Principles

PEARLS IN BLUE MUSSEL 37

and procedures of statistics. McGraw-Hill Book on digenetic trematodes of the genera Gym- Company, Inc. New York, 481p. nophallus and Parvatrema. Biol. Bull. Stunkard, H. W. and J. R. Uzmann. 1958. Studies 115:276-302.

Proceedings of the National Shellfisheries Association Volume 68-1978

BLUE CRAB PREDATION ON CULTCHLESS OYSTER SPAT

George E. Krantz and John V. Chamberlin

UNIVERSITY OF MARYLAND

CENTER FOR ENVIRONMENTAL AND ESTUARINE STUDIES

HORN POINT ENVIRONMENTAL LABORATORIES

BOX 775

CAMBRIDGE. MARYLAND 21613

UNIVERSITY OF MARYLAND

DEPARTMENT OF AGRICULTURE ENGINEERING

COLLEGE PARK, MARYLAND

ABSTRACT

Many cultchless oyster spat (Crassostrea virginica) planted experimentally on natural bottoms in Maryland portions of the Chesapeake Bay in 1975 and 1976 have been destroyed by blue crabs (Callinectes sapidus). High mortalities (79 to 99%) were observed within one month after oysters ranging from (3 mm-40 mm) in diameter were placed directly on natural bottom or in trays without closed tops. Oysters in enclosed trays at these locations exhibited no unusual mortality. Several patterns of shell destruction (chipped shell edges, a single puncture over the adductor muscle, broken area of original spat attachment, and opposing notches on the shell edges) were observ- ed among the oyster shells collected from the field. The same destruction patterns were observed when cultchless oyster spat were placed in aquaria containing adult blue crabs. Cultchless oyster spat which produce high quality half-shell oysters if grown in protected containers, may be of little value when planted on natural bottom in Maryland to sustain native oyster stocks, or to obtain an economical return from a planted private lease.

INTRODUCTION

Crabs of various species are important predators on oyster spay. Stone crabs (Menippe mercenaria) are a serious problem in the States bordering the Gulf of Mexico. So are green crabs (Carcinus maenas) and mud crabs (Neopanope texana) in New England states (Galtsoff, 1964). Mud crabs were incriminated in very high mor- talities (as high as 50 percent) of Connecticut oyster spat less than 10 mm long by MacKenzie (1970). Several authors have also noted that the blue crab, Callinectes sapidus. can cause high levels of mortality in natural oyster and clam populations in the Middle Atlantic and Gulf States (Lunz, 1947; Menzel & Hopkins, 1955). Menzel &

Hopkins (1955) observed the feeding behavior of blue crabs in floating cages and found that a single crab can consume up to 19 oyster spat per day.

Castagna (1970) suggests that predation by blue crabs is the major constraint in the culture of hard clams on natural bottoms. Without some form of protection, clam mortality can approach 100 per- cent (Menzel et al, 1976; Castagna et al, 1970).

The same constraint — predation by crabs — has been encountered in our attempts to grow cultchless oyster spat on the natural bottom in the Chesapeake Bay. It is obvious that large popula- tions of blue crabs in our coastal waters could completely neutralize oyster mariculture opera- tions, especially if spat are too small when planted or if the spat are not protected. We have planted

38

BLUE CRAB PREDATIONS

39

cultchless oyster spat of various sizes to determine the optimum size at which to plant hatchery- reared spat and to insure enough survival for a positive economic return from the cost of the spat.

FIELD STUDIES

Our studies were conducted on the upper Eastern Shore of the Maryland portion of the Chesapeake Bay. Mean annual salinity of this region ranges from 8 to 12 ppt. This area has no significant populations of oyster drills, Urosalpinx cinerea and Eupleura caudata. The most probable predators on oyster spat in this region are the flat worm, Stylochus ellipticus; the mud crab, Rithropanopeus harrisii; the blue crab; and some fish species: oyster toadfish, Opsanus tau; croaker, Micropogon undulatus; spot, Leiostomus xanthorus; cow-nosed ray, Rhinoptera bonasus.

Our first planting study utilized 4 1/10 acre plots on hard sandy bottom. The bottom had very little habitat for mud crabs or other members of the oyster bed community. The plots were planted with 125,000 cultchless spat in late September, 1975, each plot receiving oysters of \ :", ' : ", V and 1" (6mm, 12 mm, 20 mm, 25 mm) in length. The spat were produced by the Dupuy (1973) technique, were removed from Mylar sheets when they were about 1 1" (5 to 15 mm), and grown in fiberglass trays in the hatchery until planted.

After two weeks on the bottom, the spat were observed by scuba divers. All of the 6 mm spat were missing and very little shell remained. Mor- tality in the other plots ranged from 30 to 60 per- cent in sample locations. By the following spring (May 1976) no live oysters could be found. All of the recovered shells had broken margins with large portions of the shell missing. We thought severe storm conditions might have tumbled the shells on the bottom, broken the edges of the shell and killed the oyster spat.

In 1976 ten different locations where the bottom was protected from wave action were planted with clutchless spat.

One such area was planted in July with cultchless oysters ranging from 19 to 38 mm long. Ten thousand spat were planted on a 100 sq. ft. area marked by stakes on an active commercial oyster bed. Within two weeks, all the clutchless spat were dead. Blue crabs were observed to be

abundant at this location whereas mud crabs were not found. Broken shells from the clutchless spat were found in the planted area.

In another location, trays of cultchless oysters ranging from 3 to 40 mm in length were placed ad- jacent to a natural oyster bar. Half the trays had fine-mesh wire tops, while half were unprotected. Within one month 99.7 percent of the unprotected cultchless spat were dead, while the protected had no detectable mortality. The shells of the dead spat had the same type of damage that was observed at previous planting locations.

These are two of the most severe cases of preda- tion we have observed but significant losses were incurred at other planting sites even though spat were planted in early November, 1076, after most blue crabs had become dormant.

PATTERNS OF SHELL DAMAGE BY THE BLUE CRABS

To confirm the hypothesis that blue crabs were involved in the losses of oysters in the field we fed cultchless oysters of various sizes to crabs in aquaria. Blue crabs from 4" to 6" (100 mm to 150 mm) carapace width were found to be capable of consuming cultchless oysters up to 40 mm in length. Blue crabs from 65 mm to 80 mm in carapace width were unable to consume cultchless oysters larger that 1" (25 mm) in length.

Methods used by blue crabs to break oyster shells vary from oyster to oyster. Small oyster spat (3 mm to 15 mm) were simply crushed and the meat separated from shell as the oyster was eaten. Frequently the entire crushed oyster was in- gested and no shell fragments could be found.

The next most common feeding strategy was to chip the shell margin with the chelae so that the mouth parts and chelae tips could extract the oyster meat. Figure 1 is a scale drawing to show this type of damage. Note that the spat shell has very irregular edges and most of the recent shell growth was removed, often in large chips. These oysters look as if they had been damaged by tumbling along a hard bottom. This was the predominant pattern of damage observed in our 1975 bottom planting of cultchless oysters.

A characteristic of cultchless spat produced by the Dupuy mylar technique is thinness of the shell over the area where the lower valve of the spat

40

G.E. KRANTZ AND J.V. CHAMBERLIN

FIG. 1. Shells of cultchless oyster spat showing blue crab damage to shell margins.

FIG. 2. Shells of cultchless oyster spat showing blue crab damage to soft, flattened areas produced on lower valves by the Mylar technique of spat collection.

was attached to mylar. This area is easily broken by the human fingernail and contributes to mor- tality while handling small spat in oyster hatch- eries. Blue crabs frequently crushed this area (Figure 2) in their attempts to open the cultchless oysters. Another approach was to crush the um- bo, or hinge area (Figure 3), which then permitted the crab to extract the oyster meat. A fairly com- mon type of observed damage was the removal of large chips from opposite sides of the oyster shell as if the crab compressed the shell laterally be- tween the digits of the chelae (Figure 4).

The most puzzling damage we observed in the planting studies, were spat with a single round hole in the shell (Figure 5). Up to 10 percent of the

shells in some samples had this type of damage. Initially we thought that another predator was in- volved in the losses of the hatchery-reared spat. However, feeding studies in the laboratory show- ed that blue crabs were quite effective at punching holes. Occassionally we found shells which had multiple holes. Figure 5 shows that the holes are not round but are slightly irregular. Frequently the inner shell surface of the hole and the "walls" showed irregular fractures on the margins. The damage differs from the more regular, smooth and slightly conical holes made by oyster drills.

Perhaps the most important observation made during the study was that to damage the cultchless oyster spat all that was required was that blue

FIG. 3. Shells of cultchless oyster spat with umbo area crushed by blue crabs.

A

FIG. 4. An example of chips removed by blue crabs from opposite edges of cultchless oyster shells.

FIG. 5. Holes punched through the shell of oyster spat by blue crab.

BLUE CRAB PREDATIONS

41

crabs pick up and manipulate the oyster spat in their claws. In this situation the edges of the cultchless spat were accessible, in contrast to spat on oyster shell cultch. The fragile shell edges of cultchless oysters were easily chipped even by small blue crabs and clutchless spat were easily manipulated to the mouth of the crab.

Laboratory feeding studies showed that blue crabs had much greater difficulty manipulating a large piece of cultch to which oyster spat were at- tached. Frequently the crab's claws were unable to contact the spat in a manner that would damage the shell. Manipulation to the mouth was definite- ly impaired by cultch of any size.

MacKenzie (1970) noted that mud crabs in Con- necticut rarely attacked an attached spat over 10 mm (about Vz in.) but they readily consumed unattached spat up to 25 mm. His field observa- tions were confirmed by our laboratory studies.

Our present mariculture strategy is to determine what size of spat on various types of cultch best survives the blue crab predation in Maryland por- tions of Chesapeake Bay. Hopefully a combina- tion of cultch type, size of cultch, and size of oyster spat can be found that will reduce mortality to a level where a positive economic return can be realized from planting hatchery-reared spat on the natural bay bottom.

ACKNOWLEDGEMENTS

The authors wish to acknowledge Dr. John Dupuy for culturing some of the oysters used in this study, Donald Meritt for assistance in field work and for oysters he raised for the study, and Mrs. Deborah Kennedy for her drawings of damaged oyster spat.

This study was supported by Maryland Depart- ment of Natural Resources Fisheries Administra- tion.

BIBLIOGRAPHY

Castagna, M. A. , G. W. Mason, and F. C. Biggs. 1970. Hard clam culture method developed at VIMS. Va. Inst. Mar. Sci. Sea Grant Advisory Pgt. No. 4. 3 p.

Dupuy, J. L. 1973. Translation of Mariculture Research into a Commercial Oyster Seed Hatch- ery. Proc. World Mariculture Soc. 677-685.

Galtsoff, R. S. 1964. The American oyster Crassostrea virginica Gmelin: Fishery Bulletin No. 64. U. S. Fish & Wildlife Service, 480 p.

Lunz, G. R. Jr. 1974. Callinectes versus Ostrea, J. Elisha Mitchell. Sci. Soc. 63: 81

MacKenzie, C. L. 1970. Causes of oyster spat mortality, conditions of oyster setting beds, and recommendations for oyster bed management. Proc. Nat. Shellfish. Assoc. 60:59-67.

McDermott, J. J. 1960. The predation of oysters and barnacles by crabs of the family Xanthidae. Proc. Penn. Acad. Sci. 34:199-211.

McDermott, J. J. and F. B. Flower. 1952. Preliminary studies of the common mud crabs on oyster beds of Delaware Bay. Conv. Addr. Nat. Shellfish Assoc. 47-50. Menzel, R. W. and S. H. Hopkins. 1955. Crabs as predators of oysters in Louisiana. Proc. Nat. Shellfish Assoc. 46:177-182.

Menzel, R. W., E. W. Cake, M. L. Haines, R. E. Martin and L. A. Olsen. 1976. Clam Mariculture in Northwest Florida: Field Study on Predation. Proc. Nat. Shellfish Assoc. 65:59-62.

PREDATION ON HARD CLAMS,

MERCENARIA MERCENARIA, BY MUD CRABS, PANOPEUS HERBSTIP

Jack M. Whetstone mid Arnold G. Eversole

DEPARTMENT OF ENTOMOLOGY AND

ECONOMIC ZOOLOGY

CLEMSON UNIVERSITY

CLEMSON, SOUTH CAROLINA 29631

ABSTRACT

Predation on hard clams planted in protected trays was studied by examination of empty clam shells and the stomach contents of potential predators gaining entry to the trays. Decapod crustaceans comprised 88.5% of the predators collected. Panopeus herbstii was the most abundant species while Callinectes sapidus and Menippe mercenaria were only occasionally found. Gastropods were collected, but examination of empty clam shells revealed no evidence of gastropod predation. Stomach analysis revealed clam shell bits in four crab species. Shell bits were found in 15.8% of the 279 P. herbstii analyzed. Occurrence of shell bits varied but were more often found in stomachs of larger crabs.

The abundance and mean size of P. herbstii in trays varied from a peak in summer to a low in winter. Clam mortalities decreased with increases in clam size and with decreases in water temperature. Frequency of shell bits in the stomachs of P. herbstii paralleled clam mortality. Some size selection process appeared to be operating in this predator-prey system.

INTRODUCTION The potential exists for improving the hard clam, Mercenaria mercenaria, industry in South Carolina. Rapid growth of clams, vast amounts of suitable tidelands, increased market for southern clams, and recent success of the Santee fishery are some of the reasons for this optimism (Eldridge, Waltz, Gracy and Hunt, 1976; Gracy, 1974; Gracy and Keith, 1975). Clam mariculture has also been demonstrated to be feasible in pilot studies in Virginia (Castagna, Mason and Briggs, 1970), but more work remains to be done. A ma- jor problem in culturing clams is predation. The

1 Technical contribution no. 1524, published by permission of the Director. South Carolina Agricultural Experiment Station.

objectives of this study were to investigate preda- tion in protected trays in an estuary in South Carolina and to identify some important factors influencing predation.

MATERIALS AND METHODS

Hatchery clams (X= 13 mm shell length, SL) were planted in 20 protected oyster trays (10 inter- tidal, 10 subtidal) containing natural sediment in May, 1975. Basket compartments (118 x 61 x 14 cm) in the trays were enclosed with 9 mm plastic netting and lined with fiberglass insect screens to retain sediment. Each tray was covered with 9 mm plastic netting to help protect against predation. This was not completely successful because poten- tial predators apparently could enter trays

42

PREDATION ON CLAMS

43

through the crack between the cover and basket compartment. The trays were located near Clark Sound, S. C. (Lat. 32° 42' 5" N, Long. 79° 52' 2" W) in an area characterized by a soft sand (20-30% silt-clay) bottom and a salinity range of 25-30%0atlowtide.

Fifteen collections of potential predators were made from May, 1975 to December, 1976. All potential predators were saved except for the col- lections in June and July, 1975, when represen- tative samples were taken from each tray. Members of other feeding groups (e.g. filter feeding polychaetes) were not collected. Field col- lections were preserved in 10% formalin until specimens could be sorted and counted. Crusta- ceans were sexed and measured to the nearest 0.1 mm. The stomach contents of each crustacean and, in most cases, all the specimens collected were examined to determine the food types. Samples of the abundant crustaceans, Panopeus

herbstii and the snapping shrimp (Alpheus spp.), were selected for examination. Samples of P. herb- stii included representatives from each sample date, tidal location, and three class intervals of carapace widths (CW): small crabs <15 mm CW; medium crabs 15.1-25.0 mm CW; and large crabs >25 mm CW. Cardiac stomachs were excised, and the individual food items were sorted as clam shell bits, arthropod parts, plant material and grit. Finely ground food items which could not be iden- tified and sediment were classified as grit. The fre- quency of each food item was expressed as percent occurrence in individuals containing food. The quantity of a food item was estimated by the points' method (Hynes, 1950).

Clam growth and mortality in each tray was determined during the sampling period. Clam den- sity was maintained in the individual trays by replacing dead or sacrificed clams with marked clams of comparable size. Empty clam shells were

TABLE 1. Potential predators collected from centages represent totals for 15 sampling dates in stomachs is represented by + ; the absence 0

20 protected trays containing clams. Numbers and per- from 5/75 through 12/76. The presence of clam shell bits and not examined ne.

				Presence
	% of	%of	%of	Clam Shell
Number	Order	Class	Total	Bits

CRUSTACEA Decapoda Panopeus herbstii Callinectes sapidus Menippe mercenaria Eurypanopeus depressus Neopanope texana Uca pugilator Eurytium limosum Pachygrapsus transversus Alpheus heterochaelis Alpheus armillatus Alpheus normanni Stomatopoda Squilla empusa GASTROPODA Neogastropoda Urosalpinx cinerea Nassarius vibex Nassarius obsoleta Eupleura caudata TOTAL

1746	—	100.0	88.9	+
1740	100.0	99.7	88.5	+
1465	84.2	83.9	74.6	+
8	0.5	0.5	0.4	+
8	0.5	0.5	0.4	+
13	0.7	0.7	0.7	+
2	0.1	0.1	0.1	0
9	0.5	0.5	0.5	0
2	0.1	0.1	0.1	0
1	0.1	0.1	0.1	0
184	10.6	10.5	9.4	0
39	2.2	2.2	2.0	0
9	0.5	0.5	0.5	0
6	100.0	0.3	0.3	0
6	100.0	0.3	0.3	0
219	—	100.0	11.1	ne
219	100.0	100.0	11.1	ne
150	68.5	68.5	7.6	ne
10	4.6	4.6	0.5	ne
56	25.6	25.6	2.8	ne
3	1.4	1.4	0.2	ne
1965

44

J.M. WHETSTONE AND A.G. EVERSOLE

collected to determine the size of dead clams and possible cause of mortality. Materials and methods not detailed above were described previously (Eldridge et al., 1976).

RESULTS

The number of potential predators found in the trays are presented in Table 1. P. herbstii, a mud crab, was the most abundant (75%) species col- lected. Predators of clams which are generally suspected, Callinectes sapidus and Menippe mercenaria, comprised less than 1% of the specimens found. Gastropods were common, but empty clam shells did not exhibit the borehole characteristic of gastropod predation. Cracked and crushed clam shells were abundant in the trays indicating forced entry, probably by crusta- ceans.

Fourteen percent of the 362 crustaceans ex- amined contained clam shell bits in their cardiac stomachs. Only four species (i.e., P. herbstii, C. sapidus, Menippe mercenaria and Eurypanopeus depressus) of the 12 species of crustaceans ex- amined contained shell bits (Table 2). There was no significant difference (PX3.05) between the oc- currence of shell bits in the four species.

Clam shell bits were more frequent in stomachs of large crabs (Table 2). Shell bits were found in significantly greater (P<0.05) percentage of large P. herbstii than were found in medium and small ones. Although percent occurrence of shell bits in P. herbstii were significantly different between sizes, volumes (i.e. point method) were not. Variation in this trend among the less abundant species can be accounted for by the number and temporal differences of the crab populations in the trays. No shell bits were observed in medium- sized E. depressus (15.1-25.0 mm CW), however crabs this size were absent until May, 1976 when clams had a 36.6 mm mean shell length. Small sample sizes also limited statistical analyses of data on the less abundant species.

Prey size, as well as predator size, apparently affects clam mortality levels. The mean shell lengths (SL in mm) of live and dead clams (i.e. empty shells) increased linearly after time of plant- ing (x = time in months). The regressions are:

live clams x SL = 1.69x + 4.76; and

dead clams x SL = 1.31x + 3.37,

with r values of 0.99 and 0.96 respectively. These linear regressions are significantly different

TABLE 2. Analysis of crab stomach contents containing clam shell bits. Percent occurrence is based on those crabs with food. Crab size class intervals are small < 25.0 mm carapace width (CW), medium 15.1-25.0 mm CW and large > 25.0 mm CW.

Carapace	Number		Clam	% Occurrence
Crab Species Width	Examined	Food	Shells	Arthropods	Plants	Grit
Panopeus herbstii Small	92	76.1	10.0	20.0	44.3	100.0
Medium	119	87.4	15.4	18.3	38.5	100.0
Large	68	97.1	31.8	31.8	43.9	100.0
Total	279	86.0	18.3	22.5	41.7	100.0
Callinectes sapidus Small	3	100.0	0	0	33.3	100.0
Medium	1	100.0	0	0	0	100.0
Large	4	100.0	50.0	50.0	25.0	100.0
Total	8	100.0	25.0	25.0	25.0	100.0
Menippe mercenaria Small	7	71.4	20.0	40.0	20.0	100.0
Large	1	100.0	100.0	0	0	100.0
Total	8	75.0	33.3	33.3	16.7	100.0
Eurypanopeus depressus Small	4	100.0	25.0	0	50.0	100.0
Medium	6	83.3	0	0	40.0	100.0
Large	3	100.0	66.7	33.3	66.7	100.0
Total	13	92.3	25.0	8.3	50.0	100.0

PREDATION ON CLAMS

45

50.

% 40J WITH CLAM SHELLS

3Q

20.

10

LJ Large Crabs 0250mm) (ZI Medium Crabs( 15 I -250mm) Ul Small Crabs(il5.0mm)

A S 1975

0 N D

J F

		n
M A M J J A
		i	97«

S 0 N D

FIGURE 1. Percent P. herbstii containing food with clam shell bits plotted against sampling month. At any date the percentage of crabs containing food with shell bits is represented by the entire histogram. Representatives of each size class interval were found in all collections except 6/75 when no large crabs were found.

(P<0.01) suggesting clam size as an integral com- ponent of predation. As clam size increased the percentage of P. herbstii containing clam shell bits decreased (Figure 1). Also the proportion of large P. herbstii with shell bits increased from 40% in 1975 to 67% in 1976. Only large P. herbstii con- tained shell bits one year after planting. Over the same period, mortality of clams decreased from a high of 30.2% to a relatively constant low mor- tality of approximately 0.5% per month by March, 1976. (Figure 2).

The abundance and mean size of P. herbstii varied throughout the experimental period (Figure 3). This variation could be due to several cohorts

in the crab population. The mean size of crabs was greatest in the warmer months (i.e. June - Oc- tober). Laboratory observations indicate that in P. herbstii 20.0 mm CW was the minimum size necessary for successful predation upon clams above 11.5 mm SL (the approximate planting size). The percentage of the crab population above this minimum successful predator size changed significantly (P<0.05) throughout the year with the greatest percentage in the summer (i.e. June, July, and August). Also the average number of P. herbstii above 20 mm CW was significantly (P<0.05) greater during the summer than fall, winter and spring. This seasonal variation of P.

46

J.M. WHETSTONE AND A.G. EVERSOLE

jIjIaIsIoInId

1975

40

30

' 2 2

20£

I0IX

JlFlMlAlMlJ |J lAlSlOlNlO 1976

FIGURE 2. Percent mortality of clams and mean shell length of live clams grown in 20 protected trays plotted against sampling date. (P. ]. Eldridge and A. G. Eversole, unpublished data).

herbstii allows speculation into some methods of predator control.

DISCUSSION Callinectes sapidus, Carcinus maenas, and Menippe mercenaria have been documented as major predators of the hard clam, Mercenaria mercenaria (Belding, 1912; Carriker, 1951; God- win, 1968; Haven and Andrews, 1957; Loosanoff, 1946; Menzel, Cake, Haines, Martin and Olsen, 1976; Menzel and Sims, 1962). Callinectes sapidus and Menippe mercenaria have been suspected of causing 100% mortality in hard clams planted in unprotected plots in Georgia and Florida (God- win, 1968; Menzel and Sims, 1962). Haven and Andrews (1957) reported C. sapidus as the major predator of hard clams planted in suspended trays in Virginia. Other decapods have been proposed as predators of hard clams. Eldridge et al. (1976) suggested mud crabs, members of the Xanthidae

family, were important predators of clams in South Carolina.

Callinectes sapidus and Menippe mercenaria comprised less than 1% of the potential predators collected and could not be the major cause of mor- tality in our culture system. Clam shell bits were found in only four species of predators, three of which were xanthid crabs. The xanthid crab, P. herbstii. appeared to be the most important predator in this system. To date no specific men- tion has been made of P. herbstii as a predator of the hard clam, although McDermott (1960) cited a predator-prey interaction between P. herbstii and Crassostrea virginica. Carriker (1959) and Landers (1954* also cited Neopanope texana. a xanthid crab, as an effective predator of seed clams.

Prey and predator size appear to be major fac- tors in the Panopeus herbstii - Mercenaria mercenaria interaction. Menzel and Hopkins

PREDATION ON CLAMS

47

(1956) observed that the size of the Menippe mercenaria limits this xanthid crabs ability to prey on oysters. Our findings indicate that larger P. herbstii preyed more successfully on clams than smaller crabs. Also, the proportion of small and medium crabs containing clam shell bits decreased as clam size increased. A major decline in mortality was observed as clams grew larger, and the percent of P. herbstii containing shell bits declined from 50% when clams averaged 15.5 mm SL to less than 5% after clams reached an average size of 38.4 mm SL. Carriker (1959) observed that a larger initial planting size significantly reduced mortality of clams in unprotected plots. Assuming that predatory activity of P. herbstii was limited by clam size, planting larger clams should reduce mortalities.

Crabs may be selecting the smaller clams in the trays. The regression for mean shell lengths of live clams against time (months) was significantly greater than the regression for the mean shell lengths of dead clams. Furthermore this regression for live clams was significantly greater than a regression representing the projected sizes of dead clams if the dead clams had the opportunity to grow throughout each sampling interval. The pro- jected size was computed by adding the mean shell length of dead clams and an increment equivalent to shell growth of live clams for the appropriate sampling interval. McDermott (1960) suspected that P. herbstii selected thinner-shelled oysters. Cake (1970) found no prey size selectivity by C. sapidus and Menippe mercenaria when offered sunray venus clams, Macrocallista nimbosa. Smaller species of crabs may not have the ability to open the larger and /or thicker-shelled molluscs. Smaller species of crabs such as P. herbstii may therefore exhibit more prey size selection than larger crab species. If the larger crabs are more capable of preying on a wide variety of clam sizes or thicknesses, then prey size selectivity probably will not be pronounced. This may explain, in part, why Cake (1970) observed no size selectivity with C. sapidus and Menippe mercenaria.

Menzel and Hopkins (1956) found lower mor- talities in oysters in Louisiana during the winter, and predation by Menippe mercenaria was not observed below 10 °C. Similarly we observed lower mortalities in the winter months when water

% S

-14

j 1 j Ia Is lo In lo jIfImIaImIjIjuIsIoInId 1973 1976

FIGURE 3. Smoothed means of carapace widths (in mm) of P. herbstii from trays plotted against mean sampling dates. Histograms show percen- tages of sampled crabs with carapace widths larger than 20 mm.

temperatures approached 8°C. Less activity by the crabs in the winter would contribute to this lower observed mortality. Lower mortality also cor- responded to the presence of fewer P. herbstii and a smaller percentage above the minimum effective predator size during winter.

Time of planting and clam size should be con- sidered in formulating any management program for extensive clam culture operations. Planting clams in the late fall should improve survival of seed because predator populations are then at lower levels. Also, there are fewer large predators, and predators are generally less active during cooler months. In addition, during this period clams have an opportunity to grow to a size that

48

J.M. WHETSTONE AND A.G. EVERSOLE

will limit predation by the smaller crabs. If selected planting times and clam sizes were used with other techniques such as aggregate protection to effectively reduce predation, clam culture can be successful and profitable even in areas where predators abound.

ACKNOWLEDGEMENTS The senior author wishes to thank Drs. P. J. Eldridge (South Carolina Wildlife and Marine Resources Department) and A. G. Eversole for the unpublished data on clam growth and mortality. Special thanks are also due Wayne Waltz and George Steele for their assistance in field collec- tions.

LITERATURE CITED

Belding, D. L. 1912. The quahog fishery of Massachusetts. Comm. Mass. Dept. Conserv. Mar. Fish. 2:1-41.

Cake, E. W., Jr. 1970. Some predator-prey rela- tionships involving the sunray venus clam, Macrocallista nimbosa (Lightfoot) (Pelecypoda: Veneridae) along the gulf coast of Florida. M.S. Thesis, Florida State University, Tallahassee, Florida, 169 pp.

Carriker, M. R. 1951. Observations on the penetration of tightly closing bivalves by Busy con and other predators. Ecology 32:73-83.

Carriker, M. R. 1959. The role of physical and biological factors in the culture of Crassostrea and Mercenaria in a salt-water pond. Ecol. Monogr. 29:219-266.

Castagna, M. A., L. W. Mason and F. C. Briggs. 1970. Hard clam culture method developed at VIMS. Va. Inst. Mar. Sci., Sea Grant Advisory Ser. No 4, 3 pp.

Eldridge, P. J., W. Waltz, R. C. Gracy and H. H. Hunt. 1976. Growth and mortality rates of hatchery seed clams, Mercenaria mercenaria, in protected trays in waters of South Carolina. Proc. Natl. Shellfish. Assoc. 66:13-20.

Godwin, W. F. 1968. The growth and survival of planted clams, Mercenaria mercenaria, on the

Georgia coast. Georgia Game and Fish. Comm. Contrib. Ser. 9:1-16.

Gracy, R. C. 1974. Management and development of the shellfish industry in South Carolina. Ann. Rept. on Project 2-179-D in Coop, with Natl. Mar. Fish. Sen/., Dept. of Commerce. Under Public Law 88-309, 22 pp.

Gracy, R. C. and W. J. Keith. 1975. Management and development of the shellfish industry in South Carolina. Final Rept. on Project 2- 179-D in Coop, with Natl. Mar. Fish. Serv., Dept of Commerce. Under Public Law 88-309, 110 pp.

Haven, D. and J. D. Andrews. 1957. Survival and growth of Venus mercenaria, Venus campech- iensis. and their hybrids in suspended trays and on natural bottoms. Proc. Natl. Shellfish. Assoc. 47:43-49.

Hynes, H. B. N. 1950. The food of fresh-water sticklebacks (Gasterosteus aculeatus and Pyg- osteus pungitius). with a review of methods us- ed in studies of the food of fishes. J. Anim. Ecol. 19:36-58.

Landers, W. S. 1954. Notes on the predation of the hard clam. Venus mercenaria, by the mud crab, Neopanope texana. Ecology 35:422.

Loosanoff, V. L. 1946. Commercial clams of the Atlantic Coast of the United States. U. S. Fish Wildlife Service Leaflet 13:1-13.

McDermott, J. J. 1960. The predation of oysters and barnacles by crabs of the family Xanthidae. Proc. Pa. Acad. Sci. 34:199-211.

Menzel, R. W., E. W. Cake, M. L. Haines, R. E. Martin, and L. A. Olsen. 1976. Clam mariculture in Northwest Florida: Field study on predation. Proc. Natl. Shellfish. Assoc. 65:59-62.

Menzel, R. W. and S. H. Hopkins. 1956. Crabs as predators of oysters in Louisana. Proc. Natl. Shellfish. Assoc. 46:177-184.

Menzel, R. W. and H. W. Sims. 1962. Experimen- tal farming of hard clams, Mercenaria mercenaria in Florida. Proc. Natl. Shellfish. Assoc. 53:103-109.

Proceedings of the National Shellfisheries Association Volume 68-1978

SEASONAL ASPECTS OF THE BIOLOGY, DISTRIBUTION AND

RELATIVE ABUNDANCE OF THE DEEP-SEA RED CRAB

GERYON QUINQUEDENS SMITH, IN THE VICINITY OF THE

NORFOLK CANYON, WESTERN NORTH ATLANTIC1,2

Paul A. Haefner, Jr.3

VIRGINIA INSTITUTE OF MARINE SCIENCE GLOUCESTER POINT, VIRGINIA 23062

ABSTRACT

Deep-sea red crabs were collected from demersal trawl surveys of Norfolk Canyon and an adjacent open slope area in the Chesapeake Bight of the western North Atlantic Ocean. The surveys were made in each of four seasons over a period of three years. The 2539 red crabs caught ranged from 16mm to 143 mm in carapace length (CL). Rela- tionships between CL and carapace width (CW) were derived for 308 males and for 269 females. Wet weight to CL relationships were derived for 238 males and for 142 females.

Red crabs were contagiously distributed within the total depth range of capture (200- 1800m) as well as within the 300-1000 m range of most consistent catches. They were equally abundant in canyon and slope regions in the four seasons. The majority of the population inhabits bottom water deeper than 400 m arid overlain by cold, well- oxygenated water.

Over 50% of all red crabs were larger than 96 mm CL (114 mm CW), the minimum size presently acceptable to processing plants. This proportion varied by sex, season and depth. Potentailly marketable male crabs constituted 70% of total males caught, whereas less than 25% of females exceeded 96 mm CL. Seasonally the proportions were consistent for males, but varied for females. The proportion of large males was consis- tent over most of the depth range (200-1600 m). Most of the larger females were cap- tured in water shallower than 600 m.

An inverse relationship between water depth and crab size was evident for females from 200-1500 m; for males only in the 200-500 m depth range. The mean size of males caught deeper than 600 m was fairly stable.

Females were more abundant than males in samples shallower than 600 m: males dominated catches in deeper water.

A spawning cycle is suggested although ovigerous females were captured in all seaons. Peak incidence of ovigerous females occurs in November characterized by a high percentage of late stage eggs and a peak incidence of recently extruded eggs. The

Research cruises supported by National Science Foundation Grant GA-37561, J. A. Musick. principal investigator, by the University of Virginia Institutional Grant Program; and NOAA, Office of Sea Grant (NO. 04-3-158-49) tor P.A.H. participation.

Contribution No. 830, Virginia Institute of Marine Science.

Present Address: Department of Biology, Rochester Institute of Technology, Rochester, New York 14623.

49

50

P. A. HAEFNER, JR.

ovarian cycle of non-ovigerous females complements the spawning pattern. From June through November, an increase in incidence of developing ovaries urns accompanied by a decrease in advanced and mature ovaries. Absence of mature ovaries in November complements the peak incidence of ovigerous females at that time.

Relative density of red crabs was estimated for the Norfolk Canyon area and com- pared with values from other areas along the east coast of the United States.

INTRODUCTION

Interest in Geryon quinquedens Smith as a marketable species has been slowly increasing. In- itial explorations (Schroeder 1959; McRae 1961) proved that red crabs were readily captured with trawls off the east coast of the United States. More recent surveys have been concerned with estimating fishery potential and evaluating harvesting methods, both in the United States (Gray4; Haefner and Musick, 1974; Ganz and Her- rman5; Wigley, Theroux and Murray, 1975) and in Africa (Dias and Machado6; LeLoeuff, Intes and LeGuen, 1974; Intes and LeLoeuff, 1976) Technological and economic aspects of harvesting have also been under investigation (Varga, Dewar and Anderson, 1969; Meade and Gray, 1973; Holmsen and McAllister, 1974)

Other than the data obtained by Haefner and Musick (1974) and Wigley, et al. (1975), which were restricted to one period of sampling, obser- vations on the biology of Geryon have been super- ficial. This paper supplements the survey of Haefner and Musick (1974) by presenting seasonal data on distribution, relative abundance and reproductive biology of red crabs in the Chesapeake Bight area of the western North Atlantic Ocean.

MATERIALS AND METHODS Deep-sea red crabs were collected during demer- sal fish trawl surveys of Norfolk Canyon and an

4 Gray, G. W., Jr. 1969, Investigation of the basic life history of the red crab (Geryon quinquedens). Rhode Island Div. Cons. Completion Rep. (P.L. 88-309, Proj. 3-46-R), pp.36.

s Ganz, A. R. and J. F. Hermann, 1975. Investigations into the southern New England red crab fishery. Rhode Island Dept. Nat. Res. Div. Fish. Wild]. Mar. Fish. Sec. pp. 78.

6 Dias, C. A. and J. F. S. Machado. 1974. Preliminary report on the distribution and relative abundance of deep-sea red crab (Geryon sp.) off Angola, 8 p. In: Scientific papers pre- sented to the second session of the International Commis- sion for the Southeast Atlantic Fisheries (Madrid, December 1973), M. E. Bioceanol. Pescas, Angola, 12, pp. 75.

adjacent open slope area in the Chesapeake Bight region of the western North Atlantic Ocean (Figure 1.) Four surveys, one in each season, were made during a period of three years using Univer- sity of Miami vessels R/V Columbus O. Iselin and fames M. Gilliss (Table 1). The sampling gear con- sisted of 15.1 m (headrope) semi-balloon, 4- seam shrimp trawls equipped with plastic mud- rollers and steel China V-doors. The nets were nylon of the following stretch mesh: 44 mm body, 37 mm intermediate, 36 mm codend and 12 mm inner liner. Thirty-minute tows were made in depth strata less than 1000 m; tow time was one hour at deeper stations.

Initially, an equal number of tows was to have been made in each of four depth strata (75-150 m, 150-400 m, 400-1000 m, 1000-2000 m) in the can- yon and slope areas. Variations in actual depth of tows, encounters with bottom types prohibitive to trawling and cruise time limitations combined to alter the program. The realized effort is presented in Table 1. Mean trawl depths were calculated from depths recorded at start and finish and at 3- minute intervals during each tow.

All red crabs were processed at sea. Carapace width (CW, distance between the tips of the lateral spines) and carapace length (CL, distance from the diastema between the rostral teeth to the posterior edge of the carapace, along the midline) were measured to the nearest millimeter. The latter measurement was emphasized based on recom- mendations of Gray". Accuracy in weighing, which was done aboard ship, depended on sea state conditions. In most cases, weight was record- ed to the nearest gram; some larger specimens were weighed to the nearest decigram.

Females were examined for evidence of egg ex- trusion and hatching. Color of eggs was noted for most females. External eggs from selected ovigerous females were examined microscopically to relate developmental stage to egg color. Developmental stages of ovaries of selected non-

SEASONAL DEEP-SEA RED CRAB

51

76c

75<

74'

73c

38c

37c

36°

76°

75°

74c

38c

37<

36<

73c

FIGURE 1. Chart of the Cheapeake Bight region of the western North Atlantic region showing shelf (SH), canyon (C) and open slope (S) trawl survey areas in the vicinity of the Norfolk Canyon. Isobaths in fathoms.

ovigerous females were classified as described in Haefner (1977).

Temperature and salinity of near-bottom water strata were monitored at trawl stations as well as at independent hydrographic stations. A variety of instruments was used including bathyther- mographs, expendable bathythermographs, a salinity-temperature depth recorder and reversing thermometers. Dissolved oxygen concentration of near-bottom water samples was determined by Winkler titration.

RESULTS AND DISCUSSION

Profiles of near-bottom (within 5-100 m of recorded depth) temperature and dissolved ox- ygen concentration (D.O.) for the four cruises are

presented in Figure 2. In all cases the data for can- yon and slope areas are combined. The June plot reflects only data taken in conjunction with trawl stations; the other three plots include data from hydrographic and trawl stations. Only data from stations shallower than 1600 m are included.

Throughout the year temperature of near- bot- tom water was indirectly proportional to depth at depths exceeding 100 m (Figure 2). A large temperature gradient from 13 °C to 6 °C existed between 100 and 500 m. Below 500 m, temperature decreased gradually to 4° C at 1600 m.

Oxygen minima were associated with the ther- mal gradient (Figure 2). They were particularly obvious in June and November when D.O. values

52

P. A. HAEFNER, JR.

TABLE 1. Norfolk Canyon demersal trawl fishing effort, expressed as number of tows, in the ca)iyon (C) and on adjacent open slope (S) by season and by depth strata between 200 m and 1800 m.

Depth

Stratum

(m)

Spring Summer Fall Winter

4-16 June 73 9-20 September 75 13-25 November 74 22-31 January 76 QS CS CS CS

201- 300 301- 400

401- 501- 601-

701- 801-

500 600 700 800 900

901-1000 1001-1200 1201-1400 1401-1600 1601-1800

TOTAL

6	1	3	1	1	2	2	2
1	5	3	2	2	1	2	2
1	0	0	0	0	0	1	1
0	0	0	0	0	1	2	0
1	1	2	2	3	3	0	1
3	4	1	1	2	0	1	2
0	0	2	1	0	1	1	1
1	0	2	2	0	1	0	0
0	2	2	1	1	0	0	1
0	1	1	0	0	0	0	0
2	0	0	1	3	1	2	0
3	1	2	1	0	3	0	2
18	15	18	12	12	13	11	12

of 4.6 mg/liter were detected in near-bottom water within the 200-400 m depth range. Between 400 m and 800 m D.O. increased to 8 mg 'liter, which persisted to a depth of 1600 m.

Except for shelf stations shallower than 200 m, salinity of near-bottom water was consistently between 35 % and 36 % .

The 2539 red crabs caught during these cruises ranged from lb mm to 143 mm in carapace lenth (CL). For purposes of comparing carapace length with carapace width (CW) reported in other papers, the following relationships were derived for 308 males, CL range 18-138 mm:

CW = 8.74 + 1.09 CL, r = 0.98 and for the 269 females, CL range 23-116 mm: CW = 11.04 + 1.0b CL, r = 0.98

Male crabs attain a larger size than females. The largest male crab caught measured 143 mm CL and weighed 1200 g; the largest female was 123 mm CL and weighed 510 g. The following wet weight- length relationships were derived for 238 males ranging from 3 g to 1200 g:

log male weight = -3.58 + 3.14 log CL, r = 0.99 and for 142 females ranging from 4 g to 510 g:

log female weight = -3.13 + 2.88 log CL, r = 0.97.

Although the size frequency distributions (Figure 3) indicate that the red crab is obviously not vulnerable to the trawl gear, certain modal groups (CL) were recognized in nearly every season. The most obvious were the 90-130 mm group for males and the 75-110 mm group for females. Other modal groups, such as 50-90 mm for males and 50-75 mm for females were less ob- vious. Wigley, Theroux and Murray (1975), in a survey extending from offshore Maryland north- eastward to Corsair Canyon on Georges Bank, observed similar modal groups for male crabs. Their female size frequency curve, however, did not indicate a well-defined intermediate group. On the other hand, their well-defined peak for red crabs less than 30 mm CL was missing from Nor- folk Canyon samples.

Analysis of the Norfolk Canyon data using a three-point moving average, with subsequent calculation and plotting of percent cumulative fre- quency on probability paper was encouraging. Thirteen to fifteen smaller, less pronounced modes were detected which suggest molt classes within the larger modal groups. If, indeed, the smaller modes were representative of molt classes, a 6 to 11 percent range in growth increment (CL) is sug- gested. This range is comparable to four of the five

SEASONAL DEEP-SEA RED CRAB

53

TEMPERATURE (C) AND DISSOLVED OXYGEN (rug/liter) 4 6 6 10 12 14 16 .4 6 6 10 12 14

LONG CARAPACE WIDTH (mm)

FIGURE 2. Seasonal profiles of temperature (T) and dissolved oxygen (D.O.) of near-bottom water in relation to depth in the Norfolk Canyon and adjacent shelf and slope area.

observations (6.7, 8.5, 8.9, 10.4, 18.1%) of Gray4 of crabs molting in the laboratory. One red crab molting at VIMS increased in length by 13.8%.

FIGURE 3. Seasonal size frequency distributions of Geryon quinquedens. Males are plotted above the line; females below. Black areas indicate ovigerous individuals and those with egg rem- nants on the pleopods.

Vertical arrows indicate present acceptable market size. Scales for carapace length and carapace width are included for comparative pur- poses.

Crabs less than 70 mm CL were poorly represented in the June sample (3% of the catch), but they made up an increasingly larger propor- tion of the catch through September (13.8%) and November (22.9%). The proportion in January (16.4%) was largely due to the absence of crabs less than 50 mm CL.

Over 50% of all Norfolk Canyon red crabs were larger than 96 CL (114 mm CW), the minimum size presently acceptable to processing plants (Wigley, Theroux and Murray, 1975). This pro- portion varied by sex, season and depth (Table 2).

54

P. A. HAEFNER, JR.

in

O

m

(N

m

ro

o

m

m

~

o»

#

in

(N

in

rH

00

tN

rH

nO

o

o

O

«*'

o

(N

f

(N

(N

rM

10

"^

E Ol

m

m

^i"

^f

o

O

(N

r-i

On

nO

ra

O

in

>, -G

<

PL,

Z

o

00 in

—

O rH

rH

m — i

m

rO

-^

rH

o rH —

"^

H

nO

o

O

m

rH

t^

t-^.

ro

rg

o

r-4

r-)

O H

^

ID

<N

nO

t~^

o

t^

ro

O

O

o

ro

o

o

Q

41

in

O

00

t^

<o

m

m

00

IN

00

NO

fN

•£

13

a.

2

o

00

^f

vO

rH

o

t^

IN

In

m

On

rH

<N

^

2

no

00

^o

r 1

m

T

m

n^

rH

t-H

00

"^3

,—i

"*

rO

r-t

rH

•^<

^

T—t

-c

00

<N

o

O

rA

tX)

U

#

rO

o'

(N

d

o

o

O

C>

m

rH

<N

m

r*

cs

fl

>-■

£ at

o

m

(N

t^

"*

OO

On

o

r^

o

o

o

*r

ci

PL,

z

H

Vj

Vj

ro

rO

£

<

T-*

rO

E

D

CnJ

IN

t^

o

On

IN

nO

NO On

Z <

jy

1< 00

oo

m

00

nO Pn) rO

in 00

in nO

O

1 ■

13

2

o

o

fN

m

^H

"3"

0

o

in

o

o

o

"3"

bp

z

1"H

IN

m

(N

in

00

0

_s

r-i

00

[^

0

o

'S'

11

sP

i — i

in

"-1

o

d

o

IN

a a

es

E

UJ

01

o

ro

O

I V

IN

■t

ro

o

O

o

rH

o

ro

CO 2

PL,

z

(N

00

m

rH

bo

UJ

ro

ro

ro

NO

C^

r-4

o

rM

rO

^

>

£

o

rs

d

(N

d

^1"

oo

d

rO

^t"

O z

S 2

t— 1

rO

O

0 rH

00

m

nO

00

o rH m

o

nO On

o

nO ro

H

z

— 1

r0

(N

NO

r—i

ro

T~i

r~t

cu

fS

r~*

in

Cu

00

«

K 0)

tN

00

r—*

o

O

^-*

0)

#

in

ro

<>

o

o

in

o

o

d

E3

(N

ro

rH

rs]

5) S3.

£ 01

in

o

O

O

fS

■*

o

o

00

NO

o

o

■f

T-

CO

UL,

z

o

m

r0

!-H

r 1

r-*

— I

cs

2

rs

.

uj

m

m

O

■*

On

O

On

o

'S1

"bo 3

H

#

(N

o

ro

r-<

O

On

ra

o

o

ro

^Oj

vO

0

^>

tx

On

IS

nO

00

In

CS

LO

13

tj

2

00

rH

O

o

^O

M"

"*

m

m

m

o

rH

On

c 01

z

fS

CO

rH

Tf

o

ro

r-(

In

T—t

— ■

rO

-a i>

■^<

o

rr>

On

M"

3

QJ

#

<*

in

*t

r^

o

o

O^

O"

■3"

in

r-l

r*

ro

C

Ctf

'3 a*

E O)

o

00

<N

o

t^

if

o

rH

O

o

rH

o

■*

u_

z

00

ro

00

T—t

c

UJ

<N

*#

0

z

>.

D

IN

q

(N

m

r-4

IN

<v

^

■^

d

<N

t>!

(N

d

u

11 13

o

o

t^

NO

IN

00

o

S

o

m

(N

o

^0

Cv

o

00

o

o

o

o

nO

^.

z

[H

rH

rn

— i

T—l

0

Cu

rH

^—<

n

-a

S

S

—

o

Q

Q

o

Q

o

o

Q

o

o

o

o

6

o

o

O

o

o

o

o

O

o

o

o

o

rJ

(N

ro

-r

m

0

[ V

00

On

O

o

-r

D

00

< o

_CJ

_c

i

-^

r-<

rH

^

^

H

M

-^

T—4

rH

rH

rH

rH

—

r~*

rH

rH

rH

o

O

o

o

o

O

O

o

O

O

O

o

(N

ro

-T

in

vO

IS

00

o

o

rg

■*

0

r-i

D

r-<

r^

r-<

—

Potentially marketable male crabs constituted 70% of total males caught whereas less than 25% of the females exceeded 96 mm CL, a reflection of the larger ultimate size of male crabs. Seasonally, the proportions were consistent for males, but varied for females. A high proportion of females (39%) in June reflected the absence of small modal groups at that time; a low of 7.4% in November was related to a reduced modal group of large females.

These figures are markedly higher than the percentage (24%) of harvestable crabs in the northeast red crab survey (Wigley, Theroux and Murray, 1975). However, any comparisons bet- ween the two geographic areas must be qualified. Differences in size frequency distributions of the two populations are most likely related to the dif- ferences in sampling methods.

Comparison with other geographic regions is further complicated by difference in the acceptable harvestable size as well as in method of catch. For example, the minimum size of red crabs for the Ivory Coast of Africa is 84 mm CL (100 mmCW) (Intes and LeLeouff, 1976). In their survey 70% of the total catch of red crabs (38% of all males) was legal. If 84 mm CI instead of 96 mm CI was the ac- cepted minimum market size in the Untied States, the percentage of potentially marketable male crabs caught in the Norfolk Canyon survey would increase from 70% to 78% and the percentage of marketable females would increase from 50% to 65%.

The proportion of large males in the population was consistent over most of the depth range (Table 2). Most of the larger females were found in relatively shallow water, <600 m.

The inverse relationship between water depth and crab size observed by Wigley, Theroux and Murray (1975) was also evident for females from 200 m to 1500 m and for males only in the 200-500 m depth range in the Norfolk Canyon region (Figure 4). The mean size of male crabs caught deeper than 600 m was fairly constant. Apparent seasonal variations shown in Figure 4 are due to lack of samples (Table 1) and/or low number of specimens in certain depth strata.

Red crabs were contagiously distributed within the Norfolk Canyon survey area within the total depth range of capture (220-1800 m), as well as within the 300-1000 m range of most consistent

SEASONAL DEEP-SEA RED CRAB

55

I40t

I30--

I20--

<s>

£ iio-

UJ

o

<

Q. <

<

< a. <

< O

100 90 - 80--

30

100 90

x

o 80

70--

60-

50-

40

MALE

200

DEPTH (m)

FEMALE

X o

200 400 600 800 1000

DEPTH (m)

1200

1400

400 600 800 1000 1200 1400 1600

--I60

-150 "g

E --I40 -*

x

-130 q

--I20 ui o

--no <

Q. <

■-100$ o

20

110

M00 E

-90

--80

i

O

+ 70 <

Q- <

cr <

o

1600

FIGURE 4. and for all poses.

Geryon quinquedens. Relation of mean size of males and females to depth according to season seasons combimed. Scales for carapace length and width are included for comparative pur-

catches. Because there was an excess of samples with zero catch and the variance exceeded the mean number of crabs per tow (X= 23.3 within 200-1800 m; X= 39.5 within 300-1000 m), catch data were transformed (Elliott, 1971) as an index of abundance: Y = "M'4|> where x is the

number of red crabs caught per 30 minute tow in any given stratum and n is the number of tows at any given depth. Data from the hour long tows were adjusted to correspond to half hour tows.

Red crabs were significantly (P = 0.001) more abundant on the adjacent slop than in Norfolk Canyon proper (Figure 5). Although stratifying depth by 100 m and 200 m intervals emphasizes voids in actual sampling depth (Table 1), it is evi- dent that the distributional depth range is broad, from 200 to 1800 m (actual fishing depth range was 210 m to 1725 m). Most hauls were successful in catching crabs within the 400-1000 m zone (ac- tual fishing depth 405 m to 1042 m) where relative

56

P. A. HAEFNER, JR.

B i

2/5 "-

m

3/ o

1/1

XL

-B-£-

1

37/52

301- 401- 501- 601- 701- 801- 901- 1001- 1201- 1401- 1601- j

300 400 500 600 700 800 900 1000 1200 1400 1600 1800 DEPTH (m)

FIGURE 5. Relative abundance of Geryon quin- quedens by area, expressed as index of trans- formed mean number of crabs per half-hour tow within each depth stratum. Fraction above or below each bar is the ratio of the number of sta- tions at which red crabs were captured to the total number of stations in each stratum.

abundance was highest. The actual distributional depth range is most likely 210-1565 m because the single individual recorded for the 1601-1800 m stratum is suspected to be a wash-down from a previous sample.

In light of the temperature-dissolved oxygen- depth relationship, it is clear that the majority of the red crab population inhabits bottom waters deeper than 400 m which are overlain by cold, well-oxygenated water (Figures 2,5). A smaller percentage of tows caught crabs in the shallower «400 m) areas where warmer, oxygen minimum water existed.

Females were more abundant than males in samples from depths shallower than 600 m; males dominated the catches in deeper water (Figures 6,7H). This segregation of the sexes by depth has

CANYON AND SLOPE

I

a*

Ph

201- 301- 401- 501- 601- 701- 801- 901- 1001- 1201- 1401- 1601-

300 400 500 600 700 800 900 1000 1200 1400 1600 1800

DEPTH (m)

FIGURE 6. Relative abundance of Geryon quin- quedens by sex, expressed as index of transformed mean number of crabs per half-hour tow within each depth stratum.

been observed in other areas. In Rhode Island waters Ganz and Herrman (1975) observed a lack of females between 700 m and 915 m; the greatest percentage of males was caught in 685-1110 m. The survey of Wigley, Theroux and Murray (1975) indicated a marked reduction of the number of females in water deeper than 500 m, where males predominated. Basicallly the same pattern was observed off the coasts of Angola

CARAPACE WIDTH (mm) 20 40 60 80 100120 140 160 170

CARAPACE WIDTH (mm) 20 40 60 80 100 120 140 160 1 70

\j

	100	E	r~
	80		i
	60		/
TOTAL	40		"*\ /SEPTEMBER
	20

20 40 60 80 100 120 140 160 20 40 60 80 100 120 140 160

.- '00

60-

CANY0N U> 40

< 20

"\ / N0VEMB

20 40 60 80 100 120 140 I60K 20 40 60 80 100 120 140 160

\ / SLOPE

V

SIUU1 £80-	G		/
O- 60-			j
			/ JANUARY
20-		V	/

20 40 60 80 100 120 140 160 20 40 60 80 100 120 140 160

CARAPACE LENGTH (mm)

/

CARAPACE LENGTH (mm)

DEPTH (m)

FIGURE 7. Percentage of Male Geryon quin- quedens by size and depth. Data expressed as probability curves for total catch, area, season and depth. Scales for carapace length and width are included for comparative purposes.

(Dias and Machado") and Abidjan (Ivory Coast) (LeLoeuff, Intes and LeGuen, 1974; Intes and LeLoeuff, 1976): the largest male catch was recorded from 600-700m; females were largely restricted to 300-400 m. A 400-600 m zone of tran- sition was recognized wherein males and females occurred in variable, but nearly equal numbers.

The combined, four-cruise male: female ratio was 1.28:1, significantly different from the

SEASONAL DEEP-SEA RED CRAB

57

theoretical 1:1 (Chi-Square, P = 0.01). The ratio varied from cruise to cruise with significant dif- ferences (M:F) in September (1.77:1), November 2.51:1) and June (0.73:1). The 0.88:1 ratio in January was not significant.

The presentation of sex ratio data in the form of probability curves (Wenner, 1972) is more mean- ingful. Red crab data were tabulated into 10 mm size classes and the sex ratio (as percent males) calculated within each class and plotted. The "anomalous" pattern shown for total red crab catch (Figure 7A) is not appreciably different than that for location (Figure 7B, C) or season (Figure 7D-G). Scatter about the curves is due to low numbers of individuals in the smaller size classes. The "anomalous" pattern is consistent with the Fisher theory (Wenner, 1972) in which a 1:1 off- spring production is favored by natural selection, but a unique deviation from the 1:1 ratio is il- lustrated for older or larger animals.

The observed pattern is principally a function of physical size. The shape of the curve for red crabs exceeding 100 m CL is determined by the larger maximum size of male crabs. Although growth rate and longevity are presently unknown for Geryon, it is known that females reach maturity at 65-75 mm CL, (Haefner, 1977) and continue to molt to eventually attain at least 116 mm CL. Males most likely reach maturity near 65-75 mm CL but attain a larger maximum size than females, either by more frequent molts, larger molt in- crements, living longer or by a combination thereof.

The paucity of males and an accumulation of females distributed about 85 mm CL (Figure 3) shows up as a dip in the probability curve. The ac- cumulation of females in the large modal size group bracketing 85 mm CL could be due to longer intermolt intervals, smaller growth in- crements, terminal anecdysis or inhibition of molting by vitellogenesis and spawning (oviposi- tion). Such inhibition is known to occur in other crustaceans (Adiyodi and Adiyodi, 1970: Swartz, 1976).

The study of the reproductive biology of females was based on individuals exceeding 70 mm CL (85 mm CW), which includes most of the mature females (Haefner, 1977). This manipula- tion insures that most of the reproductively

mature segment of the female population is treated and reduces data bias due to variations in catch of smaller size classes. (Figure 3).

Wigley, Theroux and Murray (1976) related col- or of the external egg to developmental stage. Within the red-orange, brown, dark brown, pur- ple and black spectrum, they judged newly deposited eggs to be light red or orange; eggs become darker as they ripen.

A microscopic examination of eggs from 11 ex- truded egg masses (sponges) from Norfolk Can- yon crabs indicated a large variation in color within the earlier stages of development (Table 3). Because of the obvious difficulty in assigning specific development stage by color alone, the observed colors were grouped into early (A-C; orange-brown) and late (D; burgundy-purple- black) categories according to the scheme in Meredith (1952).

A spawning cycle is suggested for red crabs in the Norfolk Canyon area although ovigerous females were captured in all seasons (Figure 8).

35n

30-

o 25

e

E O 20

CO

<

5-

395 176 126 282

m

E R

1

□ >

A-C

JUN SEP NOV JAN

FIGURE 8. Seasonal percentage of early (A-C) and late (D) developmental stages of extruded eggs on ovigerous individuals and of egg remnants (ER) on total catch of female Geryon quinquedens >70 mm CL captured in the vicinity of Norfolk Can- yon. Blank areas indicate unidentified egg stage. Sample size indicated by numbers above the bars. Egg stages are described in Table 3.

58

P. A. HAEFNER, JR.

TABLE 3. Developmental stages of external eggs of Geryon quinquedens. Size range based on measure- ments of 10 eggs from each sponge. Modified from Meredith (1952).

Stage

Number of

Sponges

Examined

Description

Size Range (Mm)

Color

B C

D

Egg early to late blastoderm; nearly spherical. Minute eyes visible in larger embryos.

Pre-larval embryo. Eyes large. Abdomen and ap- pendages well-developed, free from head. Red- brown pigmentation present, light over entire body, intense in eyes, cardiac and gastric regions of carapace, and in dorsal abdominal melano- phores.

640-740

Eye V3-V2 size of Stage D embryo; eye the only 710-780 pigmented (brown) area. Yolk abundant. Abdo- men free. Abdomen without melanophores.

720-820

Orange

Red-orange

Red-brown Orange

Red-orange

Burgundy Purple (black)

The presence of females bearing egg remnants and the low incidence of ovigerous individuals in lune suggests a high incidence of egg hatching between January and June. This is based on the assumption that the presence of egg remnants indicates recent spawning. However, at the present time it is not known how long egg remnants remain on the pleopods. Hatching continues through the sum- mer and is accompanied by an increase of ovigerous females bearing early stage (A-C) eggs. The peak incidence of ovigerous females occurs in November. This is characterized by a high percen- tage of females with late stage (D) eggs and a peak incidence of recently extruded (A-C) eggs. The percentage of ovigerous females in the population remains high in January, and although it is somewhat less than that observed in November, it is similar in the proportional makeup of early and late stage eggs.

Information on incidence and relative abun- dance of red crab larvae in plankton samples would help to resolve the presence or absence of a spawning cycle, but such data are presently unavailable or incomplete. Winter and spring samples off the coast of New Jersey have revealed very few first and second zoeal stages (P.O. Smyth, VIMS, personal communication). This in- dicates some hatching has occurred in winter and spring. Analyses of summer and fall samples are incomplete.

The ovarian cycles of non-ovigerous females (Figure 9) complements the spawning pattern. From June to September, there was no major change in ovarian stage. In November, an increase in the incidence of developing (intermediate) ovaries was accompanied by a decrease in advanc-

I00n

E 6 O

>

o

o

z

80

60

40-

20-

24 49 30 53

31

r~

MAT

-J ADV.

INT.

I I IMM.

JUN SEP NOV JAN

FIGURE 9. Seasonal proportion of ovarian developmental stages (MAT, mature; ADV, ad- vanced; INT, intermediate; IMM, immature) among samples of non-ovigerous female Geryon quinquedens >70 mm CL captured in the vicinity of Norfolk Canyon. Sample size indicated by numbers above the bars.

SEASONAL DEEP-SEA RED CRAB

59

ed/mature ovaries. The lack of mature (ripe) ovaries in November complements the peak in- cidence of ovigerous females at that time, sug- gesting that most of the mature females have ex- truded their eggs. Continued progressive develop- ment results in the higher proportion of advanced and mature gonads observed in January.

The temperature regime appears to determine the distribution of ovigerous females, the majority of which were captured in 400-800 m depths where temperatures exceeded 5° C (Figure 2.) Ovigerous females were captured over an extensive depth range in Norfolk Canyon although the evidence shown for the 1201-1400 m stratum is due to only two individuals in a small sample (Figure 10). The highest incidence of occurence clearly fell within the 401-800 m depth range. Females with external eggs in early stages of development were found deeper than 300 m and were most frequently found within the 401-700 m depth range. The late

£ 25

95

I

18

201- 301- 401- 501- 601- 701- 801- 901- 1001- 1201-

300 400 500 600 700 800 900 1000 1200 1400

DEPTH (m)

H.

□ ■ ■

□ ,

FIGURE 10. Depth distribution of ovigerous red crabs by egg development stage (A-C, D) and of females with egg remnants (ER) on pleopods, ex- pressed as percentage of total catch of females >70 mm CL. Blank areas indicate unidentified egg stage. Numbers above the bars indicate number of females in sample. Egg stages are described in Table 3.

TABLE 4. Density of red crabs (no. /hectare) captured by 13.7 m trawl and standing crop (in thousands of crabs) in vicinity of Norfolk Canyon (36°30' —37°10 'N).

Depth (m)	September 1975 No. /ha Crop lOOO's	J. No. /ha	anuary 1976 Crop lOOO's	No. /ha	All Crop lOOO's
200-300	8.5		67.0	0		0	4.3	33.9
300-400	31.1		199.2	5.6		35.9	14.5	92.9
400-500	—		—	60.7		35.6	60.7	35.6
500-600	—		—	125.4		793.8	125.4	793.8
600-800	18.1		210.0	16.3		189.2	17.5	203.1
800-1000	23.6		337.3	51.4		734.5	30.4	434.4
1000-1200	8.6		162.5	14.4		272.1	9.2	173.8
1200-1400	31.9		703.7	—		—	31.9	703.7
1400-1600	0		0	0		0	0	0
1600-1800	0.1		4.9	0		0	0.1	4.9
All	13.2		2599.1	19.6		3859.3	15.8	3111.1

stage of development was most abundant at 401-800 m depth, but it was the only stage present in the 201-300 m depth stratum. Females with egg remnants were caught within an extensive depth range but were markedly more prevalent in shallower water (201-400 m) than in any other depth.

These data do, to a certain extent, corroborate the findings of Wigley, Theroux and Murray (1975) of an increase of spawning females with depth (to 640 m for their northeast sample), but there is no evidence from the Norfolk Canyon data to indicate seasonal differences in the in- cidence of early and late stage eggs with depth.

60

P. A. HAEFNER, JR.

The presence of stage D eggs at 201-300 m (Figure 10) is due to September collections only and the high incidence at 701-800 m is due primarily to June collections. Only the females with egg rem- nants are restricted to two seasons, but the data for June and September (Figures 8, 9) are consis- tent with that shown in Figure 10. This preponderance of females with egg remnants in water warmer than 8° C (in June, they were cap- tured in 300-400 m depth stratum) may not be coincidence. The migration of ovigerous females, with eggs in advanced stages (D) of development, into shallower, warmer water may enhance egg development and hatching. Although mortality of eggs and larval development as a function of temperature has not been determined, it is known that eggs survive and hatch at 15° C and larval development proceeds at temperatures as warm as 21 °C (Perkins, 1973).

Relative density of red crabs was estimated for the September 1975 and January 1976 surveys (Table 4) when tow distances could be determined with reasonable accuracy from Loran C readings. Area fished was then computed from tow distance and the effective fishing width of the net, estimated to be 7.6 m (Haedrich, Rowe and Polloni, 1975). This accounted for 55.4% of headrope length, a value near the lower end of the range of values computed for other types of trawls (Griswold, Kurlyandsky and Twohig, 1971).

The relative densities were then used to com- pute standing crop estimates for the immediate fishing zone (Table 4). Area of each depth stratum within the zone was determined with a planimeter using a base chart constructed by E.P. Ruzecki (VIMS).

The relative densities of red crabs in the area of Norfolk Canyon were 50-95% lower than those observed for more northeasterly areas of the con- tinental shelf and slope (Grassle et al., 1975; Wigley, Theroux and Murray, 1975) although a fair degree of overlap exists within the range of values of these three studies.

The low red crab density in the Norfolk Canyon area is most likely a reflection of the inefficiency of the collecting gear. The ineffectiveness of trawl nets in capturing contagiously distributed, motile megabenthic invertebrates has been revealed in certain gear comparison studies (R. Cooper and J. Uzmann, NMFS, personal communication). Ad-

mittedly, crab density calculations based on trawl catches are tenuous because of a number of variables (Wigley, Theroux and Murray, 1975). However, in order to compare their data with mine, I converted their trawl catches to relative densities (Table 5) and found that they fell short of the values determined by photographic sled. Their trawl densities, ranging from 19% to 78% of the

TABLE 5. Comparison of red crab densities de- termined by photographic sled and those com- puted for 4.° m trawl for total geographic zone. Data source: Wigley, Theroux and Murray (1975). Red Crab Density Percent of

Depth (m)	Sled	Trawl	Sled Density
229-320	21.1	16.5	78.4
320-412	258.3	190.2	73.7
412-503	273.2	136.5	50.0
503-640	282.0	53.8	19.1
640-914	91.4	45.5	49.8
914-1280	61.0	37.2	61.0
1280-1646	10.9	0

photographically estimated densities, were com- puted from an assumed vessel speed of 1.75 knots for a duration of 30 minutes, and an effective fishing width of 3 m for the 4.9 m (16 ft) trawl (Haedrich, Rowe and Polloni, 1975).

The small trawl is apparently more effective in capturing red crabs than is the large trawl. Although the larger net sweeps more than twice the area of bottom in a given unit of time, it is unable to capture twice the number of crabs, assuming the same density of crabs is available (vulnerable) to the gear. Although it is difficult to accept the apparent superiority of the 4.9 m trawl over the 13.7 m trawl in capturing red crabs (com- pare Tables 4 and 5), the fact is that it yielded larger catches on an equal effort basis (0-218, x = 33; 0-197, x = 23 crabs/30 minute tow, for small and large trawl, respectively). The relative ineffec- tiveness of the large net may be due to a number of factors such as lift of the net from the bottom and rolling over crabs in depressions. Net avoidance, however, is not likely a factor; a small net could be more easily avoided than a larger net.

There remains the possibility that the red crab stock in the Norfolk Canyon area may be relative- ly sparse compared with stocks distributed north-

SEASONAL DEEP-SEA RED CRAB

61

easterly along the shelf and slope. This dilemma will not be resolved until the stocks are more effec- tively evaluated by tagging, gear comparison and calibration.

Other studies either cannot be compared to ours or to that of Wigley, Theroux and Murray (1975) because of variation in gear used (Schroeder, 1955, 1959). Furthermore, they present conflicting values. For example, Grassle et al (1975) deter- mined red crab densities using photographic tech- niques in DSRV Alvin. Their values from two dives (180 crabs/ha in 4^5-490 m depth; 19.4 crabs- ha in 992-1000 m) represented 64% and 22%, respectively, of the photographic sled den- sities reported by Wigley, Theroux and Murray (1975) within the same geographic zone.

ACKNOWLEDGEMENTS

Gratitude is expressed to the following person- nel at VIMS who shared their time and expertise: E.P. Ruzecki, hydrography; E.L. Wenner, R.K. Carpenter and J.R. Bloom, Jr., data collection; F.J. Wojcik, data reduction; M. Williams, art work; W.A. Van Engel and M. Roberts, manuscript review.

LITERATURE CITED

Adiyodi, K. G. and R. G. Adiyodi. 1970. En- docrine control of reproduction in decapod Crustacea. Biol. Rev. 45:112-165.

Elliott, J. M. 1971. Some methods for the statistical analysis of samples of benthic in- vertebrates. Freshwater Biol. Assoc. Sci. Publ. No. 25, pp. 148.

Grassle, J. F., H. L. Sanders, R. R. Hessler, G. T. Rowe and T. McLellan. 1975. Pattern and zona- tion; a study of the bathyal megafauna using the research submersible Alvin. Deep-Sea Res. 22:457-481.

Griswold, B. L., Y. Kurlyandsky and P. J. Twohig. 1971. Performance of trawls used in joint U.S.-U.S.S.R. groundfish studies, Int. Comm. Northwest Atl. Fish. Res. Doc. 71/117, pp. 26.

Haedrich, R. L., G. T. Rowe and P. T. Polloni. 1975. Zonation and faunal composition of epibenthic populations on the continental slope south of New England. J. Mar. Res. 33(2):191-212.

Haefner, P. A., Jr. 1977 Reproductive biology of the female deep-sea red crab, Geryon quin- quedens, from the Chesapeake Bight. Fish. Bull. 75(1):91-102.

Haefner, P. A., Jr. and J. A. Musick. 1974. Obser- vations on distribution and abundance of red crabs in Norfolk Canyon and adjacent con- tinental slope. Mar. Fish. Rev. 35(l):31-34.

Holmsen, A. A. and H. McAllister. 1974. Technological and economic aspects of red crab harvesting and processing. Univ. Rhode Island Mar. Tech. Rep. No 28. pp.35.

Intes, A. and P. LeLoeuff. 1976. Etude du crabe rouge profond Geryon quinquedens en Cote d'lvoire. I. Prospection de long du talus con- tinental; resultats des peches. Doc. Scient. Cen- tre Rech. Oceanogr. Abidjan. VII(1):101-112.

LeLoueff, P., A. Intes and J. C. LeGuen. 1974. Note sur les premiers essais de capture du crabe profond Geryon quinquedens en Cote d'lvoire. Doc. Scient. Centre Rech. Oceanogr. Abidjan. V(l-2):73-84.

McRae, E. E., Jr. 1961. Red crab explorations off the northeastern coast of the United States. Com. Fish. Rev. 23(5);5-10.

Meade, T. L. and G. W. Gray, Jr. 1973. The red crab. Univ. Rhode Island Mar. Tech. Rep. Ser. No. 11, pp.21.

Meredith, S. S. 1952. A study of Crangon crangon L. in the Liverpool Bay area. Proc. Trans. Liver- pool Biol. Soc. 58:75-109.

Perkins, H. C. 1973. The larval stages of the deep sea red crab, Geryon quinquedens Smith, reared under laboratory conditions (Decapoda:- Brachyrhyncha). Fish. Bull. 71(l):69-82.

Schroeder, W. C. 1955. Report on the results of exploratory otter trawling along the continental shelf and slope between Nova Scotia and Virginia during the summers of 1952 and 1953. Papers Mar. Biol. Oceanogr. Deep-Sea Res., Suppl. Vol. 3:358-372.

Schroeder, W. C. 1959. The lobster, Homarus americanus, and the red crab, Geryon quin- quedens, in the offshore waters of the western North Atlantic. Deep-Sea Res. 5:266-282.

Swartz, R. C. 1976. Sex ratio as a function of size in the xanthid crab, Neopanope sayi. Amer. Nat. 110:898-900.

62 P. A. HAEFNER, JR.

Varga, S., A. B. Dewar and W. E. Anderson. Wigley, R. L.. R. B. Theroux and H. E. Murray.

1969. Survival of red crabs held on ice and ■ 1975. Deep-sea red crab, Geryon quinquedens,

refrigerated air. Halifax Dept. Fish. For. Tech. survey off northeastern United States. Mar.

Rep. No. 2, pp.3 Fish. Rev. 37(8):1-21. Wenner, A.M. 1972. Sex ratio as a function of size

in marine Crustacea. Amer. Nat. 106:321-350.

Proceedings of the National Shellfisheries Association Volume 68-1978

FISHERY BIOLOGY OF SPINY LOBSTER

(PANULIRUS ARGUS) of the GUAJIRA PENINSULA OF COLOMBIA, SOUTH AMERICA, 1969-1970

H. J. Squires1 and G. Riveros2

UNDP/FAO PROYECTO PARA EL DESARROLLO DE LA PESCA MARITIMA EN COLOMBIA, BOGOTA, D.E.

ABSTRACT

The fishery for Panulirus argus of the Guarjira Peninsula of Colombia is reported to average 550 tons (total weight) a year since 1969. However, an annual catch of 20 tons from traps in an estimated area of 40 km2 in 1971 is evidence of a potential yield of 1,000 tons annually from the total area of about 2,000 km1 of smooth bottom where turtle grass and coralline outcroppings occur in depths of 5-50 m. Spawning (and possibly molting) in almost all mature females was estimated to occur every two mon- ths. With molt increments of about 6mm cl. hypothetical gains from original weight on molting are approximately 35% for one, 70% for two. 100% for three and 150%for four molts on the average. The period of oogenesis (development of ova in ovaries) is about equal to embryogenesis (development of embryos in eggs on pleopods) and most females are ready to spawn again when eggs are hatched. About 300,000 eggs are car- ried at 75 mm cl and 1 million at 110 mm cl. Females carry eggs as early as 64 mm cl but about 50 % were immature at 71 mm cl. A minimum size of 76 mm cl (at least) is recommended for this area where smaller ones are sometimes exploited.

INTRODUCTION

Spiny lobsters (Panulirus argus and P. laevicauda) are found all along the coast of the Guajira Peninsula, the eastern part of the Colom- bian coast adjacent to Venezuela (Figure 1). A commercial fishery for lobsters occurs west of Cabo de la Vela where there is a shelf gradually sloping to the 50 m contour and averaging 20 km in width. Where lobsters are found is estimated to be 130 km long and 15 km wide, an area approx- imately 2,000 km2. All this area is not necessarily fished by the Guajiros, a tribe of Arawak Amerin- dians who dive reportedly to about 10 m deep (us-

Present addresses:

'122 University Avenue, St. John's. Newfoundland Canada

2Departamento de Planacion, Ministerio de Agricultura, Bogota D. E.

ing face masks only), nor by other coastal fisher- men who use Florida-type wooden traps in depths of 10-25 m. The landings reported since 1969 (Table 1) average about 550 tons (live weight) a year, a small but significant amount when com- pared with the total annual Caribbean catch of 13,000 tons (Gulland, 1970). The fishery provides an important cash income for the coastal peoples of the Guajira.

The coastlands of the Guajira Peninsula are low-lying and semi-desert with several large lagoons, some of which are used for solar salt pro- duction and have rocky or sandy substrate, while others are fringed at least partly by mangroves and have mud substrate. Just offshore throughout the area are extensive beds of turtle grass (Thalassia testudinum) interspersed with smooth sandy bottom and outcroppings of coralline rock

63

64

H.J. SQUIRES AND G. RIVEROS

3°W

R IOH ACH A ■'Camarones

11«

COLOMBIA

L

/

PENINSULA DE LA GUAJIRA

VENEZUELA

Spmy lobster t i shi ng area

'/W^fl© 6 8°

FIGURE 1. Map of the Guajira Peninsula of Colombia showing fishing grounds for spiny lobsters, Panulirus argus and P. laevicauda, and place names used in the text.

or fossil oyster reef. Immediately west of Cabo de la Vela is an indentation of the coast (Figure 1) where large numbers of small lobsters are reported throughout the year, and it is possible that current patterns encourage settlement of lobster larvae from the plankton. Navigation charts show cur- rents of up to 1.2 knots setting westward off Cabo de la Vela. These would be under the influence of northeast and easterly trade winds in the area. Also near the coast are weak counter currents (0.9 knots) setting eastward along the shore of the Guajira Peninsula. It is likely, therefore, that a weak anti-cyclonal gyre is set up just west of Cabo de la Vela.

The present study of the spiny lobsters in this area was undertaken to make an estimate of their

potential yield to a fishery and to determine some essential features of their biology such as repro- duction, growth and feeding habits. -

TABLE 1. Annual catches of Panulirus argus from the Guajira Peninsula of Colombia, 1969-74.

Year

Total weight tons

1969 1970 1971 1972 1973 1974

800 870 540 300 270 420

BIOLOGY OF SPINY LOBSTER

65

mJJ

$ 5

FIGURE 2. Histogram of monthly catches of Panulirus argus handled by a processing com- pany, including those caught by Florida-type wooden traps and diving near Riohacha, Guajira Peninsula of Colombia, 1971-72,

METHODS

Monthly trips were made to Riohacha (Figure 1) where freezing and cold storage facilities were located and the spiny lobster catch being process- ed for export (total cold storage was about 20 tons and freezing capacity about 3 tons daily in 1970). Data on catch were supplied by the processing company and INDERENA (Instituto de Desarrollo de los Recursos Naturales Renovables).

Measurements and examinations of the spiny lobsters were made from random samples of live catches brought to the company for processing. Samples were taken by setting aside all specimens from a segment of a heap of lobsters emptied from burlap bags on the concrete floor of the processing room. Each lobster in the samples was measured and examined as follows: Lengths and weights to the nearest mm or g.

Each lobster was weighed whole (TW) on a commercial balance (accurate to about 5 g);

carapace length (CL) from the anterior edge of the carapace between the two large supraorbital spines to the posterior edge in the mid-dorsal line with vernier calipers; total length (TL) on a measuring board (stretched with back against the board) with the supraorbital spines against the headboard (3 cm high) and the length read where the tip of the telson reached on the board; the ab- domen (tail) was then separated from the thorax (head) and weighed on the balance (AW), then stretched on the measuring board back down with the anterior edge of the first segment pressed against the headboard and the length (AL) read where the tip of the telson reached on the board; the greatest width of the abdomen (at the second segment) was taken on the measuring board by pressing the ventral edge of the pleuron of one side against the headboard and reading the width where the edge of the pleuron of the other side reached on the board. Examination for maturity, etc.

Ovaries were removed from the thorax and their colour recorded. Greater diameter of a few ova from part of the ovary was measured under magnification (10X) on a transparent mm grid. When eggs were carried on the pleopods they were removed before weighing the abdomen. The color of the egg mass was recorded and the phase of em- bryo development noted under magnification (10X). Hardness or newness of the shell, presence of epifaunal growths, etc., were also recorded (New shell was defined as recently molted but suf- ficiently rigid to permit capture and handling. It was bright in color and free from epifauna or discoloration and buckled under light pressure). Stomachs seen to contain food were taken out and preserved in formalin for later examination.

Temperatures and salinities were taken with a portable salinometer from a beach site, from the head of the jetty at Riohacha and from lobster fishing stations by the R/V CHOCO (March 9-16, 1970.

Landings from lobster traps and from diving for one year were provided by one of the lobster pro- cessing companies at Riohacha.

BIOLOGY

Length-weight relationships.

Regression equations for total length at each carapace length in males and females were as follows:

66

H.J. SQUIRES AND G. RIVEROS

Male TL = 2.34 CL + 39.4 (N = 500); Female TL = 2.40 CL + 47.9 (N = 493).

The rate of increase in total length relative to carapace length appeared to be greater in females than in males over the range of commercial sizes (Figure 3). However, in spite of the large numbers used an appreciable overlap in the regression coef- ficients is apparent in a "t" test (Spiegel, 1960) showing a range of 2.24 - 2.44 in males and 2.29 - 2.51 in females so that both could be similar (Table 2).

TL ■ 2-40CL* 39-4 N >493

%■

66 86 106 126

CARAPA I IENGIH-""

FIGURE 3. Regression of Total Length (TL) on Carapace Length (CL) in male and female Panulirus argus from the Guajira Peninsula of Colombia.

Regression equations for abdomen length relative to carapace length give divergent slopes as follows (Figure 4):

Male AL = 1.34 CL + 32.1 (N = 500);

Female AL = 1.56 CL + 17.1 (N = 493). The slope for females is the steeper indicating that the increase in length of abdomen relative to carapace length is greater in females than in males. The ranges in coefficients shown by a "t" test sup- port this, being 1.27 - 1.41 in males and 1.51- 1.61 in females (Table 2).

■ I. sec L + 17'1

1-34CL432 I N> 600

Xr

66 «6 106

Carapace length min

FIGURE 4. Regression of Abdomen Length (AL) on Carapace Length (CL) in male and female Panulirus argus from the Guajira Peninsula of Colombia.

TABLE 2. Regression equations and "t" test of re- gression coefficients of length and weight relation- ships of Panulirus argus from the Guajira Penin- sula of Colombia. Calculation of "t" test from Spiegel (1961):

Coefficient a, ±

^/N^I

m

N = 500 (M) or 493 (F); and s, , =

where t = 1.65; S,2 — fl|Sxy

N

Sx =

Regression equations

Male TL = 2.34 CL + 39.4

Female TL = 2.40 CL + 47.9

Male AL = 1.34 CL +32.1

Female AL = 1.56 CL + 17.1

Male TW =3.29 AW -3.5

Female TW = 3.15 AW -20.9

Range of regression coefficients at 95% confidence limits 2.24—2.44 2.29 — 2.51

1.27 — 1.41 1.51 — 1.61

3.28 — 3.30 3.13 — 3.17

Regression equations for total weight (TW) at each abdomen (tail) weight (AW) are useful for converting the "tail" weight of commerce to total

BIOLOGY OF SPINY LOBSTER

67

weights for estimates of yields from lobster populations. They are as follows:

Male TW = 3.29 AW -3.5 (N =500);

Female TW = 3.15 AW — 20.9 (N = 493).

(Figure 5)

In these equations the coefficients have ranges of 3.28 - 3.30 and 3.13 - 3.17 in males and females, respectively (Table 2) so may be considered to be different at the 95% confidence limit. For practical purposes the conversion of tail weight to total weight may be done by a factor of 3.2 if it is known that the numbers of males and females in the catches are equal (Squires, 1974). On the average the tail weight is about 31% and 33% of the total weight in males and females, respective- ly, or 32% with males and females combined (Table 3).

150 250 350 450 550

»BD0 HEN WEIGHT 19!

FIGURE 5 Abdomen Panulirus Colombia.

. Regression of Total Weight (TW) on (Tail) Weight (AW) in male and female argus from the Guajira Peninsula of

2500 2000

1500-

lOOO 800

f 600.

r 500

o

" 400 *

4 300

Loq T W.-5-655.2-79 2 LogCL N:4»J I »0 94

109 TWs. 2- 287* 2-5 78 Log CL NmSOO '-0-90

40 SO 70 100 150 200

CARAPACE IENGTH-MM

FIGURE 6. Regression of Total Weight (TW) on Carapace Length (CD in male and female Panulirus argus from the Guajira Peninsula of Colombia.

Equations for obtaining total or abdomen weights from carapace lengths (Riveros, 1972) have less practical use. They are as follows:

Male Log AW = 2.5146 Log CL — 2.6782

(N = 500); Female Log AW = 2.4331 Log CL — 2.4302

(N=493). (Figure 7).

Male Log TW = 2 . 5780 Log CL — 2 . 2873

(N = 500); Female Log TW = 2.7921 Log CL — 2.6555

(N=493) (Figure 6).

Maturity

Size at first maturity. In females the minimum size of mature lobsters (bearing eggs on the pleopods) was 64 mm CL in these samples. However, the minimum size found with large ova ready to be laid was 55 mm CI, but if this lobster molted before spawning (which is likely) it would be about 61 mm CL when carrying eggs. Only 5% of the females were immature at 71 mm CL and no juveniles or immatures were more than 94 mm CI

68

H.J. SQUIRES AND G. RIVEROS

TABLE 3. Average abdomen and total weiglits at average carapace and abdomen lengths in 450 male and 450 female Panulirus argus from the Gnajira Peninsula of Colombia.

Carapace	Abdomen	lengths	Abdomen weights	Total	weights	Percent-	AW of TW
length	Male	Female	Male	Female	Male	Female	Male	Female
mm	mm	mm	g	g	g	g	%	%
51	100	97	41	53	131	146	31	36
56	107	104	52	67	168	190	31	35
61	114	112	65	82	210	237	31	35
66	120	120	70	99	256	291	31	34
71	127	128	95	119	309	354	31	34
76	134	136	113	140	368	420	31	33
81	140	143	132	170	431	515	31	33
86	147	151	153	180	500	574	31	33
91	154	159	177	217	579	663	31	33
96	161	167	203	247	664	757	31	33
101	167	175	230	279	757	858	30	33
10b	174	182	260	314	852	968	31	32
111	181	190	292	352	957	1088	31	32
116	187	108	326	392	1069	1214	30	32
121	194	206	362	434	1187	1346	30	32
126	201	214	400	479	1319	1530	30	31

*

* 100

I

° 80-1

Log AW » -24302*2-4331 Log CL N*493

Log A W = -2.6782*2-5146 Log C M.SOO

30 40 60 801OO

C AH APACF LENGTH MM

FIGURE 7. Regression of Abdomen (Tail) Weight (TW) on Carapace Length (CL) in male and female Panulirus argus from the Guajira Peninsula of Colombia.

(Riveros, 1972). Abdomen widths were greater in proportion to carapace length by about 6% after maturity (Table 4).

In males the smallest matures were 66 mm CL and very few juveniles (5%) were larger than 85 mm CI (Riveros, 1972). The 66 mm CI cor- responds to an average tail weight of 79 g or 2.9 oz in males and 99 g. or 3.6 oz in females. Minimum accepted commercial size is a 4 oz (114 g) tail

TABLE 4. Average width of abdomen in female Panu- lirus argus when immature and when mature; expressed as a percentage of CL and TL (total length).

Carapace	Total	Abdomen width
length	length	as percent CL	of TL	Maturity
mm	mm	%	%
64	194	78	26	Mature
		72	24	Immature
67	200	75	25	Ovigerous
		69	24	Immature
70	207	71	24	Mature

BIOLOGY OF SPINY LOBSTER

69

which would be from a male of 76 mm CL and a female of 71 mm CL. (A minimum size of 76 mm CI was recommended to Government for inclusion in legislation to protect the lobster resource).

Male maturities. After males became mature (with large vasa deferentia purplish in color) they continued to retain full maturity irrespective of season (Riveros, 1972). Maturities could be categorized into three phases (Table 5) with con- siderable overlap in sizes of specimens.

Female maturities. Four phases of maturity of ova in ovaries (oogenesis) were defined as follows: I, immature or juvenile; II, maturing; III, mature and IV, spent and recovering (Table 6). In all phases there were oocytes or small ova present in the ovarian stromae, but these were beginning to be predominantly orange in color and slightly larger in Phase IV.

Macroscopic phases of embryo development (embryogenesis) in eggs on pleopods could be seen

TABLE 5. Phases of maturity in 500 male Panulirus argus from the Guajira Peninsula of Colombia, 1969- 70.

Phase Category Testes Vasa deferentia Diameter Spermatophore Range 50% of

condition size and ofv.d. glands of of phase

and color color 5th legs CL at CL

mm mm mm

>1 Small 40-85 65

1-2 Medium 62-110 83

3-4 Large 66-158 100

I	Immature	Small.	Small,
		translucent	translucent
II	Maturing	Medium,	Medium.
		opaque	opaque
		yellowish	purplish
III	Mature	Large,	Large,
		opaque	purplish
		whitish

TABLE 6. Phases of maturity of ova in ovaries (oogenesis) of Panulirus argus from the Guajira Peninsula of Colombia, 1969-70.

Phase	Category	Ovary size and color	Oocytes or small ova	Greater diameter of ova mm	Range of CL in category mm	50% of phase at CL mm
I	Immature	Small, whitish translucent or	Present, translucent	0.1-0.2	40-90	68
II	Maturing	cream to pink Orange predominating, but some	Present, translucent	0.3-0.4	50-105	85
III	Mature	opaque and whitish Bright red	Present, translucent	0.4-0.5	55-130	85
IV	Spent and recovering	Dull reddish orange, few large red degenerating	or opaque Not present; small orange ova predominating	0.2-0.3	65-125	85

70

H.J. SQUIRES AND G. RIVEROS

TABLE 7. Macroscopic phases of maturity of embryos (embryogenesis) in eggs on pleopods o/Panulirus argus from the Guajira Peninsula of Colombia, 1969-70.

Phase

Category

Color of eggs	Greater	Phase of
	diameter	ovary
	of egg	development
	mm
Bright red	0.5	IV
Orange	0.6	II
Brownish	0.7-0.8	III

I Full yolk, no eye spot

II Eye spot a flat black crescent

III Eye spot round, limb structure visible; prenaupliosoma

under low magnification (10X). Three phases were defined: the first in which no eye spot was present, the second with the eye spot merely a slit or black crescent and the third when the eye spot was rounded and limbs of the prenaupliosoma could be distinguished (Table 7). These phases cor- responded with Phases IV, II and III, respectively, of ovarian development indicating that by the time embryos were ready to hatch the ova were almost ready to be extruded. In a very few large females carrying eggs the ovaries showed only an early phase of development. Incidentally, the ear- ly phases of embryo development in eggs on pleopods were seen more frequently than the later phase (Table 8).

Abdomen width as an indication of maturity. Immature females had narrower abdomens than mature females of the same CL. As a percentage of CL average abdomen widths were about 67% in

TABLE 8. Incidence of macroscopic phases of em- bryo development (Table 6) in samples of Panul- irus argus (August, 1969 to May, 1970) from the Guajira Peninsula of Colombia.

Month	Phase I	Phase II	Phase III	Number
	(red)	(orange)	(brown)	examined
	%	%	%

Aug	67	19
Sept	64	14
Oct	64	36
Nov	-	-
Dec	50	36
Jan	-	-
Feb	100	0
Mar	0	0
Apr	-	-
May	67	0

15

21

0

14

0 100

33

27 14 14

14

3

1

immature and 72% in mature females. In the 234 females included in this comparison the change from immature to mature abdomen widths oc- curred at about 70 mm CL (Table 4).

Fecundity. Counts of eggs from 19 females from these samples by Herazo (1971) gave average numbers of just over 300,000 at a size of 75 mm CL and more than one million at 110 mm CI (Figure 8).

12-

10-

o

z a-\

in o o

CO

2

3 Z

4-

2-

70 90

CARAPACE

I 10 LENGTH

— i

130

MM

FIGURE 8. Number of eggs at each Carapace Length (CL) in Panulirus argus from the Guajira Peninsula of Colombia (Counts of eggs from pleopods of 19 females).

BIOLOGY OF SPINY LOBSTER

71

Stomach contents

Of the 93 stomachs examined (only those with food are included) 40 percent contained fish re- mains (bone and tissue fragments). Crustacean re- mains (fragments of crabs and shrimps) were in 30% of the stomachs, and molluscan shells (bivalves and gastropods) in 20%. The rest com- prised low incidence of fragments of echinoderms and plants. Sand was present in small quantities in all. Molluscan shells and echinoderm remains were frequently seen in intestines but only stomachs were examined in detail.

Shell condition

The highest percentages with new shells were in May (18%), July (6%) and September (10%). In other months the percentages varied from 1-6% (Table 9).

Temperature and salinity

Average water temperatures on the fishing grounds were 27 C from 1-10 m deep and 24 C from 10-20 m during periods of observation (Squires et al., 1974). Salinities observed were

34-37%0.

TABLE 9. New shells in monthly samples of Panulirus argus from the Guajira Peninsula of Colombia, 1969-70.

Month

May

lune

July

Aug

Sept

Oct

Nov

Dec

Jan

Feb

Mar

Apr

May

Percent with	Number
new shell	examined
%
18	164
0	10
8	68
3	91
10	143
8	72
1	69
0	21
1	73
0	21

195

ESTIMATE OF YIELD AND DENSITY OF SPINY LOBSTERS ON THE FISHING GROUNDS

A comparison of landings from trap fishing and from diving indicates that in 1971 and 1972 the divers contributed more than 85% to the landings (Figure 2). In the month to month fluctuations in landings September-November and March-June produced 8,000-16,000 kg a month while July and August had less than 3,000 kg and December the least with only 1,500 kg (Riveros, 1972). These fluctuations were reportedly the results of changes in fishing effort because of weather conditions or tribal activity of the Guajiros.

Fishing with traps, operated over an area of ap- proximately 40 km-' near Riohacha (about 20 km by 2 km), caught about 20,000 kg in 1971. About 200 traps were hauled daily with not less than three days between hauls for any one trap. The maximum traps in use was 1,000. Catch per area was:

20,000 kg 40 km2

or 500 kg per km*

Assuming that lobsters could have been obtained from a total area of potential fishing grounds be- tween depths of 5 and 45 m (conservatively 2.000 km- in area) between Cabo de la Vela and Camarones (Figure 1), the annual yield would be about 1000 tons. The total catch in 1971 was reported as 540 tons (tail weight =170 tons).

From a potential yield of 1,000 tons and an average weight of about 400 g for each spiny lobster at commercial sizes, the density of lobsters on the grounds would be about 1 to every 400 m2. This estimate is obtained from the total number of commercial lobsters in 1,000 tons (about 2,- 500.000) plus an equal number of pre-commercial sizes, and assuming that they were all evenly distributed over the shelf area.

DISCUSSION

The fishery

Average

5 Total

927

Although the average catch of fishermen using traps off Riohacha was approximately 2 tons each it did not exceed the average catch of the Amerin- dians diving for lobsters. The latter do not have a strong incentive to catch more. Their need for cash income is more than satisfied by the rate of fishing, and they fish only when there is no fiesta

72

H.J. SQUIRES AND G. RIVEROS

or other tribal activity. Although their "free" div- ing (with face mask only and with no fins) is relatively effective compared with trapping, they cannot dive during windy weather when the sand is stirred up and visibility is low in the water and sharks are more dangerous. Trapping would reduce the hazards. The investment in larger boats and traps, however, would require a much larger catch to provide the same level of income to these fishermen.

Fishery regulations prescribe a minimum size, a "veda" or closed season and non-retention of egg- bearing females. Because several market outlets are available, the Guajiros sell only to buyers who will take all sizes of lobsters. The closed season is not observed. There is, however, a natural season all but closed to the divers during about three months of heavy winds, and since there are uniformly warm water conditions throughout the year, there is no season such as a breeding season when closure would be useful. Also, since most adult females are either carrying eggs or ready to lay eggs, saving the egg-bearing ones would not be different from saving the non-eggbearing. The on- ly regulation, therefore, that would be useful is the one related to minimum size which would add ap- preciably to the total weight of lobsters harvested each year if kept strictly (Table 10). A minimum size suggested is 76 mm CL which could be in- creased in subsequent years if accepted by the fishermen.

Weight gains per molt.

Hypothetical weight gain per molt may be estimated from average weights at given carapace

lengths when the increment in length with each molt is known (Squires, 1970; Squires et al, 1971; Riveros, 1972). The average molting increment of 8 mm CL estimated by Riveros (1972) may be ex- cessive in view of the estimates given by others for this species. Munro (1973) reviewing several works from different areas suggests 30 mm per year in four molts; Peacock (1974) for Antigua and Barbuda, 21-35 mm in 4-5 molts and Beau- mariage and Little (1975) 15 mm, presumably in two molts in Florida. Olsen et al (1972) documented slower growth in females than in males: 4-7 mm per molt compared with 4-8 mm. In this paper, 6 mm increments in CL per molt in males and females is taken for calculations of weight gains (Table 10).

Starting at 59 mm CI (178 g) weight gains of males would be 35% in one, 63% in two, 105% in three and 142% in four molts. Corresponding gains in females would be 39%, 74%, 111% and 159% of the original weight of 190 g at 59 mm CI. Percentage gains in abdomen weight were slightly less (Table 10). Olsen et al (1972), however, postulates that molting in females is less frequent than in males so that increases in weight would take place over a longer time.

Demonstrating weight gain in lobsters (if left to molt) is a way to convince fishermen how much weight is lost to a fishery if lobsters smaller than regulation size are taken. Also it shows them that small lobsters will soon gain weight if left on the grounds where they may be caught again later after release.

TABLE 10. Hypothetical gains in weight of male and female Panulirus argus using average CL increments of 6 mm in each molt and an original CL of 59 mm (Calculated from regression equations of total weight (TW) and abdomen weight (AW) on carapace length (CD). (Figures 6 and 7).

CL	Average	Average	Percent gain on original	weight
	total	weights	abdomen weights	Total	weights	Abdomen weights
	Male	Female	Male	Female	Male	Female	Male	Female
mm	g	g	g	g	%	%	%	%
59	178	190	70	85
65	240	265	87	107	35	39	25	26
71	290	330	100	126	63	74	43	48
77	365	400	121	147	105	111	73	73
83	430	490	138	174	142	150	97	104
89	515	600	le>2	20o	189	216	131	142

BIOLOGY OF SPINY LOBSTER

73

Reproductive potential.

Chittleborough (1974) found that individual mature females of P. longipes cyngus spawned ap- proximately every second month when kept in aquaria under high temperature conditions similar to those prevailing in tropical areas. Berry (1971) also noted repeated spawnings of P. homarus ap- proximating once every two months while the warm temperatures of summer prevailed in South Africa. Squires (1973) suggested that reproductive cycles under tropical conditions were short, and that spawning in mature female lobsters of the Guajira occurred every 2.5 months throughout the year. Olsen et al. (1972), from direct observations of female P. argits in the Virgin Islands during underwater studies, remarked that "females may reproduce several times a year".

During the present investigations the comple- tion of development of embryos in eggs (em- bryogenesis appeared to be approximately equal to the development of ova in ovaries (oogenesis). Also, the monthly percentage of mature females carrying eggs was frequently in excess of 50% (seven times out of eleven), and the average percentage was 50% in the eleven samples examin- ed (Table 11). Although data were obtained from fairly small numbers and samples were not obtain- ed in some months, every second month of 1970-71 had the greater percentage carrying eggs (Table 11). Since the samples were taken almost exactly one month apart and from approximately the same area, the effects of synchrony in bi- monthly spawning might be indicated from these data. Apparently most mature females spawned every second month.

New shells.

Although an average of only 5% of shells ex- amined appeared to be new, some were present in most months (Table 9). The low percentages may have been caused by the following: the inability of the examiner to recognize new shells, the selection by divers against soft shells, discarding of injured soft-shelled lobsters before bringing to the factory and possibly by recently molted lobsters being more secretive in behaviour. As indicated by Chit- tleborough (1974), it is likely that molting follow- ed hatching of larvae from the eggs, and preceded egg- bearing. Where spawning was frequent,

TABLE 11. Percent of mature female Panulirus argus ovigerous in monthly samples from the Guajira Peninsula, Colombia, 1969-70.

Months	Percent	Number of Mature
	ovigerous	femali	;s examined
	%
Apr	66		29
May	49		35
June	63		8
July	-		-
Aug	73		44
Sept	56		43
Oct	73		40
Nov	-		-
Dec	40		35
Jan	18		11
Feb	53		27
Mar	27		11
Apr	-		-
May	23		64
Average	50 Total		347

therefore, lobsters with new shells would be pre- sent in all months. In the first six months of this study (May-October, 1969) each alternate month had a higher percentage of new shells (Table 9). This appeared to correspond with the months following those in which peaks of spawning oc- curred (Table 11).

ACKNOWLEDGEMENTS

We wish to thank Mr. Gabriel Durana, Manager of "Crustaceos de la Guajira" at Riohacha and Mrs. Durana for their hospitality and assistance; also Mr. Bernard Herazo for data on egg counts.

LITERATURE CITED

Beaumariage, D.S. and E.J. Little. 1975. Status report of Florida's research on spiny lobster. Proc. Gulf & Car. Fish. Inst. 28:102-107.

Ben-Tuvia, A. y C. E. Rios. 1970. Informe de un crucero del CHOCO a la isla de Providencia y los bancos adyacentes de Quitasueno y Serrana

74

H.J. SQUIRES AND G. RIVEROS

en los territories insulares de Colombia. Pro- depesca en Colombia Communicaciones 2, 45

P-

Berry, P.F. 1971. The biology of the spiny lobster Pamilirus homarus off the east coast of South Africa. Invest. Rep. Oceanogr. Res. Inst. 28:1-75.

Chittleborough, R. G. 1974. Western rock lobster reared to maturity. Aust. J. Mar. Freshwat. Res. 25:221-225.

Gulland, J. A. 1970. The fish resources of the ocean. FAO Fish. Tech. Pap. 97, 425 p.

Herazo, B. 1971. Reproduccion de Pamilirus argus en la costa de la Guajira colombiana. Tesis de Licenciado en ciencias. Fundacion Universidad deBogota(JTL),61p.

Munro, J.L. 1974. The biology, ecology and bionomics of Caribbean reef fisheries. VI. Crustaceans (spiny lobsters and crabs). Res. Rept. Zool. U. W. 1.3, 57p.

Olsen, D.A., W. Herrnkind and I. G. Koblick. Ms. 1972. Ecological study for the development of lobster management techniques. PRINUL (Puerto Rico International Undersea Labora- tory) Special Report No. 1, 71 p.

Peacock, N. 1974. A study of the spiny lobster

fishery of Antigua and Barbuda. Proc. Gulf and Carib. Fish. Inst. 26:117-130.

Riveros, G. 1972. Spiny lobsters Pamilirus argus and P. laevicauda on the Caribbean coast of Colombia with particular reference to their biology and the fishery for both species on the coast of the Guajira Peninsula. M.Sc. thesis, Memorial University of Newfoundland, 115 p.

Spiegel, M.R. 1961. Theory and problems of Statistics. Schaum's Outline Series in Mathematics. McGraw-Hill. 359 p.

Squires, H. J. 1970. Lobster (Homarus americanus) fishery and ecology in Port au Port Bay, Newfoundland, 1961-65. Proc. Nat. Shellfish. Assoc. 60: 22-39.

Squires, H. J. 1973. El potencial reproductive de los crustaceos decapodos. Bol. Museo del Mar (Univ. Bogota) 5: 3-7.

Squires, H. J. Ms. 1974. Length and weight rela- tionships in shrimps. Govt. Consultation on Shrimp Resources in CICAR area. FAO/FIR: SR/74/W.P. 5, 7p.

Squires, H. J., G. E. Tucker and G. P. Ennis. 1971. Lobsters (Homarus americanus) in Bay of Islands, Newfoundland, 1963-65. Fish. Res. Bd. Can. Manuscr. Rept. Ser. (Biol.) 1151, 58 p.

PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION

75

ABSTRACTS OF THE TECHNICAL PAPERS PRESENTED

AT THE 1977 NSA CONVENTION

MEXICAN MOLLUSCAN FISHERIES

OF THE GULF OF MEXICO AND THE CARIBBEAN : 1970-1975

Erik Baqueiro and Craig B. Kensler

Centro de Ciencias

del Mar y Limnologia

Universidad National Autonoma de Mexico

Mexico 20, D.F., Mexico

Molluscan fishery statistics for the Mexican Gulf and Caribbean have been analyzed for 1970-1975 for each molluscan group or by-pro- duct landed, and for each of the six coastal states. The statistics listed the landings by the following "classes" of molluscan groups or by-products: sea clam, river clam, squid, mixed shells (industrial), conch meats, mixed molluscan meats, oyster with shell, oyster meats and octopus. During 1970-1975 total molluscan landings of 189,131 metric tonnes, valued at $29,637,244 (USA), were reported for the area — averaging about 31,522 tonnes, worth about $4,939,541 (USA), per year. The most im- portant molluscan fishery of the region during the period was for "oyster with shell" (mainly Crassostrea virginica), followed far behind by the octopus fishery (Octopus vulgaris and O. maya.) The oyster with shell fishery represented about 84% by weight and 50% by value of total molluscan landings, including by-products, from the Mexican Gulf and Caribbean region during 1970-1975. The octopus fishery represented 8.3% by weight and 25.1% by value of total molluscan landings. Veracruz was by far the most important state for molluscan landings. During the period, 51.8% of the total weight of landings, and 31.9% of total value, from the region were reported from the state of Veracruz. The state of Tabasco was se-

cond in total weight landed (29.5%) and the state of Campeche was second in total value of landings (23.1%) for the entire region.

A DISEASE SURVEY OF

NEW ENGLAND SOFT-SHELL

CLAMS, MY A ARENARIA

Robert S. Brown

Marine Pathology Laboratory

Department of Animal Pathology

University of Rhode Island

Kingston, Rhode Island 02881

A field survey for possible pollution-related diseases of the soft-shell clam was initiated in January, 1976. To date, more than 1500 clams from 10 sites of differing types and degrees of pollution have been histopathologically examin- ed. Lesions noted included: disturbances of growth (neoplasia, hyperplasia, hypoplasia), reac- tion to injury (hemocytosis, inflammation), pre- sence of parasites (bedsonia, protozoan, meta- zoan, and accumulations of orange-brown pigmented bodies (both intra- and extracellularly).

The prevalence of these lesions varied between sites suggesting environmental influences. Cytologic examination, coupled with a signifi- cantly increased mortality during a 10 month study indicated the malignancy of the neoplastic disease.

HISTOCHEMICAL ANALYSES OF PIGMENT

ACCUMULATIONS IN

MERCENARIA MERCENARIA L. AND

MY A ARENARIA L.

76

ABSTRACTS

Robert S. Brown and Carole J. OToole

Marine Pathology Laboratory

Department of Animal Pathology

University of Rhode Island

Kingston, Rhode Island 02881

Histopathologic analysis of Mercenaria col- lected from Massachusetts, Rhode Island and New Jersey demonstrated accumulations of pigments of three types: 1) irregular, 2-20 (im diameter, orange-brown staining (H&E) bodies, present ex- tracellularly in the alimentary tract epithelium and gonadal connective tissue, and intracellularly in renal epithelium; 2) spherical 7-12 ^m diameter eosinophilic concretions, present intracellularly in renal epithelium; and 3) 30-200 pm diameter melanotic casts present in renal tubular lumens.

Only type 1 pigment was found in Mya col- lected throughout New England, and was present in lesser amounts than in Mercenaria from adja- cent locations. This pigment had morphological characteristics of lipofuscins (brown staining bodies of oxidized lipids) found in vertebrates. All three pigment types had histochemical charac- teristics of lipofuscins.

Based on the examination of over 200 Mercenaria and 1500 Mya, the accumulation of these pigments appears to be a normal physiological process, although abnormally large accumulations were noted in clams from certain environments.

CULTURE OF HATCHERY-SPAWNED

MERCENARIA MERCENARIA IN

MASSACHUSETTS

H. Arnold Carr

Massachusetts Division

of Marine Fisheries Sandwich, Mass. 02563

Coastal towns, which have the prime respon- sibility for managing shellfish in Massachusetts, are finding hatcheries to be one of the best sources for hard clams (M. mercenaria). During 1976, 12 towns bought hatchery stock between 5-14 mm (longest diameter). A variety of on-bottom and off -bottom systems was used to isolate the clams from predators. Growth varied with system de-

sign, density of clams, and substrate type. Sur- vival prior to winter, though 90% or more, must be qualified. During the winter, survival was 80-90% in rafts and widely variable in pens lying in intertidal and shallow subtidal areas. The suc- cess of most transplants into natural, unprotected bottom appears related to the number of green crabs observed in the transplant site. Although the theoretical yield of this technique is high, the ac- tual benefit to a town management program re- mains to be proven.

ULTRASTRUCTURAL EVIDENCE THAT

GASTROPODS SWALLOW SHELL RASPED

DURING HOLE BORING

Melbourne R. Carriker

College of Marine Studies, University of Delaware Lewes, Delaware 19958

Observations are reported on the ultrastructure of shell material rasped by Urosalpinx cinerea follyensis Baker from boreholes in the valves of Mytilus edulis Linne and transported normally to the stomach through the buccal cavity and esoph- agus. Duration of the period of chemical activity by the accessory boring organ and rasping by the radula were determined with a valve model. Pel- lets of shell raspings were removed from the stomach and, after fracturing to reveal the in- terior, and coating with metal, were studied with the scanning electron microscope. Shell raspings were compared with prisms and lamellae in frac- ture surfaces of normal shell of M. edulis and shell etched with ethylenediamine and sodium hypo- chlorite to reveal the form of shell units clearly.

The study provided ultrastructural evidence for the first time that U. cinerea swallows shell rasped from the borehole during penetration of prey. Both prisms and lamellae were identified in the pellets removed from the stomach. Noticeable dissolution of the organic matrix, and to some ex- tent also of the mineral portion, of prisms was evi- dent, features which facilitate removal of shell by the snail during rasping. If the long axis of prisms occurs parallel to the surface of the borehole, the radula tends to rasp out long fragments of shell; if prisms are placed at right angles to the surface, the

PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION

77

radula breaks prisms into small pieces. The envel- ope of mucoid material which coats pellets un- doubtedly reduces, or prevents, laceration of the epithelium of the alimentary canal as shell fragments pass down the tract. A gross approx- imation of the percentage of shell in the borehole which is rasped and swallowed during the process of hole boring is 14 % .

THE INTERACTION OF WATER SOLUBLE

FRACTIONS (WSF) OF SOUTH LOUISIANA

CRUDE OIL AND DERMOCYSTID1UM

(LABYRINTHOMYXA) MAR1NUM AT

VARYING TEMPERATURES IN THE

AMERICAN OYSTER, CRASSOSTREA VIRGINICA GMELIN

Keith Cooper

Marine Pathology Laboratory

Department of Animal Pathology

University of Rhode Island

Kingston, Rhode Island 02881

Sammy Ray and Jerry Neff

Moody College of Marine Sciences and Maritime Resources Texas A&M University Galveston, Texas 77550

Over 500 oysters infected with D. marina were exposed to 100% WSF crude oil in closed seawater systems with water temperatures varying between 16°- 28c C. Elevated water temperature resulted in greater mortality and significantly higher levels of D. marina in infected oysters. Oysters exposed to the WSF had significantly greater mortality and decreased levels of D. marina than non-exposed oysters. Concommitant temperature elevation and exposure to WSF significantly increased mortality and the rate of uptake of napthalenic hydrocar- bons.

TETRACYCLINE AS A BIVALVE SHELL MARKER

N. Dean Dey and Ellis T. Bolton

University of Delaware

College of Marine Studies

Lewes, Delaware 19958

Crassostrea virginica, Mercenaria mercenaria and Mytilus edulis were exposed to the antibiotic tetracycline in ambient sea water. The antibiotic was dissolved in filtered sea water, to which algal food was added. The animals were allowed to feed on this mixture, which encouraged rapid shell growth. The marking period lasted one week to several months. Tetracycline was incorporated concurrently with the rapid deposition of new shell. The inlaid tetracycline fluoresces a vivid yellow-orange under UV light (Mineralight, 254nm). Marking was vivid in Crassostrea and Mercenaria and vague in Mytilus.

Marking was vague at 0.5 and 5.0 mg L"1 but vivid at 25 to 200 mg L"1. No deaths or mor- phological defects were noted at any concentra- tion tested. The amount of inlaid tetracycline was related to the quality of algal food. Algae species known to be good food sources produced the greatest incorporation of tetracycline. Marking of Mercenaria larval and spat shells was visibly bright at 25 and 50 mg L"1 and visibly evident at 0.5 and 5.0 mg L"1. under the dissecting micro- scope. Experiments with five tetracycline ana- logues showed that tetracycline phosphate com- plex produced the best results with tetracycline HCL, oxytetracycline, doxycycline and minocycline producing progressively poorer results.

The mark remains undiminished in vivo for at least seven months, and its permanence is sug- gested by structural changes to shell crystals on the ultrastructural scale. The persistence of the mark and its distinctive fluorescence suggest the utility of tetracycline in studies of shell growth and morphology and in field investigations. Com- mercially, this technique could be used as an identification, for example, by hatcheries to com- bat poaching on leased grow-out grounds.

OYSTER REEF CULTIVATION FOR CULTCH MATERIAL

William J. Eckmayer

Alabama Marine Resources Laboratory Dauphin Island, Alabama 36528

A modified oyster dredge was used as a method of utilizing existing buried shell for cultch. SCUBA

78

ABSTRACTS

and random m2 quadrats were used before and after dredging to sample Bayou Cour Reef, Kings Bayou Reef, and Shellbank Reef in Mobile Bay and Bon Secour Bay, Alabama. The reefs were dredged at a rate of 6.4 km/hour covering 0.4 ha in one hour and 15 minutes.

The modified dredge was found to destroy half- shell while exposing the half-shell resulting in an increase in shell weight. The bottom was disrupted and the exposed shell appeared to settle into the bottom with a subsequent loss of relief followed by an accumulation of silt.

Spat set increased following dredging, but loss of relief makes this method unsatisfactory for reefs over a soft substrate. Man-made or rehabilitated oyster reefs on hard substrate may benefit from this method.

MARKING CLAMS WITH RUBIDIUM1 Arnold G. Eversole

Clemson University Department of Entomology and

Economic Zoology Clemson, South Carolina 29631

Hard shelled clams, Mercenaria mercenaria, were successfully marked with rubidium by rear- ing seed clams in artificial seawater with an aqueous solution of rubidium chloride. Elevated levels of Rb* in samples of tissue were detected by atomic absorption spectroscopy. Clams exposed for 48 and 96 hours to 10.0, 1.0, 0.1 and 0.01 g/1 RbCl contained levels of Rb* significantly higher than endogenous levels. Biological activity, measured as siphon extension and survival of clams, was not affected by the presence of RbCl at concentrations less than 10.0 g/1. Also, sig- nificantly higher levels of Rb* were present in tissue for up to 3 weeks when clams were exposed to 1.0 g/1 RbCl for 96 hours then transferred to uncontaminated water.

Diatoms, Phaedactylus tricornutum, exposed for 24,48 and 96 hours to 10.0, 1.0 and 0.1 g/1 RbCl had significantly higher levels of Rb* than diatoms cultured without RbCl. Clams cultured for 96 hours in vessels containing diatoms exposed to 1.0 g/1 Rb CI had significantly more Rb* than

clams grown with unlabelled diatoms. Clams ex- posed to 1.0 g/1 RbCl solutions with and without diatoms contained levels of Rb* significantly higher than those clams reared only with labelled diatoms. No significant difference was detected between clams grown with and without diatoms at 1.0 g/1 RbCl.

Mud crabs, Panopeus herbstii, were offered Rb* labelled clams for 96 hours then sacrificed or transferred to containers with unlabelled clams for 7 days. Control crabs were fed unlabelled clams and sacrificed at 96 hours and 7 days. Fecal strands of experimental crabs had elevated levels of Rb* with a peak at approximately 132 hours. Four tissues dissected from control and experimen- tal crabs were analyzed and significantly higher levels of Rb* were found in the hepatopancreas of experimental crabs after the first 96 hours. No ap- parent difference was detected at 7 days.

This study was supported by the South Carolina Experiment Station and Clemson University Faculty Research Commit- tee.

SUSPENSION CULTURE OF EUROPEAN OYSTERS (OSTREA EDUL1SL.)

Reg Gillmor

Ira C. Darling Center

University of Maine at Orono

Walpole, Maine 04573

Operations have appeared along the coast of Maine during the past few years for the culture of European oysters. Suspension techniques are used for the grow-out of cultchless, hatchery-reared seed, and suspension systems most often involve one or another variation on the Japanese long- lin- ing theme. The first system to appear commercial- ly employs floating modules consisting of stacks of wooden-framed plastic mesh trays short- tethered to a long hauser line. These are tended from a raft which can hoist modules aboard for servicing — cleaning, grading, etc. — one at a time. A second system was recently introduced which makes use of culture nets imported from Japan where they were developed for the rearing of pearl shells and scallops. In this system the gear is in-

PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION

79

dividually buoyed and suspended, either near the surface or at mid-depth, beneath a long line. A service raft is also utilized and in this case there ex- ists the possibility of batch-servicing a number of net modules simultaneously. A third technique has appeared for nursery grow-out in well pro- tected areas. The culture unit is a rectangular, wooden-framed tray enclosed with plastic mesh on both top and bottom. These are strung together end to end, floated in rows, and flipped regularly so that each side is alternately exposed and submerged, a procedure which controls fouling. The equipment and methods of several oyster farms are illustrated and discussed indicating the varying approaches to common problems such as retrieval and resuspension of modules, prevention and removal of fouling, overwintering and others. The development of the Maine oyster aquaculture industry as a whole is briefly reviewed.

GROWTH RESPONSES OF EUROPEAN

AND AMERICAN OYSTERS

(OSTREAEDULISL.

AND CRASSOSTREA VIRGINICA G.)

TO INTERTIDAL EXPOSURE

Reg Gillmor

Ira C. Darling Center

University of Maine at Orono

Walpole, Maine 04573

In view of the use which is made of the intertidal zone in the rearing of oysters in other areas of the world (e.g., Japan, New Zealand) studies were undertaken to determine its potential usefulness to the culture of European and American oysters in Maine. As part of a seed "hardening" experiment, spat were suspended at 1 subtidal and 5 intertidal levels (9-49% mean aerial exposure time). Mon- thly instantaneous growth (whole dry weight) rates (k) indicated American oysters to be "break- ing even" with exposure; i.e., over the range of levels tested a given percentage change in immer- sion times was matched by an equal change in k so that growth relative to immersion time was ap- proximately the same at all levels. The k vs. ex- posure curve for European oyster seed was also linear but differed in having a more rapid decline in growth with increasing exposure. A similar pat-

tern of growth was exhibited by yearlings of the 2 species suspended for 5 months at 10 levels centered more or less on mean low water. Because of the closer spacing and more restricted range of exposure levels (0-30%) in this experiment, however, a better resolution of the growth response at the lower levels of exposure was possi- ble, and an interesting feature of European oyster growth not distinguishable in the seed study emerged: Growth at levels of exposure up to 10-12% equalled that of subtidal oysters, in- dicating that growth relative to immersion time was actually greatest in the groups exposed at low levels. At higher levels k values declined rapidly. Growth of American oysters again appeared to correspond with immersion time in a one-to-one fashion. The O. edulis result suggests a potential use of the lower intertidal for the "market hardening" — conditioning for prolonged valve closure — of European oysters to extend their shelf life, a problem which has been troubling Maine aquaculturists. Such a procedure need not entail any loss of growth.

THE ONSHORE SURF CLAM RESOURCE

ALONG THE SOUTHERN NEW JERSEY

COAST

Harold H.Haskin

Oyster Research Laboratory

New Jersey Agricultural Experimental Statio)i

and Department of Zoology

Rutgers University New Brunswick, New Jersey

With the general decline of the offshore surf clam beds under increasing dredging pressure over the past decade, an increasing proportion of the New Jersey catch has been taken from the inshore beds, close to the beaches of Atlantic and Cape May Counties. This shift in fishing effort has caus- ed concern for the maintenance of the inshore populations and, since 1972, has led to studies of population size, distribution, recruitment and growth.

During this period the standing stock within the three-mile limit has been declining continuously and is now estimated at less than 4-million

80

ABSTRACTS

bushels. In general the more dense populations are close to the beaches and, within the three-mile limit, decrease in density with distance offshore. The clams also show two distinct gradients in size: (1) smaller clams inshore; (2) smaller clams up the shore in the stretch from Cape May to Beach Haven inlet.

Over several summer seasons settling of juven- iles has been general in the inshore area with den- sities ranging up to several hundred per square meter. Early mortality rates, however, have been high and by end of summer young of the year are virtually non-existent. Principal predators include Limulus. Ovalipes and Pagurus. No substantial recruitment has occurred in the past five years.

MORTALITY PATTERNS AND DISEASE RESISTANCE IN DELAWARE BAY OYSTERS

Harold H. Haskin and Susan E. Ford

Oyster Research Laboratory

New Jersey Agricultural Experimental 5tatio>i

and Department of Zoology

Rutgers University New Brunswick, Neio Jersey

Since 1964, 29 groups of oysters have been laboratory-bred for resistance to kill by Minchinia nelsoni, then selected and tested for survival in en- zootic waters in Delaware Bay. Survival at the end of a 33-month test period of four filial generations of selected stocks has been compared to that of un- selected groups. Compared to 16 groups of suscep- tibles the survival ratio of F, resistants is 4.4; this ratio increases to 5.0, 6.6 and 8.9 for F:, F, and F4 respectively. Native set originating from parents selected on planted grounds in the lower bay, have a cumulative kill falling between susceptibles and Fi resistants, with a survival ratio of 2.9 compared to the unselected stocks.

Differential mortality ratios are established dur- ing exposure to the first complete summer infec- tive period and are maintained with little variation for the remainder of the test period. Despite a plateauing of cumulative mortality curves for all stocks after the initial kill, seasonal mortality rates remain higher for susceptibles than for resistant groups after nearly three years of selection.

Non-predation kill on Delaware Bay native seed stocks planted on low bay leased grounds during the past 12 years has generally demonstrated sur- vival comparable to laboratory-reared resistant groups.

THE OYSTER INDUSTRY OF VIRGINIA 1931 to 1975

Dexter S. Haven, William J. Hargis, Jr. and Paul C. Kendall

Virginia Institute of Marine Science Gloucester Point, Virginia 23062

This study investigates the cause or causes of the decline in productivity of Virginia's oyster in- dustry over the years and especially the catas- trophic decline since 1960. Included in this report are detailed studies on landings and oyster densi- ty, a documentation of the decline in spatfall in many areas since 1°60, reviews of predators and diseases, repletion activities, laws and other aspects.

The principal reason for the decline since I960 has been the direct or indirect impact of MSX, but other factors associated with degradation of the environment may also have operated. In many in- stances rising labor costs and antiquated tech- nology have prevented a recovery of the industry.

Recommendations for improving the industry are given.

AGE, GROWTH, REPRODUCTION

AND DISTRIBUTION OF THE BAY

SCALLOP,

AEQU1PECTEN IRRAD1ANS

1RRADIANS (LAMARCK).

IN THREE EMBAYMENTS

OF EASTERN LONG ISLAND, NEW YORK,

AS RELATED TO THE FISHERY

Mary T. Hickey'

Department of Marine and Environmental Science

C. W. Post Center. Long Island University

Greenvale. New York 11548

Bay scallops are an economically important fishery on eastern Long Island. The bay scallop

PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION

81

populations in three embayments in the Town of East Hampton, N.Y., were studied for weight, growth, and condition of the adductor muscle, reproduction, shell growth, abundance and distribution.

The spawning effort is maximized at the expense of other functions. Spawning occurs primarily in tune and is temperature related. The adductor muscle index and condition decrease during gonad maturation and spawning, and shell growth near- ly ceases during the reproductive period. The greatest rate of weight gain of the adductor muscle and greatest shell growth occur during the post- spawning period, July through October.

Commercially acceptable growth appeared to be density and depth related. Modal shell length of the harvestable year class was 60-65 mm at the beginning of the fishing season in October. State and local minimum size regulations were evalua- ted on a biological basis and management recom- mendations were discussed.

Present address: 4415 Independence St., Rockville. Md. 20853.

OYSTER SETTING - EVOLUTION OF COMMERCIAL HATCHERY TECHNIQUE

Herbert Hidu

Richard D. Clime

Samuel R. Chapman

Ira C. Darling Center

University of Maine

Walpole, Maine 04573

Cultchless setting has obviously revolutionized hatchery economics. Bill Budge's development has led to a variety of proprietary and apparently non-proprietary approaches to the process. Ultimate optimization in cultchless setting must depend on a complete understanding of physiolog- ical and behavioral responses at setting. It is clear that oyster larvae can delay metamorphosis and set in response to environmental stimuli. Our ex- periments indicate that the setting response is released by rapid temperature increases, water- borne pheromones from adult oysters, and an unknown property associated with calcium car- bonate of shell. All are used to advantage in an ef-

ficient cultchless process. Several calcium carbonate related materials appear to be stimula- tory including tropical beach sand, foraminiferal sand (we thank Mr. Thomas W. Dignes for sug- gesting the use of foraminiferal sand as a cultch material), polished marble and marble chips. It is possible to orient larvae with hinge-side upper- most on vertical polished marble slabs thus in- creasing the efficiency of removal without dam- age. A complete understanding of the reponse to calcium carbonate may allow us a system of oyster metamorphosis in the absence of substrate.

AN OYSTER FAMILY TREE: ANCESTRY OF CRASSOSTREA VIRG1NICA

Sewell H. Hopkins

Department of Biology

Texas A&M University

College Station, Texas 77843

A hypothetical family tree of oysters is presented, based on the authoritative text of H.B. Stenzel ("Oysters," volume 3 of 3, Part N, Bivalvia, Mollusca 6, Moore's Treatise on In- vertebrate Paleontology, 1971). However, I do not believe that our oyster descended from the giant Eocene "Crassostrea" gigantissima as some geologists argue (Sohl and Kauffman, 1964, and others). As an ancestor for our oyster I prefer Crassostrea glabra (Meek and Hayden, 1857). This species was abundant in widely separated parts of the huge brackish Laramie Sea that ex- tended across the interior of Western North American, from Canada to Mexico, in very late Cretaceous time. It was associated with a few other species of marine origin (species of Anomia, Brachidontes and Corbula, along with some species of genera which have both marine and fresh-water affinities: Corbicula, Neritina, Melania, and other). In other parts of the same sea there were many species of genera which are now found only in fresh water, indicating that salinity must have been very low. C. glabra seems to have been a true Crassostrea, usually lacking the marginal denticles that Stenzel called "chomata," and was very similar to living Maryland oysters of

82

ABSTRACTS

the deep-water type. Descent from an oyster of a low-salinity brackish sea may account for the fact that C. virginica tolerates lower salinity than any other living oyster. C. gigantissima, besides hav- ing chomata when young and thus not being a Crassostrea by Stenzel's definition, was associated with many marine species as you would expect if it grew in salty coastal lagoons.

BACTERIOLOGICAL STUDIES OF LONG ISLAND SHELLFISH HATCHERIES1

Louis Leibovitz

Department of Avian and Aquatic Animal

Medicine

New York State College of Veterinary Medicine

Cornell University, Ithaca, N.Y.

High concentrations of bacteria in shellfish hatchery media have been reported as a cause of larval mortality. This study was undertaken to characterize the dominant bacterial flora of the Long Island shellfish hatcheries. Although the study was initiated at a single hatchery during 1972, the current report covers a three-year period (1974-1976) and includes 5 hatcheries. Bacterio- logical samples from sick and healthy oyster and clam larval cultures and their ingredients were taken at a standard working dilution onto Plate Count Agar media with synthetic sea salts. Each cultural sample was replicated in triplicate. The hatchery ingredients samples included incoming ultraviolet treated and untreated bay and well water, stock algal cultures, and pooled algal cultures. Samples were taken throughout the year at regular monthly intervals. Qualitative and quantitative counts were made of the dominant populations of each culture and isolated colonies were identified as pure cultures.

A total of 1,279 cultures were taken and 710 pure bacterial isolates were identified. Sixty-six percent of all bacteria isolated and identified from intake water samples were gram negative, and 15.7 percent were gram positive. Well water had a higher percentage of gram positive organisms than did bay water intake. Thirty-three percent Pseudomonas, 10.4 percent Flavobacter or Cytophaga, 8.7 percent Acinetobacter, 4.9 per- cent Aeromonas, 3.8 percent Enter obacteriacea,

3.2 percent Vibrio and 1.5 percent Achromobacter spp. were the dominant gram negative isolates found in all water sampled. Well water counts were lower but contained a higher percentage of Aeromonas, Vibrio, and Flavobacter or Cytophaga and less Enterobacteriacea than did bay water. Ultraviolet treatment had no effect on bacterial counts of hatchery water intake. Stock algal cultures were frequently contaminated with a wide variety of bacteria. The percentage of Acinetobacter and Flavobacter or Cytophaga spp. increased in algal cultures. Sick or dying oyster larval cultures are characterized by a sharp de- crease in the percentages of Pseudomonas, Flavobacter or Cytophaga, Acinetobacter and Enterobacteriacea spp. and a rise in percent gram positive bacteria. Sick clam larval cultures were associated with a drop in the percentage of Flavo- bacter or Cytophaga, Aeromonas, and Enterobacteriacea spp. The percent of unidentified and lost bacterial cultures increased in sick oyster and clam larval cultures. Each hatchery larval culture media had its own distinctive bacterial flora. High bacterial counts were associated with warm weather, increased storage and culture time, and high organic content. The frequency and dis- tribution of Vibrio spp. during an outbreak of vibriosis is reported. Water quality studies were conducted during the study. The usefulness of routine bacteriologic sampling for diagnostic pur- poses is discussed.

1. This research was sponsored by the New York Sea Grant Institute under a grant from the Office of Sea Grant. Na- tional Oceanic and Atmospheric Administration (NOAA), U. S. Department of Commerce.

WATER QUALITY STUDIES OF LONG ISLAND SHELLFISH HATCHERIES '

Louis Leibovitz and John Hamlin Gordon II

Department of Avian and Aquatic

Animal Medicine

New York State College of

Veterinary Medicine

Cornell University, Ithaca, N.Y.

Since the health of shellfish larvae and the ef- ficacy of shellfish hatchery production is depend-

PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION

83

ent upon the characteristics and stability of the water supply, a study of the water quality of bay and well water intakes of 5 Long Island shellfish hatcheries was undertaken to define the cycle and range of water quality changes throughout a pro- duction year. Samples were taken at regular monthly intervals and included measurements for ammonia, nitriate, nitrate, orthophosphate, total organic phosphate, total inorganic phosphate, dissolved oxygen, pH, salinity and temperature.

Ammonia well values ranged from a low of 0.5 to a high of 5.2 and a mean of 1.67 mg/liter. Bay intake ammonia values ranged from 1.7 to 7.0 and a mean of 4.52 mg/liter through the year. Am- monia values rose through the spring and summer months with a sharp decline at the end of the year during early winter. A precipitous short-term drop in ammonia levels in July was associated with an outbreak of vibriosis and high larval mor- tality at one hatchery.

Nitrate values ranged from 0.5 to 5.0, mean 1.77 mg/liter for well; and 0.5 to 6.1, mean 4.52 mg/liter for bay intake. During spring and sum- mer months low levels of nitrate were found, followed by a sharp increase in fall and winter.

Orthophosphate levels ranged from 0.03 to 0.095, mean 0.05 mg/liter for well water and 0.04 to 0.40, mean 0.186 mg/liter, for bay intake. In the presence of oxygen, orthophosphate and iron levels are interrelated. An extremely high short- term orthophosphate level was associated with disease at one hatchery utilizing fossil fuel thermal effluent.

Iron levels in bay water increased markedly in the spring of the year and decreased later. Iron levels in well water were low and at a stable level. Copper fluctuated throughout the year, but were always higher than reported for ambient oceanic levels.

Well intake had lower pH values than bay in- take. Both became more alkaline during spring and summer, and more acid during fall and winter months.

More work is needed to select and define the criteria for water quality tests at shellfish hatch- eries.

1 This research was sponsored by the New York Sea Grant Institute under a grant from the Office of Sea Grant, Na- tional Oceanic and Atmospheric Administration (NOAA), U.S. Department of Commerce.

A COMPARISON OF HINGE-LINE

MORPHOGENESIS IN LARVAL SHELLS OF

MYTILUS EDUL1S L. AND

MODIOLUS MODIOLUS (L.)

Richard A. Lutz

Yale University

Department of Geology and Geophysics

New Haven, Connecticut 06520

Optical and scanning electron microscopic ex- amination of laboratory reared cultures of larval Mytilus edulis (blue mussel) and Modiolus modiolus (northern horse mussel) revealed similar hinge-line morphogenesis (straight-hinge stage through metamorphosis) in both species. The hinge apparatus (provinculum) of both mytilids increased in size and complexity throughout larval development, with progressive lateral thickening characteristic of the family Mytilidae. The number of "teeth", while highly variable between in- dividuals with similar shell dimensions, increased significantly with both total shell and provin- culum length. Ultrastructural studies of the teeth revealed marked ridged structures which became increasingly apparent with larval growth. In neither species did lateral teeth develop prior to settlement and metamorphosis.

Significant differences at the population level were found between Mytilus edulis and Modiolus modiolus upon regression and quantitative com- parison of each of the following: larval length vs. provinculum length; larval width vs. provinculum length; provinculum length vs. number of teeth; and larval length vs. number of teeth. Such dif- ferences should prove useful in discrimination of these two mytilids within the plankton.

CULTURE OF THE SEA SCALLOP,

PLACOPECTEN MAGELLANICUS (GMELIN)

IN NEWFOUNDLAND

K.S.Naidu

Fisheries and Marine Service

Fisheries and Environment Canada

3 Water Street

St. John's, Newfoundland

A1C1A1

84

ABSTRACTS

Techniques used to collect naturally-produced sea scallop (Placopecten magellanicus) spat using artificial substrates and their subsequent growth in hanging culture at a near-shore locality in New- foundland are described.

Of the various substrates employed polyethy- lene film enclosed in onion bags appeared to be the most suitable. Using this substrate the mean number of live scallop spat collected increased with depth to about 12 m; deeper than this, the numbers settling decreased. Collectors suspended in less than 8 m harboured considerable numbers of starfish. Numbers of live spat and starfish in collecting units were inversely correlated. The mean size (±1 S.D) of live spat (dorso-ventral axis) at the time of collection was 12.4 ±0.81 mm with a range of 5.0 to 19.1mm. There was a slight but significant increase in size with depth, the dif- ference in size being approximately 1.0 mm over the 12 m range sampled.

Growth of scallops in suspended culture was ex- ceptionally good. Preliminary results indicate that shell heights of 12, 24, 36, 48 and 60 month-old animals to be in the order of 36, 66, 80, 95 and 107 mm respectively. Corresponding shell heights of wild scallops from the sea bottom in the same area were x, 27,46, 66 and 85 mm (x, 12 month-old scallops, were not represented in the study). Mor- tality is approximately 5% per annum. Animals under two years appear to sustain higher mor- talities than those above it.

The relatively sedentary nature of the animal opens up the possibility of resource enhancement of the species through sea ranching the coastal zone. This less expensive but longer route is being examined as well.

ENGINEERING CONSIDERATIONS IN THE

DESIGN OF

OYSTER DEPURATION PLANTS

Bruce J. Neilson

Virginia Institute of Marine Science Gloucester Point, Virginia 23062

Oyster depuration studies were conducted using four commercial size tanks with varying en- vironmental factors and initial bacterial levels. All four tank designs were suitable for the process,

but two were superior. The primary design fea- tures which appear to be important are the resi- dence time of the water and the circulation pattern within the tank. Fecal coliform levels in the water were observed to rise as soon as the oysters were placed in the tank. Depuration occurs more rapid- ly if the bacteria released by the shellfish are rapid- ly removed from the tank. The time for a 90% reduction in concentration of dye injected into the tank was found to be roughly twice the theoretical residence time, which is obtained by dividing the total volume by the flow rate.

The residence time may be decreased by increas- ing the flow of water. An alternative method is to increase the number of oysters held in the tank. Tests showed that depuration was equally success- ful in identical tanks holding 1.8 and 3 bushels of oysters if the flow of water per bushel of oysters was the same for both tanks.

Good circulation in the tank is needed to pre- vent dead areas which could become oxygen de- pleted. Dye tests showed that the residence time was greatly reduced when a water pump was used to increase the flow within the tank. Air lift pumps can accomplish the same result. If the system is properly designed, the resulting current will be smooth and gentle and will not resuspend biode- posits.

OIL AND THE OYSTER IN DELAWARE BAY

George S. Noyes,

Harold H. Haskin,

and Cindy Van Dover

Oyster Research Laboratory

New Jersey Agricultural Expt. Station

and Department of Zoology

Rutgers University

Potentially hazardous effects of petroleum and its derivatives on an economically important population of the American oyster, Crassostrea virginica, were investigated to assist development for the Delaware estuary, of more precise water quality standards, compatible for both industry and survival of oyster beds.

Adult and larval oysters were chronically ex- posed to crude and refined oils adsorbed onto fine kaolin clay. Threshold concentrations for mor-

PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION

85

talities and relative toxicities for different oils were determined.

Adult oysters exposed to 0.3 ppm Nigerian crude showed mortalities twice those of controls. No. 2 fuel oil was approximately twice as toxic as Nigerian crude, which in turn was twice as toxic as Iranian light crude.

Larvae, cultured in natural seawater, developed and metamorphosed in control treatments, but showed increasing mortality, delay in develop- ment, and inhibition of metamorphosis at and above concentrations of 0.5 ppm Nigerian crude oil. No. 2 fuel oil was highly toxic to larvae at con- centrations as low as 0.25 ppm. Larvae did not show greater sensitivity to oils when compared to adults.

THE ROLE OF URONEMA

MARIN UM (PROTOZOA)

IN OYSTER HATCHERY PRODUCTION

Linda Plunket

Ira C. Darling Center

University of Maine at Orono

Walpole, Maine 04573

In Maine's hatchery production of oysters, significant mortalities at the early juvenile stage have been associated with ciliate protozoan in- festations. The predominant ciliate was isolated from live infested oysters in the Spring of 1974 and was identified as Uronema marinum. Feeding-type experiments were carried out to determine what food source this ciliate was utilizing in the tank en- vironment. Uronema marinum was found to be a bacteriophage and not a histophage; thus, the high mortalities accompanying the early juvenile stage cannot be primarily attributed to the ciliate in- festations.

The survival and growth of hatchery reared bivalves have been investigated in a system of fiberglass raceways through which seawater was continuously pumped. Flow rates of 50 to 60 1/min were maintained in tanks 10 m X 1.3 m with variable depths. The bay scallop, Argopecten irra- dians, was successfully grown in the raceways from a size of 5 mm. Varying raceway depth be- tween 7.5 and 30 cm had no effect on scallop growth. Scallops at stocking densities of 5, 10, 20 and 40 1/raceway and culled back to these den- sities at bi-weekly intervals grew at rates inversely proportional to population size. Scallops stopped growing when the water temperature declined to 10°C. The surf clam, Spisula solidissima, was successfully grown in the system from a starting size of 1 mm. Spisida smaller than 20 mm grew equally well in sand and without a substrate. Surf clams larger than 20 mm grew faster in a sand substrate than without and some shell abnormali- ties were observed in clams grown without substrate. Surf clams exhibited a growth rate in- versely proportional to stocking density for biomasses of 1, 4 and 8 1/raceway. The excellent growth rate of these bivalve species in raceways suggests that the use of this system may be a valuable step in the production of shellfish seed.

STATUS REPORT ON THE COMMERCIAL

BLUE CRAB FISHERY

OF THE CAROLINAS AND GEORGIA

R.J. Rhodes1, M. Wolff2, J.L. Music3

1. South Carolina Wildlife

and Marine Resources Department

Division of Marine Resources

Charleston, S.C.

THE USE OF PUMPED RACEWAY SYSTEMS

FOR THE INTERMEDIATE

GROW-OUT OF HATCHERY REARED

BIVALVES

Edwin W. Rhodes and Ronald Goldberg

U.S. Department of Commerce - NO A A

National Marine Fisheries Service

Northeast Fisheries Center

Milford, Connecticut 06460

2. North Carolina Department of

Natural and Economic Resources

Division of Marine Fisheries,

Morehead City , N.C.

and

3. Department of Natural Resources,

Coastal Resources Division

Brunswick, Georgia

86

ABSTRACTS

The annual landings, operating unit data, and ex-vessel price for the commercial blue crab (Callinectes sapidus) fisheries in North Carolina, South Carolina and Georgia between 1960 and 1976 were analyzed to document recent trends in this fishery and discuss possible factors causing the decline in landings since 1970. Between 1970 and 1976, the total annual landings of blue crabs in the Carolinas and Georgia have declined at an average rate of 2 million pounds per year (1970 landings = 35 million pounds), while 3 thousand traps per year were added to the fishery between 1970 and 1975.

The ex-vessel price for blue crab has increased from $.05 per pound in 1970 to $.11 per pound in 1976 with an increase of 20% in the adjusted price per year compared to 1970. This trend suggests a quantity-effect dominant ex-vessel price function. Inaccuracies in the N.M.F.S. reporting system and the lack of recreational and commercial harvesting catch-effort data severely limits the analysis of the current trends. Future trends in the harvesting sec- tor and management research considerations will be discussed.

THE DEVELOPMENT OF THE HINGE LINE

IN TROPICAL MUSSEL

LARVAE OF THE GENUS PERNA

Scott E.Siddall

University of Miami

School of Marine and Atmospheric Science

4600 Rickenbacker Causeway

Miami, Florida 33149

There is confusion in the literature regarding the synonomy of several species of the genera Perna and Mytilus. To resolve this problem, I reared lar- vae of Perna perna from Venezuela and Perna viridis (= Mytilus viridis) from the Philippines through metamorphosis. All living material was quarantined in the laboratory. The results of my scanning electron microscopic examination of these larvae and the larvae of Mytilus edulis clear- ly differentiate hinge line development in the two genera. In both Perna species at optimal tempera- tures and salinities, hinge teeth are well developed

as early as 15 hours after fertilization. Shell length to height ratios decrease as development proceeds with no significant differences between the species. As the larvae approach metamorphosis, the central hinge teeth become more numerous and develop columnar structures. The provin- culum broadens laterally in larvae of both genera. Approximately 20 days after fertilization, a series of larger hinge teeth lateral to the central teeth develops in both P. perna and P. viridis in con- trast to larvae of the genus Mytilus. Differences in the crystal structure between the lateral and cen- tral teeth are apparent. I have examined metamor- phosing larvae of the only other member of the genus, Perna canaliculus from New Zealand and found similar lateral hinge teeth. The results of my work support the placement of Mytilus viridis in the genus Perna.

AGE AND MORPHOMETRIC VARIATION IN SUBTIDAL POPULATIONS OF MUSSELS

Robert S. Steneck1, Richard A. Lutz2. and Robert M. Cerrato2

department of Paleobiology Smithsonian Institution Washington, D.C. 20560

'Department of Geology and Geophysics

Yale University

New Haven, Connecticut 06520

The structure of two subtidal syntopic mussel populations (Mytilus edulis and Modiolus modiolus) was investigated. A total of 703 mussels were collected along two 10-meter transects at depths of 5 and 10 meters off Rutherford Island, Maine. Population densities for Mytilus of 50.4/m2 and 144. 4/m2 were recorded at 5 and 10 meters, respectively. Population densities of Modiolus were 71.2/m2 at 5 meters and 15.2/m2 at 10 meters.

A total of 263 Mytilus specimens of all sizes were randomly selected and ages accurately deter- mined through examination of annual patterns on acetate peels of sectioned shells. The maximum age (24 years) encountered within the population is considerably greater than that reported to date

PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION

87

within the literature. The age-frequency distribu- tion for the sampled population showed marked cyclic abundance patterns with 49.4% of the population restricted to yearclass multiples of 7. This observed periodicity may be a reflection of a regularly occurring natural event such as a preda- tor-prey oscillation.

Classical modal analysis of size frequency distributions cannot differentiate between annual

recruitment and cyclic abundance patterns with periods greater than 1 year. The potential for fre- quent interpretive errors by ecologists and paleon- tologists using modal analysis to infer age struc- ture is discussed.

Modiolus population structure is inferred by comparing crustose coralline incrustation rates for Mytilus of known ages with incrustations on Modiolus.

88

ABSTRACTS

NSA PACIFIC COAST SECTION

BREEDING DISEASE RESISTANCE INTO THE PACIFIC OYSTER

fohn H. Beattie', William K. Hershberger', Kenneth Chew1 and Conrad Mahnken2

1 University of Washington

College of Fisheries

Seattle. Washington 18105

and

'National Marine Fisheries Service

Seattle, Washington

Since the 1970s, the University of Washington, College of Fisheries has been involved in a cooperative study with the National Marine Fisheries Service and the Washington State Department of Fisheries investigating the causes of Pacific oyster (Crassostrea gigas) summer mor- talities. Studies have included laboratory testing using elevated water temperatures, nutrient enrichment of sea water, and isolation of im- plicated bacteria from moribund oysters to simulate summer conditions.

During the past three years, the College of Fisheries has also developed a selective breeding program as an effort toward the prevention of future kills through the introduction of disease resistant stocks. Survivors of laboratory high temperature challenges (above 18 C) were used as parents to produce potentially resistant F, families at the Poulsbo, Washington Sea Farms hatchery during 1974 and 1975. Challenges of these stocks in October, November, and December of 1976 in- dicated two families to be significantly more resis- tant to mortality conditions than imported Japanese stocks.

Thirty additional F, families were produced during 1976. F2 families have been produced in 1977 using the 1974 and 1975 F, stocks as parents.

Some of these 1975 families have been planted in areas of historic summer mortalities in Washington (Rocky Bay and Mud Bay). Represen- tative stocks of the 1975 F2 test animals have been given to two hatcheries for large-scale production and testing. Tissues are also being examined elec- trophoretically for specific genetic information.

OYSTER LARVAE MORTALITY IN SOUTH PUGET SOUND

RickD.Cardwell

Washington Department of Fisheries

Salmon Research and Development

Box 600, Pt. Whitney Road

Brinnon Washington 98320

In the summer, particularly late summer, marine waters from many areas of the Puget Sound basin as well as Willapa Bay are found to kill Pacific oyster (Crassostrea gigas) embryos and larvae when samples of these waters are brought to the laboratory and used to culture newly- fer- tilized embryos for 48 hours. Oyster larvae mor- tality can occur in waters unaffected by point and non-point sources of pollution and is most severe and widely distributed in several inlets in inner- most Puget Sound (SPS).

More than 15 years of biomonitoring the marine waters of SPS for acute toxicity to oyster embryos has disclosed the following: larval mortality is a seasonal phenomenon; in Budd and Eld Inlets, ap- preciable (>20%) mortality commences around June, peaks in late August to early September, and subsides in late October. The scope and severity of the mortality varies greatly between inlets, with Eld Inlet causing the highest average mortality, and Budd, Case, Totten, Carr, and Hammersley Inlets causing lesser mortality, respectively.

PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION

89

Within a particular inlet, mortality increases from the entrance to the head and varies significantly with water depth. Waters 10 to 30 ft in depth usually are more toxic than those at the surface.

The evidence suggests that the toxicity is caused by toxic metabolics of natural planktonic consti- tuents. Although more than one organism is im- plicated, recent evidence suggests that high den- sities of the dinoflagellate Ceratium fusus can ex- plain a significant percentage of the variation in larval mortality. Removal of most of the Ceratium by low pressure— not vacuum— filtration through a 10/^m screen greatly reduces toxicity.

Work is underway to define the identity of some of the suspected toxic agents and to relate varia- tions in circulation, water quality, and biological productivity between inlets to observed larval mortality.

Research has been supported in part by the Marine Ecosystems Analysis (MESAl-Puget Sound Project of NOAA.

HELICOPTER CRABBING

Darrell Demory

Oregon Department of Fish and Wildlife

Newport, Oregon

The newest inovation in crab fishing was off to a flying start off the southern Oregon coast early in 1977 when a helicopter was used to pull oversiz- ed crab pots. The helicopter angled with a 6-foot grappling hook on a 25-foot line. The 110-foot double-buoyed pot line was snagged and the 6- foot wide pot was pulled out of the ocean and flown to the landing site. The crabs were removed from the pot and the pot was rebaited and flown back to the fishing site. A loop of line attached to a small hook that was in turn attached to the grap- pling hook enabled the pilot to release the pot by putting slack in the line after the pot hit the sea- bed.

Small or female crabs caught in this manner were flown back to the water in a dump container that was lowered to the water where it was tipped over, sliding the crabs out.

Claims by some conventional crab fishermen that small crabs were being "rained across the beach" proved to be grossly exaggerated. Several

days of observation by Department of Fish and Wildlife staff and the State Police showed that the operation damaged very few crabs and was pro- bably less damaging than the boat fishery.

Although only 40 pots were fished, some boat fishermen fear an adverse economic impact on the fishery. The manager of the Oregon All Coast Fishermen's Marketing Association stated publicly that helicopter crabbing is too efficient .

Two helicopter operations made landings in 1977, but it is possible that others may soon become interested.

DISEASE CONTROL IN A MOLLUSCAN SHELLFISH HATCHERY

Richard A. Eissinger

International Shellfish Enterprises, Inc. Moss Landing, California

The emphasis in shellfish hatchery disease con- trol has been on treatment rather than prevention. This paper presents several operating concepts on disease prevention management employed at In- ternational Shellfish Enterprises in Moss Landing, California.

Brood stock must be carefully and frequently selected and histologically monitored to assure the customer that he will receive a disease, parasite and predator free seed product.

Routine sanitation practices through use of isolated culture systems, the careful avoidance of cross-contamination through sterilization of all culture implements, and the routine disinfection of the hatchery water delivery system will help to prevent transfer of disease from one hatchery location to another.

Daily monitoring for the presence of bacterial pathogens is extremely important in order to discover a disease before it becomes un- manageable. Monitoring becomes especially im- portant in screening algal cultures before they are used for feeding.

Antibiotic therapy is used only as a last resort to control disease. When using antibiotics, an effec- tive choice of antibiotic can only be made with the use of antibiotic susceptibility testing, and proper dosages can only be determined through the use of tube dilution sensitivity testing.

90

ABSTRACTS

Sanitation and bacterial monitoring can con- tribute significantly to stabilizing shellfish hatch- ery production output and helps to assure the customer that the product he buys is of the best possible quality and highest survival value.

SQUAXIN ISLAND MANILA CLAM RESEEDING STUDIES

]im Glock

University of Washington

College of Fisheries Seattle, Washington 98105

As a part of a series of studies to determine the future potential for reseeding beaches with small clam seed, Manila clams (Venerupis japonica) seed clams were planted at Squaxin Island, (Puget Sound) in May 1976 and have been monitored for growth and recovery. Various protective mechan- isms tested in this study have demonstrated dif- ferential effects on recovery rates. DuPont Vexar plastic netting, used at different sites, maintained recoveries of 75%, 97%, and 29% after 15 mon- ths. Preliminary migration studies have -shown that the clams move on the beach after planting: the potential use of this migratory movement to increase the concentration of clam stocks is also being investigated.

OYSTER BREEDING: WHERE CAN IT GO

William K. Hershberger

College of Fisheries University of Washington Seattle, Washington 98105

Based on the results that have been obtained with other cultured organisms, particularly with agriculturally-raised species, and the genetic variability that has been found in oysters, it should be possible to make significant strides in in- creasing commercial production of this animal. Before this possibility can be realized a concerted effort must be made, and a systematic program must be designed to initiate the process of domestication. By definition this procedure in- volves changing an organism for maximum per- formance under conditions that are amenable to

efficient culture and harvest by man. Present methods of commercial oyster culture employ on- ly part of the domestication process in that, generally a harvestable commodity is grown on the basis on uncontrolled reproduction and semi- controlled growing beds. These methods have some significant implications concerning the in- teraction of the genetic constitution of the oyster, its environment, and its production potential. These will be discussed with the use of standard genetic models and programs suggested to further enhance the production of oysters utilizing genetic methodology and systematic breeding procedures.

A COMPARISON OF SURVIVAL, GROWTH,

AND YIELD OF PACIFIC OYSTERS

(CRASSOSTREA GIGAS) FROM SEED

OBTAINED FROM DIFFERENT SOURCES

Chris R. Jones

Washington State Department of Fisheries

Shellfish Laboratory

600 Pt. Whitney Rd.

Brinnon, Washington 98320

Experimental plantings of the Pacific oyster (Crassostrea gigas) were conducted over a 3-year period with the objective of establishing the relative quality of various sources of oyster seed and to develop methods to improve survival and growth, especially during the first growing season. The results were mixed, depending on the area where grown and the methods used; but using conventional cultural practices, it appears that Japanese oyster seed will generally produce higher yields compared to equivalent groups of hatchery- produced or wild domestic seed. Japanese seed, however, may suffer higher mortality during the second year in some locations, indicating resistance to mortality for domestic oysters. The variables of seed size, seed density, hardening methods, and off-bottom rearing were evaluated. The initial size of the seed was found to be impor- tant in some cases, but the problems of small size were often more than offset by increased numbers of spat per shell. Density-dependent mortality was found to be linear through the range of densities studied. Studies also showed that some alternative methods of seed handling could improve the

PROCEEDINGS OF THE NATIONAL SHELLFISHERIES ASSOCIATION

91

ultimate yield. In particular, it was found that off- bottom rearing for the first growing season more than doubled the production by the end of the se- cond season. The improvement was especially ap- parent for the domestic seed. Some of the "conventional wisdom" regarding the relative quality of seed and the factors that determine quality were found not to be generally true, and it was apparent that each oyster grower must deter- mine the factors that are most important for his particular cultural situation.

PRELIMINARY REPORT ON MANILA CLAM

RESEEDING PROGRAM AT FIVE PUGET

SOUND BEACHES

Mark Miller1, Kenneth Chew1,

Charles D. Magoon2, Lynn Goodwin',

and Chris Jones2

'University of Washington

College of Fisheries Seattle, Washington, 98105

'Washington Department of Natural Resources Seattle Washington

iWashingto>i Department of Fisheries Brinnon, Washington 98320

Receiving support from the Washington State Sea Grant Program, the College of Fisheries along with the Washington Departments of Fisheries and Natural Resources have continued to investigate the feasibility of planting hatchery spawned Manila seed clams on Puget Sound beaches. Dur- ing the past l1 2 years the major emphasis of the project has been centered on determining the causes for low recoveries of planted clams. Studies carried out to gain a better understanding of the problems involved have included providing pro- tection for planted clams, planting clams in modified substrate, determining clam predators and testing for movement of planted clams. The rearing of small seed clams in suspended sand- fill- ed trays has also been conducted for the purpose of obtaining advanced size clam seed for subse- quent grow out on the beach. While the economics of clam reseeding do not yet favor commercial ap- plication, there seems to be a potential for this which may be realized by further research.

SEED OYSTER PRODUCTION IN THE SALTON SEA CALIFORNIA

Thomas L. Richards

Biological Sciences Department

California Polytechnic State University

San Luis Obispo, CA 93407

The Salton Sea is an inland body of salt water located in the southern California counties of Riverside and Imperial. The sea is approximately 70 kilometers long and 25 kilometers wide with the water surface about 65 meters below sea level. The salinity currently ranges from 33 0/00 to 37 0/00. The ionic content of Salton Sea water is not the same as that found in oceanic water as the salt comes from concentrated irrigation waste water. Water temperatures, taken three meters below the surface, range from winter low of 10 °C to a sum- mer high of 37°C.

Attempts in the 1950's to establish oysters in the Salton Sea were unsuccessful. Adult pacific oys- ters, Crassostrea gigas, will live about 30 days in Salton Sea water. It appears that even though the Salton Sea water has abundant plankton the water anesthetizes ciliary action and adult oysters are unable to feed. Experiments utilizing Salton Sea water to spawn oysters and culture larvae have shown that all stages survive well and a commer- cial spat fall can be obtained. Adequate spawning and culture temperatures are present from May through November each year. During this period natually occurring phytoplankton counts range from 0.5 to 3 million cells/liter.

Plans for a commercial oyster hatchery at the Salton Sea have been reviewed and modified by local, state, and federal regulatory agencies and all necessary permits have been granted. Sufficient financial support has been obtained to begin con- struction, and it is planned that the facility will set its first commercial oyster larvae during the late spring of 1978.

PRELIMINARY STUDIES ON THE

DEVELOPMENT

OF A SYNTHESIZED DIET FOR JUVENILE

OYSTERS, CRASSOSTREA GIGAS

M.A. Toner

92

ABSTRACTS

The growth of juvenile Pacific oysters fed equal amounts (based on ash free dry weights) of algae (Pseudoisochrysis paradoxa), lysed algae, dried oyster meat and encapsulated lysed algae was monitored in three experiments. Dried oyster meat and encapsulated lysed algae were also fed as sup- plements to lysed algae at three levels: 25, 50 and 75 percent of the total ash free dry weight of the diet. The growth of oysters fed lysed and whole algal cells with and without the addition of sodium sulfamethazine (Sulmet" ) was also com- pared. Growth was measured as the difference in ash free dry meat weight between treatments and an initial sample frozen at the beginning of the ex- periment.

In the first experiment, oysters fed algae alone showed an increase in ash free dry weight that was significantly greater than that of oysters fed dried oyster meat alone and as a supplement to algae. In the second experiment, oysters fed lysed algae treated with or without Sulmet"1 showed a signi- ficant decrease in ash free dry weight from the in- itial sample. Oysters fed algae with and without treatment with Sulmet" did not show a signifi- cant increase in ash free dry weight. In the third experiment, oysters fed starch encapsulated algae and algae with and without supplementation with encapsulated algae increased significantly in ash free dry weight as compared to the initial sample.

Encapsulation appears to be a process which could be used as a delivery system for feeding syn- thesized diets to juvenile oysters.

CLAM RESOURCE MEASUREMENT FOR ESTIMATION OF POLLUTION DAMAGE'

J. R. Vanderhorst and P. Wilkinson

Battelle Pacific Northwest Laboratories

Marine Research Laboratory

Sequim, Washington

Commonly used statistical procedures are geared to assure we do not state in error that a clam population has diminished in face of pollu- tion while little attention has been given to the corrollary need by private and public resource managers to have assurance that a given reduction in the clam resource will be detected. This paper deals with the sampling requirements for detection

of change in population abundance of native little neck clams (Protothaca staminea) on specific plots in the north Puget Sound region. Example data are presented which indicate that from 75 to 20,000 samples are required for detection of a 40 percent to 5 percent change in abundance on sample plots respectively. A pre-mapping procedure of sample stratification which effectively reduced the sam- pling requirement by one-half is also discussed.

This paper is based on work performed under U.S. Energy Research and Development Administration Contract EY- 76-C-06-I830.

MUSSELING IN ON A NEW MARKET Paul Waterstrat

University of Washington

College of Fisheries Seattle, Washington QS105

The Mussel Project is currently involved in 3 areas of study: mussel set prediction; hatchery development: and market investigation. Monitor- ing of plankton and spat collectors has revealved no significant mussel sets for spring and summer 1977 at study sites in Seabeck and Clam bays. Plankton and water quality factors of tempera- ture, salinity, and phytoplankton abundance will continue to be monitored to provide a historical baseline for the study sites.

The investigation of hatchery techniques for rearing and setting mussel larvae has been under- taken to provide information on the larval development and settlement of local mussel stocks and to develop a commercially feasible hatchery system. The use of larval dimensions for iden- tification of planktonic mussel larvae has been strengthened by information gained from the hatchery revealing that dimensions of larvae ob- tained from Puget Sound stocks of Mytilus edulis are comparable to those from New England stocks.

Recent newspaper and magazine articles coupl- ed with the publishing of a mussel cookbook have increased public awareness of the delights to be found in mussel dishes. A recent survey of 300 restaurants in Washington State reveals a signifi- cant interest in mussels by restaurant owners and

PROCEEDINGS OF THE NATIONAL SHELLFISHER1ES ASSOCIATION

93

identifies poor consumer awareness, inadequate supply and product quality to by the major im- pediments for development of this market. Ex- amination of mussels offered in retail markets reveals that a poor product quality may be a great detriment to the advancement of mussels as a household food item.

PIGEON POINT SHELLFISH HATCHERY: PAST, PRESENT AND FUTURE

Anthony Weaver

Pigeon Poi)it Shellfish Hatchery 921 Pigeon Point Road Pescadero, CA $4060

The Pigeon Point facility was established in 1965 as a shellfish hatchery for the mass culture of bivalve molluscs from egg to seed-sized juvenile stages. Cultchless seed was developed at Pigeon Point in 1968. Commercial success attained from this development stimulated further expansion in 1Q70 in which a new building was built at Pigeon Point and a cultch-setting facility was established in Moss Landing.

The expansion brought about serious produc- tion problems: larval mortality and inability to grow sufficient algae to feed the spat. During this period the Pacific Northwest seed market was pen- etrated, but the inability to produce quality seed quickly resulted in marketing failures.

The production failures became increasingly severe and in 1°74 the operation went into a well- financed research and development mode result- ing in a successful pilot production program two years later.

In January, 1976, a highly favorable technical evaluation was made of the facility by an outside group of mariculture scientists from such institu- tions as Scripps, Woodshole, U.C., and O.C.U. The group also included individuals from com- mercial mariculture businesses.

On the basis of this evaluation, the facility was redesigned and successfully put into commercial operation in November, 1976.

Today, Pigeon Point Shellfish Hatchery is again producing high quality cultchless oyster seed at a yearly rate of about 30 million seed. A wide varie- ty of species and sizes ranging from 2-3 mm and 25-40 mm (1 to 1.5 inches) are available.

While growers in Oregon have had a great deal of success with the seed, the majority of the pro- duct is planted in Europe and the Eastern United States.

In the future, Pigeon Point intends to increase production levels significantly and add new spe- cies to the inventory list. We hope that more growers in the Pacific Northwest will try our large cultchless seed in their growouts. We would great- ly appreciate suggestions of how we might be able to fit into the oyster industry in this region. Samples of our products are available at limited cost.

VIBRIO ANGUILLARUM

AND LARVAL MORTALITY

IN A CALIFORNIA COASTAL SHELLFISH

HATCHERY

Ronald Thurber Zebal

Pigeon Poitit Shellfish Hatchery Pescadero, Ca. $4060

The Pigeon Point Shellfish Hatchery has had a mixed history of success in developing profitable oyster spat production primarily because of larval mass mortalities due to unknown causes. The predominant bacterial species recently isolated from mobid oyster veliger larvae produced under hatchery conditions was identified as Vibrio anguillarum. Infection of healthy larvae, and re- peated recoveries of this bacterium from diseased larvae confirmed this bacterial species to be one cause of serious epizootics within the hatchery. Routine screening for this organism and observa- tion of larval behavior, coupled with appropriate hatchery design and management techniques have controlled this cause of larval mortality.

In the paper "Mercenaria culture using stone ag- gregate for predator protection" by Castagna and Kraeuter, Vol. 67, 1977, there is an error in the last sentence before the results. The sentence states, "Once the area had been stabilized, small clams were broadcast over the aggregate at an average density of approximately 31/sq. m." It should have read 3100/sq. m.

94

MBL

WHOI

12/19/07

71c

: . .

'

S .

■

■ iii

mflPMPjHflMWwwnc

)INGS

LFISHERIES "ION

VOL. 69

...

,•' " A

■ t „ ... P .. ;

|**V*.r,'

IW,

•' • . :

I I •

I • I

I i i < r

■ i i

i • »

! .

■

I

35

PROCEEDINGS

OF THE

NATIONAL

SHELLFISHERIES

ASSOCIATION

CONTENTS Volume 69 — June 1979

List of Abstracts by Author of Technical Papers Presented at 1978 NSA Annual Meeting, New Orleans, Louisiana, and NSA Pacific Coast Section, Portland, Oregon v

Gordon Gunter

The Grit Principle and the Morphology of Oyster Reefs 1

Richard F. Dame

The Abundance, Diversity and Biomass of Macrobenthos on North Inlet,

South Carolina, Intertidal Oyster Reefs 6

D. S. Haven, J. P. Whitcomb, J. M. Zeigler and W. C. Hale

The Use of Sonic Gear to Chart Locations of Natural Oyster Bars

in Lower Chesapeake Bay 11

James W. Glock and Kenneth K. Chew

Growth, Recovery, and Movement of Manila Clams, Venerupis japonica

(Deshayes) at Squaxin Island, Washington 15

Neil Bourne

Razor Clam, Siliqua patula Dixon, Breeding and Recruitment

at Masset, British Columbia 21

Peter J. Eldridge, Arnold G. Eversole, and Jack M. Whetstone

Comparative Survival and Growth Rates of Hard Clams, Mercenaria mercenaria,

Planted in Trays Subtidally and Intertidally at Varying Densities

in a South Carolina Estuary 30

Steven A. Murawski and Frederic M. Serchuk

Shell Length-Meat Weight Relationships of Ocean Quahogs,

Arctica islcmdica, from the Middle Atlantic Shelf 40

David Dean

Impacts of Thermal Addition and Predation on Intertidal Populations of

the Blue Mussell, Mytilus edulis L 47

Leslie E. Haley

Genetics of Sex Determination in the American Oyster 54

Robert E. Palmer and Melbourne R. Carriker

Effects of Cultural Conditions on Morphology of the Shell of

the Oyster Crassostrea virginica 58

Lynn Goodwin, Warren Shaul, and Conrad Budd

Larval Development of the Geoduck Clam (Panope generosa, Gould) 73

•

.

MBl. WHOI LIBRARY

UH 1ABB $

. ^ . '" ; 4 4 A

!

- ■

?

! I

' ',

■ ; ".,.', i ;

■

. .

■

. ■ . ■■

I "':A I

	'.	• ■-■ ,
	■

■

■

4 . ;

■

1

I A,i I t •

I r 1 r r;

j * iv- » *

I i i iW> iV $i

fc •-

'

■

J

< ■ " ■ ..'' :' s ■

. :■ ■ ! i/ ' ■: i

!

i . : ■

!

■ ■■■ • . ■ ; . : r. \ '

-

f r p f |f i- .

■■ .

Ii &.J

I

•

■ •'■ * ' ■

i • II I t. 11 • • i

iv- \ » I »

IF » ' •

i

■

■

■

I

i.A'!4 I

\ .

% |

it ..i. i *

- .

'

B

;'$ ,j| .?;

, ■■. . K ■ •■;

.

!

•	M		El	Bv '^Bt ^ BE ; Ba
■	;M ' V M
\
i -i

:

' • ' ' • 'I

.

I

.

. !