Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effects of artificial diet on gonad size, egg size, egg quality, and larval vitality in green sea… Westman, M. Marja de Jong 1994

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1994-0118.pdf [ 2.25MB ]
JSON: 831-1.0087415.json
JSON-LD: 831-1.0087415-ld.json
RDF/XML (Pretty): 831-1.0087415-rdf.xml
RDF/JSON: 831-1.0087415-rdf.json
Turtle: 831-1.0087415-turtle.txt
N-Triples: 831-1.0087415-rdf-ntriples.txt
Original Record: 831-1.0087415-source.json
Full Text

Full Text

THE EFFECTS OF ARTWICIAL DIET ON GONAD SIZE, EGGSIZE, EGG QUALITY, AN]) LARVAL VITALITY INGREEN SEA URCHINS,Strongylocentrotus droebachiensis,OF AQUACULTURE SIGNIFICANCEbyM. Marja de Jong WestmanB.Sc. (Hons.), University of British Columbia, 1975THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIES(ZOOLOGY)We accept this as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIADecember 1993© M. Marja de Jong Westman, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.Department of OOt04/The University of British ColumbiaVancouver, CanadaDate \1442&245/ ‘1’t) /999DE-6 (2/88)ABSTRACTEffects of long-term dietary conditioning on gonadal growth, lipid, and moisture content,and on egg size, egg-energy content, larval morphometry, and larval development of thegreen sea urchin Strongylocentrotus droebachiensis were studied in the laboratory. Onealgal and seven artificial diets were tested over a nine-month period, coinciding with theurchin’ s annual reproductive cycle. Formulated diets differed in protein level and in thepresence or absence of different supplements, such as algal storage carbohydrates(mannitol and algin), cholesterol, and B-carotene. Diets were: 1)10% protein (LO-PRO),2)10% protein+ mannitol (LO-PRO+M), 3)10% protein+algin (LO-PRO+A), 4) 20%protein (HI-PRO), 5) 20% protein+cholesterol (HI-PRO+C), 6) 20% protein+B-carotene(HI-PRO+B), 7) 20% protein+B-carotene+cholesterol (HI-PRO+C/B), and 8) driedseaweed Nereocystis luetkeana (KELP). Urchins reared on the low-protein dietformulations had significantly smaller gonad indices than urchins conditioned on the high-protein and algal diets. Gonadal lipid and moisture contents varied significantly over timebut not among dietary treatments. Significant differences in egg sizes were observed.HI-PRO produced the smallest eggs and HI-PRO+C the largest, while other diet groupsproduced intermediate-sized eggs. Egg organic-carbon levels also differed significantlywith the largest amounts being present in HI-PRO+C/B, and the lowest in LO-PRO andHI-PRO. Larvae cultured from eggs of selected diet groups (LO-PRO, HI-PRO, HIPRO+B, HI-PRO+C/B, and KELP) showed significant differences in morphometry andin rates of development. HI-PRO+B and HI-PRO+C/B were the first to reach thefeeding pluteal stage at 3 days and the 2-arm pluteus at 5-6 days. The feeding pluteal stagewas attained by LO-PRO and HI-PRO at 4 days and the 2-armed larvae at 7-8 days.Larval development times for LO-PRO, HI-PRO, and HI-PRO÷C/B were comparable byIIthe 4-arm stage and remained so throughout development. HI-PRO+B larvae consistentlyhad the shortest development times, reaching the fully-formed 8-arm larval stage 3 daysearlier than larvae from other diet groups. HI-PRO+B larvae also had the highest rate ofmetamorphosis success. They showed significantly higher diary band/body-length ratiosthroughout all developmental stages, indicating greater larval feeding effectiveness andsuggesting that this was the key to their overall success. All larvae arising from adults fedthe artificial diets were competent; however, larvae from adults fed KELP showed severeabnormalities and subsequent early mortality. With regard to an aquaculture program, Iconclude that HI-PRO÷B is the best of the diets tested, in that it supported high gonathigrowth for the roe market, and produced healthy broodstock with resultant vigorouslarvae.ifiTABLE OF CONTENTSABSTRACT iiTABLE OF CONTENTS ivLIST OF TABLES viiLIST OF FIGURES viiiACKNOWLEDGEMENTS ixGENERAL INTRODUCTION 1SECTION 1: FISHERIES, RECRUITMENT BIOLOGY AND AQUACULTUREOF SEA URCHINS.Sea Urchin Fisheries 2Recruitment Success of Sea Urchins 3Aquaculture Potential ofGreen Sea Urchins 4SECTION 2: EFFECTS OF ADULT DIET ON GONADAL GROWTH, LIPIDAND MOISTURE CONTENT.INTRODUCTION 5MATERIAL AND METHODS 9Collection and Main:e,wnce ofAnimals 9Dietary Conditioning and Diet Formulations 9Feeding Program 13General Methods 14Test Diameters 14ivWet Weights . 14Gonad Index 14Sex Determination 15Lipid and Moisture Contents ofthe Gonads 15RESULTS 16Effect ofDiet on Survival ofAdults 16Effect ofDiet on Lipid and Moisture Contentsof the Gonads 16Effect ofDiet on Test Growth 18Effect ofDiet on Gonadal Growth 18DISCUSSION 22SECTION 3: EFFECTS OF BROODSTOCK CONDITIONiNG ON EGGS ANDLARVAE.INTRODUCTION 27MATERIALS AND METHODS 33Spawning and Fertilization 33Egg Size 34Egg Organic-Carbon 35Larval Culturing 35Larval Development 36Larval Morphometry 37RESULTS 39Effect ofAdult Diet on Egg Size 39Effect ofAdult Diet on Egg Energy Content 39Effect ofAdult Diet on Fertilization Success 42VEffect ofAdult Diet on Timing ofLarvalDevelopment 42Effect ofAdult Diet on Larval Survival 44Effect ofAdult Diet on Larval Morphometry 44Effect ofAdult Diet on Metamorphosis Success 54DISCUSSION 56SECTION 4:GENERAL CONCLUSIONS 63APPLICATIONS TO AQUACULTURE 64REFERENCES 66viLIST OF TABLES1. Chemical composition of artificial diets prepared as conditioningdiets for adult Strongylocentrotus droebachiensis 112. Summary of lipid and moisture values of gonadal tissue fromfemale Strongylocentrotus droebachiensis sampled in Novemberand March 173. Gonad indices for Strongylocentrotus droebachiensisconditioned on eight dietary treatments for nine months 204. Summary of egg sizes and egg organic carbon levels ofStrongylocentrotus droebachiensis conditioned on eight differentdietary treatments 405. Summary of egg fertilization success of eggs arising fromStrongylocentrorns droebachiensis reared on five selecteddiet groups 416. Summary of larval rudiment diameters as measured in 8-arm andcompetent Strongylocentrotus droebachiensis larvae 55viiLIST OF FIGURES1. Gonad index values as a function of dietary treatment and time foradult Strongylocentrotus droebachiensis conditioned on eightdietary treatments for nine months 192. Developmental stages of a sea urchin 313. Larval measurements to assess morphometry of developinglarvae 384. Comparisons of development rates and metamorphic success of larvaeproduced by diet-conditioned adultStrongylocentrotus droebachiensis 435. Relationship between overall larval length and stages of developmentfor Strongylocentrotus droebachiensis larvae 456. Relationship between body length and stages of development forStrongylocentrotus droebachiensis larvae 477. Relationship between total arm length and stages of development forStrongylocentrotus droebachiensis larvae 488. Relationship between ciliated band length and stages of development forStrongylocentrotus droebachiensis larvae 509. Relationship between the ratio of ciliated-band length and body lengthand stages of development for Strongylocentrotus droebachiensislarvae 5110. Photographs of Strongylocentrotus droebachiensis larvae arising fromfive selected dietary treatments 53vmACKNOWLEDGMENTSI would like to thank my supervisor Dr. Thomas Carefoot for his advice and helpduring the course of this study and a special thanic you to committee member Prof. BerylMarch for her guidance and interest in this project. Also, I appreciated the advice of theother committee members, Dr. George Iwama and Dr. Tim Parsons. I would also like tothank Dr. Ed Donaldson and Dr. I. Stockner for allowing me the use of facilities at theD.F.O. lab in West Vancouver where all the experiments were conducted. Thanks too, toLinda Townsend at the Nanaimo Biological Station for supplying the phytoplanktoninocula, to Maric Goliner for his advice regarding phytoplankton culturing and larvalphotography, to Walter Hagen for his help with the gonad lipid analyses, to Dr. Pci-YuanQian for his aid with the larval culturing experiments and to Tony Westman for his helppreparing the figures included in the thesis.xGENERAL INTRODUCTIONSea urchins have been fished in British Columbia’s waters since 1970 for theprovision of roe to the Japanese sushi trade. Stocks here are presently considered plentiful;however, populations in other areas of the world are in serious decline and it is likely thaturchins in general cannot sustain current fishing pressures. With the expected depletion ofB.C. wild stocks, the increased demand for roe, the frequent small size and poor quality ofharvested roe, and the green sea urchin’s biological suitability as a species for aquaculture,I propose that Strongylocem’rotus droebachiensis is a valid candidate for an aquacultureprogram. The present study was designed to find an artificial diet which would lead toboth good gonadal growth required for the roe market and vigorous larvae needed for asuccessful breeding program. The thesis is organized into four sections. Section 1reviews the history of the fisheries, urchin recruitment biology and the suitability of thegreen sea urchin for aquaculture. Section 2 explores important aspects of the reproductiveand feeding biology of the urchin and deals with experiments conducted on the relationshipof adult diet to gonadal growth. Section 3 investigates the effects of selected adult diets onegg size and energy content, and larval morphometry and development. Section 4concludes the fmdings and discusses their applicability to aquaculture.1SECTION 1: FISHERIES, RECRUITMENT BIOLOGY AND AQUACULTURE OFSEA URCHINS.Sea Urchin FisheriesSea urchins are fished worldwide for their roe. Historically, world sea urchinlandings have been dominated by Japan, with large contributions coming from Californiaand Chile (Keegan et al. 1984). With the exception of Japan, which successfully sustainsan aquaculturally enhanced wild sea urchin fishery (Anon. 1984), sea urchin stocks in otherareas of the world, such as California, Washington, Mexico, and Chile, have beenoverexploited and are in serious decline (Sloan 1984, Gonzalez et al. 1987). BritishColumbia’s urchin roe industry is relatively new and exists solely to satisfy the Japanesemarket for fresh roe in the sushi trade (Breen et al. 1976, Conand and Sloan 1989). Redsea urchins began to be harvested along our coast in 1971, greens began in 1987, andsubsequent years have seen a steady increase in landings for both (Anon. D.F.O. Stats.1985, 1991a,b).Three sea urchin species inhabit B.C. waters: the red, Strongylocentrotusfranciscanus; the purple, S. purpuratus; and the green, S. droebachiensis. Only red andgreen urchins are actively harvested (Breen et al. 1976, Anon.D.F.O.Stats 1985 and199 la,b). Although British Columbia’s sea urchin population is generally thought to beunderexploited (Conand and Sloan 1989, Anon. D.F.O. Stats. 1991a,b), pockets ofoverharvesting have already been recorded (Bell, B. and 3. Hereford: pers.comm.). TheB.C. fishery is managed in a preemptive fashion. Harvesting is controlled through a suiteof area closures, seasonal closures, area landing quotas, and size limits (Frances DicksonD.F. 0.: pers. comm). However, surveillance and enforcement of regulations are difficultbecause harvesters are highly mobile over large coastal areas (Conand and Sloan 1989).2The short history of the fishery is probably the key factor in the present“healthiness” of the resource in B.C. In other areas worldwide, however, overfishingcombined with naturally low and sporadic recruitment are causing the decline of sea urchinstocks (Tegner and Dayton 1977, Gonzalez et a!. 1987, Pearse and Hines 1987, Sloan eta!. 1987, Conand and Sloan 1989). Fishing not only reduces the broodstock, but may alsoaffect recruitment of larvae and juveniles (Pearse and Hines 1967, Tegner and Dayton1977, Sloan et al. 1987). This is because adult urchins of several species are known toprovide a “nursery” for juveniles under their protective canopy of spines (Tegner andDayton 1977, 1981). Grazing by adults also keeps clear the surrounding rock and frees upcoralline algal surfaces known to act as inducers for urchin larvae settlement andmetamorphosis (Cameron and Schroeter 1980, Johnson and Mann 1982, Hart andScheibling 1988, Rowley 1989). Fishing, therefore, not only depletes the stocks in termsof actual numbers of animals, but also has long-term effects on the dynamics of localpopulations.Recruitment Success ofSea UrchinsEven without the pressures of fishing, recruitment of untouched wild populations islow and sporadic (Tegner and Dayton 1977, Pearse and Hines 1987, Sloan eta!. 1987)In Oregon, Ebert (1983) observed that S. purpuratus showed no or poor recruitmentfrom 1964-1973 and, based on differences in class sizes, he concluded that recruitment ofthe magnitude seen in 1963 had not occurred since at least 1959. This suggests, then, thatover a 14-year period there were only two successful recruitments. Similar trends withother species have been noted worldwide (Tegner and Dayton 1981, Ebert 1983, Gonzalezet al. 1987, Hart and Scheibling 1988a).3In conclusion, since wild sea urchin populations have difficulty sustaining afishery and to guarantee British Columbi&s long-term success in an expanding roe market,an aquaculture program should be seriously considered.Aquaculture Potential ofGreen Sea UrchinsThe green sea urchin is biologically suitable for aquaculture. Adults in culture canpotentially provide a regular and reliable annual source of seed. Juveniles grow quicklyand reach sexual maturity in two years (Swan 1966). Despite living in high densities (up to350 animals m2;Mottet 1976), urchins resort to cannibalism only in periods of low foodsupply (Himmelman and Steele 1971), and are relatively disease free. Lost spines can beregenerated, damaged ones repaired and woundings of 1 cm or less in the test healed within2 weeks (Ebert 1967,1968; Edwards and Ebert 1991). Finally, urchins can tolerate a widerange of physical and biological stresses, such as salinities 50-120% of normal seawater,temperatures of 2-24°C, quiet and wavy water, substantial water loss, and starvation(Moore 1966).4SECTION 2: EFFECTS OF ADULT DIET ON GONADAL GROWTH, LIPID ANDMOISTURE CONTENTS.INTRODUCTIONUrchins are generalist feeders able to obtain and utilize a wide variety of foodincluding drift and attached algae, diatoms, zooplankton, and fish (Himmelmann and Steele1971, Lawrence 1975, Vadas 1977, Larson er at. 1980, Dc Ridder and Lawrence 1982,Johnson and Mann 1982, Himmelman and Nedelec 1990, Kenner 1992). Diet varieswith locality and season (Gonor 1973, Lowe and Lawrence 1976, Dc Ridder and Lawrence1982). In the field, urchins aggregate in areas of high algal abundance, yet they appear torespond as strongly to animal matter as to plant (MCClintock etal. 1982). Urchins have akeen sense of chemoreception. Vadas (1977), Larson etat. (1980) and Himmelman andNedelec (1990) have described sensitive chemoreceptive responses of Strongylocentrotusdroebachiensis to algae, noting positive directional and behavioural (increased activity oftube-feet and lantern teeth) responses to preferred algae and avoidance of algal speciespossessing chemical defenses.Nutritional requirements of sea urchins are poorly known. However, the presencein gut extracts of carbohydrases (arnylase, alginase, laminarinase, ceilulase and, to a lesserextent, mannosidase) and proteases, lipases, and phospholipases suggest that they canutilize a wide variety of foodstuffs (Lawrence 1982, Obrietan et at. 1991).Gut microflora are plentiful in echinoids, and appear to play a role in digestion andalso nutrition (Ferguson 1969, Fong and Mann 1980, Lawrence 1982). It is thought thatthe bacteria acquired from the foods eaten provide needed digestive enzymes and act assites of nitrogen fixation (Lawrence 1975, 1982). In their study of barren-ground seaurchins, Johnson and Mann (1982) suggest that microflora are not as important in5supplementing organic nitrogen with bacterially-fixed nitrogen as they are in synthesizingessential amino acids and reorganizing the composition of dietary amino acids to moreclosely resemble the amino acid complement of the urchin.Absorption efficiencies of echinoids vary with food type (Lawrence 1975, Loweand Lawrence 1976, Vaths 1977), with algae containing high levels of soluble proteins andcarbohydrates being absorbed most efficiently. Although lipid is absorbed well thereappears to be little digestion and absorption of structural carbohydrates (Lawrence and Lane1982). In their studies of the effect of diet on growth in S. droebachiensis, Vaths (1977),Larson et a!. (1980), and Keats et a!. (1984), noted that absorption efficiencies forpreferred algal species were generally greater than those for non-preferred.Gonad size in sea urchins differs between years and geographical locations,indicating that growth of the gonad is influenced by local environmental factors. Inparticular, sea urchins are sensitive to both food quantity and quality, showing differencesin gonadal and somatic growth in relation to nutrient supply (Greenfield et al. 1958, Gonor1973, Lawrence 1975, Vadas 1977, Himmelman 1978, Larson et a!. 1980, Johnson andMann 1982, Thompson 1982, Keats etal. 1984, Andrew 1986, Edwards and Ebert 1991,Levitan 1991, Kenner 1992). Gonadal development is often used as an index of thenutritional condition of the animals, not only because of the dependence of gonad growthon diet quality, but because sea urchins have a relatively large annual investment inreproductive effort. At maturity they are essentially a sac of gonads surrounded by acalcareous test.The gonad acts as the main organ of nutrient storage for sea urchins (Giese 1966,Gonor 1973, Walker 1982). Large amounts of protein and lipid accumulate in the gonadsprior to gonad development (Giese eta!. 1959) and, during conditions of low food supply,6the gonad can decrease in size (Johnson and Mann 1982) as nutrient and energy materialsaie allocated to maintenance functions. These nutrient stores are used for gametogenesis,and it has been shown that an inverse relationship exists between the size of the cells of thegonad which store nutrient reserves (nutritive phagocytes) and the gametogenic cells(Walker 1982). Once nutrients are positioned in the gonad, the gonads are relatively self-sufficient in terms of their nutritional requirements for gametogenesis (Walker 1982).Himmelmann’s (1978) study of the reproductive cycle of Strongylocentrotusdroebachiensis population at First Narrows, Vancouver, B.C. over a four-year periodshowed seasonal fluctuations in gonad biomass related to the annual spawning cycle. Heshowed that the gonad index peaked in February and March and dropped abruptly fromMarch to June as a result of spawning, a pattern also known for other west coast locationsand for different species (Boolootion 1966, Mottet 1976, Vadas 1977).Gonadal development for the green sea urchin can be divided into fivesuccessive stages: 1) resting, when gonad size is at a minimum following spawning, 2)growing, which is accompanied by active feeding and occurs during the late summer, fall,and winter, 3) premature, when the gonad is large in size but not mature, and colourdifferences are evident between sexes (this stage provides the best commercial quality forsushi), 4) mature, when the gonad weight is at a maximum (but because of the high watercontent, this stage has little commercial value), 5) spawning, when sperm and eggs areextruded from the gonads and there is related drop in gonad size.7Effects of specific dietary components on gonadal growth in urchins have not beenextensively studied; however, findings from previous studies provide suggestions as towhich nutrients are worthy of testing. For example, it is known that brown algae supporthigh rates of gonadal growth (Vadas 1977, Larson et al. 1980, Keats et al. 1984) but it isnot known which nutrient components of the algae are responsible. Lilly (1975) correlatedthe growth of the tropical sea urchin Tripneustes ventricosus to protein and caloric levelsof food and concluded that level of protein was more important. Miller and Mann (1973)proposed that urchins must consume large quantities of nitrogen-poor foods to obtainsufficient protein for their nutritional needs, and Lowe and Lawrence (1976) suggested thatthe growth and reproductive rates of sea urchins are probably more dependent on amountof protein than upon amount of energy acquired in feeding. Tests of protein levels ongonadal growth can be done readily using artificial diets.Given the future potential need of sea urchin farming, development of suitableartificial diets becomes an important issue. Yet, to date, no successful artificial diets havebeen formulated for sea urchins. Because of the dramatic changes in gonad size throughoutthe reproductive cycle it is likely that differential gonad growth could be observed fromdiets of varying nutritional composition, and this is the approach used for the first part ofthe present study. The study utilizes artificial diets of known composition, withformulations based on potentially important nutritional components and additives. Specificresearch questions investigated are: 1) assessment of the feasibility of using artificial dietsin the maintenance of adult sea urchins, and 2) assessment of selected nutritionalcomponents on gonadal growth.8MATERIALS AND METHODSCollection and Maintenance ofAnimals1040 adult Strongylocentrotus droebachiensis were collected using SCUBA atdepths ranging from 5-10 m from Caulfeild Cove, West Vancouver, B.C. The collectionsite consisted primarily of sloping solid rock formations covered in fine sediment. Densepopulations of urchins were present and little algal growth was observed. Urchinscollected were of a similar size, with test diameters ranging between 5-7 cm. Thepopulation, therefore, probably represented a similar year class of animals 6-7 years old(Miller and Mann 1973, Ebert 1975). The animals were transported “dry” in plastic meshbaskets to West Vancouver Laboratory, Department of Fisheries and Oceans, a 10 minutedrive from the collection site. All experiments were conducted at this facility.Urchins were randomly divided into eight groups with 130 each, and reared in 200L indoor tanks supplied with a continuous flow of unfiltered seawater. Seawater waspumped from 18 m depth outside of the lab. Temperatures varied from 6-14°C andsalinity from 28-32%o over the nine-month study period. Artificial light was provided byoverhead “daylight” fluorescence and photoperiods followed ambient outdoor conditions.Handling of animals was minimized since both locomotory and feeding responses areknown to be greatly reduced with excessive handling (Vaths 1977).Dietazy Conditioning and Diet FoimulationAnimals in each tank were provided one of eight diets: seven artificial diets and onealgal diet consisting of air-dried bull kelp Nereocystis luetkeantj. The artificial dietsdiffered in protein concentration and in the presence or absence of algal storagecarbohydrates (mannitol and algin) , cholesterol, and B-carotene (see Table 1 for diet9compositions). The two basal diets (LO-PRO and HI-PRO) contained two concentrationsof protein respectively, 10% and 20% d.w. The 10% value was selected on the basis thatit represents the low end of the range of protein levels found in algal species known to beeaten by sea urchins (Carefoot unpubL).Three variations of 10% protein diets were formulated, designated LO-PRO. Twocontained additions of either mannitol (LO-PRO+M) or algin (LO-PRO+A), while the otherhad no additives (LO-PRO). Mannitol and algin are commonly occurring storagecarbohydrates in brown algae. Mannitol is a component of the polysaccharide laminarin,and algin is a complex phycocoiloid occurring in cell walls and intercellular spaces.Mannitol and algin can comprise 25-50% and 14-40%, respectively, of the dry weights ofphaeophytes (Percival and McDowell 1967). Both mannitol and algin are known to bereadily absorbed by echinoids (Boolootian and Lasker 1964), and Lawrence (1975) andVadas (1977) showed that both absorption efficiences and gonad growth were highest insea urchins fed the brown keips Macrocystis and Nereocystis (both genera rich in thesestorage carbohydrates). Additions to the “basal” 10% diet were done without changing thebalance of vitamins, lipids, and other “essential” ingredients by varying the amount ofcornstarch. As noted, carbohydrates are believed to be nutritionally less important tourchins than proteins and other nutrients.10Table 1: Chemical composition of artificial diets (% dry weight) used in experiments todetermine the effect of adult diet on gonadal growth, egg size and quality, and larval vitalityin the green sea urchin Strongylocentrotus droebachiensis. Symbols are:LO-PRO = 10% protein; LO-PRO+M = 10% protein plus mannitol; LO-PRO+A = 10%protein plus algin; HI-PRO = 20% protein; HI-PRO+C = 20% protein plus cholesterol;HI-PRO+B = 20% protein plus beta-carotene; HI-PRO+C/B = 20% protein pluscholesterol and beta-carotene.©DIET LO-PRO LO-PRO+M LO-PRO+A HI-PRO HI-PRO+C HI-PRO+B HI-PRO+C/BCornstarch 64.7 55 55 31.7 31.5 31.7 31.5Mannitol 0 9.7 0 0 0 0 0Algin 0 0 9.7 0 0 0 0GroundWheat 15 15 15 40 39.8 40 39.8Fish Solubles 10 10 10 14 13.9 14 13.9(Condensed)Albumen 2 2 2 6 6 6 6Herringoill 2 2 2 2 2 2 2Corn oilLecithin 0.3 0.3 0.3 0.3 0.3 0.3 0.3Binder 3 3 3 3 3 3 3Vitamins! 3 3 3 3 3 3 3Minerals *Cholesterol 0 0 0 0 0.5 0 0.5B-carotene 0 0 0 0 0 0.006 0.006Total % 100 100 100 100 100 100 100* Composition given below:VITAMINS I MINERALSVITAMINS % MINERALS % *Riboflavin 0.31 Ca (CaHPO4) 61.86Ca Pantothenate 0.62 Mg (MgSO4) 5.41Niacin 1.24 Zn (ZnO) 0.19Biotin 0.02 Mn (MnSO45H2O) 0.17Folacin 0.08 I (1(103) 0.03PyridoxineHCl 0.15 Fe (FeSO4) 0.71Thiamine 0.19 Cu (CuSO4) 0.02Choline Chloride 16.47 Co (CoCl6H2O) QJ-(_a-Tocopherol 12.37 100%acetate* percentages refer tothe mineral elements11Four 20% protein diets were formulated, designated HI-PRO. Three of thesevaried in the contents of B-carotene and cholesterol. One diet consisted of a 20% proteinbase plus B-carotene (HI-PRO÷B). B-echinenone is the principal carotenoid in sea urchinsbut is derived from dietary B-carotene (Vevers 1966, Ferguson 1969, Tsushima andMatsuno 1990). In addition, it is thought that carotenoid-containing pigment cells play anutritional role in the development and maturation of the egg (Burke and Bouland 1989).On these bases, I considered that B-carotene may potentially have a positive influence onthe egg pigment cells and their function. Also, brown algae which are known to supportgood gonadal growth in sea urchins contain high levels of B-carotene. The role of B-carotene as a precurser to vitamin A and this vitamin’s role in the synthesis of glycoproteinsadded to its interest as a component to test. A second 20% protein diet containedcholesterol (Hl-PRO+C). Cholesterol was selected because it is a dietary requirement forsome lower animals and is the most commonly occurring sterol in echinoids. A third 20%protein diet contained a combination of B-carotene and cholesterol (HI-PRO +CIB). Thefourth 20% protein diet contained no additives (HI-PRO). Finally, the bull kelpNereocystis luetkeana was chosen as a representative natural diet (KELP) because it hasbeen found to give both high somatic and gonadal growth in Strongylocentrotusdroebachiensis (Vadas 1977). However, because of its annual cycle, live bull kelp wasnot available for the entire study and dried kelp was chosen instead. The kelp cameprepared as air-dried fronds from the Barkley Sound Kelp Company. Prior to shipment, itwas harvested wild, dried indoors at 26°C for 10 hours, and then dried at roomtemperature for a further 11 hours.Artificial diets were prepared in pellet form. Diet preparation consisted of thefollowing steps: (a) preparation of “vitamin-mineral pre-mix”, (b) measuring and mixing ofdry ingredients (cornstarch, ground wheat, albumen), (c) measuring and mixing of liquid12)ingredients (corn oil, herring oil, fish solubles, lecithin), (d) combining liquid and dryingredients, (e) addition of binder, and (I) pelleting, done in the following procedure:1. Components were measured to make up 5 kg of each diet. Diet preparation wasdone twice through the study period, once in July and again in November.2. Minerals and vitamins were measured individually.3. Minerals were ground together using a mortar and pestle.4. Vitamins were mixed by combining and shaking in a plastic bag.5. Vitamins and minerals were then mixed with 1 kg of cornstarch to form thepre-mix. This vitamin-mineral premix was then refrigerated until later use.6. Required amounts of dry ingredients (cornstarch, groundwheat, albumen) weremeasured for each diet.7 Required amounts of liquid ingredients (fish solubles, corn oil, herring oil, lecithin)were measured for each diet.8. Required cholesterol was added directly to the corn oil and mixed, prior to beingadded to wet ingredients.9. Dry ingredients were mixed together for 15 minutes. After thorough mixingof dry ingredients, liquid ingredients were then added while mixer was functioning.10. Calcium ligno-sulfonate binder was then mixed with 250 ml of water and added toother ingredients while mixing continuously.11. Once the desired consistency of mixture was reached, diets were processed in alaboratory size California pellet mill (at U.B.C.) or steam pelleter (at D.F.O. Lab).12. Large quantities of pellets were stored at -18°C and quantities required for weeklyfeedings were kept in a cold room at +5°C.Feeding ProgramUrchins were not fed two weeks prior to the feeding program to allowacclimatization and to promote uniformity of initiation of feeding. Each of the experimentaldiets was assigned to 1 tank of urchins. Feeding was three times weeldy thoughout thenine month experimental period. Individuals received approximately 0.9 gram dry weightof food each feeding thy and, depending on the diet, were given either pellets or algalsheets. Pellets were offered dry, but kelp sheets were pre-soaked prior to feeding. Rationlevels took one week to establish and were determined on the basis of the maximumamount of food ingested without excessive waste accumulating at the bottom of the tanks.This feeding regime was maintained for a period of nine months and coincided with the13urchinstannual reproductive cycle, being initiated in July at the end of the annual spawningcycle, and ending in late March when the gonads had once again reached maturation.All diets appeared equally attractive to the sea urchins. When food was placed inthe tanks I observed directional responses toward both the pellets and kelp, and an increasein tube-feet and lantern teeth activity. Importantly, I noted that the urchins could sense,catch, manipulate, and ingest the artificial diet pellets.General MethodsTo assess the effect of adult diet on gonadal growth, measurements of test diameter,whole animal wet, drained weight (without coelomic fluid), and gonad weight wererecorded for 10 animals of mixed sexes from each dietary treatment at six-week intervals.Sampling was done from early July, when animals were in the resting stage of theirreproductive cycle, to late March, when animals had reached spawning condition. At eachsampling period, measurements and dissections were done within 48 hours.Test DiametersTest diameters were measured by placing the tips of needle-nose calipers betweenthe spines on the test from one interambulacral region to the opposite ambulacral region,across the widest portion of the test.Wet WeightsWhole animal wet weights were recorded after air-drying for 10-15 minutes,allowing excess seawater to drain from the surface of the test.Gonad IndexNormally, a gonad index is calculated as the ratio of gonad weight to whole bodyweight (Himmelmann 1978, Larson etal. 1980, Keats et al. 1984, Andrew 1986).14However, because coelomic fluid levels vary widely (relative to the size of the gonad and tothe amount of food in the gut), and to reduce errors caused such by variations, I comparedgonad weights to drained test weights instead of to whole wet weights (after Vadas 1977).Drained test weights were calculated by piercing the peristomial membrane and allowing thecoelomic fluid to drain out. The animals were then reweighed to obtain the drained weightmeasurement. The gonads were removed by cutting around the mouth and up the test toone-third its height, removing the oral section, cutting the attached mesenteries, and slidingthe gonad sections out and onto absorbent towelling. The gonads were blotted and air-dried for 3 minutes, cleaned of debris, and weighed. Gonad indices were calculated as theratio of gonad weight to drained test weight x 100%.Sex DeterminationSexes were differentiated on the basis of colour of gonadal tissue in December,February, and March. At these times, white sperm or orange-yellow eggs were readilyobserved.Lipid and Moisture Contents ofthe GonadsAnalyses of lipid and moisture were conducted on female gonadal tissue only, sincepast research showed that lipid values were higher in female gonadal tissue than in male(Giese et al. 1959, Lawrence and Lane 1982). Gonad samples taken from the Novemberand March sampling periods were analysed for lipid content following the procedure ofBligh and Dyer (1959): wet gonadal tissue was homogenized with a mixture ofchloroform, methanol, and distilled water. This separated into two layers, with thechloroform layer containing all the lipids and the methanolic layer containing all the non-lipids. A portion of the lipid extract was evaporated to dryness and the weight of the lipidextract determined. Lipid value was calculated as dry-weight. Percent moisture wasdetermined by drying a 2 gram sample of gonadal tissue for 16 hours at 95°C.15Two-way ANOVA’s combined with Newman-Keuls (NK) multiple comparisontests were used to compare gonad indices, and lipid and moisture levels (after arc-sintransformation of percentage values) among the eight dietary treatments as well as tocompare gonadal growth of males and females.RESULTSEffect ofDiet on Survival ofAdultsUrchins maintained on all diets with the exception of the LO-PRO diet showed>95% survival, with losses occurring only sporadically from August to November. Earlymortality was notable for the LO-PRO group, with 23 of 130 individuals dying betweenAugust 8-October 8. The highest losses were observed three weeks after feeding started,and no losses occurred after November in any diet group.Effect ofDiet on Lipid and Moisture Contents ofthe GonadsLipid levels, although not showing any effect of diet (F7,49 = 0.29; p = 0.95) didvary significantly with time (F1,49 = 6.9; p 0.01), averaging 21.4 ± 1.7% in Novemberand 22.7 ± 1.3% in March for all diets (see Table 2). Although significant, such smalldifferences are unlikely to have biological relevance. Moisture content was found to varysignificantly over time (F1,49 = 1012.6; p < 0.0001) but not among dietary treatments(F7,49 = 1.9; p 0.09). Moisture increased from an average of 69.7 ± 1.5% in Novemberto 82.4 ± 1.5% in March (see Table 2). No interactive effects of treatment and time onlipid (F-j, = 0.98, p = 0.45) or moisture content (F7,49 = 1.5, p = 0.18) were detected.The differences in gonadal moisture levels between November and March were largeenough to indicate a substantial component of water being added as “growth” during the16Table 2: Lipid (% dry weight) and moisture concentrations of female gonadal tissue ofStrongylocentrotus droebachiensis conditioned on eight dietary treatments for a nine-month period and sampled after five months (November) and after nine months (March).Values are mean ±S.D, with sample size in brackets.% Lipid % MoistureNov/1991 March1992 Nov/1991 March1992% d.w. (N) % d.w. (N) %w.w. (N) %w.w. (N)LO-PRO 21.2±2.9 (4) 22.6±0.4 (2) 69.9±0.7 (4) 83.3±0.6 (2)LO-PRO+M 21.6±1.8 (4) 21.3±1.0 (3) 70.3±2.9 (4) 84.2±0.6 (3)LO-PRO+A 21.5±0.8 (3) 22.9±1.4 (4) 68.8±0.3 (3) 83.5±1.5 (4)HI-PRO 21.5±0.5 (4) 23 .7±1 .0 (3) 69.8±2.1 (4) 80.2±1.1 (3)HI-PRO+C 21.2±1.1 (3) 23.0±0.2 (2) 69.1±1.4 (4) 82.1±0.3 (2)HI-PRO+B 19.8±2.1 (3) 23 .4±0.3 (3) 70.1±0.3 (3) 8 1.6±0.8 (3)HI-PRO+C/B 22.4±2.4 (3) 22.4±1.5 (3) 69.0±0.2 (3) 8 1.3±1.0 (3)KELP 22.3±1.2 (3) 22.2±2.6 (3) 70.6±2.0 (3) 83.2±0.2 (3)17later stages of the reproductive cycle, a phenomenum known to occur during gonadalmaturation in sea urchins.Effect ofDiet on Test GrowthThere were no significant differences in mean test diameters over time (F1,59 =0.21, p = 0.64) or among dietary treatments (F7,159= 1.2, p 0.29) during theexperimental period. These findings statisfied Gonor ‘s (1972) concerns that a gonad indexis useful only when comparing animals of similar size from different treatments becauseanimals with different body size do not have the same gonad:body-size ratio. The overallsimilarity in body size in the present study removed this error source and also ensured thatspecimens were at an age where they would have all been reproductively active and insynchrony for at least two years (Boolootian 1966, Gonor 1973).Effect ofDiet on Gonadal GrowthThe eight dietary treatments produced highly significant differences in gonadgrowth (F7, 558 = 14.9, p<c000l) (Fig.1, Table 3), with urchins maintained on allformulations of the HI-PRO diets and the KELP diet exhibiting significantly higher fmalgonadal indices than urchins maintained on all LO-PRO formulations (NK tests showedthat values of all HI-PRO formulations separated together and independently from values ofall LO-PRO diets, p<0.05). Time effects were also significant (F6,558 = 221.7,p<<.0001), with the data segregating into five statistically homogeneous subgroupsrepresented by July-August, October, November-December, February, and March(NK<0.05). Initial gonad index values in July at the start of the experiment ranged from15.9-17.8% and final gonad indices in March ranged from 36-39%. The interaction oftreatment and time was significant (F42,558 = 1.7, p 0.004). The gonad sizes reached bythe LO-PRO+M group at 18 weeks, and the LO-PRO and LO-PRO+A at 24 weeks, werenot significantly different from those reached at 12 weeks for all the HI-PRO and KELP1840.76356 —1 • LO-PRO30/ 2 :LO-PRO+M‘ / 6’ 3Z 25/.;:Lfli-oF 7 I 5 HI-PRO+C= :Z ..h’••• —__6 HI-PRO+B20 dF ci HI-PRO+C/B// 8 KELP15I I I I I I IJuly Aug Oct Nov Dec Feb MarchTIME OF YERHFigure 1: Gonad indices (%) of Strongylocentrotus droebachiensis conditioned on eightdietary treatments from July 1991 to March 1992 and sampled at six-week intervals.Numerical designations of diets are as follows: (1) LO-PRO, (2) LO-PRO+M, (3) LOPRO+A, (4) HI-PRO, (5) HI-PRO+C, (6) HI-PRO+B, (7) HI-PRO÷CIB, and (8) KELP.Gonad index rankings are noted for diets in December, at the time of best market qualityand in March, at time of maximum gonad index prior to spawning.19Table3:Gonadindices (%)for adultStrongyloceiurotusdroebachiensisconditionedoneight dietarytreatmentsfor anine-monthperiod.TenanimalsperdietweresampledeverysixweeksfromJuly1991toMarch1992.Values aremeans±S.D.DietdesignationssameasinTable1.LO-PROLO-PRO+MLO-PRO+AHI-PROHI-PRO+CHI-PRO+BHI-PRO÷C/BKELPJuly15.9±3.716.3±3.516.9±3.316.8±3.616.4±2.616.1±3.116.7±4.817.8±4.4August17.3±4.516.9±3.815.9±2.117.5±3.218.2±3.618.4±4.517.8±3.217.1±2.8October21.1±3.116.9±3.716.4±3.221.7±2.721.9±2.621.5±2.618.8±3.920.5±2.2November21.5÷2.921.9±4.119.0±1.926.9±1.925.9±3.826.4±2.724.2±4.125.3±1.8December20.5÷5.423.3±5.320.7÷4.627.5÷4.227.9±2.830.1±4.424.5±2.925.8±2.3February29.9±5.526.4±7.027.8±5.831.6±7.732.1±4.133.3±6.435.9±4.634.4±6.0March32.2÷4.631.4±6.032.9±5.838.8±6.235.9±6.138.9±5.539.4±6.438.6±6.2C.”diet groups (NK>O.05). However, the results of the interaction were most definitive inDecember when the gonads of HI-PRO+B were 50% larger than the gonads produced bythe LO-PRO dietary treatments. This has substantial aquaculture significance as this is thetime of the best market quality.A second analysis of the data from the months of December, February and Marchwas conducted to determine if there were any differences in gonadal growth between malesand females. Only at these sampling periods were sexes easily differentiated by simpleobservation of the gonads. There was a strong sex effect (F1,241 = 119.1, p = 0.0001),but no interactive effects of diet and sex (F7,241 = 0.9, p 0.53), or diet, sex, and time(F14,241 = 0.7, p = 0.72). The final mean gonadal index of males was 29% and offemales, 34%, based on sample sizes of 141 males and 101 females.21DISCUSSIONAll dietary treatments produced gonad indices higher than previously recorded inStrongylocentrotus droebachiensis. Even the lowest value of 31% attained by individualson the LO-PRO+M diet was 4 percentile units higher than the peak gonadal index of 27%observed by Vadas (1977) for the same species fed the kelp Nereocystis luetkeana (knownto produce the highest rates of gonadal growth of several algae tested). It is likely that thelarge sizes attained by all groups resulted from the quality of the diets, consistent feedingregime, and low-stress conditions of laboratory culture. Since all artificial and kelp dietsinduced similar positive behavioural and feeding responses, and appeared to be eaten at thesame rates, I assume that the differential gonad production recorded here was a result ofdifferences in dietary treatments and not of differences in ingestion rates.The significant difference in gonadal growth between low and high-protein dietarytreatments suggests that differential gonadal growth resulted from protein level. Althoughno other studies have been done to make qualitative comparisons, Fuji (1962) determinedthat different algal species were required in different proportions to sustain the body weightof Strongylocentrotus intennedius. The algal species with the highest protein level wasrequired in the least amount. The present fmdings support the ideas of Lilly (1975), Millerand Mann (1973), and Lowe and Lawrence (1976) who suggest that a positive relationshipexists between protein level in the food and feeding behaviour in sea urchins. Theseauthors conclude that the growth and reproductive output of sea urchins may be moredependent on the amount of protein than upon the energy acquired in feeding. In thisregard, Lilly (1975) found that Tripneustes ventricosus improved its protein:calorie intakeratio by selectively absorbing protein components from all of its foods except Ulva, andthat in so doing compensated for low levels of protein occurring in its natural diet of algae.22The high rates of gonaclal growth recorded in the KELP group cannot be as clearlycorrelated with protein levels in the food. Protein levels in Nereocystis luetkeana areknown to average 12% dry weight and range from 11-14% on a seasonal basis (Whyte etaL 1981). The protein level in this diet, therefore, was closer to the LO-PRO protein levels(which yielded a 32.2% gonad index as compared with 38.6% for KELP). Other nutrientcomponents of the algae most likely contributed to its overall quality. For example, bothmannitol and algin are key storage carbohydrates of Nereocystis , and can constitute 3-8%and 19-23% dry weight, respectively, of the kelp. It is known that these two componentsare readily absorbed by sea urchins (Boolootian and Lasker 1964). It is also known thatalgin would have been at least partly depolymerized in the drying process of the algae, thusmaking its components even more readily digestible (Whyte, I.: pers.comm.). Larson etal. (1980), Lawrence (1975), and Vadas (1977) have shown that algae absorbed at thehighest efficiency support the highest growth in StrongylocentrotiLs droebachiensis andrelated sea urchins. Nevertheless, however suitable the KELP diet may have been for theproduction of large gonads, it proved inadequate as a diet for broodstock. It will be seenthat the larvae cultured from this dietary treatment were severely abnormal and had highrates of mortality (see Section 3).The supplements of mannitol, algin, B-carotene and cholesterol, although notproducing significant differences in gonad indices, may still be important growth factors.It should be noted that the largest gonadal sizes were attained by HI-PRO+C/B (39.4%)indicating that the combination of the two supplements of B-carotene and cholesterol mayhave been synergistically beneficial to gonad growth. Mannitol and algin may also havebeen advantageous since mortality rates in the LO-PRO+M and LO-PRO+A diet groupswere negligible compared to high losses recorded for the LO-PRO diet group, which hadno additives.23Gonadal lipid levels measured in this study (2 1-23% d.w.) for Strongylocentrotusdroebachiensis were comparable to those (21-27% d.w.) recorded by Giese et al. (1959),Greenfield et al. (1958), and Giese (1966) in S. purpuratus. If lipid levels in S.droebachiensis peak coincidentally with peak gonad indices as shown by Greenfield et al.(1958) for S. purpuratus, then the unseasonally high lipid values recorded in Novemberfor S. droebachiensis may have been responsible for the larger than normal gonad sizesreached in March, as more gametes would likely have reached maturity, given the highlevel of energy-providing lipids in storage.Gonadal growth in Strongylocentrotus droebachiensis followed the expected cycle,with gonads increasing in size throughout the fall and winter and maturing in the spring(Boolootian 1966, Himmelman 1978). Initial gonadal enlargement in sea urchins is knownto result from increase in size and number of nutritive phagocytes involved in the storage ofnutrients for gametogenesis. Further enlargement results from an increase in the numberand size of gametes with a related decrease in volume of the accessory cells (Gonor 1973,Walker 1982, Pearse et al. 1986) and increase in gonadal moisture content (Greenfield etal. 1958, Miller and Mann 1973, Gonor 1973). Based on these earlier works doneprimarily on S. purpuratus and on the results shown here, it is likely thatS. droebachiensis follows the same pattern. The absence of large change in lipid levelsbetween November and March indicates that the initial phase of nutrient build-up occursbefore November, after which the lipids are likely transferred to the developing gametes.Increases in gonadal size after November are due both to an increase in the number ofmature gametes and to an increase in moisture (from 70% in November to 82% in March,see Figure 1).The apparent slowdown of growth between November and December (see Figure1) for all dietary treatments (except HI-PRO+B) may have marked a switch between the24processes of nutrient storage, and gametogenesis and associated water uptake. Keats et aL(1984) also noted a slight decline in gonadal growth between October and November for awild, but well-fed population of S. droebachiensis.The lack of significant differences in test diameters observed during the course ofthe study were expected since the experimental period chosen corresponded to the urchin’sgonadal maturation cycle, and somatic growth in echinoderms is known to be seasonal. Ithas been demonstrated that echinoids allocate little to somatic growth during periods ofgonadal enlargement (Boolootian 1966, Lilly 1975, Lawrence and Lane 1982, Thompson1982, Pearse et at. 1986). Also, rates of somatic growth decrease with age and, althoughjuveniles are known to grow relatively more quickly, urchins of the mature sizes used inthis study would show much slower rates of growth (Boolootian 1966).Ration levels established throughout this study cannot be readily compared toprevious work because of the artificial nature of the diets and their pellet form. However,Himmieman and Nedelec (1990) recorded feeding rates of Strongylocentrotusdroebachiensis ranging from lows of 0.06-0.21 dry gd1 to highs of 1.09-1.22 dry g•d1,depending on type of algae offered, values generally somewhat higher than the 0.4dry gd1 ration levels used in the present study. The larger gonads produced in the presentstudy (33-39%) as compared with those recorded by Himmelman (1978) (17%) weredoubtless a result of the nutritionally better pelleted food, the difference being made moreemphatic by the fact that the ration was generally smaller in the present study.All artificial diets used here proved successful in the maintenance of adult seaurchins. This is in contrast to results of experiments by Nagai and Kaneko (1975) whoreared Strongylcentrotus puicherrimus for five months on an artificial diet composed offish and soybean meal, yellow corn, yeast, soybean oil, agar-agar, and vitamins (with a25crude composition of 32.4% protein, 7.3% fat, 10.6% fiber, and 10% ash). Theexperiment was unsuccessful, with animals showing periodic severe declines in feedingrates and all dying by five months. The same authors tested standard farm-fish pellets andfound that they were rejected as food by the same urchin species.In applying the findings of this study to an aquaculture program, the best diet forroe production is HI-PRO+B, for even though HI-PRO+C/B grew slightly larger gonadsby the study’s end in March, at the time of the most valued market product in DecemberHI-PRO+B provided the largest gonads (see Figure 1). The high moisture levels of 81-84% in March (for all diet groups) render the gonads a less valuable product and, based onthe pattern of gonadal growth observed in this study, the best roe for the sushi marketwould be from a November-January harvest At this time the gonads reach their maximumsize both prior to the maturation of the gametes and to the uptake of water before spawning.Remarkably, the gonad sizes of HI-PRO+B were 50% larger than those attained by LOPRO for the December sampling period. It will later be seen (Section 3), that larvae rearedfrom HI-PRO+B diet were more successful than larvae reared from the other diets,including HI-PRO+CIB, further attesting to the superiority of the former as a broodstockconditioning diet for these urchins.26SECTION 3: EFFECTS OF BROODSTOCK CONDITIONING ON EGGS ANDLARVAEINTRODUCTIONIn echinoderms, egg size and chemical content, as well as larval size anddevelopment, are known to vary not only among individuals from different populations butamong individuals from the same population (Turner and Rutherford 1976, Turner andLawrence 1979, McEdward 1984, Emlet eta!. 1987, McEdward and Carson 1987,George 1990, George eta!. 1990). In sea urchins, because both fertilization anddevelopment are external, parental investment in the gametes is complete at spawning.Adult diet can therefore potentially affect both egg quality and subsequent stages of larvaldevelopment and, indeed, differences in characteristics of the echinoderm eggs and larvaehave been attributed to variations in adult food supply (Thompson 1982, George 1990,George eta!. 1990).Evaluation of egg quality in echinoderms is frequently based on the size of the eggand its biochemical composition. Echinoderm eggs may differ in size because ofdifferences in chemical content (Turner and Lawrence 1979, Emlet et a!. 1987), or be thesame size despite differences in chemical content (George eta!. 1990). Turner & Lawrence(1979) found that egg volumes were significantly different among individuals of the samespecies and between years for 11 species of echinoderms. McEdward and Carson (1987)also showed that mean egg volumes of the starfish Solaster stimpsoni differedsignificantly among females from different locations. While neither of these studiesexamined the reasons for the observed differences, George (1990) attributed variations inegg sizes of Arbacia lixula to differences in the diets of the two adult populations studied.27Just as adult diet has been shown to affect egg size and chemical content inechinoderms, it is likely that characteristics of the eggs carry over to the larvae, affectingtheir larval morphometry and development. However, studies to date present conflictinginformation. For example, McEdward (1984) compared developmental times for sevenspecies of co-occurring echinoids with a range of egg sizes and determined thatdevelopment times varied relative to egg size. In contrast, Amy (1983) found noconsistent relationship between egg size and rate of early embryonic development in studiesof five species of tropical sea urchins, and George et al. (1990b) showed that eggs ofsimilar sizes produced larvae with different developmental times in two populations ofArbacia lixula. There appears, then, to be ample reason to re-investigate the effect of adultdiet on larval vitality in sea urchins, and this is the intent of the present section.This Section, then, follows the broodstock conditioning experiments (Section 2),and assesses not only the effects of adult diet on egg quality and larval morphometry anddevelopment, but identifies a broodstock diet which will produce healthy larvae suitable fora culture program.Of the eight dietary treatments used to condition the broodstock in Section 2, onlyfive were selected for the larval rearing experiments in this Section. Diet groups wereselected on the basis of the results on gonadal growth (Section 2), as well as egg-energycontent and characteristics of the adult diet used. The dietary treatments of LO-PRO, HIPRO, HI-PRO+B, HI-PRO+C/B , and KELP were chosen because: 1) the previousgonadal growth experiments showed significant differences in gonad index values betweenthe low-protein, and high-protein and algal diet groups, 2) the egg-energy levels of LOPRO and HI-PRO diets were significantly different from those of HI-PRO+B, HIPRO+C/B, and KELP, and 3) the low- and high-protein diet formulations and KELP diet28were all represented in this selection, as well as diets containing additives considered mostlikely to produce differences in the larvae (from Section 2).Prior to the MATERIALS AN]) METHODS section, I have included a descriptionof the development and morphometry of Strongylocentrotus droebachiensis larvae.Larval Development of Strongylocentrotus droebachiensisThe following overview of larval development for strongylocentrotid sea urchinsincludes information from Kume and Dan (1968), Strathmann (1971), Gardiner (1972),Barnes (1987), Emlet etal. (1987), and Kozloff (1990).Fertilization and development are external in Strongylocentrotus droebachiensis.Gametes are released “broadcast” fashion and are fertilized in the surrounding seawaterwhere they undergo development as planktotrophic larvae. Following hatching, the larvafeeds and drifts in the plankton until ready for metamorphosis into the juvenile. Thedevelopmental process proceeds in several clearly defined stages (Fig.2): (1) unfertilizedegg (Fig.2a), (2) fertilized egg (Fig.2b), (3) blastula (Fig.2c), (4) gastrula, (5) prism(Fig.2d), (6) pluteus, (7) 2-arm pluteus (Fig.2e), (8) 4-arm pluteus (Fig.2f), (9) 6-arm pluteus, (10) 8-arm pluteus (Fig.2g), (11) competent larvae (Fig.2h), and (12)metamorphosis. Descriptions of each stage are as follows:1) Eggs are homolecithal with relatively small amounts of yolk distributed evenlythroughout the cytoplasm. Embedded in the outer jelly coat of the eggs are specialized cellscontaining carotenoid pigments (Gardiner 1972, Burke and Bouland 1989, Tsushima andMatsuno 1990), which may function in the maturation process of the egg. During the finalevents of oogenesis, pigment cells migrate in the jelly coat away from the surface of theoocyte. The migration of the pigments cells correlates positively with the attainment offmal egg diameter.292) Fertilization in sea urchins takes place after completion of meiosis. Its successdepends on maturity of both nucleus and cytoplasm. Fertilization is signified by the liftingof the cell membrane from the surface of the egg, a process requiring approximately oneminute for complete separation following sperm penetration.3) After ten cleavages a ciliated blastula forms, easily identifiable by its round shapeand outline of cilia.4) Gastrulation is marked by a flattening of the embryo, invagination of theendodermal plate, and formation of the primitive gut.5) The prism stage is characterized by a pyramidal-shaped motile larvae. At thisstage important changes occur in body organization, such as formation of thespicularskeleton and stomodaeum, and differentiation of the digestive system.6-10) The pluteal stage is recognized by the enormous elongation of the skeletalrods and the stretching of the aboral end. Feeding begins at this stage. Echinoplutei feedon suspended particles such as small flagellates, diatoms, and detritus. Throughout thevarious pluteal stages there is a trend of increasing overall larval size and number of arms.Echinoplutei develop the anterolateral, postoral, posterodorsal, and preoral arms in thisorder to attain the 2-arm, 4-arm, 6-arm and 8-arm stages, respectively (Fig.2f-g). Eachlarval arm is supported by a skeletal rod and is outlined by a band of cilia (Fig.2g). Theciliated band is continuous around the arms and, along with the bands located around themouth, function in both feeding and locomotion. By the 8-arm stage, sections of theciliated band between the arms become separated from the rest of the band andform theheavily ciliated epaulettes which are specialized for locomotion (Fig.2g).30(d)(b)(h)Figure 2 (a-h): Eggs and developing larvae of a sea urchin.(a) unfertilized egg, (b) fertilized egg showing fertilization membrane, (c) ciliated blastula;(d) prism larva, (e) 2-arm pluteus, (f) 4-arm pluteus showing the anterolateral, postoralarms and mouth, (g) 8-arm pluteus showing the preoral and posterodorsal arms, ciliatedband outlining all arms, heavily ciliated epaulettes and gut, (h) pluteus preparing formetamorphosis, seen from the left side and showing adult rudiment. (Drawings modifiedfrom Kume & Dan 1968, Strathmann 1971, and Gardiner 1972.)(a)fertilizationmembrane cilia(c)anterolateral arm(e)armmouth(f)armposterodorsalarmgut(g)adult rudiment31(11) Once the 8-arm pluteus is fully formed, preparation for settlement andmetamorphosis begins. The “echinus rudiment” forms in a coelomic pouch on the left sideof the larva and is the center of formation of the adult mouth and its surrounding structures(Fig. 2h). As metamorphic competency approaches, an increase in the size of therudiment, a loss of buoyancy from the increased weight of this same structure, ashortening of the larval arms, and a formation of 5 primary tube-feet, spines, and a part ofthe adult test, are apparent.(12) Metamorphosis is preceded by settling behaviour. During settling, the larvatests the substratum by everting the adult rudiment and walking about on the primary podia.Sea urchin podia are known to possess sensory structures (Burke 1980). A number ofenvironmental cues must be identified by the larva and a prolonged period of contact withthe selected substratum is required before metamorphosis is initiated (Cameron andHinegardner 1974, Burke 1983). While metamorphosis can be delayed until a suitablesubstrate is found, the ability to metamorphose declines over time (Emlet et al. 1987).Once metamorphosis proceeds, it is irreversible.During metamorphosis the rudiment unfolds and the larval body collapses into whatwill become the aboral surface of the adult. The epidermis, ciiary bands, epaulettes,mouth, and esophagus of the larva undergo histolysis. Histolysis occurs as little as 2-5minutes after initiation of metamorphosis. Metamorphosis is complete when the juvenilehas attained the typical urchin shape and five primary podia are visible on its outer surface(Cameron and Hinegardner 1978, Burke 1982).Larval development rates are modified by temperature, food quality, and foodavailability (Emlet et al. 1987, Gonzalez et al. 1987, Hart and Scheibling l988a) and, inresponse to differences in these factors, large variations in the timing of developmentoccur. The timing from postfertilization to metamorphosis for Strongylocentrotus32droebachiensis is known to range from 30-150 days (Ebert 1983, Himmelman 1975,Pearce and Scheibling 1991). While echinoplutei not unexpectedly have higher rates ofgrowth, development, and metabolism at higher temperatures, Hinegardner (1967) foundthat Strongylocentrotus droebachiensis larvae show abnormal development above 10°C.Larval MorphometiyEchinoplutei undergo remarkable changes in form throughout development (Kumeand Dan 1968, Cameron and Hinegardner 1978, Strathmann 1971). McEdward (1984)suggested that because of the relationship between larval size, shape, and feedingcapability, changes in larval morphometry will have important functional consequences.The most significant of these is the influence of ciliated-band lengths and the ciliated-bandto body length ratio on the feeding effectiveness of the larvae (Strathmann 1971,McEdward 1986). Increase of ciliated bands relative to body length will tend to increasefeeding efficiency in sea urchin larvae.MATERIALS AND METHODSSpawning and FertilizationIn late-March 1992 when I observed evidence of a phytoplankton bloom known tostimulate spawning in wild populations of green sea urchins (Himmelman 1975), thedietary conditioning experiments were terminated, and spawning and fertilizationexperiments begun. Viability of eggs for larval culture were determined by the followingprocedure. Sample urchins were injected with 0. 1M acetyicholine solution through the33peristomial membrane into the perivisceral cavity. All injected urchins started to shed theirgametes within 2 minutes after injection. Once sex was established, selected individualswere washed thoroughly with 0.45iim-filtered seawater to remove surface bacteria,protozoans, and other debris. Spawning females were inverted over 250 ml plastic beakerscontaining approximately 150 ml of 0.45 seawater, and the newly-releasedgametes were allowed to drip into the water and settle. Excess seawater was poured offand the eggs were washed 3-4 times with fresh 0.45 .im seawater and filtered through 100.1m nitex mesh. The resulting egg suspensions were used in the larval developmentexperiments. Two males were selected and sperm collected “dry” via pipette. These spermsamples were combined and then diluted by adding 1 drop to 5 ml of filtered seawater.The resulting sperm suspensions were then used in the subsequent larval developmentexperiments.To assure that eggs from all diets were at a comparable level of maturation,fertilization success was determined. Eggs of six sea urchins from each diet group werefertilized with sperm mixture. Two minutes after mixing sperm and eggs, the gameteswere washed with freshly-filtered seawater to remove excess sperm. Then, one drop ofmixed gametes was placed on a depression slide and the number of fertilized eggs(indicated by formation of fertilization membrane on the egg surface) was counted againstthe number of unfertilized eggs. This process was repeated three times for each urchin.Egg SizeEgg size was expressed as volume, estimated from measurements of the long andshort axes of each of twenty eggs from each of six females per diet. The volume (V) ofthe prolate spheroid eggs was calculated using the following formula: V=(4/3)irA2B,where A and B are the radii of the short and long axes of the egg, respectively.34Egg Organic-Carbon ContentEnergy content of the eggs was measured using a micro-modification of aprocedure described by Parsons eta!. (1984). The procedure determines the amount oforganic carbon in a sample and can be interpreted as a measure of the energy stored in thatsample. Organic content was measured by dichromate oxidation against a glucosestandard. Three replicates of 150 eggs from each of six females from each diet group wereused. Eggs were briefly rinsed in distilled water to remove salts, incubated in concentrated(70%) phosphoric acid (lml, 15 mm, 100-110°C) to remove residual chloride, thenoxidized with potassium dichromate (0.484%) in concentrated sulfuric acid. Reduction indichromate concentration indicated the amount of organic carbon oxidized. The organiccontent was determined as weight of glucose (ug) yielding equivalent reduction indichromate concentration and this was converted to equivalent energy in Joules using thefollowing equation lug C 3.9 x 102J. The energy content of an individual egg wasobtained by dividing the energy content of the sample by the number of eggs. This methodis appropriate for determining energy contents based on the known similarity of resultsobtained by wet oxidation and bomb calorimetry (Carefoot 1985).Larval CulturingTo determine the influence of egg size and quality on larval development threebatches of larvae from the selected diet groups of (LO-PRO, HI-PRO, HI-PRO+B , HIPRO+C/B, and KELP) were cultured. Egg suspensions from each diet group werepoured into 4L nalgene plastic trays containing 2 L of 0.45 iim seawater and combinedwith 2 ml of diluted sperm suspension. The gametes were left to settle as a thin layer onthe bottom of the tray. Trays were then placed in a 15 cm-deep seawater table and leftundisturbed for 24 hours, at which time the ciliated blastula stage was reached. A constantseawater flow into the water table maintained the temperature of the cultures at 9±1°C.Once the ciliated blastula stage was reached, 2000 embryos from each culture were35transferred to 1L plastic beakers containing 900 ml of filtered seawater (approximately 2larvae• ml- 1)• Three replicate cultures were prepared in this way for each diet group.Larvae were maintained on a 1OL/14D photoperiod provided by daylight fluorescent bulbs.Larvae were maintained in suspension by continously stirring with electrically-operatedpaddles as described by Leahy (1986) and Strathmann (1971). When the larvae reachedthe feeding pluteal stage, phytoplankton (the diatom Chaetoceros gracilis and the flagellateNannochioropsis sp. cultured at 12°C; l2hrs L: l2hrs D in 0.45um filtered and autoclavedseawater provided with nitrate, phosphate, and silicate nutrient base of imi: 1L) wasprovided ad libitum. Prior to each feeding the algal culture medium was removed bycentrifuging the samples at 5000 rpm for 10 minutes. Phytoplankters were resuspendedin filtered seawater before being offered to the larvae. The seawater in the beakers waspartially (3/4 original volume) changed every 2 days via pipette through a meshscreen. Excess uningested algae were removed from the beaker bottoms directly viapipette.Larval DevelopmentThe time taken for development was calculated by observing 100 larvae per culturebeaker and recording their stage of development. Larvae were randomly sampled bypipette, viewed microscopically, and the developmental stage of each recorded as follows:1) ciliated blastula, 2) gastrula, 3) prism, 4) pluteus, 5) 2-armed pluteus, 6) 4-armedpluteus, 7) 6-armed pluteus, 8) 8-armed pluteus, 9) competent larvae, and 10)metamorphosis (see Fig.2). Observations were done daily up to the 6-arm stage and every2 days thereafter. It was assumed that once 75% of the larvae sampled had reached asimilar stage of development the stage had been attained by that diet group.Times to metamorphosis and rates of metamorphic success were determined byselecting 30 larvae from each culture and inducing them to metamorphose on petri dishes36conditioned in seawater for two weeks to promote the growth of bacteria. Although manyexperimentors have successfully induced settlement and metamorphosis in marineinvertebrate larvae with neurotransmitters (Burke 1983), Cameron and Hinegardner(1974) determined that metamorphosis in S. droebachiensis could be simply achieved byproviding competent larvae with a hard substratum coated in a thin layer of bacteria. Whenlarvae were fully formed and showed evidence of the adult rudiment, induction experimentswere conducted. Tests of competency were repeated twice in a period of two weeks and,each time, larvae were examined at 24,48, 72, and 96 hours with the number of activelymoving, settled, and metamorphosed larvae being noted. Larvae were consideredcompetent when culture showed 60% metamorphic success (after Pearce and Scheibling1991).Larval MorphometiyAll morphometric measurements followed the procedures described by McEdward(1986) and included such dimensions as (Fig.3): 1) overall larval length (postoral arm tipto the posterior end of the larva)(Fig.3A), 2) total arm length (summed over all arms)(Fig.3 B÷C+D+E), 3) body length (anterior tip of the oral field to the posterior tip of thelarval body) (Fig.3F), 4) rudiment diameters (Fig.3G), and 5) ciliated band (Fig.3H).Measurements were made on the 2-, 4-, 6-, and 8-armed pluteus stages. At eachmeasurement time 20 larvae from each group were randomly selected, placed on amicroscope slide, killed with 70% ethanol diluted with culture medium, and measuredusing a compound microscope equipped with an ocular micrometer.37Figure 3 (A-H): Morphometric measurements of echinoplutei of Strongylocentrotusdroebachiensis.A: overall larval length (postoral arm tip to the posterior end of the larva)B, C, D, E: total arm length (summed over all arms x 2)F: body length (anterior tip of the oral hood to the posterior tip of the larval body)G: rudiment diameterH: ciliated band (indicated by double lines) (measured as inner and outer lengths of eacharm plus distance between arms of each pair)A38One-way ANOVA’s followed by Newman-Keuls (NK) multiple comparison testswere used to compare egg size, egg energy content, and fertilization rates (after arc-sintransformation of percentage values) as well as to compare all larval morphometry thta ateach developmental stage for the selected dietary treatments.RESULTSEffect ofAdult Diet on Egg SizeDiet had significant effects on egg size (Table 4, ANOVA:F7,952 8.1, p<0.0001). NK analysis showed that eggs of the HI-PRO diet were the smallest, whilethose of HI-PRO+C were the largest (p <0.05). Egg sizes of the other six diet groupswere intermediate and formed a statistically homogenous subgroup (NK, p>O.O5).Effect ofAdult Diet on Egg Energy ContentDiet had significant effects on egg energy content as indicated by organic carbonlevels (Table 4, ANOVA:F7,136 = 17.0, p <0.0001). Eggs of individuals reared on eitherthe LO-PRO or HI-PRO diets contained lowest levels of organic carbon, while highestamounts were recorded in eggs produced from adults fed HI-PRO+C/B. Organic carbonlevels of the other five groups formed an intermediate statistically homogenous subset(NK, p >0.05). Egg size and egg energy content were not correlated (Pearson ProductMoment correlation analysis: r = 0.17 and p = 0.25).39Table 4: Size and organic carbon content of eggs of adult Strongylocentrotusdroebachiensis conditioned on eight dietary treatments for nine months. Values fororganic carbon content are mean ± S.D. of six females per diet (with three subsamples of150 eggs each per individual). Values for egg size are mean ± S.D.of 20 eggs from eachof six females per diet. Results of ANOVA and NK multiple comparison tests are alsoincluded. Values which did not differ at the 0.05 level are joined by a solid line. (Valuesfor the NK test are shown in ascending rank order from left to right, in accordance withdietary numerical designations.) Diet symbols same as in Section 2.Diet Egg size Organic carbon(mm3 X 10-3) (ug C I egg)(N=120) (N=18)(1) LO-PRO 2.26 ± 0.73 0.31 ± .027(2) LO-PRO+M 2.26 ± 0.22 0.38 ± .038(3) LO-PRO÷A 2.20 ± 0.33 0.37 ± .029(4) HI-PRO 2.07 ± 0.27 0.31 ± .046(5) HI-PRO+C 2.39 ± 0.33 0.39 ± .021(6) I-ll-PRO+B 2.29 ± 0.23 0.39 ±. 036(7) HI-PRO+C/B 2.28 ± 0.26 0.41 ± .035(8) KELP 2.18 ± 0.28 0.39 ± .021ANOVA (F 7 952 = , p =) 8.14 <0.0001 (F 7,136 = , p =) 17.00 <0.0001NKtest 48321765 4j32568740Table 5: Fertilization success of eggs produced by adult Strongyiocentrotu.s droebachiensisconditioned on diets selected for larval culturing experiments. Values are mean ± S.D. ofsix females per diet (with three subsamples of 10 eggs each per individual). Results ofANOVA and NK multiple comparison tests are also included. Values which did not differat the 0.05 level are joined by a solid line. (Values for the NK test are shown in ascendingrank order from left to right, in accordance with dietary numerical designations.) Dietsymbols same as in Section 2.Diet Fertilization Success(%)(N=18)(1) LO-PRO 91 ±4(4) HI-PRO 94±2(6) HI-PRO+B 90±2(7) HI-PRO+C/B 96±2(8)KELP 90±3ANOVA (F4,85 =, p= ) 5.95, <0.001NKtest 8 6 1 4 741Effrct ofAdult Diet on Fertilization SuccessFertilization success for the five dietary treatments selected for culture experiments(LO-PRO, HI-PRO, HI-PRO+B, HI-PRO+CIB, KELP) were compared. Althoughanalysis showed significant differences among the groups (Table 5, ANOVA: F4,85 = 5.9,p <0.001), fertilization success of all diet groups was high (90% and above). Success ofeggs arising from adults reared on LO-PRO, KELP, and HI-PRO+B were notsignificantly different and were lower than rates observed for HI-PRO and HI-PRO+C/Bdiets (NK, p<O.05).Effect ofAdult Diet on Timing ofLarval DevelopmentFigure 4 shows rates of larval development among the five diet groups (LO-PRO,HI-PRO, HI-PRO+B, HI-PRO+CIB, and KELP). Differences in rates of developmentwere noted as follows: diet groups HI-PRO+B and HI-PRO+C/B were the first to reachthe feeding pluteal stage at 3 days, and the 2-arm pluteus at 5-6 days. The feeding plutealstage was attained by LO-PRO and HI-PRO at 4 days, and the 2-arm stage at 7-8 days.This initial pattern in timing of development did not persist, as LO-PRO, HI-PRO, and HIPRO+CIB larvae were comparable by the 4-arm stage and remained so throughoutdevelopment. HI-PRO+B had the shortest development time, reaching the fully-formed 8-arm larval stage 3 days earlier than other diet groups. A larger rudiment size attained byHI-PRO+B when measured at 40 days (seen later in Table 6) also indicates that this larvalgroup developed quicker than the larvae of the other diet groups. Early stage larvae fromKELP-conditioned adults were abnormally shaped and therefore it was difficult todetermine when larvae reached each developmental stage. Rates for this group were basedthen on times when the larvae showed evidence of some new morphomethc developmenthowever abnormal or stunted.426- arm Met amorphos isI , I I Ipluteus 4-arm 14% 43%2-arm 6-arm 8-arm MetamorphosisI I I I I I I Ipluteus 4-arm 75% 86%2-arm 6-arm 8-arm MetamorphosisI I I I I I I Ipluteus 4-arm 10% 34%2-arm 6-arm 8-arm MetamorphosisI I I I I I I Ipluteus 4-arm 15% 40%“2-arm” “8-arm” MetamorphosisI I I I I I“pluteus” “70%”I I I I I I I I I I IFigure 4: Larval development rates (in days) for larvae arising from adultStrongylocentrotus droebachiensis reared on selected artificial and natural diets for ninemonths (HI-PRO+C/B, HI-PRO+B, HI-PRO, LO-PRO, KELP).H I - P R 0+ C lB2 - armHI - PRO+B8 - armHI - PROLO - PROKELP0 10 15 20 25 30 35 40 45 50 55DAYS43Effect ofAdult Diet on Larval SurvivalLarval survival rates of the LO-PRO, HI-PRO, HI-PRO+B, and HI-PRO+C/Bdietary treatments ranged from 92-95% (based on numbers of larvae counted at 20 dayspost-fertilization when all groups had successfully achieved the 6-arm stage). Larvae fromthe KELP-dietary treatment, however, showed abnormal development (see Fig.4) andsubsequent high mortality. At 20 days following fertilization only 15% of the initial larvaecultured from KELP broodstock were alive. Due to low numbers of surviving larvae andtheir abnormal shape, morphometric measurements were not done; however,developmental rates of surviving larvae were assessed. Those few KELP larvae which didsurvive eventually attained the fully formed 8-arm shape and were capable ofmetamorphosis. Interestingly, a replicate larval culture for the KELP diet, established onemonth from the first, showed simiiar larval abnormalities. I concluded from this that thelarval abnormalities resulted from some feature of egg quality and not from my culturingmethods.Effect ofAdult Diet on Larval MorphometryThe morphometric data are plotted against the stage of development in Figures 5-9.This format allows a comparison of the magnitude of change in larval form at equivalentstages throughout development.Overall larval lengths were significantly different among diet groups at eachdevelopmental stage (Fig.5, ANOVA and NK statistics shown on figure). At the 4-, 6-,and 8-arm stages, groups HI-PRO and HI-PRO+C/B were significantly smaller thanlarvae from groups HI-PRO+B and LO-PRO. Larvae of LO-PRO were consistentlylargest throughout development from the 2-arm to 8-arm stage. Larvae of HI-PRO, exceptat the 6-arm stage, were smallest throughout development.4410004003002001000 I I I I0 2-arm 4-arm 6-arm 8-armSTRGE OF DEVELOPI%IENT4671 47 61 7461 4761ANOVA (F3,76= p=) 3.0 <0.05 25.2 <0.0001 8.7 <0.0001 12.0 <0.0001Figure 5: Relationship between overall larval length (urn) and stage of development forStrongylocentrotus droebachiensis larvae. Values are mean ± S.E. of 20 larvae measuredfrom each diet group. S.E. values not visible are contained within the dimensions of thegraph points. Results of ANOVA test and Newman-Keuls multiple comparison tests arealso included. Values which did not differ at the 0.05 level are joined by a solid line.Values for NK tests are shown in ascending rank order from left to right in accordance withdietary numerical designations. Designations are: (1) LO-PRO, (4) HI-PRO, (6) HIPRO+B, (7) HI-PRO+C/B.45Figure 6 shows that body length varied significantly throughout developmentalstages among diet groups. Initial differences did not persist throughout development, andrankings of groups varied over time (ANOVA and NK statistics shown in figure). Forexample, body lengths of LO-PRO, HI-PRO+B, and HI-PRO+C/B, were significantlysmaller than HI-PRO at the 2-arm stage, but by the fully-formed 8-arm stage, LO-PRO andHI-PRO+C/B were significantly larger. The only clear pattern of growth showed up in theLO-PRO and HI-PRO treatments, where HI-PRO was consistently small from the 4-to 8-arm stages, and LO-PRO was consistently large.Significant differences in total arm lengths (summed over all arms and presented asa total mean ±S.D., Fig.7) were observed among diet groups throughout development(ANOVA and NK statistics shown in figure). As noted for body lengths, differencesrecorded early in development did not persist over time. At the 4-arm stage, larvae of HI-PRO and HI-PRO+C/B were significantly smaller than those of the LO-PRO group. HIPRO larvae were smaller throughout development; however, larvae from the same HI-PRObase protein level plus supplements, HI-PRO+C/B and HI-PRO+B , were larger at laterstages. HI-PRO+B was the only diet which maintained significantly high values for totalarm lengths from the 4-arm stage on.460 6 1)L0-PRO(7)HI-PRO+C/$- •(6)HI-PRO+0.5 —. - - - (4)H1-PRO-.-- --——- __.___‘—_0.20.10 I I0 2-arm 4-arm 6-arm 8-armSTAGE OF DEVELOPMENT1674 4761 _47 1 46L1ANOVA (F3, 76 ,P) 3.9 <0.05 7.4 <0.001 17.0 <0.0001 23.0<0.0001Figure 6: Relationship between larval body length (mm) and stage of development forStrongylocentrotus droebachiensis larvae. Values are mean ± S.E. of 20 larvae measuredfrom each diet group. S.E. values not visible are contained within the dimensions of thegraph points. Results of ANOVA test and Newman-Keuls multiple comparison tests arealso included. Values which did not differ at the 0.05 level are joined by a solid line.Values for NK tests are shown in ascending rank order from left to right in accordance withdietary numerical designations. Designations are: (1) LO-PRO, (4) ff1-PRO, (6) HIPRO+B, (7) HI-PRO÷CIB.4765 ,•(6)H1-PRO+13,“(7)HI-PRO+C// (1)LO-PRO(4)HI-PROIIf0 2-arm 4-arm 6-arm 8-armSTAGE OF IJEVELOPMENT4761 74j 1476 4176ANOVA(F3,76=,p=) 5.2 <0.01 32.4 <0.0001 26.3 <0.0001 26.2 <0.0001Figure 7: Relationship between total arm length (mm) (summed over all arms) and stage ofdevelopment for Strongylocentrotus droebachiensis larvae. Values are mean ± S.E. of 20larvae measured from each diet group. S.E. values not visible are contained within thedimensions of the graph points. Results of ANOVA test and Newman-Keuls multiplecomparison tests are also included. Values which did not differ at the 0.05 level are joinedby a solid line. Values for NK tests are shown in ascending rank order from left to right inaccordance with dietary numerical designations. Designations are: (1) LO—PRO, (4) HI-PRO, (6) HI-PRO+B, (7) HI-PRO+CIB.48Ciliated band lengths were signifcantly different among diet groups from the 4-armstage (Fig.8, ANOVA and NK statistics also shown in figure). High values wereconsistently recorded in HI-PRO+B larvae and low values in HI-PRO larvae.Significant differences were observed in ciliated-band to body-length ratio amongthe larvae from the various diet groups and, as expected, these ratios increased with thestage of larval development (Fig.9, ANOVA and NK statistics shown in figure). Aconsistent trend, however, appeared only in one diet group. From the 2-arm to fullydeveloped 8-arm stage, HI-PRO+B larvae had significantly larger ratios than larvae fromother diet groups. For the other diet groups, initial ratio differences did not persistthroughout development495.(6)HI-PRO+$—• (7)HI-PRO+CI4. !/(1)LO-PRO,“ / F (4)H1-PRO=.I—I/I:z 3-.‘I.’ I /,( 1 -—0 I I I0 2-arm 4-arm 6-arm 8-armSTHGE OF DEVELOPMENT4761 4716 7416 4176ANOVA (F3,76= p) 8.0 <0.0001 14.3 <0.0001 21.9 <0.0001 25.0 <0.0001Figure 8: Relationship between ciliated band length (mm) and stage of development ofStrongylocentrotus droebachiensis larvae. Values are mean ± S.E. of 20 larvae measuredfrom each diet group. S.E. values not visible are contained within the dimensions of thegraph points. Results of ANOVA test and Newman-Keuls multiple comparison tests arealso included. Values which did not differ at the 0.05 level are joined by a solid line.Values for NK tests are shown in ascending order from left to right in accordance withdietary numerical designations. Designations are: (1) LO-PRO, (4) HI-PRO, (6) HIPRO+B, (7) HI-PRO+C/B.509.L716 4176 4716 1 L46ANOVA (F3,76= ,p=) 3.89 <0.05 52.4 <0.0001 52.4 <0.0001 40.1 <0.0001Figure 9: Relationship between ratio of the ciliated band length to body length and stage ofdevelopment for Strongylocentrotus droebachiensis larvae. Values are mean ± S.E. of 20larvae measured from each diet group. S.E. values not visible are contained within thedimensions of the graph points. Results of ANOVA and Newman-Keuls multiplecomparison tests are also included. Values which did not differ at the 0.05 level are joinedby a solid line. Values for NK tests are shown in ascending order from left to right inaccordance with dietary numerical designations. Designations are: (1) LO-PRO, (4) l{1-PRO, (6) HI-PRO+B, (7) HI-PRO+C/B.876•(6)HI-PRO+.‘ (4)H[-PRO5.=Iz=Iz=z4.-‘I ,(7)H1-PRO+CI)3(1)LO-PRO/‘_ /1/3.200 2-arm 4-armSTOGE OF DEVELOPMENT6-arm 8-arm51Rudiment diameters differed significantly among the diet groups at both the 8-armpluteal and competency stages (Table 6). At the 8-arm stage, HJ-PRO+CIB larvalrudiment sizes were significantly larger than those of HI-PRO+B, HI-PRO, and LO-PRO.When measured prior to testing metamorphic success, larvae of all diet groups showedsignificant differences in rudiment size, with HI-PRO+B being the largest and HI-PRO thesmallest.Figure 10 is a series of photographs of representative larvae taken duringdevelopment for a comparison of overall body shape. Principal features to note are theabnormalities of KELP larvae taken at 15 days postfertiization including stunted arms,enlarged mouth, and abnormal body shape (Fig. lOa). Figures 10(b-f) show sample larvaefrom each diet group taken at 41 days postfertilzation when all groups were nearing thefully-formed 8-arm pluteal and competent stages. Of particular note is the large overalllength of the LO-PRO larvae owing to both the length of the arms and extended body(Fig. lOb), the small overall size of RI-PRO larvae owing to short arms and small bodylength (Fig. lOc), the large overall size of HI-PRO+B larvae owing to comparatively longarms, moderate body length, and large body width enclosing a well-developed rudiment(Fig. 1(M), moderate overall size of the HI-PRO÷C/B larvae (Fig. 1(M), and attainment of anormal 8-arm stage by some of the KELP larvae (Fig.lOf).52Figure 10 (a-f): Photographs of Strongylocentrotus droebachiensis larvae from adultsconditioned on five selected dietary treatments for nine months (magnification 340x).(a) KELP larvae 15 days postfertilization, (b-f) 8-arm pluteal larvae of each diet group at 41days postfertilization: (b) LO-PRO, (c) HI-PRO, (d) HI-PRO-i-B, (e) HI-PRO+ C/B, (1)KELP.(a) (b) (c)(d) (e) (f)53Effect ofDiet on Metamorphosis SuccessBased on the foregoing comparison of rudiment sizes which indicated that all larvaewere approaching competency, groups were assessed at 45 and then again at 55 dayspostfertilization. Large-scale differences in metamorphic success were observed (seeFig.4). At 45 days values for metamorphosis success were 15, 10, 75, and 14% for LOPRO, HI-PRO, HI-PRO+B and HI-PRO+C/B, respectively, indicating that HI-PRO+Blarvae were the first to reach competency. At 55 days, the values were 40, 34, 86, and43%, for LO-PRO, HI-PRO, HI-PRO+B, and HI-PRO+C/B respectively, indicating notonly that HI-PRO+B larvae were again the most competent but also that other larvalgroups were maturing and nearing competency since the overall percentage ofmetamorphosis increased between testing periods. So irresistable was the urge to settle inthe HI-PRO+B larvae that they could be seen metamorphosing indiscriminately on thewalls of their culture beakers throughout the testing period. The results obtained fromthese experiments further showed that the development rates of HI-PRO+B larvae wereshorter than those of other dietary treatments.In summary, based on assessments of development rates and percentagemetamorphosis, the larvae produced from adults conditioned on the HI-PRO+B diet werethe most vigorous and, thus, this diet could be concluded to be the best for a cultureprogram.54Table 6: Rudiment diameters of Strongylocentrotus droebachiensis larvae from selectedadult diet groups measured at the 8-arm stage and at 40 days postfertiization. Values aremean ± S.D. of 20 larvae measured from each group. Results of ANOVA and NewmanKeuls multiple comparison tests are also included. Values that did not differ at the 0.05level are joined by a solid line. Values of NK tests are shown in ascending rank order fromleft to right in accordance with the dietary numerical designations.Diet Group Rudiment Diameters at 8-arm stage and at 40 days Postfertilization (um)8-arm 40 days Postfertilization(1) LO-PRO 183 .0±18.8 227 .0±20.5(4) HI-PRO 17 8.0±20.4 222.0±20.9(6) HI-PRO+B 172.0±20.5 264 .0±9.9(7) HI-PRO+C/B 189.0±11.4 252.0± 14.8ANOVA (F3,76= , p=) 3.21 <0.05 27.6 <0.0001NKtest 6 4 1 7 4 1 7 655DISCUSSIONThis study not only showed that eggs arising from different adult diets exhibitedqualitative differences in size and energy content which, in turn, influenced larvalmorphometry and development, but it also contributed information useful to a sea urchinaquaculture program. The most important was the identification of a broodstock diet, HIPRO+B, that produces larvae with high rates of development, survival, andmetamorphosis. Frequently marine invertebrate culture operations fail at the larval stage,because it is the most vulnerable phase of the life cycle with exact requirements oftendifficult to meet in a laboratory situation. Therefore, the formulation of a successfulbroodstock diet is an important contribution to any future sea urchin farming. Thesignificant variations in egg size, egg energetics, and larval development observed in thisstudy support previous studies which indicate that adult diet affects egg quality (Thompson1982, Walker 1982, George 1990, George et al. 1990) and, in turn, larval vitality (Georgeet al. 1990). However, the present study has added information on certain nutrientsupplements to broodstock diets which are beneficial to gamete nutrition. It has alsosuggested a relationship between egg quality and larval morphometry, both importantelements contributing to the vitality of the larvae.Since fertilization rates were high for all diet groups, differences observed in eggquality and egg size most likely resulted from nutritional differences in adult diet rather thanfrom differences in timing of the gametogenic cycle. The natural spawning period ofStrongylocentrotus droebachiensis is from late-March to June (Himmelman 1978) and thehigh fertilization rates obtained in late-March (90% and above) in the present experimentindicate that individuals from all dietary treatments were in reproductive synchrony andpeak spawning condition. Dietary conditioning, therefore, although significantlyinfluencing the extent of gonadal growth (see Section 2) had no discernable effect on the56timing of the reproductive cycle. Keats et al. (1984) and Pearse et al. (1986) alsodetermined that neither food type nor amount affected the timing of the reproductive cycle.Dietary Treatments Produce Large EggsEggs of Strongylocentrotus droebachiensis normally range from 136- 168um indiameter, increasing in size with increasing latitude (Emlet et al. 1987). Since myexperimental animals were collected at the southern end of their Alaska-Washingtondistributional range, I expected their eggs to be at the small end of their size range.However, eggs produced by all dietary treatments were large (means of 164-l7Oum),probably reflecting the high quality of the artificial diets and the consistent feeding regimeemployed. Interestingly, eggs arising from HI-PRO+C reached a diameter of l7Oum, asize not previously recorded for this species. Supplemental cholesterol included in thisdietary treatment may have been responsible. Cholesterol is known not only to play asignificant role in the structuring of cell membranes, but in echinoderms is known also tobe present in large amounts in the yolk of the eggs (Ferguson 1969).Dietary Treatment Influences Egg QualityThe significant differences in egg energy content recorded in this study add supportto previous studies which show an allocation of resources to offspring linked to adult diet(Thompson 1982, George et al. 1990). Interestingly, because the eggs from both the LOPRO and HI-PRO diets were lowest in organic carbon level, the supplements of mannitol,algin, B-carotene, and cholesterol in the other dietary treatments appear to have contributedin some way to egg energy content.Proteins and lipids are the most important energy reserves in eggs of echinoderms(Emlet et al. 1987), frequently ranging between 54-74% and 21-38% of the dry weight,respectively (Turner and Lawrence 1979). A single egg of Strongylocentrotus57droebachiensis has been shown to contain 0.Olug carbohydrate, 0.1 lug of lipid, and0.24ug of protein (Turner and Lawrence 1979). Variations in the concentrations of thesenutrient components are influenced by adult diet (Walker 1982, Thompson 1982, George etaL 1990). For example, George eta!. (1990) noted that populations of Arbacia lixulahaving access to higher quality algae (defined by higher levels of proteins, carbohydrates,and lipids) produced eggs with higher protein levels than did those having access to poorerquality algae. Also, Thompson (1982) observed differences in egg lipid levels inStrongylocentrotus droebachiensis and related this to various ration levels of kelp andmussel flesh fed to adults. It is possible that the differences observed in egg energy contentin my study were due to differences in the levels of proteins rather than to differences inlipids, since lipid concentration in the gonads of S. droebachiensis was found to vary only1% between November-March (see Section 2). Also, George et al. (1990) observed thategg lipid levels were relatively similar while protein levels varied significantly in eggs oftwo differently fed adult sea urchin populations.Egg Quality is Enhanced by B-caroteneThe high quality larvae resulting from the HI-PRO+B diet may have resulted fromthe inclusion of B-carotene in the eggs. Sea urchin eggs are known to contain a variety ofcarotenoid pigments. Researchers have observed the fate of pigment cells during oogenesisand early embryogenesis but have not confirmed their function. Burke and Bouland(1989) observed viable pigment cells at the surface of Dendraster excentricus eggs duringoogenesis and suggested that the carotenoid-filled pigment cells may assist both in thematuration processes of the egg and, if acting like the follicle cells of other echinderms,may also have a role in nutrition (Gardiner 1972). Kume and Dan (1968) site referenceswhich noted that the orange pigment-containing granules covering the surface ofunfertilized sea urchin eggs stay at the surface of the swimming blastulae, suggesting thattheir presence may be essential for development.58Eggs Inadequate from KELP broodstockAlthough eggs produced by the animals on the KELP diet showed high fertilizationsuccess and apparently normal cleavage to the ciliated blastula, further development wasabnormal. Early larvae maintained an abnormal prism shape for a prolonged period, andthe “2”-and “4-arm” larvae had enlarged mouths and inferior arm development that resultedin overall stunting (see Fig. lOa). Early mortality rates were high, probably related to theinablity of the larvae to feed effectively. Interestingly, those few larvae which survived theearly stages finally achieved a normal shape and were capable of metamorphosis. Theabnormalities of embryos and larvae of the KELP-fed group possibly resulted from somenutrient inadequacy of the eggs, not from poor health of the broodstock per se since theysurvived well and grew comparatively large gonads (see Section 2). Lack of certainnecessary micronutrients such as vitamins and B-carotene may be implicated, since theseare known to be at least partially destroyed in the drying of the algae (pers. comm. IanWhyte). These results suggest that such dried kelp is inadequate for use as a broodstockdiet in sea urchin aquaculture.D(ferences in Egg Quality not Egg Size translate into Differences in Larval Developmentand MorphometryBoth egg size and biochemical composition have been shown to affect the timing oflarval development in echinoderms (McEdward 1986, Emlet eta!. 1987, George etaLl99O). In his comparison of three related echinoid species, McEdward (1986)determined that egg volume was inversely related to development time. However, basedon the work of Amy (1983), who found no consistent relationship between egg size andrate of embryonic development in five species of echinoids, and of Dickie (1989), whodetermined that development times did not vary consistently with egg diameter in fourspecies of echinoderms, it seems likely that egg size is secondary to features of egg59“quality” with respect to larval developmental rate. This idea is supported by the fmdingsof my study which showed that the smallest and largest eggs of the broodstock treatmentgroups (HI-PRO and LO-PRO diet groups, respectively) reached the feeding pluteal stagesimultaneously.The lack of a consistent relationship between egg size and timing of developmentseen in previous studies may be, in part, a factor of the different methods used to assessegg size. Egg size is usually measured as diameter or volume. Turner and Lawrence(1979) point out that egg diameter has many shortcomings because it is difficult to compareeggs of different shapes and suggest instead to use volume. Gallagher and Mann (1991)also consider egg volume to be a more descriptive parameter than egg diameter, pointingout that any small changes in egg diameter will lead to large volumetric changes and,importantly, a large egg containing the same percentage of lipid, for example, as a smalleregg could contain more than double the absolute quantity of lipid. Therefore, researchersrelating egg diameter to larval development may achieve different results from those relatingegg volume to larval development.A relationship between egg quality (measured as amount of organic carbon) andrates of larval development was seen in my study. High quality eggs of HI-PRO+B andHI-PRO+CIB reached the feeding pluteus stage one day earlier than the poorer qualityeggs of the LO-PRO and HI-PRO diet-groups. Eggs with the highest amounts of organiccarbon (HI-PRO+CIB) were first to reach the 2-arm pluteus stage at 5 days, followed (inorder of organic carbon content) by NI-PRO+B at 6 days; HI-PRO at 7 days, and LOPRO at 8 days. However, such differences in timing found during the initial stages oflarval development did not persist throughout development. For example, the low-energyeggs of the LO-PRO and HI-PRO diets reached the 4-,6-, 8-arm, and competency stagessimultaneously with the highest-energy eggs of HI-PRO+CIB. The HI-PRO+B larvae60had the fastest rate of development over all the stages, reaching 4-,6-, 8-arm stages at 2,1, and 4 days, respectively, ahead of LO-PRO, HI-PRO and HJ-PRO+CIB and, based onpercentage metamorphosis, were competent at least 10 days before the others as well.These findings are supported by George et aL (1990), who observed 2-3 days difference inthe timing of larval development arising from eggs differing in protein levels, with initialdifferences disappearing as time to metamorphosis approached.Although George et aL(1990) determined that egg quality can affect subsequentlarval development, her study did not include the role of differences in larval morphometryin speed and success of development. However, she did suggest that on the basis ofMcEdward’s (1986) work, better-quality eggs (containing higher levels of protein) likelydeveloped into larvae with higher feeding efficiencies, and that this, in turn, wouldcontribute to their overall success. My study provides more conclusive evidence of arelationship between the parameters of egg quality, larval morphometry, and rates andsuccess of development, as explained in the following.Larval form is suggested to have potentially important consequences on rate andmagnitude of subsequent larval growth and development (McEdward 1986). Strathmann(197 1) has shown that the length of the ciliated bands in sea urchin larvae profoundlyinfluences the feeding effectiveness of the larvae by the fact that the maximal larvalclearance rates are linearly related to ciliated band length. Strathmann noted that ciliatedband length increases 7-9 fold during arm formation, considerably increasing the rate offeeding and translating presumably into a coffespondingly decreased development time.McEdward (1986) developed this idea one step further in his comparative study of larvalmorphometry of echinoplutei, when he proposed that to maintain or increase feeding withrespect to metabolism during development, plutei must increase feeding structuresdisproportionately faster than body size. Echinoplutei characteristically show a pattern of61positive allometric growth of the ciliated band relative to body length, and McEdward(1986) suggested that the greater the difference between the two, the greater will be thefeeding effectiveness and thus the greater rate of development. The relationships betweenciliated band length, ratio of ciliated-band to body-length, and timing of developmentobserved in my study support these points of view. Developmental times tometamorphosis were shortest for larvae from the HI-PRO+B diet which, from 4-to 8-armstages, had the longest ciliated-band measurements and highest ciliated-band to body lengthratio. In contrast, developmental times were longest (during competency tests theyconsistently recorded the lowest percentages of metamorphosis) for larvae from the HI-PRO diet. These exhibited shortest ciliated band lengths at the 2-, 4-, and 8-arm stages andsmallest ciliated-band to body length ratios at the 2-,4-, and 6-arm stages.Interestingly, some of the differences in early development rates did not persistthroughout development. For example, the larvae produced by LO-PRO which, in turn,produced low quality eggs, took the longest to reach the 2-arm stage, but by the 4-armstage they caught up to the larvae from HI-PRO and HI-PRO+C/B and had similar rates ofmetamorphosis. George eta!. (1990) observed a similar trend when comparing larvaereared from eggs of different protein content. This trend may be explained by the ability ofechinoplutei to respond to unfavourable environmental conditions by making adaptivechanges in shape. For example, Hart and Scheibling (1988b) reared groups of green seaurchin larvae under different temperature and food regimes, and observed a capacity of thelarvae to increase arm length in response to food limitation. It may be possible then, thatlarvae can similarly respond to initial inadequacies arising from egg quality. In this regard,the LO-PRO larvae exhibited the longest arm lengths at the 2- and 4- arm stages and, bybeing able to feed more effectively, were perhaps able to attain the same growth rates aslarvae arising from better diets and eggs. This would be a productive line for futureresearch.62GENERAL CONCLUSIONSThe use of artificial diets proved effective for both the maintenance and growth ofadult Strongylocentrotus droebachiensis and, as well, for conditioning diets forbroodstock. Differences in the adult dietary treatments produced nutrition-related variationsin gonadal growth, egg size and quality, and overall larval vitality, suggesting that adultdiet most likely affects all stages of the life cycle. In general, diets which were mostbeneficial to the adults were also most beneficial for the eggs and young (except the KELPdiet which, although producing large gonads, produced poor quality larvae). For example,the low protein diet (LO-PRO) gave rise to comparatively small gonads, intermediate-sizedeggs but with low organic carbon levels, and larvae which initially were slow to developand had intermediate levels of metamorphosis. In contrast, the high protein dietsupplemented with beta-carotene (HI-PRO+B ) resulted in large gonads, large eggs withintermediate amounts of energy, and the most vigorous larvae both in terms of developmenttimes and success of metamorphosis.With respect to a possible aquaculture program for the green sea urchin, S.droebachiensis, the most successful diet of the ones tested proved to be the high proteindiet supplemented with beta-carotene (HI-PRO÷B), for not only did this treatment producelarge gonads suitable for the roe market, but it also was the best diet for rearing ofbroodstock as the offspring had the shortest development times and highest rates ofmetamorphosis. It is often the failure of larval culturing which prevents the successfulfarming of marine invertebrates, so the production of healthy (high survival rates) andvigorous (short development times) larvae from the artificial HI-PRO+B diet is animportant fmding.63SECTION: 4 APPLICATIONS TO AQUACULTUREA report on the management of B.C.’s sea urchin fisheries presently being draftedby D.F.O. (December 1993) states that wild stocks can not sustain present fishingpressures. The latest statistics indicate a decline in stocks leading to more stringent landingquotas and seasonal closures. In 1992 landings of the red urchin fished along the northcoast reached a peak of 12,018 tonnes and for the first time in its 20-year history the northcoast fishery will no longer function year-round in 1994. Now weekly, daily, and areaclosures will be in effect, quotas applied, and allowable landings reduced to 7,811 tonnes.Quotas are also proposed for green urchins in 1994, and area and weekly closures will alsobe in effect. Targeted landings of green urchins will be 640 tonnes, down from 984 tonnesin 1992. Interestingly, diver-catch-per-unit effort has also declined by 40% since 1989,indicating that easily accessible stocks are being fished out. All of these statistics lendcredence to the potential of urchin farming in the future.Interestingly, it is the roe of the green urchin which commands the highest marketprice in Japan. Unlike red sea urchin roe which is processed locally prior to shipment,green urchins are shipped live to Japan and sold as Japanese tini. British Columbiafishermen in 1993 received $0.77 per kilogram for red urchin roe and $4.58 per kilogramfor whole green urchins.The present study produced several fmdings of importance to aquaculture: 1) itestablished that sea urchins will grow marketable gonads on artificial diets, 2) itdetermined that a single diet could be used successfully to produce both marketable roe andvigorous larvae, 3) it confirmed that the green sea urchin is suitable as a species foraquaculture in that it is tolerant of high densities, disease resistant, and exhibits lowmortality in the laboratory, 4) it determined not only that the best harvestable roe is64achieved in November-January, but also that the HI-PRO+B diet had the largest gonads atthis time (indicating also that this diet would be most cost effective), and 5) it establishedthat females produce larger gonads and would, therefore, be more profitable to rear.Furthermore, these findings if combined with the work of Pearse et al. (1986), whodetermined that the reproductive cycle of the sea urchin is under photoperiodic control,would enable the aquaculturist to time the farmed population to mature at different times ofthe year and provide a marketable product year round.Even though the development of a successful artificial diet was achieved in thisstudy, many other factors must be considered to determine if a sea urchin culturingprogram is worthwhile. A cost-comparison of land-based culture, enhanced ocean fishery,and natural fishery needs to be prepared, as well as a comparison of the costs of artificialand natural diets. Importantly, too, taste and colour assessments must be done on the roeproduced from the artificial diets to assess their palatability.Sea urchin gonads are known to be highly nutritious. One hundred grams of freshgonads can contain 71% water, 16% protein, 9% fat, 2% carbohydrate and 2% ash, alongwith 20 mg calcium, 300 mg phosphorus and 2 mg of iron and, appealing to the presenttrends in the health food market, 3,000 I.U.of vitamin A, 7,500 I.U. of B-carotene,0.30mg Bi, 0.4 mg B2, and 2.5 mg of nicotinic acid, and is represented by 148 calories.Development of local and North American markets, through advertising the nutritionalbenefits of urchin roe and developing ways of making the roe more appealing to the NorthAmerican palate, would be an invaluable contribution to the success of sea urchinaquaculture in British Columbia.65REFERENCESAmy, R.L. 1983. Gamete sizes and developmental time tables of five tropical sea urchins.Bull.Mar.Sci. 33:173-176.Andrews, N.L. 1986. The intertaction between diet and density in influencingreproductive output in the echinoid, Evechinus chioroticus. J.Exp.Mar.Biol.Ecol. 97:63-79.Anon. 1984. On the natural seeds collection, intermediate culture and release of the seaurchin, Strongylocentrotus intennedius. J.Hoklcaido Fish.Exp.Stn. 41:270-315.Anon. 1985. British Columbia Catch Statistics. D.F.O. Canada.Anon. 1991a. Green sea urchins. Management plan. Pacific Region. D.F.O. Canada.Anon. 199 lb. Management plan for the 1991 red sea urchin fishery. Pacific Region.D.F.O. Canada.Barnes, R.D. 1987. Invertebrate Zoology. 5th ed. Saunders College Publishing.N.Y.Bligh, E. & W.J. Dyer. 1959. A rapid method of total lipid extraction and purification.Can.J.Biochem.Physiol. 37:911-917.Breen, P.A., D.C. Miller & B.E. Adkins. 1976. An examination of harvested sea urchinpopulations in the Tofino area. Fish.Res.Bd.Can., Publ.No.1401.Boolootian, R. & R. Lasker. 1964. Digestion of brown algae and the distribution ofnutrients in the purple sea urchin Strongylocentrotus purpuratus.Comp.Biochem.Physiol. 11:273-289.Boolootian, R. 1966. Reproductive physiology. In: Physiology of Echinodermata. JohnWiley & Sons. N.Y. pp.561-606.Burke, R.D. 1980. Podial sensory receptors and the induction of metamorphosis inechinoids. J.Exp.Mar.Biol.Ecol. 47:235-249.Burke, R.D. 1982. Echinoid metamorphosis: retraction and resorption of larval tissues. In:Proc.Int.Echinoderm Conf. Ed. J.M.Lawrence. A.A.Balkema Rott. pp.513-5 19.Burke, R.D. 1983. The induction of metamorphosis of marine invertebrate larvae: stimulusand response. Can.J.Zool. 61:1701-1719.Burke, R.D. & C. Bouland. 1989. Pigmented follicle cells and the maturation of oocytes inthe sand dollar, Dendroster excentricus. Develop.Growth & Differ. 31:431-437.66Carefoot, T.H. 1985. Calorimetry. In: Handbook of Phycological Methods EcologicalField Methods: Macroalgae. Eds. M.Littler and D. Littler. Cambridge University.Cambridge. pp.479-491.Cameron, R.A. & R.T. Hinegardner. 1974. Initiation of metamorphosis in laboratorycultured sea urchins. Biol.Bull. 146:335-342.Cameron, R.A. & R.T. Hinegardner. 1978. Early events in sea urchin metamorphosis,description and analysis. J.Morph. 157:21-32.Cameron, R.A. & S.C. Schroeter. 1980. Sea urchin recruitment: effect of substrateselection on juvenile distribution. Mar.Ecol.Pro.Ser. 2:243-247.Church, D.C. & W.G. Pond. 1988. Basic Animal Nutrition and Feeding. 3rd edition.Wiley & Sons. N.Y.Cognetti, G. 1982. Nutrition of embryos. In: Echinoderm Nutrition. Eds. M. Jangoux andJ.M. Lawrence. A.A. Balkema. Rott. pp.469-476.Conand, C. & N.A. Sloan. 1989. World fisheries for echinoderms. In: MarineInvertebrate Fisheries: Their Assessment and Management. Ed. J.F. Caddy. Wiley& Sons. N.Y.pp.647-663..De Ridder, C. & J.M. Lawrence. 1982. Food and feeding mechanisms: echinoidea. In:Echinoderm Nutrition. Eds. M. Jangoux & J.M. Lawrence. A.A.Balkema. Rott.pp.57-115.De Ridder, C. & J.M. Lawrence. 1982. Digestive systems: echinoidea. In: EchinodermNutrition. Eds. M. Jangoux & J.M. Lawrence. A.A.Balkema. Rott. pp.213-234.Dickie, L., M. Hart & R. Helling. 1989. Prefeeding larval development time is notcorrelated with egg size in regular echinoids (Strongylocentrotus species). Invert.Reprod.Devel. 15:229-232.Ebert, T. 1967. Growth and repair of spines in the sea urchin Strongylocentrotuspurpuratus. Biol.Bull. 133:1:141-149.Ebert, T. 1968. Growth rates of the sea urchin, Strongylocentrotus purpuratus relatedto food availability and spine abrasion. Ecology. 39:6.Ebert, T. 1975. Growth and mortality of post-larval echinoids. Amer. Zool. 15:755-775.Ebert, T. 1983. Recruitment in echinoderms. In: Echinoderm Studies. Eds. M.Jangoux &J.M. Lawrence. A.A. Balkema. Rott. 1:169-203.Edwards, P.B. & T.A. Ebert. 1991. Plastic responses to limited food availibility and spinedamage in the sea urchin Strongylocentrotus purpuratus. J.Exp.Mar.Biol.Ecol.145:205-220.Emlet, R.B. & LR. McEdward, R.R.Strathmann. 1987. Echinoderm larval ecologyviewed from the egg. In: Echinoderm Studies. Eds. M. Jangoux & J.M. Lawrence.A.A. Balkema. Rott. 2:55-136.67Ferguson, J. 1969. Feeding, digestion and nutrition in echinodermata. In: ChemicalZoology. Eds. M. Flork. & B. Scheer. Academic Press. N.Y.pp.71-96.Fong, W. & K. Mann. 1980. Role of gut flora in the transfer of amino acids through amarine food chain. Can.J.Fish.Aquat.Sci. 37:88-96.Fuji, Akira. 1962. Studies on the biology of the urchin v. food consumption ofStrongylocentrotus intermedius. Jap.J.of Eco. 12:5:181-186.Gallagher, S.M. & R. Mann. 1986. Growth and survival of larvae of Mercenariamercenaria and Crassostrea virginica relative to broodstock conditioning and lipidcontent of eggs. Aquaculture. 56:105-121.Gardiner, M.S. 1972. The Biology of the Invertebrates. McGraw-Hill. N.Y.George, S.B. 1990. Population and seasonal differences in egg quality Arbacia lixulaInvert.Reprod.Devel. 17: 111-121.George, S.B., C. Cellario & L. Fenaux. 1990. Population differences in egg qualityArbacia lixula. Proximate composition of eggs and larval development.J.Exp. Mar.Biol.Ecol. 141:107-118.Giese, A.C., L. Greenfield, H. Huang, A. Farmanfarmaian, R. Boolootian and R. Lasker.1959. Organic productivity in the reproductive cycle of the purple sea urchin.Biol.Bull. 116:49-58.Giese, A.C. 1966. On the biochemical constitution of some echinoderms. In: Physiologyof Echinodermata. Ed. R. Boolootian. Wiley & Sons. N.Y.pp.757-797.Gonor, J.J. 1972. Gonad growth in the sea urchins, Strongylocentrotus purpuratus andthe assumptions of gonad index methods. J.Exp.Mar.Biol.Ecol. 10:89-103Gonor, J.J. 1973. Reproductive cycles in Oregon populations of the echinoid,Strongylocentrotus purpuratus. 1. Annual gonad growth and ovarian gametogeniccycles. J.Exp.Mar. Biol.Ecol. 12:45-64.Gonzalez, L.P., J.C. Castilla, & C. Guisado. 1987. Effect of larval diet and rearingtemperature on metamorphosis and juvenile survival of the edible sea urchin,Loxechinus albus. J.Shellfish Res. 6:109-115.Greenfield, L., A.C. Giese, A. Farmanfarmaian & R.A. Boolootian. 1958. Cyclicbiochemical changes in several echinoderms. J.Exp.Zool. 139: 507-524.Hart, M.W. & R.E. Scheibling. l988a. Heat waves, baby booms, and the destruction ofkelp beds by sea urchins. Mar.Biol. 99: 167-176.Hart, M.W. & RE. Scheibling. 1988b. Comparing shapes of echinoplutei using principalcomponents anaylsis, with an application to larvae of Strongylocentrotusdroebachiensis. In: Proc.Echinoderm Conf.Victoria. Eds. R. Burke, P. Mladenov,P. Lambert, & R. Parsley. A.A. Balkema. Rott. 277-284.Himmelman, J.H. & D.H.Steele. 1971. Foods and predators of the green sea urchinStrongylocentrotus droebachiensis in Newfoundland waters. Mar.Biol. 9: 3 15-322.68Himmelman, J.H. 1975. Phytoplankton as a stimulus for spawning in three marineinvertebrates. J.Exp.Mar.Biol.Ecol. 20:199-214.Himmelman, J.H. 1978. Reproductive cycle of the green sea urchin, Strongylocentrotusdroebachiensis. Can.J.Zool. 56:1828-1836.Himmelman, J.H. & H. Nedelec. 1990. Urchin foraging and algal survival strategies inintensely grazed communities in eastern Canada. Can.J.Fish.Aqua.Sci. 47:1011-1026.Hinegardner, R. 1967. Care and handling of sea urchin eggs, embryos, and adults. In:Methods in Developmental Biology. Eth F.H.Wilt & N.K.Wessells.pp. 139-155.Hinegardner, R. 1969. Growth and development of the laboratory cultured sea urchin.Biol.Bu1L 137:465-475.Johnson, C.R. & K.H. Mann. 1982. Adaptations of Strongylocentrotus droebachiensisfor survival on barren grounds in Nova Scotia. In: Proc.Int.Echinoderm Conf.Ed. J.M.Lawrence. A.A. Balkema.Rott. pp. 277-283.Keats, D.W., D.H. Steele & G.R. South. 1984. Depth-dependent reproductive output ofthe green sea urchin, Strongylocentrotus droebachiensis in relation to the nature andavailability of food. J.Exp.Mar.Biol.Ecol. 80:77-9 1.Keegan, B. & B. O’Connor. 1984. Echinoderm fisheries of the world: A review. In:.Proc.5th Echinoderm Conf. Galway. Ed.B.F. Keegan and B.O’Connor. A.A.Balkema.. Rott.Kenner, M. 1992. Population dynamics of the sea urchin Strongylocentrotus purpuratus ina central California kelp forest: recruitment, mortality, growth and diet. Mar.Biol.112:107-118.Kozloff, E. 1990. Invertebrates. Saunders College Publishing. Philadelphia.Kume, M. & K. Dan. 1968. Invertebrate Embryology. Nolit Publishing House.Yugoslavia.Larson, B.R., R.L.Vadas & M. Keser. 1980. Feeding and nutritional ecology of the seaurchin Strongylocentrotus droebachiensis in Maine, U.S.A. Mar.Biol. 59: 49-62.Lawrence, J.M. 1975. On the relationships between marine plants and sea urchins.Oceanogr.Mar.Biol.Ann.Rev. 1975. 13:213-286.Lawrence, J.M. 1982. Digestion. In: Echinoderm Nutrition. Eds. M.Jangoux and J.M.Lawrence. A.A.Balkema. Rott. pp.283-316.Lawrence, J.M. & J.M. Lane. 1982. The utilization of nutrients by postmetamorphicechinoderms. In: Echinoderm Nutrition. Eds. M. Jangoux and J.M. Lawrence.A.A.Balkema. Rott. pp.331-371.Leahy, P.S. 1986. Laboratory culture of Stongylocentrotus purpuratus adults, embryos &larvae. In: Methods in Cell Biology. Ed.T.E.Schroeder. Academic Press.N.Y.Vol.27. pp.1-369Levitan, D.R. 1991. Skeletal changes in the test and jaws of the sea urchin Diadernaantillarwn in response to food limitation. Mar.Biol. 1991. 111:431-435.Lilly, G.R. unpubi. 1975. The influence of diet on the growth and bioenergetics of thetropical sea urchin, Tripneustes ventricosus. Ph.D.thesis.U.B.C.Lowe, E.F. & J.M. Lawrence. 1976. Absorption efficiencies of Lytechinus variegatus forselected marine plants. J.Exp.Mar.Biol.Ecol. 1976. 21: 223-234.McClintock, J.B., T.S. Klinger & J.M. Lawrence. 1982. Feeding preferences of echinoidsfor plant and animal food models. Bull.Mar.Sci. 32:365-369.McEdward, L. R. 1984. Morphometric and metabolic analysis of growth and form of anechinopluteus. J.Exp.Mar.Biol.EcoL 82: 259-287.McEdward, L. R. 1986. Comparative morphometrics of echinoderm larvae. II. Larval size,shape, growth, and the scaling of feeding and metabolism in echinoplutei.J.Exp.Mar.BioLEcol. 96:267-286.McEdward, L. R. & S. F. Carson. 1987. Variation in egg organic content and itsrelationship with egg size in the starfish Solaster stimpsoni. Mar.Ecol.Prog.Ser.37: 159-169.Miller, R.J. & K.H.Mann. 1973. Ecological energetics of the seaweed zone in a marinebay on the Atlantic coast of Canada. III. Energy transformations by sea urchins.Mar. Biol. 18:99-114.Moore, H. 1966. Ecology of Echinoids. In: Physiology of Echinodermata. Ed.R.Boolootian. Wiley & Sons. N.Y.pp.73-83.Mottet, M. 1976. The fishery biology of sea urchins in the family Strongylocentrotidae.Wash.Dept.Fish.Tech.Rep. 20:1-66.Nagai, T. & K. Kaneko. 1975. Culture experiments on the sea urchin Strongylocentrotuspuicherrimus fed an artificial diet. Mar.Biol. 29:105-108.Obrietan, K., M. Drinkwine & D.C. Williams. 1991. Amylase, cellulase and proteaseactivities in surface and gut tissues of Dendraster excentricus, Pisaster ochraceousand Strongylocentrotus droebachiensis. Mar. Biol. 109:53-57.Parsons, T.R., Y. Maita, & C.M. Lalli. 1984. A manual of chemical and biologicalmethods for seawater analysis. Pergamon. N.Y.Pearce, C.M. & R. Scheibling. 1991. Effect of macroalgae, microbial films, andconspecifics on the induction of metamorphosis of the green sea urchinStrongylocentrotus droebachiensis. J.Exp.Mar.Biol.Ecol. 14: 147-162.Pearse, J.S. 1980. Synchronization of gametogenesis in the sea urchin Strongylocentrotuspurpuratus and S. franciscanus. In: Advances in Invertebrate Reproduction. Eds.W.H. Clark & T.S. Adams. Elsevier. N.Y. pp. 53-69.Pearse, J.S., V.B. Pearse, K.K. Davis. 1986. Photoperiodic regulation of gametogenesisand growth in the sea urchin, Strongylocentrotuspurpuratus. J.Exp.Zool. 237:107-118.70Pearse, J.S. & A.H. Hines. 1987. Long-term population dynamics of sea urchins in acentral California kelp forest: rare recruitment and rapid decline. Mar.Ecol.Prog.Ser.39:275-283.Percival, E. & R.H. McDowell. 1967. Chemistry and Enzymology of Marine AlgalPolysaccharides. Academic Press. N.Y.Rowley, R.J. 1989. Settlement and recruitment of sea urchins Strongylocentrotus spp. ina sea-urchin barren ground and a kelp bed: are populations regulated bysettlement or post-settlement processes? Mar.Biol. 100:485-494.Sloan, N.A. 1984. Echinoderm fisheries of the world: A review. In: Proc.EchinodermConf.Galway. Ed. B.F. Keegan & B.O’Connor. A.A. Balkema. Rott. pp.109-124.Sloan, N.A., C.P. Lauridsen & R.M. Harbo. 1987. Recruitment characteristics of thecommercially harvested red sea urchin, Strongylocentrotusfranciscanus in southernBritish Columbia. Fish.Res. 5:55-69.Strathmann, R.R. 1971. The feeding behavior of planktotrophic echinoderm larvae:mechanisms, regulation, and rates of suspension-feeding. J.Exp.Mar.Biol.EcoL 6:109-160.Swan, E. 1966. Growth, autotomy and regeneration. In: Physiology of Echinodermata.Ed. R. Boolootian. Wiley & Sons. N.Y.pp.397-425.Tegner, M.J. & P.K. Dayton. 1977. Sea urchin recruitment patterns and implications ofcommercial fishing. Science. 196:324-326.Tegner, M.J. & P.K. Dayton. 1981. Population structure, recruitment and mortality of twosea urchins, Strongylocentrotusfranciscanus and S. purpuratus, in a kelp forest.Mar.Ecol.Prog. Ser. 5:255-268.Thompson, R.J. 1982. The relationship between food ration and reproductive effort in thegreen sea urchin, Strongylocentrotus droebachiensis. Oecologia 56:50-57.Tsushima, M. & T. Matsuno. 1990. Comparative biochemical studies of carotenoids in seaurchins I. Comp.Biochem.Physiol. 96B:801-810.Turner, R. L. & L. M. Lawrence. 1979. Volume and composition of echinoderm eggs:implications for the use of egg size in life-history models. In: Reproductive ecologyof marine invertebrates. Ed. S. E. Stancyk, University of South Carolina Press,Columbia.S.C. pp.25-40.Turner, R. L. & J.C. Rutherford. 1976. Organic, inorganic and caloric composition ofeggs, pentaculae, and adults of the brooding sea cucumber Cucumaria curata.J.Exp.Mar.Biol.Ecol. 24:49-60.Walker, C. 1982. Nutrition of gametes. In: Echinoderm Nutrition. Eds. M. Jangoux andJ.M. Lawrence. A.A. Balkema. Rott. pp.449-466.Whyte, J.N.C., J.R. Englar & T. Nishihama. 1981. Chemical constituents of Nereocystlsluetkeana. Proc.Int.Seaweed Sym. Eds. G.E. Fogg & W.E. Jones. Uni.College ofN. Wales, Gwynedd. U.K. 8:644-649.71Vacias, R.L. 1977. Preferential feeding: An optimization strategy in sea urchins.Ecol.Monog. 47:337-371.Vevers, H.G. 1966. Pigmentation. In: Physiology of Echinodermata. Ed. R. Boolootian.Wiley & Sons. N.Y. pp.267-27372


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items