Feeding and Digestion in the Phyllosoma Larvae of Ornate Spiny Lobster, Panulirus ornatus (Fabricius) and the Implications for their Culture

Transcription

1 Feeding and Digestion in the Phyllosoma Larvae of Ornate Spiny Lobster, Panulirus ornatus (Fabricius) and the Implications for their Culture By Matthew D Johnston B.Aqua (Hons) This thesis is presented for the Degree of Doctor of Philosophy The University of Western Australia Zoology DECEMBER 2006

2

3 Declaration and Authority of Access I hereby declare that this thesis contains no material which has been accepted for the award of any other degree or diploma at any university, and to the best of my knowledge contains no paraphrase or copy of material previously published or written by another person, except where due reference is made in the text of this thesis. Candidate s signature: Matthew D Johnston This thesis may be made available for loan and limited copying in accordance with the Copyright Act Candidate s signature: Matthew D Johnston i

4 Contribution to Thesis Dr Danielle Johnston prepared the Australian Research Council (ARC) application from which this study was predominately funded. The structure and scope of the study surrounding the ingestive and digestive morphology and physiology and diet development was prepared by Dr Danielle Johnston. All aspects of this thesis in relation to the formulation of the diets and experimental design were prepared by me with assistance from Drs Danielle Johnston, Clive Jones and Brenton Knott. All of the chapters included in this thesis represent multi-authored papers; however, I am the main contributor to each of these chapters and subsequent publications. In each Chapter Drs Danielle Johnston and Brenton Knott provided critical review on the original manuscript. Dr Clive Jones provided critical review of Chapters 5 and 6. In Chapter 3 assistance with transmission electron microscopy was obtained from John Murphy and Steve Parry (Centre for Microscopy and Microanalysis, The University of Western Australia), however all processing and imaging was prepared by myself. Development of a method to determine the concentration of ytterbium oxide in the four diet preference trials was devised in collaboration with Michael Smirk (Department of Soil Science, The University of Western Australia). All diet trials, formulations, dilutions and analytical work were conducted by me. As noted earlier, my supervisors, Drs Danielle Johnston, Brenton Knott and Clive Jones, reviewed this body of work. Their contributions have been further acknowledged with their co-authorship of many of the chapters/papers included in this thesis. ii

5 For my little girl Abigail Katherine Johnston Born 21/07/2006 iii

6 Frontispiece. (Top) Stage IV (instar 6) Panulirus ornatus phyllosoma. (Bottom) Adult Panulirus ornatus (photo supplied by the Queensland Department of Primary Industries, Australia). iv

7 Abstract In this thesis I investigated the ingestive and digestive morphology and digestive physiology during development of phyllosomata of the ornate spiny lobster, Panulirus ornatus. This knowledge was applied to develop a suitable formulated diet to be fed in combination with Artemia or used as a supplement to reduce production costs. The major impediment to closure of the life cycle of spiny lobsters has been an inadequate dietary regime, stemming from a lack of information on their feeding biology and ingestive and digestive capabilities. Of all spiny lobster species, P. ornatus is the best candidate for aquaculture in Australia having the shortest larval development phase (4 6 months) and fast growth rate, attaining 1 kg within 2 years of hatch. Currently, Artemia and fresh feeds such as mussel are used routinely as hatchery feeds. However, the development of a formulated diet that is palatable and delivers the correct balance of nutrients is seen as a highly attractive and cost effective alternative. An appropriate formulated diet for aquaculture of phyllosomata of spiny lobsters can be developed more effectively when the ingestive and digestive morphology, physiology and feeding behaviour are fully understood. To gain this understanding the development of the mouthparts, foregut, digestive gland and digestive physiology was examined using scanning and transmission electron microscopy, histology and biochemical techniques. Mouthpart and foregut structure indicates that P. ornatus phyllosomata manipulate and ingest zooplankton of any hardness and that the size of prey captured, manipulated and ingested would increase during development. The foregut of P. ornatus phyllosoma consists of a single chamber, with well-developed grooves, channels and setae, but lacks a gastric mill. An increase in number of lateral setae, main brushes and the development of a functional filter press at stage IV indicates an increased ability to triturate prey and sort and filter particles internally; reducing both the time spent externally manipulating prey during stages I III with the mouthparts and the vulnerability to predation in the open ocean. The digestive capacity of P. ornatus phyllosoma increased considerably during days of culture/development. Presence of enzyme secreting F-cells, lipid and glycogen storing R-cells, and intracellular digestion and absorptive B-cells in the tubules of the digestive gland from day 0 indicates that P. ornatus phyllosoma have a fully functional digestive gland and therefore the ability to digest ingested prey items/formulated diets. Also, P. ornatus phyllosomata have a diverse range of protease, carbohydrase and lipase enzymes at stage I that enables them to exploit a wide range of zooplankton. High activities of proteases and carbohydrases indicate that both protein and carbohydrate should represent a substantial component of a formulated diet for this species. Observations of the feeding behaviour of palinurid phyllosomata using Panulirus cygnus as a test species demonstrated that their raptorial feeding behaviour places high demands on the physical integrity of a formulated diet. Furthermore, the type of binder and method of drying influences the appeal and water integrity of the diets. The best binder that resulted in the least loss of dry matter was sodium alginate (2% with or without the inclusion of a sequestrant (1%)) for dry meal formulations and gelatine (6.75%) for gelatinous formulations. Diet preference trials indicated that P. ornatus phyllosomata will ingest a formulated diet directly following hatch from the egg and that preference for formulated diets is affected by diet form (texture), particle size, feeding stimulants and level of crude protein. Results have indicated that a suitable formulated diet for aquaculture of early-stage P. ornatus phyllosomata should consist of a firm dietary texture, be within a μm particle size range, contain glycine or the quaternary amino compound betaine, and a consist of a level of crude protein between 44 50%. Partial replacement trials revealed that P. ornatus phyllosoma are stimulated to feed by visual cues. Furthermore, 75% of the entire Artemia ration can be replaced with a formulated diet without having any adverse effects on survival and growth of early-stage phyllosomata. Weaning P. ornatus phyllosomata onto 100% formulated diet during stages II III resulted in reduced survival but demonstrated that diets containing 44 50% crude protein with a diverse range of marine protein sources provides optimum survival and growth. This thesis has identified both physical and nutritional components that will contribute to the successful development of formulated diets for aquaculture of this species. Ultimately, although formulated diets are ingested and provide more than adequate survival when fed in combination with Artemia during early ontogeny, greater success and the possibility of totally replacing Artemia may occur after day 32 (stage IV) due to an increased efficiency to capture and manipulate larger sized particles externally and a greater capacity to triturate prey and sort and filter particles internally. Furthermore, a general increase in specific activity of digestive enzymes at stage IV suggests the possibility of a greater capacity to digest and assimilate nutrients. v

8 Table of Contents Declaration and Authority of Access... i Contribution to Thesis... ii Abstract... v Table of Contents... vi Acknowledgements... viii Chapter 1 Introduction Scope of this Study Biological Characteristics of Panulirus ornatus Development of P. ornatus Phyllosomata and Terminology Australia s Panulirus spp., Attributes for Aquaculture Considerations in Development of Formulated Diets Thesis Structure Chapter 2 Developmental Changes in the Structure and Function of the Mouthparts and Foregut of Early and Late Stage Panulirus ornatus Phyllosomata Introduction Materials and Methods Results Discussion Chapter 3 Structure and Function of the Digestive Gland and Ontogenetic Changes in Enzyme Activity of Panulirus ornatus phyllosomata Introduction Materials and Methods Results Discussion Chapter 4 Stability of Formulated Diets and Feeding Response of Stage I Western Spiny Lobster Panulirus cygnus Phyllosomata Introduction Material and Methods Results Discussion Chapter 5 Effects of Diet Form, Particle Size, Feeding Stimulants and Protein on Diet Preference By Panulirus ornatus Phyllosomata Introduction Materials and Methods Results Discussion Chapter 6 Evaluation of Partial Replacement of Live and Fresh Feeds with a Formulated Diet and the Influence of Weaning Panulirus ornatus Phyllosomata onto a Formulated Diet during Early Ontogeny Introduction Materials and Methods Results Discussion Chapter 7 Summary Findings of the Present Study Suggestions for Future Research Literature Cited Appendix 1 Authorities and Common Names of Species Appendix 2 Description of Stages vi

9 Appendix 3 Mouthpart and Foregut Ontogeny in Phyllosomata of Panulirus ornatus and their Implications for Development a Formulated Larval Diet Appendix 4 Details of Feed Ingredients and Product Codes vii

10 Acknowledgements I thank the following individuals and organisations for their assistance during this study: my supervisors, Drs Danielle Johnston, Brenton Knott and Clive Jones, for their academic guidance and constructive comments on thesis-related issues; Australian Research Council, Queensland Department of Primary Industries and the Department of Zoology (UWA) for financial assistance; Queensland Department of Primary Industries staff; notably, Nick Sachlikidis, Larnie Linton, Shaun Mayes, Darella Chapman, Druce Horton, and Will Bowman for their technical assistance and collection of phyllosomata for both morphology and physiology; Drs Mike Hall (AIMS) and Darren Dennis (CSIRO) for collecting and supplying late-stage phyllosomata. Centre for Microscopy and Microanalysis (UWA) staff, particularly John Murphy and Steve Parry, for their assistance with scanning and transmission electron microscopy; The Western Australian Department of Fisheries for supply of aquaria and marine-based feed ingredients; Michael Smirk in the School of Earth and Geographical Sciences (UWA) for analytical assistance and devising a suitable method to determine the concentration of ytterbium oxide; Microserve Laboratories, Perth for chemical analysis of feed ingredients; Andrew Tennyson for supply of stage I Panulirus cygnus phyllosomata; my lab colleagues, Danny Tang, Patience Lindhjem, Magdalena Zofkova, Sarah Goater and Andrew Limbourn for their friendship and many entertaining moments in the laboratory and at The University Club; Katherine Carlile for her encouragement, support and companionship; my little girl Abigail Katherine who has brought me much joy and love in the last few months; and my family, parents (Peter and Denice) and sister (Nicole) for their ongoing love and support. viii

11 ix

12

13 Chapter 1 Chapter 1 Introduction 1.1 Scope of this Study My thesis investigated the ingestive and digestive morphology and digestive physiology during ontogeny of phyllosoma (plural phyllosomata) of the ornate spiny lobster, Panulirus ornatus (Fig. 1.1) (all authorities for members of the Palinuridae referred to in this thesis are listed in Table 1.1, while authorities for all other species listed in this thesis are provided in Appendix 1). Information concerning ingestive and digestive capabilities of stages I VI (hatchery reared) and stages IX X (wild-caught) P. ornatus phyllosoma provided an understanding of their ingestive and digestive capabilities and how they changed during development. This knowledge was applied to develop a formulated diet that could constitute a complete replacement or be used in conjunction with Artemia and fresh feeds such as mussel gonad to reduce hatchery production costs of the larval phyllosoma phase. To set this thesis in a biologically robust context, I present first in this Chapter the background of spiny lobsters, their taxonomy, systematics and ecology, distribution and habitat, including favourable attributes for aquaculture. Information in relation to unsuccessful commercial hatchery culture of palinurid phyllosoma then follows, outlining the limited knowledge of their diet in the wild and feeding habit, focusing on structure and function of the mouthparts, foregut and digestive physiology, issues specifically relevant to development of a formulated diet. Information pertaining to the development of formulated diets for marine larvae, predominately penaeid prawns and caridean shrimps then follows. I conclude this Chapter with an outline of the scope and structure of my thesis. 1.2 Biological Characteristics of Panulirus ornatus Systematics, Taxonomy, Life History & Ecology of Spiny Lobsters Spiny (rock) lobsters (known in French as langoustes) constitute one family of the superfamily Palinuroidea (Martin and Davis, 2001), as follows: Superfamily: Palinuroidea Latreille, 1802 Family: Palinuridae Latreille, 1802 (spiny/rock lobsters) Family: Synaxidae Latreille, 1825 (coral/furry lobsters) 1

14 Chapter 1 Family: Scyllaridae Bate, 1881 (slipper lobsters) The Family Palinuridae comprises eight genera that were originally divided between two evolutionary lines by Parker (1884), the more ancient, non-stridulating Silentes (Projasus and Jasus), and the more recently evolved, stridulating Stridentes (Linuparus, Puerulus, Palinustus, Justitia, Palinurus, and Panulirus). The eight genera evolved from an ancestral genus that originally occupied the relatively stable, deeper and cooler parts of the Tethys Sea, a continuous ocean that separated the northern landmass of Laurasia from the southern landmass of Gondwanaland (George and Main, 1967). However, fragmentation of Gondwana (120 Ma) caused by the northward drifting of the major continental plates (Indian, Australian, American and African Plates) away from Antarctica resulted in major mountain forming events such as the Himalayan and New Guinean mountains (Kennett, 1977). These mountain chains profoundly influenced the climate by creating new wind and water circulation patterns, and vastly increased riverine inputs of sediment and freshwater into coastal waters (Kennett, 1982; George, 1997). The new positioning of the continental plates would have created physical barriers to mixing of palinurid adults and phyllosomata isolating breeding populations (George and Main, 1967). Figure 1.1 Panulirus ornatus (photo supplied by Department of Primary Industries, Queensland, Australia). Panulirus is the most recently evolved genus of the Stridentes (McWilliam, 1995) and is divided between two major lineages (Table 1.1). Members of the First Major Lineage inhabit well-lit subtropical and tropical shelf-waters; the adults are slow- 2

15 Chapter 1 growing, long-lived and mate seasonally (Table 1.1). Adult females produce one or two batches of eggs each season. The eggs are small, incubation time is short (1 4 months) and the small-sized, stage I phyllosomata pass through many (15 27) instars over a relatively long time (6 13 months) (George, 2005). The Second Major Lineage represents a further radiation of Panulirus into a variety of shallow habitats in the tropics (George, 2005). The warmer conditions have resulted in faster-growing, shortlived, multiple (2 >4) spawning species with a shortened phyllosoma life stage of 4 8 months Distribution and Habitat Of the eight species in the Second Major Lineage, only P. ornatus supports a major fishery within Australia. Although P. ornatus has a wide distribution in the Indo-west Pacific, the species is found predominately in northern Australia, particularly the Torres Strait and far northern Queensland, inhabiting shallow water of muddy, sandy and rocky substrates, often near the mouth of rivers but also on coral reefs (Pitcher et al., 1997). Panulirus ornatus is not restricted to northern Australia, with the species also occurring in isolated pockets throughout the Indo-west Pacific from areas off the East African coast (southern Red Sea to Natal) to southern Japan, the Solomon Islands, Papua New Guinea, New Caledonia and Fiji (Fig. 1.2). Fig. 1.2 Geographical distribution of Panulirus ornatus (Holthuis, 1991). 3

16 Chapter 1 4

17 Chapter 1 5

18 Chapter 1 Radiation of P. ornatus into these areas and associated habitats was influenced by the major Miocene-Pliocene uplift of New Guinea as a result of the collision between the plate carrying Australia with the Asian plate (George, 1997). The elevated landscape created turbid water conditions and a low salinity shelf environment for P. ornatus to evolve (George, 1997) Life History The life cycle of P. ornatus is similar to all members of the Palinuridae, consisting of five major phases: (1) adult, (2) egg, (3) phyllosoma, (4) puerulus (post-larval stage) and (5) juvenile. Palinurid lobsters along with the synaxid and scyllarid lobsters, are the only decapod crustaceans possessing phyllosoma in their life history (Phillips and Sastry, 1980). Phyllosoma (the Greek phyllos, a leaf, and soma, a body) are dorsoventrally flattened, transparent and leaf-like, adapted for passive horizontal transport assisted by diurnal vertical migration (Ritz, 1972). In north eastern Australia, adult P. ornatus reach maturity at 1 2+ years postsettlement and typically migrate to the eastern Gulf of Papua New Guinea (PNG) and to the far northern Great Barrier Reef (GBR) to breed and spawn during the southern hemisphere summer (November to February). The eggs of P. ornatus, which are carried on the pleopods of females for up to 4 6 weeks, hatch into phyllosoma that become part of the plankton. The north eastern Australian stocks of P. ornatus phyllosomata are transported by a complex of oceanic currents, gyres and trade winds. Initially, phyllosomata are carried south-east by the Hiri Current, which flows close to the PNG continental shelf. After reaching the eastern tip of PNG, phyllosomata are transported in a clockwise direction into the Coral Sea Gyre (CSG). The southern margin of the CSG meets the slow-moving Southern Equatorial Current (SEC), and P. ornatus phyllosomata are transported toward the northern Queensland coast (Dennis et al., 2001) (Fig 1.3). At latitudes <16 o S, late-stage P. ornatus phyllosomata are carried north in the prevailing near-shelf current and during the month of May moult into pueruli (Dennis et al., 2001). The pueruli actively swim across the outer GBR and settle on the inner GBR, where they moult into juveniles and develop into adults (Dennis et al., 2001). The release of phyllosomata from December through to April coincides with timing of the Australasian monsoon, an annual meteorological event associated with the inflow of moist, north-westerly winds into the monsoon trough, producing convective cloud and heavy rainfall over northern Australia. These moisture-laden 6

19 Chapter 1 winds originate from the Indian Ocean and south Asian waters. The monsoon also signifies the cyclone season. The release of P. ornatus phyllosomata throughout this time would aid in their dispersal, as oceanographic events such as wind-induced upwelling events and cyclones enhance nutrient availability and concentrate both phytoplankton and zooplankton communities (Rissik and Suthers, 2000). Such timing would be a beneficial life history attribute, creating favourable feeding conditions for P. ornatus phyllosomata to survive and develop into pueruli. Fig. 1.3 Map of the north-west Coral Sea showing major near-surface ocean currents, breeding grounds of Panulirus ornatus (hatched area) and the Torres Strait fishery (dotted area) (adapted from Dennis et al., 2004). 1.3 Development of P. ornatus Phyllosomata and Terminology Panulirus ornatus has a relatively short planktotrophic phyllosoma phase (4 6 months) in which XI phyllosoma stages (>20 instars) precede transformation into a puerulus (Fig. 1.4) (Pitcher et al., 1997; Dennis et al., 2001). Prasad and Tampi (1957), Johnson (1971), and Duggan and McKinnon (2003) have described P. ornatus phyllosoma using plankton hauls and hatchery reared specimens; however, the descriptions are rather brief or incomplete. A complete taxonomic key of the first X phyllosoma stages of hatchery reared P. ornatus is provided in Appendix 2, where distinguishing morphological characters and illustrations for stages and instars are 7

20 Chapter 1 supplied. The provision of taxonomic keys from hatchery reared phyllosomata greatly reduces ambiguity when comparing hatchery reared and plankton hauled individuals (Matsuda and Yamakawa, 2000). The term stage has been traditionally used in studies on crustacean larvae with extended development as a relative measure of morphological development and as such somewhat arbitrary in nature, while instar refers to the inter-moult period between ecdyses. The two terms are not interchangeable as there may be several instars involved in the transition from one stage to the next. In this thesis, the term instar is similarly defined as above while the term stage refers to an arbitrary assessment based on morphological characteristics selected from hatchery reared and wild-caught specimens (Marinovic, 1996). 1.4 Australia s Panulirus spp., Attributes for Aquaculture Spiny lobsters are one of the most valuable crustacean commodities and, on average, are 65 70% more valuable per unit weight than either crabs or prawns (Hall, 2000). Fig. 1.4 Panulirus ornatus phyllosoma (stage IV; instar 6). The spiny lobster fishery is one of Australia's most important, worth $415M pa. or 15% of Australia's total fishery value in (ABARE, 2006). Almost all of the catch is exported as live or chilled product to markets mainly in SE Asia and the United States of America (USA) (Fig. 1.5). As with most fishery sectors, the total catch from wild spiny lobster fisheries cannot completely satisfy market demand and there is growing interest in increasing harvest by culturing selected spiny lobster species. In the immediate term, this could be achieved by on-growing of collected post-pueruli and juveniles taken from the wild, and the holding of adults for weight gain or niche 8

21 Chapter 1 marketing opportunities. However, spiny lobsters reared from the egg are recognised as the only sustainable option for aquaculture having no negative effects on larval recruitment of the wild fishery (Jeffs and Hooker, 2000). Fig. 1.5 Pearl Island holding tank of wild caught Panulirus ornatus (photo supplied by Department of Primary Industries, Queensland, Australia). One of the most critical points for successful aquaculture of spiny lobsters is the spawning of brood-stock and rearing of phyllosomata. Unlike penaeid prawns, where 30% of the total world s prawn production is produced by hatchery culture, spiny lobsters are yet to be cultured commercially to any significant extent in Australia (Hall, 2000). The predominant reason for this is the extended phyllosoma phase. In the Astacidea (freshwater crayfish), crayfish have a post-maternal larval phase which lasts 1 2 weeks. In contrast the shortest phyllosoma phase of Panulirus spp., is typically 4 5 months. Of the seven Panulirus spp., within Australia, four species have a phyllosomata phase of less than 7 8 months; the painted spiny lobster, Panulirus versicolor; the pronghorn spiny lobster, Panulirus penicillatus; the scalloped spiny lobster, Panulirus homarus; and P. ornatus. Of this group, P. ornatus is the predominant candidate for aquaculture in Australia due primarily to its larger size, established markets, members attaining sexual maturity at 1 2+ years, multiple spawning (Table 1.1), short phyllosoma phase (4 6 months), and fast growth rate, attaining 1 kg after 18 months post-settlement (Phillips et al., 1992; Butler and Hernkind, 2000; Smith et al., 2003; Barclay et al., 2006). The recent closure of the life 9

22 Chapter 1 cycle of this species by the MG Kailis Group (recent media release) 1 has increased interest in aquaculture of this species, not only for on-growing but restocking back into the wild fishery. 1.5 History And Culture Problems from Egg Through to Pueruli Rearing experiments involving palinurid phyllosomata were initiated over 60 years ago in Japan (Oshima, 1936). The first initial published record of successful culture was of the Japanese spiny lobster, Panulirus japonicus, which moulted to stage II when fed Artemia nauplii (Nonaka et al., 1958). The period of phyllosoma culture gradually extended to 178 days (Saisho, 1966) and then 253 days (Inoue, 1978). However, complete phyllosoma development was not achieved until it was demonstrated for the Cape spiny lobster, Jasus lalandii by Kittaka (1988). Since then culture has been achieved for the common spiny lobster, Palinurus elephas (Kittaka and Ikegami, 1988), the green spiny lobster, Jasus (Sagmariasus) verreauxi (Kittaka et al., 1997), the southern spiny lobster, Jasus edwardsii (Kittaka et al., 1988) and the long-legged spiny lobster, Panulirus longipes (Matsuda and Yamakawa, 2000). Recently the closure of the life cycle of P. penicillatus (Matsuda et al., 2006), Caribbean spiny lobster, Panulirus argus (Goldstein et al., 2006), and P. ornatus has created significant interest in aquaculture of spiny lobsters. Obviously the selection of species for aquaculture is critical, as the phyllosoma phase extends over several months to years (Table 1.1). Our understanding of the environmental conditions, natural prey and nutritional requirements of palinurid phyllosomata is limited (Phillips and Sastry, 1980; Tong et al., 1997). Furthermore, the food in the foregut of wild-caught phyllosomata, when present, is often difficult to identify (Phillips and Sastry, 1980). This limited understanding of the wild diet, their nutritional requirements and digestive capabilities stands as the major impediment to successful aquaculture of palinurid phyllosomata (Macmillan et al., 1997; Johnston and Ritar, 2001; Nelson et al., 2002; Cox and Johnston, 2003a, 2004). Currently, Artemia nauplii (Branchiopoda: Anostraca) are provided to hatchery reared Panulirus and Jasus spp. phyllosoma due to their size, movement and nutritional value (Inoue, 1965, 1978; Tong et al., 1997). Despite these favourable attributes as a food source, survival of phyllosomata has been low, particularly in the early phyllosoma stages (Phillips and Sastry, 1980; Kittaka, 1994; Kittaka and Abrunhosa, 1997). Greater success has been achieved using Artemia nauplii during

23 Chapter 1 early ontogeny and flesh of mussels and fish larvae after stages V VI (Kittaka, 1988, 1997; Kittaka and Ikegami, 1988; Kittaka et al., 1988; Kittaka and Kimura, 1989; Yamakawa et al., 1989; Ritar et al., 2002, 2003). The absolute requirement of Artemia and fresh feeds is considered to be a limiting factor to the future commercial culture of spiny lobsters. Potential drawbacks include variable nutrient composition, increased risk of the introduction of pathogens into the culture system, and cost. The development of a formulated diet(s) is recognised as a cost-effective approach to feeding palinurid phyllosomata, and meeting their various nutritional requirements during development. 1.6 Considerations in Development of Formulated Diets Development of palinurid phyllosomata is associated with substantial changes in ingestive, digestive morphology and physiology. An appropriate formulated diet can be developed most effectively when the ingestive and digestive morphology, physiology and feeding behaviour are fully understood. This understanding is also required as information in relation to the wild diet of palinurid phyllosomata is relatively unknown (Phillips and Sastry, 1980), hence inferences about the feeding behaviour and processing of prey is based primarily on the morphology of the mouthparts and digestive tract (Mikami et al., 1994; Macmillan et al., 1997; Johnston and Ritar, 2001; Cox and Bruce, 2003; Cox and Johnston, 2003a, 2004). Although considerable research has been carried out on adults and juvenile spiny lobsters, little is known about the feeding mechanism, digestion, digestive enzymes, assimilation of nutrients, structure of the digestive tract and nutritional requirements of spiny lobster phyllosomata (Patwardhan, 1935; Paterson, 1968; Maynard and Dando, 1974; Mikami and Takashima, 1994). Morphological studies of the mouthparts of palinurid phyllosomata have suggested that early-stage phyllosoma are adapted to feeding on relatively large, soft prey, such as mucilaginous zooplankton and fish larvae, but as phyllosomata develop they become better equipped to ingest more fibrous prey items, such as mussel flesh and small crustaceans (Nishida et al., 1990; Mikami et al., 1994; Macmillan et al., 1997; Johnston and Ritar, 2001; Cox and Johnston, 2003a). The pereopods effectively shred zooplankton of any hardness into smaller more manageable sizes and the mandibles crush and masticate prey prior to ingestion (Mitchell, 1971; Kittaka, 1994; Abranhosa and Kittaka, 1997; Johnston and Ritar, 2001; Nelson et al., 2002; Cox and Bruce, 2003; Cox and Johnston, 2003a, 2004). Prey encounters of palinurid phyllosomata are thought to occur through chance encounters, 11

24 Chapter 1 facilitated through a tumbling, swimming and grasping feeding behaviour (Nelson et al., 2002; Cox and Bruce, 2003). This form of feeding behaviour suggests that a high density of food particles may be required in suspension at all times. The digestive system of palinurid phyllosomata is a simple tube, comprising an oesophagus and proto-proventriculus, midgut and branching digestive gland diverticula (tubules) and hindgut. Within the first few developmental stages a filter press is absent and a gastric mill is lacking throughout this life history stage (Wolfe and Felgenhauer, 1991; Macmillan et al., 1997; Kittaka, 1999; Johnston and Ritar, 2001; Cox and Johnston, 2004). Digestion is conducted by enzymes released from the digestive glands (Higgins, 2002; Johnston et al., 2004a, 2004b). Unlike penaeid prawns, which have anterior midgut diverticula along with a small digestive gland, the digestive capabilities of spiny lobsters, clawed lobsters and caridean shrimps may be limited during the larval stages (Biesiot and Capuzzo, 1990; Kamarudin et al., 1994; Lemos et al., 1999; Hammer et al., 2000; Le Vay et al., 2001). However, an increase in volume of the digestive gland during development, facilitated by an increase in number of tubules and a longer retention time of processed prey within the digestive tract may increase their digestive capability (Lemmens, 1994; Higgins, 2002) Formulated Diets and Their Potential Role in Palinurid Aquaculture The requirement for live feeds such as Artemia nauplii and mussel gonad is considered to be a limiting factor in the commercial culture of larvae of many fish and crustacean species, including spiny lobsters (Nelson et al., 2005). Although Artemia nauplii and fresh feeds such as mussel gonad and fish larvae have proven successful for rearing phyllosomata of many species through to puerulus, inherent problems still remain, including variable nutrient composition and availability, high costs of labour and infrastructure and potential introduction of pathogens into the culture system (Ohs et al., 1998). The development of a formulated diet that is palatable and delivers the correct balance of nutrients is an attractive and cost effective alternative to Artemia nauplii and fresh feeds (Kovalenko et al., 2002). Numerous attempts have been made to develop formulated diets that effectively replace live food for larval fish and crustacean species, but most have been unsuccessful (Teshima et al., 2000; Langdon, 2003). To date, the bulk of diet development and nutrition research on decapods has concentrated on penaeid prawns, primarily Penaeus monodon, P. japonicus, clawed lobsters Homarus americanus and H. gammarus, and the caridean shrimp 12

25 Chapter 1 Macrobrachium rosenbergii. Formulated diets are used routinely in hatchery culture of many penaeid prawn species; however, such success has not been accomplished for caridean shrimp and lobsters despite numerous attempts (Farmanfarmaian et al., 1982; Harpaz et al., 1987; Kurmaly et al., 1990; Ohs et al., 1998; Floreto et al., 2001; Felix and Sudharsan, 2004; Teshima et al., 2006). Recently, a semi-moist, purified diet has been established for M. rosenbergii, replacing the requirement for Artemia after the fifth larval stage (Kovalenko et al., 2002). Nutrients contained within formulated diets typically are delivered to marine larvae in six forms (Table 1.2). The past success achieved, incorporating formulated diets for the culture of larvae of penaeid prawns but not larvae of caridean shrimp and clawed lobsters, is attributed to the initial passive filter feeding of penaeid prawns in the early proto-zoeal stages (D'Abramo, 2002; Kovalenko et al., 2002). This form of feeding allows formulated diets with low digestibility to be developed as a low assimilation of nutrients is compensated by the relatively large amount of diet continuously passing through the digestive tract. In contrast, larvae of caridean shrimp and lobsters are active raptorial feeders from the onset of hatch, using their mouthparts and foregut to manipulate food items externally and internally (Farmer, 1974; Factor, 1978; Hinton and Corey, 1979; Johnston and Ritar, 2001; Nelson et al., 2002; Cox and Johnston, 2003a). Raptorial-feeding larvae also discriminate in their selection of food (Kurmaly et al., 1990; Kovalenko et al., 2002). This form of feeding, in conjunction with a comparatively long retention time of processed food items within the digestive tract, suggests that formulated diets for caridean shrimp and lobsters must be highly digestible (Teshima and Kanazawa, 1983; Teshima et al., 2000; D'Abramo, 2002). Different feeding mechanisms between lobsters and penaeid prawns impose additional problems for development of a successful formulated diet. As lobsters are slow, intermittent feeders, using their mouthparts to capture and manipulate prey, diets need to have a firm texture and a high water resistance to nutrient leaching (Jussila and Evans, 1998; D'Abramo, 2002; Ruscoe et al., 2005). These issues have been ignored largely in the development of formulated diets for penaeid prawns, predominately due to different feeding and digestive strategies through their larval phase. Despite differences in larval feeding mechanisms the nutritional requirements of larval penaeid prawns and larval lobsters are generally thought to be similar (Nelson et al., 2005). Ultimately, commercialisation of aquaculture of spiny lobsters will greatly depend on the development of a formulated diet that is palatable, digestible and assimilated at rates equivalent at least to the live feed Artemia. There are many 13

26 Chapter 1 complexities inherent in supplying a complete nutritional package in the form of a formulated diet (Table 1.2). A thorough understanding of the ingestive, digestive, physiological and feeding habits of palinurid phyllosomata will be pivotal to efficient development of formulated diet(s). Panulirus ornatus is the prime candidate for aquaculture within Australia due to this species many favourable attributes; hence, my focus on this species in my thesis. 1.7 Thesis Structure My thesis research utilises modern morphological techniques to explore the structure/function relationship of ingestive and digestive processes of phyllosomata of P. ornatus throughout ontogeny. This information was applied to develop a formulated diet to either totally replace or reduce the demand for culturing Artemia and use of fresh feeds such as mussel gonad. My thesis comprises five data chapters that have been formatted recently to be submitted for publication, although at the time of submitting this thesis, only Chapters 2, 4, 6 and Appendix 3 are formally in press. In Chapter 2, I describe the morphology of the mouthparts and foregut, and identify key structural changes that occur during development and their possible functional significance in relation to how prey items are likely to be captured, ingested and digested in the wild, and how this information can be applied to develop a suitable formulated diet, including appropriate physical characteristics of a diet such as size, form (texture) and buoyancy. In Chapter 3, I describe the structure of the digestive gland in stage I P. ornatus phyllosomata and quantitatively determined during the first four developmental stages the types and concentrations of digestive enzymes (proteases, lipases and carbohydrases). An understanding of digestion and determination of the types and concentrations of digestive enzymes during development is critical to design of a formulated diet(s), as they provide information pertaining to likely dietary shifts and nutritional components (proteins, carbohydrates and lipids) that should be incorporated in formulated diets. In Chapter 4, I discuss the feeding behaviour of phyllosomata, using Panulirus cygnus. Phyllosomata are not easy to collect, especially >3000 km from the source, so when a large enough sample of P. cygnus were made available I took the opportunity to assess the suitability of different forms of formulated diets by assessing the role of the mouthparts in the ingestive process by digital video analysis. Stability of potential diet formulations with a range of binders is also assessed over three time periods to determine an optimum binder for later diet preference and grow-out trials. These results form part of my 14

27 Chapter 1 thesis, namely Chapter 4. The influence of diet form, particle size, feeding stimulants and the level of protein was assessed by diet preference trials in Chapter 5. A rare earth metal, ytterbium oxide was incorporated into each diet formulation and diet preferences were assessed over a 4 h period using inductively coupled plasma-mass spectrophotometry (ICP-MS). In Chapter 6, I discuss the efficacy of partially replacing Artemia nauplii and on-grown Artemia with flesh of greenshelltm mussels and a formulated diet. Co-feeding during weaning is also explored, whereby the Artemia ration was successively reduced over 4 7 days and replaced totally with a formulated diet. In Chapter 7, I provide an overview of the major findings and their significance to design of formulated diet(s), and how prey items are externally and internally processed and digested, along with suggestions for future work. 15

28 Chapter 1 16

29 Chapter 2 Chapter 2 Ontogenetic Changes in the Structure and Function of the Mouthparts and Foregut of Early- and Late-Stage Panulirus ornatus Phyllosomata Introduction The Queensland population of the ornate spiny lobster P. ornatus (Fabricius, 1798) inhabits coastal waters of north-eastern Australia and Papua New Guinea (PNG) where it is a dominant element of the tropical reef community (Pitcher et al., 1997). Unlike Australia s temperate spiny lobster species, such as P. cygnus (larval phase in excess of 9 months) and J. edwardsii (larval phase in excess of 12 months), P. ornatus has a relatively short planktotrophic larval phase (4 6 months) in which 11 phyllosoma stages (>20 instars) precede transformation into a puerulus (Dennis et al., 1997; Pitcher et al., 1997). Panulirus ornatus also has a fast growth rate, attaining 1 kg within 2 years post-hatch (Phillips et al., 1992; Butler and Hernkind, 2000; Smith et al., 2003; Barclay et al., 2006). These favourable attributes have created considerable recent interest in culturing P. ornatus through to marketable size from the egg stage. Currently, limited knowledge of the nutritional requirements and ingestive capabilities of P. ornatus phyllosomata stands as the major impediment to their successful aquaculture; in the past this limited understanding has led to the provision of unsuitable diets, resulting in high mortalities, particularly in the mid-stages of development of this species (C. Jones, personal communication, Queensland Department of Primary Industries). Ontogeny of the mouthparts and foregut provides useful information about potential prey items, processing of prey and digestive function, which in turn may aid in developing a successful formulated diet (Nishida et al., 1990; Johnston and Ritar, 2001). The mouthpart and digestive tract morphology of adult and juvenile spiny lobsters has been extensively documented (Patwardhan, 1935; Paterson, 1968; Maynard and Dando, 1974; Wolfe and Felgenhauer, 1991; Mikami and Takashima, 1994). However, less attention has been given to ontogeny of phyllosomata, with most descriptions being taxonomic/diagnostic accounts that do not highlight the developmental changes in the fine structure of the mouthparts and foregut. Attempts have been made to correlate mouthpart and foregut structure with function during 2 Presented as a paper: Johnston, M., Johnston, D., Knott, B (in press) Ontogenetic changes in the structure and function of the mouthparts and foregut of early and late stage Panulirus ornatus (Fabricius, 1798) phyllosomata (Decapoda: Palinuridae). Journal of Crustacean Biology

30 Chapter 2 ontogeny for all phyllosoma stages of P. argus (Wolfe and Felgenhauer, 1991) and J. edwardsii (Johnston and Ritar, 2001) only. Other studies of mouthpart and foregut morphology are based on single or the first few phyllosoma stages (Nishida et al., 1990; Lemmens and Knott, 1994; Mikami et al., 1994; Cox and Johnston, 2003a, 2004). Structural characteristics of the mouthpart and foregut morphology of early stage (I III) P. argus (Wolfe and Felgenhauer, 1991), J. edwardsii (Johnston and Ritar, 2001) and J. (Sagmariasus) verreauxi (Cox and Johnston, 2003a, 2004) phyllosoma suggest a preference for ingesting and digesting soft-bodied prey items such as mucilaginous zooplankton that are easily gathered by the mouthparts and internally masticated by the foregut (Wolfe and Felgenhauer, 1991; Mikami et al., 1994; Macmillan et al., 1997; Johnston and Ritar, 2001; Cox and Johnston, 2004). However, mid- and late-stage phyllosomata are likely to ingest and digest larger prey items such as fish larvae and small crustaceans as indicated by the increased setation and spination of the mouthparts, and also an increase in the number of main brushes and ampullary channels (filter channels) within the foregut (Johnston and Ritar, 2001; Cox and Johnston, 2004). The recent closure of the life cycle of this species by the MG Kailis Group 3 has increased interest in commercial aquaculture of this species. However, as hatcheries currently rely on live feeds such as Artemia, which increase production costs of producing phyllosomata, the development of a formulated diet to reduce the demand for Artemia or to be used as a supplement is recognised as a high priority. A formulated diet for aquaculture of P. ornatus phyllosomata can be developed more efficiently when the ingestive and digestive morphology and their feeding behaviour is fully understood. Formulated diets can be classified broadly into three categories: 1) microencapsulated, 2) microcoated, and 3) microbound (Tucker, 1998). The latter is most suited to the raptorial feeding behaviour of P. ornatus phyllosomata, and has proved to be the most successful diet type for other larval crustaceans that have similar feeding behaviour (Kurmaly et al., 1990; Kovalenko et al., 2002; Genodepa et al., 2004a, 2004b). The success of microbound diets for raptorial feeding crustacean larvae is related to the ability to produce larger sized particles that can be manipulated by their mouthparts into adequate sizes for ingestion. Microcoated and microencapsulated diets have proved to be highly successful for species such as penaeid prawns that filter-feed during the zoeal stages, and for larval fish that are gulp

31 Chapter 2 feeders (Jones et al., 1979; Teshima et al., 1982; Teshima and Kanazawa, 1983; Jones et al., 1984; Langdon et al., 1985; Kanazawa and Teshima, 1988; Kanazawa, 1989; Kolkovski et al., 1997; Jones, 1998; Yúfera et al., 1999; Langdon, 2003). Microcoated and microencapsulated diets are typically spherical in shape, smaller in size than microbound diets, and are thought to be difficult to be initially gathered by the mouthparts of P. ornatus phyllosomata due to its lack of an inhalant feeding current. The aim of this study was to describe the structure of the mouthparts and foregut of early- (I VI) and late-stage (IX X) P. ornatus phyllosomata, to gain an understanding of their likely feeding and ingestive and digestive capabilities. Information from the current study will assist in identifying changes in ingestive capabilities and therefore possible dietary shifts. This information has direct implications for aquaculture of this species aiding in the development of an appropriate formulated diet. 2.2 Materials and Methods Brood Stock Collection and Handling of Phyllosomata Female and male P. ornatus were collected near Trinity Inlet (16 55 S., E) (northern Queensland, Australia) and transported to the Northern Fisheries Centre, Cairns where they were conditioned on a mixed diet of frozen greenshelltm mussel, P. canaliculus, pipis (Donnax spp.), scallops, P. fumatus and fresh squid (Nototodarus spp.), and allowed to mate. Ovigerous females were removed from the culture tank into individual incubation chambers (50 l) and held at a mean temperature of 26.0 ± 0.5 C and salinity of 36 g l 1. Newly hatched phyllosomata were skimmed from near the water surface of the incubation chambers and stocked into 20 l upwellers at a density of 4 phyllosoma l 1. The rearing tanks were provided with recirculating water (at an exchange rate of 10 l h 1 ) subject to mechanical filtration (to 1 μm) and UV and ozone treatment. During this period phyllosomata were fed ad libitum on a diet of ongrown Artemia (1.5 3 mm total length) reared on T. chuii and enriched with I. galbana (Tahitian strain). At each instar moult (Table 2.1), phyllosomata (n = 10) were removed from upwellers for examination of the mouthpart and foregut morphology. Phyllosoma stages were identified using light microscopy after the method of Duggan and McKinnon (2003) to stage VI; thereafter, phyllosomata were staged using a 19

32 Chapter 2 morphological key prepared by staff at the Queensland Department of Primary Industries (Northern Fisheries Research Centre, Cairns). Table 2.1. Instar moults and corresponding stages of phyllosoma development. Instars and associated stages used in this study (*). Data of hatchery reared Panulirus ornatus phyllosomata provided by Department of Primary Industries, Queensland. Phyllosoma stage Instar Inter-moult period (days) Total length (mm) Carapace length (mm) Carapace width (mm) I 1* II 2* III 3* * IV 5* * V 7* * VI VII VIII * IX 17* * * X 20* Morphology of the Mouthparts Stage I VI (instars 1 8) phyllosomata were fixed for 2 3 h in 2.5% glutaraldehyde in 0.1 M phosphate buffer ph 7.4. Stages IX X (instars 16 20) phyllosomata were fixed in 2% paraformaldehyde and 2.5% glutaraldehyde in 0.2 M sodium cacodylate, 0.2 M NaCl, and 15% (w/v) sucrose buffer ph 7.4, after Macmillan et al. (1997) and Johnston and Ritar (2001). After fixation, samples were rinsed in phosphate buffer (3 x 10 min), dehydrated in an ethanol series, critical point dried, carbon coated, and examined with a ZEISS VP FEGSEM in the Centre for Microscopy and Microanalysis, The University of Western Australia. Paired paragnaths of Stage I and II phyllosomata were removed using insect pins prior to critical point drying to reveal the morphological structure of the mandibles. Setal classification follows Garm (2004), and terminology of the mouthpart appendages is based on Wolfe and Felgenhauer (1991). 20

33 Chapter Morphology of the Foregut Phyllosomata of each stage were fixed and dehydrated as above. After removal of their pereiopods, phyllosomata were embedded in JB4 glycol methacrylate resin and sectioned serially (transverse) at 2 μm with a Sorvall microtome. The cephalic shields of stages IX and X (instars 17 20) were trimmed on either side of the foregut to the width of the glass knife for sectioning. Sections were stained with a polychrome stain and examined on an Olympus BX50 microscope. Terminology is confusing with respect to foregut structure, and where possible terminology in this study is consistent with previous detailed morphological studies of phyllosomata of spiny lobsters (Nishida et al., 1990; Wolfe and Felgenhauer, 1991; Johnston and Ritar, 2001; Cox and Johnston, 2004). 2.3 Results Morphology of the Mouthparts The mouthparts of P. ornatus phyllosomata are obvious at hatch, with the in situ positions remaining constant between each developmental stage. The oral field of P. ornatus phyllosoma comprises a single labrum, paired mandibles, paragnaths, maxillules, maxillae, and maxillipeds 1, 2 and 3 (Fig. 2.1A,B). Density and robustness of setation increased with each successive developmental stage, and the distance between the maxillule and maxillae and between each maxilliped (1, 2 and 3) increased in relation to total length (Table 2.2). The mouth aperture increased considerably from 3 μm at hatch to 52 μm in late stage phyllosoma, with a substantial shift at stage III (instar 3) from 3.5 μm to 9.9 μm at stage IV (instar 4). The most conspicuous morphological change occurs on the maxillae, with development of flattened exopodite (scaphognathite) and endopodite at stage IX (instar 16) (Fig. 2.1B) Labrum This structure, together with the paired paragnaths and mandibles, forms a semienclosed oral chamber; its posterior aboral surface has a series of sharp, anteriorly positioned denticles (Fig. 2.1C). Ventrally, the denticles are more robust and of a larger size, forming discrete rows; however laterally the denticles are smaller and 21

34 Chapter 2 arranged in clusters of three or four. Both size and density of the denticles increases during development. Table 2.2. Comparison of mouthpart dimensions during ontogeny of P. ornatus (mean ± S.E., n = 10). Total length, measured from anterior tip of cephalothorax to posterior tip of abdomen; Carapace width, measured across the size of the labral teeth; the distance between the third and second maxilliped (mxpd 3 and mxpd 2), the second and first maxilliped (mxpd 2 and mxpd1) and the maxillules and maxillas (mx1 and mx2); mouth field, the distance measured across widest lateral edges of the mandibles; oral field, distance between mandible and maxilliped 3; mouth aperture, lateral distance between the paragnaths. Instar Total Length Distance Between Mouth Field Oral Field Mouth Aperture (mm) Mxpd 3 and Mxpd 2 (mm) Mxpd 2 and Mxpd 1 (μm) Mx2 and Mx1 (μm) (mm) (mm) (μm) ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± * 6.39 ± ± ± ± ± ± ± * 9.15 ± ± ± ± ± ± ± * ± ± ± ± ± ± ± * ± ± ± ± ± ± ± * ± ± ± ± ± ± ± 2.70 * = wild-caught phyllosomata Mandibles The mandible is asymmetrical and lack a mandibular palp (Fig. 2.1D). The gnathobase is heavily chitinised. The mandibles consist of a toothed incisor process and a well developed molar process connected by a curved spine row (Fig. 2.1D). The right mandible has four incisor teeth and the left has three. The number of spines which form the spine row increases during development. The morphology of the rows of teeth that form the molar process is similar between left and right (Fig. 2.1E). A row of sharp teeth are positioned around the perimeter of each molar process. Positioned posteriorly on the left mandible is a toothed projection [toothed recess after Wolfe and Felgenhauer (1991)] and anteriorly on the right molar process is a toothed row (Fig. 2.1E). 22

35 Chapter 2 Fig Scanning electron micrographs of the mouthparts of P. ornatus phyllosoma. A, Oral region (Instar 1) showing spatial relationships between mouthparts. Scale, 30 μm. B, Oral region (Instar 18) showing spatial relationship between mouthparts. Scale, μm. C, Aboral surface of labrum and medial margin of paired paragnaths (Instar 1). Scale, 2 μm. D, Gross structure of the right mandible (Instar 2). Scale, 10 μm. E, Right and left molar processes showing toothed-recess and row (Instar 1). Scale, 12 μm. F, Medial margin of paired paragnaths (Instar 1) showing pores and teeth. Arrows indicate positioning of the 6 small, cuticular pores. Scale, 4 μm. Inset, small pore located on aboral surface of paragnath. Scale, 400 μm. IP = incisor process; L = labrum; M = mandible; MP = molar process; Mxp1 = maxilliped 1; Mxp2 = maxilliped 2; Mxp3 = maxilliped 3; Mx1 = maxillule; Mx2 = maxilla; P = paragnath; SR = spine row; Tre = toothed recess; Tro = toothed row Paragnaths These features (Fig. 2.1A,B,F) possess three types of cuticular pore on both the oral and aboral surfaces. Large oval shaped pores (>1 μm) (Fig. 2.1F; inset) were confined to the aboral surface with the two smaller pore sizes (<800 nm) positioned medially. 23

36 Chapter 2 Dense clusters of pappose setae project into the oral cavity from the medial margin of the aboral surface. Large teeth-like projections also emerge from the medial margin of each paragnath, as well as six small cuticular pores (Fig. 2.1F). The robustness of the teeth-like projections and density of the pappose setae positioned on the medial margin increases during development Maxillules These limbs have a basal protopod with both anterior exopod and posterior endopod (Fig. 2.2A). The exopod bears stout, non-articulated spines having two setal rows (Fig. 2A; inset), and the endopod bears similar spines that articulate at their base. The number of exopod and endopod spines increases during development, from two (instars 1 8) to three (instars 16 20). The number of setae increases during development from seven along each side at stage I (instar 1) to sixteen by stage X (instar 20). Three chemosensory-like simple setae which lack a terminal pore are also present on the aboral surface of the exopod (Fig. 2.2A) Maxillae These appendages consist of a flattened basal protopod and a flattened distal exopod (Fig. 2.2B). In early stage (instars 1 8) phyllosoma the basal protopod is larger than the distal exopod, with three sensory-like setae present on the medial margin (Fig. 2.2C). The number of pappose setae present on the lateral surface of each exopod increases during development, from four in stages I III (instars 1 3) to five at stage VI (instar 8). At instar 16 (stage IX) the distal exopod has expanded and is considerably larger than the protopod and a small endite is present on the anterior edge of each maxilla. By instar 18 (stage IX) the exopod has expanded to form a scaphognathite, which overlaps the thorax. The endite is larger and a short fringe of pappose setae extends from the lateral margin of each scaphognathite (Fig. 2.2D). 24

37 Chapter 2 Fig. 2. Scanning electron micrographs of the mouthparts of P. ornatus phyllosoma. A, Aboral view of left maxillule, showing gross structure of the exopod and endopod (Instar 4). Scale, 30 μm. Inset, fine detail structure exopod spines of the maxillules (Instar 1). Scale, 2 μm. B, Right maxilla (Instar 2). Scale, 2 μm. Inset. Fine structure of pappose seta. (Instar 3) Scale, 3 μm. C, Sensory-like setae on medial margin of the maxilla protopod. Arrows indicate positioning of setae. Scale, 6 μm. D, Right maxilla and first maxilliped (Instar 20). Scale, 100 μm. E, Right first maxilliped (Instar 2). Scale, 4 μm. F, Right second maxilliped showing setal arrangement of the propodus and dactylus (Instar 5). Scale, 40 μm. Inset. Flagellate-like setae on lateral margin of the second maxilliped dactylus. Arrow indicates flagellate-like seta. Scale, 10 μm. G, Left third maxilliped showing distal setal arrangement of propodus and dactylus (Instar 1). Arrows indicate positioning of setae on propodus. Scale, 30 μm. H, Fine structure of the distal region of the unusual serrate setae (Instar 5). Joined line indicates two setal rows. Scale, 4 μm. Inset. Position of sub-terminal pore on the distal region of the third maxilliped. Endo = endopod; En = endite; Exop = exopod; Ex = exite; Mxp1 = maxilliped 1; P = paragnath; Prot = protopod; Sc = scaphognathite. 25

38 Chapter First Maxillipeds These limbs are rudimentary in early stage (instars 1 8) phyllosomata, with a short distal simple seta (Fig. 2.2E). By stage IX (instars 16 19) an endopod has formed, and by stage X (instar 20) the endopod is larger with a small proximal exite (Fig. 2.2D) Second Maxillipeds The second maxillipeds have a 5-segmented endopod (Fig. 2.1B). The propodus has six robust serrate setae, two elongated serrate setae opposing the distal setae of the dactylus, and four proximal serrate setae (Fig. 2.2F). The dactylus is stout, having three distal cuspidate setae, all of which possess a sub-terminal pore. A flagellum-like simple seta (Watling, 1989; Felgenhauer, 1992) with sub-cuticular socket inserts on the lateral margin (Fig. 2.2F; inset). The distance between the first and second maxillipeds increases successively during each instar moult, with considerable increases between instars 3 4 and within instars (Table 2.2) Third Maxillipeds The entire integument of these appendages is covered with an array of stout pectinate denticles. In the 5-segmented endopod, the propodus is elongated with an irregular array of serrate setae positioned along its length and bears eight elongated serrate setae distally (Fig. 2.2G). All of the eight distal setae have three rows of densely packed denticles distal to the annulus; the irregularly spaced denticles become smaller distally before separating into two rows (Fig. 2.2H). On the apical tip of each seta is a prominent terminal pore (Fig. 2.2H; inset). The dactylus is stout, bearing eight serrate setae with similar setal morphology as described above. The distance between the second and third maxillipeds increases successively during development. A considerable shift in the disposition of the third and second maxillipeds occurs between stages III IV (instars 3 and 4) and again late in ontogeny from stages IX X (instars 19 and 20) (Table 2.2) Morphology of the Alimentary Tract The alimentary tract comprises three morphologically distinct regions: foregut including the oesophagus and the proto-proventriculus; midgut with branching digestive gland 26

39 Chapter 2 tubules; and hindgut. In all phyllosoma stages, the oesophagus is a short, chitinised tube, surrounded by large columnar epithelial cells and circular muscle. Short spines that increase in density during ontogeny project into the oesophageal lumen. The proto-proventriculus of all phyllosoma stages lacks a gastric mill and cardiopyloric valve, and consequently there is no clear distinction of cardiac (anterior) and pyloric (posterior) chambers, but an anterior chamber is defined in this early stage of development by the two dorsal grooves separated by a prominent dorsal ridge, and two ventral grooves separated by lateral in-foldings of the ventral wall forming the anterior floor. Setal density of the anterior floor and the number of robust lateral setae and main brushes that project medially into the lumen of the anterior foregut chamber increase during ontogeny, with a considerable increase between stages III IV (instars 3 4) (Table 2.3). The posterior limit of the comb row with the lateral folds in the anterior chamber delineates the start of the posterior chamber. The lateral folds divide the posterior foregut chamber into both dorsal and ventral chambers. After stage II (instar 2), the ventral chamber forms a filter press, with well-developed opposing setal rows on both the inner and outer valve. During development the width and number of ampullary channels (filter channels) increases from two at stage II (instar 2) to twelve by stage X (instar 20) (Table 2.3). The filter press opens posteriorly into the primary ducts of the digestive gland diverticula, and a complex array of dorsal, ventral and lateral setae project medially to form the pyloric-intestinal valve. The setae of the pyloric-intestinal valve become increasingly setose and more robust during development. Based primarily on changes of the anterior and posterior chambers of the foregut during each progressive instar moult, phyllosomata have been divided into three distinct ontogenetic groups: stages I III (instars 1 3); stages IV VI (instars 4 8); and stages IX X (instars 17 20) Stages I III (Instars 1 3) The anterior chamber of stage I III phyllosomata is simple, comprising two dorsal and ventral grooves, and one lateral seta and two three main brushes (Fig. 2.3A). The lateral setae and main brushes project medially into the lumen and are similar in total length (Table 2.3). The anterior floor comprises a thin mat of short setae which increases in density during development. No filter press is present in the posterior chamber in stages I II but is evident in stage III (instar 3) phyllosomata when it comprises four ampullary channels (Fig. 2.3B). The ampullary channels of stage III 27

40 Chapter 2 phyllosomata are not well formed and their maximum width is 5 μm (Table 2.3). Posteriorly, the ventral chamber opens into the primary ducts of the digestive gland at the junction of the fore- and mid-gut (Fig. 2.3C). Dorsally, a complex array of robust setae forms the pyloric-intestinal valve. The setae become more robust during development (Fig. 2.3C). Fig Transverse sections through the foregut of instar 1 4 phyllosomata. A, Anterior foregut (Instar 1). Scale, 100 μm. B, Posterior foregut showing initial development of a filter press, arrows indicate ampullary channels (Instar 3). Scale, 100 μm. C, Posterior foregut, attachment of the ventral channels with the ducts of the digestive gland, arrow indicates pyloric-intestinal valve (Instar 3). Scale, 100 μm. D, Anterior foregut, arrow indicates increased setal density of the anterior floor (Instar 4). Scale, 100 μm. AF = anterior floor; DC = dorsal chamber; DG = dorsal groove; DR = dorsal ridge; IV = inner valve; LS = lateral setae; MB = main brushes; OV = outer valve; VC = ventral chamber; VG = ventral groove. 28

41 Chapter Stages IV VI (Instars 4 8) The anterior chamber of stage IV VI (instars 4 8) phyllosomata is more developed, with four lateral setae and an increase in number of the main brushes, which terminate posteriorly at the formation of the lateral folds. A dense setal mat is evident on the anterior floor (Fig. 2.3D) and the maximum length of the lateral setae increases considerably from stages IV VI (instars 4 8) (Table 2.3). By stage IV (instar 5) the lateral setae have increased to 17 μm in length and overlap considerably with the dense setal mat of the anterior floor, which in combination create a more efficient filtration barrier to the ventral channels (Fig. 2.4A). The lateral folds are well developed and extend into the lumen creating distinct dorsal and ventral chambers of the posterior chamber. The medial margins of the lateral folds contain short fine setae that intermesh laterally (Fig. 2.4B). The filter press is more prominent with clearly defined ampullary channels, inner and outer valves and an apical crest (Fig. 2.4C). The number of ampullary channels increases from eight at instars 4 5 (stage IV) to ten by instars 6 8 (stages IV, V and VI) (Table 2.3). The ampullary channels are screened by a dense row of dorsally-directed setae. Posteriorly, the setal arrangement that forms the pyloric-intestinal valve is more robust and ventrally the ventral groove connects with the primary ducts of the digestive gland. Table 2.3. Comparison of anterior and posterior foregut structures during development of Panulirus ornatus (mean ± S.E., n=5). AC, ampullary channel number; maximum width of largest ampullary channel in the filter press; LS, lateral setae number, length; MB, main brushes number, length. Wild caught phyllosomata (*). Instar LS Number LS Length (μm) MB Number MB Length (μm) AC Number AC Width (μm) ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± * 4 70 ± ± ± * 4 79 ± ± ± * ± ± ± * ± ± ±

42 Chapter 2 Fig Transverse sections through the foregut of instars 5 8 and phyllosomata. A, Mid foregut showing position of main brushes and overlapping of the lateral setae with anterior floor (Instar 5). Scale, 100 μm. B, Mid- late-anterior foregut showing well-developed lateral folds (Instar 5). Scale, 100 μm. Posterior foregut showing detail of the filter press. (Instar 5). Scale, 100 μm. D, Posterior foregut showing detail of the inner and outer valve and dorsally directed setae of the ampullary channels (Instar 19). Scale, 100 μm. E, Posterior foregut showing increase in number of ampullary channels (Instar 20). Scale, 100 μm. F, Junction between foregut and midgut showing position of the attachment with the primary duct and setal arrangement of the pyloric-intestinal valve. AC = ampullary channel; AF = anterior floor; DC = dorsal chamber; DG = dorsal groove; IV = inner valve; LF = lateral fold; LS = lateral setae; MB = main brushes; OV = outer valve; PD = primary duct; PIV = pyloric-intestinal valve; VC = ventral chamber. 30

43 Chapter Stages IX X (Instars 17 20) By the late stages the anterior chamber is completely developed with a row of robust cuticularised spines comprising the main brushes, which increase in both size and number between instars (Table 2.3). The lateral setae increase in relation to total length and overlap with the dense setal mat of the anterior floor, creating additional filtration to the ventral channels. The posterior chamber undergoes considerable morphological change during the final moults to puerulus. The number of ampullary channels increases from five (instar 17) to eight (instar 18) to twelve by instars 19 and 20 (Fig. 2.4D,E; Table 2.3). The density and robustness of the dorsally directed setae per ampullary channel and setae positioned on the outer valve are robust and more numerous than earlier stages. Posteriorly, the filter press opens into the primary ducts of the digestive gland and the setal arrangement which comprise the pyloric-intestinal valve are extremely spinose and robust (Fig. 2.4F). 2.4 Discussion Morphology of the Mouthparts The gross morphology of the mouthparts is similar to other Panulirus spp., (Johnson and Knight, 1966; Wolfe and Felgenhauer, 1991; Matsuda and Yamakawa, 2000) and Jasu spp., (Kittaka and Abrunhosa, 1997; Macmillan et al., 1997; Nelson et al., 2002; Cox and Bruce, 2003; Cox and Johnston, 2003a) phyllosomata. Mouthpart morphology changes little during development, suggesting both early- and late-stage P. ornatus phyllosomata ingest similar prey items and that external mastication is well developed from stage I. However, increasing setation of the mouthparts, mouth- and oral-fields, mouth aperture, and disposition between the mouthparts from stage IV indicates that the size of prey handled and ingested becomes considerably larger during development. Similar observations were reported for J. edwardsii (Johnston and Ritar, 2001) and J. (Sagmariasus) verreauxi (Cox and Johnston, 2003a) phyllosomata. Sharp spinose setae on the propodi and dactyli of the third and second maxillipeds, maxillules, and rows of sharp teeth forming the molar process suggest that the mouthparts of P. ornatus are well adapted for grasping, manipulating and masticating a wide variety of zooplankton in the Coral Sea such as calanoid and cyclopoid copepods, cirriped larvae, ectoprocts, and mysid shrimps. 31

44 Chapter 2 Prey capture and manipulation would initially be facilitated by the third and second maxillipeds (Johnston and Ritar, 2001; Nelson et al., 2002; Cox and Johnston, 2003a, 2003b). The complex array of distally positioned serrate and cuspidate setae on the propodi and dactyli of the third and second maxillipeds would aid in the anterior transfer of food toward the oral cavity (Cox and Johnston, 2003b). These setae are present at stage I and are long enough on the medial and lateral margins of the propodi and dactyli to facilitate prey entrapment (Nelson et al., 2002). Tthe second and third maxillipeds and mandibular molars are well developed at stage I and their disposition (spacing between each mouthpart) increases during development, suggests that P. ornatus are not limited to prey of a particular hardness, but they are only limited by their ability to capture prey. Hence, the types of prey ingested are not likely to change abruptly during development [such as a shift from mucilaginous zooplankton to fleshy/muscular prey items during development (Johnston and Ritar, 2001; Cox and Johnston, 2003a, 2004)], but the limbs are thought to increase in size, with a greater ability to hold larger prey close to the maxillules for further manipulation and ingestion. Similar suggestions based on feeding observations of J. (Sagmariasus) verreauxi phyllosomata was provided by Nelson et al. (2002), who suggested based on feeding evidence using mussel gonad and particles of jellyfish that mucilaginous zooplankton (jellyfish, medusae, ctenophores) are an unsuitable food source, fouling the mouthpart setae, and are likely to suffice during opportunistic feeding events only. Furthermore, mucilaginous zooplankton are of low nutritional value, consisting mostly of water and low percentage of total lipid (Nelson et al., 2000). Captured prey from the third and second maxillipeds would be held close to the oral cavity by the second maxillipeds and endopod and exopod spines of the maxillules. The endopod and exopod spines of the maxillules are robust and the number of setae along each margin of these spines increases during development, which would facilitate the holding of larger prey items and increase the effectiveness of external mastication. Increased setation of the maxillules and an ability to hold larger sized prey close to the oral field has also been suggested for J. edwardsii (Johnston and Ritar, 2001) and J. (Sagmariasus) verreauxi (Cox and Johnston, 2003a) phyllosomata. Manipulated prey would then be passed toward the incisor and molar processes of the mandibles for mastication. The teeth-like projections of the molar process are densely packed and are sharp on both opposing sides, suggesting that quite hard prey items such as copepods could be masticated effectively into sizes suitable for ingestion. Denticles on the labrum and teeth-like projections on the medial margin of each paragnath would hold manipulated particles of prey close to the oral cavity. 32

45 Chapter 2 Masticated prey particles passing into the oral cavity are likely to be well lubricated easing swallowing as a suite of cuticular pores on both the oral and aboral surface of the paragnaths were observed Morphology of the Foregut The morphology of the foregut changes considerably during development and differs significantly from that of adult palinurids. The key differences are that the protoproventriculus of all phyllosoma stages of P. ornatus lacks a gastric mill and cardiopyloric valve, and consequently there is no clear distinction of cardiac and pyloric stomachs. A filter press is also absent in phyllosoma stages I and II. The absence of a gastric mill, but presence of robust lateral setae and well developed main brushes at stage I suggests a limited capacity to triturate prey internally. Furthermore, as the number, length and robustness of the main brushes increases at stage IV, indicates an increased efficiency to triturate prey internally. However, the efficiency of the main brushes and lateral setae to triturate quite hard items is thought to be limited to small crustaceans such as copepods and is not thought to be as effective as the gastric mill in juveniles and adults that enables them to crush extremely hard prey items such as ingested bivalve shell. Ingested prey items would be reduced to a fine slurry, and sorted and filtered for final absorption of nutrients in the digestive glands. In early developmental stages (I II), the foregut is a simple chamber, with one lateral and two three main brushes, and a filter press is absent also. These features suggest that P. ornatus phyllosomata have limited ability to triturate and filter ingested particles of prey internally during these initial two developmental stages, hence ingested prey particles are likely to be masticated extensively by the molar process and ground into a fine slurry to facilitate ingestion. In stages III VI, ability to triturate ingested particles of prey to some capacity internally most likely increases, with an increase in number and length of the main brushes and an increased ability to filter particles due to the development of a filter press (stage III) with four ampullary (filter) channels on either side of the inner valve. The maximum width of the ampullary channels increases from the final instar moult of stage IV (instar 6) to stage VI from μm indicating that sizes of particles capable of assimilation in the digestive glands would increase during development. In late-stage phyllosomata (IX X) there is no further increase in number of main brushes or maximum width of the ampullary channels. The length of the main brushes increases between stages IX (instar 17) and stage X (instar 20) from

46 Chapter 2 μm and there is a considerable increase in number of ampullary channels from 5 (instar 17) to 12 (instar 20) indicating an increased ability to triturate prey to a limited capacity in the foregut and increased capacity to filter ingested prey for assimilation of nutrients in the digestive glands. This increased capacity to filter larger amounts of ingested prey in the final stages of development is likely to be associated with the accumulation of energy reserves in the form of lipid which are stored in the digestive glands as a source of energy for the metamorphosis to necktonic pueruli (Jeffs et al., 1999) Implications for Diet Development At stage IV there are key morphological changes of both the mouthparts and foregut that occur simultaneously (increase in size of the mouth aperture, lateral setae and main brushes and presence of a well-formed filter press with four ampullary channels on each side), suggesting an increased ingestive and digestive capacity. This increase in ingestive and digestive capacity, from stage IV onwards, indicates that this would be the most appropriate period to wean from live food (Artemia) and introduce formulated diets. Diet texture will be critical to development of a successful formulated diet for this species given that the complex array of setae on the endopod and exopod of the maxillipeds and endopod and exopod of the maxillules would foul easily with a too soft or not adequately bound diet. Hence, a formulated diet should consist of a texture that is quite hard/firm but still penetrable by the dactyl spines of the second and third maxillipeds. Jelly-like diets as proposed by Johnston and Ritar (2001) and Cox and Johnston (2003a) for J. edwardsii and J. (Sagmariasus) verreauxi phyllosomata are unlikely to be successful for aquaculture of P. ornatus phyllosomata due to this species tropical habit. Jelly-like diets are commonly high in moisture and bound using binders such as gelatine that soften quickly at temperatures >25 C, leading to increased particle abrasion and mouthpart fouling. During development of P. ornatus phyllosomata the mouth- and oral-fields and mouth aperture increases. Setation and spination of the mouthparts, especially those close to the oral field such as the maxillules, paragnath and labrum, also increase, which would aid in retaining larger-sized prey close to the oral cavity. Furthermore, an increase in number of main brushes, lateral setae, and number and width of ampullary channels from stage IV onwards indicates a capacity to digest larger-sized portions of prey. Hence, sizes of formulated diets are likely to change during development in 34

47 Chapter 2 coordination with key developmental changes of the mouthparts and foregut. Optimum diet size for phyllosomata is currently unknown. However, as P. ornatus phyllosomata readily accept Artemia nauplii and on-grown Artemia from day 0, diet particles <428 μm (minimum size of Artemia nauplii) are thought to be too small to be captured and manipulated by the mouthparts due to the lack of an inhalant feeding current (Genodepa et al., 2004a). In the experiments described in Chapter 5 more diet particles within the μm size range was consumed over the 4 h time period. As there is not a considerable increase in size of the oral- and mouth-fields and mouth aperture until stage IV, this size range of diet particles may be appropriate for the first three initial developmental stages. However, after stage IV based on a mouth aperture >9.9 μm, diet particles >700 μm would be more suitable, most likely ranging in size between μm. Particle sizes should gradually increase during phyllosomata development coinciding with the increasing disposition of the mouthparts but should not exceed μm, as zooplankton larger than this is unlikely in their natural environment. Diet buoyancy will be another important issue for development of a formulated diet, as P. ornatus are planktonic feeders and rarely feed at the bottom of tanks. Keeping diet particles in suspension to create movement is typically achieved by using aeration; however, additional aeration is rarely used when culturing phyllosomata as the water exchange is often quite high, but can be used to disperse both phyllosomata and feeds. Furthermore, in feeding trials using combinations of Artemia, fresh feeds and a formulated diet (Chapter 6), only those combinations whereby Artemia were included were successful, indicating that P. ornatus phyllosomata are stimulated to feed by visual cues. Hence, if a formulated diet cannot be developed to be neutrally buoyant with adequate proportions of oils, an appropriate tank design that facilitates the continuous movement of diet particles or frequent re-suspension by upwelling may be more efficient and increase the feed intake of the formulated diet. Future work is needed to explore the use of other forms of microbound diets such as complex microparticles for P. ornatus phyllosomata using both replicated feeding trials and feeding response observations. Furthermore, tank designs that promote the frequent re-suspension of diet particles or diet formulations which enable diet particles to be neutrally buoyant to stimulate feeding by visual cues will serve to optimise feed intake. 35

48 36

49 Chapter 3 Chapter 3 Structure and Function of the Digestive Gland and Ontogenetic Changes in Enzyme Activity of Panulirus ornatus phyllosomata 3.1 Introduction The larval development of palinurid phyllosomata is characterised by an extended phyllosoma phase that includes considerable changes in anatomy, physiology and behaviour. The ornate spiny lobster, P. ornatus is a tropical species with an Indo-west Pacific distribution (Pitcher et al., 1997). This species has a relatively short phyllosoma phase (4 6 months) and fast rate of growth, attaining 1 kg after 18 months postsettlement (Phillips et al., 1992; Butler and Hernkind, 2000; Smith et al., 2003; Barclay et al., 2006). Due to these favourable attributes, there has been increasing interest in the last decade for rearing this species under intensive aquaculture conditions in Australia. Recently the MG Kailis Group 4 reared this species from egg through to puerulus using live and fresh feeds. Complete closure of the life cycle has also been achieved recently for two other Panulirus spp., P. argus (Goldstein et al., 2006) and P. penicillatus (Matsuda et al., 2006). Recent success in closure of the life cycle of these three species has increased interest in developing methods for intensive aquaculture of phyllosomata of spiny lobsters. However, despite the success of these recent break-throughs, survival through to the benthic puerulus stage is still quite low and currently not at a stage of commercial viability (>65%). Lack of knowledge surrounding the wild diet has been a major impediment to successful hatchery culture of spiny lobsters, with unsuitable prey items and the introduction of disease being blamed for the consistent occurrence of high mortalities (Phillips and Sastry, 1980; Kittaka, 1994, 1997; Johnston and Ritar, 2001; Cox and Johnston, 2003b; Bourne et al., 2004). A range of approaches to identify the wild diet have had limited success; the approaches have included natural prey-choice experiments (Mitchell, 1971), laboratory trials (Nelson et al., 2002; Ritar et al., 2002), lipid signature profiles (Phleger et al., 2001), gut content analysis (Phillips and Sastry, 1980), and ingestive and digestive morphology (Macmillan et al., 1997; Johnston and Ritar, 2001; Nelson et al., 2002; Cox and Johnston, 2003a, 2004). Assessment of the appearance and activity of digestive enzymes is an effective approach to further the understanding of the

50 Chapter 3 digestive physiology and nutritional requirements of spiny lobster phyllosoma (Johnston et al., 2004a, 2004b). Furthermore, digestive physiology of crustaceans is a reflection of prey choice and feeding strategies, and can be used to indicate both feeding transitions during development and as an indicator of the type and level of the main nutrients (protein, carbohydrate and lipid) to include in a formulated diet (Martinez et al., 1999; Hammer et al., 2000; Johnston and Johnston, 2005). The digestive gland is the largest and most complex organ associated with the alimentary tract (Factor, 1995), and is involved in a diverse range of metabolic activities, including the synthesis and secretion of digestive enzymes, as well as the absorption of nutrients (Gibson and Barker, 1979; Dall and Moriarty, 1983; Lovett and Felder, 1989; Icely and Nott, 1992; Factor, 1995; Hammer et al., 2000; Johnston et al., 2004a, 2004b). In phyllosomata of spiny lobsters the digestive gland connects to the ventro-posterior region of the proto-proventriculus via the primary ducts that branch into the tubules that make up the digestive gland (Mikami et al., 1994; Johnston and Ritar, 2001; Higgins, 2002). The digestive gland tubules comprise of epithelial cells that differentiate into four cell types: E- (embryonic), R- (resorptive), B- (blister-like), and F- (fibrillar) according to the scheme of Jacobs (1928). E-cells arise via mitotic division in the distal tip of the tubules and then differentiate into the three other cell types, R-, B-, and F-cells as they migrate proximally (Loizzi, 1971; Johnston et al., 1998; Higgins, 2002). Cell function of the R-, B- and F-cells remains a source of debate, due in part to early histological misinterpretation and rapid proteolytic autolysis after death (Johnston et al., 1998). R-cells are now recognised to be responsible for the storage of lipid and glycogen (Icely and Nott, 1992; Mikami et al., 1994; Johnston et al., 1998; Higgins, 2002), and F-cells are involved in the synthesis and secretion of digestive enzymes (Dall and Moriarty, 1983; Mikami et al., 1994; Johnston et al., 1998; Higgins, 2002). The function of B-cells is not entirely clear; however, it is generally thought that they absorb fluid and small particles through pinocytosis (Mikami et al., 1994; Johnston et al., 1998; Higgins, 2002). The development of digestive function of phyllosomata of spiny lobsters has received little attention (Mikami et al., 1994; Higgins, 2002; Johnston et al., 2004a, 2004b). Higgins (2002) and Macmillan et al. (1997) reported an increase in number of digestive gland tubules during development of J. edwardsii phyllosomata indicating an increased efficiency to absorb nutrients. Higgins (2002) and Mikami et al. (1994) observed both R-, F- and B-cells from day 0 in J. edwardsii and P. japonicus, indicating that phyllosomata have the ability to digest, and presumably feed from this point. 38

51 Chapter 3 Furthermore, due to an abundance of R-cells in the digestive gland tubules, the final phase of absorption of nutrients is thought to occur intracellularly within the R-cells of the digestive gland, where digested nutrients including amino acids, simple sugars, and lipids are absorbed (Mikami et al., 1994). Knowledge of the activity and concentration of digestive enzymes of phyllosomata of spiny lobsters is restricted to J. edwardsii only (Johnston et al., 2004b). Johnston et al. (2004b) detected a range of proteases, carbohydrases and lipases in both hatchery reared and wild-caught phyllosomata, suggesting that J. edwardsii phyllosomata can readily digest protein, carbohydrate and lipid at all stages of development. Furthermore, protease and lipase activities were considerably higher than carbohydrases, indicating a carnivorous diet and that phyllosomata make best use of dietary items that are high in protein and lipid, such as zooplankton (Le Vay et al., 2001; Johnston et al., 2004b). This study investigated the structure of the digestive gland of day 0, stage I P. ornatus phyllosomata and the types and concentrations of digestive enzymes of cultured phyllosomata (days 0 32) fed a diet of known nutritional composition (ongrown Artemia) to determine their digestive capacity, possible nutritional requirements and to distinguish the relative importance of protein, lipid and carbohydrate for development of a formulated diet. 3.2 Materials and Methods Brood Stock Collection and Handling of Phyllosomata Cultured P. ornatus phyllosomata were collected in June and July 2006 from a female held at the Northern Fisheries Research Centre, Queensland Department of Primary Industries, Cairns, Queensland. The female previously received a mixed diet consisting of frozen squid (Nototodarus sp.), greenshelltm mussels (P. canaliculus), pipis (Donnax sp.) and scallops (P. fumatus). The berried female was placed in an incubation chamber (50 l) 3 4 weeks prior to the release of phyllosomata, held at a temperature of 26 ± 1 C and disinfected with 15 ppm formaldehyde in sea water for 30 min. Newly hatched phyllosomata were skimmed from near the water surface and stocked into four 250 l up-welling pulse tanks. The up-welling pulse tanks were supplied with recirculated water at 26 C, filtered to 1 μm, and disinfected with both ultra-violet radiation and ozone. De-capsulated Artemia cysts (INVE) were hatched and cultured 39

52 Chapter 3 on T. chuii and enriched for 24 h with I. galbana (Tahitian strain) and fed daily to phyllosomata at a density of 4 individuals ml Histology and Ultrastructure of the Digestive gland For histological examination, stage I phyllosomata were fixed in 2.5% glutaraldehyde in 0.1 M phosphate buffer (ph 7.4) for 2 3 h and stage X phyllosomata were fixed in 2% paraformaldehyde and 2.5% glutaraldehyde in 0.2 M sodium cacodylate, 0.2 M NaCl, and 15% (w/v) sucrose buffer ph 7.4, after Macmillan et al. (1997) and Johnston and Ritar (2001). Following dehydration in graded ethanol solutions, phyllosomata were embedded in JB4 (Pro Sci Tech, Brisbane, Australia) glycol methacrylate resin and sectioned serially (transverse) at 2 μm with a Sorvall microtome. Sections were stained with a polychrome stain and amino black, and examined with an Olympus BX50 microscope. For ultrastructure examination (n=10), stage I phyllosomata were prepared using a Pelco Bio Wave Microwave with a cold-spot stage. Phyllosomata were fixed in 2.5% glutaraldehyde (ph 7.4) for 2 3 h, washed in 0.5 M phosphate buffer (ph 7.4) and post-fixed in 1% osmium tetraoxide in 0.5 M phosphate buffer (ph 7.4). Specimens were then dehydrated in a graded ethanol series, embedded in resin (Spurr, 1969) and sectioned with a Reichert Jung Ultracut E ultra-microtome using a diamond knife. Thin sections were stained with acidified, saturated uranyl actetate in 50% ethanol for 15 min, and with lead citrate for 15 min and examined with a JEOL 2000FX transmission electron microscope at 80 kv Enzyme Extraction Cultured phyllosomata were sampled in triplicate with each sample consisting of pooled animals ranging from 800, 300, 200, 150, and 120 individuals for days 0, 5, 15, 19 and 32, respectively. Artemia samples (n=3000) were also assayed for enzyme activity to compare with phyllosoma activity. Phyllosomata samples were homogenised for 5 min in 1 ml (2 ml for Artemia samples) of chilled 50 mm Tris, 10 mm CalCl 2, 20 mm NaCl buffer ph 7.5 using an electric Ultraturrax disperser, after Johnston et al. (2004b). The homogenate was centrifuged at rpm for 10 min at 4 o C and 200 μl aliquots of supernatant transferred to centrifuge tubes and stored at -20 o C. 40

53 Chapter Enzyme Assays One enzyme unit was defined as the amount of enzyme that catalysed the release of 1 μmole of product per min, and was calculated using the appropriate molar extinction coefficient (έ) in the assay conditions or a standard curve. Specific activity was defined as enzyme activity per mg of larval protein (Units mg 1 ) and total activity was defined as enzyme activity per larva (Units larva 1 ). Protein concentration was determined by the method of Bradford (1976) using bovine serum albumin (Sigma P0914) as the standard. Photometric enzyme assays (200 μl micro-assays) were performed in duplicate at 37 o C in IWAKI flat bottom microplates and absorbance read in a Bio Tek Elx 808 microplate reader (Bio Tek Instruments, INC., Winooski, Vermont). Tests confirmed the enzyme activities were linear with incubation. All chemicals were purchased from Sigma-Aldrich (Castle Hill, New South Wales) or MP Biomedicals (Seven Hills, New South Wales) Proteases Total protease (E.C ) activity was measured according to a modified method of Pavasovic et al. (2004) using azurine cross-linked casein (AZCL-casein; Megazyme, Ireland) as substrate. Briefly, 1000 μl of enzyme extract, diluted 1:50 in 100 mm sodium phosphate buffer ph 7.0, was added to 1000 μl of an AZCL-casein solution. The mixture was incubated at 37 o C for 60 min in a water bath. The reaction was stopped by placing the tubes on ice for 10 min. Following this, 700 μl of distilled water was added to the reaction mixture which was then centrifuged at rpm for 10 min at room temperature. The absorbance of the supernatant was read at 595 nm. One unit of total protease activity was calculated from a tyrosine standard curve that was generated by diluting aliquots of a tyrosine stock solution (0.25 mg ml 1 stock solution) with 100 mm HCl. Trypsin (E.C ) was assayed using N-α-benzoylarginine-ρ-nitroanalide (BAPNA) dissolved in dimethylformamide (DMF) as substrate. Each assay contained a final concentration of 1.25 mm BAPNA in 200 mm Tris, 200 mm NaCl, 10 mm CaCl 2 and 0.2% (w/v) polyethylene glycol 6000 ph 8.0. Assays were initiated by the addition of enzyme extract and the release of ρ-nitroanaline was measured at 405 nm. Under these assay conditions the molar extinction coefficient was 9300 M 1 cm 1 for ρ- nitroanaline (Stone et al., 1991). A positive control of 3 mg ml 1 porcine pancreas trypsin in 1 mm HCl was used. 41

54 Chapter Carbohydrases Amylase (E.C ) activity was determined using a modified method of Pavasovic et al. (2004) using blocked ρ-nitrophenyl maltoheptasoside (BPNPG-7) (Alpha-Amylase Kit, Megazyme) as substrate. Each assay (total 200 μl) contained 36 μl of BPNPG-7 (54 mg+10 ml distilled water) and 162 μl of 0.1 M Malic acid, 0.1 M NaCl, 2 mm CaCl 2.2H 2 O, 0.01% sodium azide ph 7.0 buffer. Assays were initiated by the addition of 2 μl enzyme extract and the release of nitrophenol was measured at 405 nm. Under these assay conditions the molar extinction coefficient was M 1 cm 1 for nitrophenol (Pavasovic et al., 2004). Chitinase (E.C ) activity was determined using the substrate ρ- nitrophenyl N-acetyl β-d glucosaminide. Each assay contained a final concentration of 4 mm substrate in 200 mm Tris, 200 mm NaCl, 10 mm CalCl 2, and 0.2% (w/v) polyethylene glycol 6000, ph 5.0. Assays were initiated with the addition of enzyme extract and the release of nitrophenol was measured at 405 nm. Under these assay conditions the molar extinction coefficient is M 1 cm 1 for ρ-nitrophenol at ph >9 (Erlanger et al., 1961) Lipase Lipase (neutral bile salt-dependent lipase activity; E.C ) activity was determined using a method modified from Albro et al. (1985) using ρ-nitrophenyl myristate (PNM) dissolved in ethanol as substrate. Each assay contained a final concentration of 0.4 mm PNM in 24 mm ammonium bicarbonate, 37.5 mm sodium chloride, ph 8.5, and 0.5% (w/v) Triton X-100 as an emulsifying agent. Assays were initiated by the addition of enzyme extract and the release of nitrophenol was measured at A 405. Under these assay conditions the molar extinction coefficient was M 1 cm 1 for nitrophenol (Sullivan et al., 1999) Amylase/Protease Ratio The ratio of amylase to protease activity (A/P ratio), which is frequently used to determine digestive capacity of spiny lobsters (Johnston, 2003; Johnston et al., 2004b), was estimated from the ratio of α-amylase activity to trypsin activity (BAPNA as substrate). As trypsin plays a significant role in proteolysis in lobsters from the superfamily Palinuroidea Latreille (Johnston et al., 1995; Johnston, 2003, 2004b) the 42

55 Chapter 3 activity of this enzyme has been substituted for total protease activity in this calculation Statistical Analyses Mean values from duplicate assays for each pooled phyllosomata sample were compared with a one-way analysis of variance (ANOVA) to identify significant changes in specific and total activities of enzymes between days of culture (P<0.05). For each analysis the assumptions of ANOVA were checked using residual plots and tested for homogeneity using a Levene s test. Duncan s post hoc test was used to identify differences between the means for different days of culture. 3.3 Results The digestive system of P. ornatus phyllosoma consists of a fore-, mid- and hind-gut. The midgut includes the digestive gland, and is located within the cephalic shield and connects with the posterior proto-proventriculus via the primary ducts (Fig. 3.1A,B,C). The digestive gland comprise anterior, lateral and posterior lobes, from which branch a number of blind ending tubules (Fig. 3.1A). The anterior lobes consist of single tubules that extend anteriorly on either side of the foregut, whereas the lateral lobes consist of numerous tubules that bifurcate from a main branch laterally. The posterior lobes extend laterally along the mid- and hind-gut. The digestive gland tubules have a central lumen, a microvillous brush border and are lined by four epithelial cell types Digestive Gland Structure (Light Microscopy) Using light microscopy, enzyme secreting F-cells was identified by their dark-staining cytoplasm and pinocytotic channels protruding into the lumen (Fig. 3.1D). E-cells were not identified as they are a precursor of the other cell types, but are assumed to be located in the distal tips of the tubules. R-cells were present (Fig. 3.1D, E), and B-cells were identified in the lateral but not anterior lobes of stage I phyllosomata (Fig. 3.1D). Large nuclei (~10 μm) were a distinguishing character of all digestive gland cells (Fig. 3.1E). B-cells and F-cells were clearly distinguished during development, with increased vacuolation of B-cells in late-stage phyllosomata indicating an increase in number of digestive vacuoles (Fig. 3.1F). During histological processing, lipid is 43

56 Chapter 3 Figure 3.1 Photomicrographs and transverse sections through the digestive gland of Panulirus ornatus phyllosomata. (A) Photomicrograph, ventral view of stage I phyllosomata. Scale, 500 μm. (B) Posterior foregut showing attachment of the primary ducts. Scale, 20 μm. (C) Anterior foregut and digestive gland lobes (stage I) showing two lobes on the right side of the phyllosomata and one on the left. Scale, 50 μm. (D) Anterior digestive gland lobes (stage I), arrow indicates F-cell. Scale, 20 μm. (E) Mid-posterior digestive gland lobe (stage I) showing large lumen, dense microvillous brush border indicated by an intense green-colour around the perimeter of the lumen (stain, amino black). Arrow indicates cell nucleus. Scale, 20 μm. (F) Large accumulation of B-cells (stage X). Arrow indicates B-cell. Scale, 50 μm. AL; anterior lobe; B-, B-cell; DT, digestive tract; F-, F-cell; FG, foregut; LL, lateral lobe; Lu, lumen; PD, primary duct; R-, R-cell. dissolved and appears as a clear empty space (Icely and Nott, 1980). Hence, lipid was positively identified in R-cells using transmission electron microscopy whereby the phyllosomata samples were post-fixed in osmium tetraoxide. 44

57 Chapter Digestive Gland Structure (Transmission Electron Microscopy) Cell differentiation was also evident based on ultra-structural details. The digestive gland tubules were characterised by many large central vacuoles and a dense microvillous brush border (Fig. 3.2A). B-cells were not located in the distal tips of the tubules, but were evident in the mid-anterior region of the lateral lobes. B-cells were readily recognisable due to a single, large vacuole, enclosed by dense cytoplasm near the proximal cell border (Fig. 3.2B). A microvillous brush border, rough endoplasmic reticulum and mitochondria were also evident, positioned around the vacuole (Fig. 3.2B). The apical complex of mature B-cells was observed to protrude well into the tubule lumen. F-cells appeared after the distal tips of the tubules, and were characterised by densely packed rough endoplasmic reticulum and large vesicles (Fig. 3.2C). Lipid droplets were evident in the lateral lobes and are a strong indication of an R-cell; however, they are often difficult to identify as they are prone to degeneration (Fig. 3.2D) Enzyme Activity Variation in specific and total activities of all enzymes was detected during the period of study. Corresponding days of culture and the relative stage of development of P. ornatus phyllosomata at the time of sampling is listed in Table 3.1. Protease and carbohydrase total and specific activities were higher than lipases (Table 3.2). α- Amylase activity was not detected in day 0 phyllosoma and a large range of Table 3.1. Day of culture and the corresponding developmental stage of Panulirus ornatus phyllosoma. Day of culture Corresponding stage (instar) of development 0 stage I (instar 1) 5 stage I (instar 1) 15 stage II (instar 2) 19 stage III (instar 3) 32 stage IV (instar 4) 45

58 Chapter 3 Figure 3.2 Transmission electron micrographs of a digestive gland tubule of a day 0, stage I, Panulirus ornatus phyllosomata. (A) Anterior region of digestive gland tubule showing large central vacuole, 100 μm. Inset, Digestive gland lobe showing microvillous brush border and part of the lumen. Line range indicates thickness of the brush border. Scale, 100 μm. (B) Large vacuole of a B-cell, B-cell not yet grown to full size due to lack of engorgement and apical complex. Arrow indicates vacuole. Scale, 1 μm. (C) Numerous vesicles in F-cell, rough endoplasmic reticulum around perimeter of both vacuole and vesicles. Scale, 100 μm. (D) Lipid droplets, characteristic of R-cell. Scale, 200 μm. B-cell; Gol, Golgi body; Lu, lumen; Mit, mitochondria; Rer, rough endoplasmic reticulum; Vac, vacuole; Ves, vesicle. enzyme activities of on-grown Artemia fed to the phyllosomata over the culture period were generally low, and at the lower end of the range detected in P. ornatus phyllosoma (Table 3.2). 46

59 Chapter Proteases Protease total activity increased significantly (P<0.05) between days 0 and 15. Activity decreased by day 19 but increased by day 32 (Fig. 3.3A). Protease specific activity increased also during early development with a significant (P<0.05) increase between days 0 and 5 (Fig. 3.3B). Protease specific activity, decreased after day 5 and by day 32 was similar to the concentrations detected in day 0 phyllosomata. Trypsin total and specific activities increased significantly (P<0.05) between days 0 and 5, with a sharp increase in activity (Figs. 3.3C,D). Between days 5 and 32 total activity of trypsin fluctuated around units phyllosomata 1 (Fig. 3.3C). Trypsin specific activity decreased after day 5 and was detected at levels similar to those in day 0 phyllosomata by day 32. Table 3.2. Panulirus ornatus; total and specific activities of digestive enzymes in cultured phyllosoma and on-grown Artemia fed during hatchery culture. Phyllosoma data are mean activity ranges between days 0 32 and Artemia data are the mean activities at the end of the culture period. Total activity given as Units phyllosomata 1 or Units Artemia 1 ; specific activity given as Units mg 1. Enzyme Total activity of cultured P. ornatus phyllosomata (Range) Specific activity of cultured P. ornatus phyllosomata (Range) Total activity of Artemia (Mean) Protease Trypsin Amylase n.d n.d Chitinase Laminarinase Lipase n.d n.d = not detected Carbohydrases α-amylase, chitinase and laminarinase total activity increased significantly (P<0.05) between days 0 and 32 of culture (Figs. 3.4; 3.5A). α-amylase total activity increased significantly between days 0 and 5, remaining relatively constant to day 19, but increased significantly (P<0.05) by day 32, more than doubling in activity (Fig. 3.4A). These significant increases were observed also in the specific activity (Fig. 3.4B). Chitinase total and specific activity was higher than α-amylase, and increased significantly (P<0.05) between days 19 and 32 of culture (Fig. 3.4C,D). Laminarinase total activity increased between days 0 32, with a large increase in activity detected 47

60 Chapter 3 Fig. 3.3 Panulirus ornatus. Protease (A,B) and Trypsin (C,D) total and specific activity during development of cultured phyllosomata. Data are mean ± S.E (n=3). Days of culture with different superscripts are significantly different. Fig. 3.4 Panulirus ornatus. α-amylase (A,B) and Chitinase (C,D) total and specific activity during development of cultured phyllosomata. Data are mean ± S.E (n=3). Days of culture with different superscripts are significantly different. 48

61 Chapter 3 after day 15 (Fig. 3.5A). Laminarinase specific activity decreased during the first 15 d of culture but increased at day 32 to a concentration comparable to that detected in day 0 phyllosomata (Fig. 3.5B) Lipases Lipase total activity (Fig. 3.5 C) increased significantly after day 15 from to units phyllosomata 1 by day 19. Lipase total activity increased further after day 19. Lipase specific activity decreased during the first 15 d of culture but increased significantly by day 19 (Fig. 3.5D). Fig Panulirus ornatus. Laminarinase (A,B) and Lipase (C,D) total and specific activity during development of cultured phyllosomata. Data are mean ± S.E (n=3). Days of culture with different superscripts are significantly different Amylase: Protease Ratio There was a significant (P<0.05) increase in the α-amylase/protease (A/P) ratio by day 5 of culture, peaking at 0.76 (Fig. 3.6). The A/P ratio gradually decreased during days 15 and 19 of culture, with a decrease to 0.17 by day 32 (Fig. 3.6). 49

62 Chapter 3 Fig Panulirus ornatus. Ratio of α-amylase activity to trypsin activity for cultured phyllosomata. Data are mean ± S.E (n=3). Days of culture with different superscripts are significantly different. 3.4 Discussion Structure of the Digestive Gland In day 0, stage I P. ornatus phyllosoma, the digestive gland occupies a relatively small area of the cephalic shield. Higgins (2002) reported similar findings for day 0 J. edwardsii phyllosoma. Anterior and lateral lobes of the digestive gland were present and the three epithelial cell types R-, F- and B-cells were distinguished in the tubules using light histology and transmission electron microscopy. The anterior lobes were centrally located adjacent to the foregut and nerve cord and extended laterally. Posterior lobes were not visible when observing the whole animal under light microscopy, but were identified after the break-down of the foregut in stained light microscopy sections, extending posteriorly adjacent to the mid- and hind-guts (Fig. 3.1G). The lateral lobes were identified as being all other tubules, and extended laterally into the cephalic shield. The final phase of digestion in decapod crustaceans is enzymatic and is generally believed that the principal functions of the digestive gland in these species are the secretion of digestive enzymes and absorption of nutrients for maintenance and growth (Gibson and Barker, 1979; Mikami et al., 1994; Macmillan et al., 1997; Higgins, 2002). Based on their functions in other species and distinguishable morphological characters, the following functions are proposed for each of the cell types in this study (Gibson and Barker, 1979). R-cells were the most numerous cell type; similar findings have been reported for J. edwardsii (Macmillan et al., 1997; Higgins, 2002) and P. japonicus (Mikami et al., 50

63 Chapter ) phyllosoma. Spherical globules containing lipid and presumably glycogen were observed by staining with osmium tetraoxide (Fig. 3.2). The abundance of R-cells and electron dense spherical globules in this cell type suggests that R-cells play an important role in absorbing digested nutrients (proteins, lipids and carbohydrates), presumably via molecular transport through the apical cell membrane, metabolising these nutrients intracellularly (Mikami et al., 1994). B-cells were readily recognisable due to the relatively large, single vacuole enclosed by dense cytoplasm. The brush border of this cell-type was thin and contained a small accumulation of mitochondria and microtubules. B-cells typically absorb nutrients via pinocytosis and store these in vacuoles (Mikami et al., 1994). B- cell secretions were also observed, where the apical complex pinches off in a manner similar to apocrine secretion, releasing the contents of the vacuole into the lumen of the digestive gland tubule (Loizzi, 1971). F-cells were identified due to a dark staining cytoplasm when stained with amino black (stain for identification of protein on semi-thin sections) and an abundance of rough endoplasmic reticulum and small vesicles that characterise this cell type (Higgins, 2002). F-cells with their well developed rough endoplasmic reticula are involved in the secretion and storage of digestive enzymes (Mikami et al., 1994; Johnston et al., 1998; Higgins, 2002) Enzyme Profiles The contribution of exogenous enzymes from Artemia consumed by cultured phyllosomata, although possible, is unlikely. Kamarudin et al. (1994) concluded that the contribution of exogenous enzymes from live food such as Artemia to total digestive capacity of M. rosenbergii larvae is low and insignificant. Similar findings have been reported for the white shrimp Penaeus setiferus (Lovett and Felder, 1990a, 1990b), J. edwardsii phyllosomata (Johnston et al., 2004b) and for larval fish such as the Japanese sardine S. melanostictus (Kurokawa et al., 1998). Total enzyme activities of Artemia were considerably lower or below the level of detection compared to cultured P. ornatus phyllosomata, suggesting that the contribution of exogenous enzymes from ingested Artemia is likely to be minimal. The raptorial feeding biology of P. ornatus phyllosomata, using their pereopods and maxillipeds to shred Artemia into smaller, more manageable sizes for ingestion, indicates that the enzymes contained within the alimentary tract of Artemia are likely to be ingested in an inactive form (Johnston et al., 2004b). Furthermore, as the Artemia are not ingested whole it is not 51

64 Chapter 3 possible to quantify the contribution of exogenous enzymes from ingested Artemia on the digestive capacity of P. ornatus phyllosomata. All digestive enzymes analysed (protease, trypsin, α-amylase, chitinase, laminarinase and lipase) were detected in every day of culture except for α-amylase at day 0. However, α-amylase activity may be present at stage I, but below the level of detection. Expression of a range of protease, carbohydrase and lipase enzymes at stage I suggests that P. ornatus phyllosomata are capable of digesting protein, carbohydrate and lipid throughout the first 32 d of culture. Total activities (enzyme activity per phyllosoma) of the enzymes analysed throughout the 32 d trial generally increased, indicating P. ornatus phyllosoma are capable of digesting these components throughout development. However, the contribution of each of these enzymes is likely to vary during each developmental stage, reflecting morphological changes of the digestive tract, genetic cues and their feeding habit. Protease and carbohydrase activities were considerably greater than those of lipases, indicating that P. ornatus phyllosomata are likely to rely on protein and carbohydrates for energy during the initial 32 d of culture. High protease activity has been detected in J. edwardsii phyllosomata (Johnston et al., 2004b). The higher activity of carbohydrases and proteases than lipase is a possible reflection of their natural feeding biology. Although the natural diet of P. ornatus phyllosomata is unknown, higher activities of these two enzymes suggests they are capable of digesting copepods, mysid shrimps, ectoprocts and cirriped larvae, all of which are common zooplankton in the Coral Sea (Rissik and Suthers, 2000). Pelagic zooplankton in general are known to have high protein (Le Vay et al., 2001), however, as the most common zooplankton communities of the Coral Sea have a hard chitinous exoskeleton and ingest phytoplankton, high activity of carbohydrases including chitinase (carbohydrase) would facilitate digestion of these prey items. In the only other study of enzyme expression in phyllosoma of spiny lobsters, Johnston et al. (2004b) detected higher activities of proteases and lipases in J. edwardsii than carbohydrases, indicating a carnivorous diet. Although hydrolysis of lipid is likely to be important in the digestive physiology of P. ornatus, the difference in enzyme expression between these two species is likely to reflect the different abundance of prey items in their natural habitats, genetic cues, different substrates for the detection of lipase activity and Artemia enrichment. In the current study Artemia were enriched on two species of algae and not enriched using commercial fatty acid enrichments such as DHA Selco, INVE. The level of lipase activity detected in J. edwardsii phyllosomata was reported by Johnston et al. (2004b) to be considerably 52

65 Chapter 3 higher than in previous larval crustacean studies fed enriched Artemia (Lovett and Felder, 1990b; Hammer et al., 2000), indicating that lipid is utilised as an important nutrient for growth in spiny lobster phyllosoma. Despite high lipase concentrations in cultured J. edwardsii phyllosomata, the activity was considerably lower than in wildcaught J. edwardsii phyllosomata. Johnston et al. (2004b) suggested that this large increase in lipase activity between hatchery reared and wild-caught phyllosoma indicates that lipid is an important nutrient for growth and that phyllosoma are capable of making opportunistic use of lipid when it is available. The low lipase levels detected in P. ornatus phyllosomata should be reflected in diet formulations also, as protein and carbohydrate, as reflected by higher protease and carbohydrase activities indicates that these two categories of nutrients are important for growth and survival of this species Ontogenetic Changes The presence of a suite of digestive enzymes (proteases, carbohydrases and lipases) at day 0 and in 5 d old P. ornatus phyllosoma indicates that P. ornatus phyllosoma are capable of digesting prey from the onset of feeding. This is also consistent with the presence of enzyme secreting F-cells in the digestive gland of stage I phyllosoma (Figs. 3.1, 3.2). Jasus edwardsii phyllosoma are also known to produce enzymes at hatch (Johnston et al., 2004b) as are other crustaceans such as M. rosenbergii (Kamarudin et al., 1994), P. setiferus (Lovett and Felder, 1990b), and Procambarus clarkii (Hammer et al., 2000). The most significant increase in total and specific activities of the enzymes analysed occurred between days 0 5 and Specific activities (enzyme activity per mg of larval protein) of both protease, trypsin and α-amylase increased significantly between days 0 5, whereas higher concentrations of chitinase and lipase were detected between days As the same quality of food was supplied throughout the trial, the activity of digestive enzymes varied independent of the diet suggesting that the enzyme activity of P. ornatus can be modulated by the quality of their food and may indicate either a dietary shift in the wild or an increased digestive capacity due to a general increase in the number of lateral tubules and number of enzyme secreting F-cells in the digestive gland. An increase in chitinase and lipase at day 32 suggests P. ornatus phyllosoma are more proficient at digesting the chitinous exoskeleton of zooplankton and storing lipid for growth. The change in activity of both these enzymes should be reflected in diet formulations also from day 32, with 53

66 Chapter 3 increased lipid and carbohydrates including chitin being provided by ingredients such as extra marine oils and crustacean-type meals, such as krill and shrimp meals. Although P. ornatus phyllosoma are smaller than J. edwardsii the total activity of trypsin was higher at hatch in P. ornatus than in J. edwardsii phyllosoma (Johnston et al., 2004b). Trypsin activity significantly increased after 5 d of culture and remained relatively constant until the end of the trial. The trypsin levels at day 32 were, however, considerably less than the levels detected in J. edwardsii phyllosoma at a similar stage of development (Johnston et al., 2004b). High trypsin levels have been found in most carnivorous crustacean larvae (Jones et al., 1997), and generally increase in proportion to the relative amounts of protein in their ingested diets (Kamarudin et al., 1994; Johnston, 2003; Johnston et al., 2004a, 2004b; Johnston and Freeman, 2005; Johnston and Johnston, 2005). This high activity of trypsin indicates that protein is a major component of their diet and should represent a substantial component in a formulated diet for this species. Furthermore, the importance of protein was also reflected in the amylase:trypsin ratio, indicating a large increase in specific activity of both these enzymes at day 5 of culture, with the ratio gradually decreasing suggesting a carnivorous diet and highlights the relative importance of protein for growth and development. The specific activity of chitinase and α-amylase increased significantly between days 19 and 32 of culture. An increase in both these carbohydrases during this period coincides with a substantial size increase of the mouth aperture (Chapter 2) from 3.5 μm (stage III) to 9.9 μm (stage IV). This substantial increase in mouth size with an increase in number of lateral setae and main brushes in the foregut and formation of well developed ampullary (filter) channels forming the filter press suggests that P. ornatus phyllosoma are capable of ingesting efficiently larger chitinous prey items such as small planktonic crustaceans (Chapter 2). Furthermore, the marked increase in specific activity of these two enzymes as well as the increase in protease activity during this time suggests that variations in enzyme profiles of P. ornatus strongly reflect both diet and genetic morphological cues, and indicates a possible shift in diet in the wild. Hence, diet formulations should be altered at day 32 to reflect this change in digestive capacity. Laminarinase activity was detected in P. ornatus phyllosoma. This enzyme is not analysed often in larval enzyme studies. However, as high activities of laminarinase have been detected in both post-pueruli J. edwardsii (Johnston, 2003) and more recently in P. argus (Johnston pers. comm., Department of Fisheries, Western Australia), spiny lobsters may have capacity to digest laminarin. Laminarin is 54

67 Chapter 3 a polysaccharide carbohydrate very much like starch, except it functions as an energy storage compound in Laminaria and other brown alga (Johnston and Johnston, 2005). The presence of this enzyme in coordination with the high activities of α-amylase detected suggests that P. ornatus could potentially hydrolyse more complex, polymeric sugars for energy and growth. Furthermore, as P. ornatus are capable of ingesting and digesting zooplankton which ingest brown algas such as diatoms, suggests a capacity to hydrolyse even the smallest amounts of this carbohydrate contained in their zooplankton prey for energy and growth. The presence of this enzyme and high concentrations of α-amylase is also a favourable result for development of a formulated diet of this species, as the incorporation of complex carbohydrate-type binders such as k-carrageenan and sodium alginate are likely to be digestible, as both binders originate from seaweeds and are presumed to contain a substantial quantity of complex carbohydrates including the β-glucan, laminarin. The digestive capacity of P. ornatus phyllosomata increases considerably during days of culture/development. Cell differentiation in the tubules of the digestive gland from day 0 indicates that phyllosoma have a fully functional digestive gland and therefore the ability to digest ingested prey and formulated diets from stage I. Panulirus ornatus phyllosoma hatch with a diverse range of protease, carbohydrase and lipase enzymes that ultimately enables them to exploit a wide range of zooplankton. High activities of chitinase and proteases suggest that P. ornatus phyllosomata are able to digest the most common zooplankton communities in the Coral Sea such as small crustaceans which have high protein content and a hard chitinous exoskeleton. A greater capacity to hydrolyse protein and carbohydrate than lipid during the first 32 d of culture would suggest that formulated diets for this species should contain high amounts of protein (30 60%) and carbohydrate (10 25%) and smaller amounts of lipid (<15%) initially. Future studies which investigate the morphological structure of the digestive gland and activities and concentrations of digestive enzymes throughout development of P. ornatus phyllosoma will be pivotal to successful culture of this species. Furthermore, enzymology studies which compare the digestive enzyme activities of cultured and wild-caught P. ornatus phyllosoma will help to identify natural feeding transitions, highlight possible nutritional deficiencies in using Artemia and allow formulated diets to be adequately developed to contain the appropriate level of nutrients (protein, lipid and carbohydrate) for aquaculture of this species. 55

68 56

69 Chapter 4 Chapter 4 Stability of Formulated Diets and Feeding Response of Stage I Western Spiny Lobster Panulirus cygnus Phyllosomata Introduction The western spiny lobster, P. cygnus inhabits the warm coastal waters of Western Australia, ranging as far south as Cape Leeuwin (34 22 S) and as far north as North West Cape (21 45 S) (Gray, 1992). Like other spiny lobsters, P. cygnus has a relatively long larval life (9 11 months) in which nine pelagic phyllosoma stages precede a transitional non-feeding puerulus before transformation into a juvenile (Batham, 1967; Nishida et al., 1990; Lemmens and Knott, 1994). Currently, there is no commercial culture of P. cygnus, although, it is thought that on-growing post-pueruli would be commercially feasible due to the ability to catch large numbers off the Western Australian coast using sandwich collectors (Phillips et al., 2003). However, the potential impact of pueruli removal on the wild spiny lobster fishery is a contentious issue (Schaap, 1997). Rearing P. cygnus phyllosomata from hatch under culture conditions is considered the most sustainable option. Currently, culture attempts have achieved production of only early- and mid-stage P. cygnus phyllosomata (Marinovic, 1996; Liddy et al., 2003, 2004). An inadequate dietary regime has been a major impediment to achieving successful culture of phyllosomata of spiny lobsters, stemming from a lack of understanding of their feeding biology, nutritional requirements and ingestive and digestive capabilities (Matsuda and Yamakawa, 2000; Johnston and Ritar, 2001; Cox and Johnston, 2003a; Johnston et al., 2005). To achieve successful larval culture of spiny lobsters, development of a formulated diet is a high priority. The success of any larval diet is critically dependent upon physical form, which is ultimately influenced by moisture content and binder type (D'Abramo, 2002). Shape and size are other important characteristics. However, as phyllosomata of spiny lobsters are raptorial feeders, using their mouthparts to grasp and manipulate diet particles that are physically reduced into sizes that are consumed, size of the diet is recognised not to be as critical in comparison to other crustacean species that filter feed during early development. Raptorial feeding behaviour places high demands on the physical integrity (stability) of formulated diets. Strategies to enhance diet stability are commonly the 5 Presented as a paper: Johnston, M., Johnston, D. (2007). Stability of formulated diets and feeding response of stage I western spiny lobster Panulirus cygnus phyllosomata. Journal of the World Aquaculture Society 38,

70 Chapter 4 addition of different binders or the use of different extrusion manufacturing methods (Heinen, 1981; Teshima et al., 1982; Jussila and Evans, 1998; Langdon, 2003). The type of binder and method of drying have also been recognised to influence the performance of a formulated diet (Teshima and Kanazawa, 1983). A critical aspect of evaluating the appeal of formulated diets was suggested by D'Abramo (2002) to include detailed video analysis of the feeding behaviour of the target species, because although developed diets may contain the necessary feeding stimulants and nutrient profile, the diet still may remain unacceptable. The objectives of this study using P. cygnus as a test species were: (1) to compare the effectiveness of several binders, at various concentrations, in terms of dry matter loss following immersion in water, and (2) determine which form (dry meal versus gelatinous formulations) of formulated diets are most suited to the larval feeding biology of early-stage phyllosomata by assessing their feeding behaviour using digital video analysis. 4.2 Material and Methods Brood Stock Collection and Handling of Phyllosomata Ovigerous female P. cygnus were collected from the west coast of Western Australia in October 2004 and transported to the Western Australian Marine Research Laboratories, Watermans Bay, Perth. They were held at a mean ambient temperature of 19.5 ± 2.5 C. Following spawning, newly hatched phyllosomata were skimmed from near the water surface and transferred to 50 l sand-filtered seawater tanks with a flow rate of 1 l min Diet Preparation of Dry Meal and Gelatinous Formulated Diets The ingredient compositions of the six dry meal and three gelatinous formulated diets used to assess binder effectiveness are shown in Table 4.1. The nutritional requirements for P. cygnus phyllosomata currently are unknown. Hence, diet formulations were based on the known requirements of post pueruli P. cygnus, other palinurid lobsters, and the composition of diets used in previous larval crustacean studies (Kanazawa, 1981; Castell et al., 1989; Glencross et al., 2001; Genodepa et al., 2004a, 2004b; Smith et al., 2005). The quantity of each binder was based on optimal 58

71 Chapter 4 stability and inclusion levels suggested for formulated diets for marine prawns (Heinen, 1981; Teshima et al., 1982) The six dry meal diets were initially prepared by combining and thoroughly mixing the dry and moist ingredients in separate bowls. Both the dry and moist ingredients were then combined and mixed thoroughly using a Kitchen Aid Heavy Duty Plus mixer. Each of the five binding agents were either dissolved in warm water (>60 C) (k-carrageenan, gelatine, and agar), dissolved in 80% ethanol (zein) or added dry (sodium alginate). Sodium polyphosphate (1% inclusion) was used in conjunction with sodium alginate as a sequestrant to improve diet stability. The diets were then spread evenly on an aluminium tray, cut into shape using a star-shaped cookie cutter (1.5 cm point to point x 0.5 cm), and either oven-dried at 50 C for 72 h (sodium alginate and zein), gelled overnight at 4 C (gelatine and k-carrageenan) or oven-dried at a reduced temperature of 40 C for 72 h (agar) to promote gelatinisation. The three gelatinous formulated diets were prepared using a similar technique to the dry meal diets; however, the marine-based meals and wheat flour were sieved to 500 μm, and all other dietary ingredients except the binding agent were solubilised within the oils Water Stability Tests Water stability of each formulated diet was determined using a modified method of Fagbenro and Jauncey (1995). A weighted quantity (n=3) of each of the six dry meal and three gelatinous diets were placed in a 1 l glass beaker, filled with 800 ml of sea water, and then incubated in a water bath at room temperature for 4, 8 and 12 h. Aeration was provided by a 3 cm air stone, at a rate of 1 l min 1. In this way the test diets were gently but continuously agitated (Forster, 1972). After immersion, the remaining water and dislodged diet particles were removed from beakers after the method of Jussila and Evans (1998) before being oven dried at 105 C for 2 h and reweighed. Water stability was calculated as the percentage difference in sample weight (minus the initial diet moisture) after reweighing, and expressed as percentage loss of dry matter (% LDM). Diet moisture was quantified by initially placing each diet in triplicate in pre-weighed, pre-combusted aluminium foil weigh boats, heated at 105 C for 2 h, cooled in a desiccator and reweighed. 59

72 Chapter 4 60

73 Chapter Processing of Gelatinous and Dry Meal Formulated Diets One hundred day 1, stage I P. cygnus phyllosomata were removed from the 50 l sea water tank and placed in three 1 l glass beakers with a weighed sample of each formulated diet. All diets were presented as a rectangular cube (2x1x1 cm), which was achieved by pouring each diet mix into defined-shape cooking moulds. All diets were tied with a nylon string and suspended two thirds from the bottom of each beaker. This technique allowed phyllosomata uninterrupted access to each type of formulated diet in the water column. Feeding response (numbers of phyllosomata attracted to each diet) to the gelatinous and dry meal formulated diets was recorded using a Canon MVX100i digital camcorder and viewed using Quicktime 5.0. Further detailed observations of feeding response (mouthpart movement) was observed using an Olympus DP70 camera linked to a Mag view wide screen monitor and recorded using Image-Pro Plus software v Individual phyllosoma were placed in a Petri dish containing 10 ml of sea water and maintained at 19 (± 0.5 C). Phyllosomata (n=10) were presented with either a gelatinous or dry meal gelatine bound diet and the feeding response and coordinated action of the mouthparts was assessed as above Chemical and Statistical Analysis Samples of finely ground raw ingredients were analysed in duplicate by standard laboratory methods, essentially in accordance with the recommendations of the AOAC (1984), at the Microserve Laboratory, Perth. Dry matter (DM) was determined by oven drying at 105 C to constant weight; ash by ignition at 600 C for 2 h; total nitrogen by a micro-kjeldahl technique; crude protein was calculated by multiplying total nitrogen by 6.25, irrespective of the nature of the nitrogen. Total lipid was determined by acid hydrolysis. All binder stability data are presented as mean ± S.E (n=3) and all statistical analyses were performed using SPSS Mean % LDM values from triplicate bound diets were compared with a one-way ANOVA followed by a Duncan s post hoc test to determine significant differences between the binders and the two forms of formulated diet (gelatinous versus dry meal). For each analysis the assumptions of ANOVA were tested using a Levene s test for normality, and homogeneity of variance was checked using residual plots. All data were transformed prior to analysis using an arcsin square root transformation. Differences were regarded as significant when P<

74 Chapter Results Stability of Dry Meal Formulated Diets At the completion of 4 h immersion, the dry meal formulated diets bound with 2% sodium alginate, with and without the addition of sequestrant (sodium polyphosphate) and 2% gelatine, had the greatest binding ability with less than 25% loss of dry matter (Fig. 4.1A). Dry meal diets bound with 5% k-carrageenan, 3% zein and 3% agar lost in excess of 25% dry matter after 4 h immersion and was significantly (P<0.05) different from both diets bound with sodium alginate. After 8 h immersion, the dry meal diet bound with 2% sodium alginate had significantly better stability (P<0.05) in water than the other dry meal diets, and was the only dry meal diet to lose less than 20% dry matter (Fig. 4.1B). The dry meal diet bound with 5% k-carrageenan performed poorly, with dry matter loss in excess of 40% and had significantly poor stability (P<0.05) in comparison to the other dry meal diets. Dry meal diets bound with 3% zein, 3% agar, 2% gelatine and 2% alginate + 1% sequestrant all had similar dry matter loss after 8 h and were not significantly different, ranging from 24 31%. At the completion of 12 h immersion, both 2% sodium alginate with and without the addition of 1% sequestrant and 2% gelatine had the greatest binding potential, with less than 30% loss of dry matter (Fig. 4.1C). Dry matter loss from the diets bound with 5% k-carrageenan, 3% zein, and 3% agar all had poor stability, with dry matter loss in excess of 35 54%. 62

75 Chapter 4 Figure 4.1. Percentage loss of dry matter (% LDM) of six dry meal and three gelatinous formulated diets using different binders over a 4(A), 8(B) and 12(C) h immersion period. Data are mean ± S.E; n = 3. Different alphabetical superscripts denote significant differences between binding agents (P<0.05). Carrag = k-carrageenan; Zein = zein; Agar = agar; Gel = gelatine; Al + Seq; sodium alginate + sequestrant; Alginate = sodium alginate; G Kr = gelatinous krill; G F/K = gelatinous fish krill; G Wh = gelatinous whey. 63

76 Chapter Stability of Gelatinous Formulated Diets At the completion of 4 h immersion, the gelatinous krill formulated diet had the lowest loss of dry matter. This diet also had similar stability to the dry meal diets bound with alginate (with and without the addition of a sequestrant) and gelatine (Fig. 4.1A). The gelatinous fish/krill and gelatinous whey diets had significantly (P<0.05) higher losses of dry matter in comparison to the gelatinous krill diet, with large diet particles becoming dislodged shortly after immersion (<15 min). After 8 h immersion, the gelatinous fish/krill diet had significantly the poorest stability (P<0.05), with loss of dry matter in excess of 55%. Both gelatinous diets bound with 6.75% gelatine (krill and whey) had similar stability and were not significantly different, with less than 35% loss of dry matter (Fig. 4.1B). The gelatinous whey formulated diet was the most stable gelatinous diet (34% loss of dry matter) after 12 h immersion with significantly (P<0.05) better stability (Fig. 4.1C). Stability of the gelatinous krill diet was poor after 12 h, with dry matter loss doubling from 28% to 56% (Fig. 4.1B,C). However, the gelatinous fish/krill diet had the poorest stability of all the gelatinous diets after 12 h immersion, losing 59% dry matter (Fig. 4.1C) Processing of Gelatinous and Dry Meal Formulated Diets The gelatinous krill diet was completely covered with feeding phyllosomata shortly after immersion (1 2 min). More detailed analysis of the role of the mouthparts during feeding revealed that phyllosomata easily manipulated and speared the gelatinous krill diet with the dactyli of the first pereopods, before passing the diet particles anteriorly toward the maxillipeds and maxillules (Fig. 4.2A F). There were less phyllosomata feeding on the dry meal formulated diets than the gelatinous krill diet, although of this type of diet the 5% k-carrageenan diet was more suited than diets bound with 2% sodium alginate, with and without the addition of a sequestrant, 3% agar and 3% zein. During diet manipulation, increased particle abrasion was observed in the dry meal diet bound with gelatine, which hindered the beating action of the maxillae due to fouling, while the gelatinous krill diet held together during feeding, causing no external fouling. 64

77 Chapter 4 Figure 4.2. Feeding response of stage I Panulirus cygnus phyllosomata to dry meal and gelatinous formulated diets. (A) Phyllosomata locating gelatinous krill formulated diet. Note the positioning of the dactyli of pereopod 1, which penetrate into the diet. (B) Dactyli of pereopod 1 hold the phyllosoma to the diet, while using the dactyli of maxillipeds 3 to manipulate smaller, manageable sized diet particles for ingestion. (C) Dactyli of maxillipeds 2 used to move diet particles anteriority towards the oral cavity. Arrow indicates diet particle. (D) Ventral view, showing spearing of diet particle by the dactylus of pereopod 1. Arrow indicates dactylus. (E) Dactyli of maxillipeds 2 is used to move diet particles anteriorly toward the maxillules for further manipulation. Arrow indicates diet particle. (F) Exopodite and endopodite of the maxillules are used to transfer captured diet particles between the paired paragnaths. Arrow indicates diet particle. Gel DM, gelatine bound dry meal diet; K Gel, gelatinous krill diet; MX, maxillules; MXP2, maxilliped 2; MXP3; maxilliped 3; P1, pereopod 1. 65

78 Chapter Discussion The gelatinous krill formulated diet was more suited to the larval feeding biology of stage I P. cygnus phyllosomata than the dry meal diets tested here. The gelatinous krill diet was observed to draw larger numbers of feeding phyllosomata, presumably due to this diets high level of moisture, that would allow the quick release of feeding stimulants such as certain amino acids that stimulate feeding (Tolomei et al., 2003). Consequently, this diet elicited a heightened feeding response with large numbers of feeding phyllosomata observed covering the diet surface within 2 min of immersion. This formulated diet was also the most stable when immersed in water for up to 4 h and did not foul the maxillae during ingestion, which are important characteristics of successful hatchery diets. The dry meal diet bound with 2% sodium alginate was also extremely stable when immersed in water for up to 4 h; however, these diet types crumbled during manipulation by the mouthparts, consequently fouling the setae of the maxillae. The dry meal formulated diet bound with 5% k-carrageenan was the most successful dry meal diet attracting more feeding phyllosomata than dry meal diets bound with 2% sodium alginate, 3% zein and 3% agar. The reduced feeding response in terms of numbers of phyllosomata observed feeding on this type of diet appears to be related to diet form, moisture content and the method of drying. Teshima and Kanazawa (1983) reported differences in growth and survival of Kuruma prawn larvae, P. japonicus, when fed k-carrageenan diets dried under different conditions. The increased feeding response to diets with high moisture content suggests they offer a level of palatability and eventual consumption that cannot be achieved with dry diets (D'Abramo, 2002). The soft texture of moist diets may also be more like the natural live prey of early stage P. cygnus phyllosomata, which is thought to consist of mucilaginous zooplankton, such as medusae, salps and ctenophores (Macmillan et al., 1997; Johnston and Ritar, 2001; Cox and Johnston, 2003a, 2003b). Preferred preference for soft/moist diets has been determined for a wide range of marine and freshwater prawns and freshwater crayfish (Subrahmanyam and Oppenheimer, 1970; Kitabayashi et al., 1971; Jussila and Evans, 1998; Kovalenko et al., 2002; Ruscoe et al., 2002). Preferred preference for moist/soft diets is also consistent with the recommendations by Cox and Johnston (2003) who concluded that a successful hatchery diet for culture of the green spiny lobster, J. (Sagmariasus) verreauxi should have a soft, jelly-like consistency. 66

79 Chapter 4 During feeding the gelatinous krill diet was extremely stable and, unlike the dry meal diet, did not crumble or abrade excessively. Excessive particle abrasion during feeding should be avoided as it was observed to foul the maxillae, hindering the beating action of this mouthpart. This was not observed with the gelatinous diet as the particles that did abrade during feeding became trapped only temporarily. The reduced fouling was presumed to be associated with the jelly-like texture and larger size of the abraded particles. Similar observations were reported by Cox and Johnston (2003a) for J. verreauxi phyllosomata when fed large Artemia that resulted in reduced instances of setal fouling. The incorporation of different binder types and extrusion methods are the major strategies to enhance water stability of formulated diets. Diets bound with binders are typically susceptible to leaching of water soluble nutrients. Some leaching is necessary to stimulate feeding in crustaceans due to their slow and inconsistent feeding pattern. However, excessive nutrient leaching should be avoided as it reduces the nutritional value of the diet and can have detrimental effects on water quality (López-Alvarado et al. 1994; Teshima et al. 2006). Development of water-stable diets that offer minimal leaching is often achieved at the expense of poor diet attraction and palatability (D'Abramo 2002). Information in this Chapter has provided a preliminary assessment of suitable diet form, binders and moisture content of potential formulated diets for culture of palinurid phyllosomata by using P. cygnus as a test species. It has also highlighted that the successful development of a formulated diet for palinurid phyllosomata is complex, with many factors which affect the physical integrity, palatability and attractiveness of formulated diets being interrelated. Further research examining leach rates, alternate nutritional compositions and use of feeding stimulants to improve attractiveness, palatability and consumption is required to further develop diet formulations for potential hatchery grow-out of palinurid phyllosomata. 67

80 Chapter 5 Chapter 5 Effects of Diet Form, Particle Size, Feeding Stimulants and Protein on Diet Preference By Panulirus ornatus Phyllosomata. 5.1 Introduction The ornate spiny lobster, P. ornatus, is found widely within the Indo-Pacific but is particularly abundant throughout the Torres Strait between north-east Australia and Papua New Guinea (George, 1968; Dennis et al., 1992). Of all spiny lobster species within Australian waters, P. ornatus is the best candidate for aquaculture, having a short oceanic phyllosoma phase (4 6 months) and fast growth rate, attaining commercial size of 1 kg within 2 years post-hatch (Phillips et al., 1992; Butler and Hernkind, 2000; Smith et al., 2003; Barclay et al., 2006). Collection of P. ornatus postpueruli and early juveniles from the wild and on-growing in-caged culture has become a profitable industry in many south-east Asian countries including India, Indonesia, Vietnam and the Philippines (Barclay et al., 2006). However, further expansion of these industries is limited by the seasonal recruitment of pueruli (Srikrishnadhas and Rahman, 1995; Lovatelli, 1997; Barclay et al., 2006). Rearing P. ornatus phyllosomata from the egg stage under aquaculture conditions is considered a more sustainable long-term option. MG Kailis Group recently announced (web-based media release 6 ) closure of the life cycle; however, this was achieved using live and fresh feeds which greatly increase production costs. For production of phyllosomata to be cost effective and viable, formulated diet(s) still need to be developed. To date the palinurid species which have been reared from the egg throughout the extended phyllosomata phase have been fed Artemia nauplii and fresh feeds such as squid, fish larvae and mussel gonad (Kittaka, 1988; Kittaka and Ikegami, 1988; Kittaka et al., 1988; Kittaka and Kimura, 1989; Yamakawa et al., 1989; Ritar et al., 2002). Due to the prolonged phyllosoma phase, extending over several months and typically 9 11 developmental stages, the use of live feeds generates their own problems during hatchery culture. They are a vector for the introduction of pathogenic microorganisms, lack nutritional consistency, and are labour-intensive to produce (Genodepa et al., 2004a). Furthermore, on-grown Artemia and fresh feeds are expensive to maintain, and add considerably to total production costs of producing phyllosomata. The development of a formulated diet that is palatable and delivers the correct balance of nutrients is seen as an attractive and cost effective alternative to live and

81 Chapter 5 fresh feeds (Kovalenko et al., 2002). There has been little research on development of a formulated diet(s) for spiny lobster phyllosomata. A successful formulated diet should: (1) efficiently retain nutrients; (2) possess physical and chemical characteristics that result in their ingestion; (3) be readily digested and assimilated; and (4) consist of an optimal nutrient composition for maximum survival, development and growth (Guthrie et al., 2000). Before nutritional requirements can be determined, an early and necessary step in the growth process is that a formulated diet must be available to and ingested by phyllosoma. Several physical characteristics of formulated diets including particle size, colour, taste, texture (diet form), initial water content, and buoyancy may be of significance and affect diet preferences of P. ornatus phyllosomata. Preference for diets may be studied by visual observation of feed intake, a time consuming and difficult method for larval crustaceans as they are slow, intermittent feeders, or by analysis of gut content after a period of feeding using X-radiography. Markers like chromic oxide (Austreng, 1978; Ishikawa et al., 1996), radioactive isotopes (Kolkovski et al., 1993; Genodepa et al., 2004b) and 5α-cholestane (Ishikawa et al., 1996) have been added to the diets of larval/juvenile fish and crustaceans to quantify the feed content in the gut, often in connection with nutrient digestibility studies. Although many methods have been devised for measurement of feed intake and diet preference, most of them have short-comings, especially for raptorial feeding larvae, as formulated diets are often consumed inefficiently, and/or lose some nutrients by leaching into the water (Teshima et al., 2000). Furthermore, many methods require the collection of a large quantity of egested faecal material, which is often a difficult task when larvae are, involved (Teshima et al., 2000). Recently, inert markers such as lanthanide and ytterbium markers have been used for measuring feed intake, diet preference and digestibility of nutrients in a range of juvenile fish species such as cod (Gadus morhua) (Ottera et al., 2003; Garatun- Tjeldstø et al., 2006), Atlantic salmon (Salmo salar) (Austreng et al., 2000) and gibel carp (Carassius auratus gibelio) (Xue and Cui, 2001). Ytterbium oxide has also recently been used to measure feed intake and diet preferences in J. (Sagmariasus) verreauxi phyllosomata (Cox, 2004). An appropriate marker to evaluate diet preference should ideally be: 1) homogenously incorporated into the diet, and easily and accurately analysed, even at low concentrations; 2) indigestible, not affect metabolism or react with digestive enzymes; 3) not absorbed across the epithelial walls of the gastro-intestinal tract; 69

82 Chapter 5 4) should not discolour the feed (especially important for visual feeders), and be relatively in-toxic to humans (Austreng et al., 2000). Austreng et al. (2000) verified the feasibility of using ytterbium oxide as a mono-labeled marker in nutrient digestibility studies in juvenile Atlantic salmon. The results indicated that ytterbium oxide gave accurate estimates of apparent digestibility at much lower concentrations than necessary when using markers such as chromic oxide. In addition, similar ratios of the marker in the feed and faeces indicated that ytterbium oxide was recovered at a similar rate after passing through the digestive tract. Furthermore, ytterbium oxide (1.3%) was the least soluble of a range of trivalent metal oxides tested in vitro solubility in weak acid (HCl, ph 3 as in stomach contents of Atlantic salmon) and was not soluble in water. Chromic oxide has previously been used as a marker in many fish and crustacean studies (Austreng, 1978, Ishikawa et al., 1996; Teshima et al., 2000). However, chromic oxide is absorbed in many fish species (Austreng et al., 2000), but is not absorbed in the penaeid prawn, P. japonicus (Ishikawa et al., 1996). Despite this favourable attribute for use as a marker for crustaceans, chromic oxide is toxic to man in low concentrations (Austreng et al., 2000) and also discolours the feed (hues of green). As P. ornatus are visual feeders, diet colour may affect consumption and hence diet preferences (Chapter 6, Appendix 3). Thus, as ytterbium oxide meets many more of the selection criteria for use as a marker for diet preference tests than chromic oxide, and that ytterbium oxide has been used in nutritional studies of larval fish in methodologies that have been accepted for publication, and used to measure feed intake and diet preferences in J. (Sagmaraisus) verreauxi phyllosomata (Cox, 2004). It is reasonable, therefore, to suggest that ytterbium oxide would be suitable for investigating diet preferences for early-stage P. ornatus phyllosomata. Consequently this method was used here in this study. There were four objectives to the present study whereby the inert marker ytterbium oxide was included in each diet formulation to aid in determining optimum diet characteristics for stage I P. ornatus phyllosomata.: (1) to determine a preference for diet texture (gelatinous, paste and hard), (2) to ascertain an optimum particle size range ( , , and >800 μm), (3) to establish a preference for known (betaine, glycine and taurine) feeding stimulants, and (4) to determine an optimum level of protein (44%, 50% and 50%-squid CP), 70

83 Chapter Materials and Methods The experiments were conducted over two years ( ): optimum diet form and particle size were tested in year I, and feeding stimulants and level of protein were tested in year II. Diet formulations for the preceding diet preference trial in each year were altered according to experience Brood Stock Collection, Handling and Stocking of Phyllosomata Female and male P. ornatus were collected near Trinity Inlet (16 55 S., E) (northern Queensland, Australia), transported to the Northern Fisheries Research Centre, Cairns where they were conditioned on a mixed diet of frozen greenshelltm mussel, (P. canaliculus), pipis (Donnax spp.), scallops, (P. fumatus) and frozen squid (Nototodarus spp.) and allowed to mate. Ovigerous females were removed from the culture tank into individual (50 l) incubation chambers and held at a mean temperature of 26.0 ± 0.5 o C and salinity of 36 g l 1. A sub-sample of newly hatched phyllosomata (day 0, stage I) were skimmed from near the water surface of the incubation chambers, and subject to a salinity stress test before being stocked into 3 l rearing tanks. The rearing tanks were provided with either flow-through or recirculating water (at an exchange rate of 6 l h 1 ) subject to mechanical filtration (to 1 μm) and ultraviolet and ozone treatment Test Diet Preparation and Proximate Analysis The composition of the test diets is shown in Table 5.1. Diets were formulated to contain levels of protein and energy that are characteristic of Artemia nauplii, using feed ingredients that have been used previously in diet formulations for juvenile and adult P. ornatus (Smith et al., 2003, 2005; Williams et al., 2005); the inclusion and proportion of the fish, cod liver and corn oils were based on Castell et al. (1989) and Genodepa et al. (2004a). All formulated diets were prepared by combining and mixing the dry and moist ingredients in separate bowls using a Kitchen Aid Heavy Duty Plus mixer. The inert marker ytterbium oxide (Yb ) was then added and mixed. The binding agents (gelatine and k-carrageenan) were dissolved in warm water (>60 C), allowed to cool to room temperature before being added to the diet mix. The mixed diet was then spread onto an aluminium tray and allowed to gel overnight at 4 o C. 71

84 Chapter 5 In all feeding trials except determination of a preferred particle size range, formulated diets were cut into shape using a star-shaped cookie cutter (1.5 cm point to point x 0.5 cm) and provided to phyllosomata as a ration in excess of the equivalent dry weight of Artemia nauplii (>13.04 mg l 1 ). Particle sizes were produced by grinding the gelled paste diet (Table 5.1) using a mortar and pestle and sieved to the desired particle range. For proximate analysis, samples of finely ground raw ingredients were analysed in duplicate by standard laboratory methods essentially in accordance with the recommendations of the AOAC (1984), at the Microserve Laboratory, Perth. Dry matter (DM) was determined by oven drying at 105 o C to constant weight; ash by ignition at 600 o C for 2 h; total nitrogen by a micro-kjeldahl technique (crude protein was calculated by multiplying total nitrogen by 6.25, irrespective of the nature of the nitrogen); and total lipid was determined by acid hydrolysis Determination of the Concentration of Yb 2 O 3 An estimated 1500 phyllosomata were initially stocked into individual 3 l larval rearing tanks (x 3 replicates) and allowed to feed continuously on each formulated diet. Each hour, approximately 300 phyllosomata were removed from the larval rearing tanks (for a total duration of 4 h), rinsed in sea water and frozen at -80 o C. Each treatment was sampled in triplicate and control treatments with unfed phyllosomata were collected and used to validate sample readings. For the analysis of 70 Yb concentrations in samples of the diet and pooled phyllosomata samples, the samples were initially defrosted, dried on blotting paper and weighed (wet weight). All samples and blanks were digested by adding 3 ml of concentrated nitric acid (HNO 3 ) for 2 h at 80 o C, and then a further 10 h at room temperature. Once the digests were clear, samples were transferred to volumetric flasks and diluted to 10 ml with double distilled water, and analysed using a Perkin- Elmer Elan 6000 inductively-coupled plasma-mass spectrometry (ICP-MS). The internal standards and blank samples were prepared and analysed together with the sample and diet solutions. 72

85 Chapter 5 73

86 Chapter Calculations and Statistical Analysis Preference for a diet in each experiment was assessed by the relative proportion of the marker ingested each hour over a 4 h period. diet/phyllosomata) was calculated as an equation: Diet preference (DP, μg Preference for diet containing marker DP = (f i /d i )/N; where f i is the gross amount of marker detected, d i is the gross amount of marker in the diet, and N is the number of phyllosomata used. All data are presented as means ± S.E. (n=3) and all statistical analyses were performed using SPSS The effect of diet preference for each experiment was analysed using a one-way analysis of variance followed by a Duncan s multiple ranges post hoc test to determine significant differences between each treatment. For each analysis the assumptions of ANOVA were tested using a Levene s test for normality, and homogeneity of variance was checked using residual plots. Differences were regarded as significant when P< Results Preferred Diet Form and Particle Size Range Stage I P. ornatus phyllosomata consume a formulated diet from the onset of hatch and may continuously feed on a diet over a 4 h period (Figs. 5.1, 5.2). Phyllosomata had a significant preference (P<0.05) for the hard diet after 3 h of continuous feeding (5.57 μg phyllosomata) and continued to prefer (P<0.05) this diet until the final fourth hour of the trial (12.25 μg diet/phyllosomata) (Fig. 5.1A) Stage I P. ornatus phyllosomata did not show a preference for an optimum particle size range (Fig. 5.1B). Furthermore, the amount of diet consumed over the 4 h period was substantially less than the preceding (diet form) trial (Figs 5.1A, 5.1B) Preferred Feeding Stimulant and Level of Protein The feeding stimulant betaine had significant (P<0.05) feeding enhancing effects when included at 1.5% (Fig. 5.2A). Preference was significantly (P<0.05) higher after 3 h of 74

87 Chapter 5 Fig (A) Diet preference by Panulirus ornatus phyllosomata, in which the phyllosomata received three different forms of formulated diet and (B) different size ranges. Preference is expressed as (μg diet/phyllosomata). Error bars represent mean ± S.E; (n=3). Means with different superscripts are significantly different within diet form and particle size (P<0.05). 75

88 Chapter 5 Fig (A) Diet preference by Panulirus ornatus phyllosomata, in which the phyllosomata received formulated diets containing feeding stimulants, and (B) level of protein. Preference is expressed as (μg diet/phyllosomata). Error bars represent mean ± S.E; (n=3). Means with different superscripts are significantly different within feed stimulant and % crude protein (P<0.05). 76

89 Chapter 5 continuous feeding for diets containing the feeding stimulant betaine (0.32 μg diet/phyllosomata) and further increased after 4 h (0.54 μg diet/phyllosomata) (Fig. 5.2A). Diets containing taurine resulted in a low diet preference compared to the control containing the inert filler, carboxymethyl-cellulose (1.5%) (Fig. 5.2A). Preference of formulated diets containing two graded levels of protein by stage I P. ornatus phyllosomata is shown in Fig. 5.2B. The results shows that after 2 h of continuous feeding stage I phyllosomata significantly (P<0.05) prefer a formulated diet containing 44% and 50% CP (0.156 and μg diet/phyllosomata) than the formulated diet containing 50%-squid CP (0.048 μg diet/phyllosomata). However, after 3 and 4 h of continuous feeding no preference for diets containing graded levels of protein was apparent (Fig. 5.2B). 5.4 Discussion This is the first evidence that stage I P. ornatus phyllosoma will consume a formulated diet post-hatch, indicating they are capable of capturing and ingesting formulated diets. Furthermore, physical characteristics of a diet such as form, particle size and nutritional characteristics including; feeding stimulants and level of crude protein are important considerations that affect diet preferences Diet Form (Texture) The results indicate a preferred preference for the hard diet form (Fig. 5.1A). It is highly probable that P. ornatus phyllosomata prefer the hard diet due to better palatability and availability (fast sinking rate). In Chapter 4 I determined gelatinous diet formulations were well accepted by P. cygnus phyllosoma. Similar concluding comments based on morphology of the mouthparts and foregut of J. edwardsii and J. verreauxi phyllosomata have also been made (Johnston and Ritar, 2001; Cox and Johnston 2003a). However, preference for diet texture is likely to be different within the Palinuridae and this result highlights that different forms of formulated diets are likely to be required for the different commercially important species within Australia. The hard diet form may also consist of a texture which is more like the natural live prey of P. ornatus phyllosomata. In Chapter 2 I determined that the mouthparts of P. ornatus phyllosomata are well established at hatch to capture a wide range of prey items of any hardness as there is little structural variation of the mouthparts during development. However, their ability to capture and process larger prey items increases 77

90 Chapter 5 with development, which is associated with increasing size of the mouth aperture and disposition between the mouthparts. Another plausible reason for the increased preference of the hard diet form may be associated with the availability and contact time between the diet and phyllosomata. The hard diet sank rapidly, which may be a favourable attribute, as phyllosomata spent considerable time on the bottom of the rearing tanks in the present trial. The gelatinous diet form was positively buoyant and the paste diet resuspended easily when in close proximity to the water inlet. It should be noted that although negatively buoyant particles in this experiment were successful, neutrally buoyant particles may be more appropriate for other tank designs. Buoyancy of the diet particles in conjunction with light intensity is also likely to be important during culture, as P. ornatus become photo-negative after stages III IV, suggesting that if light intensity is not varied or the larval rearing tanks are in a direct line with artificial lighting, then negatively buoyant particles may be more appropriate Particle Size Panulirus ornatus phyllosomata, like caridean and nephropid lobster larvae, are carnivorous, displaying a raptorial feeding behaviour, post-hatch. This form of feeding behaviour highlights their ability to feed on a wide range of prey items and size ranges. This ability was reflected in the results of the second trial in this study, which showed no significant preference for diet particles from the four size ranges (Fig. 5.1B). However, stage I P. ornatus phyllosomata did ingest more diet particles within the μm size range over the 4 h duration of this trial. Artemia nauplii are commonly used as the initial feed for rearing P. ornatus phyllosomata, ranging in size between μm (Genodepa et al., 2004a). Hence, as Artemia nauplii are at the lower end of this preferred size preference, consumption and feeding efficiency for aquaculture of this species may be improved if diet particles are provided within this size range or if Artemia are on-grown from the nauplius stage on algae for 1 5 days. The optimum particle size range determined in this Chapter is also within the optimum size range of preferred particle size ranges determined for other raptorial feeding crustacean larvae that use their mouthparts to capture and manipulate dietary items such, as mud crab larvae (Genodepa et al., 2004a), whereas smaller dietary particles have been determined as an optimum size for both fish larvae which are gulp feeders (Cañavate and Fernández-Díaz, 1999) and crustacean larvae which filter-feed in the initial proto-zoeal stages (Jones et al., 1979). Preference by 78

91 Chapter 5 P. ornatus phyllosomata for diet particles ranging between μm is also within the size range of the most abundant zooplankton communities within the Coral Sea that early-stage phyllosomata are likely to encounter and ingest throughout development, such as calanoid and cyclopoid copepods, mysid shrimps and cirriped larvae (Rissik and Suthers, 2000) Feeding Stimulants Panulirus ornatus phyllosomata are capable, post-hatch, of discriminating between diets containing feeding stimulants, with a significant (P<0.05) preference for diets containing betaine and glycine after 4 h of continuous feeding (Fig. 5.2A). The results also suggest that taurine may not be suitable as an individual feeding stimulant for P. ornatus phyllosomata. The incorporation of feeding stimulants to increase foraging activity and feed intake is exceptionally important for aquaculture of P. ornatus phyllosomata, as although P. ornatus phyllosomata are stimulated to feed by visual cues (Chapter 6, Appendix 3) they also have an array of chemosensory setae on the dactyli of the second and third maxillipeds and integument (Chapter 2) similar to those found on other species of spiny lobster that increase their excitatory capacity to locate and feed on captured prey (Cox and Bruce, 2003). Furthermore, as P. ornatus phyllosomata are raptorial feeders, tearing and manipulating larger-sized prey into smaller more manageable sizes for ingestion commonly leads to the abrasion of small uneaten particles (Chapter 4). This form of feeding behaviour highlights that during culture P. ornatus are highly vulnerable to microbial attack, resulting from uneaten particles and to poor water quality. The attractant ability of betaine and glycine in crustaceans has been reported by several authors. The feeding stimulant was shown to act as a dietary feeding attractant in P. monodon, with significant weight gain when added at 10 g kg 1 (Penaflorida and Virtanen, 1996) and for M. rosenbergii with substantially higher weight gain, feed conversion ratios (FCR), growth per day and feed intake, when added as little as 5 g kg 1 (Felix and Sudharsan, 2004). Food and searching behaviour have been also enhanced in a range of crustaceans, including J. edwardsii (Tolomei et al., 2003) and M. rosenbergii (Harpaz et al., 1987) when an aqueous solution of betaine has been infused in the water. Feed intake of J. edwardsii when fed a formulated diet containing glycine was found by Sheppard et al. (2002) to significantly (P<0.05) increase when compared to taurine and betaine. 79

92 Chapter 5 Taurine was found to be an ineffective feeding stimulant for addition to formulated diets for aquaculture of P. ornatus phyllosomata. Similar observations have been reported by Barbato and Daniel (1997) for P. argus with negligible stimulation provided by both taurine and glycine. The detection and excitatory capacity of these three feeding stimulants within the Palinuridae appears to be highly variable, with large interspecific differences, possibly reflecting the presence or absence of these chemicals in their natural prey (Awapara, 1962) Level of Protein Preference for diets were significantly (P<0.05) affected by different combinations of dietary crude protein (CP), with phyllosomata preferring diets containing 44% and 50% CP after 3 h (Fig. 5.2B). Preference for these two diets in comparison to the 50%-squid CP diet is in agreement with increased survival of P. ornatus phyllosomata over 28 d when fed the same two diets (44% and 50% CP) without the inert marker ytterbium oxide (Chapter 6). Preference for formulated diets containing various levels of crude protein for phyllosomata of spiny lobsters has not previously been reported. However, the levels reported here are within a similar range determined for adult P. ornatus (53% CP) (Smith et al., 2003). The reduced preference of the 50%-squid CP diet is likely to be due to reduced palatability, as although both 50% CP diets have a similar level of crude protein, the inclusion of fish meal was considerably greater in the 50% CP diet, which may possibly contain compounds such as water soluble amino acids that are more attractive to P. ornatus phyllosomata. Furthermore, in feeding trials discussed in Chapter 6, phyllosomata fed the 50%-squid CP diet had significantly lower survival than phyllosomata fed the 50% CP diet. This suggests that there is a possible nutritional deficiency, and that the 50% CP diet has a more balanced nutritional profile for growth and survival of P. ornatus phyllosomata. Little is currently known of formulated diet development for P. ornatus phyllosomata, and the results presented in this Chapter are a significant development in this field. Perhaps most importantly, this study has shown that formulated diets are ingested by P. ornatus phyllosomata post-hatch; however, considerable work is still required to improve the nutritional value of formulated diets for this species. 80

93 81

94 Chapter 6 Chapter 6 Evaluation of Partial Replacement of Live and Fresh Feeds with a Formulated Diet and the Influence of Weaning Panulirus ornatus Phyllosomata onto a Formulated Diet during Early Ontogeny Introduction The ornate spiny lobster P. ornatus is recognised as a prime candidate for aquaculture, due to its short planktonic phyllosoma phase (<6 months), high market demand in western China, and fast growth rate, attaining 1 kg after 18 months post settlement (Butler and Hernkind 2000; Jones et al. 2001; Barclay et al. 2006). One constraint hampering the successful hatchery culture of phyllosomata of P. ornatus is the scarcity of information on their feeding biology and nutritional requirements (Johnston et al. 2005). Currently, phyllosomata like other larval raptorial-feeding crustaceans, are cultured using live brine shrimp, Artemia sp., enriched on either micro-algae or commercially available enrichment products. Artemia are used almost exclusively for phyllosoma culture as they are known to stimulate a visual feeding response, are readily accepted by early-stage phyllosomata, and are easily cultured under hatchery conditions (Johnston et al. 2005). Despite these favourable attributes as a larval food source, Artemia are expensive and labour-intensive to produce, are a potential vector for the introduction of pathogens into the culture system, and have variable nutrient compositions (Callan et al. 2003; Genodepa et al. 2004a). Development of formulated diets is an attractive alternative as their nutrient composition can be modified to meet any specific ontogenetic demands and can be used also to reduce hatchery production costs, by totally or partially replacing the production of live and fresh feeds. Combining Artemia nauplii with fresh feed of other live feeds have generally produced far superior results in terms of growth and survival for early- and mid-stage phyllosomata, than feeding an exclusive diet of Artemia nauplii (Kittaka 1988, 1997; Kittaka and Ikegami 1988; Kittaka et al. 1988; Kittaka and Kimura 1989; Yamakawa et al. 1989; Ritar et al. 2002, 2003). Tong et al. (1997) suggested that combinations of fresh feeds in combination with Artemia nauplii may provide a better nutritional profile for phyllosoma development. Although combinations of mussel gonad and Artemia have been successful for culturing phyllosomata of many palinurid species, their potential use as a partial replacement feeding regime for P. ornatus has not been 7 Presented as a paper: Johnston, M., Johnston, D., Jones, C. (in press). Evaluation of partial replacement of live and fresh feeds with a formulated diet and the influence of weaning Panulirus ornatus phyllosomata onto a formulated diet during early ontogeny. Aquaculture International. 82

95 Chapter 6 reported, nor has the potential replacement of Artemia with different combinations of formulated diet. Co-feeding, a feeding process during weaning where the formulated diet ration is introduced and successively increased to replace live feeds completely is often used in hatcheries to reduce mortality associated with cannibalism and stress. Co-feeding live feeds with a formulated diet during weaning has been successful for many larval marine fish species whereby co-feeding has increased the acceptance of formulated diets, improved larval nutrition and conditioned the larvae to accept more readily the formulated diet when the live feed is withdrawn and has accelerated the weaning process (Kolkovski et al. 1997a, 1997b; Cañavate and Fernández-Díaz 1999). Cofeeding regimes have also accelerated the process of weaning in a range of crustacean species, including the mud crab, S. serrata (Genodepa et al. 2004b) and the white leg prawn, L. vannamei (Biedenbach et al. 1990). It is likely that a similar approach may be successful in the hatchery culture of P. ornatus phyllosomata, as low survival and growth typically are observed when phyllosomata are fed formulated diets exclusively at the onset of hatch (Johnston et al. 2005). A feeding regime whereby live feeds are used for the first I III developmental stages, before weaning to a 100% formulated diet ration using co-feeding, would significantly reduce hatchery production costs and would be of considerable benefit to the industry. Integral to the development of a formulated diet for any larval species is the identification of their protein requirements, as protein is required to provide amino acids for growth (Glencross et al. 2001). The optimum dietary protein levels for adult spiny lobsters reported so far are species dependent (Glencross et al. 2001; Smith et al. 2003; Ward et al. 2003). Adult P. ornatus showed a classic dose response to various levels of protein, with adequate growth at 53% crude protein (CP) (Smith et al. 2003). As protein-based marine meals are costly, an indication of a level of protein that provides optimum growth and survival of P. ornatus phyllosomata will greatly reduce hatchery feed costs. There were three objectives of the present study: 1) to evaluate the partial replacement of Artemia and greenshelltm mussel gonad with various combinations of a formulated diet; 2) to determine an optimal period to wean P. ornatus phyllosomata onto a formulated diet during early ontogeny; and 3) to ascertain whether formulated diets containing various levels of crude protein have a significant affect on growth and survival of P. ornatus phyllosomata when live feeds are withdrawn when weaning. 83

96 Chapter Materials and Methods Brood Stock Collection, Handling and Stocking of Phyllosomata Female and male P. ornatus were collected near Trinity Inlet (16 55 S., E) (northern Queensland, Australia), transported to the Northern Fisheries Research Centre, Cairns where they were conditioned on a mixed diet of frozen greenshelltm mussel, (P. canaliculus), pipis (Donnax spp.), scallops (P. fumatus) and frozen squid (Nototodarus spp.) and allowed to mate. Ovigerous females were removed from the culture tank into individual incubation chambers and held at a mean temperature of 26.0 ± 0.5 o C and salinity of 36 g l 1. A sub-sample of newly hatched phyllosomata (day 0, stage I) were skimmed from near the water surface of the incubation chambers, and subject to a salinity stress test before being stocked into 3 l up-welling rearing tanks. The 3 l rearing tanks were cylindrical in shape and made from highgrade plastic. The rearing tanks were provided with recirculating water (at an exchange rate of 6 l h 1 ) subject to mechanical filtration (to 1 μm) and ultra-violet and ozone treatment. Photo period of 12L:12D was also applied Composition and Manufacture of Formulated Diets The composition of the formulated diets for all feeding trials is listed in Table 6.1. Diets developed for the first three trials were formulated to contain levels of protein and energy that are characteristic of Artemia nauplii, using feed ingredients that have been previously used in diet formulations for juvenile and adult P. ornatus (Smith et al. 2003, 2005; Williams et al. 2005); the inclusion and proportion of the fish, cod liver and corn oils were based on Castell et al. (1989) and Genodepa et al. (2004a). All formulated diets were prepared by combining and mixing the dry and moist ingredients in separate bowls using a Kitchen Aid Heavy Duty Plus mixer. The binding agent (gelatine) was dissolved in warm water (> 60 C), allowed to cool to < 37 C before being added and mixed. The mixed diet was then spread onto an aluminium tray and allowed to gel overnight at 4 C. In the first three trials and in the fourth trial, until the end of the co-feeding phase the gelled diet was ground using a mortar and pestle and sieved to the desired particle size range ( μm). After the co-feeding phase in trial IV, a star-shaped diet was used as a stationary feed station that was produced using a star-shaped cookie cutter (1.5 x 0.5 cm). 84

97 Chapter 6 85

98 Chapter 6 Formulated diets in trial IV were developed to provide one diet at 44% and two diets at 50% crude protein (CP) (Table 6.1). The dietary protein gradient was obtained by varying the inclusion of squid and Peruvian fish meal compensated for by a respective decrease in carboxymethyl-cellulose (inert filler) (Table 6.1). All diet formulations contained a constant inclusion of Antarctic krill meal, krill hydrolysate, freeze-dried blood worms and greenshelltm mussel flesh to enhance the palatability and attractiveness of the formulated diets Artemia Production and Preparation of Mussel and Feeding Protocols Artemia cysts (INVE, Belgium) were hatched daily (18 24 h total hatching time) in 15 l conical cones at 28 C and either harvested as nauplii and fed to phyllosomata, or ongrown in 500 l conical tanks for up to 13 days on T. chuii. On-grown Artemia were enriched with I. galbana (Tahitian strain) for h prior to feeding to phyllosomata to increase the levels of highly unsaturated fatty acids (HUFAs). Both the harvested Artemia nauplii (<520 µm) and on-grown Artemia ( mm length) were rinsed to remove debris and disinfected for 25 min in 100 ppm formaldehyde before feeding to phyllosomata. The gonads of frozen greenshelltm mussel (P. canaliculus) were removed and chopped finely, washing through an 800 μm mesh onto 500 μm mesh and the resulting μm pieces were weighed and stored on ice until feeding. In all trials phyllosomata were fed once daily at 11 AM. Feed quantities were based on 4 Artemia ml 1 and formulated diet and greenshelltm mussel rations were calculated on the basis of an Artemia nauplii dry weight of 3.26 x 10 3 mg individual 1. Ten percent was added to greenshelltm mussel rations (standard ration size as determined at the Northern Fisheries Research Centre) to account for moisture (Table 6.2) Determination of Moults and Mortalities and Hygiene of Larval Rearing Vessels Moults, were counted daily by removing exuviae floating in the water column. Phyllosomata mortalities were removed and counted daily and were confirmed by counting all survivors every second day. Stages of phyllosomata development were determined according to Duggan and McKinnon (2003). At the start and end of each trial, a sub-sample (n = 5) of phyllosomata from each treatment were measured for 86

99 Chapter 6 carapace length and width and total length using a Leica camera and image analysis software before returning to each rearing vessel. Uneaten food was removed daily by siphoning. Phyllosomata were transferred every second day into clean rearing tanks after uneaten food, moults and mortalities were removed from the dirty rearing tanks Experimental Design Trial I: partial replacement of Artemia nauplii and greenshelltm mussel gonad with a formulated diet This trial examined various replacement options of Artemia nauplii (Table 6.2): % Artemia nauplii 2. 50% Artemia nauplii: 50% formulated diet 3. 50% Artemia nauplii: 50% greenshelltm mussel % formulated diet Phyllosomata were conditioned for 7 d on Artemia nauplii before receiving various combinations of live and fresh feeds and a formulated duet. Prior to the start of the trial, the stocking density was reduced from 40 to 10 phyllosomata l 1. Table 6.2 The proportions of on-grown Artemia, Artemia nauplii and formulated diet fed to Panulirus ornatus phyllosomata in trials I II. Dietary Treatment Trial Artemia Mussel Formulated Diet # l 1 (mg l 1 ) (mg l 1 ) 100% Artemia nauplii % Artemia nauplii + 50% Formulated Diet % Artemia nauplii + 50% GreenshellTM Mussel % Formulated Diet % Formulated Diet + 50% GreenshellTM Mussel % Formulated Diet + 25% Ongrown Artemia % Formulated Diet + 25% GreenshellTM Mussel % Ongrown Artemia + 50% GreenshellTM Mussel

100 Chapter 6 Trial II: partial replacement of on-grown Artemia and greenshelltm mussel gonad with a formulated diet This trial examined various replacement options of on-grown Artemia (Table 6.2): 1. 50% formulated diet: 50% greenshelltm mussel 2. 75% formulated diet: 25% on-grown Artemia 3. 75% formulated diet: 25% greenshelltm mussel 4. 50% on-grown Artemia: 50% greenshell TM mussel Phyllosomata were conditioned for 7 d on on-grown Artemia before receiving various combinations of live and fresh feeds and a formulated duet. Prior to the start of the trial, the stocking density was reduced from 40 to 10 phyllosomata l 1. Trial III: conditioning phyllosomata for 14, 18 or 21 d on live Artemia nauplii prior to weaning onto a 100% formulated diet In this trial, early stage phyllosomata were subjected to a 7 d co-feeding protocol wherein Artemia nauplii were fed for 14, 18 and 21 d before beginning successive daily 14.28% volumetric increases in the proportion of formulated diet, and respective volumetric decreases in the proportion of Artemia nauplii, with formulated diet fed exclusively from day 7 of co-feeding onward. Prior to the start of co-feeding at day 14, phyllosomata were re-stocked from 50 to 25 phyllosomata l 1. Trial IV: conditioning phyllosomata for 14 d on live on-grown Artemia prior to weaning onto three formulated diets containing two graded levels of crude protein. Phyllosomata were conditioned for 14 d on on-grown Artemia prior to weaning onto one of three formulated diets consisting of two protein levels 44, 50 and 50%-squid CP. Phyllosomata were subjected to a 4 d co-feeding protocol from day 14 with successive daily 25% volumetric increases in the proportion of formulated diet, and respective volumetric decreases in the proportion of on-grown Artemia. Prior to the start of weaning at day 14, phyllosomata were re-stocked from 80 to 30 phyllosomata l 1. 88

101 Chapter 6 After the completion of the 4 d weaning period, formulated diets were presented as a stationary feed station, whereby the formulated diet was cut into a star-shape using an appropriately shaped cookie cutter (1.5 cm point to point x 0.5 cm) and skewered through the middle of the diet using a fibre glass rod (20 mm diameter) which was attached to the bottom of the larval rearing vessel. Stationary feed stations were removed after 4 h of feeding Chemical and Statistical Analysis Samples of finely ground raw ingredients were analysed in duplicate by standard laboratory methods essentially in accordance with the recommendations of the AOAC (1984), at the Microserve Laboratory, Perth. Dry matter (DM) was determined by oven drying at 105 C to constant weight; ash by ignition at 600 C for 2 h; total nitrogen by a micro-kjeldahl technique; (crude protein was calculated by multiplying total nitrogen by 6.25, irrespective of the nature of the nitrogen); and total lipid was determined by acid hydrolysis. All survival and measurement data are presented as mean ± S.E (n=3) and all statistical analyses were performed using SPSS Mean values from triplicate diet treatments from each of the four trials were compared with a one-way analysis of variance (ANOVA) followed by a Duncan s multiple range post hoc test to determine significant differences between each dietary treatment. For each analysis the assumptions of ANOVA were tested using a Levene s test for normality, and homogeneity of variance was checked using residual plots. Differences were regarded as significant when P< Results Trial I Replacement of live Artemia nauplii and greenshelltm mussel Growth and survival of phyllosomata offered combinations of live (Artemia nauplii) and fresh feeds with a formulated diet are shown in Figure 6.1A and Table 6.3. Phyllosomata offered Artemia nauplii at the 50% replacement level with greenshelltm mussel showed significantly (P<0.05) higher survival throughout the trial from day 11 onwards. 89

102 Chapter 6 Fig. 6.2 (A) Mean survival (± S.E) of Panulirus ornatus phyllosomata fed with different combinations of Artemia nauplii, formulated diet (form.) and greenshelltm mussel. (B) Mean survival (± S.E) of P. ornatus phyllosomata fed with alternate combinations of on-grown Artemia, formulated diet and greenshelltm mussel. Columns on each day with the same superscript are not significantly different (P<0.05). 90

103 Chapter 6 On day 17, the final day of the trial, survival of phyllosomata fed the 50% combination of Artemia nauplii and greenshelltm mussel was >55%. Complete replacement of Artemia nauplii with a formulated diet resulted in significant (P<0.05) mortality from day 11 onwards. Mean size was not affected due to the short length of the trial (Table 6.3). Table 6.3 Final mean (± S.E) growth measurements (mm) of Panulirus ornatus phyllosomata during trials I IV. TL, total length; CW, carapace width; CL, carapace length. Trial Treatment Day TL 1 CW 2 CL 3 I 100% Artemia nauplii ± 0.15 a 1.10 ± 0.05 a 1.11 ± 0.06 a 50% Artemia nauplii/ 50% ± 0.04 a 1.01 ± 0.02 a 1.35 ± 0.02 a Form. diet 50% Artemia nauplii/ 50% ± 0.09 a 0.98 ± 0.04 a 1.23 ± 0.07 a green mussel 100% Form. diet 17 n.d n.d n.d II 50% on-grown Artemia/ 50% green mussel 75% Form. Diet/ 25% ongrown Artemia 75% Form. diet/ 25% green mussel 50% Form. diet/ 50% green mussel l ± 0.08 a 1.09 ± 0.02 a 1.11 ± 0.04 a ± 0.06 a 1.12 ± 0.05 a 1.41 ± 0.04 a ± 0.03 a 1.08 ± 0.09 a 1.37 ± 0.06 a ± 0.10 a 1.03 ± 0.07 a 1.09 ± 0.08 a III Weaning started at Day ± 0.11 a 1.01 ± 0.02 a 1.36 ± 0.07 a Weaning started at Day ± 0.01 a 1.30 ± 0.06 a 1.69 ± 0.07 a Weaning started at Day ± 0.12 a 1.06 ± 0.07 a 1.37 ± 0.11 a 100% Artemia nauplii ± 0.14 a 1.19 ± 0.07 a 1.60 ± 0.11 a IV 44% CP ± 0.05 a 1.37 ± 0.02 a 1.77 ± 0.06 a 50% CP ± 0.04 a 1.29 ± 0.03 a 1.66 ± 0.08 a 50%-squid CP ± 0.11 b 1.04 ± 0.06 b 1.14 ± 0.07 b 100% on-grown Artemia ± 0.05 a 1.21 ± 0.07 a 1.61 ± 0.03 a * Means with superscripts are significantly different (P<0.05). 1 TL, anterior tip of the cephalothorax to posterior tip of the abdomen. 2 CW, widest distance across the carapace. 3 CL, anterior to posterior tips of carapace Trial II Replacement of live on-grown Artemia and greenshelltm mussel Mean size and survival of phyllosomata fed alternate combinations of live feed (ongrown Artemia), fresh feed (greenshelltm mussel) and formulated diet to those of trial I are shown in Figure 6.1B and Table 6.3. Phyllosomata fed the combinations of 75% formulated diet and 25% on-grown Artemia and combinations of 50% formulated diet and 50% on-grown Artemia, consistently showed higher survival throughout the trial 91

104 Chapter 6 Fig. 6.3 (A) Mean survival (± S.E) of Panulirus ornatus phyllosomata during three different days of weaning during early ontogeny. (B) Mean survival (± S.E) of P. ornatus phyllosomata when co-fed and weaned for 4 d using graded levels of protein. Superscripts depict a significant difference between cofeeding treatments (P<0.05). 92

105 Chapter 6 from day 11 onwards. Survival at day 17, the final day of the trial, was relatively high (>70%) for combinations of 75% formulated diet and 25% on-grown Artemia and the control (50% on-grown Artemia and 50% greenshelltm mussel). The two combinations of formulated diet and greenshelltm mussel performed poorly, with significantly (P<0.05) reduced survival from day 11 onwards. At day 17, the final day of the trial, survival of phyllosomata fed a combination of 75% formulated diet and 25% greenshelltm mussel was significantly (P<0.05) lower than the combination of 50% formulated diet and 50% greenshelltm mussel. Mean size was not affected due to the short length of the trial (Table 6.3) Trial III Conditioning phyllosomata for 14, 18, 21 d on live Artemia nauplii prior to weaning The effects of weaning phyllosomata onto 100% formulated diet during early ontogeny by sequentially replacing Artemia nauplii with increasing proportions of a formulated diet on survival and mean size of P. ornatus phyllosomata is shown in Fig. 6.2A and Table 6.3. At day 28, the final day of the trial, survival of phyllosomata weaned at day 14 was significantly (P<0.05) higher in comparison to when weaning began at days 18 and 21. Mean size was not affected by day of weaning; however, moulting to stage III occurred earlier in phyllosomata weaned at days 14 and 21 (Table 6.3) Trial IV Protein level and effect on weaning at day 14 The effects of formulated diets containing various levels of protein on growth and survival of P. ornatus phyllosomata when on-grown Artemia were sequentially withdrawn from day 14 are shown in Fig. 6.2B and Table 6.3. Phyllosomata fed the 50%-squid CP diet performed poorly, with significantly (P<0.05) worse survival than all the other diets tested from day 26. At day 28, the final day of the trial, there was no difference in survival of phyllosomata when weaned onto formulated diets containing 44% and 50% CP. Although, survival of phyllosomata receiving these two protein levels was not significantly different at the completion of the trial, survival was >15% when phyllosomata received the diet containing 44% CP, and was higher than the ongrown Artemia control. Growth and moulting were also affected by the level of protein, with no phyllosomata fed the 50%-squid CP diet moulting to stage IV (Table 6.3). 93

106 Chapter Discussion Trials I II - Partial replacement of Artemia nauplii, on-grown Artemia and greenshelltm mussel The results of the second replacement trial indicate that 75% of an Artemia ration can be successfully replaced with a formulated diet without compromising survival of P. ornatus phyllosomata. Although, survival of phyllosomata in this treatment was not significantly different from the control (50% formulated diet and 50% greenshelltm mussel) both treatments had survival greater than 65% which is required for commercial viability. The results of the first two trials also indicate that when 100% of the Artemia ration is replaced with a formulated diet, survival is low. A similar survival response was displayed by day 1 post-hatch P. ornatus phyllosomata when fed a range of different textured formulated diets (Johnston et al. 2005). The reduced survival was presumed to be associated with stress, resulting in increased cannibalism. Similar findings have been reported in replacement and weaning trials with mud crab, S. serrata larvae (Genodepa et al. 2004b). Survival of phyllosomata was also significantly (P<0.05) reduced when two inert items were fed in combination (formulated diet and greenshelltm mussel). However, when Artemia was used in combination with either a formulated diet or greenshelltm mussel, survival of phyllosomata was generally higher. Ritar et al. (2002) also reported reduced survival when feeding only blue mussel to J. edwardsii phyllosomata. Reduced survival of P. ornatus when fed two inert dietary items suggests that early-stage P. ornatus phyllosomata rely on visual feeding cues to stimulate a feeding response (Chapter 2; Appendix 3). This has also been shown to be the case for many marine fish larvae (Kolkovski et al. 1997a, 1997b; Cañavate and Fernández-Díaz 1999). Kolkovski et al. (1997b) reported up to 120% increase in ingestion rates of formulated diets when the formulated diet was offered in conjunction with Artemia to gilthead seabream, S. aurata. The influence of Artemia on enhancing acceptance of formulated diets was suggested by Kolkovski et al. (1997b) to be twofold: (1) the remote influence of visual and chemical cues helps in capture of formulated diets, and (2) the biochemical composition of the live food aids larval digestion and assimilation of the formulated diet. Although Artemia are required to provide visual cues to stimulate feeding, the mode of action of exogenous enzymes from ingested Artemia on digestion and assimilation of formulated diets for early stage P. ornatus phyllosomata is less clear. 94

107 Chapter 6 The consistently high survival of phyllosomata fed combinations of Artemia and formulated diet also suggests that the nutritional requirements of phyllosomata can be met more appropriately using a partial replacement feeding regime utilising adequate proportions of formulated diet than greenshelltm mussel gonad. The nutritional benefit of combining different types of foods for spiny lobster phyllosomata has been shown in previous studies. For example, J. (Sagmariasus) verreauxi phyllosomata survived through to pueruli when fed a combination of Artemia, mussel and fish larvae using green water culture methods (Kittaka et al. 1997). Mitchell (1971) found a mixed diet of zooplankton to be a more appropriate diet for P. interruptus phyllosomata than solely using Artemia. Similar results have been reported also for mud crab (Quinitio et al. 1999; Williams et al. 1999), penaeid prawn (Jones et al. 1993) and marine fish larvae (Callan et al. 2003) Trials III IV - Determination of an optimal period during early development to initiate weaning and effects of crude protein The results of the initial weaning trial outline the difficulty faced by P. ornatus hatcheries for complete replacement of Artemia with a 100% formulated diet ration during the early stages of development. The highest survival was achieved when weaning occurred after 14 days post-hatch. Generally, survival of phyllosomata was well below what is required for commercial viability. Despite this, survival of phyllosomata within the control treatment rapidly declined after day 14 and by day 28 was significantly (P<0.05) lower than phyllosomata weaned onto a formulated diet at day 14. This result indicates that the nutrient profile of Artemia nauplii appears inadequate for growth and survival of P. ornatus and should be replaced with either on-grown Artemia or combinations of 75% formulated diet and 25% on-grown Artemia. Kittaka and Abrunhosa (1997) observed similar nutritional deficiencies when feeding Artemia nauplii exclusively to stage I P. elephas phyllosomata. Ultimately, different palinurid species will have different nutrient requirements for growth and survival; hence species-specific formulated diets will almost certainly be required to be developed for phyllosomata of spiny lobster. Survival and growth of P. ornatus phyllosomata was significantly (P<0.05) affected by the level of crude protein included in the formulated diet. Diets containing 50%-squid CP resulted in significantly (P<0.05) reduced survival. Poor survival of phyllosomata that received the 50%-squid CP diet is likely to be associated with poor palatability and nutrient deficiencies. Reduced palatability of the 50%-squid CP diet 95

108 Chapter 6 was also observed in diet preferences trials whereby an inert ytterbium marker was incorporated into the diet formulation (Chapter 5). Panulirus ornatus phyllosomata like other crustaceans have a requirement for 10 essential amino acids (EAAs). Of the 10 EAAs, there are five that are commonly deficient in developed formulated diets for crustaceans: lysine, methionine, tryptophan, isoleucine and threonine (Houser and Akiyama 1997). If the requirement for these amino acids is met, the other five are usually present in sufficient quantity to meet requirements. As the predominant difference between the two 50% CP diets was the quantities of fish- and squid-meal; total mortality of P. ornatus phyllosomata fed the 50%-squid CP diet between days suggests one or a combination of the five commonly deficient amino acids was not present at a minimum required level, halting protein synthesis. Furthermore, fish meal generally is believed to be a superior source of protein, as the amino acid composition of fish meal closely resembles the composition in the muscle tissue of spiny lobsters (Kanazawa, 1997). Most studies on protein requirements of crustaceans such as prawns, shrimps and spiny lobsters have demonstrated the optimum dietary protein levels by feeding trials containing graded levels of crude protein (Deshimaru and Shigeno 1972; Teshima and Kanazawa 1984; Glencross et al. 2001; Smith et al. 2003; Ward et al. 2003). Although the results from this study can not suggest conclusively an optimum level of crude protein to be included in a larval formulated diet for this species, 44 50% CP is similar to the optimal level determined for adult P. ornatus (53% CP) by Smith et al. (2003). Until now, live and fresh feeds were the food that promoted better overall results in culturing the early stages of Panulirus ornatus phyllosoma. The high survival (>65% for commercial viability) of phyllosomata fed a combination of 75% formulated diet and 25% on-grown Artemia in this study suggests that a considerable proportion of the costs associated with the production and rearing of Artemia and therefore the final costs of producing phyllosomata could be reduced. Diet formulations for this species are currently not at the stage where live feeds can be totally replaced, as it appears Artemia are required at least during the early developmental stages to stimulate a visual feeding response. Future research that uses stage IV phyllosomata as an initial stage to introduce formulated diets as either a complete or partial replacement will most likely have greater success as the alimentary tract and digestive gland of this species is more developed indicating a greater digestive capacity. 96

109 97

110 Chapter 7 Chapter 7 Summary 7.1 Findings of the Present Study In my thesis I focused on ingestive and digestive morphology and digestive physiology of phyllosomata of the ornate spiny lobster, P. ornatus and how these changed during development. Information on ingestive and digestive capabilities during development of P. ornatus phyllosomata was applied to develop a formulated diet which could be used as either a complete replacement for live and fresh feeds, such as commercially available Artemia and mussel gonad or used in combination with these feeds as a partial replacement or supplement to reduce hatchery production costs during the larval phyllosomata phase. My investigations applied modern morphological and staining techniques to investigate the morphology of the mouthparts and foregut of early- and late-stage P. ornatus phyllosomata. Key structural changes of the mouthparts and foregut during ontogeny were identified and their possible significance in relation to development of a formulated diet. Emphasis focused on the physical properties of a diet, including size, shape, form, and buoyancy. Mouthpart morphology changes little during development suggesting both early- and late-stage phyllosomata have similar ingestive capabilities, and that external mastication is well developed directly following hatch from the egg. A number of key developmental changes of the mouthparts and foregut occur at stage IV, including a considerable increase in size of the mouth aperture suggesting that larger sized portions of prey can be ingested. Furthermore, the number of main brushes and lateral setae increase in the proto-proventriculus, indicating an ability to triturate the ingested food bolus to a limited capacity internally and a greater efficiency to sort and filter particles due to an increase in number and width of the ampullary channels forming the filter press. As there are no other considerable morphological changes of the mouthparts and foregut after stage IV, I would recommend that future diet trials would be presumably more successful, from this point of development onwards. The ultrastructure of the digestive gland of stage I P. ornatus phyllosomata and concentration and activity of digestive enzymes (stages I IV) were evaluated in Chapter 3. Throughout stage I, P. ornatus phyllosoma had a maximum of two digestive gland lobes (anterior and lateral) on either side of the foregut. The presence of R-, F- and B-cells were identified from day 0 (hatch), indicating that P. ornatus phyllosomata have the ability to digest and absorb nutrients from both prey and 98

111 Chapter 7 formulated diets from day 0. A wide spectrum of digestive enzymes, including proteases, trypsin, α-amylase, chitinase, laminarinase and lipase were detected during the first 32 d of culture (stages I IV), indicating P. ornatus can readily digest protein, lipid and carbohydrate, including chitin during all of these stages of development. High activities of chitinase and proteases suggests P. ornatus phyllosomata are able to digest the most common zooplankton forms within the Coral Sea such as calanoid and cyclopoid copepods, mysid shrimps, ectoprocts and cirriped larvae that have high protein content and a hard chitinous exoskeleton. The low level of lipase in comparison to J. edwardsii fed commercially enriched Artemia, suggests that use of algal enrichments in this study may not provide the required levels of fatty acids needed for optimum growth and survival of this species during the phyllosoma life history stage. The mouthparts of P. ornatus are consistent with a raptorial feeding behaviour, being used to grasp and manipulate diet particles physically into sizes that are consumed, implying that this form of feeding behaviour would place high demands on the physical integrity (stability) of formulated diets. Hence, a preliminary assessment of suitable binders and formats of formulated diets using stage I P. cygnus phyllosomata as a test species were evaluated in Chapter 4. Two different forms of formulated diets were developed (dry meal and gelatinous formulations) bound with different binders and drying methods. The gelatinous krill diet appeared most suited to the larval feeding biology of P. cygnus as it elicited a heightened feeding response following immersion, and was most stable in water up to 4 h. This form of diet also did not foul the maxillae during ingestion. The type of binder included in each formulation and method of drying influenced the appeal and water integrity of the formulated diet(s), a finding similar to diet development studies for penaeid prawns (Teshima and Kanazawa, 1983). Furthermore, evaluating the appeal in terms of numbers attracted to their choice of formulated diet using digital video analysis is an advantageous technique because, although developed formulated diets may contain the necessary feeding stimulants and nutrient profiles, they may still remain unacceptable (D'Abramo, 2002). Although gelatinous diet forms were well accepted by P. cygnus phyllosomata, diet forms with a firm/hard consistency were preferred by P. ornatus phyllosomata in diet preference trials in Chapter 5 and were therefore the basis of future diet formulations for this species. Little is currently known in the area of formulated diet development for aquaculture of palinurid phyllosomata (Nelson et al., 2005). The results of the diet preference trials in Chapter 5 whereby an inert ytterbium marker was added to each 99

112 Chapter 7 diet formulation to evaluate diet preferences represent a significant development in this field. Perhaps more importantly, the findings of this Chapter have shown that formulated diets are ingested by P. ornatus phyllosomata directly following hatch from the egg. Panulirus ornatus phyllosomata clearly preferred formulated diets with a firm/hard consistency, and the raptorial feeding behaviour following hatch of this species was accentuated in the particle size trial with no clear determination of an optimum particle size range. However, P. ornatus phyllosomata ingested more diet particles within a μm size range. Preference for diet particles within the μm size range is well within the mean size class of the most abundant natural prey items in the Coral Sea that P. ornatus phyllosomata are likely to encounter (Rissik and Suthers, 2000). Furthermore, preference for particles within this size range suggests that Artemia nauplii, which are predominately used in hatchery culture, are of an inadequate size range ( μm) and should be either on-grown or replaced with an appropriate formulated diet. Preference for the hard diet form resulted from either better palatability or a higher availability caused by a fast sink rate, as phyllosomata spent considerable time feeding from the bottom and not mid-water. This form of diet is highly desirable for hatcheries, as diets with low moisture content offer advantages in terms of long-term frozen storage. Addition of feeding stimulants and level of crude protein had significant effects on diet preference by P. ornatus phyllosomata. Although feeding stimulants have been identified to have a chemosensory effect on crustaceans, this is the first indication that they could be incorporated to reduce feed costs for aquaculture of P. ornatus phyllosomata. Incorporation of the quaternary amino compound betaine or glycine could have substantial advantages for hatcheries other than the direct effects on feed intake and growth. Increased appeal of formulated diets containing betaine and glycine may reduce wastage and also leakage of water soluble components due to a quicker search and feeding response time. The amino acid taurine was found to have no effect on diet preference and instead could be classified as a feeding inhibitor for this species. The different detection and excitatory capacity of feeding stimulants was highlighted also in Chapter 5, and appears to be highly variable within the Palinuridae, with large interspecific differences possibly reflecting the presence or absence of these chemicals in their wild prey. Diet preferences were affected also by the level of crude protein (CP), and percent inclusion of different protein-based feed ingredients. Panulirus ornatus 100

113 Chapter 7 phyllosomata preferred diets containing 44% CP than any of the other diets. Preference for diets containing 44% CP is also in agreement with increased survival of P. ornatus phyllosomata over 28 d during the second weaning experiment in Chapter 6. Panulirus ornatus phyllosomata also ingested diets containing >50% CP, however, diet preferences were lower for the 50% CP diet containing a large amount of squid meal (50%-squid CP). These findings were represented also in the second weaning experiment in Chapter 6, with reduced survival of phyllosomata ingesting the 50%- squid CP diet after day 18, with 100% mortality after day 24. The high mortality and reduced feed intake indicate that the inclusion of a range of marine-based protein sources provides a more palatable, attractive and nutritionally complete diet for this species. Due to rapid mortality of P. ornatus phyllosomata when fed a formulated diet directly following hatch (Appendix 3) and the fact that P. ornatus phyllosomata are stimulated visually to feed by motile prey, provided scope to potentially replace a proportion of the requirement for Artemia with adequate proportions of a formulated diet. Various combinations of Artemia nauplii, on-grown Artemia, greenshelltm mussel gonad and a formulated diet were evaluated in Chapter 6. The results of both replacement trials provided the first significant indication that a combination of formulated diet/greenshelltm mussel gonad with Artemia may provide a more balanced level of nutrition for growth and survival of P. ornatus than Artemia alone. The rapid mortality of phyllosomata fed 100% formulated diet in replacement trial I and poor survival of phyllosomata in replacement trial II that were receiving two inert dietary items (greenshelltm mussel and formulated diet) supported earlier observations that P. ornatus phyllosomata are stimulated to feed by visual cues (Johnston et al., 2005; Appendix 3). The most significant result of the two replacement trials for hatcheries was the potential to replace 75% of the on-grown Artemia ration with a formulated diet. Survival >70% using this feeding protocol would considerably reduce hatchery costs by reducing the need for large quantities of Artemia. Furthermore, although phyllosomata were provided from two different females in both replacement trials, the use of on-grown Artemia enriched with algae provided greater survival of P. ornatus phyllosomata than Artemia nauplii, possibly due to a more balanced nutritional profile and larger size to capture and ingest. The potential use of weaning and co-feeding at various days of development was assessed in a second series of experiments in Chapter 6 to determine if an optimum time to wean P. ornatus phyllosomata onto a formulated diet could be established. Phyllosomata weaned at day 14 had significantly higher survival than all 101

114 Chapter 7 treatments including the Artemia nauplii control. However, survival <20% at day 28 indicates the importance of incorporating a small amount of live prey to stimulate feeding. Survival of phyllosomata was improved in the second co-feeding trial using on-grown Artemia and formulated diet containing 44% CP. From the results of all the trials in this thesis, a basal diet formulation for future diet development studies is provided in Table 7.1. Table 7.1. Ingredient composition of a basal diet for Panulirus ornatus phyllosomata for future diet development studies. Composition based on diet stability (Chapter 4), diet preference (Chapter 5) and long term grow-out trials (Chapter 6). Ingredient g kg 1 Krill meal 300 Squid meal 16 GreenshellTM mussel 30 Fish meal 90 FD blood worms 19 Krill hydrolysate 60 Cod liver oil 50 Corn oil 10 Lecithin, soy 15 Wheat flour 100 Wheat gluten 55 Cholesterol 20 Vitamin and minerals 20 Diatomaceous earth 150 Gelatine 50 Betaine 15 * A complete list of suppliers, their contact details and product numbers are provided in Appendix Suggestions for Future Research Although the present study has made significant contributions to the understanding of the ingestive and digestive morphology and digestive physiology of P. ornatus phyllosomata, including initial development of formulated diets for aquaculture of this species, the scope for complete replacement of Artemia and fresh feeds such as mussel gonad with a formulated diet currently remains unknown. As there are many complexities in the design of formulated diets related to ingestive and digestive morphology and digestive physiology of P. ornatus phyllosomata, the scope for future research is quite extensive. However, areas of inquiry for future research should focus on: 1) Detailed descriptions of the morphology of the mouthparts and foregut of stages VII and VIII is required as these were unable to be obtained in the 102

115 Chapter 7 present study. This information would aid in the development of a formulated diet for the mid-late stages of phyllosoma development and would help to identify possible size ranges of formulated diets based on the size of the mouth aperture and width of the ampullary channels (filter channels). Furthermore, multi-photon confocal microscopy of the foregut of early-, mid- and late-stage phyllosomata would provide a 3-D image of the proto-proventriculus and would provide a computer generated model, showing the morphological changes of the foregut, in particular the complexity of the filter press. 2) Analysis of digestive enzymes from stage V through to stage XI and the analysis of wild-caught phyllosomata would aid in identifying potential feeding transitions, and indicate the relative importance of protein, carbohydrate and lipid to include in formulated diets for hatchery reared phyllosomata to obtain similar nutrition and growth to that which they would receive in the wild. 3) The use of endogenous enzymes (in vitro enzyme digestion) extracted from P. ornatus phyllosomata at different stages of development could be investigated to determine digestibility of feed ingredients for future diet formulations. Although this method will never replace the apparent digestibility technique for determination of nutrient digestibility, the technique offers the possibility of reducing significantly the time and costs required to evaluate the digestibility of a feed/feed ingredient. This technique is most useful for protein digestibility. As marine-based protein sources are the most expensive components of a formulated diet, a technique that would allow for the selection of feed ingredients that are highly digestible and presumably assimilated more efficiently is extremely important to the future success of diet development for this species. 4) The ultrastructure of the digestive gland using transmission and light microscopy could be expanded to include other developmental stages. More importantly Rodriguez et al. (1999) detected the uptake of proteins, lipids and carbohydrates into the lumen of the digestive gland of P. japonicus phyllosomata using a range of marine- and plant-based oils, suggesting the possibility that particulate and dissolved organic matter can be utilised by P. japonicus phyllosomata, perhaps as a supplementary source of nutrients to formulated diets. A similar approach may be effective for P. ornatus 103

116 Chapter 7 phyllosoma and may reduce costs of formulated diets for this species. Particularly, as late-stage Panuliru spp., phyllosomata are known to store lipid in the digestive gland as a source of energy to actively swim from the outer reefs/continental shelf to settle on the coast as pueruli. 5) Although the physical integrity of formulated diets was assessed in the current study by evaluating the potential use of a variety of binders and drying techniques, determination of leach rates of water soluble vitamins and amino acids would be beneficial to the selection of appropriate binders, pellet extrusion methods and drying processes for future diet formulations. 6) Incorporation of the inert marker ytterbium oxide into diet formulations was an effective method to evaluate diet preferences, particularly both physical and nutritional characteristics of diets. Future use of this marker in diet formulations could be used to evaluate; (1) feed digestibility, (2) optimal ration size(s), (3) combinations or incorporation of other feeding stimulants, (4) effects of lipid levels ( 15%), and (5) protein levels <44% CP, on diet preference, rates of ingestion/consumption and nutrient digestibly. Ultimately, the physical or nutritional variables tested that gain favourable results should be incorporated in future grow-out trials to assess growth and survival. 7) As marine-based protein sources are the most expensive component of formulated diets for marine larvae, future formulations incorporating plantbased protein sources could be investigated as P. ornatus phyllosomata exhibited high protease levels and had higher survival and growth when fed a 44% CP diet, which was the lowest protein-based diet developed in this study. 8) Although it is clear that survival of P. ornatus phyllosomata was acceptable (>70%) when fed a combination of Artemia and formulated diet, their survival decreased considerably after withdrawal of Artemia when weaning. This result indicated that Artemia are required to stimulate a visual feeding response. Furthermore, this result highlights that formulated diets may never totally replace the need for production of live feeds for the aquaculture of this species, especially during the early phyllosoma stages. Diet buoyancy is a critical issue to the success of total replacement of live feeds. This could be achieved by experimenting with formulating diets with different levels and types of lipid or 104

117 Chapter 7 different deliveries of aeration into the culture vessel. However, as additional aeration is not typically provided in hatchery culture of P. ornatus phyllosomata, alternate tank designs that keep diet particles in suspension are areas of critical importance. In my thesis, I achieved all of my initial objectives and have found commercially relevant results, and have identified a number of relevant lines of research for future studies. 105

118 Literature Cited Literature Cited ABARE, Australian Fisheries Statistics Canberra, Australia, pp. 76. Abranhosa, F., Kittaka, J., Effects of starvation on the first larvae of Homarus americanus (Decapoda, Nephropidae) and phyllosomas of Jasus verreauxi and J. edwardsii (Decapoda, Palinuridae). Bulletin of Marine Science 61, Albro, P., Hall, R., Corbett, J., Schroeder, J., Activation of nonspecific lipase E.C by bile salts. Biochimica et Biophysica Acta 835, AOAC, Official Methods of Analysis. AOAC, Washington, DC, pp Austreng, E., Digestibility determination in fish using chromic oxide marking and analysis of contents from different segments of the gastrointestinal tract. Aquaculture 13, Austreng, E., Storebakken, T., Skinlo Thomassen, M., Refstie, S., Thommassen, Y., Evaluation of selected trivalent metal oxides as inert markers used to estimate apparent digestibility in salmonids. Aquaculture 188, Awapara, J., Free amino acids in invertebrates: A comparative study of their distribution and metabolism. In: Holden, J. (Ed.), Amino acid pools. Elsevier, Amsterdam, pp Baisre, J., Phyllosoma larvae and the phylogeny of Palinuroidea (Crustacea; Decapoda): A review. Australian Journal of Marine and Freshwater Research 45, Barbato, J., Daniel, P., Chemosensory activation of an antennular grooming behaviour in the spiny lobster, Panulirus argus, is tuned narrowly to L-glutamate. Biological Bulletin, Marine Biological Laboratories, Woods Hole 193, Barclay, M., Irvin, S., Williams, K., Smith, D., Comparison of diets for the tropical spiny lobster Panulirus ornatus: Astaxanthin-supplemented feeds and mussel flesh. Aquaculture Nutrition 12, Barnard, K., Report on a collection of Crustacea from Portuguese East Africa. Transactions Royal Society South Africa 13, Bate, C., Annals Magazine Natural History 7, pp Bate, C., Report on the Crustacea Macrura collected by H.M.S. Challenger during the years Report Scientific Results Voyage Challenger, (Zool), 24, pp Batham, E., The first three larval stages and feeding behaviour of phyllosoma of the New Zealand palinurid crayfish Jasus edwardsii (Hutton, 1875). Transactions and Proceedings of the New Zealand Institute 9, Berry, P., The biology of the spiny lobster Palinurus delagoae Barnard, off the coast of Natal, South Africa. Investigation Reports of the Oceanographic Research Institute of South Africa 31, Biedenbach, J., Smith, L., Lawerence, A., Use of a new spray-dried algal product in penaeid larviculture. Aquaculture 86, Biesiot, P., Capuzzo, J., Changes in digestive enzyme activities during early development of the American lobster Homarus americanus Milne Edwards. Journal of Experimental Marine Biology and Ecology 136, Bloch, M., M. Marcus Elieser Bloch's..., ausübenden Arztes zu Berlin, Oekonomische Naturgeschichte der Fische Deutschlands. Berlin, pp Boone, L., Penaeus vannamei Boone, Bulletin of the American Museum of Natural History 63, 173. Booth, J., New Zealand s rock lobster fisheries. In: Phillips, B., Kittaka, J. (Eds.), Spiny Lobsters: Fisheries and Culture. Fishing News Books, London, pp Booth, J., Phillips, B., Early life history of spiny lobsters. Crustaceana 66, Bourne, D., Young, N., Webster, N., Payne, M., Salmon, M., Demel, S., Hall, M., Microbial community dynamics in a larval aquaculture system of the tropical rock lobster, Panulirus ornatus. Aquaculture 242, Bouvier, E., Panulirus inflatus (Bouvier, 1895). Bulletin du Musuem D'Histoire Naturelle 1, 8. Bradford, M., A rapid sensitive method for the quantification of microgram quantities of protein utilising the principle of protein-dye binding. Analytical Biochemistry 72,

119 Literature Cited Briones-Forzan, P., Lozano-Alverez, E., The spiny lobster fisheries in Mexico. In: Phillips, B., Kittaka, J. (Eds.), Spiny Lobsters: Fisheries and Culture. Fishing News Books, London, pp Butcher, R., An introductory account of the smaller algae of British coastal waters. Part I: Introduction and Chlorophyceae. Vol. ser. IV (Part 1). Fisheries Investment Report: Ministry of Agriculture and Fisheries, Great Britain, pp. 74. Butler, M., Hernkind, W., Puerulus and juvenile ecology. In: Phillips, B., Kittaka, J. (Eds.), Spiny Lobsters: Fisheries and Culture. Fishing News Books, London, pp Callan, C., Jordaan, A., Kling, L., Reducing Artemia use in the culture of Atlantic cod (Gadus morhua). Aquaculture 219, Cañavate, J., Fernández-Díaz, C., Influence of co-feeding larvae with live and inert diets on weaning the sole, Solea senegalensis, onto commercial dry feeds. Aquaculture 174, Castell, J., Kean, J., D'Abramo, L., Conklin, D., A standard reference diet for crustacean nutrition research. I. Evaluation of two formulations. Journal of the World Aquaculture Society 20, Cox, S., Bruce, M., Feeding behaviour and associated sensory mechanisms of stage I-III phyllosoma of Jasus edwardsii and Jasus verreauxi. Journal of the Marine Biological Association of the United Kingdom 83, Cox, S., Johnston, D., 2003a. Developmental changes in the structure and function of mouthparts of phyllosoma larvae of the packhorse lobster, Jasus verreauxi (Decapoda: Palinuridae). Journal of Experimental Marine Biology and Ecology 296, Cox, S., Johnston, D., 2003b. Feeding biology of spiny lobster larvae and implications for culture. Reviews in Fisheries Science 11, Cox, S., Feeding biology and artificial diet development for the culture of Jasus edwardsii and Jasus verreauxi spiny lobster Phyllosomas. PhD thesis, University of Auckland, pp Cox, S., Johnston, D., Developmental changes in foregut functioning of packhorse lobster, Jasus (Sagmariasus) verreauxi (Decapoda: Palinuridae), phyllosoma larvae. Marine and Freshwater Research 55, D'Abramo, L., 2002, Challenges in developing successful formulated feed for culture of larval fish and crustaceans. In: Cruz-Suárez, L., Ricque-Marie, D., Tapia-Salazar, M., Gaxiola- Cortés, M., Simoes, N. (Eds.), Avances en Nutrición Acuícola VI. Memorias del VI Simposium Internacional de Nutrición Acuícola, 3 al 6 de Septiembre del Cancún, Quintana Roo, México., pp Dall, W., Moriarty, D., Functional aspects of nutrition and digestion. In: Mantel, L., Bliss, D. (Eds.), The Biology of Crustacea. Vol. 5. Internal Anatomy and Physiological Regulation. Academic Press, New York, pp De Man, J., On some species of the genus Palaemon Fabr. with descriptions of two new forms. Notes from the Leyden Museum 1, Dennis, D., Pitcher, C., Prescott, J., Skewes, T., Severe mortality in a breeding population of ornate rock lobster Panulirus ornatus (Fabricius) at Yule Island, Papua New Guinea. Journal of Experimental Marine Biology and Ecology 162, Dennis, D., Pitcher, C., Skewes, T., Distribution and transport pathways of Panulirus ornatus (Fabricius, 1798) and Panulirus spp. larvae in the Coral Sea, Australia. Marine and Freshwater Research 52, Dennis, D., Skewes, T., Pitcher, C., Habitat use and growth of juvenile ornate rock lobsters, Panulirus ornatus (Fabricius, 1798) in Torres Strait, Australia. Marine and Freshwater Research 48, Dennis, D., Yimin, Y., Pitcher, C., Skewes, T., 2004, Ecology and stock assessment of the ornate rock lobster Panulirus ornatus population in Torres Strait, Australia. In: Williams, K. (Ed.), Proceedings of a workshop held at the Institute of Oceanography, Nha Trang, Vietnam, CSIRO, Canberra, Australia, pp Deshimaru, O., Shigeno, K., Introduction to the artificial diet for prawn Penaeus japonicus. Aquaculture 1, Duggan, S., McKinnon, A., The early developmental stages of the spiny lobster Panulirus ornatus (Fabricius, 1798) cultured under laboratory conditions. Crustaceana 76,

120 Literature Cited Erlanger, B., Kokowsky, N., Cohen, W., The preparation and properties of two chromogenic substrates of trypsin. Archives of Biochemistry and Biophysics 95, Fabricius, J., Mantissa Insectorum sistens eorum species nuper detectes adiectis characteribus genericis, differentiis specificis, emendationibus, observationibus. 1, pp Fabricius, J., Supplementum Entomologiae Systematicae 400, pp Factor, J., Morphology of the mouthparts of larval lobsters, Homarus americanus (Decapoda: Nephropidae), with special emphasis on their setae. Biological Bulletin 154, Factor, J., The digestive system. In: Factor, J. (Ed.), Biology of the lobster Homarus americanus. Academics press, New York, pp Fagbenro, O., Jauncey, K., Growth and protein utilization by juvenile catfish (Clariasgariepinus) fed dry diets containing co-dried lactic acid - fermented fish silage and protein feedstuffs. Bioresource Technology 51, Farmanfarmaian, A., Lauterio, T., Ibe, M., Improvement of the stability of commercial feed pellets for the giant shrimp (Macrobrachium rosenbergii). Aquaculture 27, Farmer, A., The functional morphology of the mouthparts and pereopods of Nephrops norvegicus (L.) (Decapoda: Nephropidae). Journal of Natural History 8, Felgenhauer, B., External anatomy and integumentary structures. In: Harrison, F., Humes, A. (Eds.), Microscopic Anatomy of Invertebrates. Wiley-Liss, New York, pp Felix, N., Sudharsan, M., Effect of glycine betaine, a feed attractant affecting growth and feed conversion of juvenile freshwater prawn Macrobrachium rosenbergii. Aquaculture Nutrition 10, Floreto, E., Brown, P., Bayer, R., The effects of krill hydrolysate-supplemented soya-bean based diets on the growth, colouration, amino and fatty acid profiles of juvenile American lobster, Homarus americanus. Aquaculture Nutrition 7, Forskål, P., Descriptiones animalium, avium, amphibiorum, piscium, insectorum, vermium. Havniae, Carsten Niebuhr, pp Forster, J., Some methods of binding prawn diets and their effects on growth and assimilation. ICES Journal of Marine Science 34, Garatun-Tjeldstø, O., Ottera, H., Julshamn, K., Austreng, E., Food ingestion in juvenile cod estimated by inert lanthanide markers - effects of food particle size. ICES Journal of Marine Science 63, Garm, A., Revising the definition of the crustacean seta and setal classification systems based on examinations of the mouthpart setae of seven species of decapods. Zoological Journal of the Linnean Society 142, Genodepa, J., Southgate, P., Zeng, C., 2004a. Diet particle size preference and optimal ration for mud crab, Scylla serrata, larvae fed microbound diets. Aquaculture 230, Genodepa, J., Zeng, C., Southgate, P., 2004b. Preliminary assessment of a microbound diet as an Artemia replacement for mud crab, Scylla serrata, megalopa. Aquaculture 2004, George, R., Description of Panulirus cygnus sp. nov., the commercial crayfish (or spiny lobster) of Western Australia. Journal of the Royal Society of Western Australia 5, George, R., Tropical spiny lobsters, Panulirus sp., of Western Australia (and the Indo- West Pacific). Journal of the Royal Society of Western Australia 51, George, R., Tectonic plate movements and the evolution of Jasus and Panulirus spiny lobsters (Palinuridae). Marine and Freshwater Research 48, George, R., Evolution of life cycles, including migration, in spiny lobsters (Palinuridae). New Zealand Journal of Marine and Freshwater Research 39, George, R., Main, A., The evolution of spiny lobsters (Palinuridae): A study of evolution in the marine environment. Evolution 21, Gibson, R., Barker, P., The decapod hepatopancreas. Oceanography and Marine Biology: An Annual Review 17, Girard, C., A revision of the north American Astaci, with observations on their habits and geographical distribution. Proceedings of the Academy of Natural Sciences of Philadelphia 6,

121 Literature Cited Glencross, B., Smith, M., Curnow, J., Smith, D., Williams, K., The dietary protein and lipid requirements of post-puerulus western rock lobster, Panulirus cygnus. Aquaculture 199, Goldstein, J., Matsuda, H., Butler, M., Success! Caribbean spiny lobsters, Panulirus argus is cultured from egg to juvenile for the first time, The Lobster Newsletter, pp Gray, H., The western rock lobster Panulirus cygnus. Book 1: A Natural History. Westralian Books, Geraldton, pp Groeneveld, J., Stock assessment, ecology and economics as criteria for choosing between trap and trawl fisheries for spiny lobster Palinurus delagoae. Fisheries Research 48, Guthrie, K., Rust, M., Langdon, C., Barrows, F., Acceptability of various microparticulate diets for first-feeding walleye Stizostedion vitreum larvae. Aquaculture Nutrition 6, Hall, M., Closed cycle breeding of Crustacea with special reference to rock lobsters. Fisheries Research Development Council, Hobart, pp Hammer, H., Bishop, C., Watts, S., Activities of three digestive enzymes during development in the crayfish Procambarus clarkii (Decapoda). Journal of Crustacean Biology 20, Harpaz, S., Kahan, D., Galun, R., Moore, I., Responses of freshwater prawn, Macrobrachium rosenbergii, to chemical attractants. Journal of Chemical Ecology 13, Heinen, J., Evaluation of some binding agents for crustacean diets. Progressive Fish Culturist 43, Herbst, J., Cancer (Astacus) polyphagus Herbst, Versuch einer Naturgeschichte der Krabbenund Krebse 2, 90. Higgins, K., The structure and function of the digestive gland of rock lobster (Jasus edwardsii) phyllosoma. M.Sc. thesis, University of Tasmania, pp Hinton, D., Corey, S., The mouthparts and digestive tract in the larval stages of Homarus americanus. Canadian Journal of Zoology 57, Holthuis, L., FAO Species Catalogue. Vol. 13. Marine lobsters of the world. An annotated and illustrated catalogue of species of interest to fisheries known to date. FAO, Rome, pp Houser, R., Akiyama, D., Feed Formulation Principles. In: D'Abramo, L., Conklin, D., Akiyama, D. (Eds.), Crustacean Nutrition: Advances In World Aquaculture Volume 6. The World Aquaculture Society, Baton Rouge, Louisiana, pp Hunter, E., Biology of the European spiny lobster, Palinurus elephas (Fabricius, 1787) (Decapoda, Palinuridae). Crustaceana 72, Hutton, F., Description of two new species of Crustacea from New Zealand (Panulirus edwardsii). Transactions and Proceedings of the New Zealand Institute 7, Icely, J., Nott, J., Accumulation of copper within the hepatopancreatic caeca of Corophium volutator (Crustacea: Amphipoda). Marine Biology 57, Icely, J., Nott, J., Digestion and absorption: Digestive system and associated organs. In: Harrison, F., Humes, A. (Eds.), Microscopic Anatomy of Invertebrates. Wiley-Liss Inc., New York, pp Inoue, M., On the relation of amount of food taken to the density and size of food and water temperature in rearing the phyllosoma of the Japanese spiny lobster, Panulirus japonicus (V. Siebold). Nippon Suisan Gakkaishi 31, Inoue, M., Studies on the cultured phyllosoma larvae of the Japanese spiny lobster Panulirus japonicus. I. Morphology of the phyllosoma. Nippon Suisan Gakkaishi 29, (in Japanese). Inoue, N., Sekiguchi, H., Distribution of late-stage larvae of Panulirus japonicus in the Kuroshio subgyre. Marine and Freshwater Research 52, Ishikawa, M., Teshima, S., Kanazawa, A., Koshio, S., Evacuation of inert markers in digestibility determination, 5α-cholestane and chromic oxide in the prawn Penaeus japonicus. Fisheries Science 62, Jacobs, W., Untersuchungen über die Cytologie der Sekretbildung in der Mitteldarmdrüse vonastacus leptodactylus. Zeitschrift Fur Zellforschung Und Mikroskopische Anatomie 8,

122 Literature Cited Jeffs, A., Hooker, S., Economic feasibility of aquaculture of spiny lobsters Jasus edwardsii in temperate waters. Journal of the World Aquaculture Society 31, Jeffs, A., Willmott, M., Wells, R., The use of energy stores in the puerulus of the spiny lobster Jasus edwardsii across the continental shelf of New Zealand. Comparative Biochemistry and Physiology, Part A 123, Johnson, M., On palinurid and scyllarid lobster larvae and their distribution in the South China Sea (Decapoda, Palinuridae). Crustaceana 21, Johnson, M., Knight, M., The phyllosoma larvae of the spiny lobster Panulirus inflatus (Bouvier). Crustaceana 10, Johnston, D., Ontogenetic changes in digestive enzyme activity of the spiny lobster, Jasus edwardsii (Decapoda; Palinuridae). Marine Biology 143, Johnston, D., Freeman, J., Dietary preference and digestive enzyme activities as indicators of trophic resource utilization by six species of crab. Biological Bulletin 208, Johnston, D., Ritar, A., Mouthpart and foregut ontogeny in phyllosoma larvae of the spiny lobster Jasus edwardsii (Decapoda: Palinuridae). Marine and Freshwater Research 52, Johnston, D., Alexander, C., Yellowlees, D., Isolation and characterisation of a trypsin from the slipper lobster Thenus orientalis Lund. Archives of Biochemistry and Biophysics 324, Johnston, D., Alexander, C., Yellowlees, D., Epithelial cytology and function in the digestive gland of Thenus orientalis (Decapoda: Scyllaridae). Journal of Crustacean Biology 18, Johnston, D., Ritar, A., Thomas, C., 2004a. Digestive enzyme profiles reveal digestive capacity and potential energy sources in fed and starved spiny lobster (Jasus edwardsii) phyllosoma larvae. Comparative Biochemistry and Physiology, Part B 138, Johnston, D., Ritar, A., Thomas, C., Jeffs, A., 2004b. Digestive enzyme profiles of spiny lobster Jasus edwardsii phyllosoma larvae. Marine Ecology Progress Series 275, Johnston, M., Johnston, D., Digestive capabilities reflect the major food sources in three species of talitrid amphipods. Comparative Biochemistry and Physiology, Part B 140, Johnston, M., Johnston, D., Knott, B., Jones, C., 2005, Mouthpart and foregut ontogeny in phyllosomata of Panulirus ornatus and their implications for development of a formulated larval diet. In: Hendry, C., Van Stappen, G., Wille, M., Sorgeloos, P. (Eds.), Larvae th Fish and Shellfish Larviculture Symposium, European Aquaculture Society, Special Publication No. 36, Gent, Belgium, pp Jones, C., Linton, L., Horton, D., Bowman, W., Effect of density on growth and survival of ornate rock lobster, Panulirus ornatus (Fabricius, 1798), in a flow-through raceway system. Marine and Freshwater Research 52, Jones, D., Crustacean larval microparticulate diets. Reviews of Fisheries Science 6, Jones, D., Holland, D., Jabborie, S., Current status of microencapsulated diets for aquaculture. Applied Biochemistry and Biotechnology 10, Jones, D., Kamarudin, M., Le Vay, L., The potential for replacement of live feeds in larval culture. Journal of the World Aquaculture Society 24, Jones, D., Kanazawa, A., Abdel Rahman, S., Studies on the presentation of artificial diets for rearing larvae of Penaeus japonicus Bate. Aquaculture 17, Jones, D., Kumlu, M., Le Vay, L., Fletcher, D., The digestive physiology of herbivorous, omnivorous and carnivorous crustacean larvae: A review. Aquaculture 155, Jussila, J., Evans, L., Growth and condition of marron Cherax tenuimanus fed pelleted diets of different stability. Aquaculture Nutrition 4, Kagwade, P., Fecundity in the spiny lobster Panulirus polyphagus (Herbst). Journal of the Biological Association of India 30, Kamarudin, M., Jones, D., Le Vay, L., Abidin, A., Ontogenetic changes in digestive enzyme activity during larval development of Macrobrachium rosenbergii. Aquaculture 123, Kanazawa, A., Penaeid Nutrition. In: Pruder, G., Langdon, C., Conklin, D. (Eds.), Proceedings of the Second International Conference on Aquaculture Nutrition: Biochemical and Physiological Approaches to Shellfish Nutrition. Louisiana State University, Baton Rogue, pp

123 Literature Cited Kanazawa, A., 1989, Microparticulate feeds for penaeid larvae. In: (Eds.), Advances in Tropical Aquaculture, Aquacop Ifremer, Actes de Colloque, Tahiti, French Polynesia, pp Kanazawa, A., Research In Asia. In: D'Abramo, L., Conklin, D., Akiyama, D. (Eds.), Crustacean Nutrition: Advances In World Aquaculture Volume 6. The World Aquaculture Society, Baton Rouge, Louisiana, pp Kanazawa, A., Teshima, S., 1988, Microparticulate diets for fish larvae. In: Sparks, A. (Ed.), New and innovative advances in biologie/engineering with potential for use in Aquaculture, National Oceanic and Atmospheric Association, National Marine and Fisheries Service, Seattle, pp Kennett, J., Cenozoic evaluation of Antarctic glaciation, the Circum-Antarctic Ocean, and thier impact on global paleoceanography. Journal of Geophysical Research 82, Kennett, J., Marine Geology. Prentice Hall, Englewood Cliffs, New Jersey, pp Kitabayashi, K., Shuda, K., Kakamura, K., Ishikaua, S., Studies on formula feed for Kuruma prawn on the growth-promoting effects of protein level in a diet an reexamination of ingredients used. Bulletin of Tokai Regional Fisheries Research Laboratory 65, Kittaka, J., Culture of the palinurid Jasus lalandii from egg stage to puerulus. Nippon Suisan Gakkaishi 54, Kittaka, J., Larval rearing. In: Phillips, B., Cobb, J., Kittaka, J. (Eds.), Spiny Lobster Management. Fishing News Books, London, pp Kittaka, J., Culture of larval spiny lobsters: A review of work done in northern Japan. Marine and Freshwater Research 48, Kittaka, J., 1999, Importance of three major factors for successful phyllosoma culture. In: Hart, P., van Barneveld, R. (Eds.), Proceedings of the 1999 FRDC Rock Lobster Propagation Workshop, January 1999, Fisheries Research and Development Corporation, Hobart, Australia, pp Kittaka, J., Culture of spiny lobsters. In: Phillips, B., Kittaka, J. (Eds.), Spiny Lobsters: Fisheries and Culture. Fishing News Books, London, pp Kittaka, J., Abrunhosa, F., Characteristics of palinurids (Decapoda; Crustacea) in larval culture. Hydrobiologia 358, Kittaka, J., Ikegami, E., Culture of the palinurid Palinurus elephas from egg stage to puerulus. Nippon Suisan Gakkaishi 54, Kittaka, J., Kimura, K., Culture of the Japanese spiny lobster Panulirus japonicus from egg to juvenile stage. Nippon Suisan Gakkaishi 55, Kittaka, J., Iwai, M., Yoshimura, M., Culture of a hybrid of spiny lobster genus Jasus from egg stage to puerulus. Nippon Suisan Gakkaishi 54, Kittaka, J., Ono, K., Booth, J., Complete development of the green rock lobster, Jasus verreauxi from egg to juvenile. Bulletin of Marine Science 61, Kolkovski, S., Arieli, A., Tandler, A., 1997a. Visual and chemical cues stimulate microdiet ingestion in sea bream larvae. Aquaculture International 5, Kolkovski, S., Koven, W., Tandler, A., 1997b. The mode of action of Artemia in enhancing utilisation of microdiet by gilthead seabream Sparus aurata larvae. Aquaculture 155, Kolkovski, S., Tandler, A., Kissil, G., Gertler, A., The effect of dietary exogenous digestive enzymes on ingestion, assimilation, growth and survival of gilthead seabream (Sparus aurata, Sparidae, Linnaeus) larvae. Fish Physiology and Biochemistry 12, Kovalenko, E., D'Abramo, L., Ohs, C., Buddington, R., A successful microbound diet for the larval culture of freshwater prawn Macrobrachium rosenbergii. Aquaculture 210, Kurmaly, K., Jones, D., Yule, A., Acceptability and digestion of diets fed to larval stages of Homarus gammarus and the role of dietary conditioning behaviour. Marine Biology 106, Kurokawa, T., Shiraishi, M., Suzuki, T., Quantification of exogenous protease derived from zooplankton in the intestine of Japanese sardine (Sardinops melanostictus) larvae. Aquaculture 161, Langdon, C., Microparticle types for delivering nutrients to marine fish larvae. Aquaculture 227,

124 Literature Cited Langdon, C., Levine, D., Jones, D., Microencapsulate feeds for marine suspensionfeeders. Journal of Microencapsulation 2, Latreille, P., Histoire naturelle générale et particulière des Crustacés et des Insectes. Ouvrage faisant suite à l'histoire naturelle générale et particulière, composée par Leclerc de Buffon, et rédigée par C.S. Sonnini, membre de plusieurs Sociétés savantes 3, pp Latreille, P., Des langoustes du Muséum National d'histoire Naturelle. Annales Muséum Histoire naturelle Paris 3, Latreille, P., Familles Naturelles du Règne Animal. Exposés succinctement et dans un ordre analytique, avec l'indication de leurs genres, pp Le Vay, L., Jones, D., Puello-Cruz, A., Sangha, R., Ngamphongsai, C., Digestion in relation to feeding strategies exhibited by crustacean larvae. Comparative Biochemistry and Physiology - Part A: Molecular and Integrative Physiology 128, Lemmens, J., The western rock lobster Panulirus cygnus (Decapoda: Palinuridae): Feeding biology and energetics of the puerulus stage. PhD thesis, The University of Western Australia, pp Lemmens, J., Knott, B., Morphological changes in external and internal feeding structures during the transition phyllosoma-puerulus-juvenile in the western rock lobster (Panulirus cygnus, Decapoda: Palinuridae). Journal of Morphology 220, Lemos, D., Hernandez-Cortes, M., Navarrete, A., Garcia-Carreno, F., Phan, V., Ontogenetic variation in digestive enzyme activity of larvae and postlarvae of the pink shrimp Farfantepenaeus paulensis (Crustacea: Decapoda: Penaeidae). Marine Biology 135, Liddy, G., Phillips, B., Maguire, G., Survival and growth of instar 1 phyllosoma of the western rock lobster, Panulirus cygnus, starved before or after periods of feeding. Aquaculture International 11, Liddy, G., Phillips, B., Maguire, G., Effects of temperature and food density on the survival and growth of early stage phyllosoma of the western rock lobster, Panulirus cygnus. Aquaculture 242, Linnaeus, C., Systema Naturae per Regna tria Naturae, secundum Classes, Ordines, Genera, Species, cum Characteribus, Differentiis Synonymis, Locis, (ed. 10) 1, Linnaeus, C., Systema Naturae per Regna tria Naturae, secundum Classes, Ordines, Editio Duocecima Reformata. Tom. 1. Part II. Systematicae. Naturae (ed. 12) 1, Loizzi, R., Interpretation of crayfish hepatopancreatic function based on fine structural analysis of epithelial cell lines and muscle network. Zeitschrift Fur Zellforschung Und Mikroskopische Anatomie 113, López-Alvarado, J., Langdon, C., Teshima, S., Kanazawa, A., Effects of coating and encapsulation of crystalline amino acids on leaching in larval feeds. Aquaculture 122, Losado-Tosteson, V., Posado, J., Losado, F., Size and reproductive status of fished spotted spiny lobster, Panulirus guttatus in Morrocoy National Park, Venezuela: A preliminary report. Marine and Freshwater Research 52, Lovatelli, A., Status of aquaculture in Vietnam. Aquaculture Asia 2, Lovett, D., Felder, D., Ontogeny of gut morphology in the white shrimp Penaeus setiferus (Decapoda, Penaeidae). Journal of Morphology 201, Lovett, D., Felder, D., 1990a. Ontogenetic changes in enzyme distribution and midgut function in developmental stages of Penaeus setiferus (Crustacea, Decapoda, Penaeidae). Biological Bulletin 178, Lovett, D., Felder, D., 1990b. Ontogenetic change in digestive enzyme activity of larval and postlarval white shrimp Penaeus setiferus (Crustacea, Decapoda, Penaeidae). Biological Bulletin 178, Macmillan, D., Sandow, S., Wikeley, D., Frusher, S., Feeding activity and the morphology of the digestive tract in stage-i phyllosoma larvae of the rock lobster Jasus edwardsii. Marine and Freshwater Research 48, Marinovic, B., Behavioural and physiological investigations with respect to light and temperature on the phyllosoma larvae of the western rock lobster Panulirus cygnus George, PhD thesis, The University of Western Australia, pp Martin, J., Davis, G., An updated classification of the recent Crustacea. Natural History Museum of Los Angeles County Science Series 39, Los Angeles, CA, pp

125 Literature Cited Martinez, I., Moyano, F., Fernandez-Diaz, C., Yufera, M., Digestive enzyme activity during larval development of the Senegal sole (Solea senegalensis). Fish Physiology and Biochemistry 21, Matsuda, H., Yamakawa, T., The complete development and morphological changes of larval Panulirus longipes (Decapoda, Palinuridae) under laboratory conditions. Fisheries Science 66, Matsuda, H., Takenouchi, T., Goldstein, J., The complete development of the pronghorn spiny lobster Panulirus penicillatus (Decapoda: Palinuridae) in culture. Journal of Crustacean Biology 26, Maynard, D., Dando, M., The structure of the stomatogastric neuromuscular system in Callinectes sapidus, Homarus americanus and Panulirus argus (Decapoda: Crustacea). Philosophical Transactions of the Royal Society Series B 268, McWilliam, P., Evolution in the phyllosoma and puerulus phases of the spiny lobster genus Panulirus White. Journal of Crustacean Biology 15, Mikami, S., Takashima, F., Functional morphology of the digestive system. In: Phillips, B., Cobb, J., Kittaka, J. (Eds.), Spiny Lobster Management. Blackwell Scientific, Oxford, pp Mikami, S., Greenwood, J., Takashima, F., Functional morphology and cytology of the phyllosomal digestive system of Ibacus ciliatus and Panulirus japonicus (Decapoda, Scyllaridae and Palinuridae). Crustaceana 67, Milne Edwards, H., Histoire naturelle des Crustacés, comprenant l'anatomie, la physiologie et la classification de ces animaux 2, Milne Edwards, H., Jasus (Sagmariasus) verreauxi (H. Milne Edwards, 1851). Annales des Sciences Naturelles, Paris, Zool., 16, 255, 290. Mitchell, J., Food preference, feeding mechanism, and related behaviour in phyllosoma larvae of the California spiny lobster, Panulirus interruptus (Randall). M.Sc. thesis, San Diego State College, pp Nelson, M., Cox, S., Ritz, D., Function of the mouthparts in feeding behaviour of phyllosoma larvae of the packhorse lobster, Jasus verreauxi (Decapoda: Palinuridae). Journal of Crustacean Biology 22, Nelson, M., Bruce, M., Nichols, P., Jeffs, A., Phelger, C., Nutrition of wild and cultured lobsters. In: Phillips, B. (Ed.), Lobsters: Biology, Management, Aquaculture and Fisheries. Blackwell Publishing, pp Nelson, M., Phelger, C., Mooney, B., Nichols, P., Lipids of gelatinous Antarctic zooplankton: Cnidaria and Ctenophora. Lipids 35, Nishida, S., Quigley, B., Booth, J., Nemoto, T., Kittaka, J., Comparative morphology of the mouthparts and foregut of the final-stage phyllosoma, puerulus, and postpuerulus of the rock lobster Jasus edwardsii (Decapoda: Palinuridae). Journal of Crustacean Biology 10, Nonaka, M., Oshima, Y., Hirano, R., Rearing phyllosoma of the Ise Lobster. Annual Report of Shizouka Prefectural Fish Experiment Stations. pp Ohs, C., D'Abramo, L., Buddington, R., Robinette, H., Roethke, J., Evaluation of a spraydried artificial diet for larval culture of freshwater prawn, Macrobrachium rosenbergii, and striped bass, Morone saxatilis. Aquaculture Nutrition 4, Olivier, M., Écrivesse, Astacus. Encyclopedie methodique, Histoire Naturelle Insectes 6, Oshima, Y., Feeding habit of Ise lobster. Suisan Gakkai Ho 7, (in Japanese). Ottera, H., Garatun-Tjeldstø, O., Julshamn, K., Austreng, E., Feed preferences in juvenile cod estimated by inert lanthanid markers - effects of moisture content in the feed. Aquaculture International 11, Parke, M., Studies on marine flagellates. Journal of the Marine Biological Association of the United Kingdom 28, Parker, J., On the structure of the head in Palinurus with special reference to the classification of the genus. Transactions of the New Zealand Institute 16, Paterson, N., The anatomy of the Cape rock lobster, Jasus lalandii (H. Milne-Edwards). Annals of the South African Museum 51, Patwardhan, S., On the structure and mechanism of the gastric mill in Decapoda. 4. The structure of the gastric mill in Reptantous Macrura. Proceedings of the Indian Academy of Science (B) 1,

126 Literature Cited Pavasovic, M., Richardson, N., Anderson, A., Mann, D., Mather, P., Effect of ph, temperature and diet on digestive enzyme profiles in the mud crab, Scylla serrata. Aquaculture 242, Penaflorida, V., Virtanen, E., Growth, survival and feed conversion of juvenile shrimp (Penaeus monodon) fed a betaine/amino acid additive. Bamidegh 48, 3-9. Phillips, B., Sastry, A., Larval ecology. In: Cobb, J., Phillips, B. (Eds.), The Biology and Management of Lobsters. Academic Press, New York, pp Phillips, B., Melville Smith, R., Cheng, Y., Estimating the effects of removing Panulirus cygnus pueruli on the fishery stock. Fisheries Research 65, Phillips, B., Palmer, M., Cruz, R., Estimating growth of the spiny lobsters Panulirus cygnus, P. argus and P. ornatus. Marine and Freshwater Research 43, Phleger, C., Nelson, M., Mooney, B., Nichols, P., Ritar, A., Smith, G., Hart, P., Jeffs, A., Lipids and nutrition of the southern rock lobster, Jasus edwardsii, from hatch to puerulus. Marine and Freshwater Research 52, Pitcher, C., Dennis, D., Skewes, T., Fishery-independent surveys and stock assessment of Panulirus ornatus in Torres Strait. Marine and Freshwater Research 48, Prasad, R., Tampi, P., On the phyllosoma of Mandapam. Proceedings of the National Institute of India (B) 23, Quinitio, E., Estepa, F., Alava, V., 1999, Development of hatchery techniques for the mud crab Scylla serrata (Forskål): Comparison of feeding schemes. In: Keenan, C., Blackshaw, A. (Eds.), Mud Crab Aquaculture and Biology, ACIAR Proceedings, ACIAR, Canberra, pp Quoy, J., Gaimard, J., Voyage autor du monde, entrepris par Ordre du Roi, sous le ministère et conformément aux instructions de S. Exc. M. le Vicomte du Bouchage, secrétaire d'état au départment de la Marine, excécuté sur les corvettes de S.M. l'uranie et la Physicienne sur les pendant les annés 1817, 1818, 1819 et Zoology Randall, J., Panulirus interruptus (Randall, 1840). Journal of the Academy of Natural Sciences, Philadelphia 8, Rissik, D., Suthers, I., Enhanced feeding by pelagic juvenile myctophid fishes within a region of island-induced flow disturbance in the Coral Sea. Marine Ecology Progress Series 203, Ritar, A., Smith, G., Dunstan, G., Brown, M., Hart, P., Artemia prey size and mode of presentation: Effects on larval survival and growth of phyllosoma larvae of the southern rock lobster (Jasus edwardsii). Aquaculture International 11, Ritar, A., Thomas, C., Beech, A., Feeding Artemia and shellfish to phyllosoma larvae of southern rock lobster (Jasus edwardsii). Aquaculture 212, Ritz, D., Behavioural response to light of the newly hatched phyllosoma larvae of Panulirus longipes cygnus George (Crustacea: Decapoda: Palinuridae). Journal of Experimental Marine Biology and Ecology 10, Rodrigues Souza, J., Strussmann, C., Takashima, Sekine, Shima Absorption of dissolved and dispersed nutrients from sea-water by Panulirus japonicus phyllosoma larvae. Aquaculture Nutrition 5, Roscoe, M., Biology and exploitation of the rock lobster Jasus tristani at the Tristan da Cunha Islands, South Atlantic, Investigational Report, South African Division of Sea Fisheries 118, Ruscoe, I., Jones, P., Jones, C., A comparison of moist and dry diets fed to red claw crayfish Cherax quadricarinatus, in tanks. Freshwater Crayfish 13, Ruscoe, I., Jones, C., Jones, P., Caley, P., The effects of various binders and moisture content on pellet stability of research diets for freshwater crayfish. Aquaculture Nutrition 11, Saisho, T., Studies on the phyllosoma larvae with reference to oceanographic conditions. Memoirs of the Faculty of Fisheries, Kagoshima University 15, (in Japanese). Schaap, A., Tasmania looks to rock lobster culture. Austasia Aquaculture 11, Sheppard, J., Bruce, M., Jeffs, A., Optimal feed pellet size for culturing juvenile spiny lobster Jasus edwardsii (Hutton, 1875) in New Zealand. Aquaculture Research 33, Siebold, P., Fauna Japonica Crustacea. De Historiae naturalis in Japonia statu 15,

127 Literature Cited Silberman, J., Sarver, S., Walsh, P., Mitochondrial DNA variation in seasonal cohorts of spiny lobster (Panulirus argus) postlarvae. Molecular Marine Biology and Biotechnology 3, Smith, D., Williams, K., Irvin, S., Response of the tropical spiny lobster Panulirus ornatus to protein content of pelleted feed and to a diet of mussel flesh. Aquaculture Nutrition 11, Smith, D., Williams, K., Irvin, S., Barclay, M., Tabrett, S., Development of a pelleted feed for juvenile tropical spiny lobster (Panulirus ornatus): Response to dietary protein and lipid. Aquaculture Nutrition 9, Spurr, A., A low-viscosity epoxy resin embedding medium for electron microscopy. Journal of Ultrastructure Research 26, Srikrishnadhas, B., Rahman, M., Growing spiny lobsters A profitable venture. Journal of Seafood Export 26, Stebbing, T., South African Crustacea. Marine Investigations in South Africa 1, Stone, S., Betz, A., Hofsteenge, J., Mechanistic studies on thrombin catalysis. Biochemistry 30, Streets, T., Descriptions of five new species of Crustacea from Mexico. Proceedings Academy Natural Sciences, Philadephia 1871, Subrahmanyam, C., Oppenheimer, C., The influence of feed levels on the growth of grooved penaeid shrimp in mariculture. Proceedings of the World Mariculture Society. 1, Sullivan, E., Leahy, J., Colwell, R., Cloning and sequence analysis of the lipase and lipase chaperone- encoding genes from Acinetobacter calcoaceticus RAG-1, and redefinition of a proteobacterial lipase family and an analogous lipase chaperone family. Gene 230, Temminck, C., Schlegel, H., Pisces. In: von Siebold, F (Ed.), Fauna japonica sive descriptio animalium, quae in itinere per Japoniam jussu et auspiciis superiorum, qui summum in Indian Batavia imperium tenent suscepto annis collegit, notis, observationibus et adumbra, pp Teshima, S., Kanazawa, A., Effects of several factors on growth and survival of the prawn larvae reared with microparticualte diets. Bulletin of the Japanese Society of Scientific Fisheries 49, Teshima, S., Kanazawa, A., Effects of protein, lipid and carbohydrate levels in purified diets on growth and survival rates of the prawn larvae. Bulletin of the Japanese Society of Scientific Fisheries 50, Teshima, S., Ishikawa, M., Koshio, S., Nutritional assessment and feed intake of microparticulate diets in crustaceans and fish. Aquaculture Research 31, Teshima, S., Kanazawa, A., Sakamoto, M., Microparticulate diets for the larvae of aquatic animals. Mini Review and Data File of Fisheries Research 2, Teshima, S., Koshio, S., Ishikawa, M., Alam, S., Hernandez Hernandez, L., Protein requirements of the freshwater prawn Macrobrachium rosenbergii evaluated by the factorial method. Journal of the World Aquaculture Society 37, Tolomei, A., Crear, B., Johnston, D., Diet immersion time: Effects on growth, survival and feeding behaviour of juvenile southern rock lobster, Jasus edwardsii. Aquaculture 219, Tong, L., Moss, G., Paewai, M., Pickering, T., Effect of brine-shrimp numbers on the growth and survival of early-stage phyllosoma larvae of the rock lobster Jasus edwardsii. Marine and Freshwater Research 48, Tucker, J., Marine Fish Culture. Kluwer Academic Publishing, Boston, pp Walbaum, J., 1792 Petri Artedi renovati. Part 3. Petri Artedi sueci genera Piscium in quibus systema totum ichthyologiae proponitur cum classibus, ordinibus, generum characteribus, specierum diffentiis, observationibus plumiris. Redactis Speciebus 2. Ichthyologiae, part III, pp 723. Ward, L., Carter, C., Crear, B., Smith, D., Optimal dietary protein level in juvenile southern rock lobster, Jasus edwardsii, at two lipid levels. Aquaculture 217, Watling, L., A classification system for crustacean setae based on the homology concept. In: Felgenhauer, B. (Ed.), Functional Morphology and Feeding and Grooming in Crustacea. A.A. Balkema, Rotterdam, pp

128 Literature Cited Williams, G., Wood, J., Dalliston, B., Shelley, C., Khu, C., 1999, Mud crab (Scylla serrata) megalopa larvae exhibit high survival rates on Artemia-based diets. In: Keenan, C., Blackshaw, A. (Eds.), Mud Crab Aquaculture and Biology, ACIAR Proceedings, ACIAR, Canberra, pp Williams, K., Smith, D., Irvin, S., Barclay, M., Tabrett, S., Water immersion time reduces the preference of juvenile tropical spiny lobster Panulirus ornatus for pelleted dry feeds and fresh mussel. Aquaculture Nutrition 11, Wolfe, S., Felgenhauer, B., Mouthpart and foregut ontogeny in larval, postlarval, and juvenile spiny lobster, Panulirus argus Latreille (Decapoda, Palinuridae). Zoological Scripta 20, Xue, M., Cui, Y., Effect of several feeding stimulants on diet preference by juvenile gibel carp (Carassius auratus gibelio), fed diets with or without partial replacement of fish meal by meat and bone meal. Aquaculture 198, Yamakawa, T., Nishimura, M., Matsuda, H., Tsujigado, A., Kamiya, N., Complete larval rearing of the Japanese spiny lobster Panulirus japonicus. Nippon Suisan Gakkaishi 55, 745. Yúfera, M., Pascual, E., Fernández-Díaz, C., A highly efficient microencapsulated food for rearing early larvae of marine fish. Aquaculture 177,

129 117

130 Appendix 1 Appendix 1 Authorities and Common Names of Species Species name Common name Authority Carassius auratus gibelio Gibel carp (Bloch, 1782) Gadus morhua Atlantic cod Linnaeus, 1758 Homarus americanus American lobster H. Milne Edwards, 1837 Homarus gammarus European lobster (Linnaeus, 1758) Isochrysis galbana Parke, 1949 Macrobrachium rosenbergii Giant river prawn (De Man, 1879) Mytilus edulis Blue mussel Linnaeus, 1758 Oncorhynchus mykiss Rainbow trout (Walbaum, 1792) Pecten fumatus Commercial scallop Reeve Penaeus japonicus Kuruma prawn Bate, 1888 Penaeus monodon Tiger prawn Fabricius, 1798 Penaeus setiferus White shrimp (Linneaus, 1767) Penaeus vannamei White leg shrimp Boone, 1931 Perna canaliculus Green mussel (Gmelin) Procambarus clarkii Louisiana crayfish (Girard, 1852) Salmo salar Atlantic salmon Linneaus, 1758 Sardinops melanostictus Japanese sardine (Temminck and Schlegel, 1846) Sparus aurata Gilthead seabream Linneaus, 1758 Scylla serrata Mud crab Forskål, 1775 Tetraselmis chuii Butcher,

131 119

132 Appendix 2 Appendix 2 Description of Stages Panulirus ornatus phyllosomata are classified into 11 stages (of which the first 10 are described here) based on the key characteristics presented in Tables A2.2. A detailed summary of the developmental sequence of the key characteristic of each stage using specimens from this study is provided in Table A2.2. Detailed descriptions of the mouthparts (in sequence, a single labrum, paired paragnaths, mandibles, maxillules, maxillae, and maxillipeds 1 3) are provided in Chapter 2 and consequently are omitted here. However, the appearance of gill buds, pleopods and uropods is provided. All images are photomicrographs and supplied for stages I V and IX X. Table A2.2 Distinguishing characteristics for each developmental stage of Panulirus ornatus (table adapted from the Queensland Department of Primary Industries, Cairns, Australia). 1. Eyestalk unsegmented Stage I, Instar 1 2. Eyestalk segmented Stage II, Instar 2 3. Exopod of third pereopod setose Stage III, Instar 3 4. Fourth pereopod bud Stage IV, Instar 4 Fourth pereopod unsegmented Stage IV, Instar 5 Fourth pereopod bifid Stage IV, Instar 6 5. Fourth pereopod setose Stage V, Instar 7 6. Antennule bud Stage VI, Instar 8 Antennule with three segments Stage VI, Instar 9 Antennule with four segments Stage VI, Instar 10 Abdomen segmented Stage VI, Instar 11 Uropod separate from abdomen Stage VI, Instar Uropod bifid Stage VII, Instar Second maxilliped with exopod bud on basis, pleopods bifid Stage VIII, Instar 14 Second maxilliped bud on basis larger Stage VIII, Instar Second maxilliped exopod with two pairs of plumose setae Stage IX, Instar 16 Fifth pereopod with two segments Stage IX, Instar 17 Fifth pereopod with three segments Stage IX, Instar 18 Second maxilliped with three pairs of plumose setae Stage IX, Instar Fifth pereopod with five segments, second maxilliped with four pairs of plumose setae Stage X, Instar

133 Appendix 2 121

134 Appendix 2 Stage I Stage I (Fig. A2.1 A,B). Mean total length 1.5. Carapace length 0.8 mm. Carapace width 0.8 mm (measurement data based on hatchery reared phyllosomata, supplied by the Queensland Department of Primary Industries). Eyestalk unsegmented. Antennule uniramous, unsegmented, with five distal setae. Antenna uniramous, unsegmented, with three distal setae, slightly shorter than the antennule. First and second pereopods, all biramous; exopods of first and second pereopods clearly developed with five pairs of plumose setae. Third pereopod biramous, exopod a small bud. Fig. A2.1 Panulirus ornatus. (A) Stage I phyllosoma, ventral view. Scale, 500 μm. (B) Unsegmented eyes stalk of stage I phyllosoma. Scale, 500 μm. (C) Stage II phyllosoma, ventral view. Scale, 500 μm. (D) Six pairs of plumose setae on exopod of second pereopods (stage II). AB, Abdomen; A1, Antennule; A2, Antenna; E, eye; EN, endopod; EX, exopod; MXP3, third maxilliped; P1-3, first to third pereopod. Stage II Stage II (Fig. A2.1A,B). Mean total length 1.9 mm. Carapace length 1.1 mm. Carapace width 0.9 mm. Eyestalk segmented. Antennule with six distal setae. First and second pereopod exopods with six pairs of plumose setae. Third pereopod exopod bud larger than in Stage I, plumose setae absent from exopod. 122

135 Appendix 2 Fig. A2.2 Panulirus ornatus. (A) Stage II phyllosoma, ventral view. Scale, 500 μm. (B) Six pairs of plumose setae on exopod of second pereopods (stage II). ES, eye stalk; MXP3, third maxilliped; P1-3, first to third pereopod. Stage III Stage III (Fig. A2.3A,B). Mean total length 2.4 mm. Carapace length 1.5 mm. Carapace width 1.1 mm. Eyestalk segmented, 400 μm in length. Antennule longer than antenna. First and second pereopod exopods with seven pairs of plumose setae. Third pereopod exopod with four pairs of plumose setae. Fourth pereopod present as a small bud. Fig. A2.3 Panulirus ornatus. (A) Stage III phyllosoma, ventral view. Scale, 500 μm. (B) Abdomen and buds of fourth pereopod (Stage III). Scale, 750 μm. AB, abdomen; P1, first pereopod; P2, second pereopod; P3, third pereopod; P4 BUD, fourth pereopod bud. 123

136 Appendix 2 Stage IV Stage IV (Fig. A2.4A D). Separated into three instars based on morphological changes of the fourth pereopod during the three moults from stage IV to V. Mean total length mm. Carapace length mm. Carapace width mm. Antennule with six distal setae, and one medial seta. First pereopod exopod with nine to twelve pairs of plumose setae, depending on instar, dactylus ends in spine, with three small setae. Second pereopod exopod segment one with small distal spine, segment two with pairs of plumose setae (depending on instar). Endopod propodus with serrate setae along length, nine serrate setae distally. Dactylus approximately two times longer than that of first pereopod. Dactylus with long robust spine and four smaller setae. Fourth pereopod bifid, depending on instar. Fig. A2.4 Panulirus ornatus. (A) Stage IV phyllosoma, ventral view. Scale, 500 μm. (B) Buds of fourth pereopod. Scale, 750 μm. (C) Tips of fourth pereopod. Scale, 750μm. (D) Bifid fourth pereopod. Scale, 750 μm. MXP3, third maxilliped; P1 4, first to fourth pereopods. P4 BIFID, fourth pereopod bifid; P4, fourth pereopod. 124

137 Appendix 2 Stage V Stage V (Fig. A2.5A B). Mean total length 4.2 mm. Carapace length 2.9 mm. Carapace width 1.9 mm. Antennule six distal elements, inner flagellum recognised as a small protrusion, with two setae. Antenna longer than antennule. First pereopod exopod second segment with 12 pairs of plumose setae. Second pereopod exopod segment one with two small medial spines, segment two with 12 pairs of plumose setae. Endopod dactylus is shorter, and more robust. Third pereopod exopod segment two with eight pairs of plumose setae. Endopod propodus with serrate setae. Fourth pereopod exopod longer than endopod, exopod segment two with three pairs of plumose setae. Fifth pereopods appear still as small buds. Fig. A2.5 Panulirus ornatus. (A) Stage V phyllosoma, ventral view. Scale, 500 μm. (B) Fourth pereopod showing three pairs of plumose setae. Scale 200μm. MXP3, third maxilliped; P1, first pereopod; P2, second pereopod; P4, fourth pereopod. Stage VI Stage VI (photo absent due to limited number of specimens). Mean total length mm. Carapace length mm. Carapace width mm. Antennule peduncle three-segmented, bud of inner flagellum increased in size. Antennae longer than antennule, unsegmented, five setae medially, three sub-terminally and six distally. First pereopods dactylus with long distal spine, surrounded by four setae. Second exopod segment with 13 pairs of plumose setae. Second pereopods dactylus longer than first pereopod, distal spine, with the two opposing long setae and seven small setae. Second exopod segment with 14 pairs of plumose setae. Third pereopod dactyus with distal spine and three small setae. Second exopod segment with 10 pairs of plumose setae. Fourth pereopod biramous, two small medial spines and seta 125

138 Appendix 2 near endopod/exopod branch, exopod two-segmented with five pairs of plumose setae. Endopod four-segmented, with seta and distal spine. Fifth pereopod buds slightly larger. Abdomen segmented. Stage IX Stage IX (Fig. A2.5 A D). Mean total length mm. Carapace length mm. Carapace width mm. Antennule five-segmented, peduncle threesegmented. Antennae longer than antennule, two four segmented. Second maxilliped with two-three pairs of plumose setae. First pereopod exopod, two-segmented with pairs of plumose setae. Second pereopod exopod, two-segmented with pairs of plumose setae. Third pereopod exopod, two-segmented with pairs of plumose setae. Fourth pereopod exopod, two-segmented with pairs of plumose setae. Fifth pereopod endopod one four segmented, extending to posterior of abdomen. Abdomen segmented, with four biramous pleopods. Rudimentary or short (unilobed) gill buds on third maxillipeds and first fourth pereopods. Stage X Stage X (Fig. A2.7 A,B). Mean total length 15.5 mm. Carapace length 11 mm. Carapace width 7.1 mm. Antennae longer than antennule, four-segmented. Second maxilliped with four pairs of plumose setae. First pereopod, exopod two-segmented with 27 pairs of plumose setae. Second pereopod, exopod two-segmented with 27 pairs of plumose segmented. Third pereopod, exopod two-segmented with 26 pairs of plumose setae. Fourth pereopod, exopod two-segmented with 22 pairs of plumose setae. Fifth pereopod five segmented. A full compliment of gill buds is present on the second and third maxillipeds and first fourth pereopod. Second maxilliped with a bilobed bud on the dorsal side of the coxa. For the third maxilliped and the first fourth pereopods a bilobed bud is present on the dorsal side of the coxa, a unilobed bud is on the edge of the thorax and two unilobed buds are present on the edge and the dorsal side of the thorax. Uropod with serrations on the lateral margins of both rami. 126

139 Appendix 2 Fig. A2.6 Panulirus ornatus. (A) Stage IX (Instar 17) phyllosoma, ventral view. Scale, 500 μm. (B) Stage IX (Instar 17) ventral view of abdomen and segmentation of fifth pereopod. Scale, 200 μm. (C) Stage IX (Instar 19) ventral view. Scale, 1 mm. (D) Stage IX (Instar 19) ventral view of abdomen showing segments of fifth pereopod. Scale, 200 μm. A1, antennule; A2, antennae; ES, eye stalk; MXP3, third maxilliped; P1 5, first to fifth pereopods; S1 4, segments one to four of fifth pereopod. 127

140 Appendix 2 Fig. A2.7 Panulirus ornatus. (A) Stage X phyllosoma, ventral view. Scale, 500 μm. (B) Stage X, ventral view of abdomen and fifth pereopod. Scale, 200 μm. A1, antennule; A2, antennae; G Bud; gill bud; MXP3, third maxilliped; P1, first pereopod; P3, third pereopod; P4, fourth pereopod; P5, fifth pereopod; S1 5, segments one to five of fifth pereopod. 128

141 129

142 Appendix 3 Appendix 3 Mouthpart and Foregut Ontogeny in Phyllosomata of Panulirus ornatus and their Implications for Development a Formulated Larval Diet Introduction An inadequate dietary regime has been a major culture impediment of phyllosomata of spiny lobsters, arising due to a lack of understanding of their larval feeding biology and nutritional requirements (Johnston and Ritar, 2001). This study identifies key structural changes of the mouthparts and foregut of phyllosomata of P. ornatus in the early- and mid-stages of their ontogenetic development, and infers how these changes may influence the physical characteristics (size, shape and texture) of a formulated larval diet. A feeding trial using three different textured diets was conducted to determine a preferred diet texture for early stage phyllosomata. Materials and Methods Ovigerous female brood stock collected from Princess Charlotte Bay (Northern Queensland, Australia) were held in flow through Nally bins (26 o C) until egg hatch. The newly hatched phyllosomata were transferred to 10 l up-weller recirculation tanks and maintained at 26 C and 36 salinity. For morphological examination of the mouthparts and foregut during ontogenetic development, phyllosomata were fed to satiation on a sole diet of enriched Artemia. For the feeding trial, phyllosomata were reared in 3 l mini up-wellers (50 phyllosomata/l). Phyllosomata that were used in the feeding trial were fed one of three different textured formulated diets, and control tanks were fed on-grown Artemia as above. At each developmental stage, phyllosomata were removed from the culture vessels for examination and staged after the method of Duggan and McKinnon (2003). For examination of the mouthparts, phyllosomata (n=10) were fixed for 2 3 h at room temperature in 2.5% glutaraldehyde in 0.1 M phosphate buffer ph 7.4. Phyllosomata were then prepared for scanning electron microscopy and examined at high vacuum with a LEO VP FEGSEM. Presented as a paper: Johnston, M., Johnston, D., Knott, B., Jones, C., Mouthpart and foregut ontogeny in phyllosomata of Panulirus ornatus and their implications for development of a formulated larval diet. In: Hendry, C., Van Stappen, G., Wille, M., Sorgeloos, P. (Eds.), Larvae th Fish and Shellfish Larviculture Symposium. European Aquaculture Society, Special Publication No. 36, Gent, Belgium, pp

143 Appendix 3 For histological examination of the foregut, phyllosomata were fixed and prepared as above before being embedded in JB-4 glycol methacrylate resin. The foregut was serially sectioned (transverse) at 2 μm on a Sorvall microtome and stained with a polychrome stain. Digital images were observed and captured using an Olympus DP70 camera and Image Pro Plus v.10 software. The ingredient composition of the three formulated diet textures used in the feeding trial is shown in Table A3.1. As the nutritional requirements for P. ornatus phyllosomata are currently unknown, diet formulations were based on the known requirements of adult P. ornatus and larval penaeid prawns. Table A3.1 Composition of the 3 micro-bound diet textures. * %DW = dry weight Ingredient Gelatinous Paste Hard Krill meal Squid meal Fish meal Krill hysrolysate Fish oil Corn oil Soy lecithin Cholesterol Vitamin and mineral premix Wheat flour Gelatine * A complete list of suppliers, contact detail and product numbers are provided in Appendix 4. Results and Discussion The mouthparts of P. ornatus phyllosomata are well developed at hatch and the gross mouthpart structure does not change significantly throughout larval development (Fig. A3.1 A,B), especially the mandibles (Fig. A3.1 C). Density and robustness of setation of the mouthparts does however, increase during larval development, especially on the medial surface of the paired paragnaths and the number of spinose projections on the first maxillules also increases in mid-stage phyllosomata (Instar 8). Increased setation of the mouthparts in mid-stage phyllosomata suggests that they are able to ingest larger fleshier prey, by generating a stronger feeding current by the presence of extra pappose setae on the maxillules and are able to maintain fleshy prey items close to their buccal cavity by trapping food particles on the increased setation of the paragnaths. The foregut of early instar phyllosoma (1 2) is quite simple, with no filter press, but is well armed with robust lateral setae and main brushes (Fig. A3.1 D). Filter press development occurs at instar 3 and is a significant structural development, occurring earlier in larval development than other spiny lobster species. Filter press complexity 131

144 Appendix 3 increases in mid-stage phyllosomata, with an increase in number of ampullary channels (Fig. A3.1 F). Developmental changes of the foregut after instar 3 suggests that there may be a shift in dietary regime from soft gelatinous food items to larger fleshy prey, as the internal triturative and filtration ability of phyllosomata improves throughout larval development. Fig. A3.1. Scanning electron micrographs of the mouthpart and transverse histological sections of the foregut of phyllosomata of Panulirus ornatus. (A) Oral region (Instar 4) showing spatial relationship of mouthparts. Scale, 60 µm. (B) Molar process (Instar 1) showing complexity of spinose projections at hatch. Scale, 2 µm. (C) Anterior foregut (Instar 1). Scale, 20 µm. (D) Posterior foregut, Instar 8, showing detail of inner and outer valve setae of the filter press. Scale, 20 μm. AF, anterior floor; DLS, dorso-lateral setae; IV, inner valve L, labrum; LS, lateral setae; M, mandible; MX1, maxillule; O, outer valve; P, paragnath. As the early morphological structure of the mouthparts and foregut of phyllosomata suggests that their ingestive and digestive capacity is well suited to soft gelatinous dietary items, a feeding trial investigating the effect of different formulated diet textures and Artemia on survival and growth was conducted (Fig. A3.2). Phyllosomata survival when fed only Artemia was significantly higher than all three formulated diets. Initially phyllosomata were attracted to the formulated diets and were observed manipulating the diets with their first pereopods and appeared to move diet particles anteriorly toward their mouth. Despite this initial feeding response, 132