Early weaning of southern flounder, Paralichthys lethostigma, larvae and ontogeny of selected digestive enzymes

There is considerable interest in rearing Southern flounder, Paralichthys lethostigma, for commercial production and for stock enhancement. Both goals depend upon excellent larval nutrition for the production of robust juveniles. The current use of live prey for larviculture is an expensive and time consuming process that can be alleviated by weaning larvae onto dry feed. A study was conducted to assess the potential for early weaning of southern flounder larvae onto a microdiet (MD). In addition, the activity of selected digestive enzymes was measured during ontogeny to evaluate the digestive capabilities of the larvae over time. Pancreatic enzyme activities (U larva^− 1) were very low or undetectable at hatching and a marked increase in activity was not observed until the larvae reached 4 mm (~ 11 dph) in standard length for chymotrypsin (24–44,000) and 6 mm (~ 25 dph) for amylase (< 1–24), trypsin (1–18) and bile salt-dependent lipase (0–443). Acid protease activity (~ 1.0) was detected once the larvae were 8.5–9.0 mm (37–39 dph) in length although a sizeable increase in activity (> 10.0) was not observed until after complete metamorphosis (> 11.0 mm; 40–45 dph). Feeding regimes employed for the weaning study consisted of a live feed control (C) and a combination of live feed and MD in which the addition of the MD was initiated on 11 dph and live feed terminated on 17 (T17), 23 (T23) or 29 (T29) dph. At the end of the study (35 dph), mean standard length and the percent of settled fish were significantly greater for fish in the control treatment (8.3 mm; 21.1%) than for fish fed any combination of live prey and MD (6.4 mm; 2.0%). Average survival was 27.7% and no significant differences were noted among treatments. However, the number of fish exhibiting spinal deformities, lordosis, was significantly lower in the control and T29 treatments (1.7%) than the T17 and T23 treatments (25%). The results of this study indicate that southern flounder larvae readily wean onto dry feed prior to the onset of metamorphosis. However, decreased growth and a high incidence of lordosis emphasize the need for the development of a more appropriate MD for this species when digestive enzyme activities are relatively low and gastric digestion is absent.     

Effects of early weaning strategies on growth,survival and digestive enzyme activities in cobia (<Emphasis Type="Italic">Rachycentron canadum</Emphasis> L.) larvae

The effects of weaning strategies of cobia (Rachycentron canadum L.) larvae to commercial microdiets, either from rotifers or from Artemia, on growth, survival and enzymatic digestive capacity, were investigated. In the first experiment, cobia larvae were weaned from rotifers by co-feeding with a microdiet (Otohime) from 8, 13 or 20 days post-hatching (dph). The larvae in the control treatment were fed rotifers (2–12 dph), Artemia nauplii from 7 dph, and co-fed with the microdiet from 20 dph. In the second experiment, the larvae were weaned from Artemia, which was fed to the larvae from 7 dph, by co-feeding with a microdiet (NRD) from 8, 13 or 18 dph. The larvae in control treatment were fed rotifers, then Artemia to the end of the experiment (28 dph). Weaning of cobia larvae onto a microdiet directly from rotifers significantly reduced growth, survival and digestive capacity of the larvae and did not lead to larval acceptance of the microdiet, compared to those weaned from Artemia in the first experiment. Early weaning of cobia larvae onto NRD microdiet (on 8 or 13 dph) from Artemia in the second experiment also reduced growth, survival rate and gut maturation index, compared to those fed live feed. With available microdiets, weaning of cobia larvae could start from Artemia at around 18 dph in order to obtain comparable growth, survival and gut maturation to larvae fed live feed.     

Activities of digestive enzymes and histology of digestive system during larval development of devil stinger (Inimicus japonicus)

Devil stinger is a valuable demersal scorpaenid fish while the rearing of stinger larvae still relies on live prey. This study was conducted to illustrate the development of the main digestive enzymes and digestive system during larval development of this species to provide evidence for the application of artificial feeds. Enzymatic and histological assays were conducted from 1 day post hatching (dph) to 36 dph in larvae. The result showed that the selected digestive enzyme activities increased significantly after 15 dph. Specifically, the total trypsin activities increased significantly from 18 dph to 33 dph. The total pepsin and amylase activities increased significantly first and thereafter decreased significantly. The lipase activities followed the similar pattern with trypsin. With regard to the histological study, the stinger larvae open their mouth to first feeding at 3 dph and turned into totally exogenous nutritional stage at 6 dph. In addition, mucous membrane, rich in goblet cells, was widely distributed in oesophagus epithelium at 18 dph. The height and amounts of gastric gland in cardia and main body of the stomach increased gradually with the development of stinger larvae after 15 dph. The intestine length of stinger larvae was short, and goblet cell was abundant in anterior intestine after 12 dph, not the posterior intestine. The ontogeny of liver and pancreas started from newly hatched stage, and the differentiation of liver was prior to pancreas. The above findings would provide evidence for the use of artificial feeds from the larval stage of stinger larvae (at least from 21 dph).     

Evaluation of weaning performance of California halibut (Paralichthys californicus) larvae using growth,survival and digestive proteolytic activity

This study evaluated weaning success of California halibut, Paralichthys californicus, larvae onto a microdiet at various stages of development utilizing growth, survival and digestive enzyme activity. Weaning onto a microdiet was evaluated at 16, 26, 36 and 46 days posthatch (dph). Alkaline and acid proteases and leucine aminopeptidase activities were measured after weaning and compared between the weaned treatment and Artemia‐fed controls. Survival was significantly lower in the microdiet‐fed treatments compared to the control groups. Growth was significantly reduced in all weaning treatments compared to the control, except for the 46 dph group. No differences in enzyme activities were detected between treatment and diet at 16 and 26 dph; however, activities were higher for the microdiet‐fed larvae at 36 and 46 dph. This study demonstrates that California halibut larvae possess a differentiated and effective digestive system early in development and can be weaned with relative success (>40% survival) before completion of the metamorphosis (i.e., 36 dph). The lack of weaning success at an early date cannot be entirely because of the absence of a functional stomach but could be related to, among other factors, the low‐microdiet ingestion rates observed and higher leaching of smaller microdiets.     

Changes in digestive enzyme activities during larval development of leopard grouper (Mycteroperca rosacea)

The leopard grouper is an endemic species of the Mexican Pacific with an important commercial fishery and good aquaculture potential. In order to assess the digestive capacity of this species during the larval period and aid in the formulation of adequate weaning diets, this study aimed to characterize the ontogeny of digestive enzymes during development of the digestive system. Digestive enzymes trypsin, chymotrypsin, acid protease, leucine–alanine peptidase, alkaline phosphatase, aminopeptidase N, lipase, amylase and maltase were quantified in larvae fed live prey and weaned onto a formulated microdiet at 31 days after hatching (DAH) and compared with fasting larvae. Enzyme activity for trypsin, lipase and amylase were detected before the opening of the mouth and the onset of exogenous feeding, indicating a precocious development of the digestive system that has been described in many fish species. The intracellular enzyme activity of leucine–alanine peptidase was high during the first days of development, with a tendency to decrease as larvae developed, reaching undetectable levels at the end of the experimental period. In contrast, activities of enzymes located in the intestinal brush border (i.e., aminopeptidase and alkaline phosphatase) were low at the start of exogenous feeding but progressively increased with larval development, indicating the gradual maturation of the digestive system. Based on our results, we conclude that leopard grouper larvae possess a functional digestive system at hatching and before the onset of exogenous feeding. The significant increase in the activity of trypsin, lipase, amylase and acid protease between 30 and 40 DAH suggests that larvae of this species can be successfully weaned onto microdiets during this period.     

Combined effect of weaning time and co‐feeding duration on growth and survival of peled Coregonus peled (Gmelin) larvae

The study investigated the combined effect of weaning from live feed to a commercial dry pellet at 10, 15, 20, 25 or 30 days posthatching (dph) and co‐feeding for 1, 3 or 6 days on survival and growth of Coregonus peled larvae. Additional groups fed only live Artemia sp. nauplii (ART), and only Biomar LARVIVA ProWean 100 (DRY) were included. A final survival rate of 66.4%–85.5% was observed in groups weaned after 20 dph. Final body weight (BW) and total length (TL) were significantly lower in groups weaned at 10 and 15 dph, regardless of the duration of co‐feeding. Larvae reached 29–37 mg BW and TL of 17.7–19.0 mm in groups weaned at 20, 25 and 30 dph. The recommended minimum duration of feeding with live food, based on these results, is 20 days. Based on the significantly higher yield of larvae weaned after 20 dph irrespective of co‐feeding duration, it can be concluded that abrupt weaning to dry food after 20 days of feeding with live prey can provide adequate production while reducing the effort and costs associated with live feed.     

Lipolytic activities in developing turbot larvae as influenced by diet

The development of neutral lipase and phospholipase activities was studied in larval turbot fed live prey. Activities of neutral lipase and phospholipase (activity larva⁻¹) increased significantly between days 6 and 24 after hatching in turbot larvae. The specific activities of both enzymes (activity μg protein⁻¹) decreased in older larvae. Feeding of a microdiet for 3 days (days 10–13) affected the lipolytic activity of neutral lipase and phospholipase negatively, compared to the larvae fed on rotifers. Since neutral lipase activities in whole larval homogenates and in the gut were significantly lower, it suggests a reduced synthesis rate and a reduced secretion of the enzyme in larvae fed the microdiet. A correlation between neutral lipase and phospholipase activities was found in larvae fed rotifers, but not in larvae fed the microdiet. This may indicate different regulating and stimulating mechanisms for these enzymes. The contribution of exogenous enzymes from ingested live prey to the total larval enzyme activity was about 6% for neutral lipase and 10% for phospholipase on day 6. The exogenous prey enzymes accounted for only 2% of the total activities in 12-day-old turbot larvae, suggesting that enzymes from prey did not contribute considerably to the digestion of lipids. This revised version was published online in August 2006 with corrections to the Cover Date.     

Feeding Larval Red Drum on Microparticulate Diets in a Closed Recirculating Water System1

A feeding protocol was developed for red drum larvae based on combining a commercial microparticulate diet (Kyowa Fry Feed) with live prey (rotifers) in a closed, water reuse system. In five trials, growth and survival were measured on larvae reared on a combination of live and microdiet for 1–5 d and then microdiet alone. Results in each trial were compared to control larvae reared on live rotifers Brachionus plicatilis and brine shrimp nauplii Artemia salinas. The most satisfactory combination was feeding live food and microdiet together for the first five days and then completely discontinuing live prey, eliminating the need to feed brine shrimp to the larvae. Growth rates of larvae fed progressively larger sizes of the microdiet were as good as larvae reared on live prey. Both groups metamorphosed to the juvenile stage at less than one month. Survival rates on the five day live food and microdiet combination were a remarkable 60% from egg to the juvenile stage. The successful weaning of red drum to microdiets paves the way to produce a semipurified diet to test nutrient requirements of larval fish.     

Early weaning of winter flounder (Pseudopleuronectes americanus Walbaum) larvae on a commercial microencapsulated diet

Like most small marine fish larvae, the stomachs of winter flounder Pseudopleuronectes americanus are undeveloped at first feeding and have relatively reduced digestive capacity. This work was undertaken to test whether larvae at the onset of stomach differentiation (larval size about 5.5 mm) could be early weaned onto a commercial microencapsulated diet. We assessed the effect of early weaning by first comparing growth performance (standard length, total protein content and age at metamorphosis) of larvae fed enriched live prey from first feeding to a size of 5.5 mm and then reared on three different feeding regimes until metamorphosis: (1) live prey (LP) as a control group; (2) mixed feeding of live prey and microencapsulated diet (LP‐ME); (3) exclusively microencapsulated diet (ME) after fast weaning over 4 days (to a larval size of 6.2 mm). No differences were observed between larval development in the two first groups, which began metamorphosis at 40 days old. The larvae of the third group showed significantly slower growth that resulted in a delay of 4 days in the onset of metamorphosis. Differences in live prey availability between the treatments and the short transition period to allow the larvae to adapt to the new diet were identified as possible contributing factors to the slower growth and to the delay in metamorphosis of early weaned larvae. In a second experiment, the transitional weaning period was increased until the larvae were 6.6 mm in length. Weaning at that size resulted in no slowing of growth or delay in metamorphosis, suggesting that the feeding schedule was adequate.     

Effects of two commercial microdiets on growth and survival of barramundi (Lates calcarifer Bloch) larvae within various early weaning protocols

Because of high costs and labour requirements along with the highly variable nutritional value of live feeds, we investigated the possibility of early weaning for barramundi (Lates calcarifer Bloch) larvae aimed at reducing the use of Artemia. Two commercial microdiets, Gemma Micro (Skretting, Australia) and Proton (INVE, Belgium) were compared for growth and survival of larvae using three weaning protocols, until 33 days posthatch (dph). Enriched rotifers were fed to larvae in all protocols through mouth opening until 21, 18 and 30 dph (protocols 1, 2 and 3, respectively). At 13 dph, enriched Artemia metanauplii were introduced to weaning protocols 1 and 2, and continued until 29 and 24 dph, respectively, whereas protocol 3 did not receive Artemia. Microdiet was initiated at 20, 16 and 13 dph in protocols 1, 2 and 3, respectively. Barramundi larvae grew successfully to 33 dph when co‐fed rotifers and microdiet, and significantly larger larvae resulted from feeding Gemma Micro rather than Proton, when Artemia were not used. However, larvae weaned onto Proton using a longer period of Artemia provision were significantly larger than larvae reared according to all other protocols. Survival was significantly higher in all Gemma Micro protocols when compared with Proton protocols. This was in part due to higher cannibalism when using Proton compared with Gemma Micro (22.8 ± 0.9% and 9.2 ± 0.6%, respectively). Cannibalism became more noticeable in all protocols when the larvae reached 7–8 mm standard length and further increased after the cessation of live feed. Tank biomass production was the highest when larvae were weaned onto Gemma Micro including a short period of Artemia provision as a result of a combination of high growth and survival. However, similar biomass production resulted when larvae were weaned directly from rotifers onto Gemma Micro and/or from a prolonged Artemia period onto proton. The success of weaning barramundi larvae directly to microdiet from rotifers, thus eliminating the need for Artemia, was influenced by the microdiet. Relatively higher levels of free amino acids and lipids were believed to contribute to increasing larval growth and survival. Larvae that were fed Gemma Micro showed higher growth when Artemia were utilized for a shorter period, while Proton‐fed larvae benefited from an extended Artemia feeding period.     

The effect of different rotifer feeding regimes on the growth and survival of yellowtail kingfish Seriola lalandi (Valenciennes, 1833) larvae

First feeding success is critical to larval marine finfish and optimization of live feed densities is important for larval performance and the economics of commercial hatchery production. This study investigated various rotifer feeding regimes on the prey consumption, growth and survival of yellowtail kingfish Seriola lalandi larvae over the first 12 days post hatch (dph). The common practice of maintaining high densities of rotifers (10–30 ind. mL^?1) in the rearing tank was compared to a low density feeding technique, where 5–8 ind. mL^?1 of rotifers were offered. A ‘hybrid’ feeding regime offered rotifers at the high density treatment until 5 dph and the lower feeding densities thereafter. There was no significant difference in larval survival (hybrid: 28.9 ± 7%, low density: 17.3 ± 5% and high density: 17.2 ± 9%) or growth (hybrid: 6.12 ± 0.18 mm, low density: 6.03 ± 0.10 mm and high density: 6.11 ± 0.23 mm) between treatments. Rotifer ingestion was independent of rotifer density throughout the trial and increased with larval age, with larvae at 4 dph ingesting 22 ± 1.5 rotifers larvae^?1 h^?1 and by 11 dph ingesting 59 ± 1.6 rotifers larvae^?1 h^?1. These data demonstrate that from first feeding, yellowtail kingfish larvae are efficient at capturing prey at the densities presented here and consequently significant savings in rotifer production costs as well as other potential benefits such as facilitation of early weaning and improved rotifer nutritional value may be obtained by utilizing lower density rotifer feeding regimes.     

Incidence of systemic granulomatosis is modulated by the feeding sequence and type of enrichment in meagre (Argyrosomus regius) larvae

Systemic granulomatosis is the most frequent disease in juvenile and adult meagre, but studies regarding the first appearance of granulomas in larvae do not exist. In order to evaluate this, meagre larvae were fed four different feeding regimes as follows: RS and RO (rotifer enriched with Easy DHA Selco or Ori‐Green from 3 to 30 dph respectively), RAS and RAO (rotifer enriched with Easy DHA Selco or Ori‐Green from 3 to 21 dph and Artemia enriched with Easy DHA Selco or Ori‐Green from 12 to 30 dph respectively). All treatments were also fed with commercial microdiet from 20 to 30 dph. At 30 dph weight, length, specific growth rate and survival were significantly higher in Artemia‐fed larvae, regardless of the enrichment. Microscopic first appearance of granulomas was observed in 20 dph larvae fed RS and RO. At 30 dph granulomas and thiobarbituric acid reactive substances (TBARS), values were significantly higher in RS and RO‐fed larvae than in RAS and RAO‐fed larvae. The results showed that granulomas first appeared in meagre larvae at 20 dph when fed rotifers only. Conversely, a reduced appearance of granulomas and lipid peroxidation occurs when Artemia is included in the feeding sequence reinforcing the hypothesis of a nutritional origin of the systemic granulomatosis.     

Nile tilapia broodfish fed high‐protein diets: Digestive enzymes in eggs and larvae

The aim of this study was to assess the activity of gastric, pancreatic and intestinal digestive enzymes during the embryonic and larval development of Nile tilapia (Oreochromis niloticus) GIFT strain Aqua America^® 1 obtained from a broodstock fed two levels of crude protein (CP). A total of 72 females and 24 males, 10 hapas, two CP levels (32% and 38%) and six phases of embryonic (cleavage, blastula, gastrula) and larval (hatching, 7 and 10 days post hatch, dph) stages were used. The eggs were collected in cleavage, blastula and gastrula stages, 300 mg was collected, and kept in cryogenic tubes in liquid nitrogen. For the samples at larval stage, the remaining eggs were separated according to crude protein level and kept in hatcheries and samples were collected on 7 and 10 dph the same way as before. A total of 48 samples were collected: at each protein level (32% and 38% CP), four samples were collected in each phase of embryonic and larval development. Statistical differences were not observed during embryonic development for acid proteases, trypsin, amylase and lipase activity at both levels of crude protein (32 and 38% CP). Differences for acid proteases were noticed on 7 dph; trypsin and amylase on 7 dph and 10 dph. Significant differences on blastula and 7 dph for protease; as for aminopeptidase, there was significant difference on 7 dph. The data indicated early appearance of digestive enzymes in Nile tilapia broodfish receiving 32% CP taking into account the rapid growth and development of this species.     

The effect of dietary exogenous digestive enzymes on ingestion,assimilation, growth and survival of gilthead seabream (Sparus aurata,Sparidae, Linnaeus) larvae

The success of microdiets commonly used in the cultivation of marine fish larvae is limited to serving as partial replacements for live food. This limited success is thought to be associated with a reduced digestive ability due to an incompletely developed digestive system. The enhanced growth obtained from live food has been partially attributed to the digestive enzyme activity of the food organism. The present study was designed to test the effect of an exogenous digestive enzyme incorporated. into a microdiet on the growth of Sparus aurata. Larval gilthead seabream, 20–32 days old, were fed ¹⁴C labelled microdiets containing a commercial pancreatic enzyme at different concentrations (0, 0.1 and 0.05g / 100 g dry diet). Rates of ingestion and assimilation were measured and their relationship to dry weight was determined. Our results show that the success of the microdiet as a food for larval gilthead seabream was limited by the larva's low ingestion rate which only approached its maintenance requirement. In addition, the presence of digestive enzyme in the microdiet enhanced its assimilability by 30%. Larval growth over ten days was 0, 100 and 200% on microdiet free of added enzymes, one with added enzymes and a live food regime, respectively. It is our opinion that successful development of microdiets for Sparus aurata must be based on diets improved both in digestibility and attraction to the larvae. Further studies are now underway to determine the nutritional requirements of gilthead seabream larvae using the experimental method developed in the present study. This research was carried out in partial fulfillment of the requirements for the M.Sc. degree.     

The Effect of Temperature on First Feeding, Growth, and Survival of Larval Witch Flounder Glyptocephalus cynoglossus

Witch flounder Glyptocephalus cynoglossus has recently been identified as a candidate species for aquaculture in the northeastern United States and the Canadian Atlantic Provinces. This study investigated the optimal temperatures for witch flounder larval first feeding and for long term larval culture from hatching through metamorphosis. Maximum first feeding occurred between 15.0 and 16.2 C. Larvae did not survive beyond first feeding when reared at mean temperatures of 5.1, 10.4, or 19.5 C and were unable to initiate feeding at mean rearing temperatures below 6.0 C. At a rearing temperature of 15.0 C in 16-L tanks, mean larval survival to 60 days post hatch (dph) was 14.1%. Mean overall length-specific growth rate for larvae reared to 60 dph at 15.0 C was 3.5%/d and mean absolute growth was 0.62 mm/d. Subsequent larval growth at 15.6 C began to taper off towards 70 dph at the onset of weaning which overlapped with larval metamorphosis. Growth plateaued at 85 dph, followed by a rebound between 90 and 95 dph. Survival was 100% when weaning onto a dry, pelleted diet was initiated at 70 dph with a 10-d live diet co-feeding period. These results are favorable and encourage the further pursuit of commercial witch flounder culture.     

Structural development of Pacific red snapper Lutjanus peru from hatching to the onset of first feeding

Successful rearing of larval fish requires culture conditions and feeding strategies matching the ontogenetic status of larvae. This study describes the external morphology and development of organs and structures involved in the feeding process (i.e. sensorial organs, mouthparts and digestive system) from hatching until first feeding in Pacific red snapper. Hatching occurred 26 h after fertilization at 26°C and total length (TL) was 2.45 ± 0.08 mm. The larvae showed an undifferentiated eye and digestive tract. At 48 hah, TL was 3.31 ± 0.12 mm. Yolk and oil globule were still present. The mouth was still closed, but the Meckel's, quadrate, hyoid and hyomandibular cartilages were present. The retina was formed by 5 layers, and a thin layer of pigment epithelium was observed in the outer nuclear layer (ONL). At 70 hah, TL was 3.44 ± 0.22 mm. A remnant of oil globule was still present. The mouth and anus were open. At 93 hah, the number of cones in the ONL of the retina have increased and there was more pigment in the pigment epithelial layer. A joint between Meckel's and the quadrate cartilage and also a joint between the hyomandibular cartilage and the skull were present. The presence of live feed was detected in the digestive tract of these larvae. Based on these observations, the Pacific red snapper larvae is functional to start ingesting live feed between the 3rd and 4th day after hatching.     

Is it possible to successfully rear meagre (Argyrosomus regius Asso 1801) larvae without using rotifers?

In hatcheries, meagre Argyrosomus regius larvae still depend on an adequate supply of rotifers and Artemia, as no artificial diet can totally fulfil their nutritional requirements. However, production of live feed is highly expensive and demands intensive labour and specific facilities. This study investigated the effect of a dietary regime without the use of rotifers, to simplify the meagre larval rearing protocol. Two feeding treatments (T₁ & T₂) are compared to investigate their effects on survival and growth of meagre larvae. In T₁, larvae were fed rotifers from 2 to 5 days post hatch (dph), and Artemia from 4 to 15 dph. In T_2, larvae were kept under dark conditions and fed Artemia from 6 to 15 dph. Standard larval length (SL) was significantly higher in T₁ (p < .01) until 8 dph in comparison with larvae reared initially without rotifers. No significant difference in SL was found among treatments (p = .187) at 15 dph. Significant difference was found among treatments in survival rate at 15 dph (p < .003). The survival rate observed at 15 dph in T₂ (30 ± 4.2%) represents an important finding, although the highest survival rate was observed in T₁ (45.0 ± 3.4%). This study showed that it is possible to conduct larval rearing of meagre without using rotifers. Nevertheless, further research efforts are still needed to improve these results in comparison with the common larval rearing protocol.     

Estimation of the period sensitive for the development of abnormal morphology in brown sole <Emphasis Type="Italic">Pseudopleuronectes herzensteini</Emphasis> fed live food enriched with docosahexaenoic acid

The aim of the present study was to clarify the effect of a feeding regimen of live food enriched with docosahexaenoic acid (DHA) on the rates of abnormal morphology in larval brown sole Pseudopleuronectes herzensteini. Live foods were enriched with oil emulsion to produce both high-DHA and low-DHA content. Larvae in group 1 (control) were fed low-DHA-content live foods throughout the experiment. Conversely, larvae in group 2 were fed high-DHA live foods from 15 days post-hatching (dph), when larvae were D stage. Larvae in group 3 were fed high-DHA live foods from 25 dph, when larvae consisted mainly of E stage. The occurrence of normal morphology development in group 2 was significantly higher (P < 0.05) than those of groups 1 and 3, suggesting that normal morphology of brown sole is determined during the critical larval period (D–E stages).