Abstract

Three different egg hatching mechanisms were observed under laboratory conditions in Euphausia pacifica Hansen, Thysanoessa spinifera Holmes and Thysanoessa inspinata Nemoto: backward, forward and flipping. Like all broadcast spawning euphausiids, these species usually hatch as nauplius 1 (N1). Some hours before hatching the vitelline membrane breaks and the embryo is freely suspended within the chorion; later the embryo takes on a slightly oval shape. When ready to hatch, the N1 pushes against the chorion with the posterior part of the abdomen producing a protuberance. No spine or egg tooth is present to break the chorion. The pressure breaks the chorion, and the nauplius pushes itself backwards with the first and second antennae and mandible to slide from the chorion. After about three quarters of the body is outside, the nauplius brings all the appendages together to move backwards without becoming stuck in the chorion. This is the backward hatching mechanism. The vitelline membrane remains within the egg after the nauplius leaves the chorion. Hatching takes 5–20 s, and most of the eggs in a clutch hatch during <2 h. Several eggs of E. pacifica hatched as meta-nauplii (MN) (>200 h after spawning) or as calyptopis 1 (C1) stage (>232 h), rather than as N1. Delayed hatching of embryos also was observed in T. spinifera as nauplius 2 (N2) (>120 h) or as MN stage (>180 h), and in T. inspinata as N2 (106 h) after spawning. Eggs with larvae in stages of development beyond N1 have not been observed from preserved zooplankton samples. However, eggs spawned in the field and incubated in the laboratory also had extended development and late hatching but with low frequency (<0.06%). It is proposed that, if the backward hatching mechanism fails, alternate hatching mechanisms can be used by the euphausiid. There is high flexibility in their hatching modes. The N2 and MN break the chorion with the first and second antennae, hatching forwards, and the C1 breaks it with the telson spines and by flipping of the abdomen, resembling the decapod hatching mechanism. Delayed hatching using the forward and flipping mechanisms were associated with low hatching success in comparison with the backward hatching mechanism.

INTRODUCTION

There are 86 species of euphausiids in the oceans, of which 57 are believed to shed their eggs freely into the sea; the remaining 29 species protect their embryos by attachment to the posterior pairs of thoracic legs (Mauchline and Fisher, 1969). Euphausiids of the genera Bentheuphausia, Euphausia, Thysanoessa, Meganyctiphanes and Thysanopoda spawn freely and hatch as nauplius 1, while the genera Nematoscelis, Nyctiphanes, Pseudeuphausia and Stylocheiron brood their eggs and hatch in the early metanauplius phase as pseudometanauplii or as metanauplii (Mauchline and Fisher, 1969; Brinton et al., 2000). Nematobrachion and Tessarabrachion, having three and one species respectively, were included in the past with the genera that brood their eggs (Mauchline and Fisher, 1969). However, this has been questioned and there are no direct observations of reproduction for these four species. Data from samples collected during eight oceanographic cruises (n = 222) along Baja California (1986–1991) México, showed that adult females of Nematobrachion flexipes Ortmann were present at 50 stations. From those 50 stations, 246 immature females, 42 females with spermatophore attached, and nine females with ripe oocytes inside the carapace were recorded. No female with a brood sac was observed (Gómez-Gutiérrez, unpublished data). This information suggests that this species is a broadcast spawner. Mauchline (Mauchline, 1985) had the same conclusion for Nematobrachion boopis Calman in the Rockall Trough. Therefore, the presence of ovigerous sacs has been confirmed for only 26 euphausiid species of the family Euphausiidae (Casanova, 1984; Mauchline, 1985; Brinton et al., 2000; Ross and Quetin, 2000).

As a rule, the emergence of a larva from its egg membranes is a critical period of the life cycle of crustaceans. For example, Quetin and Ross (Quetin and Ross, 1989) showed that all embryonic stages of Euphausia superba Dana can regulate metabolism over a wide range of oxygen concentrations, but hatching eggs cannot regulate and require high oxygen levels. This suggests the vulnerability of this stage to variability in environmental oxygen concentrations. Hatching of crustacean eggs generally occurs with a sudden rupture of the chorion (Davis, 1968). There is very little information in the literature on the eclosion of euphausiid, perhaps because hatching is a simple process, lasting just a few seconds and requiring almost continuous observation under laboratory conditions, and because most researchers are most interested in egg production rates and hatching success (Komaki, 1967; Ross et al., 1982; Iguchi and Ikeda, 1994). Three reviews of hatching of invertebrate eggs reported no adequate description of hatching for the Order Euphausiacea (Davis, 1968, 1981; Anderson, 1982). There are studies of the general morphology and biometry of the eggs (Suh et al., 1993) and egg hatching success (Ross, 1981; Ross et al., 1982; Iguchi and Ikeda, 1994). However, except in brief and general descriptions by G. O. Sars (Sars, 1898) and Ponomareva (Ponomareva, 1963), hatching behavior has not been reported for broadcast spawners. Lebour (Lebour, 1950) and Ponomareva (Ponomareva, 1969) described the hatching of the eggs of Stylocheiron carinatum G.O. Sars, a tropical species that broods its eggs on its thoracic appendages.

Currently we are incubating clutches of eggs from mature females of the euphausiids Euphausia pacifica Hansen, Thysanoessa spinifera Holmes, and Thysanoessa inspinata Nemoto collected from the Oregon upwelling system in order to describe the early embryology of the eggs and to estimate egg production rates, egg development rates and egg hatching success at several temperatures. During those experiments we recorded the development and hatching of the eggs of these three species using a video camera system. In this paper I describe the hatching mechanism of nauplius 1 (N1) and report delayed hatching by other means of some of the eggs as nauplius 2 (N2), metanauplius (MN) and even calyptopis 1 (C1). This is the first report of an N2, MN or C1 inside the egg of any free spawning species of the order Euphausiacea.

METHOD

Mature females were collected at night using a 1 m diameter net with a 333 μm mesh during seven oceanographic cruises along the Newport Hydrographic line (NH line, 44°38′N) and four oceanographic cruises along the coast of Oregon and California, all part of the NE Pacific US GLOBEC sampling program. Sampling was conducted during May 18, June 27, July 18 and 30, August 5, September 10 and 18, 2001 and March 4, April 10 and May 9, 2002 at the NH5, NH15, NH20 and NH25 stations, where the number indicates the number of nautical miles from the coast (9.3–46.3 km). Additional shipboard experiments were conducted off the Oregon and northern California coasts during May 28–June 18, 2002 (Mesoscale-III cruise). Samples were obtained by double oblique tows from 10 to 20 m depth. The zooplankton catch was diluted and transferred into 20 l insulated containers filled with surface seawater. Females were transported to the cold room at Hatfield Marine Science Center (HMSC, Newport, OR) within 2–3 h after collection for the NH line cruises. During the Mesoscale-III cruise, ship incubations were done inside a cold room at 10°C. The female reproductive status stages I–IV of egg growth within the ovary described by Ross et al. (Ross et al., 1982) was applied. Mature females ready to spawn (stage IV) are easily recognized by a purple band for E. pacifica and blue–green band for T. spinifera just under the pericardial area of the cephalothorax. Healthy appearing, mature females (maturity stage IV) of both species were sorted out using a plastic spoon. Females of T. inspinata with colored ovaries never were observed (even though they spawned in the laboratory). Therefore, observing the brown spermatophore attached to the thelycum was the main selection criterion for females of T. inspinata included in the egg production experiments. The eggs of this species are heretofore undescribed (Brinton et al., 2000).

Specimens were placed into 1 l bottles filled with sieved (<64 μm) surface seawater from the stations where the females were collected. One female was placed in each bottle. From 2 to 38 replicates were made per station according to the availability of purple females in the zooplankton samples. Bottles were incubated under constant temperature 10.5 ± 0.5°C, 12 h dark and 12 h low light. This temperature was selected because it is the average value observed for the upper 20 m of the water column over the inner and middle shelf off Oregon. The females were monitored every hour after incubation to detect when they spawned. If a female spawned, she was removed from the bottle with a spoon. The females were measured for total length (mm), and the eggs were counted directly from the bottle without exposing them to the air. Fecundity was expressed as brood size: number of eggs produced per female per event (eggs fem−1). Development of the eggs was monitored using a digital video camera (CV-730 super resolution) adapted to a stereomicroscope and a digital camera (Olympus Camedia 3040, 3.3 × 106 pixels resolution). Video and digital pictures were recorded, using Capture Vision (v 1.52) or Camedia software, every 10 min during the first 8–10 h of egg development, every hour between 10–15 h, and every 2–4 h after 15 h until the end of the experiment.

Development of the eggs after spawning was classified according to the stages described by Ponomareva (Ponomareva, 1963), George and Strömberg (George and Strömberg, 1985), and Quetin and Ross (Quetin and Ross, 1984, 1989) which are defined in Table I.

RESULTS

Normal hatching process

From a total of 92 E. pacifica females incubated between May 18, 2001 and May 9, 2002, 41 produced eggs that hatched. Most of the eggs in a brood hatched almost synchronously. The average time to hatching for the first N1 of each brood, at 10.5°C (±0.5°C), was 35 h (range 27–50 h; SE ± 2.3 h, broods = 9), and more than 50% of the N1 had hatched by ∼40 h after spawning (range 29–60 h; SE ± 3.1 h, broods = 11).

A sequence of eight pictures (Figure 1A–H) shows the hatching mechanism for E. pacifica. The entire process takes 5–20 s for an individual egg. The eggs of E. pacifica, during developmental stages eLB and lLB, have a vitelline membrane around the embryo, a perivitelline space, and the outer chorion. Mauchline and Fisher (Mauchline and Fisher, 1969) believe that the chorion is really two membranes tightly laminated. When the egg is in developmental stage TW, the vitelline membrane breaks and the embryo is freely suspended within the chorion. At this stage the appendages show weak movements. A few hours before hatching the embryos change to a slightly oval shape. There is no apparent increase of egg size caused by an osmotic process. During normal hatching, after the embryo breaks the vitelline membrane (egg stages lLB to TW), the N1 pushes with its appendages against the chorion and the posterior part of the abdomen produces a protuberance (Figure 1A). In the live embryo, the posterior part of the abdomen appears light red. Perhaps a chemical substance derived from the red spot, maybe an enzyme, helps to break the chorion. No spine or egg tooth is present to break the chorion; rather pressure from the abdomen breaks it (Figure 1B). The nauplius pushes itself backwards by means of the first and second antennae and mandible and slides out of the egg (Figure 1C–D). After about three quarters of the body is outside, the nauplius brings all the appendages together to move backwards without becoming stuck in the chorion (Figure 1E–H). The vitelline membrane never leaves the chorion (Figure 1H), and it is easily seen inside after the nauplius is out.

From a total of 121 T. spinifera females incubated between May 18, 2001 and May 9, 2002, 51 produced eggs that hatched. From a total of six T. inspinata females incubated during the May 28–June 18, 2002 Mesoscale-III cruise five produced eggs, and the embryos of three clutches hatched. In general, T. spinifera and T. inspinata have the same backward hatching mechanism described for E. pacifica.

Nauplius stuck in the chorion

Occasionally some of the N1 were held back by their appendages and unable to pass completely through the hole in the chorion (usually <1% of the nauplii). Those animals tried for hours to push themselves out, working constantly. Some stuck larvae were observed to have developed to N2 and a few to MN. Most of these animals died before freeing themselves from the chorion. In one case an N1 broke the chorion at the anterior part of the egg with the appendages, then tried to push itself forwards. However, this N1 remained stuck with the chorion around its abdomen until it died.

Late hatching of euphausiid larvae

Two females of E. pacifica (EP1 and EP2) collected at NH20 on July 30, 2001, and two females collected at NH25 on April 10, and May 9, 2002 produced eggs that hatched as MN or C1. Eggs produced by those females were monitored every 2–4 h during 247 h (10.3 days). The EP1 female produced 194 eggs, but only one egg hatched normally as N1, after 38 h. However, 13 more hatched as MN ∼190 h after spawning. The rest of the eggs did not hatch. Photographs of each developmental phase of those eggs with delayed hatching are shown in Figure 2A–F. The N2 appear inside the egg ∼96 h (4 days) after spawning (Figure 2B). The N2 showed two pairs of spines on the posterior margin. The MN was first observed inside the eggs ∼103 h (4.3 days) after spawning (Figure 2C). This was easily recognized by the two striated bundles of eyestalk retinal elements typical of this phase, and because in some specimens the carapace was visible inside the egg, fringed with spines on the anterior margin and first half of the lateral margin. The mandibles were reduced and seen as small buds in the ventral view of the larva; the first antennae were unsegmented but the second antennae showed segmentation developing. The MN showed lines in the thorax between blocks of tissue forming the cephalothoracic appendages (rudiments of the mandible, maxillae 1 and 2, and maxillipeds). The tissue of the abdomen showed no segmentation. The formation of the more complicated internal organs of the larvae also took place. The first MN hatched 190 h (7.9 days), the last 239 h (10 days) after spawning. The hatching of the MN was different from that of N1. The larvae inside the eggs as N1, N2 and MN were very active, moving their appendages continuously. The chorion began to disintegrate ∼8 days after spawning, so the MN was able to break it easily in different areas with the first and second antennae, hatching forwards. Usually the appendages were the first parts extending outside the eggs (Figure 2D–E). These MN had a healthy appearance (Figure 2F) and they molted later to C1 with normal morphology and swimming behavior. The developmental progress of the larvae inside the eggs coincided well with that of the only specimen that hatched as N1 and swam freely in the petri dish as it developed to C1.

A second female (EP2) produced 121 eggs of which 67 (55%) hatched normally as N1, mostly 35–36 h after spawning. Additional embryos developed inside the egg to N2. Of the 54 remaining eggs, eight did not hatch, but 46 embryos continued to develop inside the egg. Overall hatching success was 85%. Just hatched N1 were transferred to a new beaker, and most of them developed to N2 at ∼85 h (3.5 days), to MN at ∼99 h (4.1 days), and to C1 at ∼204 h (8.5 days) after spawning at 10°C. The rest of the eggs (46) showed development similar to those that hatched normally as N1, and hatched in stage C1. When all the animals that hatched as N1 had developed to C1, only four of the 46 individuals with delayed hatching had developed to C1. The rest remained in the MN phase inside the egg, indicating a slower development rate. After 233 h, the first C1 was observed inside the eggs (Figure 2G). By then the eggs had increased in diameter considerably, and the chorions looked in better shape than those of the EP1 brood. The first body part that the C1 extended outside the egg was the abdomen, which flipped actively as the larvae tried to exit from the eggs. It is probable that the C1 slit the chorion using their telson spines (Figure 2H). The hatching C1 were healthy and similar in morphology and swimming activity to larvae hatched several days before as N1 (Figure 2I). Experiments were stopped 247 h after spawning (∼10.3 days). Eggs produced by female EP2 increased in outer diameter throughout development until reaching C1 inside the egg. The average diameter for eggs in stage SC was 0.411 mm (n = 22, range 0.390–0.439 mm), for the eggs in stage TW 0.421 mm (n = 8, range 0.415–0.439 mm), and for the eggs with the C1 inside, 0.526 mm (n = 5, range 0.510–0.530 mm). The average volume of the egg increased 2-fold, from 3.64 × 10−5 cm3 for the SC egg to 7.62 × 10−5 cm3 for the egg with the C1 inside.

Two additional E. pacifica females with broods of 35 and 60 eggs collected on April 10 and May 9, 2002, respectively, also showed late development. Only six and two animals hatched as metanauplius and calyptopis 1 ∼65 h (2.7 days) and 201 h (8.4 days) respectively after spawning at 10.5°C. None of the four broods of eggs with delayed hatching of E. pacifica hatched in stage N2.

During April 10 and May 9, 2002, 76 females of T. spinifera were incubated at 10.5°C, 21 of them produced eggs that hatched as N1, and six produced eggs that hatched as N2 or as MN. One female produced 430 eggs of which six embryos hatched as N2 and nine as MN (3.5% hatching success). But 152 eggs with N2 and 70 eggs with MN inside never hatched, and the ‘late hatchers’ were deformed in comparison with animals that hatched backwards on schedule as N1. Figure 3 shows the delayed hatching of the eggs of T. spinifera. The sequence shows the morphology of a spherical egg with the N1 inside at ∼45 h after spawning (Figure 3A). An elongation of the egg is typical of the twitching egg developed in this case 83 h after spawning (Figure 3B); the N2 inside the egg appeared ∼94–119 h after spawning, however the posterior part of the nauplius does not have typical spines but a bifurcate end (Figure 3C). About 1 h later, six N2 hatched, breaking the chorion with their appendages, resembling the E. pacifica metanauplius forward hatching mechanism (Figure 3D). Several embryos continued to develop inside the egg as metanauplius between 140 and 180 h after spawning (Figure 4E–G), and at ∼157 h a deformed larva with morphology intermediate between MN and C1 hatched using the abdomen (Figure 3F). Nine MN hatched ∼182 h after spawning, using the forward hatching method. However, those animals did not have the typical morphology of the MN that hatched earlier as nauplius 1 (Figure 3H).

Figure 4 shows the late hatching of a T. inspinata. The mother produced 68 eggs (SC stage = 360–370 μm diameter) with very low hatching success (2.9%). The N1 (TW egg stage) was observed at ∼39 h (Figure 5A) and the N2 at 113 h after spawning. Very long spines are characteristic of the N2 of this species (Figure 4B).

A comparison of the brood size (eggs fem−1) and hatching success, expressed as the percentage of eggs that hatched, of the broods from females that hatched backwards (normal hatching mechanism) and delayed hatching (forward or flipping hatching mechanism) for E. pacifica and T. spinifera is shown in Figure 5. The data set includes only the females collected at the same day and station where normal and late hatching were observed simultaneously in the laboratory. These graphs show that the average brood sizes and hatching success for both species were smaller for the broods with late hatching (forward and flipping hatchers) than for the broods with only normal hatching (backwards) (Figure 5A,B). However, a t-test for differences of the means between these two groups showed marginal significant differences only for brood size and hatching success of E. pacifica (one-sided P value = 0.069, and 0.024, d.f. = 9). The T. spinifera comparisons were not statistically significant (one-sided P value > 0.16, d.f. = 25). Because T. inspinata had a small sample size of females incubated (n = 5) with normal and late hatching, it was not compared statistically.

DISCUSSION

Hatching mechanism

This study demonstrated that ‘backward hatching’ is the normal hatching mechanism for broadcast spawning euphausiids. When this mechanism fails two unusual hatching mechanisms can potentially be used by the euphausiid embryos: the ‘forward hatching’ used by N2 and MN, and the ‘flipping hatching’ used by C1. Backward hatching is associated with relatively greater brood size and hatching success than extended development with forward and flipping hatching. Although Ross (Ross, 1981), Ross et al. (Ross et al., 1982), Suh et al. (Suh et al., 1993), Iguchi and Ikeda (Iguchi and Ikeda, 1994) and Summers (Summers, 1993) perhaps observed the mechanism of normal hatching for E. pacifica and T. spinifera during their experiments on egg production and hatching success, or for the description of early nauplii stages, they did not describe hatching because that was not among their goals. Ross and Quetin (Ross and Quetin, 1982), George (George, 1984), George and Strömberg (George and Strömberg, 1985), Marshall and Hirche (Marshall and Hirche, 1984), and Quetin and Ross (Quetin and Ross, 1984, 1989) also certainly observed nauplii of E. superba hatching. In fact, Ross and Quetin (Ross and Quetin, 1982) (their Figure 2) and George (George, 1984) (his Figure 1) showed photographs of nauplii hatching backwards, but they did not describe the process. There is little information on hatching for other species of the Order Euphausiacea (Davis, 1968,1981; Anderson, 1982). In G.O. Sars’s (Sars, 1898) description of the eggs of Meganyctiphanesnorvegica M. Sars, in Ponomareva’s (Ponomareva, 1963) review of euphausiid biology in the North Pacific, and in the landmark reviews of euphausiids by Mauchline and Fisher (Mauchline and Fisher, 1969) and Mauchline (Mauchline, 1980) the hatching process is described only briefly and in general terms. For example, G.O. Sars (Sars, 1898) reported that: ‘the movements become more energetic, and at last cause the rupture of the thin external coating of the sphere, whereupon the young individual escapes in a very imperfect condition, as a rather simple nauplius.’ Ponomareva (Ponomareva, 1963) describes the hatching process as ‘the fully formed nauplius ruptures the egg membrane and emerges’. Ponomareva (Ponomareva, 1969) reported for Stylocheiron carinatum G.O. Sars the only description of the hatching of eggs for species that brood their eggs. She said, ‘the perfectly formed nauplii extended and contracted the anterior part of the appendages, pining the egg membrane by short dense spines and then, parting their appendages (swimming movements), ruptured the nearest part of the egg membrane and subsequently emerged into the surrounding water.’ This description suggests that the nauplius of S. carinatum had forward hatching. However, she did not show drawings of the hatching process. It is interesting that Ponomareva (Ponomareva, 1963) and Lebour (Lebour, 1950) reported a nauplius emerging from the egg of S. carinatum rather than a pseudometanauplius or a metanauplius, which are more usual hatching stages in euphausiid species with an egg sac (Komaki, 1967; Brinton et al., 2000).

Iguchi and Ikeda (Iguchi and Ikeda, 1994) also reported nauplii of E. pacifica hatched out almost synchronously. Hatching of the three species observed here appears to be primarily mechanical, but the observation of the light red coloration at the posterior part of the abdomen of the N1 just before hatching suggests that an enzyme could play a part in the dissolution of the chorion. An enzymatic mechanism has been observed in embryos of other marine crustaceans (De Vries and Forward, 1991; Saigusa and Terajima, 2000). Saigusa and Terajima (Saigusa and Terajima, 2000) reported that embryos of the estuarine crab Sesarma haematocheir have a pair of glands of 30 μm diameter located in the dorsal thorax adjacent to the site where the egg case is ruptured. They recognized two active substances released outside the egg case: one is a caseinolytic protease and the other an unidentified enzyme. The activity of the former was at a very low level compared with a standard casein-lytic enzyme. It might be digesting the thin, sticky internal layer enclosing the embryo but would not act on the thick, tough layer constituting the main component of the egg capsule (Saigusa, 1996). The second substance, unidentified so far, may soften the egg case. They observed that the egg case seemed to be softened during a few hours preceding hatching (Saigusa and Terajima, 2000). The N1 of E. pacifica and T. spinifera, unlike the N2, does not have two pairs of spines on the posterior margin (Suh et al., 1993; Summers, 1993) to help open the chorion of the egg. I believe that hatching is primarily mechanical because the nauplius completely fills the perivitelline space and can exert pressure against the chorion. The possibility of enzymatic weakening of the chorion is an open question.

The empty chorion of the euphausiid species is flexible, and the hole the nauplius makes for hatching is not easily detected. The vitelline membrane remains inside the chorion after the nauplius completely escapes from both envelopes. Hatching of E. superba, reported by Ross and Quetin (Ross and Quetin, 1982) and George (George, 1984), must have a similar mechanism. Hatching of euphausiids is basically different from that of other holoplanktonic crustaceans including Anostraca, Cladocera, Copepoda, Branchiura, Mysidacea, Isopoda, Amphipoda and pelagic Decapoda (Lucifer), which are extensively discussed by Davis (Davis, 1968, 1981) and Anderson (Anderson, 1982). In Calanoid copepods, for example, Marshall and Orr (Marshall and Orr, 1954) and Peterson (Peterson, 1980) reported the chorion splits by pressure from within caused by expansion of the vitelline membrane. The inner membrane expands until it is much larger than the original egg, and within this blister membrane the unattached nauplius begins to swim actively. They believe that substances of high osmolality secreted by the embryo cause the inflow of environmental water, increasing the pressure inside the egg. The thin membrane is finally burst by the activities of the nauplius, which beats upon the membrane with the setae of its appendages. The euphausiid nauplius breaks the vitelline membrane inside the egg and this membrane remains within the egg even after the nauplius leaves the chorion. Comparative studies leave no doubt that the basic pattern of development in the Crustacea includes a small yolky egg, modified spiral cleavage, hatching as a nauplius larva and anamorphic development of the post-naupliar somites during a series of larval stages with intervening moults (Anderson, 1982; Dahms, 2000). However, the hatching mechanism among different crustacean orders has important differences, perhaps reflecting their phylogenetic separation.

Nauplii stuck in the empty chorion

A low percentage of the E. pacifica eggs observed (<1%) remained stuck in the chorion. Stuck larvae have also been observed for the eggs of Branchiura, in which nauplii trapped by the springiness of the split chorion struggle for many hours before dying (Davis, 1968). The N1 at hatching has setae on the first and second antennae, which sometimes entangle with the chorion. An N1 could be stuck inside the chorion due to the small size of the hole, inadequate extension of the antennal appendages to push out of the chorion, or weakness. An N1 can develop to N2 and MN while stuck in the chorion because those phases do not require exogenous food (Mauchline and Fisher, 1969; Suh et al., 1991). The chorion is flexible enough to permit the increase in size of the larvae without breaking. Usually N2 and MN have a fast escape response to moving objects. Being stuck in the chorion wastes energy on escape efforts and certainly makes them more vulnerable to predation in the field.

Late hatching

This is the first time an abnormally delayed hatching has been observed for any of the broadcast spawning euphausiids. The N2 and MN break the egg chorion in different areas with the first and second antennae hatching forwards and the C1 breaks it with the telson and the abdomen, resembling the decapod hatching mechanism (Davis, 1968; Saigusa and Terajima, 2000). Delayed hatching was observed in three euphausiid species that inhabit relatively different distributional patterns: E. pacifica is widely distributed in the North Pacific, the Japan Sea, and the Yellow Sea (Suh et al., 1993). T. spinifera is considered a neritic species and occurs along the western coast of North America from the Southeastern Bering Sea, to as far south as mid-Baja California (28°N) during particularly cold springtimes (Brinton, 1962; Brinton et al., 2000). T. inspinata is an oceanic species and is usually recorded in low abundances across the North Pacific in the zone 40–45°N; it is known to distribute in the California Current southwards to Oregon (Brinton et al., 2000). It is well known that eggs of these species hatch as N1 (Ross, 1981; Summers, 1993), as do the eggs of the other species of Euphausiacea that shed their eggs in the ocean (Brinton et al., 2000). Iguchi and Ikeda (Iguchi and Ikeda, 1994) reported the hatching time of eggs of E. pacifica as a function of temperature and estimated that at 10.5°C the hatching time would be 1.26 days (30 h). I observed that normal hatching of the three species ranged between 35 and 45 h at this temperature (unpublished data). The eggs with delayed hatching of E. pacifica, T. spinifera and T. inspinata hatched at ∼9–10 days (220–247 h), 8 days (182 h), and 4 days (106 h) after spawning. The timing of hatching is controlled primarily by temperature (Ross et al., 1988; Iguchi and Ikeda, 1994), but the physiological mechanism underlying the timing of hatching for normal and delayed eggs is unknown because normal and delayed hatching were observed from females collected from the same day and location (station), and incubated under the same laboratory conditions. Nevertheless, it is possible that laboratory conditions, like a ‘wall effect‘, where the eggs are incubated in a bottle or petri dish, could promote delayed hatching.

Previous studies of egg development of M. norvegica M. Sars, Thysanoessa raschi M. Sars, Thysanoessa inermis Kroyer, and E. superba did not show any abnormality or suggest they could hatch at phases older than N1 (Sars, 1898; Taube, 1909; Ponomareva, 1963; Ross and Quetin, 1982; Marshall and Hirche, 1984; George and Strömberg, 1985). However, Zelikman (Zelikman, 1961) (his Figure 2) showed a drawing of a nauplius breaking the chorion with the second antennae and mandible suggesting a forward hatching mode for T. inermis. Apparently egg development of species that shed their eggs into the ocean may be more flexible than previously thought. Therefore broadcast spawning euphausiids have several alternate hatching mechanisms (backward, forward and flipping). It would be interesting to know how a larva could molt inside the egg. This certainly occurs in eggs of species that brood their eggs and hatch as pseudometanauplius or metanauplius. Saigusa and Terajima (Saigusa and Terajima, 2000) using light and electron microscopy observed two thin layers (ex1 and ex2) inside the eggs of the crab S. haematocheir, which were identified as exuvia deposited by the embryos at late stages of development. It is possible that the advance from metanauplius to calyptopis also involves a molt inside the egg, leaving the exuvia within, but that was not observed.

The N2, MN and C1 are typically motile phases. Similar to larvae stuck in the chorion of the eggs, the eggs with delayed hatching time had the disadvantage of a lack of motility to avoid predators. In the hypothetical scenario that delayed hatching of N2 and MN occurs in the field, it might have a small ecological impact because those phases do not have a mouth and do not feed. They obtain their nourishment from the remainder of the yolk present in their body. However, the molt from MN to C1, when the feeding appendages are added, seems to be critical (Ross, 1981; Ross et al., 1988). The first calyptopis has functional mouthparts and feeds by filtering suspended matter, including phytoplankton and marine snow, from the surrounding water (Suh et al., 1991; Dilling et al., 1998). According to Ikeda (Ikeda, 1984), the C1 is the stage with the highest metabolic activity and the lowest energy store in the body; therefore survival of this stage is very sensitive to food limitation. There is a critical period after the larvae develop the ability to feed, also known as the point-of-no-return or (PNR), when the starved larvae cannot recover if re-fed and eventually die (Ross and Quetin, 1989; Quetin and Ross, 1989). If the C1 spends more time inside the egg before hatching, it would have less time to feed, so the survival of those individuals could be reduced. Ross and Quetin (Ross and Quetin, 1989) calculated that the PNR of E. superba larvae is between 9 and 15 days after they molt from MN to C1 for temperatures between −1 and 2°C. Because E. pacifica, T. spinifera, and T. inspinata larvae usually inhabit waters with temperatures >8°C the metabolic expenses could be greater and the PRN could be shorter than 9 days.

The egg brood size and hatching success of both species were slightly lower for the broods with delayed hatchers than the broods which had normal backward hatching (Figure 5A,B). The hatching success values reported in this study were relatively lower than hatching success reported by Ross (Ross, 1981) and Iguchi and Ikeda (Iguchi and Ikeda, 1994) for E. pacifica (usually >95%). It is uncertain if eggs of E. pacifica, T. spinifera, and T. inspinata with abnormally delayed hatching occur in the field. However, eggs spawned in the field and incubated in the laboratory occasionally had embryos that hatched forward as N1, a typical hatching mode for delayed hatching. In addition, eggs spawned at the field and incubated under laboratory conditions had delayed hatching but with low frequency (0.06%, n = 1751 eggs). Euphausiids eggs with an N2, MN or C1 inside have never been seen in a seven year (1996–2002) biweekly sampling program along the Newport hydrographic line (L. Feinberg and W. T. Peterson, personal communication). I did not observe any in the examination of biweekly samples for the same stations collected in 1970–1972. The verification of the occurrence of delayed hatching of the eggs of these three species in the field and evaluation of the impact on their population dynamics awaits future studies. Researchers need to observe eggs for many days beyond normal hatching to avoid an underestimate of the egg hatching success. Also, future observations of hatching mechanisms for other euphausiid species around the world will show whether the hatching mechanism and occasional late development observed for E. pacifica, T. spinifera, and T. inspinata are general among euphausiids that spawn freely.

Table I
Egg stage  Description 
Single cell (SC) Recently spawned, perivitelline space minute, and no sign of cleavage. 
Multiple cell (MC) Initial stage of cleavage to multicellular blastula. 
Gastrula (G) Formation of two layers of cells enclosing a central cavity (archenteron). 
Early limb bud (eLB) The embryo is transformed into a nauplius, the naupliar appendages still connected with the body by a membrane and limb primordia visible in lateral view as ridges. 
Late limb bud (lLB) The distal ends of the limbs have become free, tube-like structures. 
Twitch (TW) The nauplius has taken shape, no membrane surrounding it, appendages freely suspended from the body and, in a live egg, the nauplius moves the appendages and has a pulsating heart. 
Egg stage  Description 
Single cell (SC) Recently spawned, perivitelline space minute, and no sign of cleavage. 
Multiple cell (MC) Initial stage of cleavage to multicellular blastula. 
Gastrula (G) Formation of two layers of cells enclosing a central cavity (archenteron). 
Early limb bud (eLB) The embryo is transformed into a nauplius, the naupliar appendages still connected with the body by a membrane and limb primordia visible in lateral view as ridges. 
Late limb bud (lLB) The distal ends of the limbs have become free, tube-like structures. 
Twitch (TW) The nauplius has taken shape, no membrane surrounding it, appendages freely suspended from the body and, in a live egg, the nauplius moves the appendages and has a pulsating heart. 
Fig. 1.

Sequence of video pictures of the normal backward hatching process of the eggs of the euphausiid E. pacifica under laboratory conditions at 10.5°C (±0.5°C, living specimens). The entire hatching process takes just 5–20 s. Scale bar = 100 μm.

Fig. 1.

Sequence of video pictures of the normal backward hatching process of the eggs of the euphausiid E. pacifica under laboratory conditions at 10.5°C (±0.5°C, living specimens). The entire hatching process takes just 5–20 s. Scale bar = 100 μm.

Fig. 2.

Delayed hatching of the eggs of E. pacifica ‘EP1’ and ‘EP2’ females collected in the NH20 station during July 30, 2001. The sequence shows the morphology of the egg of stage TW with the nauplius 1 inside the egg 36–51 h after spawning (A). The nauplius 2 developed between 70 and 96 h after spawning (B) and metanauplius between 90 and 220 h (C). From this point the broods of EP1 and EP2 had different development and hatching mechanism. About 200 h after spawning the chorion of the eggs of the EP1 brood began to disintegrate and all the eggs hatched as metanauplius (DF). For the EP2 brood, calyptopis 1 was observed 232 h (G) and they hatched around 347 h after the spawning (H). 63 eggs of the EP2 brood hatched as N1 65 h after spawning and 46 eggs hatched as C1 (I). Scale bar = 100 μm.

Fig. 2.

Delayed hatching of the eggs of E. pacifica ‘EP1’ and ‘EP2’ females collected in the NH20 station during July 30, 2001. The sequence shows the morphology of the egg of stage TW with the nauplius 1 inside the egg 36–51 h after spawning (A). The nauplius 2 developed between 70 and 96 h after spawning (B) and metanauplius between 90 and 220 h (C). From this point the broods of EP1 and EP2 had different development and hatching mechanism. About 200 h after spawning the chorion of the eggs of the EP1 brood began to disintegrate and all the eggs hatched as metanauplius (DF). For the EP2 brood, calyptopis 1 was observed 232 h (G) and they hatched around 347 h after the spawning (H). 63 eggs of the EP2 brood hatched as N1 65 h after spawning and 46 eggs hatched as C1 (I). Scale bar = 100 μm.

Fig. 3.

Delayed hatching of the eggs of T. spinifera female collected in the HH-4 station during April 10, 2002. The sequence shows the morphology of the egg stage TW with the nauplius 1 inside the egg at ∼45 h after spawning (A). Twitching egg developed 83 h after spawning (B), nauplius 2 inside the egg 94 h (C), nauplius 2 hatching 120 h (D), metanauplius inside the eggs 140 h (E), deformed larvae between metanauplius and calyptopis 157 h (F), metanauplius inside the egg 180 h (G), and free metanauplius >180 h after spawning (H). Scale bar = 100 μm.

Fig. 3.

Delayed hatching of the eggs of T. spinifera female collected in the HH-4 station during April 10, 2002. The sequence shows the morphology of the egg stage TW with the nauplius 1 inside the egg at ∼45 h after spawning (A). Twitching egg developed 83 h after spawning (B), nauplius 2 inside the egg 94 h (C), nauplius 2 hatching 120 h (D), metanauplius inside the eggs 140 h (E), deformed larvae between metanauplius and calyptopis 157 h (F), metanauplius inside the egg 180 h (G), and free metanauplius >180 h after spawning (H). Scale bar = 100 μm.

Fig. 4.

Delayed hatching of the eggs of T. inspinata female collected in the RR-7 station during June 13, 2002. The sequence shows the morphology of the egg stage TW with the nauplius 1 inside the egg at ∼39 h after spawning (A), and nauplius 2 hatching ∼113 h after spawning (B). Scale bar = 100 μm.

Fig. 4.

Delayed hatching of the eggs of T. inspinata female collected in the RR-7 station during June 13, 2002. The sequence shows the morphology of the egg stage TW with the nauplius 1 inside the egg at ∼39 h after spawning (A), and nauplius 2 hatching ∼113 h after spawning (B). Scale bar = 100 μm.

Fig. 5.

Comparison of the brood size (eggs fem−1) (A) and hatching success, expressed as the percentage of eggs that hatched from the spawned brood size (B), of the broods from females that hatched backwards (normal hatching mechanism, empty bars) and late hatching (forward or flipping hatching mechanism, black bars) for E. pacifica and T. spinifera under laboratory conditions. Dataset includes only females collected at the same date and station where normal and late hatching was observed under the laboratory conditions. Vertical lines are standard error (Sx).

Fig. 5.

Comparison of the brood size (eggs fem−1) (A) and hatching success, expressed as the percentage of eggs that hatched from the spawned brood size (B), of the broods from females that hatched backwards (normal hatching mechanism, empty bars) and late hatching (forward or flipping hatching mechanism, black bars) for E. pacifica and T. spinifera under laboratory conditions. Dataset includes only females collected at the same date and station where normal and late hatching was observed under the laboratory conditions. Vertical lines are standard error (Sx).

1
Present Address: Oregon State University, College Of Oceanic And Atmospheric Sciences, Oregon State University, 104 Ocean Administration Building, Corvallis, Or 97331-5503, USA

This research is part of the PhD thesis of Jaime Gómez-Gutiérrez (COAS-OSU) and was supported by funds provided by a Mamie Markham Research Award from Oregon State University, Hatfield Marine Science Center (HMSC, Newport, OR) during 2001–2002. Thanks to Charles Miller (COAS), William Peterson (NOAA), Margaret Knight (SIO), Edward Brinton (SIO), Annie Townsend (SIO), Masayuki Saigusa (Okayama University), and Cheryl Morgan (HMSC) for their valuable comments. I am in debt to Leah Feinberg, Tracy Shaw, Julie Keister, Anders Roestad, Mitch Vance and Jesse Lamb for their collaboration in the laboratory and the collection of the live samples in the rough Oregon Sea. The author JGG is also supported by an SNI fellowship, by COFAA-IPN, and a PhD CONACyT grant (122676) to study at Oregon State University. The US GLOBEC program (NA860P0589) provided ship time and staff assistance. This is publication number 343 from the NE Pacific US GLOBEC program.

REFERENCES

Anderson, D. T. (1982) Embryology. In Abele, L. G. (ed.), The Biology of Crustacea, Embryology, Morphology, and Genetic. Academic Press, London, Vol. 2, pp. 1–41.
Brinton, E. (
1962
)
The Distribution of Pacific Euphausiids
 . Bulletin of the Scripps Institution of Oceanography, University of California,
8
,
51
–270.
Brinton, E., Ohman, M. D., Townsend, A. W., Knight, M. D. and Bridgeman, A. L. (2000) Euphausiids of the World Ocean. Series: World Biodiversity Database CD-ROM Series Window version 1.0. Expert Center for Taxonomic Identification, Amsterdam, Netherlands.
Casanova, B. (
1984
) Phylogénie des Euphausiacés (Crustacés Eucarides).
Bull. Mus. Natn. Hist. Nat.
 ,
4
,
1077
–1089.
Dahms, H. U. (
2000
) Phylogenetic implications of the crustacean nauplius.
Hydrobiologia
 ,
417
,
91
–99.
Davis, C. C. (
1968
) Mechanisms of hatching in aquatic invertebrate eggs.
Oceanogr. Mar. Biol. Ann. Rev.
 ,
6
,
325
–376.
Davis, C. C. (
1981
) Mechanisms of hatching in aquatic invertebrate eggs. II.
Oceanogr. Mar. Biol. Ann. Rev.
 ,
19
,
95
–123.
De Vries, M. C. and Forward, R. B. Jr. (
1991
) Mechanisms of crustacean egg hatching: evidence for enzyme release by crab embryos.
Mar. Biol.
 ,
110
,
281
–291.
Dilling, L., Wilson, J., Steinberg, D. and Alldredge, A. (
1998
) Feeding by the euphausiid Euphausia pacifica and the copepod Calanus pacificus on marine snow.
Mar. Ecol. Progr. Ser.
 ,
170
,
189
–201.
George, R. Y. (
1984
) Ontogenetic adaptations in growth and respiration of Euphausia superba in relation to temperature and pressure.
J. Crust. Biol.
 ,
4
,
255
–262.
George, R. Y. and Strömberg, J. O. (
1985
) Development of eggs of Antarctic krill Euphausia superba in relation to pressure.
Polar Biology
 ,
4
,
125
–133.
Iguchi, N. and Ikeda, T. (
1994
) Experimental study on brood size, egg hatchability and early development time of the euphausiid Euphausia pacifica from Toyama Bay, Southern Japan Sea.
Bull. Sea Nat. Fish. Res. Inst.
 ,
44
,
49
–57.
Ikeda, T. (
1984
) Development of the larvae of the Antarctic krill (Euphausia superba, Dana) observed in laboratory.
J. Exp. Mar. Biol. Ecol.
 ,
75
,
107
–117.
Komaki, Y. (1967) On the early metamorphosis of Nematoscelis difficilis Hansen (Euphausia, Crustacea). Inf. Bull. Plankt. Jap., Special volume, 101–108.
Lebour, M. V. (
1950
) Some euphausiids from Bermuda.
Proc. Zool. Soc. Lond.
 ,
119
,
823
–837.
Marshall, H. P. and Hirche, H. J. (
1984
) Development of eggs and nauplii of Euphausia superba.
Polar Biol.
 ,
2
,
245
–250.
Marshall, S. H. and Orr, A. P. (
1954
). Hatching in Calanus finmarchicus and some other copepods.
J. Mar. Biol. Ass. UK.
 ,
33
,
393
–4–1.
Mauchline, J. (1980) Biology of mysids and euphausiids. In Blaxter, J. H. S., Russell, F. S. and Younge, M. (eds.), Advances in Marine Biology. Academic Press, London, Vol. 18, pp. 1–681.
Mauchline, J. (
1985
) Growth and production of Euphausiacea (Crustacea) in the Rockall Trough.
Mar. Biol.
 ,
90
,
12
–26.
Mauchline, J. and Fisher, L. R. (1969) The biology of the euphausiids. In Russell, F. S. and Younge, M. (eds.), Advances in Marine Biology. Academic Press, London, Vol. 7, pp. 1–454.
Peterson, W. T. (1980) Life history and ecology of Calanus marshallae Frost in the Oregon upwelling zone. PhD thesis. Oregon State University.
Ponomareva, L. A. (1963) Euphausiids of the North Pacific. Their Distribution and Ecology. Academy of Sciences in the USSR Institute of Oceanology. US Department of the Interior and the National Science Foundation. Translated from Russian. pp. 1–154.
Ponomareva, L. A. (
1969
) Investigations on some tropical euphausiid species of the Indian Ocean.
Mar. Biol.
 ,
3
,
81
–86.
Quetin, L. B. and Ross, R. M. (
1984
) Depth distribution of developing Euphausia superba embryos, predicted from sinking rates.
Mar. Biol.
 ,
79
,
47
–53.
Quetin, L. B. and Ross, R. M. (
1989
) Effects of oxygen, temperature and age on the metabolic rate of the embryos and early larval stage of the Antarctic krill Euphausia superba Dana.
J. Exp. Mar. Biol. Ecol.
 ,
133
,
103
–127.
Ross, R. M. (
1981
) Laboratory culture and development of
Euphausia pacifica. Limnol. Oceanogr.
 ,
26
,
235
–246.
Ross, R. M. and Quetin, L. B. (
1982
) Euphausia superba: fecundity and physiological ecology of its eggs and larvae.
Antarct. J. USA
 ,
17
,
166
–167.
Ross, R. M. and Quetin, L. B. (
1989
) Energetic cost to develop to the first feeding stage of Euphausia superba Dana and the effect of delays in food availability.
J. Exp. Mar. Biol. Ecol.
 ,
133
,
103
–127.
Ross, R. M. and Quetin, L. B. (2000) Reproduction in Euphausiacea. In Everson, I. (ed.), Krill Biology, Ecology and Fisheries. Fisheries and Aquatic Resources Series. Blackwell Science, Cornwall. pp. 150–181.
Ross, R. M., Daly, K. L. and English, T. S. (
1982
) Reproductive cycle and fecundity of
Euphausia pacifica. Limnol. Oceanogr.
 ,
27
,
304
–314.
Ross, R. M., Quetin L. B. and Kirsch, E. (
1988
) Effect of temperature on development times and survival on early larval stages of Euphausia superba Dana.
J. Exp. Mar. Biol. Ecol.
 ,
121
,
55
–71.
Saigusa, M. (
1996
) Two kinds of active factors in crab hatch water: Ovigerous-hair stripping substance (OHSS) and a proteinase.
Biol. Bull.
 ,
191
,
234
–240.
Saigusa, M. and Terajima, M. (
2000
) Hatching of an estuarine crab, Sesarma haematocheir: from disappearance of the inner (E3) layer to rupture of the egg case.
J. Exp. Zool.
 ,
287
,
51
–523.
Sars, G. O. (
1898
) On the propagation and early development of Euphausiidae.
Arch. Math. Naturv.
 ,
20
,
1
–41.
Suh, H. L., Toda, T. and Terazaki, M. (
1991
) Diet of the calyptopis of the euphausiid Euphausia pacifica in the Yellow Sea.
Mar. Biol.
 ,
111
,
45
–48.
Suh, H. L., Soh, H. Y. and Hong, S. Y. (
1993
) Larval development of the euphausiid Euphausia pacifica in the Yellow Sea.
Mar. Biol.
 ,
115
,
625
–633.
Summers, P. L. (1993) Life history, growth and aging in Thysanoessa spinifera. MSc thesis, University of Victoria. 209 pp.
Taube, E. (
1909
) Beitrage zur Entwicklungsgeschichte der Euphausiden. 1. Die Furchung des Eies bis zum Gastrulation. Z. Wiss. Zool., Abt., A,
92
,
427
–464.
Zelikman, E. A. (
1961
) Morphology and the early stages of development species of Barents Sea euphausiids.
Trudy. Murmansk. Boil. Sta.
 ,
3
,
23
–35.