.gif)
Some Studies on the New Zealand Oysters
2. Reproduction and Development of O. Lutaria
2. Reproduction and Development of O. Lutaria
Ostrea lutaria reproduces for at least eight months of the year, between August and March and possibly for longer when favourable conditions are present. Sections of the gonads show that sexual products are present all the year and that a variety of sexual phases are represented including hermaphrodite individuals. The first functional phase appears to be that of the male while the female phase does not develop until the third summer. When ovulation occurs, the eggs are shed from the gonad and retained in the inhalent mantle chamber where fertilisation and development take place. The New Zealand mud-oyster is thus an incubatory species.
During the present study, all stages of incubation were observed in oysters held in the laboratory which had been induced to liberate their larvae by a rise in sea water temperature. The larvae are liberated when they are provided with a shell and pigment spots and are capable of swimming. A study of the plankton was made to find free swimming larvae. This investigation was carried out over a period of one year, including two summers and resulted in the identification of the larvae of two species of mussel and four other bivalve larvae, but no oyster larvae. The process of settling and metamorphosis were also observed in the laboratory.
The larvae were studied in the first instance as live material and these observations were supplemented by whole mounts stained with acetic acid-alum-carmine. Attempts to section the embryo were unsuccessful.
Maturity and Sexual Phases
Sexual cycles of many marine organisms are more often than not governed by the complexity of their reproductive organs and the environmental conditions during the spawning months. However, the anatomical simplicity of the reproductive organs of oysters and other bivalves in general reflects their simple sexual reactions.
Sperm balls are generally considered to be characteristic of hermaphroditic and larviparous species of oyster. Sperm balls are characteristic of O. lutaria so that it could be expected also to be hermaphroditic and larviparous; this proved to be the case. The argument that sperm balls indicate hermaphroditism and the larviparous condition in oysters that possess them is taken further to indicate that such oysters are also protandric, and have a rhythmical alternation of female and male phases throughout the remainder of their lives. As regards the last two characteristics, it must be mentioned here that although O. lutaria appears to be protandric it is not proven and although sex change does occur, the lack of tagged individuals has made it impossible at this stage to say whether the sex change is a rhythmical alternation.
The young were never observed to become sexually mature in the summer during which they attached themselves. Maturity was only reached during the second summer when oysters measured approximately 20mm in length. In general, this agrees with the observations of Sparck (1925) on O. edulis when he noted that the young of this species may in exceptional circumstances produce eggs in the second summer. Individuals of O. lutaria were in their third summer at least and more commonly in their fourth summer before egg development commenced.
Although no experiments were carried out with O. lutaria as regards temperature and time requirements for maturation, it is noted that an oyster collected in August at a water temperature of 10.0°C, liberated larvae in the laboratory. The fact that the surface water temperatures in the previous few weeks was below 10.0°C. (average 9.4°C.) clearly indicates that O. lutaria is capable of maturation and spawning at temperatures as low as 10.0°C. Whether this was an exceptional case of spawning is unknown and it can only be said here that during the course of this study no other oysters were collected which spawned at similar temperatures.
Sexual phases: During the course of this study several young oysters were collected. These young oysters ranged in size from 4mm to 25mm, the latter being about one year old. The gonads gave no appearance of development until the oyster was at least a year old, or 20mm in length. Smears made of these gonads produced on all occasions spermatids and mature sperm balls. Whether or not the primary gonad is bisexual is not known since the limited material was insufficient to enable the author to clarify this point. Protandry is considered to dominate the sexuality of larviparous and oviparous species of oysters (Stafford, 1913; Coe, 1931, 1932) and it would appear that further investigation on O. lutaria will prove it to be protandric.
The sequence of sexual phases is hard to study in the oyster for two reasons: (1) there is a need for tagged oysters so that an all year round check can be made on the gonads that were originally known to be either male or female, and (2) the sequence is further complicated by the fact that different regions of the same gonad may have different sexual phases at the same time. Thus, the tubules further away from the main area of proliferation, i.e., beneath the hinge may be in an advanced state of spermatogenesis while the area of the gonad nearer the pericardium may contain ripe eggs.
In the present study sections and smears were made of just over 100 gonads to show the types of sexual phase exhibited by O. lutaria. The phases were as
page 11
follows: hermaphrodite individuals with ripe sperm and eggs; females but with some early proliferation of spermatogonia; males, with no trace of oogenesis; males with oogonia present; males with oocytes lining the wall of the follicles. From this list it can be seen that no females were observed which did not show some phase of spermatogenesis. Orton (1927, p. 976) regards all oysters in which ripe ova are present as essentially functional females, irrespective of the amount of spermatogenesis occurring in the secondary and primary follicles.
The number of sexual phases that O. lutaria may pass through in a year, or even in a life-time, is unknown. In O. edulis however, the indifferent phase is followed by spermatogenesis. Thus, the lack of eggs in young O. lutaria suggests that in the New Zealand mud-oyster also spermatogenesis succeeds an indifferent phase. The exact age of O. lutaria when oogonia begin to develop is unknown. Nonetheless, the gonads of oysters in their third summer exhibit a characteristic male phase with ripe sperm morulae in the lumen of the follicles and young oogonia and oocytes lining the walls of the follicles. The development of this first observed female phase proceeds concurrently with the development and further maturation of the sperm morulae.
In the transitional phase between male and female phases the developing oocytes enlarge, thus leaving the ripe sperm morulae isolated in the lumen of the follicles. Spermatogonia that would appear to belong to a subsequent male phase lie between the developing ova round the wall of the follicle. The fact that ripe sperm balls and mature eggs occur together in the same follicle does not mean that self-fertilisation will necessarily result, for the individual spermatozoa of the sperm ball are tightly held together until they are liberated by contact with the sea water. Only if the eggs and sperm are liberated simultaneously from the gonad will there be opportunity for self-fertilisation. If self-fertilisation does occur, then the ripe sperm and eggs have been developed in an hermaphrodite gonad.
Hermaphroditism: Coe (1932, p. 136) notes that hermaphroditism is not rare in the genus Ostrea as at least ten of the more than sixty described species are known to be monoecious and viviparous. There has been a suggestion that hermaphrodite oysters "are derived from individuals which spawned incompletely, at the end of the previous season leaving behind in the gonad a fair number of unripe eggs which were retained throughout the winter'' (Cole, 1942). The true factors governing the formation of hermaphrodite follicles are still unknown. Cole's suggestion when considered relative to the presence of hermaphrodite follicles at the beginning of a season seems reasonable, but it is hard to see how it can account for hermaphrodite follicles being present all through the breeding season.
Korringa (1941, p. 46) notes that "as the eggs develop, the oysters remain functioning as males for a considerable time and often sperm production continues until about ten days before the discharge of the eggs. When we consider that a few days after the shedding of the eggs the gonad again contains sperm morulae, we may conclude that the purely female phase can be very short, about three weeks". In O. lutaria the follicles that possessed developing or mature eggs always possessed in addition developing spermatozoa. Thus the gonads of three year old and older oysters of O. lutaria always appeared to be hermaphroditic and in a variety of intersexual forms.
Gonads after ovulation: Serial sections were prepared of the gonads of oysters that had liberated larvae in the laboratory. Sections were also made of gonads while the parent oyster was still incubating veliger larvae. In all these sections it could be seen that the secondary follicles of the gonad had collapsed, having become very irregular in outline and considerably lessened in size. The
page 12
genital canal connecting the lumen of the secondary follicles was also noticed to be considerably narrower than that found in sexually mature oysters. Phagocytes were present throughout the primary and secondary follicles and in the surrounding connective tissue. A few degenerated eggs were noticed scattered in the follicles.
Orton (1927, p. 974) shows that it is normal for unspent eggs to be voided from the gonad after the initial main act of spawning. He also considered that if the residual ova were fertilised upon spawning, they would then give rise to a successful but spurious second spawning. Thus, two stages of developing larvae would be likely to be found within the inhalent chamber. Such a condition was not observed in O. lutaria.
Orton (1927, p. 974) suggests that in O. edulis degeneration with absorption of eggs usually and possibly always occurs, while Coe (1932, p. 133) states that in O. lurida the few remaining ripe ova may eventually undergo degenerative changes and phagocytosis. This latter course of development occurs in O. lutaria where phagocytes containing the nuclear material of the unspent ova occur in the lumen of the follicles and similar phagocytes without the nuclear contents occur in the surrounding connective tissue. Phagocytes were present in the follicles of oysters that had spawned some ten to fourteen days earlier. This persistence of phagocytes in the follicles of O. lutaria is in direct contrast to that recorded for other oysters.
Coe (1932) and Orton (1927) state in O. edulis, O. lurida and other incubatory oysters, spermatogenesis begins immediately after the eggs have been liberated from the follicles and that the follicles have abundant mature sperm balls at the time when the larvae are fully developed and ready to be liberated. Spermatogenesis in these oysters continues until approximately eight weeks after oxidation.
In the twelve post-spawned gonads that were sectioned, there was complete absence of spermatogenesis and oogenesis. The twelve gonads were representative of early summer and late summer spawners. Since there were no adequate facilities for keeping the spawned oysters alive indefinitely, the interval between ovulation and the commencement of spermatogenesis is unknown. The length of time elapsing between the first and second female phase is also unknown.
Resting period and wintering condition: After the ripe sperm balls and eggs have been discharged the appearance of the body alters considerably, for the tissues become flabby, watery and translucent. This translucent condition which enables the brownish digestive diverticular tubules to be observed, remains until the next sex phase is entered upon. Oysters that appeared to be in a resting phase were observed throughout the year but with more frequency during the months of March, April, May, June and July. During these months, the resting condition may last longer than in the summer months. Sections of the gonads during these months showed that oogenesis and spermatogenesis still occurred but not to the point where ripe eggs and sperm were present. Cole (1942, p. 343) notes that the "stage at which an individual passes the winter is conditioned by the phase it reaches at the close of the previous season".
The duration of the spawning season for O. lutaria is eight months at least and possibly all the year round when favourable conditions are present. Thus there is never a long period of over-wintering but rather a resting period with the spermatocytes and oocytes ready to mature when the conditions become more favourable. Since mature eggs and sperm were never found in the sections made of over-wintering gonads, in O. lutaria it may be assumed that phagocytes have broken down and absorbed any ripe products left from the previous season. The functional sexual phase in the following season will be that of the dominant sex cells in the resting follicles.
Spawning and Fecundity
Spawning in the oyster involves the discharge of eggs and sperm from the gonads into the sea. The sperm pass down the suprabrancial chamber and enter the sea by way of the exhalent chamber. The eggs, however, pass immediately from the suprabranchial chamber through the spaces of the gills and the gill ostia into the inhalent chamber. The eggs lie on the gill plates, just posterior to the labial palps; the inhalent chamber thus acts as a brood chamber. The length of time during which the eggs are retained is dependent on the larvae reaching a certain stage of development and on external factors such as temperature.
Korringa (1941) refers to the liberation of larvae from the inhalent chamber to the sea as "swarming" and considers that this is in effect only the delayed completion of the spawning act. The release of eggs from the gonads in incubatory oysters has never been observed, observations being confined to the liberation of developed larvae.
The spawning season of an oyster varies in length according to local conditions and can only be observed in incubatory oysters by a regular examination of adults throughout the summer season. The first oysters examined from Wellington Harbour were collected on 18/12/60, the water temperature being 11.0°C. The gonads of all the oysters were undeveloped and appeared to be in an "overwintering" or resting period. However, from this date, the water temperature rose steadily to 18.5°C. by 28/12/60 when approximately 36 oysters were collected from Evans Bay; two of these oysters liberated larvae on the following day. Further examination of the adults showed that spawning continued until early March. The last collection of oysters that contained larvae was on 1/3/61, at a water temperature of 16.0°C. Examination of over 100 oysters during the following weeks failed to produce any incubating oysters or even oysters with well developed gonads.
Examination of adults was recommenced on 22/8/61 when the water temperature was as low as 10.0°C. This collection of oysters followed a succession of severe southerly storms that lasted for several weeks. The average water temperature during this cold weather was 9.4°C. as already mentioned above. Of the 31 oysters examined from this collection, one only was carrying larvae. The appearance of the gonads of the remaining oysters suggested various stages of development; some ripe, others less so. This is notable for two reasons: (1) the gonads were actually capable of maturation at temperatures as low as 10.0°C. and (2) oysters were spawning for at least eight months of the year, that is, from August to March.
It would seem possible then, since the water temperature rarely goes very far below 10.0°C. in Wellington Harbour (lowest record was 8.8°C. on 11/8/61) that spawning takes place throughout the year. This does not mean that spawning takes place with the same frequency and regularity throughout the year, but rather will there be periods of more active spawning during the summer months when there are higher water temperatures. Before it is possible to state with certainty that O. lutaria spawns throughout the year, it will be necessary to examine oysters collected from April to July. It is equally possible that any of these months and maybe all of them will be found to be periods of over-wintering, but maturation of the gonads and spawning would be possible during warmer months.
A spawning season of eight months is considerably longer than recorded for other incubatory oysters, except where oysters occur 'in warmer water (i.e., the Mediterranean) than in New Zealand and spawning takes place over the whole year.
page 14
.gif)
Text-fig. 2.—Fig. A—Unfertilised egg with sperm at the receptive spot. Fig. B—Late morula. Fig. C—Gastrula. Fig. D—Early trochosphere. Fig. E—Slightly older trochosphere. Fig. F—Trochosphere with mouth and velum developed.
Abbreviations: bias, blastopore; m., mouth; meg., megamere; mic., micromere; or.c., oral cilia; pol.b. polar body; prot., prototroch; sh.gl., shell gland; sp., sperm; sp.t., sperm tail; unf.e., unfertilised egg; vel., velum.
.gif)
Ostrea angasi Sowerby. Fig. 1, external surface of lower valve; 2, external surface of upper valve: 3, internal surface of lower valve; 4, internal surface of upper valve. Approx. ½ natural size.
.gif)
Ostrea lutaria Hutton. Fig. 1. external surface of lower valve; 2, external surface of upper valve; 3, internal surface of lower valve; 4, internal surface of upper valve. Approx. natural size.
.gif)
Ostrea lutaria Hutton. Fig. 1. external of lower valve: 2, internal of upper valve. Ostrea heffordi Finlay Fig. 3, internal surface of lower valve: 4, internal surface of upper valve; 5, internal surface of upper valve; 6. internal surface of lower valve: 7, external surface of lower valve; 8, external surface of upper valve. Approx. natual size.
Korringa (1941, p. 93) reviews the little that has been written on the number of larvae that may be spawned by incubatory and non-incubatory species of oysters. There have been no estimates of the number of eggs or larvae produced by O. lutaria until the present account but marketable oysters (approximately four to five years old), produce between one million and two million larvae while oysters of two or three years produce about 500,000. At the end of the breeding season, oysters of three and four years may produce as few as 2,000 larvae. It is possible that these oysters were spawning at the time of collection and that consequently only the remaining larvae were spawned in the laboratory. In general however, there is a tendency for oysters to produce fewer larvae at the end of the breeding season.
Fertilisation and Cleavage
Fertilisation and cleavage in O. lutaria take place in the inhalent chamber of the parent oyster. The sperm balls rotate rapidly and move through the sea water by means of the lashing sperm tails. After a few minutes in contact with the sea water, the spermatozoa begin to break free from the sperm ball. The heads of the spermatozoa become detached from the central mass of the ball and move outwards and away; the sperm tail is the last part of the spermatozoon to break free. The complete disintegration of a single sperm ball containing approximately 2,000 spermatozoa takes about five minutes. These observations were made by teasing a small portion of a male gonad into a watch glass containing fresh sea water and observing under a microscope.
As fertilisation has never been observed in incubatory oysters, it is assumed that the spermatozoa enter the mantle cavity of a female oyster with the inhalent water current and subsequently fertilise any eggs lying on the gill plates. Three separate attempts at artificial fertilisation of the eggs of O. lutaria were unsuccessful and likewise attempts by other workers to fertilise the eggs of incubatory species of oysters have also been unsuccessful.
However, on one of these three attempts, sperm were noticed clustered around a particular area of each egg. This area corresponds to the flat surface of the egg that had been lying next to the wall of the follicle. This area is thought to correspond to the receptive spot (Text-fig. 2, A).
Before syngamy takes place and possibly just after the sperm nucleus has penetrated the egg, the formation of polar bodies takes place. In some eggs, great clusters of incompletely divided polar bodies were observed while other eggs had only one, two or three polar bodies present. In some instances the polar bodies were noticed to persist through the stages of cleavage and were present on the prototroch of the trochosphere (Text-fig. 2, D).
The two, four, eight celled stages of cleavage were not observed in O. lutaria; the earliest stage observed was the blastula. The result of the early stages of cleavage is that the oosperm becomes differentiated into a megamere (deutomere) surrounded by a large number of micromeres (blastomeres). Finally the micromeres increase to such an extent that they arch over and partly surround the megamere so that the gastrula is formed by epiboly (Text-fig. 2, B and C).
Development and Liberation
.gif)
Text-fig. 3.—The relationship of height to length in free-swimming larvae of Crassostrea virginica, Ostrea edulis and O. lutaria. Figures for C. virginica and O. edulis taken from Yonge (1960, pl. IV and V)
on the broader, anterior end of the embryo. These cilia are between 80μ and 100μ tall and encircle the polar bodies if the latter are still present. This anterior region forms the prototroch which later develops into the velum (Text-fig. 2, E and F). The invagination forming the mouth represents the ventral region. The next noticeable event observed was the formation of the shelled larva and thus the trochosphere develops into the veliger larva.The Veliger
Larval shell: The valves of the shell appear on either side of the body on the dorsal surface. As the valves increase in size, they grow together and meet along the hinge line. The fully formed larval shell or prodissoconch is equivalvular, each valve being discoidal with a straight hinge line and moderately convex externally. Distinct concentric growth lines are present, some of which are more prominent than others. The edges of the valves are entire and distinct and further apart ventrally. As the larva approaches the settling stage, the straight hinge becomes slightly and evenly curled and the lower left valve becomes deeper so that both valves are no longer equivalvular.
In the early stages of shell growth, the initial tendency is for the shell to elongate rather than to deepen. When the shell measures approximately 0.42mm × 0.28mm it deepens more rapidly so that when the larva is ready to settle, the shell measures approximately 0.47mm × 0.40mm. These measurements are in direct contrast to those given for C. virginica and O. edulis where the lengthening and deepening of the shell keep pace. The height-length relationship of the larval shell of O. lutaria, O. virginica and O. edulis is shown in Text-fig. 3.
The velum: The velum is very well developed, conspicuous and protrudes anteriorly from between the valves of the shell. With the development of the velum, velar retractor muscles become differentiated (Text-fig. 4, A). These muscles originate in the velum and are inserted onto the shell. The general outer surface of the velum is ciliated as was the prototroch. The marginal cilia are 100μ tall and are more powerful than the cilia covering the outer surface of the velum (Text-fig. 4, E and F).
The alimentary canal: Associated with the velum is the mouth which is ventral in position. The cilia of the velum become continuous with the cilia surrounding the mouth but at the same time become noticeably shorter until they are only 20μ in length (Text-fig. 4, C). The mouth opens into the oesophagus which passes back dorsally between the base of the velum and the foot. The stomach occupies the central mass of the body bounded anteriorly by the velum and posteriorly by the gill buds and the rectum. The right and left sides of the stomach are surrounded by the lobes of the digestive diverticula. In O. lutaria, the coiling of the intestine is first visible where it passes downwards and backwards as the rectum (Text-fig. 4, E). The rectum terminates in the anus which opens into the mantle cavity behind the posterior mantle suture.
The gills: The gills are first visible as a series of knobs extending from the mantle margin beneath the foot to a position near the posteroventral margin of the stomach. As the larva develops, the gill knobs become more distinct and segmented, the outermost knobs being smaller than the innermost knobs. In a fully developed larva there are approximately nine or ten gill filaments in the series. The filaments separate at the time of settling, and active cilia were noted on the surfaces of the larger filaments.
.gif)
Text-fig. 4.—Fig. A—Trochosphere with velar retractor muscle visible. Fig. B—Trochosphere with shell developing. Fig. C—Trochosphere at later stage. Fig. D—Trochosphere with position of alimentary canal visible. Fig. E—Straight-hinged veliger. Fig. F—Straight-hinged veliger in swimming position.
Abbreviations: A., anus; a.a., anterior adductor muscle; d.d., digestive diverticulum; fo., foot; g., gill; l.v., left valve; m.e., mantle edge; mo., mouth; o.c., oral cilia; rec., rectum; r.m.e., right mantle edge; r.v., right valve; v.e., hinge region of valve; vel., velum; vel.r., velar retractor muscle; v.h., hinge.
muscles was particularly noticed when a 20% solution of cocain was added to the sea water containing the larvae. The solution caused the adductor muscles to close the shell so tightly that the velum was cut off.The mantle: In the veliger the first region of the mantle to be observed was the margin (the mantle lobes are not easy to observe until the larva has become fully developed). In the early veliger, the mantle margin is yellow and the mantle lobes can be seen in the living larva moving backwards and forwards beneath the valves of the shell.
The foot: In the early veliger, the foot is represented by a small projection on the ventral surface, beneath the mouth (Text-fig. 4, A). As the veliger develops, the foot lengthens until finally it is as long as the valves. When not in use, the foot is withdrawn and contained entirely within the mantle cavity beneath the mouth. As the foot begins to protrude, a groove becomes apparent on the under surface. This groove is ciliated as is the entire outer surface of the foot. The two sides of the groove are capable of coming together and closing over the groove, thus forming a canal. As the foot protrudes even further, the groove becomes shallower. The fully extended foot is long, slender and strap-shaped. It is capable of moving anteriorly over the velum, posteriorly behind the gills and over the outer surfaces of the valves. Posteriorly behind the foot, is a heel. This heel is only observed when the foot is fully extended. Stafford (1913, p. 46) refers to this heel as the byssus-papilla and although the byssus gland was found in O. lutaria, the duct leading to the papilla was not (Text-fig. 5, B).
Otocysts: Stafford (1913, p. 50) observed about a dozen small otoconia in each cyst. The right and left cysts are situated at the base of the foot near the surface of the first gill filament. On a few occasions when the veliger foot was fully extended, structures that could have been the otocysts were observed, but confirmation of the presence or absence of otocysts awaits further investigation.
Nerve ganglia: The only nerve ganglia observed in O. lutaria were the pedal and cephalic ganglia. The former are situated in the proximal region of the foot and are only visible when the foot is fully extended. The cephalic ganglia are associated with the apical sensory organ which is situated in the central region of the velum (Text-fig. 5, A and B).
Pigment spots: The pigment spots are sometimes called eyespots but are better known as the former since their photosensitivity has not been confirmed. The pigment spots are paired, and lie on the lateral walls of the right and left mantle lobes just anterior to the proximal region of the gill buds. They are almost black in colour, irregular in outline and measure approximately 12μ in diameter (Text-fig. 5, A, B and C).
Duration of the incubation period.
Because artificial fertilisation was unsuccessful with O. lutaria, the duration of the incubation period was studied by regular examination and measurement of larvae liberated by oysters held in the laboratory.
.gif)
Text-fig. 5.—Fig. A—Veliger with foot fully extended. Fig. B—Veliger with velum and foot in swimming position. Fig. C—Veliger with velum and foot withdrawn. Fig. D—Spat settled for approximately 72 hours.
Abbreviations: A., anus; a.a., anterior adductor muscle; an.s., anterior suture; a.s., apical sensory organ; b.g., byssus gland; diss.s., dissoconch shell; e.s., eyespot; fo., foot; g.p., gill plate; h., hinge; he., heart; int., intestine; lab.p., labial palp; m.e.m., mantle edge; mo., mouth; p.a., posterior adductor muscle; p.g., pedal ganglion; prod., prodissoconch; p.s., posterior suture; rec, rectum; r.v., radial vessel; vel., velum; vis.m., visceral mass.
total time taken for an early veliger to develop into a veliger with pigment spots is unknown, but observations made on oysters held in the laboratory in sea water between 18.0°C. and 20.0°C., suggest that it is five to eight days. A fully developed veliger measuring 0.47mm × 0.35mm is liberated after about a further six days. Thus the total time of development within the mantle chamber appears to be about 21 days.The stage of development reached by the larva under natural conditions before it is liberated is discussed in the section on the free swimming larva, but briefly if O. lutaria follows other incubatory species of oysters, the larvae will be liberated when they have attained a size of about 0.32mm × 0.26mm. Larvae of O. lutaria with these measurements were liberated in the laboratory, but more fully developed larvae complete with foot and pigment spots were also observed being liberated from the parent indicating that if conditions are favourable the parent oyster will incubate the larvae through all the stages of development until the latter are ready to settle.
Liberation: In O. lutaria the developing larvae are liberated by violent contraction of the posterior adductor muscle of the parent. This fast closure of the shell forces the larvae out through a gap in the inner pallial fold as in other incubatory species.
Liberation of the larvae in some cases was completed within an hour; in other instances the liberation took as many as four days to complete. The only regular feature of liberation that was observed was that the majority of larvae were liberated in the first two or three rapid closures of the shell and that subsequent liberations yielded very much lower number of larvae. Towards the end of liberation, larvae appeared to "spill" out of the shell and over the left valve to the dish beneath. The last few larvae were always held in strands of mucus that was probably secreted by the gills during the process of feeding.
Free Swimming Larvae
No free swimming oyster larvae have so far been found in plankton collections studied from Wellington Harbour. Under laboratory conditions however, the larvae were observed to be released in various stages of development which may or may not be the case in the field. Oysters were collected from Evans Bay and were quickly removed to the laboratory where they were placed in 7 ½ in culture dishes (3–4 oysters per dish) and covered with sea water. The temperature of this sea water varied between 18.5°C. and 20.5°C. and was usually 2–4 degrees warmer than the temperature prevailing in Evans Bay. In every instance this rise in temperature of the sea water was sufficient to induce the oysters to liberate their larvae. All stages of development were liberated from parent oysters held in the laboratory including embryos and trochospheres, early straight-hinged veliger larvae and fully developed larvae possessing a pair of pigment spots and well developed foot.
The measurements of free swimming larvae of O. edulis vary between 0.16mm and 0.20mm (Erdmann, 1934, p. 6, and Korringa, 1941, p. 101) whereas larvae of O, lutaria at a comparable stage of development vary between 0.31mm and 0.33mm. Swimming in these latter larvae is by means of the velum and is restricted to horizontal movement on the bottom of the dish and short vertical movements. In the horizontal movement, the velum is uppermost and the larvae turn anti-clockwise. The movements as seen by the naked eye is very slow, almost imperceptible. The measurements of seven fully developed larvae of O. lutaria are given to show the size variations in height of larvae with similar length:
page 22
0.42mm × 0.28mm; 0.42mm × 0.35mm; 0.42mm × 0.36mm; 0.43mm × 0.31mm; 0.44mm × 0.34mm; 0.45mm × 0.37mm; 0.47mm × 0.39mm.
Since the size of the straight-hinged veliger of O. lutaria is greater than that of O. edulis, it could be anticipated that the fully developed larva of O. lutaria would be larger than that of O. edulis. Measurements of the larvae of O. edulis which have just developed the pigment spot vary between 0.26mm and 0.30mm (Cole, 1939, and Korringa, 1941) and larvae ready to settle measure between 0.27mm and 0.31mm (Erdmann 1934, Cole 1939, and Korringa 1941).
Fully developed larvae of O. lutaria (Text-fig. 5, A, B and C) liberated in the laboratory have been observed to swim rapidly in both horizontal and vertical directions, to rest on the surface of the water and to creep on the bottom of the dish. Larvae that are about to settle do not necessarily confine themselves to crawling and in fact they alternate swimming with crawling phases, always being in a crawling phase immediately prior to attachment. This last type of movement is the exploratory phase described by Cole (1938, p. 478) in which the foot serves as a locomotory organ on the substratum immediately prior to attachment and apparently acts as a tactile organ.
Furthermore, Cole (1938, p. 471), notes that in O. edulis the eyespot (pigment spot) develops only a few days before attachment. Thus it would appear, that if in O. lutaria the foot and pigment spots are also developed only a few days before attachment then the larvae of O. lutaria are ready to settle and become attached immediately after being liberated. Fully developed veliger larvae which do not settle have also been liberated. These larvae are easily recognised by their pigment spots but do not exhibit an exploratory phase which is typical of attaching larvae. These larvae are the same size as larvae that settle and become attached.
Duration of free swimming stage.
If it is assumed that under normal conditions the larvae of O. lutaria are liberated complete with well-developed foot and pigment spots, then it is apparent that the duration of the free swimming stage will be reduced accordingly. During the course of this study only three oysters of the 565 held from time to time in the laboratory liberated fully developed larvae which subsequently became attached. The free swimming larvae of these three oysters became attached as follows: one, five days after a temperature rise from 15.5°C. to 18.5°C.; the second, three days after a temperature rise from 18.0°C. to 20.0°C.; and the third, two days after a temperature change from 16.0°C. to 19.5°C. The temperature change was caused by removing oysters from Evans Bay to warmer sea water in the laboratory. In all cases mentioned, the rise in temperature was sufficient to induce oysters to liberate their larvae prematurely. In other instances, the successful attachment of the larvae depends on their state of development at the time of liberation.
During experiments some larvae were held at 21.5°C. Of these, a few became attached within two hours. The majority died and about 20 per cent were still in an exploratory phase 18 days later. This indicates that the larvae are capable of postponing attachment for some considerable length of time, but the experimental temperature of 21.5°C. is higher than that recorded in Evans Bay so that it is quite possible that the larvae behaved abnormally.
In conclusion, it seems reasonable to assume that any pelagic free swimming stage of the larvae of O. lutaria if it does exist, lasts a few days only, but that attachment may be postponed if suitable conditions are not present.
Attachment and Metamorphosis
The exploratory, crawling phase preceding attachment has already been mentioned. Several oysters liberated veliger larvae that subsequently became attached but the actual process of attachment was not observed. Larvae of O. lutaria were observed crawling above the surface water level and these larvae subsequently became firmly attached to the glass wall of the bowl. Also, larvae which became detached floated to the surface of the water and remained alive for as long as 24 hours. Such larvae, it was observed, never attempted resettlement, possibly because the byssus gland cement was secreted during the initial attempt at attachment. During the hours following attachment while the foot was still present, spat were observed defending their attachment area by means of the foot. If several spat settled in very close proximity they kept the foot fully extended. Consequently spat are never attached less than their own distance apart (which is also equal to the length of the foot).
The influence of light on the settling process.
The pigment spots of fully-developed larvae are eye-like structures as shown by Cole (1938) and Erdmann (1934). The black colour of the spots is due to heavy pigmentation in the epithelium. The spherical cup formed by the epithelium is filled with a gelatinous matrix, the aperture being closed by a lens-like body. The function of the eyespots has been the subject of considerable controversy. To date, observations on O. edulis and O. lurida indicate that the pigment spots are not light sensitive at all. On several occasions, fully-developed larvae of O. lutaria were placed in a glass tube and subjected to a bright light; half of the tube being shaded. The larvae did not exhibit a negative phototaxi. In contrast, the larvae of C. virginica have been demonstrated by some authors to be photosensitive. The larvae when stimulated continue to move away from the light until they reach a shaded site, and this is thought to account for the preference shown by the settling larvae for attachment in shade.
Experiments investigating the influence of light on settling and attachment have been carried out by many authors on various oysters and as yet there is no general agreement on the subject. Korringa (1941, p. 192) concludes that "although light appears to be no orientating factor in the settling behaviour of O. edulis under field conditions, it is not impossible that light influences the settling process in cases where the intensity of the light exceeds a certain degree."
Metamorphosis.
Knowledge of the metamorphosis of the larva of the American oyster, V. virginica, is considerable, owing mainly to the work of Stafford (1913). Cole (1937, 1938) describes the less well-known metamorphosis of incubatory oysters. The following is a brief account of the anatomical reorganisation of the spat of O. lutaria when it takes on the fixed life of the adult oyster.
The shell: The dissoconch is the post larval shell; it is formed during the 24 hours following attachment. It appears around the distal margin of the prodissoconch, but is very thin and indefinite at first. The dissoconch shell growth can be divided into two phases, (1) the silphologic (spat) phase, and (2) the adult shell phase. The silphologic shell phase has further been divided into five stages by Jackson (1888) most of which have been identified in O. lutaria. The lower left valve as it grows, becomes closely attached to the attachment area by becoming flattened. Lateral wings appear on the valves, usually on the anterior side. Growth continues until the lower valve reaches the margin of its attachment when it
page 24
proceeds to grow upwards so that the general form of the adult shell with concave lower valve and flattened upper valve is soon apparent. Another feature of the dissoconch silphologic stages is the subnacreous layer which is formed unevenly at first until it covers the entire inside lining of both valves.
The foot: The foot persists for the first 24 hours following attachment but then rapidly degenerates and is carried forward by the rotation of the mouth until it reaches a mid-ventral position. The statocysts are thought to persist in the adult oyster of other species but no trace was ever found of them in O. lutaria.
The velum: The velum persists as an identifiable organ for about 12 hours following attachment. It is contained within the valves and the movement of the cilia is quite clear. The velum finally collapses and shrinks in size, becoming converted eventually into the outer palps of the adult oyster (Text-fig. 5, D). Cole (1938) describes how the velum of O. edulis shrinks after attachment and is carried forward and upward by the rotation of the body so that about 24 hours after attachment, the velar remnants "consist of a little of the typical epithelium of the thickened edge and some of the muscle fibres of the interior ... subsequently the much thickened upper lip spreads out laterally and gives rise to short rounded lobes which project, one on each side, at the upper corners of the mouth ". These are the rudiments of the outer palps of the adult oyster. The outer palps are initially large but gradually shorten as the inner palps are formed. The beginning of the formation of the inner palps was never observed in O. lutaria and as far as the writer is aware, has not been observed in other species of oysters. However, they are present in spat measuring 5.5mm in length.
The mantle: The mantle edge is visible in a fully-developed larva and in the attached spat. Associated with the mantle lobes are prominent radiating vessels (Text-fig. 5, D) and in spat measuring 5mm they appear to act as pulsating vessels or accessory hearts. Tentacles were not observed in recently settled spat. In older spat measuring 5mm the mantle lobes are well defined being united anteriorly beneath the hinge and posteriorly with the gills. The mantle margin has three folds as in the adult but the similarity ends there. The outer fold has no tentacles and is unpigmented, and the middle fold is more well-developed, being tentacular. There are two types of tentacles arranged in no particular order. The majority are small and regular, as tall as they are broad and the remainder are twice as tall as they are broad and are scattered amongst the smaller more regular tentacles. Both types of tentacles are brown apically. The inner fold which is the pallial curtain is present as a simple flap, smaller than the outer fold and similarly without tentacles.
The gills: At the time of attachment, the larva has between eight and ten gill filaments (Text-fig. 5, C). This is in contrast to the six filaments recorded for O. edulis by Erdmann (1934) and seven filaments recorded by Cole (1937, p. 413). Cole further notes that eyed-spat have eight gill filaments; spat measuring 0.35mm have ten gill filaments and that the 11th, 12th and 13th gill filaments appear on the left side before the spat is 90 hours old. Such spat measure approximately 0.6mm in diameter. At this stage there are also seven gill filaments on the right side. Spat of O. lutaria measuring 0.6mm (Text-fig. 5, D) in diameter have been settled for only 72 hours in contrast to the 90 hours of O. edulis described by Cole (1937). Such spat have 11–12 gill filaments on the left and about five gill filaments on the right (not figured). Thus the larvae of O. lutaria possess more gill filaments is in keeping with the much larger size of the settling larvae of O. lutaria. Yonge (1926) figures a spat of O. edulis which had probably been settled for five or six days. 1.2mm in length, with 20 gill
page 25
filaments on the left and 13 gill filaments. Spat of O. lutaria of a similar stage were never found. However, spat measuring 5mm in diameter had 90 gill filaments on the left. The filaments were simple and similar in structure, there being no transitional or principal filaments or even any indication of plication. The filaments are spaced their own distance apart and are ciliated laterally and apically and are connected apically to adjacent filaments by a thin membrane. In these spat the filaments had split longitudinally so as to form the respective inner and outer lamellae. This splitting did not include the head of the filament. Two rows of interfilamentar junctions are present on the left inner demibranch but none could be seen on the corresponding right demibranch. The interfilamentar junctions transform the linear filamentous spaces into a series of fenestrae.
Stafford (1913, p. 68) stated that weight and pressure acting unequally upon the two sides of the gill soon effects a marked difference in the gill growth rate. The left gill in O. lutaria grows much faster than the right gill. The right and left gills of a recently settled spat correspond to the inner demibranchs of the adult. The measurements of a spat with the beginnings of the right outer demibranch are unknown but the left outer demibranch is seen to appear from the mantle and the gill axes of the left inner demibranch in a spat measuring 5mm × 5.5mm. The appearance of the left outer demibranch in other incubatory species of oysters is unknown as far as the author is aware. Stafford (1913, p. 68) also mentions that in C. virginica the right outer demibranch is formed when the spat is 2.5mm tall and the left outer demibranch is formed when the spat is 3mm tall. A spat 3mm tall has 50 gill filaments in the left inner gill. This is considerably less than the 90 filaments that O. lutaria has when the outer left demibranch appears.
Stafford furthermore notes that a spat of C. virginica 1.5mm in height has 23 gill filaments which is very similar to the 1.2mm spat of O. edulis which has 20 gill filaments. Therefore it is quite reasonable to estimate that a spat of O. edulis measuring about 3mm in height will show incipient left outer demibranch. In contrast, the spat of O. lutaria takes longer to develop all the demibranchs even though it settles at a much larger size and possesses more gill filaments at the time of attachment.
Adductor muscles: The fully developed larva possesses an anterior and posterior adductor muscle more or less symmetrically placed with regard to the hinge and of similar size. However, after attachment the anterior adductor muscle is moved upwards and outwards towards the edge of the prodissoconch where it finally disappears (Text-fig. 5, D). The posterior adductor muscle moves downwards and towards the centre of the viscera. In a spat measuring about 0.5mm in height, the posterior adductor muscle is situated between the edge of the prodissoconch and the dissoconch, until it finally moves entirely onto the dissoconch (Text-fig. 5, D). As the muscle moves centrally, it enlarges but does not as yet show any differentiation into quick and catch areas. The muscle is also becoming characteristically lunate as it moves ventrally on to the dissoconch.
Pigment spots: The pigment spots are lost during the 20 hours immediately following attachment.
Spat axes: In the larva, the mouth and the anus develop ventrally but when the larva becomes attached the anus moves dorsally. The mouth also rotates dorsally to rest against the ventral side of the anterior adductor muscle, and the posterior adductor muscle develops on the ventral side of the intestine. The palps and gills have also rotated dorsally from a former ventral position. The hinge of the prodissoconch is dorsal but after metamorphosis a reorientation of the body takes place so that the hinge becomes anterodorsal.