Previous Page Table of Contents Next Page


Hatchery production of bivalve is a well-established and relatively routine operation. Figure 16 presents a schematic representation of two traditional bivalve hatchery methodologies in common use today: the Milford and the Wells-Glancy methods. The primary difference between these methods is in the techniques employed for algal production. The Wells-Glancy method (Wells 1927) relies on blooms of natural phytoplankton populations initially segregated by filtration or centrifugation. The Milford method (Loosanoff and Davis 1950; Loosanoff 1954) introduced broodstock conditioning and the controlled culture of specific phytoplankters as rations in bivalve larviculture. Large commercial hatcheries generally use a combination of these methods: the Milford method for larval rearing and the Wells-Glancy method for postset populations. Several hatcheries bypass the use of raw seawater completely by employing saltwater wells. This alleviates the need for filtration and sterilization and provides a uniformly stable culture medium for both larvae and algae. It does require, however, a complete dependence on cultured algae for hatchery operations.

The success of larviculture and subsequently the success of a hatchery is a great extent dependent on the quantity and quality of algal production. Despite their importance, the nutritional requirements of adult and larval clams are only partially understood. Ukeles (1971) provided a review of nutritional aspects of shellfish culture and reiterated the rather pragmatic nature of algal culture support systems for lamellibranch larviculture. Early culture research indicated that certain phytoplankters were better than others as food for larvae. Cole (1937) concluded that only minute naked flagellates were utilized by swimming bivalve larva. Davis and Guillard (1958) and Walne (1970) indicated that bivalve larvae do well on such naked flagellates as Isochrysis and Monochrysis but do best with mixtures of these two and Dunaliella and/or Platymonas. Today, naked Chrysophytes are the most commonly used algae in bivalve culture. Isochrysis galbana and Pavlova lutheri are often mixed in and other recently isolated species are gaining popularity (e.g. Tahitian Isochrysis). Kilada (1985) showed that Isochrysis galbana proved to be a better food for the Manila clam (Tapes semidecussatus) than Skeletonema costatum. He also showed that Chlamydomonas coccoides appears to be of very poor food value for the same bivalve species

Figure 16

Figure 16. Schematic diagram of the Milford and Wells-Glancy methods of bivalve larval culture. (Source: Manzi, 1985).

Hatcheries using blooms of natural phytoplankton to produce larval culture media try to limit phytoplankton size by filtering incoming water. Water thus filtered through a 5- or 10-um bag or subjected to centrifugation and then forced to bloom will produce relatively dense concentrations of mixed nannoplankters (~3–5 um in diamter). These natural “batch” cultures are then used to provide feed to bivalve larvae. Although cost effective and relatively easy to perform, the Wells-Glancy technique is not used by many large commercial facilities. Selective monoculture of algae (Milford method) is employed by most large-scale commercial hatcheries. Axenic algal stock cultures are maintained in test tubes or small flasks and are used as inocula for larger cultures. Each culture stage serves as inocula for progressively larger cultures until final culture volumes are attained. These cultures are intermittently cropped to provide food for the larvae. The final production cultures can often be kept in a log growth phase for weeks by appropriate cropping and thus provide many times their original volume in food for bivalve larvae. These high-density algal cultures are dependent on nutrient supplements (i.e., sources of N.P and other elements) added to sterile seawater cultures. These can be simple commercial fertilizers (Loosanoff and Davis 1963) or complex formulations for specific algal groups (Guillard 1958; Ukeles 1971). In comparison with the Wells-Glancy method, the Milford method requires much more energy, labour, and material per unit volume of algae produced, but its common use in commercial facilities testifies to its value and to some degree its cost effectiveness (DePauw, 1981; Epifanio, 1979).

5.1 Algae species

A number of species of phytoplankton have been found suitable for rearing bivalve larvae. A few species are universally appropriate, while others can be used only with certain species or at specific periods of larval development. Table 1 lists some of the more commonly cultured phytoplankton species and larvae for which they have been found to be suitable.

The above table is not complete, however it provides an idea of the range of bivalve species able to utilize some of the more easily cultured algae. Small naked flagellates such as I. galbana and M. lutheri give uniformly good results through all larval stages. Clam larvae seem to be able to utilize a broader range of species than oysters. Phytoplankton species with cell walls are poorer food and, in fact, cannot be used by some species. Some Chlorella spp. can be toxic to larvae at high concentrations.

5.2 Principles of culture growth

The growth of monospecific cultures in sterile media follows a predictable pattern. Cells increase in number by asexual fission until either the nutrients in the media are exhausted or toxicity of accumulated metabolites causes the culture to collapse. The growth curve of a typical culture can be represented by the curve shown in Figure 17 A. The curve can be divide into several distinct phases:

Table 1. Bivalve species and their suitable food type.
Isochrysis galbanaCrassostrea gigas
Crassostrea virginica
Tapes semidecussatus
Pseudoisochrysis paradoxa
Monochrysis lutheri
T. semidecussatus
Mercenaria mercenaria
M. campechensis
Ostrea edulis
Area transversa
Panope generosa
Saxodomus nuttili
Argopecten irradians
Ostrea lurida
Tetraselmis chuii,
Tetraselmis suecica
Best for feeding spat of any bivalve species. Also used for older larvae.
Thalasiosira (Cyclotella)
pseudonana, 3H-strain
Excellent for larvae of oysters larger than 180 microns. Also good for spat and broodstock.
Chlorella sp.M. mercenaria
O. edulis
A. transversa
Mytilus edulis
A. irradians
Dunaliella euchloraC. virginica
C. gigas
M. mercenaria

Lag phase.     The length of the lag phase depends upon the particular species' inherent growth rate and the quantity of cells inoculated. A large inoculum will shorten the lag time, however in practice a culture vessel is usually inoculated with 5–10% of the media volume. The relationship between number of cells in the inoculum and lag time can be sketched approximately as shown in Figure 17 B. A common error is to inoculate with a culture that has not entered it slog phase, with the result that the lag phase is extended and cell production will be low.

Log phase.     During this period of culture growth, cell density increases exponentially. A culture should be harvested during its log phase, as it has been shown that the algal cell protein content is maximum at a point about half way through the log phase. In a hatchery the culture is usually harvest somewhat higher on the log curve, just before the culture enters its peak phase.

Peak phase.     The culture stops growing for the reasons mentioned above. Another cause in large culture volume can be “shading” or insufficient illumination, particularly if the circulation in the tank is poor. Carbon dioxide can also become a limiting nutrient at very high densities. Since the condition of algal cells at this point is not good, the culture should not be used for anything. Bacteria and fungi also begin to appear at this stage as well as clumping. As the culture enters the final period of this phase, it begins to “foam”. In large volume tanks, the culture can often be rescued at this point by harvesting forty to fifty percent of its volume and replacing with fresh media. A healthy culture has a characteristic odour, which will change during the peak phase. A change in odour may also indicate imminent collapse (crashing).

Crash.     This phase is characterized by clumping and sedimentation. Protozoa begin to appear and the culture becomes useless.

5.3 Media preparation

A variety of media have been developed for culturing a variety of phytoplankton species. For large scale production, the media needs to be as simple and inexpensive as possible. In general, media formulae involve some form of nitrogen and phosphorus, usually as a nitrate and phosphate salt, a mixture of trace metals and a vitamin mix. The sodium salt of ethylenediaminetetraacetic acid (EDTA) is added to the trace metals to prevent precipitation, particularly of iron compounds. EDTA is a chelator, and it has also been observed to have a bacteriostatic effect, perhaps due to the chelation of trace metals necessary for bacterial growth. The addition of some form of silicate is necessary when culturing diatoms.

For large volume cultures, particularly “green-water” culture, the trace metals and vitamins are omitted and nitrogen and phosphorus are added in the form of commercial fertilizers. This may be urea, NPK and TSP. Often the addition of EDTA will have a beneficial effect.

Figure 17.

Figure 17. (A) Typical growth curve of a phytoplankton culture, and (B) the relationship between number of cells in the inoculum and Lag time.

Media for stock and starter cultures requires sterilization in an autoclave. However, autoclaving raises the pH of most media. Therefore, before autoclaving, the ph of the media should be adjusted to 6.5–6.9 by the addition of a 10% solution of HCl. The autoclaved media should have a pH in the range of 7.8–8.3.

Intermediate volumes of media, for example, 20 litre carboys, can be pasteurized by heat or OCl- (hypochlorite). Filtration by ceramic or other media can also be used. Hypochlorite pasteurization is only effective if the volume of inoculum is at least 10% of the media volume.

If heat pasteurization is used, the temperature of the media is brought up to 80 °C. This can be accomplished by a heat exchanger or by an immersion heater placed directly in the carboy. If an immersion heater is used it should be of quartz on Vycor glass. If many carboys have to be heat pasteurized, a heat exchanger is more practical.

As the column of the inoculum increases, less efficient methods of water treatment can be employed. For tanks of several hundred to several thousand liters, diatomaceous earth filters are very effective, although the initial cost may be high. However, these filters are readily fabricated. Filter bags with an effective pore size of 5 microns are also used. Both of these should be proceeded by sand filtration. Hypochlorite pasteurization is also applicable to large volumes. Whenever OCl- is used, a test kit should be available to determine a) if the free chlorine concentration is adequate (5 ppm) and b) all chlorine has been eliminated, either by neutralization by sodium thiosulfate or by overnight aeration. These kits are readily available from swimming pool supply companies.

5.4 Culture inoculum and starters

The proper maintenance of stock cultures is particularly important, since they provide the basic inoculum. Stock cultures must be kept in a clean environment and inoculation done with strict sterile technique. The inoculation area should be in a separate room, preferably separated from other rooms.

Starters are maintained in 20 ml screw cap test tubes. Cool daylight fluorescent tubes provide adequate illumination. Most of the species suitable for larvae foods can be maintained at 25 °C.

Two flask levels can be used between the stock cultures and carboys. These flask cultures are also kept in the same room with the stock culture tubes. Inoculations are made at weekly intervals. One ml of the stock culture is used to inoculate a new test tube and nine ml are inoculated into 90 ml of new media in 125 ml Erlenmeyer flasks. Fifty ml from these flasks are inoculated into 250 ml of media in a 500 ml flask. Four of these flasks are used to inoculate one carboy.

The carboys, containing about 8 liters of media, are kept in their own room, air conditioned to 25 °C. In a large facility, there may be two levels of carboys. The first uses heat pasteurized media and is inoculated as described above. The second level uses OCl- pasteurized media and is inoculated with two liters from the first level, so that one carboy from the first level can be used to inoculate nine on the second. This procedure greatly reduces the cost of pasteurizing media and introduces a margin of safety since extra carboys of heat pasteurized media can be prepared. Carboys are best illuminated by overhead fluorescent lamps and must be vigorously aerated. If a dry air-type blower is used, air filtration should not be necessary. Larvae may be fed directly from carboys or they may be used to inoculate larger tanks. One or two hundred litre tanks can be used to inoculate 1,500 litre tanks filled with filtered seawater.

5.5 Monitoring culture growth

With experience, monitoring can be done by color and some commercial hatcheries use only this method. More precise feeding of larvae requires that the cultures being fed be counted. It is usually not necessary to count stock and flask cultures.

The simplest counting device is a haemocytometer. In a large hatchery, counting by this method would be tedious. A turbidity meter or nephelometer may be used to measure the optical density of the culture sample. The instrument is calibrated with the haemocytometer for each species. Readings from the turbidity meter are then read as cell numbers from the calibration curve.

5.6 Large volume cultures

Culturing single species in large outdoor tanks requires adequate illumination and good circulation. The latter is particularly important so that all cells are exposed to adequate illumination. Dead spots in the tank must also be eliminated, since cells will accumulate and clumping or bacterial growth can begin in these dead spots. Air-stones are inadequate and airlifts or perforated pipes must be used. The shape of the tank should also be designed to promote circulation and elimination of dead spots. The efficiency of airlifts is influenced by the ratio of emergence to submergence, air input volume and the size of bubbles.

Previous Page Top of Page Next Page