A knowledge of the physiology of fertilization of fish eggs seems to be most important not only for an understanding of the basic phenomena of fertilization in vertebrates but also for its bearing on fish culture. In contrast to echinoderm materials which have been so extensively explored, relatively little research has been done in this field. The absence of suitable material as well as appropriate methods has been the prime cause of the failure to promote physiology of fertilization in fish eggs.

For physiological studies on fertilization and artificial activation, it is desirable to keep ripe unfertilized eggs in a fertilizable condition for considerable periods of time. This had not been possible because fish eggs lose their capacity for fertilization within a short period of time after they are removed from the female and immersed in the water of its spawning environment. This property is in strong contrast to that of sea urchin eggs which retain their fertilization capacity for a long time in sea water. The short duration (usually a few minutes) of the fertilizable period of fish eggs rendred it impossible to subject unfertilized ova to treatment with any agent prior to artificial fertilization. Therefore, the "dry" method has been widely used for artificial fertilization in fish culture since the nineteenth century. Although this method is most excellent for practical use, it is not adequate for exact physiological analysis of the fertilization reaction.

The writer found in 1939 that the ripe unfertilized egg of the medaka, Oryzias latipes, a fresh-water fish belonging to Oryziatidae, retains its fertilizability for a long period after it is immersed in an isotonic Ringer's solution instead of fresh water, and that normal fertilization can take place in the solution. Since the process of fertilization is normal in the isotonic Ringer's solution, it is possible to note the step-by-step change in the egg before and after fertilization. Furthermore, it is possible to subject the unfertilized egg to various treatments prior to fertilization without limiting its fertilizability. The procedure for keeping *Partly extracted from T. Yamamoto, 1961: Physiology of Fertilization in Fish Eggs. Intern. Rev. Cytol. 12: 361-405, edited by G.H. Bourne and J.F. Danielli. Academic Press Inc., New York. unfertilized ova in fertilization condition may be called the "isotonic method of fertilization." With the ova of the medaka both methods can be successfully employed. By these methods it has become possible to investigate the physiology of fertilization of Oryzias egg in the same way as is possible with sea urchin eggs.

On the basis of a series of studies on fertilization and artificial activation of Oryzias eggs, the writer in 1944 advanced the theory of the "fertilization-wave." The most prominent feature of the cortical changes during fertilization in teleostean eggs is the breakdown of the cortical alveoli, which begins near the animal pole and ends at the vegetal pole. Physiological study of the phenomenon in Oryzias eggs led to the view, however, that the easily visible breakdown of the cortical alveoli is an indication of a more deeply seated physiological change which is essential for the completion of the physiological reaction. The fundamental concept which the writer adopted is that the ripe ovum is an irritable system with the capacity of excitation-conduction. It has become clear that some invisible change of the nature of an "impulse" must be provoked at the cortical layer either by sperm or activating agents, which propagates itself in succession from the point of stimulation and causes the breakdown of the cortical alveoli. This impulse is the writer's "fertilization-wave." It has also been experimentally proved that the fertilization-wave is conducted with decrement in the cortical protoplasmic layer but not through the yolk.

There followed the important corollaries, viz., (1) Although the sperm may stimulate the ovum, perhaps by its chemical substance, at the point of its attachment (animal pole) all the subsequent cortical reactions are entirely due to ovogenous substances, viz. reactions are realized by virtue of the conduction of the fertilization-wave. (2) Any theory that the sperm is the bearer of a substance directly causing the cortical changes at all points on the surface of the egg is inadequate. Although extensive studies had been made of the physiology of fertilization of sea urchin eggs, the occurrence of the fertilization-wave at the time of fertilization or artificial activation has not been demonstrated. On the basis of observations carried out on sea urchin eggs stuck on a glass surface, Runnstrom and Kriszat (1952) postulated that an impulse caused by the sperm contact propagates itself in the cortical layer. This fact has been demonstrated beyond doubt for sea urchin eggs by the researches of Sugiyama (1953a, b) and Allen (1954). The writer believes that the fertilization-wave or the fertilization-impulse will undoubtedly be demonstrated in the future for ova of all animals including the human.

The isotonic method of fertilization

It has long been known that ripe unfertilized ova of teleosts lose their fertilizability within a short period after they are immersed in their spawning media, i.e., fresh water in the case of fresh-water fishes and sea water in marine fishes. The unfertilized ova of the medaka (Oryzias latipes), a fresh-water fish, also become incapable of fertilization within a few minutes after they have been immersed in fresh water.

It was found by the writer (1939a, 1943a, b, 1944a) that the ripe ova of the medaka retain their fertilizability for several hours after they have been kept in an isotonic Ringer's solution (M/7.5 NaCl, 100 + M/7.5 KCl, 2.0 + M/11 CaCl2, 2.1; pH adjusted to 7.3 by NaHCO3)*. This Ringer's solution was formulated on the basis of an actual measurement of the osmotic pressure of the egg, which was found to be isotonic to M/7.5 NaCl solution (Yamamoto, 1941). The Ringer's solution is not only excellent for preservation of ripe ova for a long period, but it can be used as an insemination solution. Ova of Oryzias are capable of fertilization and development in the Ringer's solution.

In view of the fact that ova of the medaka provide unparalleled material in physiology of fertilization of fish and that my researches have been chiefly on this material, it is taken for granted that there is need for a few sentences on the method of obtaining ripe ova from the medaka.

The female medaka exhibits a maturation cycle of 24 hours and spawns a batch of eggs at dawn of every day during a certain period of the breeding season. In order to obtain ripe ova, females known to have spawned should be removed from males on the day before the experiment is to be performed. While ripe eggs could be stripped out from the fish, it is wise to sacrifice the females and remove the ovary gently. This is because ordinary stripping methods frequently injure ova and cause them to manifest mechanical activation. On the day of the experiment, the fish is found to contain ripe eggs which would have been spawned had they been left with males. The belly of the fish is incised after pithing the brain, and the ovary removed and put into Ringer's solution. The ripe unfertilized eggs which can be distinguished by size and translucency are isolated with glass needles and kept in the Ringer's solution. Only a few eggs (5-6%) show autoactivation in Ringer's solution. These are discarded. An adult male is cut open and the testis is also isolated and kept in the Ringer's solution. Insemination of ova in Ringer's solution is performed by teasing the testis in Ringer's solution.

*Later known as Yamamoto's solution. This property of Oryzias eggs, retaining fertilizability for a long period, is somewhat different from that of ova of Cyprinidae and most fishes (except Salmonidae) in which 100 percent of ripe ova manifest autoactivation even in an isotonic Ringer's solution.

Cortical Alveoli and their disappearance during fertilization

Sars (1876), a Norwegian ichthyologist, seems to have been the first to observe the cortical alveoli in teleostean ova, although he erroneously identified them as "oil bladders" in the unfertilized ovum of the cod (Gadus morrhua). To Ryder (1884), however, belongs the credit for having called attention to the fact that the "vesicles" (now cortical alveoli) are distinct components of the cod ovum which differ from oil droplets, and that they disappear after fertilization. He stated that the disappearance of the "vesicles" is a manifestation of fertilization. Herfort (1901) described the presence of the cortical alveoli and their disappearance after fertilization in the ovum of the river lamprey Petromyzon (Lampetra) fluviatilis. All of them, however, entirely failed to realize the significance of the phenomenon in fertilization.

Kagan (1935) observed the "platelets" embedded in the cortical layer of the unfertilized egg of the killifish (Fundulus heteroclitus) and their disappearance in either fertilized or activated eggs. He, however, did not observe a progressive breakdown of the platelets during fertilization. According to him, the platelets of unfertilized eggs disappear gradually on immersion in sea water. The peculiar behavior of the unfertilized Fundulus egg in sea water as described by Kagan (1935) led Rothschild (1958) to express some doubts as to the identity of the platelets and the cortical alveoli. Kagan stated that the unfertilized eggs are "activated" by immersion in sea water and that the disappearance of the platelets occurs about one hour after stripping, and that the process itself is much slower than in fertilized eggs. Wessels (1953), also, stated that "activation" by sea water proceeds differently from fertilization, viz., the platelets disappear in random fashion, some persisting long after the formation of the perivitelline space. These observations lead me to the view that the so-called "activation" of the Fundulus egg by sea water is not a typical activation but either over-ripeness or incomplete activation, in which some cortical alveoli remain unchanged in places. Typical activation is characterized by a sudden triggering off of the cortical reaction which is completed within a few minutes. Kagan observed that in some eggs the remaining platelets come together in clusters which stream toward the blastodisc, and that the size of the blastodisc is inversely proportional to the quantity of unbroken platelets. On the basis of these observations he arrived at the erroneous conclusion that the platelets were directly concerned in the formation of the blastodisc. Kagan's observations seem to indicate that incomplete breakdown of the platelets may interfere with the process of ovoplasmic segregation resulting in a small blastodisc, and that unbroken platelets are moved toward the animal pole by the protoplasmic stream during the formation of the blastodisc. From these considerations, it is almost certain that the platelets of the Fundulus egg are none other than the cortical alveoli of other fish eggs.

Mechanism of chorion separation

In order to understand the mechanism of separation of the chorion following fertilization and artificial activation, it is necessary to investigate the permeability of the chorion as well as the nature of substances released from the ovoplasm. Yamamoto (1936) in Oryzias eggs found that the chorion is permeable to such crystalloids as salts, sugars, and dyes. When the fertilized eggs are immersed in a concentrated solution of salts or sugars, a temporary depression of the chorion results but the chorion eventually resumes its tension and spherical form. The temporary depression is provoked owing to the fact that the rate of the outflux of water is more rapid than that of influx of a crystalloid. A subsequent study (Yamamoto, 1939b) pertaining to the mechanism of formation of the perivitelline space demonstrated that a permanent depression of the chorion persists when the fertilized egg is placed in the egg white of the hen. This shows that unlike crystalloids the chorion is impermeable to egg albumin. Unpublished experiments indicate that the same is true of gum arabic.

When a slit is made on the chorion "Schlieren" appear, indicating that a colloid oozes out. it was concluded that the chorion is a dialyzing membrane and that it is under tension owing to the colloid osmotic pressure of the perivitelline substance. The situation is the same as that in sea urchin eggs, as demonstrated by Loeb (1908).

Since the breakdown of the cortical alveoli is subsequently followed by separation of the chorion the writer inferred that the formation of the perivitelline space may be due to release of a colloid from cortical alveoli during fertilization and artificial activation (Yamamoto, 1944a). This is in harmony with the writer's observation that a slight but significant decrease (7%) in volume of the ovoplasm takes place after fertilization (Yamamoto, 1940).

Actual observation of the release of colloid from cortical alveoli during fertilization and artificial activation in teleosts is very difficult because the process is completed instantly and the released colloid is transparent.

The bottom series of figures in Fig. 6-4 is based on a side view of the process as observed when the alveolus is focused at the equator of the ovum. In this profile view the outline of the releasing colloid cannot be observed in the intact egg. When the unfertilized egg is stained with 0.002 percent neutral red in Ringer's solution for 4 hours, the content of the cortical alveolus stains ochre red. When the stained egg is activated by pricking with a needle, release of the colored colloid from the cortical alveolus can be observed. Combining the surface and side observations, it has become clear that the obscure ring of the double rings observable in the surface is none other than the outline of an aggregate of colloid released from the cortical alveolus. As the released colloidal aggregate disperses, its outline becomes gradually invisible. In the medaka, the released aggregate of the colloid soon becomes a homogeneous solution.

When we observe the course of the breakdown of cortical alveoli of the Oryzias egg under a microscope equipped with a condenser, the color of the content of cortical alveoli becomes light pink during the breakdown process.

The fertilization-wave or impulse

To account for the progressive breakdown of the cortical alveoli beyond the point of sperm attachment or site of artificial stimulation, e.g., pricking, three possibilities can be considered. First, the breakdown of an alveolus itself at any point may cause in some way the breakdown of the adjacent alveoli, thus eliciting the progressive breakdown. Figuratively speaking, this mechanism may be likened to a leaning row of books in which, when the first falls down, the others follow in succession. This possibility, however, can be ruled out by the following centrifuge experiment: By centrifugal force, the cortical alveoli of the Oryzias ovum are forced to accumulate in the centrifugal side. There is no relationship between the egg axis and the direction of the accumulation (Fig. 6-5). Centrifuged eggs retain their capacity for fertilization and artificial activation. Even those centrifuged eggs which are alveoli-free in the animal hemisphere can be fertilized. The alveoli-free cortex of centrifuged eggs can be stimulated by pricking. An important fact observed is that only the cortical alveoli embedded in the protoplasm at the centrifugal side break down while those accumulated in the yolk remain unchanged during fertilization or activation.

As the third possibility, an intracellular or intracortical diffusion of the sperm-bearing substance, or of a locally produced substance from the cortex in the case of pricking, might be postulated. That such a diffusion concept cannot be tenable is proved by the following experiment and consideration. When the vegetal pole, which has lowest sensitivity, is pricked with a 5� needle point, some eggs show no cortical response. In such cases, not a single alveolus at the vegetal pole breaks down. Such eggs, however, show complete breakdown of all the cortical alveoli on stimulation with a 15-20-� needle point, or on repeated stimulations with a 5� needle point. If some cortical substance which causes the breakdown of the cortical alveoli were produced at the site of pricking and diffused either intracellulary or intracortically, some alveoli at the site of pricking should break down even in the case of unsuccessful pricking with a 5� tip point. No such condition is realized. Hence, the diffusion concept can be ruled out.

The matter under consideration is of great significance in the concept of fertilization and artificial activation. From the concept of the fertilization-wave, there follows immediately this important corollary: If a chemical reaction between the sperm and egg substance ever takes place, it may be restricted to the point of sperm attachment only. In other words, the sperm substance is not directly responsible for the cortical changes at all points of the surface of the ovoplasm. These cortical changes are the result of reactions of ovogenous substances entirely.

Among pioneers in the physiology of fertilization, F. R. Lillie (1914) rightly stressed this non-diffusion concept. According to him, cortical changes in the entire surface of echinoderm eggs, except for the point of sperm attachment, are due to activation of an ovogenous substance (fertilizin) but not to the sperm-bearing substance. Consequently, he assumed that if one molecule of fertilizin is activated, a fertilizin-activating change spreads beyond the point of attachment of the successful sperm. Unfortunately, however, he gave no indication of the mechanism by which one activated molecule (fertilizin) activates the neighboring molecules. Although Lillie's fertilizin theory as such cannot be wholly adequate, he seems to have grasped the fundamental nature of the cortical changes.

In this context, it may be remarked that the bulk of the sperm of the medaka is roughly 1/1,000,000 of the ovum. It is difficult to suppose there is a stoichiometrical relationship (molecule to molecule) between the sperm-bearing substance and the whole of a reactive substance contained in the entire surface of the ovum, to account for cortical changes during fertilization. If the diffusion of sperm substance were valid it would be logical to expect to find a large sperm for large eggs (e,g., salmonids and Elasmobranchia) and a small sperm for small eggs. This is not the case.

The intrinsic phenomenon occurring during fertilization and artificial activation is the fertilization-wave or impulse which propagates itself through or on the surface of the protoplasmic layer, and eventually causes easily visible breakdown of the cortical alveoli. Any theory of fertilization that presupposes direct reaction between the sperm-bearing substance and egg substance at all points of surface of the ovum seems to be inadequate. In our theory, it is sufficient for the sperm to stimulate (perhaps by chemical action) the ovum at its attachment point only. Thereafter, the ovogenous cortical reaction takes place automatically by virtue of the self-propagating fertilization-wave. Our theory is consistent with cellular physiology inasmuch as we regard the ovum as an irritable system with the capacity for excitation-conduction.

It is interesting to see whether the fertilization-wave or impulse manifests itself as a detectable electric change: Maeno et al (1956) observed in Oryzias eggs a change in potential appearing simultaneously or shortly before the cortical change induced by insemination or mechanical stimulation. Hori (1958) observed in the same form a single transient negative spike prior to the breakdown of the cortical alveoli, which is followed by a gradual decrease of potential. The latter change seems to be associated with the progressive breakdown of the cortical alveoli.

It may be difficult to separate the electric change associated with the fertilization-wave from those related to other cortical changes such as the breakdown of cortical alveoli or granules. It is remarkable that in sea urchin egg Uehara and Katou (1972) has succeeded in detecting the change of membrane potential which is unequivocally connected with the ferilization-wave.

Role of Calcium ions in fertilization and activation

Calcium ions play an important role in fertilization of fish eggs. If ripe unfertilized Oryzias eggs are washed with Ca-free Ringer's solution, no fertilization takes place. However, eggs shown to have been infertile in Ca-free Ringer's solution can be fertilized when returned to Ringer's solution. The presence of Ca ions is, therefore, sine qua non for fertilization. The Ca ions can be replaced by Mg ions. The former are, however, far more effective than the latter. It seems that Ca ions are prerequisite for the union of the sperm and the ovum. It was found that a very low concentration of Ca ions is sufficient for successful fertilization. Thus fertilization is still 67 percent successful in isotonic solution which contains 0.0001 M CaCl2 per liter (Yamamoto, 1944a). In normal spawning, a trace of Ca ions in fresh water plus those contained between the preformed chorion and the ovoplasm and its surface may be sufficient to ensure successful fertilization, since natural fertilization takes place soon after oviposition before Ca ions in the surface of the egg have completely diffused out.

The question may be asked as to whether Ca ions are prerequisite for the activation process initiated either by the sperm or artificial stimulants. To answer the question, experiments along this line have been performed by the writer (Yamamoto, 1954a). A few unfertilized eggs (5-6%) of Oryzias reveal autoactivation even in isotonic Ringer's solution. The percentage of autoactivation markedly decreases in Ca-free Ringer's solution. No eggs show autoactivation in an isotonic oxalate solution or in oxalated Ringer's solution. When unfertilized eggs are treated with isotonic oxalate solution they lose the capacity for cortical changes on stimulation by pricking. These eggs show the cortical response without further stimulation when transferred to Ringer's solution. The presence of calcium ions favors the cortical response of eggs on stimulation with sodium oleate and saponin. Only a few eggs show activation if the eggs are first treated with chemical agents dissolved in Ca-free Ringer's solution and then transferred to Ca-free media.

A number of experiments were performed in which calcium ions were removed by means of sodium oxalate. When the unfertilized eggs are immersed in an isotonic solution of sodium oxalate (M/11) for 10-15 minutes, none of the eggs show any visible cortical changes on stimulation with pricking at the animal pole. If these eggs are transferred to Ringer's solution at 10 minutes after pricking, 62-65 percent of them undergo typical cortical changes without further pricking. The fact that the oxalate solution itself is not an activating agent was shown by a number of experiments in which normal eggs were immersed in the oxalate solution for various periods and then transferred to Ringer's solution. Not a single egg showed activation. We are forced to consider that some events occur in oxalated eggs when they are pricked in oxalate solution. In this respect oxalated eggs differ fundamentally from anesthetized eggs, which also fail to respond to stimulation, but display no cortical changes on return to Ringer's solution. It requires further stimulation to induce cortical changes.

It may be postulated that the activation process comprises a series of events which are connected in a catenary way. Physiological studies favor the view that reception of stimulation and evocation of excitation can be separable processes (see Hashida, 1930, 1931). To account for the stimulation of oxalated eggs it is postulated that stimulation is established by pricking in oxalate solution but the later processes are blocked by the absence of Ca ions. It follows that Ca ions are not necessary for the primary phase (reception of stimulation) but necessary for later phases (fertilization-wave or breakdown process of the cortical alveoli or both).

Calcium ions are not only unnecessary for the primary phase in activation but there is some evidence to show that Ca ions are released from the cortex on stimulation (see Yamamoto, 1954).

We want to know a series of events occurring from the sperm-egg contact to completion of cortical changes. Taking into consideration of the forgoing facts, the cortical changes involved in activation of Oryzias egg may tentatively be expressed by the following catenary reaction: Stimulants �� Reception of stimulation �� Evocation of excitation (fertilization-wave or impulse) �� Breakdown of cortical alveoli �� Release of colloid from alveoli �� Separation of chorion by colloid osmotic pressure. A more detailed chain reaction will be given later. Dettlaff (1958) studied the role of Calcium ions in the stimulation of eggs and spread of the cortical reaction by using eggs of Acipenseridae.

Dark-red granules (a-granules) around cortical alveoli

Careful observation of the ripe unfertilized egg of Oryzias indicates that there exist specific granules in the interstitial protoplasm surrounding the cortical alveoli. The presence of these granules is well defined when one observes the ripe unfertilized egg at a magnification of 400 or higher. They are approximately 0.3-0.5 � in diameter, and the largest solid granules among those in the protoplasm and look dark red. In the preliminary account on these granules (Yamamoto, 1951b), they were denominated as "cortical granules A." These granules may more properly be called dark-red granules or the a-granules (Yamamoto, 1958, 1961, 1962). They can be observed when an ordinary microscope with a condenser is adjusted to focus on the equator (maximum diameter) of a cortical alveolus. Otherwise they cannot be detected because of spherical aberration of the cortical alveolus, which exerts a lens-like effect.

It may be noted that the a-granules are not found on the envelope of the isolated cortical alveoli freed from protoplasm. This indicates that the granules are contained in the protoplasm and are not firmly affixed to the enveloping covering of the cortical alveolus, although they are closely associated with it in vivo. Besides these a-granules there are numerous pink granules different from the a-granules, which we call b-granules. The b-granules are fainter in color and smaller in size as compared with the a-granules. In the small alveoli-free area adjacent to the animal pole only b-granules are found, the a-granules being absent there. In just ripe eggs, the cortical alveoli are embedded so compactly in the protoplasmic layer that there is no remaining unoccupied protoplasmic area except around the animal pole. Careful observation on the interstitial protoplasmic layer surrounding the cortical alveoli revealed that it contains both a- and b-granules (Fig. 6-6). In overripe eggs, however, some cortical alveoli have spontaneously broken down here and there, with the result that vacant protoplasmic areas appear in places. In these vacant areas there are only b-granules.

It was found that the outline of the a-granules becomes indistinct before dissolution of the envelope of the cortical alveoli during fertilization and artificial activation. Apparently the a-granules start to dissolve before breakdown of the cortical alveoli and finally disappear. When the dissolution of the cortical alveoli is continuously observed during the dissolution process under a microscope with a condenser, their content is seen to change from colorless to pale pink as if some constituent of these granules were dissolved into the cortical alveoli. After fertilization or artificial activation, there are no a-granules in the protoplasmic layer, while the b-granules remain apparently unchanged (Fig. 6-6). Following the writer's observation on living materials Aketa (1962) demonstrated cytochemically that the a-granules are rich in ribonucleic acid.

Having obtained indirect evidence of enzymic dissolution of the envelope of the cortical alveolus, the chain of events occurring during the fertilization reaction and artificial activation as outlined in the preceding section can be expressed more exactly as this: Stimulants -> Reception of stimulation -> Evocation of excitation (fertilization-wave or impulse)-> Activation of the enzyme-> Enzymatic dissolution of alveolar envelope-> Release of alveolar content ->  Separation of chorion by colloid osmotic pressure.

Experiments of isolated alveoli presented previously suggest that the enzyme responsible for dissolution of the alveolar envelope resembles esterase rather than lipase. The time course of the breakdown of the cortical alveoli, as outlined in Figure 6-4 suggests that the enzyme may first dissolve the outermost region of the cortical alveolus as well as the plasma membrane at the common tangent.

(Reference to the present Chapter are given at the end of the next Chapter)