Eggs of some archaic fishes, such as Polypterus, sturgeon or the holosteans, resemble amphibian eggs. Not only has the similarity been found in the amount of yolk (mesolecithal types of eggs), which indicates the manner of egg cleavage (holoblastic) in these vertebrates, but also in the morphological composition of yolk. Fat droplets, yolk platelets and an unidentified fluid in which they are suspended, in fishes sometimes called ichthulin, are three main components of yolk in these eggs.

Yolk platelets have been an object of interest to various observers because of their unique structure (Hwan Sun, 1962; Ward, 1962; Karasaki, 1963) and the problems of their formation (Lanzavecchia, 1960; Ward, 1962) and breakdown when being utilized (Karasaki & Komoda, 1959; Hwan Sun, 1962; Jurand & Selman, 1964). Most information concerning their structure and physicochemical properties has, however, been acquired by observations of amphibian yolk platelets. When in 1946 Holtfreter described the behaviour of yolk platelets of the frog (Rana pipiens) at different osmotic concentrations he found that yolk platelets are marked by birefringence and that during intracellular digestion they can split into smaller discs, which still show birefringence. A similar effect can be achieved by treating yolk platelets with weak acids or alkali (Gross & Gilbert, 1956; Ringle & Gross, 1962), the smaller discs thus obtained still having the analogous properties of anisotropic substances. These observations made it possible to demonstrate that amphibian yolk platelets are composed of a substance having a laminar structure. Observations were made using the electron microscope by Karasaki & Komoda (1958), who described the hexagonal crystalline structure limited to the core of the platelet, while the superficial portion surrounding the core has a granulo-fibrous structure enclosed by a semi-permeable membrane.

Yolk platelets of the fishes mentioned at the beginning of this paper behave similarly to the amphibian ones, which they also resemble in appearance (Grodziński, 1958, 1963, 1968). The authors studied yolk platelets of the eggs of Amia and Lepisosteus (Holostei) to determine to what degree the ultrastructure of these platelets resembles that of the amphibian yolk platelets.

Mature ovarian eggs of two species of fishes, the gar-pike (Lepisosteus osseus) and bowfin (Amia calva), caught in Wisconsin in the United States, were studied. The eggs were obtained from females in the month of May, in the middle of the spawning season. They were quite ripe; their size (2.2-2.5 mm in diameter) and appearance corresponded to the description of freshly layed eggs of Amia and Lepisosteus (Dean, 1895; Whitman & Eydeshymer, 1897). Eggs were fixed whole or crushed and the yolk mass, flowing out of them, was fixed. Parallel fixatives applied were 4 % formalin neutralized with magnesium carbonate, cacodylate-buffered 5 % glutaraldehyde and 2 % aqueous solution of osmium tetroxide with 0.02 M calcium chloride added. The eggs were left in the aldehyde fixatives for 24 h at 4 °C. In the osmium tetroxide solution they were fixed for 4 h at 4 °C. After fixation all the preparations were dehydrated through a graded series of ethanol and then kept in 96 % alcohol for several weeks. The material collected was fixed and partly dehydrated in the Institute of Biophysics, University of Chicago, and sent in 96 % alcohol to Poland. The dehydrated material was embedded in Epon 812 (Luft, 1961) and polymerized at 60 °C. Ultra-thin sections were cut with a diamond knife on a Porter-Blum microtome. The sections, silver in colour, were picked up on copper grids coated with carbon film and ‘stained’ in uranyl acetate and lead citrate (Reynolds, 1963). They were examined with a Japanese Electron Optics Co. electron microscope, type JEM-5Y. Measurements were taken using an eyepiece micrometer to an accuracy of 0-1 mm on positives enlarged 150000 times.

Light microscope observations

In the yolk of Holostean fishes one can distinguish fat droplets and yolk platelets in the light microscope. The latter behave characteristically in NaCl solutions of different osmolarity and react with some supravital dyes (Grodziński, 1963, 1968).

The yolk platelets of both species investigated, as a rule, are elliptical. Their dimensions vary within wide limits. The long axis of the platelet of Amia measures 5-6µ, of Lepisosteus 3-22µ. In the yolk of the latter fish some yolk platelets are rectangular in shape and 20-25 µ long (Fig. 1a).

Fig. 1.

A single yolk platelet from an oocyte of Amia in 0.18 M-NaCl solution (a), and in 1 M-NaCl solution (b,c). A single yolk platelet from an oocyte of Lepisosteus (d) in 1 M-NaCl solution, x 1000.

Fig. 1.

A single yolk platelet from an oocyte of Amia in 0.18 M-NaCl solution (a), and in 1 M-NaCl solution (b,c). A single yolk platelet from an oocyte of Lepisosteus (d) in 1 M-NaCl solution, x 1000.

The homogeneous appearance of the yolk platelets changes after a few minutes in a hypertonic (1 M) solution. A borderline appears between the transparent sheath and greyish core (Fig. 1b). Shortly after, the core disintegrates into extremely fine granules, which start rapid whirling movements. At the same time the sheath contracts and rounds up, forming a rectangular disc (Fig. 1c). In isotonic or slightly hypotonic solutions (0.12—0.18 M) the sheath moves apart from the core and forms two ear-like protrusions in Lepisosteus, whereas in Amia it keeps its previous outlines (Fig. 1d).

The supravital basic dyes, such as neutral red, colour the core. The acid ones, e.g. trypan red, do not enter the platelet. The sheath exhibits therefore two qualities: semipermeability and elasticity. The core is of a stiff structure, but may break up into fine granules.

Electron microscope observations

Yolk platelets are composed of three readily distinguishable components. The central portion of the platelet is a structurally homogeneous dense oval core. Stained with lead or uranium ions, the platelet core increases in density. Another component of the yolk platelet is the layer surrounding the core, which differs from the core in its lower density and granular structure (Figs. 2, 3). The whole is enclosed by the third component, a single membrane (Fig. 3).

Fig. 2.

Components of yolk of an oocyte of Amia. The yolk platelets (YP) vary in size and shape. Among the yolk platelets are very dense fat droplets (LP). x 1700.

Fig. 2.

Components of yolk of an oocyte of Amia. The yolk platelets (YP) vary in size and shape. Among the yolk platelets are very dense fat droplets (LP). x 1700.

Fig. 3.

A single yolk platelet from an oocyte of Amia. The core has two components, the central one occupying the greater part of the platelet and the peripheral cortex of somewhat greater density. The core is surrounded by the superficiail layer (SL), limited in turn by the membrane (Mb), x 6000.

Fig. 3.

A single yolk platelet from an oocyte of Amia. The core has two components, the central one occupying the greater part of the platelet and the peripheral cortex of somewhat greater density. The core is surrounded by the superficiail layer (SL), limited in turn by the membrane (Mb), x 6000.

The seemingly homogeneous structure of the core of the yolk platelet observed under high magnification reveals distinct zonal differences in density. The core consists of a less dense central part and a periphereal layer of a greater density, giving it the shape of a regular oval (Fig. 3). Besides this difference in density the two zones differ also in the organization of the macromolecules of which they are built.

The cores of most yolk platelets observed showed a compact granular structure with no special regular organization. The central part differed from the cortical part only in density. However, in some yolk platelets we observed regular striations (Fig. 4), consisting of light and dark bands. The striations of the core were at varying angles in individual platelets and no regularity in relation to the long axis of the yolk platelet could be observed. The measurement of the light and dark bands was very difficult, for the boundary between these bands, being blurred, was hard to define precisely. Another difficulty was more fundamental: measurements were being made on bands magnified 1500000 times, which appeared to be built of globules connected by blurred narrowings. These narrowings, however, seem to be optical artifacts caused by the resolving power being insufficient to separate individual particles arranged linearly, which under low magnification look like a band. This is true of both dark and light bands. An attempt was therefore made to measure the diameters of these particles. In the case of light particles the diameter ranged between 68 and 70 Å, and the narrowing between two such particles measured about 47 Å. Similar values were obtained for dark particles and their narrowings. The cores of a few yolk platelets looked as if they had two systems of striation, intersecting each other at an angle of 47° (Fig. 5). This image could, however, be changed by shifting the focus at the section to the rear, in which case only one system of bands was seen, as in Fig. 4. The value for the angle (47°) does not seem to be constant, since various angles of intersection were observed in particular platelets. In some of the gar-pike platelets examined the linearly arranged particles gradually changed their arrangement into a hexagonal one at the opposite margin of the platelet (Fig. 6). In this arrangement we observed a hexagonal figure built of light particles about 47 Å thick, the sides of this figure being 68 Å long. If dark particles were assumed to be the starting point, the figure obtained was also hexagonal, but this time it was built of dark particles, 68 Å thick and arranged round a central particle. In such a figure all the dark particles were 47 Å apart.

Fig. 4.

Fragment of the central part of the core of a yolk platelet of Amia. The distinct striation consists of dark and light bands alternating regularly at intervals of 65 and 47 Å. x 128000.

Fig. 4.

Fragment of the central part of the core of a yolk platelet of Amia. The distinct striation consists of dark and light bands alternating regularly at intervals of 65 and 47 Å. x 128000.

Fig. 5.

Fragment of the central part of the core in a yolk platelet of Amia. The central part looks as if there were two systems of striation intersecting at an angle of 47°. This apparent image is formed due to the hexagonal arrangement of dark (dense) particles, x 128000.

Fig. 5.

Fragment of the central part of the core in a yolk platelet of Amia. The central part looks as if there were two systems of striation intersecting at an angle of 47°. This apparent image is formed due to the hexagonal arrangement of dark (dense) particles, x 128000.

Fig. 6.

Fragment of the core in a yolk platelet of Lapisosteus. The linear arrangement of particles at the lower edge of the platelet passes gradually into the hexagonal system at the upper edge, x 168000.

Fig. 6.

Fragment of the core in a yolk platelet of Lapisosteus. The linear arrangement of particles at the lower edge of the platelet passes gradually into the hexagonal system at the upper edge, x 168000.

Cortical layer of the core

The cortical layer of the core differs from the central part in density. Surrounding the central part, it surrounds the polygonal outlines of this part so as to give it the shape of an oval (Fig. 3). The structure of the cortex is distinguished by an irregular arrangement of granules or short fibrils, which vary in density and are very closely packed (Fig. 7). A gradual disorganization of the regular structure of the central part can be observed at its boundary with the cortex, which might suggest that the cortical layer is a portion of the central part, the particles of which are not arranged in a crystalline manner. The measurements of dark and light particles of the cortex also agree with those in the central part, which may also indicate their identity. In the zone bordering upon the superficial layer the granulo-fibroid structure of the cortex of the core is less dense and has irregular outlines. No membrane separating the core from the superficial layer was observed (Fig. 7).

Fig. 7.

Fragment of the core of an Amia yolk platelet. It shows the border area between the central part and the cortex, this last zone having a distinctly denser structure than the central part. The crystalline arrangement of particles in the central part passes gradually into the structural disorder of particles of the cortex (arrows), x 168000.

Fig. 7.

Fragment of the core of an Amia yolk platelet. It shows the border area between the central part and the cortex, this last zone having a distinctly denser structure than the central part. The crystalline arrangement of particles in the central part passes gradually into the structural disorder of particles of the cortex (arrows), x 168000.

Superficial layer

The superficial layer surrounding the core of a platelet varies in thickness; in some platelets it may even appear to vanish at one pole and increase its thickness by many times at the opposite one (Fig. 2). This suggests the possibility of displacement of the core within the superficial layer, or may be due to the fact that the section plane runs tangentially to the long axis of the platelet. The density of the superficial layer is considerably less than that of the core. This layer has a loose organization and it is made up of fine granules and fibrils, about 130 Å thick, arranged irregularly. The organization observed here seems to have been brought about by the coagulating action of the fixatives (Fig. 2, 3). The superficial layer of the yolk platelet, and consequently the whole platelet, is surrounded by a single membrane 180-200 Å thick. In our preparations this membrane is very badly preserved. Poor fixation is indicated here by lack of continuity or by vesiculation of the membrane. Yolk platelets derived from crushed preparations and those kept in alcohol too long were completely devoid of the membrane and, occasionally, of the superficial layer.

The yolk of phylogenetically newer fishes (mainly teleosts) is organized in the form of lipoprotein spheres (without a crystalline core) like the yolk of birds or reptiles (Grodziński, 1939, 1949, 1951, 1954, 1956). This type of yolk might be called ‘bird type’ . The phylogenetically older groups of fishes (sharks, sturgeons or holosteans) have yolk organized in the form of platelets (with a crystalline core) like amphibians. Therefore we propose to call this type of yolk organization the ‘amphibian type’ .

The yolk platelets of a holostean consist of three main components: the crystalline core, the surrounding superficial layer and the single membrane, which encloses the whole platelet. This organizational pattern can also be seen in another group of vertebrates, amphibians, e.g. Triturus, Diemictylus, Rana and Bufo (Ward, 1962; Karasaki & Komoda, 1963). Our attention was concentrated mainly on one element of the platelet, the core. The central part of the core, composed of two types of substances varying in density and arranged alternately, shows a clear crystalline structure, shown by its periodic striation. However, this would raise doubts whether our images of yolk platelets are images of actual molecular architecture or whether they are effects of the diffraction of electrons by this architecture. Karasaki (1963) rejects the diffractive interpretation of the striation on the assumption that the particles which make up the central part of the core of an amphibian yolk platelet are sufficiently large and can be resolved with the electron microscope. Karasaki had at his disposal the results of chemical analyses of amphibian yolk (Wallace, quoted by Karasaki, 1963), which allowed the rough determination of the size of particles on the basis of their molecular weight. No data concerning the chemical composition of holostean yolk have been published as yet. Nevertheless, great similarities in molecular organization of yolk platelets between holosteans and amphibians have provided grounds for us to compare these structures and state that there is an analogy between them in the size of particles of which they are built. Consequently we would suggest that the striations observed in the central part of the core of the holostean yolk platelet is an image of the actual molecular structure and not a diffraction effect.

The values obtained by us for the widths of the light and dark bands differ from those given by the authors who worked on amphibians (Karaski & Komoda, 1959, 1963; Ward, 1962; Jurand & Selman, 1964), but the differences do not seem to be significant. Under high magnification and at better resolution the uniform bands split into regularly arranged spherical particles, whose diameter corresponds more or less exactly to the width of the band. If one manages to find yolk platelets sectioned at different angles, then the particles making up the band are arranged into parallelograms, rhomboids, or, in the case of the greatest inclination of the platelet, a hexagonal pattern. On the basis of these observations we may venture the conclusion that the particles are arranged in long hexagonal ‘pillars’ .

Our observations showed that the density of the particles can be further increased by the use of uranium or lead salts. This phenomenon is due to the specific adsorption of these ions on phosphate particles present in protein and lipid molecules in the yolk platelets.

A small difference in density and lack of crystalline structure distinguish the central part of the core from the thin cortical layer. Lack of a sharp demarcation line between these zones and the existence of some continuation between the regularly organized molecules of the central part and the molecules of the cortex scattered in a disorderly manner permit the supposition that we are concerned here with the same components. The measurements of the cortical particles agree with those of the particles of the central part of the core.

The core of the yolk platelet is surrounded by the superficial layer and the whole is enclosed by the membrane, which structurally resembles the membranes making up other organelles of the oocyte. The presence of this membrane, though sometimes called into question (Ringle & Gross, 1962) is confirmed by some physico-chemical properties of holostean yolk platelets as well as by the morphological evidence. Such phenomena as the swelling or shrinking of the superficial layer in NaCl solutions of different osmolarity, the penetration of alkaline dyes and the blocking of acid dyes can be explained only by the presence of a semi-permeable membrane (Grodziński, 1968). The poor state of preservation of the membrane or sometimes even its absence in some of the platelets observed by us is due to the poor conditions of fixation or dehydration of the material.

The yolk of holostean fishes (Lepisosteus and Amid) consists of fat droplets and yolk platelets. This type of yolk can be called the ‘amphibian type’ . The yolk platelets are composed of three main components: a crystalline core, a superficial layer surrounding it and a membrane, which encloses the whole platelet. The platelet core shows distinct dark and light periodic bands alternating at regular intervals of 65 and 47 Å. Yolk platelets sectioned at various angles reveal corresponding changes in the arrangement of bands. High magnification and good resolution made it possible to show that the bands are made up of dark and light spherical particles, 65 and 47 Å in diameter. These particles are arranged hexagonally in the core of the platelet.

Le vitellus des Poissons Holostéens

Le vitellus des Holostéens (Lepisosteus et Amid) consiste en gouttelettes lipidique et en plaquettes vitellines. Les auteurs ont nommé ce type de vitellus le ‘type amphibien’ . Les plaquettes vitellines sont formées de trois composants principaux: un noyau cristallin, une couche superficielle l’ entourant et une membrane qui entoure la plaquettes entière. Le noyau de la plaquette présente des bandes périodiques distinctes, sombres et claires, alternant à intervalles réguliers de 65 et 47 Å. Des plaquettes vitelline sectionnées selon des angles variés révèlent des modifications corrélatives dans la disposition des bandes. Un fort grossissement et une bonne résolution ont permis de constater que les bandes sont formées de particules sphériques sombres et claires, de 65 et 45 Å de diamètre. Ces particules sont disposées en hexagones dans le noyau de la plaquette.

We are grateful to Professor W. Bloom, from the Department of Biophysics, University of Chicago, for his personal and financial help in supplying us with the material for investigations, and J. Bigaj for his skilful technical assistance.

Dean
,
B.
(
1895
).
The early development of gar-pike and sturgeon
.
J. Morph
.
11
,
1
25
.
Grodziński
,
Z.
(
1939
).
Zur Morphologie des Huhnereidotters unter normalen und experimentalien Bedingungen
.
Bull. Acad. Polonaise Sci. Cracovie
1939
, pp.
317
55
.
Grodziński
,
Z.
(
1949
).
The fat in the yolk of the Sand-lizard Lacerta agilis L
.
Bull. Acad. Sci. Cracovie
1949
, pp.
367
81
.
Grodziński
,
Z.
(
1951
).
The morphotic components in the yolk of the turtle Geoclemys reevesi Gray
.
Bull. Acad. Sci. Cracovie
1951
, pp.
68
116
.
Grodziński
,
Z.
(
1954
).
Zóltko rózanki Rhodeus amarus
.
BL Folia Morph
.
4
,
13
26
.
Grodziński
,
Z.
(
1956
).
Some physico-chemical properties of the yolk of thermophilic fish
.
Acta biol. Exper
.
17
,
1
,
41–60
.
Grodziński
,
Z.
(
1958
).
The yolk of the dogfish
.
Acta biol. Cracoviensia
1
,
55
68
.
Grodziński
,
Z.
(
1963
).
The yolk morphology of Acipenser giddenstadti colchicus Brandt
.
Acta biol. Cracoviensia
6
,
97
110
.
Grodziński
,
Z.
(
1968
).
The yolk of the Holostean fishes
.
Acta biol. Cracoviensia
(in the press)
.
Gross
,
P. R.
&
Gilbert
,
L. I.
(
1956
).
Chemistry and ultrastructure of amphibian yolk platelets
.
Trans. N. Y. Acad. Sci
.
19
,
108
18
.
Holtfreter
,
J.
(
1946
).
Experiments on the formed inclusions of the amphibian egg. I. The effect of pH and electrolytes on the yolk and lipochondria
.
J. exp. Zool
.
101
,
355
406
.
Hwan
Sun
(
1962
).
Electron microscopic studies on structural changes of developing cells of the anuran embryos
.
Embryologia
7
, no.
3
,
185
200
.
Jurand
,
A.
&
Selman
,
G. G.
(
1964
).
Yolk utilized in the notochord of newt as studied by electron microscopy
.
J. embryol. exp. Morphol
.
12
, no.
1
,
43
50
.
Karasaki
,
S.
(
1959
).
Electron microscopic studies on cytoplasmic structures of ectoderm cells of the Trituras embryo during the early phase of differentiation
.
Embryologia
4
,
247
77
.
Karasaki
,
S.
(
1963
).
Studies on amphibian yolk. 1. The ultrastructure of the yolk platelet
.
J. Cell Biol
.
18
, no.
1
,
135
51
.
Karasaki
,
S.
&
Komoda
,
T.
(
1958
).
Electron micrographs of crystalline lattice structure in yolk platelets of the amphibian embryo
.
Nature, Lond
.
181
,
407
.
Lanzavecchia
,
G.
(
1960
).
The formation of the yolk in frog oocytes
.
Proc. 4th Eur. Reg. Conf. E. M. Delft
2
,
746
9
.
Luft
,
J. H.
(
1961
).
Improvements in epoxy resin embedding methods
.
J. biophys. biochem. Cytol
.
9
,
409
14
.
Reynolds
,
E. S.
(
1963
).
The use of lead citrate at high pH as an electron-opaque stain in electron microscopy
.
J. Cell Biol
.
17
,
208
12
.
Ringle
,
D. A.
&
Gross
,
P. R.
(
1962
).
Organization and composition of the amphibian yolk platelet. I. Investigations on the organization of the platelet
.
Biol. Bull. mar. biol. Lab., Woods Hole
122
,
263
75
.
Ward
,
T. R.
(
1962
).
The origin of protein and fatty yolk in Rana pipiens. II. Electron microscopical and cytochemical observations of young and mature oocytes
.
J. Cell Biol
.
14
,
309
41
.
Whitman
,
C. O.
&
Eydeshymer
,
A. C.
(
1897
).
The egg of Amia and its cleavage
.
J. Morph
.
12
,
309
415
.