1. Accelerated cine-photographs show that the unfertilized trout egg exhibits rhythmical movements within its chorion after immersion in tap water. The movement can be described as a periodic axial perturbation or ‘precession’, similar to that which takes place in a slowing-down top. The egg does not, however, spin at the same time. The frequency of the motion is about 3·4 min. at 16·8° C.

  2. There is a strict correlation between the frequency of the egg precession and the frequency of the impedance changes that are observed when the egg is placed in a conductivity cell in an alternating current bridge. The impedance changes exhibit a characteristic ‘impedance pattern’, the frequency of which is identical with that of the precession.

  3. The oscillations are neither initiated nor affected by the measuring current.

  4. The question as to whether the electrical changes are entirely caused by the dielectric heterogeneity and precession of the egg, or whether they are partly due to changes in the vitelline membrane, associated with the egg movements, is discussed. The former is the simpler hypothesis, and there is some evidence in support of this interpretation from impedance measurements made with the electrodes in different planes relative to the egg. It was difficult to reach a firm decision on this point because the electrical method of recording the changes is more sensitive and accurate than existing optical methods.

  5. Examination of the fine structure of the impedance cycle, by means of the ellipse technique and enlarged time-impedance loci at particular frequencies, shows that there is a very small resistance change associated with the impedance oscillations, which is not in phase with the capacitance change. The possibility of these resistance changes being due to variations in the position of the egg relative to the measuring electrodes; combined with small deviations from Maxwell’s formula, or to resistive heterogeneity in the vitelline membrane, or to changes in the resistance of the vitelline membrane, is discussed.

  6. The variation of frequency with temperature has been investigated. In one typical run, the temperature characteristic was 15,200 which corresponds to a Q10 of 2·47. The average value of the temperature characteristic for the whole season was 16,200. There are no obvious ‘breaks’ in the curve relating log velocity to the reciprocal of the absolute temperature; the curve obeys Arrhenius’s equation except at temperatures above 18° C., when irreversible changes tend to set in. There is no very convincing reason for associating this temperature characteristic with developmental processes, particularly as measurements are made on unfertilized eggs. On the other hand, the possibility of all unfertilized eggs undergoing abortive partheno-genetic activation on immersion in tap or stream water is discussed.

  7. Sodium cyanide (M/300, M/500, M/100 in tap water, pH adjusted) reversibly reduces the frequency and amplitude of the impedance changes, which suggests that cytochrome is concerned in the metabolic system responsible for the egg movements. The frequency is more sensitive to cyanide than the amplitude. There are indications that this compound has some effect on the electrical properties of the vitelline membrane, but this has not been systematically examined. Sodium azide, which has no effect at pH 7·4, cannot be used at pH 6·5 because this concentration of H+ in the external medium has a marked, though not permanent, effect on the egg movements and impedance changes.

  8. Neither phlorizin (M/100, M/200, M/500 in tap water, pH not adjusted), nor sodium fluoride (M/25, pH not adjusted) have any effect on the egg movements or the impedance cycle. This suggests that the energy for these changes is not glycolytic in origin; but there is a possibility that the molecules of these inhibitors do not penetrate the vitelline membrane.

  9. The action of the inhibitors does not as yet permit any comparison to be made with ciliary or amoeboid movement, or with muscular contraction, as the experiments do not enable a distinction to be made between the utilization of oxygen in the breakdown phase or the resynthesis phase of the metabolic cycle.

  10. The following substances have no effect on egg movements or impedance changes:

    • Laboratory-distilled water. There is a possibility that distilled water affects the properties of the vitelline membrane, in the sense that some exosmosis of electrolytes occurs when the egg is in the medium. This has not been systematically examined as it has no direct bearing on the egg oscillations.

    • M/25-CaCl2 in tap water.

    • 0·001 M-chloretone in tap water.

      • There is no appreciable periodic change in egg volume.

When a trout egg is placed in a conductivity cell in an alternating current bridge, rhythmical changes of the system’s impedance are observed (Hubbard & Rothschild, 1939). These changes are reversibly inhibited by phenyl urethane (Rothschild, 1940). Analysis of the impedance cycle into its resistive and capacitative components shows that it is mainly due to changes in the capacitance of the system (Rothschild, 1947b). Similar changes occur in fertilized and unfertilized salmon eggs (Rothschild, 1947 a). Few measurements have been made on the capacitances of living cells, and in particular on their plasma membranes, while data on changes in capacitance are even scarcer. The most important are :

  • On fertilization in Arbacia and Hipponoë esculentus eggs, when the plasma membrane capacitance increases by more than 100% (Cole, 1938).

  • When H. esculentus eggs swell in hypotonic sea water, the plasma membrane capacitance decreases (Cole, 1935).

  • During the passage of an action potential along Nitella the capacitance decreases by 15% (Cole & Curtis, 1938).

  • During the passage of an action potential along the squid giant axon the capacitance is said by Cole & Curtis (1939) to decrease by 2%.

  • When the mammalian erythrocyte undergoes haemolysis by water or chemical lysins there are changes in the frequency-dependent properties of the membrane capacitance (Fricke & Curtis, 1934).

  • According to Cole (1928, p. 53), ‘In conversation Dr Hugo Fricke said that he had found a change in the capacity of a frog egg upon fertilization…’.

The stability of the capacitance under varying environmental conditions and during cellular reactivity has contributed to the concept of the plasma membrane as a thin, inert, and electrically insulating lipoid matrix; these properties may be responsible for the large capacitance (1 μF.cm. −2) that is usually observed. It is, therefore, important to consider the changes in membrane structure, or in the system as a whole, which might be responsible for the observed rhythmical changes in capacitance. The interpretation of these phenomena is made more difficult by the discovery, described in this paper, that both unfertilized and fertilized eggs undergo rhythmical movements (Pl. 9). Apart from any detailed interpretation, the existence of these changes gives rise to a number of questions, of which the following are important:

  1. On what metabolic system or systems are the changes dependent?

  2. Why do the changes stop a certain number of days after fertilization?

  3. Is it possible that the changes are not spontaneous, but are due to the treatment of the egg during measurements?

  4. What variations in the external environment have an effect on the changes?

  5. If these effects are peculiar to cells which are morphologically similar to the trout or salmon egg, what morphological peculiarity is responsible?

  6. Do the changes merely indicate that the egg, or some parts of it, are in an abnormal or moribund condition?

Answers to some of these questions will be found in this paper, though it is obvious that the problems raised by these rhythmical changes cover so wide a field that complete or detailed explanations are unlikely to be forthcoming for some considerable time. The relatively short period during which trout eggs are available each year makes a general survey desirable, so that those lines of inquiry which are likely to be most profitable can be established. The following experiments have been performed and will be dealt with in this paper:

  • The effect of temperature changes on frequency.

  • Examination of the variation in frequency with age of eggs, and of the variations in frequency between eggs from the same and from different females.

  • The effects of sodium cyanide, sodium azide, phlorizin, sodium fluoride, chloretone, distilled water, and calcium chloride. The effects of chloretone and CaCl2 were examined for reasons which will be discussed later. The effects of NaCN and NaN3 in inhibiting the cytochrome oxidase system, and of phlorizin or NaF in inhibiting glycolysis, are well known. The experiments with inhibitors represent the beginning of an attempt to identify the metabolic system or systems on which the changes depend. This is a complicated problem, and one which requires specialized biochemical attention.

  • The effect of varying the plane of measurements relative to the egg, to see if there is any correlation between the form and frequency of the change and the morphological characteristics of the egg.

  • Examination of the possibility of there being a resistance change as well as a capacitance change in the system, and of the phase relationship of these two parameters.

  • Photographs and cine-photographs of unfertilized trout eggs have been taken. I am much indebted to Mr K. Williamson for assistance in this work.

Before describing the experiments, it is advisable to consider what physical changes in the egg as a whole might be responsible for the observed impedance changes. The first of these is a periodic change in the membrane dielectric constant, which would be measured as a change of capacitance. The second possibility is a periodic change in egg volume, which is the subject of experiments described in this paper. If the egg swells, the surface membrane must become thinner, unless the egg synthesizes ‘membrane material’ in such a way as to keep the membrane thickness constant, which is improbable. If the membrane becomes thinner, the capacitance will increase, and therefore if the egg swells and shrinks rhythmically, the capacitance will vary in a rhythmical manner. The simplest explanation of periodic changes in volume would be that the egg takes in and expels water periodically; but this raises difficulties because Gray’s experiments (1932) show that the surface of the trout egg is virtually impermeable to water. A third possibility is that the egg changes its shape periodically. The electrical properties of a suspension of spheres are different from those of a suspension of ellipsoids, and it is therefore conceivable, though unlikely, that a periodic change in shape might account for the observed impedance changes. The last possibility is that different parts of the egg surface have different dielectric constants or capacitances, and that the position of these parts relative to the measuring system is not constant. It has long been known that fertilized eggs of certain fresh-water fish exhibit movements. These seem first to have been observed by Rusconi (1840), who thought the movements were due to cilia, and quotes W. Sharpey as saying in his article on cilia in R. B. Todd’s Cyclopaedia of Anatomy and Physiology (1836, p. 606) that this is the case. I have not been able to find any statement to this effect in the article. Ransom (1867) made a systematic investigation of fish-egg movements, but was unable to find any movements in the trout egg. In recent years, fish-egg movements have been investigated by Yamamoto (1940) and to a lesser extent by Kuhl (1939). Detailed references to earlier papers on this subject will be found in the bibliographies at the ends of these two papers. Although movements have on rare occasions been observed in unfertilized fish eggs (Wintrebert, Painlevé & Yung Ko-Ching, 1929), no systematic experiments have been carried out on them, and there has been a tendency to associate such movements with developmental processes. In his most recent paper, Yamamoto (1940, p. 75) says: ‘There are a host of observations to indicate that the eggs of animals reveal irregular amoeboid movement shortly after fertilization’, while in a paper on the movements of goldfish eggs (Yamamoto, 1934), it is stated that the movements start at the four-cell stage, and that the variations in these movements with temperature are similar to those observed during developmental processes. Needham (1942, p. 75), referring to these and other movements in embryos, says: ‘We are still far from any understanding of their meaning, but it will be remembered that protoplasmic contractility plays a large part in echinoderm fertilization….’

Unfertilized trout eggs undoubtedly undergo periodic movements, and it is probably due to their relatively low frequency, which is about one complete cycle every 3 min. at room temperature, that they have not been observed during experiments on these eggs in recent years. In fact, if one looks at a trout egg under a microscope in the ordinary way, no changes are noticeable, and it is only by special methods of observation, accelerated films, or by impedance records, that these changes become apparent. Such movements make the interpretation of the rhythmical impedance changes very much more difficult.

Experiments a–c, described above, and their interpretation, apply equally well to periodic egg movements or periodic egg impedance changes. The extent to which the electrical changes are in part due to structural alterations in the egg surface, or to changes in egg volume, or entirely to the movements of the egg as a whole, is examined in detail in this paper; but, as is mentioned above, impedance measurements on moving systems are exceptionally difficult to interpret.

Unfertilized eggs of the brown trout (Salmo fario), the rainbow trout (Salmo irideus), and the shasta variety were used. There seemed to be no significant electrical differences between these three types of egg. Some eggs were obtained by the usual method in the laboratory; others were sent by post from the Midland Fishery in test-tubes containing about fifty eggs in spring or tap water.

Fertilized eggs were not used this season because no difference between the impedance cycle in fertilized and unfertilized eggs has been detected. It is often taken as a criterion of health if the eggs can be fertilized at the end of the experiment. This is impossible in the case of trout eggs for well-known morphological reasons. Moreover, the moment a trout egg becomes unhealthy, the normal impermeability of the vitelline membrane disappears and salts diffuse out of the egg interior. The electrical method of measurement (q.v.) is ideally suited to following this process. Normally, the experimenter can tell when a trout egg is unhealthy or moribund, because white patches, caused by the precipitation of intracellular globulin, can be seen through the chorion, the patches corresponding to areas of permeability in the vitelline membrane. But unhealthiness, characterized by the exosmosis of salts, can be observed much earlier by electrical than by visual means, since the slightest variation in the electrical conductivity of the external medium, quite apart from changes in membrane resistance, is perceptible in an a.c. bridge.

A geometrically idealized diagram of a trout egg is shown in Text-fig. 1. For purposes of subsequent discussion, three axes at right angles to each other, with their origin at the centre of the egg, are included in the diagram; By the usual convention, the observer at A is looking in the negative z direction, while the observer at B is looking in the positive z direction. The slab E, on the left side of the egg, is one of the electrodes, and is parallel to the x–z plane, but at right angles to the x–y and y– planes. Electrodes in this position are described as being in the ‘normal’ position; electrodes parallel to the plane pqrs are described as being in the ‘vertical’ position.

Text-fig. 1.

Diagram of a trout egg, not to scale. A, observer looking in the negative z direction;-B, observer looking in the positive z direction; E, one electrode; O, origin of rectangular co-ordinates ; pqrs is a plane parallel to the x-y plane.

Text-fig. 1.

Diagram of a trout egg, not to scale. A, observer looking in the negative z direction;-B, observer looking in the positive z direction; E, one electrode; O, origin of rectangular co-ordinates ; pqrs is a plane parallel to the x-y plane.

If the egg in Text-fig. 1 is rotated through 180°, so that the blastodisc is at the south pole, the whole egg will rotate within the chorion until the blastodisc is again at the north pole, because the cap of oil droplets round the blastodisc, and possibly the blastodisc itself, make the animal pole the lightest part of the egg. The actual position of these oil droplets is not quite clear. According to Gray (1932), they are within the vitelline membrane and blastodisc, but it seems possible that they adhere fairly firmly to the insides (or undersides) of these two structures.

Measurements were made with an a.c. bridge of conventional design. The bridge was energized by a heterodyne beat oscillator with continuously variable frequency up to 50 kcyc./sec. The detector was an oscilloscope and tuned amplifier. For some experiments the output of the oscillator was split into two parts; one part supplied the bridge and the other was applied to the x plates of the oscilloscope. The output from the bridge was applied to the y plates, and the phase relationship was so arranged that the resultant Lissajous figure was an ellipse which tilted for standard arm resistance unbalance and widened for capacitance unbalance.

Various types of conductivity cell were used and are shown in Text-fig. 2. The first of these, a, consists of two platinized platinum circular shell electrodes, attached to Zeiss micromanipulators, and dipping into any suitable container. This is used for preliminary experiments. The second type, b, is a cubical glass container, two opposite faces of which are internally covered with platinized platinum plates, the electrodes. A ground-glass plate fits on the top of the container. The base has three small conical glass supports in the centre, on which the egg sits. This cell was thermostatted, and the way this was done is shown in Text-fig. 2c. A third type of conductivity cell is shown in Text-fig. 2d. It is a thermostatted version of cell a. Text-fig. 2 e shows the electrode arrangement when the impedance changes are measured in different planes. Only opposite pairs of electrodes are placed on the egg surface during an experiment. A special conductivity cell is shown in Text-fig. 3. In it the base, upon which the egg sits, can be moved about in the x-y plane, relative to the electrodes, and at the same time the electrodes, which are fixed relative to each other, can be rotated round the egg. The movement of the egg in the x-y plane, and that of the electrodes, are controlled from the same drive, so that these two movements, in different planes, have a fixed phase relationship. The object of this device will be explained later.

Text-fig. 2.

Conductivity cells described in text. A trout egg, radius 0·23 cm., is shown in position in each case. Thermostatting arrangements are shown in c and d.

Text-fig. 2.

Conductivity cells described in text. A trout egg, radius 0·23 cm., is shown in position in each case. Thermostatting arrangements are shown in c and d.

Text-fig. 3.

1 perspex block; 2, electrodes ; 3, spindle ; 4, connecting rod held in contact with lever 5, by the spring 6, which also holds 5 against the eccentric cam 7; 8, spindle carrying glass vessel 9. Rotation of 8 rotates the egg between the electrodes and rocks the electrodes on the spindle 3.

Text-fig. 3.

1 perspex block; 2, electrodes ; 3, spindle ; 4, connecting rod held in contact with lever 5, by the spring 6, which also holds 5 against the eccentric cam 7; 8, spindle carrying glass vessel 9. Rotation of 8 rotates the egg between the electrodes and rocks the electrodes on the spindle 3.

The bridge unbalance wave-form

The existing method of recording the impedance change has the disadvantage that the frequency of the actual impedance variation is half the frequency of the maximum unbalance values in the record. This is shown in Pl. 10 a, which is a record of the bridge unbalance. The actual impedance change derived from this record is shown by the dotted line in Pl. 10 a.

This effect is shown diagrammatically in Text-fig. 4, in which it will be seen that when values of Rp and Cp in the bridge standard arm are such as to balance the egg in its conductivity cell at some arbitrarily selected time during the impedance cycle, the resulting record has double the ‘frequency’ of the actual change; furthermore, according to the moment in time at which balance occurs, the shape of the impedance cycle appears to vary, though in fact it does not. This disadvantage in the method of recording changes would be obviated by taking records entirely outside the region of bridge balance, the impedance changes being studied unilaterally in the envelope curve; but the technical problems associated with this method of measurement make it preferable to adhere to the original method.

Text-fig. 4.

i. The graphs a-e show a sinusoidal variation of impedance (magnitude) with time. The horizontal line in each case represents bridge balance (ohms). is the ohm-time co-ordinate at which balance is first effected. B2,B3, B4 and B6 represent successive bridge balances, ii. Detector record at high input frequency. B1,B2 and B3 represent bridge balances, iii. Detector record at low input frequency showing separation of carrier wave. B3, B4 and B5 represent bridge balances, ai, ii and iii; bi, ii and iii to ei, ii and iii show the effect of varying t in the co-ordinate Bt, first on the value of | Z | necessary to effect balance and secondly on the recorded change.

Text-fig. 4.

i. The graphs a-e show a sinusoidal variation of impedance (magnitude) with time. The horizontal line in each case represents bridge balance (ohms). is the ohm-time co-ordinate at which balance is first effected. B2,B3, B4 and B6 represent successive bridge balances, ii. Detector record at high input frequency. B1,B2 and B3 represent bridge balances, iii. Detector record at low input frequency showing separation of carrier wave. B3, B4 and B5 represent bridge balances, ai, ii and iii; bi, ii and iii to ei, ii and iii show the effect of varying t in the co-ordinate Bt, first on the value of | Z | necessary to effect balance and secondly on the recorded change.

Apart from this we are not interested so much in the recorded impedance cycle as in the electrical changes in the system which produce this impedance cycle. These may be different from the observed bridge unbalance records, and it is upon the former that it is necessary to concentrate. The method of deducing how the various electrical parameters which make up the impedance vary is outlined in the following section.

Fine structure of the impedance cycle

The relationships between changes in some parts of the system comprising the unknown arm of the bridge and the equivalent changes in the bridge standard arm are somewhat complicated. They are easier to understand if one considers a much simpler network than that to which an egg in tap water is equivalent. Suppose that the unknown arm of the bridge contains the network ,and that by means of some device the values of R and C are made to vary sinusoidally, i.e.
formula
We use the letter Ω instead of the more normal ω for the angular frequency, so as not to cause confusion between the frequency of the variation in R and C and the frequency of the oscillator energizing the a.c. bridge. θ is the phase difference between R and C; it may be zero. The impedance of this varying system, at any frequency ω, will be
formula
while the phase angle across the terminals will be
formula

Neither | Z | nor Ψ is necessarily sinusoidal, so a particular shape of observed impedance variation does not automatically mean that variations in the unknown arm components are of the same form. Even if the capacitance alone changes in the unknown arm, the resulting impedance change of the network need not be sinusoidal. Consequently, the shape of the impedance change, which is measured across the terminals of the complete network in the unknown arm, provides no direct information about the form of the variations in individual components in the unknown arm.

In the very simple case examined above, the phase relationship between R and C will be exactly reproduced by Rp and Cp in the bridge standard arm, because the networks in the standard and unknown arms are identical. But if the values of Rp and Cp are used to plot the impedance locus, which involves converting Rp and Cp into their equivalent series resistances and reactances at each frequency, the relative phases and wave forms of Rs and Cs (from which the reactance I/ωCs is derived) may differ from those in the original network. There is, however, a method of establishing whether or not variations in the components in the unknown arm are in phase. Suppose that the unknown arm does consist of a resistance R and capacitance C in parallel, and that the values of R and C vary as in Text-fig. 5 a. There is no point at this stage in making R and C vary sinusoidally, since according to their values, Rs and I/ωCS may contain harmonics. We shall consider four conditions: R and C are in phase, C leads R by , C leads R by , R does not vary, but C does. The last condition determines the arc of the circle on which the static values of R and C lie, and is shown in Text-fig. 5 bI. The other conditions are shown in Text-fig. 5 b II, c and d. From these diagrams it will be clear that if the parameters in the unknown arm are not in phase, the criterion of this condition is that the time locus of the impedance (or admittance) vector terminal at the selected frequency will enclose part of the RSX8 plane and will not be a line. That this must be so is obvious from Text-fig. 5a; the ordinates at t1 and t2 cut the C2 curve at equal C2 values, but R is different in each case. The same applies to t3 and i4, so that in all cases where the phase difference between C and R is greater than o and less than π, the impedance or admittance vector will trace a loop and not a line with time. The width of the figure traced out is a function of the phase difference between C and R for given values of C and R.

Text-fig. 5.

a. Variation of resistance, R, and capacitance C with time. is in phase with R while C2 and C3 are ൞π and ½π out of phase respectively, b. Semicircle I: ⊙, variation of capacitance alone; ⊡, simultaneous in phase variation of resistance and capacitance. Semicircle II, change in impedance locus due to resistance change alone, c. Simultaneous variation of resistance and capacitance, ൞π out of phase, d. Simultaneous variation of resistance and capacitance, ½π out of phase.

Text-fig. 5.

a. Variation of resistance, R, and capacitance C with time. is in phase with R while C2 and C3 are ൞π and ½π out of phase respectively, b. Semicircle I: ⊙, variation of capacitance alone; ⊡, simultaneous in phase variation of resistance and capacitance. Semicircle II, change in impedance locus due to resistance change alone, c. Simultaneous variation of resistance and capacitance, ൞π out of phase, d. Simultaneous variation of resistance and capacitance, ½π out of phase.

‘Precession’

Accelerated cine-photographs show that the unfertilized trout egg undergoes rhythmical movements within the chorion (Pl. 9). This movement can loosely be described as a periodic axial perturbation, or wobble, about a particular point on, or in, the egg. This point may be at the south pole of the egg, or within the egg, on the line joining the north and south poles. The movement is somewhat similar to that part of the motion of a top which is known as precession. The analysis of the motion of a precessing sphere is rather difficult, both conceptually and mathematically, and a detailed consideration of this phenomenon is unnecessary here. The simplest way of considering this motion is by means of the idealized egg diagrams in Text-fig. 6. In these diagrams, the orbits of a spot on the egg surface (in reality an oil droplet) is delineated, while the egg is examined in different planes. To obtain a complete picture of the motion of a precessing sphere by visual means, an observer must look at the egg, or take accelerated cine-photographs of it, in at least three different directions.

Text-fig. 6.

a. Precession of the egg about the south pole; three positions, i, ii and iii, are shown; iv, conical surface generated by the north-south axis. The observer at A sees the globule G describing an elliptical path; the observer at B sees a stationary globule G’. In iv, the three positions i, ii, iii, of the egg show the enveloping surface over which G travels, b. Precession of the egg about the point P on the north-south axis. Note that in this case, the observer at B sees the globule G describing an ellipse, in the opposite sense to that seen by the observer at A.

Text-fig. 6.

a. Precession of the egg about the south pole; three positions, i, ii and iii, are shown; iv, conical surface generated by the north-south axis. The observer at A sees the globule G describing an elliptical path; the observer at B sees a stationary globule G’. In iv, the three positions i, ii, iii, of the egg show the enveloping surface over which G travels, b. Precession of the egg about the point P on the north-south axis. Note that in this case, the observer at B sees the globule G describing an ellipse, in the opposite sense to that seen by the observer at A.

Although it is desirable to express these periodic movements of the trout egg in an unambiguous or formal way, so as to avoid words like ‘wobble’, this paper is concerned with the physiological properties and electrical concomitants of these movements. Formal analysis will therefore be dealt with elsewhere.

Correlation between egg movements and impedance changes

The time taken for an oil droplet to describe one complete orbit is identical with the duration of one complete impedance pattern. The meaning of ‘complete impedance pattern’ is explained in the next section, entitled ‘Wave-form of the impedance change’. Whenever an egg exhibits rhythmical movements, there are associated rhythmical changes in the impedance of the system; the converse is difficult to prove. The electrical method of measurement is so much more sensitive and accurate than visual examination that it is hard to establish the presence or absence of egg movements when the electrical changes have been reduced in amplitude and frequency by special treatment. We cannot as yet say that the impedance changes are exclusively the result of the movements, or that they are not the result, but a concomitant, of the movements, as in the case of muscle. This question is discussed later in this paper.

Wave-form of the impedance change

Examination of Pl. 10 a shows that every fourth bridge unbalance has a different amplitude from the previous three. In other words, the impedance cycle has an amplitude modulation every other cycle. The time interval between two consecutive modulations is the ‘complete impedance pattern’ referred to earlier on; the time interval is identical with the frequency of the egg movement.

This modulation has been further investigated by examining the relationship between the impedance change measured with the electrodes in the normal (lateral) position, and with the electrodes in the vertical position (Text-fig. 2e). The results of these measurements are shown in Pl. 10 b.

It will be seen that whereas the impedance change in the lateral plane consists of a ‘fundamental’ change modulated every other cycle, the change in the vertical plane, though in fact of the same frequency, could be described as a ‘fundamental’ change of half the frequency of the lateral change, with a harmonic of double the frequency.

The way the changes in the two planes are described is largely a matter of choice, because the frequency of the complete impedance pattern is the same in both cases, and the difference between them can be interpreted.

Spontaneity

The evidence that the changes in the impedance of the system are neither caused nor affected by the measuring current is :

  • If the measuring current makes the impedance of the egg or of the whole system vary rhythmically, or alters any existing rhythmical changes in the egg, one would not expect the impedance variation to continue its normal course during an interruption of the measuring current. Pl. 10 c shows the effect of interrupting the measuring current at various times and for various durations. When the current is switched on after an interruption, the impedance variation picks up exactly where it should do if unaffected by the measuring current; it behaves, in fact, as if the current had never been switched off.

  • There is an almost linear relationship between the amplitude of the impedance variation and the voltage across the unknown arm of the bridge. The maximum and minimum values of Rp and Cp remain reasonably constant when the voltage across the unknown arm is varied (Table 1), showing that in any particular egg the amplitude of the impedance variation depends only on the impressed voltage. The lack of discontinuities in the relationship between impressed voltage and the values of RP and Cp supports the conclusion that the current energizing the bridge is not responsible for, nor has any effect on, the observed changes. The fact that fertilized and unfertilized eggs do not undergo these changes when first put into water suggests that they may be caused by the abrupt change of environment when the egg leaves the female and enters a medium with entirely different properties, though we shall see later that the comparative lack of Ca++ in the external medium does not seem to be the responsible factor. When therefore it is said that the effect is spontaneous, all that is meant is that these rhythmical changes are neither produced nor affected by the measuring apparatus.

Table 1.

17 January 1947. UF egg, 3 days old, v =3000. Maximum and minimum values of RPand Cp

17 January 1947. UF egg, 3 days old, v =3000. Maximum and minimum values of RPand Cp
17 January 1947. UF egg, 3 days old, v =3000. Maximum and minimum values of RPand Cp

The effect of changes in temperature

The frequency of the impedance pattern is fairly constant in eggs from one female (Table 2), though eggs from different females show a good deal of variation (Table 3). It would therefore be convenient to measure a number of Q10’s on eggs from one female; but this is hard to do, because of the time each temperature run takes, and the difficulty of doing more than one or two runs at the same time. A typical graph of the variation in frequency with temperature is shown in Text-fig. 7, while in Text-fig. 8 six separate runs are recorded, to show the extent of the variation between different eggs. In both cases the logarithm of the frequency per hour is plotted against the reciprocal of the absolute temperature ; it will be seen that the Arrhenius equation is reasonably well satisfied from the lowest temperatures to about 18°C. Above 18° C. there is a marked flattening of the curve, or ‘break’, though the behaviour of the eggs at these higher temperatures suggests that the change in the slope of the Arrhenius plot is more likely to be a sign of irreversible injury than of a ‘critical point’ (Crozier, 19246). At temperatures above 18° C. eggs behaved in an erratic manner. A number died during the experiments, and only a few runs were done at these higher temperatures. In Text-fig. 7 the average value of the temperature characteristic μ, between 4·15 and 18·1° C., is 15,200, calculated from the Arrhenius equation, which corresponds to a Q10 of 2·47. The slight spread at the lower temperatures, which is visible in Text-fig. 8, is real; it reflects the fact that the rhythmical changes are affected by these low temperatures, and not always in a reversible manner. The mean value of the temperature characteristic for seventeen runs (54 measurements of μ), from c. 18 to 5°C., was 16,200.

Table 2.

29 January 1947. Impedance pattern frequencies in unfertilized eggs from the same female trout. T°C. 16·8

29 January 1947. Impedance pattern frequencies in unfertilized eggs from the same female trout. T°C. 16·8
29 January 1947. Impedance pattern frequencies in unfertilized eggs from the same female trout. T°C. 16·8
Table 3.

Impedance pattern frequencies in unfertilized eggs from different female trout

Impedance pattern frequencies in unfertilized eggs from different female trout
Impedance pattern frequencies in unfertilized eggs from different female trout
Text-fig. 7.

Variation in frequency of impedance change with temperature. Ordinates, log frequenc abscissae, reciprocal of absolute temperature. The temperatures against each co-ordinate are in °

Text-fig. 7.

Variation in frequency of impedance change with temperature. Ordinates, log frequenc abscissae, reciprocal of absolute temperature. The temperatures against each co-ordinate are in °

Text-fig. 8.

Six temperature runs to show individual egg variations. Ordinates, log frequency; abscissae, reciprocal of absolute temperature. Approximate temperatures in °C.

Text-fig. 8.

Six temperature runs to show individual egg variations. Ordinates, log frequency; abscissae, reciprocal of absolute temperature. Approximate temperatures in °C.

The effect of sodium cyanide and sodium azide

M/300, M/500 and M/1000 solutions of NaCN in tap water, with pH adjusted to that of ordinary tap water (c. 7·4), were used. KCN was avoided because of the well-known action of K+ ions on mem brane resistance (Cole & Marmont, 1942 ; Hodgkin & Huxley, 1946). Care has to be taken in setting up and interpreting experiments to test the effect, or lack of effect, of chemical inhibitors on the impedance cycle. The addition of NaCN to tap water alters its conductivity, and control experiments must therefore be made in tap water with conductivity no lower than that of the test solution. Moreover, it is obviously essential that the control tap water should not have any effect on the impedance cycle. This latter point presented difficulties in the case of the controls for the experiments with tap water containing NaN3 (q.v.), but not in the case of the cyanide experiments. In the latter, the control solution was tap water to which NaCl was added until the conductivity was about the same as that of the pH-adjusted, NaCN-tap water solution. NaCl has no effect on the impedance cycle, at any rate in the low concentrations needed for these purposes.

Experiments to determine the effect of inhibitors on the electrical properties of a single cell are difficult to compare with conventional manometric experiments in which, for example, the actual percentage of cyanide-sensitive respiration can be determined. In these experiments there are four possible effects of NaCN which may be expected: reversible reduction in amplitude, reversible alteration in frequency, reversible abolition, irreversible abolition.

To attempt a more strictly quantitative investigation would, at this stage, be unprofitable. It must be remembered that though a trout egg is the size of a small pea, the amount of actual metabolizing tissue present is exceedingly small.

M/1000-NaCN reversibly reduces the frequency of the oscillation. Pl. 11 shows this effect, the egg beating at cycle/min. before treatment; it also shows various phases in the reduction of frequency during immersion in M/1000-NaCN tap water, and the egg recovering its normal frequency, when replaced in its normal environment. These experiments also show that the frequency of the oscillation is much more sensitive to cyanide than is the amplitude, which barely alters even when the frequency is markedly reduced. Stronger solutions of NaCN are needed to reduce the amplitude or to abolish the oscillation altogether; even if M/1000-NaCN would eventually achieve this result, there are several factors which make it difficult to continue this type of experiment for long periods of time. The oscillation can be abolished reversibly by M/300-NaCN, but the experiments are difficult because this concentration tends to kill the egg and it is not easy to take an egg out of the cyanide solution after the oscillation has become invisible but before the egg is irreversibly damaged. This has only been achieved once or twice, though by washing in tap water one can get eggs to recover after a large reduction in amplitude through treatment with M/300-NaCN solutions.

Suppose that NaCN lowers the membrane resistance and causes salts to diffuse out of the egg interior into the external medium. The impedance change, which it will be remembered is mainly capacitativo, would then be shunted by a reduced resistance. Can the reduction in amplitude be accounted for by such changes in resistance, the oscillation not, in fact, being cyanide-sensitive at all? As regards the decrease in amplitude of the oscillation, this possibility cannot be absolutely discarded without quantitative measurements which present difficulties. All that can be said is that this interpretation is improbable. In any case, the frequency, as well as the amplitude, is affected by NaCN, and a reduction in the values of the electrical parameters can have no effect on frequency.

It would be natural to extend these results by experiments with NaN3 instead of NaCN. Some preliminary experiments of this type have already been reported (Rothschild, 1940) and a more detailed examination has now been made. NaN3 only exerts its effect on the cytochrome oxidase system at pH 6·6 or lower (Keilin, 1936). If the effect of NaN3 in tap water at pH 6·5 is to be investigated, the controls must also be studied at pH 6·5. When an egg is transferred from ordinary tap water to tap water whose pH has been reduced to 6·5 with HC1, the oscillation changes in a remarkable way. This is illustrated in Pl. 10d. The amplitude and frequency are reduced almost instantaneously to a very low value. Recovery as regards amplitude gradually sets in, but the frequency seems to be affected for a longer time. This effect of pH reduction raises certain interesting questions, but the azide experiments at this pH necessarily had to be abandoned, as, without controls, they would have been meaningless.

The thermostatting of trout eggs during electrical measurements is most easily done in a conductivity cell of the type shown in Text-fig. 2d, which is open at the top. Some preliminary experiments in this cell gave deceptive and unreproducible results owing to the volatility of HCN, which is the compound responsible for the effects described above. Below pH 8 almost all the cyanide is present in the form of HCN. All experiments were therefore carried out in the sealed cell (Text-fig. 2b) after preliminary experiments with open cells were found to be quite inconclusive. The use of sealed conductivity cells delays temperature equilibration to a certain extent, so that it is difficult to take records of the impedance cycle immediately after an egg has been put into the test solution; but in the case of the relatively weak cyanide solutions the effect is sufficiently gradual to make this delay unimportant. There are indications that stronger cyanide solutions, and possibly even weak ones, have an immediate effect on the electrical properties of the vitelline membrane. A systematic investigation of this effect can only be carried out with eggs which are not beating, According to Gray (1932, p. 289), however, M/100-KCN ‘has relatively little effect on the stability of the eggs if the latter are absolutely fresh and not subjected to mechanical disturbance’. It is evident that a further examination of the effect of KCN and NaCN on eggs which are not beating is necessary.

The effect of phlorizin

Experiments were carried out with phlorizin to see if the changes were in any way dependent on the phosphorylation stage of anaerobic glycolysis, M/100, M/200 and M/500 solutions of phlorizin in tap water (without pH adjustment) were used. These solutions had no effect on the impedance cycle, and unfertilized eggs left for 6 days in M/500-phlorizin in tap water continued oscillating without any signs of being affected by this solution. It can therefore be concluded either that the changes are not associated with phosphorolysis, or that the phlorizin molecule does not penetrate.

The effect of sodium fluoride

Eggs were left in M/25-NaF in tap water for 3 hr.; this treatment had no effect on the rhythmical changes. Eggs which were left in this solution for 16 hr. died though control eggs were quite healthy, and this may be compared with the results with phlorizin. Though the short-period experiments with fluoride and phlorizin may be comparable, this is not necessarily the case with the longer period experiments, as the egg batches were different. The NaF experiments were done much later in the season, when trout eggs are more delicate than earlier on.

The effect of calcium chloride

Though there are several reasons for thinking that the trout egg impedance cycle does not belong to the same class of phenomena as the action potential, it was thought advisable to see if an increase in the Ca++ concentration round the egg had any inhibiting effect. Lack of calcium in the external medium induces spontaneous rhythmical discharges in squid nerves (Arvanitaki, 1939). These spontaneous and rhythmical potential changes are abolished by restoring the normal Ca++ concentration in the external medium. The student of fertilization in fresh-water animals can hardly fail to appreciate the marked change of environment to which an egg is subjected when laid. Within the female trout, the egg is in contact with a fluid in which Δ is about–0·48° C., while after being laid, the egg is in an environment in which Δ is about–0·02° C., and the Ca++ concentration is 0·002 M. The sudden change when the egg passes from one medium to the other, might, by analogy with nerve, cause rhythmical changes in the vitelline membrane.

The experiments to find out if the restoration of Ca++ in the external medium has any inhibitory effect on the impedance cycle were carried out with M/25-CaCla in tap water round the egg, instead of the usual tap water. The conductivity controls were carried out as described inthe previous section on the effect of NaCN. The experiments showed that M/25-CaCl2 in tap water has no effect on the rhythmical changes.

The effect of distilled water

The eggs were placed in a special perfusion conductivity cell in which laboratory-distilled water was continuously led in at the bottom and siphoned off at the top. This system ensures that the egg is in contact with distilled water; but because of the effect of the chorion in hindering diffusion some time must elapse before the vitelline membrane is in actual contact with the new medium. Since, however, the conductivity of the system varies continually while diffusion of distilled water into the perivitelline space and diffusion of tap water out into the external medium is going on, it is easy to tell when the egg is in a pure distilled water environment. Distilled water has no effect on the rhythmical changes ; an egg was left in it for 48 hr. and was unaffected by this treatment.

There is a possibility that the increase in conductivity of the distilled water round the egg at the beginning of the experiment is not entirely due to the outward diffusion of tap water from the perivitelline space, but to exosmosis from the egg proper. This possibility has not been investigated as it is not relevant to the experiments described in this paper.

The effect of chloretone

Phenyl urethane reversibly abolishes the impedance change (Rothschild, 1940). The experiments were carried out with relatively high concentrations of this substance (6 × 10−4M), and its inhibitory action can therefore be attributed to a general ‘narcotic ‘effect. The reversibility of the inhibition suggests that at this concentration the narcotic inhibits some dehydrogenase system in the cell by non-specific adsorption. Experiments with 0·001 M-chloretone were carried out to see if the flavoprotein theory of inhibition (Quastel, 1943) had any relevance in the case of the metabolism associated with the rhythmical changes. The results were negative, although the cyanide experiments suggest that the cytochrome system is concerned in these oscillations.

Volume changes

Estimates of egg volumes were obtained by making planimeter measurements of egg areas on photographic enlargements. The measurements were accurate to about 2 %. No rhythmical changes in area, nor therefore in volume, were observed.

Fine structure of the impedance cycle

The impedance changes can roughly be divided into two classes: first, a very regular type with a characteristic amplitude modulation every other cycle (Pl. 10 a); and secondly, a much less regular type (Pl. 12 a), in which the equivalent parallel resistance and capacitance do not go through balance at the same instant in time (i.e. they are out of phase), but in which again there is a typical impedance pattern. Previously (Rothschild, 1947 a, b), the variation in impedance has only been resolved into its equivalent maximum and minimum resistive and reactive components. By use of the ‘ellipse’ technique the analysis can be extended to cover the whole cycle, and though it has not so far been possible to correlate the results of such analyses with the rhythmical egg movement, it seems worth while to record briefly the ‘electrical’ results of such experiments. Pl. 12 a shows one of the second type of impedance variations, while from the ellipse traces of this impedance change, shown in Pl. 12 b; it can be seen that the equivalent parallel resistance and capacitance are out of phase. Resistance balance occurs when the major ellipse axis is vertical, capacitance balance when the minor ellipse axis has its minimum width. If the resistance and capacitance changes are in phase, the ellipse degenerates into a line when the major ellipse axis is vertical; this is evidently not the case. The values of the equivalent parallel resistance and capacitance at any time in the cycle can be deduced from the ellipse traces, and these can be converted into the equivalent series resistance and reactance form, from which the familiar circle diagram can be set up. Text-fig. 9b and d show the variation with time of the locus of the co-ordinate Rs, jXs, at one frequency only, for regular and irregular types of impedance variation. These diagrams are, in fact, small parts of circle diagrams drawn on a very large scale (c. × 125). Even in Text-fig. 9b there is a very slight suggestion of a-resistance change in the system, which is not in phase with the main capacitance change, and is more marked in the irregular type (Text-fig. 9 d). It is interesting that the main electrical difference between the two types lies in the degree to which the capacitance and resistance changes are out of phase, which can also be seen in the graphs showing the variation of the equivalent parallel resistance and capacitance with time, in Text-fig. 9.* Geometrically, the difference in the wave-form of the two types of impedance may be connected with the symmetry or asymmetry of the precession with respect to the two electrodes. This is discussed in more detail later.

Text-fig. 9.

a. Normal dynamic circle diagram for regular type of impedance change ; R0, 3487 Ω. R, 2115 Ω. Xs at the characteristic frequency, 633 Ω. Frequencies in kilocycles, b. The same as a, but enlarged 125 times. The broken line is that part of the arc of the circle surrounded by a dotted line in a. In this time impedance locus at 1000 cycles, the points on the curve are 7·5 sec. apart. c. Variations in Rp and Cp (the resistance and capacitance equivalent to the egg in the conductivity cell) with time, from which b is obtained. Note that Rp and Cp, reach their maxima and minima at the same time. The ordinates represent the tilt of the ellipse in the case of RP and the width of the ellipse in the case of Cp. d. Time impedance locus for the irregular type of impedance change shown in Pl. 12. e. Rp and CP variations from which d is obtained. Note different phase relationships.

Text-fig. 9.

a. Normal dynamic circle diagram for regular type of impedance change ; R0, 3487 Ω. R, 2115 Ω. Xs at the characteristic frequency, 633 Ω. Frequencies in kilocycles, b. The same as a, but enlarged 125 times. The broken line is that part of the arc of the circle surrounded by a dotted line in a. In this time impedance locus at 1000 cycles, the points on the curve are 7·5 sec. apart. c. Variations in Rp and Cp (the resistance and capacitance equivalent to the egg in the conductivity cell) with time, from which b is obtained. Note that Rp and Cp, reach their maxima and minima at the same time. The ordinates represent the tilt of the ellipse in the case of RP and the width of the ellipse in the case of Cp. d. Time impedance locus for the irregular type of impedance change shown in Pl. 12. e. Rp and CP variations from which d is obtained. Note different phase relationships.

Correlation between egg movements and impedance changes

The egg can be considered as a sphere with uniform electrical properties, except for a cap, composed of the blastodisc and oil droplets, which has a different capacitance or dielectric constant from the rest of the egg. The vitelline membrane has a high resistance, which has not so far been measured, and a capacitance of the same order as that found in other biological material. The internal resistance of the egg is between 100 and 200 Ω cm. (Gray, 1932; Rothschild, 1946). When the egg is undergoing the rhythmical movements, the cap describes an approximately circular or elliptical path.* At any given moment the whole system has certain electrical properties, and is equivalent to some particular resistance and capacitance in parallel, at a particular frequency. Suppose that the electrodes are in the lateral position, and that the movement of the cap is exactly symmetrical with respect to the two electrodes; further, that the magnitude of the impedance of the system is ohms at this moment, when the cap is in the position shown in Text-fig. 10a. When the cap has reached the position shown in Text-fig. 10 b, the impedance of the system will be p + δp ohms, where δp may be positive or negative. When, however, the cap reaches the position shown in Text-fig. 10c, the impedance of the system will be p ohms again, by virtue of the symmetry of the system egg + electrodes. For the same reason, the impedance will be p + δp again when the cap is in the position shown in Text-fig. 10d, and of course will be p when the cap returns to its original position, having moved through one complete cycle. In these circumstances, as can be seen in Text-fig. 10, the result is that the impedance oscillation has double the frequency of the egg oscillation. If, however, the track of the cap is not exactly symmetrical with respect to both electrodes, the impedance records should reflect the frequency of the egg oscillations and this is what actually happens. The frequency of the amplitude modulation, which is half the apparent frequency of the impedance cycle, is equal to the frequency of one complete egg oscillation.

Text-fig. 10.

Theoretical variations in impedance during precession of the egg. The egg drawings are purely diagrammatic. Ordinates, impedance (magnitude); abscissae, time.

Text-fig. 10.

Theoretical variations in impedance during precession of the egg. The egg drawings are purely diagrammatic. Ordinates, impedance (magnitude); abscissae, time.

When the electrodes are in the vertical plane, the cap is never symmetrically placed with respect to both electrodes. It alters its position relative to the electrode above the north pole of the egg, but its position relative to the south pole remains nearly constant. Consequently, the frequency doubling effect is very much reduced ; and the impedance records correspond in frequency to that of the egg oscillation and are of smaller amplitude. As might be expected, there is a very slight effect due to the variation in the position of the cap relative to the south pole electrode. The possibility of the amplitude modulation (lateral electrodes) or the small ‘harmonics’ (vertical electrodes) being due to electrical changes in the vitelline membrane is discussed later.

The movement of the egg should not cause any change in the resistance of the system if Maxwell’s equation for the resistance of a suspension of spheres is obeyed. This equation is
formula

Where r1 = resistivity of external medium, r =resistivity of whole system, =resistivity of egg, ρ = volume concentration.

The position of the egg in the measuring apparatus does not come into this equation, and egg movements should not therefore cause any change in the resistance of the system unless the egg is heterogeneous with respect to its resistive properties. The existence of the polar cap of oil droplets suggests that if the egg (and therefore the cap) moves, the capacitance of the system will vary—and this is confirmed by the experiments.

Final proof that this interpretation is correct could be obtained by taking an egg which is not exhibiting any impedance cycle and moving it and the electrodes in such a way that the variation in orientation of the egg relative to the electrodes is similar to that which takes place when the electrodes are fixed in space and the egg precesses. The apparatus shown in Text-fig. 3 was designed for this purpose, but experiments with it, though tending to confirm this interpretation, were not entirely successful. Superficially it may seem that this method of confirming the electrical effects of the egg precession is rather roundabout, and that the same result could be achieved by keeping the electrodes fixed and artificially making the egg precess. This is impossible because of the property the egg has of always orienting itself so that the blastodisc and the oil droplets are at the north pole. For example, if an egg in its chorion is turned upside down so that the blastodisc is at the south pole, rotation occurs at once and the blastodisc comes back to the north pole. This phenomenon, which is common to many fish eggs, has caused certain authors to liken fish eggs to ‘Stehaufmannchen’ (Kuhl, 1939).

In a previous paper (Rothschild, 1946), it was shown experimentally that Maxwell’s equation does not hold exactly for a system of this type, and it can also be shown on theoretical grounds that this is so. Correspondingly, it is found that very small changes in resistance do occur when an egg is moved about in a conductivity cell. Possibly the small changes in resistance that were described in the ‘Results’ section under ‘Fine structure of the impedance cycle’ (in the case of the regular type of impedance change) may in part be due to such deviations from Maxwell’s equation.

The question arises as to whether the recorded impedance changes are entirely due to the ‘electro-geometrical’ changes discussed above, or to electrical changes in the vitelline membrane itself. The capacitance change is almost certainly in the first category. It is possible that the small resistance changes in Text-fig. 9 might be due to resistance changes in the vitelline membrane, but there is negligible evidence in support of this possibility. If the movements of the egg are due to some contractile property of the vitelline membrane, and if the contraction of the vitelline membrane is associated with changes in its resistance, the concomitant impedance changes might be expected to occur at the beginning of each impedance cycle, or at any rate not all the way through the egg cycle. This is exactly what happens, if we care to interpret the impedance changes shown in Pl. 10 in the appropriate way. Superimposed on the impedance changes due to the dielectric heterogeneity of the egg surface, there is, according to this view, some other change which increases the amplitude of the impedance variation; the effect has a significantly smaller duration than the whole cycle. I do not think that this complicated interpretation is justified at present. The simplest hypothesis is that the impedance changes are caused by the dielectric heterogeneity and movement of the egg, and that the amplitude modulations are due to the precession not being symmetrical with respect to both electrodes ; but it is difficult to establish this with complete certainty.

Causes of the precession

In other fish eggs; similar movements have been correlated with waves of contraction in the vitelline membrane, passing over the whole egg (Ransom, 1867; Yamamoto, 1933). No such contraction waves have as yet been observed in unfertilized trout eggs. The movements must presumably be due to periodic changes in the centre of gravity of the egg, of a peculiarly regular nature.

Metabolism

Investigations into the metabolism of the egg during the rhythmical changes are made somewhat difficult by the impermeability of the vitelline membrane and even perhaps by the large amount of yolk within the vitelline membrane, in which diffusion may be slow. The fact that cyanide reversibly reduces the frequency and amplitude of the changes, and in sufficiently high concentrations inhibits it altogether, indicates that the cytochrome system is involved; but we cannot conclude, as has been done for analogous reasons in the case of ciliary movement (Gray, 1928), that because cyanide exerts its effect slowly, the cytochrome system is involved in a resynthesis stage in the metabolic cycle and not in the breakdown of the energyproducing compound. The same difficulties would apply to the interpretation of experiments carried out in anaerobic conditions, because of the possibility of oxygen reserves in the yolk. The experiments with phlorizin and NaF, which can be interpreted as indicating that the changes are not associated with any glycolytic metabolism, are in the same category.

There are two ways in which the egg movements seem to differ from ciliary movement. First, although a decrease in the pH of the external medium has a marked and immediate effect on the oscillations, recovery sets in later. The influence of pH changes has not been examined in detail, and such experiments would be of some interest. The experiments carried out at low pH’s were a byproduct of those with sodium azide. Secondly, the presence or absence of Ca++ in the external medium is without effect. This might suggest that calcium does not play a part in the reaction responsible for the egg movements; but this seems unlikely as calcium lack (i.e. distilled water) has a marked effect on the eggs, making them much more sensitive to mechanical shock than when in the normal environment. Equally, M/25-CaCl2 in tap water has no inhibitory effect, which disposes of the possibility that the changes are comparable to those that occur in nerves when the Ca++ concentration of the medium is reduced.

The origin and significance of these egg movements may be elucidated by studying the effects of other chemicals on them. The influence of veratrine, narcotics at different concentrations, CO in the dark and light, Mg++, K+, and H+ would be of interest.

Temperature

These experiments were carried out to obtain data for comparison with other rhythmical biological phenomena, rather than to try to interpret the mechanism of the changes by attaching some significance to the absolute value of the Q10 or temperature characteristic. There is a marked similarity between the temperature characteristic of the egg oscillation and that of ciliary movement. This is evident if we compare the results described in this paper with those obtained and described by Gray (1928). Crozier’s description of Gray’s data on the temperature characteristics of ciliary movement and their O2 consumption gives a somewhat different picture, however, as he states that there is a significant ‘break’ at 15°, above which μ= 11,800 and below which μ= 16,700 (Crozier, 1924a). From this it might be concluded that the mechanism responsible for the variation in the frequency of egg oscillations with temperature is different from that responsible for the variations in ciliary movement with temperature, because no break occurs in the former at 15°. Such a conclusion is hardly justifiable when it is remembered that Gray found no break at 15° and drew a straight line, with a correspondingly smaller slope, through the points obtained at different temperatures.

A convenient table for comparison of the μvalues for different biological processes is to be found in the Appendix, ‘Index to temperature characteristics’, in Barnes’s Textbook of General Physiology (1937). In this table protoplasmic streaming (cyclosis) is said to have a temperature characteristic of ‘4700–10,300’, amoeboid movement 10,800, and the pulsation of smooth muscle in rabbit intestine 8000 and 16,150. The pulsation of contractile vacuoles can be shown to have a mean value of 16,400. Trout-egg development is said to have μ equal to 24,500, while the first cleavage of the frog’s egg has μ equal to 10,800 or 21,900. From these data one might conclude that it is unwise to dogmatize about the nature of some biological process solely from a consideration of the variation in velocity of the process with temperature. Yet Yamamoto (1931, p. 161), who investigated what are undoubtedly similar movements in the fertilized egg of the Japanese medaka Oryzias latipes, and found temperature characteristics of 10,000–11,000, 21,000–22,000 and 27,000–29,000, between 7 and 39°C., concludes that the movements ‘unquestionably differ from those vital activities known to be associated with oxidative processes… and that the temperature characteristics ‘agree well with those heretofore obtained for the rate of development or growth phenomena’; while, as mentioned in the Introduction, Needham implies that these changes are in some way a result of fertilization and that their function is to stir up the yolk for the benefit of the embryo. These views are not easy to reconcile with the fact that the oscillations occur in unfertilized eggs, unless we postulate that all unfertilized fish eggs sustain a parthenogenetic stimulus when placed in fresh water. I have tentatively suggested that this may be the case (Rothschild, 1935; Hubbard & Rothschild, 1939), and Trifonowa’s experiments (1934) show that the eggs of fresh-water fish are somewhat liable to be parthenogenetically activated—though abortively—when placed in fresh water. There is some evidence, which I hope to publish elsewhere, that this is so in the case of frogs’ eggs; but the data on temperature characteristic recorded in this paper neither support nor controvert this possibility in the case of the eggs of trout.

There is a further point about these egg movements which should be mentioned. It has been tacitly assumed that the whole egg precesses, or that after being turned upside down the whole egg rotates until the blastodisc is at the north pole. The same applies to the frog’s egg, which rotates within the perivitelline space so that the white pole is at the bottom of the egg. If the egg as a whole did not rotate or precess, but only the vitelline membrane moved, the optical results would be the same, and it is not easy to distinguish between these alternatives. Ancel & Vintemberger (1933) have suggested that something of this sort happens in the frog’s egg.

Ancel
,
P.
&
Vintemberger
,
P.
(
1933
).
C. R. Soc. Biol
.,
Paris
,
114
,
1035
.
Arvanitaki
,
A.
(
1939
).
Arch. int. Physiol
.
49
,
209
.
Barnes
,
T. C.
(
1937
).
Textbook of General Physiology
.
Philadelphia
.
Cole
,
K. S.
(
1928
).
J. Gen. Physiol
.
12
,
37
.
Cole
,
K. S.
(
1935
).
J. Gen. Physiol
.
18
,
877
.
Cole
,
K. S.
(
1938
).
Nature, Lond
.,
141
,
79
.
Cole
,
K. S.
&
Curtis
,
H. J.
(
1938
).
J. Gen. Physiol
.
22
,
37
.
Cole
,
K. S.
&
Curtis
,
H. J.
(
1939
).
J. Gen. Physiol
.
22
,
649
.
Cole
,
K. S.
&
Marmont
,
C.
(
1942
).
Fed. Proc
.
1
,
15
.
Crozier
,
W. J.
(
1924a
).
J. Gen. Physiol
.
7
,
123
.
Crozier
,
W. J.
(
1924b
).
J. Gen. Physiol
.
7
,
189
.
Fricke
,
H.
&
Curtis
,
H. J.
(
1934
).
J. Gen. Physiol
.
18
,
821
.
Gray
,
J.
(
1928
).
Ciliary Movement
.
Cambridge
.
Gray
,
J.
(
1932
).
J. Exp. Biol
.
9
,
277
.
Hodgkin
,
A. L.
&
Huxley
,
A. F.
(
1946
).
Nature, Lond
.,
158
,
376
.
Hubbard
M. J.
&
Rothschild
,
LORD
(
1939
).
Proc. Roy. Soc. B
,
127
,
510
.
Keilin
,
D.
(
1936
).
Proc. Roy. Soc. B
,
121
,
165
.
Kuhl
,
W.
(
1939
).
Z. wiss. Zool
.
152
,
305
.
Needham
,
J.
(
1942
).
Biochemistry and Morphogenesis
.
Cambridge
.
Quastel
,
J. H.
(
1943
).
Trans. Faraday Soc
.
39
,
1
.
Ransom
,
W. H.
(
1867
).
Philos. Trans
.
157
,
431
.
Rothschild
,
N. M. V.
(
1935
).
Dissertation
,
Trinity College
,
Cambridge
.
Rothschild
,
LORD
(
1940
).
Nature, Lond
.,
145
,
744
.
Rothschild
,
LORD
(
1946
).
J. Exp. Biol
.
23
,
77
.
Rothschild
,
LORD
(
1947a
).
Nature, Lond
.,
159
,
134
.
Rothschild
,
LORD
(
1947b
).
J. Exp. Biol
.
23
,
267
.
Rusconi
,
M.
(
1840
).
Arch. Anat. Physiol. Heft
11
,
185
.
Trifonowa
,
A.
(
1934
).
Acta zool., Stockh
.,
15
,
183
.
Wintrebert
,
P.
,
Painlevé
,
J.
&
YUNG
Ko-Ching
(
1929
).
C. R. Acad. Sri
.,
Paris
,
189
,
208
.
Yamamoto
,
T.
(
1931
).
J. Fac. Sri. Imp. Univ. Tokyo, Sec. iv, Zool
.
2
(
part 3
),
153
.
Yamamoto
,
T.
(
1933
).
J. Fac. Sri. Imp. Univ. Tokyo, Sec. iv, Zool
.
3
(
part 2
),
111
.
Yamamoto
,
T.
(
1934
).
J. Fac. Sri. Imp. Univ. Tokyo, Sec. iv, Zool
.
3
,
275
.
Yamamoto
,
T.
(
1940
).
Annot. zool. jap
.
19
,
69
.

PLATE 9

Photographs, at intervals of 49 sec., of one complete egg cycle. Note that in ii and v, though the egg appears to be; in the same position relative to the chorion, the oil globules under the proximal egg surface are in focus in ii, but out of focus in v. This means that if the camera had been in the x-z, instead of the y-z plane, the egg would not have been in the same position relative to the chorion in ii and v.

PLATE 10

a. Normal impedance change. T ° C. 16 ·8. Time-marker, minutes. The dotted line shows one ‘complete ‘impedance change, b. Measurements of impedance in different planes relative to the egg. T ° C. 16 ·8. Time-marker, minutes, i, electrodes in normal position; ii, electrodes in vertical position, c. Effect of current interruptions. T °C. 16 ·8. No interruption, time-marker, minutes; interruption lasting 5 sec.; interruption lasting 10 sec. d. Effect of pH. i, in tap water, pH 7 ·3; ii, after 6 min. in tap water, pH 6·3. T° C. 16·8. Time-marker, minutes.

PLATE II

Effect of sodium cyanide (M/1000 in tap water). C. 16·8. Time-marker, minutes, i, before treatment; ii-v, 28, 78, 170 and 225 min. after immersion; vi, vii, recovery in tap water, after 3 and 15 hr.

PLATE 12

a. Irregular type of impedance change: time-marker, minutes. T° C. 16·8. b. Ellipse traces for a showing Rp (tilt) and Cp (width) out of phase.

PLATE 9

ROTHSCHILD—RHYTHMICAL IMPEDANCE CHANGES IN THE EGG OF THE TROUT

PLATE 9

ROTHSCHILD—RHYTHMICAL IMPEDANCE CHANGES IN THE EGG OF THE TROUT

PLATE 10

ROTHSCHILD—RHYTHMICAL IMPEDANCE CHANGES IN THE EGG OF THE TROUT

PLATE 10

ROTHSCHILD—RHYTHMICAL IMPEDANCE CHANGES IN THE EGG OF THE TROUT

PLATE 11

ROTHSCHILD—RHYTHMICAL IMPEDANCE CHANGES IN THEN EGG OF THE TROUT

PLATE 11

ROTHSCHILD—RHYTHMICAL IMPEDANCE CHANGES IN THEN EGG OF THE TROUT

PLATE 12

ROTHSCHILD—RHYTHMICAL IMPEDANCE CHANGES IN THEN EGG OF THE TROUT

PLATE 12

ROTHSCHILD—RHYTHMICAL IMPEDANCE CHANGES IN THEN EGG OF THE TROUT

*

Alternative interpretations of the variations in the length of the ‘loop ‘minor axis are possible. Analytically, the variation in loop area and tilt with the amplitude and phase relationships of R and C, when the latter are undergoing S.H.M., is somewhat complicated. As this is a purely electrical problem, it is not considered in this paper.

*

Actually, the path is not circular or elliptical in the ordin ary sense, because the movement is not in one plane. The movement does not take place on the surface of the sphere (the egg), but on the surface of an oblate spheroid, which is the surface enclosing the precessing sphere. What is seen through the microscope is the plane projection of the movement on this surface. A circular orbit will appear elliptical in these circumstances.