The Permeability of the Shell of the Egg of Teleogryllus Commodus Measured with the Aid of Tritiated Water

The eggs of Teleogryllus commodus, like those of many other species of insects, absorb water during their development. The water enters through the shell at a specific stage in the egg’s development, and there is an as yet unresolved problem as to how the control of the flow of water is effected (Browning, 1967). Recently it has been shown that the isolated egg-shell of Locusta migratoria migratorioides is permeable to water at all stages in the egg’s development, and the same is probably true of T. commodus (Browning, 1968.) The objections raised by McFarlane (1966) to earlier work by Browning & Forrest (1960), using deuterated water to assess permeability, are therefore probably not valid. Nevertheless Browning & Forrest (1960), for technical reasons, were unable to separate the water held in the shell from that which had penetrated to the yolk and embryo, and as McFarlane (1966) has pointed out, this limits the conclusions that can be drawn from their work. The present experiment, using tritiated water, which can be detected in much lower concentrations than was possible with deuterated water and the equipment then available, aims to obtain information on the following points: (1) the relative permeability of the shell at times when little or no net flow of water is occurring, compared with the permeability during the period of rapid water uptake; (2) the effect of the metabolic rate of the egg on permeability; and (3) the relative quantities of water entering the shell and entering the contents of the egg (using the word ‘contents’ to include the embryo, embryonic fluids and the yolk).

Eggs were obtained overnight from a large culture of crickets and placed at 12° C. for 3 weeks, so that they would complete their diapause and develop promptly and uniformly when incubated at 30° C. These eggs were then placed at 30° C. and each day for the first 5 days five samples of 20 eggs each were taken from them. The time spent at 30° C. before the samples were taken will be referred to as the ‘pre-incubation period’. Each sample was put in a Conway vessel on filter paper, and the paper was moistened with enough tritium oxide solution (specific activity 100μcml.^-1) to wet the eggs and form a meniscus on their sides, without completely immersing them. The lids were sealed on the vessels with grease and one vessel was placed at each of five temperatures, namely 12, 17, 21, 28 and 30° C. Prior to the eggs being placed in them, the vessels had been standing at their respective temperatures overnight, and the eggs were transferred rapidly, so that they should reach the temperature of the incubator as quickly as possible.

The temperatures at which the eggs remained while they were in contact with the tritiated water were chosen to span the temperatures of 30° C., at which the eggs develop rapidly, and 12° C., at which no appreciable morphogenesis occurs. On the third day (that is, after 2 days pre-incubation at 30° C.) the eggs had begun to absorb water, and they continued to increase in size and weight while they were in contact with tritiated water at the higher temperatures, whereas there was no evidence of continued water absorption at the lowest temperature (cf. weights of eggs given in Table 1). Thus on the third day, at 30° C. the measurements represent the flow of ³H₂O^* entering the egg along with a net influx of water, whereas at 12° C. they represent the flow of ³H₂O through a shell that was in the condition to absorb water, but through which no net flow was occurring.

After the eggs had remained for 1, 6 and 24 hr. in contact with tritiated water, a sample of five eggs was taken from each dish. The eggs were dried quickly on filter paper, weighed and plunged into individual baths of liquid paraffin. Each egg was then cut in half with scissors under the paraffin and the contents of the shell were sucked up into a fine pipette. The contents of the pipette were ejected into 8 ml. of scintillation fluid in a vial, the end of the pipette was rinsed several times in scintillation fluid by sucking it up and down, and finally the end of the pipette was crushed against the bottom of the vial, so that any material adhering to the glass remained in the vial. The two pieces of the shell were transferred to another vial of scintillation fluid with forceps. The vials from 1 day’s observations were usually retained together; and the activity of the 150 vials, together with the necessary blanks and standards, was estimated in a Packard automatic scintillation spectrometer. Preliminary observations had shown that no appreciable change in the apparent activity occurred when vials were allowed to stand on the bench from between 10 min. and 2 days. No quenching effect of the small quantities of paraffin used could be detected.

The contents of the eggs sometimes spilled as the eggs were cut in half. The yolk, especially of young eggs, tended to spread on the glass of the vessel, under the paraffin, and it was difficult to collect it all. Another problem encountered with young eggs was that the shell sometimes crumpled as the contents were sucked out. It was then not possible to suck up the whole contents; some remained inside the shell. If the problem was severe, the egg was rejected and a new one obtained, or a blank left in the results. These difficulties no doubt accounted for some of the variability in the results. There is also reason to suppose that some eggs had damaged shells, although each egg was inspected when it was weighed and obviously damaged ones rejected.

Table 1 sets out the average quantity of tritiated water of specific activity 100 μcml.^-1 recovered from both the contents and shells of the eggs on each of the 5 days and at each sampling time. The table also includes the mean fresh weight of the group of five eggs from which the results were obtained. It is clear that results were highly variable, but a statistical analysis of all the figures showed highly significant differences between temperatures, sampling times and days of incubation. The results for contents and shells were analysed separately because it was clear from the outset that the contents contained significantly more radioactivity than did the shells under all circumstances, on the average over five times as much. There is no doubt that the tritium oxide penetrated rapidly through the shells and entered the contents of the eggs.

The results obtained after the eggs had been in ³H₂O for only 1 hr. were more variable than those obtained after longer exposures. This is also true of the ratio between the quantity of water that had entered the contents and the quantity held in the shell. These ratios, calculated for each sampling time are shown in Table 1, and it is clear that the ratio became fairly stable after 6 hr. immersion and the variability, as expressed by the standard error, fell considerably. This probably reflects the variability in the shells of different eggs in their ability to hinder the passage of water into and out of the egg, which may be expected to have a greater influence during the initial period.

During the first 2 days the total quantity of water that penetrated the eggs was small. The eggs were also small (about 0·65 mg.), and there was little change in weight. On the third day, however (that is, after 2 days pre-incubation), the eggs had all begun to absorb water (average weight about 0·85 mg.), and the total quantity of tritiated water absorbed subsequently was greater. The eggs treated at 30 and 25° C. completed their water absorption during the 24 hr. period, but in those treated at lower temperatures, and in which the rate of water absorption was appreciably slower (no change in weight could be detected at 17 or 12° C.), much less radioactivity was found in the eggs. Nevertheless, even the eggs that had gained no weight while exposed to tritiated water had still gained appreciable radioactivity. On the last 2 days of the experiment, when the eggs had already completed their water absorption during the period of pre-incubation at 30° C., large quantities of ³H₂O were found in all eggs, and again the quantity detected depended on the temperature at which the eggs were held in tritiated water.

The figures in Table 1 are simply the gross quantities of ³H₂O recovered, calculated from the counts obtained, and give little information on the permeability of the sheD, because its area changes considerably during the period of water absorption. In Table 2 the flow-rates through the shell are shown. These figures have been calculated from the quantitity of ³H₂O recovered from the contents of the eggs after each timeinterval and take into account the area of the shell of each egg (see Appendix). In general the rates of flow were high during the first 6 hr., but fell considerably when the whole 24 hr. period was considered, and as the concentration of ³H₂O inside the shell became more nearly equal to that outside (see Table 3). There were no significant differences detectable between the over-all flow-rates into eggs that had not been preincubated compared with those pre-incubated for only 1 day, nor among those preincubated for 2, 3 or 4 days, but the rates after the short pre-incubation times were significantly lower (P = 0-05) than those after the longer times. It may be concluded that the shell is rather less permeable during the first two days than it is subsequently.

Table 3 shows the proportion of the total water in the egg that was replaced by exchange from the outside during each of the sampling periods. These values are for the whole egg, because it was not found practicable to estimate the proportion of water in the shell and its contents separately. To estimate the proportion of water in whole eggs, the eggs were weighed individually on a torsion balance of 2 mg. capacity, reading to o-ooi mg., dried for 24 hr. at 105° C., and weighed again quickly on the same balance. The weight of water in each egg was then plotted against its fresh weight and a line was fitted through the points by eye. Very little variation was found about this line, which represented 250 eggs whose fresh weights varied from 0·52-1·-40 mg., so that little error was involved in reading from this graph the weight of water contained in an egg whose fresh weight was known. Taking the weight of each egg, and the quantity of ³H₂O in the whole egg, the proportion of water in the egg that had been replaced by the bathing solution was calculated and the results are summarized in Table 3. Only when the eggs had been pre-incubated for 2 days and were in the condition to absorb water, and only when placed in tritiated water at the higher temperatures at which water absorption actually occurred, was almost all the water in the egg replaced in 24 hr. Otherwise, only about half the water at the higher temperatures, and much less at the lower temperatures, was replaced during the 24 hr. of contact with ³H₂O.

The results in Table 1 show that ³H₂O penetrates the shell of the egg of Teleogryllus commodus rapidly, a small proportion remaining in the shell, but most accumulating in the internal tissues. The isolated egg-shell has been shown to be permeable to water when it is used as an osmometer membrane (Browning, 1968), and it thus appears likely that, in this experiment, much of the water penetrated the shell by mass flow, rather than by ionic exchange, as was postulated by McFarlane (1966).

In the section on Methods the great variability found within groups of eggs treated similarly, and between groups, was mentioned, and some of the probable reasons for this were stated. However, it is possible that some of the variability is biological, rather than due to experimental technique, and may reflect the variable structure of the shells of different eggs. Phillips & Dockrill (1968) found a similar variability in the permeability to larger molecules of a cuticular structure in Schistocerca gregaria. Such variability may appear more likely if the function of the shell is primarily to provide a strong container for the tissues within, rather than to act as an impermeable barrier from the environment This question will be discussed later.

A marked difference was found in the amount of water that entered the eggs at different temperatures. Over-all the average temperature coefficient relating the flowrate to temperature (Q₁₀) was about 3 when allowance is made for highly aberrant values (for example after 1 hr. on day 3 in Table 2). It is perhaps unlikely that such a high value would be due simply to the increased rate of osmotic flow as the temperature increased; rather, some metabolic phenomenon seems to be implicated. All eggs on any day were at the same stage of development when they were placed in ³H₂O, all having spent the same time developing at 30° C. However, those that were placed in ³H₂O at 30° C. continued their development without interruption, whereas the development of all the others was retarded at the lower temperatures. For example at 30° C. the egg requires 2 days to reach the stage at which water is absorbed, and absorption occurs in about 12 hr., whereas at 12° C. over 30 days are required to reach the stage of water absorption and absorption occupies another 12 days. Thus the general metabolism of the eggs is greatly slowed down at lower temperatures, and the concurrent slowing in the rate of flow of water through the shell may well be due to the rate at which water, once inside the shell, is transported through the egg. Such transport can be thought of as a kind of biological stirring mechanism, which would operate much more slowly at 12 than at 30° C.

At 30° C. the rate of transfer through the shell, as measured by the amount of tritiated water taken up by the egg, is sufficient to allow the egg to absorb all the water it requires during morphogenesis in about 12 hr. Even at the lowest rate of flow observed during the period when the eggs had begun to absorb water (i.e. at 12° C. on the third day), the amount of water required could flow through the shell in less than 3 days, whereas some 12 days are needed for complete absorption at this temperature. At least at the lower temperatures, some mechanism other than the permeability of the shell must be restricting the entry of water.

It has been shown that anoxia during the period of water absorption stops the further flow of water into the egg of crickets (Browning, 1965; McFarlane & Kennard, 1960). Lack of oxygen would not be expected to interfere with osmotic flow, but it would be expected to stop the growth of the egg. Both McFarlane (1963) and Browning (1967) have shown that a profound change occurs in the endochorion of the shell just before water absorption begins. If this change is thought of as a growth process, causing the shell to become greater in area (rather than as signalling only a change in permeability, as McFarlane thought), an increased flow of water into the egg would be expected, and lack of oxygen would be expected to prevent continued absorption. Both of these conditions are observed. This may not be a general explanation for water absorption, since some eggs probably continue to absorb water even under anoxic conditions (Dr A. D. Lees, personal communication; T. O. Browning, unpublished observations).

If we can conclude that the shell of the egg is indeed permeable to water at all stages during its development, the question arises why the egg does not absorb water during the first few days of its development, nor during the period it spends at low temperature prior to incubation. During the early part of development the osmotic potential of the solution within the egg is likely to be greater than it is after water has been absorbed and dilution occurs (Grellet, 1966; Laughlin, 1957). It has been shown that when the shell of the young egg is made into a balloon partly filled with a solution of glucose equivalent in concentration to the solution within the egg, and the balloon is immersed in water, it absorbs water but the shell does not stretch or split (Browning, 1968). Under these circumstances a sufficient hydrostatic pressure must build up within the shell to oppose the net entry of water by osmosis. Perhaps a similar mechanism operates in the living egg.

A large group of eggs was incubated at 30° C. and from these 10 eggs were taken each day. Each egg was weighed and placed on a glass slide and its outline was traced using a camera lucida and a low-power compound microscope with reflected light. The image of a graticule 2 mm. in length and divided into 200 parts was then traced on to a strip of celluloid and this was used to measure the egg using its outline. Measurements were made to the nearest o-oi mm. The method was as follows. A line representing the longest axis of the egg was ruled on each tracing and, assuming the egg was circular in cross section, the diameter of the egg was measured at regular intervals along the axis at right angles. The object was to cut the egg into a series of thin slices of equal thickness, each of which could be regarded as a flat cylinder, except the two end ones. One of the latter had a thickness equal to that of the slices, and the other was the residual piece after all slices of predetermined thickness had been cut. Its thickness was either equal to or less than that of all the others.

Treating all the slices except the first and last as cylinders whose height was known and whose diameter was the mean of the diameters at each end, the area of the curved surface was calculated. The two end slices were treated as segments of spheres whose depths and the diameters of whose bases were measured, and from these measurements the areas of the curved surfaces were calculated. The total area of the shell was taken as the sum of the areas of all the curved surfaces.

Probably the main source of error in this procedure lies in treating the slices as cylinders. This is particularly true near the ends where the diameter is changing rapidly. To estimate the error likely to be involved in the method, the outline of one egg was transected to give 81 slices, as many as could be accurately spaced along the axis, and the area was calculated from these. The same outline was again transected using slices 5 times and 10 times the thickness of the original one. This produced 17 and 9 slices respectively. The surface area was again calculated on these two bases. Assuming that the area calculated using 81 slices was correct, the error involved in using 17 slices was +2·4% and with only 9 slices it was +5·2%. Since the error involved in using the intermediate thickness of slice was only about 2–3 % of the best estimate obtained with five times as much work, the intermediate thickness was used to estimate the surface areas of the shells of 10 eggs on each day of incubation at 30° C.

The weight of each egg was then plotted against its calculated surface area, and a straight line was ruled through the points by eye. The variability about this line was very small. The line was then used to read off the surface area of eggs whose weight was known.

I wish to thank Miss Meredith Porter, B.Ag.Sc., for her able assistance with this work. Financial support from the Australian Research Grants Committee is gratefully acknowledged.