The Energy-Sources in Ontogenesis: II. The Uric Acid Content and the General Protein Metabolism of the Developing Avian Egg

In the preceding paper of this series (32) a report was given of experiments in which the urea present in the developing hen’s egg was estimated as accurately as possible on each day of incubation. In the experiments now to be described the uric acid has also been estimated, and it has been found possible to draw definite conclusions from the resulting figures about the utilisation of protein by the developing embryo. The primary object of these researches has been to test the truth of the hypothesis that in the course of ontogenesis there is a succession of energy-sources, carbohydrate being first made use of, then protein, and finally, fat. It was shown in the preceding paper that a large amount of the previous work on the chemistry of embryonic development fell into line and formed a connected whole if that hypothesis was adopted.

As regards the hen’s egg, which was the material used in these investigations, there was a large gap in our knowledge, for data about the nitrogenous excretory products of the embryo were very few in number. Although we possessed a certain amount of moderately exact information about the movements of fat and of carbohydrate, we were still in the dark as far as protein was concerned. Systematic estimations of the urea production had never been done, and the work of Fridericia(13) was the only systematic examination of the uric acid production. His figures were for two reasons unsatisfactory: in the first place, they did not begin until the eleventh day of development, by which time many important things might have happened, and in the second place, they were obtained by the old methods of copper and silver precipitation. These necessitated a separation of the uric acid from all the purine bases and, as has been found by many workers, such a technique is not reliable. An additional criticism is that he only used about 200 eggs, while in the experiments now reported, more than double that number have been analysed.

Two methods were made use of in estimating the uric acid in order to allow for the fact that owing to the growth of the embryo the whole scale of uric acid production is outside the total range of one method. As a micro-method for the early stages that of Benedict and Franke(3) was used, and for the later period the ammonium chloride precipitation method of Hopkins (21).

The eggs were all laid by White Leghorn hens and incubated as nearly as possible under the standard conditions suggested by Murray(31), that is to say, temperature constant at 38·8 ± 0·4° C., humidity constant at 67·5 ± 2·5 per cent., and continuous ventilation by warm air. The eggs were aired every morning for 15–20 minutes and rolled once a day.

It was found that for every stage in development from the fourth day onwards, excellent separation of white and yolk from the embryo and its membranes could be obtained by the following method. Before being opened the egg was held over a powerful electric lamp, and the area occupied by the white marked on the shell in pencil. The shell was then cracked inside the circle so formed, the crack enlarged with forceps, and the albumen allowed to flow out through the opening. The yolk was then freed from the allantois and allowed to escape in the same way. In nine cases out of ten the embryo and its membranes were left in the shell free from yolk and white. They were then tumbled out into a tared vessel and weighed when a sufficient number had been collected. The whole material was ground up with well-washed quartz sand in a mortar and extracted three times with distilled water acidified with acetic acid. During the first extraction the flask was held in a boiling water bath so that all the protein was coagulated. The separation from the protein could be made exceedingly clean by adjusting the reaction of the liquid with acetic acid to give a greenish-yellow colour with brom-cresol-purple, and when the extractions were finished all the extracts were evaporated down on a water-bath together to a low bulk. As a rule a very small quantity of protein separated out during this process but it was removed by centrifuging or filtration, and on the clear pale yellow liquid the estimation was done. In the case of the micro-method the arsenophosphotungstic acid reagent and the sodium cyanide were added to a small amount of the solution, and the resulting blue colour compared with that the of standard solution of uric acid similarly treated. The micro-method in question had been evolved in the first instance for the purpose of estimating uric acid in urine, and there was a certain danger lest the maximum reading should not represent the amount of uric acid present but simply the maximum colour obtainable with the quantities of reagents added. It was easy to overcome this difficulty, however, by doing in each case a series of readings, diluting the original solution 100 per cent, every time. If the results came to just half each other, then the highest reading could be accepted as correct, but if in any given instance the result was not halved although the strength of the solution had been halved then, obviously, the method was not recording the full value of the uric acid present.

If the macro-method was being used, then the pale yellow tissue extract was fully saturated with ammonium chloride and made alkaline with strong ammonia. This precipitated all the uric acid and also a certain amount of ammonium phosphate. After standing for three or four hours, the precipitate was filtered off, washed with saturated amrrionium chloride solution, removed with hot distilled water into a beaker, strong hydrochloric acid added, and left for the uric acid to crystallise out over night. Next morning, the uric acid was filtered off, dissolved up again in distilled water made alkaline with sodium carbonate, and titrated against standard permanganate.

The results for the uric acid are tabulated in Table I. Column 1 gives the number of days of incubation, column 2 the number of eggs used for the determination in question, and column 3 the amounts of uric acid present every day in mg. per embryo. Column 4 gives the uric acid present in mg. per cent, of wet weight of embryo, and column 5 does the same thing for dry weight of embryo. These calculations were made using the figures for wet and dry weight of White Leghorn chick embryos given by Murray (31). The data are graphically represented in Figs. 1–3. Fig. 1 shows the gradual increase in mg. per embryo of uric acid. It will be seen that the points fall on a very regular curve. In Fig. 2 is shown the mg. per embryo wet weight and here it is very significant that a plateau appears. In the first seven days of development the uric acid mg. per cent, wet weight is exceedingly small in amount, but from the seventh to the eleventh day it rises rapidly until on the twelfth day it attains a constant level which it does not leave. There is thus a specially intensive production of uric acid between the seventh and the eleventh days of incubation. It will be seen that the points are fairly close together, and the only serious divergence occurs on the eleventh day. There can be little doubt but that these differences are due to unequal rapidity of development, some embryos being more advanced than others at the beginning, as, for instance, “body-heated eggs.” The only way to avoid such errors is to take a very large number of eggs, and as far as possible that was what was done. The ascending curve is, of course, the place where the greatest divergences would be expected, if they were due to inequalities on the time-scale.

Fig. 3 gives the uric acid in mg. per cent, dry weight of embryo. The curve reaches a definite peak on the eleventh day after which it descends and seems to be reaching a steady level by the time of hatching, at about 460 mg. per cent. The relation between the results now reported for uric acid in mg. per cent, of wet weight of embryo and those which can be calculated from the data given by Fridericia(13) is shown in Table II. On the eleventh, twelfth and thirteenth days, his figures, though some 20 mg. per cent, higher than mine, yet manifest a constancy, but after that they begin to mount until the time of hatching, with a slight drop on the twentieth day. This cannot possibly be due to differences in breed or race.

The explanation seems to be afforded by the work of Le Breton and Schaeffer (26). In the course of a long and very careful research on the purine metabolism of the developing embryo, they had occasion to examine the method used by Fridericia for the separation of uric acid from purine bases and concluded that it was rather unsatisfactory. It appeared that even under the best conditions, from 6 to 34 per cent, of the total purine bases might be lost from the purine fraction and add themselves on to the uric acid fraction. Le Breton and Schaeffer accordingly made use themselves of the processes of Hopkins (21) and Ronchèse (40). But it is somewhat significant that in Fridericia’s estimations, there is found a plateau on the purine base curve beginning just exactly at the point when the uric acid curve rises and draws away from that described in this paper. It certainly seems as if the explanation of the difference in our results might lie in this direction.

In Table III are shown the smoothed curve values for uric acid in absolute mg. per embryo, mg. per cent, wet weight, and mg. per cent, dry weight. Beside them, for purposes of ready comparison, are placed the smoothed curve values for the data for urea in absolute mg. per embryo, mg. per cent, wet weight and mg. per cent, dry weight, contained in the preceding paper. Figs. 4–6 show graphically the rather striking relationships between them. In Fig. 4 is shown the urea and the uric acid in mg. per cent, wet weight of embryo plotted on the same scale. It shows the very interesting fact that on the third, fourth, fifth, sixth and seventh days of incubation, the uric acid is rising distinctly more slowly than the urea, and, indeed, columns 3 and 7 of Table III show that in absolute quantities per egg there is less uric acid than urea until the eighth day is reached. Between the seventh and the eighth day, as is seen in Fig. 4, the uric acid rises tremendously in amount and overtaking the urea almost attains its final constant value. These relationships are better seen in Fig. 5 which gives the mg. per cent, wet weight for both uric acid and urea, the abscissa being arranged so as to get them both on to the same graph. When this is done it is obvious that though both urea and uric acid rise in course of time to a constant level, the urea starts rising much earlier than the uric acid and reaches its maximum level a day or so before. Inevitably this is reflected on the mg. per cent, dry weight curve, shown in Fig. 6, only now in a still more striking way, for peaks appear, and it is seen that the urea is in advance of the uric acid by two days. It is as yet too early to speculate on the possible biological significance of this precedence of uric acid by urea in the nitrogen excretion of the developing embryo, but the fact itself seems certain.

A further point on Table III is worthy of mention. For the combined estimations of uric acid and urea, 776 eggs have been analysed, a figure which compares very favourably with the 192 eggs used by Fridericia, the 80 of Le Breton and Schaeffer, and the 476 of Murray.

Among the various deductions which can be made from the data now at our disposal, the first one may be the relative behaviour of the urea and the uric acid. In the hen, the excreted nitrogen is mostly in the form of uric acid, and the urea takes only a very small share of it. At what stage in embryonic development is the adult relationship reached? In Table IV the ratio $uric acidurea$ will be found. In column 2 it has been calculated for the absolute amounts of uric acid and urea present in the egg each day, in column 3 for the absolute mg. excreted each day per embryo. This is assuming that the absolute increment found each day in the uric acid and urea curves consists of material all of which has been excreted into the allantoic liquid. In point of fact, a very small proportion of the nitrogenous end-products might be expected to remain behind in the embryo, but this correction is probably exceedingly slight. The ratio is shown graphically in Fig. 7. It will be seen that from the fourteenth day onwards the ratio is constant at about 16, that being the adult level. Before the seventh day the value of the ratio is less than unity because more urea is present and more urea is excreted daily than uric acid. The adult ratio is therefore seen to be attained well before hatching. Another manner of expressing the relationship, which leads to a result slightly different, is shown in columns 4 and 5 of Table IV and also in Fig. 8. Here, assuming that the urea and the uric acid together make up the total nitrogen (which is certainly not absolutely true, but within 98 per cent.) the partition between them has been calculated as mg. excreted by the embryo each day of urea and uric acid nitrogen in per cent, of the total nitrogen in mg. excreted by the embryo each day. Between the fourth and fifth days, the uric acid only accounts for 9·4 per cent, of the total nitrogen, but so rapidly does the change occur that between the eighth and ninth days it accounts for as much as 90·7 per cent. It has therefore practically reached its adult level, as is shown by the comparative standards to the right of the-graph. The shaded parts at the bottom represent urea and the rest uric acid ; they are figures taken from H. Meyer (29), von Knieremto), Schimansky (44), Meissner (28) and Steudel and Kriwuscha (48).

It is clear that in the first week of development the relationship of urea to uric acid is altogether different from what holds in the adult, but that in the last week of development the adult value is rigidly adhered to. These results throw light on the finding of Aggazzotti (2) that the amniotic fluid in the chick passes from pH 7·2 to pH 4·4 in the last half of incubation, whereas before the ninth day, it has been constant at about pH 7·2.

Since the amount of urea and uric acid produced by the embryo during its development are now known, we can calculate the amount of protein combusted by the embryo.

In Table V the figures are given by means of which the calculation is made. The absolute mg. of uric acid and of urea excreted per embryo per day are given in columns 2, 3, 4 and 5, and shown also in Fig. 9. The points are seen to rise on a uniform curve, and show the same relationship to each other for the whole latter half of development. From them it is easy to calculate the mg. of urea and uric acid nitrogen excreted per day per embryo (columns 7 and 6 in Table V) the mg. of total nitrogen excreted per day per embryo (column 8 in Table V) and so, multiplying by the 6·25 factor, the mg. of protein catabolised per day per embryo (column 9 in Table V). These two last columns naturally follow a simple ascending curve and so are not plotted, but their totals give figures of importance for calculations. Then in columns 12 and 13 are given the mg. of protein broken down each day per cent, wet weight and dry weight of embryo. In order to prepare the figures of columns 12 and 13 it was necessary to have the weights for the intermediate periods between days, and these, calculated from the weight curves of Murray (31) are given in columns 10 and 11. The mg. protein combusted daily per cent, of wet weight and dry weight of embryo are plotted with abscissae arranged so as to have them in one graph, in Fig. 10. It will be seen that in both cases the peak occurs at eight and a half days of development, the significance of which fact is sufficiently indicated by the labels relating to the respiratory quotient. The curve for the dry weight declines considerably more than that for the wet weight after the peak is passed, and this is due to the well-known fact that the embryo gets drier as it grows older.

Another interesting point which arises out of Table V is the following. Sakuragiu(41), in a paper to which reference will again be made, found that the coagulable protein diminished during development from 1·846 mg. nitrogen per cent, to 1·698 mg. per cent.; 148 mg. are therefore lost per cent, of egg contents which amounts to 67 mg. per average egg, a figure showing very close agreement indeed with that seen at the bottom of column 9 in Table V, namely 68·74 mg-nitrogen excreted. Thus 100 per cent, of the coagulable protein disappearing is accounted for by nitrogenous waste-products found, and considering the variability of eggs, as well as the technical difficulties associated with work on them, this agreement is as near as can be expected. Apparently all the protein combusted is coagulable protein, probably albumen, the ovomucoid not taking part in a breakdown for purposes of energy supply. According to Bywaters (6) there is no preferential absorption of ovoalbumen and ovomucoid; they enter the embryo in the same proportions throughout development. Experiments are being continued in this laboratory to try to throw some light on the significance of ovomucoid.

There are various figures already in the literature regarding the loss of protein nitrogen during embryonic development. These are summarised in Table VI. Krogh (25) and Tangl and von Mituch(51) found that the hen’s egg lost no nitrogen whatever in gaseous form during incubation, and many workers, Tangl(51), Idzumi(22), Szneroffna (49) and Aggazzotti(1), observed that there was no alteration in the amount of total nitrogen present during development. But this was only to be expected, since the nitrogenous waste-products cannot escape, and the really significant figures are those for coagulable nitrogen. From Table VI it appears that even in widely different classes of animals the same percentage of the protein nitrogen present at the beginning of development is utilised to provide energy during that process.

Szneroffna’s ratio was calculated. It was assumed that only minute errors would be introduced by calling the urea N plus the uric acid N the nitrogen present in the allantoic fluid, and two different sets of figures for the total nitrogen in the embryo were made use of. Actually three sets were available, those of Fridericia(13), Le Breton and Schaeffer (26) and Murray (31), but only the last two were used because Fridericia’s fell on a curve, the points of which were averaged from widely divergent individual differences.

In Table VII the results are tabulated. Column 3 contains the figures of Murray for total nitrogen of embryo, column 2 those of Szneroffna herself, and column 4. those of Le Breton and Schaeffer. Column 6 contains the mg. of urea and uric acid nitrogen found by me, and column 5 the nitrogen in the allantoic fluid found by Szneroffna. It is to be noted that Szneroffna’s figures usually largely exceed mine, which would be the case had any blood escaped into the allantoic fluid during its collection. The important difference between the nitrogen figures of Le Breton and Schaeffer on the one hand, and Murray on the other, is that the former excluded the membranes in their estimations, while the latter probably included them. We therefore have a way of determining what part the membrane is playing in the protein metabolism, if any. These relations are shown in the form of a graph in Fig. 11. The newly-calculated ratio does not quite become a constant during the last ten days of incubation although it approximates to one, and it never reaches the low figure obtained by Szneroffna. The extreme smallness of the protein catabolism during the first six or seven days is reflected on this curve in the extreme height of the ratio, but just as Fridericia’s figures only began after the steady level had been reached, so Szneroffna’s missed the early descent. The more intense the protein metabolism, the lower the ratio because the greater the amount of nitrogen excreted, and so it looks as if the metabolism of protein was more intense in the absence of the membranes. From this it is perhaps legitimate to conclude that the part they play in the combustion of proteins is very slight.

Since we are now in possession of information concerning the amount of protein catabolised during the various stages of development in the hen’s egg, it would be possible, uniting the protein figures with those already in the literature for fat and for carbohydrate, to compute the respiratory quotient from purely chemical evidence. It could then be compared with the experimental respiratory quotients obtained by Bohr and Hasselbalch(5) and by Lussanna(27).

Unfortunately, in the present state of our knowledge, such a calculation can be only a rather poor approximation, and from the resulting curve no strict conclusions can be drawn. In the first place we have no accurate information as to the carbohydrate combustion. However, it is significant that Sakuragiuc finds the loss in total carbohydrate during development to be approximately equal to the loss in free glucose during the first ten days. If a curve is constructed from the data given by Pavy(35), Tomita(53), Satô (42), Bywaters(6), Idzumiua) and Sakuragiuc, averaging out all their points for the disappearance of free glucose during the first ten days of development, it is found that there is a loss of 166 mg. per egg. Now the figure for loss of total carbohydrate found by Sakuragiuc is 130 mg. per egg.

Accordingly we may conclude that of the 166 mg. of free glucose disappearing during the first ten days of development all but about 35 mg. are combusted. The difficulty arises when it is necessary to distribute this difference of 35 mg. over the first ten days. It is possible to gain some idea as to how this distribution should be done by using the lactic acid figures of Tomita(53). He found a peak of lactic acid on the fifth day, and it is significant that at its highest point it reaches 34 mg. per egg, thus almost exactly accounting for the 35 mg. glucose disappearing but not combusted. In Table VIII the average values of the six workers mentioned above have been brought together in column 2, and the daily differences between them in columns 5 and 6. In column 3 are placed the figures for the lactic acid present each day according to Tomita, and in column 7 the lactic acid accumulating each day. If this is subtracted from the glucose disappearing the corrected curve for sugar combusted is obtained and is shown in the last column. We may suppose that the lactic acid so formed is converted after the fifth day into alanine for some synthetic purpose. During this early period, however, the protein burnt is so small in amount that this correction does not affect the R.Q., so in Table X the uncorrected curve will be found.

Our knowledge of the combustion of fat is unsatisfactory. The position is experimentally a difficult one, for we have to deal with a system containing enormous amounts of fat, and yet in which the tiniest changes are of great theoretical importance. It is not surprising that the methods in use up to the present time have not succeeded in solving the problem. In Table IX are collected together the figures obtained by averaging out all the points for total fat in the egg obtained by Eaves (9), Idzumi (22), Murray (31) and Sakuragi (41). Column 2 shows the amounts oHat actually present in the egg each day during incubation, and column 4 the fat lost each day, the smoothed curve figures of which are shown in column 5. In column 6 are the figures calculated by Murray (31) from his estimations of CO₂-production, assuming that all the CO₂ produced each day was derived from the combustion of fat. These reveal a considerable divergence, and it is to be noted that from the seventh to the fourteenth day the fat lost as determined by the averaged chemical analyses, is considerably in excess of that lost as determined from the CO₂-output, even supposing that all the CO₂ was derived from fat, which is not true. The figures of Bohr and Hasselbalch (5) for CO₂-output would give an even worse divergence, for during this particular period they were lower than those of Murray. The explanation for this missing fat must be either that during that period it is used for other purposes than combustion, or perhaps more probably, that the estimations of fat are wrong. However, since we are purposing to calculate the respiratory quotient, we cannot make use of Murray’s computed fat loss curve, because it was itself derived from respiratory data; and must, therefore, adopting the chemical curve, neglect the error in question.

The fact that there do exist these considerable errors in the curves for fat and carbohydrate loss, however, make it impossible to lay any stress on the figures for percentage utilisation shown in Table X. Columns 2, 3 and 4 of that table give the combustion of foodstuffs in absolute mg. per embryo per day, and their sum the total amount of foodstuff catabolised each day in column 5. At the bottom of each of these columns there will be found its total. In columns 6, 7 and 8 the protein, fat and carbohydrate combusted each day in percentages of the total foodstuff combusted each day, are shown. For the reasons given above no significance can be attached to the exact shape of these curves. Finally, the calculated respiratory quotient is given in column 9.

Fig. 12 shows the way in which the respiratory quotient calculated from chemical analyses only, assuming carbohydrate as 1·0000, protein as 0·801and fat as 0·707, agrees with the respiratory quotient determined experimentally by Hasselbalch (19), by Bohr and Hasselbalch (5) and by Lussannau(27). The approximation is inexact but suggestive. It is interesting to note that of the high respiratory quotients of the first five days, the only point actually given by Bohr and Hasselbalch was that at 0·890 for the fourth day. In their tables, however, there were several blank spaces, and when the respiratory quotients were calculated for these with the aid of the oxygen utilisation curve of Hasselbalch (19) five other high points came to light. These are all shown on the graph.

At the beginning of development the albumen contains a certain amount of alkali reserve, which might be expected to bind CO₂ and so to depress the apparent R.Q. This has been measured by Aggazzotti (2) and by Healy and Peter(20), who obtained very divergent results in titrating to different indicator end-points. I have calculated what the correction for the R.Q. should be, on the basis of the highest figures of Healy and Peter : the resulting points are seen in Fig. 12 and show the effect of the alkali reserve to be quite small.

Since at no time does the protein reach a level of more than 3-5 per cent, of the total foodstuff catabolised, it hardly affects the respiratory quotient curve. As far as the observable respiration is concerned, the burden is borne by the fat and, to a lesser degree, by the carbohydrate.

The totals at the bottoms of columns 2, 3, 4 and 5 in Table X are of some interest. To make them comparable with other data in the literature the following corrections may be made :

View Large

From these figures two comparisons may be made. In the first place, they agree well with the results of some previous workers in the absolute loss of dry weight by the egg during its development, this corresponding more or less (since the protein is not burnt altogether away) to the foodstuff catabolised. Table XI shows the results brought together.

Secondly, it may be enquired what part of the total energy of development is provided by fat, protein and carbohydrate. In Table XII the available figures are shown, though, like those in Table XI, they are only rough. It is evident that there is no agreement among different classes of animals as to the substance from which they shall principally derive their energy during their development. And it may well turn out that some of them do not utilise a fat, a protein, or a carbohydrate, but a sterol instead.

Since the preceding paper of this series was written the hypothesis of a succession of energy-sources in embryogenesis, carbohydrate preceding protein and protein preceding fat, has received further confirmation from recent work. The points which have arisen will now be dealt with.

(1) The undoubted utilisation of carbohydrate by the hen embryo in the early stages of its development naturally leads to the question of whether embryonic tissues vary in glycolytic power with their age. Negelein(34) has recently gone into this problem and gives a curve relating the anaerobic glycolysis to the age of rat embryos. There is a high peak at 0·-47 mg. dry weight, after which time the activity steadily declines till at birth (11·0 mg. dry weight) the glycolytic power is less than a third of its former value. “Der anaerobe Glycolyse,” says Negelein, “istum so kleiner je alter der Embryo ist.” The aerobic glycolytic power also manifests a peak at 0·47 mg. dry weight, but Negelein considers that this is not a physiological phenomenon.

(2) The silkworm embryo seems to come into line with other embryos as regards early carbohydrate utilisation, according to the recent work of Pigorini(38), who estimated the glycogen in the embryos of Bombyx mori throughout their development. Tallarico(50) has brought forward evidence showing that the lipase of the hen’s egg increases very markedly in activity towards the end of incubation, especially after the ninth day.

(3) Attention was drawn in the previous paper to the fact that the calorific quotients obtained by Meyerhof (30) on the eggs of Arbacia pustulosa and Aplysia limacina did not agree with the view that carbohydrate was being utilised as an energy-source in the very earliest stages of development. His calorific quotients varied about 2·6 and he concluded that there was combustion of fat, though the theoretical figures are 3·2 for protein, 3·3 for fat and 3·5 for carbohydrate.

Very recently, Rogers and Cole (39) have made more accurate estimations of the heat-production of Arbacia eggs and have obtained higher values than any previously recorded. Their technique was modelled on the micro-calorimetric methods applied to muscle by A. V. Hill, and as the chief trouble in such work is leakage of heat, it is probable that their values are the most accurate we have. The ratio of heat-production between fertilised and unfertilised eggs they found to be the same as that of Meyerhof and Shearer (47). They themselves drew no conclusions about the calorific quotient, but clearly higher values for heat-production will bring the C.Q. more into line with the theoretical values, so it was worth while to calculate what the C.Q. would be on the new figures of Rogers and Cole. Reference to Fig. 13 will show that the new calorific quotients are higher than the theoretical range, but not so far from it as the old ones, and what is significant is that they point to a combustion of carbohydrate both before and after fertilization.

(4) Indirect evidence about the utilisation of protein by the embryo can be gained by the work of Scheminskito). Scheminski determined the resistance of the trout egg to electric currents during its development. The whole period was 55 days, and for the first 30 days there was practically no change in the resistance, but after that time it rose tremendously, the strength of current required to produce precipitation of the egg globulin in 1 minute increasing six times in the last 25 days of development. The effect of the current was to render the egg-membrane permeable to kations, which would diffuse out and cause the globulin to be precipitated (cf. Gray(18)). Jarisch(23) showed that lipoids and fats in systems poor in salt favour the precipitation of globulin, so if the current dismisses the kations from the egg, the precipitation of globulin will be more favoured the more fatty substances there are present. Scheminski’s curve becomes, then, in some measure, an index of the amount of fat absorbed by the embryo, and the fact that it is of so gradual a slope during the first two-thirds of development may be interpreted as showing a greater intensity of fat absorption (and combustion?) towards the end of development than towards the beginning. These findings may be compared with those of Gage and Gage (14) on the hen embryo, discussed in the previous paper.

Tomita (53) and von Gräfe(17) were not the only workers who drew attention in the past to the evidence showing that fat was not the only energy-source of the chick embryo. Dröge(8) considered that protein must take a share in the work, and Sakuragi(41) specifically went into the question of the other energy-sources of the embryo. In the German summary to his Japanese paper, he says, “Obwohl bisherige Autoren, welche sich mit Stoffund Energiewechsel von bebriitenden Hühnereiern beschäftigen, die Bedeutung des Kohlehydrates für Energiewechsel ganz vernachlassigten, glaubt der Verfasser, dass der schon vorhandene Trauben-zucker in den ersten Bebrütungstadien besondere Wichtigkeit und grosse Bedeutung dafiir hat, und dass der erste chemische Vorgang in den bebrütenden Eiern in der Zersetzung von Traubenzucker besteht.” Sakuragi estimated the free and combined sugar, the fat, the various fractions of nitrogen, and the glycogen at the different stages of development and interpreted his figures as showing that throughout development carbohydrate was combusted, the fat at the late stages being turned into carbohydrate before being burnt. His arguments for this process were not very convincing but his experimental results were very valuable indeed.

In the present paper, further evidence has been brought forward which seems to show that there is in the developing embryo a succession of energy-sources, carbohydrate being the first one to be used, then protein and then fat. The intensity of production of urea, and of uric acid, the intensity of combustion of protein, have all been shown to be greatest from the seventh to the eleventh day of development, in other words in the second quarter of incubation. The position can best be expressed by saying that the points of maximum intensity of combustion of the three classes of foodstuff probably follow each other in the order, carbohydrate, protein, fat.

It is interesting to note that in a closed system such as the hen’s egg the combustion of protein is much less than that of the other two foods. Fat and carbohydrate can be burnt completely away while the products of protein metabolism remain as nitrogenous waste which cannot be got rid of. In an egg such as that of the trout, however, where the nitrogenous end-products can escape into the surrounding medium, we find that as much as 63 per cent, of the total energy used during development may come from protein.

It is also interesting that in agreement with Gayda’s findings with the egg of the toad (15) the period when it is most expensive to double the weight of the chick embryo seems to coincide with the period of maximum intensity of protein combustion. In Fig. 14 is shown the mg. per cent, of wet weight of embryo protein combusted per day, and as a background, the number of gram-calories evolved per day per gm. wet weight of embryo during periods in which the weight of embryo is doubled. The agreement is suggestive, and, if further criticism should leave this relationship established, it would seem to be due to the specific dynamic action of the protein combusted. If we accept the findings of Sendju(45) that the amino-acids principally combusted are tyrosine and tryptophane, it may in the future be possible to calculate what the extent of this specific dynamic action should be. According to Seth and Luck(46) the S.D.A. of an amino-acid is proportional to its power of raising the blood amino nitrogen, and hence to the rapidity of its absorption from the gastro-intestinal tract. In the case of the egg, where the aminoacids presumably pass directly from the albumen into the blood-stream, there is at present no means of calculating what the S.D.A. ought to be. Possibly in the egg, where the influence of the intestinal wall is eliminated, all amino-acids may have the same S.D.A,

The ultimate nature of the succession of energy-sources presents a problem of great interest. It is possible that carbohydrate is first combusted because it requires no preparation. Proteins must be deaminated, fats must be desaturated, and perhaps the embryo in its early stages cannot do either of these things : but, on the other hand, glucose lies ready for use, and it is significant that what is then combusted is free, not combined, carbohydrate. There is already strong evidence that the power of desaturation of fats only arises at a comparatively late stage of development, e.g. the tenth to the fifteenth day in the chick (Needham (33), p. 21). And we may look on the unsaturated fatty acids present in egg-yolk as a preparation for these conditions.

Or it may be that some conception of “ease of combustion” will prove helpful. Quastel and Whetham (56) studying the action of B. colt on various organic substances, found that carbohydrates were much better hydrogen-donators than substances of protein or fatty type. The following figures, taken from their paper, are very striking :

The succession of energy-sources might in some such way be related to a changing oxidation-reduction potential of the embryonic cells, and a microinjection study of such cells in tissue-culture, using rH indicators, would now be in order.

An ontogenetic successiop of carbohydrate, protein and fat, could appear in three ways :

1. Combustion of foodstuffs by the embryo.

2. Absorption of foodstuffs by the embryo.

3. Storage of foodstuffs by the embryo, i.e. its constitution.

The following papers in this series will be devoted to investigations dealing with the relations between combustion and absorption on the one hand and combustion and constitution on the other. For a discussion of this subject from a somewhat different angle, reference should be made to the second paper of Murray (31).

Assuming, for the time being, that an ontogenetic succession of energy-sources exists, we may enquire whether it is due to the influence of the embryo or to the influence of its food supply. In other words, whether it is embryogenie or ovogenie, whether the embryo combusts protein because it cannot obtain carbohydrate, or because it must by its functional constitution, do so.

A certain amount of evidence already exists which contributes to a solution of this problem. It will be seen from Table VIII that the free sugar—for reasons given above probably the only carbohydrate fraction burnt by the embryo—does not entirely disappear until the twelfth day of development. Yet, as Fig. 10 shows clearly, it is between the eighth and the ninth day that the peak of utilisation of protein occurs ¹. The embryo then by no means awaits the exhaustion of its carbohydrate supplies before beginning to combust protein. This fact is strong evidence in favour of the view that the embryo and not the supply of food at its disposal is in command of the situation.

In order, however, to make more certain of this, some injection experiments were carried out. Eggs were injected in the manner described in a foregoing communication (Needham (33)) with a solution of glucose containing 500 mg. per c.c. The only modification in the method of injection was that the glucose was injected into the air-space, from which it was quickly absorbed ; this process considerably reduced the mortality of embryos. As will be seen from Table XIII and Fig. 15 no significant effect was produced upon the uric acid curve. If the embryo had been burning protein because carbohydrate was absent or not easily obtained, then the uric acid curve should have been depressed after the injection of glucose, but this was never the case.

We may conclude provisionally that the succession of energy-sources in ontogenesis is embryogenie and not ovogenic, that it is in some way intimately bound up with the metabolic potentialities of the growing embryo. The embryo does not behave as regards its nourishment as does the bacterial cell.

My thanks are due to Prof. Sir Frederick Hopkins for his constant interest, to my wife for much valuable help, and to the Government Grant Committee of the Royal Society for a grant towards the cost of these researches.