The Energy-Sources in Ontogenesis: I.—The Urea Content of the Developing Avian Egg

At the present time it is generally believed that fat is the energy-source specially associated with embryonic growth. This view, which has arisen almost entirely from work on the hen’s egg, is due to the researches of Pott (34), Liebermann (25), and Tangl and von Mituch(44). Estimations of the fat in the embryo and that in the rest of the egg showed that some had been lost in transit, actually from 2.1 to 2.76 gms. per egg on an average. Calculation demonstrated that this figure corresponded satisfactorily with the carbon dioxide evolved throughout incubation and the heat produced during the same time. Needless to say the fit of the figures was not very exact but was thought to be sufficient to show that fat was the only source of energy. For example, Libermann (25) had shown that the fat of the egg contained 71.67 per cent, of carbon, and the amount used per egg Tangl and von Mituch(44) found to be (on an average) 2.11 gms. According to Hasselbalch(19), the total amount of CO₂ produced during incubation amounted to 5.939 gms., i.e. 1.62 gms. of carbon. From this the calculated amount of fat used was 2.26 gms., which agreed moderately well with Liebermann’s figure but rather less so with one which they themselves found, viz., 2.76 gms. Another line of investigation which gave great support to this view was the work of Bohr and Hasselbalch (4), who measured carefully the respiratory quotient of the hen’s egg during development. From the seventh day onwards their figure was very constant at about 0.73, so that unless something very unusual was going on, fat must be considered as the only energy-source.

It might always have been doubted, however, that fat was the only important source of energy: there were hints to the contrary in the literature. William Harvey (18) had said, “and therefore the yolke seems to be a remoter and more deferred entertainment than the white, for all the white is quite and clean spent before any notable invasion is made upon the yolke.” Another important observation was that of Prévost and Le Royer (35) in 1825, who obtained from the allantoic fluid of a seventeen-day old chick a nitrogenous substance which gave an insoluble nitrate and resembled urea in all particulars. There had clearly been some catabolism of proteins. And since at about the same time Jacobson (21), Sacc(37), and Stas (41) found uric acid in considerable amounts in the allantoic and amniotic fluids of developing chicks during the third week of incubation, there was no doubt about it. Moreover, Wetzel (52) from a study of the chemical composition of many different types of eggs before and after development came to the conclusion that all three kinds of foodstuff had been burnt.

In a previous paper (30) I pointed out that whether or not fat was the predominating energy-source it could certainly not be considered to be the only one, for many arguments of great force pointed to bodies of carbohydrate nature as being the energy-sources of the initial stages. These arguments can be summarised as follows :—

1. In the first eight days of incubation of the hen’s egg there is a striking fall in the amount of free glucose. The curves, which will be found in the papers of Satô (38), Idzumi (20), Bywaters (5), and Tomita(46), all show a rapid fall from about 0.4.gm. per cent to 0.1 or less.

2. Simultaneously with this disappearance of free glucose, the lactic acid in the egg, which has previously been low, reaches a peak and immediately afterwards descends to its previous level. The curves given in Tomita’s paper (5) show that from an initial level of 0.02 mgm. per cent it attains on the fifth day a maximum of 0.13 mgm. per cent and regains its original value about the fourteenth day. That it was intimately connected with glucose he proved by injecting glucose and observing an increase of lactic acid.

3. If the figures of Bohr and Hasselbalch (4) are closely examined, it will be seen that although the respiratory quotient during the last two weeks of incubation is certainly 0.73 on an average, this is not the case with the earlier stages. On the contrary, figures as high as 0.91 and 0.81 were obtained, although in small number. Nothing could fall better into line with the results of Tomita; if anything these early figures are low, for two reasons. Firstly, Hasselbalch (19) showed that the egg, especially the yolk, gave out small amounts of oxygen during the first twelve hours of development and, secondly, the egg probably possesses an alkali reserve which would tend to absorb the first small amounts of expired CO₂. These high figures in the early stages did not escape the notice of Grafe (15), who in his review of the subject in 1910 thought that the high respiratory quotient might be associated in some way with the laying down of morphological structure.

4. The disappearance of glycogen in very early stages of development has been investigated by Lewis (24) and Konopacki (22). The former grew tissue-cultures of cells from very young chick embryos and could never find glycogen present in them except if they were taken before the first fifty hours of development were over. Konopacki, working on the frog, obtained exactly similar results. He found that after fertilisation and the formation of the perivitellus the glycogen greatly diminished in quantity and remained very low until the neurala stage was reached, after which the glycogen rose again. Similar results have recently been reported by Vastarini-Cresi(47). All these workers made use of microchemical methods.

5. Evidence pointing in the same direction is seen in the work of Warburg, Posener, and Negelein (49). They occupied themselves in examining the glycolytic behaviour of neoplasmic and embryonic tissues, and found a very marked preferential consumption of carbohydrate on the part of 3 to 5 day chicks. The production of ammonia when the tissue was suspended in bicarbonate Ringer in vitro was greater than that of any adult tissue save brain or retina, and corresponded to the similarly high position of embryos of this age as regards glycolytic power. But if sugar was provided for it, it would immediately cease to give off ammonia and would utilise carbohydrate. This fact, and the fact that the power of glycolysis is very high in early development and much lower in the adult condition both confirm strongly what has been said before.

6. Perhaps it is no coincidence that the respiratory quotient which is found in the early cleavage stages of small marine eggs such as those of the sea-urchin is in the near neighbourhood of unity. Warburg (48) found a respiratory quotient of 0.9 for the first twenty-four hours of the development of Arbacia pustulosa eggs, and Shearer (40) working on those of Echinus microtuberculatus obtained a figure of 0.95. A higher figure still was found by Fauré-Fremiet (9), whose results with the eggs of Sabellaria alveolata work out at an R.Q. of 1.0. As indices of the nature of the combustions going on, these figures ought to be accepted with the utmost caution, for there may be curious gas effects at fertilisation, e.g. the “neutral gas” of Bialascewicz and Bledowski (2). But it is natural to suppose that a stage which lasts five days in the chick may last only a few hours in lower animals which develop faster, so that the time to look for utilisation of carbohydrate would be from fertilisation to the neurula stage.

7. What would appear to be data antagonistic to the view now put forward are found in the monograph of Le Breton and Schaeffer (23). Calculating from the results of Bohr and Hasselbalch (4) on the chick they found that the number of gram calories produced per gram of embryo wet weight per hour was much higher in the initial stages of incubation than towards the end. They associated this fact with the curves which they obtained for the chemical nucleoplasmic, ratio, i.e. nitrogen, but they did not draw the conclusion that it looked as if fat was being used as the energy-source in the beginning. A rough calculation shows, however, that though this is so at first sight, it is not really the case. From the figures of Eaves (8) and Idzumi (20), it is found that about 0.3 gm. of fat are combusted by the chick on the fourteenth day. In this calculation a small correction must be made for the amount of fat stored in the embryonic tissues; this is done by means of the accurate embryonic fat figures kindly lent me by Dr H. A. Murray. The still smaller corrections necessary to allow for the fat set free from the lipoids at this time (Plimmer and Scott (33)) and the fat simultaneously removed as esters of cholesterol (Mueller (27)), practically cancel each other out. In the.fourteenth day, therefore, the fat combusted amounts to approximately 0.038 gm. per gram of embryo wet weight On the other hand the figures of Satô(38), Bywaters(5), and Tomita(46) show that in the fifth day about 0.04 gm. of free glucose is combusted, which works out at 0.18 gm. per gram of embryo wet weight, or five times as much as the amount of fat similarly used on the fourteenth day.^* Le Breton and Schaeffer’s curve is therefore explained by the fact that in the early stages of incubation relatively more carbohydrate is used per gram of embryo than is fat per gram of embryo later on. If these figures were related to dry weight instead of to wet weight the difference would be more striking, for the older the embryo becomes the drier it gets. This argument is almost the same thing as saying in Tangl’s (43) terminology that the relative and specific “Entwicklungsarbeit “is higher in the earlier stages than it is later on. It is also interesting to note that in Le Breton and Schaeffer’s curve the gram calories per gram of embryo wet weight per hour descend very rapidly to the end of what from the above evidence we might call the carbohydrate period. On the fourth day the figure is 46.6 and on the seventh day only 9.6, after which the fall is not nearly so rapid and on the eleventh day the figure is still four. The inflection in the curve comes when the glucose has disappeared and not long after the peak of lactic acid. The descending curve of Le Breton and Schaeffer is reflected again in that of Shearer (40), who studied the oxygen-consumption of embryonic tissue in vitro. The metabolic rate, i.e. the c.c. oxygen consumed per gram wet weight per hour calculated from his figures, falls almost exactly on the curve of Le Breton and Schaeffer.

These arguments seem to show fairly clearly that in the hen’s egg and in other eggs there is a preliminary period when carbohydrate is used as the principal source of energy. There are, however, other data in the literature which are not explained upon this hypothesis, such as the curious results relative to the calorific quotient obtained by Meyerhof (26) and Shearer (40), and the work of Bialascewicz and Mincoffna (3) on the frog. For a long time, too, all the data have pointed to an exclusive combustion of carbohydrate by mammalian and viviparous embryos. These may have evolved special metabolic habits to meet their special conditions, and have perhaps prolonged their carbohydrate period to cover the whole of their development. In other cases, the period of carbohydrate utilisation may have been shortened to a few hours; if this were the case certain contradictory facts at present in the literature would be explained. It was pointed out above that the fit of the figures in Tangl and von Mituch’s calculation was not very exact ; so this cannot be adduced as an argument against the existence of a “carbohydrate period,” since during its predominance the embryo is very small and so also is the total turnover of matter and energy. But there are further arguments, which not only confirm those which have been given above, but which point to the existence of a period of protein combustion, midway between the utilisation of carbohydrate and the utilisation of fat.

1. Simon Gage and-Susanna Gage(13) fed laying hens on Sudan III and found that although red eggs were laid, the embryos showed no trace even of a pink colour until the tenth day of incubation. After that time they became rapidly more coloured until at hatching they were quite red. One would suppose that in order to be combusted, fat would have to be absorbed into the embryo and would take the dye with it, as indeed did happen after the tenth day. Confirmation of this is found in the figures of Dr H. A. Murray, referred to above, for fat storage in the chick. At the eleventh day it begins to rise and draws away from its previous level. Both these facts point to an awakening of fat metabolism at the mid-point of incubation.

2. The work of Riddle (36) on the yolk and the yolk-sac is interesting in this connection. From the fifteenth to the twentieth day the yolk is the only supply of the chick, and his chemical study of it during this period shows that there is a preferential absorption of fat. The neutral fat decreases markedly in the yolk and the protein substances increase.

3. Bialascewicz and Mincoffna (3) working on the frog’s egg found that up to hatching there had been practically no loss of fat, but that a loss of protein could be recognised. They did not look for any preliminary period of carbohydrate consumption, but concluded that protein during the early period had undoubtedly served as a source of energy.

4. The earliest statement that protein in the hen’s egg must be a source of energy is due to Schneroffna (39). She found a constant relationship between the nitrogen in the embryo and the nitrogen in the allantoic fluid, a ratio which always worked out to about 17. Every 17 gms. of nitrogen absorbed by the embryo corresponded exactly to 1 gm. excreted. She concluded that this indicated protein as a source of energy, and gave figures to show that there was a peak of total nitrogen in the embryo at the ninth day. She interpreted this as showing a peak of protein absorption at that point—certainly in view of what we have already said, the point at which it might be expected to come. But there is no a priori reason for wishing to associate the point of greatest protein absorption with the period of greatest protein combustion (cf. Bywaters (5)).

5. Fauré-Fremiet and Dragoiu (10) made an extended study of the frog’s egg. They agree with Bialascewicz and Mincoffna in finding a loss of protein before hatching but they also observe a loss of glycogen and of fat, indicating that all three substances have acted as sources of energy. The same conclusion was arrived at by Tangl and Farkas (45) for the egg of the silkworm, Bombyx mori. None of these workers observed any stages in this. During most of the development Fauré-Fremiet and Dragoiu obtained an R.Q. of 0.6, which agrees well with the 0.65 and 0.7 of Parnas and Krasinska (32) and of Bialascewicz and Bledowski respectively. This state of affairs is like that seen in the hen’s egg, where for slightly over two-thirds of the total incubation-period, the R.Q. indicates combustion of fat. Moreover, Dakin and Dakin (7) found a utilisation of proteins during the development of the eggs of the plaice, and Greene (16) observed the same thing in the king-salmon.

6. If urea and uric acid are indicators of protein metabolism, so also is the phenomenon of specific dynamic action. Evidence that there is a period in development when this phenomenon appears is contained in the recent work of Gayda(53). Using a differential calorimeter, he measured the heat given out by developing toad embryos throughout their development. When he calculated the heat given out during each period in which the weight was doubled, he obtained a curve with a peak in the centre, to which the values rose and from which they descended. Thus development was more economical at the beginning and end of development than at the middle. When this is compared with the figures of chemical analyses (9, 28, and 2), it is seen that there is a predominant combustion of protein during the middle part of development, proper allowance being made for the slower development of the toad as compared with the frog. I have made similar calculations for the hen, using the figures of Bohr and Hasselbalch (4), and Dr Harry Murray. Exactly the same peaked curve appears, and shows that there is a point of least economical development about the ninth day; this may well be due to the specific dynamic action of protein combusted at that period.

There is thus a great deal of evidence in the literature which points to a succession of stages, each with a definite energy source. The only facts which diametrically oppose the view that in the earliest stages carbohydrate is utilised are those brought forward by Meyerhof (26), in the paper referred to above. He found that the calorific quotient of Arbacia pustolosa eggs immediately after fertilisation was 2.6, and concluded that fat was being utilised, although the classical figures of Rubner are 3.2 for protein, 3.3 for fat, and 3.5 for carbohydrate. Until a closer approximation to one of these is reached, it would surely be better to suspend judgment. With Aplysia limacina the figure was 3.0 which was little better. Meyerhof supported his opinion that fat was being combusted in the earliest stages by showing that the intensity of staining with sudan III diminished as development proceeded, though significantly enough it made a rapid fall after the pluteus stage. He looked for a carbohydrate in echinoderm eggs but could not find any with the usual tests—by no means proof that none was there. It does not seem that these data are sufficient to counterbalance the facts which have already been mentioned.

Probably other substances besides protein, fat, and carbohydrates may be utilised to supply energy in some forms of life. For example, the recent discovery by Heilbronn of Liverpool, of great amounts of spinacene, a cholesterol-like substance, in teleostean eggs, may lead to the solution of the problem of the energy-source of these embryos. What they combust has so far been quite unknown (see ref. 30, pp. 20 and 21).

Gräfe in 1910(15) thought that there might be some connection between the period of carbohydrate utilisation in the chick’s development and the fact that at that time all the organs were being formed, so that profound morphogenetic changes were going on. In the paper already referred to (30), I suggested that the carbohydrate period at the beginning and the fat period at the end might be associated with preponderance of change of shape, internal and external, on the one hand, and change of size on the other hand. But perhaps the time has not yet arrived for correlations of this kind. In the first place, “augmentation” and “framing,” “growth” and “differentiation” are not nearly such simple processes as they have been considered to be in the past, for as Murray (29) has shown, morphological and chemical differentiation are two quite distinct things, as are also morphological and chemical growth. Morphological differentiation and morphological growth change most rapidly at the beginning of embryonic life, while chemical differentiation as expressed by metabolic rate and changes in concentration of chemical substances, and chemical growth as expressed by the increase of dry solids relatively to water, change most rapidly in the later stages (see also Ref. 30, p. 50). Secondly, the data are at present only fragmentary and until many more are collected, it must remain unwise to correlate facts too hastily. On the other hand, it would be very interesting if any connection appeared between a succession of energy-sources in ontogenesis and the numerous observations of susceptible stages in development, such as those of Stockart (42) and of Parnas and Krasinaka (32). This work has brought out with exceptional clearness the fact that development may be discontinuous and in all cases passes through critical stages when disastrous effects will follow an interference innocuous at other times. Such a critical stage is the beginning of gastrulation. Stockard says : “The present extremely crude state of our knowledge of the chemistry of development will permit of no satisfactory statement of the principles underlying differences in developmental rate.” Perhaps the speculation may be permitted that critical stages in development may turn out to be associated with changes from one type of substance to another type as a source of energy. An intermediate link in the chain of events would be the rapid growth-rate of one or more organs, leading to a teratological result if development was at that moment interfered with.

But what are needed first of all are investigations to decide whether the conception of a succession of energy-sources is well-founded or not If indeed it occurs in the chick, will it be found to occur also in all types of embryo ? How far will it be found to be affected by the adaptations of particular species to the peculiar necessities of their environment ? These are the problems which must first be answered.

But assuming that the conception of a succession of energy-sources should prove to be well-founded, its exact interpretation would still remain uncertain. The ontogenetic procession could be either “ovogenic” or “embryogenic.” On the former view the energy-sources would succeed one another simply because the dynamic equilibria and the relative concentrations of substances in the yolk and white necessitated it. The embryo would play a passive part, combusting protein and fat only since it could not get carbohydrate. On the latter view the succession of energy-sources would be intimately connected with the changing potentialities of the growing embryo. Energy must be the same from whatever source it comes, but the embryo—on this view playing an active part—would combust such and such substances at such and such periods of its development because it would not have at those times the capacity for combusting others. The molecular orientations on its intracellular surfaces would differ at different stages of its development, perhaps in relation to its differentiations and growths, or at diverse times its enzyme systems would vary profoundly in activity.

At present there is not enough evidence to allow us to make a choice between these views. The ovogenic hypothesis would commit us to the belief that if sufficient carbohydrate were present during the protein and fat combustion periods the utilisation of these latter by the embryo would greatly diminish or disappear. On the embryogenic hypothesis we should have to believe that however much carbohydrate were present during the protein period the embryo would continue to combust protein, for a close relation would exist between its source of energy and its stage of development. In favour of the ovogenic hypothesis might be cited the case of the viviparous embryo, which is believed to combust carbohydrate throughout its development (3). But dangers beset any direct comparison between embryos in ovo and in utero. Mammalian embryos have an unlimited supply of nutriment, owing to their continuous perfusion system (1, 18), avian embryos do not; so that in one case the proportion of embryo to nutriment does not alter and in the other case it does. Mammalian embryos can have all their combustible material supplied to them in solution ; if the avian embryo lived in the same style it would need an egg vastly larger than its present size to contain its physiological sugar solution. The fat of the avian egg is tabloid food. Moreover, the placenta, as the reviews of Murlin (28) and Zuntz (51) show, exercises a far-reaching selective influence on the food of the embryo, such as is not known to exist in the avian or amphibian egg.

The active autohegemonic powers of growth which the embryo has been shown to possess by the experimental embryologists such as Hertwig, Roux, and Driesch might seem to favour the embryogenic hypothesis. In its support could also be adduced the fact that during the protein period in the avian egg, there is plenty of carbohydrate being absorbed in the bound form of ovomucoid. But accurate estimations of the carbohydrate content of the embryo must be done before this point can be elucidated. Perhaps light will be thrown on it by the researches of Hanan on the embryonic blood-sugar, a preliminary report of which has already appeared (17). Attempts will also be made in this laboratory to study the effects of injecting glucose. If teratological results can be avoided (and it is believed that they can (31)), the demonstration of a protein or fat-sparing action would help to decide between the two hypotheses.

Although the developing chick has been so much investigated, there remains a very wide field for observations upon it. Since we already have many data obtained on the chick which bear directly upon the question at issue, it was considered best to concentrate here for the time being. With this end in view, experiments have been carried out on the urea content of the egg during incubation, and these are reported in this paper. Determinations of the uric acid content of the egg are at present in progress and a subsequent paper in this series will be devoted to them. Eventually a study of the comparative glycolytic power of embryonic tissues at varying stages of incubation will be undertaken.

It is in the highest degree remarkable that so few researches have been carried out on the urea and uric acid in the hen’s egg. Probably this has been due to the absence of accurate and simple estimation methods, especially in view of the practical difficulties attending work on eggs. Gori (14) made quantitative estimations of the urea in the eggs of Torpedo ocellata and of various fishes, but he did not follow the changes that occurred throughout incubation. The only other study is that of Fridericia (12) who estimated the uric acid in the embryo and its spaces from the eleventh day onwards, but neglected the urea as being a priori unimportant, since the nitrogen excretion of the adult hen is 90 to 95 per cent in the form of uric acid. He did, however, make a few estimations of the urea present on the seventeenth day of incubation, using Schöndorff’s laborious method, and found the results to be low and variable. It was clearly necessary to go into the matter afresh, using a more modern method and trying to get a complete curve for the whole of incubation. In any case the fact that his estimations began only at the eleventh day renders them of little use for our purpose, since the most interesting changes might be expected to occur before that point.

The eggs used in these experiments were those of White Leghorn hens placed in the incubator on the same day that they were laid. For analysis the shell was broken open at different places at different periods of incubation. At the beginning it is best to open it along one long side and to let the yolk and albumen flow out into a basin, leaving the embryo and its membranes in the shell, from which they can be easily removed. In the earliest stages of all, however, it is best to break the whole egg contents into an evaporating basin and remove the embryo from the top on which it floats. Later there supervenes a period in which the yolk membrane is extremely fragile, so that the greatest care has to be used in opening the egg. As the yolk flows out it must be eased with a pair of forceps to avoid tearing the membrane against sharp points of the broken shell. Towards the end of incubation the yolk membrane again gets tougher, and as by this time the allantois has grown completely round the interior of the shell, it must be cut and turned back all round the opening before the yolk is pulled out. In this way there is a minimal loss of the allantoic and amniotic fluids.

The tared vessel containing the embryo, its membranes, and their contents, is weighed and the whole material carefully and quantitatively ground up with fine washed sand. Then it is washed into a large test-tube, using plenty of wash water. This is placed in a water-bath at 100° to coagulate the protein, which is removed by a prolonged series of decantations, or more quickly by filtering through closely woven muslin. The combined extracts are then evaporated down with gentle heat to a small bulk, perhaps 25 c.c., and in this liquid the urea is estimated by the method of Folin and Wu(11). Another preliminary procedure has to be adopted after the fourteenth or fifteenth day, for owing to the growth of the bones, it becomes impossible to grind up the chick satisfactorily with sand. The embryos are placed in a Bolinder mincing machine, which in a short time reduces them to a uniform pulp. The mass is then extracted twice with acetone in the proportion of 10 gms. of tissue to 30 c.c. of solvent. No further urea is extracted by a third or fourth repetition of this process. The coagulated tissue is filtered off and the acetone recovered by distillation, so that a watery solution is left which can be treated in exactly the same way as described above. Any traces of acetone are expelled during the evaporation to small bulk.

In Folin’s method the urea is decomposed by urease acting for thirty minutes at 50° in presence of a few drops of phosphate buffer mixture at pH 7.0, a saturated solution of borax added, and the ammonia driven over by rapid distillation into ice-cold 0.05 N hydrochloric acid, where it is estimated colorimetrically by nesslerisation. The blanks with this method are excellent, but for good results it is essential to have freshly acetone-precipitated soya bean. [See Van Slyke and Cullen (50)]. Each estimation must be accompanied by a duplicate without urease in order to know the amount of free ammonia present in the sample, and this value must be deducted from the other.

The results are all tabulated in Table 1. Columns D and E are calculated from the very accurate series of figures obtained for wet and dry weights of embryos by Dr H. A. Murray, who kindly put them at my disposal. The daily excretion, calculated from column C in column F shows a steady rise until the eleventh day, followed thereafter by a series of oscillations. The daily excretion rate naturally follows this.

The results are shown graphically in figs. 1 and 2. In fig. 1 is seen the mgms. per cent, weight plotted against the time and also the mgms. of urea related to each embryo. The mgms. of urea per embryo rise steadily as might be expected, but the mgms. per cent, of wet weight follows a much more interesting curve. It rises steadily also, until the ninth day, at which point it ceases to rise and becomes quite stationary for the whole of the rest of development. In other words, as far as the wet weight is concerned, the production or excretion of urea is very intense after the fourth day and before the ninth day. At later periods although excretion of urea is still going on, it only just succeeds in keeping abreast of the wet weight. It is certainly interesting that this intensive period of urea production occurs exactly between the carbohydrate period and the period associated with the predominance of fat metabolism. The effect might, of course, be due to many other causes than to a specially intense combustion of protein during this period. For example, it might be due to a limiting factor such as the incapacity of the embryonic liver at this stage to turn urea into uric acid. That the developing liver can act in this way is probable from what we know of the fat-desaturation process in embryonic metabolism (30). If this was the case, however, an inflection in the curve of absolute mgms. per cent, per embryo should appear at the time when the liver takes on this function, but this is not the case. It is true that the liver may exert such an influence to a slight extent, but this does not seem to affect the urea curve as a whole. The activities of the enzyme arginase must also be suspect for the present, for we have unfortunately no data as to its presence or distribution in embryonic tissues. It is unlikely that any urea escapes into the yolk and the albumen thus vitiating the curve. Tests made on such fractions on the eighth and sixteenth days gave no indications of the presence of urea there. It would be strange if the embryonic excreta passed into its food.

In fig. 2 the urea content is seen related to the dry weight of the embryo. Here the effect is the same but the shape of the curve quite different, for we have a peak at the ninth day instead of an inflection. After the ninth day has passed, the urea excretion far from keeping pace with the increase in dry matter quite fails to do so and drops away from it down to a possibly constant level of 30 mgms. per cent. But as before it is the ninth day which is prominent.

The results of these experiments are very significant as regards the protein metabolism of the embryo, but at the same time altogether incomplete. Until the full curve for uric acid is known, we shall not be in a position to discuss the significance of the ninth day peak, nor shall we be able to calculate—most important of all—whether the amount of protein combustion as indicated by the urea and uric acid excreted will account for the gradual lowering of the respiratory quotient from 0.9 to 0.73. At the same time it is extremely interesting that the curves for the urea content of the egg should show a peak and an inflection at the ninth day. If urea can here be regarded even as a subsidiary straw to show which way the metabolic wind is blowing, the fact that its intensive production takes place between the period of carbohydrate utilisation and that of the utilisation of fat is very important and must take its place beside the other facts which were discussed earlier in this paper. For their explanation a synoptic conception is required.

1. Attention is drawn to the numerous facts in the literature of chemical embryology which point to a succession of energy-sources during ontogenesis, carbohydrate preceding protein and protein preceding fat. This question is discussed.

2. The urea content of the hen’s egg has been investigated from the fourth to the nineteenth day of incubation. There is a period of intensive urea production from the fifth to the ninth day. After that point the excretion of urea fails to keep pace with the growth and differentiation of the embryo.

3. It is pointed out that this period comes exactly between the stage at which carbohydrate is known to be utilised as an energy-source, and that at which the same may be said of fat.

My thanks are due to Professor Sir Frederick Hopkins for his great interest and kindness, to Dr H. A. Murray and Miss M. Stephenson for the stimulus of their conversation, and to the Government Grant Committee of the Royal Society for a grant towards the expense of these researches.