THE BEHAVIOR OF CHLORIDES IN THE CELL SAP OF NITELLA

1. A method is given for determining the chloride content in a drop (less than 0.03 cc.) of the cell sap of Nitella. 2. Chlorides accumulate in the sap to the extent of 0.128 M; this accumulation can be followed during the growth of the cell. The chloride content does not increase when the cell is placed for 2 days in solutions (at pH 6.2) containing chlorides up to 0.128 M. 3. The exosmosis of chlorides from injured cells can be followed quantitatively. When one end of the cell is cut off a wave of injury progresses toward the other end; this is accompanied by a progressive exosmosis of chlorides.     

THE NERVOUS MECHANISM OF COORDINATION IN THE CRINOID,ANTEDON ROSACEUS

1. Stimulation causes Antedon to swim by means of alternate oral bending and dorsal stroke of the arms. Two arms of a given ray move alternately so that while one is executing the aboral stroke its mate is flexing ventrally. This implies reciprocal inhibition. 2. Recriprocal inhibition between the two arms of an isolated ray can be abolished by the use of either strychnine or nicotine. 3. Coordination between the rays is referable to the conducting properties of the nervous pentagon which connects the five rays. In this system an impulse loses in effectiveness as it passes from the point of origin. 4. When Antedon is made to rest oral face down on the floor of an aquarium, oral flexion of all the rays, swimming movements, and righting result. Antedon is therefore negatively stereotropic with reference to its ventral side. 5. Excitation of the dorsal cirri results in aboral bending of all the rays. Stimulation of the cirri inhibits ventral flexion to the extent of preventing righting movements while on the other hand stimulation of the ventral surface inhibits the grasp reflex of the cirri. Thus oral and aboral sides of Antedon exhibit dynamic symmetry although structurally dissimilar.     

SIMILARITY OF THE KINETICS OF INVERTASE ACTION IN VIVO AND IN VITRO. II

1. The pH-activity relationship of invertase has been studied in vivo and in vitro under identical external environmental conditions. 2. The effect of changing (H⁺) upon the sucroclastic activity of living cells of S. cerevisiae and of invertase solutions obtained therefrom has been found, within experimental error, to be identical. 3. The region of living yeast cells in which invertase exerts its physiological activity changes its pH freely and to the same extent as that of the suspending medium. It is suggested that this may indicate that this intracellular enzyme may perform its work somewhere in the outer region of the cell. 4. In using live cells containing maltase, no evidence of increased sucroclastic activity around pH 6.9, due to the action of Weidenhagen''s α-glucosidase (maltase), was found.     

CHEMICAL CHANGES IN THE ADDUCTOR MUSCLE OF THE CHELIPED OF THE CRAYFISH IN RELATION TO THE DOUBLE MOTOR INNERVATION

An investigation has been made of the phosphate and lactic acid changes in the adductor muscle of the cheliped of the crayfish Cambarus clarkii upon stimulation of the isolated axons for the fast and slow contractions at determined frequencies. The data obtained point to the following conclusions: 1. When the mechanical effects of the two types of contraction are the same, the chemical changes are of the same order. If the mechanical effects are different, the chemical changes likewise are not equivalent. This is especially to be seen in the case of stimulation at 50 shocks per second: a slowly rising, long continued, strong slow contraction takes place with no apparent change in the phosphate content; a quickly rising fast contraction occurs with a large increase in the phosphate. 2. Since equivalent chemical changes accompany equivalent mechanical action, the two types of contraction do not differ in the essential mechanism of the chemical changes involved, and only one type of contractile substance is present. 3. Even when a contraction has taken place to the maximum extent obtainable, only enough phosphate is found to correspond to one-fifth to one-third of the available phosphagen.     

THE R?LE OF THE ACTIVITY COEFFICIENT OF THE HYDROGEN ION IN THE HYDROLYSIS OF GELATIN

1. The hydrolysis of gelatin at a constant hydrogen ion concentration follows the course of a monomolecular reaction for about one-third of the reaction. 2. If the hydrogen ion concentration is not kept constant the amount of hydrolysis in certain ranges of acidity is proportional to the square root of the time (Schütz''s rule). 3. The velocity of hydrolysis in strongly acid solution (pH less than 2.0) is directly proportional to the hydrogen ion concentration as determined by the hydrogen electrode i.e., the "activity;" it is not proportional to the hydrogen ion concentration as determined by the conductivity ratio. 4. The addition of neutral salts increases the velocity of hydrolysis and the hydrogen ion concentration (as determined by the hydrogen electrode) to approximately the same extent. 5. The velocity in strongly alkaline solutions (pH greater than 10) is directly proportional to the hydroxyl ion concentration. 6. Between pH 2.0 and pH 10.0 the rate of hydrolysis is approximately constant and very much greater than would be calculated from the hydrogen and hydroxyl ion concentration. This may be roughly accounted for by the assumption that the uncombined gelatin hydrolyzes much more rapidly than the gelatin salt.     

THE MECHANISM OF THE INHIBITION OF HEMOLYSIS

The principal conclusion of this investigation is that the inhibitory effect of plasma or serum on hemolysis by saponin and lysins of the same type is similar in nature to the inhibitory effects of certain sugars and electrolytes, which again are similar to the acceleratory effects produced by indol, benzene, and other substances already studied. All these effects, both inhibitory and acceleratory, are the result of reactions between the inhibitors or accelerators and those components of the red cell membrane which are broken down by lysins. The inhibitory effect of plasma on saponin hemolysis has a number of properties in common with the inhibition produced by sugars and electrolytes and with accelerations in general. (a) The temperature coefficient is small and negative. (b) The extent of the inhibition depends on the type of red cell used in the hemolytic system. (c) The most satisfactory measure of the extent of the inhibition, the constant R, is a function of the concentration of lysin in the system, and (d) R is a linear function of the quantity of inhibitor present. It is also shown that the inhibitory effect of plasma, and serum is not entirely dependent on its protein content. The process underlying the phenomenon of lysis and its acceleration or inhibition seems to be one in which the lysin reacts with a component or components of the cell membrane in such a way as to break down its semipermeability to hemoglobin, and in which the accelerator or inhibitor also reacts with the same component in such a way as to increase or decrease the effectiveness of the lysin in producing breakdown. The membrane is considered as being an ultrastructure made up of small areas or spots of varying degrees of resistance to breakdown, the resistances being distributed according to a negatively skew type of frequency curve, and the process of lysis seems to begin with the least resistant spots breaking down first. These spots may be arranged in some regular spatial pattern, and the membrane has also to be regarded as possessing spots of varying rigidity of form. The accelerator or inhibitor changes the resistance of every reactive spot in the ultrastructure by a factor R, which suggests that acceleration and inhibition are results of some over-all effect, such as that of changing the extent to which lysin is concentrated at the surface or partitioned between the material of the membrane and the surrounding fluid. Some kind of combination between the accelerator or inhibitor and the material of the ultrastructure is presumably involved; at first the combination seems to be a loose one and partly reversible, but later some of the loose links are replaced by more permanent combinations involving the same types of bond as are broken down by the lysins themselves.     

THE KINETICS OF PENETRATION : XVI. THE ACCUMULATION OF AMMONIA IN LIGHT AND DARKNESS

The accumulation of ammonia takes place more rapidly in light than in darkness. The accumulation appears to go on until a steady state is attained. The steady state concentration of ammonia in the sap is about twice as great in light as in darkness. Both effects are possibly due to the fact that the external pH (and hence the concentration of undissociated ammonia) outside is raised by photosynthesis. Certain "permeability constants" have been calculated. These indicate that the rate is proportional to the concentration gradient across the protoplasm of NH₄ X which is formed by the interaction of NH₃ or NH₄OH and HX, an acid elaborated in the protoplasm. The results are interpreted to mean that HX is produced only at the sap-protoplasm interface and that on the average its concentration there is about 7 times as great as at the sea water-protoplasm interface. This ratio of HX at the two surfaces also explains why the concentration of undissociated ammonia in the steady state is about 7 times as great in the sea water as in the sap. The permeability constant P'''''' appears to be greater in the dark. This is possibly associated with an increase in the concentration of HX at both interfaces, the ratio at the two surfaces, however, remaining about the same. The pH of sap has been determined by a new method which avoids the loss of gas (CO₂), an important source of error. The results indicate that the pH rises during accumulation but the extent of this rise is smaller than has hitherto been supposed. As in previous experiments, the entering ammonia displaced a practically equivalent amount of potassium from the sap and the sodium concentration remained fairly constant. It seems probable that the pH increase is due to the entrance of small amounts of NH₃ or NH₄OH in excess of the potassium lost as a base.     

AUTOPHAGIC DEGRADATION OF GLYCOGEN IN SKELETAL MUSCLES OF THE NEWBORN RAT

Large amounts of glycogen accumulate in rat skeletal muscle fibers during the late fetal stages and are mobilized in the first postnatal days. This glycogen depletion is relatively slow in the immature leg muscles, in which extensive deposits are still found 24 hr after birth and, to some extent, persist until the 3rd day. In the more differentiated psoas muscle and especially in the diaphragm, the glycogen stores are completely mobilized already during the early hours. Section of the sciatic nerve 3 days before birth or within the first 2 hr after delivery does not affect glycogen depletion in the leg muscles. Neonatal glycogenolysis in rat muscle fibers takes place largely by segregation and digestion of glycogen particles in autophagic vacuoles. These vacuoles: (a) are not seen in fetal muscle fibers or at later postnatal stages, but appear concomitantly with the process of glycogen depletion and disappear shortly afterwards; (b) are prematurely formed in skeletal muscles of fetuses at term treated with glucagon; (c) contain almost exclusively glycogen particles and no other recognizable cell constituents; (d) have a double or, more often, single limiting membrane and originate apparently from flattened sacs sequestering glycogen masses; (e) are generally found to contain reaction product in preparations incubated from demonstration of acid phosphatase activity. The findings emphasize the role of the lysosomal system in the physiological process of postnatal glycogen mobilization and appear relevant in the interpretation of type II glycogen storage disease.     

THE ISOELECTRIC POINTS OF THE PROTEINS IN CERTAIN VEGETABLE JUICES

The state in which a protein substance exists depends upon the nature of its combination with acids or bases and is changed by change in the protein compound. The nature of the compound of a protein that exists at any hydrogen ion concentration can be ascertained if the isoelectric point of the protein is known. Accordingly information regarding the isoelectric points of vegetable proteins is of importance for operations in which it may be desirable to change the state of protein substances, as in the dehydration of vegetables. The Protein in Potato Juice.—The hydrogen ion concentration of the filtered juice of the potato is in the neighborhood of 10^–7 N. Such juice contains the globulin tuberin to the extent of from 1 to 2 per cent. The character of the compound of tuberin that exists in nature was suggested by its anodic migration in an electric field. The addition of acid to potato juice dissociated this compound and liberated tuberin at its isoelectric point. The isoelectric point of tuberin coincided with a slightly lower hydrogen ion concentration than 10^–4 N. At that reaction it existed most nearly uncombined. The flow of current during cataphoresis was greatest in the neighborhood of the isoelectric point. This evidence supplements that of the direction of the migration of tuberin, since it also suggests the existence of the greatest number of uncombined ions near this point. At acidities greater than the isoelectric point tuberin combined with acid. The compound that was formed contained nearly three times as much acid as was needed to dissociate the tuberin compound that existed in nature. At such acidities tuberin migrated to the cathode. Though never completely precipitated tuberin was least soluble in the juice of the potato in the neighborhood of its isoelectric point. Both the compounds of tuberin with acids and with bases were more soluble in the juice than was uncombined tuberin. The nature of the slight precipitate that separated when potato juice was made slightly alkaline was not determined. The Protein in Carrot Juice.—The isoelectric point of the protein in carrot juice coincided with that of tuberin. Remarkably similar also were the properties of carrot juice and the juice of the potato. Existing in nature at nearly the same reaction they combined with acids and bases to nearly the same extent and showed minima in solubility at the same hydrogen ion concentrations. The greatest difference in behavior concerned the alkaline precipitate which, in the carrot, was nearly as great as the acid precipitate. The Protein in Tomato Juice.—The protein of the tomato existed in a precipitated form near its isoelectric point. Accordingly it was not present to any extent in filtered tomato juice. If, however, the considerable acidity at which the tomato exists was neutralized the protein dissolved and was filterable. It then migrated to the anode in an electric field. The addition of sufficient acid to make the hydrogen ion concentration slightly greater than 10^–5 N again precipitated the protein at its isoelectric point. At greater acidities migration was cathodic.     

THE INTRACELLULAR SITE OF SYNTHESIS OF MITOCHONDRIAL RIBOSOMAL PROTEINS IN NEUROSPORA CRASSA

The intracellular site of synthesis of mitochondrial ribosomal proteins (MRP) in Neurospora crassa has been investigated using three complementary approaches. (a) Mitochondrial protein synthesis in vitro: Tritium-labeled proteins made by isolated mitochondria were compared to ¹⁴C-labeled marker MRP by cofractionation in a two-step procedure involving isoelectric focusing and polyacrylamide gel electrophoresis. Examination of the electrophoretic profiles showed that essentially none of the peaks of in vitro product corresponded exactly to any of the MRP marker peaks. (b) Sensitivity of in vivo MRP synthesis to chloramphenicol: Cells were labeled with leucine-³H in the presence of chloramphenicol, mitochondrial ribosomal subunits were subsequently isolated, and their proteins fractionated by isoelectric focusing followed by gel electrophoresis. The labeling of every single MRP was found to be insensitive to chloramphenicol, a selective inhibitor of mitochondrial protein synthesis. (c) Sensitivity of in vivo MRP synthesis to anisomycin: We have found this antibiotic to be a good selective inhibitor of cytoplasmic protein synthesis in Neurospora. In the presence of anisomycin the labeling of virtually all MRP is inhibited to the same extent as the labeling of cytoplasmic ribosomal proteins. On the basis of these three types of studies we conclude that most if not all 53 structural proteins of mitochondrial ribosomal subunits in Neurospora are synthesized by cytoplasmic ribosomes.     

THE KINETICS OF CELLULAR PROLIFERATION IN REGENERATING LIVER

The study concerns the kinetics of cellular proliferation in the different cell populations of the normal and regenerating rat liver. A detailed analysis is presented, which includes techniques of in vivo labeling of DNA with tritiated thymidine and high-resolution radioautography, of the temporal and spatial patterns of DNA synthesis and cell division in the parenchymal cells, littoral cells, bile duct epithelium, and other cellular components in the liver during the first 64 hr of regeneration after partial hepatectomy. The analysis of cell population kinetics indicates that (a) the rate of entry of parenchymal cells into synthesis, after an initial burst of proliferative activity, was an orderly progression at 3–4%/hr; (b) most cells divided once and a few twice, a large proportion of the cell deficit being replaced by 72 hr after the onset of proliferation; (c) T_s was ~8.0 hr; T_gg2+m/2, 3.0 hr; and M, ~1.0 hr. Littoral cell proliferation began about 24 hr after the onset of parenchymal cell proliferation; the rate of entry of littoral cells into synthesis was greater than 4%/hr. Interlobular bile duct cell proliferation lagged well behind the parenchymal and littoral cell populations both in time and extent of proliferation.     

THE DISTRIBUTION OF EXOGENOUS FERRITIN IN TOAD SPINAL GANGLIA AND THE MECHANISM OF ITS UPTAKE BY NEURONS

Toad spinal ganglion cells are individually enclosed in sheaths consisting of one or more attenuated layers of satellite cell cytoplasm surrounded externally by a basement membrane. Narrow (~150 A) extracellular channels separate these layers from one another and from the underlying neuron. In both in vivo and in vitro experiments it was found that molecules of ferritin, a water-soluble protein, are to some extent able to pass across the basement membrane and through these channels to reach the neuronal plasma membrane. Ferritin particles arriving at the neuronal surface are engulfed by the neuron in 0.1 to 0.2 µ "coated" vesicles. The concentration of ferritin in these vesicles is higher than in the perineuronal space. The ferritin incorporated into the neuron is segregated, apparently intact, in multivesicular bodies. It is inferred that the 150A channels in the satellite cell sheath are patent, aqueous spaces through which molecules with a diameter as large as 95 A are able to pass, and that these neurons are capable of taking up whole protein from their immediate environment by the process of pinocytosis.     

DIFFERENTIATION OF THE JUNCTIONAL COMPLEX OF SURFACE CELLS IN THE DEVELOPING FUNDULUS BLASTODERM

The structure of the junctional complex between surface cells was investigated in blastula, mid gastrula, late gastrula, and early embryo of the teleost fish Fundulus heteroclitus. In blastulae, the intercellular complex is simple and consists of an apical region where the adjacent membranes are closely apposed (40–60 A) and in places touch, an intermediate zone with a wider intercellular space (> 100 A), and incipient desmosomes. In gastrulae, there are frequent points of fusion of membranes along the apical zone of the complex. Dilatations and an increased number of desmosomes in different stages of development are found along the intermediate zone. In mid gastrula, a close or gap junction with an intercellular space of 20 A occurs below the level of the desmosomes. In late gastrula, the gap junction is reduced in extent and desmosomes are better developed. In the early embryo, the basic organization of the complex is the same, although the deeply situated close junctions are no longer apparent and desmosomes and their associated system of filaments are well developed. At this time, the junctional complex is comparable to that of many epithelia and consists of an apical zonula occludens, a short zonula adherens, and deeply situated maculae adherentes.     

THE METABOLIC SYSTEMS INVOLVED IN DISSIMILATION OF CARBOHYDRATE RESERVES IN BAKERS' YEAST

Evidence is presented showing that the dissimilation of carbohydrate reserves in two strains of bakers'' yeast, Saccharomyces cerevisiae, is a purely respiratory process. Endogenous respiration is KCN-labile. Our own experiments together with various accounts and data given in the literature show that the same "oxygen-transporting mechanism" functions in both endogenous and exogenous metabolism. However, the lack of sensitivity of the endogenous system of reactions to low concentrations of monoiodoacetic acid, the absence of anaerobic CO₂ production, and the absence of alcohol production, demonstrate that fermentation is not involved in the dissimilation of the carbohydrate reserves. Throughout the experiments the endogenous respiration behaved functionally as a unitary system of reactions. The O₂ consumption and CO₂ production were parallel at all times; i.e., the R. Q. was consistently 1. Monoiodoacetic acid and KCN in concentrations from 10^–5 to 10^–1 molar affected both O₂ uptake and CO₂ production to the same extent. The only agents known to alter the value of the R. Q. were those which disrupted the normal protoplasmic structure, viz. grinding the cells with sand, plasmolyzing them with toluol and hypertonic salt solutions, or pressing them in a hydraulic press. These agents brought about a vigorous anaerobic CO₂ production accompanied by an accumulation of alcohol in the medium. The unitary character of endogenous respiration is exhibited only when the normal structure of the cell is kept intact; apparently it depends upon the maintenance of a chambered (or compartmental) architecture of the cell.     

THE VAPOUR PRESSURES OF AQUEOUS SOLUTIONS WITH SPECIAL REFERENCE TO THE PROBLEM OF THE STATE OF WATER IN BIOLOGICAL FLUIDS

Data for the depression of vapour pressure are presented for the following aqueous solutions: NaCl (0.03 to 0.1 molar), KCl (0.03 to 0.1 molar), urea (0.05 to 0.5 molar), sucrose (0.05 to 0.10 molar), lactic and succinic acids, creatine, CaCl₂ (0.05 molar), and mixtures of these substances with one another and with certain other solutions (gelatin, gum acacia, sea water, LiCl, etc.). The relation of the depression of vapour pressure of a mixed solution to that of solutions of the individual constituents was investigated in order to ascertain to what extent such studies may be used for the determination of the degree of hydration, or of the state of water, in solutions. Organic substances (urea, sucrose, etc.) showed anomalous results which were markedly affected and unpredictable in mixed solutions. They are, therefore, unsuited for the study of water binding. In the case of solutions of inorganic substances—LiCl and CaCl₂—the principle of the additive nature of colligative properties is also only approximately true—except perhaps in very dilute solutions. The limitations of the colligative method for determining the degree of hydration have been defined in accord with the above findings. Studies of the vapour pressures of mixtures of gelatin or gum acacia with NaCl or KCl demonstrated that hydration in gelatin is relatively small at pH = 7 and undetectable in gum acacia solutions. The view, therefore, that hydrophilic colloids are strongly hydrated has not been substantiated. The passage from the sol to the gel state also was not accompanied in gelatin or in blood by any appreciable change in the degree of hydration of the hydrophilic colloids present in these substances.     

THE STRUCTURE OF THE COLLODION MEMBRANE AND ITS ELECTRICAL BEHAVIOR : IX. WATER UPTAKE AND SWELLING OF COLLODION MEMBRANES IN AQUEOUS SOLUTIONS OF ORGANIC ELECTROLYTES AND NON-ELECTROLYTES

1. Dried collodion membranes are known to swell in water and to the same limited extent also in solutions of strong inorganic electrolytes (Carr and Sollner). The present investigation shows that in solutions of organic electrolytes and non-electrolytes, the swelling of dried collodion membranes is not as uniform, but depends on the nature of the solute. 2. The solutions of typically "hydrophilic" substances, e.g., glycerine, glucose, and citric acid, swell collodion membranes only to the same extent as water and solutions of strong electrolytes. In solutions of typically carbophilic substances (e.g., butyric acid, valeric acid, isobutyl alcohol, valeramide, phenol, and m-nitrophenol) the swelling of the membranes is much stronger than in water, according to the concentration used. For the brand of collodion used the swelling in 0.5 M solution was in some cases as high as 26 per cent of the original volume, as compared to 6 to 7 per cent in water. Therefore, in these solutions the "water-wetted dried" collodion membrane is not rigid, inert, and non-swelling, but behaves as a swelling membrane. 3. The solutes which cause an increased swelling of the membranes are accumulated in the latter, the degree of accumulation being markedly parallel with the degree of their specific swelling action. 4. The anomalously high permeabilities of certain carbophilic organic solutes reported by Michaelis, Collander, and Höber find an explanation in the specific interaction of these substances with collodion. 5. The use of the collodion membrane as a model of the ideal porous membrane is restricted to those instances in which no specific interaction occurs between the solute and the collodion.     

THE NATURE OF THE LATENT PERIOD IN THE PHOTIC RESPONSE OF MYA ARENARIA

The latent period in the response of Mya to illumination varies inversely as the duration of the exposure to which it is subjected. The reciprocal of the latent period, measuring the velocity of the process which underlies it, is a linear function of the exposure period. Since the duration of the exposure represents the amount of photochemical activity, it is concluded that the substances formed at that time act to catalyze a chemical reaction which determines the duration of the latent period. This explanation is in accord with the previous work on the photochemical reaction and with the effect of temperature on the latent period. As a result of the combined investigations there is presented a concrete hypothesis for the mechanism of photic reception in Mya.     

THE FINE STRUCTURE OF EPENDYMA IN THE BRAIN OF THE RAT

The ciliated ependyma of the rat brain consists of a sheet of epithelial cells, the luminal surface of which is reflected over ciliary shafts and numerous evaginations of irregular dimensions. The relatively straight lateral portions of the plasmalemma of contiguous cells are fused at discrete sites to form five-layered junctions or zonulae occludentes which obliterate the intercellular space. These fusions occur usually at some distance below the free surface either independently or in continuity with a second intercellular junction, the zonula adhaerens. The luminal junction is usually formed by a zonula adhaerens or, occasionally, by a zonula occludens. The finely granular and filamentous cytoplasm contains supranuclear dense bodies, some of which are probably lysosomes and dense whorls of perinuclear filaments which send fascicles toward the lateral plasmalemma. The apical regions of the cytoplasm contain the basal body complexes of neighboring cilia. These complexes include a striated basal foot and short, non-striated rootlets emanating from the wall of each basal body. The rootlets end in a zone of granules about the proximal region of the basal body, adjacent to which may lie a striated mass of variable shape. All components of the basal body complex of adjacent cilia are independent of each other.     

OBSERVATIONS ON THE TRANSPORT OF CARBON DIOXIDE IN THE BLOOD OF SOME MARINE INVERTEBRATES

Although the results we have recorded merely serve to indicate the possibilities of this interesting field of investigation, we have sufficient data to enable us to draw certain general conclusions. In the first place it is evident that the bloods of the more highly developed marine invertebrates, such as the active Crustacia and the Cephalopods, are specially adapted for the carriage of carbon dioxide. The quantity of carbon dioxide taken up by the blood of Maia, Palinurus, or Octopus at any given tension of the gas is, in general, about twice or three times as great as that which is taken up by sea water under the same conditions. On the other hand, the blood of a slow, creeping form, such as Aplysia, or of a sessile animal such as the ascidian Phallusia shows no more adaptation for the carriage of carbon dioxide than does sea water. But our estimations of the CO₂ content of the blood as it circulates in the bodies of these more active invertebrates show that the conditions of transport of this gas differ considerably in some respects from those which obtain in mammals. For the invertebrate blood in the body contains only a relatively small quantity of carbon dioxide, averaging in the forms we examined from 3 to 10 cc. per 100 cc. of blood. This forms a marked contrast with the condition found in mammals where even the arterial blood contains about 50 cc. of CO₂ per 100 cc. of blood. The invertebrate, therefore, works at a very low CO₂ tension. There is a twofold significance in this circumstance. In the first place, it means that only the first portion of the carbon dioxide dissociation curve is in use in the respiratory mechanism. Now an inspection of our curves will show that at these low carbon dioxide tensions the dissociation curves tend to be steeper than at higher tensions. As we intend to show in a later paper it can be proved mathematically that, other things being equal, a blood with a carbon dissociation curve of moderate steepness, i.e. one in which the carbon dioxide content of the blood increases fairly rapidly with increase of carbon dioxide tension, is a more efficient carrier of the gas from the tissues to a respiratory surface than a blood in which the dissociation curve is either steeper or flatter. It would seem as if the active invertebrates avoid the use of too flat a part of their CO₂ dissociation curves by working over the initial steeper portion. Furthermore, it is seen that over the range of this initial steep portion of the curves the changes of reaction produced by the uptake of carbon dioxide are much smaller than at higher tensions of the gas; for these initial portions of the curves are more nearly parallel to the lines of constant reaction calculated for a temperature of 15°C. according to Hasselbalch''s method (10) on the assumption that the whole of the combined CO₂ is in the form of sodium bicarbonate. It is evident also that on this assumption the hydrogen ion concentration of the blood of invertebrates (with the exception of the tunicates) would appear to be practically the same as that of the warm-blooded vertebrates—a conclusion confirmed by the direct measurements of Quagliariello (9). On the other hand, our measurements do not lend support to the idea put forward by Collip (4) that in order to maintain an appropriate faintly alkaline reaction an invertebrate needs to retain carbon dioxide in its blood at a comparatively high tension. This idea was based on the observation that at comparatively high CO₂ tensions the blood of invertebrates contains considerably more sodium bicarbonate than does sea water. But our curves show that this is no longer true at the lower values of carbon dioxide tension, the amount of sodium bicarbonate falling off more rapidly in the blood than in the sea water with diminution of the carbon dioxide tension so that in order to maintain an appropriate reaction in the blood only a comparatively small tension of CO₂ is required. The largest amount of carbon dioxide that we found present in the circulating blood of any of the types examined was 9.7 cc. per 100 cc. of blood in the case of Maia, and in most cases the amount was considerably less. But even this lowest value corresponds to a tension of CO₂ of only about 3 mm., so that the tension gradient across the gill membrane must be even less than this. We would emphasize rather the circumstances that as the portion of the dissociation curve over which the reaction is approximately constant is of but small extent, it is necessary that in an active form like Octopus the carbon dioxide produced should be removed rapidly lest an accumulation of it should cause the limits of normal reaction to be exceeded; and this need is correlated with the extreme efficiency of the respiratory apparatus in this animal. It is interesting to notice that the mammal which, in order to obtain an appropriate reaction in the blood, has to work at relatively high carbon dioxide tensions where the dissociation curve is comparatively flat, secures a steeper physiological CO₂ dissociation curve in the body, and with it a more efficient carriage of carbon dioxide and a more constant reaction in the circulating fluid, in virtue of the effect of oxygenation on the carbon dioxide-combining power of its blood (3, 6). Returning now to the consideration of the actual form of the dissociation curves we have obtained—it is a significant fact that it is in those forms such as Maia, Palinurus, and Octopus whose bloods are rich in proteins—particularly hemocyanine—that the initial steep portion of the curve is observed. This suggests that in these forms the blood proteins act as weak acids and expel carbon dioxide from the blood at the low tensions which include the physiological range, just as in vertebrates the hemoglobin similarly displaces carbonic acid from its combination with alkali metal. On the other hand the cœlomic fluid of Aplysia contains no pigment and only 0.00672 per cent of protein nitrogen (Bottazzi (11)) and shows no initial rapidly ascending portion of the CO₂ dissociation curve. This is supported by the observation of Quagliariello (9) that the acid-neutralising power of the blood of an invertebrate is roughly proportional to its protein content. It seems as if the proteins of invertebrate blood like the blood proteins of vertebrates, exist in the form of sodium salts which are capable of giving up sodium for the transport of carbon dioxide as sodium bicarbonate. That this is so in the case of hemocyanine follows from the fact that the isoelectric point of this pigment occurs at a hydrogen ion concentration of 2.12 x 10^–5 N, i.e. at a pH of 4.67 (Quagliariello (12)) so that in the alkaline blood of the invertebrates possessing it, hemocyanine will act as a weak acid. It is probable that the initial steep portion of the carbon dioxide dissociation curves which we have found to be of such importance in the respiration physiology of Octopus, Palinurus, and Maia is produced by the competition of this acid with carbonic acid for the available sodium of the blood.