Experimental Biology and Medicine

The most potent source of the antirachitic factor is fish oil, that from the liver of the cod having been found to be most active. In the course of an assay of various foods, etc., it was ascertained that the liver of the puffer fish (Spheroides maculatus) not only contains a large quantity of oil, but that this oil has exceptional antirachitic power. The oil was prepared from the liver by means of steam, by a method similar to that which is made use of in extracting oil from the livers of the cod. On testing this puffer fish oil biologically, it was found to be at least 15 times as potent as the most active specimens of cod liver oil. Such was the case in preventive tests in which the Sherman-Pap-penheimer diet was fed, as well as in curative tests in which the McCollum diet was employed as a basal ration. No doubt if the stearine had been removed, the activity of the oil would have been proportionately increased.

No investigations have been carried out to ascertain whether the remarkable antirachitic potency of fish oil is due to the diet of the fish or to the intensity of the solar rays to which they are subjected in nature. In order to study this question, oil was prepared from the livers of puffer fish which had been maintained in an aquarium in a tank for a period of 3 months. The fish were of the same batch which had been previously tested. During this period of captivity no active ultra-violet radiations had access to the fish, as the windows of the building were closed at all times. In spite of the fact that the fish were deprived for 3 months of activating rays, quantitative tests showed that their livers contained oil equal in potency to that which had been prepared shortly after they had been removed from the sea.

One may say that under the conditions of these experiments, a lytic principle active for B. coli “D” is acted upon by ultra violet light in much the same way as are two strains of known filterable viruses, i. e., herpes and Levaditi's so-called encephalitis virus. Exposure to ultra violet light at a distance of one foot for forty minutes is sufficient to attenuate or destroy both the bacteriophage and the two filterable viruses employed in these experiments. In a substrate of normal rabbit serum all three are protected from the action of ultra violet light.

In recent years much light has been thrown upon the chemical nature of some typical enzymes, both by the direct method of purification and analysis and by other methods which, while indirect, have phoven none the less instructive.

Pancreatic amylase has been purified by the methods developed in this laboratory 1 to a maximum enzymic activity practically constant in many independent purification experiments. This material in the dry state is an amorphous white powder having the ultimate composition, and showing the color reactions of typical proteins. Moreover, the enzyme material showed, on analysis by the Van Slyke method, 2 the same forms of nitrogen as are found in typical proteins, and in typical quantitative proportions. Like the purified malt amylase previously described by Osborne, 3 this pancreatic amylase on heating in water yielded a coagulated albumin and left in solution a proteose or peptone.

Our purified pancreatic amylase preparation has much the highest enzymic activity of any material of which we have found adequate record. In thirty minutes at 40°, this material splits 20,000 times its weight of starch and forms 10,000 times its weight of maltose; and notwithstanding its gradual inactivation in solution, this material has, in sufficiently long experiments, digested as much as 4,000,000 times its weight of starch and formed as much as 2,800,000 times its weight of maltose. 4

This great enzymic activity was shown at a dilution of 1 :100,000,000, whereas the most delicate tests for protein are probably not valid at dilutions greater than about 1:100,000. Thus the failure of protein reactions in solutions enzymically active does not show, as many writers assume, that the enzyme is of other than protein nature, since as in this case, the enzymic activity may constitute a test 1000-fold more delicate than any reaction which can be employed as a test for the presence of protein material.

The isolation of a crystalline substance from an active sub strate always leaves existent the doubt as to whether the activity is due to the crystals themselves or to adhering traces of the potent factor. The complete inactivation of Bios 223 crystals by forming the crystalline benzene sulfonamide and the recovery of activity by hydrolysing this inactive crystalline product seems to remove all reasonable doubt and to demonstrate that the yeast stimulating power of Bios 223 resides in the crystals themselves. The nature of the addition compound also suggested that the NH₂, group is concerned in the exercise of this function.

We also take pleasure in acknowledging the suggestions and advice of Professor John M. Nelson of Columbia University in developing these results.

Erythrocytes of the domestic fowl, washed free from plasma, were suspended in salt solution, one part in twenty. This suspension was then injected through a fine glass cannula into one of the short gastric branches of the splenic artery of a rabbit, under ether anesthesia. One minute after beginning the injection a portion of spleen at one end was clamped off, excised and placed in Helly's fixing solution for histological study. After two minutes a second sample was taken. After fifteen minutes the remainder of the spleen was perfused through the original cannula, first with a little salt solution and then with Helly's fixing solution, until the spleen became firm and yellowish in color.

Histological study of the one-minute sample showed abundant avian erythrocytes in the spleen pulp of the marginal zone of the Malpighian corpuscles with relatively less cells of this type in the spleen sinuses. At two minutes the relationship was similar but the number of nucleated erythrocytes in the sinuses and in the pulp at a distance from the Malpighian corpuscles had increased. In the last specimen, perfused with fixing solution at the end of fifteen minutes, the distension and perfect fixation permit a clear view of the intimate structural relationships. Here one sees the foreign blood cells in the pulp of the marginal zone and of the pulp cords but also in the interior of the sinuses. Some of them have already been ingested by phagocytic cells.

These observations are directly opposed to the view of Helly. 1 that all transfused foreign erythrocytes reach the spleen pulp after traversing the venous sinuses, and are quite in accord with the observations of Gray 2 Mall 3 and Weidenreich, 4 who have ascribed the conspicuous injection of the marginal zone of the Malpighian corpuscles to the termination of abundant arterial capillaries in the intercullular spaces of the splenic pulp in this region.

Studies were carried out on venous bloods obtained from limbs in which the circulation was cut off for periods of 2 to 5 minutes by means of a sphygomomanometer cuff. Such bloods were studied for their cell volume, oxygen capacity, and plasma proteins. Control Moods were obtained from the opposite limb where no compression was made.

No changes in the constituents studied were observed in cases where “complete compression” was used and we are forced to conclude that short periods of so called “complete block” do not produce a shift of water either from or to the tissues.

The practical application of the above is emphasized.

In the majority of patients who died with symptoms of status thymo-lymphaticus in the wards of the Babies'Hospital, the prostration of the patient was so marked that it was out of all proportion to the general physical findings. The clinical picture closely simulated a state of sugar shock from overdosage of insulin. It was, therefore, decided to determine the blood sugar level in such patients. In three cases where the diagnosis of status thymo-lymphaticus was confirmed by autopsy a sample of blood was taken within half an hour of death while the patient was in convulsions. The blood sugars were 52, 57, and 42 mg. per 100 cc. (Table I.) All three are considerably below the normal blood sugar of infants, which we have found to be between 80-120 mg. per 100 cc. as estimated by the Myers-Bailey modification of the Lewis Benedict method.

We have determined the blood sugar within half an hour of death in a series of patients with diseases other than status thymo-lymphaticus. (Table II.) The values are all normal or above normal. Hypoglycemia is not therefore a constant finding before death.

The blood sugars of six patients in convulsions of some hours'duration produced by conditions other than status thymo-lymphaticus were normal or above normal. (Table III.) It does not therefore seem probable that the convulsions alone produced the hypoglycemia.

The blood sugar was estimated in six cases of suspected status thymo-lymphaticus where roentgenograms showed a large thymus. The patients were not in convulsions at the time the sample of blood was taken. With the exception of one, which was slightly decreased, the Mood sugars were normal. (Table IV.)

The hypoglycemia, therefore, was temporary rather than continuous.

Acute suprarenal insufficiency is suggested as the immediate cause of the sudden death in status thymo-lymphaticus.

Experiments begun in 1916 by one of the authors 1 to test in vitro by the tissue culture method the comparative action of certain bacterial substances on tissues and on bacteria have been continued. In the previous studies human tissues were used and a uniform exposure of one hour was employed. In the present experiments rabbit tissue (spleen) was used throughout, and the time of exposure in the first series of experiments was twenty minutes. The antiseptics tested have included alcohol, iodine, mercuric chloride, mercurochrome, acriflavine, protargol, albargin, gentian violet, neosalvarsan, and hexylresorcinol.

The spleen of a freshly killed rabbit was divided into pieces of about 1 mm. in diameter. After washing in physiological salt solution, one lot of the fragments was placed one minute in a suspension of Staphyloccoccus aureus, and then put in graded solutions of the antiseptic for 20 minutes. Tissue cultures were then prepared in hanging drops of homologous plasma, following two washings of the tissue in salt solution. A second set of cultures were made from non-infected tissues similarly exposed with appropriate controls of untreated tissues. The comparative effect of these antiseptics on splenic cells and staphylococci is shown in the following table.

It is seen that in the majority of instances, particularly in the case of several of the newer antiseptics, bacteria proved decidedly

more resistent than cells. The most favorable results were obtained with iodine, mercuric chloride, and neosalvarsan. It is possible, however, that even in these cases the surviving tissue cells may have come from the center of the small fragments where they were partially protected.

A second series of experiments was carried out in which the antiseptic was added directly to the plasma culture medium.

The anhydrosugar, glucosane (C₆H₁₀O₅), and its polymer, tetra-glucosane, are widely used in Germany in the dietary treatment of diabetes. These substances are reputed to be oxidized by the diabetic since, according to Nothmann and Kühnau 1 and others, no extra urinary sugar results after its administration, no rise in blood sugar occurs thereafter, and an antiketogenic and protein sparing action follows its ingestion. It is likewise claimed to be glycogenic.

In the present paper we have investigated the behavior of “Salabrose,” a commercial tetra-glucosane preparation obtained from Germany, and have administered it orally, subcutaneously, and intraperitoneally to normal and phlorhizinized dogs. In order to determine whether glucosane might be excreted unchanged in the urine or feces following its administration, it has been quantitatively determined by reconversion to glucose by boiling for a short interval with dilute mineral acid. It was found that boiling for 15 minutes with 20 or 30 per cent HC1 destroyed appreciable quantities of this substance, while boiling with a HC1 concentration of 5 per cent of the same length of time completely hydrolyzed it to glucose. After neutralization the glucose obtained by this hydrolysis was determined by the usual methods. In diabetic urines this procedure was found to be satisfactory, the glucosane being equivalent to the total glucose (after hydrolysis) minus the preformed glucose.

When 30 gram doses of salabrose were fed to normal or phlorhizinized dogs no increase in the glucose excretion was noted, a result in harmony with the earlier reported German work. However, no rise in respiratory quotient was noted in normal or phlorhizinized dogs following its oral administration, indicating that only a slight oxidation, if any, occurred after its ingestion. The same amount of glucose, when fed to a fasting dog, resulted in a prompt rise in respiratory quotient.

It was demonstrated in a recent series of experiments 1 that strontium lactate administered orally to parathyroidectomized dogs greatly delays the onset of tetany, and may even bring about a permanent cure if the treatment is long continued. Preliminary experiments, carried out in conjunction with the strontium work, indicated that magnesium salts act similarly to strontium. The present paper is concerned with this phase of the work.

If parathyroidectomized dogs are given a 5 per cent solution of magnesium lactate in 100 cc. doses three times daily by stomach tube, they may be kept free from tetany for long periods, occasionally for three weeks. Some of the animals may develop tetany after eight or ten days. Once tetany becomes manifest, and the serum calcium falls below 7 mg. per 100 cc., the magnesium exerts little effect except to allay the more violent symptoms, probably through its anesthetic effect. However, if the animal is fed a pint or more of milk each day, along with the magnesium, the tetany soon disappears, generally within three or four days, and if the milk is discontinued the dog remains normal and shows no tetany symptoms so long as the magnesium is administered. Examination of the serum calcium shows that the calcium has risen to 7.5 or 8 mg. per 100 cc., which is above the level at which tetany becomes manifest in dogs.

The experiments indicate clearly that magnesium, per se, exerts little influence on tetany except through its action in sparing calcium. Somehow, the magnesium prevents the serum calcium from falling below the tetany level. A striking illustration of the sparing action of magnesium on calcium is shown in the following experiment. Parathyroidectomized dogs were allowed to develop tetany and then given the regular magnesium treatment.

In the course of some experimental work upon the effect of injections of ovarian and placental hormones into monkeys, operations have been performed at several intervals of the menstrual cycle. In four monkeys both ovaries were removed; in a fifth, large follicles were injured. The effect of these operations upon the time of appearance and the duration of the next menses provide some evidence as to the cause and nature of menstruation.

The ovaries were removed from the first monkey on the first day of the menstrual cycle (dated from the appearance of bleeding). The operation apparently had little effect for the menses continued to the fourth day. No second bleeding followed this one during a control period.

Double ovariectomy was performed upon the second and third monkeys on the 10th and 14th days of the cycle respectively. At this time the skin of the external genital organs and surrounding regions was considerably reddened. Ovulation had recently occurred in both cases and the corresponding tubes were removed and the ova recovered. 1 Apparently typical menstrual bleeding began on the 4th day after operation in the second monkey and on the 5th day in the third monkey. The duration of these menses was 8 and 3 to 4 days respectively.

Both ovaries were removed from a fourth monkey at a time when the regions surrounding the genitalia were greatly swollen and reddened. Vaginal smears for several days before the operation were free from leucocytes. Although the time of operation in this animal cannot be dated in the cycle from an observed menstruation, the condition described would indicate a time near or after the middle of the cycle, probably included between the 10th and 20 days.

The technique of intrauterine surgery by which experimental methods may be applied to mammalian embryos has recently been described. 1 The results of a preliminary investigation show that rat embryos, operated within the last eight days of development, generally survive the operation, remain viable and are born with the rest of the litter. It is possible to rear such animals although it is difficult to do so.

The first point of attack to which this method is being applied is the study of the factors influencing limb development. The first phase of this problem, the capacity for regeneration, forms the basis of this communication.

The anesthetized pregnant animal is placed upon the operating table with the dorsum toward the operator. An incision is made extending from the costal line to the anterior margin of the hind leg, through the dorsal skin and musculature of the body wall. This admits the operator to a field in which the embryos are clearly visible through the uterine wall. This method of approach obviates excessive handling of the viscera and consequent operative shock.

The incision through the uterine wall is made just large enough to permit manipulation of the limb and its extrusion through the opening. The limb is then excised at the determined level by means of fine iridectomy scissors, ligature or cautery.

The authors in a study of the effects of pyloric obstruction in rabbits 1 noted a large loss of fixed base as well as of chloride ion into the stomach. Gamble and Ross 2 found sodium in vomited stomach secretions from a dog, following experimental obstruction of the pylorus, to the extent of about one-half the equivalence of the chloride ion loss. The experiments to be here reported were undertaken for the purpose of determining whether or not fixed base is contained to an appreciable extent in gastric juice produced under conditions permitting normal gastric function. So far as we are aware the only published measurements of fixed base in gastric juice are those of Rosemann. 3 These were obtained from two large samples collected from dogs by the “sham feeding” method of Pawlow. The average amount of fixed base found in this “hunger juice” was 20 cc. 0.1 N per 100 cc., and of this approximately 11 cc 0.1 N was sodium and 9 cc. 0.1 N potassium. The chloride ion content of the samples was in average 147 cc. 0.1 N. The secretion of an alkaline juice by the mucosa of the pyloric antrum was noted by early workers and has recently been confirmed by several investigators. 4 , 5 No measurements of fixed base in this secretion have been reported.

In this study gastric secretions were obtained from isolated pouches constructed in the fundic and pyloric regions of the stomach. These pouches were provided with external fistulae but were found to retain the secretions entering them until removed by catheterization. In the table are given data obtained from samples of juice withdrawn from a pouch in the fundus at intervals following ingestion of high protein food (meat or fish) and of low protein food (cereal and cream or bread and water).

Although not commonly recognized, there exist two curious and clearly marked discrepancies in the specific action of typhoid immune serum and typhoid antigen. In 1912 Pfeiler and Rehse 1 isolated from an epidemic of the disease, fowl typhoid, an organism which they regarded as a new species and to which they gave the name, Bacillus typhi gallinarum alkalifaciens. They demonstrated the agglutinability of this organism to a high titer in typhoid (human) immune serum; also a slightly reduced agglutinability of B. typhosus in B. T. G. A. immune serum. I have shown elsewhere 2 that the organism discovered by Pfeiler and Rehse was not a new species, but the old B. gallinarum of Klein, which has masqueraded under various names. In 1915 Theobald Smith and Tenbroeck 3 confirmed the findings of Pfeiler and Rehse, using in this case the Bact. sanguinarium of Moore, which also is identical with B. gallinarum of Klein. In addition, they showed a similar serologic and antigenic relation between B. typhosus and another avian paratyphoid, the Bact. pullorum of Rettger, the causative agent in chick septicemia. I 2 have repeated the experiments of Pfeiler and Rehse, and of Smith and Tenbroeck, and have studied the serologic relationships of these avian types, not only to B. typhosus, but also to a number of other cultures coming from avian or mammalian sources. A resumé of the reactions usually observed, so far as they concern the subject of this paper, are presented in the following table.

In view of these peculiar and easily observed antigenic relationships between B. typhosus and the avian paratyphoids it was a matter of interest to ascertain to what degree, if at all, similar relationships might be present in the lytic principle reactions, active and passive, between the same cultures.

In studying possible modifying effects on spermatogenesis, young adult male rats were injected subcutaneously and intraperitoneally with (1) freshly macerated testis suspensions and (2) with a suspension of rat spermatozoa. Subsequent matings were made with all injected males and the testes were later studied histologically. Agglutination tests were not made.

It was found that one to five relatively heavy injections of these materials produced no specific effect on testicular activity within the limits of the experiment (three days to four months after the last injection). Smears from testes of previously injected males showed sperm motility, and litters obtained from matings proved the spermatozoa physiologically normal.

Formed materials injected into the abdomen at times resulted in the production of cysts or abscesses sometimes located in the scrotum. When these latter were sufficiently large to cause displacement of the testis into the abdomen, the testis was found to be degenerate. Such degeneration, however, was not due to serological influences of the injected materials, but solely to the mechanical displacement of the organ; removed from the scrotal regulatory influences degeneration of the testis follows rapidly. In the abdomen degeneration is brought about through the exposure to higher than normal temperature for the testis and this degeneration should be carefully distinguished from supposed serological influences.

The multiplication of yeasts in pure nutrient solutions has already received much study; as a result there are many discrepancies in conclusions which have been reached. This brief report deals with two phases of the question: 1, the effect of concentration of an accessory containing substance, and 2, the effect of aeration on the multiplication of yeasts in pure nutrient solutions.

In both of these studies, Fulmer and Nelson's Medium E 1 was used. To study the first part of the problem Fred, Peterson and Davenport's yeast water medium 2 was added to medium E in different dilutions, as follows: 1, 100 cc. medium E + 10 cc. sterile water; 2, 100 cc. medium E + 10 cc. yeast water medium; 3, 100 cc. medium E + 10 cc. of a 1-10,000 dilution of yeast medium, and 5, 100 cc. of a medium E + 10 cc. of a 1-1,000,000 dilution of yeast medium. These cultures after sterilization and inoculation were incubated under the same conditions. Saccharomyces anomolous, Torula colliculosae, Willia satunus, Zygosaccharomyces mandschuricus and Pichia farinosus were used.

The results showed that when the yeast water medium was present in greatest amounts there was an acceleration in multiplication, but when it was added only in minute traces, there was little difference between these flasks ad the control. There was no alteration in multiplication accelerating effect by “hating the yeast medium for 60 hours at I 5 lbs. pressure (121 .5° C). It is believed that the acceleration in multiplication observed when large amounts of yeast water were added was due to chemically definite food substances and that one need not place the explanation on the presence of “bios.”

The data collected under the conditions of these experiments indicate a marked difference in resistance of pure yeasts to sodium benzoate. On dextrose agar and in dextrose broth even in the presence of 0.40 per cent of sodium benzoate, growth was appreciable. Torla communis did not grow after the concentration of the benzoate in the broth had reached 0.15 per cent. One strain of Sacch. ellipsoideus was also inhibited at this concentration. Another strain showed a greater resistance; it grew in the presence of even 1 per cent but the growth was slower and less abundant. The fungi used seemed to be able to tolerate the sodium benzoate better when growing on a solid medium containing it than when growing in a liquid medium. This is probably explained by the fact that on the solid medium the organisms may use up the disinfectant during the early stages of growth so that the cells which develop later in the history of the culture do not have to endure such a high concentration as in the beginning. In liquid media such as dextrose broth, the sodium benzoate is in solution and therefore may come into closer contact with the cells. Two of the strains of budding fungi, Mycoderma uini and a pure culture from vinegar (No. 18) were able to grow apparently as well in the presence of one per cent of sodium benzoate as in the controls.

In sterilized apple juice the results were more striking. The sterile apple juice was almost as good a medium for the yeasts as the sterile dextrose broth since in the control tubes without any preservative the growth was abundant. However, when sodium benzoate was added the growth was markedly inhibited since none of the fungi grew after 0.01 per cent had been added.

Our experiments show the presence of albumin globulin and alkali soluble protein in the colon.

The toxicity of the water solubk protein shows marked deviation from that of the colon itself.

The work on the colon bacterial proteins will be continued on much larger amounts of the dried organism.

In the course of our investigation of the body righting reflexes in the fowl, a detailed report of which will appear shortly, we had one animal which as a result of a cranial operation assumed a peculiar posture at certain times. It would keep its head down and tail up, so that the long axis of the body was at an angle about 60° to the horizontal. When in such a condition, the animal invariably walked backward and continued to do so till it would hit the wall of the room. But when it kept its head in a normal position it could and did walk forward.

In order to find out the mechanism of this phenomenon we studied a number of normal Leghorn and Sebright Bantam fowls in which the lower portion of the neck was kept flexed upon the trunk by means of a gauze bandage. We found that these animals, without exception, in attempting to move walked backward. After the bandage was removed they immediately returned to the normal method of progression. Such test could be performed repeatedly on the same animal with similar results.

To determine whether this retrogression was not due to voluntary movements of the nature of escape reaction, we performed these tests on the decerebrated fowl. The results were the same as in normal animals, indicating that this phenomenon is of a reflex nature (forced movement).

We canclude, therefore, that normal progression in the chicken is dependent upon the posture of the neck, or the tonic state of the neck muscles. This ability is lost as soon as the posture of the neck is changed (neck ventroflexed upon the trunk), and progression is replaced by retrogression.

1. Blood from cases of hypertension contains a pressor substance.

2. The blood proteins by virtue of their anaphylactic reaction and the intact vagi of the experimental animals inhibit the otherwise obvious pressor response. Henlce these factors should be eliminated in such experiments.

3. No pressor effect of any kind was ever seen by us from the injections of comparable amounts of saline, rabblits'blood or human blood of people with normal blood pressures.

Rapidity of digestion of the protein contained in different cereal breakfast foods was determined in human subjects, partly by the emptying time of the stomach as measured by the total nitrogen in its contents obtained at successive intervals after a cereal breakfast, and partly by the rate of nitrogen excretion through the kidney following the meal. The former method obviously gives information chiefly upon the motility of the stomach.

Each of four subjects ingested 25 grams of dry cereal, after cooking a uniform length of time, with cream and sugar and in some cases coffee. Before the meal a Rehfuss tube was introduced and the fact established that the stomach was empty. The tube was left in place and one hour after the meal the contents of the stomach were completely removed by the use of large amounts of wash water. Each subject repeated the experiment on the following morning with the same cereal and with exactly similar technique, except that the stomach was emptied at one and a half hours. On succeeding mornings the time was 2 hours, 2 1/2 hours, etc., until the stomach was empty. The next week each subject took a different cereal with another change for the third, and still another for the fourth week. The cereals employed were a milled wheat product, a whole wheat product and two oats products.

The differences in stomach emptying time amongst the four cereals were small, much less than those found amongst the different subjects for the same cereal. An average of all the subjects for all the cereals indicated that at the end of one hour 35 to 45 per cent of the ingested cereal protein had passed out of the stomach, at two hours about 75 per cent and at three hours the average stomach was empty.

1. The rate of glycogen formation in the liver during the absorption of fructose is slightly greater than during the absorption of glucose and leads to a higher glycogen maximum.

2. The liver plays a larger role in removing fructose from the blood stream than glucose.

3. Large doses of insulin almost completely suppress the glycogen formation from fructose.

4. Galactose is very slowly converted into glycogen and is excreted to a large extent in the urine.

1. The intravenous tolerance of male rats, fasted previously for 48 hours, was close to 2.5 gm. glucose per kilogram per hour.

2. Large doses of insulin raised the tolerance to 3.0 gm. glucose per kilogram per hour.

3. The tolerance for fructose, when infused into the femoral vein, was close to 0.35 gm. per kilogiram per hour.

4. The tolerance for fructose, when infused into a mesenteric vein, was between 0.6 and 0.8 gm. per kilogram per hour.

5. Insulin had practically no effect on the intravenous tolerance for fructose.

Four normal persons, two men and two women, have served as subjects for this investiagtion. On each, the following observations were made once a week throughout the year (February, 1925—February, 1926) under basal conditions between 8 and 10 a. m. The observations and their apparent course throughout the year follow:

1. Oral temperature; no certain periodicity.

2. Reclining systolic pressure: all show a marked depression in the spring and in three cases there are also low points in midsummer and late fall—early winter.

3. Reclining pulse: lowest in summer.

4. Alveolar air:

(1) Oxygen, uncertain, with perhaps a tendency to be lowest in summer;

(2) Carbon dioxid, highest in the summer in all subjects and rhythmically coincident with the menstrual cycle in the women, the low points coming just before the onset of menstruation.

5. Pulmonary ventilation: both the tidal and minute volumes show a marked depression in the spring with a rise to maximum in late summer and early fall.

6. Expired air: the percentages of oxygen and carblon dioxid vary inversly, the former showing a marked low point in spring and a maximum between July and September, while the latter is highest in the spring and lowest during late summer and early fall.

7. Carbon dioxifd per minute: nlo evident periodicity.

8. Oxygen per minute and calories per square meter per hour: lowest in the period, July to September.

9. Respiraltory Quotient non-protein) : with one exception seems to be highest in the summer and early fall.

Under the same basal conditions and either just preceding or following the above determinations 20 cc. of blood were taken from an arm vein.

This is to report the effect on the blood sugar level of cats of a “heat liability,” produced by giving ice water by stomach tube, according to the method devised by Cannon. Eight animals have been used and all have given concordant results, of which those for one cat shown in the figure are typical. From inspection of the curves (and deducting the effect of the similar procedure in which water at body temperature was given), it appears that such a heat liability induces a pronounced mobilization of carbohydrates and increase of blood sugar value. In spite of the fact that the fall in body temperature is immediate, this mobilization of sugar seems to get under way slowly and is only noticeable after an hour or more. From then on it rises slowly to a maximum during the next hour or two and then, even more slowly, falls; in only two cases out of the eight did the blood sugar value return to normal within the five or six hours during which the animals were kept under observation.

More than two years ago we began experiments to determine whether it would be possible to prolong the life of adrenalectized animals by the administration of adrenal preparations.

Adrenalin injected intramuscularly or subcutaneously every hour or two after the removal of both adrenals did not prolong life. The average length of life for the 7 cats studied was 19.7 hours. The amounts injected at one time were from 0.25 cc. to 0.5 cc. of 1:1000 adrenalin intramuscularly or subcutaneously. The adrenals had been removed from these cats at one operation.

The remainder of the experiments were conducted upon cats whose adrenals had been removed at different operations from two to ninety days apart. The operation was through the lumbar path.

Five cats which were injected intravenously or intramuscularly with material obtained by ether extraction of the dried cortex (beef) lived an average of sixty hours. Two cats injected intravenously or intramuscularly with glycerol extract of the cortex lived an average of forty-five hours.

Nine cats which were injected intramuscularly or intraperitoneally with alcohol soluble material from the cortex lived an average of forty hours. One animal receiving intraperitoneal injections of olive oil extract of the cortex lived ninety-three hours.

The average length of life of all of these cats (17), not including those injected with adrenalin, was 49.8 hours.

On the other hand, 22 adrenalectomized cats injected intravenously, intramuscularly, intraperitoneally or subcutaneously with whole sallinle extracts of cortex lived an average of 146.6 hours.

1. Our values except those for levulose agree with the data of previous investigators.

2. Levulose has a higher reduction value than glucose.

3. The difference in behavior of galactose and glucose toward Folin-Wu copper solution is not due to difference in velocity of reduction or differences in pH, but to structural differences in the two sugars.

4. Glucose picric acid standards used in the Benedict-Lewis methods are stable for a period of at least two years.

One of the most important factors in the preparation of diabetic diets, as well as in the study of the relationships between diet and sugar metabolism, is the carbohydrate content of the injested foodstuffs. The most extensive data on this subject is found in Bulletin No. 28 of the U. S. Department of Agriculture. Unfortunately the carbohydrates in the majority of cases have been determined by difference and hence do not differentiate between the free and hydrolyzable sugar content. Furthermore, since the carbohydrate content of fruits and vegetables varies between wide limits, it is often necessary to analyze the diet used, rather than depend upon figures gathered at random. Of the various procedures for the analysis of the carbohydrate content of foodstuffs the method of Dutcher and Thomas for reducing sugars in plants seemed to adapt itself most readily to the rapid analysis of food materials.

1. Dutcher and Thomas 1 method for analyzing the free sugar content of plant material has been adopted to the analysis of the free carbohydrates in general foods and fruits.

2. Lloyd's reagent may be substituted for mecururic nitrate as a clarifying agent and the Benedict-Osterburg or Lewis-Benedict methods are applicable in the presence of alcohol.

3. A four-hour extraction period with 75 per cent alcohol is sufficient to extract completely 2,000 mgs. of glucose, fmctose, galactose, xylose, and arabinose.

4. Maltose, lactose, raffinose and sucrose are extracted by 75 per cent ailcohol.

5. Sucrose is not appreciably hydrolyzed by a four-hour extractioln with 75 per cenlt alcohol, nor as pointed out by Dutcher and Thomas, is it hydrolyzed by 15'boiling with picric acid and Na₂ CO₃.

6. The analysis of a few common food materials for free and hydrolyzable carbohydrates is presented.

When normal urine is treated with phenylhydrazine a reddish compound can be isolated. This compound is non-crystalline, appearing under the microscope in the form of round masses. This substance on treatment with chloroform can be separated into two fractions, the one soluble in chloroform, an amorphous deep red powder, the other insoluble in chloroform, a light yellow compound. The latter compound crystallizes from water ion clusters of short stocky needles. It also may be purified by recrystallization from 70 per cent ethyl alcohol. The ozazone melts at 150-153°C (corr.) with decomposition, and is soluble in pyridene. On treatment with acids the compound yields furfural. The substance contains nitrogen, but so far no satisfactory analyses have been obtained. It does not correspond in its properties or crystalline structure to known hexose, pentose or glucuronic acid ozazone. Further work is in progress on the identification of this compound.

As a result of the recent experiments of Bollman, Mann and Magath 1 there is now little doubt that the liver plays the leading rôle in urea formation. Nevertheless very little is known concerning the mechanism of urea formation. To be sure we have long known from the experiments of Kossel and Dakin 2 that argininie may be split directly into urea and ornithine under the action of a specific enzyme, arginase, but relatively this accounts for only a small amount of the urea formed.

We wish to record observations of autolysis experiments carried out some years ago 3 (1914) which we have recently confirmed and extended 4 . The original motive for initiating these experiments was the availability of a specific method for estimating urea (Marshall). We were able to show then that whereas urea formation took place in autolyzing liver tissue, no change in the concentration of urea occurred in muscle tissue similarly treated.

Since our first experiments were carried out somewhat similar results have been obtained by several workers. Hoagland and Mansfield, 5 contrary to their expectations, obtainea entirely negative results in muscle autolyses. Fosse and Rouchelman, 6 employinlg the xanthydrol method, were able to demonstrate a six fold increase of urea in autolyzing liver tissue, a change which mas entirely inhibited by boiling the liver. Fosse 7 believes that the intermediate substance in urea formation is cyanic acid. McCance 8 has endeavored to ascertain the optimum conditions for urea formation in autolyses. He reports urea formation not only in liver but also in spleen.

Data illustrating the results which were obtained in our earlier experiments are given in Table I. As will be noted there is no evidence of any urea formation in the muscle autolyses, but in the liver autolyses colncentration of the urea increased more than ten fold.

The mydriatic effect of cocaine is regarded by some to be due to a stimulation of the peripheral sympathetic nervous mechanism supplying the radial muscle of the iris. By others it is regarded as the result of a direct weakening of the circular muscle of the iris.

The first group of experiments here reported were performed in an attempt to demonstrate by actual measurements such a weakening of the circular muscle by cocaine. Accurate pupillary measurements were obtained by measuring the image of the pupil on a scale in the eye-piece of a measuring telescope. Series of measurements made in this way were employed for the plotting of curves showing the development of drug effects.

Albino rabbits were the animals employed. In the study of cocaine, two drops of a four per cent solution were placed in one eye only. After five to ten minutes there was placed in each eye a definite amount of solution of physostigmine sulphate. Frequent measurements were made of both pupils, and curves were plotted to show the effect exerted by the cocaine on either the degree of the physostigmine effect or on the rate of its development.

Thirty-six experiments were carried out, some on animals from which the superior cervical sympathetic ganglion on one side had been removed, and some on animals on which this operation had been performed on both sides. A few experiments were on unoperated animals. This work gave no evidence of a weakening of the circular muscle of the iris by cocaine.

The second group of experiments is a study of the effect of cocaine on the isollated sphincter nluscle of the iris, mounted in Locke-Ringer solution for the recording of its muscular activity as a kymographic tracing. Steers and large dogs were used.

In previous experiments 1 it was pointed out that when betamethyl hydantoin is given subcutaneously to dogs only fifteen per cent of the compound reappears unchanged in the urine. A similar amount is excreted as methyl hydantoic acid. But while all of the nitrogen of the compound was excreted, the remainder was unaccounted for in the nitrogen partition. Immense amounts of oxalic acid were found in the urines during experimental days, and this suggested hydrolysis of methyl hydantoic acid to methyl urea and glycollic acid, with oxidation of the latter to oxalic acid. Supplementary experiments have thrown doubt upon excretion of a large amount of methyl urea, hence the possibility of origin of the oxalic acid found from a substance such as methyl parabanic acid, in the method of determination, is being considered.

Experiments have also been carried out in which hydantoin and hydantoic acid were administered. In keeping with the findings of Lewis 2 no evidence of urea or ammonia formation was obtained. When hydantoin was given there was, however, an increase in organic acid excretion corresponding to elimination, as hydantoic acid, of one-half of the compound administered. Similarly, the apparent rise in creatinine excretion, due to elimination of unchanged hydantoin, was only half as great as it would have been had all of the compound been excreted unchanged. The rise in total nitrogen again accounted for all of the compound, and the amount of benzal hydantoin, 3 isolated from the urine after hydrolysis, justified the assumption that all of the compound is excreted either unchanged or as the corresponding acid.

While hydantoin therefore undergoes partial hydrolysis to the corresponding hydantoic acid, it is very resistant to further destruction in the body, as compared with the previously investigated methyl derivative.

“Baker's yeast” as obtainable in Ceyloln and in England is not a pure culture of yeast, but is generally composed of two or more species of Saccharomyces an'd one or two or several species of bacteria diving in close association. The whole yeast as a rule gas-fermented more carbohydrates than any of the isolated germs did, either Saccharomyces or bacteria. 4 bacillus, which produced only acidity, not gas, in maltose, would produce gas when Saccharomyces was added to it, was inactive on maltose and was isolated from the same sample.

In recent years I have used for my experiments mostly pathogenic bacteria and also a pathogenic yeast, Cryptococcus graciloides.

Simple acid fermentation of certain carbon compounds caused by micro-organisms which never produce gas may in a number of instances be transformed into fermentation with the productioln of gas by adding a micro-organism which, alone, is inert, viz., produces neither acidity nor gas on those particular carbon compounds. For example :B. dysenteriae Flexner produces acidity, never gas, in maltose and mannitol; B. morgani produces neither acidity nor gas in those two carbohydrates; the mixture B. dysenteriae Flexner + B. morgani producrs, however, acidity and gas. For the phenomenon to take place several conditions are necessary; the most important one seems to be that the added bacillus, though inert on those particular carbon compounds, must be capable of producing fermentation with gas in glucose.

Earlier experimental observations 1 and certain clinical experiences indicated the advisability of making an electrocardiographic study of the behavior of the heart under the influence of propylene, the results of which are here reported.

In four dogs and three cats the administration of propylene in concentrations from 25 per cent (less than sufficient to induce or maintain anesthesia) up to 40-50 per cent caused ectopic∗ ventricular beats, which promptly ceased when the gas was discontinued or the concentration lowered. With higher concentrations the ectopic beats usually became more and more numerous, often so much so that an ectopic ventricular tachycardia resulted. Under these conditions also discontinuance of the gas or a sufficient lessening of its concentration resulted in a return to the normal mechanism. In control observations with ether, chloroform, nitrous oxide, ethylene, and acetylene in varying concentrations ectopics were not observed.

While cooperating with Drs. Isidore Cohn and I. I. Lemann during the course of some experimental work on rabbits, observations were made on the blood platelets following splenectomy and transplantation of the spleen.

Under ether anaesthesia the spleen was removed and the whole organ immediately transplanted beneath the rectus sheath of another etherized animal. After a period of 25 to 35 days the animals were killed. At post-mortem the grafts were found to have been completely destroyed and largely replaced by a necroticmass, all transplants being considered non-functional.

Blood platelet counts were made before operation and at intervals of 24 to 48 hours after operation until the animal was killed, covering the period of 25 to 35 days. The average number of platelets in all animals before operation was found to be 375,000 per cm. of blood. Forty-eight hours after operation there was an average of 475,000 per cmm. in the spleenectomized group, and an average of 573,000 per cmm. in the group of recipients. During the first ten days after operation this primary increase was sustained, the number of platelets being approximately the same in both groups on the tenth day, averaging 476,000 per cmm. in the donor group and 466,000 per cmm. in the group of recipients. After the tenth day the number of platelets gradually decreased in the recipient group, reaching the preoperative level about the twentieth day and remaining normal thereafter. In the other group the increase was sustained throughout the entire period of postoperative observation. Two animals of this group were observed over a period of 35 days after operation, at the end of which there was approximately 500,000 platelets per cmm. present in each case, or approximately 125,000 per cmm. more than before operation.

The very similar chemical properties, correlated with closely corresponding structural configurations of the pentoses and hex-oses, make difficult the elaboration of chemical methods for the determination of one in the presence of the other. The resistance of pentoses to fermentation by ordinary baker's yeast has furnished a means that has been used at times to separate the two types of substances for analytical purposes. In the present study this procedure has been employed to determine the pentoses in the blood following their administration intravenously. The reducing power of normal blood can not be removed completely by fermentation with yeast, but since this residual reduction seems to remain small and fairly constant no real difficulty arises. It is scarcely unjustifiable to assume that in the presence of pentoses the difference from the usual unfermented reducing power of the blood furnishes a reliable index of the amount of pentoses present.

The details of this series of experiments may be summarized as follows: One gram of xylose in 15 cc. of solution was injected into the marginal ear vein of a 2 to 2.4 kg. fasting rabbit. The blood samples (2 to 3 cc.) were collected at the following intervals: one the preceding day, one just before and one just after the experiment, others after one, two, four and eight hours, respectively. These were analyzed for non-protein nitrogen, blood sugar and reducing power after fermentation. For this latter determination the whole blood was incubated one hour at 37 C. with a suspension of yeast. Then the proteins were precipitated according to the Folin-Wu procedure, and the reducing power of the filtrate determined by the Shaffer-Hartmann method, the results being expressed in terms of glucose.

Recent work by Ashman 1 and by Ashman and Woolley 2 has demonstrated that the compressed or injured cardiac muscle of the turtle may, after excitation, recover its conductivity along a curve which passes through a supernormal phase. Thus an impulse traversing the compressed muscle will travel most rapidly during the supernormal period. If the degree of compression is sufficient only impulses arriving at the compressed region during the supernormal phase are transmitted; earlier or later ones are blocked. These experimental observations add much weight to the argument for a supernormal phase in the recovery of conductivity in the human hearts discussed below.

Lewis and Master 3 reported two cases of block in the human heart which they interpreted as evidencing a supernormal phase. In their first case typical complete heart block was interrupted at times by a ventricular systole in response to an auricular impulse and only when the latter arose during late systole or early diastole. Auricular impulses arising later in diastole were blocked.

We have obtained electrocardiograms of two clinical cases of heart Mock which show phenomena attributable to a supernorind phase. Our first case is similar to Lewis and Master's excepting that, (a) a series of auricular impulses are transmitted if the auricles are rapid; (b) ventricular standstill supervenes when, as a result of auricular slowing, conduction fails, and (c) the period of ventricular quiescence is interrupted at irregular intervals by idioveiitricular contractions which do or do not permit conduction depending upon their time relations with the auricular systoles. As in Lewis and Master's case, Conduction time increases slightly as the interval between the K-wave and the subsequent P-wave widens.

In reviewing the iodine determinations of drinking water from the State of Utah made in this laboratory (some of these determinations being made by J. C. Hathaway, H. H. Vanderslins, Clarence Dahl) it was found there was not a very close relationship between iodine and freedom from goiter. In some cases a number of determinations were made on the same water supply and the iodine content was found to vary at different times.

In the case of the water supply of Stanford University a chemist who had worked on these iodine determinations actually collected and evaporated the water and yet it varied considerably during different months.

Since the enlargement of the thyroid is the result of iodine deficiency during several years, even though the child has not changed its residence during this period, the immediate iodine content of the water supply would not be expected to bear a very close relation to the size of the thyroid. If, however, a large number of cases are taken it is always found that where there is a low average iodine content of the water there is higher average incidence of goiter among school children as shown in the following table:

In this table there are twenty-six towns. Thirteen with low iodine show an average of 52 per cent goiter in the schools. The thirteen with high iodine show an average of 39 per cent goiter in the schools. One town, Huntsville, in Weber County, with 41.1 per cent goiter in the school, showed on one analysis high iodine (7.07)) and in another analysis low iodine (0.22). It was included in the table, and yet its inclusion in either the high or the low would not materially affect the average.

Three rats were fed on diets of purified casein, glucose (cerelose) and cod liver oil, with a salt mixture of NaCl, CaCO₃, K₂HPO₄ and Fe₂(SO₄)₃. Analysis of the ash of samples of these diets showed that they contained 1.2 mg. magnesium per 100 gm. of diet. The rats were kept on the diets 20, 20 and 100 days respectively. The ash of the rats contained 23, 14 and 31 mg. magnesium, or 0.030, 0.029 and 0.026 per cent of their respective body weights. The corresponding control rats contained 24, 15 and 36 mg., which was .039, .043 and .037 per cent of their respective body weights. Estimating that rat No. 3, which weighed 52 gm. at the beginning of the experiment, contained about .040 per cent magnesium, or 21 mg. magnesium, it gained approximately 10 mg. During the same period, it ate about 1000 gm. diet, containing 12 mg. It must, therefore, have absorbed and retained nearly all the magnesium it received. This is in marked contrast to its inability to utilize all of the Ca and P of a diet, even when they are present in inadequate amounts.

In the present paper I wish to make a brief report of results obtained in an effort to produce a pneumococcic antitoxin. This work is the outgrowth of our observation that bacterial toxins are rendered non-toxic by soaps of some of the unsaturated fatty acids, particularly sodium ricinoleate, and that the antigenic properties of such detoxified toxins remain unimpaired. We have further called attention to the fact that pneumococci, streptococci and certain other micro-organisms are either killed in a short time in aqueous solutions of sodium ricinoleate, or lose their powers to infect. Larson and Nelson 1 showed further that there is a rapid development of agglutinins in rabbits against both pneumococci and streptococci following the injection of soaped cultures of these organisms.

Ten cc. or more of a virulent broth culture of soaped pneumococci may be injected intraperitoneally into rabbits without causing an infection. Agglutinins appear in the blood stream as early as 24 hours following such injections. This observation led us to attempt to produce an antipneumococcic serum by this method. Pneumococci of various types were passed through a series of mice in order to increase their virulence as much as possible. They were then grown in veal infusion broth which contained 1 to 2 per cent peptone. Parke-Davis'and Witte's peptones were used, since these products have been found by others to be best for toxin production. The whole cultures were treated with sodium ricinoleate solutions of a pH of 7.8, so the final concentration of the soap was 1 per cent. The pH of the cultures was adjusted to 7.4 before the soap was added.

Rabbits and sheep were then immunized with these soaped cultures. Sheep were given as much as 200 cc. at one time.

It is well known that hemorrhage causes fluid to enter the blood from the tissues. Likewise it is well known that colloid-free fluids on injection rapidly leave the blood and pass to the tissues. It was the object of this study to determine, if possible, whether certain tissues are of more importance than others in these interchanges. The experiments were carried out on cats anesthetized with ether. The tissues chiefly worked with were skin (clipped free from hair), muscle, intestine, spleen, liver and blood. Both kidneys were ligated off. Corresponding parts on each side of the body were prepared and when all was ready, the small pieces from one side were snipped off and dropped into stoppered weighing bottles. Then the experimental procedure was performed and after thirty minutes the corresponding pieces from the opposite side of the body were taken. The samples were all dried to a constant weight at 102°. The experimental procedure consisted in a hemorrhage of 15 cc. per kilo or in the injection of 25 cc. of the various solutions per kilo. The calculations as to the amounts of the various tissues in the body of the cat were based on the figures of Voit. 1 The amount of blood in the cat was taken as 4.6 per cent of its weight and the amount of water entering or leaving the blood calculated from hemoglobin and hematocrit alterations.

The results are as follows: Following the hemorrhage all tissues examined gave up water, but of the amount of water entering the blood, about 78 per cent came from skin and subcutaneous tissue, 10 per cent from muscle, 8 per cent from the liver, 2 percent from the intestine and .03 per cent from the spleen.

Lumbricus terrestris, under the conditions stated, survived a loss of 43 per cent in body weight during 5 hours of exsiccation but did not survive longer periods with losses of 50 to 60 per cent. Nevertheless, the curves of body weight during exsiccation and after replacement in water appear similar in all cases (up to 24 hours). Measurernetits of cross-sectional areas indicate that during exsiccation in the earlier stages the epidermis, body muscles, celomic cavity intestinal wall and lumen lose roughly in proportion to their size. But in extreme exsiccation a relatively larger proportion is contributed by the celomic cavity and the intestinal lumen. The loss in the tissues appears chiefly from the intercellular spaces, as the cells (excepting the epidermis) show but slight changes in size and structure.

We have determined lactic acid in a number of serous fluids and in samples of blood taken at the time of paracentesis. Determinations were made immediately by the method of Clausen.

The exudates all showed a considerably greater lactic acid content than that of the blood, the highest values and greatest excess being found in frankly purulent fluids and in sero-fibrinous exudates (streptococcus) which later became purulent. The transudates yielded relatively low figures, usually slightly below the blood values.

The accumulation of lactic acid in inflammatory exudates may be due to bacterial action or to the action of leucocytes on glucose and other metabolites, as noted by Levene and Meyer. 1 In malignant exudates its presence, in concentrations greater than that of the blood, would be explained by the observation of Warburg, 2 whose study of the respiration of malignant tissues showed a production of lactic acid by cancer cells many times greater than that of normal tissues.

To 3 grams of wheat starch, ground according to the method of Alsberg and Perry, 1 100 cc. of distilled water was added at room temperature. This solution was centrifuged for 10 minutes and after removal of the precipitate, centrifuging was repeated twice. To one part of the amylose solution thus obtained two parts of 96 per cent alcohol was added the same afternoon and the precipitate with supernatant liquid was allowed to stand for about 3 weeks. In the second week to a part of the precipitate and liquid an equal volume of 96 per cent alcohol was added, giving an alcohol concentration of 80 per cent.

After 3 weeks the precipitates were examined in alcohol under the microscope and were found to contain many spherocrystals. If alcohol is added in excess to the amylose solution a refractive myelin precipitate, consisting of smaller or larger irregular lumps, is formed. This myelin precipitate was later partly present as such but all gradations between amorphous masses and perfect spherocrystals could also be found. The latter consist of separate, very refractive radial needles attached to a central point or nucleus. The needles are about 1μ thick and up to 25μ long. The spherocrystal is formed by several or numerous needles. If the crystal is less perfect, lumps of the less refractive myelin precipitate are still present at its center, so that the needles project only half their length, or the top of the needle is not pointed but formed by a lump that is still amorphous, or the needle itself remains thick and less refractive than the perfect needle, with a more refractive axis in the center. The precipitate from 80 percent alcohol showed more numerous and more perfect spherocrystals than the precipitate from 64 per cent alcohol.

It has been shown for the first time that small amounts of tuberculin fractions prepared from protein-free synthetic substrates are capable of sensitizing normal, non-tuberculous guinea pigs. In these animals, typical skin reactions can he elicited subsequently by intracutaneous injection of minute amounts of homologous as well as heterologous fractions and of unfractionated tuberculin form which such fractions have been prepared.

Clinical trial of these fractions in juvenile patients has demonstrated for the first time that the substances have diagnostic value. Positive skin tests were perfectly correlated with positive clinical and laboratory findings of tuberculous infection. Such was not the case for ordinary old tuberculin in over 20 per cent of young patients in a group of 150 who were studied routinely. In a number of instances the observations suggested that the reactions might be correlated with the degree of activity or with arrested tuberculous infection.

The present studies point to a method which makes an accurately standardized tuberculin available for routine clinical work. It is believed that the active principle of tuberculin can best be isolated by this or by a similar method.

The observations on sensitization suggest that these may be applied to studies of immunity, and by extension, to therapeutic studies in the experimental animal. Such work, now in its third year of progress, will be reported in the future.