Experimental Biology and Medicine

Trendelenburg and Goebel 1 demonstrated that serum from thyro-parathyroidectomized cats, when used as a nutrient fluid for the isolated heart of the frog, diminished the amplitude of contraction of the heart, as compared with serum from normal cats. They attributed this effect to a reduction in the total calcium of the serum, with a corresponding decrease in ionized calcium, but were unable to rule out the possibility of the formation of an un-ionized compound of calcium, presumably by combination with an organic acid.

Two of the authors have shown that the sensitivity of the frog's heart to changes in the concentration of calcium is specific for ionized calcium, bound calcium being inert with respect to the preparation, and have utilized this property of the heart for direct quantitative estimations of Ca⁺⁺ concentrations in biological fluids. 2 They have also shown 3 that the ionization of calcium in the protein-containing fluids of the human body follows, as a first approximation, a simple mass-law relationship, expressed by the equation

from which it is possible to calculate Ca⁺⁺ concentrations, from analysis for total calcium and total protein, with a degree of accuracy at least as great as that of direct observation by the frog heart method.

Using the methods of direct observation and of calculation of Ca⁺⁺ concentrations, the authors have studied the influence of thyro-parathyroidectomy and of the administration of the parathyroid hormone upon the ionization of calcium in the serum of the cat. The cat was chosen as the experimental animal because its serum was found to be especially favorable for observations by the frog heart method, which is not the case with the dog, and because the state of calcium in its serum was found to be accurately described by the mass-law relationship above referred to, which is not the case with the rabbit.

Dixon, 1 Dale 2 and Loewi 3 originally established acetylcholine as the humoral agent that produces motility of the gastro-intestinal tract. Later research, recently summarised by Cannon 4 and Dale 5 has shown that acetylcholine is liberated by stimulation not only of parasympathetic nerves and sympathetic ganglia, but probably even with stimulation of striated muscle. Fröhner 6 reported that eserine was commonly used to produce increased peristaltic activity of the gastro-intestinal tract in animals; and since the work of Hunt, 7 Fühner 8 and Simonart, 9 it has been known that eserine added to acetylcholine greatly fortifies the action of acetylcholine. Matthes 10 and Loewi and Engelhart 11 have added to our knowledge that eserine prevents the hydrolysis of acetylcholine by blood and tissues through inhibiting an esterase. In the light of this knowledge, the authors investigated the effect of eserine and acetylcholine on ileus following peritonitis (to be published). In this research it was found that in peritonitis acetylcholine given intramuscularly in small doses produced either no effect or a very small one, but it was always effective when given with eserine. The present research was undertaken to investigate whether or not the stimulation of salivary secretion would produce motility of the intestine through liberation of acetylcholine from the stimulated salivary glands. Babkin, Alley and Stavraky 12 have demonstrated that acetylcholine is liberated into the venous blood when the chorda tympani was stimulated.

Healthy dogs under light ether anesthesia were used. Both chordae tympani were prepared and stimulated with weak faradic currents from a Harvard inductorium. Balloons were placed in the stomach, ileum and upper part of the colon, and connected with water manometers. Blood pressure was recorded in the usual way.

It appeared to the author that it would be of considerable importance if the fundamental behavior of cardiac and skeletal muscle could be more closely correlated by the explanation of an apparent difference in their behavior following changes in initial fiber length.

Evans and Hill 1 showed that the heat produced during a short tetanus of the frog's sartorius increases as the muscle is extended rising to a maximum at about the in situ resting length, and then decreases upon further extension.

Starling and Visscher, 2 utilizing the isolated heart-lung preparation, showed that the oxygen consumption of the mammalian ventricle rises with increasing diastolic volume, but found no indication of a maximum with subsequent decline, although they explored a very wide range of alterations in ventricular volume. It is obvious that these experiments involved the measurement of the total oxygen requirement of the heart, while in the myothermic experiments quoted above only that excess metabolism due directly to activity was determined.

Feng 3 reported that the resting heat production and oxygen consumption of the frog's sartorius both rise with increase in the length of the muscle. These observations might have been anticipated in the light of the experiments of Clark and White, 4 who showed that the oxygen consumption of the quiescent, filled auricle of the cold blooded heart approached that of the beating, empty auricle.

The dangers of prolonged tourniquet application are as follows: 1. Interference with nutrition of the part sufficient to cause gangrene; 2. Paralysis, possibly due to prolonged pressure on the nerves; 3. Circulatory depression which frequently occurs after the release of a constrictor. Paolucci 1 and Fogliani 2 observed in animals that this circulatory depression was followed by death in many cases.

The present study is an attempt to evaluate the possible factors producing this fall of blood pressure and death. The factors considered were: 1. Return of blood to the limb (if this had been expressed before application of the tourniquet); 2. Reactive hyperemia (Bayliss); 3. Absorption of toxic metabolites; 4. Swelling of the limb by transudation.

Experiments were done on 19 dogs anesthetized with sodium barbital. One hind leg was constricted at the level of the hip joint for periods of from 2 to 20 hours. Six animals recovered after tourniquet applications of 2 to 6½ hours while 13 dogs died after constrictions of 3 to 20 hours. Seventeen percent of a control series of dogs died from the prolonged anesthesia.

In the animals dying, the hind quarters were symmetrically removed after the method of Blalock. 3 There was an average increase in weight of the ligated limb of 3.54% of the body weight. Plasma protein estimations showed the edema fluid to resemble diluted blood plasma. Johnson and Blalock 4 found that a loss of plasma equal to 2.4% of the body weight in 28 hours was sufficient to cause the death of dogs. In 3 cases gas was present in the tissues; Clostridia were found in a high percentage of cases from both the normal and the ligated leg. These were not tested for exotoxin production and their identity with B. welchi was therefore not definitely established (Trusler and Reeves 5 ).

The report of Marie 1 that sodium citrate administered by mouth or intravenously reduced the blood cholesterol of rabbits 60% led us to try its effect upon blood cholesterol of other species. It is of interest in this connection to note that dogs with a hyperpyrexia induced by the administration of 10 mg./kg. dinitrophenol showed an increase in blood cholesterol, the initial value of 161 mg. % being elevated to 285 mg. % 2 hours after injection of the drug. These observations led us to continue the study with albino rats.

Five groups of 5 rats each were placed upon a diet similar to that suggested by Chanutin and Ludewig, 2 in which extracted casein was substituted for extracted dried beef. To this cholesterol deficient diet, cholesterol was then added to the extent of 2%. The first group of rats received no supplement and served as control animals. In the second group, each animal received a daily subcutaneous injection of 1 mg. dinitrophenol per 100 gm. body weight. To the third, 0.2 gm. sodium citrate per 100 gm. body weight was given in aqueous solution by stomach tube. The fourth and fifth groups received NaHCO₃ and citric acid solutions, respectively, by stomach tube, in order to determine, if possible, the ion responsible for any change in the distribution of cholesterol. These 2 substances were given in amounts furnishing separately the same quantities of sodium and citrate ions as received by the animals in Group 3 on the sodium citrate supplement. Each animal of the fourth group, therefore, received 0.14 gm. NaHCO₃, and of the fifth group 0.11 gm. citric acid daily, per 100 gm. body weight. At the end of 3 days the rats were killed, the blood from each group pooled and analyzed for total and free cholesterol, total fatty acids, and lipoid P.

The Bordet Gengou reaction was designed to test the hypothesis of the identity of bacteriolytic and hemolytic complement, 1 although Bordet recognized later that the identity could not be regarded as absolute, since he was able to make specific anticomplements for the sera of various animals. 2 Muir and Browning 3 suggested that the bactericidal action of normal serum might be due to some moiety of complement which they termed “bacteriophilic”. Gordon and his coworkers 4 have been unable to obtain a serum which is devoid of hemolytic complement and yet possesses bactericidal action, though the reverse might be true of sera absorbed with bacteria, unless absorption were continued too long or the number of bacteria used were overwhelming.

A series of tests was made to determine which bacteria in the School collection could be depended upon to resist the bactericidal action of certain non-immune sera, and which bacteria were sensitive to it. The results with pooled human sera did not vary, but occasionally an organism was killed by some rabbit sera and not by others (e. g., B. enteritidis, B. typhosus Rawlins). Table I is a. resume of these tests.

The technique used was as follows: The bacterial suspensions, from 18-hour cultures on agar slants, were made in saline containing 0.1% gelatin. Very few bacteria survived incubation in saline, but the addition of this very small amount of gelatin removed the toxicity of the saline. The first dilution was made to correspond with Tube 1 of the McFarland nephelometer, the final dilution was one ten-thousandth of the first. The number of bacteria on control plates without serum was 1,000 to 2,000 per cc., 1 cc. being the amount of suspension used in the tests. The mixtures of bacteria and serum were incubated for 4 hours in a waterbath at 37.5°C, and the entire mixture was plated by pouring.

At a given initial fiber length the mechanical energy of contraction of the heart is increased under the influence of calcium. This is due not only to an increase in the fraction of the total energy liberated which is converted into useful work but also to an increased total energy liberation. The increase in efficiency occurring with administration of small doses of calcium salts indicates that the effect is a beneficial one, and might be of importance in contributing to the therapeutic effect of calcium in cardiac disease.

Since Locock 1 reported the successful use of bromides in several cases of epilepsy, this drug has been used extensively in the treatment of this disease. When these patients have been so treated, it has been noted that the concentration of the bromide ion in the blood is not always related to the amount of the drug administered; or is this concentration parallel to the clinical course, either in regard to the control of the convulsions or to the development of bromide intoxication.

Since the amount of bromide in the blood necessary to prevent epileptic convulsions in man varies, a determination of the concentration necessary to prevent experimentally produced convulsions in rabbits was made. The convulsions were induced by the intravenous injection of 2% thujone in 6% gum acacia, using 0.15 cc. per pound body weight. Seizures always followed such a dose, providing the injections were not repeated within one week. Rabbits were fed 5 grains of sodium bromide daily by stomach tube. When a blood bromide level of 240 mg. per 100 cc. of blood as measured by the Wuth 2 method was reached, injection of the thujone no longer produced a convulsion. In some animals a lower concentration sufficed to prevent induced seizures. (Table I.)

The total blood halide remained unchanged even when there was a 38% replacement (Table I) of the chlorides by the bromide; one may readily conclude that bromide replaces chloride, ion for ion.

To determine the site of action of the bromide, the amount in the brain was compared to that in the blood. The Berglund 3 method of analysis was used. The chloride content of the brain varied in rabbits, but the average was 166 mg. per 100 gm. of brain.

In a previous communication 1 it was shown that when large doses of irradiated ergosterol were administered to normal dogs there was a very great increase in metabolic rate. Since it appears that the rat is relatively insensitive to toxic effects of this substance, it seemed advisable to investigate its effect on the metabolism of the rat. Three rats of each sex were selected and trained for several months in a metabolism chamber. They were kept in separate cages and fed on the Wistar normal diet. The cages were kept in the same room where the apparatus was operated during a period of nearly 10 months, and the rats were handled only by the operator. It is believed, therefore, that extraneous factors were eliminated as completely as possible.

The apparatus used consisted of a closed circuit with a motor-driven hydraulic pump that delivered approximately 90 cc. of air each half minute. The respiration chamber consisted of a glass jar. Volume change was recorded on a kymograph by means of a writing needle attached to the bell of a small spirometer cut into the circuit. There was no new principle introduced into the arrangement.

The surface area was calculated by the formula derived by Lee, 2 S = KW^0–60 in which K = 12.54, W = weight in gm, and S = surface area in cm². It was mentioned in the earlier publication that one of the difficulties in calculating the surface area in the dog was the varying weight. Since the rat weight also varied it was thought necessary at first to use Lee's modification of Cowgill's formula, 3 S = KW^0–61 × but comparisons of surface area calculated by the 2 formulae on the same animal made it apparent that for purposes of comparison over a period of time there was not enough difference to justify the use of the more complicated formula.

The lack of knowledge concerning the cause of the nausea and vomiting which follows surgical operation prompted these studies. We have examined 115 specimens of postoperative vomitus from 62 cases chosen at random to include anesthetics and operations of many types. Over half had had ether either alone or in combination with other anesthetics. These factors seemed to bear no relationship to the results tabulated below.

Cushing 1 has recently called attention to a possible nervous mechanism in the production of peptic ulcer. In this connection it is interesting to note that of 11 specimens with an acidity over 20, 7 were obtained following operations for brain tumors. This observation might be explained by Cushing's postulation of a center in the cerebrum which is capable of promoting gastric secretion.

Not only the cases that actually vomit have this anacidity. Postoperative aspiration was done by nasal catheter in 15 patients, In these specimens there was no free acid. This lack of acid is not accompanied by a corresponding lack of enzymes. In all cases where free acid was present peptic activity was present, and in most of the cases (91 of 115) where the acid was absent peptic activity was detected after acidification with 0.4% hydrochloric acid. Rennin was present in 80 of 115 specimens.

Evidence of duodenal regurgitation is present as witnessed by the fact that of 115 cases, only 57 had bile in the stomach, while a considerably greater number contained pepsin but were not acid. Specimens from 75 cases were found to contain trypsin. Twenty-five of these 75 samples contained no bile.

It still remains the general opinion that heat production in the body depends mainly on the muscles, although it is admitted that a certain amount of heat is liberated by the chemical processes occurring in viscera, especially the liver. In comparison to the amount of heat produced by the muscles it is considered insignificant. Moreover there are different chemical processes that occur in the liver, which actually absorb heat, instead of producing it. In other words, any noticeable increase in the body temperature is attributed to the muscular activity.

Since the production of heat is constantly associated with the activity of the muscles and especially when the muscles are in action, it is also generally accepted that any tremor of muscles is accompanied with heat production. The bacterial chill, in which we have a contraction of muscular fibers and rigor, is accompanied with an elevation of body temperature. This increased heat production is usually attributed to apparent muscular activity. Briefly, to the muscular tremor, whenever it occurs, we ascribe heat production and associate the rise in temperature of the patient with bacterial chill to this cause.

In disagreement with this general opinion Petersen and Müller 1 presented experimental evidence indicating that the chill in the dog produced by continuous injection of colon bacilli is always associated with an increase in the splanchnic activity on the one hand and muscular tremor and leukopenia on the other. But this muscular activity is not accompanied by increased temperature in the muscles themselves.

For our temperature readings of a mucous membrane of gastrointestinal tract we used a Queen's potentiometer.

Ten dogs were used. Under nembutal anesthesia, the abdomen was opened and an opening in the stomach or intestine was made. For measurements of temperature of mucous membrane of the stomach the ends of 3 thermocouples (devised by Bachem), covered with shellac and embedded in celloidin, were introduced into the stomach and fastened to the inner surface by a ligature. For recording the temperature of the mucous membrane of the intestine the 3 thermocouples mounted on a piece of a rubber tubing of suitable diameter, and fastened with a rubber band were introduced into the intestine (duodenum). For measuring the temperature of the rectum the thermocouples were introduced through the anus. Heparin was injected intravenously (10 mg. per kilo), the femoral artery opened and a long glass thermometer introduced through it into the abdominal aorta. Ice packs and hot moist towels were applied to the skin of the chest of dogs.

The results are shown on the accompanying graphs. The abscissa represents the time from the beginning of the experiment, the ordinates—temperature in centigrade. Dotted lines represent the findings of the temperature in the abdominal aorta, the straight line—the temperature of the surface of the mucous membranes. The perpendicular lines show the time of application of cold and heat to the body surface of the dog.

Epidemiologists have observed that (a) a percentage of persons exposed to the risk of infection with Brucella-type abortus or suis-show either agglutinins or complement fixing antibodies in their serum and (b) clinical undulant fever is relatively infrequent in the occupational groups who come continuously and intimately in contact with Brucella organisms. Various inquiries leave no other explanation than that the Brucella abortus and probably the Brucella suis and melitensis as well are organisms endowed with a moderate degree of virulence, which rather frequently infect persons who are exposed either by contact through the skin or by alimentary ingestion. These infections are quite often subclinical and remain latent. Surveys in occupational groups have shown that the serologic tests fail to furnish a true picture of the number of unrecognized infections and therefore the allergic skin tests have been employed. Although a much more sensitive reagent to detect the existence of latent infections, the skin allergy is a variable state and in part influenced by the occupational activity. With the aid of this test one may recognize the allergic but not necessarily the latent infected and immune human being. The observations made by Huddleson, Johnson and Hamann 1 that the polymorphonuclear cells of persons who have had undulant fever in the past years or are actively infected or have no definite history of the disease may show a strikingly increased phagocytic activity offers a third procedure to detect latent infections. With the aid of a slightly modified Veitch technique 2 the usefulness of the method has been verified.

In studies on the streptococci and lactobacilli now in progress it became necessary to carefully examine the methods used in staining bacteria. The staining technique of the bacteriologist, for the most part, has advanced very little beyond that employed by the early workers. Perhaps Dobell 1 did more to point out certain fallacies of these methods than any other, yet his work is rarely quoted. The seemingly accepted technique is to dry on a slide the film of liquid containing the organism. After flaming, aqueous solutions of dye are flooded over the films. The delicate cytoplasm of the cell must necessarily be somewhat distorted by such methods. There are some descriptions of vital staining but generally little has been accomplished. We attempted to study the subject, using, as suggested by Dobell, a very large cell. In this case a Gram-positive organism from the gut of a guinea pig was used.

The method of staining was to add a water solution of the stain directly to either a broth or saline suspension of the organisms. The stained bacteria were examined in both wet and dry preparations. Certain stains revealed granules of varying arrangement and size within the cell. From the work of Guillermond 2 these were in all probability metachromatic granules. (Table I.) Dobell's results would agree since he says, “In my experience, only non-living structures in the cells (metachromatic granules, etc.) can be stained during life. But doubtless much depends on the stain itself!”

Gay and Clark 3 suggested that “vital” staining may really only represent a stage in a process of injury by the stain. In an attempt to test the possibility of injury to the cell experiments were designed to determine the toxicity of the various stains.

At the suggestion of Dr. H. W. Wade we studied the toxicity in rats of single subcutaneous doses of a number of representative types of ethyl esters of the fatty acids of chaulmoogra oil. Preparations included were: (1) a purified ethyl chaulmoograte supplied by Dr. R. Wrenshall, fractionated by redistillation in vacuo at 10 mm. Hg., containing a small amount of ethyl hydnocarpate as indicated by the iodine number; (2) mixed ethyl esters of fatty acids of Hydnocarpus wightiana oil prepared at Culion, containing more than 80% ethyl chaulmoograte as indicated by polariscopic measurements; (3) a similar preparation containing 0.5% iodine, prepared by Cole's method 1 ; (4) crude ethyl chaulmoograte containing 4% creosote, a preparation used in India, supplied by Dr. E. Muir from clinical supplies at Calcutta; and (5) “Chaulmestrol” (Winthrop), a proprietary preparation of mixed ethyl esters of fatty acids of chaulmoogra oil. The pertinent findings summarized in Table I agree with Martins' report 2 of 40 cc./kg. as being lethal for Leptodactylus ocelatus.

Examination of the tabulated data indicate that: (1) purified ethyl chaulmoograte preparations are more toxic than the ethyl esters of the total fatty acids of chaulmoogra oil, presumably because of higher content of ethyl chaulmoograte; (2) iodizing ethyl chaulmoograte with 0.5% iodine has little effect on toxicity; (3) addition of 4% creosote increases the toxicity, commensurate with the amount of creosote present; (4) the degree of unsaturation of crude ethyl chaulmoograte is not a sufficiently sensitive index of toxicity to be reliable; and (5) if subcutaneously administered ethyl chaulmoograte exerts its toxic effect through liberation of Na chaulmoograte into the circulation, the summation of rates of hydrolysis and diffusion cannot much exceed that causing a disappearance of 2 millimols per Mol of the injected ester per hour, since Na chaulmoograte kills rats in intravenous acute doses 3 of 0.1–0.125 gm./kg.

Van Slyke and Hiller 1 and Van Slyke and Robson 2 have reported the presence in gelatin of a new amino acid which is precipitable by phosphotungstic acid. On the basis of the elementary analysis of the copper salt, a positive test for a pyrrol group, and the ratio of amino to total nitrogen, the product was considered as being probably dihydroxy-pyrrol-alanine. According to the criteria which have been suggested by Vickery and Schmidt; 3 the evidence brought out by Van Slyke and his coworkers is not sufficient to consider this substance as one of the accepted amino acids. The experiments herein described deal with a number of attempts to isolate and identify the product described by Van Slyke and his coworkers.

Three lots of about 1 kilo each of a good grade of gelatin were treated essentially as described by Van Slyke and his coworkers except that the bulk of the arginine was first removed by means of flavianic acid. On decomposing the phosphotungstic acid precipitate a product was obtained which was extremely soluble in water but no evidence was obtained to the effect that this substance was dihydroxy-pyrrol-alanine. In the first experiment there was present in this fraction 884 mg. of nitrogen of which 340 mg. were liberated by treatment with nitrous acid in 3 minutes and 374 mg. in 30 minutes. Flavianic acid gave no precipitate with a fairly concentrated solution even on standing for 24 hours at 0°. The substance represented by the amino nitrogen was almost quantitatively precipitated by rufianic acid and by picrolonic acid. The rufianate was amorphous and decomposed without melting. The picrolonate was crystalline and fractional recrystallization from water and dilute alcohol resolved it into at least 3 fractions which differed widely in melting point, solubility, and somewhat in crystal form. None of these fractions could be isolated in pure form.

Emery and Atwell 1 showed that the administration to normal animals of extracts of whole sheep pituitaries provoked a marked increase in adrenal weight due predominantly if not exclusively to the hypertrophy of the cortex. Independently, we had obtained the same effect in rats from implants of beef anterior pituitary tissue. The present communication reports a striking diminution in this effect when a preceding ablation of the thyroids has been performed.

Fresh beef hypophyses were washed in alcohol before dissection, and to eliminate all chance of infection, both merthiolate (1 :10,000) and hexylresorcinol (1 :13,000) were added to many of the batches immediately after grinding. Of 214 adult male and female rats, some of which were unilaterally adrenalectomized, 41 had been previously hypophysectomized, 69 including many first thyroidectomized received pituitary implants, 67 including thyroidectomized were left untreated, and a group of 37 normals was sacrificed and right and left adrenals weighed separately.

Results. 1. Implants of adequate amounts of finely ground beef hypophyses, whether freshly prepared or stored at 0°C. for several months, induced marked adrenal hypertrophy in every case in which the animal possessed intact thyroids, but had little or no effect on thyroidectomized rats (Tables I and II).∗ 2. Two grams of anterior pituitary implants increased the adrenal weight of rats with intact thyroids to at least double that of the controls, larger or smaller amounts producing proportionate effects. 3. In one case where 60 days had elapsed between the last treatment and autopsy, the adrenal had maintained its increased size, being palpable during the whole period, and weighing at autopsy 80 mg., that of a sibling control being 26.5 mg. 4. The adrenal cortex was chiefly affected by the preparations employed.

Herein it is demonstrated that H. pertussis U. B. A., an undenatured extract prepared by mechanical disruption of Phase I H. pertussis bacilli and subsequent filtration through a collodion membrane, is able to stimulate the formation of complement fixing antibodies, agglutinins and precipitins.

In previous communications 1 it was pointed out that definite immunity could be developed against the virus of poliomyelitis using virus rendered non-infective by formalin. However, the amount of formalin used produced considerable skin irritation. Since then it has been shown that virus suspensions inactivated with 0.1% formalin, at incubator temperature, were also antigenic for monkeys, and at the same time, gave practically no skin irritation. The majority of the animals showed both humoral immunity, as tested by the neutralizing power of the serum for the virus of poliomyelitis; and tissue immunity, that is resistance to intracerebral inoculation of active virus.

The inactivated antigen produced no reaction whatsoever, neither symptoms, temperature rise nor cerebrospinal fluid changes developed upon repeated inoculations of large doses given both intracerebrally and intraperitoneally, each dose representing the equivalent of thousands of infective doses of living virus. Moreover, during vaccination the animal suffered no untoward local or systemic reactions, so it was felt that the vaccine could be given to humans with perfect safety.

However, before giving it to children, it was deemed advisable to try it upon ourselves, not that we had any misgivings about the possibilities of infection, but rather to determine whether the vaccine produced any disagreeable local or general reactions. Accordingly, 6 volunteers from the Bureau of Laboratories, Department of Health, New York were given 5 cc. doses of a 10% virus suspension, inactivated with 0.1% formalin for 16 to 48 hours. Three were given one dose, two, 2 doses and the third, 3 doses.

Human urine and extracts of urine have been known for some time to contain both pressor and depressor substances. 1 Attempts have been made to associate them with the presence of arterial hypertension with small success. Since the properties of these substances are little known, we have undertaken a study of them.

Cats anesthetized with ethyl urethane (7–10 cc. of a 25% solution given subcutaneously) or ether were employed as test animals. Blood pressure was recorded from the femoral artery and the vagus nerves were severed. Injections of the warmed fluid were made into a femoral vein. Forty-one animals have been employed for the experiments.

Injections of 5 cc. of fresh normal urine usually produced elevation of blood pressure of from 6 to 20 mm. of Hg. Boiling the urine for 1–3 minutes did not affect the pressor action, nor did preliminary atropinization of the animal. Mixtures of urea and sodium chloride, or sodium acid phosphate of the same specific gravity as the urine, did not appear to duplicate the vascular action of urine.

Often the same test animal which yielded a pressor response to normal urine exhibited depression of the blood pressure on injecting the same urine at another time during the course of the experiment. Some animals exhibited only depressor responses. We have been unable to predict which response will occur, though the pressor was the more common. At times the animals became entirely refractory to urine injections. The variability of response led us to suspect that the functional intactness of the central nervous system was essential for a vascular response to be elicited. Decerebrate and pithed cats were found to be refractory to urine injections, whereas the same animals before pithing had responded actively. It appeared, then, that the vascular action of urine was not peripheral, but central.

As far as we have been able to discover, there have been no published investigations of action potentials in heart muscle cultures. This is a report on a method which it is hoped will furnish information on the mechanism of irritable tissues in general and particularly those which are excited spontaneously.

In general, the potential difference between the micro-electrodes making contact with a muscle culture is impressed on the grid of a vacuum tube and measured by the deflection of a string galvanometer in the plate circuit.

The cultures were prepared by the usual hanging drop technique. Small pieces of the ventricle of 16-day rat embryos were explanted in a medium of rat blood plasma and rat embryo extract. The ages of the cultures varied from 2 days to 2 months. The present records were taken from the older cultures containing outgrowths of muscle which had differentiated in the manner described by Goss 1 and were presumably without nervous tissue elements. Spontaneous contraction occurs in practically all cultures at one time or another and may be rhythmic or irregular. There is a great variation in the duration of the resting and acting periods and also in the frequency of rhythmic activity.

The electrode systems are micro-pipettes, 2–5 mu in diameter at the tip, with chloride coated silver wire coils placed in the large ends and the whole filled with mammalian Ringer solution. These pipette electrodes are mounted in micro-manipulators 2 and enter the moist chamber, on which the culture cover slip is placed, through a water trap as used by Chambers. 3 Such an electrode system has a high electrical resistance which precludes the direct use of a rapid recording instrument even if polarization difficulties could be overcome. The electrodes are connected to the grid circuit of a type 32 vacuum tube, operating at its free grid potential so as to draw a minimum of grid current.

In previous studies 1 it was found that removal of the spleen in albino rats free from Bartonella muris infection is followed by an increased elimination of copper, which produces a negative copper balance. In a study of the metabolism of rabbits before and after splenectomy we found that removal of the spleen causes an increase in copper excretion in this species as well.

Two rabbits were splenectomized at the age of 2 ½ months, after a foreperiod of 3 weeks, while another one was kept as a control under identical conditions. They were given alfalfa hay, oats and copper-free water, which was also used for all chemical determinations and for washing everything that came in contact with the animals. Urine was collected daily, feces twice a week. The urine was analyzed for total nitrogen (Kjeldahl), total sulphur (Benedict), total and inorganic sulphates (Folin), calcium (volumetric), phosphorus (Fiske-Subbarow), and copper. 2 Feces, hay and oats were analyzed for total nitrogen, calcium, phosphorus, copper and iron. 3

Our experiments show that splenectomy causes no change in the metabolism of nitrogen, sulphur, phosphorus and iron in rabbits. The calcium excretion proceeds unchanged after splenectomy, which also tends to show that the spleen is not involved in the regulation of calcium metabolism.

While the excretion of copper in the intact animal was remarkably constant, it increased in the splenectomized animals a week after the operation. The increase takes place in the feces, since the greater part of copper is excreted by the gut. The excretion of copper by the kidney does not seem to be influenced by splenectomy, though a larger amount of the total copper excreted is found in the urine in rabbits than in rats. As shown in the table, the copper intake on a diet of hay and oats is insufficient to maintain a positive copper balance in a growing rabbit, but the loss of copper from the body increases considerably after splenectomy.

Attempts to reproduce the Shwartzman phenomenon with B. pertussis have been reported by several authors. (Gross, 1 Shwartzman, 2 Mishulow, Mowry and Scott. 3 ) However, the potency of the nitrates seemed to have varied for unknown reasons and always remained low as compared to similar preparations from B. typhosus, 4 meningococcus 5 and other organisms. 6

During some work with serum neutralizations of B. pertussis toxic substances it occurred to the author of this paper to employ Toomey and McClelland's brain veal infusion medium 7 for the preparation of the toxic factors necessary for the production of the Shwartzman phenomenon.

In these experiments the brain medium was prepared essentially according to the method of Toomey and McClelland with various peptones (neopeptone, proteose-peptone (Difco), Witte's peptone). The H-ion concentration was adjusted either to pH 7.3 or 7.8. The media were seeded with 24-hour-old cultures on “chocolate agar” slants. The strain used was subcultured for at least 3 successive days preceding these inoculations.

The organisms grew luxuriantly in the brain media similar to Toomey and McClelland's description. The strain employed was M 12 (group B), kindly supplied to us by Miss Mishulow of the New York City Board of Health Laboratories.

Berkefeld filtrates of 4, 7 and 17 days' brain medium cultures were tested in rabbits for the elicitation of the Shwartzman phenomenon. 0.25 cc. undiluted filtrate was injected intradermally into rabbits and followed in 24 hours by a single intravenous injection of the same filtrate undiluted or in varying dilutions. Comparative studies of the brain culture filtrate with “chocolate agar” culture washings were made. The results may be described in brief as follows:

Following total gastrectomy in monkeys, a “secondary” type of anemia cccurreil 72 hours after operation, and persisted. Evidence of regeneration (reticulocytosis, nucleated red blood ceils, polychromatophilia and basophlllc stippling) was noted. The specific blood picture of human pernicious anemia did not develop within the 6 month period of study, although there was a decrease in the number of red blood cells, a slight tendency for the color index to rise above the preoperatiye value, a leukopenia and a slight increased degree in size dispersion of the red blood cells.

In spite of an immense amount of work on the retention of calcium and phosphorus under various conditions the amount or ratio of these elements recommended for pregnant or lactating animals is still almost empirical. Before dependable knowledge of the optimal ratio and amounts of these elements during such periods can be obtained numerous animals should be studied through their whole reproductive life. In order to test thoroughly any one calcium and phosphorus mixture it would be desirable (1) to secure as rapid reproduction as possible, and (2) to provide constant quantities of these elements during the animal's entire reproductive history. We have made such a study during the past 2 years, using female rats, and report in this preliminary communication one of our more important findings.

Method. Virgin female rats weighing approximately 200 gm. at 100 days of age, were allowed to raise their first, “qualifying” litter on stock food. The females were remated at the end of 21 days' lactation and placed on the experimental diet.∗ All litters were reduced to 6 pups. When the young were 21 days old they were killed, and the mother immediately remated—thus allowing no rest period between cycles. In this way 10 successive gestations and lactations were studied. Five females were placed on each mineral combination, and the average “success” of the 5 mothers was used as a criterion of the adequacy of the calcium and phosphorus mixture. The “success” of the rats was gauged by averaging, over 10 reproductive cycles, the weight of the young at 21 days of age.

It has been stated that blood-cerebrospinal fluid barrier permeability is capricious. This is possibly a confession that the meaning and mechanism of permeability are yet incompletely understood. It is not surprising that this is so for careful study of standard texts and original papers since 1910 leads one to conclude the theories of origin, function and fate of cerebrospinal fluid are far from agreed upon.

Studies of variations in the permeability of the barrier have been prosecuted from two points of view. Foreign substances have been introduced, more commonly into the blood stream, occasionally into the subarachnoidal space, and their fate traced. Such substances have been crystalloids, as bromide; colloids, colored and uncolored, and McKinley and Holden 1 utilized the bacteriophage. Again, study has been made of blood elements not normally found in the cerebrospinal fluid whose presence there may be taken as evidence of increased permeability. Of these may be mentioned induced antibodies and naturally occurring antibodies such as hemolysins and isohemagglutinins. Three references exist dealing with this last indicator of permeability.

Weil and Wall 2 found no isohemagglutinins in 12 cerebrospinal fluids from such conditions as syphilis of the central nervous system, hemiplegia, brain tumor and myoclonic encephalitis. On the other hand Herman and Halber 3 found isohemagglutinins in 11 of 80 fluids examined. These fluids came from more than 30 widely different types of pathology, although most involved the central nervous system either organically or functionally. In 6 of 22 cases of bacterial meningitis the cerebrospinal fluids contained isohemagglutinins but the 6 cases of tuberculous meningitis included were all negative. Two of 3 cases of cord tumor, 2 of 4 cases of sclerosis, and 1 of 3 cases of chorea accounted for the remaining 5 positives. Parr 4 examined 104 cerebrospinal fluids at Beirut and found 6 containing isohemagglutinins.

The writer 1 described experiments indicating that a strain of diphtheria bacillus under investigation required for its growth, among other things, a substance associated with the “proline” fraction of amino acids obtained by Dakin's butyl alcohol extraction method. Further work with this material has shown that not a single factor, but 3 separate substances were involved in the accelerating effect there described. It was learned to begin with, that the action was not entirely limited to the “proline” fraction, but that the “monoamino” fraction even more frequently exerted it, and occasionally the 2 added together gave better results.

Before any material progress was made in separating and recognizing these factors, it became necessary to introduce a quantitative method for the estimation of relative amounts of growth. The method finally used was a micro kjeldahl determination of the amount of nitrogen in the bacterial growth on a measured amount of material for a definite time. The details of this procedure will be described elsewhere.

Careful examination of the effect produced by the proline fraction showed that proline itself was evidently not concerned, but that the acceleration in growth was apparently produced by minute quantities of ethyl alcohol used as solvent in handling the fraction. Pure ethyl alcohol exerted the same effect, and maximum acceleration of growth with our particular strain of organism is reached with a concentration of about 0.5%. Larger quantities of course become inhibitory. In place of ethyl alcohol, sodium acetate, glycerol, or maltose all appear to enhance the growth when added to the same control solutions.

During a study of the influence of environmental temperature upon the metabolic rate and certain organ weights of the albino rat a number of these animals were maintained at room temperature of 5° and 10°C. When they were killed it was noticed that the kidneys of many were swollen, soft and mottled in appearance. Fig 1. Histological examination revealed an interesting lesion in their kidneys. Besides its inherent interest it seemed desirable to report the occurrence of this lesion because of the common belief that exposure to cold is often associated with the onset of one form of Bright's disease in man. That the renal lesion reported here was acutally due to the cold environment seems certain for only 0.015% of the kidneys from 26,000 animals of the same stock raised at ordinary temperatures have shown any renal pathology.

Eighteen young adult rats, 9 males and 9 females were kept 3 in a cage at a constant environmental temperature of 5°C. for 90 days. By this time 7 were dead. Two of these were examined and both had mottled soft kidneys. Of the 11 which remained 4 showed the presence of the renal lesion grossly. Twenty-four young adult rats, 12 males and 12 females were kept under the same conditions as the first group at a temperature of 10°C. for 90 days. When they were killed 6 showed quite obviously the renal lesion to be described here.

Microscopic examination reveals depressed areas in the superficial part of the cortex which appear to be due to partial collapse of the tubules in these areas with necrosis and desquamation of the epithelial cells. Some of the cells are shrunken and clumped together so that they appear as partly fused cellular masses in which the nuclei are very hyperchromatic and closely crowded.

The rate of absorption of glucose from the intestinal tract of the rabbit decreases with time after the glucose is administered. This is apparently related to the decreasing amount of glucose entering the intestine from the stomach each hour for there is a relation between the rate and amount of glucose available for absorption.

Houssay and Biasotti, 1 Barnes and Regan 2 and others have demonstrated the prolongation of life and amelioration of the diabetes that follows total pancreatectomy in the hypophysectomized dog. During the last few months observations have been made upon 2 cats in which total pancreatectomy was performed 2 weeks after hypophysectomy. There was no loss of weight in the period between the 2 operations. One animal died in hypoglycemia on the 35th day after pancreatectomy, the other was killed on the 85th day while still in good health for liver glycogen determination (0.8 g %). Neither of these animals received insulin at any time. Both of the animals lost almost 50% of their body weight during the periods noted. They were somewhat apathetic in their behavior and showed considerable muscular flaccidity although capable of ordinary activity with no apparent distress. Their appetite was capricious but their food intake was usually ample for the caloric requirements of normal animals.

In 9 control animals the average length of life after pancreatectomy or withdrawal of insulin was 4 days (range 2–7 days). The glycosuria averaged 5–10 gm. daily in spite of the fact that the animals usually refused food after the second day.

Pellagra and pernicious anemia are now regarded as special types of deficiency diseases. They have in common such clinical manifestations as achylia gastrica, glossitis, peripheral neuritis, and central nervous system changes. Goldberger and his associates 1 , 2 believe that pellagra is caused solely by a diminished intake of some specific food substance (“vitamin G”). Rolph, 3 Turner, 4 and others have described cases developing secondary to lesions of the gastrointestinal tract. While it has likewise been suggested that in certain instances the lack of some substance in the diet may cause pernicious anemia, it usually follows the lack of the essential secretion (“intrinsic factor”) in the gastric juice which changes food into an anti-anemic substance. 5 A short time ago Spies and Payne 6 produced remissions in 2 patients with pernicious anemia, by giving an incubated mixture of beef muscle and achylic gastric juice from acute pellagrins. This observation indicated that the “intrinsic factor” was present in gastric secretions from pellagrins in an amount adequate to form the antianemic substance. They suggested in this study that the usual pellagrin apparently develops his disease as a result of inadequate food ingestion, whereas the usual pernicious anemia patient developed the anemia from the failure of his gastric juice to make an antianemic substance from food.

Since pernicious anemia and pellagra are both related to diet and its subsequent assimilation by the body, there have been recent attempts to recognize more specifically those substances in food which are important in the pathogenesis of the 2 diseases. Strauss and Castle 7 , 8 have found that the substance in food (“extrinsic factor”) concerned with the development of pernicious anemia is associated with so-called “vitamin G” which other workers have considered as the “antipellagric vitamin”.

It has been shown 1 , 2 , 3 that the loss of sodium chloride and possibly other inorganic ions contributes to the development of the severe symptoms which follow adrenalectomy, and that the administration of these ions will partially or completely prevent the development of these symptoms.

Since disturbances of the oestrous cycle have repeatedly been studied in adrenalectomized animals, 4 , 5 , 6 and since various authors came to the conclusion that cortical extracts have a specific influence on the sex organs, 6 , 7 , 8 it seemed of importance to establish whether the administration of salt would correct the disturbance of the cycle, as our preliminary experiments seemed to indicate. 9

Forty-six normally cyclic rats were adrenalectomized. Fifteen of these were untreated, and went into a dioestrus which lasted until death. The remaining 31 were given the Rubin salt mixture. 3 Of the 18 animals maintained in good health by salt, 14 remained normally oestric; the animals which died in spite of treatment were nearly all dioestric until death. The completeness of the operation was assured by the death which followed soon after the removal of the salt, and by the absence of any cortical tissue at autopsy.

In general, then, it seems to be true that where the salt mixture maintains the life and weight of the adrenalectomized animal, oestrus appears at normal intervals. In this connection we would like to emphasize that the salt mixture maintained only 18 of the 31 animals treated.

The oxidation of thiol-imidazole sulfur to sulfate by FeCl₃ is frequently used in the conversion of thiol-imidazole compounds to imidazoles. One would assume from the literature that cystine sulfur is not similarly oxidized by this reagent. 1 This suggested the utilization of this reaction as a means of testing for the possible presence of a thiol-imidazole compound in insulin. So far only about 70% of the sulfur of insulin has been accounted for. The highly specific Sullivan reaction indicates the presence of about 8.4% cystine, whereas if all the sulfur of insulin (3.2%) were present as cystine, 12.0% should be present.

To test the feasibility of this approach experiments with cystine were first undertaken. These experiments showed contrary to our expectations that cystine sulfur was quite readily oxidized to sulfate by FeCl₃. Twenty mg. of cystine heated with 2.5 cc. of 6.5% FeCl₃ for 1 hour in the boiling water bath gave a positive test for sulfate. Under the same conditions no sulfate was formed from cysteic acid.

Other experiments on cystine were conducted in which the compound was heated in FeCl₃ (6.5%) with HCl varying in concentration from N/10 to 10%. All samples showed appreciable amounts of sulfate after being heated for an hour. The amount of oxidation decreased in a roughly proportional manner with the increase in acid content, yet we found no conditions under which no sulfate was obtained.

To determine the amount of sulfate formed a 5 gm. sample of l-cystine was suspended in 625 cc. of a 6.5% FeCl₃ solution and was heated on a boiling water bath for a total of 21 hours. Samples were removed at intervals for sulfate determination. The oxidation followed roughly an exponential curve of low coefficient. At 21 hours approximately 60% of the sulfur was accounted for as sulfate.

It is of interest that these therapeutic agents used in the treatment of angio-spastic disease should antagonize the action of each other on the blood vessels of the frog.

Trichloracetic acid and chloral hydrate are reagents that serve in the detection of ergosterol and of carotene. Ergosterol with the 2 reagents yields an initial red and a final blue color, while carotene gives only the blue color.

Emotional states produce a marked relative lymphocytosis in healthy persons.

Investigations on the Salmonella were carried out employing the method recently described by us 1 for studying the specificity of absorption, particularly from polyvalent phages, by heat-killed bacilli. In principle the method depends on the fact that the absorption of phage by heat-killed bacilli is largely irreversible and consequently the residual phage may be studied both qualitatively and quantitatively without removing the absorbing organisms.

In general, it was possible with this procedure to differentiate some of the following groups, for instance, the suipestifer group which absorbed specifically the homologous phage and also the suipestifer fraction 1 of the polyvalent anti-paratyphosus A phage. However, some of the suipestifer strains, particularly those of the European type, were characterized by a capacity to absorb actively also other fractions, for example, the paratyphosus B fraction of the polyvalent anti-paratyphosus A phage when B. aertrycke (of the paratyphosus B group) was used to test for residual phage. (Table I.) This same type of suipestifer absorbed the anti-paratyphosus B phage to a moderate degree, but less intensely than the paratyphosus B group of organisms, while other suipestifer strains showed no absorption.∗

The paratyphosus A group of organisms and also some of the European suipestifers employed were the most active absorbers of the anti-paratyphosus A phage when the homologous strain was the test organism. The distinction between these could be made readily since the paratyphosus A strains failed to absorb actively the anti-suipestifer phage.

In contrast to the ease with which differences within the suipestifers could be demonstrated, was the failure to distinguish the typhosus-enteritidis group from the aertrycke-paratyphosus B group, since in the several combinations of phages and test organisms available, both groups absorbed equally well. (Presumably this observation is in harmony with the findings recorded by Furth and Landsteiner, 2 that some anti-sera of the 2 groups gave weak to moderate precipitin reactions also with the heterologous extract.)

The specific inhibition of bacteriophage by bacillary extracts recently described by us 1 , 2 and confirmed by Kligler and Olitski, 3 Burnet, 4 and Gough and Burnet 5 may be employed to study the chemical properties of these complex cellular products with the aid of various manipulations such as fractionation and hydrolysis.

In a previous communication, 2 the observation was made that crude saline extracts of B. aertrycke boiled in N/2 alkali, still retained their phage inhibiting capacity and, consequently, it seemed desirable to study in detail the sensitivity of these substances to alkaline and acid treatment before attempting purification. For this purpose, a crude saline extract of B. aertrycke was employed and tested for phage inhibition before and after boiling in N/2 sodium hydroxide for 5 minutes∗ and subsequent neutralization. Such alkali treated extracts showed marked increase in phage inhibiting capacity although they gave a precipitin titre that was slightly lower than that of the untreated material. This observation, therefore, indicated that the degree of inhibition of phage did not run parallel to the precipitin titre of the crude extract. Striking confirmation of this view was found when these extracts were subjected to mild hydrolysis in N/15 and N/30 hydrochloric acid at 80° C. This procedure was adopted because boiling for 5 minutes or longer in N/2 acid completely destroyed both the phage inhibiting reaction and also the capacity to react with precipitins.

A crude extract of B. aertrycke was dissolved in saline, passed through a Seitz filter and then treated with N/5 hydrochloric acid in sufficient quantity to produce a final concentration of N/15 acid and held at 80°C. Samples were removed at various intervals, neutralized, brought to a dilution of 1:500, centrifuged to remove the precipitates, and sterilized by heating for one hour at 80° C.

The glycolytic enzyme of muscle extract after becoming inactive on standing can be reactivated by thioglycolic acid to a higher state of activity than that originally present. Lipmann's hypothesis of the enzyme being inactivated by oxidation and reactivated by reduction is corroborated and specified by the evidence that this oxidation-reduction is concerned with a sulphydril group of the enzyme. There is evidence for the assumption that glycolytic enzyme of muscle seems to be a sulphydril compound which loses its enzymatic property by oxidation to the disulphide state and is reactivated by reduction to the sulphydril state.

The writer has reported his phytopharmacological studies on blood sera from various clinical conditions and described characteristic phytotoxic reactions produced by such specimens from certain diseases. These reactions were found useful in establishing a differential diagnosis and also in evaluating the results of therapeutic procedures. 1 Such specific phytotoxic reactions were particularly noted in pernicious anemia, 2 leprosy, 3 and pemphigus. 4 , 5 This method of investigation has been repeatedly employed by Professor Leon Tscherkes of Odessa, who has confirmed Macht's findings. Some time ago, in a personal communication to the author, Professor Tscherkes wrote that he had discovered that blood sera from still another pathological condition exhibited a marked phytotoxic property for the seedlings of Lupinus albus—namely, trachoma. Since this discovery was of considerable importance, the present writer undertook a study of blood sera from a series of trachoma cases in the United States in order to corroborate the findings of Tscherkes.

Specimens of blood from trachoma patients in all stages of the disease were obtained from different sources. A few blood sera were obtained from sporadic cases in Baltimore; for the most part, however, specimens were secured through the courtesy of public health officers in Tennessee, Kentucky and Indian Territory. According to the method described in detail elsewhere, 6 these blood sera were tested on seedlings of Lupinus albus; and the phytotoxic indices of trachoma blood were compared with those given by normal human sera and by sera from various other ophthalmological diseases obtained, with permission of Professor Wilmer, from the Wilmer Institute of the Johns Hopkins University. The results obtained with specimens from 100 trachoma cases are indicated below. At the time of examination, in 30 cases, the disease had endured for periods ranging from one month to one year; in 10 cases, it had persisted from 10 to 20 years; in 8 cases for more than 20 years; and, in 3 cases, for more than 40 years.

Studies on the nervous mechanism involved in copulatory ovulation have disclosed several interesting facts. First, the sympathetic chain has been shown not to be essential for ovulation and pregnancy in the cat 1 or rat. 2 Secondly, mid-brain removal or section of the brain stem immediately after copulation does not prevent ovulation in the rabbit if the anterior pituitary is left intact. 3 These facts suggest that a nervous stimulus reaches the pituitary during copulation and that the pathway is probably independent of the sympathetics.

The purpose of this study has been to determine the effect of parasympathetic drugs on the process of ovulation. The first step was to determine whether or not these drugs alone would cause ovulation.

Adult female rabbits having been isolated 2 to 5 weeks and showing the typical vulval swelling and purple congestion of full oestrus were used. Four were treated with atropine sulfate (2 ½ mg./kg. intravenously); 2 with pilocarpine (1–3 mg./kg. every 2 hours for 6 hours); and 3 with physostigmine (¼ mg./kg. every hour for 6 hours). Laparotomy was performed in each case 48 hours after the last injection. Ovulation was not induced in any case. Ovaries in many animals showed signs of a decrease in oestrin secretion as was indicated by paling of the uteri and vulva. These rabbits, instead of ovulating, went out of oestrus immediately and refused to copulate for several days. This indicates that these drugs are inhibitory in their action rather than stimulatory.

To test this hypothesis, a series of 11 rabbits was injected with atropine as in the former series and allowed to copulate. The time which elapsed before copulation was different for each individual (1, 3, 4, 5, 7, 10 minutes, etc.), so that a complete series was obtained in which the time of drug action varied from 1–30 minutes.

Novak 1 has stated that the good results he has obtained from the administration of the anterior pituitary-like hormone (Antuitrin-S) in cases of functional uterine bleeding were possibly due to changes produced in the blood by the action of this substance. To determine the presence or absence of such changes, 6 patients were given Antuitrin-S by hypodermic injection, using 1 cc. 3 times weekly. Three of these were injected for a period of 2 weeks, the other 3 for 3 weeks. Complete blood examinations (involving the formed elements only) were made before the injection of the product was begun, and again after all the injections had been completed. These examinations included the hemoglobin, red blood count, white blood count and differential, platelet count, bleeding time, coagulation time, and tourniquet test. It was felt that comparison of the figures obtained before and after treatment might throw some light upon the effect of Antuitrin-S upon the blood elements, as well as explain any improvement in functional uterine bleeding noted, if dependent on blood changes. The patients selected were all fertile women who presented no demonstrable abnormalities other than the uterine bleeding.

The study of the results of the injections demonstrated that the effect on the blood was so slight as to be negligible. In most cases there was some drop in the hemoglobin after the administration of the hormone, the greatest difference being 13% and in only 1 case a rise of 4%. The other variations were too unimportant to require comment. It is interesting to note that the platelet count almost uniformly was lower after treatment. This is surprising, if the fact that a reduction in platelets is usually found associated with an increased tendency toward hemorrhage, is accepted.

Analysis of the basic amino acids of thyroglobulin by direct isolation methods indicates that this protein yields 0,62% of histidine, 8.22% of arginine and 1.93% of lysine.

Using a modification of the technique of Friedman and Lapham, 1 involving intravenous injection of urine and the inspection of the ovaries after a period of 48 hours, an effort was made to determine the minimum amount of urine required to produce a positive reaction in cases of normal pregnancy. Observations were made between the fifth and the fourteenth weeks. All animals used were virgin does between 3 and 4 months old and were obtained from the same source. One hundred and nineteen determinations were made on 31 pregnancies. Of 39 fractional tests, 2 were positive with .05 cc. of urine; 10 with .1 cc., and the remaining 27 required .5 cc. or more.

Sixty-two similar determinations were made on 10 cases of excessive vomiting, 4 of which were of the pernicious type. Of 16 tests on these patients, one, associated with a hydatidiform mole, was positive with .01 cc. and .0075 cc.; 4 were positive with .0125 cc.; 6 with .05 cc. and 4 with .1 cc.

Due consideration was given to the concentration of urine as evidenced by the specific gravity, since I 2 have previously shown that this is a factor in the amount of urine required to produce a positive reaction.

In 5 of the vomiting cases, followed to date, it was found that the amount of urine required to produce a positive reaction increased as the symptoms improved.

Although the amount of urine required to produce a positive reaction varies considerably in the normal cases, it would seem that the discrepancy between that group and the vomiting cases is too great to be explained by simple concentration of urine or normal variation.

Completely denervated sweat glands may be made to secrete by pilocarpine (Langley and Anderson, 1 Langley 2 ), in spite of the fact that section of a peripheral nerve may lead to a much diminished secretion, which Burn 3 attributed to the section of the somatic motor fibers rather than to the sensory or sympathetic fibers in the nerve. Burn believed that this was due to circulatory effects produced by muscular inactivity and found that it paralleled the decreased vasodilator response to histamine. Furthermore he found that pilocarpine following deafferentation of the hind limb in the cat produced a diminished response, accompanying the decreased movement and impaired circulation that was present for a time following the operation. This diminution disappeared later on. Burn, 3 , 4 further showed that the exaggerated sweat response following sympathectomy was due mainly to the removal of the sympathetic control of the blood vessels and also in some animals to a hypersensitiveness of the denervated glands and vessels to pilocarpine.

Guttmann 5 has reviewed the literature on this problem of sweat secretion in man and has described it as seen in a number of Foerster's patients. He has suggested that the sweat glands have a secretory innervation through the sympathetic chain as well as directly by way of the ventral roots to the spinal nerves, and an inhibitory innervation by way of the dorsal roots (Dieden 6 ). Langley 7 denied vigorously that there could be any direct secretory innervation to sweat glands aside from that passing through the sympathetic chain in the cat. He also found a facilitory instead of an inhibitory effect in the cat on stimulation of the 6th and 7th lumbar dorsal roots in case the secretion was already in progress, due, he thought, to the presence of the accompanying dorsal root vasodilatation.

A simple apparatus is described for measuring the metabolic rates of smaller laboratory anirnals. It requires no unusual materials and can be made in a few hours by anyone able to solder. Its accuracy and simplicity are derived from the fact that by keeping the size down to a minimum, the contained gas volume is very small and the necessity for rigorous temperature control and circulation of the air is thereby eliminated. The values reported in the literature, with more complicated apparatus and groups of animals measured over longer intervals of time, are reproducible in this simple apparatus using single animals and observation-periods of relatively few minutes.

A fairly uniform bacteriologic technic is in current use for the demonstration of “hormones,” “vitamines” and “accessory growth-factors.” As a typical example of this technic, Nicolle and Césair 1 inoculated 10% broth-dilutions of specific immune serum with homologous bacteria, and invariably obtained a much more luxuriant growth in 10% homologous anti-serum than in normal or hetero-immune serum controls. From this they concluded that the specific “antibodies” in question are in reality specific “growth-stimulating hormones” for the corresponding bacteria.

In technics of this type there are numerous arbitrary experimental constants which may conceivably condition the end-result. Prominent among these arbitrary constants are the dosage, age, growth-phase and nutritional condition of the inoculum, and the time and method of reading the end-result. Most “hormone” data, for example, are recorded in terms of relative turbidity at the end of 24-hours incubation.

In order to test the reliability of this arbitrary time-factor, parallel “viable” and “total” population growth-curves were plotted for Streptococcus hemolyticus in 100 cc. beef-infusion broth and broth containing 0.1% beet juice. Typical curves thus obtained are recorded in Figure 1.

From these curves, it is seen that the conventional selection of the 24-hour turbidity (or “total” population count) as the recorded reading gives data apparently proving that beet-juice contains a powerful growth-stimulating hormone for streptococci. An arbitrary selection of the 2 to 5-day “viable” count, however, would give equally convincing evidence that beet auximone is a fairly efficient surgical antiseptic, causing premature death of 99% of the potentially “Viable” streptococci.

There are marked differences in the susceptibilities of different samples of presumably normal human plasma-clot to the thrombolysin formed or secreted by certain strains of S. hemolyticus. While one clot may be completely liquefied within 30 minutes after admixture with a given lytic culture, the clot from a second, presumably normal individual may show no sign of softening even after 24-hours incubation with this culture. 1

To test the possibility that this difference may be due to quantitative differences in fibrin-content, parallel fibrinolytic titrations were made with different concentrations of the same isolated fibrin. Typical data thus obtained are recorded in Table I.

From this table it is seen that the rate and completeness of fibrinolysis with a given lytic filtrate varies inversely with the fibrin-content. Complete insusceptibility to lysis, however, does not result from an increase in fibrin-content even to 4 times that of the average normal human plasma-clot. This result is in line with the statistical evidence reported by Hadfield, Magee and Perry. 3 These investigators found no correlation between fibrin-content and fibrinolytic susceptibility in different blood samples.

To test the possibility that the observed differences in susceptibility may be due to qualitative differences between “normal” and “immune” fibrins, parallel fibrinolytic titrations were made with fibrins isolated from susceptible and resistant plasma-clots. A typical comparison of this type is recorded in Table II.

From this table it is seen that serum-free fibrins isolated from resistant and susceptible plasmas are very nearly identical in their fibrinolytic susceptibilities. The slightly delayed softening of the “immune” fibrin is presumably due to adsorbed antibodies. This is confirmed by the fact that susceptible fibrinogen redissolved in and afterwards reisolated from resistant serum, has approximately this same slight degree of insusceptibility. This conclusion is in line with data reported by Fuchs 4 in his study of “normal” and “immune” fibrins by other technical methods.

Stricker 1 first showed that stimulation of the dorsal roots of spinal nerves produced a vasodilatation in peripheral blood vessels. This observation has been confirmed by Bayliss, 2 Ranson and Wightman 3 and others. Hinsey and Gasser 4 showed that the fibers mediating vasodilatation on stimulation of the dorsal roots were ones whose potentials were found in the C-spike of the action potential. Vasodilator reflexes over the dorsal roots were demonstrated in positive experiments by Bayliss 5 and Fofanov and Tschalussow. 6 Recently Bishop, Heinbecker and O'Leary 7 and Zuckerman and Ruch 8 have again obtained positive results which tend to show that the vasodilator fibers in the dorsal roots are accessible to reflex stimulation.

The observations of Hinsey and Cutting, 9 Hinsey, 10 and of Zuckerman and Ruch 8 showed that the skin temperatures in deafferented hind limbs were lower than in the opposite normally innervated limbs. These observations could either be attributed to an overaction of the vasoconstrictor pathways or to an absence of a dorsal root vasodilator mechanism in the colder limb. Zuckerman and Ruch state that when a limb is connected only with the spinal dorsal roots, it shows a different behavior than the opposite completely denervated limb (method of denervation is not stated; unless a bilateral abdominal sympathectomy was performed in addition to cutting the lumbosacral spinal roots, the denervation would not have been complete). The former fell in temperature much less than the denervated limb and usually less than the normally innervated skin. Hinsey observed that when a deafferented limb was sympathectomized, it became warmer than the opposite normally innervated limb.

Evans et al. 1 have shown that there is a wide distribution of a substance in the urine and blood serum of mammals which resembles prolan, in that when it is mixed in vitro with the pituitary synergist the effect upon the ovary of the immature rat is enhanced. Evidence that a similar prolan-like substance is present in the blood serum of the non-pregnant and pregnant mare is given in Table 1.∗ Further the data indicate that the prolan-like substance is distinct from the substance in the blood serum of the pregnant mare which is so highly effective in stimulating the gonads when given alone. Following our nomenclature in previous publications we will term the latter substance the gonad-stimulating hormone.

Referring to Table 1 it may be seen that equivalent amounts of non-pregnant and mid-pregnancy mare serum are equally effective in stimulating the gonads of immature rats when combined with the pituitary synergist. The amount of gonad-stimulating hormone present in the 2 types of sera differs greatly. Two cc. of the mid-pregnancy serum when given alone induces follicular development and luteinization whereas 10 cc. of the non-pregnant serum produces no response. In fact, we have shown previously that doses of 60 cc. of non-pregnant serum are without response on the immature rat. Thus it is clear that there is no relationship between the response of the serum when given alone and the response of serum plus synergist. This fact is further demonstrated when the response of early pregnancy blood serum highly potent in gonad-stimulating hormone is compared to non-pregnant serum. The amount of early pregnancy serum injected contained 2 to 4 rat units of gonad-stimulating hormone while the non-pregnant serum contained none. Still the response of the early pregnancy serum plus synergist was less than the non-pregnant serum plus synergist.

Ucko 1 has described a saponin-like substance present in normal plasma but in much larger amounts in active cases of primary anemia. It is more toxic to the reticulocytes than to the mature cells. It seemed to us that the presence of such a substance, rather than a structural defect in the reticulocytes of patients with primary anemia, might explain the fall of 30–60% in the reticulocyte count on incubating sterile defibrinated blood of active cases of primary anemia. 2 Blood from normal persons, from secondary anemia cases, or from cases of primary anemia during remission does not have a fall in reticulocyte count when incubated in the same way. 2 , 3

We therefore took blood from 4 active cases of primary anemia just before starting intensive parenteral liver therapy and preserved the plasma on ice for 4–7 days. Washed cells obtained at the height of the reticulocyte rise were then mixed with twice their volume of fresh plasma, and also, in the same proportion, with plasma obtained before treatment. These tubes were incubated at 37° C. All the blood was handled with sterile technique and counts were made only if there was no bacterial growth on incubation.

In one case, the mildest and least jaundiced (Van den Bergh, indirect, 1.25 units), the blood taken during the active stage had to be transported and cells and plasma were not separated until several hours after being drawn. This patient had 2.4 million red cells per cu. mm. before treatment and in this case the reticulocyte count in the tubes containing pre- and post-treatment plasma was the same after 24 hours incubation. The other 3 patients all had counts under 1.4 million and indirect Van den Berghs of 1.5 to 3 units before treatment. At the time the reticulocyte peak was reached the Van den Berghs had fallen to 0.6 to 0.8 units. In all of these the reticulocyte count fell on incubating the cells in plasma taken before treatment, but not in that taken during the period of high reticulocytosis. In the first 2 cases the proportion of cells to plasma was as 1 to 2 during incubation; in the third case a 1 to 10 suspension was also observed and showed a more striking fall in reticulocyte count in the plasma taken during the acute stage (Table 1).

Using an adaptation of Griffith's and Collins' method for measuring systolic pressure 1 in rats which had developed hypertension as a result of renal insufficiency, 2 we have studied the effect of destroying the central nervous system on the blood pressure of normal and of hypertensive rats. The blood pressure was determined under light ether anesthesia, and then, after inserting a tracheal cannula connected with a device for arterial respiration, the rat was pithed and decerebrated. A cord, passed around the neck under the skin and trachea, was drawn tight crushing the cervical spine and ligating the carotid and vertebral vessels. Through a thoracic laminectomy a wire with a beaded tip was passed up and down the spinal canal. In this way the central nervous system can be destroyed quickly and with little blood-loss, and the preparation, without artificial warming, will last for 5 hours. The blood pressure remains constant during the first half hour, falling slowly thereafter. We followed the pressure during the period from 2 to 30 min. after pithing and decerebration in 10 normal and in 10 hypertensive male rats. The control level in the former was 117.6 mm. Hg average, varying from 85 to 136 mm.; in the latter the preoperative range was 151 to 209 mm., average 176.4 mm. After destroying the central nervous system the average level in the normals was 55.3 mm. Hg., range 42 to 66 mm., while the hypertensive group averaged 54.5 mm., range 45 to 66 mm. In the animals with destruction of the nervous system epinephrine and pituitrin, given intraperitoneally, caused a marked but transient rise in pressure. We therefore conclude that whatever pressor substance may be present in the plasma of rats with renal hypertension has no vasoconstrictor effect but acts through the vasomotor center.

It has been reported 1 that repeated injections of aqueous extracts prepared from fresh whole pituitary glands (sheep and cattle) or from fresh anterior lobes, had a slight but distinct stimulating effect upon the growth of the Passey mouse melanoma. With the Flexner-Jobling rat carcinoma, the Bash ford mouse carcinoma 63, and the Rous chicken sarcoma, no stimulating action was observed.

It was further shown in this institution that the aqueous extracts of the Passey mouse melanoma gave a distinct intermedin reaction on a small fish, called the Elritze (Phoxinus laevis)∗ On the other hand, aqueous extracts of the Flexner-Jobling rat carcinoma, the Sugiura rat sarcoma, the Heiman rat fibroadenoma 308, and the Rous chicken sarcoma gave either a negative or a very weak color reaction.

These experiments suggest that the melanoma-stimulating substance found in pituitary extracts and the intermedin-like substance present in melanoma tissue might be identical.

Since in preparing the extracts of the pituitary, both the anterior and middle lobes were used, there is a good reason to suppose that the hormone of the intermediary lobe of the hypophysis may possess much greater melanoma-stimulating action than that of the anterior pituitary. The following experiments were carried out with the object of testing this supposition.

Twenty young adult albino mice bearing 7-day-old Passey mouse melanoma were each injected subcutaneously with 0.05, 0.1, 0.2, or 0.3 cc. of intermedin† 3 times a week, over a period of 78 days, Each cubic centimeter contained from 2000 to 2500 fish units. The intermedin injections seemed to have no ill effect on the general health of these animals.

In the careful studies of Miller 1 on the effects of general anesthesia on the muscular activity of the gastro-enteric tract, it was shown that the relatively light anesthesia usually maintained with ethylene causes no marked changes in the tone or amplitude of contraction of intestinal muscles. On the other hand during the surgical stage of anesthesia with ether there is marked loss of tonus and almost complete inhibition of rhythmic and peristaltic contractions in stomach, small intestine, and colon. Divinyl oxide has been shown to have pharmacological properties resembling ethylene and ether, to which it is related chemically. 2 Its general physiological effects are less severe than those of ether, although it is a more powerful anesthetic agent. Since the action of divinyl oxide on intestinal movement has not yet been reported upon, it became of interest to determine what its relation might be to ether and ethylene in this regard.

Segments about 2 cm. long from the jejunum of a freshly killed rabbit were suspended by the Magnus method from a muscle lever, in oxygenated Locke's solution at 37.5° C. The drugs were added to the solution to saturation. Repeated trials were made with each drug in varying sequences of application on intestinal segments of 8 different rabbits.

Ether was always found to cause an immediate and marked loss of tone and inhibition of movement of such a muscle preparation (Fig. 1). Upon washing the ether out of the bath, normal tone and contractibility are promptly recovered. Ethylene bubbled to saturation in the bath was noted to cause a slight and transitory loss of the tone of the muscle, with occasional increase in amplitude of contraction (Fig. 2). Divinyl oxide added to saturation in the bath was observed uniformly to increase the tonicity of the intestinal segment.

A pure culture of chicken sarcoma is described (Strain 13) that elaborates a highly virulent virus or viruses capable of producing both sarcoma aiid erythroleukosis.

1. Intramuscular administration of acetylcholine in dogs results in an increase in the blood aniylase titre. 2. Physostigmine markedly increases the blood amylase response to acetylcholine. 3. Previous atropine admiinistration inhibits the blood amylase response to acetylcholine.

That the soluble specific substance of the pneumococcus may remain in tissues infected with this organism even after the inflammatory exudate has disappeared was first suggested by the studies of Dochez and Avery 1 on the excretion of “S” substance in the urine of cases of lobar pneumonia in human beings. They found that the soluble specific substance was excreted in some cases for days or weeks after recovery from the disease. In some instances this continued excretion of “S” could be correlated with delayed resolution, but in other cases this was not the explanation. It would seem probable that in these latter instances “S” remained stored in the pulmonary tissues for some time after the other constituents of the inflammatory exudate had apparently disappeared.

A direct approach to this problem cannot be made in human beings because of obvious difficulties. However, the availability of an experimental animal in which a pneumococcus pneumonia may be induced makes possible the determination of “S” substance in the lungs at various intervals after recovery. The present report is a record of preliminary studies of the persistence of “S” substance in the lungs of dogs recovered from experimental lobar pneumonia.

Healthy adult male dogs were infected with Type I pneumococci by the method of Terrell, Robertson, and Coggeshall. 2 Some of these dogs had been previously “Vaccinated” by spraying formalized Type I pneumococci into the bronchial tree. Two dogs in the group had been similarly treated with a 0.9% salt solution as controls. The “vaccination” procedure had been carried out in the course of a separate experiment but the dogs after recovery were utilized for this study. The other animals in the series had received no preliminary treatment.

A study of the effect of ultraviolet irradiation on the amount of hemoglobin of the blood was made on 3 pens of White Leghorn chickens (10 chickens each pen) grown in complete absence of sunshine. The room, 20′× 30′, received moderate lighting from two 60 watt Mazda lamps. Small brooders supplied with 60 watt carbon filament lamps furnished the brooder heat for individual pens.

Pen 1, the normal controls, received 2 minutes daily irradiation from a quartz mercury arc lamp at 30 inches distance. Pen 2 had no ultraviolet irradiation. Pen 3 received no ultraviolet until after hemoglobin studies on rachitic chickens were made.

The irradiated chickens developed smooth plumage, yellow shanks and bills, and the male birds large red combs. (These organs do not develop until later in the females.) All of these chickens showed every sign of normal growth.

The non-irradiated chickens were less than 2/3 the size of the controls (Table I). Many showed extreme weakness and deformed joints and bones due to rickets. The feathers were rough and the shanks and bills were light, almost white, in color.

At 7 weeks of age hemoglobin studies were made. These average readings were for the 3 pens, 10.3; 8.41; and 6.52 gm. Hb per 100 cc. blood, respectively. (Determinations were made by the Hellige Solid Plane Hemometer reduced to the standard 14.5 gm. Hb per 100cc. of blood.)

For many years it was believed that infection with one species of avian malaria conferred protection against subsequent infection with that species only, and one of the tests by which a given species could be differentiated from another was to inoculate a subpatent or chronic infection of the former with the suspected new species. Recently it has been shown by Gingrich 1 and others that this is not always the case, but this method is still an important test of specificity.

The species used in the present work was originally described by Novy and MacNeal 2 from the common robin in 1904, and the strain employed was isolated from a catbird caught in Syracuse in April, 1934. The other species with which P. vaughani was crossed were secured from sources previously mentioned by the author. 3 , 4 Female canaries were used throughout.

Before inoculating a bird with a latent infection the continued presence of an infection was checked either by microscopic examination or (since this is usually negative) by subinoculation into fresh birds. Inoculations were generally intravenous, and large doses of parasites were usually used. Ten series of 3 birds each (in addition to one or more controls in each series) were used, 5 being chronic cases of P. vaughani and the others consisting of one series each of P. cathemerium, praecox (relictum), circumflexum, elongatum, and rouxi. Each series of chronic vaughani infections was inoculated with one of the other species named, and the experiment was then reversed, using chronic infections of the latter.

The results show that a subpatent (chronic) infection with P. vaughani confers little if any protection against subsequent infection with the 5 other species named, and the reverse of this relationship appears to hold true also, except that there is some indication that a pre-existing infection with P. praecox (relictum) gives a partial immunity to subsequent infection with P. vaughani.

We have previously reported 1 some preliminary observations on totally adrenalectomized-depancreatized cats. Several workers in the past have attempted this double operation upon dogs and in no instance have the animals survived more than a few hours. Such results as have been obtained have led to the conclusion that total adrenalectomy had no influence upon the subsequent course of the diabetes. During the past year we have completed the double operation upon a dog which was observed over a period of 5 weeks.

The dog (a male weighing 11.2 kilos) was totally depancreatized on March 9th, 1934, the left adrenal removed on March 20th, 1934, and the right on May 4th, 1934. The animal lived until June 8th, 1934.

Following the pancreatectomy the animal was easily stabilized on a diet of 300 gm. liver, 50 gm. raw pancreas and 20 gm. sugar daily with 20 units of insulin administered in 2 doses. On this regime the urinary glucose averaged about 4 gm. per day. The fasting overnight blood sugar ranged from 191–254 mg. %. During this period of 10 days the weight decreased to 10.3 kilos.

After the unilateral adrenalectomy the daily glycosuria still averaged 4 gm. on the same diet and insulin, while the blood sugar ranged around 270 mg. %. Reduction of the insulin to 10 units daily keeping the diet the same resulted in an increase of the urinary glucose to 16 gm. daily. The animal was then restored to the previous regime and maintained on it in good health for a further 6 weeks. During this time no alterations occurred in the character of the diabetes while the weight at the end of the period was 9.1 kilos.

Experimental problems requiring the repeated taking of blood pressure have, in the past, required the use of fairly large animals. Various indirect methods suggested for small animals have, as a rule, proven unsatisfactory. Griffith and Collins 1 reported an indirect method for obtaining blood pressure in man by capillary observation with simultaneous arterial compression. This method has now been modified for use in small animals.

White rats under nembutal anesthesia were used. The dorsum of one foot was depilated, either with barium sulphide or, better, with a razor. A special blood pressure cuff was then wrapped about the thigh and fastened with adhesive. This cuff is 2.7 cm. wide and encloses a circular pressure bag 2.4 cm. in diameter. This cuff conforms in design to a portion of a ring enclosed between 2 concentric circles, the outer circle having a radius of 5 cm. and the inner a radius of 2.3 cm. The ‘ring’ or cuff is thus seen to be 2.7 cm. wide, and any convenient length may be used, 25 cm. measured along the greater circle being quite satisfactory. The pressure bag is inflated from a rubber bulb which is compressed by a thumb screw, the whole system being connected through a T-tube to a mercury manometer. This type of cuff shows little tendency to slip, but if it does a stitch or two can be taken through the skin.

The foot is now gently supported by a mould of plasticine, and a drop of immersion oil placed on the shaven area. A strong light, cooled by passing through water, is focused on this area, and it is observed under the low power objective of the microscope.

Oerskov 1 reported that in the cultures of certain fluorescens strains isolated from milk, if the media contains saccharose, masses of tiny non-stainable granules develop which multiply independently from the bacteria. The cultures on ordinary agar plates correspond in every respect to the cultures of other fluorescens strains; on saccharose agar plates the colonies consist mostly of an amorphous substance in which strands of bacteria are embedded. By direct examination under the microscope or by the usual bacterial staining methods in the amorphous substance, no form elements are visible. In dark background preparations it seems to consist of innumerable tiny granules small enough to pass a Berkefeld filter. If the development of the colonies is examined under the microscope, it is seen that the bacilli and the amorphous substance start to grow separately. The amorphous substance grows in tiny round transparent colonies with a more refractile zone at the center. Most of these colonies coalesce later with the colonies of the bacillus and those which remain free grow only to a size of 0.1–0.2 mm. Oerskov did not succeed in growing the granules separately from the bacteria but when a bacillus emulsion was killed by moderate heat and planted, the granule colonies grew between the dead bacilli. The conclusion of Oerskov is that a virus-like organism is growing in symbiosis with the bacillus.

We succeeded at the first attempt in isolating Oerskov's bacillus and were able to verify his most important observations. In the following points our observations supplement those of Oerskov or are at variance with them.

The granules can be cultivated separately from the bacillus. The granule colonies are of 2 distinct types; one is smaller, perfectly transparent and gives no growth in transplants; the other is somewhat larger and, as already indicated by Oerskov, it contains a more refractile central zone.

Impaired functioning of the digestive system was for a long time known to be one of the earliest and most prominent manifestations of experimental vitamin-B deficiency. A diminution of gastric secretion was also reported repeatedly (Miyadera, 1 Shinoda, 2 Farnum, 3 Gilman 4 ). The data concerning the gastric secretion were however not very consistent, and furthermore there were considerable difficulties in interpretation for the reason that the experiments quoted above were complicated by many other factors, particularly by a loss of appetite and usually a high degree of inanition in the experimental animals. Recently it was demonstrated in this laboratory that a marked diminution of gastric secretion occurs in dogs on a diet lacking vitamin B even under conditions where the animals preserve an excellent appetite and well maintain their body weight (Babkin, 5 Webster and Armour 6 , 7 ). The withdrawal of yeast, as a source of vitamin B, from the standard diet practically abolished the gastric secretion in response to sham-feeding, to histamine, and to intra-intestinal administration of alcohol, as well as to introduction of food into the duodenum. The importance for normal gastric secretion of an adequate supply of vitamin B₁ was emphasized recently by Cowgill and Gilman. 8 The fact that with a diet lacking vitamin B the gastric glands are resistant to any kind of stimuli suggests that some of the factors concerned in vitamin B complex may play a vital part in the intimate mechanism of gastric glandular activity. In such a case one might expect the presence of such a factor also in the product of activity, i. e., in the gastric secretion.

The experimental study which we have carried out along these lines has indeed provided ample evidence that at least the anti-neuritic principle, vitamin Bi, is a normal constituent of canine gastric juice.

Two theories have recently received most attention as to the etiology of traumatic shock. These are (1) the absorption of metabolic toxins from the traumatized area (Cannon 1 ) and (2) the local loss of blood and/or plasma (Blalock, 2 Phemister 3 ) The work of various investigators showing certain tissue extracts to be circulatory depressants has frequently been cited in favor of the “toxic theory”. The fact has been largely overlooked, however, that one may obtain both pressor and depressor substances from most tissues, and that either type may predominate according to the method of extraction (Collip 4 ).

The present experiments are upon the effects on the blood pressure of extracts from traumatized limbs obtained by an hydraulic press. This method of extraction probably does not alter the proportions of the various hemodynamic substances present in the tissue.

Twelve experiments were performed on dogs as follows: One animal was anesthetized, and one hind leg traumatized repeatedly by means of a padded hammer until the arterial blood pressure remained at a low level, usually 60 to 86 mm. Hg. The periods of traumatization (2–3 minutes) were interspersed with 10 minute periods for observation of the arterial blood pressure level. In no case did the traumatization break bones or rupture the skin. This leg was then amputated, skinned, and the muscles minced and extracted in an hydraulic press at a pressure of 375 kg./cm². A considerable amount of bloody fluid ran off during the dissection of the leg; this was given to the animal along with the muscle extract in 5 cases, and separately in 7 cases. In all cases the extract was administered to a second animal by means of an Anrep viviperfusion flask to prevent alteration of the circulation volume of the recipient.

Evidence is offered that the essential factor for the production of pregnancy cells in the anterior pituitary of the rat is contained in the luteinized ovary and, further, that this factor is not sex specific.

Twenty-four specimens of Rana sylvatica and Rana pipiens hypophysectomized at the tail-bud stage, and kept in the laboratory for 525 days, whose normal mates had metamorphosed at 65–70 days, were divided into 3 groups. Seven animals were kept as controls, 7 were treated with Collip's adrenotropic 1 hormone, batch AI, and the remaining 10 with his thyreotropic 2 hormone, batch TG. As a result of his own bio-assays (on rats) Collip stated that the thyreotropic hormone was not free from adrenotropic material but that the adrenotropic hormone was ‘fairly pure’. The hormones were administered by intraperitoneal injection, using special fine needles, in 0.05 cc. doses, given once daily, 6 days a week. For the injection an animal was removed from its aquarium and placed on a pad of absorbent cotton wet with ice water. It was then immobilized, ventral side uppermost, by a smaller sheet of wet cotton having a slit cut in it through which the injection could be made. Following the injection the animal was quickly returned to its aquarium.

In 4 days both of the treated groups were dark in color. The adrenotropic group was decidedly darker than the thyreotropic. The only other change noted in the first 5 days was a slight reduction of the tail fin in a few cases.

On the 6th day the hind legs of one animal in the thyreotropic group appeared to be lengthening and by the 10th day all but 2 of this group showed definite growth of the hind legs. From then on the changes normally seen at metamorphosis appeared rapidly. Forelegs emerged from the 11th day on, the tail shortened, the mouth became a transverse slit, characteristic frog spots (R. pipiens) appeared and from the 14th day animals were living out of the water.

In a previous communication 1 we have announced the cultivation of a filterable virus from 2 cases of typical influenza, one occurring in an epidemic of the interpandemic type in 1930 and the other during an outbreak of similar character in 1932. These 2 cultures when inoculated intranasally in human volunteers produced respiratory infections of the general type of the common cold but of considerable severity and with a tendency to greater constitutional reaction than is commonly observed after inoculation of cultures of common cold virus.

In March, 1934, there occurred in a small convalescent home for rheumatic children a local outbreak of influenza. The symptoms of the early cases were almost identical, consisting of malaise followed by fever as high as 104°F. and lasting for 3 days. During the height of the disease there was headache, general aching, nausea and vomiting. This was followed by marked weakness for several days. In the later stages of the epidemic the picture was that of a severe cold with only a slight rise in temperature. The leucocyte count was elevated rather than lowered as in some outbreaks of influenza.

Chick embryo tissue cultures were made from filtered nasopharyngeal washings obtained from 2 of the early cases of the disease. One of these cultures in the 18th generation, 2 months after isolation, was inoculated intranasally into 3 human volunteers under rigid quarantine. The results are presented in the following protocols.

Volunteer 93, a heavy, muscular man with no recent history of respiratory disease was inoculated intranasally with 3 cc. of culture R18. The inoculation was repeated 5 hours later. The next morning, 18 hours after the first inoculation, the subject manifested nasal obstruction, sneezing, mucoid nasal discharge, a productive cough and a slight malaise.