Experimental Biology and Medicine

Although it has been demonstrated 1 , 2 , 3 that guinea pigs can be readily sensitized to neoarsphenamin, Frei 1 reported that he could achieve no hypersensitiveness to this drug in rabbits. In view of the findings of Sulzberger and Mayer 4 that guinea pigs bred in different countries reacted differently to the same brand of neoarsphenamin, experiments have been carried out to determine whether the rabbits bred in Peiping could be sensitized by this compound.

The brand of neoarsphenamin used throughout the work was that prepared by Hoechst Company, Germany, The ampules used for each experiment were from the same batch. 0.15 gm. of the drug was dissolved in 10 cc. of doubly distilled water and the solution diluted to 100 cc. with sterile normal saline. Only freshly prepared solutions were employed.

Observations have been made on 50 native bred adult male rabbits; individual body weights varied from 1350 to 2320 gm. All animals were albinos and the majority were of a delicate build with a slender head and moderately short ears. Thirty rabbits had been inoculated with Treponema pallidum from 4.5 to 34 months previously. At the time of the present experiments, only 4 showed active syphilitic lesions, the remainder being in the latent stage of the infection. Twenty-four hours before each experiment, the skin over both flanks was shaved; only ordinary soap was used. In each instance, the right flank was used for the first, and the left for the second injection, both being administered intradermally. The second or “testing” injection was made 4 weeks after the administration of the sensitizing injection. In the case of 8 normal and 12 syphilitic rabbits, 0.1 cc. was used for both the sensitizing and the testing doses.

It is well recognized that variations in the intensity and constancy of hypersensitive cutaneous reactions occur in the human subject with respect to different regions of the body. As far as similar variations in animals are concerned, however, few observations appear in the literature. The object of the present study was to ascertain whether different skin areas of guinea pigs and rabbits sensitized to would react in a similar or in a different manner to subsequent intradermal injections of this drug.

The experiments included 9 adult male albino rabbits and 12 adult guinea pigs; in the latter animals the skin of the flanks and abdomen was non-pigmented. The solution was prepared and the animals were shaved according to the technique described in the preceding report. 1 The rabbits were sensitized and tested for hypersensitiveness with 0.2 cc. and the guinea pigs with 0.1 cc. The injections which were given with a sharp tuberculin needle and a. tuberculin syringe calibrated to 0.01 cc, were made as equally deep in the epidermis of the flank and abdomen as possible; the wheals formed were of approximately the same area and depth. The test for hypersensitiveness was made 5 weeks after the first or sensitizing injection. The reaction was read 24 hours after injection, the width and breadth of the involved skin area being measured.

The results shown in the accompanying table indicate that there was a distinct variation in the reactions on the flank and abdomen in every animal, those on the flank being larger than those on the abdomen. Furthermore, in 8 of the 12 guinea pigs and in 4 of the 9 rabbits, there was practically no reaction in the abdominal area in contrast to the definite flank reaction in each case.

Two strains of mice, both albino Mus musculus and here designated as H T M and E L T M, are known to differ in the number of tailrings as previously reported (Fortuyn 1 ). They also show differences in the number of young per litter which the mother, in the absence of the father, is able to raise to an age of one month. In each strain this has been tested in the offspring of a single pair so as to reduce variability. Table I gives the size of the litters. In 45 first litters of H T M the average number of young which lived with the mother to an age of one month is 4.75 ± 0.48 (mean error). For 28 second litters this figure is 5.60 ± 0.48. In 32 first litters of E L T M the average number of young, one month after birth, is only 1.53 ± 0.37, in 39 second litters it is 1.74 ± 0.37. The differences with their mean errors amount to 3.22 ± 0.53 for the first litters and 3.86 ± 0.60 for the second, and are therefore statistically significant in both cases. This difference in fecundity is explained by the fact that in H T M only 5 first litters and 2 second ones are completely destroyed by the mother, whereas in E L T M these figures are 18 and 19. Moreover, in those cases where the litter is not completely destroyed, more young survive in H T M, which may have litters of 10 and 11 young than in E L T M which never had litters of more than 7 young. No observations have been made on the question of whether death occurs shortly after birth through the inability of the young to suck, or because the mother has no milk, or kills the young independently from either condition.

Citrated whole blood has been shown to be inadequate for the study of pneumococcus growth inhibition on account of the detrimental effect of sodium citrate upon the organisms on the one hand, and upon the leucocytes on the other. A modification of the original method of Robertson and Sia is proposed. Because of the simplicity and the satisfactory results it gives, this technique may find a wider application for similar studies with other organisms.

One of us previously showed that the pneumococcus soluble specific substance exerted a specific antiphagocytic action on the blood in vitro. 1 This finding has since then been confirmed by others. 2 , 3 , 4 In vivo attempts, on the other hand, have not so far given similar striking results. By injecting Type II soluble substance together with an attenuated Type II pneumococcus in mice, Felton and Bailey 5 found that it exhibited an antagonistic effect on the defense of the animals or an actual increase in the virulence of the organism from 10-100 folds. We wish to record the results of a study along similar lines but with more striking effects.

The soluble specific substance employed in this work was prepared from a Type II pneumococcus according to the method of Heidelberger and Avery. 6 The chief organism used was a strain of a typical Type II pneumococcus, highly virulent for white mice, but relatively avirulent for rabbits.

Preliminary experiments with a modified method 7 for demonstrating growth-inhibitory and pneumococcidal action of the blood before and after the intravenous injection of Type II pneumococcus soluble substance, showed that while rabbit blood normally was capable of destroying large numbers of Type II pneumococci, blood taken after the injection of the substance had entirely lost this property.

The effect of the intravenous injection of the soluble substance on the course of pneumococcus bacteremia in rabbits was next studied. It was found that when a sublethal dose of the Type II pneumococcus was injected intravenously into normal rabbits, the organisms quickly disappeared from the blood stream; whereas in rabbits receiving soluble substance and similar amounts of pneumo-coccus culture, an increase in the bacteremia ensued and frequently terminated in the death of the animals.

A medium consisting of 1% dextrose beef infusion agar pH 7.8 and type specific antipneumococcus serum is presented. Pneumococci of the homologous type when grown in the depths of the immune serum-agar plate develop well defined annular opacity surrounding each colony. The development of the annular opacities has been found to be strictly type specific.

Six rabbits with transparent chambers introduced into the ears according to the Sandison 1 technique were studied. The blood vessels studied included both preformed, and newly formed ones of from 10 days to 4 months old. The reaction to ephedrine of these various types of vessels was essentially the same.

Ephedrine hydrochloride in 0.1% to 1.0% solutions in 0.9% saline was introduced by intravenous injection into the marginal vein of the unobserved ear or the saphenous vein of the leg and continuous observation with the microscope under low power was made during the first half hour after the injection.

Ephedrine in doses from 0.1 to 0.8 mg. (0.1% solution) injected into the blood stream did not produce a change in the vessels of the ear chamber. A slight narrowing of the central artery and a slowing of the circulation in the vascular bed was observed when the amount injected was 1 mg. (1% solution). The flow in a few capillaries became reversed, while in a few others it was accelerated, although the calibre of these showed no changes. The effect took place in a half to one minute after the injection and disappeared in about 5 minutes.

With a dose of 2 mg. (1% solution) the constriction of the central artery was more marked. The wave of constriction started from the central portion and travelled toward the periphery. This appears to differ from the effect exerted by epinephrin, which usually produced a wave starting from the opposite direction, namely from the periphery to the center. The arterioles, venules and capillaries again showed no changes in the size of the lumen. A slowing or reversal of the flow in a number of these minute vessels was apparently secondary to the changes in the arteries.

Seidell and Fenger 1 observed very striking seasonal fluctuations in the amount of iodine in the thyroid glands of hogs, sheep and beef, more being present and the gland being smaller during the summer than during the winter season. Fenger 2 confirmed these results. Martin 3 working with sheep glands arrived at similar conclusions. Loeb 4 in studying the compensatory hypertrophy of the thyroid gland of the guinea pig noted that this hypertrophy was considerably less in the summer than in the winter months.

We may then assume that during the colder periods of the year the thyroid gland is much more active, perhaps owing to the fact that more demands are made on this organ than during the warmer season. As a result of the lessened need for thyroid hormone during the summer months, iodine, instead of being used in the metabolism of the gland is stored up in the acini.

It has now been established that the functional state of the thyroid gland is to a marked extent influenced by a hormone of the anterior pituitary. Following previous observations of Allen, 5 Smith, 6 and Uhlenhuth and Schwartzbach 7 in amphibian larvae, Loeb and Bas-sett 8 observed that in mammals (guinea pigs) injections of extracts of the anterior pituitary cause a marked hypertrophy and hyperplasia of the acinus cells of the thyroid, a liquefaction and absorption of colloid, indications of an increased hormone action (loss of weight), therefore, in general, changes corresponding to those observed in Graves disease. To these changes corresponded a marked rise in basal metabolism under the influence of anterior pituitary extracts. (Siebert and Smith. 9 )

Under these conditions it was conceivable that the seasonal variations in the thyroid gland were due to corresponding variations in the activity of the anterior pituitary.

Several reports deal with the preparation of dry antisera. Evaporation in vacuo and precipitation in the cold with alcohol or acetone are the methods most used. According to reports dealing with the latter method, the low temperature has been assumed to be the factor preventing denaturation and loss of antibody activity. We have found that the concentration of the organic precipitant is an equally important factor. There exists a critical concentration of the organic solvents methyl, ethyl and propyl alcohol and acetone, in the range of 60% to 75% concentration at which denaturation of serum proteins is maximal. As the concentration is increased from about 75% the degree of denaturation is decreased until at final concentrations of 90% or above serum proteins can be precipitated by these organic solvents at room temperatures without loss in solubility or antibody activity.

Applying this phenomenon the following method has been used to prepare dry immune sera. To 10 or more (but not less) volumes of acetone add slowly with shaking one volume of the serum. Collect the precipitate on a filter, wash once with acetone followed by 3 washings with anhydrous ether, the precipitated mass being stirred with a wooden spatula after each ether addition. Approximately 5 volumes of ether to each original volume of serum is required for each washing. The final white mass is spread out on the filter paper and placed in the 37°C. incubator for about one hour. The resulting dry mass is readily pulverized with the wooden spatula to an extremely light, white, fluffy powder. This powder is slowly (due to slow wetting) though completely soluble in distilled water or physiological sodium chloride solution. There has been detected no loss in agglutinating activity, hemolytic activity or antitoxin content.

In the fall of 1930, we began a study of the origin of immunity to diphtheria among the Central and Polar Eskimos and of the nature of their throat flora. The materials were collected from 115 Eskimos by Dr. Peter Heinbecker during the late summer of 1930 and in a report soon to be published 1 we have stated that no hemophilic organisms were found. At the beginning of the current year we continued this study with a new supply of approximately twice the number of throat cultures and with additional safeguards against the dying off of these more delicate organisms, in case they should be present in the original materials.

From this latter group of throat cultures, 18 strains of Gram negative hemophilic bacilli possessing the accepted characteristics of Hemophilus influenzae were isolated, and this preliminary report is made to supplement and correct our statement already referred to, while a more detailed and complete description of these strains will appear later in the Journal of Infectious Diseases.

All these strains refused to grow on plain agar and only occasionally produced scanty growth on infusion agar to which filtered raw tomato juice had been added. On the contrary, all strains grew well on fresh blood agar and still better on this medium after the addition of tomato juice. The “satellite” phenomenon was very marked with many of the cultures, especially around colonies of hemolytic staphylococci, thereby strengthening the evidence that they required both a vitamine-like factor and hemoglobin for growth. Several strains were hemolytic and all strains tested showed nitrate reduction, indol production, ability to ferment dextrose, and bile solubility.

Of 5 strains saved for a more complete study, 4 were hemolytic, capsulated, and agglutinable by type “A” serum of Pittman. 2

In testing the therapeutic value of the lysed cultures of B. diphtherias it became necessary to destroy the slight amount of toxin present with bacteriophage in some of the filtrates. Since bacteriophage is comparatively resistant to heat it was thought that simple exposure to heat may destroy the toxin without destroying the phage present in the lysates. This, however, we were not able to accomplish. Exposure to heat sufficient to destroy all the toxin caused almost complete destruction of the phage. In a study of the mechanism of inactivation of phages by alcohol we concluded that the effect of alcohol was not due to the direct destruction of the active agent, but to the denaturation of the protein vehicle on which the lytic agent was adsorbed. 1 Suspecting that somewhat similar relation may exist in the inactivation of phages by heat, we attempted 2 series of experiments.

In the first we made use of our earlier finding that the addition of polyvalent cations to the medium containing phage protected it from inactivation by alcohol. 2 We found that the addition of enough CaCl₂, for instance, to bring its concentration between 0.002 m and 0.015 m, some degree of protection against heat inactivation may be secured. Optimum concentration of CaCl₂ differs with each phage, and it thus becomes necessary to find a suitable concentration for each phage and even for each batch of the same phage, a laborious and time-consuming procedure.

In later experiments we used another method suggested by Spiro that glycerin, carbohydrates, urea, urethan, etc., prevent denaturation of proteins during coagulation. 3

Kendall employed a culture medium 1 consisting presumably of whole undenatured protein, devoid of all products of digestion which he referred to as peptones. He reported the transformation of ordinary bacteria into primitive forms capable of passing filters presumably impervious to the passage of normal bacteria. 2 Kendall concluded that he was dealing with unusual forms of bacteria, not only on the basis of their filterability, their appearance under the dark-field microscope, and their irregular staining, but also because of his inability to obtain sub-cultures by transplantation of the “filterable” forms to ordinary media, whereas sub-cultures were obtained by similar transplantation to his own, so-called K medium.

We here summarize some of our experiments to acquaint ourselves with this apparently new and important phenomenon.

The appearance of viable and cultivable bacteria in the filtrates of bacterial cultures has been repeatedly recorded. It is generally understood, that the mere passage of cultivable bacteria through filters does not predicate any radical change in their original size. It has been repeatedly shown that under a variety of conditions, normally effective filters may become permeable to bacteria of ordinary and even large size.

Even the first few experiments convinced us that in the case of cultures grown on K medium, the frequency of passage and the numbers of viable units of bacterial protoplasm appearing in the filtrates at each filtration were considerably greater than in the case of cultures grown on ordinary media. Furthermore, the filterability of cultures grown in K medium was the greater the more profuse was the growth attained at the time of nitration. Increased growth could be secured by the addition of carbohydrates or peptone to the K medium and in either case the filterability was increased, thus it is evident that the exclusion of peptones from the medium is certainly not responsible for the increased filterability of Kendall's cultures.

Observing carefully the progress of lysis under the microscope we noticed that disappearance of individual bacteria under the influence of phage is usually preceded by more or less marked swelling of the cells. 1 A more detailed inquiry indicated that the swelling itself was due to imbibition of water which in turn appeared to be the result of intracellular digestion of bacterial cytoplasm, 2 and consequent increase of osmotic pressure within the cells. The actual disappearance of cells from the field was therefore interpreted as due to bursting. 3 We found further that if bacteria are grown on media in which free water was immobilized by the addition to the medium of hydrophilic colloids, the swelling was prevented and consequently lysis did not occur. On the contrary, in place of a marked fall in the number of bacteria (as would be the case if lysis had occurred), we observed actual increase in the number of bacteria when compared with control cultures grown on the same medium but in the absence of phage. 4 While the above observations were made with Gram-positive and Gram-negative bacteria, the changes were always more constant and more marked in the latter and therefore Gram-negative bacteria were used to illustrate the phenomenon. 5 Recently there appeared a paper in which the authors, 6 working with cultures of Staphylococcus, failed to observe either the swelling or the increased rate of growth of these organisms in the presence of phage. It was imperative to reinvestigate the subject.

The experiments here reported were performed to show that phenomena observed by us during the growth of Gram-negative bacteria in the presence of phage occur also with Gram-positive bacteria.

In a previous paper the author in collaboration with Richardson 1 described variations in the cytolytic changes occurring in undifferentiated and differentiated tissues which were induced to undergo prolonged proliferation in an unchanged medium in vitro. The present study was undertaken to extend similar comparative observations upon the rate of cytolysis in pure isolated strains of cells, each possessing different inherent growth rates. Under experimental conditions Parker and Fischer 2 depicted different strains of “Fibroblasts” possessing varying growth potencies, and, for the purpose of these experiments similar tissues were selected and isolated according to the methods described by these investigators. All tissues employed were taken simultaneously from the same embryo of 12 days' incubation and throughout this investigation were explanted under exactly identical conditions from the moment of their isolation in vitro. The selected fragments were obtained from the following regions : osteoblasts from the supra-orbital; chondrioblasts from the periostium of the sphenoid (fast growing); and fibroblasts from heart and leg muscles (slow growing).

The first cytolytic response to the accumulations of cyto-toxins in the unchanged medium was morphologically expressed in the slow growing strains after 65-70 hours incubation in vitro. Similar examinations, however, of the fast growing strains after the same periods of growth showed them to be more resistant to the factors incurring cytolysis, as degenerative phenomena were not yet apparent. Even when the fast growing osteoblasts and the slow growing muscle-fibroblasts were implanted side by side in the same culture medium, they not only showed the same relative difference in their growth rates, but exhibited the same time variations as when cultivated in their isolated conditions.

1. The ovulating dose of an anterior pituitary extract for rabbits is very much less than that of Prolan when the potency of these 2 substances is standardized on immature female rats. 2. This difference in quantitative activity of Prolan and anterior lobe extract suggests a physiological difference between the 2 substances.

The use of Roentgen rays as an adjuvant in the treatment of human poliomyelitis has been widely recommended by a number of clinicians 1 while others 2 have expressed their doubt regarding the value of this form of therapy. Experimentally, the problem has apparently never been approached save for some earlier work of Amoss, Taylor and Witherbee. 3 These experiments, although originally undertaken for a different purpose, served to demonstrate that monkeys subjected to large doses of X-ray developed a more fulminant type of infection than did the controls.

In view of the conflicting clinical reports and in the absence of pertinent experimental data, it appeared worthwhile to investigate whether poliomyelitic monkeys upon exposure to very small doses of X-ray, such as are used successfully in a variety of acute inflammatory conditions, would not benefit from such treatment. There seemed to be at least a possibility that the quicker regression of paralysis alleged to have happened occasionally in human cases after early treatment with X-ray, might have been due to an acceleration in resorption of the perineuronal edema, mechanical pressure of which is generally conceded to exert a deleterious effect on the ganglion cells long before actual destruction has resulted in permanent damage. 4 At the same time it was interesting to determine whether raying of virus in vitro with much larger doses would affect the virulence of the infectious agent.

The experimental data submitted in this paper deal with observations on a total of 12 rhesus monkeys, exclusive of controls, which are distributed over the following 3 groups:

The Brown-Pearce rabbit tumor is a transplantable malignant epithelioma carried in this laboratory for more than 100 generations by intratesticular inoculation. It has been used extensively for the study of animal constitution and it has been found possible to alter the susceptibility or resistance of animals to this tumor by various surgical and environmental procedures. About a year and a half ago attempts were made to alter the resistance or susceptibility of animals to inoculation with this tumor by the use of material derived from the tumor itself. The essential feature concerned the preparation of the material for use in conjunction with a regular tumor inoculation.

The first attempt resulted in a marked enhancement. 1 A rabbit which had died 5 to 10 hours previously of the Brown-Pearce tumor was placed for 2 weeks in the ice-box (26-32 °F.) At the termination of this period, a normal saline emulsion of the primary tumor (testicle) was made and 0.3 cc. inoculated into the right testicles of each of a group of normal young adult rabbits. Two weeks later 0.3 cc. of a normal saline emulsion of fresh Brown-Pearce tumor was inoculated into the left testicles of each of the same group .of rabbits and also into each of a control group not previously inoculated.

This first experiment was repeated 7 times, and the only change in the experimental procedure was to remove the tumor from the rabbit, imbed it in paraffin, and preserve the imbedded tumor in the ice-box before using. All experiments were terminated at the end of a 2 months' period after the last inoculation and the results in every experiment confirmed the original observations that a significant enhancement of malignancy had occurred.

In male germ cells the well-known dictyosomes and their derivatives, the acroblasts, are vigorously blackened by silver or osmium impregnation methods. Therefore they have been termed “Golgi bodies” and are accepted as complete homologues of the Golgi-apparatus of mammalian nerve and gland cells. The Golgi-apparatus is believed to be concerned with the function of secretion. An example frequently cited is that the acrosome of the animal sperm is secreted by the dictyosome complex, involving the tacit assumption that this complex is the homologue of the Golgi-apparatus. Recently this homology has been questioned. Parat 1 and his coworkers, for example, consider the dictyosomes to be chondriosomes of large and active type, partly for the reason that the dictyosomes (Parat's “lepidosomes”) of the living cell stain with Janus green nearly as vigorously as the chondriosomes. Both the staining reaction and the implied homology have been doubted. 2 , 3 It therefore seemed desirable to reinvestigate the problem.

The present account presents some results of intra-vitam staining of metamorphosing male germ cells of various insects. The species used were 2 Orthoptera of the family Acrididae, Rhomaleum micropterurn and Melanoplus femur-rubrum, and 3 gryllid Orthoptera, Œcanthus nigricornis, Nemobius fasciatus and Gryllus assimilis var. *check*luctuosus. In the experiments with Janus green an hemipter, Euschistus euschistoides, was also used.

We have shown that several of the organic cyanides 1 when injected subcutaneously into rabbits daily for 3 or more weeks produce thyroid hyperplasia. Methyl cyanide produces the greatest thyroid reactions, and young rabbits (3 to 4 months) are more reactive than adults to the same doses per kg.

Chronic bilateral exophthalmos has occurred in a significant number of prepubertal rabbits after methyl cyanide has been given for varying lengths of time. Exophthalmos has developed as early as the 20th day after beginning daily injections of 0.1 cc. of methyl cyanide in 3 months old rabbits. Of the 2 breeds used (Dutch and Belgian) we have observed it most frequently in young Dutch rabbits. It has not been detected in adults (6 months and over) in either strain. Representative instances are given in Table I.

So far we have observed only moderate degrees of exophthalmos. It occurs earliest in those rabbits which develop thyroid hyperplasia the quickest and appears to be proportional to the degree of thyroid hyperplasia. Rabbits that have failed to develop thyroid hyperplasia, or rabbits in which the hyperplasia has been slight have not developed exophthalmos. When exophthalmos is developing the rabbits become nervous and fidgety. This is frequently associated with soft feces and an increase in the volume of urine.

No attempt has been made to determine what rôle the hormones of the posterior pituitary and chromaffin tissue may play in the production and maintenance of the exophthalmos. There is, however, some hypertrophy and great hyperemia of the medulla of the suprarenals in these animals.

Wide discrepancies exist in the literature as to the life-span of adrenalectomized rats. Most of the work between 1900 and 1930 indicated that approximately 50% of these mammals survived adrenal ablation for a month or indefinitely. This survival was attributed to the presence of accessory cortical tissue. Recently, however, Pencharz, Olmsted and Giragossintz 1 , 2 reported that the rat is no exception to the rule that adrenalectomy is fatal in mammals. Freed, Brownfield and Evans 3 stated definitely that adrenalectomy was uniformly fatal in rats if one-quarter inch of the pedicle was removed along with the gland. Kutz 4 stated that out of 57 animals of his strain operated at 4 weeks age, 56 were dead by the tenth day. While these results were being reported other workers have been reporting a high percentage survival. The present study was undertaken in an attempt to give some explanation to this chaos of evidence.

A series of 156 adrenalectomies were done to investigate the effects of different types of operation in animals of different ages and different strains. Rats from 35 days to 10 months age were used from 5 different colonies. To demonstrate that the animals were really dying of adrenal insufficiency a series from all colonies were revived from their terminal coma with the cortical hormone, kindly supplied by Drs. W. W. Swingle and J. J. Pfiffner. Also to check the effect of surgical trauma a series of control. operations were done. Rigid asepsis was maintained in the operations. A balanced diet containing all necessary food factors was used. No animal was used unless in apparent perfect condition of health.

Numerous theories have been advanced and investigations made to account for differences in voice quality based upon a study of speech sound after it leaves the mouth. Such studies have been in many instances unreliable because formerly it was almost impossible to make an accurate study of the intra-oral, pharyngeal, and laryngeal mechanisms, which are among the important structures used to produce speech and singing.

About 20 years ago it was appreciated that the X-ray, by outlining the forms and sizes of speech and singing cavities, could be the means of uncovering many of the hidden secrets of phonation. Myers 1 demonstrated a certain technic by which these structures were visualized on an X-ray plate. Later, Stephen Jones 2 devised the chain technic, which consisted of the passing of a small chain through the nostril of a subject who was instructed to swallow the loose end. An X-ray plate was then made to show the position of the chain in relation to the soft tissues. Russell 3 used a fine thread which apparently was impregnated with some substance opaque to the X-ray. The subject was made to swallow the loose end of the thread, which was supposed to remain by capillary attraction on the middle portion of the tongue. We are unable to find mention by any of the numerous investigators that the palate was satisfactorily outlined by any marker except the normal bone as shown on the X-ray film.

It must be appreciated that a mechanical means employed in outlining the forms and sizes of speech and singing cavities, such as the chain technic, gold foil, or the thread technic, could interfere with the normal functioning of the anatomic structures which produce speech.

The cyanosis which develops in an arm with the circulation totally or partially occluded, may fail to appear when the part is surrounded by an atmosphere of oxygen. The skin of the forearm is usually maintained free of cyanosis under these conditions, while the fingers especially and the hand are quite refractory. This investigation, suggested by the work of Shaw and his coworkers, 1 is quite consistent with their finding that oxygen is absorbed by the skin. However, contrary to the interpretation which they place upon their work, we must conclude that, under the conditions of our experiments, the oxygen finds its way into the blood of those vessels which give color to the skin.

The reactive hyperemia, which invariably appears upon the release of a circulatory occlusion of an arm in air or nitrogen, is absent in an arm which has been in an atmosphere of oxygen during the period of circulatory arrest, except in those parts which were cyanosed or “dusky” before the release, i. e., the fingers and hand. The oxygen, therefore, would seem to prevent the dilatation of the skin vessels which prevails in an arm with arrested circulation.

The failure to obtain a reactive hyperemia under these conditions was examined in the light of the hypothesis of Lewis, 2 which ascribes the control of the small vessels of the skin to an “H” substance, a normal metabolite, the identity of which with histamine Lewis believes to be established by a wealth of convincing evidence.

The findings here recorded would seem to indicate that either histamine is not produced or it is destroyed under prevailing conditions. The idea as developed by Lewis precludes the first possibility, since if one assumes that dilatation of the small vessels of the skin under a variety of physiological conditions is due to the “H” substance, one must of necessity, as indeed Lewis does, postulate that the substance is a normal skin constituent, present in but small amounts normally, but piling up when the circulation is arrested.

In studies involving the fractionation of blood phosphate compounds by means of hydrolysis with N HCl certain difficulties were encountered in the determination of the resulting orthophosphate by Kuttner's colorimetric procedure. 1 The results were usually too low, and analyses of known quantities of P under similar conditions likewise gave values below those expected. Although inclined to attribute our difficulties to the losses of HCl during hydrolysis, we had the same trouble even when we prevented such alterations in concentration. We had no more trouble when we substituted N H₂ SO₄ for the N HCl. Our analyses on known quantities of P were also entirely satisfactory when H₂ SO₄ instead of HCl was used for the hydrolysis. This led us to investigate the probable influence of different salts in high concentration on the orthophosphate values determined by the Kuttner method. We discovered subsequently that Rimington 2 had already pointed out the effect which the concentration of various salts used as anticoagulants may exert upon the quantitative determination of P in blood by Brigg's procedure. We, nevertheless, present these results to call attention once more to this important matter, and thus save other investigators the time and trouble it has cost us to find out this simple fact. The warning is perhaps especially needed now, inasmuch as hydrolysis by N HCl is generally employed in the study of various phosphate fractions in blood or muscle.

On neutralizing this extra acid before carrying out the colorimetric test a rather high salt concentration is produced, which in our experiments, on account of the final dilution used, amounted to 0.2 M. Not all salts interfere with the colorimetric reaction, as will be shown presently, but the chlorides and nitrates do interfere seriously.

Plasma protein deficiency results in a decrease of the osmotic pressure of the blood and has been held to be part of the explanation of edema in nephrosis and certain cases of undernutrition. If this postulation be correct the capillary pressure would be in excess of the osmotic pressure of the blood and as a result of the increased transudation a state of oligemia should be expected to prevail under such conditions. That this is actually the case in nephrosis has been shown by Darrow 1 and Waterfield. 2 The behavior of blood volume in nutritional edema has, however, not received the attention it deserves. The present report deals with the blood volume changes in both of these diseases, special attention being directed towards correlating such volume changes with the fluctuation of the plasma protein.

The plasma protein was determined by the micro-kjeldahl method and the blood volume by inhalation of carbon monoxide according to the technic reported previously. 3 Hematocrit readings were made by the method of Osgood. 4 In the cases of nephrosis the 24-hour urine output for the day of examination was also ascertained. In all, 5 cases of nephrotic kidney condition and 7 cases of nutritional edema were studied. The findings are given in Tables I and II.

On examining the data two points deserve special emphasis. First, in both diseases, when not complicated by severe anemia, the blood volume expressed in cc. per square meter of the surface area (calculated from patient's height and wight) was very much lower than the normal standard determined by the same method. 3 This is in agreement with the previous work on nephrosis 1 , 2 and further corroborates the belief that the edema of these 2 diseases may be placed under the same category.

Twenty-six animals were operated upon for the production of jejunal ulcers using a slight modification of the technic of Mann and Williamson 1 for depriving the jejunum of its normal alkalinity.

In each case the duodenum was transected between the pylorus and the ampulla of Vater and both ends inverted with silk. The jejunum was then divided a few inches beyond Treitz's ligament and again both ends inverted. Following this a gastro-enterostomy was performed between the distal end of the jejunum and the stomach and an enteroenterostomy performed between the closed duodenojejunal loop and the upper ileum about 24 inches below the gastroenterostomy, allowing the bile, pancreatic juice and duodenal secretions to be emptied back into the small bowel at this point. These changes were made in the original procedure because it was felt that the nutrition of the animals would be better preserved if the drainage were done higher in the small intestine, thereby eliminating this factor from the ultimate results. Also since a lateral anastomosis has a better blood supply than one made end-to-end, there would be less chance of criticism on this score as a possible source of ulcer formation.

Ten of these animals formed typical chronic ulcers varying in size from 1/2 to 2 1/2 cm. in diameter. The period necessary for the formation of the ulcers varied from 42 to as long as 428 days with the average of 118 days. The ulcers were multiple in 3 cases. They always formed on the wall of the jejunum opposite the gastroenterostomy stoma and never involved the suture material which was of silk throughout, except in one instance, where the edge of a very large perforating ulcer touched it in one spot.

Since the original work of Mann and Williamson 1 much has been written concerning the experimental production of duodenal ulcers by depriving the duodenum of its normal alkalinity. It, therefore, becomes of interest to know if any one of the 3 secretions, bile, pancreatic juice, or intestinal juices, is specifically responsible for protection against ulcer formation, or whether the presence of all of them is necessary.

In an attempt to determine this point 3 series of experiments were done. In the first series 5 operations were performed under aseptic technic and the common bile duct isolated, tied, and divided in each case. Following this a cholecyst-enterostomy was performed at a point on the small bowel about 18 inches below Treitz's ligament, thereby excluding the bile from the duodenum and draining it into the jejunum at this point. The anastomoses and closures were done with silk. Two of these 5 animals developed an acute duodenal ulcer at 76 and 119 days postoperative respectively. In each case the ulcer was situated exactly at the pyloric ring and was about 3 mm. in diameter. Both died of hemorrhage into the bowel without warning and the ulcer was only discovered at autopsy. Pathologically these ulcers were very different from chronic ulcers produced by duodenal drainage. They were smaller, more punched out, showed no sign of secondary infection, had hemorrhagic bases ulcerating into a vessel in each case, and presented no evidence on the serosal surface by which ulcer formation could have been predicted. Microscopically, they showed very little infiltration and no scar tissue formation. Both animals were in good general condition at the time of death. The 3 others were well nourished and showed no sign of ulcer formation at the date when they were sacrificed, 234, 236, and 239 days postoperative respectively.

Prevailing methods for the determination of the fragility of erythrocytes are based upon the theory of their hemolytic stability when brought into contact with chemicals such as saponin and bile salts, with specific sera, or with hypotonic salt solutions. Theoretical as well as practical considerations have directed the development of these studies toward improving the technique for determining the resistance of the red blood cells to solutions containing different concentrations of various salts. Of the many variations in the original method of Ribierre, 1 that of Simmel, 2 or one of its modifications, 3 appears to be the best. However, the necessity of freshly prepared hypotonic salt solution of very exact concentrations, together with the labor in using the counting chamber and the necessity for setting up a control test with a known normal blood at each observation, are definite disadvantages.

We have devised a fragility test in which the difficulties mentioned are largely eliminated. The testing of the relative fragility of the patient's cells in varying dilutions of his own plasma is the essential principle introduced in this new method.

Technique. Test tubes of 20-30 cc. capacity and capable of withstanding the high speed of the centrifuge are fitted with rubber stoppers. To each is added 2 mg. of powdered heparin, an amount sufficient to prevent the clotting of 10 cc. of blood for approximately 24 hours. After weighing out this quantity a few times to visualize the approximate volume involved, the quantity of heparin added to each test tube may be estimated without disturbing the accuracy of the test. Ten cc. of blood are drawn from one of the arm veins, using a dry (not rinsed in salt solution) 10 cc. syringe and needle.

Wickman 1 , 2 dwelt at some length upon the lesions of the central nervous system associated with acute poliomyelitis, and as a result of the accuracy of his original observations his descriptions of the changes encountered have been little altered in subsequent years. Wickman noted that the vascular lesions of acute poliomyelitis were more marked in the veins than in the arteries, and that the frequently encountered round cell infiltration was situated in the lymph channels of the vessel wall. It was also recognized, however, that the lymphocytic infiltration was often of sufficient intensity to extend beyond the adventitial limits of the vessels into the surrounding tissues. Flexner and Amoss 3 pointed out that in cases of acute poliomyelitis induced by intravenous administration of virus the vascular lesions were more extensive than in instances of infection by other routes.

The lesion described below was encountered in the meningeal vessels of a Macacus cynomolgus which developed acute poliomyelitis after having been fed with virus-infected milk. The material used was a 10% milk suspension of recently glycerinated brain and spinal cord from monkeys which had succumbed to typical poliomyelitis. Six daily feedings of 30 cc. were administered by mouth with the aid of a medicine dropper, the milk being fed slowly so that the monkey could swallow it easily.

Symptoms were first apparent 11 days after the last feeding, at which time the animal moved about slowly and showed obvious weakness of the left leg.

Eventual interepidemic control of typhus fever will depend upon the effective sanitary utilization of the epidemiological facts now available. Even with our almost complete knowledge of transmission and animal reservoir, however, the circumstances under which typhus epidemics occur usually exclude any hope of adequate control. Students of typhus, therefore, have long felt that in this disease particularly a method of specific prophylaxis was needed.

The Rickettsia provaceki may now be accepted as the established etiological agent, and all rational methods of vaccination must be based on (1) active immunization with Rickettsia, killed, attenuated or in subinfectious doses; (2) immunization with living Rickettsia, sensitized or neutralized with convalescent or immune serum; and (3) passive prophylaxis with such serum alone.

The ideal method at which first attempts should be aimed is active immunization with killed Rickettsia material. The first fact to be ascertained is whether or not such a thing is possible.

Active immunization of guinea pigs with carbolized Rickettsia was first accomplished by da Rocha-Lima, 1 confirmed by Weigl, 2 Breinl, 3 and Rosenberger. 4 Doerr and Schnabel 5 obtained negative results, possibly, as Otto, 6 suggests, because of deterioration of their vaccines. This is rendered likely by Kemp, 7 who found that our own early vaccine material lost potency within about 4 weeks. The above writers, as well as Otto, believe that the results obtained with louse vaccines, in contrast to the negative attempts made with virulent tissues—brain, blood, etc.,—are attributable to the high concentration of Rickettsia in the louse vaccines.

In our first vaccinations with formalinized tunica material of Mexican typhus with Batchelder 8 we were encouraged chiefly by the results of Spencer and Parker 9 and of Conner 10 with Rocky Mountain Spotted Fever, which favored the belief that active immunization with dead Rickettsia was feasible provided a sufficient concentration could be attained.

1. There is conclusive evidence that iotloacetic acid slows the rate of lactic acid production in the anaerobic recovery period, and the slowing is in proportion to the concentration of the poison. 2. Iodoacetic acid does not accelerate removal of lactate under anaerobic conditions.

It is well known that homoplastic implants of the anterior lobe of the pituitary induce sexual precocity in many animals. The present investigation continues the work of Wolf, 1 on frogs. The same technique of injection is used. Since, by giving daily implants of anterior lobe of the pituitary to mature frogs that had ovulated normally in April, it was possible to induce a second ovulation as early as October, it seemed of interest to determine how soon a second ovulation could be secured by this method, following a previous artificially induced one. By thus beginning the experiments with an artificially induced ovulation, the season of the year is apparently negligible, and the exact time of the first ovulation is accurately recorded.

The frogs were heavily parasitized and many died of “red-leg”. Fifty mature female leopard frogs (Rana pipiens), were given implants of 2 anterior lobes of the pituitary daily. Only 5 survived the entire experiment. After about the fifth implant, 44 were still living and had ovulated. The number of eggs discharged in every case exceeded four thousand. The second day after ovulation, an observational operation was performed in many to see if any mature ova remained undischarged. In no case were there more than 18 mature ova in the body cavity, ovaries, oviducts or uteri. (Postmortem examinations made on those that died of red-leg after ovulation confirmed these findings.) After 2 months, only 5 frogs were still living. Daily implants of anterior lobe were again given. After the third implant, one of the frogs died of red-leg.

Sure and Smith 1 , 2 have reported preliminary results on the influence of vitamin B deficiency on the concentration of lipids in the blood of the albino rat, indicating that lipemia is a symptom complex in this avitaminosis. Further detailed studies from this laboratory, 21 however, do not substantiate such findings. The reason that led us to subject to a critical analysis the lipemia in polyneuritic nursing young of the albino rat, 2 interpreted as due to vitamin B deficiency, is the fact that no appreciable changes were found in the concentration of either fatty acids, cholesterol, or phospholipids in weaned rats suffering from vitamin B (B₁) deficiency. Furthermore, an analysis of the data of Sure and Smith on lipid changes in the blood of weaned rats on diets deficient in the vitamin B complex, 3 disclosed too many great variations on the same animal from week to week, particularly in the fatty acid determination. It was discovered that carbon dioxide emanating from a close gas burner (which was being used some days during the week) was influencing the fatty acid titer considerably; therefore, all subsequent fatty acid determinations were henceforth carried out in a different room kept as free from the influence of carbon dioxide as possible, a precaution discovered in our laboratory (A. E. C.) before the same disturbing factor was called attention to in a recent paper by Himwich, Friedman, and Spiers. 4 From then on we ceased obtaining great variations in the weekly determinations of blood fatty acids. The work on the vitamin B complex was, consequently, repeated on adult rats depleted until total collapse, and even during the last stages of the avitaminosis, associated with marked inanition, no significant changes were detected in the concentration of either fatty acids or cholesterol of the blood.

Studies by Landsteiner and Van der Scheer 1 and by Landsteiner 2 have shown that 2 different species-hybrids may be distinguished from their parental species by the use of sera immunized against the respective blood cells, with subsequent agglutinin-absorptions. The experiments to be presented involve a cross of females of the domesticated Ring dove (Streptopelia risoria) with males of an Asiatic genus (Spilopelia chinensis), commonly called P'earlneck. Antisera were prepared by injecting rabbits with erythrocytes from individual representatives of each genus and of the hybrid. The agglutinations were performed by adding to 0.1 cc. of the immune serum in its varying dilutions (by halves) one drop of a 2.5% suspension of the red blood cells. For the absorptions, twice the volume of the serum diluted according to its original titre was added to a given volume of washed, packed red blood cells. The mixture was agitated gently at intervals, allowed to stand at room temperature for 2 hours and in the ice box overnight. All absorptions were repeated until complete at a dilution of 1:30. Readings were generally made after 2 hours at room temperature.

Without giving the inter-agglutinations of the parental genera and the hybrid, to be published elsewhere, it may be said that each was readily distinguishable from the others after reciprocal absorptions had been made.

Certain serological inter-relationships of the parental genera and the hybrid are presented in the table. On the basis of these reactions, it is seen that the cells of the hybrid resemble those of the Pearlneck genus more closely than those of the Ring dove.

Zondek, 1 Heidrich, Fels and Mathias 2 and Hady 3 have reported obtaining a positive Ascheim Zondek reaction in cases of chorionepithelioma testis.

Stimulated by these reports we have had the opportunity to study a case of testicular tumor admitted to the surgical ward of the University of Pennsylvania Hospital.

The patient, H. A., a white male, age 24, was admitted with a history of a firm mass the size of a golf ball in the right testis of two months duration. There was a slightly visible and definitely palpable intumescence and induration about 4 cm. in diameter under each areola of the breasts.

The right testicle and the distal end of the spermatic cord were removed at operation and a tumor 3 cm. in diameter involving the lower pole of the testis and infiltrating the seminiferous tubules was found.

Three months later, with no clinical evidence of the presence of metastasis of the tumor, a morning specimen of urine was obtained and 24 cc. of urine were injected intravenously into an isolated nonpregnant female rabbit according to the method described by Friedman and Lapham. 4 On sacrificing the rabbit several corpora lutea and corpora hemorrhagica were found in both ovaries resembling exactly the reaction obtained on injection of urine of pregnancy.

The pathological report by Dr. Joseph MacFarland on microscopic examination of the tissue after the test was teratoma of the testis.

In view of these results it is quite evident that this reaction may be of value in establishing a diagnosis of teratoma testis and also in the determination of the presence of metastatic tumor tissue of this character.

Extirpation experiments upon the embryonic forelimb of the rat demonstrating their inability to regenerate have been described. 1 In order to carry on this type of investigation with a more primitive animal the pouch young of the Virginia opossum, Didelphis virginiana, were subjected to a series of experiments to determine their capacity for regeneration.

The anesthetized female with new born young was stretched dorsicumbent on an operating table and the extremities fastened down. The pouch was held open by retraction and the pouch young 2 were operated upon with fine iridectomy scissors.

These preliminary operations were made in order to ascertain whether any particular locality was preferred for such treatment. Gross examination of these specimens after fixation in Bouin's fluid showed no signs of regeneration. In these experiments 1 to 5 digits or 1 foot from a posterior limb, or 1 posterior limb, or the tail were removed from a series of 16 animals. The specimens were killed and preserved at intervals from 33 to 315 days after the operation.

Unfortunately, due to the difficulties of maintaining the opossum in the laboratory, half of the experimental material was lost. Further work is planned using the young immediately at birth and, as before, confining the experiments to the posterior region of the body, which is less developed at the time of birth.

The work of Cori and others has definitely established the ability of epinephrine materially to increase the blood lactic acid and blood sugar. Cori and Buchwald 1 produced positive increments in blood lactic acid and sugar in man by continuous intravenous injections of epinephrine. Similar results were obtained in rabbits by subcutaneous injections.

Ephedrine has produced results similar to those given by epinephrine. Himwich, et al., 2 and Nitzescu and Munteanu 3 report work upon amytalized dogs in whom they obtained increases in blood lactic acid after ephedrine injections of 40-50 mg. per kilo. Wilson 4 found a hyperglycemia in dogs and rabbits after the subcutaneous injection of ephedrine in doses of 10-30 mg. per kilo.

The experiments herein reported were performed upon normal students. Ephedrine sulfate was ingested in aqueous solution in doses of 60-90 mg. Venous blood samples were drawn at intervals without stasis and the blood lactic acid determined by the method of Friedemann and Kendall. 5 Blood sugar was determined by Benedict's method. (Table I.)

In view of the calorigenic action of epinephrine and the similar effects of epinephrine and ephedrine upon blood lactic acid, it was decided to study the effect of ephedrine upon the respiratory metabolism. The Benedict-Universal metabolism machine was used in following the changes in oxygen consumption and CO₂ output after ephedrine ingestion. The metabolic rate was determined in the post-absorptive or “basal” state after at least 30 minutes rest in bed. Immediately after the basal run a blood sample was taken for lactic acid analysis. Ephedrine, in doses of 1 mg. per kilo was then ingested and after the effects had become pronounced (as indicated by heart rate) the metabolism was again determined. Positive increments of blood lactic acid and oxygen consumption were obtained.

This communication describes a method which has been developed to produce renal insufficiency by surgical procedure. Rabbits were used. The method is carried out by putting a loop of thread about one renal artery, when the animal is about 10 days old. This is done by placing a wire of 0.4 mm. diameter alongside the artery and tying a silk thread down upon the 2 together, and then withdrawing the wire. This leaves a loop about the artery of a diameter little larger than the artery. The artery soon grows up to this size and then any increase in the blood flow and size of that kidney is prevented. The other kidney overdevelops and so takes care of the increased renal demands of the animal. When the animal is 3 to 4 months of age, this large kidney is removed, leaving the animal with a small kidney, which cannot hypertrophy because of the tie about the artery. Too much time must not elapse before the second operation, lest atrophy of the small kidney supervene.

In this way any degree of kidney insufficiency can be produced. If the reduction in renal tissue be large (four-fifths) we find a rise in blood non protein nitrogen to 50-80 mg. per 100 cc. of blood, and a large volume of dilute urine is excreted. Greater reduction in kidney tissue results in higher blood non protein nitrogen (120-150 mg. N per 100 cc.), symptoms like uremia, low plasma chlorides, and death.

Clinical and experimental data have been accumulating indicating a functional relationship between the anterior lobe of the hypophysis and the pancreatic islets. (See especially Houssay et al.) Due to the importance of such a relationship we think it justifiable to note briefly some remarkable findings bearing on this matter.

Two litters of purebred dachshund, one consisting of 2 males, the other of 2 females were secured. One male and one female were injected daily intraperitoneal with the anterior hypophyseal growth hormone, 21 free of gonad stimulating hormone, for a period of about 8 months. Skeletal growth and body weight greatly exceeded that of the litter-mate controls, the weight of the injected animals soon being double that of their controls. Skin overgrowth and folding were present in both but was particularly prominent in the male. After 8 months of daily injection the male developed polydypsia, polyuria, polyphagia and became emaciated. The animal suffered from skin infection with loss of hair. He was inactive and evidently sick. Fehling's test of the urine was strongly positive. Only a trace of albumin was present in the urine. Fasting blood sugar was 232 mg. %. When injection was stopped for one week the animal improved, but rapidly failed again on resumption of injection. At present, 4 months after cessation of injection, the animal still has sugar in the urine. The test for albumin is now negative. The volume of urine has decreased and the animal has almost completely recovered.

The female dachshund has now been injected one year. Though responding markedly to the growth hormone she has never shown clinical or laboratory evidence of disturbance of carbohydrate metabolism. Urine and blood sugar have remained normal.

Following the observation of Castle and his coworkers 1 , 2 that normal human gastric juice activates beef to produce an antianemic substace, it seemed worthwhile to test bovine gastric juice for this property. The present report deals with a study that is similar to those of Castle but different in that bovine gastric juice was used instead of the stomach contents from human beings.

We obtained fresh gastric juice from the fourth stomach pouch of adult cattle. This material was filtered until a water-clear product was obtained and then incubated with beef at 37°C. for 2 hours. The incubated product was strained and portions administered by stomach tube to 4 patients having classical pernicious anemia.

The diagram shows a characteristic remission of pernicious anemia (case 1). The reticulocytes, red blood cells, and the hemoglobin all behaved in the manner described by Minot. 3

These experiments demostrate conclusively that the unpurified gastric juice taken from the fourth stomach pouch in cattle contains a potent anti-anemic factor. This work closely parallels the work of Castle and, in addition, makes it possible for us to continue studies without resorting to human sources for gastic juice.

1. The calcium contained in unrefined and refined dicalcium phosphate is utilized much more efficiently than are equivalent quantities of calcium supplied in the form of carbonate or citrate. 2. That this is not due solely to the fact that phosphorus is a limiting factor in these diets is demonstrated by the fact that the calcium of refined dicalcium phosphate is utilized about 70% more efficiently than is the calcium in tricalcium phosphate.

The substitution of refined dicalcium phosphate or unrefined dicalcium phosphate for calcium carbonate in the ricketsproducing diet resulted in the prevention and cure of rickets. Disregarding the effect of an external supply of phosphorus in the diet, these experiments indicate that from the standpoint of bone formation, refined dicalcium phosphate or unrefined dicalcium phosphate are infinitely superior to calcium carbonate in its various forms, such as precipitated chalk, crushed limestone, or crushed shells. Hence it is proper to refer to refined and unrefined dicalcium phosphate as being indicated in cases of rickets, or leg weakness of chicks.

From the standpoint of solubility in hydrochloric acid solutions of the concentration and at the temperature of gastric juice, unrefined dicalcium phosphate is greatly superior to bone meal. The calcium and phosphorus of unrefined dicalcium phosphate are also more available than these elements in the form of tricalcium phosphate. Refined dicalcium phosphate bears a similar relation to these salts as unrefined dicalcium phosphate does, although the solubility figures for the latter are slightly higher.

The subcutaneous injection of urine specimens or saline emulsions of stools obtained from patients ill with poliomyelitis brought about the death of guinea pigs in greater numbers than similar specimens obtained from normal individuals.

Of 201 guinea pigs injected with stool suspensions from infantile paralysis patients, 66% died and 95% had reactions ranging from a simple local induration to a generalized cystic condition of the abdomen.

Of 120 guinea pigs injected with stool suspensions from normal adults, 41% died and 60% had local or generalized skin reactions.

Of 42 guinea pigs injected with stool suspensions from normal infants, 3% died and 21% had local or generalized skin reactions.

Of 145 guinea pigs injected with suspensions of stools from patients ill with diseases other than infantile paralysis, 23% died and 50% had local or generalized skin reactions.

Of 214 guinea pigs injected subcutaneously with urines from infantile paralysis patients, 14% died and 61% had local or generalized skin reactions.

Of 51 guinea pigs injected with urines from normal patients, none died and but 5% showed local reactions.

Of 139 guinea pigs injected with urines of patients with diseases other than poliomyelitis, 7% died and 12% had local reactions.

Convalescent poliomyelitis serum previously injected intraperi-toneally into guinea pigs prolonged their lives, prevented massive local reactions or even protected them from death when later they were injected with stool emulsions or urine specimens from poliomyelitis patients.

That there was a specific toxic factor in the stools and urines of poliomyelitis patients was evidenced by the fact that serum taken from patients at the height of the disease (51 out of 87 cases) did not protect, while convalescent serum taken from the same patients some 21 days after the infection had started, protected guinea pigs which were later injected with stool suspensions or urine specimens obtained from poliomyelitis patients.

The colon bacillus was not the prime factor causing the death of guinea pigs after subcutaneous injections of standardized emulsions of stools from infantile paralysis patients. Nevertheless, this organism might have some part in the disease picture since convalescent serum protected coli injected guinea pigs to a slight extent, prolonging the lives of the animals, though not ultimately saving them. This slight protection might be due to an antibody in the serum, perhaps, an agglutinin.

Twenty-three coli-paratyphoid organisms, obtained partly from the American Type Culture Collection and partly isolated by us were used to determine the relative agglutinating capacity of serum obtained at the height of the disease and convalescent serum obtained 21 days after the onset of the disease. Both specimens were taken from the same patient.

Of 88 cases whose sera were thus tested, 55 showed marked increases in the agglutination titer of the convalescent sera as compared to the sera obtained in the acute stages. Thirty-one cases had sera of equal agglutination titer, while 2 cases had higher titers with the acute serum than with the convalescent serum.

In infantile paralysis, the superficial abdominal reflex is lost or modified in whole or in part, often before somatic paralysis appears. Theoretically then this loss could not be due to a spinal nerve involvement but might be due to sympathetic nerve paralysis. Since we have believed that the disease is essentially of gastrointestinal origin, the possible connection between the sympathetic system and the gut was given consideration. Experiments with autonomic drugs were performed to show this connection.

In typical cases of the spinal type of the disease, the patients were injected in the paralyzed and non-paralyzed areas with atropine, histamine, adrenalin, etc., without definite results. Salt solution was injected intradermally over the paralyzed areas, but the comparative data were useless.

When pilocarpine was injected subcutaneously, remote from the part paralyzed, the patients began to sweat first over the paralyzed area and only later over the corresponding normal side. A few minutes later the unparalyzed side stopped sweating, but the paralyzed side usually continued to be moist.

Since the sweat glands are solely innervated by the thoracolumbar outflow, their stimulation was then attempted by adrenalin. When adrenalin was given such a patient at the height of the sweating response to pilocarpine injections, the sweating ceased promptly over the unaffected area, but the patient continued to sweat over the paralyzed side, showing that the thoracolumbar sympathetic outflow of that particular segment was paralyzed. Since this reaction occurred often before somatic paralysis, it directed attention to the sympathetic system as the first part of the nervous system to be involved in this disease.

Blood examinations were made on a number of syphilitic patients 20 who were divided into 2 groups. One group consisted of 58 patients who had received no treatment up to the time of the blood examination. Thirty-two were seen in the stage of an active primary infection, 18 had active secondary lesions, and 8 had signs and symptoms of tertiary disease. The second group consisted of 55 patients who had received varying amounts of specific treatment, ranging from less than one full course of arsphenamine and mercury to several such courses. In 20 of these, treatment was instituted during the primary stage of the disease, in 16 treatment was begun during the secondary stage, and in 19 during the tertiary stage.

Since the patients were ambulatory and repeated counts could be obtained only with great difficulty, one complete blood examination was made on each individual. Each examination included a total red and white cell count made with standardized pipettes, a hemoglobin determination by the Newcomer method, a platelet count by the Ringer-heparin method of Casey, 1 , 2 and a differential white cell count made with the supravital technique, 100 cells being counted on each of 2 smears. The blood findings in these 2 groups were then compared, and the mean and standard error of the mean was determined for each blood element. A difference was considered to be significant when the probability of its occurrence by chance was less than one in 100.

In the treated group as compared with the untreated patients, the hemoglobin level and the absolute and relative numbers of lymphocytes were significantly higher, while the total white cells, the number of platelets, the absolute and relative numbers of neutrophils, and the absolute and relative numbers of monocytes were significantly lower.

It was shown by Kay 1 that the plasma phosphatase increases in certain clinically observed cases of bone diseases. We have confirmed and extended these observations. 2 In an investigation of experimentally produced bone lesions we have reported the effects of chronic and acute hyperparathyroidism. 3 In view of the many analogies observed in bone resorption, whether produced by parathormone or by other agents, it was desirable to extend our observations to the osteoporoses produced by low-calcium diets and by ammonium chloride administration to animals on low and high calcium diets.

Four litters of dogs were used—3, 6, 8, and 18 months old, respectively. The experiment was continued for about 11 weeks. The animals received a diet of fresh, lean horse meat supplemented with cod liver oil and tomato juice. This is a low-calcium diet. Some of the animals received a calcium supplement (2.5 gm. each of bone meal and calcium lactate per kilo of food) equivalent to between 0.5 and 2.5 gm. of calcium daily. On this diet the controls grew rapidly; the bones were normal upon autopsy. On the low-calcium diet, the growth was equally rapid; the bones, however, were thinned. Ammonium chloride was administered by stomach tube in a 1% solution. At the end of the experiment some dogs received daily as much as 1 gm. per kilo. In the youngest litter bone softening and deformity resulted on a low-calcium diet; in all dogs administration of ammonium chloride generally caused loss of appetite.

The method for determination of phosphatase has been described. 4 , 5 Serum phosphatase rather than plasma phosphatase has been determined in this series of tests to avoid the inhibition of phosphatase activity by oxalate. 5

Following the discovery of Minot and Murphy that raw liver is highly effective in the treatment of pernicious anemia, liver fractionation was undertaken by Cohn, Minot and their associates. The Cohn liver fraction G, in which approximately two-thirds to three-quarters of the potency of the original liver is recovered in 3 to 4% of its weight, has been generally employed in the oral treatment of pernicious anemia. Use of this preparation, involving the consumption of 10 to 14 gm. of material daily, entails some inconvenience to the patient. In attempting to isolate and identify the active principle Cohn, West and others have made very highly concentrated preparations but the procedures employed have not found practical application on account of their complexity and the losses of active principle entailed.

We here describe a method whereby the bulk of material administered orally to pernicious anemia cases to produce a given therapeutic effect may be substantially reduced, not by employing fractional methods as heretofore, but by so treating the raw liver or liver fractions (Cohn Fraction G) as to increase their therapeutic efficacy several fold.

These experiments were started immediately after Castle's demonstration that a thermolabile agent derived from the normal human stomach could interact with beef muscle to produce an active product which when administered to pernicious anemia cases induced remission with reticulocyte rise and increase in red cell count, while control pernicious anemia cases receiving normal human gastric juices alone and other control cases receiving beef muscle alone exhibited no such effects. Since fresh untreated muscle exerted no effect whatever, while a given weight of fresh muscle previously treated with normal stomach juice exerted a very substantial effect approximately equal to that exerted by an equal weight of fresh raw liver, it appeared important to determine whether a similar pre-treatment with normal stomach juices or stomach tissues would cause a corresponding rise in the therapeutic efficacy of the already active raw liver or liver fractions.

The accuracy of the determination of the blood group depends essentially upon the sterility of the typing serums and the freedom from formation of precipitates. The usual method of preserving stock typing serums is in a liquid form with the addition of 0.5% phenol. Lattes 1 has shown that serums preserved with phenol as well as many other preservatives, become turbid and form precipitates, and advises the use of sterile serums. The length of time that these aseptically collected serums will remain uncontaminated is variable and indefinite. Prati, 2 Grove and Crum 3 have demonstrated that liquid typing serums are very subject to contamination by a mustard bacillus and thus develop in the contaminated serums a non-specific property of human blood cell agglutination.

Desiccation may offer a desirable means of preservation. The value of desiccation is dependent upon its effect in time on the agglutinins in the serums. Slides were prepared as follows: Near the end of the slide was placed 0.01 cc. of type II serum, according to the Moss classification, containing 0.5% phenol and allowed to dry as a drop in the open air. Similarly near the other end of the slide 0.01 cc. of type III serum containing 0.5% phenol was placed. Each serum was marked below the drop to identify its group. The slides were divided into 2 lots, one lot being held at room temperature and the other at 8°C. Liquid phenolized typing serums were prepared at the same time for controls. These were tested at intervals.

Rosenthal 4 demonstrated that typing serums could be preserved in a liquid form by staining and that they required no particular aseptic precautions in handling and storing.

In a previous paper 1 it was demonstrated that a left ventricular arhythmia produced from insufflations of benzol in tracheotomized animals was contingent on impulses which descend the median part of the lateral columns of the spinal cord and reach the heart by way of the stellate ganglia. The purpose of this report is to record some observations which suggest that this arhythmia is not dependent on the nutrient supply of the heart. This arhythmia is normally preceded by a moderate rise in carotid pressure, a change which Markwalder and Starling, 2 Anrep and Segall, 3 Hochrein, Keller and Mauke 4 would place first, as a means for producing an increased coronary supply. The arhythmia never appeared when blood pressure was low (40 mm. and below). Double vagotomy in no way delayed the onset or shortened the interval of this arhythmia, except possibly for a short interval immediately after sectioning. Porter, 5 Mass, 6 Wiggers, 7 Anrep and Segall, 3 Greene 8 and others demonstrated that the constrictor fibers for the coronaries run in the vagi and Anrep and Segall 3 have demonstrated that double vagotomy causes a considerable increase in the coronary flow.

If a change in the nutrient supply to the nerve-intact heart is the cause of the arhythmia from insufflations, it should be due to an increase in the coronary flow rather than to a decrease. To demonstrate the effect on the pulse of a brief interval of anemia of the left ventricle, the left interventricular artery and the left ventricular vein were ligated close to the atrium in a number of animals for several minutes, and in a few animals the right interventricular artery was also ligated.

It has been known for a long time that the sera of children and monkeys, who have suffered from poliomyelitis, are capable of neutralizing the specific virus, although the frequency of this occurrence appears by no means to be as regular as was formerly believed 1 More recently a high percentage of normal adults, giving no history of a previous attack or of contact with the disease, have also been found to possess virucidal substances in their blood. 2 A number of authors, by indirect inference, have deduced from epidemiological criteria, that this capacity is acquired through exposure to subclinical infections. 3 In contrast to this viewpoint, we 4 have recently presented direct experimental evidence which strongly suggests that resistance to poliomyelitis as expressed by the fall of morbidity and the rising level of serological activity with increasing age, is predominantly a function of normal physiological maturation and to a large extent seems to develop independently of previous contact with the specific antigen. In further support of such a conception—or rather in opposition to the general validity of the exposure theory—the following observations are reported which bear on the presence of virucidal substances in the normal sera of certain susceptible and nonsusceptible animals.

Rhesus monkey. There is general agreement in the literature that the serum of immature monkeys is uniformly devoid of neutralizing power. We can confirm this point from a summary of 16 tests on a corresponding number of individual samples of serum obtained from normal rhesus monkeys ranging in probable age between 1 1/2 and 3 years. In no case did we observe any neutralization effect with any of the described sera when amounts of 0.6 to 0.8 cc. of serum were tested against 0.6 cc. or 0.4 cc. respectively of a 10% virus suspension (contact 1 1/2 hours at 37°C, overnight icebox).

We desire to record an observation in regard to the relation of the vagus nerves to appetite, an observation indicating that afferent sensations mediated by the vagi may be responsible for anorexia in abnormal gastro-intestinal states. Our reasons for recording an observation on one animal are as follows: (a) The animal preparation on which the observation was made is difficult to care for in that the feeding requires almost the full-time attendance of one individual and there is no likelihood of our being able to repeat the observation on other animals in the near future; and (b) the observation has a direct bearing on the interpretation of the results of those who are now investigating the subject of appetite, especially as related to Vitamin B deficiency, a problem in which we are not interested immediately.

In September, 1930, a dog was prepared 1 with a pouch of the entire stomach with vagi intact and a jejunal fistula for feeding. The pylorus was cut across, the duodenum closed and the pyloric opening brought to the outside, the jejunal fistula having been made previously. Nothing was allowed the dog by mouth for 3 months. At this time we tried to induce the dog to eat, but failed. The dog was on a diet supplemented with vitamines, including Vitamin B concentrate. On rare occasions we were able with much effort to induce the animal to eat an ounce or 2 of food. This continued throughout the period of one year, during which the body weight and nutrition were maintained solely by jejunal feeding. We thought that the appetite mechanism had been “reconditioned”, i. e., the dog, “satisfied” with the jejunal method of feeding, had lost the desire to eat by mouth, since the animal was doing well and a desire to take food by mouth could not be stimulated by administration of the various vitamins in large quantities.

Folin and his associates 1 demonstrated that uric acid injected intravenously into dogs in standard amounts disappears from the blood in a characteristic manner, i. e., 70% is destroyed within the first 10 minutes and in 2 hours is down to 2 mg. per 100 cc. of blood. They obtained the same results in Eck fistula dogs and concluded that the process occurs within the blood stream, possibly through some “essential factor” elaborated in the liver. Bollman, Mann and Magath 2 have found that extirpation of the liver in dogs produces an increase in the uric acid of the blood, tissues and urine. When uric acid was injected intravenously into hepatectomized dogs, disappearance was considerably slowed and the uric acid could almost quantitatively be recovered from the urine. It was their opinion that destruction of uric acid in the dog is dependent on the liver.

During studies on the function of the reticuloendothelial system we observed that when 8% india ink in normal saline was continuously injected intravenously into normal dogs, there was an appreciable increase in the uric acid of the blood. For example, after 6 hours, when 600 cc. of ink had been injected into a 14 kilo animal, the blood uric acid level was 3 mg. per 100 cc. This could be due to an increased production of uric acid or to a diminished destruction. Goldzieher and Sherman 3 observed an increase in urea following ink blockade in mice, rabbits and dogs. They interpreted this as due to an increased activity of the reticuloendothelial system. We therefore studied the rate of disappearance of intravenously injected uric acid in animals which had been previously blocked.

The uric acid was injected as in the procedure of Folin, Berglund and Derrick, 100 mg. per kilo body weight, viz., a 2% solution with 0.5% lithium carbonate.

Knowledge of the significance of histamine in biological processes would be materially increased if assays of higher specificity and sensitivity were available. In the absence of such tests it is necessary to look for an increased number of reactions both chemical as well as biological, indicative of histamine. Such reactions might aid by their variety rather than by their specificity.

Experiments with various smooth muscle contractants, forms of shock, etc., now in progress in this laboratory made it advisable to compare the reactivity of chicken and mouse intestine to histamine and acetyl choline. This ratio is well known to be 2:1 in the guinea pig. 1 This figure agrees on the average with our own observations. However, we have not infrequently dealt with individual variations such as guinea pig intestinal strips with a histamine-acetyl choline ratio of 5 :1 or 1:1. It should be remembered that the iliac end of the guinea pig intestine is generally conceded to be more reactive than the duodenal.

The action of histamine and acetyl choline upon the mouse and chicken intestine does not appear to have been studied previously. The reactivity of the mouse uterus to histamine has been shown to vary with the concentration. 2 , 3 , 4 Low concentrations (1:1,170,000) stimulate contraction, while high ones (1:1,250) inhibit. Generally speaking, the use of intestinal strips in preference to other smooth muscle seems advantageous in some instances, because a number of strips of the same intestine can be used and checks can be run conveniently.

In the present determinations the Schultz-Dale technique was used. The bases were added to the isolated intestinal strip suspended in 25 cc. oxygenated Tyrode's kept at 38-40°C. Dilutions neutralized to phenolphthalein were made in Tyrode's solution so that 1 cc. contained a unit amount of histamine di-hydrochloride 20 or acetyl choline hydrobromide. 21 Absence of hydrolysis of the acetyl choline was insured.

That the liver is practically indispensable for the development of anaphylactic shock in the dog has been indicated, principally by Manwaring and his colleagues, 1 Voegtlin and Bernheim, 2 and Simonds and Brandes. 3 Manwaring, 4 in addition, has reported that during anaphylactic shock in the dog, physiologically active substances, called by him “hepatic anaphylatoxins,” appear in the circulating blood so that by appropriate cross circulation experiments their presence can be demonstrated by the effects they produce in the recipient animal. A number of considerations led us to suspect that if transportable physiologically active substances are set free during anaphylactic shock in the dog, they might appear in the thoracic duct lymph. Petersen and Levinson 5 have noted the marked increase in the rate of flow of the thoracic duct lymph during shock. That this increase in flow is secondary to the shock phenomena occurring in the liver is indicated by the work of Simonds and Brandes, 6 who reproduced similar effects by mechanical obstruction of the hepatic veins. The points of similarity between the manifestations of anaphylactic shock and mechanical obstruction of the hepatic veins in the dog have been studied in considerable detail by Simonds and Brandes, and warrant the conclusion that a considerable proportion of the increased lymph flow occurring during shock is of hepatic origin. If this is the case, it would seem possible that substances such as the hepatic anaphylatoxins of Manwaring might appear in the lymph.

As one of the most delicate methods of testing for the presence of certain smooth muscle stimulating substances, the isolated surviving strip of guinea pig intestine was used.

Hypertrophy of the pancreatic islets in diabetes mellitus is usually explained as a compensation resulting from hypoinsulinism. In the thyroid gland experimental evidence has been obtained by Loeb 1 and his associates that excess hormone is inhibitory. Such observations lend strength to the idea that cell regeneration is a response to functional need.

By the method previously used by one of us 2 the mitotic activity of the islet cells was determined in serial sections of the chrome-formol fixed tissue. The results in detail are shown in the table. The injections were made at 1 or 2 day intervals with the last and largest dose about 18 hours before autopsy. Usually the pancreases from the 2 or 3 animals were mounted together and from 100 to 800 islets were enumerated. The controls were rats of the same age and they often were litter mates. The insulin treatment inhibited the proliferative activity of the islet cells although the growth of the animals continued normal and their weights were sometimes greater than those of the controls.

Previously we have shown 1 that fresh macerated liver injected in a prescribed way stimulated hepatic regeneration. We have now obtained results which show that by the injection of macerated kidney the mitotic rate of the tubular epithelium may be increased. The stimulant effect appears in the averages. The causes of the individual variations were usually undetermined. The compilation (Fig. 1) includes all treated animals that lived 14 days. Many negative results obtained by methods other than those described and not included in this report, are as yet unexplained.

With a few exceptions the experimental rats weighed from 100 to 150 gm. Kidneys from somewhat larger rats were crushed in a mortar for intraperitoneal injection. Seven normal rats were injected with 2.5 gm. crushed kidney and killed on the fourteenth day. Eight other normal animals were injected on the first and fifth day with a total of 4.5 gm. and killed on the fourteenth day. In a second series a unilateral nephrectomy was performed; 14 of these were killed after 2 weeks. Thirteen nephrectomized animals were injected either once on the day after the operation or on the second and fifth day with autopsy on the fourteenth day.

Transverse sections were taken through the middle of the kidney and the number of the mitoses per section determined. The influence of the intraperitoneal injection is shown in the figure. Seven normal rats with one injection had an average of 4 mitoses per section and 8 with 2 injections had an average of 7.6 as compared with 0.75 in 50 normal untreated animals.

While studying the effect of acute esophageal obstruction in the dog, 1 it was noted that the production of a closed loop of the esophagus was invariably fatal. Although post mortem evidence indicated that death was due to an aspiration pneumonia or to rupture of the loop with subsequent pleuritis and mediastinitis, the possibility of toxin absorption from the esophageal loop was considered. The following experiments were performed to ascertain the absorptive ability of the esophagus under various conditions.

Dogs were used. Under sodium pentobarbital anesthesia, a closed loop of the esophagus was prepared by ligating the esophagus in the neck and in the thorax just above the diaphragm. The substances to be tested were introduced into the lumen of the esophageal loop and the absorption estimated either by chemical tests of the urine and saliva, or by observation of the characteristic physiological effects.

Five experiments were performed in each group. Potassium iodide (in 10% solution), sodium ferrocyanide (in 10% solution), and sodium nitrite (in 5% solution) were all slowly but definitely absorbed under the conditions of these experiments. Histamine hydrochloride (in doses up to 3 mg. per kilo), strychnine sulphate (in doses up to 5 mg. per kilo), cocaine hydrochloride (in doses up to 15 mg. per kilo), ephedrine hydrochloride (in doses up to 10 mg. per kilo), nicotine (in doses up to 2 mg. per kilo), and epinephrine hydrochloride (in doses up to 10 mg. per kilo) were not sufficiently absorbed to produce characteristic effects. Phenolsulphonphthalein, methylene blue, and antipyrine were not sufficiently absorbed during the course of 2 hours to be detected in the urine. Distention of the esophageal loop with air pressure up to 60 mm. of mercury after introduction of the above drugs which were not normally absorbed, caused various reflex disturbances of respiration and blood pressure, but did not apparently facilitate absorption.

Various experimenters 1 have shown that young rats can be vaccinated per os against multiple lethal doses of virulent pneumococci. Ross has shown that the blood serum of such orally immunized animals contains antibodies which may be demonstrated by means of the passive immunity conferred upon other animals to which it is transferred. Besides this indication of a general humoral resistance to infection, we have seen no attempts at explanation of the mechanism of the immunity of such animals, and it was thought that studies of the leucocyte response, together with blood culture examinations, might throw some light upon the subject.

The method consisted of making periodic tests of the peripheral Wood of a series of normal rats and comparing the results with those of a series of rats previously immunized with egg-white and Type I pneumococcus autolysate, as already described. 2 Both sets of rats were inoculated intraperitoneally with 100 fatal doses of the homologous strain of pneumococci. Blood was obtained under sterile precautions from the tail of the animal; examinations were continued as long as the unvaccinated rats survived (2 to 4 days).

From work done on normal persons, it was expected that the leucocyte counts would be subject to wide fluctuations. This proved to be the case. One feature of the leucocyte counts, however, was constant; the immunized animals, after the first 6 hours following injection, showed an appreciably higher white cell count than the unvaccinated controls. This may be attributed to increased mobilization of leucocytes without any considerable diapedesis into the peritoneum. The lower counts in the unvaccinated animals probably reflect the loss of leucocytes which go to make up the exudate in the peritoneal cavity. This loss may go on until a leucopenia occurs, especially if the rats do not die until the third or fourth day.

During experiments upon the intestinal absorption of B. prodigiosus and various types of yeasts in rats, it was found that many of the plate cultures of the contents of different levels of the gastrointestinal tract were sterile. To determine whether these plates were actually sterile or whether forms of microbes were present in a phase which-did not grow appreciably, we subjected such apparently sterile cultures to the serial plating technic described by Hauduroy. 1 About 0.5 cc. of sterile nutrient broth is taken up in a capillary pipette, the slender part of which is about 3 inches long. After discharging all the broth onto the surface of the plate from which transfer is to be made, the pipette is held so that the side of the Bunsen flame just touches it about 2 inches from the tip. Softening of the glass and the weight of the tip operate to bend the pipette into a spreader which is used to scrape the discharged broth back and forth across the surface of the plate, thereby washing off whatever microbic elements may be present. The same pipette is used to draw up the washings, to discharge them upon the fresh plate, and to spread them over its surface. We have found this technic superior to that of using 2 pipettes and 2 spreaders for this simple operation. Practically all of the apparently sterile plates selected for this study were derived from the stomach or duodenum of the animal; an occasional plate made from the first part of the jejunum was also sterile. These plates were all observed for at least 48 hours (many of them for 60-72 hours) and did not show detectable growth under a hand lens or low power of the microscope.

In a series of dogs, segments of the small intestine 5 to 10 cm. in length were isolated from the remainder of the intestinal tract and transplanted onto the surface of the abdomen. These segments were taken from various portions of the bowel between the upper duodenum and terminal ileum. Continuity of the remainder of the intestinal tract was re-established. When the animals had completely recovered from the operative trauma electrograms were made of the isolated segments. Two electrodes were placed at varying distances apart on the longitudinal axis of the bowel segment and were connected with a string galvanometer. The electrograms obtained were constant in their rate for given segments and reveal some rather characteristic phases. The electrograms persisted after the segment had been completely separated from the central nervous system. They also persisted, although with slight changes in certain phases, whether the segment was visibly active or quiet. This has been discussed more fully elsewhere. 1

Four dogs have been studied to determine the relationship of respiration and of mechanical activity of the bowel segment to the electrogram. Platinum wire electrodes covered with cotton and saturated with normal saline solution were used. The string of the galvanometer was relatively loose, deflecting 2 cm. per millivolt. A pneumograph was so arranged as to record respiratory movements simultaneously with electrograms on the same film. These showed that there was no relationship between respiration and the electrical complexes obtained from the isolated bowel. (Fig. 1.) Balloons were then inserted into the isolated bowel segments and connected to record contractions of the bowel on the same film with the electrogram. A very small balloon was used and the electrodes were placed on the surface of the bowel directly over it.

Removal of the stomach necessitates anastomosis of esophagus to duodenum; this is made difficult in the dog by the shortness of the abdominal part of the esophagus. In the rodent this part is much longer. It therefore occurred to one of us that such an operation might be feasible in the rat, provided that some other technique than sewing might be used in making the anastomosis. We had on hand a number of small cannulae made of magnesium and intended for blood-vessel surgery. According to Lexer 1 these were introduced into surgery by Payr. The technique we have developed consists essentially of tying the cut end of the esophagus over one end of the cannula and the end of the duodenum over the other. This unsurgical method could not be expected to work in larger animals, but in the rat it gives very promising results.

We have thus been able to perform gastrectomies on 31 rats, and at present have 8 living. These have survived their operations by 8, 8, 8, 8, 21, 21, 25, 43, and 51 days respectively. Of the 23 that have died, the survival times were 1 (9), 2 (3), 3 (2), 4 (1), 5 (1), 6 (3), 7, 8, 9, 13, and 20 days respectively. In all rats now living the filling of the intestine by way of the anastomosis during the ingestion of an opaque meal has been observed by fluoroscopy. In the accompanying roentgenogram is shown the anastomosis in the rat which is now in its 43rd day; this rat has had acute esophageal obstruction 4 times, but is now able to eat anything it cares for. Four of these rats have regained their pre-operative weights.

White rats show esceetlingly great variability in potentiality for the protluctiotl of oocyits during an infection with Eiweria miyuivii. If a similar situation exists among other species of either mammals or birds, the implications are far-reaching and important, for they may serve to explain many of the cases where one worker claims to have controlled or cured coccidiosis with a particular food or remedy, while another secures contrary results with the same methods.

In the investigation of Ch'an Su, a commercial preparation of the venom of a Chinese toad, and the poisonous secretions of 10 additional species of toads collected from different parts of the world, the authors found variations in the potency, physical constants, and chemical composition of some of the principles that were isolated from them. The secretion of each species of the toads contains at least from 3 to 5 distinct principles that belong to 5 classes of compounds: cholesterol, epinephrine, bufagin, bufotoxin, and bufotenine.

Cholesterol occurs in all of the secretions. Spectrographs examinations of several different samples showed the presence of ergosterol in various amounts. It has already been proved that after irradiation the sterol from Ch'an Su has an antirachitic action in rats. 1

Epinephrine occurs in B. marinus 2 Ch'an Su, 3 and appears to be present in B. arenarum, B. regularis, B. formosus, and B. bufo gargarisans. The amount varies from 2 to 5% of the dried secretion.

Bufagin is a word used generically, since it was observed that most of the bufagins, although similar in their action and elementary composition, are quantitatively different from each other as shown by their minimal emetic dose in cats and pigeons, respectively, and also by the average fatal do^e in cats (Table I). Besides, their melting points are not the same throughout. The bufagins possess the essential features of the members of the digitalis group. The action of the bufagins is not persistent, for the cat can eliminate completely a minimal emetic dose within one to few hours. The cat unit of each principle is less than 1 mg.

Wieland and Alles 1 isolated bufotoxin from the skin of B. vulgaris or B. bufo bufo. In our study of Ch'an Su and the secretions of 10 additional species of toads, we obtained a buf otoxin from the dried venom of each species. The word buf otoxin is used generically for the same reason as given for bufagins. Chemically, the bufotoxins are formed by the conjugation of one molecule of suberyl-arginine with one molecule of a specific buf agin found in the respective secretions. Their pharmacological action is similar to that of bufagins, being different only in degree (Table I), as shown by the systolic standstill in the myocardiogram of frogs, typical changes in the electrocardiograms of cats, and emesis in cats and pigeons. Locally, they are bitter to the taste.

Handovsky 1 isolated an “alkaloid” from B. vulgaris, and named it bufotenine. In our investigation of Ch'an Su, we 2 obtained cinobu-fotenine, and found it to have a marked pressor action. It appears now that the secretions of most toads contain bufotenines—a name also used generically. The bufotenines are organic bases having an indole ring, and form organic salts such as the flavianates, which have been used in this investigation. Three bufotenine flavianates, prepared from Ch'an Su and the secretions of B. fozvleri and B. bufo gargarizans, respectively, have a marked pressor action in pithed cats (Table I); those separated from the secretions of B. formosus, B. bufo bufo, and B. viridis viridis are slightly less active; and the remaining bufotenines have little blood pressure raising property. All of them have an oxytocic action, and with one exception, a stimulating action on isolated intestines.

The usual dilution methods for determining the precipitin titer of antiserum only roughly approximate the relative antibody strength of a given serum and give no information with regard to the absolute quantity of antibody. Analysis of precipitates from the precipitinogen-precipitin reaction for their nitrogen content by the micro-Kjeldahl method has been used by Wu and his collaborators 1 and Heidelberger and Kendall. 2 Precise measurements of the antibody per unit volume of antiserum, and exact studies on the ratio of antigen to antibody in the precipitate have thus been made possible. This method requires at least 0.5 cc. of serum for each test and cannot be used when only small volumes of antiserum are available.

We have found that the following procedure gives satisfactory results for determining the quantity of precipitin in a small volume of rabbit anti-crystalline-egg-albumin serum. The method was suggested by figures presented by Heidelberger and Kendall. It depends on the determination of the amount of antigen necessary to precipitate the precipitins completely from a given volume of an antiserum. This determination of the mg. of antigen necessary completely to precipitate the antibodies from a sample of such serum is easily accomplished without analytical methods because the point of maximum precipitation proves to be that point where antigen as well as antibody is completely precipitated and neither antigen nor antibody remains in the supernatant fluid. We have called this optimum ratio of antigen to antibody the neutralization point. Furthermore, at this optimum point antigen and antibody combine in a ratio of about 1;13. The description of the method will clarify these points.

Method. The stock solution of crystalline egg albumin was diluted with saline to provide known concentrations of the antigen expressed in mg. of nitrogen per cc.