Search

Conclusions1. Streptomycin appears to have no adverse effect on the 10- to 11-day chick embryo when injected into the allantoic sac.2. Streptomycin is demonstrable in bac-teriostatic or bactericidal concentrations in the allantoic fluid 48 hours after a single injection of 2,500 units into the allantoic sac. Only small amounts appear in the amniotic fluid.3. Streptomycin in the concentrations used appeared to have no adverse effect on the growth of either Influenza A or B virus in the allantoic sac.4. Streptomycin is highly effective in preventing gross contamination of allantoic fluids of chick embryos injected with un-filtered throat washings and suspensions of human stool.5. Influenza A virus, when highly diluted and mixed with unfiltered throat washings or stool suspensions containing added streptomycin, can be readily recovered after growth in the allantoic sac of the developing chick embryo.6. The results indicate that the loss of virus which may follow removal of contaminants by filtration may be a...

Observations in six horses with heaves established a clear relationship between attacks of heaves and the feeding of hay. Severe acute attacks were accompanied by striking changes in the eosinophil count and the sedimentation rate. The variation in the severity of heaves in relation to the feeding and withholding of hay is accounted for by assuming that attacks result from a transient obstructive lesion in the bronchial tree or lung caused by hypersensitivity to some component of hay. The observations strongly suggest that heaves is a respiratory allergic disease in the horse. The relation of heaves to pulmonary emphysema in the horse is discussed.

Ten asymptomatic patients with normal pulmonary function were selected for a double-blind trial of immunotherapy in cat-induced asthma. Each patient had a positive prick test to cat pelt extract and also a positive bronchial challenge response to the same extract. Patients were randomly assigned to active treatment or placebo groups and received weekly or biweekly injections over a 3 to 4-month period. The 5 patients who received the active treatment received a cumulative dose of cat pelt extract that ranged from 16.4 to 44.8 mg of total solid containing 1.7 to 4.7 mg of cat allergen 1. Apparent systemic reactions were observed in 3 patients who received the placebo and 3 patients who received the active treatment. The 5 patients who received the active treatment showed a reduction in skin reactivity to cat pelt extract as well as a significant mean reduction in bronchial sensitivity to the same extract. The 5 patients who received the placebo showed no significant changes in skin reactivity or bronchial sensitivity to cat pelt extract. Bronchial response to histamine did not change significantly in either the active treatment or the placebo groups.

A purified fraction of cat pelt extract was prepared and subjected to electrophoresis in 7% acrylamide gel. The bulk of the allergenic activity was recovered from the prealbumin portion of the gel that contained a single band of protein (band B). Antiserum to band B reacted with both band B and albumin; after absorption with albumin, only antibody against band B could be detected. Electrophoresis of band B protein in 10% acrylamide gel resolved four protein bands; one of these bands was associated with the bulk of allergenic activity and reacted with absorbed antiserum to band B. The absorbed antiserum (anti-allergen antiserum) reacted in gel diffusion with a single component of crude cat pelt extract; this component had the molecular size characteristics of allergen. Anti-allergen antiserum removed significant amounts of allergenic activity from both crude cat pelt extract and purified fractions of cat pelt extract. It is proposed that the allergen defined by this antiserum be designated cat allergen 1. The radioallergosorbent test was modified so as to provide an assay for allergen. The modified test provided a useful supplement to the skin test as an assay for allergen and showed that, of various cat tissue extracts, pelt extract was the most abundant source of allergenic activity. Extract of cat brain also contained significant amounts of allergen.

Aqueous extracts prepared from lyophilized, defatted rat, mouse, guinea pig, and rabbit pelts elicited intense wheal-and-flare responses in the skin of a high proportion of patients who were clinically sensitive to these animals. The major allergens in each extract were nondialyzable. Skin test reactions to rat, mouse, and guinea pig serum were common in patients allergic to these animals. The fractions of rat, mouse, and rabbit pelt extract showing maximum allergenic activity contained proteins with the electrophoretic mobility of serum albumin. Fractions of guinea pig pelt extract with maximum allergenic activity were of prealbumin mobility and contained little stainable protein. On Sephadex G-100 gel filtration, most allergen from rat, mouse, and guinea pig pelt extracts was recovered in fractions containing proteins with a molecular weight range of 10,000 to 25,000 daltons. Allergen in rabbit pelt extract had a slightly higher molecular weight range of 18,000 to 38,000 daltons.

Summary and Conclusions A series of experiments is presented in which type III pneumococci were grown in various ways in artificial media, both with and without sulfapyridine, in order to determine whether the changes which take place in the medium or in the organisms during the course of growth have any effect on subsequent growth in such media. When sulfapyridine was added during the various phases of active growth from a small inoculum, there followed, in turn, a period in which multiplication continued, a bacteriostatic and then a decline phase. The latter effects were limited only by the population of the culture at the time the drug was added, and the amount of multiplication that preceded was fairly constant within that limit. The growth of pneumococci in fresh media was dependent on the conditions under which they had been grown previously. Organisms in the static or in the decline phase of sulfapyridine action continued to die when transferred to fresh drug-containing media, as if the media had not been changed. When transferred to drug-free media they grew like a fresh inoculum of similar size that had previously grown without drug. Within the limits of the experiment, therefore, sulfapyridine action is reversible. The action of sulfapyridine effects a change in the organism and not in the medium. This change was shown to take place even in the presence of a large inoculum when growth inhibition was not apparent, and it became manifest when small numbers of the organisms were transferred to fresh drug-containing media. Pneumococci were grown repeatedly from fresh small inocula in sulfapyridine broth in which all the phases of growth, bacteriostasis and bactericidal action had previously been carried to completion. All the resulting growth curves were essentially the same. The inhibition or killing of pneumococci was not accompanied by any stable change in the media capable of modifying the growth of fresh organism subsequently inoculated. The population of a culture that had attained the static phase of sulfapyridine action was doubled by the addition of fresh organisms that had grown without sulfapyridine. The resulting population-curve of that culture from that point on was the same as that of a control culture of the new inoculum in fresh sulfapyridine broth. This suggests that no unstable changes are produced during the growth of pneumococci in sulfapyridine-containing media that are capable of modifying the growth of freshly inoculated organisms.

Summary Washed specific precipitates were prepared from goat antiserum to rabbit γ-globulin and rabbit antibody to a series of protein antigens. The ability of these aggregates to bind antigen specifically was studied with I131-labeled materials. It was found that aggregates prepared with rabbit antibody to ribonuclease, hen egg albumin, bovine serum albumin and human γ-globulin could, with sufficiently long incubation, take up approximately 80% of theoretic amounts of antigen. Aggregates prepared with rabbit antibody to hog thyroglobulin and busycon hemocyanin bound only small amounts of antigen. The results suggest that the dimension of the lattice formed by goat antibody and rabbit γ-globulin permit antigens up to the size of human γ-globulin to enter and react with specific rabbit antibody sites. The larger antigens are almost totally excluded and react probably only at the surface of the aggregate. Each molecule of rabbit antibody is positioned in the aggregate so as to be capable of binding close to two molecules of antigen under optimal conditions.

INJECTIONS of allergenic extracts as a means of mitigating the effects of seasonal exposure to pollens, to be referred to as injection therapy, has been practiced for over fifty years. The number of patients in this country receiving injection therapy is unknown, but it must be very large — probably of the order of 1,000,000. The mere magnitude of this medical enterprise might suffice as evidence for its efficacy except that objective criteria for success or failure are lacking and that factors other than injection treatment may profoundly influence the course of pollenosis and the opinion of both patient and . . .

Spirographic tracings obtained during the performance of vital capacities are discussed from the point of view of a number of artifacts and the possible significance of the manner in which the speed of expiration changes. The rate of this change may be determined by plotting time on the abscissa and the logarithm of the volume remaining to be expired along the ordinate. Analysis of the curves obtained indicates: 1.1. During forced expiration air leaves the normal lung with a speed which changes (i.e., slows) at a constant or almost constant rate. 2.2. The rate of change in the speed of expiration may vary during a single expiratory effort among patients with chronic lung disease depending on poorly understood intrapulmonary factors. 3.3. In some patients with pulmonary disease, no single measurement or index derived from the so-called timed vital capacity can be regarded as truly representing the manner in which air is expired from the lungs. However, this is readily recognized with a kymographic tracing made with the paper moving at the rate of 0.5 cm. per second or more.

Disodium cromoglycate, a chromone derivative, has been shown by others to prevent or diminish the asthmatic attack induced by allergenic exposure and to diminish or relieve the symptoms of asthma in a number of controlled clinical trials. A double-blind crossover study with 27 trials was carried out: In 17 trials the drug was identified as helpful, in 4 the placebo was regarded as helpful, and there was no opinion in 6. This is strong evidence that disodium cromoglycate is effective in relieving symptoms in asthma (p =

The accumulation of eosinophils 18 to 24 hours after the application of allergenic extract to a superficially denuded area of skin (skin window technique) was observed in 20 patients, and in 10 the results were related to the intensity of hypersensitivity as judged by the scratch test and by end-point titrations in the skin. All three techniques gave a fairly satisfactory correlation with the occurrence and severity of symptoms on exposure to the relevant allergen. However, the numbers of eosinophils seen correlated better with symptoms than did either the scratch test or end-point titrations. Experience indicated that, with the technique used in this study, concentrations of extract in the range 1:1,000 to 1:10,000 weight by volume give satisfactory preparations without risk of systemic allergic reactions. Basophils were seen in many preparations, and in some their numbers exceeded the number of eosinophils. However, the number of these cells was less clearly related to skin reactivity and symptoms and appeared to be relatively independent of the concentration of extract used. Their role in the allergic response seems even more enigmatic than that of the eosinophil.

0 BSTRUCTIVE pulmonary disease is one of the most important causes of dyspnea, and is a common reason for patients to seek medical attention. Since mild obstructive disease often goes unrecognized, there is need for a means of detect,ing its presence and grading its severity. When severe, it is often confused with congestive heart failure, and the common coexistence of the two imposes the urgent need for a ready means of assessing the pulmonary component of the patient’s disease. Although obstructive pulmonary disease produces a great many characteristic aberrations in function, from the clinical standpoint, the most practicable tests are those which measure the maximal expiratory rate. A kymographic tracing of a maximal forced expiration (expirogram)l serves this purpose well and we believe that, among the procedures suitable for day t,o day use in the practice of medicine, this is the best available measure of airway obstruction. Expirograms in severe emphysema or asthma present an abnormality so gross as to be recognizable at a glance. However, there are many instances which differ from the normal in minor respects and their evaluation requires carefully selected standards to distinguish normal variation from disease. Several parameters, all derivable from the expirogram, have been used as measures of the abnormality in obstructive disease. Some of these are shown in Fig. 1. The volume expired in the first second (F’SV)2 and the rate of expiration between the points where 200 CC. and 1,100 cc. have been cxpired” have three disadvantages. First, they deal with the early phase of the expiratory effort at which time the bronchial lumen is most patent and the effect of an obstructing lesion is minimal. Second, patients often fail to exhale as rapidly as possible during early expiration either because of pool motivation or technique. Finally, although of doubtful clinical importance, the inertia of most spirometers makes this portion of the curve differ somewhat from the true maximal expiratory flow rate. Examination of many expiratory tracings from patients with obstructive disease and apparent,ly

ConclusionsEvidence is presented indicating the presence of two antibodies for crystalline insulin in the serum of a patient who was both allergic and resistant to crystalline insulin. It was also shown that the patient was allergic but not resistant to a preparation of human insulin. The relation of the two antibodies to the patient's response to crystalline and human insulin is discussed.

1.1. Changes in the vital capacity were studied following the injection of acetyl-beta-methyl choline and histamine in forty asthmatic subjects. 2.2. Analysis of the results with respect to the pulmonary responsiveness to the two drugs on the one hand and various clinical features on the other indicate that: (a) patients with severe asthma were most responsive to the two drugs and those with mild asthma were least responsive; (b) patients with asthma of very recent origin or those whose asthma had ceased or markedly diminished were either slightly responsive or not responsive at all; (c) age, sex, the presence of hay fever, eczema, or urticaria in the subject or the subject's family appeared to have no influence on the susceptibility to the two drugs. It is concluded that of all the factors considered, asthma was the only one which appeared to have a direct influence on the pulmonary responsiveness to histamine and acetyl-beta-methyl choline.

Certain hemolytic disorders are characterized by the attachment of a globulin to the erythrocyte membrane, this globulin possessing the characteristics of an antibody for antigenic determinants associated with the surface of the red cells (1). Cells with attached globulin, i.e., sensitized cells, may agglutinate spontaneously when placed in serum, albumin solutions, or other protein-containing mediums, but usually do not agglutinate when suspended in saline. Erythrocytes, so sensitized, are encountered in the blood of some patients with idiopathic acquired hemolytic anemia and erythroblastosis fetalis (2–4). Similar sensitization may follow exposure of normal red cells in vitro to nonagglutinating antibody, occasionally present in the serum of those “iso-immunized” by pregnancy or transfusion (5–7). The adherent globulin may be demonstrated most readily by the so-called “Coombs test,” the basis of which is the agglutination of these cells by an antiserum prepared in rabbits injected with human serum.

It is concluded that the total eosinophil count of healthy nonatopic individuals is lower and exhibits a significantly narrower range than the published data for "normal" values would indicate. The data also suggest that skin reactivity to extracts of common airborne allergens might serve as a basis for reassessment of the genetic transmission of atopy.

A controlled study of 17-ketosteroid and 11-oxysteroid excretion in ambulatory patients with bronchial asthma has been carried out during intervals when the patients had been without steroid medication for at least two weeks. No significant correlation was noted between steroid excretion and the asthmatic state as verified by serial timed vital capacities and daily score sheets. No significant correlation was noted between 17-ketosteroid excretion and 11-oxysteroid excretion. Total 11-oxysteroid excretion was not abnormal in bronchial asthma. Significant reduction of total 17-ketosteroids was found in both male and female asthmatic patients, as compared to age- and sex-matched controls, but not in patients with hay fever. Fractional 17-ketosteroid excretion as measured by androsterone-glucuronide and etiocholanolone-glucuronide analyses tended to confirm total 17-ketosteroid excretion. The administration of cortisone resulted in a further decrease of androsterone-glucuronide and etiocholanolone-glucuronide excretion for the duration of therapy.

Summary The collodion-particle technic described by Cannon and Marshall was compared with the gelatin-interface test and the complement-fixation test. The comparisons were made using rabbit antisera for ovalbumin, ragweed-pollen extract and horse-serum. The collodion-particle technic consistently permitted the greatest dilution of serum without loss of the ability to give an easily visible antigen-antibody reaction. The titers obtained with the collodion-particle technic paralleled those obtained with the gelatin-interface and the complement-fixation tests.

Experience with an apparatus permitting continuous recording of the respiration is described. The value of simultaneously observing the oxygen saturation with an oximeter is emphasized. The changes which accompany induced asthmatic attacks in subjects breathing ambient air in a closed system included (1) a temporary decrease in tidal volume and (2) an increase in the functional residual capacity. These changes were associated in some instances, with a demonstrable decrease in the efficiency of the lung. Relief of an induced asthmatic attack with aerosolized Isuprel caused an increase in ventilation and a return of the functional residual capacity toward the initial level. Anoxia accompanied severe induced attacks of asthma. The system described was readily adapted to the detection of the decrease in ventilation which follows the inhalation of oxygen by patients with respiratory acidosis.

I N THIS article the allergenic activity of a fraction of ragweed pollen is shown to be removable by means of a precipitate made by mixing rabbit anti-humangamma-globulin serum and serum from a patient or subject who had received a series of injections of ragweed pollen extract. The procedure depends on the capacity of antibody to combine with antigen when the antibody itself is present as an insoluble immunologic precipitate. It has been shown by others t,hat rabbit antiequine serum will precipitate soluble equine tetanal antitoxin and that rabbit antiequine serum does not bring about dissociation of diphtheric toxinantitoxin complexes,l, 2 nor does precipitated antit,oxin lose its antitoxic activity.” More recently it was shown that a precipitate made with rabbit antiequine serum and equine ant,iovalbumin serum would combine with ovalbumin4 but that in similar experiments equine antihemocyanin serum, presumably because of stearic hindrance, failed to combine with hemocyanin.”

Summary The influence of various salt concentrations in the presence of 5 per cent glucose, on agglutination of red cells by influenza A virus (PR8) was studied. No visible agglutination occurred in concentrations of 0.030M NaCl or less and at a concentration of approximately 0.003M, union of virus and red cells was inhibited. Virus was eluted when combination of virus and red cells was permitted to take place and the electrolyte was then removed by washing in a salt-free solution. Concentrations of salt as high as 0.583M failed to cause noticeable inhibition of agglutination of red cells by virus although such salt-concentrations completely inhibited agglutination of red cells by antiserum. For comparative purposes, agglutination of red cells by antibody was likewise studied with respect to the effect of various salt-concentrations. The results were similar to those which have been obtained by others and differed from those obtained with influenzal virus in that interference with agglutination was much less marked with low and much more marked with high salt-concentrations.

The rate at which the blood of asthmatic subjects, purposely rendered hypoxic, became saturated upon breathing 100 per cent oxygen was determined using the Millikan oximeter. A continuous record of respiration was obtained by having the subjects breathe in a closed system with spirometer and kymograph. The findings indicate that information concerning the functional state of the lungs may be obtained by this means. The rate of saturation was readily modified by varying the degree of asthma with epinephrine injected subcutaneously or with aerosolized histamine.

A comparison of the immunologic response of allergic and normal individuals to dextran and SVII polysaccharide revealed no differences between the groups in respect to precipitin antibody or skin reactivity produced to these antigens. Both groups gave similar skin reactions when tested with histamine, serotonin, and Compound 4880.

The increasing importance of aerosols in industry, in the treatment of pulmonary conditions and more recently in the diagnosis and experimental study of bronchial asthma has intensified the need for a method permitting measurement of the amount of aerosol retained in the human lung. This problem has been approached in various ways. Landahl and Hermann (1) measured the difference in concentration of aerosol in the inspired and expired air. They showed that the relative amount of aerosol removed is markedly influenced by the size and physical properties of the particles and also by the rate and depth of respiration. Using aerosols containing radioactive sodium, Talbot, Quimby, and Barach (2) estimated the quantity of aerosol retained by using a Geiger counter over the axilla and the foot. Wilson and LaMer (3) extended this method to include the use of aerosol droplets of uniform size. They determined the total amount of radioactive sodium retained by the subject by measuring the difference in the amount delivered to the subject and the amount in the expired air. The quantity absorbed was estimated from the concentration of radioactive sodium in the blood. The use of blood concentrations as an index of the amount absorbed from the lungs requires assumptions regarding the distribution of the substance in the body. By assuming that the concentration of radioactive sodium in the interstitial fluid was 25 per cent of that found in the blood 15 minutes after exposure, Wilson and LaMer estimated that between 31 and 63 per cent of an inhaled aerosol composed of particles 0.7 microns in diameter was retained in the alveoli. In these experiments, 71 to 80 per cent of the administered aerosol was removed by the subject, the difference presumably being trapped in the upper respiratory tract. More direct methods for measuring the amount of material retained in the lung have not been described. In the experiments to be described, we determined the pulmonary retention of inhaled aerosols by measuring the difference between the amount in the inspired and expired air and also by measuring the amount absorbed. By using two methods simultaneously, it was hoped that possible errors in measurement would be more easily detected and the validity of both methods could be determined.

An aqueous extract prepared from lyophylized, defatted cat pelts elicited intense wheal-and-flare responses in the skin of patients exhibiting asthma or rhinitis on exposure to cats. On electrophoresis of the extract in agarose, protein bands with the mobility of albumin, α 2 -, β-, and γ 1 -protein were observed. Allergenic activity was localized in two distinct regions with albumin and α 2 -mobility, respectively. Allergenic activity remained bound to DE52-cellulose columns equilibrated with phosphatebuffered (0.01M, pH 7.4) saline (0.05M), but could be eluted by increasing the salt concentration to 0.25M. Following gel filtration on Sephadex G-100, allergen was recovered in fractions containing proteins with a molecular weight range of 30,000 to 60,000 daltons. A purified fraction of cat pelt extract obtained by sequential fractionation on DE52-cellulose and Sephadex G-100 columns was subjected to polyacrylamide disc gel electrophoresis; 5 protein bands were observed. Allergenic activity was recovered in fractions containing two of these bands. Five of twelve subjects with strong skin reactivity to cat pelt extract also reacted strongly to cat serum; the major allergen in cat serum was found to have the electrophoretic mobility of α 2 -protein.

Several workers have shown that the application of the appropriate allergenic extract to superficially denuded skin sites in patients with pollinosis is followed by the appearance of eosinophils in the exudate. Studies were done to determine the time course of migration of eosinophils to such sites. The influx of eosinophils was found to be most intense at about 12 hours, on the average, after application of the allergen and then to decrease significantly 24 hours after application. Since the administration of corticosteroids in the treatment of allergic and other conditions causes a rapid reduction in the number of eosinophils in the blood, studies were also done to determine whether cortisol applied locally would influence the accumulation of eosinophils at a skin site. The results indicate that cortisol has a pronounced inhibitory effect which appeared to be specific in that the inhibition could not be explained in its entirety as a mere consequence of an anti-inflammatory effect of the steroid.

Atopic and nonatopic subjects were exposed intranasally to a protein antigen, and observations were made on the development of skin reactivity of the immediate type, the occurrence of symptoms, and the appearance of skin-sensitizing capacity in the serum. These were induced exclusively or with greater frequency in the atopic group, and this was not attributable to antigen administered in the course of skin testing. Four among 10 nonatopic individuals developed delayed-type skin reactivity which was not observed among the atopic subjects. The ABO blood group secretor status was observed with equal frequency in allergic and nonallergic individuals and was unrelated to appearance of skin reactivity. It is suggested that atopic individuals differ from nonatopic in having a greater mucosal permeability to inhaled antigens.

TN an earlier report1 dealing with orally administered bronchodilator preparations, it was shown that information equivalent to that available by the conventional double-bind technic could be gained by the use of medications coded by color and dispensed in a single bottle. Coding by color appeared to have the advantage of simplicity and ease of administration. To validate this technic and to check its ability to supply reliable data, further inquiry was desirable. The design of such a study would use the same group of drugs coded by two different sets of colors, evaluated by the same patient. The repetitious nature . . .

The anti-A and anti-B titers in group O individuals may vary independently following injection of group-specific substances, although such independent variation was not always seen. The rise in anti-B which followed the injection of A sub stance in some instances may have been due to traces of B activity in the A substance or to nonspecific stimulation of antibody which may follow the injection of any antigen. The part played by such factors was not determined. It is concluded that the presence of A and B agglutinins in O individuals is not dependent on a single cellular characteristic responsible for the simultaneous formation of the isoagglutinins.

1.1. Fifty-one patients with hay fever, 10 of whom had seasonal pollen asthma, who had failed to benefit from specific pollen therapy and the antihistaminic drugs or other agents, were treated with one or more four-day courses of oral cortisone. In 42 patients, treatment with cortisone appeared to provide complete to satisfactory relief; 9 patients failed to obtain relief. 2.2. It is concluded that oral cortisone in the dosage used is effective in the treatment of hay fever.

ALTHOUGH the hazards of many therapeutic procedures are reasonably well known, clinicians need to be reminded occasionally that measures thought to be perfectly safe may be dangerous under certain special circumstances. Oxygen in high concentrations, as often administered to patients with a variety of respiratory, cardiovascular and neuromuscular conditions, may at times be more harmful than beneficial. From the manner in which oxygen is prescribed, however, one would doubt that its potentially deleterious effects are widely recognized, although they have been known for some time. Oxygen given in high concentrations (50 to 100 per cent) for long periods may be . . .

I NSULIN resistance occasionally develops in patients with diabetes, who may require as much as 1,000 to 5,000 units per day to control hyperglycemia. It is also seen in patients undergoing insulin-shock therapy for psychiatric disorders, and therefore, it may occur in the absence of diabetes. The unresponsiveness in most instances has remained unexplained. Infection, the commonest cause of increased insulin requirement, does not seem t,o account for resistance to insulin in most of the cases. Endocrine disturbances have been postulated, but again, with the exception of diabetes, they have rarely been demonstrated. Evidence has been presented that an immune mechanism plays a role in some cases of insulin resistance.l-” In a previous study,? it was demonstrated that responsiveness to human insulin was present in a patient who was highly resistant to cryst,alline (beef and pork) insulin; and, furthermore, the resistance seemed to develop during periods in which insulin wa$ given and to disappear during periods in which insulin was withheld. This patient was also allergic to insulin. Injections of insulin caused swelling and generalized urticaria, and the serum contained skin-sensitizing ant,ibody. However, allergy and resistance are not, necessarily concomitant clinically, and in this patient the degree of resistance and hypersensitivity varied independently, These observations led to an attempt to induce insulin resistance on an immunological basis. Twenty-four rabbits were in jetted repeatedly over a period of many months with 25 to 50 units of crystalline (beef and pork) insulint incorporated int,o an adjuvant mixture,s containing mineral oil, Falba,$ and killed acid-fast bacilli.5 Insulin tolerance tests were done in the fasting state with 0.1 p/kg. given intravenously before immunization and at intervals thereafter. Table I summarizes the immunizing procedure in the 4 rabbits that became resistant and shows a comparison of the hypoglycemic effect of insulin before and after the development of resistance. Of the 19 rabbits similarly treated which did not become resistant, only 4 survived. Two died early in the course of the study because of overdosage with insulin; several died with diarrhea. Tuberculintype hypersensitivity of high degree to one or more of the constituents of the adjuvant mixture may have been the cause of death in some of the others. In no case was there evidence of diabetes as judged by weight, loss or hyperglycemia. One of the rabbits (No. 5) was studied in detail, and the results obtained have been reported e1sewhere.O A normal glucose tolerance test and the repeated