Experimental Biology and Medicine

Living non-pathogenic bacteria appear in the thoracic duct lymph when these microorganisms are placed in the lumen of the duodenum in an alkalinized bile solution in the post-digestive dog. 1 , 2 Various concentrations of bile have been placed in the stomach of dogs through a fistulous opening and the gastric contents removed and titrated at intervals 2 to 3 hours after the bile was administered. Dehydrated bile (Difco) has been used, 100 cc. volume of fluid, concentration of bile varying from 0.5 to 10.0%. It was found that 5.0% and higher concentrations caused vomiting within a short period of time, followed by diarrhea. There was usually a period of active HCl secretion following the emptying of the stomach in this manner. One half of the required amount of acid was added to neutralize the alkali in 100 cc. of a 5.0% bile solution to see if this mixture would leave the stomach rapidly and not stimulate acid secretion. (Line “D” in graph.) One per cent bile was found to be the optimum concentration to rapidly leave the stomach and be followed by a certain period of acid deficit. (Line “B” in graph.) Three dogs were used for these experiments. Each curve on the graph is an average of 5 experiments using 3 dogs, a total of 15 separate experiments. Experiments were conducted during the forenoon. Animals were fed in the afternoon and used on alternating dogs 18 hours after feeding.

These animals had, in addition to the gastric fistula, a caecal fistula. 2 B. prodigiosus was suspended in water and the various bile solutions and the relative concentration in the caecum was determined. This method has been used to determine the relative bacterial killing power of the intestinal tract. 2

1. The development of a green color in old (1 week to 1 year) apomorphine hydrochloride solutions appears to be accompanied by 110 qualitative or quantitative changes in the pharma- cological actions of the drug.

2. Dogs receiving an emetic dose of apomorphine daily for 4 months exhibit no tolerance or hypersusceptihility. No wuseaizt conditioned salivary reflex similrr to that of morphine appears.

In spite of many chemical and pharmacological studies of apomorphine during the past 50 years, there still seems to be considerable doubt as to whether or not the green coloration which appears in apomorphine solutions is evidence of deterioration or of the formation of a new toxic substance. Guinard, 1 Eggleston and Hatcher 2 found little change pharmacologically in solutions of apomorphine which had become a deep green. Heinekamp, 3 however, believed that such apomorphine solutions were definitely less potent. The United States Pharmacopoeia X, and some text books (e. g., Trendelenburg 4 ) recommend that solutions should not be used, especially if green immediately or shortly after the dissolving of the solid drug. Harnack 5 declared that the oxidation with subsequent loss of potency occurred with some apomorphine preparations and not with others. The experiments reported below were, therefore, in part, undertaken to determine the effect of oxidation of the dissolved drug (as shown by the green color of the solution) upon its pharmacological characteristics.

Three different crystalline preparations∗ of apomorphine hydrochloride were used for making both the new and the old solutions which were injected into 7 healthy unanesthetized dogs of from 7 to 11 kilos for 1 to 4 1/2 months. Four of the 7 animals were injected daily for at least 4 months. The procedure finally adopted was to inject the animals once daily. Fresh solutions and green aged solutions of a given sample of apomorphine were administered on alternate days, subcutaneously. The minimal effective dose of 0.15 to 0.20 mg. per kilo was found to be the same as that reported by Eggleston and Hatcher. The data in Table I are representative of the results with injections over a period of 3 or 4 months.

90 to 95% of the viable bacteria placed upon the mucous membrane of the nose are rendered non-viable within 5 to 10 minutes. Repeated application of bacteria to this mucosa is not associated with an exhaustion of this disinfecting mechanism. Cultures obtained from bacteria that have survived the initial period of disinfecting action are removed as promptly as the original strain from the nasal mucosa. Repeated applications did not lead to the development of a resistant strain.

“S” and “R” types of B. coli are removed at the same rate. Isolated instances of the appearance of an “R” variant from an “S” type of B. coli and also of lysogenic B. coli colonies were observed.

We were unable to demonstrate growth inhibiting substance in nasal washings and secretions in vitro.

Subjects in warm rooms showed a retardation in the disappearance of the bacteria from the nasal mucosa.

The average rate of disappearance of bacteria sprayed upon the mucous membrane of the nose has been determined for a series of animals, including guinea pigs, rabbits, dogs and humans. The technic finally used was the following: Broth cultures were diluted 1:200 with saline. A spray delivering 0.15 cc. (±0.01) was used to distribute the bacteria in each nares. A swab touched upon the surface of the respiratory portion of the nose at several places in a uniform manner and streaked well over the surface of an agar plate. This technic was repeated at intervals varying from one minute to one hour. An average of 400 experiments in normal humans, using 42 different subjects is shown in the accompanying graph. The average number of viable bacteria sprayed was 20,000,-000. The swab taken immediately after spraying contained an average of 2,000 bacteria. After 5 minutes the number was reduced to about 50 colonies, 10 minutes about 15 colonies, after 15 minutes about 8 to 10 colonies, after 30 minutes about 5 to 10 colonies. At times none were present after this period. After one hour sometimes 1 or 2 colonies appeared, but ordinarily all of the sprayed bacteria were absent from such agar cultures. The same results were obtained with the other animals used by us. There were 185 animal experiments. If the same bacteria are sprayed in the same number in both sides of the nose, immediate swab only being obtained from one side, and after 5 minutes specimen collected from both nares, the rate of reduction in both sides being the same. This type of experiment has been carried out many times with several different time intervals. The mechanical removal of bacteria by swabbing the mucosa does not play a demonstrable rôle in this reaction.

Human serum has no complement for gonococci (Scaltritti 1 ). This fact gives us a new viewpoint in the treatment of gonorrhoeal and blenorrhoeal conditions. With the expectation of a mild reaction I (Pribram) tried this treatment in a 4-year-old girl who had a gonorrhoic infection for 4 weeks. I introduced a few drops of a fresh animal serum into the urethra and the vestibulum of the vagina. The discharge became abundant the next day. The microscopic inspection revealed pus cells filled with gonococci. I stopped the treatment in this case and continued with an argyrol salve in the usual way. The first attempt showed that we are able to induce an acute exacerbation of a subchronic gonorrhoeal process by means of the application of fresh animal serum. Besides this it became evident that the serum had a strong chemotactic action on human phagocytes and increased their opsonic index.

The further investigations were carried on by Dr. Jonas on male adults with acute and chronic gonorrhoeal affections. The method consisted in a very cautious instillation of a few cubic centimeters of fresh animal serum into the urethra. The serum was allowed to remain in the urethra for a few minutes. The discharge, absent or very poor in chronic cases, became abundant and showed many phagocytized gonococci. It disappeared after 1 to 3 days almost completely. We had the opportunity to treat 2 patients with a fresh, hitherto untreated gonorrhoea. No other treatment than the local application of fresh animal serum on 5 successive days was instituted in this case. The abundant discharge of phagocytosis following the treatment stopped completely after the instillation of the serum had been discontinued.

Further studies will be necessary in order to state exact quantities of the complement which have to be used, furthermore the most effective serum species must be determined and the number of instillations which are necessary for a complete cure.

The series of investigations was carried out on normal individuals between 18 and 30 years of age. It includes 52 women and 50 men. The sedimentation index was determined by the Cutler method and the fibrin by the Wu colorimetric method, with slight modifications. The fibrin determinations were done in duplicate.

The average sedimentation index for women was 7.635 (mm.) ±.345; the average index for men was 2.380 ± .164. The difference between the 2 sexes is 5.255 ± .382. This difference is 16.3 times as large as its probable error, and is probably significant. The average fibrin content for women was .300 (gm. per 100 cc. plasma) ± .004. For men it was .288 ± .004. The difference between the 2 sexes is .012±.006. Since this difference is only twice as large as its probable error, it cannot be considered certainly significant.

Treatment of the sedimentation indices and the fibrin contents by the statistical method gave a correlation coefficient of .312±.084 for the women. This is 3.83 times as large as its probable error and is probably significant. The empirical mean fibrin values as ordinates are plotted against the sedimentation indices in the graph. While certain of these means, particularly that for the fibrin content of .325 (index 1), deviate rather widely from the regression line, it must be remembered that the total number of cases is 52, and that certain of these means are based on very few analyses.

The correlation coefficient for men was .328±.085, which is 3.86 times as large as its probable error, and is probably significant. The empirical mean fibrin values plotted against the sedimentation indices, and the fitted regression line is shown in the graph.

The difference between the correlation coefficients for the 2 sexes is only .016±.120.

Increasing attention is being paid to the physico-chemical properties of plant saps. Earlier studies were devoted almost exclusively to the problems of the osmotic concentration of the plant saps, particularly as related to environmental conditions. Dixon and Atkins 1 pointed out that in order to secure uniform samples of sap the plant tissues must be made permeable by freezing. They used liquid air for freezing the material, and later Gortner and Harris 2 modified this method by using a simple salt and ice bath. Since that time a voluminous literature has accumulated 3 in regard to the influence of geographical environment and climate on the osmotic properties of plant saps. In most of the work carried out by Harris, and Harris and Gortner the electrical conductivity of the tissue fluids has been recorded, in addition to osmotic concentration, but it is only recently that a new factor has been shown to be of importance.

Newton and Gortner 4 devised a technic for the measurement of “bound water” i. e., the water which appears to be associated with hydrophilic colloids and consequently is differentiated from the free water which is free to dissolve organic and inorganic solutes. Newton 5 , 6 , 7 has demonstrated that a measurement of bound water content is of great importance in the study of winter hardiness of plants, and in similar unpublished work Newton has likewise demonstrated that measurements of bound water are of importance in studying problems of drought resistance. Robinson 8 , 9 showed bound water to be a factor in ecological problems of insects. Gortner 10 has called attention to some of the relationships involving free and bound water as affecting biological problems.

The futility in expecting to obtain an unaltered metabolic product of a complex character like human or animal melanin after prolonged boiling with caustic or concentrated mineral acids, was pointed out by Gortner 1 . The method of isolation of urinary melanin by direct precipitation of the urine either with basic lead acetate or with barium hydroxide suffers from the disadvantage that a great variety of compounds besides the melanin is taken down by either method; which fact does not facilitate the purification.

Due to the great scarcity of melanuria in cases of melanosarcoma in man, the problem of a simple and at the same time mild method for the isolation of melanin from human urine is only seldom attacked. The fact that we are engaged in the isolation of another cyclic urinary compound 2 made it possible for us to use the occasion of obtaining melanotic urine from a middle aged man with multiple metastases from a melanosarcoma of the skin. The abundance of melanin produced and circulated in the organism in this patient was demonstrated not only by the occurrence of melanin in his urine but also by a unique diffuse melanotic discoloration of the skin giving a slate colored character particularly to face and hands.

For the isolation of the melanin the following procedure has been employed: The urine is evaporated on a water bath in a current of air and the residue is extracted with absolute methyl alcohol. The extraction is carried out in a tall cylinder or beaker and the salts allowed to settle slowly. This process is repeated several times with fresh washings of methyl alcohol until the final washing is straw colored or colorless.

It has previously been reported that grouse are susceptible to experimental tularemia. 1 , 2 Inoculation of an abrasion through the skin of a ruffed grouse results in a fatal septicemia.

Preliminary experiments which have been carried out on the susceptibility of the Hungarian partridge to experimental tularemia, indicate that this bird is highly susceptible to the disease, a fatal infection resulting from an open inoculation of a skin abrasion.

The pigeon appears to be relatively more resistant. In this bird an intramuscular injection causes the formation of a local lesion, and in some cases this is followed by a generalized invasion by the organism. Even with the production of a septicemia, no clinical symptoms have been observed in the pigeons in our series and those not killed have recovered.

The ring-necked pheasant also appears to be relatively resistant to experimental tularemia.

The domestic chicken appears to be absolutely resistant to the disease. Organisms injected intramuscularly do not appear to invade and produce a general infection.

The growth in length of the human tibia in the latter part of prenatal life has been studied quantitatively from observations, made chiefly by Corrado 1 and by Tamassia, 2 on a series of 152 fetuses. The data include the total body-length, the total length of the tibia, the length of the tibial diaphysis, and the combined lengths of the superior and inferior (cartilagenous) tibial epiphyses. Regression formulae have been computed for certain relationships between these values and are given, with other data, in Table I. Fig. 1 shows the distributions of these observations and their regressive lines.

These formulae indicate that, in the period under consideration, the growth in length of the tibia and the lineal growth of its various parts are directly proportional to the growth in total or crown-heel length. The empirical formulae for the total length of the tibia and for the length of the tibial diaphysis, with respect to total body-length, are characterized by negative second constants, as are the various empirical formulae which have been developed for the external dimensions of the lower extremities (Calkins and Scammon 3 ). The empirical formulae for the epiphyses, on the other hand, show positive second constants. Therefore, while the lengths of the tibia as a whole and of its diaphysis are becoming relatively, as well as absolutely, greater, the epiphyses are becoming relatively shorter, although absolutely longer.

Fig. 2 shows the calculated growth in length of the tibia and its several parts in the last trimester of prenatal life. The relation of the lengths of these structures to age, in fetal or lunar months, has been computed by the use of the empirical formula of Scammon and Calkins 4 for the relation between time and total body-length in the fetal period.

The chemical nature of amyloid, as well as the mechanism of its formation, has been very obscure. Experimental investigation of the problem has until recently encountered the difficulty that amyloid could not regularly be produced in animals. Thanks to the work of Kuczinski, 1 Jaffé, 2 Letterer 3 and others, this difficulty no longer exists. By parenteral injection of sodium caseinate and various other proteins and non-protein substances, as well as by feeding excessive amounts of protein, amyloidosis is produced in mice in the majority of those animals. In the course of the work of these investigators, certain observations have been made that greatly increase the interest of the problem of amyloid disease.

In experimental amyloidosis as well as in the spontaneous condition as it occurs in man, excessive breakdown of tissue proteins is known to take place. There is evidence that this is followed by an increased production of serum globulin, which apparently acts as the mother substance of amyloid. 3 It has been stated that on further injection of casein or other substances employed, globulin is precipitated in the tissues in altered form and that the resulting deposits constitute amyloid. The well recognized close relation of globulin to antibodies suggests that the whole process of amyloid formation is akin to an immunological reaction. This suggestion is at least worth testing, since, if true, experimental production of amyloid would be a most useful method in studying the mechanism of certain immunological reactions, having the unique advantage of producing easily recognizable and characteristic morphological changes.

The best line of attack appears to be study of changes in blood chemistry during the course of production of amyloidosis. For this purpose an animal larger than the mouse is highly desirable. Guinea pigs and rabbits have, in the hands of other workers, given negative results. The rat was, therefore, selected as a possibly suitable animal.

A series of white rats was raised upon a cholesterol-free diet over a period of 3 generations since the question had arisen as to whether or not cholesterol was synthesized in these rats in the absence of sterols. The chemical end of the research was conducted by Knudson and Randles 1 and the experimental evidence ascertained seemed to justify their conclusion that this chemical was, to a certain extent, built up by the rat body. 2 At the conclusion of their experiments, the rats were killed and the tissues fixed in Zenker, Bouin, Meves, and formalin, and stained with a variety of the usual laboratory stains, chief among which was a combination of hematoxylin and Mercurochrome-220 Soluble (Hynson, Westcott and Dunning, Baltimore, Md.).

These rats of the third generation were weak, small, and undernourished, and lived only from 11 to 17 weeks after birth, i. e., from 7 to 13 weeks after weaning. At the time of weaning they weighed from 26.0 to 30.0 gm., and at the time of death from 74.0 to 80.0 gm.

There was no evidence at the time of autopsy of either inflammation or degeneration in any of the body tissues. All of the body organs were reduced in size and in weight but preserved a fair normal proportion to the mass of the body in comparison to the controls.

All of the tissues and organs of the rats were studied in detail histologically. It was evident that the diminutive size of the various viscera had been attained less by a reduction in number of the individual functional units composing them, than by a decrease in the size of the cells forming these units. Hence, the tissues and organs presented the appearance of structural compactness, without a conspicuous absence of the characteristic cells ordinarily composing them.

Previous investigations have repeatedly shown that graded responses in rate of growth may be obtained from the feeding of quantitatively graded allowances of either vitamin A or vitamin B, in each case as supplement to a basal diet adequate in all other respects. And in the case of vitamin B it has been shown, first by Osborne and Mendel 1 and later in a different way by Sherman and MacArthur, 2 that the larger the experimental animal the greater the amount of vitamin needed to maintain a given rate of growth, or the less the rate of growth resulting from the feeding of a fixed limited allowance of the vitamin. This is now found to be true of vitamin A also, larger experimental animals, of a given age and sex, growing more slowly upon a fixed allowance of this vitamin, the allowance in all of these cases being less than would have sufficed to support a fully normal rate of growth.

The bearing of this finding upon the accuracy of the quantitative determination of vitamin A by means of the feeding method is being discussed elsewhere. The purpose of the present communication is rather to invite consideration of the bearing of these and some related observations upon the nutritional rôle of vitamin A in relation to growth and health, with special reference to the importance of this factor in the food of children.

Thus the facts now demonstrated upon experimental animals (albino rats) that the vitamin A requirement increases both with size of body and with rate of growth will mean in the feeding of children that the large child needs a liberal allowance of this vitamin because of his size and the small child needs a liberal allowance to induce and support a more rapid rate of growth.

Some years ago it was shown that hen's eggs are especially rich in the antirachitic factor, an observation which has been confirmed repeatedly. We have since found that the eggs of other animals, for example those of various fish and of serpents, likewise contain a large amount of this specific factor, the concentration varying considerably. In the hen the concentration is greater in the egg than in the liver. Several investigators 1 have shown that the content of the antirachitic factor in hen's eggs varies according to their diet and the degree of exposure to ultra-violet rays, and furthermore than when this factor is notably deficient a marked lack of hatchability results. This failure to hatch has been attributed to defective absorption of calcium from the shell into the yolk.

From last summer until this spring one group of hens has been placed on a diet devoid of the antirachitic factor (a grain mixture, including yellow corn, also meat scrap and 1 part of sodium chloride) but otherwise complete, whereas another group has been given daily supplements of cod liver oil. Our experience may be summarized by the statement that the group on the deficient diet laid comparatively few eggs even during the productive period of March and April, that these eggs contained very little of the antirachitic factor, that fertile eggs failed to hatch, and that the livers of these hens were also lacking in this substance. It should be added, however, that the blood serum of both groups of hens was found to contain a high percentage of calcium in the spring, that is during the productive season, in spite of the fact that one of the groups was laying and hatching normally, whereas the other was almost inactive in this respect.

In a recent communication it was reported that in the cod there is a definite relationship between the percentage of calcium in the blood and the maturation period, the calcium rising to a markedly high level at the spawning season. 1 It was found likewise, in agreement with the investigations of others, that there is an inverse relationship between the size of the liver and the size and development of the roe, the liver becoming smaller as the ovary fills up and grows larger. At this time the liver may lose 50% or more of its fat. In many instances the weight of the fully matured ovary far exceeds that of the liver; for example in 3 fish, the ovary weighed 400, 345, and 270 gm. respectively, whereas the liver weighed 29, 105 and 60 gm. On testing the roe for antirachitic potency it proved in every instance to be active; this held true for the roe of the cod, sheepshead, flounder, carp, buffalo fish and smelt. The milt of the cod and of the carp had similar protective value. When the respective weights of the organs, of the liver and the ovary, were taken into consideration it was found that the latter contained quite as much as, or even more, of the antirachitic factor than the liver. In view of these facts it seems that the liver may be the source of the specific factor in the roe.

In the course of these experiments it was found that the antiracitic potency of the liver of the cod is in inverse ratio to its content of oil. Much to our surprise tests of the ether extract of livers, very small and poor in fat, showed them to be exceptionally potent—far exceeding in potency any oil which has been heretofore reported. The details of these tests will be published elsewhere.

In the course of previous studies, 1 , 2 it was found that type I pneumococcus culture broth contained a substance which resembled the hemotoxin described by Cole 3 , 4 and later by Avery and Neill. 5 Unlike the substance described by Cole, which was contained within the cells and was only set free by their dissolution, the hemolytic substance in the culture broth appeared, in some cases, as early as 4 hours after incubation was begun. The titer of the lytic substance, as indicated by the amount of culture broth required to produce an appreciable amount of hemolysis of 1 cc. of 1% washed sheep cells, was usually greatest after from 16 to 24 hours’ incubation.

Culture broth of a group-IV pneumococcus strain, comparatively avirulent in its present form but capable of having its virulence raised by animal passage, and of a completely avirulent, atypical pneumococcus strain, No. 189, isolated from a horse which died during immunization with the standard type-I strain, No. 5, also contained the hemolytic substance.

The hemolytic substance in the culture broth, like that of Cole, is thermolabile and is destroyed by heating for 30 minutes at 55° C. Its activity is diminished by standing at cold room temperature and also, in a much shorter time, at 37° C.

Red cells, stroma, and leucocytes absorb the substance from the culture broth. The substance is also absorbed from the culture broth by shaking with animal charcoal and alumina. Shaking alone, however, greatly reduced the lytic power of the culture broth.

Normal horse serum has an inhibitory effect on the hemolytic power of the substance, while pneumococcus immune serum has a much more marked effect, which is apparently species rather than type specific.

The present comparative study seems to the writer to warrant the conclusion that the tooth organisms discussed here’ constitute a type entirely separate from B. acidoplzilus of the intestinal tract.

During the course of an investigation in which dogs were fed a diet deficient with respect to vitamin “B” one animal developed severe convulsions. Death seemed imminent. The only chance to save its life was the immediate parenteral injection of a vitamin “B” concentrate (Cowgill 1 ). Vitamin powder (Harris) was dissolved, neutralized, sterilized by boiling for 3 minutes and the solution injected subcutaneously. Practically all symptoms were alleviated within a few hours.

A few days later an abscess developed at the site of the injection. Palpation revealed gas crepitus. It was then recalled that the material injected had not been sterilized sufficiently to destroy resistant spores. A provisional diagnosis of “gas infection” was therefore made. The abscess was incised and drained and a pure culture of Bacillus aerogenes capsulatus isolated. The animal was given large doses of material rich in vitamin “B” daily and recovered rapidly. Repeated cultures subsequently made from the site of the wound proved to be negative for B. welchii.

The following questions arose: (1) are normal dogs immune to this organism? (2) if normal dogs possess “natural” immunity, is it possible for a lack of a dietary essential to bring about a breakdown of this immunity? (3) was the vitamin “B” therapy responsible for the very rapid development of immunity? An attempt was made to answer these questions by experiment.

Dogs, presumably normal, were inoculated with varying quantities of cultures of this organism. No untoward symptoms were observed. The animals were then fed a diet complete in every known respect except for what has hitherto been called vitamin B. Positive blood cultures were obtained. Vitamin “B” therapy was then instituted and the blood cultures became negative. This procedure was repeated a number of times over a period of 15 months and the same results obtained.

Studies were made of excised kidney and muscle tissue of normal and depancreatized dogs, in order to answer the following questions:

(1) With normal animals what is the respiratory metabolism of these tissues when removed from the body?

(2) In excised diabetic tissue (a) is the power to oxidize carbohydrate lost? (b) is the power to split carbohydrate into lactic acid lost?

The main features of the technique were (1) the employment of the Barcroft-Warburg manometers in the study of micro-respiration, (2) the comparison of tissues of normal and diabetic animals, (3) parallel determinations of the metabolism of the tissues of the same animal, and (4) the development of a technique for obtaining relatively untraumatized mammalian muscle, by stripping the fasciculi lengthwise, instead of mincing.

The results are as follows:

(1) The figures for oxygen consumption of kidney and muscle were remarkably constant. For instance in normal muscle the extreme variations in 24 consecutive determinations were from 2.3 to 3.1 cu. mm. per dry mg. per hour. These correspond to values obtained in both tissues by the method of analyzing arterial and venous blood simultaneously in the intact organ. The respiratory quotients were all within physiological limits and indicated that various mixtures of foodstuffs were oxidized. In 11 triplicate determinations the highest figure was 0.911. This cannot be reconciled with the widespread idea that only carbohydrate can be oxidized by muscle.

(2a) The average respiratory quotients were as follows:

These figures show that isolated diabetic tissue can oxidize only a small amount of carbohydrate.

(2b) Diabetic tissue retains the power of splitting carbohydrate into lactic acid. Therefore it is not a failure to glycolize which depresses the oxidation of carbohydrate in diabetes.

This study was made upon the perfused liver of the bullfrog (Rana catesbiana). It was found that in the isolated livers, whether excised or left in situ, thyroxin as a rule produced a slight fall in the amount of sugar given off into the perfusate, as against the amounts of sugar obtained during the preliminary control periods.

When the body of the frog was perfused with thyroxin through the systemic aorta there was a marked rise in the sugar of the perfusate draining from the hepatic vein. (Graph 1.) This increase lasted only a limited length of time and was then followed by a gradual decline of the sugar concentration to the control level. The fall occurred in spite of prolonged perfusion with thyroxin. The same results were observed in a frog with the adrenals excluded from the circulation.

When the liver was tied off so that it could be perfused separately from the body, and the body perfusates, collected from the posterior vena cava, were passed through the isolated organ, there was a sharp and prolonged increase in the hepatic sugar. There was no further rise, however, as a result of perfusion with body perfusate which had thyroxin added to it before passing it through the body. In an experiment with 2 frogs which were cross-perfused, the fluid from the vena cava of the one frog emptying through a rubber tube connection into the liver of the other, there was a 22% rise in the hepatic sugar. At the end of the experiment this liver was found to contain only 6.10 mg. of glycogen.

These results seem to point to the conclusion that the effect of thyroxin upon the carbohydrate of the liver is probably an indirect one. The manner in which this influence is brought about is not yet clear.

Last year we 1 demonstrated for the first time that normal guinea pigs, when exposed to an organic dust such as horse dander, could become sensitized through inhalation. Guinea pigs, thus sensitized, when again placed in contact with the same dust after a suitable incubation period, demonstrated unmistakable signs of anaphylaxis which we termed “respiratory anaphylaxis” and which we believed to be identical with bronchial asthma in the human being. This was offered as a method for studying asthma experimentally in the animal.

Recently, Figley and Elrod 2 reported the occurrence of a large number of cases of asthma resulting from the inhalation of castor bean dust “pomace” thrown into the air from the pipes of a castor oil factory. To determine whether pomace was an anaphylactogen and could be the cause of asthma, we proceeded to study the problem of pomace asthma in the guinea pig. The pomace came from the above castor oil factory.

Thirty-seven normal guinea pigs were placed for several hours in glass cages in which pomace was kept in circulation by currents of air. After an incubation period of about 3 weeks, 2 of these animals were given an intravenous injection of 0.5 cc. of a crystal clear alkaline extract of this pomace. Both animals showed definite anaphylaxis, one dying within a few minutes and demonstrating typical anaphylactic lungs, and the other going into collapse with marked dyspnea and finally recovering. These 2 animals established the fact that sensitization could be accomplished by inhalation of pomace.

Of the remaining 35 animals, 17 died 6 to 7 days after the sensitizing inhalation of pomace, with definite evidences of ricin poisoning. However, besides the classical abdominal pathology of ricin poisoning, we noted in many of the animals a severe hemorrhagic condition of the lungs. We first attributed this lung condition to the entrance of ricin through the .nasal passages but we obtained similar results later when the pomace extract was injected.

Experiments previously reported 1 have shown that the cut ends of the spinal cord can be prevented from reuniting by the removal of a comparatively large amount of the nervous system and the implantation into the wound of the anlage of the limb and pronephros.

The responses of such an embryo during early stages are strictly in keeping with the findings of Coghill and Harrick. The uninjured portions of the nervous system behave as individual units and take up their function with a very slight delay over the normal. As the animals differentiate, a complex series of reactions occurs by means of which stimuli are transmitted across the gap in the nervous mechanism. This transfer of stimuli takes place through the intermediation of the musculature which is in that region.

The first integration of response was noticed in the reactions occurring after stimulation in the head region. The sensory pick-up is mediated through the uninjured medulla giving rise to a motor response through the 7th cranial nerve which activates the gular musculature and the arcuate muscles of the gills. This reaction is transmitted to the anterior trunk region where the sensory pick-up acts in the pectoral region to transmit the stimulus to the uninjured cord of the trunk. This motivates the musculature of the flank and tail and gives rise to distinct swimming movements.

The limbs which have been transplanted into the wound area develop synchronously with the normal limbs. They are generally single and although they develop in close proximity to the gills, show no trace of the developmental dominance of that tissue. In every case the laterality of these limbs has been that of the side from which they were taken.

This report deals with the effect upon growth and viability of streptococci, of the addition of various nitrogen compounds to synthetic media. Details of the technique involved have previously been described. 1 , 2 The substances tested include inorganic nitrogen compounds, amino nitrogen compounds, benzene and imidazo derivatives and alkaloids.

We found that (NH₄)₂HPO₄ is more available for streptococcus than (NH₄)₂CO₃, NH₄NO₃ and (NH₄)₂SO₄. Previous investigators make no mention of such findings as applied to streptococcus metabolism. However, Dolt, 3 in his study of about 40 simple media for the growth of B. coli has chosen 2 as the best. One of these contains (NH₄)₂HPO₄ as the sole source of N. Fermi 4 shows that this ammonium salt is easily utilized by B. prodigiosus and B. fitzianus.

The relation of amino acids to the growth of streptococcus is not constant. For example, one series of experiments, in which amino acids were added to a basal mixture, showed glycocoll more beneficial than aspartic acid; in another series of experiments, the same amino acids were added to another basal mixture, aspartic acid was in the lead and glycocoll fell behind. Similar variable effects were obtained between aspartic acid and alanine and glycocoll and alanine. Tyrosine was always less available than glycocoll in these studies. The relation between tyrosine and aspartic acid was not so regular. It is evident that no definite conclusion can be drawn as to the comparative utilization of a certain group of compounds when a single basal mixture is investigated.

When amino acids were compared with other N compounds 2 tendencies were noticed. In some cases the former were utilized less rapidly than the latter. There were, however, comparatively few instances in which this was true.

The filtrate of Chicken Tumor No. 1 when kept in contact for 1 to 5 hours at room temperature with finely minced or dried muscle from susceptible fowl is generally either completely inactivated or has its potency greatly reduced. This partial or total inactivation was pronounced in 34 out of 46 tests. Carefully controlled experiments have shown that the 12 cases in which there was no demonstrable reduction in potency of the filtrate after muscle contact was explained by the fact that excessively active filtrates were used, for the inactivating effect of the muscle became evident when such filtrates were tested in more dilute form.

The specific nature of this reaction was shown by the fact that not only did muscle tissue of such resistant animals as the rabbit and pigeon have no effect on the activity of the tumor filtrate but the brain, kidney and liver of the susceptible chicken were without activity in this respect.

Desiccation of chicken muscle pulp after contact with filtrates has failed to release the tumor agent in an active form from the muscle. Either the union of the agent with the muscle substances is so strong as to render dissociation difficult or the phenomenon may be one of inactivation.

In the experiments reported previously 1 material was used for the peptic digestion consisting mainly of egg albumin but also containing egg globulin. Since it was thought that possibly the latter was responsible for the effects observed, experiments were made with preparations of egg globulin. The results were in agreement with the former experiments in that the animals sensitized with the peptic digestion products were more sensitive to these than to unchanged globulin and vice versa. This result was especially pronounced in one batch of guinea pigs, the lethal intravenous dose for the animals sensitized with the digestion product being 1 cc. of a 1-2% solution of the digested material∗ and about 1 cc. of a 0.2% solution of globulin, while the respective values for the animals sensitized with globulin were about 4% and 0.0125%. In another batch of animals this difference was still definite but less considerable. But, after precipitating a 5% solution of the digestion product with an equal volume of 10% trichloracetic acid and precipitating the supernatant fluid with alcohol the resulting substance was much more toxic for the guinea pigs sensitized with the digested material (lethal dose 0.5%) than for those sensitized with globulin (lethal dose 4% or more). Similar results were obtained by sensitizing with digested heat-coagulated egg globulin. These animals were killed by an i. v. injection of 1 cc. of 0.25% to 0.125% of the digestion product after fractionation with trichloracetic acid and by 1 cc. of a 0.1% solution of unchanged globulin.

In some cases, but irregularly, the animals sensitized with peptic products succumbed to injections of 1% ovomucoid. This substance, however, can hardly be essential for the results reported since digested products of heat-coagulated globulin were more active in producing shock than those of non-coagulated globulin in animals sensitized with digestion products of the latter preparation, and also on account of the results of experiments with guinea pigs sensitized with ovomucoid itself.

The index of refraction of a mixed solution of gelatin and of hydrochloric acid as measured by an immersion refractometer has been stated to be an additive function of the indices of refraction of the individual solutions. 1 Using the Zeiss-Löwe interferometer, which is a differential refractometer capable of measuring small changes in refractive index to which an immersion refractometer is insensitive, the authors have found that the refractive indices of mixed solutions of gelatin and of hydrochloric acid are not additive.

However, the refraction of mixed solutions of gelatin chloride and of hydrochloric acid are additive. Consequently, when a solution of gelatin is titrated with an acid, an abrupt change in slope of the curve (refraction as function of acid added) is noted at the stoichiometric point. The authors have found this to be the case also when an amino acid is similarly titrated, and recently the interferometer method of following simple acid-base neutralizations has been reported elsewhere. 2

The data for a gelatin titration are given in Table I and are shown graphically in Figure I. The gelatin used was a pure specimen obtained from the Eastman Kodak Company. Its ash content was 0.03%.

The 2 branches of the curve intersect at the point equal to about 7.9 cc. of acid. Consequently the equivalent weight of this specimen of gelatin is about 1111. For the benefit of the reader interested in a description of the instrument used by the authors, the excellent paper of Adams 3 is recommended.

Phosphate added to the diet causes rapid healing of rats made ricketic on a high calcium-low phosphorus regimen which has not been irradiated and is poor in vitamin D. This is evidenced by histological sections of the bones, blood serum analyses, bone analyses and studies of the metabolism of calcium and phosphorus. 1 As a result of the alteration in diet not only is the rickets cured but a transient tetany is also produced. The blood serum phosphate becomes extremely high—the calcium low. If the animals survive the tetany the blood serum calcium and phosphorus return to values considered normal in 2 weeks. Similarly, the feeding of phosphate to normal dogs is known to produce tetany (Salvesen, Hastings and McIntosh 2 ).

When the diet for rats was altered by the addition of phosphate it was not considered a high phosphorus diet. It was known from the studies of McCollum et al. 3 to produce normal rats. The question accordingly arises whether non-ricketic animals shifted from a diet of low phosphorus to one of moderate phosphorus content would develop tetany. If so, would this be of short duration like that observed in the cure of rickets? Or is there something in the condition called rickets which predisposes to an exaggerated effect of phosphate?

Plan—Rats fed with diets of a high calcium low phosphorus content and a liberal supply of anti-ricketic agents were compared to those fed on the same diet without vitamins. After 3 weeks on this diet all were given enough phosphate to make the Ca/P ratio 1/1. Seven and 14 days later the blood serum was analyzed for calcium and inorganic phosphorus, histological preparations of the bones were made, and the percentage ash in the fat-free bones was determined.

Observations on individual differences in human blood aside from the 4 blood groups 1 , 2 support the idea that in transfusion the blood of the donor and recipient is, as a rule, not identical in biochemical constitution even though belonging to the same blood group.

In the literature instances are reported in which the first transfusion was uneventful while a repeated transfusion with the same compatible donor was followed by severe symptoms, 3 , 4 , 5 , 6 , 7 and others. There are, however, only casual remarks on the occurrence of atypical isoantibodies under such conditions.

We examined recently the sera of 4 individuals of group 0 and 3 of group A.∗ In 5 of these individuals, who had been transfused several times with the blood of the same donor of their own group, no abnormal agglutinins were observed. The 6th individual, of group 0, was transfused with blood (500 cc.) of the same group and the transfusion was repeated, using the same donor, since in the routine test no incompatibility was detected. Actually no untoward manifestation followed. On careful observation with a more delicate technique, however, an abnormal isoagglutinin was found in the serum of the transfused individual (after the 1st and 2nd transfusion) that reacted on numerous bloods and stronger on several others than on that of the donor. The reactions took place with bloods of group 0 and, as could be shown after removal by absorption of the common isoagglutinins, also with bloods of the other groups. Some of the results are recorded in the table.

In a number of phenoltetrachlorphthalein liver function tests, determinations of the bilirubin content of the blood showed a rise of the icterus index following the intravenous injection of the dye. When bromosulphalein was used no rise in the index took place. At first the reason for the rise in index following the use of the phenoltetrachlorphthalein was thought to be its toxic action upon the liver. Dr. S. M. Bassett, however, suggested that, inasmuch as distilled water was used to dilute a 5% to a 1% solution in the case of phenoltetrachlorphthalein, and no dilution was made in the case of bromsulphalein, the increase in icterus index might be due to the laking of red blood cells by the distilled water. Accordingly injections of distilled water up to 100 cc. were given intravenously, and, as surmised, a rise in the icterus index was produced.

From this observation the idea arose that this use of distilled water might serve as a liver function test. The rationale was as follows: if the normal liver removes bilirubin from the blood (the source of bilirubin being the hemoglobin from destroyed red blood cells), it is reasonable to suppose that a disordered liver will remove it less efficiently. If, then, in the normal person after the injection of distilled water the icterus index rises from 2 to 4 points one half hour to an hour after injection, with an average rise of 2.5 points, and returns to its basal value in 5 hours, a greater increase in icterus index might conceivably be expected in the individual with a disordered liver, or the return to normal might be prolonged.

Adequate dosage of the water was ascertained by experiment. 25 cc. are used in the case of individuals weighing from 100 to 110 pounds, with an increase of 5 cc. for every pound of body weight up to a maximum dosage of 50 cc.

Gravimetric determinations of total sulphur, non-protein sulphur, total sulphates and inorganic sulphates have been made upon the humors of oxen eyes. Mean figures are based upon from 6 to 14 analyses in each instance. The total sulphur of the aqueous humor was 3.74 mg. per 100 cc., and of this 64% was found to be nonprotein sulphur. Sulphate sulphur formed 54% of the nonprotein sulphur. Ninety % of the total sulphates were present as inorganic sulphates.

In the vitreous filtrate 4.05 mg. per 100 cc. of total sulphur were found, and of this total sulphur, 54% was non-protein sulphur. Seventy-four % of the nonprotein sulphur was sulphate sulphur. The inorganic sulphates formed 83% of the total sulphates.

The sodium nitroprusside test for glutathione has been applied to normal lenses, both human and oxen, and to cataractous lenses. All of the normal lenses gave a deep magenta color. Of 64 cataractous lenses, 26 (40%) showed no color, 19 presented a trace of color about the periphery, and in 19 a color definitely lighter than in the normal lenses was obtained.

The mercuric atom of the double mercuric-sodium chloride salt in carbonate solution is reduced by glucose at room temperature, giving a fine crystalline precipitate of mercurous chloride. This precipitate bears a quantitatively reproducible relation to the concentration of glucose which can be easily measured in the Junior Scopometer. 1 With low concentration (circa. 5 mg. glucose per cc.) the turbidity is equivalent to 8 scale divisions per mg. on a wedge calibrated for urinary proteins precipitated by Exton's reagent and with high concentration (100 mg.) somewhat less than one scale division. This gives a wide range in its practical application and sensitiveness where this is most important for clinical interpretation.

The microscopic crystalline precipitate consists of stubby needles when precipitated from a pure solution in the presence of certain colloids. Without the colloid the needles form crosses, and in the presence of acetate ions these fill out giving square plates with cross figures. The crystal habit and size are reasonably controllable even in urinary filtrates. Other substances also reduce the mercuric atoms. Of those which may be present in urine the volatile aldehyde and ketones reduce the mercury before the solution is made alkaline and the mercurous chloride from this source may be removed before the glucose reduction is initiated by the addition of carbonate.

Creatinine and many other substances are removed by a preliminary clarification. Mercuric salts have been found convenient for this purpose, each salt and combination of salts removes most of the urinary constituents in definite fractions. This matter is being investigated in detail and will be the subject of subsequent contributions since it is considered of practical value in clinical chemistry.

This work is part of a series of experiments to determine whether subcutaneous injections of Mercurochrome-220 Soluble (Hynson, Westcott and Dunning, Baltimore, Md., U. S. A.) rendered adult white rats more susceptible to the lethal effects of X-ray energy. On alternate days, 1.0 to 2.0 cc. of a 0.5% solution in normal saline was injected into the dorsal subcutaneous tissue of 12 adult rats, all of the same size. Different areas were selected for the successive injections to avoid a possible necrosis of the tissues from too concentrated and prolonged action of the stain. Thus each rat received a total of from 6.0 cc. to 8.0 cc. of this solution. Upon the 4th day, there was noticeable a distinct reddish tinge to the skin, ears, eyes, mucous membranes, and to the feces and urine as well. All of the animals were killed after the 4th injection; i. e., 7 days after the beginning of the experiment.

At autopsy, the connective tissues, fat, blood, skin and various organs of the digestive tract and of the urinary tract were distinctly stained with the dye, as has been reported by other investigators. 1

Upon microscopic examination, the kidneys presented histologic changes more marked in the glomeruli and in the cells lining the convoluted tubules. The vessels of the glomeruli and of the adjacent portions of the tubules were distended and filled with deeply stained red blood cells. The columnar lining of the convoluted tubules presented various stages of cloudy swelling, necrosis, and exfoliation. The most interesting and most constant of the histologic findings were referable, however, to the cecum and ascending colon.

Young and Hill 1 have reported colitis and enteritis present in animals receiving intravenous therapy, and, likewise, Todd 2 has recognized a form of catarrhal colitis.

In the ethylene glycol method of grading collodion membranes 1 the grading is brought about not, as Pierce supposed, by ethylene glycol alone but by ethylene glycol plus the alcohol retained in the membranes after 24 hours drying.

Ethyl alcohol is readily absorbed both by ethylene glycol and by calcium chloride, therefore the amount of alcohol retained by the membrane is determined in part by the per cent of glycol in the collodion solution employed, and in part by (1) the quality of calcium chloride used in the outlet from the drying chamber and (2) the accessibility of this chloride to the alcohol vapors from the chamber. By varying the diameter of the opening between the floor of the drying chamber and the container of calcium chloride, wide variations were produced in the “weight before washing” and in the “wet weight” of membranes prepared from collodion solution of the same ethylene glycol content, while with the calcium chloride constant in all respects the grading by means of ethylene glycol was good.

This study was undertaken to determine the extent to which the task of excretion of nitrogenous katabolites could be intensified in the organism by simultaneous reduction of kidney tissue, reproduction with lactation and high protein intake and to observe the physiological adjustments made to these conditions. Unilateral nephrectomy was performed on adult female rats which were placed immediately on the experimental rations and, as soon as the regular oestrous cycle was reestablished, were mated. The weight of the litters at birth and at weaning were compared and lactation closely observed during this time. Some of the mother rats were killed after the first litter was weaned and others were bred one or more times further, and then killed.

Reproduction on the protein diets was fairly successful but lactation was definitely deficient until certain changes were made in the experimental food. In Table I is given a summary of the high protein rations and control low protein rations fed.

Renal enlargement calculated on the basis of increase in weight of the remaining (left) kidney over the “normal” kidney weight corresponding to the actual heart weight ranged from 219 to 29% in the animals on the protein-rich rations and from 58 to 7% in those on the low-protein control food. The degree of renal enlargement appeared to be due largely to 2 factors, namely, to the concentration of dietary protein and to the extent to which the maternal organism was called upon to satisfy the nutritional demands of the suckling young. The length of time on the ration after nephrectomy, the age of the animal, and gestation unaccompanied by lactation, were comparatively negligible factors in the renal enlargement under the conditions of these experiments.

Studies on the life history of Blepharisma undulans were carried on for just 6 years, all the animals employed being the progeny of a single individual isolated November, 1921. The work was planned to determine, in particular, the viability of this infusorian in the absence of fertilization, and the effects of fertilization on the rate of reproduction, in an attempt to broaden the results from our earlier studies on Paramecium aurelia 1 and Spathidium spathula. 2 A preliminary statement of the characteristics of this race of Blepharisma undulans was published in 1922, 3 and a summary of the chief experimental results was presented in 1927. 4

The present communication is to reiterate the results previously presented in the light of the final year of study of the culture.

In brief, the conclusions derived from Paramecium and Spathidium that these infusorians are capable of unlimited reproduction, under favorable environmental conditions, without recourse to fertilization, may now be extended to Blepharisma; and, furthermore, the conclusion that fertilization accelerates the division rate, as previously demonstrated in Uroleptus 5 and in Spathidium, 2 may now be extended to Blepharisma.

These studies have been aided by a grant from the Bache Fund of the National Academy of Sciences.

The otic vesicle of various species of Rana was extirpated completely and in part. In one series the ear was grafted homoplastically to the flank, and later regrafted to the orthotopic position. The animals were carried through metamorphosis, to study the effects of these operations upon the formation of the middle ear. It was sought to determine the extent which differentiation of the inner ear influences the formation of the plectrum and tympanum, and the periods at which this influence may be effective.

The operations were performed on the left ear. Equilibration was found to be abnormal, the larvae swimming in counter-clockwise spirals or in clockwise circles. The larvae at rest were found to lie upon the left side. Normal equilibration was regained in some of the older larvae, but a number of metamorphosed animals showed a sharp inclination of the head to the left.

The grafted ear vesicle differentiated more or less completely. In some cases considerable differentiation of the otoliths took place. Normal differentiation of the Anuran ear in heterotopic position has recently been described.

The metamorphosed animals had a fully developed tympanic membrane on the operated side. The annulus tympanicus was present. A plectrum was formed when but a small rudimentary ear was present, its medial end fused with the ear capsule, which showed no operculum. The tympanum and Eustachian tube were formed normally.

Marked discrepancies in regard to the normal survival period in cats, following bilateral adrenal extirpation, are apparent throughout the literature dealing with the adrenal cortex. Zwemer 1 states the average survival period for a series of 43 cats to be 53 hours with 26 and 111 hours as the extremes. Swingle 2 finds the average survival period for bilaterally epinephrectomized cats to be about 60 hours. Houssay and Lewis 3 found that animals from which the last vestige of cortical tissue had been removed survived but from 24 to 26 hours. Hartman, 4 in seeking an active cortical extract, considered 90 hours as a marked prolongation of the life of bilaterally operated cats. The above references will serve as evidence that no definite criterion has as yet been established in work of this nature.

In a series of 31 animals the author has found the average survival period to be 115 hours, and concludes that the normal life-span of untreated bilaterally epinephrectomized cats must be placed slightly in excess of 100 hours. These findings have been completely corroborated by the recent work of other investigators.

Hartman 5 has placed the average survival period at from 5 to 6 days. Marine and Bauman 6 also agree that the life-span should be placed at somewhat in excess of 100 hours (5.3 days). Professor Swingle, in recent communications to the author, states that he would place this period in excess of 100 hours. Zwemer, in a recent conference with the writer, has stated that he would at present place the life-span of control animals at approximately 100 hours.

The author is inclined to attribute the comparatively short survival periods obtained by various workers to imperfections in operative technique, excessive visceral manipulation during operation, or to excessive handling and excitation.

For the purpose of comparing the growth of normal and heteroplastically transplanted organs, eye vesicles (with lens ectoderm) and limb discs were transplanted between Amblystoma punctatum and A. tigrinum. The operations were done in the “tail-bud” stage, and the grafts were placed in the normal position and orientation. To standardize the feeding, both species were daily fed the maximum amount they would eat.

Measurements of normal and transplanted eyes were made with a micrometer ocular at regular intervals. As accurate measurements of the developing limb are impossible, pairs of animals with heteroplastically transplanted limbs were killed from time to time as a comparison.

Measurements made as soon as the outline of the eye became sufficiently visible, before either species had reached the feeding stage, showed that the normal and transplanted eyes were of the same size. This relation is maintained during most of the larval period. As the time of metamorphosis approaches, accurate measurements of the eye can no longer be made, but apparently the normal tigrinum eye now becomes slightly larger than the transplanted eye on the punctatum, and the transplanted punctatum eye becomes larger than its control. In other words, the 2 eyes on the tigrinum, normal and transplanted, become larger than the 2 on the punctatum.

In normal development, the fore-limb of Amblystoma punctatum becomes visible externally about the same time as the gills, while that of A. tigrinum does not appear until just before the animal begins to feed. When the limb disc is transplanted heteroplastically, the transplanted limb and the normal limb of the donor species appear at the same time. During the major portion of larval life the transplanted limb and the normal limb of the donor species maintain the same size. Eventually, in late larval life, the normal limb of the tigrinum becomes somewhat larger than the limb transplanted to the punctatum, and the transplanted punctatum limb becomes larger than its control. The growth relations of the eye and limb after heteroplastic transplantation are therefore the same.

Students of vitamin “B” deficiency are well aware that animals fed a complete diet except with respect to vitamin “B” lose their appetite for the deficient ration and rapidly decline in body weight. It is also well known that an increase of blood concentration occurs in starvation. The question arises whether a deficiency of what has been spoken of hitherto as vitamin B is associated with anhydraemia, and if such is the case, whether this can be accounted for by the starvation phenomenon which so frequently prevails.

Consequently, a series of albino rats weighing between 200 and 250 gm. were placed on a vitamin “B” deficient diet. Control animals were given on an average the same amount of food and water daily as the vitamin “B” deficient rats consumed. Blood samples were obtained by cardiac puncture from each rat of both groups from week to week during the course of the vitamin deficient and realimentation periods. Hemoglobin estimations were made by the Cohen and Smith 1 method.

The results indicate that a definite increase in hemoglobin occurs during the vitamin “B” deficient period followed by a very distinct drop during the first few days of realimentation. The parallel changes that developed in the control group of rats indicate that the phenomenon just described may be accounted for by the diminished food and water consumption incident to “B” avitaminosis. Other experiments performed on another species (dog) and extending over a period of approximately 1 year appear to confirm these results.

In an attempt to throw some light on the disturbances of gastric motility occurring during the deficiency of what has been spoken of hitherto as vitamin B these experiments were undertaken. Dogs were deprived of water and parallel observations on gastric motility, and blood concentration as indicated by hemoglobin estimations, were made. Except for anorexia and decrease in general activity no abnormal symptoms were in evidence. All of the dogs developed an increase of approximately 20% in hemoglobin (estimated by the method of Cohen and Smith 1 ) and complete gastric atony. Liberal administration of water produced a rapid decrement in blood concentration, followed by a return of gastric motility to normal within a few days. The results indicate a striking parallelism to exist between anhydraemia and gastric motility. The significance of this phenomenon in relation to disturbances of gastric motility occurring in “B”-avitaminosis will be discussed elsewhere.

1. Kahn antigen base when concentrated, chilled and used according to a formula described above yields thoroughly satisfactory emulsions that retain their antigenic properties undiminished in precipitation tests for syphilis for 3 days.

2. These emulsions are further superior to Kahn antigen dilutions in tests done at low room temperatures in that under these conditions they give no false clumping.

The methods for studying the changes of bronchiolar tone in intact animals are often complicated by changes in the circulation which make their interpretation uncertain. It would therefore be desirable to check them under conditions in which the circulation would be excluded, namely outside of the body. The use of ring preparations of tracheal or bronchial muscle for this purpose has not been altogether satisfactory, partly because these are confined to the trachea and larger bronchi which might react differently from the smaller bronchioles that play the major part in the bronchial reactions; and partly because it is difficult to reproduce the natural conditions of tension. These objections are avoided in the following method, which is based on measuring the rate of flow of a Locke solution through the bronchial tree, allowing the fluid to drain off by filtration.

The arrangement is shown diagramatically in Fig. 1. A Mariotte bottle (M) filled with Locke solution is connected with a Woulff bottle (W), also filled with Locke solution, which sits in a bath (B) regulated to keep the temperature of the solution at 39° to 40° C., as indicated by the thermometer (Th). The outflow passes to a tube (T). The upper limb of this is furnished with a short piece of rubber tubing (R) which may be closed by a screwclamp (S). The lower limb is connected with a cannula tied into the short stump of the trachea of the excised lung (L). This should be used either promptly after the death of the animal, or else after lying on ice.

In starting the experiment, the T tube is raised so that no fluid flows from the Mariotte bottle; the screwclamp (S) on the upper limb is opened, and the lung is attached to the lower limb. The T is then gradually lowered so that just enough liquid flows into the bronchi to distend the lung approximately to its normal size.

In previous determinations of the refractory period of the mammalian heart an artificial rhythm has been established by means of a series of break induction shocks at a rate necessarily more rapid than that of the normal sinus rhythm. This rhythm has then been interrupted by break shocks timed by some mechanical device to fall at various points in diastole. Such measurements have, therefor, yielded the refractory period of an artificially induced excitation.

The authors have devised an apparatus utilizing the action current after amplification with which it is possible to interrupt the sinus rhythm at any interval following a normal excitation. The amplifier is designed to be connected to the terminals of the galvanometer string without affecting the performance of the latter. The electric impulse to be amplified is on the order of 1 millivolt for 1/50 second, and the current supplied to the amplifier must be negligible as compared with about 1 microampere, the galvanometer current. A 5-stage, resistance-coupled amplifier is used, with 4 UX-240 tubes and 1 UX-112A. The string terminals are connected directly in the first grid circuit. The 4 coupling condensers are large, so as to give the entire amplifier a discharging time constant of 1/6 second. The first 2 plate circuits have condensers in parallel, so as to obtain a charging time constant of 1/50 second. In this way, the amplifier is rendered free from impulses of much longer or shorter duration. A 10,000-ohm magnet carries the plate current of the last tube, normally less than 1/2 milliampere because of a large negative grid voltage. An impulse of 1/2 millivolt for 1/50 second, applied to the amplifier, is amplified to over 20 volts for 1/10 second at the magnet terminals, about 200,000 times the applied impulse.

Although chicory is widely used as a beverage ingredient, very little information as to its action is available in the literature. The popular belief, perhaps founded on statements by those commercially interested, that it serves as a mild laxative, suggested testing its effects on isolated intestine. It has been shown that coffee infusions increase the activity of isolated intestinal segments 1 to a greater extent than could be explained by their caffeine content. 2 Kymographic tracings were made from segments of rabbit duodenum in the usual fashion. The preparation tested was a fresh 10% infusion of ground and roasted chicory, made in Tyrode's solution. In numerous experiments we have observed no indication of stimulation of either tone or amplitude of contraction, but with concentrations greater than would be present in the gastro-intestinal tract of individuals drinking chicory-coffee infusions there was uniformly a more or less marked depressant effect.

Sherman and Campbell 1 have shown that in a simplified ration composed of whole wheat and whole milk powder marked improvement in growth, reproduction, lactation and general vigor resulted from increasing the milk from 1/6 to 1/3 of the total weight.

We have fed albino rats simplified rations patterned after diets suitable for children, including milk, cereal food (represented by white flour), an assortment of vegetables, a butter substitute (equaling butter in vitamin A content) and meat residue. The composition of these diets is shown in Table I.

On each diet a 5th generation has been secured, but the present report covers only 3 generations; on each diet approximately 40 animals have been carried to the age of 1 year, and on each over 1000 young have been born. As many young as survived have been reared to the age of 28 days. The 3 diets with varying proportions of vegetables but no egg, supported excellent growth, fair reproduction, but very poor lactation. Growth and fertility were slightly better in the 2nd and 3rd generation than in the 1st, but the percentage of young reared decreased with each successive generation, owing to poor lactation.

On the 4th diet, in which egg was used, there was better growth of the young, earlier maturity of the mothers, higher fertility, and better lactation. These findings are summarized in Table II.

These experiments show that a mixed diet already adequate but not containing egg can be improved by the addition of egg more than by an increase in the amount of vegetables, and suggest special supplementing values for the egg, which require further investigation.

Through the kindness of Dr. H. C. Bingham and Professor R. M. Yerkes an opportunity was presented to examine for parasites 4 chimpanzees maintained by the Institute of Psychology at Yale University. These animals came from various parts of Africa and have been at the laboratory of the Institute since September 12, 1925, before which time 2 of them were for 3 months in a zoo in Philadelphia.

Since March, 1926, occasional samples of stools from the chimpanzees have been examined, always with the result of finding in large numbers motile forms of a Balantidium morphologically indistinguishable from Bal. coli. Cysts occurred only very rarely. Although these ciliates have swarmed in the intestines of the animals, at least since examinations first began, and have been passed in great numbers, presumably in every stool, the chimpanzees are healthy and show no visible signs of suffering from their infection. The stools have been more often soft than well formed, and have occasionally been diarrhoeic, but consistency has varied with the animals’ diet. In stools kept at room temperature the ciliates have remained alive for several hours, some as long as 32 hours after leaving the host.

Balantidia in anthropoids have been studied by Brooks, 1 Noc, 2 Brumpt, 3 Joyeaux, 4 Christeller, 5 Hegner and Holmes, 6 Ziemann 7 and Cunha and Muniz. 8 As a result of their observations it appears that the ciliates may parasitize orangutans, macaques, capuchins, baboons and chimpanzees; and that the animals are in most cases carriers, not apparently suffering from the presence of the parasites, but may in some cases develop ulcerative dysentery. Dobell and O'Connor state that these animals react to Balantidium in much the same way as does man and Hegner 9 says “it has long been known that the organism is pathogenic to primates.” Most of the authors have supposed that the species with which they dealt was Bal. coli.

Hemolytic strains of staphylococcus producing a varying amount of yellow pigment are found in pyogenic processes, in septicemias, and also in non-suppurative types of dermatitis and on apparently normal skin. The majority of strains from these varied sources cross-agglutinate in monovalent serums. In an attempt to differentiate them by the method of agglutinin absorption an interesting relationship has been brought out which is in accord with the hypothesis of a multiple antigenic composition for such organisms.

Agglutinating sera were made in rabbits and reciprocal agglutinating absorptions carried out. To obtain constant results it was found necessary to use fresh suspensions of standard strength and even so, specific reactions seemed obtainable only in narrow quantitative limits. If too heavy suspensions were used absorption was obtained which was apparently non-specific. This was due probably to adsorption of the anti-bodies by the bacteria, similar to that which can be accomplished by thick finely divided insoluble suspension of many substances. Tentative grouping could be obtained by testing the varied serums in an optimum dilution, but for clear cut results the following technique was found most satisfactory: A series of dilutions of the serum from 1-10 to 1-12,800 was prepared and each dilution absorbed with a standard dose of fresh cocci. After absorption and centrifugation the supernatant fluid was divided into 2 parts and tested for agglutination; one part against the strain used for absorption and the other against the strain used for production of the serum. With completely analogous strains complete absorption appeared at the same dilution, that is, if all the antibodies for the absorbing strain were removed those for the immunizing strain were also removed. Such a reaction, however, was not always reciprocal and a given strain often absorbed all the agglutinin from the serum from another strain when a suspension from this second strain would not remove all the agglutinin from a serum prepared against the first. In reciprocal tests with three strains A, B and C, it was found that:

A study of the immunologic reactions with a tobacco mosaic virus to an antiserum from rabbits has been undertaken with the hope of throwing more light upon the nature of the virus.

Antisera were produced separately by intravenous injections of rabbits with sap extracted from healthy Turkish tobacco plants, and sap from plants affected with tobacco mosaic virus. In several series of alexin-fixation experiments, each antiserum was titrated with its homologous and heterologous antigen. The higher titer for a given antiserum was obtained with the homologous antigen. These results are in agreement with Dvorak, 1 who compared the precipitin reactions of an antiserum to healthy potato sap with those of an antiserum to sap extracted from mosaic diseased potato plants.

Precipitin-absorption tests further demonstrated the presence of a highly specific precipitin in the antiserum to virus-sap.

The preventive action of specific antiserum upon virus-sap in vitro was demonstrated by the inability of antiserum-virus mixtures, that had stood at room temperature for several hours, to induce mosaic disease when inoculated into healthy tobacco plants. Under the same set of conditions, normal rabbit serum and antiserum to healthy tobacco sap did not exhibit the preventive action upon virus that the specific antiserum showed. Fresh antiserum to virus-sap exerted a stronger inhibiting action than the same antiserum, inactivated by a temperature of 56° C. for one-half hour. Heated antiserum could be reactivated by the addition of fresh rabbit or guinea pig serum, so that the reactivated antiserum produced the same preventive action upon virus as the active antiserum.

In a former paper 1 it was shown that when the marine alga, Valonia, was placed in solutions of methylene blue dissolved in sea water at pH's 5.8 and 9.0, the dye penetrating into the sap was methylene blue itself and not some lower homolog, such as trimethyl thionin, formed by the oxidation of methylene blue in the outside solution. The identification of the dye was made by means of the spectrophotometer.

In order to test this conclusion still further, another set of experiments was made using the fresh water alga, Nitella. The method was similar to that used in the experiments with Valonia. The same sample of dye was also used. All necessary precautions were taken during extraction to keep the sap from becoming contaminated by dye from the rest of the cell. The plants were also tested for injury by transferring samples from the experimental solutions to tap water. These were found to be alive (green and turgid) 2 months after experimentation, when they were discarded.

In the experiments herewith reported selected Nitella cells of the same size were placed in solutions of methylene blue for 2 hours at 25° C., after which the sap was extracted, placed in a hollow ground slide with a flat-bottomed depression, covered by a cover slip, and spectrophotometric measurements made.

It was found that the dye penetrating into the sap had a maximum absorption at 660 mμ and a secondary maximum at 610 mμ. These are the absorption maxima for methylene blue. It is therefore concluded that methylene blue penetrates Nitella sap as such. This corroborates the previous results obtained by the writer with Valonia.

In previous papers 1 , 2 it was shown that horse proteins injected intravenously into normal dogs are retained quantitatively in the canine circulation for at least 6 days. During this retention, however, the proteins undergo marked denaturization. By the end of 4 days they will call forth no recognizable anaphylactic reaction even on massive transfusion into hypersensitive dogs. By the end of 6 days they cease to act as sensitizing agents for normal dogs. We have endeavored to follow subsequent changes in these proteins.

Titration by means of a specific horse-protein-precipitating rabbit serum indicate that the proteins are further altered, denaturized or inhibited. By the end of 30 days they give little or no demonstrable reaction with rabbit precipitin, if tested in low dilutions.

That this negative reaction is not due to an absence of horse proteins from the 30-day blood sample is indicated by continuing the titration to higher dilutions (Table I). In dilutions above 1:1000, the 6-day and 30-day samples give practically identical precipitates and turbidities.

The horse proteins thus further altered are apparently retained quantitatively for at least 4 months (latest titration thus far made). Within the limits of the experimental error the 4-month and 30-day blood samples give identical precipitates in all dilutions.

The measurement of blood flow throughout the cardiac cycle has been accomplished by use of instruments of rather low efficiency due to their heavy construction and the use of more sensitive instruments has been restricted by the difficulties attending their use. We have attempted to construct a simple instrument for use in studying reflux. The lever of the aortic haemodromograph is of fine aluminum wire, the fulcrum a water-tight strip of rubber dam fixed between 2 metal plates. An oval aluminum vane, 6×8 mm., fixed at one end of the lever projects into the aorta at right angles to the direction of flow; the connection is made air-tight by an oval shoe and threaded washer (Fig. 1). The length from vane to fulcrum is 2 cm., from fulcrum to the strand of rubber used to damp the lever is 5 cm., and a fine brass wire projects 3 cm. above this, its shadow registering on the moving film. The haemodromograph for the carotid is incorporated in the side of a brass tube, 4.5 mm. in diameter, which is introduced between cut ends of the vessel. The lever is damped 5 mm. from its fulcrum, the vane extends 4 mm. into the blood stream and the fine wire extends 3.5 cm. from the fulcrum. The calibration of the apparatus, by measuring deflection produced by known rates of constant flow, is theoretically simple and accurate. In practice with our rather crude instruments we have found such calibration misleading, as the deflection is not constant from day to day, and small wisps of fibrin deposited on the vane produce a marked deviation. At present it can be used only for studying the shape of the curve of rate of flow. The instruments used have a vibration frequency of 50-80 per second; decrement per beat is from 20-30%. The sensitivity and decrement can be varied by varying the tension of the elastic strand. The relationship between deflection of the lever and rate of flow is not linear. With increasing velocity the sensibility of the instrument for a given increment of speed is increased.

Recent results obtained by Smith and Engle 1 by daily hypophyseal homeo and heterotransplants in mice and rats led to an investigation of the effect of daily injections of fresh rabbit pituitary gland on an hypophysectomized puppy. Aschner 2 published striking photographs of hypophysectomized puppies of various ages which had remained infantile in appearance and size until sacrificed.

Evans and Long 3 found that fresh anterior lobe fluid given intraperitoneally to immature rats produced gigantism. Later, this extract was found to stimulate hypophysectomized rats to normal growth but failed to affect their sex glands. Smith, 4 using fresh pituitary glands of rats, was able completely or in large measure to cure the disabilities arising from hypophysectomy in immature rats. Smith and Engle 5 were able to induce precocious sexual maturity both in immature and in hypophysectomized rats, in 2 animals within 36 hours, by these fresh daily transplants. By means of the intracranial approach of Dandy and Reichert 6 a total hypophysectomy was performed on a 6 weeks old female puppy and the base of the brain in the infundibular region cauterized by heat. This animal was constantly kept with her control litter mate, also female. After months of observation, she remained infantile, failed to grow, retained the puppy face and hair, failed to erupt permanent teeth and exhibited infantile genitalia, indicative that a total hypophysectomy had been performed.

Six months after hypophysectomy when 7 1/2 months old, weighing 4.2 kilos and when the control had attained adult size and weighed 9.3 kilos, fresh whole rabbit's hypophysis was given subcutaneously every day by the technic described by Smith and Engle. 1 Within 48 hours the vulva were definitely swollen and within 3 days became so greatly distended as to protrude 3 cm. from the body.

Starch, as it occurs in nature, is not soluble in cold water, and hence one cannot determine the specific rotation of the untreated substance. The several soluble starches, which have formed the basis of numerous optical studies, represent natural starches altered either by heating or treatment with reagents, or by treatment with enzymes, or by some combination of the above procedures.

The reported values of the specific rotation of the various starches range from 216.0° to 182.0°. This wide range of results indicates that the determinations were made on starches hydrolyzed to a varying extent, or on untreated solutions so opalescent as to prevent accurate evaluation of the specific rotation.

It was shown by Alsberg, Griffing, and Field 1 that clear solutions could be prepared from purified starch by dry-grinding in a pebblemill, dispersion in cold water, and centrifugation. This technique is particularly effective with wheat and corn starches. These solutions are water-clear, dialyze through collodion, and give an intense blue with iodine. Such solutions are always very dilute. 2 , 3

The specific rotation of these solutions has been determined. The source of light employed was an electric lamp of the type described in Circular 44, U. S. Bureau of Standards, page 34. Determinations of the concentration of the starch solutions were made by the method of Munson and Walker.

The average of a number of determinations gave 196.0° for the specific rotation of the cold-water-soluble portion of ground corn starch, and 194.7° for wheat starch, with the light-source specified. This difference is not significant in view of the fact that it lies in the range of analytical error in determining the concentrations of such dilute solutions. These determinations afford no evidence that the cold-water-soluble fractions of ground corn and wheat starches differ in character.

A new species of Euplotes, a marine hypotrichous ciliate, found in San Francisco Bay can be readily induced to encyst and excyst at will by the following method.

By means of a micro-mouth pipette, from 25 to 35 organisms picked at random from a thriving culture are placed on a slide or coverslip within a drop of the culture medium. The slide (or coverslip) is then placed in a moist chamber to allow gradual evaporation to take place. The rate of evaporation under these conditions at about 22° C. gives the best results. Within 1 to 10 hours, or even less time, encystment occurs. This process includes a gradual decrease in motility of the organisms until they become fixed to the slide, whereupon they gradually assume a rounded form. External organelles at first can be recognized within the cyst wall but later disappear from view. Excystment takes place when a small drop of fresh culture medium, tap, or distilled water is added to the drop containing the cysts. The time required for complete excystment depends upon the age of the cyst and varies from a few minutes to 1 or 2 hours. A contractile vacuole reappears and functions slowly soon after the addition of water. The cyst wall breaks, usually just above the contractile vacuole. The contents stream to the outside, external organelles make their appearance, and soon the organism again becomes free-swimming.

The above method has proved to be an excellent means for studying the structural changes which occur during cystment. The cov-erslip is inverted upon a depression slide, the edges sealed with vaseline, and now the slide may be placed under a microscope. There the processes of encystment and excystment can be studied with any magnification, including oil immersion.

Recently we 1 have shown that regeneration of the peridental membrane takes place in the guinea pig and rabbit following artificial destruction of this tissue. The sequence of events leading up to this repair was outlined. Injury had been induced by passing a dental bur through the alveolar process across the peridental membrane and into the upper incisor tooth on its lingual surface. Thus, the peridental fibres were severed at this point. The rodent was chosen for that work since it has continuously growing teeth with the result that uninjured tooth surface will move forward to a position opposite to the site of injury to the alveolar bone. Thus repair may be followed more easily by microscopic technique.

A similar series of experiments then was performed upon cats. This was done in order to learn whether a like sequence of repair may be demonstrated in an experimental animal still higher in scale and in which the tooth is stationary. The animals were 4 or 5 months old and had received the usual animal house diet. The operation was performed by Dr. F. V. Simonton of the College of Dentistry of this university and his cooperation is thus acknowledged.

Following the operation, the animals were killed with chloroform as follows: 10 days, and 7, 8, 13, 17, and 20 weeks. The experimental tissues were then dissected out and sections cut according to the method outlined by one of us (A. W.). 2

Examination of the sections made shows a picture closely similar to that found in rodent tissue. In certain of our animals killed at the close of the 7 and 8 week periods, it was evident that infection had developed. This had resulted in destruction of tissue adjacent to the site of the cut and both hard and soft tissues had been involved.

In a previous report 1 it was stated that typical ovario-testis could be produced in adult male rats by grafting the ovaries of immature females into the testis, subsequently giving daily transplants of the anterior lobe, using Smith's technique. 2

Eight to 12 days of treatment is sufficient to produce numerous ovarian follicles of mature condition in the ovarian graft, but no changes have been observed in the host animal. If the daily transplants are continued for 18 days, the ovarian mass engrafted in the testis is greatly increased. The large follicles are more numerous, and in many cases, the thecal cells have hypertrophied, forming a layer of lutein cells around the periphery of the follicle.

As shown in Fig. 2, the response in the tissue of the mammary gland of the treated male rat bearing an ovarian graft is markedly different from that of its litter-mate (Fig. 1), which had received the same number of daily transplants of anterior lobe, but which did not bear an ovarian graft.

These rats possess normal seminiferous tubules, and have mated with normal females during the experimental period. All animals bearing grafts have sired 1 litter, many 2 litters, and a few 3 litters, during the period of the experiment.

In view of the finding that skin can readily be transplanted in young chicks 1 it seemed desirable to see if a similar procedure would be equally feasible in other commonly used laboratory birds. To this end a few experiments were made with newly hatched ducks and pigeons.

Skin from 2 young mallards supplied grafts for 6 white Pekin ducklings. Of these, 1 graft to the back and 1 to the breast were entirely successful, 2 to the top of the head and 1 to the back failed, and 1 duckling was lost before the fate of its graft could be determined. Mallard skin grown on a Pekin host has produced juvenile mallard plumage.

Through the kindness of Professor C. S. Stoltenberg 20 pigeons were available for the second experiment. Four autoplastic grafts were all successful. Of the exchange grafts between full brothers and sisters, 3 became fully established and 7 others lived for 9, 11, 12, 15, 26, 30+ and 30+ days. When donor and host were not closely related, the grafts lived for 7, 7, 9, 12, 17 and 25+ days.

Because of the rapid growth of the young and the ease with which the condition of the graft may be determined from day to day without the necessity of sacrificing the host, the pigeon seems especially favorable for the study of graded compatibilities of tissues. The data thus far available indicate a sharper differentiation of individuality factors in the pigeon than in the fowl. There is evidence that the grafts which are doomed to fail may nevertheless have varying life expectancies depending upon the part of the body to which they are transplanted.

The observations described in the following preliminary report were made by subjecting animal cells to the action of sound waves produced by a rapidly vibrating quartz crystal. The oscillating current was generated by a 250 watt tube connected in a modified Hartley circuit. The crystal lay between a heavy lower electrode and a light upper electrode of copper gauze in a crystallizing dish containing the dielectric liquid, usually xylene. The crystal vibrated at the rate of approximately 750,000 times per second.

Effects in Vitro. In this type of experiment the animals were placed in a test tube or similar vessel containing water and dipped into the dielectric over the vibrating crystal. Free-swimming Triturus larvae are rapidly killed, the gill filaments being torn away and areas of denuded surfaces exposed. Confirming the findings of Wood and Loomis, 1 the larger protozoa are killed by this method, while the smaller protozoa, being able to swim between the nodes of the stationary wave system set up in the water, escape unharmed. Spirostema, being relatively large, are quickly killed, Paramecia less rapidly, while the smaller protozoa are unaffected after many minutes. By means of a simple arrangement it was possible to observe the cells through the microscope while subjecting them to the sound waves. A drop containing the protozoa was placed on a microscopic slide and a cover glass laid over it. The slide was then laid across the oscillator dish. The sound waves were transmitted in this manner to the cells through the medium of the oscillator dish and the slide. Neglecting morphological details for the present, cell destruction is due to an actual disruption of the whole or of a part of the cell, to a gelation or coagulation of the protoplasm, and to a combination of the two effects.

The effect of liver extract in pernicious anemia may be analyzed from the point of view of the immediate reaction, as well as of the sustained reaction. The immediate reaction includes among other things 3 distinct morphological features: (1) the normoblastic response, (2) the reticulocyte response, and (3) the liberation of the mature erythrocytes. 1 It is a well established fact that the bone marrow in pernicious anemia is hyperplastic and contains large amounts of erythrocytes in various stages of development. In other words, there appears to be a difficulty in the maturation of the erythrocytes necessary for their discharge from the marrow into the circulation. This faulty maturation is in character probably morphological rather than chemical, for the cells in the marrow are completely filled with hemoglobin and products of hemoglobin have been deposited in the various tissues. Thus the hemosiderosis, in the classical interpretation of pernicious anemia looked upon as a proof of the hemolytic character of the disease, might indicate a destruction of erythrocytes in the foci where they were formed before they were ever delivered into the circulating blood. When liver therapy is instituted, the bone marrow presumably is stimulated to mature and release the cells. A slight retardation in the maturation as compared with the discharge thus explains the 3 features of the immediate reaction. Experiments on the effect of liver extract upon normal individuals are here presented as a confirmation of the view of the mode of action of the liver extract to accelerated the discharge of cells into the circulation.

Six normal individuals (doctors and assistants in our laboratory) whose erythrocyte counts were between 4,500,000 and 5,800,000 were given 3 vials of liver extract (343 Lilly), or the equivalent of 300 gm. of raw liver, daily for a period of 10 days.