Experimental Biology and Medicine

For some years we have been making studies with Cladocera on the occurrence of mutations and their inheritance in parthenogenesis. Four mutant characters have been studied in Daphnia longispina. The inheritance is complete, since every parthenogenetic young from a mutant individual genetically possesses the mutant character (except, of course, for the occasional interference of another mutation). This is in accord with expectation, in view of the absence of chromatic reduction and segregation in the maturation of the parthenogenetic (diploid) eggs in Cladocera.

It was of further interest to obtain, if possible, sexual reproduction and hybrid offspring between mutant clones and the “wild” type. Sexual eggs (haploid and requiring fertilization) are frequently obtainable, and males may usually be obtained when desired, but it is not easy to obtain, simultaneously, ovarian sexual eggs and functional males from just the stocks desired. When sexual eggs and active males are obtained at the same time, fertilization frequently does not occur, and the fertilized eggs when obtained give exceedingly poor hatches, so that many fruitless attempts preceded the obtaining of the desired hybrids.

Forty hybrids have been or are now being studied and some data obtained on the inheritance of the two mutant characters involved in the stock used in the crosses. The characters involved are sex-intergradedness and “excavated” head. Twenty-nine of these hybrids are from wild type × mutant , and 11 are from the reciprocal cross. Nineteen of these hybrids genetically possess the mutant character, excavated head, and 19 show sex intergradedness. It seems probable, therefore, that each of these characters is a dominant mendelian character and is heterozygous in the mutant clones used in the crosses. Further breeding and study will be necessary before the full extent of the inheritance of these characters is determined.

Recently, in the course of some experiments on the viability of monocytes, 1 it became necessary to preserve, for further study, cells which had been vitally stained with neutral red. The cells consisted of various types of white blood cells present in the peritoneal exudate of the guinea pig. These cells had been under observation in hanging drop preparations such as are commonly used in tissue culture work, 2 and since the usual methods of preservation seriously affected or entirely removed the vital stain, it was necessary to apply a new method. The one finally devised, as described below, not only retains the vital dye but permits other staining methods, such as Wright's method for differentiating blood cells, to be superimposed upon the vitally-stained cells.

In brief, the method consists in taking the cover glass from a hanging drop preparation in which the cells have been well stained vitally, either by direct contact with neutral red on the cover slip, 3 or by exposure to a dye bath consisting of a few crystals of neutral red dissolved in unbuffered Ringer's solution in the incubator at 38° to 40°C, and immersing the preparation for one half to one hour at 38° to 40° C. in the Ringer-formalin fixative, suggested by Fischer, 4 which is buffered as noted below. The tissue is then washed in running water for three hours, rinsed with distilled water and permitted to dry. After drying, the preparation may be counterstained with a one-tenth of one per cent aqueous solution of methyl green, as a nuclear stain, or with Wright's blood stain. The preparation is again dried and may then be mounted in damar, or, if necessary, it may be cleared according to Fischer's 4 method, by subjecting it, before mounting, to passage for two minutes each through a graded series of acetone-xylol mixtures consisting of 5 per cent, 30 per cent, 70 per cent, xylol, and two changes of pure xylol.

During the examination of some old stock cultures in November, 1924, there was found an agar slant of Bacillus anthracis which had the appearance of undergoing lysis by bacteriophage. The film of growth was interrupted by well defined partly confluent denuded areas, many of them containing small centrally located secondary colonies. Transplants of the culture to agar gave similar appearances after incubation for two days at 37° C. and one or more days at room temperature. Transplants to broth did not show clearing but the plaques again appeared when the broth cultures were transplanted to agar. An older agar slant culture from which the culture first mentioned above had been inoculated did not show plaques nor did fresh transplants of this culture, but when this “negative” strain was inoculated with material from one of the plaques of the “positive” strain plaques appeared in subsequent transplants. Filtrates of broth cultures of the positive strain were inactive when added to the negative strain.

Plaques on agar plates examined under the microscope were found to contain a nucleus of free spores, the remainder to the plaque being made up of pale remnants of bacilli and amorphous granular detritus. The matrix of growth surrounding the plaques was composed of typically curled but apparently sporefree anthrax filaments. Fishings from the centers of plaques yielded cultures of typical sporulating Bacillus anthracis. Fishings from the matrix yielded cultures of non-sporulating bacilli otherwise like B. anthracis, hereafter referred to as the nonsporulating strain.

After prolonged growth on agar, on potato or in broth at 37°C. and at 20°C. the non-sporulating strain failed to produae visible spores and did not resist heating for twenty minutes at 75°C. Cultures of this sporulating strain resisted similar exposure to heat.

Pure strains of fibroblasts or of epithelial cells increase in mass in an unlimited manner, when they are cultivated in plasma and embryonic tissue juice. For 14 years, colonies of a strain of fibroblasts have doubled in size every 48 hours in such a medium. Pavement and thyroid epithelium also have been found to manufacture unlimited amounts of protoplasm from the constituents of embryonic juice. Neither epithelial cells nor fibroblasts multiply in serum proteins, egg albumin, crystallized egg albumin, amino acids from embryonic juice, or artificial mxtures of amino acids for a longer time than in Tyrode solution. So far, embryonic juice is the only material which has been found to maintain epithelial cells and fibroblasts in a conditoti of true cultivation.

Investigation of the chemical nature of the nutritive materials in the embryo juice has led to the conclusion that the nitrogenous substance utilized by the tissues is the protein itself. The amino acids and other ultra-filtrable and dialyzable constituents slightly stimulated the migration and multiplication of the cells, but failed to produce an increase in the mass of the tissues. Since the protein of the embryo juice is utilized by the cells, it seems evident that it must be hydrolyzed and some intermediate product absorbed. Complete digestion by trypsin and pepsin produced only toxic substances.

However, some of the higher cleavage products formed by the partial hydrolysis of protein have been found to produce the same effect as the protein of the embryo juice, causing continuous multiplication of cells and increase in the mass of tissue for long periods of time. In fact, some preparations of these protein hydrolytic products have given a larger increase in the mass of the tissue than has ever been obtained by embryo juice.

The structure of thyroxin as recently determined by Harrington 1 represents the substance as a derivative of di-iodotyrosine and as containing two benzene rings. It is, therefore, of some interest to know what effect, if any, other derivatives of di-iodotyrosine containing two aromatic nuclei, would exert on basal metabolism.

Two such substances have been examined, namely, di-iodotyrosyl-di-iodotyrosine and the cyclic anhydride (diketopiperazine) of di-iodotyrosine. The former substance may be prepared by the action of iodine in feebly alkaline solution upon tyrosyl-tyrosine, or by the regulated action of alkali on the cyclic anhydride of di-iodotyrosine. The latter substances cannot be obtained by the direct action of iodine upon tyrosine anhydride, but may be readily attained by the action of iodine chloride in acetic acid upon the anhydride. Both substances are cream colored microcrystallin compounds, which in their solubilities and general chemical deportment, colour reactions, etc., show considerable resemblance to thyroxin. The anhydride of di-iodotyrosine melts at about 245 to 250 degrees (uncorr.) while di-iodotyrosyl-di-iodotyrosine does not melt sharply but begins to decompose above 200 degrees. The substances were prepared for these experiments by H. D. Dakin.

The metabolism experiments were carried out on a normal male individual of about 101 kilos, 184.5 cm. in height. All determinations were done in triplicate under basal conditions, either the Tissot or Benedict machine being used. All Substances investigated were taken intravenously.

Basal metabolic rates determined on five days during the week preceding the injection of the cyclic anhydride of di-iodotyrosine varied between −9 and −12 per cent. Twenty-four hours after the injection of 21 mg. of the substance the basal rate was −9 per cent, and at the end of 48 hours was −7 per cent.

The physiological action of wheat bran and a number of products isolated from bran was studied on dogs, which received as a basal ration a “synthetic” diet of casein, sucrose, and lard, together with the necessary mineral salts and vitamins. The purpose of the investigation was to determine what constituents of bran are responsible for its laxative effect. After a control period of 8 days the dogs were given the basal ration, supplemented with the bran or other material, the laxative power of which was to be determined. The frequency of defecation and the total weight of air-dried feces per eight-day period served as criteria of laxation.

Washed bran (starch-free) was laxative when ingested in amounts ranging from 10 per cent to 0.5 per cent by weight of the food intake. The average minimum effective dose was about 2 per cent (200 to 275 milligrams per kilo, body weight). About two grams of additional water was excreted in the feces for every gram of bran eaten.

The crude fiber of bran was a much more potent laxative than bran itself. The fiber when fed in such small daily doses as 0.5 gram to dogs weighing 7 to 9 kilos caused a marked rise in the number of defecations per eight-day period and produced an increase in the fecal output equal on the average to three times and, in some instances, to five or six times the weight of the ingested fiber. In general, bran itself did not possess this fecesforming power. Extensive experiments on 10 animals indicate that the laxative action of wheat bran is due at least in part to its crude fiber content.

The laxative factor of bran was not removed by acid and alkali hydrolysis or by prolonged, successive treatment with water, sodium chloride solution, hot 70 per cent alcohol, and weak sodium hydroxide solution.

None of the ordinary constituents of body fluids interferes with the test outlined below, save a substance in the urine, as yet unidentified, but which is probably a small amount of oxalic and perhaps glyoxalic acids, which have been reported as occuring freely in urine by Fürbinger 1 and by Granström 2 , respectively. Creatinin, creatin, uric acid, etc. are negative. Allantoin, however, may be separated from the other constituents of the urine by precipitation with ammonical silver nitrate. Then the allantoin is subjected to the test. This procedure is necessary only in the case of urine for blood and other body-fluids show no interfering substance.

To 25 mg. filtered urine, add, if necessary, 10 per cent hydrochloric acid to slight acidity; if the urine is initially acid, this is unnecessary. Then add, drop by drop, 1 per cent aqueous solution silver nitrate until no more precipitate of silver urate is obtained; ten drops are usually sufficient. Let stand five minutes, filter, saving the filtrate. Or, one may use a 50 mg. centrifuge tube and precipitate by means of the centrifuge.

A test for the presence of uric acid on this filtrate may be made by Folin's test. Then add, drop by drop, concentrated ammonium hydroxid to the filtrate to obtain the precipitate of allantoin silver nitrate; in case, after the addition of two drops of the ammonia, no precipitate appears, add a drop lor two of silver nitrate solution as before, to insure that all of the silver nitrate first used was not carried down as silver urate. An excess of ammonia causes solution of the precipitate.

A precipitate, varying with the amounts of allantoin, may be obtained by centrifuging. Procedure for all fluids: To a few mls. of the fluid, or to the precipitate obtained above after decantation of the supernatant fluid, add one small flake of indol, a commercial product.

There is no difference of opinion that only young and actively growing bacteria are subject to transmissible lysis. Old or dead bacteria do not undergo lysis and do not contribute to the increase in concentration of the active lytic substance, but rather, on the contrary, bring about a measurable reduction of it.

Recently, Twort 1 has stated (somewhat similar to the evidence offered a year earlier by Gratia 2 , 3 ) that dead staphylococcus may undergo lysis if, in addition to a suitable bacteriophage, there is also present live staphylococcus. We have confirmed these observations, and in addition have endeavored to ascertain the explanation of the mechanism of this phenomenon. We found that in order to elicit the phenomenon it is necessary to control the numbers of live and dead bacteria in the mixture. An excess of dead bacteria interferes with lysis by adsorbing the active lytic substance before it has the opportunity to initiate lysis of the live bacteria, and thus all solution is prevented. The phenomenon is specific, that is, the lysis of live bacteria will initiate lysis of dead bacteria of the same species only. In its extent and completeness, lysis of dead bacteria occurs best with stapylococcus, the organism which easily undergoes autolysis under appropriate conditions. In the case of colon bacillus, which undergoes autolysis very slowly, the lysis of the dead bacteria is less regular and less complete.

If the experiment is performed so that a suitable semi-permeable membrane is interposed between the dead and live bacteria, dead bacteria are never dissolved, in spite of the lysis of living bacteria on the other side of the membrane. The active substance determining the lysis of dead bacteria is not diffusible, while the principle initiating the lysis of live bacteria diffuses freely and is demonstrably present on both sides of the membrane.

The fact has been observed that prior to lysis in the presence of the bacteriophage, bacteria usually undergo more or less marked swelling. The extent of swelling and the relative proportion of swollen bacteria, as well as the actual relation between swelling and lysis are difficult to establish by direct microscopic examination, because the swelling and lysis of bacteria go on simultaneously and continuously, and because the degree of swelling of individual bacteria varies to such an extent that results of such an analysis must, of necessity, be highly subjective. Moreover, in the case of cocci it is very difficult to decide whether or not swelling takes place at all.

It seemed, therefore, that since swelling of bacteria increases the relative volume occupied by the solids in the medium, the swelling of the total bacterial mass (with proper correction for growth) should bring about an alteration in viscosity of the solution.

Measurements were malde both by means of a capillary viscosimeter of Ostwald and in the torsion viscosimeter of du Noüy. It was found that in general the viscosity of the mixture of bacteria with a corresponding bacteriophage increases steadily up to the time when visible lysis sets in, at which time the viscosity begins to diminish until it gradually reaches the originallevel.

In general, the greater the relative concentration of phage, the sooner the maximum viscosity is attained. The greater the number of bacteria (within the limits compatible with the concentration of phage present) the greater is the percentage of change in viscosity. At its maximum, the increase in viscosity of the mixture has varied in the experiments thus far performed, between 14 and 50 per cent, depending on the relative concentration of bacteria and bacteriophage.

Since it has been established that exposure to sunlight or to ultra-violet rays from artificial sources is able to protect animals or infants from rickets, the question arises whether this protective quality can be transmitted through the milk by the mother to the young. As the result of some experiments on cows, Luce 1 concluded that “the anti-rachitic value depends on the diet of the cow and possibly also on the degree of illumination to which she is exposed.” More recently Steenbock, Hart and their associates 2 showed that the irradiation of a goat led to a decided increase in the antirachitic potency of their milk.

This is a question of importance in relation to the etiology of infantile rickets, for although this disorder occurs less frequently among nursing than among bottle-fed infants, nevertheless it has been found in from 1/3 to 1/2.of the breast-fed infants in the temperate zone. In the following experiments, undertaken to elucidate this question, rickets was primarily induced in a series of rats by means of the Sherman-Pappenheimer low phosphorus diet. After radiographs had shown that rickets was present, a ration of 25 cc. of human milk was substituted for the rickets-productive dietary. When the milk had been fed in this amounlt for a period of 9 days the animals were again radiographed, their blood analyzed for inorganic phosphorus, and the bones examined microscopically. It was thus found that this quota of milk failed to induce healing as evidenced by the radiologic as well as the microscopic picture. The inorganic phosphorus in the blood was very low-1.98 mg. per cent.

The woman was then irradiated every other day by means of a mercury-vapor lamp.

Growth of yeast cells in sugar solutions, in light and darkness was studied by measurements of CO₂ evolved. Uniform suspensions of Fleishman's baking yeast were made in solutions of glucose or more often of sucrose, and the growth in sunlight and in the dark at the same temperatures was studied in two sets of fermentation tubes. Ordinarily suspensions of 0.5 or 1 per cent by weight of yeast in a 5 per cent or 10 per cent solution of sugar was employed. It was found that fermentation took place more rapidly in the dark than in the light, at the same temperature. The effect of sodium benzoate on the growth of yeast and fermentation was studied by adding the drug in concentrations of 1:1500 to 1:500. It was found that sodium benzoate in concentration of 1:1000 produced practically no inhibition in the growth as compared with normal yeast suspensions when kept in dark- and that even concentration of 11500 of benzoate caused but very slight inhibition of growth in the dark. On the other hand when exposed to direct sunlight even 1:1000 and 1:1250 of sodium benzoate produced a distinct retardation of fermentation, and the degree of inhibition was very much greater than in control experiments with sunlight alone, without the benzoate. The inhibitory action of the sodium benzoate was increased a hundred and more per cent in the light. In diffuse sunlight, growth of yeast in benzoate was also less than in darkness at the same temperature.

Examination of the glass from the fermentaticon tubes by spectro-photography with a mercury vapor quartz lamp revealed that it was permeable to wave lengths as short as 3000 Angstrom units. By the use of suitable filters in connection with the above experiments ilt was found that it was the shorter rays of the sun, that is, those which were cut out by amber and brown colored filters which were respnsible for the above photosensitizing effect on sodium benzoate.

The influence of a number of fluorescein derivatives on the growth of yeast in sunlight and in the dark was studied, in a manner similar to that described in the preceding communication. A 0.5 per cent or 1 per cent suspension of yeast in 5 per cent or 10 per cent of cane sugar was generally employed and the quantity of CO₂ evolved was measured. To such suspensions of yeast in sugar solutions, the following fluorescent compounds were added in concentrations varying from 1:50,000 to 1:250,000.

The sodium salt of fluorescein itself; the potassium salt of tetra-brom-fluorescein (eosira); the potassium salt of di-bromfluorescein; the sodium salt of tetra-iodo-fluorescein (erythrosin); the potassium salt of tetra-chlor-fluorescein, the chlorine being introduced into the phthalic residue; the potassium salt of a chlorinated fluorescein with the chlorine introduced into the resorcin component; and the potassium salt of sulphone fluorescein. These compounds were prepared at the request of the author by Dr. E. White through the courtesy of the research laboratories of Hynson, Westcott, and Dunning Co.

It was found that in concentrations 1:100,000, none of the dyes produced any appreciable effect in yeast fermentation in the dark. In stronger concentrations (1:50,000) eosin produced a slight inhibition. When exposed to direct sunlight, in concentrations of 1:50,000 all of the dyes produced a slight inhibition of the yeast. When however, the same dyes were aidded to yeast suspensions in direct sunlight together with small quantities of sodium benzoate a remarkable synergistic effect was noted. The inhibitory power of the combinations on the fermentation power of the yeast was pttentiated two hundred and even more per cent, as compared with either dye or sodium benzoate alone. Thus solutions of eosin 1:100,000 plus sodium benzoate 1:1000, in direct sunlight produced an inhibition two or three times as great as that produced by either eosin or soaium benzoate alone.

The results of recent work 1 on the possibility of producing anti-bodies by means of substances apparently belonging to the class of lipoids have stimulated renewed investigation on the cause of the production of the Wassermann reagins.

Several main hypotheses relative to this subject have been considered. One of these implies that the reagins are no antibodies at all; others suppose that they are antibodies for spirochetes with an affinity also to lipoids of common origin; still another assumes that antibody formation is brought about by lipoids of the infected organism.

In analogy to the experiments on the production of hetero-genetic antibodies by mixtures of proteins and alcoholic extract of organs, 1 Sachs and his coworkers 2 thought of the possibility that the production of antibodies is due to a combined action of components of the spirochetes with lipoids derived from the infected animals. As a support of this hypothesis they report experiments in which it is shown that Wassermann positive sera can be produced in rabbits by the injection of alcoholic extracts of rabbit organs along with diluted pig serum.

Among our own experiments dealing with the question were such concerning the possibility of obtaining Wassermann positive sera by injections of dead trypanosomes (T. equiperdum). It had been shown previously that infection of rabbits with trypanosomes (T. equiperduwz) often results in the development of a positive Wassermann reaction. 3 Experiments by Klopstock in which he succeeded in eliciting Wassermann reactions by injecting dead Spirocheta pallida have been published very recently. 4

For the selection of rabbits the Sachs Georgi flocculation test was employed and only such whose sera gave completely negative reactions were taken for the experiments. One set of animals was treated with dead trypanosomes which had been kept over night in the icebox in a saline soluti'on containing 0.25 per cent phenol.

In experiments upon the cardio-vascular responses following repeated occlusions and restorations of the circulation of the head arteries in cats, it has been repeatedly observed that the spinal level of blood-pressure, that is, the level of pressure maintained by the activity of the spinal cord alone, when the functional activity of the medulla has been eliminated, may vary between 35 and 75 millimeters of mercury.

Certain criteria are necessary to determine whether blood-pressure within this range is actually spinal pressure, since in some animals, a pressure as low as 50 or 60 mm. is sufficient to restore medullary activity, whereas in other animals, a pressure to 70 to 75 mm. may be ineffectual in bringing this about. Blood-pressure, then, may be considered spinal when

(1) its level is unaffected by occlusion or release of the head arteries.

(2) no signs of any change in level of pressure or return of any medullary reflex is observable from half to three quarters of an hour after circulation has been restored to the head arteries.

With these conditions in mind, a series of experiments was carried out, with varying numbers of successive occlusions, from 1 to 14, in normal cats. In these animals, at some stage in the occlusion series, the head arteries were permanently ligated following an anaemic rise and fall of blood pressure. Blood pressure recorded on the chart in each case, is the mean pressure.

In these experiments, it was found that the height of the spinal pressure 'has a definite relationship to the number of occlusions which have been done on the cat, as the accompanying curve shows. That is, when the first or second occlusion is permanent, spinal pressure is about 70, whereas, when the 13th or 14th occlusion is permanent, the spinal pressure is about 20 mm.

The effect of repeated induced intravascular hemolysis upon the blood of rabbits was studied in the following manner: Ten to 12 cc. of sterile distilled water were injected slowly into the marginal ear veins of 6 rabbits at 48 hour intervals. The number of injections varied, the maximum being 34. Recently, as a control experiment, Patterson and Kast 1 described an anemia in rabbits produced by the intravenous injection of sterile distilled water, but as their experiments were only continued for 19 days, they apparently did not observe that the animals subsequently developed an increased resistance to such injections.

The following constituents of the blood were studied; erythrocytes, hemoglobin, leucocytes and reticulocytes. The resistance of the erythrocytes against hypotonic salt solution was also determined. The examinations were made immediately preceding each injection. Control observations were made upon normal rabbits and rabbits which had received physiological salt solution in doses corresponding to the distilled water.

After various intervals following the repeated injections of water, there was a diminution of 40 per cent to 50 per cent in the number of erythrocytes. However, despite continued repeated injections the erythrocyte count returned approximately to the original figures. There was a slight to moderate poly-chromatophilia and anisolcytosis with an occasional normoblast and hemacytoblast. The number of reticulocytes showed a slight increase; the highest figure was 7 per cent. The variations in the percentage of hemoglobin were less marked than in the number of erythrocytes. The total number and differential count of the leucocytes remained within normal limits. The resistance of the erythrocytes against hypotonic salt solution did not change beyond the limits of normal variations in rabbits.

During the past twenty-five years the work of many investigators has shown that the ovaries exercise control over the rapid temporary growth and hyperemia of the uterus which is characteristic of “heat” in mammals. Various extracts made from ovaries, and from other distinct tissues, have also long been known to bring about these particular uterine changes when injected into mammals (Marshall and Jolly, 1905; Lane-Claypon and Starling, 1906; Sonnenberg, 1907; Adler, 1911). The separation from the ovary of a lipoid fraction capable of inducing these uterine changes is also well established (Iscovesco, 1912; Hermann, 1913; Fellner, 1913, 1921; Seitz, Wintz and Fingerhut, 1914; Frank and Rosenbloom, 1915); and recently both the separation and the testing of this hormone have been more satisfactorily accomplished (Allen and Doisy, 1923; Frank and Gustavson, 1925). Still other hormones have probably been separated from the ovary (Iscovesco, 1912; Seitz, Wintz and Fingerhut, 1914; Papanicolaou, 1924), but the one considered here is the utero-stimulant” of Iscovesco, “das spezifische Ovarialsekret” of Fellner, the “follicular hormone” of Allen and Doisy, and the “female sex hormone” of Frank and Gustavson.

No previous study has been made of the action of this hormone when injected into a bird—nor into any animal other than a mammal. Allen, and co-workers 1 , however, obtained from fowl ovaries an extract which gave positive tests for the hormone when injected into rats. They were unable to obtain the hormone from egg-yolks. Fellner 2 extracted the hormone from the eggyolks of the fowl and from the ovaries of fish, and obtained the typical uterine reactions in tests made on rabbits. In our rather limited studies we have obtained positive results on oviducal enlargement (little or moderate) and hyperemia in 3 of 7 tests with immature (3.5 to 6.5 mo.) ring doves.

Brief reference is made to the more important earlier studies on the action of a utero-stimulating substance in mammals.

In a small group of tests of the utero-stimulating action of a placental extract, and of a substance prepared from the liquor folliculi of the sow by the Allen-Doisy method, we have obtained a few positive results-enlargement and hyperemia of the oviduct of virgin doves. This supplies the last necessary fact in proof of the lack of specificity, as between birds and mammals, of this substance.

Heavy dosage failed to affect a response in the virgin oviduct in some cases. Responses in the ovary and on sex behavior in virgin and mature doves have not been observed. An insufficient number of cases was studied but our results indicate a less ready, and a less varied, response of the bird to this substance than has been found for mammals.

A number of investigators have made studies on filtrable toxic products produced by bacteria in young cultures. These products have been variously called soluble toxins, × substances and endotoxins. In order to produce constant effects it has been necessary to inject the filtrates intravenously into rabbits and mice. The use of a small laboratory animal and a method of inoculation which would allow a somewhat greater range in the materials introduced would aid in a more detailed study concerning the nature of these substances.

It was thought that an organism more highly virulent for the mouse than those used by other observers might produce a filtrate in young cultures which would be toxic intraperitoneally. A strain of B. Friedlander (non-lactose-fermenting) was chosen which consistently caused death in about 12 hours when introduced intraperitoneally into mice. This culture was grown for various periods in veal infusion Witte peptone broth with and without 0.1 per cent dextrose. Sterile filtrates of these cultures were used for injections.

Eighteen-hour cultures produced no effect on the mice. Older cultures, 42 hours, 66 hours, 5 days, 10 days, and 20 days, while more toxic, did not produce constant effects. In some cases mice receiving 1 cc. of the filtrate survived, while others receiving 0.5 cc. died in 8 to 24 hours.

Eighteen-hour cultures grown anaerobically, and aerobic cultures in Martin's media, produced no effect in the animals. The intraperitoneal injection of such substances as gum tragacanth, bile, and 1:500 calcium chloride, as peritoneal irritants, followed by the injection of a 42-hour filtrate likewise caused no reaction.

Rabbits inoculated intravenously with 3 cc. of the filtrate of this strain, grown in veal infusion Witte peptone broth, did not show any signs of discomfort with the 18-hour cultures.

In a previous paper 1 were reported details of experiments on the intravascular use of heparin 2 , 3 , 4 in etherized dogs. These experiments, together with other experiences in preservation of blood samples used for various physical and chemical analyses, have demonstrated the value of this material in laboratory work. It seemed advisable, therefore, to investigate the possibilities of its use in transfusion when, for any reason, there might be delayed injection.

Numerous experiments were undertaken in which large doses of heparin—5 to 60 mg. per kilo of body weight—were injected intravenously in normal dogs and rabbits. In no instance was there any demonstrable gross disturbance of any body function.

It occurred to us, however, that the effects of repeated injections should be investigated. For this purpose, 20 mg. per cc. of heparin were dissolved in Ringer solution and injected into the peritoneal cavity of adult healthy guinea pigs, in relatively large doses, 20 to 60 mg. In a few instances symptoms characterized by prostration and weakness occurred within a few minutes after the first injection. After periods varying from 2 to 3 weeks, transcutaneous and intracutaneous tests were made with a similar solution of heparin, all of which proved negative. A second injection of a comparable dose of heparin was now made. In most cases this proved entirely negative, but in a few instances there occurred symptoms suggestive of anaphylaxis.

These results, together with those noted following the first injection, suggested two possibilities: 1. There might be a protein fraction in the heparin that sensitized certain more susceptible animals. 2. The heparin possessed toxic properties of undetermined nature.

Subsequent employment of fresh lots of heparin uniformly failed on both first and second injections to produce either toxic or anaphylactic symptoms, so that it is possible that the particular lot first used had not been satisfactorily purified.

Experimental work on dogs, cats, and pigs has thus far yielded the following positive findings. The cells of the III, IV, and VI cranial nuclei of the dog and cat can be separated into two distinct sizes, hitherto unrecognized, both having “motor” tigroid substance, and being in general diffusely intermingled throughout the nuclei. The smaller cells do not preponderate in any portion of the III, IV, or VI nuclei, nor in any of the subnuclei of the III, except Perlia's median nucleus which, in the dog, is made up entirely of the smaller cells. In the dog, the proportion of the sizes of the cells by actual count correlates roughly with the sizes of fibers in the peripheral trunks, more especially in the case of the third cranial pair. There is in the dog an excess of cells in the central nuclei of the extraocular muscles over the number of fibers in the peripheral homologous trunks, roughly, 30 to 40 per cent for the Nn. III and IV, and 10 per cent for the N. VI. Adequate clumps of cells along the peripheral nerves for the mediation of kinaesthetic sense of the extraocular muscles have not been demonstrated. The accumulated evidence of experiments with degenerations thus far indicates the possibility that the smaller cells in the central nuclei may mediate kinaesthetic sense of the extraocular muscles. The problem is being attacked further.

In the numerous observations made in recent years on the chemical changes which set in after death in muscle, little attention has been paid to those which occur in mammalian muscle during the first few minutes after the cessation of blood flow. The increase which occurs in free phosphorus after two hours incubation of muscle juice or chopped muscle, and from which the so-called lactacidogen is calculated, has not been clearly linked up with the changes which take place in glycogen, lactic acid and free sugar. In the theories of Meyerhof, 1 A. V. Hill, 2 and their co-workers, 3 with regard to muscular contraction, it is considered that lactic acid is very rapidly formed from glycogen. But this cannot occur directly; there must be some intermediate carbohydrate, and it was to obtain evidence of this that present investigation was undertaken. That such a substance may be formed in muscle during the action of insulin is suggested by various observers. 4

The muscles were rapidly dissected from one hind limb of a rabbit immediately after stunning, and frozen to a brittle mass with liquid air. After thorough pulverization in an iron mortar, quantities of from 3 to 5 gm. each of the powdered muscle were placed in weighed vacuum tubes. After all the tubes had again been weighed, equal volumes of 0.9 per cent NaCl solution were added to each, and they were evacuated and well shaken at room temperature. At intervals of 3 to 5 minutes tubes were then taken for the determination of glycogen, free sugar, total carbohydrate, free phosphorus and lactic acid. The results of a typical experiment are shown in curve form.

The free sugar in this particular experiment was that found in the Schenck filtrate, but the results are the same in numerous others in which the sugar was dissolved out from the muscle by 80 per cent alcohol.

The extraordinary separation of the pubic bones of the guinea pig in late pregnancy has been recorded by several investigators. Todd 1 has made the most recent contribution to this subject and has thoroughly discussed previous research as well as described the gradual changes that take place at the symphysis during normal pregnancy and the subsequent return to the post-parturient condition typical of multiparous females. The writer has been interested in this phenomena for the last four years, chiefly from the standpoint of its physiological explanation and possible correlation with a similar condition which has been studied in the pocket gopher. 2 In the pocket gopher the pubic bones are resorbed at the symphysis before pregnancy occurs, and the reaction is governed by the ovary, while in the guinea pig relaxation of the pubis occurs during pregnancy and little or no bone is resorbed, but the connective tissue at the symphysis is greatly increased, allowing a marked separation of the bones.

It has been possible to produce changes in the pubic ligament of virgin guinea pigs by experimental procedure and these are apparently identical with those occurring normally during pregnancy. If 2 cc. of blood serum of pregnant rabbits or guinea pigs are injected subcutaneously into virgin guinea pigs during early post oestrum a noticeable relaxation of the pubic ligament can be discerned within six to eight hours by movements at the pubis. These movements gradually become more pronounced during the next eighteen hours and the ligament may not return to its normal condition for two or three days. The blood of males and nonpregnant females does not bring about this reaction, and the blood of parturient females loses its effectiveness very noticeably within the first eight hours, negative results being usually obtained from blood drawn 24 hours after the young are born.

A recent note in the PROCEEDINGS by Deuel, Mandel and Waddell 1 leads us to report one of a series of experiments which we undertook some time ago, using a preparation of caramelized oatmeal with normal diabetic patients as suggested by Grafe. 2 Grafe had used caramel as a substitute for ordinary carbohydrates as far back as 1911. 3 It was found, however, that its use was liable to produce diarrhea.

The preparation which we used was Satrose, made by Messrs. Schering. It is said to contain nitrogen 2 per cent; fat 2.3 per cent; cellulose 8 per cent; ash 2 per cent, and carbohydrate 75 to 80 per cent. It is a brown powder with the taste of burnt brown paper. Although we gave it in the way suggested by the makers, we were unable to induce many of our patients to take it. This, combined with the fact that we were not able to make out any notable effect on the metabolism, lead us to abandon further work with it. From the standpoint of practical dietetics with the diabetic it does not appear to possess any advantage.

We give the results of administering 50 grams of Satrose and 50 grams of glucose to the same diabetic patient in the post-absorptive state.

In 1918 Smith 1 published electrocardiographic tracings following ligation of coronary vessels in the dog. In one tracing, five minutes after tying the circumflex branch of the left coronary artery, there occurred, among other changes, marked elevation of the T-wave, with the T-wave originating on the down stroke of the R-wave, when the latter had reached about one-half the distance to the base line.

In 1920 Pardee 2 published a similar electrocardiogram obtained from a patient four hours after an obstruction of a coronary vessel.

More recently, Rothschild, Mann, and Oppenheimer 3 renewed the interest in the peculiar change of the R-T interval as an early diagnostic sign of coronary obstruction. They reported observations on four patients shortly after acute coronary occlusions in which among the first changes in the electrocardiogram there was the characteristic elevation of the R-T interval above the base line.

In 1909 Eppinger and Rothberger 4 observed in dogs that partial or complete absence of the desecending limb of the R-wave was a relatively frequent and characteristic occurrence following the injection of silver nitrate solution into tlie muscle of the left ventricle whether at the base or apex. The T-wave became high and broad and originated either at the peak of the ascending limb of the R-wave or from the lower end of the descending limb which failed to reach the base line. They explained this effect on the basis of impaired contraction of the mass of circular muscle fibres in the structure of the left ventricle.

It is universally believed that the ciliated epithelium of the oviducts and uterus of mammals play the chief role in the transport of the ovum from ovary to uterine cavity. A few authors, however, seem to regard the contractions of the uterine muscle as being more important

The above belief is based on two facts, (1) the direction of the ciliary movement being vaginal-ward and (2) the appearance of small particles in the uterus or vagina, after being introduced into the general peritoneal cavity.

In attempting to establish the corollary, namely that if the direction of the cilia be reversed, the ovum could not progress beyond the part reversed, we have found that the reversal of the uterus offers no impediment to the passage of the ovum.

The method employed was to divide the uterine horn (of rabbits) on each side, in two places 2.5 to 3.0 cm. apart and to reunite the cut ends after reversing each segment. The animals were then allowed to heal (a laparotomy being performed in some cases to determine the condition of the anastomosis) and were later mated.

Six animals thus experimented upon have all become pregnant. Three of the animals were subjected to observation before term, the remainder were left to complete their pregnancy.

It was found that implantation occurred most frequently within the reversed segment and in one instance in the portion of the uterus between the segment and the cervex uteri. It is thus established, that an ovum may travel across two anastomoses apparently against an opposing ciliary movement.

It has been shown 1 that the secreted concentration of HCl (i. e., the concentration of acid as it is secreted by the cell) may not vary between 0 and a concentration higher than the observed maximum concentration (about 0.6 per cent), but may vary between a certain value (at least 0.29 per cent) and the latter, if it does so at all.

The restriction of the limits of variation strongly supports Pavlov's contention, namely that HCl is secreted at a constant concentration. If this be true, it should be possible to test the validity of the assumption by observing the factor which converts the amount (weight) of HCl into volume (cc.) of acid secreted. If the factor behaves as a constant, the concentration must obviously be invariable.

Employing simultaneous equations of the following form, hx + py + m = c (x and y are factors for converting known HCl (h) and pepsin (p) values in terms of cc. (c), and m is the volume of mucus) taken from two successive observations over an equal period, in any one Pavlov- or Heidenhain-pouch dog, the following values of x were obtained:

0.15 (3), 0.16 (2), 0.17 (7), 0.18 (6), 0.19 (7), 0.20 (6), 0.21 (4), 0.22 (2), 0.23 (2).

The mean of x = 0.19 and the S. D. = 0.02. (The figures in brackets give the frequency of each observation in the series.)

It will be seen that x is relatively constant, probably as constant as can be expected from the methods available for the determination of pepsin and mucus.

The values for y, however, varied greatly (0.0001 to 0.14) and gave extremely low volumes for pepsin. If, therefore, pepsin be neglected and the Cl not accounted for by acid (i. e., total Cl=HCl Cl) be considered to be NaCl which has diffused through the gastric mucosa as an isotonic solution, and if this volume of non-acid fluid replaces py in the above formula, the solution of the simple equation gives the following values of x:

It has been demonstrated 1 that polymorphonuclear leucocoytes neither activate nor contribute to the proteolytic power of pepsin when added in quantities (100,000 to 500,000 per cc.) commonly occurring in the gastric juice of dogs. It has now been found that lymphocytes behave in a similar manner, (see Tables I and II). Lymphocytes (lymphocytes 93 per cent, monocytes 4 per cent, polymorphs. 3 per cent) from the dog's thoracic duct lymph, were, after repeated washings in saline, suspended in distilled water and immediately added either to dog's gastric juice or to known dilutions of Merck's pepsin and the whole adjusted to constant volume and acidity. The ferment activity was estimated by Mett's method.

These results do not lend support to the theory 2 , 3 that the lymphocytes play a role in the peptic digestion of gastric juice.

1. Pseudoephedrine in a 10 per cent solution is an uncertain mydriatic and cycloplegic. The age of the individual does not seem to be a factor. There does exist, however, an individual difference, probably due to a variation in the rate of the absorption of the drug.

2. The action of pseudoephedrine upon the eye is like that of homatropine, rather than like adrenalin, which it closely resembles chemically.

3. In 10 per cent solution pseudoephdrine is slightly toxic when instilled into the eye.

4. There is no clinical evidence that pseuodephedrine contracts the blood vessels as does ephedrine which confirms the laboratory studies upon frogs performed by M. Fujii. 2

5. Pseudoephedrine in a 10 per cent solution apparently has no place in the treatment and examination of ophthalmic diseases. A stronger solution would undoubtedly be a more active mydriatic and cycloplegic but would be too toxic to justify its use.

6. The results of these experiments do not wholly agree with those of Lewin-Guillery. 3

The dysentery amoeba of the monkey and that of man are so similar in structure that some regard them as probably being the same species, though other investigators have given them different species names. In an attempt to procure experimental evidence on this question two monkeys found to be amoeba-free were fed cysts of E. dysenteriae of man. Examination of the feces during a period of three months and of histological sections following autopsy showed the presence of cysts and trophozoites of E. dysenteriae, morphologically and racially indistinguishable from the amoeba fed, thus indicating that the dysentery amoeba of man may be experimentally established in the monkey. Subsequently, two cats, six months and seven months of age respectively, found negative for protozoa by preliminary examination, were given rectal injections of monkey feces containing cysts of the dysentery-like amoebae of a naturally infected monkey, using the technique of Boeck and Drbohlav. 1 Two months later, autopsy of the cats showed the amoebic infection to be well established in the upper colon and cecum, where there was excessive mucous, pronounced hyperemia, and a distinct thickening of the gut wall. Trophozoites of amoebae, many of which contained red blood corpuscles, were found in great numbers in these regions and histological sections showed necrosis of the mucosa and penetration of the amoeba into the tissue. Amoebae from the infected colon of the cats were planted into the L. E. S. medium of Boeck 1 and at the time of writing seventeen successful subcultures have been made on alternate days and the cultures show no signs of dying out.

In a previous communication 1 there was reported the isolation of organisms belonging to the genus Salmonella from foxes dying of an epidemic disease. These organisms have been shown to be very pathogenic for foxes. When injected into healthy foxes a disease is consistently produced, death usually occurring in from 14 to 20 days. The organism has been isolated from the spleen of an animal killed 5 days after injection, but no gross pathological changes had yet occurred. In artificially infected foxes the pathology at necropsy is characterized by an intestinal inflammation and enlargement of the spleen, lungs usually being normal. The spleen is sometimes enlarged 20 times, by weight. The same picture has been produced by a direct injection of material from carcasses of ranch animals.

Further experimental work has been carried out in vaccination of ranch animals with these organisms, and further studies of disease occurring in these animals, using for transmission experiments foxes which also have been immunized with a Salmonella vaccine. An epidemic on a large ranch had previously been studied in which the picture of a Salmonella infection was the characteristic pathology. The organisms were isolated from many animals. Vaccination of the entire ranch of more than 500 animals appeared to control the epidemic, which was mainly among foxes 4 months old.

During the ensuing months there were a few scattered deaths, and 6 months later there was an increase in the number of deaths sufficient to indicate the presence of a mild epidemic. In the second epidemic quite a different picture was presented from the one previous. There were no external signs of disease. Some animals were found dead; others were found in convulsions, dying shortly afterward, and a few were sick for several days before death.

It has been reported by a number of investigators 1 , 2 , 3 that rats become rachitic on diets high in P and low in Ca.

Two rats were kept on a normal diet (A) containing 302 mg. P and 414 mg. Ca per 100 gm. diet. Another group of two were given a diet (B) low in P (124 mg. P and 372 mg. Ca per 100 gm. diet). A third group of two were given a diet (C) containing 1044 mg. P and 19 mg. Ca per 100 gm. diet. After three weeks the rats were photographed by X-rays and ashed. The ash was analyzed for Ca, P and Mg. The X-ray photograph showed that the bones of the rats on low Ca, high P were normal, except for a slight osteoporosis. In marked contrast to those on normal diet and diet low in Ca, both rats which received diet (B) became rachitic.

The percentages of Ca and P in the ash of those on diet C were slightly less than in normal rats () of their own age. The P/Ca ratio in the ash of the rats was normal (0.72). The rachitic index figured by McClendon's formula 4 was (A) 5.2, (B) 1.4, and (C) 2.0.

In a previous paper 1 data were presented which led to the belief that the comparative resistance of certain plum varieties to brown rot is due to mechanical resistance to the entrance of the fungus. The more resistant varieties had a tougher skin and a firmer flesh; and, at the same time, they had a higher crude fiber content. Preliminary calculations of correlation by rank between skin and flesh texture indicated that the two factors, tough skin and firm flesh, varied together. If this premise could be proved to hold: generally, it would be of value in further work on plums, since only one of the two factors would need to be measured.

A considerable number of measurements of these two constants are now at hand. They cover the three seasons of 1923, 1924, and 1925; they involve over 45 varieties of plums grown at the Fruit Breeding Farm of the University of Minnesota; and they have been taken at all stages of ripeness. These data are presented in the accompanying table.

Each sample, representing a variety at a particular stage of ripeness, consisted of from 4 to 6 plums. The puncture test on the skin and the penetration test on the flesh, as described in the previous paper, were made from 3 to 6 times on each plum. The values used in the calculations were sometimes the averages of the readings on the individual plums, at other times the averages for the lot. In the 1925 material stages III and IV were used both separately and together. In the other cases all stages were included in one calculation.

It is apparent from the magnitude of the coefficients that a very significant negative correlation exists between the values for the two constants.

The present paper briefly presents certain facts which extend present knowledge regarding the effects of some of the following external conditions on the velocity of respiration and photosynthesis in the marine kelp Nereocystis. Some of the facts do apply, and the others may apply to respiration and photosynthesis of plants in general.

Uniform strips 1 × 10 cm. were cut from the frond of the kelp. The oxygen exchange in respiration and photosynthesis was determined by Winkler's method. Specially made bottles of 25 cc. volume were used. One strip was placed in each bottle for a determination. Three duplicate bottles with strips were used for the same concentration of oxygen, cyanide, etc. Each figure in the tables is, therefore, the average of three duplicate, simultaneous determinations. During the tests the experimental bottles were suitably immersed in the open sea, which afforded a constant temperature to within ±0.5° C. during any one experiment. Light intensity was the same in all the tests of any one experiment, but varied more or less from one experiment to another. Suitable control experiments showed that the effects reported are not due to differences in hydrogen ion concentration.

Effect of oxygen concentration on the rate of respiration and photosynthesis. It is a familiar fact that increase in concentration of oxygen increases the rate of oxygen consumption in many kinds of plants. It is not such a generally recognized fact that increase in concentration of oxygen decreases the rate of photosynthesis. These two effects of oxygen concentration on Nereocystis are shown by the three experiments performed simultaneously, and given in Table I.

Column 1 shows the initial concentration of dissolved oxygen in sea water at which the rates of respiration and photosynthesis were tested.

Destruction of uric acid, which is rapid and marked in the normal dog, 2 does not occur if the liver is entirely removed. Complete removal of the liver in the dog produces a very great increase in the uric acid content of the blood and tissues, and also in the urine. Uric acid injected into the dehepatized dog remains unchanged in the blood and tissues and is excreted unchanged in the urine. The destruction of uric acid in the dog seems to be entirely dependent on the presence of the liver, 1 since no uric acid is destroyed in the absence of the liver and no other means of influencing the destruction of uric acid has been demonstrated. Intravenous injection of standard amounts of uric acid into dogs with hepatic insufficiency is followed by a delay in the disappearance of the excess uric acid from the blood, and by an increase in the amount of uric acid excreted in the urine. Both the delay in the disappearance of the excess uric acid from the blood and the amount of uric acid appearing in the urine are greater, the greater the amount of damage or reduction of hepatic tissue. Two objections should be raised to the use of injections of uric acid as a test of hepatic function. First, injections of large amounts of uric acid produce severe lesions in the kidneys and interfere with excretion; and second, lesions of the kidneys may produce retention of uric acid, although in the dog the presence or absence of the kidneys is without effect on the rate of disappearance of uric acid from the blood.

The human uterus undergoes a marked reduction in length and weight in the first few weeks following birth. This was first described by Lyubetski, 1 and later, independently, by Bayer 2 and by Conte. 3 This reduction takes place through hypoplasia and hypotrophy of the uterine muscle, together with a disappearance of the marked natal hyperemia of the organ. It is supposed to be caused by the withdrawal at birth of a hormone produced by the placenta, the ovary or the tissues of both of these structures, (Aschner, 4 Herrmann, 5 Fellner, 6 , 7 Frank, 8 , 9 and others.) Little is known of the fetal growth of the uterus which precedes this natal reduction. The present study is based upon a total of 207 observations on the length of the uterus, 89 of fetal uteri, 62 of uteri of infants stillborn or dying within 48 hours after birth, and 56 uteri of children over 2 days and under 1 year of age.

A graphic analysis of this material is shown in the accompanying figure, in which the mean uterus length for 5 cm. intervals of total or crown-heel body-length, is plotted against the total body-length (as indicated on the base-line scale), and the computed age, (as indicated on the upper scale of the graph). Starting with an average length of about 2.5 mm. in the 5 to 10 cm. interval of crown-heel length, the viscus grows at a slow and fairly constant absolute rate, with respect to the total body-length, until approximately 7 lunar months (about 35 cm. crown-heel length). At this stage the average length of the uterus is approximately 17 mm.

In 1923, Dr. C. E. Tennant 1 reported a surgical case in which 15 stones having a total weight of 73 grams were removed from a kidney. He noted that these stones were composed chiefly of cystine, which, upon purification, crystallized in hexagonal plates. Inasmuch as this material offered an unusual opportunity of again investigating the old question—is stone cystine identical in chemical composition with protein cystine—we secured, through the kindness of Dr. Tennant, a number of the kidney stones, and have analyzed them and prepared certain organic derivatives of the “stone” cystine. Our data, in summary, are:

1. 5.20 grams of the cystine stones yielded 4.84 grams, or 93 per cent of pure cystine crystallizing in typical hexagonal plates. Qualitative tests on the filtrate from the cystine crystallization indicated that small amounts of calcium and phosphate were present. Neuberg and Mayer 2 state that “protein” cystine crystallizes in hexagonal plates but “stone” cystine crytallizes in needles. We have found “protein” cystine to crystallize in the typical hexagonal plates, whereas our “isomeric” 3 cystine, prepared from “protein” cystine by long boiling with 20 per cent HCl crytallizes in microscopic needles. We have found “protein” cystine to crystallize in the typical hexagonal plates, whereas our “isomeric” 3 cystine, prepared from “protein” cystine by long boiling with 20 per cent HCl crytallizes in microscopic needles.

2. The cystine crystals analyzed for 11.63 per cent nitrogen (theory 11.65 per cent) and 26.55 per cent sulfur (theory 26.67 per cent), and a 1 per cent solution in approximately 0.1 N HCl had a specific optical rotation of [α_D ²⁰] = −242.6°. Neuberg and Mayer 2 report −224° for the optical rotation of “protein” cystine and −206° for “stone” cystine.

Our previous studies have shown the feasibility of measuring the velocity of blood flow in health and disease by injecting the active deposit of radium into one of the antecubital veins, and determining the time of arrival of the active deposit in the arterial vessels about the elbow of the other arm.

The path of the active deposit coursing through the body necessarily included the veins of the arm, the pulmonary circulation, and the artery of the other arm. The velocity measured was a somewhat complex expression, therefore, of the peripheral as well as of the central blood flow.

This report deals with an attempt to determine the separate velocities along these paths.

A detecting device was placed in a lead block. This block had a hole 5 cm. in diameter bored through its center. The radiations of radium C as it was injected into the veins and as it flowed towards the right heart could not penetrate the lead block and could therefore set up no disturbance in the detecting device. The hole of the lead block with the detector set into it, was placed over the right heart. When the active deposit reached the right heart, the emergent radiations easily traversed the air and tissues between the right heart and the detector, and set up characteristic disturbances in the detecting device. These disturbances were amplified by three electrode vacuum tubes, and were finally registered automatically by a recording pen galvanometer. Knowing the time of injection of the active deposit and knowing the time of its arrival into the right heart, we have a measure of the velocity of the venous blood flow. Figure I is the record of a typical determination.

In previous experiments on embryonic limb transplantation in amblystoma, 1 it was shown that when the fore limb is removed, the sensory ganglia of the brachial nerves (third, fourth and fifth) undergo a decrease in the number of cells of approximately fifty per cent. Conversely, when a limb is grafted to a strange region, the ganglia of the spinal nerves supplying the heterotopic limb undergo an increased cellular development of twenty-five to forty per cent.

Since it is known that spinal ganglia contain the cell bodies of sensory fibers to skin, as well as to body musculature, and since cellular hypoplasias result when the peripheral field is diminished, the supposition is that both skin loss and muscle loss play a role in the resulting hypoplasias.

In an attempt to determine quantitatively the part played by skin and by muscle in the development of spinal ganglia, two embryos were fused together laterally, and in such a way as to reduce greatly the area of skin on the fused sides of the two component embryos. By comparing the area of skin belonging to the fused sides of the components with that covering the free sides, the percentage reduction of skin resulting from the fusion has been obtained for homologous regions of two pairs of “parabiotic twins.” 2 By comparing the weight of the musculature belonging to the fused body walls of the two components with that of the two free body walls, the percentage reduction in the volume of muscle was likewise obtained. The cellular reductions in the spinal ganglia, which supply the fused body walls, were also calculated for those nerves concerned with the homologous regions under comparison.

Designating the fraction of cells in the ganglia related to skin by a, and the fraction related to muscle by b, the data obtained from the quantitative study of skin loss, muscle loss, and cellular loss in the homologous regions of the laterally fused animals, have been utilized to obtain values for a and b, which indicate that approximately sixty per cent of the cellular losses in the spinal ganglia, after fusion, is due to skin loss, and that approximately forty per cent is due to muscle loss.

The reaction to epinephrin in the obese has been studied from a metabolic standpoint in a series of cases and compared with a group of normal controls. The subjects were studied in the basal condition, that is twelve to fifteen hours after the last meal, with a period of preliminary rest of one half hour before readings were begun. A solution of adrenalin (P. D. & Co.) in the tablet form, containing 0.625 mg. of adrenalin, were injected intramuscularly in each subject and the metabolism, respiratory quotient, ventilation, pulse rate and blood pressure noted 10, 20, 30, 60, 90, 120, and 150 minutes after epinephrin injection. The tablet form of epinephrin was used so as to insure greater constancy in epinephrin content, the solution having a tendency to deteriorate on standing.

Expressed on a basis of percentage rise above basal, it was found that the metabolism rose 26 to 28 per cent above basal in 20 to 30 minutes after epinephrin injection, in both the obese and normal subjects, which compares favorably with Boothby and Sandiford's work. 1 The ventilation per hour rose to 38 per cent in the obese and to 42 per cent in the normal controls. The respiratory quotients however showed the greatest variation, rising to 11 per cent above basal in 10 minutes in the obese, while a level of 19.6 per cent was reached in the normal subjects in the same length of time. The average basal respiratory quotient in the obese was .754, while in the normal it averaged .810. The pulse rate rose to 40 per cent in the obese, but only to 26 per cent in the normal in 30 minutes after injection.

Patients with exophthalmic goiter and normal controls were given a carefully weighed diet deficient in calcium (0.1 gm. per day) but adequate in total calories. On this regime a negative calcium balance was established in all subjects. Determinations were made of total calcium, nitrogen and phosphorus in urine and feces. Frequent basal metabolic rates and determinations of calcium and phosphorus of blood were also made.

Three typical, rather severe cases of exophthalmic goiter showed a very high calcium excretion—one of them five times the average found in a series of controls. Phosphorus excretion was also increased though not as markedly as calcium. This high excretion was maintained with a high basal metabolic rate, but as the basal metabolic rate fell (following the ingestion of Lugol's solution and operation), the calcium excretion also fell markedly and approached normal.

One myxedema patient showed a calcium excretion below normal. Two of the normal controls also took thyroid and thyroxin in amounts sufficient to raise metabolism twenty per cent. On this diet, inadequate in calcium, the calcium excretion rose definitely with the metabolic rate.

In all of these subjects the blood calcium and phosphorus was normal.

The significance of these findings and their possible relation to parathyroid activity is being further investigated on cases of myxedema, adenoma of the thyroid, parathyroid tetany, and on animals.

Those who find inadequate the belief that the sympathetic nervous system controls the tonus of skeletal muscle have been confronted by the difficulty of offering an alternative view satisfactory to themselves and to their opponents. It is the purpose of this communication to suggest a mechanism of tonic contraction compatible with the all-or-none principle of muscular activity. The present conception takes due account of the long-sustained character of postural reactions; it recognizes the lengthening and shortening reactions, and it conforms with the well-recognized cooperative interaction between voluntary and tonic responses. Finally, it takes into account the probable function subserved by the sympathetic nerve supply of skeletal muscle

To examine tonic reactions of skeletal muscle, it is essential to simplify experimental conditions to the greatest possible extent. Consequently, if we utilize the muscles of a decerebrate preparation, which are admittedly tonic, we must exclude all possible extraneous reflex influence. Magnus 1 has found that decerebrate rigidity continues to exist in preparation, the brain-stem of which has been sectioned just above the vestibular nuclei and in which the VIIIth cranial and the I, II, and III cervical nerves on both sides have been severed. The rigidity of a decerebrate preparation continues to exist after complete denervation of the skin in any given limb, and in any given extensor muscle, after section of all nerves except the nerve supplying that muscle. If one examines the vastocrureus muscle of a decerebrate preparation; the skin and all other muscles of both hind extremities of which have been denervated, one finds that so long as the muscle is stretched the rigidity continues to exist. Detach the tendon and allow the muscle to shorten a few millimeters, and the “rigidity” vanishes.

The writer wishes to report briefly the results of experiments carried out during the last seven months in attempting to repeat and amplify the experiments of Gye 1 on the Rous chicken sarcoma. The latter interpreted his results as meaning that two factors were involved in the production of this tumor; first a specific, non-living, chemical factor found only in extracts of the neoplasm; and second, a filtering virus, capable of being cultivated in vitro under certain conditions. The latter was also obtainable from cultures of mammalian tumors. Neither substance alone was capable of producing a tumor, together they were active. The media used in cultivation of the virus contained broth, rabbit serum and chick embryonic tissue. No controls using un-inoculated, but incubated media of this type to replace virus were quoted in the original paper. In personal communications, Gye lias reported to the writer that such controls have been carried out and in his hands have proved negative. Murphy, 2 on the contrary, has recently presented evidence that media uninocu-lated with tumor, but containing either chick embryonic tissue or mouse placenta, i. e., rapidly growing tissues, and incubated anaerobically, could be substituted successfully for virus cultures. Until agreement can be reached on this and other points, judgment of the interpretations to be placed on the phenomenon may well be reserved.

In this laboratory the repetition of the type experiments of Gye have proved to be far from easy. The experiments thus far have used something over two hundred chickens and have served to point out the difficulties to be expected, and perhaps to suggest means of overcoming them.

There are marked individual variations in susceptibility to small doses of tumor extract manifested even by pure Barred Rock chickens, which we have used exclusively.

Intracardiac injections of dyes and of India ink in living cats, rabbits and rats have resulted in complete filling of the capillaries of these hearts. Numerous counts have shown approximately 1100 capillaries to each thousand muscle fibers, or about one capillary to each muscle fiber. But when for any reason the heart dilated during the injection, very few of the capillaries were injected, though the larger vessels were completely injected.

The same results were obtained when human hearts were injected through the coronary arteries. Distension of the chambers prevented injection of the capillaries, but when steps were taken to prevent dilatation of the chambers complete injections were obtained.

At the same time it was noted that perfusion of the coronary arteries in dead hearts resulted in distension of the chambers, and when the walls were so stretched 80 per cent to 90 per cent of the perfusate escaped directly into the chambers of the heart, while only 10 per cent to 20 per cent returned by way of the coronary sinus and veins. These findings suggest that during dilatation of the heart the chief route of blood flow is through the arteries to the Thebesian vessels and thence into the chambers of the heart.

A modification of the Langendorff method of coronary perfusion has given complete injections of the capillaries in cat and rabbit hearts when the hearts were beating strongly. This method when applied to human hearts obtained within three or four hours post mortem has given complete injection of the capillaries in certain areas of the heart. There is approximately one capillary per muscle fiber in the human heart, except in the auricle where the number is not constant and the supply less abundant.

Several parasites, a flagellate of the genus Giardia and also microsporidia have been reported in intestinal round worms, but up to the present time the latter have not been shown to be concerned in the transmission of disease.

Blackhead, an infectious disease of turkeys and other poultry, is caused by a flagellate, Histomonas meleagridis. It is transmitted experimentally and to some extent in nature by the direct ingestion of material contaminated with freshly passed discharges containing the protozoon. It appears, however, to be much more frequently transmitted indirectly by some phase distributed on the soil, evidently in association with the eggs of the caecal worm, H. vesicularis.

The presence of the protozoon in the egg of Heterakis is indicated by experimental evidence of various sorts.

1. Heterakis eggs kept in 1.5 per cent nitric acid until embryonated produce blackhead when fed to young birds isolated from all other sources of infection, although this treatment renders the material bacteriologically sterile.

2. That there occurs no resistant phase of the blackhead protozoon apart from the worm egg, capable of resisting the 1.5 per cent acid, is shown by the invariably negative results obtained by the repeated feeding of susceptible birds with the discharges of blackhead carriers, after treatment in 1.5 per cent acid. Furthermore, no resistant form has been demonstrated micro-scopically.

3. Heterakis material will only produce blackhead after the ova have blecome embryonated and capable of hatching. Samples of the same material fed before the eggs are ripe invariably furnish negative results.

4. The feeding of male Heterakis also furnishes only negative results although ova-containing females of the same lot produce blackhead.

The disease usually follows the feedinig of large numbersl of Heterakis, especially when the latter are pooled from several different birds.

In 1917 Myers and Killian 1 described a new simple method for the measurement of the diastatic, or starch-splitting, activity of blood. In this method two samples of 2 cc. each of oxalated blood, diluted with a given amount of water, to one of which was added 1 cc. of a 1 per cent solution of soluble starch, were heated in a water bath for 15 minutes at 40° C. and at the end of the digestion period the sugar content of each sample was measured. The difference, representing the amount of glucose liberated from the starch by the blood diastase, was converted into per cent of the original amount of starch (10 mg.), and this percentage figure was spoken of as the diastatic index or as diastatic units. They reported a considerable number of findings from the examination of the blood of healthy and diseased individuals, and later DeNiord and Schreiner, 2 Brill, 3 Watanabe, 4 and Lewis and Mason 5 reported clinical studies in which this method was used with slight modification. None of the results reported, however, gave evidence of constant variations in this diastatic power of the blood, which could be definitely correlated with the physical condition of the individuals studied. Karsner, Koechert, Wahl, 6 and Cohen 7 have reported studies of the variation of the blood diastase in animals under certain experimental conditions, using this method.

A study of the diastatic activity of the blood of infants and children, normal and otherwise, was undertaken with the hypothesis that the blood of infants suffering with various types of severe nutritional disturbances might show significant variations in the activity of the enzymes presumably responsible, in part, for the utilization of food substances.

The most sensitive method hitherto used for measuring the carbon dioxide output of tissues in vitro is the method of Tashiro, 1 depending upon the detection of a crystal of barium carbonate in a solution of barium hydroxide under the microscope. It was found that 1 × 10^-7 of a gram of carbon dioxide gave a visible precipitate in 10 minutes. An unknown amount of carbon dioxide was measured by rinding such a dilution that a visible precipitate was just formed in 10 minutes. This method has not been found useful in practice in the hands of others.

The next most sensitive method is the indicator method of Osterhout, 2 which in the hands of Parker 3 has been found adequate for the detection of 1 × 10^-6 grams. It takes, however, 20 × 10^-6 grams to produce the standard color change in the indicator. If this change takes place in 20 minutes, Parker finds himself able to detect a perceptible color change after 1 minute, i. e., from the effect of 1 × 10^-6 grams.

At the Cleveland meeting of the Federation, December, 1925, the writer demonstrated a carbon dioxide method based upon measurements of the conductivity of barium hydroxide, with which it is possible to detect 1 × 10^-7 of a gram of carbon dioxide. The method was designed for the estimation of the carbon dioxide output from a stimulated nerve. Owing to an unavoidable delay in completing the experiments on nerves it has seemed advisable, in response to requests from other workers, to publish a brief description of the method, which is not new in principle. It was pointed out to me at Cleveland that it had been used by Spoehr 4 for relatively large amounts of CO₂.

In previous communications, Mellon 1 , 2 has shown that the principle of ion antagonism is applicable to bacteria. Under the conditions of the experiment with strain N.D.—67 the principle operated in a Na: Ca ratio as high as 100 or 150 to 1. This strain, which is quite stable in H₂O and NaCl solutions, is quickly precipitated by CaCl₂ solutions in strengths as dilute as 10^-6 cm. The physico-chemical mechanism suggested was reversal of the aqueous-lipoid system in the wall of the cell whereby CaCl₂ made the lipoid layer the external phase. This theory is in accord with the work of Clowes 3 with the simple oil-water emulsions.

It was determined to test the validity of this conception by the surface tension method. Mudd 4 has shown that the tubercle bacillus when placed at the interface between an aqueous and a lipoid or a lipoid solvent will always be drawn into the latter by virtue of the large amount of wax contained in this organism. It was thought that the occurrence of the same phenomenon would be strong evidence for a reversal of the aqueous-lipoidal relations of our N.D.—67 by CaCl₂.

Accordingly the organisms were placed in CaCl₂, solution in concentrations ranging from 10^-1 to 10^-6 and this solution shaken with organic lipoidal solvents, such as cyclohexanol, amyl alcohol, capryl alcohol and pelargonic acid. In every instance the organisms were found on the organic side of the phase boundary when the latter formed. They did not pass aver to the aqueous side under the time of observation, which was several days.

A series of higher alcohols was used that showed a progressively increased solubility in H₂O as follows: Di-butyl carbinol < than Di-propyl carbinol which is < than Di-ethel carbinol.

Previous work 1 has indicated that, contrary to the generally accepted view, the fecal lipids (particularly the sterols) do not reach the intestine by way of the bile, but enter it below the absorbing portion. In order to shed further light upon this question, bile fistula operations have been performed upon 3 of the normal dogs used in previous work with essentially lipid-free diets 1 , 2 and these animals have been employed in similar experiments, the results obtained while the dogs were still normal, serving as controls. In the case of 2 of the animals, 6 such control experiments had been carried out, while 13 were available in the case of the 3rd. No attempt was made to keep the bile tract sterile in one of the animals, and there was considerable infection during the experiment. The dog ate well, however, appeared normal, and maintained its weight. There was no essential difference in the results from this animal and those from the others in which the bile tract was kept sterile. During the experiments the dogs were prevented from licking the bile either by means of a muzzle (in one experiment with the infected dog) or by collection of the bile in a balloon kept in place by a binder. There was no marked loss of weight in any case.

The diet was biased on the work of Cowgill 3 and contained casein (23.2 parts), sugar (50.0 parts), bone ash (1.4 parts), Liebig's meat extract (1.8 parts), salt mixture (1 part), and Vitavose∗ (2.2 parts). The casein was continuously extracted with hot alcohol for several days, and that used in most of the experiments contained [less than 2 parts of lipids in 10,000 as determined by digestion of the protein with strong alkali for 24 hours, followed by acidification and extraction with ether and petroleum ether.

The investigation of habit formation in thyroidectomized sheep and goats by the maze method has yielded results difficult to analyze because of the complicated responses elicited. The conditioned motor reflex method of Bekhterev has, therefore, been adopted.

In a preliminary experiment a conditioned reflex to a tactile stimulus was established in three animals, two thyroidectomized sheep, aged three and four years respectively, and the twin control of the three year cretin. Tactile stimuli were applied to a spot on the rump, at the rate of thirty per minute, for periods varying from two to ten seconds. With the final stimulus, a far-adic current was applied to the left foreleg of sufficient intensity to evoke a defensive movement. The first definite leg movement, in response to the tactile stimulus alone, occurred at the tenth combination in the control and in the three year cretin, and at the seventeenth combination in the older cretin. At the end of the ninth day and forty-fifth combination, one milligram of thyroxin was administered to each thyroidectomized animal. In spite of this, one and one half months later the younger cretin died and the training of the other animals was then discontinued, after two hundred thirty-eight combinations of conditioned and unconditioned stimuli.

After an interval of six and one half months the tactile stimulus without reenforcement evoked a vigorous conditioned reflex in both normal and cretin sheep. This is shown in Fig. 1, the tracing from the cretin appearing at the left. Continued tactile stimulation without application of the faradic current soon elicited a less extensive response (see Fig. 2) and finally, after one hundred nine periods of stimulation, distributed over sixteen days, the conditioned reflex failed to appear in either animal on the seventeenth day.

The adrenals are remarkably susceptible to diphtheria toxin. Elliott 1 found that a sufficient dose of diphtheria toxin to produce death within seventy hours after injection usually caused a marked depletion of the epinephrin content of an adrenal with an intact nerve supply.

We have attempted to show an increased output of epinephrin after the injection of diphtheria toxin by means of the degang-lionated iris (superior cervical ganglion, removed at least a week before the test). Diphtheria toxin ranging from 0.0005 cc. per kilo to 0.0119 cc. per kilo was injected intraperitoneally into fourteen cats. This was usually given in one injection.

The size of the pupil (as compared to the control), the rectal temperature and heart rate were determined at intervals after the injection.

With the larger doses some animals died in twenty to twenty-six hours. Animals lived longer with the smaller doses, although most of them died from the injection within a few days.

In none of the cats was there an unquestioned ailatation of the deganglionated pupil.

The adrenals were fixed with formaldehyde and potassium bichromate and stained with Sudan III.

With the range of dosage used there appeared to be no definite relation between the amount of toxin and the depletion of the medulla. Occasionally, the medulla was depleted, judging by the staining. Usually, however, epinephrin was present in patches or in limited groups of cells.

Likewise, there appeared to be no definite relation between the dosage and the amount of lipoids in the cortex except that with the larger doses, lipoids were absent or nearly so in the zona reticulairis.

There has been lacking a suitable, simple method for demonstrating the muscular hypertonicity of anaphylactic shock in the intact organism. The following method answers the purpose admirably.

A sensitized pigeon of about 300 to 400 gm. body weight, and deprived of its food over night, is tied, back down, by fixing its legs and wings conveniently to a small operating board such as is used for guinea pigs. A head holder is unnecessary. Then a balloon made from a fish-skin condom, tightly attached to the end of a rubber catheter (No. 20 French) so as to form a balloon of about 10 by 4 cm. when distended, is introduced directly into the crop. The introduction is facilitated by first collapsing the balloon and then bringing it in close apposition to the catheter, which is stiff and carries the balloon with it. The catheter is now attached to a T-tube, which in turn is joined by rubber tubing to a small tambour. Then the balloon is distended by blowing air through one arm of the T-tube, and the system adjusted so as to keep the crop moderately well distended with air and yet permit a record of tonus changes from the tambour on a slow kymograph. Presently large peristaltic waves may begin, but these are not indispensable to the experiment. As soon as a control record is secured, the antigen is injected into a wing vein (right) in the axillary region, using an ordinary Luer syringe and needle of 23 or 26 gauge.

Almost immediately after injection of from $0.2 to $0.5 cc. of the antigen, there is a sharp and marked rise in tonus accompanied by other symptoms (lachrymation, salivation, dyspnea, convulsions, etc.) of anaphylactic shock.

Harder 1 has made a careful study of the group of microorganisms, generally classed by him as the trichobacteria, to which he ascribes the deposition of certain sedimentary iron ores. Although Harder 'devotes himself largely to these trichobacteria, he mentions finding a number of bacteria (eubacteria) which were also able to precipitate ferric hydroxide and other ferric salts from media containing iron salts of organic acids. Since these iron depositing bacteria are soil types, it is not surprising that to a greater or lesser degree they should be found in milk and its products.

The ferric ammonium citrate media of Harder was used both as a broth and as a solid medium. On the agar were obtained colonies described by Harder as typical; that is, large, irregular, iron-incrusted colonies. Other types as well were found. Of the organisms forming these colonies, cocci and bacilli seem to occur in about equal numbers. There were very few spore formers, no motility, and no liquefaction of gelatin. Dextrose is fermented but lactose and saccharose only rarely so. As a class these organisms were able to utilize the salts of organic acids for their source of carbon, the pH of the medium rising distinctly. Thus they correspond to the group of alkali bacteria studied by Ayers. 2

The knowledge of the presence of these organisms in milk is believed to be new and significance is attached to them, since frequently undesirable qualities are found in the milk when they are present in considerable numbers.

During the studies of iron depositing bacteria as found in milk (see preceding paper) and being impressed with the possible relation of these organisms to hard pan formation, some iron hard pans such as abound in certain sections in the state of California were inoculated into the ferric ammonium citrate medium of Harder. 1 After standing two months a precipitate appeared which is typical of the reaction of iron depositing bacteria in this medium. Plates poured with this medium showed colonies of bacteria (bacilli) and actinomyces. Observation of these pans under the microscope disclosed as well some very minute threads which on staining suggested crenothrix. Not all of the pans, however, showed this appearance.

It is recognized that this is hypothetical, but Morrison and Sothers 2 have suggesteld this very thing, although with no apparent evidence to substantiate their suggestion. And recently A. C. Swinnerton 3 states that bacteria “may be important factors in the cementation of sand and gravel materials.”

Work along this line is being continued.

A better understanding of the relationship of physical measurements to body weight is necessary before a more certain basis for predicting normality of weight can be found. At present height is practically the sole basis. But height is a very stable measure in which adults are probably more nearly alike than for any other physical measurement. Stature has a co-efficient of variability of only 4 per cent, weight has around 12 per cent, and pelvic diameter as representative of bony body widths has approximately 9 per cent. If the bony widths of the body have so much variability as compared with bony lengths should they not be taken into account in estimating the normal weight, which of course is a cubic factor?

A group of adult hyperthyroid cases were measured before, and again, six months after operation. Critical linear measures were carefully taken. The effort was for “bony measurement.” Data for 14 cases, 11 women and 3 men, are summarized. The average gain in weight was 9.1 kilograms, an increase of 17 per cent of the preoperation weight. The calories per kilogram decreased 37 per cent. Chest girth, a cross-sectional measure, increased 9 per cent, but the bony lengths gave almost no change; height, + 0.1; sternal notch height, +0.2, and sitting height, +0.4 per cent. Widths were not quite as fixed during this marked weight change as were lengths: shoulders (acromion) +2.3; chest (transverse diameter) +2.7; chest depth, +4.2, and pelvic maximum diameter, +0.5 per cent. The latter measure is remarkable constant with the individual.

Since adults show relatively large differences among themselves in the linear skeletal widths, and, these measures show themselves quite constant in the individual adult, even under marked and rapid weight change, it is evident, if further data confirms, that some skeletal widths should appear in the formula for predicting normal weight.

In the early stages of the malt diastase hydrolysis of starch, Lintner and Düll 1 obtained crystal clusters which gave the characteristic starch reaction with iodine, were insoluble in cold water but soluble in hot, and had a specific rotation of 196°. Beijerinck 2 dissolved starch paste by autoclaving and obtained microscopic needle clusters on cooling. These were biref ringent when viewed between crossed nicols, but did not show the black cross which is so characteristic of starch grains. The writers' associate, Van de Sande Bakhuyzen, 3 found that aqueous solutions of starch, prepared by the method of the writers 4 without the use of reagents or heat, give a precipitate with alcohol consisting of clusters of microscopic needles, which when viewed between crossed nicols are birefringent, without showing the black cross.

The writers repeated Beijerinck's experiments and verified his results. However, by autoclaving at a somewhat higher temperature, 150° to 160° C, and by the use of some other modifications of his method, they obtained clusters of larger needles, not merely from potato starch, but also from wheat, maize, canna and arrow-root starches. These crystals are but little soluble in cold water, more readily in hot. They give the iodine reaction.

Viewed between crossed nicols, the clusters are birefringent in all cases. In the case of wheat, maize, and arrow-root, the needles are very small. In the case of the potato and canna, they are relatively large—some of the clusters reaching a diameter over one-half that of the natural granules. The potato and canna clusters, moreover, show the black cross between crossed nicols about as plainly as the natural granules. In the canna clusters, the coni trast between the black cross and the white of the cluster is not so sharp as in the potato clusters.

Although the skin is not the main receptive organ for the bacteria of the pasteurella group, experiments conducted on guinea pigs have established the possibility of protecting the entire integumenturn of the animals inside of 18 hours, but not the subcutaneous tissues, when non-toxic filtrates prepared from seven days old cultures grown in shallow layers, containing a growth inhibiting substance or “antivirus”, are inoculated intracutaneously.

1. The condition shown in Fig. 1 represents a non-inflammatory type of parodontoclasia in a dog.

2. The occurrence of the retrograde changes in Fig. 1 in relation to a diet known to produce osseous degeneration, together with a reversal of the type of bone change with a reversal of the dietary fault, postulates that nutrition is an important etiologic factor.

3. The microscopic appearance of the tissues shown in Fig. 1 resembles diffuse atrophy in human tissues as described by Gottlieh. 4

4. The condition described as resulting from the basic diet cccurred without any obvious degenerative changes in the long bones and in an animal apparently in good health, robust and well-nourished. This is in conformity with the occurrence of parodontoclasia in humans who seem to be in perfect health.

5. The freedom from inflammatory reaction of the soft tissues may be due in part to the high resistance to infection char-acteristic of the tissues of the dog, and in part to lack of function. The dogs were fed on pap, requiring practically no mastication. It is possible that vigorous use of the tissues in which the skeletal support was so defective would lead to trauma, hemorrhage, and invasion of micro-organisms.

6. The bone changes represented in Fig. 1, namely, resorption of the crest of the alveolar septum, lateral excavations, opening of the bone marrow spaces to the side, osteoporosis, coalescing of marrow spaces and widening of the sockets of the teeth crownwise, all compare with the appearances familiar to parodontists in roentgenographs of human parodontoclasia.

Although a fairly large number of determinations of the blood volume have been made in cases of chronic nephritis with hypertension, very few are reported in cases of parenchymatous nephritis (nephrosis). Plesch, 1 using his infusion method, reported one case in 1909 in which the blood volume was about half the normal amount. Using the carbon monoxide method, Plesch 2 reported in 1922 a less marked decrease in the blood volume in one patient. In this paper, he also reported two cases of chronic nephritis with edema, which showed increasing blood volumes with decreasing edema. Linder, et al., 3 used the dye method on a number of cases of nephritis. They found slightly low blood volumes in two cases of parenchymatous nephritis.

The following cases are reported because of the strikingly low figures found. In all but one case, more than one determination was made.

Blood volumes were determined by the method of Keith, Rountree and Geraghtye 4 Three of the cases were typical instances of parenchymatous nephritis in every respect. One case (No. 2) was not entirely typical as the N. P. N. was elevated and blood cholesterol was normal. Otherwise the Mood of all the cases showed high cholesterol concentration, and only slight or no elevation of the non-protein-nitrogen. The urine of these patients was concentrated, showed large amounts of albumin, casts and only occasional red blood corpuscles. There was no elevation of blood pressure. Patient No. 2, on admission, was suffering from pneumonia, as well as chronic nephritis. This patient's urine was of high specific gravity, showed large amounts of albumin, but no red blood corpuscles. His whole blood non-protein nitrogen concentration was 115 mg. per 100 mm. and the whole blood cholesterol was 155 mg. per 100 mm.

The influence of posture on the renal output of water, bicarbonate, chloride, inorganic phosphate, inorganic sulphate, urea, ammonia, creatinine and titratable acid, on urine pH, on blood pressure, pulse rate, circulation rate and rate of metabolism has been studied. The procedure was as follows: The subject took no food or water after ten p. m. of the evening preceding an experiment. At eight a. m, he emptied the bladder, discarding the urine, drank 200 cc. of tap water and stood or lay for two hours, at the end of which he voided, drank another 200 cc. of water and lay or stood for another two hour period. At the end of each two hour period he emptied the bladder and drank another 200 cc. of water. The order of lying and standing periods was varied in the various experiments. Each experiment was continued for four consecutive two hour periods. Nine experiments were performed on three subjects, four on subject H, aged 18 years, three on subject D, aged 30, and two on subject S, aged 50. The circulation rate for a ten minute period was determined by Henderson's and Haggard's method 1 in at least one lying and one standing period. During the period of circulation rate determination, ten arterial blood pressure readings were made, using the auscultatory and graphic (Erlanger instrument) methods simultaneously. Each blood pressure and pulse rate figure given in the table is thus the average of ten readings. No food was taken during the experiment and no water except the 200 cc. at the beginning of each period.

The circulation rate, blood pressure and metabolism figures are not presented because of lack of space. In every case the general circulation rate was greater in the recumbent posture and in all the experiments except those on subject S the mean arterial blood pressure was lower in the recumbent.

In 1904 Erlanger and Hooker 1 suggested that the product of the pulse pressure (P. P.) and the pulse rate (P. R.) might be used as an index to the circulation rate (C. R.), or, expressed differently, the pulse pressure as an index to the stroke volume (S. V.), provided changes in systolic time and elasticity of the arteries do not disturb too much the direct relation of the P. P. to S. V.

In order to ascertain what the relation is, four normal individuals were studied under four relatively normal conditions. The conditions studied were the recumbent and standing postures and light and post-heavy exercise. Arterial pressure determinations could not be made during heavy exercise; therefore, they were made immediately afterwards.

The arterial pressure readings were made simultaneously by the oscillatory (Erlanger) and the auscultatory methods. The circulation rate in liters per minute was determined by the Henderson and Haggard 2 method, involving the inhalation of ethyl iodide. Ten arterial pressure determinations were made to each observation of the circulation rate and the average of these are used as a basis of comparison.

When the S. V. is plotted against the P. P. it is found that a straight line passing through the points inaicating the mean values for standing and light exercise almost, if not quite, passes through zero and also the mean of heavy exercise, indicating that the relationship between these is one of direct proportionality. On the other hand the mean of the recumbent observation always falls to the left of this line. The recumbent pulse rate and diastolic pressure are in each case decidedly lower than in the other states. The longer systolic time of the slower P. R. and the reduced coefficient of elasticity associated with the lower diastolic pressure both would have the effect of reducing the P. P. produced by a given stroke volume, and both would, therefore, tend to put the mean of the recumbent reading to the left of the line passing through the mean standing and exercise readings.

Because of the lack of consistent data bearing on the action of adrenalin on the pyloric sphincter, a further investigation of the problem has been undertaken. The results so far obtained have been sufficiently constant that a preliminary report, indicating their general character, seems justified. The investigation has been confined to a study of the immediate effects of the drug administered intravenously to anesthetized dogs. Several methods have been employed to observe the reactions of the sphincter muscle without significant differences in the results.

In dogs that have been anesthetized with morphin and ether, and allowed an hour or more for recovery from operative procedures, adrenalin produces regularly a prompt relaxation of the sphincter muscle. After a variable interval, the duration of which depends on the dosage, the tonus recovers and may temporarily reach a level slightly higher than before the drug was given. On the other hand, if observations are made immediately after operating on the animal or if ether alone is used for anesthesia, the administration of adrenalin usually causes, immediately, a slight increase in the tonus of the sphincter.

No very pronounced changes in the tonus of the sphincter have been observed as a result of administering adrenalin. The greatest increase in tonus obtainea with adrenalin was much less than that which results from morphin, or from vagus stirnulation. When a decrease in tonus is obtained it is generally somewhat more pronounced, but is decidedly less than is seen in the small intestine when adrenalin is given.

Although Gruber and others have found that adrenalin may contract or relax gastrointestinal muscle, depending on the dosage, no such influence of dosage is apparent in the results of these experiments so far, but further study of this point seems desirable.

A study has been made of the excretion of free benzoic acid, hippuric acid, and total combined benzoic acid in the urine of normal adult rabbits following the administration of sodium benzoate. The maximum rate of synthesis of hippuric acid occurred after the ingestion of 0.5 gm. of benzoic acid per kilo. Larger doses of benzoic acid increased the rate of excretion of combined benzoic acid but not that of hippuric acid. In the 24 hour period following the administration of 1.0 gm. of benzoic acid per kilo, the average excretion of hippuric acid in six rabbits was 82 per cent of the combined benzoic acid, the individual variations ranging from 65 to 90 per cent. These urines contained an ether-soluble, non-fermentable reducing substance which gave a positive naphtho-resorcin test. Therefore, it was concluded that 10 to 35 per cent of the combined benzoic acid excreted by the six rabbits was benzoyl glycuronic acid. It has long been known that the rabbit excreted benzoyl glycuronic acid following benzoate administration, but it has generally been assumed that this substance was present only in traces. These experiments indicated that the extent to which the rabbit detoxi-cates ingested benzoate by conjugation with glycuronic acid has been underestimated.

Bylina, 1 studying pancreatic secretion, applied hot water (70° R.) to the gastric mucosa in order to produce a temporary achylia. An achylia, which lasted from six to seven days, resulted from one application.

It occurred to us that permanent achylia might be produced by application of hot water at intervals. Our method has consisted in injecting into the stomach via a gastrostomy from 250 to 300 cc. of water at 70° to 75° C, allowing it to remain in the stomach for one minute. This has been repeated at six to ten day intervals. The gastric secretory response to a meal, 1 mg. of histamine and 2 mg. of histamine, has been followed.

Our experiments have been under way for two months. We have found that the gastric glands neither respond to a meal, nor to 1 mg. of histamine. In some tests, in which we used 2 mg. of histamine to excite the gastric glands, one week after the last application of water, we obtained from 5 to 10 cc. of mucous which contained free acid. We find that it is difficult to maintain the nutrition of these animals, which difficulty we hope to overcome.

Our observations show that hot water, lower in temperature than that used by Bylina, can be used to cause a temporary achylia, and that the achylia can be maintained by intermittent applications of the hot water at from six to ten day intervals.

In previous papers 1 we have called attention to the fact that a distinct balance exists in the autonomic orientation between the peripheral region of the body and the splanchnic area. Muller 2 has studied the relation of this balance to the distribution of leukocytes, while in our own work with the lymph we have studied the alternations in permeability. In the present series of observations we have studied the changes in the blood sugar during malarial fever (artificial inoculation) with particular reference to the crisis.

In the fasting patient we have determined that the period of the chill and rise in temperature is associated with a decrease in the blood sugar level. This corresponds to the time of a peripheral leukopenia and a diminished concentration of blood protein. With the crisis and fall in temperature, blood sugar increases, leukocytes increase in the peripheral blood stream and the blood proteins increase. These pictures can be regarded as a result of a reversal of the autonomic balance. In the first stage we have a peripheral sympathetic tonus with a splanchnic parasympathetic orientation; in the period of defervescence, a peripheral parasympathetic and splanchnic sympathetic tonus. We cannot, of course, exclude the effect of metabolic utilization of the sugar. We believe, however, that the magnitude of the sugar mobilization during the period of defervescence would indicate a splanchnic sympathetic effect.

A dog with the tail of the pancreas transplanted under the skin, and a Thiry fistula of the jejunum was demonstrated. The continuous secretion of the pancreatic transplant was collected for two hours. It amounted to 0.4 cc. Normal twentieth hydrochloric acid was applied by perfusion to the mucosa of the Thiry fistula after washing it with warm 0.9 per cent salt solution. After a latent period of 5 minutes a copious flow of juice from the pancreatic transplant resulted, amounting to 0.5 cc. in twenty minutes.

This experiment clearly demonstrates that when dilute hydrochloric acid (from N/10 to N/20) is applied to the intestinal mucosa something (a hormone, we believe) is caused to enter the blood or lymph stream which excites the cells of the pancreas to form pancreatic juice.

This preliminary paper is devoted to a description of the technique and some suggestive results of the culture in vitro of the tissues of the flatworm, Planaria dorotocephala. Practical sterilization of the material has been accomplished by exposing the worms, in a small amount of water, to ultra-violet radiation from a quartz mercury arc lamp at a distance of 35 cm. during 4. minutes. After such an exposure the bacteria associated with the worms are practically all killed or prevented from reproducing, without injury to the worms. The hanging-drop, as well as the petri-dish technique were used for the tissue cultures.

Buffer solutions composed of the common physiological salts, in order to be tolerated by the Planarian explants, must be between 1/8 and 1/25 the concentration of chick serum. The optimum concentration of Planaria is 1/10 to 1/12 the chicken concentration. Intact worms, however, tolerate indefinitely all concentrations between well-water and buffer solutions containing from 1/4 to 1/5 the salt-content of vertebrate-isotonic solutions. The organ of osmotic regulation seems to be the external epithelium of the worm.

In hanging-drop cultures the cells survive in fluid media for as long as 10 to 15 days; they are pseudopodially active, and show some cell division, but not active, sheet-like proliferation. When support and tension are afforded to the explant by the addition of a drop of agar, a conspicuous sheet-like outgrowth of parenchyma may be seen within 12 hours. The cells will migrate along fine silk strands, but do not spread in sheets. Under certain conditions, cells in petri dish cultures may survive as long as 64 days.

Serum or tissue extracts of the following animals have been introduced into the culture media: tapeworm, clam, snail, isopod, frog, sheep, without injurious effect upon the explant.

Koppanyi, Ivy, Tatum and Jung 1 recently found that when the pancreas of the chicken is removed, a Von Mering-Minkowski diabetes mellitus is present for seven or eight days only. After this period the blood sugar returns to normal and sugar disappears from the urine.

Since it was not known whether or not the pancreas and other tissues of the chicken contained insulin, it was decided to assay some of the tissues of the normal chicken for insulin content. Such knowledge would be necessary before one could arrive at any interpretation of the insulin content of the tissues of depan-creatised chickens. Also, this question is involved in any explanation that might be offered for the above strange and interesting observation.

Through the co-operation of Dr. Jones and Mr. Templeton of the Research Division of Swift and Company, we were able to remove the pancreas, liver and kidneys from one hundred chickens within one hour after death.

Insulin was prepared from these tissues by Fisher's 2 modification of the Doisy-Shaffer 3 method.

Three hundred and thirty grams of pancreas, 240 grams of kidney, and 712 grams of liver were extracted, the liver containing a high percentage of lipins was more difficult to work with. After filtering off the toxic portion, the insulin was precipitated, dried, weighed and made up to the following volumes: 1.8 gram of precipitate from the pancreas was made up to a volume of 42 cc. 1.5 gram of precipitate from the kidney and 1.8 gram of precipitate from the liver were dissolved in 52 cc. of distilled water respectively. The solutions were preserved with cresole.

The work presented is part of a series of investigations of the metabolism of undernourished children of school age, who showed no pathological complications on medical examination, except that they were underweight for their height. The normals were not only normal in weight, but were healthy in every respect.

In this study the children were put first on a high and then on a low protein diet of at least 6 days each, and for the last 3 days 24-hour specimens of urine were collected and analyzed for total nitrogen, ammonia nitrogen, creatinine, and creatine. The nitrogen of the foods was also determined.

There is no marked difference in nitrogen absorption in the two groups. Both of them absorbed more nitrogen on a high protein diet, although the percentage absorption was lower. Ammonia excretion was decidedly higher in the undernourished children both computed in terms of mg. per kilo of body weight and in percentage of total nitrogen. This high ammonia excretion is probably due to the greater demand for alkali for neutralization of the excessive fatty acids produced in the intestines and eliminated as soaps in undernourished children. The alkali loss in the tissues is made up by ammonia formed in the body and excreted in the urine.

Creatinine excretion is independent of protein intake and there is no difference in the amount excreted by the two groups of children. On the other hand creatine seems to vary with protein intake in both groups, although the undernourished children excrete more creatine on low protein diet than the normals. The ratio of creatine nitrogen to total nitrogen output is remarkably constant. The higher creatine excretion in the undernourished children on a low protein diet suggests that they excrete more endogenous creatine.

The Mockade of the reticula-endothelial system of rabbits with saccharated iron oxide had no apparent effect on typhoid agglutinin formation.

Testes of rats have been exposed to graded doses of X-ray (determined physically and biologically) and studied over periods of time ranging from one day to six months. Doses up to ten erythema have been given at one time. The extent of degeneration produced in the seminiferous epithelium depends roughly upon amount of radiation. However, the degree of degeneration for a given dosage after a given number of days is not constant. Actively dividing cells, spermatocytes and then spermatogonia are the most easily injured. Spermatozoa disappear for a short period of time about ten days after exposure. Their absence marks the time required for conversion of spermatogonia into spermatocytes, these to spermatids, and the latter to spermatogonia.

Doses exceeding 3 erythema cause aspermatogenesis with degeneration of all seminiferous epithelium excepting so called “Sertoli” cells. In all experiments with exposure up to seven and one half erythema these cells have persisted. Doses of eight erythema and over have resulted in death. The persisting cells gradually change in structure and from them are produced spermatogonia. Cell types intergrading between these so called “Sertoli” cells of degenerated testes, which are in realtiy Indifferent cells, and spermatogonia have been found in these experiments. Regeneration takes place from these persisting Indifferent cells. They are, therefore, essentially germ cells. Within one month considerable recovery has taken place and in three or four months the testes are again normal. The time required for regeneration following extensive injury is about three months.

Since germinal epithelium has not been completely destroyed by radiation, as evidenced by cell division and regeneration, it cannot be said that this tissue does not take part in producing the sex hormone in X-rayed animals.

In a previous paper 1 it has been shown that single-cell strains of paratyphoid bacilli of the R and S types can be converted one into the other by suitable treatment. The S type so obtained is more stable in salt solution and more virulent than the corresponding R type, and also possesses the easily recognizable character of producing smooth, projecting colonies on an agar plate.

Rabbits have been inoculated with single-cell strains and the serum tested for agglutinating qualities. The following results have been obtained. With a single-cell smooth strain of the Schottmuller type (209 S) the serum agglutinates the homologous organism in about 1:500 dilution with a heavy flocculent precipitate. It also agglutinates other Schottmuller strains in the same manner in similar dilution. Its action on rough strains is not quite the same. Although a smooth strain serum, it agglutinates all rough strains of the Schottmuller type in higher dilution than it does the corresponding smooth strains. The precipitate, however, is finely granular instead of flocculent. It also agglutinates at least some smooth strains of the Aertrycke type in nearly as high dilution as the Schottmüller strains, but the precipitate is always granular.

The serum obtained with a single-cell rough strain (209 R) behaves very differently. In no instance has a “rough” serum agglutinated a smooth strain of its own group. All R strains tested, whether Schottmüller or Aertrycke type, are agglutinated by the R serum, usually in about the same dilution and always with the formation of a granular precipitate. A flocculent precipitate was never produced.

In two experiments S cultures derived from single-cell R strains by the method previously described 1 have been used. Both are of the Aertrycke type (223 and 239). The results are as follows:

Acute cocaine poisoning in the rabbit causes significant respiratory derangements in such a way as to lead to death of the animal. Since artificial respiration constitutes a very satisfactory method of treatment in this animal and since Richet 1 and Feinberg 2 found independently that decortication as well as decerebration raised the minimal lethal dosage of cocaine, it was thought that possibly other parts of the nervous system might contribute to the harmful effects of cocaine. This report, therefore, involves a study of the possibility of vagus involvement.

Rabbits were anesthetized with ether, double vagotomy and tracheotomy performed. After recovery from the anesthetic the animals were poisoned by subcutaneous injections of cocaine hydrochloride. It was found that this procedure raised the average minimal fatal dose from 100 to 125 mg. per kilogram in the intact animal to approximately 175 mg. per kilogram in the operated animal. Tracheotomy alone was sufficient to be equally efficacious, whereas vagotomy alone was without effect in modifying the average minimal fatal dose.

These facts are interpreted to mean that in the intact rabbit, respiratory embarrassment must exist, either through spasm of the glottis or other obstruction, in such a way that the asphyxial element thereby induced very markedly contributed to the central injury produced by cocaine.

The studies of Swoboda 1 in this laboratory indicate that cystin added to certain media stimulates yeast growth. This suggested a more intensive study on the actual utilization of various forms of sulphur in yeast growth.

The forms of sulphur studied were sulphate, sulphide, cystine, cystein, cysteinic acid, taurine and taurocholic acid. These were added to the usual synthetic medium as employed by Williams, 2 Miller, 3 and Swoboda, 1 with and without the “biose vitamine” and without the usual sulphate content. The yield of yeast was determined by weighing the yeast obtained after a growth period of eighteen hours. The saccharose and asparagin used in the media were purified to remove sulphur containing substances as completely as possible.

The results obtained were as follows:

1. Inorganic sulphate is the best form of sulphur for yeast growth, especially if magnesium is present with an ample amount of the biose vitamine. Neither magnesium nor sulphate could be substituted for the biose vitamine.

2. Sulphide sulphur added as hydrogen sulphide in concentrations of 0.0022 mg. per 125 cc. may be used as a sulphur source by growing yeast. In higher concentrations it retards yeast growth, especially when a good sulphate containing medium is employed.

3. Cystine stimulates yeast growth slightly in concentrations of 1 to 4 mg. per 125 cc. medium. In higher concentrations and especially in a good sulphate containing medium it retards yeast growth. Cysteine behaves similarly. Possibly the harmful effect of these is in part due to their partial conversion into hydrogen sulphide.

4. Cystine sulphur is partly utilized in synthesizing new yeast protein, but a part of it also remains in the medium as sulphate.

5. In otherwise sulphur free media cysteinic acid stimulates yeast growth very slightly, but in a sulphate containing medium it retards the same.

The question of the identity of pepsin and rennin has been under consideration for many years, Pekelharing, 1 Pawlow 2 and others, holding to the view that pepsin and rennin activities were not distinct, with Hammarsten, 3 and more recently Fenger, 4 claim that the two activities are different in character.

The study made in this laboratory was carried on in order to help bring further proof as to the identity of these two activities.

A very active preparation of rennin was made by the method of Fenger. 4 The dried product from this method of preparation showed on the average a milk-curdling power of 1:1,000,000 in ten minutes on fresh pasteurized milk at 40° C. Reprecipitation of this material gave a final product 41.66 times as active as Armour's commercial rennin. It was more active than could be consistently prepared by the sodium chloride salting out process and it had the advantage of being free from absorbed sodium chloride. The product can be kept for a long time without losing activity.

Studies were then carried out in order to determine if a more active product could be prepared. This was attempted by absorbing the rennin activity from the calf preparation, on solid purified casein suspended in solutions at various pH values. In these studies it was observed that the rennin activity could be absorbed to the extent of 90 to 95 per cent when the pH value of the solution was from 4.4 to 4.6. This was shown to be true by testing the filtrates, after absorption, for milk curdling power. It was also observed that the absorbed rennin activity could be recovered from the casein to the extent of 75 per cent to 80 per cent by bringing the solution to a pH value of from 6.4 to 6.6 by means of NaHCO₃.

It is a well known fact that the normal position of the head is maintained through impulses coming mainly from the otoliths (labyrinthine righting reflexes on the head). An animal with both labyrinths destroyed will not be able to preserve the normal posture of the head when held in the air in various abnormal body positions. According to Magnus, 1 the cat, the dog, and the monkey are exceptions to this rule, in that through visual impulses they are still capable of righting their head even after bilateral labyrinthectomy.

In our experiments we have employed normal puppies, rabbits, guinea pigs, fowls and pigeons. All these animals are capable of maintaining the normal posture of the head, as indicated by the direction of the snout or beak, when held in the air in the side position or in the supine position. The latter involves a very sharp ventroflexion of the head, the snout or beak being turned towards the tail. With one eye kept closed by pasting a strip of adhesive tape over it, all these animals are capable of preserving the normal posture of the head when held in the air in the lateral position in such a fashion that the open eye is uppermost. If, however, the closed eye is uppermost, the head is held in the normal position only for a few seconds and then gradually begins to sink sideways, until it assumes a position characteristic of an animal without labyrinths. The difference in the two lateral positions is especially striking in the case of the fowl.

When the animal is completely blindfolded (by pasting adhesive tape on both eyes) it loses completely the symmetrical righting reflexes on the head when held in the air in the supine position.

Corymorpha palma is a large, unbranched tubularian hydroid abundant on the Southern California coast. Many different lines of evidence show that in this form, as in others, the physiological axis is primarily represented by a quantitative differential, a gradient, in metabolic activity and associated conditions in the protoplasm. From the most active or 'high' end of this gradient the apical structure, the hydranth, develops and from lower levels the stem and base.

The forms resulting from the reconstitution of isolated pieces of the stem of Corymorpha differ with length of piece, level of body, physiological age and condition of animal, etc., and also with differences in various external conditions. The forms produced fall into three chief groups: (1) uniaxiate, a single individual or an apical part, which may develop from either distal or proximal end of the piece, or under certain conditions, from the side; (2) biaxiate, with an apical end developing from each cut surface and other parts in order from each apical end, as far as the length of the piece permits; (3) intermediate forms, biaxiate as regards the apical ends, but with one or more proximal or basal regions developing from the side of the stem between the two apical regions. In lots of similar pieces from similar stems the frequency of uniaxiate and intermediate forms is much higher in pieces allowed to lie undisturbed on the bottom of the container than in pieces supported above the bottom on loose absorbent cotton. The frequency of uniaxiate and intermediate forms is higher in pieces undisturbed on the bottom than in pieces moved about and turned over every few hours. Pieces above a certain length lying on the bottom and subjected to the action of depressing agents, e. g., alcohol, LiCl, chloretone, ethyl urethane, give a higher frequency of uniaxiate and intermediate forms and in such forms a higher frequency of large proximal parts and multiple basal ends, than pieces in well aerated water.

The subject of this investigation was a child of thirteen years who, at the age of four, swallowed strong acids, and in consequence developed almost complete cicatricial stenosis of the esophagus, the upper end of the occlusion being about 5 cm. below the level of the sternum. During the nine years the child had taken nourishment almost exclusively by a gastrostomy tube. For months at a stretch not even a drop of water could pass through the esophagus into the stomach. Occasionally an opening of 2 to 3 mm. diameter appeared, allowing water and milk to pass from the mouth into the stomach. Evidently a chronic spasm of the injured portion of the esophagus had developed on the top of the mechanical stricture. During an investigation of possible means of controlling the esophageal spasm, the child frequently swallowed small quantities of barium milk while fluoroscopic observations were being made. There was little or no permanent dilation of the end of the esophagus above the occlusion. The presence of the barium milk in this region led at once to vigorous movements of the esophagus resembling the movements of the small intestine above an obstruction, as originally described by Cannon. 1 One could make out local rings of constriction (segmentation movements) at different levels, as well as regular peristalsis. There frequently appeared vigorous antiperistalsis, forcing some of the barium milk towards and even into the mouth. The esophagus was able to empty itself completely in this way. We could secure no evidence that the mechanism was under voluntary control. The antiperistalsis did not induce any special sensation, such as nausea. Material in the upper end of the esophagus was felt as “something stuck in the threat,” whether or not antiperistalsis was present.

This investigation was undertaken in the hope that light would be thrown on the subject of desaminase action by the various tissues of the body. The subject is in a more or less unsettled state, some workers 1 even disclaiming a true deaminizing action in tissues, believing that the ammonia production is due to deami-dase action. Still others 2 take the view that the amino acids instead of yielding ammonia, are attacked in the carbon chain itself, thus being broken down and oxidized, the products formed producing cyanic acid which can be converted into urea by the addition of ammonia formed by deamidase action.

In this work an attempt was made to study the action of ammonium salts, or of mixtures of amino acids and peptides, or of amino acids alone upon ammonia and urea formation or utilization by tissues in vitro.

The tissues used were obtained from recently killed dogs or from the abattoir, in this case using hog tissue. These were minced as soon as possible, mixed with a buffer phosphate Solution and after an hour the juice was filtered and pressed out. Of this well mixed fluid, 25 cc. portions were taken and incubated with the additions referred to. All the necessary control estimations were made and the methods used were critically studied before using them on the problem. The tissues studied were liver and kidney.

The main observations made are as follows:

1. Ammonium salts added to the tissue extracts inhibit ammonia production. The chloride is most inhibiting. The lowered ammonia production is usually accompanied by increased urea production. Thus, in four experiments ammonium as acetate, phosphate or lactate stimulated urea production, while in two experiments all the ammonium salts retarded urea formation, the phosphate being the least inhibiting.

The statements in the literature regarding the hemodynamic effects of epinephrine injected hypodermically are few and conflicting. Some investigators 1 report feeble pressor effects; others 2 are as certain that there is no effect unless perchance epinephrine is injected quite directly and accidentally into a small venule. Even those who report feeble positive effects (Meltzer and Auer) were probably mistaken in the facts and interpretation, as we will point out more specifically in our detailed report.

In the course of some work in which epinephrine was administered subcutaneously in dogs under light paraldehyde (and morphine) anesthesia an appreciable rise of blood pressure was noted during the 3 to 4 minutes following the injection. Repetition of the experiments yielded questionable results; for there was either no immediate rise in blood pressure following such injection or the progressive rise in blood pressure could be readily explained on the basis of recovery from the light paraldehyde anesthesia of the animals. However, it was soon noticed that gentle massage of the injected area effected an appreciable hemodynamic effect and this led to a systematic investigation of the phenomenon.

Four to twelve kg. dogs were used for the most part under light paraldehyde anesthesia (1 to 1.5 cc. per kg.) reinforced as found necessary by subcutanecus injections of morphine sulphate (1/4 to 3/4 gr.). Blood pressure records were taken from the carotid artery. Hypodermic injections of epinephrine were made in a great many regions of the body in doses of 1/2 to 6 cc. (Parke Davis & Co.) in original dilution or more or less diluted with distilled water. In most instances the area so injected was gently or vigorously massaged immediately after injection and at various time intervals following injection.

If a definite amount of an unabsorbable compound, such as ferric oxide, be mixed with a diet and the feces be analyzed for iron and for any food constituent, such as calcium, it is possible to determine the percentage absorption or “utilization” of the latter without an accurate separation of the feces. If the ratio of calcium to iron in the food be 10:1 and in the feces 5:1, calcium unabsorbed amounts to 50 per cent and calcium abserbed would of course also be 50 per cent.

Using this method, it has been shown that lactose in contrast with other common sugars, markedly increases the absorption of calcium. Lower fatty acids have a more favorable effect than higher fatty acids. Antirachitic substance produces the expected result.

In the same way, if animals fed an iron containing diet are killed at the height of digestion and analyses made of the intestinal contents at different levels, the percentage absorption of food products in different parts of the intestines may be estimated.

Experiments of this type on rachitic albino rats on diets with or without cod liver oil give interesting results. Using a calciumhigh phosphorus-low diet rats with or without cod liver oil show considerable calcium absorption in the upper small intestine. This type of rickets is not due, therefore, to a failure of calcium absorption. Excretion of absorbed calcium into the lower bowel does, however, bring about a negative or subnormal calcium balance.

Phosphorus, on the other hand, is actually excreted into the upper intestines. Animals given cod liver oil make up for this loss by absorption from the cecum and large intestines. Rachitic animals do not and, therefore, the phosphorus balance remains negative.

On a milk diet, high in both calcium and phosphorus, both are absorbed to a considerable degree in the small intestine.

The great variation in the susceptibility of different animal species to oral injections of botulinum toxin may conceivably be due either (1) to the destruction or adsorption of the toxin by the intestinal contents, or (2) to the difference in permeability of the intestine to the toxin. Type A botulinum toxin was used in these experiments.

There was no evidence that botulinum toxin was adsorbed by the intestinal contents of guinea pigs in vitro, even where the pH was shifted with M/3000 HCl or Na₂HPO₄.

A loop of small intestine in each of six rabbits was injected with botulinum toxin and perfused with blood from the same animal for intervals varying from 30 minutes to 2 hours and 20 minutes. Very small quantities of toxin were demonstrated in the perfused blood, often only sufficient to produce symptoms of botulism in mice receiving 0.5 cc. quantities. No toxin was ever demonstrated in 0.1 cc.

In each of two hogs a loop or small intestine was perfused in a similar manner. In one case a very small amount of toxin was found in the blood, a mouse receiving 0.5 cc. of serum taken after an hour and a half of perfusion died in 4 days. In the other case toxin was not demonstrated. The toxin introduced into the ligated loop of intestine showed little if any decrease in potency during the course of the perfusion experiments.

Hogs were found to be very resistant to large oral doses of toxin; in some cases as much as ten million M. L. D.'s for mice were fed without producing any ill effects. Toxin was not demonstrated in one cc. of blood taken after 90 minutes from a hog which had been fed nine million (mouse) M. L. D.'s.

By total urinary phenols we mean substances which react with the reagent in the Folin and Denis colorimetric method. The subject studied in this experiment abstained from meat, fish, coffee, tea, alcohol and tobacco; an average figure was taken for a four day sample during such weeks in which there was no sickness, accident or vacation.

Although the level for 1923 is higher than that for 1924, yet, within each year we find about 100 mg. more phenols excreted per day during the summer months of June, July and August, than during the winter—meaning December, January and February; the spring and autumn show intermediary figures, and those for the autumn would probably have been higher had not September been vacation-month. If we combine the two years of 1923 and 1924, we find the following:

Spring, 60 days, average 411 mg.;

Summer, 71 days, average 463 mg.;

Autumn, 60 days, average 388 mg.; and, finally, Winter, 84 days, average 345 mg.; or for 3 winters, 124 days, average 353 mg. of phenols.

As to the cause of such a seasonal variation we at first thought that the analyses were perhaps carried out at somewhat higher temperatures during the summer and that this might be a factor. Tests showed that a standard set at 20° Centigrade compared with a similar standard set at 30° Centigrade gives an increased color in the latter equivalent to a 20 per cent error; however, a certain sample gave 333 mg. of phenols at 30° Centigrade and 338 at 20° when compared with the same standard at the corresponding temperatures. We always kept the standard and the unknown at the same temperature, so this error was not introduced.

Secondly, because the phenols in an exaggerated manner followed the Nitrogen fluctuations of 1924-25, we have included the records of the winter 1925-26, which demonstrates that Nitrogen is not the main factor; for with 12.66 g of N per day for the winter against 21.75 for the summer, the phenols remain low.

It has long been known that virulent strains of pneumococci or other bacteria are not readily ingested by leucocytes and that avirulent strains are generally ingested readily. It has also been established that the presence of acids, alkalies, salts and other reagents modify the velocity or the extensiveness of phagocytic reactions.

In studies on the parallel relations between virulence, electro-phoretic potential difference (P. D.), agglutinability and other characteristics of bacteria, strains of pneumococci of different virulence for white mice were examined in considerable detail. 1 The pneumococci designated A are virulent Type I organisms. The strains B and C are variants of A which were derived by Blake and Trask by growing strain A in the presence of specific anti-serum. The minimum fatal doses for white mice are: 0.5 × 10^-7 cc. for A; 0.5 × 10^-3 cc. for B; and 0.5+ cc. for C. The P. D. is highest on A and lowest on C.

The experiments reported here were undertaken to determine the extent to which A, B and C strains of pneumococci are taken up by phagocytes under conditions which were designed to decrease or increase the P. D. It has been established 1 that the virulence for white mice is altered in a parallel manner when the P. D. is increased or decreased.

It has been found that in the presence of normal rabbit serum or of immune horse serum, the sequence of decreasing phagocytic indices is: C, R, A. For the opsonic indices, the series is: A = E > C. The enhancement of phagocytosis by the presence of inheated or heated immune serum is approximately proportional to the concentration of the serum.

It has been found' that the sequence of the relative magnitudes of the P. D. values for the A, B and C strains is inverted by frequent washings with water from A > B > C to C > 13 > A. It was found, correspondingly, that the sequence of the phagocytic indices was also inverted.

Measurements of electrophoretic velocities on strains of pneu-mococci and other organisms indicate that many vital phenomena are closely associated with the electrokinetic P. D. between the organism and its menstruum.

It has become evident from the work of Falk and his associates 1 that there are significant parallelisms, direct or inverse, between electrophoretic potential, virulence, agglutinability and other characteristics of bacteria. For pneumococci of the types 1, 2, 3, 4 it has been found that the P. D. is higher the greater the virulence for white mice and vice versa.

The work here reported was undertaken to determine whether or not P. D. would parallel virulence of various strains of diphtheria bacilli as well as to obtain comparable data in regard to P. D. on avirulent or pseudodiphtheria bacilli.

The cataphoretic potentials on corynebacteria were determined by the method described and used by Falk, Gussin and Jacobson. 1 For reasons discussed by them results are expressed in terms of observed velocities (μ/sec) instead of expressing the electrical P. D. at the interface between the bacterium and the menstruum in millivolts.

P. D. in observed velocities (μ/sec) may be converted into millivolts by multiplying by the factor 1.3. Ten measurements of velocity (5 with one and 5 with reversed orientation of the electrical field) were made at stationary layers V_s133, 497, in the same medium in which the bacteria were grown. The same methods were used in measuring velocities of cells in distilled water.

The electrokinetic P. D. on 65 strains of corynebacteria parallel the virulence (toxigenicity) of these bacteria. Measurements of potential on 48 hr. veal broth cultures of toxigenic strains give potential differences of 4.2 to 1.0 (μ/sec) with an average of 3.4 (μ/sec).

In earlier publications 1 we discussed at length the parallel relations between the virulence and the electrophoretic potentials of pneumococci. We reported that variant strains (Blake) of type I pneumococci, which differ in their virulence for mice, also differ in a parallel manner in their electrophoretic potentials.

In the studies reported here we have undertaken to determine whether “rough” colony varieties could be produced from the Blake Type I pneumococcus (designated as A) and its variants (designated as B and C respectively). These variants are derivatives of the A strain, and were obtained by Blake and Trask by growth in the presence of specific anti-serum. We have also undertaken a series of experiments to determine changes in virulence and P. D. upon successive passage through white mice of cultures of significantly different virulence and P. D.

We have found that the sequence of decreasing virulence for white mice, decreasing P. D. and increasing agglutinability is: A, B, C. The A, B and C strains form “smooth” colonies on peptone, serum or blood agar plates. Strains which give “rough” colonies on blood agar plates have been produced from the A, B and C cultures by growth in broth to which specific anti-serum had been added. After 23 transfers in serum broth, none of the cultures were completely converted to the rough varieties. Organisms of both the S and R varieties could be recovered. Some of the S varieties, which were recovered after 23 transfers in broth containing anti-serum, showed the virulence and the electrophoretic potential characteristic of the original cultures. Some of the S varieties show reduced virulence and potential. The rough varieties recovered after 12 transfers of B and C strains in broth plus anti-serum showed the same virulence and potential as the original B and C cultures.

This is the sixth of a series of studies on radiation, four of which have concerned themselves with differential modification of embryonic development. The previous work includes experiments on Arbacia eggs and sperm, and on Fundulus eggs. The results in all the experiments agree in their more general features, namely, those regions which at the time of exposure have the highest rates of physiological activity, are the most readily modified in their development. Such modifications may be brought about by exposures made at various intervals after fertilization, as well as before fertilization. Experiments with hens' eggs yield the same general results, when exposed before and after incubation.

The complete spectrum of the Hg vapor arc (running at 110 v. D. C. at a distance of 10 inches) served as a source of radiation. Eggs were exposed at intervals, up to 64 hr. after incubation, as well as before incubation. Since it was found that exposures through the shell were entirely without effect, it was necessary to remove a small part of the shell and underlying membrane, and thus expose the blastoderm directly to radiation through the very thin layer of albumen which covered the surface of the yolk of the egg. Such a thin layer does not screen off the effective radiation. Exposures were from 1/5 min. to 10 min. in duration. It was found that later stages of development required less radiation to produce a given result, and would be killed with doses of 5 to 10 min. which produce only developmental modifications when exposures are made Before incubation.

Before incubation, or reincubation, the piece of shell which had been removed was replaced, and the egg sealed with paraffin. Development was usually allowed to continue for three days, although some eggs were removed from the incubator after 24 or 48 hours.

A study of the problem of organic precursors of the hematin part of hemoglobin shows that no attention has been directed toward the possible relation of vitamines therein, whether this be as a direct precursor, or as involved in the use of such possible precursors as chlorophyll.

The problem was to determine whether rats could be rendered anemic by the absence of one or more vitamines in a synthetic diet and whether phaophytin, obtained from chlorophyll, functions as a hemoglobin precursor in anemic rats, however this anemia may have been brought about. Rats kept upon such diets were systematically followed as to weight, blood hemoglobin and red blood cell count. The tentative conclusions arrived at as applied to rats are the following:

1. Synthetic rations containing 18 per cent casein and deficient in Vitamines A, B and E do not produce anemia but do produce an unusually high hemoglobin content.

2. Phaophytin may not be substituted for any one of the Vitamins A, B, or E.

3. Casein at a 10 per cent level produces a lower hemoglobin content than does casein at an 18 per cent level.

4. Casein extracted with acetic acid and aerated is less efficient in maintaining hemoglobin than is casein extracted with lukewarm alcohol.

5. Casein is much more efficient in the maintenance of hemoglobin than is gluten.

6. Experimental anemia may be produced on synthetic rations using wheat gluten as the protein and this anemia may be at least partially and temporarily alleviated by the aadition of 0.2 to 1 per cent phaophytin to the ration.

7. It appears that proteins contain an important precursor of hematin and that since casein contains this in greater quantity than wheat gluten, the indications are that possibly tryptophane is involved.

The main problem attacked was the question of a difference in the endocrine secretions of the testes of the Sebright and the Leghorn. The Sebright male is so-called “hen-feathered” as it has a feathering similar to that of the female, while the Leghorn male is male-feathered. The work of Morgan 1 (and later of Eliot, unpublished) shows that the castration of the male Sebright is followed by the formation of capon feathers (usually called male feathers); thus the testis of the Sebright, like the ovary of the hen, and unlike the testis of the Leghorn, inhibits the formation of male feathers. Morgan concluded from this result and from later histological studies, in conjunction with Miss Boring, 2 that the Sebright and the Leghorn testes have different endocrine secretions, and that the Sebright testis contains cells indentical with the so-called luteal cells of the ovary, the presumed source of the internal secretion of the latter, which are responsible for hen-feathering of the male. Thus the Sebright was regarded as hermaphroditic in respect to its endocrine apparatus. The relation of the so-called luteal cells of the testis to henfeathering, however, has been denied by M. s. Pease, 3 Goodale, and Nonidez, 4 who arrived at the conclusion that there is no morphological difference between the testes of different breeds to account for a presumed difference in their endocrine functioning. This led Prof. F. R. Lillie to suggest that the problem be attacked experimentally. The work was done on 10 Autotransplantation in the Leghorn, 2 Autotransplantation in the Sebright, 50 Cross-transplantation of Sebright testis into the Leghorn capon, 38 Cross-transplantation of Leghorn testis into the Sebright capon.

Autoplastic grafts grow readily in both breeds, and, when sufficiently developed, substitute completely, both physiologically and psychologically, for the normal glands.

The recent announcement by Iwama 1 that a considerable number of medullated fibers from the vagus enter the sympathetic trunk near the inferior cervical ganglion, especially on the right side, and that some of these fibers run into the abdominal viscera via the splanchnic nerves, immediately brings into question considerable visceral neurology based upon experimental stimulation of the splanchnic and the vagus in the cat, at least. According to Iwama's observations it is impossible to stimulate the right splanchnic of the cat without also stimulating some vagus fibers.

It also occurred to us that there might be enough variability in number of vagus fibers that reach the gastric regions through the splanchnic nerves of different species of animals to be a factor in the results obtained by Carlson and others 2 following stimulation of the peripheral end of the splanchnic, such as only inhibition of the cardia in the rabbit and either contraction or inhibition in the cat depending on the state of tonus at the time of stimulation.

The preliminary work necessary to get this question upon a quantitative basis gave results at such variance with those expected that this note is deemed advisable although only four cats and two rabbits of our series bear directly on the subject.

Section of the right vagus in the mid-cervical region of the rabbits produced no Wallerian degeneration in the thoracic sympathetic trunk and hence none in the splanchnic nerves.

In three cats (right vagus cut in two and left vagus in one) results were similarly negative. In a fourth cat, however, with right vagus cut above the anastomosis with the inferior cervical sympathetic ganglion, two degenerated medullated fibers were clearly identified in the right thoracic sympathetic trunk above the origin of the splanchnic nerve.

Desiring to obtain some data on the output of certain urinary constituents in dogs during short intervals of time, we thought we could get rid of the influence of anesthetics by decerebration. However, we were greatly surprised to find that on decerebration, the flow of urine ceased, although there had been a good flow previously. There was no material alteration of blood pressure and respiration was kept up. Subsequent investigation showed this to be a constant phenomenon. Under normal conditions, after decerebration, there are a few drops of urine at the previous rate, and then a marked slowing in the rate amounting in most cases to a stoppage, and lasting as long as the experiment, in one case, seven hours after decerebration.

In our experiments, dogs have been used exclusively. They were given morphine and ether. Cannulae were inserted into the ureters close to bladder and the flow of urine recorded by allowing the drops of urine to fall on a lever which closed a circuit and thus recorded the flow by means of an electromagnet. Blood pressure was recorded from the left carotid, and the right carotid and the two vertebrals were usually ligated or clamped. The skull was trephined and the urine output recorded this way for a time. A few results may be given. Thus one kidney which produced 75 drops of urine in the thirty minute period previous to decerebration, produced 9 drops in the following thirty minutes. Another kidney which produced 138 drops in 30 minutes previous to decerebration, produced only 24 drops during 30 minutes following decerebration. Another kidney which formed 49 drops in 15 minutes previous to decerebration produced only 9 drops in the following 35 minute period.

When a dye solution is run into the frog's kidney in such a way that, in the lower portion of the kidney, it runs into tubular capillaries and into a large number of glomerular capillaries, while in the upper portion of the kidney it runs into tubular capillaries, and into only a few glomerular capillaries, the amount of dye excreted follows the number of active glomeruli receiving dye solution, although both halves of the kidney are producing urine.

It seems reasonable to assume that a number of factors influence the size of the human individual at various stages of development. Pearson and his co-workers have shown the great importance of hereditary factors. They have found that in general such environmental factors as they have been able to consider play a much smaller rôle in determining the characteristics of the individual. Nevertheless it seems important to investigate all such factors in as great detail as possible, with a view to disentangling those factors which are innate in the zygote from which the individual develops from those which are extrinsic to the individual, and to measuring quantitatively the relative importance of each. It seems especially desirable to consider certain morphogenetic factors.

The purpose of this paper, which is one of a series dealing with the physical characteristics of the newborn infant of various races, is to consider the possible relationships between the age of the parents and the length and weight of the newborn child.

The data represent measurements on infants of the white race taken at the Sloane Hospital for Women, New York City, and recorded in the archives of the Obstetric Divisions of that institution. For opportunity of using these records we are indebted to the late Dr. W. E. Studdiford, who was Superintendent at the time this phase of the investigation was undertaken.†

Since it is possible that the average age of parents and the average dimensions of the newborn infant differ from nationality to nationality, we have considered separately the charactceristics of infants born of Austrian, English, German, Irish, Italian and Russian parents. Both parents were born in the countries designated. By the examination of about 43,000 records assembled over a period of about 30 years (September, 1890, to September, 1921), it was possible to secure the following numbers of measurements of length and weight.

In a preceding paper 1 evidences for the existence of a low positive correlation between the age of the parents and the length and weight of the newborn infant of various nationalities have been adduced. The present paper is devoted to a consideration of the relationship between pregnancy order and birth order on the one hand, and the length and weight of the infant on the other. The source of the data is the same as that of those employed in the preceding investigation.

By pregnancy order I understand merely the serial order of the pregnancy (Gravida). By birth order I understand merely the order of birth of the child (Para). These differ by the number of miscarriages which may have occurred previous to the birth under consideration.

In Diagram 1 and 2 the heavy bars in each of the four panels denote zero correlation, as shown on the scale of ordinates.

Examination of the correlations for pregnancy order and the lengths of boy and girl infants shows that in 9 of the 12 available series the coefficients are positive, whereas in 3 of the series they are negative in sign. For the relationship between pregnancy order and weight of child, all of the coefficients are positive in sign. Thus for both weight and length, 21 of the 24 coefficients indicate a positive correlation between pregnancy order and the weight or length of the child.

The average values of the coefficients measuring the relationship between pregnancy order and the characteristics of the child are:

Pregnancy order and length of son, $r = +.0914

Pregnancy order and length of daughter, $r = -.0117

Pregnancy order and weight of son, $r = +.2102

Pregnancy order and weight of daughter, $r = +.1393

Two organs of the human body undergo a remarkable reduction in size following birth, these are the uterus and the suprarenal glands. The neonatal loss in mass of the suprarenals seems to have been first noted by Scheel. 1 It is due in great part, if not entirely, to a degeneration of the inner and middle layers of the cortex. This process was first described by Starkel and Wegrzynowski 2 and, shortly after, independently by Thomas, 3 Kern, 4 and Armour and Elliott. 5

It has been pointed out that the neonatal loss in size of the uterus is preceded by a period of marked growth of this viscus in the latter part of fetal life, and that this neonatal involution reduces the organ to essentially the dimensions which would have obtained had the early fetal growth rate remained unchanged. Since the suprarenals undergo a neonatal involution which is concomitant with that of the uterus, it seems desirable to determine whether they also show a period of increased growth in the latter part of prenatal life.

The present study is based on observations on the weights of 1087 pairs of suprarenal glands; 425 of fetuses, 338 of newborn infants, stillborn or dying within 2 days after birth, and 324 of children over 2 days and under 1 year of age.

The course of growth of these structures is shown in the accompanying figure in which the suprarenal weight is plotted against the total body weight (as indicated on the base line scale) and the computed prenatal age and observed postnatal age, as indicated on the upper boundary line of the graph. The observations on fetal material indicate that the growth of the suprarenals before birth is proportional to the growth in body-weight for the relation between these weights is approximately rectilinear throughout the fetal period.

Eight to twenty months after complete thyroidectomy in the rabbit, well defined skin lesions appear on the dorsum of the feet, which have the characteristics of the lesion ascribed to pellagra. Pellagra-like lesions may also be apparent around the neck and in a circumscribed area about the mouth and nose. Xeropthalmia may also occur. These deficiencies appear in the hypothyroid rabbits on diets that are adequate for normal rabbits.

Rabbits and rats on an adequate diet and thyroidectomized 2 to 3 weeks after birth show disturbances in the development of bones during the growing period, which appear identical with the pathological processes ascribed to rickets.

Wenckebach, 1 in an article upon cinchona derivatives in the treatment of heart disorders, stated that “an intravenous injection of quinin may stop an attack of paroxysmal tachycardia in many cases.” This therapeutic suggestion was tested 10 times among 5 patients suffering from organic heart disease and paroxysmal auricular tachycardia, during a paroxysm. In all cases the effect of posture and the various forms of vagal pressure were first essayed. This was without avail in all. Continuous electrocardiography upon a lead selected from the diagnostic electrocardiogram was begun before the injection commenced and continued until 2 minutes after its completion. The injection occupied about 5 minutes. The dihydrochloride salt of quinin in 0.3 to 0.6 gram doses diluted to 10 cc. was used.

In six injections among four of these patients there was no alteration produced in the tachycardial rhythm governing the heart beat. In the fifth patient, the attacks came to an end with a single auricular premature contraction preceding the onset of the normal sinus rhythm, approximately one and one-half minutes after the injection was completed. This was repeated in three more paroxysms. This patient was having paroxysms lasting about one hour and a half four to six times a day. The injection of the drug did not lessen the frequency of the attacks. In one instance, not included among these ten, a paroxysm came to an end just as the quinin was about to be administered. This points to the caution with which inferences regarding the positive effects of therapy in this condition must be surveyed. In all of these patients no more paroxysms appeared after the oral administration of quinidin.

Rabbits thyroidectomized 2 to 3 weeks after birth have been allowed to develop in a hypothyroid condition for 10 to 16 weeks. At that time marked symptoms of cretinism are apparent. These rabbits have then received daily doses of dessicated thyroid by mouth or intravenous injections of Kendall's thyroxin, until the end of the growth period. If the thyroid treatment is then discontinued and the animal allowed to live for a long period of time, usually eight months or more, the posterior extremities become spastic. Reflexes are exaggerated and a slowly progressive paralysis occurs, which results in complete inability to use the hind legs. Histological study of the spinal cord shows degenerative changes. The blood picture is that of primary anemia.

Among 12 patients with varying types of heart disease presenting persistent premature contractions, including all the important etiologic, anatomic, and function combinations, epinephrin in a dose of 1 cc. in a solution of 1-1000 was injected hypodermically 18 times. Electrocardiograms of one minute duration on a previously selected lead were taken every five minutes; five before and five or more after the injection.

A considerable increase in the average number of premature contractions occurring per minute resulted in all cases beginning five minutes after the injection and enduring thirty minutes to one hour. Rise in blood pressure and slight increase in the heart rate in most instances accompanied this. Those with auricular premature contractions frequently developed ventricular premature contractions and vice versa, where they were not in evidence before the injection. Those with left ventricular premature contractions often developed right ventricular premature contractions and vice versa. Two or three ventricular premature contractions occurring together were not infrequent. In each patient with auriclar premature contractions, groups of four and five premature contractions in succession appeared. In one individual, free of premature contractions by means of quinidine, many appeared following it. In another, freed of premature contractions by means of digitalis, none appeared following injections of epinephrin. Apparent rhythmicity in the recurrence of the premature contractions often occur following injection. No variation from the basic normal sinus rhythm predominating ever occurred.

Singer 1 reported upon the favorable therapeutic effect of calcium, administered intravenously, in conjunction with digitalis in the conjestive type of heart failure.

Segall and White 2 concluded that calcium chloride may be employed as a diuretic in this type of heart failure.

Eight patients with congestive heart failure in whom rest in bed and digitalis did not completely remove the edema present, were treated with 15 grams of calcium chloride daily, administered orally, in divided doses, over periods varying from one week to 10 days. No significant changes followed in heart rate, blood pressure, electrocardiography, urinary output or body weight. In five of these, marked diuresis and significant loss of body weight with disappearance of the edema present followed the administration of other diuretics. In one of these, another diuretic was tried both before and after the trial of the calcium with excellent response at both times, though none occurred during the period of exhibition of the calcium. (This individual only became edema-free when treated with the combination of rest in bed, digitalis and diuretics. When the diuretics were withdrawn, the edema of the dependent portions of his body gradually returned.)

In two more patients with congestive heart failure of similar functional condition, given daily injection of calcium chloride in 0.5 gram doses, results similar to those obtained by oral administration of the calcium occurred.

(1) A female, age 23, with mitral stenosis and insufficiency of rheumatic etiology, predisposed to paroxysms of auricular tachycardia, never failed, when not previously subjected to the influences of other drugs, to have a paroxysm of auricular tachycardia following 2 injections of 1 cc. of epinephrin, 15 minutes apart. 1 When quinidin sulphate, 0.2 grams, twice daily, was administered previously, a paroxysm failed to appear upon exhibition of epinephrin in the same manner. This was performed twice. The first time after 6 weeks of this quinidin therapy, and 7 days later during which time no quinidin was administered, a paroxysm of tachycardia followed the exhibition of epinephrin as described. Fourteen weeks of quinidin therapy followed this and at the end of this time no paroxysm appeared after the epinephrin, yet five days later during which time no quinidin was administered, a paroxysm of tachycardia followed administration of the epinephrin.

(2) A male, age 65, with arteriosclerotic heart disease, subject to attacks of paroxysmal auricular fibrillation during a period of normal sinus rhythm, was similarly treated with epinephrin. Auricular fibrillation appeared with very little change of rate about 15 minutes after the 2nd injection of epinephrin. Quinidin was utilized to re-establish the normal sinus rhythm, which reappeared 12 hours after administering 0.2 grams every two hours. The dosage was then reduced to 0.2 grams four times daily. After four days administration of quinidin in this manner, during which time the normal sinus rhythm prevailed, epinephrin exhibited in the same manner was not followed by alteration of the rhythm controlling the heart beat. It appears as if quinidine offers a protection to the heart against the inception of paroxysmal auricular tachycardia and fibrillation induced by means of epinephrin.

As low protein diets have been recommended in treatment of diabetes by Joslin, 1 Petren, as quoted by Joslin, 2 Newburgh and Marsh, 3 and others, a study of long continued low protein diet was made in this case. The patient was a man, age 32, with a history of moderately severe uncomplicated diabetes for the past 2 1/2 years.

Observations extended over a period of 6 months. The diets were carefully weighed and the nitrogen in the food calculated from the Atwater tables. No analyses of food were made. Total nitrogen in the urine was determined daily by the Kjeldahl method. Nitrogen partition products were not determined. Nitrogen in stools was determined for two ten day periods by the same method. Daily weights were recorded.

The patient received a diet of 2000 calories with 50 grams of protein for 15 days before reduction of protein was started. The protein was gradually lowered by drops of 10 grams each until a 20 gram level was reached. Three or 4 days were allowed at each level for the patient to adjust himself to his new diet. The patient remained at this level for 106 days. The calories remained the same during the entire period. Carbohydrate and fat were changed from time to time so that the fatty acid-glucose ratio varied between 0.9 and 3.0. In this change the carbohydrate varied from 132 grains to 30 grams in successive steps, the calories being replaced by fat. The nitrogen excretion was not noticeably affected. This suggested that fat served to spare protein as efficiently as carbohydrate. Insulin was required during the entire period. There was no glycosuria. Moderate acetonuria appeared on fatty acid-glucose ratios above 1.5.

In studying the toxicity of carbon tetrachloride it was found that this drug in pure form causes very severe central necrosis of the liver which heals with scar formation. 1 , 2 The toxicity of a large dose of carbon tetrachloride (4 cc. per kilo) is greatly increased by the addition of alcohol. 3 Experiments were carried out to determine the effect of repeated doses of carbon tetrachloride alone and carbon tetrachloride given with alcohol in order to study the toxicity of such repeated doses and the effect on the liver. Ten dogs were used; some were given the therapeutic dose of carbon tetrachloride (3 cc), others 4 cc. per kilo, and others the same doses of carbon tetrachloride but with the addition of approximately 25 cc. of 50 per cent alcohol, and finally a control series was given the same dose of alcohol alone. The dogs were given these doses of the drug over a period of approximately sixteen weeks. No signs of intoxication in any of the dogs were seen. They maintained or gained weight and were killed for autopsy in apparently perfect condition. (One or two dogs died from unknown causes within the first day or two.) The livers of the dogs receiving alcohol showed nothing macro-scopically or microscopically. All the other dogs showed definite signs of an early cirrhosis of the liver. Macroscopically the liver showed a uniformly granular scarred surface in which there were numerous larger nodules and hyperplastic hepatic cells. Microscopically there was extensive scarring of the liver and distortion of lobulation by fibrous tissue. There was also abundant cellular infiltration and some regeneration of bile ducts. There was no apparent difference in the livers of the dogs receiving carbon tetrachloride alone or both carbon tetrachloride and alcohol.

During an investigation pertaining to the mineral deficiencies of milk, it was observed that the young, born of milk fed rats in February and March, 1926, were smaller, and their rate of growth was far less rapid, than that of any which we had heretofore noted in a long series of milk fed animals. These young were furthermore much smaller than young born of these same mothers during October and September, 1925. The appearance of the young in all groups (6) were so strikingly similar, even though the milk modifications were somewhat unlike, that it seemed as if the difficulty must be inherent in the milk, and not in the added inorganic substances. This was further emphasized by the fact that in two cases in which the mother rats were being used as controls, the animals were receiving only milk with the customary iron and iodine additions.

To determine what constituent of this particular “winter” milk was lacking, we reduced the number of young in each litter to four, and added to the milk rations of the mothers the following substances respectively: Ca₃ (PO₄)₂, Vitamin B preparation, Yeast, and cod liver oil. In no case did we get satisfactory growth. Slight gains were observed with three of the cod liver oil groups. One group receiving the cod liver oil addition died, as did also the group receiving the Ca₃(PO₄)₂.

Besides the small and spindling appearance of these young, it was observed that the skin had a peculiar pink, yellow color, and the fur was very thin and peculiarly wiry. In spots it was quite gone, giving the animals a moth-eaten appearance. A somewhat similar condition in the fur of animals receiving a purified ration, containing cotton seed oil for its chief fat component, seemed to have been corrected by irradiating the food mixture with ultra violet light from a quartz mercury vapor lamp.

Rats fed on diets adequate in all other respects but low in the Vitamin A factor, develop, in the course of time, infection of the upper respiratory tract. The first symptoms of the dietary deficiency are snuffles, loss of appetite, and failure to gain. These animals invariably die unless appropriate dietary treatment is introduced fairly early in the course of the infection. Macroscopic examination of the various organs of these animals, with the exception of the aural and nasal passages, which have been found filled with pus, shows nothing consistently abnormal. Occasionally the lungs are infected, often the intestines are filled with gas and greatly distended. Death in these animals seems to be the result of an infection superimposed on tissues which have been altered by the dietary deficiency.

The excessive reaction to infection of the upper respiratory tract (nasal and aural passages) of the artificially fed baby and the low incidents of such infection amongst breast fed babies, suggests that in the former case as in the rats, we are dealing with an infection superimposed on tissues which have been altered by a dietary deficiency.

The excessive reaction to infection of the upper respiratory tract (nasal and aural passages) of the artificially fed baby and the low incidents of such infection amongst breast fed babies, suggests that in the former case as in the rats, we are dealing with an infection superimposed on tissues which have been altered by a dietary deficiency.

The possibility of the occlusion of Vitamin A in the calcium soap excreted by the artificially fed baby, and the consequent failure to absorb enough to meet the physiological needs, was suggested by Daniels and Armstrong in 1923. 1 Acting upon this suggestion a comparative study of the Vitamin A excretion by way of the gut, in an artificially fed and in a breast fed infant, has been made.

A year ago a simple technique of estimating the pH of urine was suggested, 1 in which application was made of the bicolorimeter and the phthalein dyes, phenol red, brom cresol purple and brom cresol green. At that time we concluded that as far as the matching of colors went, the method had a probable error of pH ± 0.02 to 0.04, but we realized that the factors of temperature and dilution must exert some influence on the true pH. Obviously one desires to know the pH of the undiluted urine at body temperature. During the past year we have been trying to ascertain how far colorimetric determinations on diluted urine at room temperature differ from electrometric determinations on undiluted urine at 38°. In connection with an acid-base balance study we felt that we should know just how far the present colorimetric determinations might deviate from the true pH values and if a correction factor might be applied which would hold for all urines.

Since our preliminary report was made and the present study begun, Hastings, Sendroy and Robson 2 have reported a study of this question. They carry out their determinations at 38° diluting the urine 1-5. They state that the dilution error is the important error and amounts to about 0.1 pH, which must be subtracted from the colorimetric result. They believe that with this correction their results fall within 0.1 of the actual pH.

In routine work it is quite troublesome to carry out colorimetric estimation at 38°. In the present study it has been found that the factor necessary to correct for the determination being made on the diluted urine at room temperature amounts to roughly 0.2 pH, about half of which is a temperature correction and half a dilution (salt and buffer) correction.

In 1916 Benedict 1 reported a single experiment in which the ingestion of caffeine lead to an increased output of uric acid as determined by the then new Benedict-Hitchcock colorimetric method. A year later Mendel and Wardell 2 studied this question with the same method, employing not only caffeine, but also coffee and tea. They observed an unmistakable increase in the uric acid output in all cases. The statements in the older literature are quite contradictory. Many texts still state that the ingestion of methylated xanthines is without effect on the excretion of uric acid. For this reason it seemed desirable to again study this problem.

In the present studies metabolism experiments have been carried out with the three common methylated xanthines, viz., caffeine, theobromine and theophylline, employing several methods for the estimation of uric acid. In all experiments the subjects were placed on purine-free constant diets until the uric acid excretion reached its endogenous level. The methylated purine to be studied was then added to the diet for a period of three or four days and finally the purine-free diet was continued for three days longer. Estimation of uric acid excretion were always made by the Benedict-Franke colorimetric method; in certain cases they were made also by the Benedict-Hitchcock colorimetric method and bly the Krüger-Schmidt precipitation method.

With two subjects the excretion of uric acid was markedly increased throughout the period of caffeine ingestion, but when the caffeine was discontinued, the uric acid fell back to its endogenous level within 48 hours. With two other subjects the excretion of uric acid was markedly increased on the first day of caffeine ingestion, then gradually decreased until on the last day of caffeine ingestion the uric acid was back to its endogenous level, where it remained practically constant throughout the final period.

Methods for the estimation of guanidine and its methyl derivatives have been subject to considerable criticism. A study of the color reaction for guanidines recently described by Marston 1 was undertaken in order to determine its biochemical applicability. Although Marston specifically states that his reagent will keep indefinitely, in our hands deterioration and precipitation invariably set in shortly after mixing the three ingredients, sodium nitroprusside, potassium ferrocyanide and sodium hydroxide.

A reagent has been developed, which, although it contains the same ingredients as employed by Marston, has the advantage that it keeps well and gives within five minutes a full color development which does not fade or become turbid for more than an hour, thus allowing ample time for color comparison.

The reagent we now employ is prepared as follows: 6 gm. sodium nitroprusside and 8.5 gm. potassium ferrocyanide are dissolved in water and made up to 100 cc. About 15 to 20 minutes before using, one volume of this solution is mixed with one volume of 10 per cent sodium hydroxide and two volumes of 3 per cent hydrogen peroxide. It has ken found convenient to add 1 cc. of the prepared reagent to 4 cc. of the unknown guanidine solution and compare in a colorimeter with standards similarly prepared. With this technique quantities of guanidine bases as small as 0.2 mg. may be estimated.

Creatine and creatinine give a faint coloration with the reagent but do not interfere materially in the estimations. Uric acid and ammonia interfere by hindering the color devlelopment with the bases. Urea produces about one tenth of the color given by guanidine, a fact not mentioned by Marston. Methyl urea, β-methyl hydantoin, β-methyl liydantoic acid and glucose yield no color, while ethyl alcohol gives a very faint reaction.

The use of kaolin in removing creatinine from dilute solution is disadvantageous in that the adsorbed creatinine cannot be released again for identification. An adsorbing agent which can be decomposed under conditions that will not destroy creatinine therefore appears desirable.

If one adds to 10 cc. of the strongest creatinine standard for blood determinations (1 mg. of creatinine in 100 cc. of saturated picric acid), 0.1 cc. of 9.5 per cent potassium chloride, and 1 cc. of 10 per cent phosphotungstic acid, a finely divided yellow precipitate forms, which settles out very slowly, but can be centrifuged out quickly. The supernatant liquid is decanted, and the precipitate suspended in 10 cc. of saturated picric acid. On adding 1 cc. of 10 per cent sodium hydroxide, the precipitate dissolves and the color developed matches that of 10 cc. of the same standard, made alkaline directly with the same amount of sodium hydroxide. Controls with the potassium precipitate in absence of creatinine are negative.

Complete adsorption of creatinine is practically coincident with appearance of the precipitate. Supersaturation is occasionally encountered. If the precipitate fails to appear, the difficulty is overcome, when working with larger amounts, by adding a drop of potassium chloride solution after the phosphotungstic acid. The precipitate is more difficult to centrifuge out if all of the potassium chloride is added after the phosphotungstic acid.

The precipitate obtained from a liter of saturated picric acid containing 10 mg. of creatinine, on adding 10 cc. of 9.5 per cent potassium chloride, and 100 cc. of 10 per cent phosphotungstic acid, can be suspended in the centrifuge tube in which it has been collected, in 30 cc. of normal sulfuric acid. On diluting the mixture to about 50 cc. and shaking with ether, an emulsion forms, which separates on a moment's centrifuging into three sharp layers, predominantly ether, sulfuric acid, and phosphotungstic acid solutions.

The obstruction was produced, 20 cm. below the pylorus, by means of a ligature tied over an elastic compressor on the outside of the abdominal wall. Such an obstruction causes almost no trauma to the intestine, and can be released at any time desired without employing a second anesthetic and laparotomy, with their complicating effects.

If the obstruction at this level is not released, the period of survival averages seventeen hours. The variations from this average time are within three hours.

If the obstruction is released after a duration of fifteen hours or less, the animal survives. If, however, the obstruction is released after a duration of sixteen hours or more, the animal does not survive.

The sharp line of demarcation between the duration which is fatal, and that which is followed by recovery is quite striking. Also, the recovery of the animals from an obstruction of twelve to fifteen hours duration was remarkably rapid. Even though such animals before release of the obstruction gave evidence of being in a very serious condition, a striking degree of recovery is shown within one to three hours following the release of obstruction. If the animal's condition was due to absorption of a highly toxic substance from the obstructed content, such rapid recovery would hardly be expected. If the condition were due, however, to depletion of chloride 1 or the loss of fixed base, 2 the rapid recovery could be the result of rapid reabsorption of these substances which accumulate in the fluid distending the stomach, since the non-operative release of obstruction allows the obstructed content to move rapidly on to the lower parts of the canal where absorption is rapid.

Two experiments were carried out on dogs with Thiry-Vella fistulae of ileum. The operations for formation of the fistulae were performed several weeks before the experiments were started. Graphic records of the intestinal contractions were made by introducing into the fistula a sausage-shaped rubber balloon fastened on a catheter and filled with water; the catheter was connected with a Brodie bellows-recorder which made the tracing on a slowly moving kymograph. Records were made in this way without anesthesia or operative procedure or discomfort to the animal.

Experiment 1. Ascending doses of morphine sulphate: A female collie dog, weighing 21 kg. received daily hypodermic injections for 71 days and during this time the dose was increased from 1 mg. per kg. to 25 mg. per kg. There was some constipation and loss of appetite but the dog remained in good condition, losing only one kg. in weight. Narcosis and vomiting disappeared after the fourth week, but salivation continued throughout the experiment. The dog became shy and restless as the experiment progressed.

Graphic records of the effect of the injections of morphine on the intestinal contractions were made at intervals of two to three days throughout the experiment. There was no change in the character of the reaction at any time. We have reported in a recent paper 1 that hypodermic injections of morphine in dogs greatly increase the muscular activity of the intestine. In this experiment each injection produced the usual reaction; marked increase in tone and in amplitude and frequency of the contractions. From time to time small doses (0.15 mg. per kg.) were injected before the regular daily dose, and these invariably produced stimulation of the intestinal contractions.

The mode of catabolism of the diamino acids is obscure. It has been shown that ornithine is a sugar former in the diabetic organism, while lysine is not. It appears that alpha-amino acids are frequently broken down through the stage of a fatty acid having one less carbon atom. 1 If ornithine and lysine took this route, they would form gamma-amino butyric and delta-amino valeric acids respectively. As it seemed of interest to determine the fate of these substances in the diabetic animal, they were synthesized and administered to phlorhizinized dogs. The results would indicate that delta-amino valeric ackt fails to form extra glucose while gamma-amino butyric acid is a sugar former. The possibility is therefore to be considered that one of the paths of catabolism of the diamino acids is through the stage of mono-amino acid with one less carbon atom and with the amino group in the terminal position.

Inverted and bleeding nipples and chronic inflammatory lesions of the human breast are frequently noted as forerunners of cancer of this structure. These conditions often present the aspect of the “indifferent vorstadium” described by Ribbert 1 or the picture commonly alluded to as the so-called precancerous stage.

The experiments herein noted were performed with the view of ascertaining what results would follow the inversion of nipple flaps in the mammary gland of the female rabbit.

Eight female rabbits purchased in the open market were employed and their histories and ages are therefore not known. Two of these rabbits were lactating and one other was not full grown. Twenty-eight operations were performed. Eighteen different nipple flaps were inverted and ten of the nodular masses thus produced were bisected and reinverted in other areas. The operations, as a whole, consisted in cutting a rectangular flap in which the nipple was included and the basal attachment of the flap was not disturbed, thus maintaining the circulation from that source. The depth of the flap extended down to the areolar tissue, thus including the skin and gland structure beneath. It may be mentioned that in the quiescent breast, the gland of the rabbit is rather rudimentary in its gross aspect; where lactation is present the gland is well developed and it was sectioned completely through in order to procure the flap. The tip of the nipple was clipped off and the stump cross sectioned and scraped. The underlying gland was, also scraped and partially sectioned. The flap thus prepared was inverted with the nipple downward and then drawn beneath the overlying structures by means of a retaining suture.

Evidence is derived from epidermal ridge configurations, in four subjects admitting diagnosis of congenital amputations (as opposed to agenesia), which points to the existence of the affection probably prior to the eleventh week of gestation. Three of the cases are considered in another publication, 1 where they are recorded as numbers 20, 508, and 509; a fourth, a still-born infant presenting multiple digital amputations in varied degrees, is comparable to 509.

It has been shown 1 that the alignment of epidermal ridges, hence their fashioning of patterns and of patternless series of ridges, is accomplished through the medium of growth forces obtaining in early fetal development. Such forces in growth vary locally, in accordance with the irregular molding of palmar and plantar reliefs. The influence on alignment hardly can be effective after ridges are initially elaborated (eleventh week), although it is possible that the regulation through growth may manifest its effect before the ridge anlagen actually appear. There is no evidence to show that a ridge arrangement, once effected, can be altered, barring, of course, its participation in the generalized increase of size. With this premise of permanence, coupled with the demonstrated genetic relation existing between the form of a part and the character of its ridge configuration, it is warranted to assume that a congenitally defective hand or foot will display configurations conforming to its particularized molding if the abnormality existed during the critical period of ridge determination. Such conformity is invariable in abnormalities which are initially present in the member (such as syndactylism, ectrodactylism, etc.). The conformity occurs also in the examples of amputation here reported, to the extent that the configurations are individually unique, being so far modified from the normal.

A simple slide test has been developed which seems to possess certain advantages over previously described serum tests. The stock antigen is stable. Inactivation of serum is unnecessary. The time required to make the test is two minutes. No special laboratory equipment or facilities are required.

Antigen preparation: Fresh veal hearts are selected, and the superficial fat removed. The muscle is ground in a sausage grinder, and spread on paper and dried by an electric fan. After it is completely dried, it is powdered in a mortar and extracted with ether. Four hundred cc. of ether are used to each 100 gm. of powdered heart and allowed to act for ten minutes, shaking frequently. The ether is filtered off and 300 cc. are again added to the heart and treated in the same manner. The heart muscle is again treated with 300 cc. of ether, and again a third time, the ether in each instance being filtered and discarded. The heart muscle is now dried free from ether, and for each gram of muscle, 5 cc. of 95 per cent alcohol is added and maceration is allowed to continue for three days at room temperature, after which the alcohol is filtered off and made up to the original volume with 95 per cent alcohol. This constitutes the defatted alcohol heart extract for the antigen. Six decigrams of cholesterol and 3 cc. of glacial acetic acid are added for each 100 cc. of the alcoholic extract. This is filtered after solution is effected and constitutes the finished special antigen for this test. This antigen seems to be stable for at least several weeks.

Technic: One cc. of antigen is measured into a test tube and 2 cc. of distilled water into a second tube (normal saline can be used, but the solution is unstable).

In the present communication we wish to report the results obtained in an effort to produce toxic effects in animals with the streptococcus scarlatinae and to briefly discuss the toxemia and exanthem of the disease in the light of these experiments. The work was undertaken because of our previous failure to induce acute toxic nephritis in the rabbit either with large doses of viable culture or massive quantities of culture filtrate. In fact we were unable in the earlier experiments to infect the rabbit even with large amounts of scarlet fever streptococci. The fact that in human scarlet fever there is so frequently a nephritic complication, presumably toxic in origin, led us to attempt to induce experimentally the kidney lesion. We assumed at the time the specific streptococcus in vitro would yield a soluble toxin.

Three separate isolations of the scarlet fever streptococcus were employed in our present study upon the nature of the toxic principle. Two cultures, one designated “Harrison”, the other “Tyler”, were supplied us by Dr. Dick of Chicago, while the third culture was one of our own which had been recovered here from the blood of a case of scarlet fever.

The causal excitant of measles exists in the circulating blood during the febrile stage of the disease. The Berkfeld filtered virus blood induces in the monkey, rabbit and guinea pig the symptom complex of human measles.

A living agent capable of causing in the experimental animal a concomitant leucopenia and pyrexia may be propagated and subcultured in vitro through several generations upon hydrocele or semi-solid plasma Ringer medium. In the cultures prepared from the unfiltered human measles blood of reacting animals there frequently occurs a small gram positive coccus, which in subplants to blood agar medium grows readily and under aerobic conditions. Cultures of this coccus are likewise capable of inducing in the rabbit a leucopenia and pyrexia following an incubation period of seven to eight days. The coccus culturally and tinctorially corresponds to the culture reported by Tunnicliff. Since the COCCUS was cultured from the unfiltered measles blood, we prefer to reserve opinion at this writing regarding its significance in measles.

The effect of strenuous competitive athletics on the heart, that is, the question of true physiological work hypertrophy, is a matter of considerable interest. A study, therefore, of the heart of the greyhound was undertaken. The hearts and body weights of ten of these thoroughbred dogs have been studied and compared with the established normal heart-weight-body-weight ratio of .00798 or 7.98 grams of heart per kilogram of body weight, for ordinary mongrel dogs. 1 The largest heart was found in the oldest and most successful racer, “Victorious Red,” aged three years, with a ratio of 17.3. Five greyhounds that had had their schooling, whose age was about two years and who had raced not more than three times, ranked according to ratio as follows: 16.2, 15.1, 14.5, 13.9, 13.6. Two dogs were full grown, aged eighteen months each, but had not been schooled and had never raced. The one of the better pedigree showed a ration of 13.4, while the other averaged 11.9. Two pups about one-third grown, aged six months each, though leading a caged up, sedentary existence, showed ratios of 11.5 and 11.3, which are above the maximum value of 9.98 found in a series of two hundred ordinary mongrel dogs. 1 The average for the greyhound series is 13.4, which is considerably above that of the stag, 11.5, which has up to this time topped the heart weight-body tables for mammals. Roentgenograms and electrocardiograms of Derby winners in the greyhound races confirmed the anatomical findings. The study suggests that the greyhound, by virtue of generations of strenuous exercise of coursing, or as a result of selection, is endowed with a proportionately large heart at birth and that this heart responds to schooling and training by hypertrophying to an unusual degree.