Experimental Biology and Medicine

The importance of the upper respiratory tract as a portal of entry for pathogenic micro-organisms amply justifies attempts to increase its resistance against their invasion. We have recently shown 1 that the local introduction of antigen into an area of mobilized histiocytes leads to the local formation of specific antibody. We have extended such studies to the nasal mucosa, assuming that a similar response might be obtained there following repeated applications of an antigen. This assumption was based on the probable mobilization of cells of inflammation in the mucosa, with consequent fixation of antigen among differentiated histiocytes.

Rabbits were treated intranasally at daily intervals of 2 to 13 days with a formolized vaccine of Bact. paratyphosum B., either by insufflation (2), by instillation alone (12), or by instillation subsequent to earlier instillation of ox bile (1). The animals were then allowed to rest for 1 to 12 days, and were sacrificed. The nasal mucosa, lung, liver, spleen, and blood serum were mixed with 15 parts of a solution of equal parts of glycerol and 0.85% solution of sodium chloride. They were then ground in a mortar, with the exception of the serum, and extracted at 37°C. for 7 days. These extracts were titrated simultaneously against a living suspension of Bact. paratyphosum B. Six of the 19 animals were perfused with citrated salt solution immediately after death to remove the blood as far as possible from the organs to be extracted. Most of the blood was removed from all except the spleen.

The content of agglutinin in the nasal mucosa in animals treated daily by insufflation or instillation for at least 11 days was always distinctly higher than that of either the spleen or liver.

Changes in blood cholesterol in both clinical and experimental hypo- and hyperthyroidism have been reported recently by several workers. Briefly stated, hypothyroidism is accompanied by an increased blood cholesterol; hyperthyroidism by a decreased blood cholesterol.

This study was undertaken to confirm these findings in experimental hypo- and hyperthyroidism in rabbits, and to see if any correlation exists between cholesterol changes and changes in the number of red blood cells and Hb.

The technique for preparing cretin rabbits has been previously cited. 1 Blood counts were made in the usual way; Hb was determined colorimetrically 2 and is expressed in gm. per 100 cc. blood. Cholesterol was determined on whole blood by Sackett's 3 modification of Bloor's method. All determinations were done in duplicate.

Three cretin and 6 normal animals were used in the first series. These were fed from 0.318 to 0.59 gm. desiccated thyroids daily for 2 or 4 weeks. Thyroid feeding caused a decrease in the blood cholesterol of all rabbits of this series, but this decrease is not accompanied by a consistent change in the number of red blood cells or in Hb.

A second series was studied, using 2 cretins and 4 of the animals in series I. After thyroid feeding, the losses of weight and the decrease in blood cholesterol of the 2 cretins are almost identical. There is no corresponding change in red blood cells and Hb. In the 4 normal animals, the weight loss is again roughly uniform, but again, there is no corresponding change in cholesterol and in red blood cells or Hb.

The results of these experiments seem to indicate that the changes in blood cholesterol which follow thyroid feeding in both cretin and normal rabbits, show no correlation with changes in the number of red blood cells, or the amount of Hb.

It has been proposed by Forssman, 1 Meyer, 2 Wernicke and Sordelli, 3 Taniguchi, 4 Landsteiner, 5 and Landsteiner and Simms 6 that some heterogenetic antigens may be composed of 2 chemically different substances, one being represented by a lipoidal substance containing the specific group but possessing no antigenic properties. The other component is of the nature of a protein. Since Landsteiner and Simms 6 reported investigations only with simple mixtures of these 2 systems, and Klopstock and Selter 7 those with mixtures of lecithin and diazotised atoxyl, it was of interest to study the possibility of chemically combining lipoidal substances with proteins and their degradation products to obtain a product of a definite entity which then could be used for immunological experiments. From a different aspect combination of lipoidal substances with proteins or their degradation products is interesting because of the colloidal nature of such condensations in an aqueous system. Such a physical state of the antigen is claimed to be important in immunity reactions.

The present report deals with experiments in which we prepared condensation products of cholesterol derivatives with certain amines, amino acids, and peptides. There are also indications of successful condensations with proteins.

Cholesterol, especially purified over the dibromide, was esterified with 4-nitrobenzoylchloride. The ester obtained was reduced with a Pt-catalyst as described by Shriner and Ko 8 to form the corresponding cholesteryl-4-aminobenzoate. For condensation with the above mentioned substances (amines, amino acids, peptides, etc.), diazotisation of the amino benzoyl ester was studied. A number of difficulties were encountered. It was found, for example, that while the hydrochloride of the cholesteryl-4-aminobenzoate can also be prepared by carefully grinding it with an excess of conc.

The method for preparing thyroglobulin has not been modified since Oswald described the procedure in 1899. He made saline extracts of fresh thyroid glands and precipitated the protein with an equal volume of ammonium sulfate. The process was time consuming since the precipitate had to be washed many times. Any other globulins present were also carried down and contaminated the final product. In attempting to prepare hog thyroglobulin, I found that the final preparation was frequently insoluble in saline or even dilute alkali. It was desirable to develop a more rapid method which would give a product suitable for intravenous injection. This has been accomplished by precipitating the thyroglobulin at the isoelectric point.

The fresh glands may be extracted with 0.1 M sodium acetate and the protein precipitated with dilute acetic acid. It has been found that sodium acetate will remove just as much of the iodine from fresh glands as sodium chloride. The optimum mixture for precipitation can be determined by setting up the following series of tubes:

To each tube is added 1 cc. of the sodium-acetate extract. In a few minutes the optimum precipitate can be observed. It has been found that the maximum quantity of iodine is precipitated in the tube giving the greatest quantity of precipitate. For hog thyroglobulin, tube No. 3 gives the best results. Using these proportions, a considerable quantity of extract can be precipitated at one time.

If a soluble preparation is desired, it is necessary to remove the precipitate and redissolve it in a short time. If it is allowed to stand for 24 hours, the protein is denatured and cannot be redissolved. However, after one or 2 hours, considerable precipitate has settled out.

The evidence presented seems to be in accord with the concept that bacterial properties may vary independently of each other.

A series of 15 cats were given intraperitoneal injections of nembutal (sodium iso-amytal, penta-barbital), the average dose being 60 mg. per kilo. Urine samples were collected at intervals after the injection and later analyzed for reducing substances by Sumner's method. 1 In certain instances the results thus obtained were checked by the use of Benedict's quantitative reagent. Simultaneously blood samples were taken and later analyzed for “blood sugar” by the Randle-Grigg modification of the Folin-Wu Micro Method. 2 These observations were made in the course of experiments to determine the effect of stimulation of the tuber cinereum upon carbohydrate metabolism. Rather extensive surgical procedures under deep nembutal anesthesia were required in the experiments. The bladder was exposed and emptied and the urethra cannulated from one to 2 hours after the injection of nembutal. The first sample of urine was taken about an hour after the urethra was cannulated and from 2 to 3 hours after the administration of nembutal. This sample always showed the presence of an abnormal quantity of reducing substance, usually the maximum obtained during the experiment, the concentration decreasing progressively in subsequent samples. This maximum reduction where the lowest maximum was obtained, was equivalent to that produced by 240 mg. % of glucose. In the experiment where the largest maximum was found, the reducing substance was equivalent to that produced by 1575 mg. % of glucose. An average maximum value, equivalent to that produced by 1045 mg. % of glucose, was obtained.

The amount of reducing substance in the urine was definitely related to the time interval following the injection of nembutal and not to the “blood sugar” concentration. Often in the later stages of the experiment the reducing substance would be decreasing in amount while the blood sugar was increasing.

Fogelson 1 and Atkinson 2 have observed clinically that the administration of “gastric mucin” to gastro-duodenal ulcer patients was followed by a relief of the ulcer symptoms and an increase of body weight. Kim and Ivy 3 found that one ounce of “gastric mucin” a day prevented the occurrence of duodenal ulcer in biliary fistula dogs, in which the incidence of duodenal ulcer is from 40% 4 to 60%, 5 and that the animals on “gastric mucin” maintained their body weight much better than the control dogs not receiving mucin. It is well known that the administration of alkaline powders is of benefit in the management of patients with gastro-duodenal ulcer. We desired to ascertain if the administration of alkaline powders would have the same effect as “gastric mucin” in preventing the occurrence of duodenal ulcers and in maintaining body weight in biliary fistula dogs.

Ten healthy dogs were used in each series. The biliary fistula was made according to the method of Rous and McMaster. 6 All the animals were kept in single cages and were fed a stock diet of cooked yellow corn meal, bone soup and bread with mucin or alkali twice a day. The dose of “mucin” was 15 gm. with each meal. The dose of alkaline powder consisted of 1 gm. each of sodium bicarbonate and calcium carbonate with each meal. This dose of alkali was used because it has a slightly greater “buffering action” than 15 gm. of the “mucin” used in these experiments. All of the animals showed an absence of bile in the feces during life and in the stomach and duodenal contents immediately after being sacrificed. They secreted daily from 5 to 15 cc. of bile per kilo body weight, which is “normal” for biliary fistula dogs. 3

The gross and microscopic anatomy of the molars was studied in 2 groups of rats:

A. Completely hypophysectomized. 22 animals. Age at operation: 36 to 64 days. Post-operative life: 63 to 459 days. B. Litter mate controls. Some were unoperated but most were unsuccessfully subjected to operation. Normal oestrous cycles and normal rate of growth. 15 animals.

Significant alterations were observed only in group A. The gross findings in these animals were (a) retarded eruption; (b) shorter length of roots. The histologic findings are summarized below.

1. The epithelial attachment is absent in a number of cases for varying lengths near the cemento-enamel junction so that the enamel is bounded directly by connective tissue or by a cementum spur of varying length and width. In 7 animals the enamel shows here some areas of resorption.

2. The enamel is normal in structure except for areas of resorption near which the organic matrix is sometimes seen in decalcified sections.

3. The dentin appears, as a whole, normal in structure. It is resorbed when the resorption of the enamel or cementum has extended beyond the dentin boundary.

4. The pulp usually shows a reduced blood supply. In some cases it shows advanced atrophic changes and contains calcified globules. The size of the pulp chamber and pulp canal appears to be larger than normal.

5. The cementum shows an abnormal frequency of resorption which is most prominent at the bifurcation level and least prominent at the apical portion of the root. In some of the animals of longest post-operative life a great number of globules of various sizes that stain like the cementum are seen at the cementum surface which faces the periodontal membrane. In some cases similar but larger globules are seen in the periodontal membrane.

Intravenous pressure may be estimated clinically either by direct or indirect methods. Direct methods, while more accurate, have certain disadvantages—impossibility of frequent determinations in the same patient, the necessary prerequisite of strict asepsis, the possibility of clot formation in the needle, the difficulty sometimes encountered in puncturing a vein in obese persons, and occasionally apprehension of patients which results in an elevated venous pressure.

The indirect or bloodless methods are free from most of these objections but the prevailing procedures have other disadvantages. The most serious objections are that the end-point is not sharp and correct estimations are therefore difficult, that the impossibility of applying such methods to obese persons in whom the veins are not visible above the surface of the skin, and that the end-point must be determined quickly.

The method we have developed is based on the principle that the superficial veins on the dorsum of the hand can be visualized easily in a darkened room by placing a small light, such as an ordinary pocket flashlight, against the palmar surface, preferably in the interosseous spaces. By this transillumination the superficial veins appear as black bands which can be obliterated by moderate pressure, thus distinguishing them from shadows cast by ligaments or bone. The mouth of a small glass funnel is covered by thin rubber dam held securely in place by thread tied tightly in a previously prepared groove near the rim. A small hole is then cut in the center of the rubber dam through which the vein to be examined is visualized.

The mydriasis which the administration of pilocarpine causes in the rat (Waddell, 1 Koppanyi 2 ) is one of the numerous paradoxical responses elicited by this drug. During the course of an investigation on the nature of iris sphincter tone in this animal, it was noticed that the topical or subcutaneous administration of pilocarpine resulted in corneal anesthesia. Inasmuch as it has been shown that substances capable of acting as surface anesthetics would cause the pupil of the rat to dilate (Barnard 3 ) it was thought, at first, that we held the key to the explanation of the pilocarpine mydriasis. Certain facts, however, indicate that pilocarpine does not owe its pupillo-dilator properties to the corneal anesthesia following its administration. The mydriasis is of a greater extent than that following the application of some local anesthetic. Furthermore, pilocarpine constricts the pupil of the mouse, whereas local anesthetics such as hexylresorcinol and butyn dilate the pupil in this animal as they do in the rat (Heidgen and Barnard 4 ).

Using graduated hairs for stimulation of the cornea, and accepting an absence of the winking reflex as a criterion for anesthesia, the effect of subcutaneous administration of pilocarpine was determined.

The number of experimental animals was, 34 albino rats, 9 albino mice, 5 guinea pigs, and 2 rabbits. Before the administration of pilocarpine the application of a pledget of cotton to the cornea would result in a closing of the eye, it being sufficient merely to touch the cornea very lightly. All animals were thus tested both before the administration of pilocarpine and after the removal of the pilocarpine anesthesia by atropinization. This constituted the control.

The heart of Limulus consists essentially of 2 divisions, an atrium and a ventricle, each of which are contractile. The electromyograms obtained with the cathode ray oscillograph as a recording mechanism indicate that the fully contracted state is reached by a process of successive additions. In the atrial and ventricular muscle fibers the greatest amount of shortening occurs at the onset of a contraction, the further additions being of smaller and more nearly equal degree. The electromyogram consequently has a form somewhat comparable to the RST complex of the mammalian electrocardiogram. In contrast the electromyogram of a small group of isolated ventricular muscle fibers after stimulation with a single shock has the form of an ordinary diphasic muscle record, the rate of progression being 1 to 2 cm. per second.

The electroneurogram of the median cardiac nerve shows 2 groups of oscillatory discharges beginning 30 to 80 sigmas before the start of electrical activity in the musculature of the atrium and ventricle. The start of the ventricular electromyogram precedes the start of the visible ventricular contraction by several hundred sigmas. The oscillatory discharge can be shown to consist of potentials derived from 2 sources. From a correlation of the electrical and histological studies it is inferred that these 2 sources of potential are the axons of the large ganglion cells and the axons of the small ganglion cells. For reasons previously presented (Heinbecker) the large ganglion cells are considered normally to be the pace-maker cells, the small ganglion cells being responsible for the nerve impulses which directly innervate the cardiac muscle. Normally activity in the adult heart of Limulus is directly neurogenic in origin. The heart muscle can, however, be shown also to be rhythmically contractile without the intervention of cardiac ganglion cells.

The relative concentration of Vitamins B₁ and B₂ in normal and experimental rats was determined by feeding dried rat tissue as the source of these vitamins. “Normal rat tissue” was prepared from stock rats, 200-300 gm. in weight. The rats were decapitated, the carcasses (except skin, feet, tail, and intestines) were hashed in a meat grinder and the hashed material was dried in a warm air dryer (50°C.). In the same manner “B₁ rat tissue” was prepared from 20 rats after 30 days on the Evans and Burr diet supplemented with cod liver oil and tikitiki. The average starting and final weights of this group were 186 gm. and 172 gm. respectively, indicating cessation of growth and a loss of weight during the 30-day period on the diet deficient in B₂. “B₂ rat tissue” was similarly prepared from 20 rats after 30 days on the Evans and Burr diet supplemented with cod liver oil and autoclaved liver. The average starting and final weights of this group were 184 gm. and 148 gm. respectively, indicating cessation of growth and a loss of weight during the 30-day period on the diet deficient in B₁. These 3 rat tissue preparations were fed as sources of B₁ and B₂ to young 50 gm. male rats fed the Evans and Burr diet supplemented with cod liver oil and either tikitiki or autoclaved liver.

Normal rat tissue was not a good source of either B₁ or B₂ but was definitely richer in the latter. Better growth resulted with 0.3 gm. of this tissue as a source of B₂ (Group 8) than with 0.7 gm. as a source of B₁ (Group 7).

Normal rat tissue contained more B₁ than either the B₁ or the B₂ rat tissue preparations.

These results were in marked contrast to those obtained when the same tissue preparations were used as a source of vitamin B₂. Normal (Group 8) and B₂ (Group 12) rat tissue were of nearly equal value as sources of B₂.

As Bartley and Newman have shown, there can be recorded from the cerebral cortex, over considerable of its surface, a continuous and extremely complicated series of electrical oscillations. The record as it appears on the cathode ray oscillograph, with adequate sensitivity to show its complexity, is far too intricate for any such casual analysis into specific waves as is possible in nerve trunks.

That this activity consists of a complication of more simple elements is indicated by the following experiments: If in the rabbit's or dog's cortex shallow cuts (1 mm. deep) are made near a recording electrode (an indifferent electrode being located on the bone of the skull), the form of the record changes, becoming simpler. Cutting off the blood supply from an area causes its activity to cease in 5 to 10 minutes, without any obvious simplification. Ether anesthesia may cause marked simplification of the record at a stage shortly before all activity ceases. The simplest large waves have a duration of 30 to 100 sigma, but in the complex picture there appear to be some as short as 10 sigma. The amplitude of individual waves of the normal record may be as much as 1/2 mv. increasing upon stimulation of the animal to several times this. No oscillations can be found even at a sensitivity of 1/2 meter per mv., that have as short a duration as nerve-fiber action potentials. The long refractory period, low frequency, long duration and suppression by anesthesia at a concentration that does not affect nerve fibers indicate that nerve cells spontaneously active in the cortex are involved, and suggest that the potentials themselves may be cell potentials rather than nerve-fiber potentials. Further, responses to direct electrical stimuli are too long in duration for nerve-fiber potentials, and are of the order of duration of the spontaneous waves.

Prolonged administration of either theelin or theelol to adult female rats does not destroy the reproductive function, as is evidenced by their ability to bear and rear normal litters. The treatment with the luteinizing substance from pregnancy urine produces complete inhibition of estrus together with profound ovarian luteinization but does not permanently impair the reproductive mechanism.

One of the greatest difficulties in carrying out any problem in vitro is that of establishing a morphological criterion by which the various cell elements composing the growth can readily be classified. When dealing with pure culture strains of “Fibroblasts” and epithelial cells it is a comparatively simple matter to distinguish between the 2 cell types, as they not only exhibit distinct differences in their morphology, but also in their mode of growth. But when attempting to distinguish these 2 different types of cells in a mixed growth colony it becomes more difficult, as the epithelial elements composing the advancing free edge of the new growth invariably isolate themselves from the epithelial sheet and migrate into the culture medium as separate units, assuming a contour similar to fibroblasts (see Fig. 1). But far greater difficulties confront the investigator when he has to identify cells of the mesenchyme type, such as osteoblasts, chondrioblasts, and heart fibroblasts which, although possessing varying functional activities, exhibit similar morphological values.

Parker and Fischer, 1 realizing this difficulty, successfully demonstrated that mesenchyme elements in tissue cultures, which would be designated morphologically as “Fibroblasts”, are found to possess different inherent growth potencies when cultivated under similar conditions in vitro. On these grounds they rightly contended that “the physiological properties of the cell should constitute the first claims to any definition.”

Recently Horning 2 found that tissue culture cells which exhibit the same morphological characteristics and behavior under normal conditions in vitro, differ, however, in their reactions to similar pathological conditions, as the rate of cytolysis was in all cases found to be dependent upon the inherent growth energy of the given strain.

Policard and Doubrow, 1 using the technique of microincineration, made a comparative study of corresponding normal and malignant tissues and found significant differences in their ashed remains. Recently it has been possible to improve the technique so that one can examine incinerated preparations with oil immersion objectives and study individual cells with great facility (Scott 2 , 3 ). Using these improvements, Horning and Scott 4 studied the distribution of the inorganic salts in the developing chick embryo and found some points of interest in connection with the Cohnheim “Embryonal Theory” of tumor proliferation. Our belief that the method of microincineration is capable of yielding information of a new kind relative to malignancy has led us to study some human medullary duct carcinomata of the breasts as well as several of the scirrhous types,† together with the transplantable mice tumors, M. 63, S. 37 and 180.

The preliminary observations of the microincinerated tumors confirmed the previous statement of Policard and Doubrow that cancerous tissue remains carbonized longer than does normal tissue. Another constant feature of the human and rodent neoplasms is the relatively greater content of ash of the tumor when it is compared with similar normal tissue. Observations with oil immersion objectives show that the nuclei contain more ash than do nuclei of normal duct tissue. This deposit is marginated and evidently represents the ash remains of an hyperchromatic condition similar to that described by Horning and Richardson 5 in malignant growths. It is interesting to note that these nuclear salts contain visibly more iron oxide than do those of normal cells. The cytoplasmic ash deposit is more abundant than in the normal.

One of the difficulties in carrying out surgical compressive therapy for various lung diseases is that in some individuals the mediastinal contents are so little stabilized that such procedures as artificial pneumothorax and theracoplasty merely result in crowding the affected lung over into the opposite side of the chest instead of producing the desired compression of pulmonary cavities. The theoretical considerations, together with the demonstration of the importance of having a fixed mediastinum in such conditions, upon which is based the modern treatment of acute empyema have been described by Graham.

The method producing fixation must be innocuous, i. e., should not produce “Pick's syndrome” of adhesive mediastino-pericarditis, and furthermore the duration of the various types of fixation should be of differing degree.

The anterior mediastinal space of rabbits is potentially large with ready extension to the anterior portion of the superior mediastinum. This space accommodates 15 cc. of fluid. Roentgenograms of injections of this space with radiopaque fluids demonstrate the feasibility of filling this space with fluid substances which either by action as mechanical barriers or by evocation of a benign, slowly progressive, productive inflammation would secure fixation of the mediastinum in the mid-line sufficient to resist large alterations of intrapleural pressure.

The standard of control was the shift of the mediastinum (adult rabbits) in response to 20 cc. of air introduced into the left pleural cavity. Uniformly in the 6 control animals such a pneumothorax succeeded in shifting the mediastinum far over to the right so that the left cardiac border coincided on the roentgenograms with the left margin of the vertebral column.

Since proteins are usually species specific it might be expected a priori that soy urease would be chemically distinct from jack bean urease. In order to decide this question we have treated clear solutions of soy urease with jack bean antiurease. The antiurease was obtained by injecting crystalline urease into rabbits, followed by purification of the antiurease according to our method. 1 In all cases a precipitin reaction was obtained even when the soy urease contained as little as 1 unit per cc. Furthermore, we have found that soy urease is inactivated by the addition of jack bean antiurease. This inactivation is not so great as the inactivation of jack bean urease by jack bean antiurease. However, the difference is probably due to the presence of colloidal impurities in the soy urease. Table I shows some of our results.

As further confirmation of the inhibitory effect of jack bean antiurease on soy bean urease we have employed animal tests. Soy urease is as toxic to rabbits as is jack bean urease. If soy urease were distinct from jack bean urease one could not expect that injection of jack bean antiurease would confer passive immunity against soy urease. We have made tests, using 4 rabbits. Two rabbits were first given 80 jack bean antiurease units into the ear vein. Then all 4 rabbits were given 50 units of soy urease intraperitoneally. The 2 unprotected rabbits died within 5 hours. The 2 protected rabbits were not affected.

After orthotopic transplantation of the gonadic primordia of Amblystoma maculatum (Shaw) the writer 1 found a differentiation of the gonad according to the sex of the donor rather than the sex of the host. When an ovary and a testis developed together in any host, however, the testis, whether that of the graft or that of the host, usually induced a modification of the ovary, reducing it in extreme cases to a rudimentary structure containing relatively few germ cells (freemartin ovary). These results were in marked contrast to those of Burns 2 who found only unisexual ( or ) combinations in 80 pairs of animals of the same species which had been joined in parabiosis in embryonic stages comparable to those used by the writer for transplantation of gonadic primordia. To determine whether the local (Buffalo) strain of A. maculatum would yield similar results under the conditions of parabiosis, 60 pairs were joined in the spring of 1931. Since the operations were performed late in the season on embryos of rather low viability, only 16 pairs survived to ages of 50 days or over. In 15 of these the gonads of both members were sufficiently differentiated to permit positive identification of sex. The combinations of sexes in these pairs were found to be as follows: 1 1; 6; 8 . In all 6 heterosexual pairs the ovaries of the female member had undergone modification under the influence of the male twin. The central ovarian cavity was absent and the cortex reduced, though in no case to a completely sterile condition. In 2 cases the development of hilar and medullary germ cells in one ovary of the female was sufficient to give that gonad the character of a retarded testis; the other ovary in each of these animals, however, exhibited only the usual freemartin state with little or no evidence of reversal.

The influence of posterior pituitary extracts on mineral and water exchange was studied in 2 boys aged 9 and 5 years. The environmental and metabolic conditions, except for changes induced during experimental periods, were kept as constant as possible throughout the study. Each day was divided into 4 six-hour periods beginning at 6:00 A.M. At the beginning of each period, the subject voided, was weighed, and then given an accurately prepared meal of known weight and composition. The 4 meals of the day were identical in every respect and were prepared from simple foods of relatively constant mineral content. Water balances were calculated by the method suggested by Newburgh, Johnston and Falcon-Lesses. 1 On ashed urine sodium was determined by the uranyl acetate method of Barber and Kolthoff 2 and potassium by the chloroplastinate method of Shohl and Bennett. 3 In view of the short periods utilized, mineral excretion in the stool was disregarded and average urinary excretion on control periods was used as a base line for determining fluctuations from normal during experimental periods. The posterior pituitary preparation pitressin was administered in all instances by the subcutaneous route.

The results of the 2 experiments presented in Table I clearly demonstrate that pitressin has a pronounced effect on mineral as well as water exchange.

The retention of water during the period of antidiuresis is associated with a large increase in urinary sodium and chlorid excretion. Potassium output is diminished in one experiment but is unaltered in 2 other studies. The effect of pitressin lasts about 3 hours after the last injection and is followed by an enormous water loss, exceeding the storage of the antidiuretic period by several hundred grams. The increased sodium and chlorid excretion persists throughout the period of diuresis.

Irwin's “multiple partition coefficient” theory 1 has been examined with reference to the adequacy of its theoretical and experimental bases. This theory proposes that the rate of diffusion of a substance from (A)/to (C) in a system: aqueous phase (A)/ non-aqueous phase (B)/ aqueous phase (C) depends upon the partition coefficients between adjacent phases. Suppose the diffusing substance to be a dye originally dissolved in (A) : with time its fugacity in (B) will approach but not exceed that in (A) (which we may assume to be kept constant). Its stoichiometric concentration in (B) will assume a value which characterizes the partition coefficient, but its fugacity from (B) will be independent of the partition coefficient. The rate of diffusion out of (B) into (C) will depend upon both its fugacity and its stoichiometric concentration in (B), and also of course in (C). The rate of diffusion across either or both phase boundaries may also be affected by local conditions at the interface peculiar to the nature of the 2 phases and the diffusing substance. For these reasons the hypothesis, insofar as it is based on partition coefficients alone, is physically unjustifiable, and we would expect it to have a very limited applicability.

The supposed experimental proof 1 of the hypothesis rests upon parallelism between the relative rates of uptake of a relatively small number of dyes into the sap of living cells of Valonia or Nitella, and their uptake by the “sap” of an artificial cell. This artificial cell consists of a horizontal glass tube bearing 3 upright arms; the horizontal tube is filled with CHCl₃, which separates sea water in one end arm from natural or artificial sap in the other.

The mechanism by which parathyroid extracts increase the level of the blood calcium is almost completely unknown. Sendroy and Hastings 1 have shown that in vitro such extracts have no effect on the solubility of slightly soluble calcium salts. Similarly, we found that parathyroid extract did not increase the calcium content of serum incubated with fresh bone preparations. Bodansky, Blair and Jaffe 2 have shown that in the living animal the increase of the blood calcium on prolonged treatment is brought about by dissolution of the bone salts.

Since the effect of the extracts does not seem to be on the bone directly, the idea suggests itself that some other organ or tissue may be also involved in the action of the parathyroid hormone. Because the liver carries on so many of the chemical functions of the body, it was decided to first make some attempts to test whether the liver might be involved in the action of parathyroid extracts. The choice of the liver was made more attractive by the findings of Minot and Cutler 3 that calcium reserve is a protection against acute carbon tetrachloride and chloroform poisoning.

We determined the response of dogs with experimentally induced liver injury, to parathormone injection. 20 To date only phosphorus has been employed to produce liver damage, and in these animals it has been found that the normal response to parathormone disappears.

A typical protocol of the results obtained with phosphorus poisoning is the following:

These results necessarily must be taken with caution but they do seem to point to a connection between the liver and parathyroid action. The effect of other liver poisons is now being studied.

Arginine was first suggested as the mother substance of creatine by Czernecki. 1 From theoretical considerations the hypothesis has been generally regarded as an attractive one. It is in accordance with the fact that arginine, creatine and creatinine are the most abundant of the guanidine derivatives present in the animal organism. It is possible to postulate a series of plausible reactions by which the conversion of arginine to creatine might conceivably be accomplished. 1 It is also in harmony with the mutually exclusive occurrence of arginine and creatine as demonstrated by Kutscher and Ackermann. 2 These investigations have shown that creatine, a characteristic constituent of vertebrate muscle, is replaced by arginine in invertebrate muscle. Indeed the corresponding phospho esters, in which the muscle arginine and creatine largely occur, are even functionally equivalent. 3 , 4

Nevertheless, efforts to demonstrate the origin of creatine from arginine have been unsuccessful, almost without exception. Numerous investigations, described in Hunter's monograph, 5 and others referred to by Hyde and Rose 6 have been entirely negative in result or open to serious criticism upon some crucial point.

Doubtless one of the greatest difficulties in experiments in vivo has been the intervention of arginase, as a result of which administered arginine suffers rapid conversion to ornithine and urea. This competitive mechanism leaves little if any arginine available for the various reactions which are alleged to lead to creatine formation.

The work of Heymans 1 , 2 on the mechanism of the bradycardia of digitalis seems to dispose of the older theory that this drug stimulates the vagal centers directly. It supports Straub's 3 suggestion that the slowing is due to a direct myocardial action, the sinus pacemaker becoming more responsive to vagal tone. In support of this viewpoint, Rothberger and Winterberg 4 have claimed that digitalis lowers the threshold to electrical stimulation of the vagus trunks, and Weger has observed 5 that it makes the parasympathetic endings in the intestine more irritable to stimulation by pilocarpine. However, Weiss 6 regards the slowing after digitalis as due to reflexes originating in the viscera innervated by the vagus. In order to obtain data regarding the validity of the first of these theories, we have attempted to demonstrate the presence or absence of an altered vagus sensitivity after digitalis, by comparing the responses in pulse rate to parasympathetic stimulants before and after administration of this drug.

The slowing caused by choline or acetyl choline in 6 anesthetized and unanesthetized dogs and cats, including one dog in which the adrenals were removed and the stellate ganglia and upper thoracic sympathetic trunks destroyed, was so slight that these drugs seemed of no value as tests for vagus sensitivity. We, therefore, selected physostigmine as a stimulant of the vagus endings. In 3 dogs, the vagi were cut aseptically, and after a 24-hour recovery period, 0.3 mg. per kilo of physostigmine was injected intravenously and the pulse rate changes recorded. The next day, 20 mg. digitalis per kilo, as the dilution of an active tincture, was given intravenously, and 2 hours later physostigmine was reinjected in the same dosage.

The colloidal dyes congo red, vital red, rose bengal, and mercurochrome, tried under a variety of conditions, failed to protect the majority of guinea pigs and pigeons against the typical symptoms of anaphylactic shock. These negative results do not support the positive claims of Nikolaiev and Goldberg, who used congo red and trypan blue. It is suggested that the results with these colloidal dyes point to fundamental differences in the tissue reactivities of proteins and of drugs and toxins.

Paraldehyde has many features recommending it as a hypnotic agent. Conversations with Prof. C. W. Muehlberger and the late Prof. A. S. Loevenhart suggested a preliminary survey, at least, of the biochemorphology of paraldehyde and chemically related substances in the effort to determine whether or not there might be promise of finding a compound in the series with properties approaching more closely to those of an ideal clinical soporific. The compounds so far secured have been studied before, 1 but separately and not as a series. Acetal and methylal received clinical trial but were considered unsatisfactory. 2

In Table I is a list of the compounds we investigated, with their structural formulae, certain physical characteristics, and toxicity data. The latter were obtained by administration to rabbits by stomach tube of dilute water emulsions of the respective substances. Table II is a summary of the comparative depressant action of these aldehyde derivatives as determined in 150 experiments on rabbits, in which observations were made and recorded by methods modified from those described by Magnus. 3

These related compounds have equal rapidity and duration of action when given in doses of equal effectiveness. They are all excreted by the lungs in considerable amounts. Peripheral vasodilatation and increased heart rate are produced by all of them, but their effect on pupil size is inconstant. As may be seen from the summary data presented in Table II, none of these agents is particularly analgesic in action.

Some conclusions can be drawn concerning the biochemorphic† aspects of these substances. Very striking is the difference in the effect of lengthening the carbon chain on the central, and on the terminal carbon atoms.

It was previously reported that sodium monoiodoacetate (hereinafter called IA) inhibits lactic production in Streptobacterium casei as it does in higher forms. 1 This organism is especially advantageous for a comparative study of lactic acid fermentation, since a single compound, e. g., glucose, will suffice as a source of energy. The glucose supplied is quantitatively converted to lactic acid, and in the absence of nitrogenous compounds—essential for growth—there seems to be no other significant metabolism. 2 Furthermore there is evidence that the course of lactic acid formation by Streptobacterium casei is qualitatively the same as in other forms. 3

The production of lactic acid is a reaction of great importance and of widespread occurrence, being the physiological equivalent of respiration in some forms and its supplement in others. 4 Like oxidation it involves a chain of reactions which are primarily dehydrogenations. The results of our experiments serve to elucidate further some of the properties of the enzymes concerned in lactate fermentation and to indicate still another analogy with biological oxidation.

Lactic acid production was determined by the method of Warburg. 5 The reaction vessels were filled with a gas mixture consisting of 5% carbon dioxide and 95% nitrogen. The required amounts of IA were placed in the sidearms of the vessels and added after a preliminary control period. In the several controls, which received no IA, after a brief induction period the rate of lactic acid formation remained constant for more than 7 hours. In all cases the organisms were suspended in a solution of 2.0% glucose and 0.5% sodium bicarbonate. Little change in pH occurred during an experiment, the lowest final value noted being pH 6.7.

When IA is added to such a system there is a definite “latent period”, which is doubtless a function of cell membrane permeability and diffusion rate of IA.

Several investigators have recorded electrocardiograms on the embryonic chick, and, in the recent report of Sands, 1 a monophasic tracing was obtained at the 35th hour of incubation. A typical rapid initial ventricular phase and slow final phase, i. e., Q-R-S and T deflections were not obtained until between 50 and 72 hours.

It has been determined by Duval and others recently that ingrowth of nerve fibres does not take place until approximately the 6th day, but that the heart generally starts beating at the 10 somite stage or about 29 hours.

We attempted to obtain electrical records of the heart activity, using German silver electrode contacts along an approximate axial lead of the heart. A resistance coupled tube amplifying mechanism, the Nichols-Chase electrocardiograph apparatus, was used for amplifying and recording the action currents in most of the embryos. The Einthoven string galvanometer was occasionally employed on the older preparations.

Digitalis in an aqueous solution, as Digifoline, Ciba, equivalent to 0.1 digitalis leaf per cc. was applied directly to the exposed pulsating heart without changing the electrodes. Variations of temperature were difficult to control, but the surface was kept moist with water approximately at 37°C. Adrenalin and other drugs were applied, but the results of these applications will be reported at a later date.

The records obtained showed a multiphasic curve similar to an adult type electrocardiogram as early as the 35th hour of incubation, an embryo of 12 somites. Digitalis caused a definite change in the form and direction of the final or slow phase of the ventricular deflection. The electrical potential developed was approximately 0.5± millivolts.

The phenomenon of local skin reactivity to bacterial filtrates described by one of us (Shwartzman 1 ) was later also reproduced in the liver, kidney (Shwartzman 2 ), testis, intestines, lymphatic glands, lungs, thymus, guinea pig liposarcoma (Gratia and Linz 3 ), stomach (Karsner, Ecker and Jackson 4 ), and knee joints (Moritz and Morley 5 ). It was elicited with a great variety of microorganisms (Shwartzman 6 ) and also with vaccine virus as the preparatory factor (Gratia and Linz 3 ). The animals in which the phenomenon was observed were rabbits (Shwartzman 1 ), horses, goats (Shwartzman 8 ), and guinea pigs (Gratia and Linz 3 ). It could not be reproduced in mice and rats (Shwartzman 7 ). Assuming that malignant tumors may be of parasitic etiology, Gratia and Linz 3 thought that the hypothetical virus should then be capable of inducing a state of reactivity in the tumor tissue and thus render it susceptible to reacting factors in the blood stream. Five guinea pigs bearing liposarcoma were injected intravenously with B. coli culture filtrate. Two guinea pigs which died 24 hours later and 2 killed 48 hours later showed at autopsy hemorrhagic lesions in the tumor tissue and no lesions in other organs. The fifth guinea pig was left alive for further observations. They selected guinea pigs because of their susceptibility to the phenomenon of local reactivity to bacterial filtrates.

The bacteriophage which was used for this study was found to produce a complete lysis of 2 virulent strains of B. coli which we obtained from patients with peritonitis. One one hundred billionth of a cc. of this bacteriophage caused lysis of approximately one billion bacteria of strain “T” in 10 cc. of broth. The minimal lethal dose of the “T” strain of B. coli for mice at the time of the experiments was 1-5 million organisms which killed within 5 to 12 hours after injection of an actively growing, 3-hour, 2% dextrose cooked meat medium culture. Fifty million bacteria which constituted 10-50 M.L.D.'s were suspended in 0.5 cc. of broth and inoculated intraperitoneally into a series of mice. 0.5 cc. of bacteriophage was injected simultaneously into 2 of these mice, and into 2 more mice at varying intervals up to 4 1/2 hours after the bacterial inoculation. This approached closely the lethal period for control animals. In the control series plain broth injections were given at the same intervals as phage. The table shows one of these experiments. From this table one can see that 0.5 cc. of bacteriophage protected all of the mice when given up to 3 1/2 hours after bacterial inoculation. Only 50% of mice could be saved when bacteriophage was given after 4 hours. In similar experiments there was occasional recovery when phage was inoculated after 4 1/2 hours or later. Control animals receiving B. coli filtrate instead of phage were not saved. The bacteriophage itself proved to be innocuous, for 2 cc. and 4 cc. of it injected respectively into the peritoneal cavity of 2 mice did not produce any harmful effect. Both mice survived and the one inoculated with the larger dose suffered only slight discomfort for 1 or 2 hours, probably due to distention from the quantity of fluid injected.

The following study was undertaken to determine the influence of subtotal gastric resection and double vagotomy upon gastric acidity in dogs. Portis and Portis 1 showed that the total acidity remained unaltered after subtotal gastrectomy in dogs. Hartzell 2 found a marked reduction in acid after supra-phrenic double vagotomy, but 2 years later Vanzant 3 working with the same animals reported that the acid had returned to normal values.

Complete studies have been made upon 5 dogs. Preliminary to resection and vagotomy, a Pavlov pouch was made. Then, 2 to 3 weeks later when the acidity of the pouch was stabilized, the second operation was performed. This consisted of isolation and division of the anterior and posterior vagal trunks on the abdominal portion of the esophagus and resection of the distal portion of the stomach from a point approximately 3 cm. proximal to the incisura angularis to just beyond the pyloric ring, thus insuring complete removal of the antrum and pylorus. The stump of the duodenum was closed and an anterior Polya gastro-jejunostomy was established. For the study of gastric secretion a test meal consisting of lean meat and water was used. Specimens were obtained for analysis from the intact stomach before operation; from the pouch and main stomach 12 days after the first operation; and from the pouch and gastric remnant for 6 to 12 weeks after the second operation. The samples were aspirated at hourly intervals for periods of 4 to 7 hours after the administration of the test meal and were examined for free, combined, and total acidity by titrating against N/10 sodium hydroxide. In addition, the effect of atropine upon the secretory response to a test meal was studied before and after subtotal removal of the stomach and section of the vagi.

In a previous communication 1 observations were presented which indicate that in the dog pancreatic secretion is continuous during interdigestive periods, as it is in rabbits and ruminants. In another paper 2 we have discussed the inhibitory effect of ether anesthesia on continuous secretion, which explains the absence of continuous flow in the acute experiments of Bayliss and Starling. It has been claimed that continuous secretion in permanent fistulas is due to lack of neutralization of gastric hydrochloric acid by the pancreatic juice diverted from the intestine. 3

In order to determine whether this factor has an effect on continuous secretion the following experiment was performed. A dog was prepared with a system of altercursive intubation as described by Elman and McCaughan 3 by means of which the pancreatic juice was continuously returned to the intestine through the biliary tract. This consists of an intercommunicating system of tubes leading from the pancreas to the gall bladder in such a way that the flow of juice can be observed on the outside. By this procedure the pancreatic secretion was allowed to enter the duodenum through the ampulla of Vater and neutralization of the gastric juice could take place as in the normal animal. Elman 4 states that the gastric hypersecretion which he reports as accompanying complete drainage of pancreatic juice to the outside, does not occur under these conditions.

It has been shown 1 that specific serological reactions can be obtained with simple chemical substances, consisting of an inhibitory effect upon the precipitation of azoproteins containing corresponding chemical groups. The explanation of these inhibition reactions was obviously that the simple substances combine with the antibodies, thereby preventing the formation of a precipitate with the antigen. Direct evidence of such a combination has been claimed by Klop-stock and Selter. 2 Recently, Marrack and Smith 3 have shown that on addition of the corresponding immune serum, less of an azodye made from p-aminobenzene arsinic acid passed through a collodion membrane than when a normal serum was added. The specificity of the phenomenon was checked with another azodye (methyl red).

We have found that it is possible to obtain direct precipitation reactions with substances of simple chemical constitution. 20 The immune sera used were prepared by injecting azoproteins made from the para-amino derivatives of succinanilic, adipanilic, and suber-anilic acids. For the reactions, azodyes were used made by diazotizing the compounds mentioned above, and coupling with resorcinol or tyrosine. The concentration of the test solutions was 1:10,000 and 1:50,000.

In their specificity the reactions, particularly the group reactions of the dyes made from p-amino-adipanilic and p-amino-suberanilic acid, correspond fully to the reactions observed with azoproteins, which will be described in a later communication. There was a further correspondence in that in both cases the reactions could be inhibited specifically by the addition of the nitroanilic acids. Weak but definite precipitin reactions were also obtained with azodyes prepared from aminotartranilic acid and p-arsanilic acid, and the homologous immune sera.

Data are submitted on 38 animals in which 4 types of strangulation obstruction were produced and followed until death ensued in 3 of the 4 groups. In Group 11, the strangulated loops were resected before death, at a time when the animals appeared moribund.

In the first 2 groups in which the veins were occluded, the arteries were patent or only partially occluded. In these groups the animals apparently died from loss of blood into the bowel and peritoneal cavity. In the last 2 groups in which the arterial occlusion was complete the animals lived about 4 times as long, and undoubtedly died from factors other than the loss of whole blood. Just what caused death in these last 2 groups we are unable to say at this time.

In the last year we studied spinal anesthesia for the dog, and a number of difficulties have been overcome by the utilization of a simple piece of apparatus and the adoption of a certain routine in making the puncture. Our first problem was the development of a method of maintaining the dog in a proper position for the injection. The vertebral interspaces of the dog are very narrow, as compared to those of man and the insertion of the needle is exceedingly difficult unless the dog is kept in a flexed position. In experiments in which premedication with sedatives is inadvisable we have found that the dog can be kept quiet in a comfortable position by the use of a modification of the Delahanty table. This apparatus is readily prepared by the use of a board measuring 1 inch thick, 12 inches wide and 5 feet long. As in the Delahanty table 4 holes are cut through the board in such a manner that the dog's legs may be easily put through them. To insure maximal comfort the holes should be padded. A pillow which maintains the spinal column in a flexed position is placed under the belly. Another pillow is placed between the fore and hind limbs after which the latter are secured. A canvas corset which is attached to the median line of the board is bound about the animal and tied along the dorsum. Lateral movement is reduced by a wooden slat inserted along the corset on each side. By means of this apparatus the dog may be inclined at any desired angle. When the board is placed at an angle of about 45°, flow of spinal fluid is facilitated (Fig. 1).

An extensive literature has developed with respect to the mechanism of pain in gastric and duodenal ulcer. For this reason, in this brief report, we shall make no reference to previous studies but merely describe some observations made during the course of an operation under local anesthesia upon a patient with a duodenal ulcer.

The patient, a well developed white male 37 years of age, first experienced epigastric distress of the ulcer type in 1920. A diet of bland food together with the daily ingestion of powders (presumably alkalies) afforded some relief, but in April, 1922, his distress became so marked that an operation was advised and performed elsewhere. At operation a duodenal ulcer was found and a posterior gastro-enterostomy done. This gave more or less complete relief until 1924, at which time epigastric pain returned more severe than before and persisted intermittently until we first saw the patient in March, 1928. The symptoms and findings at this time were those of a gastro-jejunal ulcer. In April, 1928, the abdomen was opened by one of us (L.R.D.). A jejunal ulcer was found as well as the scar of an old ulcer on the anterior wall of the duodenum about 1 1/2 cm. distal to the pylorus. There was no evident pyloric stenosis. The jejunal ulcer was excised, the gastroenterostomy disconnected, and the openings in the stomach and jejunum closed. Following this operation the patient remained free from distress only until October, 1928, when he returned to the hospital, this time complaining of symptoms similar to those of 1920. The periods of distress bore the time relations to food taking characteristic of duodenal ulcer and the distress itself was completely relieved by food and adequate alkalies.

The hypodermic injection is the undesirable feature in insulin therapy, and therefore limits its usefulness. Many workers have used oral therapy, viz.: (1) Administering insulin by all enteral and respiratory routes. 1 (2) Adding to insulin: serum, defibrinated blood, bile acids, bile salts, acids, or alcohol, to prevent its destruction or inactivation by the proteolytic ferments of the gastro-intestinal tract. 2 (3) Employing substitutes which possess insulin-like properties. 3 (4) Incorporating the insulin in enteric coated capsules to make it available for absorption after passing the stomach and jejunum. 4 All these have proven to be either toxic or of dubious value. 5

We approached the problem by altering the permeability of the membrane to stimulate absorption of the insulin molecule. The method was based upon the following principles: 1. The membrane should not be exposed to ferments that inactivate insulin. 2. Permeability may be increased by elevating the temperature of the membrane. 6 3. The rate of absorption can be speeded by heating the insulin solution. 4. A “piling up” of the molecule on the membrane may be prevented by lowering the surface tension of the insulin solution. 5. All adhering materials should be mechanically washed away from the membrane. 6. The surface of the membrane should be kept at a favorable pH.

The method consisted of: (1) The nose was irrigated with normal saline, at approximately pH 4, at 40-45°C. This was immediately followed by: (2) Either a solution of Saponin was applied to the mucous membrane, or a few drops of Saponin added to the insulin solution. (3) Insulin at 50°C., was then sprayed directly into the nose with an atomizer, the stem of which had been previously heated. (4) Small plegets of cotton were then inserted into the nares to prevent the patient from blowing out the solution.

Although urease is very toxic when injected into mammals it has very little effect upon the hen unless urea is injected simultaneously. This is to be accounted for by the low urea content of hen's blood, which has been found to be about 1 mg. per 100 cc., or even less, in the 8 hens tested. From 800 to 17,000 units of urease were injected at one time, into a wing vein or directly into the heart. The white leghorn hens were observed following the injection and samples of blood and feces were analyzed for ammonia, urea, uric acid and urease. The injected urease disappeared from the blood within 4 hours after the injection and did not appear in the feces. Following injection of urease the blood urea disappeared entirely, its place being taken by ammonia. The uric acid content of hens' blood (2.0-4.0 mg. per 100 cc.) was not apparently affected by destruction of the urea, as might be expected if urea were a necessary precursor of uric acid. Examination of the feces showed that the reaction becomes slightly alkaline after the injection of urease.

Antiurease was formed in 8 hens by 4 to 10 injections of urease (each injection containing from 500 to 5,000 units) over a period of 30 to 50 days. Antiurease could be demonstrated in the blood about 14 days after the first injection. There was an incubation period of 7 days after the last injection. The amount of antiurease found in the hens, as determined by the method of Kirk and Sumner, 1 varied between 5 and 24 antiunits per cc. of serum. The chicken antiurease which has been obtained is similar to rabbit antiurease.

Although silicic acid has long been known to occur in practically every tissue of the living organism there has been no evidence that it is in combination with organic matter as are, for example, sulphuric and phosphoric acids. Drechsel 1 long ago reported its occurrence as an ester of cholesterol in bird feathers but there has been no confirmation of his observation since.

Kraut 2 has recently reported that the silicic acid content of human blood is constant in any one individual at different times, that it varies considerably in different individuals and that a temporary variation can be brought about in any one individual by the administration of silicates.

By the electro-dialysis of gelatin, ox tendon, horse and ox-blood, and human urine, the writer has recovered, at the cathode, small amounts of a jelly-like substance, containing both silicic acid and organic matter, and giving a test for carbohydrates. This substance is insoluble 20 in water, in 5% hydrochloric acid and in dilute alkali. It chars when heated on platinum and discolors concentrated sulphuric acid when heated. A suspension in water slowly gives the Molisch test for carbohydrates. It is changed by heating with 10% sodium hydroxide so that the organic matter and a part of the silicic acid go into solution, while the residue now left is soluble in 5% hydrochloric acid and contains calcium, iron and silicic acid. It does not give a test for cholesterol.

The organic fraction of the substance appears to be comparatively small and it cannot be stated to what extent a more complex substance was disintegrated by the electro-dialysis.

Unlike the present substance, colloidal silica is negatively charged and migrates to the anode.

Zucker and Matzner 1 showed that when rats develop rickets as the result of being placed on a standard rickets-inducing ration, high in calcium and low in phosphorus, the feces become more alkaline and that on the administration of cod liver oil the reaction veers quickly to the acid side. These observations have been amply substantiated. Grayzel and Miller 2 showed later that the reaction of the intestinal contents of dogs is acid throughout almost the entire length of the gut, but that this reaction tends to become alkaline when rickets is brought about. As in the case of the rat the reaction reverts to normal on the administration of cod liver oil or treatment with ultra-violet irradiation.

Our investigation was carried out on white rats weighing approximately 50 gm., i. e., about 4 weeks of age. Its main purpose was to ascertain whether rickets leads to a change in the flora of the intestinal tract and whether, after the animal is treated with ultra-violet rays and healing of the rachitic lesion is brought about, there was any associated modification of the flora. The rats were fed the rickets-inducing ration of McCollum and some were irradiated with a mercury vapor lamp. Immediately after the animals were killed by trauma they were autopsied and segments of the intestinal tract were tied off; that of the small intestine extending from the duodenum to the lower ileum and that of the large intestine extending from somewhat below the caecum to the beginning of the rectum. The content of each of these segments was squeezed out and the H-ion reaction obtained as soon as possible by a colorimetric method (Clark and Lub indicators). The material was then suspended to a standard density in normal saline solution, gram-stained films prepared and cultures made by methods adequate to determine quantitatively and qualitatively the character of the flora.

Waddell, Elvehjem, Steenbock and Hart 1 showed that anemic rats on a milk diet receiving a supplement of 0.5 mg. iron daily failed to materially increase their hemoglobin levels. Acid extracts of ash residue from liver, lettuce and corn when fed in amounts sufficient to supply 0.5 mg. iron daily, however, induced hemoglobin regeneration. Subsequently, Hart, Steenbock, Waddell and Elvehjem 2 presented evidence that the response elicited by vegetable and meat tissue ash is due to the presence of copper in small but sufficient amounts to act as a supplement to the iron. They also stated that their experiments failed to demonstrate the existence of an organic factor necessary for hemoglobin synthesis. Later work by Elvehjem, Steenbock and Hart 3 failed to substantiate the claim of Drabkin and Miller 4 that glutamic acid may serve to stimulate hemoglobin regeneration in anemic rats on a diet of whole milk with additional iron.

Myers and Beard 5 , 6 report hemoglobin regeneration with but 0.25 mg. iron unsupplemented by other metals when fed with a milk diet. Larger doses, and supplemental metals as copper, manganese, arsenic, etc., gave more rapid hemoglobin regeneration.

This paper reports results obtained in experiments on rats with nutritional anemia when a milk regimen is supplemented with spinach, apricot, or liver.

Albino rats at the time of weaning (28 to 30 days old) were placed in individual glass cages on a diet of whole raw cow's milk (collected in glass) for a period of 70 to 80 days.

The experiments described below include (A) in vitro digestion experiments on irradiated and non-irradiated proteins with pepsin and trypsin respectively and (B) growth and metabolism studies on rats kept on low protein diets where the protein was either irradiated or non-irradiated.

A. In vitro Digestion Experiments on Irradiated and Non-irradiated Purified Proteins with Pepsin and Trypsin, Respectively. 1. Tryptic Digestion of Casein. The irradiated product was prepared by exposing dry casein (Merck according to Hammarsten) for 2 1/2 hours at a distance of 18 inches to the mercury quartz lamp in a current of air. The digestion experiments with irradiated and non-irradiated casein were carried out as described by Helmer. 1 At the end of the digestion period the undigested protein was precipitated and the refractive indices of the filtrates read.

As shown in Table I, the non-irradiated casein is somewhat better digested than the irradiated. The refractive index readings were confirmed by actual determination of nitrogen in the filtrates.

2. Peptic Digestion of Egg-White. Dried egg-albumin was prepared and digestion experiments carried out according to the method described by McMeekin and Freeland. 2 A portion of the dried, powdered material was irradiated for 4 hours in a current of air; the longer irradiation period being used because of the coarse physical state of the protein. Here again the amount of digestion was slightly greater for the non-irradiated protein than for the irradiated (Table II).

B. Growth and Metabolism Studies on Rats Kept on Diets Low in Irradiated and Non-irradiated Proteins. 1. Group Experiments. In spite of the very slight differences in digestibility between the irradiated and non-irradiated proteins, the final test of actual feeding experiments was obviously the next step.

The immediate cause of thyroid hyperplasia in all probability is a relative or an absolute deficiency of iodine. 1 The fundamental or essential cause of goiter is unknown, but the search for the essential cause, as we have often suggested, appears to resolve itself into determining the cause or causes of the iodine deficiency. As most iodine deficiencies are relative rather than absolute, the search further limits itself largely to determining the factors which create the increased needs of the organism for the iodine containing hormone. The simplest way of increasing the need of the thyroid for iodine would be by depressing the utilization of oxygen in the tissues, and the discovery by Chesney and Webster 2 that the prolonged feeding of cabbage caused thyroid hyperplasia in rabbits appeared to offer a practical means of testing this hypothesis.

It has been shown that there are great seasonal and climatic variations in the goitrogenic activity of cabbage, 3 that drying in a current of air or in vacuo causes a loss of the goitrogenic agent, 4 that prolonged steaming does not impair and under certain conditions may increase its goitrogenic power, 5 that boiling for 30 minutes at pH 3.0 (HC1) does not injure it, 6 that the goitrogenic substance may be extracted from cabbage with ether and other ethereal solvents 7 and that this substance is but slightly extracted by prolonged aqueous leaching.

Since all the Brassicae so far tested may produce goiter and since mustard oils (isothiocyanates) are the most characteristic constituents of these plants it was thought that their goitrogenic activity might be connected in some way with these substances or with their cyanide precursors. Several of the mustard oils (allyl, ethyl, phenyl) have been fed to rabbits with negative results.

In the cortex of the rabbit a small area can occasionally be found that is supplied by one artery and one vein. Such a region can thus be isolated by incisions except for the tongue of tissue where the vessels enter, without serious interference with the blood supply. A metal plate slipped under this tissue and connected to ground serves as an indifferent electrode, and the end of a fine wire resting on the cortical surface serves as a test electrode. When the region so isolated is not active no record is picked up from activity in the rest of the cortex. In 2 cases such a preparation has shown activity, and its pathway could be traced via the tongue of intact tissue. The activity consisted of a succession of wavelike action potentials, about 3 per second, lower than those of the adjacent intact cortex and much less complex. The rotating interrupter of the oscillograph apparatus was adjusted to a speed almost synchronous with these waves, but slightly slower, so that successive waves appeared to progress slowly across the screen. If a stimulus is sent in to the tissue at each revolution of the interrupter, it will fall later and later in successive waves. When stimulated during the negative phase or immediately afterward, no response is elicited, the response becoming larger the later the stimulus falls in the cycle. The response consists of a wave but little shorter than the spontaneous wave, and inhibits the following wave, in which case there is a compensatory pause, but it does not otherwise alter the rhythm which is imposed from without the circumscribed region. The response to stimulation is too protracted to be assignable to nerve fibers directly stimulated, and is presumably due to nerve cells.

The technique of Stewart and White 1 for the determination of “blood fat” and its modifications by Himwich, Friedman and Spiers 2 has yielded higher figures than have the methods developed by Stoddard and Drury, 3 and Stewart, Gaddie and Dunlop. 4 The essential technical difference of these methods is that in the Stewart and White 1 procedure the liberated fatty acids are titrated directly, whereas, by the other methods the fatty acid precipitate is washed prior to titration. For this reason it is possible that the values obtained by the Stewart and White 1 technique may include substances washed away in the other procedures. Certain fatty acids of the blood may be among those so washed away, for example, arachidonic and other unsaturated fatty acids. On the other hand, extraneous acids may also be lost in the process of washing, such as phosphoric or organic acids.

The quantitative aspects of arachidonic and other unsaturated fatty acids of the blood are to be investigated in this laboratory. Stewart, Gaddie and Dunlop 4 suggested that phosphoric acid liberated from lecithin is the substance responsible for the difference in results obtained by these methods but we have not found it possible to determine any significant increases in inorganic phosphorus as a result of the Stewart and White 1 procedure. On the other hand, the possibility remained that inorganic acids arising from the breakdown of glucose might account for some of the divergencies in results. Therefore, determinations of the “fat content” of the plasma with and without the addition of glucose were made by the method of Stewart and White. 1 It was found that the titration value of “fat” rose when glucose was added to plasma.

No successful attempt to inhibit or prevent development of a metazoan parasite by injection of immune serum has been recorded. It has previously been shown that the rat can be artificially immunized against infection by onchospheres of the tapeworm Taenia taeniaeformis by injections of worm material 1 ; and that rats infected with the larval stage of this worm, Cysticercus fasciolaris are thereby protected against superinfection. 2 Serum from artificially immunized rats was later found to confer a high degree of protection against infection by onchospheres of this cestode. 3 The results of 2 experiments show that complete protection has now been secured by the injection of serum from rats infected with cysticerci.

In the first experiment 36 rats from 4 litters were evenly divided into 3 groups of 12 each. All animals were given equal portions of a uniform suspension of onchospheres by stomach tube. The 12 rats of group C were reserved for controls; 2 hours after feeding onchospheres the animals of group A received pooled serum from infected rats, and those of group B serum from artificially immunized rats. In the 2 latter cases 1 cc. of serum per 25 gm. of body weight was injected intraperitoneally. All rats were autopsied 37 days later; summarized data are shown in Table I.

Similar results were obtained in Experiment 2, in which 31 rats from 4 litters were used. Sixteen control animals received injections of serum from uninfected rats, and 15 experimental animals received pooled sera from infected rats. One cc. of serum was given per 25 gm. of body weight before, and again immediately after feeding onchospheres. All animals were autopsied 32 days later; the data are summarized in Table II.