Experimental Biology and Medicine

Sixty isometric contractions of isolated frog gastrocnemii against a tension lever at a rate of 10 per minute increased lactic acid 38 and hexosemonophosphate (as hexose) 28 mg. % over resting values of 23 and 51 mg. %, respectively. Sixty contractions at a rate of 60 per minute increased lactic acid 42 and hexosephosphate 56 mg. %. The average amount of work produced in the 2 series of experiments was practically the same. It may be seen that the average amount of lactic acid formed per 60 twitches was not appreciably changed by the rate of stimulation, while hexosephosphate increased much more with rapid than with slow stimulation. A 10 second isometric tetanus produced on an average an increase in lactic acid of 37 and in hexosephosphate of 64 mg. %. For an about equal amount of lactic acid produced, the hexosephosphate increase was greater in the tetanus than in the single twitches at a rate of 60 per minute.

Owing to variations in the amount of work performed by individual muscles, it was possible to construct curves by plotting lactic acid or hexosephosphate against work. Lactic acid was proportional to the work both in single twitches and in tetani and the curves intersected zero work at the basal value of lactic acid as determined experimentally. Hexosephosphate was not proportional to the work; it rose rapidly with small amounts of work and less so as more work was produced, both in single twitches and in tetani.

Oxidative removal of lactic acid and hexosephosphate was excluded by performing some of the experiments in nitrogen. The difference in time between slow and fast stimulation (5 minutes) is too small to allow hexosephosphate to disappear, as shown in experiments in which the rapidly stimulated muscle was fixed 5 minutes after the end of stimulation. Hexosephosphate and lactic acid account for most of the glycogen broken down during contraction. The greater increase in hexosephosphate with rapid stimulation represents an extra breakdown of glycogen, while the relation of hexosephosphate to work suggests that as work is increased proportionately less glycogen is broken down.

The antigenic relationships between diphtheria bacilli and certain derived variants have been investigated by means of the complement fixation reaction.

Two strains of virulent diphtheria bacilli, No. 19 and Park No. 18 and avirulent variant strains derived from these strains were used in the experiments, and in addition, a strain of C. hofmanni.

Two of the variant strains, Park No. 8 phage a and Park No. 8 phage b, were derived from young cultures of Park No. 8 to which specific bacteriophage had been added. Four of the variant strains were derived in various ways from the No. 19 culture and are designated as follows: 19 ‘antiserum’, derived from culture in broth containing homologous antiserum, 19 ‘broth 45°C.’ from broth culture incubated at 45°C; 19 ‘LiCl’, from culture in broth containing 0.5% lithium chloride; 19 ‘phenol’, from culture in broth containing 0.1% phenol.

These variant organisms were found to be stable in type, grew very slowly in broth, and were completely lacking in fermentative ability. The variant colony was much smaller than the typical colony of diphtheria colony. Microscopically, the variant organisms were hardly distinguishable from the parent organisms. Usually they stained more solidly, but many barred and granular forms were observed.

Sera taken from some of the rabbits before immunization failed to fix complement except when large amounts of serum were used. The course of immunization consists of 8 to 10 injections of heat-killed organisms. The sera obtained were inactivated by heating them at 60-62 °C. for 30 minutes.

The sera were used in 4 amounts: 0.025, 0.012, 0.006, and 0.003 cc. All antigens were used in such an amount that 4 times that amount was not anticomplementary. The smallest amount of serum giving a strong reaction with the homologous antigen was used as a base and the reactions in this amount with other antigens is reported in the tabulation.

Fromherz 1 has recently investigated the effect of veratrine and other drugs upon the electrical response of medullated nerve, using a moving coil galvanometer as a measuring instrument. He detects little difference in response between normal and veratrinized nerve, if the nerve is suspended in oxygen throughout. If, however, veratrinized nerve be asphyxiated and allowed to recover in oxygen, he finds that the after-potential after a single shock or short tetanus may last as long as 10 minutes. We have reinvestigated the action of veratrine on the sciatic nerve of R. pipiens with the cathode ray oscillograph, and find that if the concentration is suitably adjusted to the temperature, typical veratrine after-potentials as measured out to one-half second can be obtained whether the nerve is in air or in oxygen. The results of Graham and Gasser 2 are therefore not due to partially asphyxial conditions, as suggested by Fromherz 1 and Hill. 5

In order to study the after-potential throughout its course we have modified the usual oscillographic technique by employing a voltage amplifier of the direct-coupled type, especially designed to avoid amplifier drift (the theory of this amplifier has been published by Schmitt 3 and the details of the circuit as modified for electrophysiological use will be published elsewhere). By this method the resting potential is followed continuously so that due allowance for drift in resting potential can be made in estimating the duration of the after-potential. After a shock the potential changes can be followed for the first 3-4 seconds by photographing a slow sweep of the oscillograph beam; thereafter the potential may be measured every few seconds for as long as desired by reading the balancing potentiometer, using the oscillograph at high amplification as a null instrument. Preliminary experiments have revealed after-potentials lasting as long as 6 minutes after a single stimulation of veratrinized nerve in oxygen after previous asphyxiation. Without asphyxiation the potential seldom lasts longer than 10-30 seconds.

The mitotic activity of the thyroid gland in the guinea pig can be used as a quantitative measure of the stimulating effect exerted by various agencies on this organ. 1 In the following investigations we extended this method to the adrenal gland of the guinea pig and we determined the number of mitoses in this gland in the normal animal and under various experimental conditions. We soon found that there is a remarkable similarity in the mitotic activity in various sections through this organ at the same time, so that it is possible to obtain the measure of the mitotic activity of a gland by counting the mitoses found in 3 longitudinal sections through the center of this organ, each section being approximately 6 mikra thick. Counting the number of mitoses in a larger number of sections does not cause a significant alteration in the figures. The large majority of all mitoses are found either in the zona glomerulosa or in the outer portion of the zona fasicularis and it seems that, in the former area, mitoses occur especially in younger individuals. In these investigations, we excluded all animals which showed any sign of infection. In normal guinea pigs weighing 190-200 gm. the average number of mitoses per section is about 5; this number declines with increasing weight and age. In guinea pigs weighing 400-600 gm. it is about 1.5 mitoses per section. This series of normal animals includes 87 cases. There is no appreciable difference between males and females. The total number of mitoses in one adrenal gland of a guinea pig weighing 195 gm. was 1340, in the adrenal gland of a guinea pig weighing 340 gm., 930, and in a guinea pig weighing 515 gm., 165. Canalis stated that in sections of suprarenals from normal guinea pigs there might be 5 mitoses, in others as many as 12 per section. 2 Hoerr states that in sections cut 3 to 4μ one may expect to find on the average 2 to 8 mitoses in a longitudinal section through the center of a gland of a young adult guinea pig (350-450 gm.). 3

In view of preceding observations that injections of anterior pituitary glands of cattle cause a marked stimulation of the thyroid gland in guinea pigs, 1 leading to a loss of weight by the animals 2 and to an increase in the basal metabolic rate, 3 it was of interest to investigate the effects of such extracts on the carbohydrate metabolism.

Artundo and Mazzocco 4 found no change in the glycogen reserves of white rats after injections of fresh rat hypophysis in saline. Eitel and Loeser 5 have reported studies on guinea pigs, using anterior pituitary extracts containing thyroid stimulating hormone. They noted a fall in liver glycogen, reaching minimal values in 11 days, with a slow increase subsequently. They report no change in muscle glycogen. Our results are somewhat different in certain respects. We likewise used guinea pigs, which had served for the large majority of experiments carried out by Loeb and Siebert. 6 Male guinea pigs (190-200 gm.) kept under optimal conditions were used in all experiments. The extract was that used previously in this department. 7 Injections were made intraperitoneally with sterile precautions. The animals were killed 24 hours following the last injection and glycogen determined by a modification of the Pflüger method, 8 blood sugars after Shaffer and Somogyi. 9 Two or more animals were employed for each interval recorded and all glycogen determinations were made in duplicate. The points in the accompanying graphs represent averages.

For several years, attempts have been made to protect chicks against nutritional encephalomalacia 1 (1) by supplementing Diet 108† with various natural foodstuffs; (2) by varying the proportion of the ingredients. A summary of these experiments is given in Table I.

Although there have been indications that the addition of certain foodstuffs such as wheat, corn, lettuce, spinach and grass conferred partial protection, the results have been by no means consistent. Various modifications in the proportion of the mineral and organic ingredients of Diet 108 have likewise failed to induce complete protection. It has been noted that reduction of the lard content regularly lowers the incidence of the disease, but even complete omission of lard from the diet does not give absolute protection. This indicates that the disease is not due primarily to toxicity of the lard.

When certain vegetable oils were substituted for lard in Diet 108, all the chicks remained free from the disease. Experiments demonstrating this are presented in Table II. Thus far Crsco, Wesson Oil, Mazola and peanut oil at the level of 20% have afforded complete protection. Olive oil has been less efficacious. Fractionation of these oils is now being carried out in the attempt to gain further knowledge as to the nature of the protective factor.

Attempts to concentrate the vitamin G of food materials 1 , 2 , 3 before completion of the work here reported, have been complicated by the postulation of a multiplicity of factors in vitamin G.^4-14 While the question of this possible multiple nature of the vitamin still remains very unsettled, much progress has been made in the concentration of the vitamin. 15 , 16 , 17 The present paper reports additional evidence of the existence of some unknown variant in vitamin G research and describes a method of concentration of vitamin G which is being used as an initial procedure in the subsequent concentration work of Booher. 17

The method used in the evaluation of various materials for their vitamin G content was essentially that of Bourquin and Sherman. 18 During this investigation the small anti-coprophagy harness previously described 19 was first used to minimize the error due to coprophagy. In a group of 37 animals, half of which were harnessed, the average total gain at the end of the fifth week for the group restricted in respect to coprophagy was 18.8 ± 1.2 gm., while in the parallel group, unrestricted in respect to coprophagy, the corresponding gain was 29.4 ± 2.2 gm. (Deviation measure = P.E.). Observed activity of the animals and the greater precision in growth records of the harnessed animals indicates that the harness per se did not restrict growth.

Concentration Experiments. A concentrate of vitamin G from dried skimmed milk powder was obtained as follows: 500 gm. of air dried skimmed milk powder were refluxed successively and with constant agitation by means of a stream of purified nitrogen with 2000 cc, 1250 cc, and 750 cc. of neutral boiling 93-94% ethyl alcohol (by weight) for 30, 30, and 15 minutes respectively.

Aronson 1 isolated a strain of tubercle bacillus from an iguana which was found dead in the Philadelphia Zoo. A number of tubercles were seen in the lung of this animal and also in the liver. The spleen contained no gross tubercles. A single tubercle was found in the upper pole of the left kidney. In guinea pigs injected intracutaneously with this culture Aronson found that a small ulcer occurred at the site of inoculation. The glands became enlarged and succulent but went on to healing. The organism was found to be pathogenic for the chameleon and salamander and also for the frog. It was not pathogenic for the snake. As regards the frog Aronson's findings have been confirmed by Holly 2 and by the writers with a transfer of this organism which Aronson sent to us sometime ago for further study.

Examination of stained slide preparations reveals the organism to be acid- and alcohol-fast. Ziehl-Neelsen stained vertical sections from growing colonies made with a technique reported by us 3 show some non-acid-fast rods in addition, but the tiny non-acid-fast rods and granules so conspicuous in vertical sections of human and bovine tubercle bacillus colonies are entirely lacking. It would seem then that this organism multiplies principally by simple fission. According to Henderson and Aronson 4 the organism is culturally and serologically identical with Mycobacterium marinum which was isolated from salt water fish.

In view of the pathogenicity of this organism for frogs it occurred to the writers to determine whether a similar situation existed for tadpoles. If this proved to be the case the tadpole would make a valuable experimental animal for certain problems as the entire animal may be sectioned and all of the organs examined in the same creature at the same time. It would also permit investigations concerning the reaction of embryonic tissue toward infection by a member of the Mycobacterium group.

The question has frequently been raised 1 , 2 as to whether there are vascular zones other than the carotid sinus and the aorta that make an afferent contribution to cardiovascular reflexes. A region that might well be suspect is the mesenteric area and we have accordingly investigated the problem of the afferent nerve discharge therefrom especially with relation to blood pressure.

As a means of approach impulses were recorded in the peripheral end of the splanchnic nerve or in its terminal branches which run along the mesenteric arteries. Characteristically there is a discharge of impulses synchronous with each pulse as in Fig. 1, although in certain cases there is a considerable amount of activity in the intervals between succeeding systoles. The general picture of this afferent discharge is similar to that observed 3 in the carotid sinus and aortic nerves.

The modifications effected in the discharge by variations in intravascular pressure are likewise similar to those found in the case of the carotid sinus and aorta although less constant and predictable. When a portion of the mesenteric artery is perfused at steady pressures it is found that the end organs begin to discharge at a certain minimal pressure, the impulse frequency and number of active end organs increasing with further increases in pressure. In the intact preparation the degree of discharge is largely influenced by the magnitude of the pulse pressure, its rate of change, and especially by the blood volume. An increase in the latter increases the frequency of the afferent impulses. An increase in mean blood pressure, especially if the previous pressure is not too high, also produces an increase in the discharge although in some cases an increase in blood pressure resulting from an injection of adrenalin gave an initial decrease in the number of impulses. This may be due to a change in the relation of the vessels to the end organs.

The sensitization of guinea pigs with simple compounds such as primulin, 1 salvarsan, 2 and p-phenylene diamine 3 has been studied by several authors, but the development of the subject has been somewhat hindered by difficulties encountered in reproducing some of these experiments. 4

As already observed by Mayer, we found that guinea pigs rubbed repeatedly with salve containing p-phenylene diamine showed increased redness and scaling, in comparison to controls. Furthermore, increased sensitivity was obtained by the administration of a salve containing p-nitroso-dimethylaniline and by repeated intracutaneous injections of a saline solution of p-phenylene diamine, and some such effects were noticed, also, after injecting p-nitroso-dimethylaniline and p-amino-phenol, using approximately 1 mg. of the substances per injection.

Following this, smaller doses of 0.02 mg. were injected on the back twice a week over a longer period, a method similar to that advocated by Kolle 5 for the production of hypersensitiveness to salvarsan, in animals. With this method definite sensitization effects were obtained, besides salvarsan, with other compounds such as p-phenylene-diamine and p-nitroso-dimethylaniline. Thus, on reinjection on the sides about 7 days after the last injection of a course consisting of up to 20 injections at one or 3 day intervals, a number of the guinea pigs showed pinkish, more or less elevated areas within 24 hours, averaging 3/4-1 cm. in diameter. Likewise, with 1:2:4 dinitro-chlorbenzene∗ distinctly increased reactions, noticeable even after less than 10 injections, were observed in most of the guinea pigs after daily intracutaneous injections of the substance. In a preliminary experiment with suberanilic-resorcine dye, 6 2 out of 6 animals treated in the manner described showed definite reactions, so that one may assume that sensitization is possible with this substance, also.

Bordet was the first to recognize a relationship between thrombin and complement. Fuchs 1 has recently claimed that the mid-piece of complement is identical with prothrombin. Since the complement content of plasma and serum are apparently the same, the change of prothrombin to thrombin does not seem to influence complement. A close association of prothrombin and thrombin with complement is indicated by:

1. Both complement and thrombin are non-specific. Complement obtained from one animal will react with the amboceptor and antigen of another species. Thrombin from various animals will clot fibrinogen irrespective of its source.

2. Both thrombin and complement, particularly the mid-piece, are associated with the globulin fraction of the serum proteins.

3. Prothrombin, thrombin, and complement are inactivated at 56°C, and the resulting serum or plasma becomes antithrombic and distinctly anticomplementary (the plasma more than the serum).

4. Serum or plasma treated with an emulsion of magnesium hydroxide becomes antithrombic (anti-clotting), and anticomplementary, with the effect more pronounced in the plasma than in the serum.

5. Heparine inactivates both complement and thrombin.

6. Calcium inhibits both complement and clotting. For the reactivation of prothrombin in oxalated plasma, there is an optimal concentration of calcium, below and above which clotting is retarded. Absence of calcium does not inhibit complement.

That prothrombin is not identical with complement is indicated by: 1. Alumina cream will remove prothrombin from plasma but will not alter the complement content. 2. Patients have been found whose plasma is anticomplementary, but have a normal clotting time.

It is widely appreciated that the majority of infants up to 3 months of age and frequently longer possess immunity to certain diseases, such as measles, diphtheria, scarlet fever, and poliomyelitis. Explanations for these facts have been varied. Most of the experimental data, however, points to the transplacental transmission of antigen or antibodies as being the most acceptable explanation for the presence of immunity in the new born infants, De Souza and Jorge 1 believing that the placenta may produce non-specific immune substances, utilized cord blood in the prophylaxis of measles. Their studies made during the years 1929 to 1932 showed that with 4-10 cc. of this blood they obtained 67-100% protections in 146 infants and children exposed to measles in hospital wards. Four children exposed at home and treated prophylactically developed ordinary measles. Finkelstein 2 had excellent results using 40-60 cc. of puerperal serum, 84.5% protections in 148 children, against 79.2% with similar doses of convalescent serum in 311 children. Others 3 had less favorable results, using 20 cc. of cord blood.

McKann and Chu 4 believed that by extracting the placenta, substances were obtained which represented maternal blood, fetal blood, and probably extracts from the placental tissue itself, whereas blood obtained from the umbilical cord represented fetal blood only, and contained only those substances involved in immunity to which the placenta is permeable. Using the globulin extract of placenta for measles prophylaxis they obtained excellent results. Forty-three susceptible children, hospitalized for other conditions, and exposed to measles, were injected intramuscularly with placental extract. In no instances did the injected placental extract fail to prevent or modify the measles in the susceptible contacts.

Data are presented which indicate that the type of carbohydrate used in vitamin B and G deficient diets is an important factor in determining whether or not coprophagy will vitiate results when accurate assays of vitamins B and G are desired. Coprophagy appears to be of little consequence when sucrose is used as the sole source of carbohydrate, while dextrinized corn starch offers the possibility of introducing serious errors and incorrect conclusions if rats resort to coprophagy when this carbohydrate is used.

Experiments are now in progress in which an attempt is being made to find an explanation for the results described, and the study is being extended to include other types of carbohydrates.

In a previous publication from this laboratory 1 the danger of coprophagy, as it affects the assay of vitamins B and G, was discussed. Similar observations had been reported by Roscoe. 2 More recently Booher and Kaneko 3 have submitted data which led them to conclude that their assays for vitamin B (B₁) were not vitiated by coprophagy. The last mentioned investigators used raw corn starch as the source of carbohydrate. Since Roscoe had reported that she was unable to obtain characteristic growth responses from coprophagous rats when raw corn starch served as the source of carbohydrate, it seemed highly probable that the type of carbohydrate might be an important factor worthy of further study. This possibility has been further suggested by data obtained in this laboratory during a somewhat related investigation in which it was found that the quantity of vitamins B and G required to produce a unit increment of growth was considerably greater when sucrose was fed as the source of carbohydrate than it was when dextrinized corn starch was used.

The experiments here described were initiated to determine the extent to which growth responses in coprophagous rats could be affected by substituting carbohydrates of various types in the usual basal ration deficient in vitamins B and G. Among the carbohydrates under investigation at present are: corn starch, dextrinized corn starch, glucose, sucrose, and lactose. While the results to date are insufficient for a complete statement, the differences obtained with diets containing sucrose and corresponding diets containing dextrinized corn starch are sufficiently marked to justify a preliminary note at this time.

The method used for estimating the vitality of tissues by measuring their electric resistance has been applied, in the present investigation, to a suspension of yeast heated gradually until all the cells were killed.

A suspension of 3 gm. of yeast (Fleischmann's yeast) in 3 gm. of a N/5 solution of NaCl, was heated from 15°C. to 93° and then cooled to 15° in an Ostwald's conductivity cell immersed in a water bath, and the electric resistance of the suspension was taken at each degree. The time between 2 readings was one minute. A microphone Hummer, working on a 6 volts battery, delivered the alternating current of 1000 cycles, used in the conductivity bridge. The suspension was stirred before each measurement by moving the electrodes up and down. Previous tests showed that the concentration N/5 did not affect the resistance of the cells to an appreciable amount.

The results, plotted on the curve (Fig. 1), show:

1. From 15° to 62° and from about 60° to 15° the resistance behaves like the resistance of a salt solution, and the curve coincides with the resistance curve of a salt solution of convenient concentration.

2. From 62° to 67° the resistance falls abruptly. This irreversible drop is probably due to the death of the majority of the cells.

3. From 67° to 93° and from 93° to about 60° the curve shows a resistance higher than would be expected from a salt solution; at about 60° it coincides again with the typical curve of such a solution. This phenomenon suggests that above 67° the few surviving cells are killed.

At the 1932 meeting of the National Tuberculosis Association, we reported favorable results from the treatment of tuberculous guinea pigs with viosterol and ascending doses of tuberculin. 1 In the present work a similar though more extensive investigation was performed upon albino rabbits. In addition to the viosterol-treated animals, certain groups were treated with artificial sunlight irradiation to determine whether the irradiation might not have an effect similar to that produced by viosterol.

The rabbits were inoculated subcutaneously with a lethal dose of bovine tubercle bacilli (0.01 mg. of a culture, 0.001 mg. of which was known to produce tuberculosis in the rabbit) and were divided into 6 groups of 8 animals each. The respective groups received treatment as follows: I, viosterol; II, viosterol plus tuberculin; III, ultraviolet irradiation and maize oil; IV, ultraviolet irradiation and maize oil plus tuberculin; V (tuberculin controls), tuberculin and maize oil; VI (untreated controls), maize oil. The amounts of maize oil given to the groups that did not receive viosterol were in every case equal to those given to groups I and II as a medium for the viosterol.

Treatment was started 21 days after the inoculation. The original 10,000 X viosterol solution(Link missing) was diluted with maize oil to contain 1.46 mg. in 0.05 cc.; this amount per pound of body weight was given weekly by intramuscular injection. Tuberculin (bovine, O.T.) was given subcutaneously simultaneously with the viosterol in weekly ascending doses from 0.00001 mg. to 18.5 mg. for 34 consecutive weeks. Irradiation was given every second day. For the source of light a carbon arc lamp drawing 66 amperes with 48.5 volts across the arc was used. The intensity of radiation of visible and ultraviolet light from the arc was equal to 80 watts per square meter at a distance of one meter with 38% of this coming from rays shorter than 4000 Ångstrom units. The initial dose for the irradiation was 2 minutes at a distance of one meter from the arc.

Carotene can readily be detected in as low a quantity as 0.01 mg. dissolved in chloroform by means of acetic anhydride and concentrated sulphuric acid. The reaction is similar in method of performance and in final color production to the one employed in the Liebermann-Burchard test for cholesterol. The difference in the duration of the green color developed may serve, however, to differentiate one compound from the other. The green color obtained with carotene develops its maximum intensity at once and fades within 2 minutes, leaving a dirty yellow brown color. The green color obtained with cholesterol develops its maximum intensity at the end of 15 minutes.

Carotene does not interfere with the quantitative estimation of cholesterol in the blood. The extraction with chloroform of a mixture of blood and plaster of Paris removes the cholesterol, but does not release the carotene present.

We have postulated some common chemical configuration, actually present or easily produced as a result of chemical interaction, in sterols, in carotene, in the fat-soluble vitamins A, D, and E, in terpenes, and in 5-membered monoheterocyclic compounds. 1 We have recently demonstrated that a mixture of formaldehyde and concentrated sulphuric acid constituted a very sensitive reagent for the detection of carotene in chloroform solution. 2 Whitby, 3 employing the same reagent, reported for cholesterol a color reaction which, however, is not identical with the one obtained for carotene. In this communication we report our findings with reference to the Liebermann-Burchard reaction with carotene. The reaction is the familiar one used in the qualitative detection and in the quantitative estimation of cholesterol. We have already shown that this reagent reacts with heterocyclic compounds like thiophene, furfurane and derivatives. 4

Carotene gives a positive response with the Liebermann-Burchard reagent, using 10 cc. of chloroform solution, 2 cc. of acetic anhydride and 0.2 cc. concentrated sulphuric acid. With low concentrations of carotene a green color is obtained, which matches in color the one developed when cholesterol is treated with the same reagent. In higher concentrations the green color is tinged faintly blue, but can still be matched with the color obtained with cholesterol. Copper-colored crystals of carotene exposed to sunlight and air turn yellow brown. Carotene, thus oxidized, does not give the characteristic color when dissolved in chloroform and treated with acetic anhydride and sulphuric acid. A brown color instead of the green develops. Old solutions of carotene in chloroform also react in the negative. Baumann and Steenbock 5 have recently shown that carotene in organic solvents deteriorates on standing. We therefore used freshly prepared chloroform solutions of carotene for our tests.

We have reported 1 the successful cultivation of the European strain of Typhus rickettsia in flask cultures containing guinea pig tunica. These cultures have now passed successfully through 7 subcultures during a period of 4 months.

In view of successful immunization of guinea pigs with formalized rickettsia suspensions from infected lice 2 and Zinsser's 3 experience with formalized rickettsia suspensions of the Mexican type from infected X-rayed rats, it appeared desirable to test the possibility of using flask tissue cultures for this purpose. Our efforts to immunize guinea pigs with formalized infected brain tissue proved unsatisfactory. Dead virus produced at best only a slight degree of immunity, when large amounts (10 gm.) of formalized infected tissue were injected. Solid immunity could only be obtained with tissue formalized for 1-2 hours, that is tissue still containing live (attenuated?) virus. The difficulty seems to lie in the quantity of organisms, or in other words the amount of antigen, injected. The cultures appeared to offer a satisfactory solution to the problem. On the one hand, the organisms are present in fairly large quantities and, on the other, the cultures are more easily handled than lice.

Guinea pigs have been successfully immunized with formalized suspensions of fresh virulent cultures as well as with old cultures no longer infective for guinea pigs.

Experiment 1. The material from 2 flasks of an 11-day-old virulent culture, second generation, was mixed, centrifuged, the supernatant fluid decanted, the sediment thoroughly triturated in a glass mortar and resuspended in the clear supernatant fluid. A formalin solution was added to the suspension to give a 0.1% concentration and the material left overnight at ice box temperature. Each of 2 guinea pigs was given 3 injections intraperitoneally of 1.5 cc. of the diluted formalized material, at intervals of 3 days. Each animal received the material of one culture. None of the animals developed any sign of fever or even a slight rise in temperature. The temperature during the immunization did not exceed 39°C, that is, it remained within the normal range of temperature of our guinea pigs.

The factors determining the course, grouping, distribution, and size of peripheral nerves in development are still very much in the dark. One way of investigating these factors is direct experimentation on the embryo (Harrison, 1 Braus 2 ). Another way is to study the development of nerves in regenerating limbs of adult urodele Amphibians. Limb regeneration in these animals is known to be a sort of second development (P. Weiss), differing from the first development in the mass of tissue involved, the time relations, the conditions of muscle differentiation and of nerve supply. A comparison of nerve patterns established during regeneration with those formed during embryogenesis can be expected to yield some clue as to which factors may, and which may not be significant. If merely the nerves supplying a limb are severed, nerve regeneration into the limb takes place from the proximal stumps, the outgrowing fibers mainly following the old degenerated peripheral pathways which have persisted as such. If the whole limb is amputated, differentiation of a new limb occurs out of an undifferentiated blastema in which no old pathways are present to guide the outgrowing nerve fibers. The question arises as to what sort of pattern will be assumed by the nerves in the regenerating limb.

In metamorphosed individuals of the newt, Eurycea bislincata (Green), of varying age and size, the left fore limb and the right hind limb were amputated at their bases. The limb nerves were cut at about the level of the girdle plexus. Eight animals were kept for from 176 to 202 days, and were then preserved for dissection. Regeneration had taken place in all of the hind limbs and in 5 of the fore limbs. The regenerates varied from one-half to three-quarters the length of controls on the opposite side. They were slenderer in proportion but otherwise well formed. Some of the animals died before observations were made on the function of the limbs, but in all those observed there was normal coordination of movements. In one case the extension of the elbow was somewhat impaired.

It has been previously reported 1 that injections of an oestrin-free placenta extract, containing the A. P. L. factor, into immature female rats induced a marked increase in the weight of the ovaries and a moderate increase in the weight of the pituitaries. Histologically, the pituitaries of the injected rats showed a slight to moderate decrease in the level of the eosinophiles; a considerable number showed evidence of granular loss. The most evident changes were found in the basophiles, which showed marked evidence of granular depletion.

Injection of pregnancy urine extracts into immature female rats gives similar results. Thirty rats, 21 to 25 days old, were given daily injections of pregnancy urine extract† (25 to 75 R. U. daily for 6 to 15 days). Twenty-eight littermate sisters served as controls. The ovaries of the injected rats weighed from 50 to 80 mg., while those of the controls weighed from 15 to 25 mg. The pituitaries of the injected animals were usually, but not invariably, increased in weight. Complete serial sections of all pituitaries were cut, at 2 micra, in the horizontal plane. One section from each ribbon of 15 was mounted and stained. Cell counts were made on 5 sections, equidistant apart from each set of serial sections (110,039 cells were counted on the 208 sections studied). The anterior lobes of the injected animals were definitely altered from those of the controls. The eosinophiles were almost invariably reduced in number, and many of them were swollen and showed granular loss. The most marked changes were found in the basophiles, which showed reduction of granules. There was an increase in the percentage of the chromophobes. Mitoses were found in the chromophobes and eosinophiles of both the injected and control animals; they were more abundant in the injected animals. Mitotic division has never been observed in the basophiles.

It has been shown 1 that rigid exclusion of fat from the diet of the rat induces a characteristic deficiency disorder, since the animal is apparently unable to synthesize certain fats which are necessary. The present study was undertaken to determine whether the histological fat distribution in the various organs is affected in this disorder. The organs studied include the liver, kidney, trachea, lung, skin and tela subcutanea.

Through the courtesy of Doctors Burr and Brown, 26 typical test rats that had been maintained several months on the fatless diet (No. 550B, plus vitamin supplements) were available, together with 5 rats “cured” by the subsequent addition of the necessary fats to the diet. These rats were killed by chloroform and autopsied in the usual way. The fats were studied chiefly in formalin-fixed, frozen sections stained with a solution of “oil red O” in diacetin (which presumably stains all fats and lipoids), though in some cases other methods were used. The technique is explained in detail in a separate paper dealing with the normal distribution of fats at various postnatal stages in the rat. 2 Both test rats and controls were females of the Wistar strain.

For comparison, 3 types of controls were used. (1) Eleven direct controls were reared on the same fatless diet (550B) with the addition of the necessary vitamins and fat. This is a high carbohydrate diet, containing 84% sucrose. (2) Eight stock diet controls were reared on our normal colony diet (McCollum's diet I), containing 67.5% ground wheat, 15% casein, 10% whole milk powder, and 5.2% butter-fat. (3) Since the test rats were retarded in growth and somewhat emaciated, 5 inanition controls were held by simple underfeeding of the stock diet at a body weight corresponding to that of the test rats.

We have concluded (MacKay and MacKay 1 ) that in the albino rat the kidney weight has practically the same relation to body surface at all ages. In a broad sense this is true. However, a close inspection of the data presented at that time, omitting the less reliable figures obtained after 400 days of age, shows a slight but definite fall in the kidney weight: body surface ratio with increasing age. It has seemed desirable to call attention to this fact and to point out the cause.

Attention has been drawn (MacKay and MacKay 2 ) to the decrease in the caloric intake with increasing age so that with a given protein concentration in the diet the older the rat the less protein is ingested each day in relation to body surface. The protein taken in the food has been shown (MacKay and MacKay 3 ) to be an important factor in determining the weight of the kidneys. This relationship is expressed for casein as protein intake = 0.0183 renal weight — 2.75. With this formula there are 150 mg. of kidney per 100 sq. cm. body surface with a zero protein intake. Calculated from the latter factor any kidney weight above 150 is then due to the dietary protein. Some of our data have been corrected in this manner and reduced to a zero protein intake basis. The original and corrected data are plotted in Fig. 1.

When corrected for the effect of the dietary protein the kidney weight (average of both kidneys) in relation to body surface is more nearly the same at all ages. Each point represents an average of from 12-24 male rats.

In the course of studies involving allantoin, difficulty was experienced in finding a suitable method for the determination of this substance in urine. The method recently introduced by Larson, 1 while including such favorable features as rapidity and the avoidance of the hydrolysis of allantoin, was not found satisfactory under varying experimental conditions. Since the Folin ammoniacal copper reagent employed in this method is not specific for allantoin, further removal of interfering substances from the urine prior to its addition to the copper reagent seemed desirable. A method involving this modification is the basis of the present report.

Following the precipitation of interfering substances with phosphotungstic acid and lead acetate, the removal of excess of the heavy metals is effected with hydrogen sulphide. The allantoin is then precipitated with mercuric acetate, the precipitate washed and redissolved with hydrochloric acid. The solution is then freed of mercury by means of hydrogen sulphide, and the allantoin in a relatively pure solution is determined colorimetrically with the Folin ammoniacal reagent adopted for this purpose by Larson.

The amount of phosphotungstic acid to be added to the urine is determined by preliminary tests as follows: To a series of 6 tubes add 2 cc. of neutral urine. To the first add one cc. of 10% phosphotungstic acid and to the others increasing amounts, namely 1.5, 2.0, 2.5, 3.0, and 3.5 cc. The tubes are shaken, allowed to stand for 5 minutes, and then filtered. To each of the clear filtrates, a few drops of 10% phosphotungstic acid are added. The least amount of phosphotungstic acid that yields a filtrate remaining clear despite the further addition of the acid determines the amount necessary for precipitation of interfering substances from the urine sample. A large excess of phosphotungstic acid must be avoided since it leads to bulky precipitates, which cause a loss of solution volume. If no one of the filtrates remains clear after the addition of the acid, the urine is too concentrated and must be diluted.

The influence of insulin on purine metabolism has been the subject of few investigations. Although Kurti and Gyorgyi 1 found no constant effect of insulin on the endogenous excretion of uric acid in man, they reported that insulin led to an increase in blood uric acid during the ingestion of nucleic acid. Quick 2 has shown that 30 units of insulin along with a high carbohydrate diet had no immediate effect on the uric acid output by man. In the dog, Taubmann 3 reported that insulin led to an increased excretion of allantoin, whereas Ogawa 4 found a decrease following the administration of this hormone.

In the present investigation the dogs were fed a Cowgill diet and insulin was injected after nitrogen and allantoin equilibrium had been established. Allantoin was determined by the method previously described. 5

An increased elimination of allantoin following the injection of insulin has been observed in 6 dogs. A typical result is shown in Table I. Dog L was catheterized twice daily, at 8:35 a. m. and at 1:35 p. m., and fed daily at 1:45 p. m. On January 11th, 7 units of insulin were injected at 8:45 a. m. During the 5-hour interval following the injection of the hormone, the excretion of allantoin doubled. Since this effect is produced 19 hours after the ingestion of a purine-poor diet, it would seem that insulin is capable of affecting the endogenous metabolism of purines. In view of the well-known action of insulin hypoglycemia in evoking a secretion of epinephrine, the effect of the latter upon the excretion of allantoin is being investigated.

Tillett and Garner 1 have recently described a hitherto unknown antihuman fibrinolytic function of hemolytic streptococci. This function is demonstrable with S. hemolyticus of human origin, but is not demonstrable with apparently identical environmental or veterinary streptococci. Since the lytic factor is not active against normal rabbit fibrin, many of the results of streptococcus research on rabbits are presumably not applicable to human pathology. We have, therefore, extended the Tillett-Garner tests to include plasma-clots from other animal species. The object of these tests was to find a laboratory animal suitable for a direct study of the topographical distribution 2 and in vivo fibrinolytic function of S. hemolyticus.

The S. hemolyticus used in these tests were 2 human (C 203 and K 96) and 2 veterinary (P 454 and K 158 E) strains kindly furnished by Dr. Lancefield of the Rockefeller Institute. To these we have added a number of local strains, one of special interest being J 1, originally isolated from a clinical case in San Francisco. This strain contains the human-diagnostic carbohydrate fraction recently described by Lancefield. 3

The fibrinolytic tests were made by the Tillett-Garner technic. To 1 cc. 20% oxalated plasma was added 0.25 cc. 0.25% CaCl₂-solution, followed by 0.5 cc. 18-24 hour broth culture of the microorganism to be tested. A semi-opaque agar-like clot is usually formed within 10 minutes. In nonlytic or negative tests no softening of this clot is demonstrable after 24 hours incubation at 37°C. In positive tests, beginning (+), well advanced (++), or complete (+++) liquefaction is noted within from 30 minutes to 2 hours.

We have found that adult rats exposed to low oxygen tension, upon return to normal atmospheric conditions, develop a marked anemia, regardless of whether or not the usual polycythemia occurs.

Sixty-six adult and 11 young rats were exposed to air pressures ranging from 650 to about 360 mm. of mercury for periods of from 2 to 14 days. Red cell counts were made in each case before exposure, immediately after exposure, and then every 48 hours. Two drops of blood were used for the counts and excess bleeding was prevented by temporarily ligating the tail.

The surviving rats responded to the treatment as follows: (1) Twenty-one of the adult rats showed an increase in red cells varying from one to 4 million, following the exposure to low air pressure. Then, within 48 hours, there was a drop amounting to about 45% to 80% of the excess erythrocytes. Later, these rats all developed an anemia lasting in some cases longer than a month, with red cell counts as low as 5 million below the normal. (2) Thirteen adult rats, after similar exposure periods, did not show any apparent increase in the number of circulating red cells. But here again an anemia developed, generally setting in earlier than in the rats of Type 1. (3) Seven young rats, 40 to 60 days of age, with blood counts ranging from 5 to 6 1/2 million, showed increases after the exposure of from 1 1/2 to 4 million red cells. These rats did not develop an anemia with respect to the first blood count. They did show a slight drop after the exposure, followed later by a rise.

Cases of anemias following polycythemias have been reported by many clinicians. “Plethoric anemias” following transfusion of large amounts of blood have been noted.^1-6 In these cases, it has generally been assumed that the cause is an “exhaustion of the bone marrow,” a hyperactivity of “blood destroying organs,” or an inhibition of the bone marrow by some unknown factors.

The minimal dose capable of producing chronic intoxication in the albino rat has been determined for sodium fluoride, sodium fluosilicate, barium fluosilicate, and sodium aluminum fluoride (natural cryolite). Male rats ranging in weight from 30 to 60 gm. were placed on a basic diet consisting of corn meal 73 gm., linseed oil cake meal 10 gm., alfalfa 2 gm., casein 10 gm., codliver oil 3 gm., bone ash 1 1/2 gm., sodium chloride 1/2 gm. For each compound studied 6 control rats were placed on this diet, and for each dosage level 6 rats were placed on the same diet to which a weighed amount of the fluorine compound had been added and evenly distributed by means of a mechanical mixer. Analyses showed even distribution of the toxic agent throughout the basic diet. For each compound a series of dosage levels was employed, each dose being one-half the preceding dose.

The criterion of toxic action was bleaching of the upper and lower, especially the lower, incisor teeth. Table I summarizes the data. In the case of each compound, the dosage level indicated means that bleaching was present in all of the rats on that dosage, but absent when one-half of that dose was administered.

When the fluorine content of the molecule is taken into consideration it is seen that despite a marked variation in water solubility the chronic toxicity of the compounds, expressed as parts of fluorine per million parts of food, is remarkably constant. For all practical purposes, it may be stated that in the compounds studied fluorine is capable of producing chronic intoxication, as indicated by bleaching of the rat incisors, in concentrations as low as 14 to 16 parts per million regardless of the solubility of the compound containing the fluorine. Depending upon the amount of food consumed by the albino rat, this means that from one-half to 1 mg. of fluorine per day per kilo of body weight is capable of producing definite signs of injury to the rat incisors.

Difficulties encountered in administering sufficient amounts of the usual chaulmoogra oil derivatives uniformly to effect a cure in leprous subjects have led to the attempted application of other forms of therapy. Thus, recent clinical developments 1 , 2 again have aroused interest in the efficacy of certain non-chaulmoogry (Link Missing 1) drugs, synthetic dyes in particular, against leprosy. In the course of examining the possible antileprotic activity of certain such compounds by a standardized technique, 3 determination of their toxicity in the test animals was a necessary preliminary. As trypan blue, gentian violet, and brilliant green have been used in an empirical manner in the clinic before a critical pharmacological evaluation of their worth has been made, it is advisable to present the results of this toxicity study on experimental animals, before toxic doses may unavoidably be given to human leprous subjects. The intravenous dosage of these dyes, as reputedly used by Ryrie 1 and Ryles, 2 is within the possible lethal range as estimated from experimental animals. While these workers do not give the sources of the dyes used by them, nor their standards of purity, so that it cannot be judged whether or not their dyes are comparable in purity to those used by us, it is significant that Ryrie 1 admits the occurrence of dangerously severe acute toxic effects in his patients.

Previously reported toxicity data on these 3 dyes are so inadequate and variant that a recent encyclopedic work on chemotherapy 4 omits all reference to their toxicity. Discrepancies in some cases may be explained by variations in purity and by the use of old solutions. Photodynamic action is not a significant factor in the toxicity of these dyes as far as we could determine. We will later make a more extended report surveying the therapeutic effectiveness as well as the chronic and acute toxicity of these agents. Such a study is indicated in view of enthusiastic endorsement of dye therapy in leprosy on grounds that it is as good as treatment with hydnocarpus oil derivatives, safe, cheap and not unduly painful.

The trypsinization of the leprous tissue bits, before attempting the cultivation of the contained B. leprae, affords at once the split-protein products which we consider the necessary nutrient factor. The trypsin apparently exerts no harmful effect upon the Hansen bacillus as evidenced by its unmistakable growth in 4 to 6 weeks. This initial in vitro multiplication occurs in the trypsinized leprous tissue without the aid of other nutrients and under ordinary atmospheric conditions at 37°C. Growth continues in subculture on ordinary agar medium as long as the trypsinized leprous material is transferred. The crucial point in the initial cultivation of B. leprae away from the influence of the host tissue is experienced at this juncture. After the complete utilization of the nutrient material of the trypsinized host tissue, the acid-fast bacillus of Hansen will only multiply in a medium that is enriched with sufficient quantities of protein cleavage products.

Many attempts have been made by various investigators to culture Hansen's bacillus of human leprosy. Undoubtedly, in a number of these attempts the investigators have mistaken the increase in number of acid-fast rods, which occurs in removed autolysing pieces of leprous tissue independent of any specially supplied cultural conditions, for multiplication in vitro upon artificially prepared nutrient medium. The failure to appreciate the facts that B. leprae continues to multiply and that growth ceases when the leprous tissue autolysate is no longer available, explain the increasing difficulty to obtain macroscopic growth of the specific microorganism in transplants through successive generations, and accounts for the absence of growth in subculture in or upon a medium that does not contain the split protein products (amino-acids).

One of us (Duval 1 ) reported that B. leprae seemed unable to utilize whole protein in vivo. This was substantiated by repeated observations that the Hansen bacillus would grow in transplanted leprous tissue after autolysis had commenced. The initial culture away from the influence of the host-tissue was possible if the artificial culture medium contained certain split-products of protein. In our recent studies upon the cultivation of B. leprae, we have modified Duval's 2 earlier method. An improved procedure as described below is recommended.

The uncontaminated subcutaneous leprous nodule, rich in acid-fast bacilli, is the material of choice. Contaminated tissue is unsatisfactory, because one is likely to injure or destroy the viability of the Hansen bacillus when killing the contaminants. The use of sodium hydroxide, or the like, therefore, is not recommended as a germicide. The skin over the selected leprous nodule is cleansed with soap and water and then rinsed with alcohol. Mercurochrome, iodine or other antiseptics should not be employed. The overlying skin is next incised and the exposed nodule is dissected free and removed under strict asepsis.

The existing literature concerning pulse wave velocity deals with the effect of disease, age, sex and physiologic and pharmacologic stimuli on pulse wave velocity in groups of individuals. Spontaneous fluctuations in the same individual from time to time have received slight consideration. 1 , 2 , 3 , 4

We have measured repeatedly during a period of 6 months the pulse wave velocity in 6 normal individuals of 22 to 45 years of age. A simplified modification of a technic previously described 5 was used, which employs carbon grain microphones and 2 Einthoven string galvanometers. Carotid and either radial or index fingertip sphygmograms, together with respiratory movements, were recorded simultaneously and measurements made from reference points on the ascending limbs one-fifth of the vertical distance from beginning of upstroke to the peak. Brachial blood pressure was taken immediately following the taking of the record. With a camera speed of 10 cm. per second measurements were made with accuracy to 0.002 second. Subjects were examined at approximately the same time of day, in the sitting position (unless indicated otherwise) with the arms resting horizontally on supports. Preliminary rest periods varied from 5 to 45 minutes. Room temperatures were fairly constant. Noises and other demands upon the subjects'attention were kept at a minimum.

Careful scrutiny of Table I shows that for the group as a whole there is no apparent correlation between pulse wave velocity and systolic and diastolic pressure. When pulse wave velocity for each subject was plotted against systolic and diastolic blood pressures the following relationships were noted:

Frouin 1 found that pancreatectomy produced less glycosuria than usual, in dogs from which a considerable amount of adrenal tissue had been removed. Later work 2 has indicated an interrelationship between the adrenal and the pancreas. The adrenal cortex is necessary for the development of severe diabetes. 3 The medulla, however, may be significant in preventing insulin shock. 4

If the cortex is necessary for the full development of diabetes, is its cortin-producing or some other function responsible?

We have studied 9 cats in an attempt to throw light upon this problem. In 3 animals the right adrenal and part of the pancreas, then several days later the left adrenal and the rest of the pancreas, were removed. Two died in less than 24 hours, with blood sugars of 49 and 61 mg. per 100 cc. respectively. The third survived 6 days. Injected with cortin it showed a small increase in blood sugar from 98 to 156 mg., and slight glycosuria.

The remaining adrenal of 2 pancreatectomized, unilaterally adren-alectomized cats was denervated after establishment of diabetes. Eight days after denervation one animal showed reduction in diabetes; the other showed no reduction for 3 weeks and then only after cautery of the medulla and much of the adrenal cortex.

Four cats were kept alive by cortin injections after removal of the second adrenal, the pancreas having been removed in a 2-stage operation at the time of the adrenalectomies.

One cat showed some reduction in the glycosuria after the second adrenal was removed. Death occurred on the eighth day, perhaps from insulin, which was injected for the first time on that day.

The second cat refused to eat following the removal of the second adrenal. The high blood sugar (279 mg. per 100 cc.) and the glycosuria (maximum 290 mg. of glucose per hour) did not fall during the time of observation (4 days). He was killed in another experiment.

A definite correlation between phospholipid content and function in muscle was shown by the results of a comparison of the lipid content of various muscles of wild vs. laboratory rabbit, and hen vs. pigeon. In every case the trained muscles, such as the muscles used in running and jumping by the wild rabbit, had a much higher content of phospholipid than the same muscles in the untrained laboratory animal. The muscles used in flying by the pigeon had a much higher phospholipid content than the same muscles in the domestic hen. The cholesterol content of the muscles in the 2 groups of animals was not definitely different so that the ratio, phospholipid/cholesterol, was always higher in the trained muscle than in the untrained muscle.

Tables I and II will illustrate the differences found.

A series of experiments was made to study the chemical changes with reference to lactic acid and glycogen in the myocardium following ligation of the anterior descending branch of the left coronary artery of dogs and to correlate these with the anatomic and electrocardiographic changes observed.

Sixteen dogs were used. Essentially similar operative techniques were employed as in the experiments previously reported. 1 , 2 The tissues for chemical studies were excised from the heart in situ in the living animal under amytal anesthesia and immediately plunged into liquid air. Lactic acid was determined by the method of Friedemann and Graeser 3 and the glycogen by the modified Pflüger procedure as outlined by Yannet and Darrow. 4

In the first group of experiments 10 dogs were used. After ligation of the coronary vessels the animals were sacrificed at intervals varying from 1/2 hour to 24 hours. The lactic acid content of the muscle tissue from the zone supplied by the ligated vessels was increased 100-200% over that of the control zone (right ventricle) while the glycogen content of similar zones was decreased below that of the control zone. These changes were noted in every instance irrespective of the duration of the period of ligation. Control animals subjected to the same operative procedure except that the sutures were passed about the vessels and then removed, showed no significant variation in the content of lactic acid or glycogen from the two sides of the heart.

In the second group of experiments the coronary vessels were ligated for varying periods up to 8 hours. At the end of these intervals the ligatures were removed and the animals were sacrificed 2 hours afterward. Six dogs were used in this series. In all instances the lactic acid content of the heart muscle from the zone supplied by the previously ligated vessels was decreased below that of the muscle from the control zone. The glycogen content of the affected zones, as in the first group, was also decreased below that of the control zone.

No evidence has been obtained that the adrenal gland, as an organ in the dog, is concerned with either the synthesis or the metabolism of vitamin C. If the adrenals do play any role in the metabolism of this compound the action is conferred by the adrenal cortical hormone.

Harris and collaborators 1 , 2 have shown that dogs do not require vitamin C in their diets for adequate nutrition. Others have kept dogs upon synthetic rations, presumably devoid of this vitamin, for periods of 1-4 years without symptoms of scurvy. Tissues from such animals, notably the adrenals and liver, show by chemical and biological tests the presence of large amounts of ascorbic acid. Similar tissues from guinea pigs kept upon scorbutic diets reveal a marked diminution in the content of ascorbic acid as scurvy develops.

The early observation 3 that the cortex of the adrenal gland contained large amounts of “hexuronic acid”, later shown to be vitamin C 4 , 5 , 6 has led to speculation upon the relationship of this vitamin to the functions of the adrenal cortex. The belief that the medulla 7 did not contain “hexuronic acid” supported the idea of a functional significance for this uneven distribution. It has since been demonstrated that other animal tissues contain appreciable amounts of ascorbic acid, e. g., liver, 8 corpora lutea, 9 , 10 pituitary 11 thymus. 12 Huszák 13 has shown that the adrenal medulla contains as much vitamin C as does the cortex. Such a wide distribution is hardly indicative of a specific relationship between vitamin C and the functions of the adrenal cortex. In those species that do not develop scurvy one may reasonably consider any tissue containing much vitamin C as having a possible rôle in protecting the animal from this nutritional disorder.

The question arises, is the adrenal gland in the dog concerned specifically with the synthesis and metabolic functioning of vitamin C and the consequent protection of the animal from scurvy?

For reasons which can better be stated in a more complete report, it was thought highly probable that the pituitary hormone was not elaborated in the animal body from simple chemical substances, but on the contrary, was built up from complex precursors, or, taken in from the environment in a form not far removed from the hormone itself. If such hypothesis were correct, one should be able to isolate from foods these complex substances, or precursors, which when given to an animal in sufficient amounts, would prove to be gonad-stimulating. Of the foodstuffs considered, green plants seemed to be the most likely source of such material. Consequently, alfalfa was investigated from this standpoint and found to yield an extract capable of affecting the rabbit ovary in a fashion similar to that of the material found in pregnancy urine; i. e., the production of corpora lutea and corpora hemorrhagica in unmated rabbits.

The extract is prepared by mixing 1 kilo of alfalfa meal with 10 liters of water and sufficient 3N alkali to bring the reaction to pH 9-10. After 4 hours the mixture is filtered by suction through a mat of moistened alfalfa meal, and the filtrate then subjected to the Katzman-Doisy benzoic acid method for the extraction of the active material from pregnancy urine. 1 The extracts prepared so far are highly toxic so that they can not be injected intravenously, and therefore cannot be satisfactorily assayed by the rabbit method. The material must be injected either intraperitoneally or subcutaneously, and in order to obtain a positive response it has been necessary to inject the equivalent of 300 to 500 gm. of dry alfalfa in 2 doses not more than 12 hours apart. From previous experience with injections of pregnancy urine by the intraperitoneal route, it is estimated that the yield from 1 kilo of dry material is probably more than 30 rat units and probably less than 100 rat units. The negative results obtained with the injections of inactive extracts, prepared in an identical manner, except for 24 hours (or more) initial exposure to the dilute alkali instead of 4 hours, together with the intraperitoneal injections of a variety of materials, including male urine and sterol suspensions, serve adequately as controls.

The possible causal relationship of the intestinal bacterial flora in agranulocytic angina has not been investigated. We carried out bacteriological examinations of stool specimens of a human case and recovered together with other of the intestinal microorganisms, cultures of B. enteritidis, Streptococcus hemolyticus and B. welchii. This paper deals with the results of animal inoculation with these cultures.

B. enteritidis. Six guinea pigs were injected, 2 subcutaneously, 2 intraperitoneally, and 2 intracardially, with 1 cc. each, of a saline suspension of this strain. Neutrophilic leucocytosis resulted and the animals died in 24 to 72 hours.

Three guinea pigs were fed with oats contaminated with stools containing B. enteritidis, and over the food of 3 guinea pigs suspensions of recovered culture were poured. After 6 weeks, including repeated leucocytic counts, no appreciable effects of such feedings were noted.

Two guinea pigs were injected, 1 intracardially and 1 intraperitoneally with the filtrate of broth cultures and increase of total and granulocytic leucocytes resulted.

In previous work upon production of lesions of experimental typhoid fever in the guinea pig with in vivo prepared toxic products of B. typhosus, 1 , 2 it had been noted that a marked leukopenia resulted. No observations upon the granulocytes were made. We employed this method in the present work. A fatal peritonitis is primarily produced and the resultant exudative materials including toxins and saline washings of the peritoneal cavity were collected and filtered through a Berkefeld or Seitz-Wertz filter. A modified toxic moiety is procurable. Fifteen guinea pigs were injected by this method, employing the B. enteritidis isolation. The toxic filtrate was administered in doses of 1 to 5 cc. subcutaneously, intraperitoneally and intracardially. The intraperitoneal and intracardial routes and the larger doses were more effective.

The strain of Staphylococcus aureus employed was obtained from the blood stream of a fatal chronic human case of so-called agranulocytic angina 4 days before death. This patient presented boils and 2 large abscesses.

Injections of suspensions of this culture of Staphylococcus aureus subcutaneously into 2 guinea pigs and intraperitoneally into 2 other such animals produced neutrophilic leucocytosis.

Rabbits 2, 3, and 5 were inoculated in the following manner: Rabbit 2 was injected with 2 cc. of the living suspension in the peritonsillar structures. A marked inflammatory reaction occurred and resulted eventually in necrosis. Frequent leucocytic counts revealed only a neutrophilic leucocytosis. In Rabbit 3, the upper tracheal mucosa was scarified with a hypodermic needle and 1 cc. was injected in and about the eroded area. Neutrophilic leucocytosis ensued.

Rabbit 5 was injected subcutaneously with 5 cc. of melted nutrient agar. This was allowed to solidify and 0.5 cc, of the culture of Staphylococcus aureus was introduced into the agar mass. No reduction of the total or granulocytic cells resulted. The Seitz-Wertz filtrate of broth cultures of this microorganism produced only neutrophilic leucocytosis when injected into guinea pigs.

An in vivo prepared toxic filtrate was obtained by producing peritonitis in guinea pigs with this culture. A Seitz-Wertz filtrate of the exudative material and peritoneal washings of these supply animals was inoculated into 6 guinea pigs, 3 intracardially and 3 intraperitoneally. Considerable reductions of the total counts and marked granulocytopenia was produced in these animals (see chart). It is of interest to note that one of these animals developed an erythroblastic anemia with 40,000 nucleated red cells per cu. mm., indicating a disturbance of this portion of the hemopoietic system as well as of the myeloblastic centers.

A method is suggested for all fiocculation reactions, especially for the Hecht-Müller Ballung Reaction of elective staining of the extract colloid which brings about a distinctive color effect that facilitates the reading of results without impairing their specificity. Readings are best made against a white background (white paper). Positively reacting sera show a bright red ball, and after it is formed the chosen counterstain, blue, violet, or green appears in the surrounding fluid. Negative sera show a turbid light violet to greenish-blue color, permitting an immediate differentiation.

All flocculation reactions are based upon colloidal chemical processes. What is more natural than to introduce into these reactions dye colloids for the purpose of making the flocculation more distinct, firstly by coloring the flocculi and secondly by making them more voluminous by the joining of the dye colloids in the flocculation of the extract.

Experiments of this kind have been made frequently, but have been only partially successful.

Experiments were reported¹ in which the extract colloid could be colored, whereas a contrast coloration was hardly possible of attainment. These experiments were carried out on the Hecht Flocculation Reaction (H. F. R.), on the Sachs-Georgi-Reaction (S. G. R.), and the typhoid reaction, using the Flicker Diagnostikum.

The experiments since then, carried out with well over 100 of the most diversified dyes, led to a provisional result as follows:

The goal of these studies was the development of a flocculation reaction in which a part of the coloring matter would join with the flocculated extract so that in the case of a positive reaction the fluid remaining after the flocculation of the extract would be colored differently than the fluid in a negative reaction (with the extract not flocculated).

Experiments of this sort were done particularly with the Hecht-Muller-Ballung reaction but show in their results that they may be employed in all flocculation reactions for syphilis which are done with alcoholic extracts without addition of Balsam (Sachs-Georgi-Reaction, Kahn Reaction, Kline Reaction, Hinton Reaction, etc.).

The principle of the staining consists in the following: 1. The extract is electively stained by a dye which does not leave any color in the fluid remaining after the precipitation of the extract colloid. 2. As counterstain of the salt solution (the extract-free fluid), dyes must be used which leave the extract colloid unstained or which color it but slightly. Among numerous lipoid stains, the only one of value was Sudan 3 of the I. G. Farbenindustrie.

The cardiac output of patients with the Pick syndrome was determined by means of the Grollman 1 indirect acetylene method. This procedure is applicable to such patients in spite of their reduced vital capacity if the volume of the gas mixture to be re-breathed is decreased and if the rebreathing period is slightly increased.

The cardiac output per minute per square meter of body surface Grollman 1 calls the cardiac index. He has found that under basal conditions the normal person has a cardiac index of 2.2 ± 0.3 liters per minute. Any deviation of the index from the normal 2.2 greater than 15%, when determined by the acetylene method, is considered pathological.

Mrs. S., age 54 years, with Pick's disease, who should have had an output of 3.50 liters per minute (calculated by multiplying the patient's surface area by 2.2), had an average output of 2.26 liters per minute, 36% below normal. The estimated normal output per beat is 47 cc; she had an average beat output of 20 cc, her heart rate being 114 per minute.

Mrs. M., age 32 years, was operated on for the Pick syndrome. Her output normally should have been 3.04 liters per minute. Before operation she had an average minute output of 1.79 liters, 41% below normal. The tests after operation were carried out over a 2 weeks'period starting a month and a half after operation. Although there was not a striking increase in the average minute output of 1.84 liters (39% below normal), 5 of the 12 tests after operation reached a value never attained before operation. An average of the best 6 tests after operation gives an output of 2.05 liters per minute, only 33% below normal. This tendency toward improvement indicates that her heart has a reserve power and leads to the belief that as time goes on this patient's output will continue to improve. Before operation her beat output averaged 24 cc, her average heart rate being 75 per minute. After operation her beat output averaged 26 cc, the average heart rate being 71 per minute.

Many authors have demonstrated that vitamin C exists in various organs and fluids of the body. Moreover, it has been shown that in the guinea pig the concentration of vitamin C in the body is dependent upon the intake,^1-4 and that in the tissues of the guinea pig the vitamin C concentration is lower than in species which are not dependent upon a dietary source of this vitamin. 3 In view of such evidence, it was thought worth while to investigate the urinary excretion of vitamin C. The experiments were planned to throw light upon the following questions: 1. Is vitamin C excreted by the urinary route? 2. Is the level of urinary excretion proportional to the intake or is it a measure of excess vitamin ingestion? 3. Is there a difference in urinary excretion between species which do and those which do not require a dietary source of the vitamin?

As far as we are aware, very little work on this subject has been reported, von Euler and Malm 5 state that they were unable to detect vitamin C in normal, febrile or diabetic urine. On the other hand, van Eekelen and his associates 6 found that urinary excretion ran parallel to intake, but since these authors do not give any figures, no comparison between species can be made on the basis of their work.

The method employed was that of Birch, Harris and Ray, 7 based on the titration of ascorbic acid against 2:6 dichlorophenol-indophenol. Since urine is known to contain only traces of the compounds (glutathione and other sulphydryl compounds) which form the chief source of error in the application of this method, it seemed safe to consider any reducing substance found in urine to be ascorbic acid.

1. Epiphyseal tissue will regenerate new bone in immature dogs. 2. Cancellous bone will bridge bony defects without the aid of periosteum. 3. Periosteum will hasten callus formation of cancellous bone grafts. 4. Free bone marrow and endosteum will bridge bone defects without the aid of periosteum with a rapid forming callus. 5. The clinical use of bone marrow and endosteum is suggested in bone grafting.

The supremacy accorded the periosteum in osteogenesis has led to an almost complete neglect of the rôle played by other bone elements. Work on epiphyseal tissue has indeed been very scant, but there is both clinical and experimental evidence that cancellous bone plays an important rôle in bone repair, while the osteogenic properties of bone marrow and endosteum though often mentioned have been little studied.

Epiphyseal Tissue Transplants. A set of experiments were carried out on 6 immature dogs. A quarter inch extraperiosteal resection of each fibula was filled on one side with ground epiphyseal tissue while on the control side the defect was filled with ground whole bone. The defects were bridged across in each case. The epiphyseal tissue invariably produced a large thick spongy callus. Serial slides of X-rays were presented to illustrate this callus at different stages of development, Fig. 1.

Cancellous Bone Transplants. Experiments were done on 5 adult dogs. Cancellous bone was removed from the condyle of the femur and transplanted into extraperiosteal and subperiosteal defects of the 2 fibulae respectively on each dog. These defects bridged across in each case even where there was no existing periosteum. Slides of serial X-rays were presented to illustrate how rapidly cancellous bone grafts regenerate new bone. (Fig. 2.) On the side of the subperiosteal resections the favorable presence of intact periosteum is well illustrated in a more rapid regeneration.

Bone Marrow and Endosteal Transplants. The next group of experiments was carried out on 6 adult dogs. Each fibula was resected extra-periosteally. On each side the defect was filled with endosteum and bone marrow removed from the tibia with a small curette. No regeneration occurred on the control side where no grafts were placed, while on the side containing endosteum and bone marrow rapid and complete reformation of dense bone occurred. Slides of serial X-rays were shown to illustrate this rapid osteogenesis. (Fig. 3a.) A photomicrograph was presented to show the new bone callus. (Fig. 3b.)

While the results are not complete enough to reach definite conclusions, it would seem that in view of similar and better known conditions in rodents the mechanism of the production of the “wedding dress” in Chrosomus erythrogaster is due to the action of substances having either a stimulating effect upon the sex glands of the fish, such as prolan, or having an effect similar to that of the estrus producing hormones, such as Yohimbine.

Recent reports 1 , 2 , 3 have made it probable that certain species of fish may be used successfully as test animals for the demonstration of either the gonadotropic or the sex hormones. For some time it has been known that certain male fish, such as Rhodeus amarus, Gastrosteus, Phoxinus laevis, belonging to the Cyprinidae group, assume striking color changes during the spawning period, lasting for several weeks in the spring. This so-called “wedding dress” or “mating coat” is due to the expansion of the chromatophoric cells of the skin and is considered to be an expression of heat in order to attract the females. Thus it would be comparable to the well-known estrus of the rodents.

Wunder, 1 working with Rhodeus amarus, obtained a “wedding dress” with subcutaneous injection of testis extract, variable results with prolan, and no effect with the female sex hormone. Glaser and Haempel 2 reported the castrated Bitterling as a suitable test animal for the demonstration of the male sex hormone. Zondek 3 used the expansion of the erythrophores of Phoxinus laevis as a test for the demonstration of his pituitary pigment hormone intermedin.

While these experiments were conducted on fish, domestic and very abundant in central Europe, they are not inhabitants of this country and are difficult to obtain. For this reason and for the purpose of producing the “wedding dress” artificially in an American fish, Chrosomus erythrogaster was considered the most suitable fish. This attractive and hardy fish, commonly called “red bellied dace”, belongs to the group of Cyprinidae and is found in various parts of this country from Maine to Alabama in small, clear streams, where it breeds in May and June. No record exists of its having bred in captivity. 4 The breeding dress consists in a vivid and brilliant red discoloration of the otherwise white belly and the part just below the mandible on each side. 5

The vitamin A content of the liver and kidneys of dogs fed fresh carrots is greater than that found in dogs on a meat and boiled rice diet. The level of the vitamin A concentration in the liver and kidneys is approximately 20 vitamin A animal units. Continued feeding of carrots or addition of cod liver oil apparently does not increase the vitamin A content above this level. The spectroscopic examination indicates that the carotene from the carrots was transformed to vitamin A.

The function of carotene in nutrition has been established.^1-4 The advisability of feeding large amounts of raw carrots over a prolonged period of time is still questioned. Carotinemia has been observed in the dog, 5 and in man. 6 Investigations of Moore, 2 Capper, 7 and Drummond 8 have shown that conversion of carotene to vitamin A takes place in the livers of vitamin A-free rats or fowls. Olcott and McCann 9 later confirmed these observations and further concluded that this change may take place in vitro if incubated with fresh liver tissue. Rea and Drummond 10 failed to obtain a carotinase in liver preparations from vitamin A-free rats and further obtained no increase in vitamin A content of cat livers after carotene administration.

Table I presents the results obtained by feeding dogs fresh carrots over a prolonged period of time (2 to 8 months). Five of the animals received a diet of meat, boiled rice and fresh carrots; 3 received a similar diet with the addition of 10 cc. of cod liver oil daily, and 2 were deprived of both carrots and cod liver oil. The 2 dogs receiving only meat and boiled rice succumbed at 15 and 49 days. Previous to this their appetite became sporadic, there was marked discharge from the nose and eyes and in all the symptoms were those of distemper. The remaining dogs continued to maintain or gain in weight and appear healthy and vigorous. They were killed and autopsied at varying times. No signs of carotinemia were observed in any of the dogs. The liver and kidneys were removed at necropsy and prepared for vitamin A estimation by the method of Moore. 2

The colorimetric estimation was carried out by the method of Norris and Danielson. 11 The dilutions of the extracted tissue oils were made so that the readings were below 2 blue Lovibond units.

The substitution of strontium for calcium in the rickets-producing diet produces a form of rickets which fails to respond to the very large amounts of vitamin D. The endochondral cartilage of strontium-fed animals both treated and untreated fails to calcify when incubated in artificial serum solution even with a Ca X P product of 60 which produces marked calcification in the endochondral cartilage of control rachitic rats. These observations strongly suggest an injurious effect of strontium upon the “local factor” in the hypertrophic cartilage of the provisional zone of calcification which brings about calcification at this area when conditions in the serum are favorable. The ability of vitamin D to cause an increase of the Ca X P even when healing does not take place, implies that this raising of the product is the function of the antirachitic vitamin in the cure of rickets.

This communication reports a type of incurable rickets caused by the substitution of 3% SrCO₃ for 3% CaCO₃ in a rickets-producing diet.

Rats were divided into 2 groups. To the control group the rickets-producing diet containing 3% CaCO₃ was fed while the other group received 3% SrCO₃ instead of the calcium salt. The initial weight of the animals varied between 43-60 gm. At the end of 21 days both groups of animals showed severe rickets. The control group gained 25-45 gm. each while the strontium group gained 0-5 gm. An analysis of the serum of the control rachitic animals showed a Ca X P product of 15-24, while the strontium fed animals gave a product of 30 made up of Ca = 10.6 mg. % and Inorganic P = 2.8 mg. %.∗ The average ash content of the fat-free femurs was 33.23%. The flame test was negative for strontium in these bones.

To calcify the endochondral cartilage of the tibiae of each group, artificial serum solutions having a Ca X P product which varied from 30-60 were incubated. The technique of Shipley, Kramer and Howland 1 was used. The control group showed satisfactory calcification at products of 40, 50 and 60 while the tibiae from the strontium animals showed no fresh deposition.

The remaining animals of both groups were given 300 Steenbock units of vitamin D in the form of viosterol during a period of 9 days. All of the control rachitic animals healed and the serum Ca X P product rose to between 60 and 80. The untreated group still showed a Ca X P product between 15 and 25. The strontium animals failed to respond, the rickets being even more severe at the end of the 9 days of treatment than before therapy was begun. Nevertheless, the Ca X P product of the serum rose to 62. This value was due to a Ca of 12.1 mg. % and a serum inorganic P value of 5.1 mg. %.∗ The average ash content of the fat-free femurs was 29.4%, therefore decalcification of the bones progressed. The flame test was negative for strontium.

In previous publications from this laboratory^1-5 it has been shown that fresh plant and fruit juices contain an antigoitrous agent when tested on immature rabbits maintained on a goitrogenic diet of alfalfa hay and oats. (The goitrogenic effect of this diet is greatly enhanced by cyanide.) It was shown that this antigoitrous effect was in general inversely proportional to the iodine reducing capacity and that treating the fresh juice with copper salts destroyed or lessened this antigoitrous effect. Our data further indicated that the antigoitrous agent was probably identical with ascorbic acid but because of the lack of crystalline ascorbic acid on the one hand and the presence of iodine (also an antigoitrous agent), on the other, in all these extracts we were unable to prove it.

About 2 years ago one of us 6 simplified Szent-Györgyi's and Waugh and King's method of extracting ascorbic acid and by utilizing iris leaves, which are extremely rich in ascorbic acid, has prepared sufficient quantities of this hormone in crystalline form for these experiments, which were carried out as follows: Young guinea pigs weighing between 150 and 250 gm. were placed on a diet of alfalfa hay and rolled oats. (This diet produces scurvy in from 15 to 20 days. Pigs maintained on this diet when injected with the thyrotropic factor develop more marked exophthalmos and thyroid hyperplasia and more quickly than when given greens daily in addition.) They were injected with 1/2 to 2 cc. of ox anterior pituitary extract prepared after Loeb and Bassett's method and given 25 to 100 mg. of crystalline ascorbic acid by mouth daily. The pigs were sacrificed at various intervals and the principal data of 22 are given in Table I.

In connection with an extended series of experiments on the effects of electric shock, the fact has been brought to light that shocks whose duration is a small fraction of the cardiac cycle have a marked difference in their tendency to cause ventricular fibrillation,. depending upon the phase of the cycle in which they occur. The shocks were tripped off by amplified cardiac currents after a controlled delay from a reference point in the electrocardiac cycle. The electrocardiogram together with the shock current and voltage were photographically recorded in time relation on a continuously moving film, which enabled checks to be made on the placement of the shocks in the cardiac cycle. Nearly 800 shocks with 60-cycle current of .03 and .1 second durations were administered to 92 anesthetized sheep in different parts of the cardiac cycle, with currents ranging from 1 to 17 amperes, through the pathway between the right fore-leg and the left hind-leg. A large proportion of the shocks which were given near the end of ventricular systole, as determined by their coincidence with the T-wave of the electrocardiogram, caused ventricular fibrillation and death, while but one shock (0.1 second duration) distinctly outside the T-wave resulted in fibrillation.

It seems probable that the difference in the susceptibility of the heart to fibrillation from shocks in different phases of its cycle is related to the known variation of its excitability. Further investigation is being carried forward.

The need to measure the amount of cocaine penetrating to the aqueous humor of an experimental animal after instillation of a cocaine solution into the conjunctival sac led to the following analytical procedure, which we find applicable to the quantitation of traces of many alkaloids in aqueous solution, and of certain alkaloids, particularly cocaine and strychnine, in body fluids. The method, which is accurate within the limits of colorimetric observations, is strikingly simple, and we believe will be of interest to workers in pharmacology, toxicology and experimental medicine.

A. Procedure for aqueous solutions. To a measured amount of solution containing from .5 to .05 mg. of the alkaloid is added 1 cc. of a 1% solution of phosphomolybdic acid in normal sulphuric acid. This is best done in a small test tube of about 5 cc. capacity. After mixing and standing for 15 minutes, the mixture in the test tube is centrifuged and the supernatant fluid decanted. The residue is resuspended in water, again centrifuged and the supernatant discarded. To the washed precipitate is now added 1 cc. of fifth normal sulphuric acid, 1 cc. of 2% aqueous solution of hydroquinone, and after shaking, 1 cc. of the carbonate-sulphite solution used in the Bell and Doisy 1 method for phosphate determination (15% sodium carbonate and 3% sodium sulphite). The color developed is read promptly in the small Duboscq type colorimeter against a standard prepared by treating 1 cc. of a solution of known and approximately equal alkaloid content (using the same alkaloid that is being determined) simultaneously with the same reagents. If necessary, it is a relatively simple matter to prepare several standards (5 of the small test tubes can be spun in an ordinary 50 cc. metal centrifuge cup) and the proper one can be selected by inspection.

The present paper deals with an analysis of the action of parathormone on the neuromuscular system which may be considered the main seat of the tetany syndrome (MacCallum and Vogel 1 ) The hind legs of pithed Ranae esculentae were perfused with phosphate buffered Ringer and the sciatic nerve stimulated with rhythmic condensor discharges as described previously. 2 After a state of relative fatigue had been reached in which the height of contraction remains equal over a long period of time, the perfusion fluid was exchanged for one containing parathormone, Lilly or Parke Davis, in concentrations from 1:1000 to 1:100. In all solutions the pH was carefully adjusted to that of Ringer. The phenol content proved to be without effect as control experiments indicated. In several hundred experiments parathormone invariably causes an increase in the height of contractions up to about 200% in the strongest concentrations. The effects are reversible when the solution is again exchanged for ordinary Ringer and the effects are due to the parathormone content of the extract for the following reasons:

1. Extracts of the parathyroid which according to the blood calcium test show the same number of units per cc. but contain solids of varying concentrations show an equal effect on the muscle even when the concentration of the solids varies 100%.

2. Boiling for one hour destroys neither the action of parathormone on the blood calcium (Collip and Thomson 3 ) nor our reaction.

3. Destruction of the hormone content of the extract by boiling with 10% HC1 for one hour abolished the described reaction on the muscle, as it does the blood calcium effect according to Collip.

In studies on the effect of the composition of the medium upon the cell-free alcohol fermentation by zymin, it was soon evident that the zymin is very sensitive to relatively low concentrations of ethanol. In Table I are given quantitative data on the effect of various concentrations of ethanol upon the relative rates of fermentation based upon the production of carbon dioxide. There are also given the data obtained by Rahn 1 for yeast. An analysis of the latter data shows a linear relation between the concentration of alcohol and the relative rate of CO₂ production, up to a concentration of about 10%. From this point there is a slow and practically constant rate even in the presence of as high as 14%. There is a linear relationship for the zymin up to about 1.8% alcohol from which point there is also evident the low but continuous fermentation. The equations for the linear portions of the curve are

Relative rate for yeast = 100— 2.1 X % ethanol.

Relative rate for zymin = 100—22.5 X % ethanol.

It is at once evident that the zymin is much more sensitive to the alcohol than is the yeast. Harden 2 reports that there is a 0-20% diminution of the fermentative activity of yeast juice in the presence of a 6% ethanol and a 75% decrease for 14% ethanol. In the preparation of zymin there is evidently removed some factor which protects against alcohol. The addition of the extracted material did not decrease the sensitivity. Experiments, the details of which will be published later, show that while the addition of phosphate gives the expected increase in activity there is no decrease in sensitivity toward ethanol; the addition of the appropriate concentration of ammonium chloride not only increases the fermentative activity, with or without the addition of phosphate, but very materially decreases the sensitivity toward ethanol.

In a series of 512 enumerations of leucocytes in saliva and peripheral blood in 20 normal and 17 abnormal individuals, no correlation has been observed between the total number of white blood cells in the peripheral blood and in the saliva.

During a study of the physiologic variations of the white blood cells of normal individuals 1 in which enumerations of the circulating leucocytes were made at 15 minute intervals over a period of several hours, simultaneously the salivary leucocytes were counted in an attempt to correlate and interpret changes which might be observed. In all, 367 such observations were made of 20 normal individuals. The observations were then extended to 145 determinations of 17 patients with a variety of pathological conditions. The salivary leucocytes were counted according to the method described by Isaacs and Danielian. 2 In making white blood cell counts of blood drawn from the ear, the method previously described 1 was used, in which, by counting 4 times the usual number of cells, the probable error was found to be about ±230 cells.

Our observations of the number of salivary leucocytes in normal individuals are in essential agreement with those of Isaacs and Danielian. In more than 50% of the subjects, the number of salivary leucocytes was below 25 cells per cu. mm.; counts of from 500 to 1000 or more cells per cu. mm. were occasionally observed.

However, in our data, there is no apparent correlation between the number of the white cells in the blood and those in the saliva. The salivary leucocyte counts were fairly uniformly scattered over a series of individuals whose range of white blood cell counts was from 4,000 to 12,000 per cu. mm. Marked fluctuations were found in the salivary count without any significant change in the leucocytes of the peripheral blood and vice versa. When fluctuations occurred in both counts during the same interval there was no consistent relationship of one to the other. Neither in individuals nor in the entire group was there evidence of a consistent reciprocal relationship between the numbers of leucocytes in the saliva and peripheral blood.

Macht and Dunning described the chemical and physiological properties of mono-brom-hydroxy-benzyl alcohol, or mono-brom-saligenin. 1 This compound, which has been named Bromsalizol, was the mono-brom derivative of ortho-hydroxy-benzyl alcohol. Inasmuch as 3 isomers of saligenin with the hydroxyl OH and benzyl CH₂OH groups in the ortho, meta and para positions, respectively, are possible, they and their mono-brom derivatives were prepared in these laboratories and investigated with regard to their comparative pharmacology. The structural formulae of the 3 saligenin isomers and their corresponding mono-brom derivatives are as follows:

The melting point of ortho-saligenin was 85°-87°; of meta-sal-igenin, 70°-71°; of para-saligenin, 116°; of mono-brom-ortho-saligenin, 108°-109°; of mono-brom-meta-saligenin, 124°; and of mono-brom-para-saligenin, 128°.∗

A pharmacodynamic comparison of the compounds mentioned above was made by studying (1) their effect and that of controls on the growth of Lupinus seedlings in plant physiological solutions; (2) their action on the muscular coordination and respiration of goldfish, Carassius auratus; (3) their influence on the behavior of mice, injected intraperitoneally; and (4) of rats, also injected intra peritoneally; (5) their local anesthetic effect on the cornea of the rabbit's eye and on frog-skin; (6) their action on surviving preparations of smooth muscle from the intestine of the cat and uterus of the guinea pig; (7) on the kidney function of rabbits after administration by mouth; (8) on the circulation and respiration of etherized cats after intravenous injection; (9) their antipyretic effect on rabbits and rats; and ascertaining (10) the killing intravenous dose for cats. Most of the results obtained are summarized in Table I.

After the ingestion of an oil of high vitamin A content, such as haliver oil, a marked increase in the intensity of the Price-Carr reaction for vitamin A is observed in the blood serum of normal persons. The degree of this increase may indicate the ability of the individual to absorb vitamin A. The amount of vitamin A in the serum is stated in colorimetric units which are purely arbitrary. Using a 10% solution of copper sulphate as a standard, serum of sufficient content in vitamin A to match the standard when acted upon by the antimony trichloride reagent is considered to contain 100 units of vitamin A per 100 cc. Each colorimetric unit is equivalent to 2.9 biological units.

To 12 afebrile children who were considered to have no disorder in the absorption of fats, 2 cc. of haliver oil∗ were administered while the children were fasting. Throughout the remainder of the 24-hour test period, the children ate meals with low vitamin A content at the usual times. Blood was taken for analysis before the ingestion of the oil, and 2, 4, 6, 9, 12, and 24 hours afterwards. The maximum concentration of vitamin A is observed at 4 hours and is approximately 9 times the fasting level. (Table I.) The level at 24 hours is about 50% above the original level.

Two children suffering from idiopathic steatorrhea (celiac disease) were observed during the course of their hospitalization and treatment. It was found that the rise of vitamin A in the serum is much less than in the normal children and that the maximum rise occurs later after the ingestion of the oil; also, that during the phase of clinical improvement, the rise is more marked. We may assume that the low rise is due to faulty absorption although a more rapid withdrawal from the blood stream or a reexcretion into the intestine may play a part.