Experimental Biology and Medicine

A sample of casein, purified by three reprecipitations and extraction with alcohol and ether in the cold, yields on extraction with boiling alcohol and cooling a fat-like substance, which attracted our attention. The crystals obtained proved to be free from nitrogen, melted at 52° and gave no cholesterol reactions. On treatment with boiling alcoholic potash the substance was incompletely saponifiable. Various samples of well purified casein gave 0.5-1 per cent. of this material, whereas A-free casein Harris failed us so far in this respect.

The question whether the presence of this fat-like product fed along with the casein, plays a rôle in nutrition experiments, is being taken up in a series of experiments. It follows from the above data, that casein cannot be completely freed from lipins by extraction in cold with alcohol and ether and that the method of purification adopted by the present writer with Macallum (repeated boiling out with alcohol) is probably the most suitable one to use in preparing casein for feeding experiments.

In 1914 one of the present authors reported in collaboration with v. Schoenborn 1 that pigeons kept on a vitamine-free, artificially compounded diet, show complete disappearance of the glycogen in the liver and increased blood sugar. In one particular series of experiments the blood sugar amounted to 0.29 per cent. and no glycogen was found in the liver. In a few pigeons kept on the same diet one dose of vitamine B from yeast was injected intramuscularly with the result that liver was found to contain 0.6 per cent. of glycogen and the blood 0.19 per cent. sugar. This vitamine preparation, which proved to be very potent, was prepared in the following way: an evaporated alcoholic extract of yeast was precipitated with phosphotungstic acid, and the resulting dried precipitate treated with acetone. The insoluble fraction was decomposed with lead acetate and the filtrate freed from adenine by means of picric acid.

This observation gains much interest in view of the recent communications of Winter and Smith and also of Collip on the presence of an insulin-like substance in yeast and other starting materials.

For some time past we have been working on the same problem again. We have found that crude extracts of yeast and rice-polishings possess a blood-sugar increasing rather than decreasing action. When we took, however, yeast grown in the laboratory on a medium rich in vitamine D, then centrifuged and washed the cells, and after heating to 100° injected them subcutaneously into rabbits, we obtained in a number of instances in 3-4 hours blood-sugar decreases which amounted to 30-40 per cent. of the initial value. We agree with Collip that compared with insulin the action of the yeast substance is slow, but on the other hand it lasts longer which might prove of therapeutic advantage. We have not as yet succeeded in obtaining similar results with Fleischmann's yeast treated in the same way.

The comparative values of the following solvents in extracting vitamine from yeast were studied: ethyl alcohol (50, 60, 70, and 80 per cent.), methyl alcohol (60 to 70 per cent.), propyl alcohol (70 per cent.), butyl alcohol (70 per cent.), isobutyl alcohol (70 per cent.), acetone (70 per cent.), methyl ethyl ketone (70 per cent.) and acetic acid (70 per cent.). The extracts and residues were tested on pigeons and rats, and the extracts were also tested for their content of vitamine D (yeast growth), co-ferment, total nitrogen and total solids.

If inactivity of residue be taken as a criterion, then 70 per cent. alcohol has proved to be the best solvent for vitamine among the solvents used. If, however, we stress not only the activity of the extract, but also the quantiy of nitrogenous and other impurities accompanying the extract (selecting the one that gives a minimum of such impurities), then the best among the solvents is acetone.

Vitamines B and D tend to run parallel with one another, so that, as a rule, the higher the content of vitamine B, the higher in vitamine D is the extract apt to be.

The co-ferment shows no definite relation to either vitamine B or D.

On the whole, the higher the nitrogen content of the extract, the greater the percentage of total solids, and the greater the activity of the extract.

Using these solvents on rice polishings instead of on yeast, we find that 60 per cent. alcohol is better than 70 per cent.

The extracts from rice polishings seem to be particularly active when tested on rats, and, by comparison, far less so when tested on pigeons.

1. A series of 30 constipated subjects were under observation before, during, and after treatment with B. acidophilus. A comparison of these periods shows that the number of normal defecations has been significantly increased during treatment. The usual daily dose was 1,000 c.c. containing 200,000,000 viable B. acidophilus per c.c.

2. The beneficial influence of B. acidophilus usually persists for a considerable period of time after treatment has been stopped. Patients have been observed for from one week to about one year after treatment and almost without exception all have had more normal defecations after, than before, treatment.

3. The use of lactose during and after ingestion of B. acidophilus does much to enhance the beneficial effects.

4. A transformation of the intestinal flora from a proteolytic to an aciduric type as shown by microscopic and plate counts may generally be induced. Such transformation is usually accompanied by almost daily defecations regardless of the severity of the constipation.

5. Two cases of diarrhea have been successfully treated by the ingestion of B. acidophilus.

1. In order to study the influence of physical and chemical factors, B. acidophilus milk was centrifuged and run through a Mandler diatomaceous filter. Thus the chemical constituents were little altered. When fed to constipated patients it was practically without effect. Regular B. acidophilus milk ingested subsequently resulted in an increase in the number of normal defecations.

2. B. acidophilus milk was sterilized and lactic acid added, thus again approximating the original chemical composition. When fed to constipated patients, little change was noted. Regular B. acidophilus milk ingested subsequently resulted in an increase in the number of normal defecations.

3. These data indicate that B. acidophilus therapy is essentially bacteriological rather than physical or chemical in nature.

In the present studies we have developed a technique for the measurement of the power of the bacterial cell to combine with hydrogen and hydroxyl ions. The bacteria show a general tendency to simulate amphoteric colloids and we have therefore measured their amphoteric property by the use of titration curves. A bacterial suspension in water or in a particular salt solution is titrated electrometrically with hydrochloric acid and with sodium hydroxide and similar titration curves are obtained for the menstruum alone. From these two sets of curves it is possible to calculate a series of “buffer ratios” by taking the quotients of the amounts of acid or alkali added to shift the P_H one unit in the bacterial suspension and in the menstruum alone. Such buffer ratios have been calculated for each P_H unit zone between P_H 2 and P_H 12. The titration cells which have been developed for these experiments are illustrated in the accompany figure. The buffer ratios for Bact. coli in distilled water, 0.725, M NaCl, 0.145 M CaCl₂ and in 0.580 M NaCl + 0.145 M CaCl₂ solutions are presented in Table I.

Our findings may be summarized in the following categorical manner:

1. In distilled water suspensions of Bacterium coli show a marked tendency to resist a change in hydrogen-ion concentration in the zones of P_H which are of physiological interest, i. e., P_H 4-10. In more acid and more alkaline solution this buffering power is weaker and the buffer ratios approximate unity at P_H 3-4 and at P_H 10-11.

In connection with an extensive series of studies on the effect of electrolytes upon the various properties of the bacterial cell, we have measured the electrical charge of vegetative cells of B. cereus (chosen on account of its large size) by the direct microscopic method described by Northrop. 3 In conducting these experiments a voltage of known magnitude (112 v.) is applied to non-polarizing zinc-zinc sulphate electrodes and the direction and velocity of migration of the bacteria in unbuffered suspensions determined by observing through the microscope the time taken by the bacterial cells to cross a definite space on the ocular micrometer. In order to avoid theoretical assumptions we have expressed all results in terms of this observed velocity, a high velocity in general presumably signifying of course a greater charge. The average results obtained with varying hydrogen ion concentrations generally based on four or more observations at each P_H value are indicated below.

It appears that these bacteria maintain a fairly high and reasonably consistent negative charge through the wide range of reaction between P_H 5 and P_H 10, rising somewhat with increasing alkalinity. On the acid side of this range the charge drops steadily to an isoelectric point at about P_H 2.0, the charge beyond this point being slight and variable and sometimes reversed. At reactions more alkaline than P_H 10 the charge diminishes even more rapidly to a second (alkaline) isoelectric point at about P_H 12.

Following the administration of large doses of insulin to several diabetics, to two patients with diabetic coma, to a normal fasting individual, to a fasting dog, and to fasting rabbits, and coincident with the drop in the blood sugar which regularly occurs, a marked drop has been noted in the concentration of inorganic blood serum phosphate and serum potassium. A sharp drop in the urinary output of phosphorus and of potassium accompanies the drop in the blood serum concentration, and is later followed (3-12 hours) by a well marked compensatory increase in the urinary excretion of these substances, so that the total excretion over daily periods is not much altered.

In sharp contrast to the above findings during the convulsions of insulin shock, a very marked increase in the serum phosphate and serum potassium has been found during strychnine convulsions in rabbits, in which condition, as is well known, an extreme destruction of muscle glycogen occurs.

In seeking an explanation of the above phenomena, attention is drawn to the recent work of Embden, Meyerhof, A. V. Hill, and others, which indicates that a hexose diphosphate is an intermediary between glycogen and lactic acid in the contractile process in the muscles. It is suggested that an analogous phosphate compound is formed during the process of storage of glycogen and that insulin causes or accelerates its synthesis. This would account for the disappearance of phosphate into the tissues during the period in which insulin is acting, and the subsequent increased excretion of phosphate may be due to the further conversion into glycogen of the hexose portion of the hexose diphosphate, thus leaving the excess of phosphate available for excretion. The massive, rapid breaking down of the phosphate compound during the tetanic convulsions of strychnine, would further account for the appearance of inorganic phosphate (as well as lactic acid) in the blood stream, as has actually here been shown to occur. The shift in concentration of the potassium indicates the formation of a monopotassium salt.

The electrocardiogram as ordinarily taken is a record of the differences of potential occurring between various parts of the body remote from the heart. While these differences of potential can be shown to be due to a primary electrical effect in the heart muscle itself, it is evident that they also depend on the nature of the body tissue as a conducting medium. Thus, if the body were composed of a non-conducting material, it is obvious that the primary cardiac electrical effect would not give rise to differences of potential in remote parts, while, on the other hand, if the body possessed the property of a good conductor, such as copper, it can be shown that the difference of potential between parts remote from the heart would be infinitesimal.

These considerations led us to believe that the presence of a large effusion surrounding the heart, especially a large pericardial effusion, might manifest itself by a lowering of the voltage of the electrocardiogram. This belief has been reinforced by our observation of seven cases of low voltage associated with large pericardial or pleuropericardial effusions. In some of these cases the markedly low voltage of the electrocardiogram led to a suspicion of effusion, which was later confirmed by X-ray examination, aspiration or autopsy finding.

The characteristic finding in the electrocardiogram is a decided lowering of voltage of the main deflection in all three leads. This is not necessarily associated with any constant alteration in the shape of the various waves, although it is quite conceivable that the effusion may also alter the electrical axis of the hear in some cases, or cause other changes in the propagation of the electrical disturbance.

The generally accepted statement that collagen is not hydrolyzed by trypsin unless previously swollen in acid or alkali, shrunk in hot water, or treated with pepsin, rests entirely on qualitative work by Kühne and Ewald 1 in 1887 and 1890.

Using finely sifted hide powder as a source of collagen, we found that it was readily digested by trypsin in concentrations of the protease exceeding 10 mg. per liter. The hide powder was treated in 10 c.c. conical graduated centrifuge tubes with buffer solutions at various hydrogen ion concentrations, centrifuged, and the volume of powder measured. The buffer was then replaced by a trypsin solution made up in a buffer of known P_H, and digestion carried on, with continuous shaking, at 40.00°. By suitable variation of the buffer we found that the tryptic hydrolysis was not affected by pretreatment of the collagen at different P_H's between 1.1 and 8.9 and that the optimum reaction for the hydrolysis was at P_H 5.9. The timehydrolysis curve for trypsin-collagen is of the same nature as with trypsin and proteins in general; since here the substrate is insoluble though hydrated, tryptic action appears to take place at the surface of the substrate particles, whether these be coarsely or colloidally dispersed. The degree of hydrolysis increases with increasing concentration of trypsin and decreasing size of hide powder particles. Complete hydrolysis was not reached in four periods of 20 minutes each, the experiment being then stopped owing to increasing hydrolysis in the control (hide powder and buffer solution without trypsin).

The shape of the hydrolysis-time curve suggests that the reaction is very slightly reversible. Experiments on pelt, using concentrations of trypsin up to 0.4 per cent., showed measurable hydrolysis of the collagen, although the organized skin structure inhibited diffusion of the enzyme and greatly decreased the speed of the reaction.

In 1917 Dochez and Avery 1 showed that there was contained in filtrates from pneumococcus cultures and in the body fluids of experimentally infected animals and of patients suffering from pneumonia, a soluble substance which reacts specifically in anti-pneumococcus serum of the homologous type. This substance, which was found to be thermostable, precipitable by alcohol or acetone, non-dialyzable, and not digested by trypsin, is now being subjected to a more intensive chemical study.

Eight-day, autolyzed cultures of Type II Pneumococcus in phosphate broth were concentrated to 1/15 volume and precipitated with 1.2 volumes of alcohol. The precipitate, centrifuged at high speed, yields a compact middle layer containing the specific soluble substance. By repeated fractionation with alcohol or acetone, first in neutral, then in dilute acetic acid solution, followed by repeated fractional precipitation with ammonium sulfate and final dialysis, about 1 gm. of a highly purified preparation was obtained for each 75 liters of culture used.

In its present state of purity the specific soluble substance is amorphous and yields a viscous solution in water. A 1 per cent. solution gives no biuret test, yields no precipitate with phosphotungstic acid, mercuric chloride, or neutral lead acetate, gives a faint haze with tannic acid, and is precipitated by basic lead acetate. At a dilution of 1:1,500,000 it still gives the Molisch reaction and yields a precipitate with Type II immune serum. [a]D is + 58.7°; N, 1.2 per cent.; P, trace; S, none; C, 46.2 per cent.; H, 6.1 per cent. Hydrolysis yielded 79 per cent. of reducing sugars, of which glucose was identified by the melting point and optical rotation of its phenylosazone. Earlier preparations containing more nitrogen and yielding less reducing sugars on hydrolysis were not specific at as high dilutions.

In the preceding communication, it has been pointed out that the so-called soluble specific substance of pneumococcus is nonprotein in nature, and in its present state of purification is either itself a polysaccharide, or intimately associated with the carbohydrate. Although antigenically this substance appears capable of stimulating little or no antibody response, serologically it is highly reactive and exhibits to an extraordinary degree the reactions of type specificity in antibacterial serum of the homologous type of pneumococcus. On the other hand it is possible to recover from the pneumococcus cell another substance which is protein in character and which is distinctive in its serological behavior from the soluble specific substance. From bile solutions of pneumococci dilute acetic acid precipitates a protein fraction. This precipitate is washed in water and redissolved in dilute alkali. After repeated precipitations and thorough washing the dissolved material is passed through a Berkefeld filter and reprecipitated. The final precipitate is washed rapidly with acetone and ether and dried in vacuo. The preparation so obtained is a whitish powder, readily soluble in faintly alkaline solution, possessing the properties of a mixture of nucleoprotein and mucoid. It contains about 16 per cent. of nitrogen and 0.5 per cent. phosphorus.

Solutions of nucleoprotein prepared from one type of pneumococcus (Type 11) react in about equal degree with all three types of antipneumococcus serum, and not with antityphoid or normal horse serum. This fact, if confirmed by subsequent investigation of the protein from pneumococci of other types, would indicate, on the basis of specific precipitin reactions, that all pneumococci possess in part at least a common specific protein. The protein of pneumococcus, as contrasted with the non-protein fraction or soluble specific substance is not type specific, but reacts with antipneumococcus serum regardless of type derivation. It is therefore species specific, not type specific.

The antacid most frequently used in the treatment of hyper-chlorhydria is sodium bicarbonate. But this is not only an antacid but an alkali, so that the contents of the stomach occasionally become alkaline. Moreover, the amount required to control the gastric symptoms is frequently sufficient to make the urine alkaline. Both these alkalinizations are regarded as unphysiological. Direct evidence of the occasional toxic action of therapeutic doses of sodium bicarbonate has recently become available. 1 2

The ideal gastric antacid would appear to be a substance which can neutralize hydrochloric acid but cannot make the stomach alkaline and which is excreted, unchanged, in the intestine and not in the urine. Such substances are found in the secondary and tertiary phosphates of calcium and magnesium. They are neutral substances, which dissolve in the hydrochloric acid of the stomach with the formation of the corresponding acid phosphates. An excess cannot quite alkalinize the stomach. The tertiary phosphates, as was shown by the work of Steel and Gies 1 and of Lothrop 2 for bone-ash, which is very nearly pure Ca₃(PO₄)₂, and by experiments of the author (see Table I) for Mg₃ (PO₄)₂, are precipitated in the intestine and eliminated in the stool. Dogs have received bone-ash, in many times the therapeutic dose, for years without apparent ill effect.

That the volume of the blood is not normally constant is shown by distinct diurnal variations in this value as determined by the hemoglobin percentage when the animal (dog) is in complete muscular rest and unanesthetized. Administration of ether is immediately followed by a decrease in the relative blood volume which condition persists not only throughout the period of anesthesia but for some time afterward.

When isotonic saline is injected intraperitoneally the diluting effect on the blood volume is not observed until several hours later. The oral administration of saline dilutes the blood to a greater degree in similar time than similar doses of water. As might be expected, intravenous injection of saline show the most pronounced degree of dilution. Usually, however, the normal relative blood volume is attained after approximately two hours following the intravenous injection (100 c.c. per Rg. body wt.). At this time, when large volumes of fluid are unaccounted for, neither the muscles nor the liver show a detectable increase in fluid-not-blood. Renal activity is not essential for this prompt adjustment of blood volume for the urine volume does not account for the “lost” fluid.

Two types of diets were used: one, containing casein as the protein, and being adequate for normal growth; the other, having gliadin as the protein, sufficing for maintenance but not for growth. Both foods provided 5.3 Cals. per gram and were identical in their composition except for the difference in quality of the protein.

White rats of similar ages were employed. A standard tissue damage (liver necrosis) was induced by subcutaneous injection of 1 c.c. sterile mineral oil containing 0.15 c.c. chloroform per 100 grams body weight, which had been previously determined to be the maximum non-lethal dose. The animals were killed at definite intervals, a control rat on standard food together with an experimental rat on gliadin food.

This procedure produces a necrosis of the liver cells around the central vein of the lobule appearing at its maximum 24 to 48 hours after the injection. In both series of rats there occurs an early mobilization of leucocytes and clearing away of the cellular debris with a simultaneous initiation of regeneration from the uninjured cells at the periphery of the lobule. The process of repair is most active at about 72 hours after the injection in both series. At 120 to 148 hours the regeneration is apparently complete and the liver is histologically normal.

The rate of procedure of the repair, so far as we have been able to observe, is the same for both groups of animals. This fact raises the question as to the possibility of such repair involving, as it does, tissue reconstruction upon the gliadin fooda diet upon which general body growth is impossible. A discussion of the application of these results to the theories of intermediary protein metabolism will be reserved for a later publication.

It is usually assumed that extirpation of the uterus is without any noticeable effect on the ovaries, outside perhaps of cystic degenerative changes which have sometimes been observed to follow this operation and which evidently are due to interference with the blood supply of the ovary following injury of the uterine vessels.

Experiments in the guinea pig have however convinced us that a complete or almost complete extirpation of the uterus may be followed by a very characteristic effect, namely a long continued preservation and function of the corpus luteum. Instead of beginning to degenerate fourteen or fifteen days after ovulation, the corpus luteum may remain well preserved and even show attempts at mitotic proliferation for sixty days, or perhaps even as late as eighty days after the last heat. We have not yet determined the limit of preservation of the corpus luteum under these conditions. The cyclic corpus luteum thus equals or perhaps surpasses in vitality the corpus luteum of pregnancy which latter has in the guinea pig a duration of about sixty-five days.

In order to demonstrate this effect we extirpated the uterus a few days after ovulation. The corpus luteum which develops as a result of ovulation remains preserved for a long time following this operation. If we extirpate the uterus in very young guinea pigs at a time when an ovulation has not yet occurred, the first ovulation takes place, notwithstanding the extirpation of the uterus. In various experiments we observed this ovulation to occur as early as nine days and as late as twenty-nine days following the operation. The corpus luteum which thus originates remains then preserved for a long period of time and mitotic figures may often be seen in the lutein cells. In ovaries of these young animals, which ovulated for the first time following hysterectomy, there is usually no remnant of retrogressing corpora lutea representing former ovulations visible. In one case, however, we found such a body side by side with a well preserved corpus luteum. This was probably due to the fact that one of the corpora lutea formed during an ovulation may retrogress earlier than the others.

The experimental analysis of the sexual cycle carried out particularly during the last fifteen years makes it possible to state the main factors regulating its mechanism. While this analysis rests largely on experiments and observations in rodents, and especially in the guinea pig, yet in principle conditions seem to be similar in all the mammals.

Two phases can be distinguished in the sexual cycle; the first phase is dominated by an ovarian factor other than the corpus luteum, in all probability by the maturing follicles the wall of which secretes a substance which causes various kinds of circulatory changes and growth processes and in addition certain psychical alterations. Proliferation, under the influence of this substance, occurs in the mammary gland, in the vagina and also in the uterine wall; this substance calls forth changes in the ovary which culminate in ovulation.

These growth processes usually cease in the vagina suddenly with the appearance of oestrus, while in the mammary gland they may continue for a short time longer.

Ovulation leads to the formation of the corpus luteum. The second phase of the sexual cycle is dominated by a substance or substances given off by the corpus luteum. This substance or these substances sensitize the uterus, making possible the production of decidua or of placentomata, or the normal predecidual proliferation and facilitating the fixation and development of the fertilized ovum; they cause growth processes in the mammary gland and prevent procestrus, estrus and ovulation; the corpus luteum on the other hand does not prevent the maturation of follicles, at least in rodents. The corpus luteurn substance is without a direct effect on vagina, tube or other connective tissue or epithelial structures of the guinea pig.

We can distinguish among mammals at least three types of ovaries. They are represented among the rodents by the ovaries of the guinea pig, rat and rabbit.

(1). In the first type, that of the guinea pig, in the oestrus period, preceding ovulation by a number of hours, the large majority of all follicles—all but the very small ones—begin to degenerate. In the larger follicles the granulosa becomes karyorrhectic en masse and connective tissue begins to grow into the follicular cavity after the destruction of the granulosa epithelium has progressed still further. In smaller follicles the process of destruction of the granulosa is less obvious and the granulosa cells disappear and here also connective tissue soon begins to invade the cavity. In the week following ovulation the very small follicles develop again and pass through their cycle; about eight days after ovulation the follicles have again reached full size and maturation on the one hand and atresia on the other hand can set in; this atresia usually affects only the large follicles.

(2). In the second type, that of the rat, ovulation is not preceded by a general atresia of follicles, but the large follicles mature, rupture and become converted into corpora lutea. There takes place throughout the cycle a limited atresia of follicles. In the rat the ingrowth of connective tissue into the atretic follicles is less active than in the guinea pig, in consequence of which the cavity of the degenerating follicle remains preserved during a longer period of time in the former species. In both guinea pig and rat ovulation is spontaneous. It occurs as soon as follicles have had a chance to mature and the inhibiting influence of the corpus luteum has ceased to exert itself. In the guinea pig a considerably longer period is necessary for the maturation of the follicles than in the rat, because in the former additional time is needed to allow follicles to develop from very small to large size, while in the rat only the very large follicles are destroyed at the time of ovulation and therefore maturation of a new set of follicles can take place in a very short time.

Bial's reagent, as ordinarily made up, does not keep. It is better to dissolve the orcinol and ferric chloride in alcohol, (6 gm. orcinol, 40 drops ferric chloride in 200 c.c. alcohol). Heat 5 c.c. of sugar solution with 15 drops of the above and an equal volume of fuming hydrochloric acid. A blue color develops if pentose is present.

When formaldehyde is subjected to this test a white precipitate is formed if 0.1 mg. per c.c. is present. Upon heating the precipitate turns brown. Acetaldehyde gives a precipitate which does not turn brown. With minute quantities of formaldehyde there is no precipitate but if the test is made strongly alkaline a strong green fluorescence is observed.

Orcinol can be used to detect methyl alcohol as follows:

Distill off the alcohol, using a Vigreux column and allowing as little water to pass over as possible. Place 1 c.c. of the distillate in a large hard glass test tube, which is free from scratches. Add 2 c.c. of 6.7 per cent. potassium dichromate solution and 2 c.c. of 62 per cent. sulfuric acid. Mix at once and let stand for 10 minutes. Add 15 c.c. of distilled water, mix and heat in boiling water for 10 minutes. Add 1 c.c. of a 0.5 per cent. aqueous solution of orcinol. Mix well and continue the heating for 30 minutes. If the original alcoholic distillate contained 5 per cent. of methyl alcohol a heavy white precipitate will appear after 5 minutes heating. With 0.5 per cent. methyl alcohol a faint precipitate is formed after heating for 30 minutes, or upon cooling. To show smaller quantities of methyl alcohol add a slight excess of sodium hydroxide, heat for a few minutes and filter.

Commercial raisins are most commonly prepared by drying two types of grapes, the small Sultanina (Thompson seedless raisin) and the large Muscat of Alexandria (commonly called the Muscat or Malaga raisin). When the latter is dried and sold without removing the seeds, it is known as “loose” Muscat raisin. When the seeds are removed the term “seeded Muscat raisin” is applied.

In this investigation studies have been made on the Thompson seedless raisin, the loose Muscat raisin, the seeded Muscat, raisin seed oil and dried raisin seeds. One hundred and fifty-six rats were used in the experiments dealing with the study of vitamin A and the usual technique was employed with the exception that the yeast was fed in pellet form separate from the ration. In all of the experiments the animals were confined in separate cages and at least eight animals constituted an experimental groups, on the same diet. Food intake records were kept for each individual. The raisin materials were mixed (in ground form) with the basal ration, with the exception of the raisin seed oil, which was fed separately.

A total of eighty-three rats were used in the study of vitamin B and the usual technique was followed with the exception that 5 drops of crude cod liver oil were fed separately each day instead of butter fat.

In preliminary reports 1 Osborne and Mendel demonstrated that rats can grow to considerable size on diets consisting of nine-tenths or more of protein, provided that they receive a suitable supply of vitamins A and B as well as of inorganic salts. Both casein and washed meat were used as the sources of the protein. Similar tests have since been made with rations containing about 75 per cent. of protein in the diet. It seemed unlikely that rations on which young rats grew from 60 to 260 grams could be extremely harmful to the organism. However Squier and Newburgh 2 have concluded, in harmony with a widespread popular belief, that “a high protein diet in man is a renal irritant” and Newburgh and Clarkson 3 have described the production of arteriosclerosis in rabbits on “diets containing 27 and 36 per cent. of protein derived chiefly from beef.” For this reason it seems worth while to give a preliminary account of our observations on some of the organs of rats growing on our diets very high in protein.

The only striking change was found in the kidneys, which in the animals on the high protein diets were greatly hypertrophied. The average weight of the kidneys was almost twice that of the kidneys of control animals and their size about one-third greater. Microscopic examination showed no changes of an inflammatory or degenerative nature. The exact histological condition of the kidneys and of the other organs will be reported in full in asubsequent paper. Hypertrophy of the kidneys existed without hypertrophy of the heart. The ratio of the weight of the heart as well as of the liver to the body weight was about the same in the animals fed the high protein diets as in the control animals. The ratio of the weight of the kidneys to the body weight in the animals on the high protein diets was, on the average, almost double that of the control animals.

The following experiments were suggested by purely biophysical considerations concerning absorption from the intestines and it was deemed desirable to report the same in this preliminary communication.

On purely physical-chemical grounds it can be shown that the absorption of a chemical in solution which flows through a membranous tube will depend on the speed of flow of the solution through the lumen of the tube. The authors attempted to test this postulate on the living intestine in the following manner. Loops of intestines were tied off in dogs, cats and rabbits under anesthesia, and large inflow and outflow cannulas were attached to the two ends of the same respectively. A solution of the drug was perfused through the loop under proper precautions from above downwards, maintaining the temperature of the animal constant and a record was made of the blood pressure and respiratory effects. The drugs employed in these experiments were solutions of potassium cyanide and the powerful alkaloid, aconitin. These solutions were perfused at different rates as indicated by the outflow per minute. It was found that when the solution was perfusetl with slow speed the absorption was much greater than when the rate of flow was more rapid. The drigs may be perfused so rapidly that very little or no absorption of the poisons takes place at all. As illustrations of the above may llre given experiments performed on two cats on December 21st, 1921.

Each animal was anesthetized with ether and in each after laparotmy a loop of intestine 30 cm. long was tied off. Through one of these a solution of potassium cyanide 0.5 per cent. was perfused, at the rate of 100 C.C. per minute. In the other cat a solution of the drug in the same concentration was perfused at the rate of 500 c.c. per minute. It was found that the first signs of poisoning as indicated by the respiration and blood pressure cilrves in the first cat came on much earlier than in the second cat and death occurred in the first experiment in less than one-half of the time after which it occurred in the second experiment

In much of the literature on the suparenal glands one or another relation of these organs to sex and to reproduction is more or less clearly indicated. In a study of these relationships only slight success has been obtained from efforts to transplant suprarenal tissue. 1 2 The present study was chiefly an attempt to obtain functional grafts of whole suprarenals in young doves and pigeons; or, if persistent transplants were not obtained, to repeat the transplantations at intevals during growth so that some of this tissue would be either functioning or in course of resorption during a considerable part of the period of immaturity; and then to examine several aspects of sex and of reproduction in the treated animals.

We have failed in our attempts to obtain permanent grafts. A number of eggs were obtained from each of the transplanted doves and offspring of these birds were reared. Records were made for both the parents (transplanted) and offspring during a period of 30 months and data obtained for various reproductive capacities and characteristics (age of maturity, fertility, viability, rate of egg production, number and kind of various abnormalities). An examination of these data indicates that none of these characteristics in the offspring can be associated with the transplantations made upon their parents; and that most of these reproductive characteristics were also unaffected in the transplanted birds themselves. This considerable amount of data comprising negative results need not be further considered and is left with the above brief statement. It is thought, however, that two or three by-products of these transplantation tests are of sufficient interest to record, though the data are too meagre to warrant conclusions.

Fifteen ring doves and five common pigeon young received the first transplants at 5 to 10 days after hatching (age was therefore 19-24 days). Additional transplants were made in several cases at about 25, 50, 75 and 100 days later

These experiments indicate that there is no constant relation between the total fatty acids and the chylomicrons in the blood after feeding of fat and glycerin, and in the lipemia of rabbits produced by acute hemorrhage. The determination of the chylomicrons, therefore, cannot be taken as an index of the total fat in the blood.

That the chylornicrons are associated with the fats in the blood is confirmed.

Landsteiner, in 1901, discovered three human blood groups, and soon afterward, Decastello and Sturli, pursuing the study with Landsteiner's knowledge and encouragement, found the fourth group. Landsteiner recognized the existence in human serum of two isohemagglutinins, which were designated with small letters a and b, or (by von Dungern and Hirschfeld) α and β and two isoagglutinable substances in the corpuscles which were called A and B.

Quite recently Guthrie and Huck have reported the discovery of a third pair of isoagglutination elements, which, to bring the terminology into conformity with the previously existing one, we will refer to as isoagglutinin c and agglutinable substance C.

The relation of the exceptional individuals observed by Guthrie and Huck to the original four blood groups is shown in Table I.

In the course of a class demonstration of the two original pairs of isoagglutination elements, the serum of one of us (A. F. C., a Group I individual) was absorbed, first with the washed corpuscles of a Group II individual (Levine) and then with the washed corpuscles of a Group III individual (Sutton). The supernatant fluid, after these absorptions, no longer clumped the ccrpuscles of the respective Group II and Group III individuals, but still possessed a vigorous power of agglutinating the corpuscles of a Group IV individual (Johnson) which were agglutinable by the serum of both Levine and Sutton.

This observation indicated the presence of another undescribed pair of isoagglutination elements, which we shall tentatively designate as × (isoagglutinin) and × (isoagglutinable substance).

In cases of transfusion where it is difficult to obtain a donor of the same group as the patient, a person of Group I (Jansky) has been considered suitable on account of the fact that the corpuscles of this group are inagglutinable by the isohemagglutinins. The fact that the plasma of a Group I individual contain isoagglutinins capable of clumping the corpuscles of the recipient has been ignored, because, under ordinary circumstances, the quantity of the plasma transfused is too small to affect the larger proportion (five to ten times) of the recipient's corpuscles.

In carrying out the method of direct matching of bloods described by Coca, 1 we found that the plasma of one of us (M., Group I) produced complete agglutination of ten volumes of the corpuscles of P. L. (Group II). Quantitative tests in the test tube showed that moderate agglutination of P. L. corpuscles was produced by as little as 1/20 volume of the M. plasma.

This observation reveals a risk in using a member of the so-called “Universal Donor” group for the first time without making a rough quantitative examination of the agglutinating power of the individual's plasma. It is advisable, therefore, in carrying out the direct matching according to Coca, to include a mixture of equal parts of undiluted recipient's citrated blood with the donor's citrated blood, diluted 1 to 5.

Observations made on the distended contractile vacuoles of Amoeba verrucosa subsequent to the application of the murexide test give no optical evidence of the presence of uric acid in the vacuolar fluid. This result is not in accord with the generally accepted conclusions of Griffiths 1 who reported the production of “prismatic crystals of murexide” on application of this test to Amoeba proteus, Vorticella and Paramecium. Mass cultures of Amoeba verrucosa, killed in 50 per cent. alcohol, show the contractile vacuoles fixed at partial or complete expansion in a large percentage of cases, but a critical examination of the fluid of the vacuoles has never revealed the presence of uric acid crystals precipitated there by this method. Dark field examination of such animals shows the fluid to be structureless. Treatment with nitric acid and ammonia in the manner described by Griffiths colors both cytoplasm and pellicle a lemon yellow. Furthermore, control slides, consisting of a small quantity of pure uric acid crystals in distilled water, when similarly treated, also give a negative result. Such controls give ammonium purpurate only on being evaporated to complete dryness, a procedure obviously so drastic as to render it inapplicable in the case of protozoan cells where it is essential to preserve the vacuole and its contents intact. Mass cultures so treated leave a lemon yellow residue.

Cultures of Paramecium, concentrated by brief centrifuging, and subjected to the same test also offer only negative results. Examination of the vacuolar fluid is precluded in these forms by the contraction of the vacuoles during the process of fixation.

This would indicate, therefore, either that uric acid is not an end product of the katabolic activity of these protozoa, or that the murexide test is not sufficiently sensitive to give a satisfactory optical reaction in this case. In view of the latter possibility, other and more specific methods for determining the presence of uric acid in minute quantities, are being applied at the present time.

The contractile vacuole of Amoeba verrucosa is formed by the union of a variable number of lesser globules. Subsequent to each contraction a new series of contributary globules may appear in the same location as the preceding vacuole, but this is not invariably the case. Random formation of new vacuoles is common, either at some distance from the original organelle, or close by. The walls of two functioning vacuoles may lie in contact for some time without confluence.

The wall of the vacuole is easily indented with a blunt microneedle. A sharp-pointed microneedle will induce artificial systole by perforation of the wall of the vacuole.

A large vacuole, or a group of contributary globules, freed into the water by ripping open a cell under slight pressure, retain their identity in this medium for some time. A quick upward thrust of a blunt needle against a large vacuole apparently causes its rupture into the surrounding endoplasm, for an increase in fluidity is observed in this area. This would imply the miscibility of vacuolar fluid and endoplasm.

The contractile vacuoles of Paramecium caudatum become dilated on the addition of Grubler's Alizarin blue to the culture. The distended vacuoles together with their feeders become set, gradually taking on a vivid blue color. This stain is specific for the vacuolar walls, and, although it appears long before ciliary action ceases, should be classified as a sub-mortem rather than a vital stain. These stained and distended vacuoles do not collapse when removed from the animal but retain their shape and may be manipulated and dissected with microneedles.

The specific structure and characteristic properties of the pellicle of Amoeba verrucosa may be demonstrated in the living animals by means of microdissection and injection. A high degree of resistance to mechanical pressure is exhibited by the pellicle when subjected to compression. Perforation of the pellicle without indentation of this layer can be effected only with needles having extremely fine points. Withdrawal of a needle after penetration carries the pellicle out into a long coneshaped process, loss of endoplasm being prevented by the rapid formation of a restricting endoplasmic film at the base of this cone. Distortion of the pellicle due to compression or other injury persists for some time before the membrane resumes its normal condition. The tough but extensile nature of the pellicle may be further demonstrated by inflating the cell with distilled water injected through a micropipette. From a cell ruptured under pressure the entire pellicle may be removed. In every case recorded the nucleus came away with the pellicle. Injection of various solutions usually causes an immediate inflation of the Amoeba, the fluid collecting in an area beneath the pellicle, while the endoplasm contracts into an irregular mass containing the nucleus.

Within recent years, a great deal of evidence has been accumulated from both clinical and laboratory sources which has tended to show that some kind of connection exists between the occurrence and growth of tumors and the system of organs regulating animal economy. For the most part, the evidence bearing upon this problem has been circumstantial. In a few instances a definite relationship appears to have been established as in the case of the experiments reported by Loeb 1 concerning the effects of castration on the incidence of mammary tumors in mice and those of Rohdenburg, Bullock, and Johnson 2 on the effects of operative removal of various organs upon the growth of transplanted tumors and the immunity of tumor animals. There are also therapeutic observations on the use of thyroid and of thymic products alone or combined with castration 3 which might be regarded as equally suggestive were it not for the fact that similar results have been obtained by the use of a variety of means. 4

Two years ago, we reported the occurrence of a malignant tumor in the scrotum of a rabbit infected with Treponcma pallidum. 5 Since that time, a careful study has been made of the conditions presented by this animal 6 and a long series of investigations based upon the behavior of transplanted tumors derived from this stock has been carried out which have shown that there is an essential connection between the growth of transplanted tumors and certain members of the endocrine system on the one hand, and the mechanism of animal defense on the other.

In the animal with the spontaneous tumor, there was a notable tendency to atypical epithelial proliferation in many parts of the body associated with a deficiency in the reaction of the surrounding connective tissues and a widespread tendency to degeneration. There were also very striking alterations in such organs as the thyroid the thymus, the suprarenals and the lymphoid tissues, which might have been regarded as part of a general organic deterioration attributable to one or both of the diseases present.

In the course of the work that has been carried out in this laboratory during the past 8 years, it has become more and more apparent that of the several factors concerned in determining the general course and severity of syphilitic infections, the spirochete is of minor importance as compared with the number of other influences, especially the factor of animal resistance. This is strikingly illustrated by the experimental infection in the rabbit which, as a rule, is comparatively mild but in exceptional instances may assume a more malignant form or persist in an active state with severe constitutional symptoms for months or even years and in the end cause the death of the animal.

Little has been known as to the nature or source of the forces employed by the animal in combating this disease but there is a great deal of evidence to indicate that resistance to this infection is a function of animal economy and is subject to the same influences as are concerned in the regulation of growth and development, and the maintenance of general metabolic activities. Moreover, there have been specific indications that resistance was influenced to some extent by the endocrine system.

This possibility was first suggested to us by the occurrence of marked seasonal changes in the severity of the disease but was most forcibly impressed upon us by the abnormal resistance displayed at times by pregnant and lactating females.

Sherrington's emphasis on the differences between reflex are conduction and conduction in the nerve trunk has long been familiar to physiologists. With the establishment of the all-or-none law for muscle and for nerve in recent years it was but natural that the query should be raised as to whether reflex action be bound by the limitations of the all-or-none behavior of muscle and of nerve, or whether the central nervous system so alters the final nerve impulse to muscle that it can over-ride these limitations. A single sensory impulse, for example, might always result in a volley of impulses along the final motor neurones, according to Sherrington's suggestion. 1 Variations in the number of impulses comprising this volley might, by summation of contractions, conceivably produce in the muscle degrees of response of almost any grade of fineness, and reflex contractions would not then show an all-or-none character.

The present research was undertaken with the object of comparing minimal, or close to minimal, responses of muscle secured by single shocks to its motor nerve, with similar responses obtained reflexly by stimulation of a sensory nerve.

The muscle employed was the tenuissimus, a long slender, straight-fibered muscle underlying the biceps and containing approximately one thousand fibers. 2 The nerve to this muscle contains about twenty motor neurones, each neurone presumably innervating approximately fifty muscle fibers. If this nerve be stimulated by single shocks gradually increasing in strength the response of the muscle is not gradual, but by step-like increments, there being no gradation between one step and the next. The heights of the first three increments to appear in three of such experiments are shown in Experiments I, II and III of the appended table.

Method: The lungs of a pithed cat are aerated by Meltzer's method of intra-tracheal insufflation, temporarily interrupted once per minute. A constantly inflated lobe is carefully elevated from the open thorax and its margin fixed by two serrated clips. If an area near the margin is transilluminated by a suitable optical system the alveoli with their related blood vessels can be directly observed and studied by means of a horizontally arranged microscope magnifying 50 to 100 diameters.

Owing to the wall-thickness of the intra-lobular arteries and veins and their immediate branches, the circulating blood is only visible in the smaller vessels, i. e., (a) in the pre-arterioles and arterioles passing between the alveoli, (b) in the capillary network formed on the surface of the alveoli and embracing the air cells and (c) in the smaller venules before they merge into the efferent intra-lobular veins.

Circulation in Vessels of Different Size: In the pre-arterioles the blood stream is often clearly visible. The cellular elements are in densely packed mass formation, the flow is pulsating with a rapid systolic forward movement and a diastolic decrease in velocity, amounting occasionally to a total stoppage of the flow in diastole. Expansion and relaxation of the walls is not observed. In the arterioles the stream is no longer pulsating, the individual blood cells are distinguishable and travel in ranks of two to six. In the capillaries, the blood corpuscles travel in single file, and in a continuous stream, but the rate of flow varies in different capillaries even in the network surrounding the same alveolus. In the smaller venules the stream is constant and a little more rapid than in the capillaries. Individual cells are distinguishable and eddies often occur at the junctions. In still larger venules and the smallest veins the stream is no longer constant, but a definite pulsation occurs, the onward flow tending to be reduced during systole and increased during diastole.

This communication is based on a series of experiments conducted on 5 normal men and 8 diabetics. The method of determining the respiratory exchange was carried out by the open-circuit Tissot spirometer, with gas analyses by a modified Henderson-Haldane apparatus and calculations of the indirect calorimetry by the method of Zuntz and Schumburg, as recently described by McCann and Hannon. 1 At varying intervals during the experiments samples of blood were taken and blood-sugar determinations were made according to the method of Folin and Wu. 2 Coincident pulse and blood pressure charts were also kept in most instances.

The 5 normal men showed a similar reaction to 0.5 c.c. adrenalin given subcutaneously. There was a prompt rise of the R. Q., which was most marked about 10 minutes after the adrenalin. The quotient rose to a different extent, varying with the individual. A distinct increase of heat-production occurred with its maximum degree about 30 minutes after injection, remaining above the basal figure for over an hour. There appeared an immediate and marked rise in carbohydrate metabolism with its maximum about 10 minutes after the adrenalin, as calculated from the non-protein R. Q. Coincident alveolar air analyses in one case indicated that overventilation may play a rôle in the production of the higher respiratory quotients. An elevation of blood-sugar followed the injection of adrenalin reaching a maximum in 40-50 minutes after administration. It was the most constant of any of the observed effects of the drug. The sugar remained above normal limits for over 1 1/2 hours. The pulse rate and blood pressure responded differently in different individuals.

The reaction to adrenalin of the diabetic patients differed in some respects from that in the control cases. Its appearance was the more delayed, the sicker the patient. The R. Q. did not rise nearly as high as in the controls. There was only one diab'etic patient whose R. Q. went up as many points as the minimum control. After adrenalin the eight diabetic subjects fell into two groups as regards the total heat production:—four showing an increase from 6 to 18 per cent., the rest ranging from 29 to 43 per cent. above normal. The equivalent figures for the five normal cases lay between 17 and 33 per cent. above the basal heat production. The blood-sugar did not rise as high nor as quickly as in the controls, with one exception.

This method aims to overcome variations in antigens employed in the precipitation test for syphilis proposed by the author. Two antigens prepared under identical conditions from two different lots of beef heart, will be likely to show variation in sensitiveness when tested with syphilitic sera. This variation in sensitiveness may be considerably lessened if each antigen is first mixed with salt solution in such proportion as to bring forth its maximum power for producing precipitates with syphilitic sera. Generally speaking, this is accomplished by mixing antigen with minimum amounts of salt solution which will produce opalescent mixtures. Since the mode of adding salt solution to antigen markedly affects the final product, it is obviously important to render all conditions which are likely to affect this titration as constant as possible.

The test tubes used in this titration should not differ from those used in the regular tests. In our experience, tubes of 0.8 to 1 cm. diameter give best results. The set up of the titration is as follows:

Six test tubes receive 0.5 C.C. antigen each. Six similar tubes receive respectively 0.25, 0.5, 0.75, 1, 1.25 and 1.5 C.C. salt solution (0.85 per cent NaCL). The 0.25 c.c. salt solution is now poured into the first antigen tube and the mixture immediately poured back and forth several times. The next antigen tube receives the 0.5 C.C. salt solution in a similar manner. The remaining antigen tubes receive the four increasing amounts of salt solution under similar conditions. It will soon be observed that the first two or three tubes show varying degrees of precipitation, the middle tubes, clouding and the last few tubes, opalescence.

Employing the antigen titration outlined in the previous paper, the question arose to what extent the preparation of antigen may be varied without affecting the final results. The following antigens were accordingly prepared, titrated with salt solution and smallest amounts of the latter used which produced opalescent antigen-salt solution mixtures. The tests were carried out with positive and negative sera according to procedures I and II.

Antigen 1. This was prepared as described in a previous paper of these Proceedings. 1 Dried beef heart was freed from ether extractives and subsequently extracted in 95 per cent. alcohol for 9 days in the ice box and overnight at incubator temperature. Color approximated potassium bichromate color standard.

Antigen 2. After extracting the dried heart with ether in the usual manner, boiling alcohol was poured on the dried material, shaken and extraction continued for 1 day in the incubator. Color approximated potassium bichromate standard.

Antigen 3. Same as antigen 2, except that the alcohol extraction was carried out in the water bath at 37.5° C. for 2 days with frequent shaking. Same color index as above.

Antigen 4. Same as Antigen 1, to which was added a second alcoholic extract carried out for 2 days at incubator temperature having approximately the same color range.

Antigen 5. Ether extractives were removed in the usual manner and dried heart muscle extracted with alcohol for three hours in reflex condenser.

Antigen 6. Ether extraction carried out for 1 hour in reflex condenser, followed, after filtration and washing with fresh ether and drying, by alcohol extraction also by means of a reflex condenser for 1 hour.

The development of the thyroid of the salamander Ambystoma opacum was studied on serial sections of thyroids in various stages before and after metamorphosis.

The proportion of colloid and epithelium was found by weighing separately wax models of the colloid and epithelium. During the larval period the colloid increases more rapidly than does the epithelium. From 13 per cent., shortly after hatching, it increases to 45 per cent. of the total thyroid mass just before metamorphosis. The larval period is not a period of colloid release, but of colloid elaboration and storage. At the beginning of metamorphosis, the colloid percentage drops suddenly below 30 per cent. This drop is due partly to the sudden and excessive release and disappearance of the colloid from the follicles, and partly to an excessive increase of the epithelial mass, owing to the swelling of the individual cells. With increasing age the thyroid again enters upon a stage of colloid accumulation. In animals 4.5 years of age, 56 per cent. of the thyroid mass is colloid. This relative increase is not due entirely to an absolute colloid increase, but also to an absolute decrease of the epithelium.

The colloid was studied on sections stained, according to Kraus'method, with polychrome methylene blue and acid fuchsin.

Colloid is elaborated long before follicles are formed, at a time when the cells still contain yolk and are devoid of granules, a structure frequently considered to be prerequisite to colloid elaboration. In thyroids of larvæ fed exclusively thymus gland, the cells remain permanently primitive, consisting almost entirely of nucleus, possess very small amounts of plasma and are devoid of granules. Yet these cells elaborate large quantities of colloid (55 per cent. of the total thyroid mass).

The subcutaneous injection of paraphenylenediamin hydrochlorid in the dosage of 0.19 gm. per kilo in rabbits, produces a peculiar, specific edema of the head and neck in from one to three (median, one and one-half) hours after the administration. At the same time there is a relative increase in hemoglobin and total solids of the blood due to escape of fluid from the circulation, indicating that increase in vascular permeability may be a factor in the production of the edema. With this dosage, and providing the freshly dissolved drug is injected, edema is produced almost invariably. Old and standing solutions (even for 24 hours) are uncertain and ineffective, and this, in part at least, explains the variabilities previously encountered. The following is a summary of results on 96 animals which have been used for the study, in various ways, to date.

Meissner claimed that large doses of atropin and calcium prevented the edema. We have not been able to confirm this. Maximal doses of the following agents injected in various ways did not influence the development and the course of the edema nor the blood concentration; calcium, atropin, morphin, chloral hydrate, urethan, ether, sodium bromide, cocain, ergotoxin, antipyrin, neocinchophen, sodium salicylate, quinin and cinchophen (in some rabbits). In about half of the rabbits receiving cinchophen, the production of pleural and peritoneal exudates was favored and the mortality increased. Our results with atropin and calcium agree with negative results of Gibbs who used cats instead of rabbits for studying the edema of paraphenylenediamin. The only agent which has prevented the edema thus far is nicotin in large doses hypodermically. Section and degeneration of the cervical sympathetic nerves did not prevent the edema.

A variety of agents, previously reported by Hanzlik and Karsner cause anaphylactoid reactions when injected intravenously. With many of these emboli and thrombi composed of red blood corpuscles, fibrin and platelets are demonstrable in the lungs, and hemagglutination occurs in vitro. These changes together with the alterations in chemical composition of the blood recently demonstrated in this laboratory, are regarded as objective evidences of disturbances in important physical and chemical equilibria in the fluids and tissues of the organism, and as being of fundamental importance in the explanation of reactions from a variety of agents. The present report is a summary of the effects of different agents upon the thrombocytes (platelets) in pigeons.

For differentiation, the modified Nocht stain described by Hastings 1 was used and for counting, a modification of the cresyl violet stain of Buckman and Hallisey. 2 All agents were injected intravenously at body temperature into the wing veins and blood was obtained from superficial veins of the legs.

The following agents, which cause anaphylactoid symptoms in guinea pigs, pulmonary emboli and thrombi, and hemagglutination in vitro, produced increases in number and clumping of thrombocytes in pigeons; peptone, agar-sol gel, toxified agar, Congo red, collargol, charcoal, kaolin, colloidal iron, colloidal arsenic, 50 per. cent. acetic acid and 6 per cent. acacia. Histamin, tannin and arsphenamin (in small dosage) produced doubtful or no changes in the thrombocytes, but sections of lungs and livers showed marked clumping of erythrocytes from these agents.

Histological examinations of the lungs, liver, spleen and kidneys of all animals showed congestion and thrombosis after the majority of the agents that were injected. In a few cases marked hemorrhages were found. The majority of these agents caused definite symptoms, ranging from shivering, crouching, and increase in respiration to death.

There is a growing opinion that many of the pathological conditions of the oral cavity (caries, pyorrhea, etc.) are the result of faulty diets. The acceptance of this assumption makes it necessary to know the normal constituents of the saliva. During the last year several investigators have reported on one or more of the salivary constituents but a complete correlation between the dietary and salivary constituents is lacking. The work presented here is very incomplete and represents only one phase of a complete survey of the mineral metabolism being made on a number of adult human subjects.

The results thus far, given in a brief form in the following table, have been obtained from weekly analysis of saliva samples from six healthy adults, all of whom have been on a fixed diet for a periods of eight weeks.

A strain of B. histolyticus (No. 141) received from the Pasteur Institute in Paris in March, 1921, and described by me 1 last year, was recently found capable of repeated and successive aerobic culture upon the surface of meat infusion and blood agar slants without resort to any method of reducing the oxygen pressure. The accuracy of this observation is guaranteed not alone by the unique pathogenicity, but as well by the particular combination of morphologic and cultural properties of this species which enable one to identify it without reference to its oxygen requirements.

All authors dealing with B. histolyticus have regarded it as an obligate anaerobe and my own previous failure to observe aerobic growth can only be explained by my reliance upon meat extract, instead of meat infusion, agar slants, for detecting aerobic growth until recently. Even the growth upon meat infusion agar is extremely delicate and might be easily overlooked by any but a highly critical dye. My first supposition upon observing the delicate transparent aerobic growth upon a meat infusion blood agar slant was that a contamination had occurred. This assumption was clearly denied when as many as 31 successive transplants of this culture were made upon blood and plain meat infusion agar during a period of about 60 days without altering any of its morphologic, cultural or pathogenic characteristics permanently; there were temporary differences observed, however, in the morphology of the first few aerobic cultures, but later transplants appeared to be identical in every way with a corresponding strain cultured anaerobically in brain medium.

The possibility of contamination was also excluded by finding no significant differences in well separated deep agar colonies and in the observation that subcultrrres from such deepcolonies may also he grown aerobically.

During a survey of California soils for anaerobic bacteria, we encountered a strain of B. histolyticus which is of interest as the first recovery of this species in America and one of the few records of its occurrence in soil. All of the other cultures so far described came from war wounds in France and the only recorded proof of this organism as an inhabitant of soil, aside from the fact that most war wounds are contaminated by dirt, is a statement by the Medical Research Committee 1 that, “it … has been obtained from earth.”

The soil specimen was a clay adobe from near Walnut Creek, California, which lies in a rich agricultural valley of the coast range about 12 miles from Berkeley.

The filtrate of an initial culture of this soil in a meat mash medium in the constricted tube 1 contained also the toxin of B. botulinus Type A, and it was during our effort to recover this organism that the Bacillus histolyticus was isolated.

The primary culture contained numerous obligately aerobic hay bacilli and it is interesting to note that while our usual use 2 of gentian violet easily eliminated these by selective bacteriostasis, it was impossible in six trials to eliminate a certain facultative aercbe-anaerobe which we now consider to have been none other than the B. histolyticus since that was the only organism that could be isolated from the subsequent deep agar colonies.

The isolated culture corresponds in all of its morphologic cultural, and pathogenic properties to the war mound strains received from Dr. Weinberg of the Pasteur Institute of Paris or indirectly from Dr. Kahn of Cornell University Medical School.

I wish at this time to correct a mis-statement regarding the fermentative power of B. histolyticus that appeared in my 1922 paper, 3 in which I recorded acid and gas production in glucose and uncritically accepted the records of Henry 4 and the British Medical Research Committee 5 of fermentation of glucose, levulose and maltose, which were based, like my own, on the study of a single strain. My own result may have been due to an undetected contamination. At any rate, Weinberg and Seguin 6 , Kendall, Day and Walker 7 and Kahn 8 were inclined to deny fermentation of sugars, and a more critical study of our five strains shows clearly that they ferment neither glucose, levulose, maltose, lactose, saccharose, salicin, glycerol nor inulin. That is, there is neither increase of hydrogen ion concentration nor considerable gas production.

B. histolyticus is thus the third sporulating anaerobe failing to derive its carbon from any of the commonly tested carbohydrates, glucosides or alcohols, the other two being B. tetani and B. putrificus.

The following table summarizes the outstanding differences between these three non-fermentative species of sporulating anaerobes:

B. histolyticus was first described by Weinberg and Seguin 1 in 1915 as an obligate anaerobe from war wound infections in which is may display a remarkable and peculiar lytic activity. Pure virulent cultures injected intramuscularly into guinea pigs literally digest the flesh from the bones, hence the name—histolyticus.

Notwithstanding its marked ability to dissolve living muscular tissues its proteolytic action in brain, meat, milk, serum, and egg mediums is slow though fairly complete over long periods of time. Weinberg and Seguin 1 found that sterile filtrates liquify coagulated egg white and gelatin and Blancet Pozerski 2 and Dernby and Blanc, 3 concluded that a tryptic enzyme is secreted by this species.

Weinberg and Sequin 1 also noted that sterile filtrates injected intramuscularly in fairly large doses produce large hemorrhages, apparently by destruction of the vascular walls and we have confirmed this phenomenon by injecting guinea pigs with 2 c.c. of the Berkefeld filtrate from 24 hour glucose broth cultures of several of our strains including the California strain. Sterile hematomas the size of a pigeon's egg generally appear within several days following such injections. The blood contained therein does not become hemolysed and recovery follows their drainage. We have never observed the muscular tissues to disintegrate or the skin to break, however, from the injection of filtrates as after the injection of living cultures. It seems apparent that the hemorrhage induced by the soluble secretions of this species plays an important rule in the peculiar phenomena accompanying at least experimental infection in affording a suitable focus for the growth of this germ which appears rarely if ever to invade the blood stream itself.

Several investigators have observed the effects of rearing rats on milk diets. Matill 1 and co-workers especially have studied this problem, using dried milk or concentrated solutions of milk powder for the most part in their experiments.

We have made a limited study of this problem with the following results. Using fresh, raw liquid milk exclusively 2 we have successfully raised 3 male albino rats to full size at maturity, beginning at weaning. This trial was begun August 10, 1922. Each rat was kept in a separate cage with one-quarter inch mesh wire bottom without bedding. The consumption of milk solids by these rats has varied between 35 and 100 grams per week per rat. These rats, however, on repeated trials have failed to exhibit any mating instinct when placed with females from our breeding colony, which were known to be in heat.

In a second experiment begun October 24, 1922, two colonies of rats, each containing 5 rats, with both sexes represented, on wire bottom cages, have not attained the expected size at maturity on the fresh, raw, whole milk diet. Growth was normal for the first 50 to 70 days only. The calculated average consumption of milk solids for each rat has not been as high as in the case of the rats in Experiment 1, kept in separate cages. The females have so far been entirely barren, confirming Mattill's results.

In a third experiment 3 female rats reared to partial maturity on mixed diet, each having reared one litter, were placed in the above colonies on the whole milk diet. To date, one of these rats has littered twice and the other two once, since being placed on the milk diet; for two of these litters the sire must have been one of the male rats raised on the milk diet, but the other two litters may have been sired by a male placed on the milk diet from a mixed ration.

Spiess 1 and Ninian Bruce 2 have shown that the oedema of the conjunctiva which can be produced by instilling mustard oil into the conjunctival sac of a rabbit, can be prevented by the instillation of 10 per cent. cocaine. Bruce interprets this action as being due to the inhibition of vasdilator axon reflexes to the capillaries and smaller arterioles, and he found that if the ophthalmic nerve was cut before applying mustard oil oedema could then be produced only during the period in which the vasodilator fibers were still undegenerated, and a sufficient time were allowed for the vasodilator nerves to degenerate completely before the mustard oil was applied, oedema did not result after the instillation of mustard oil. Bruce's results have been confirmed by Bardy. 1

In 1917 I 2 was able to demonstrate the following facts : (1) That the application of epinephrin, producing a local vaso constriction of the conjunctival vessels, inhibited the development of the oedema, for about an hour, but that after the epinephrin effect had worn off oeclema developed subsequently. (2) That any continued lowering of the general blood pressure to less than 50 mm. Hg. prevented or greatly retarded the development of the oedema, even if the conjunctival vessels were dilated by the local administration of 1 per cent. sodium nitrite or if lymph secretion were increased by the intravenous injection of “Witte's peptone.” (3) That intravenous lowering of the general blood pressure by the continuous intravenous injection of dilute hydrochloric acid, in spite of the acidosis produced, diminished or entirely prevented the development of the oedema from mustard oil. (4) That ligation of the carotid artery greatly reduced the development of oedema in the eye upon the side corresponding to the ligated artery; and that in the peripheral end of the ligated artery the blood pressure was less than 40 mm. Hg.

Great difficulty is experienced in extracting fat from wet or dried micro-organisms. It has been suggested that a large portion of the fat they contain may be held in physical or chemical combination by some ingredient of the protoplasm and various preliminary treatments have been developed to free it from such combination. Three of the simplest are that of Larson and Larson for bacteria 1 which depends on simultaneous drying and extraction by acetone, followed by ether extraction of the solid residue and the acetone extract; I. S. MacLean's method used on yeast 2 which consists in boiling with normal HCl, washing and then extracting with ether in a Soxhlet; and the method developed by C. R. Smith for work on edible pastes, in which he boils the sample with alcoholic ammonia, and then extracts with ether.

The work here reported was done on Oidium Lactis. In the first experiment a two weeks'growth was drained by suction and divided into three portions, one of which was treated by each of the above methods without previous drying. The residues as well as the extracts were dried to constant weight and the total dry weight of the samples obtained by addition. The MacLean method is impractical on moist samples, as an unmanageable mucilaginous brown material results from the acid treatment. The acetone method gave 1.21 per cent. ether extract; the alcoholic ammonia method gave 6.13 per cent. No further studies were made on the lipoids thus extracted, as drying to constant weight, either in an oven at 100° or in a vacuum dessicator over P₂O₃ at room temperature, results in a hard, semitransparent brown material, insoluble in petroleum ether, only a portion of which is soluble in ethyl ether.

Bacillus subtilis characteristically produces a diffuse turbidity on broth, which usually clears toward the end of the first twelve hours, the organisms floating on the surface in small islands. These later grow together producing a heavy wrinkled pellicle. It seemed therefore that a study of possible factors influencing this spontaneous migration to the surface might illuminate the subject of pellicle-formation and surface growth in general.

Equal amounts of a young diffuse culture of B. subtilis were introduced into tubes containing each 10 c.c. of ordinary broth. These were incubated and at hourly intervals for 36 hours observations were made of morphology, progress of growth, spore formation, buoyancy of the pellicle and surface tension of the medium. The surface patches appeared at 10 hours, and the pellicles were well formed at 15. The first positive heat test for spores was obtained at 20 hours, but the cultures were not consistently positive until after the 24th observation. At the 25th hour the first spores were observed in stained preparations. At all times the majority of the growth could be centrifuged to the bottom of the tube, showing that the organisms are heavier than the medium. Occasionally small rumpled fragments of the pellicle remained at the top of the centrifuge tube, but these may well have imprisoned air bubbles.

Larson 1 has shown that this organism does not form pellicles on media whose surface tension has been sufficiently reduced by soap. The question naturally arises whether the diffusely growing organisms exhaust some surface tension depressant in the media prior to their upward migration. This, however, is not the case, since the surface tension remained approximately 59 dynes throughout the period of hourly observation (36 hours) and was still the same 13 days after inoculation, at which time the pellicles sank spontaneously.

Previously we have shown that the total removal of the liver is followed by (1) a rapid decrease in blood-sugar, (2) always accompanied by a characteristic group of symptoms, and (3) the administration of glucose which abolishes the symptoms, temporarily restoring the animal to the normal state. All of this is comparable to what happens when a large dose of insulin is administered to a normal animal.

The problem investigated is the rôle the liver plays in the effect of insulin. Therefore, insulin was administered before and after the total removal of the liver in dogs. Removal of the liver did not affect the sharp precipitate drop in the blood-sugar after administration of insulin. Whereas injecting glucose in an animal with its liver intact during the condition of insulin hypoglycemia restores the sugar level permanently, after the liver is removed no permanent level can be maintained despite frequent large doses of glucose.

The same experiments were carried out on dogs before and after partial removal of the liver. While the same sharp drop in blood-sugar occurs after giving very small doses of insulin, there is a slow restoration of the curve, even though only 20 per cent. by weight of the liver remains.

From these experiments we conclude that the symptoms associated with hypoglycemia following the administration of insulin do not differ essentially from those we noted in association with the total removal of the liver, and the action of glucose seems to be identical in the two conditions. The effect of large doses of insulin in producing hypoglycemia is not changed by the total removal of the liver, nor is the hypoglycemic action of small doses of insulin modified by partial (60 to 80 per cent.) removal of this organ. However, the liver is necessary for the permanent restoration of the sugar level.

Every clinician must have observed from time to time the appearance of sugar in the urine after prolonged use of thyroid extract in patients.

Magnus-Levy 1 writes that glycosuria may be noted from large dosages of thyroid in man as well as in dogs. Since glycosuria may occur as a result of thyroid feeding, it was thought that the high percentage of iodine in the gland of the ox might be the determining factor in the appearance of sugar in the urine after prolonged use of the drug.

According to Hammarsten 2 the thyroid of man contains 0.34 per cent. iodine; the thyroid of ox contains 0.86 per cent. It was thought that thyroid which is iodine free must have the reverse effect on the blood-sugar contents. It is a well established fact that the thyroid of newborn animals does not contain iodine. Thyroids from such calves were procured from the slaughter house immediately after killing and extracts from the weighed glands prepared. The alcoholic extracts in Ringers solution were injected intravenously or subcutaneously'or administered by the stomach tube to dogs. The extracts were prepared in the same manner as the pancreatic extracts by Banting and Best.

After weighing the thyroids they were cut in very small pieces and placed in 0.2 per cent. HCl in 95 per cent. alcohol and allowed to stand for two days. They mere then macerated with quartz and filtered, or when using large amounts first macerated and then put in a press and filtered. The clear filtrate was evaporated to dryness in a warm air current. On the next day or on the third day the dry residue was emulsified with 25 C.C. Ringers solution. 25 C.C. of this solution contained the thyroid residue from 10, 20 or 50 grams gland.

As reported 1 at the last meeting of this branch, heating to 75° and 80° C. for one-half hour does not destroy the anti-diabetic substance extracted in acid aqueous media. The original observation of Murlin and Kramer that a potent extract could be prepared by boiling dog's pancreas in 0.2N HCl was confirmed 2 some time ago. Recently we have compared the yield of rabbit units obtained on the one hand by grinding macerated pancreas in 0.2N HCl in a bacteria grinder for 15 hours and, on the other hand, by heating to 75° C. for one hour or bringing just to boiling temperature.

Kimball, Piper and Allen 3 showed also at the last meeting that the active substance is precipitated by complete saturation with sodium chloride, and with either of the alcohols, methyl, propyl, butyl or amyl when added to a solution in 80 per cent. ethyl alcohol. For example, preparation No. 140, heated in 0.2N HCl to 75° C, cooled, strained through cheese cloth, neutralized to P_H of 4.1, and filtered, gave in two tests representing 10 gms. pancreas each, a fall of blood-sugar in normal rabbits of 44 and 62 milligrams. The second day the filtrate was precipitated by saturation with sodium chloride, the precipitate taken up in 70 per cent. alcohol, centrifuged to remove insoluble protein and reprecipitated with 5 vols. amyl alcohol. The excess of amyl alcohol was evaporated off before a fan and the residue taken up in sterile water. Equivalent doses representing 10 grams pancreas now gave drops in blood-sugar in rabbits of the same size of 77 and 74 milligrams showing that all the potency had been removed. The filtrate from the NaCl precipitate contains no active substance.

Fasting animals show a different counter regulation for iletin than animals which received food. Rabbits with food plus iletin may show lowering of the blood-sugar for periods of eight hours. Convulsions have been observed as long as 14 hours after the iletin injection. This indicates that iletin is not rapidly destroyed or eliminated by the animal body. No change in the tolerance for iletin was observed after a daily administration of the extract for 83 days.

Curves were presented showing the influence of adrenaline and iletin on the free sugar content of the liver and the blood-sugar. Further curves showed the influence of glucose ingestion on the free sugar of the liver and the relation of free liver sugar to glycogen synthesis with and without iletin. These curves show that the prolonged hyperglycemia after adrenaline injection is due to the fact that the sugar set free in the liver during the first hour diffuses into the blood stream very slowly. It was shown that iletin lowered the free sugar of the liver even during glucose absorption. Ingestion of glucose without iletin increases the free liver sugar. Glycogen synthesis without the administration of iletin has only been observed when the free liver sugar was above its normal value. (Normal value for animals starved for 24 hours is 0.3-0.35 per cent.) Glycogen synthesis under the action of iletin occurs from normal or even lower than normal free sugar levels. This data indicates a disturbance of the ferment equilibrium normally established in the liver and is one of the factors involved in the lowering of the blood-sugar observed after the injection of iletin.

Folin and Wu's method for blood-sugar has been adapted to the determination of galactose glucose mixtures. The method is based on the fact that copper solutions of different alkalinities give different reduction values for galactose, but the same for glucose. It was also found that Bertrand's as well as the Myer-Benedict method give the same reduction values for glucose and galactose. The use of the latter methods in combination with that of Folin and Wu is unsatisfactory due to errors that may be introduced by the discrepancies for glucose values obtained on the same blood.

Energy metabolism studies have been made in 82 periods of observation upon 21 prematurely born infants including those previously reported. 1 Both the oxygen and carbon dioxide, and thus the R. Q., were detemined for each separate period and heat calculated from the oxygen consumption. Temperature within chamber was usually kept between 24° C. and 28° C. Infant was fed (amount of feeding determined by weighing before and after), placed in the chamber and a preliminary period run to allow a lapse of 30 minutes between time of feeding and the beginning of the first period. An average minimal heat production per sq. m. per hour (Lissauer formula) of 25.0 calories on infants over 24 hours of age was obtained as compared with 26.75 calories obtained by Benedict and Talbot 1 for normals from 2 to 8 days of age. These observers found a rather constant heat production from second to seventh day, the average of all basal periods being 27.87 cal. per sq. m. per hr. as compared with 25.72 calories for our prematures.

In comparing maximum with minimum heat values for each infant an average of 44.5 per cent. increase was obtained. Percentages varied from 5.7 with slight restlessness to 88.77 with hard crying 36 minutes (3/4 the period). Comparing the increases obtained in succeeding periods of the same observation in which all the factors, except the activity, were the same, an average of 16.5 per cent. was obtained. Increases varied from 2.5 per cent. with very slight restlessness to 40.5 per cent. with crying 16 minutes of a 33-minute period. Minimal metabolism in children of comparable ages averages 31 cal. per sq. m. per hr. after feedings of from 60 to 84 grams breast milk and 24 cal. after feedings of from 25 to 35 grams.

The agent used is the red dye known as Sudan III. Its advantages are that it may be used in very small amount thus interfering very little, or not at all, with the normal processes in the alimentary tract. The amount used for man is from 50 to 70 milligrams (3/4 to 1 grain). It is soluble only in a fatty medium or in fat solvents. Taken with, or just after a meal, it is not impossible that a slight amount may be combined with the fat of the meal and absorbed to a slight extent; otherwise it is not, apparently, susceptible to the alimentary secretions. The tests have not been interfered with by any absorption that may have occurred. The red color of the dye is essential for the test.

The rapidity with which the dye, mixed with the ingesta, will pass through the intestinal tract will depend upon peristaltic activity, the secretions, the frequency of evacuations and the amount of material ahead of it. In constipation the time will be lengthened and in diarrhea it will be shortened.

The test consists in giving 50-70 miligrams (3/4 to 1 grain) of Sudan III in a gelatin capsule, just before, or after, a meal. One or two grams of the feces from the subsequent evacuations are taken and dried. A record should be kept of the time of administration of the dye and of the evacuations which follow. After drying, the material is extracted with ether. The extract from the first feces passed (before the dye appears) is used as a control. Its color is usually greenish yellow or light brown. Later evacuations, depending on conditions, begin to show a red color.

Azoturia is a disease affecting horses. There are various synonyms of which “Hemoglobinuria” and “Monday Morning Sickness” are common. It is most common in working horses. Animals that are worked hard during the week and allowed to stand in the stable over the week end without exercise and without any decrease in the ration given on working days are most often affected. When exercise begins they are apt to show stiffness and lameness in muscles of the hindquarters. Paralysis of quarters is common and animals get down. They often die as a result of the changes having taken place in the tissues and in spite of any treatment.

The disease has been attributed to several different causes such as myogenic autointoxication, the formation of an hemolytic ferment, a very profuse formation of urea and other extractives, an abnormal amount of dextrose producing a myositis and secondary nephritis. Uremic poisoning is also given as a cause. Much emphasis has been laid upon the abundance of extractives either as a result of tissue changes or as a product of digestion.

Average readings in mgms. per 100 C.C. of blood for normal horses and four cases of azoturia reported in 1921 as well as twelve cases of azoturia from which samples were obtained in 1922-1923 and not as yet reported are appended.

“Posterior paralysis” is one of the terms rather loosely used to designate a trouble in swine which is characterized by stiffness, particularly in the hind legs, inability to rise or stand, the occasional swelling of the joints and fractures of the leg bones. Some preliminary studies relative to this trouble were made the subject of a previous communication. 1

In continuation of these studies, 24 pigs, around 30 pounds in weight, were placed on experiment in the summer of 1922. The characteristic symptoms were produced on a ration of yellow hominy, middlings, and casein, made into a slop with an equal weight of skim milk. A check group receiving added minerals in the form of calcium carbonate and bone meal did not develop the trouble. The alleviation of stiffness and the resumption of growth was brought about in the group receiving no extra minerals either by the addition of the minerals in question, or, by a daily addendum of 10 c.c. of cod liver oil.

In the following winter eight out of thirteen pigs developed the symptoms on the same ration, containing, however, yellow corn meal in place of hominy,—even when calcium carbonate and bone meal mere included. Orange juice was added to the ration of three of the affected pigs, which were obviously near death. There was a marked temporary improvement followed in two weeks by a decline and death. The addition of cod liver oil produced only a slight improvement over a three weeks'period. When the cod liver oil was replaced by 5 per cent. of chopped alfalfa hay in the case of two of these pigs, permanent improvement resulted.

According to the symptoms of apparent paralysis, it was thought that the lesions might be in the lumbar enlargement of the spinal cord. Examination of the cord by the Nissl's method, did not show any tissue changes. The principal and constant lesions in the pigs fed as described in the preceding paragraph were found in the bones and in the kidneys.

On account of recrudescences of this condition in obstetric wards following long periods of quiescence the presumption obtains that the infection has never been completely eradicated from the institution. In order to verify this conjecture, attention has been concentrated on the sporadic case, especially that one occurring after months of freedom from the disease.

Careful study of such a case reveals the interesting fact that the origin of the disease in the infant was apparently by way of the mother's milk, which was heavily infected with pure hemolytic staphylococcus aureus and albus, the former predominating. The lesion on the infant's skin followed forty-eight hours after a rather generalized rash on the mother. It was of note that aside from this macular type of rash the mother was quite normal. The rash itself resembled that often caused by drugs. Cultures of the cervix blood and stool were negative for the infecting organisms, but the urine contained a few colonies. The first and last portions obtained by completely emptying the breasts with a sterile breast pump contained approximately equal numbers of the organisms, which were quite abundant, although no evidence of mastitis was present, either clinical or from examination of the milk.

Although ingesting large numbers of the organism in the milk, the infant's stools were negative for the specific germ, which ruled out the possibility of the initial thigh lesions being the result of direct contamination from the stool. Many lesions appeared on the baby's face and head about seventy-two hours after those on the thighs. All this evidence points to a mild systemic infection via the upper respiratory or intestinal tracts.

Successful feeding experiments of young guinea pigs rather confirms this idea. A certain per cent. of these animals fed with cultures of the organism in milk developed a pneumonia, the lungs yielding pure cultures of the organisms fed. One of the adult pigs that suckled the young pigs also died of this type of pneunmnia.

This study was undertaken with the purpose of getting some information on the factors influencing water retention in the human body particularly with regard to food substances. Experiments were conducted on medical students of the University, during the late winter which in Berkeley is well adapted to experimentation of this kind since there are long periods of almost constant temperature (about 65° F.) and humidity (75 per cent.) and the laboratories are not heated. Diet was controlled and constant. No liquid was taken after 6 P. M. the night before, breakfast consisting of one egg and a slice of toast with no liquid was eaten at 7 :30 A. M. and no luncheon was taken. Urine was voided at 8 A. M. and the first sample collected at 9 A. M. after which one liter of water or the experimental solution was taken and urine samples collected at hourly or half-hourly periods for three or four hours after the diuresis had apparently ceased. Determination of volume, specific gravity, total nitrogen and chlorides were made and the results of several experiments of each kind on eight students were briefly as follows:

Water alone. An increased output of urine was noticeable in 30 minutes, reached a maximum in from one to one and one-half hours and was over in four hours. Chloride excretion was irregular but in general fell as the water excretion increased. Two maxima of nitrogen excretion were noticeable—the first about 10 A. M. and the second about 2 P. M. which is noteworthy in view of the fact that no lunch was taken, indicating apparently the influence of habit.

Water and Cane Sugar. No noteworthy effect on the water output was noted until amounts of sugar of over 150 grams were given. With 180 gms. of sugar the increased urine output began and reached approximately the same maximum at the same time as with water alone but decreased much more rapidly reaching the same low point about an hour sooner but showing a second moderate maximum at from one to two hours after the first low point.

Hammett, Nowrey and Müler, 1 Hammett and Nowrey 2 and Müller and Iszard 3 have reported that germanium dioxide stimulates red cell formation. These authors experimented almost exclusively with normal animals, chiefly rats and guinea pigs. In view of the suggested therapeutic use of germanium dioxide in the treatment of anemia, we investigated the effect of this compound in a number of animals which had been previously made anemic with methyl-phenyl-hydrazine, symmetrical di-isopropyl-hydrazine 2-2'azobis-propane and other derivatives of hydrazine.

In the case of dogs, the usual course of recovery from anemia is irregular and varies considerably in different animals. To some extent this is due to temperamental differences in individual dogs and to the inability of certain of these animals to maintain themselves in nitrogen equilibrium. For these reasons, we experienced some difficulty in controlling adequately all of our experiments with germanium.

However, the evidence adduced thus far shows that the erythropoietic action of germanium dioxide, whenever such effect occurs, is transitory. The administration of germanium dioxide over a prolonged period does not appear to alter the degree of red cell regeneration occurring without the use of this compound. Not infrequently, a decrease in the red cell count may follow an injection of germanium dioxide. Where a temporary erythrocythemia was observed in our animals, there was no corresponding increase in the percentage of hemoglobin.

The appended chart is illustrative of the effect produced by germanium dioxide, administered over a period of one month to a mildly anemic puppy. In similar experiments in which no germanium dioxide was administered, red cell regeneration occuxed as rapidly, and in some instances even more rapidly than in this animal.

The relative ease with which alterations in blood volume may be produced in rabbits, renders these animals unsuitable for the study of certain phases of experimental anemia. On the other hand, they are more easily maintained in proper nutritive condition and therefore the course of recovery from anemia is more uniform in these animals than in the case of dogs.

In a number of experiments, phenylhydrazine, emulsified in olive oil, was administered subcutaneously in doses sufficient to destroy about two-thirds of the red corpuscles within a period of a week. Germanium dioxide was administered to some of these animals before this interval had elapsed to determine whether any effect was produced in retarding the rate of hemolysis. No such effect was observed. The animals receiving the germanium became as anemic as those not receiving it.

Subsequently, a number of the rabbits received repeated injections of germanium dioxide. The rate of increase of red corpuscles in these animals was compared with that in animals which were not receiving germanium dioxide. It appears from our results, that germanium dioxide in moderate amounts does not augment the rate of red cell regeneration in rabbits recovering from phenylhydrazine poisoning.

The germanium dioxide used in this work was supplied to us by Professor L. M. Dennis of Cornell University, to whom we are indebted.

The recent publication of data on the blood of pigeons in polyneuritis and stavation 1 suggested the desirability of publishing the data obtained on rats under similar conditions. The preliminary report 2 on the influence of fasting and of vitamin B starvation has been amplified and data have also been obtained on fasting rats to whom water (10 to 21 c.c. daily) was administered by stomach tube during the last days of the fast. Total solids were determined by an adaptation of Peters'method 3 and the other constituents by the methods of Folin and Wu. 4

From the summary given in the table it is apparent that the non-protein nitrogen of the blood of fasting rats is 30-40 per cent higher than that of normal animals, the increase being practically all in the urea fraction. Total dry matter, creatinine and creatine are slightly increased. Fasting rats that are given water per os during the last few days of their fast show normal values for non-protein nitrogen and solids. The blood of rats deprived of vitamin B also gives figures that are normal except that creatinine is at the fasting level and creatine slightly higher than the fasting figure; these differences have little if any significance in the present state of uncertainty with regard to blood creatine and creatinine determinations. Deprivation of vitamin B in contrast to fasting is not accompanied by an accumulation of nitrogenous end-products in the blood; whatever the ultimate cause of this increase may be in fasting, it is prevented by administration of water.

Continued subcutaneous administration of thyroxin is known to produce hyperglycemia. In normal sheep the blood-sugar is raised from about 70 mg. to over 80 mg.

A single subcutaneous administration of thyroxin has only a slight effect on the blood sugar.

A single subcutaneous administration of insulin (10 units) produced no demonstrable effect within two hours after injection.

Intravenous administration of varying amounts of insulin (5 to 15 units) produced prompt hypoglycemia in normal sheep, the sugar decreasing from about 70 to about 40 mgs. per 100 c.c. of blood within about 30 minutes. (No unfavorable symptoms were observed). The drop is regular, and when plotted is represented by a practically straight line. The slope of this portion of the curve is constant for the same sheep upon administration of varying amounts of insulin.

The portion of the curve representing recovery from hypoglycemia show characteristic differences : When 5 units of insulin were administered, the recovery to normal sugar values began immediately after the low values had been reached. With 10 to 15 units the minimum sugar values obtained were only slightly lower than for the smaller doses, although analyses were taken at sufficiently close intervals (10 to 20 minutes) to enable a fair approximation to the minimum. The larger doses, however, showed an extensive flat portion in the curves before recovery began. The results with five units were reproducible with the same sheep and similar in different sheep.

Five units were therefore selected as the standard dose of insulin to be tested against varying amounts of thyroxin under different conditions. Administered alone, the effects produced are represented by curves A and D. Curve G shows the effect of a preliminary administration of thyroxin (1 mg. intravenously). Curve C shows the effect of a similar injection of thyroxin after the recovery from insulin had begun. The simultaneous administration of thyroxin (1.5 mg.) and insulin produced a typical divergence of the ascending arm of curve E from that of curve D (insulin alone)

The present work constitutes part of a study on the pharmacology of phloridzin and its derivatives. It is based on some earlier observations 1 concerning the influence of phloridzin on the formation of glycogen in the liver under certain experimental conditions. The results of the experiments referred to seemed to indicate that phloridzin possesses a two-fold action: namely, that of causing glycosuria, and that of stimulating the formation of glycogen in the liver. These experiments showed further that animals which had been rendered diabetic by pancreatectomy, and in which the glycosuria was interfered with by removal of the kidneys, the blood-sugar content was generally lower in those which had received phloridzin, than in those which had not received it.

It therefore seemed reasonable to suppose that if a means could be found by which phloridzin could be deprived of its glycosuric action, its effect on the formation of glycogen and the reduction of the blood-sugar would be made manifest.

Notwithstanding the importance of phloridzin in the study of certain problems of metabolism, little information is available concerning its derivatives.

In view of the object of our investigation the work is comprised of two phases: first. an examination of the pharmacological effects of known derivatives of phloridzin and second, the preparation and study of new compounds of this substance.

The purpose of the present cunlmunication is to state our problem and to record the new cotnyounds which have been prepared and identified. The following is a list of the substances:

(1.) Tetrabenzoylphloridzin: This was obtained by the actin11 of benzoyl chloride on phloritlzin in pyritiin. The substance is white, and of a starchy character.

A study of the nitrogen content of various bacterial groups and subgroups was undertaken by the writers some months ago. The object was to determine, if possible, the relationship of the nitrogen content which might occur between members of a single group, and also that between the subgroups of a species where such are known to exist. While progress has been made in this study, much remains to be done, and we are continuing the investigation. A preliminary report of the results of our analyses made upon the pneumococcus, including the four serological subgroups, may prove to be of interest to others. We therefore submit some of the data, although the study is not completed.

Recognizing the many known factors which may enter and alter the results in a study of the protein content of bacteria, we have attempted to control these factors in so far as possible by keeping them constant throughout. In this way, to a certain degree, a comparison is possible between the derived chemical data.

The organisms used in this study were for the most part recovered from sputum, blood, etc., of patients in St. Luke's Hospital, although we are indebted to Dr. Avery, of the Rockefeller Institute, and to Dr. Wadsworth, of the New York State Health Department Laboratories, for a few strains of the Type II. In each instance the identity of the organism was established by its characteristic morphology, staining properties, bile solubility, and reaction to specific immune serum. Ten strains of each of the four chief groups were cultured in flasks containg about 75 c.c. of a beef infusion broth containing one per cent. of dextrose, and having an average reaction of P_H 7.4 to 7.6. Usually sufficient growth for the purpose of the study was produced within 21 hours, when incubated at 36° or 37° C.

Rabbits inoculated upon the cornea of the right eye with our virus of Herpes febrilis show constantly on about the fourth or fifth day a turning of the head toward the right side; during the succeeding days the neck is twisted strongly toward the right and held rigidly in that position. This picture was described by Doerr and Schnabel and later by Levaditi and was attributed by Doerr and Schnabel to an encephalitis, the virus probably passing from the cornea to the brain by way of the blood stream. Levaditi, on the other hand, claimed to have proved that the virus passed from the cornea along the optic nerve to the brain. Our observations make it evident to us that this symptom is due to a lesion on the right side of the pons and medulla along the distribution of sensory fibers of the fifth cranial nerve, which involves the roots of the spinal accessory nerve supplying muscles of the neck on the same side, and that this lesion is produced by the virus of herpes entering the brain from the cornea by way of the sensory portion of the fifth cranial nerve.

In such a lesion, cells of glial and ganglion types are found to contain intra-nuclear inclusion-bodies similar to those described by Lipschütz in the cornea of rabbits inoculated with virus of herpes febrilis, and which we regard as pathognomonic of herpetic lesions in general.

The following additional experiments may be cited as confirming the passage of this virus along sensory nerves:

(1). A rabbit inoculated into the skin of the right hind leg developed on the 10th day impairment in the use of this leg, progressing during the next few days to practically complete disability of this extremity. On the 11th day impairment in function of the left leg followed. At autopsy the lumbar region of the cord showed gross hemorrhages limited to the right dorsal surface of the lumbar portion of the cord along the line of entrance of dorsal root fibers