Experimental Biology and Medicine

In several communications we have reported that the injection of solutions of strychnin, morphin or acid fuchsin in cardiectomized frogs is liable to bring on convulsions of these animals. The lymph hearts continue to beat for a while after the cardiectomy. But since the lymph hearts assist the circulation only by emptying their contents into veins, it seemed to be evident that the removal of the blood heart eliminates also the circulatory function of the lymph hearts. I have therefore assumed that the above mentioned alkaloids reach the central nervous system by way of the lymph spaces, which are connected throughout the body, and which are capable of serving as a path for distribution by means of a peripheral mechanism. In a recent paper by Abel (Jour. of Pharmacology, III, 581, 1912) in which our facts were confirmed and in which it was admitted that the activity of the posterior lymph hearts can not come into consideration, the statement was made that “the appearance of convulsions in the experiments of Meltzer and his pupils with acid fuchsin, morphin and strychnin depends entirely on the integrity of the anterior lymph hearts.” This statement is supported by a report of experiments in which, after destruction of the anterior lymph hearts, in addition to cardiectomy the alkaloids under discussion did not bring on any convulsions. I shall not enter here into a discussion of the entire subject. I merely wish to let you witness some indisputable facts. You see here a series of frogs from whom the thoraco-abdominal viscera have been removed, and in addition, the lymph hearts were destroyed by cauterization. All these animals were injected about an hour ago with strychnin; the injections were made in some animals into the dorsal lymph sac and in others into the femoral sacs of both thighs.

The production of experimental pneumonia in dogs by means of intra-bronchial insufflation of bacterial cultures has now been carried out in several series of investigations. In the first series Lamar and Meltzer produced lobar pneumonia by insufflation of cultures of a virulent pneumococcus. In a second series we produced lobular pneumonia by insufflation of a virulent streptococcus and of the influenza bacillus. Besides the differences in the gross appearances of the lesions both pneumonias offered the following distinguishing points: In the lobar pneumonia of the virulent pneumococcus there was a mortality of about 16 per cent.; even in the non-fatal cases there was bacteremia present in the first twenty-four hours; the exudate was rich in fibrin and the framework of the lungs was invariably free from leucocytic infiltration. In the cases of broncho-pneumonia produced by the streptococcus or by the influenza bacillus there was no mortality; at no time was there a bacteremia; the exudate contained very little fibrin, but the framework of the lungs was invariably and often very intensely infiltrated with leucocytes.

In a third series of experiments, which has not yet been published, we studied the effects of insufflation of a non-virulent pneumococcus. This organism also caused the development of a lesion which macroscopically resembled the lesion of lobar pneumonia produced by the virulent pneumococcus. However, there was no mortality, no bacteremia, and an exudate comparatively poor in fibrin; the framework of the lung was here again practically free from leucocytic invasion.

The results obtained with the non-virulent pneumococcus led us to a study of the intra-bronchial insufflation of such a nonpathogenic organism as the B. megatherium. Ten experiments were made. The insufflation brought on a definite pneumonic lesion in every case.

Experimental pneumonia produced by intra-bronchial insufflation of virulent pneumococci may develop quite rapidly. Thus in the experiments of Lamar and Meltzer it was observed that in one instance seven hours after the injection, nearly complete consolidation affecting the greater part of one lobe was already present. We therefore now made several experiments in which the lungs were under direct observation continuously for several hours after the insufflation of the culture. The procedure has been as follows. The dog was anesthetized by the cone method, and a tube introduced into a bronchus as deep as it could be pushed; the pneumococcus culture was then injected. Immediately after, the tube was withdrawn, so that its lower end was just above the bifurcation, and the arrangement made for continuous intratracheal insufflation. The dog received about “half ether” and was continually under complete anesthesia, The thorax was now widely opened transversely and about two thirds of three or four of the lower ribs on the right side removed; the right lower lobe, which is usually the seat of the inflammation, was now exposed to full view. We shall not enter here upon details. It may suffice to state that we were able to watch the successive stages from the earliest signs of engorgement to complete hepatization. We were also able to establish changes in the auscultation phenomena in the consolidated parts, in some instances even clearly suggesting tubular breathing. We intended to demonstrate such experiments, and we prepared several animals for this purpose. The experiments were successful indeed, but we did not take the lateness of the hour of the meeting sufficiently into account. We prepared the experiments too early and all the animals died before the meeting began. The acute infection, the great loss of heat, the severe operation, and the long-lasting anesthesia are severe factors to contend with.

In studying the toxic effects of chloroform and ether, when administered by the method of intratracheal insufflation, we observed various differences in the course of the intoxication brought on by excessive doses of these anesthetics. We shall discuss here, however, only the differences in the toxic action of these drugs upon the functions of respiration and blood pressure. At the outset we have to point out, that when administering the anesthetics by the insufflation method one of their dangerous effects is here eliminated; it is the danger which is bound to result from a partial or complete paralysis of the respiratory function. Under the method of insufflation, life remains safe even when the animal is completely curarized. Observations may be carried on, therefore, even after spontaneous respiration is completely abolished. On the other hand, intratracheal insufflation carried on with ordinary, permissible air pressure, does not cause apnea, that is, the individual continues to carry on its own spontaneous respirations, which on tracings are easily distinguishable from the infrequent partial interruptions of the continuous insufflation of air.

The observations which we wish to report here briefly are as follows. When using ether, a certain dose may be administered which is amply sufficient to keep the animal completely anesthetized, while respiration and blood pressure may remain practically unimpaired for several hours. When this anesthetic dose is exceeded, the first striking effect is upon the spontaneous respiration which may be rapidly abolished. For instance, if complete anesthesia has been accomplished by a dose which we term “1/2 ether” or “3/4 ether” and now “full ether” is turned on, the respiration may stop completely within one to five minutes. At this stage blood pressure is not impaired.

In a previous communication 1 it was shown that a comparatively strong bactericidal substance can be extracted from horse leucocytes. This substance loses its bactericidal power, in whole or in part, if it is mixed with homologous or foreign sera, with pathological exudates, with cerebro-spinal fluid, with the products of aseptic tissue autolysis, or with most of the products obtained by the bacterial decomposition of tissues. It also loses its bactericidal power if it is mixed with the products obtained by the prolonged autolysis of leucocytes themselves.

With sera that are in themselves bactericidal, not only is there a loss of the bactericidal power of the leucocytic extract in such mixtures, but there is also a destruction or inhibition of the bactericidal power of the serum itself. This gives the phenomenon of two bactericidal substances, an active serum and an active leucocytic extract, added to each other, producing a non-bactericidal substance, a good culture medium for bacteria.

An analysis of the antibactericidal action of serum shows that it is due to the combined effects of three factors: (1) the antibactericidal power of the serum colloids, (2) the antibactericidal power of sodium chloride and the other neutral diffusible serum components, and (3) the antibactericidal action of the diffusible serum alkalies.

Alkalies are very strongly antagonistic to the leucocytic bacteriolysin. The addition of 11200per cent. NaOH to leucocytic extract is usually sufficient to completely inhibit its bactericidal action. Acids, on the other hand, apparently have little or no antibactericidal effect.

This antibactericidal power of serum and tissue fluids can not be overcome by increasing the amount of leucocytic extract in the mixtures, except when the serum and tissue fluids are tested in minute quantities or greatly diluted.

It was pointed out by Koch, and has since been confirmed by others, that an animal suffering from a chronic local tuberculous lesion is more resistant than the normal to inoculation with tubercle bacilli.

This heightened resistance has been studied by numerous workers. Calmette, for example, found that tubercle bacilli, injected into the subcutaneous tissues of tuberculous cattle, soon showed marked involution and degenerating forms, but that the bacilli did not completely disappear from the site of the inoculation and the neighboring lymph glands by the end of 120 days. A rapid lysis or destruction of the bacilli was not observed by these workers.

Recently Beycke and Much 1 and others have applied the Pfeiffer technique to a study of this phenomenon. They have found that tubercle bacilli, injected into the peritoneal cavities of tuberculous guinea pigs, undergo rapid degeneration and a rapid decrease in number.

They interpret these changes as a rapid lysis of bacilli. If this interpretation is correct, it will necessitate a revision of current ideas concerning the resisting power of the tubercle bacillus.

We have repeated and extended their work. We have found that tubercle bacilli, injected into the peritoneal cavities of tuberculous guinea-pigs, will occasionally give the degeneration forms and the non-staining forms they describe; and that, under certain conditions, there may be a complete disappearance of the bacilli from the peritoneal fluids within as short a period of time as three hours.

Whether this disappearance is due to an actual lysis of the tubercle bacilli, or to other causes, we have not yet determined. As evidence in favor of lysis we have the observation that all of the normal control guinea pigs injected intraperitoneally with the test suspensions of tubercle bacilli have died from a fulminating type of visceral tuberculosis, within a period of from three to four weeks, while most of the tuberculous guinea pigs, receiving the same test doses, have survived for a longer period of time.

In order to clear the way for a study of the transfer of protein in the involuting tail of the tadpole, it was necessary to reëxamine the rôle which has been ascribed by Metschnikoff, Barfurth, Bataillon, Mercier and many others to the leucocytes in producing the phenomenon. Looss believes on histological ground, that if leucocytes play any part in the process of tissue atrophy and absorption, they play a minor and secondary one rather than a primary rôle. The process of atrophy begins prior to the invasion of the leucocytes into the muscle of the tail and this has been described by Mercier and others, who hold that phagocytosis is the principal factor. Moreover, it has been described for other involuting organs, such as in the metamorphosis of insects, the absorption of the gills of amphibia, etc. It has been suggested for the involution of the mammalian uterus, likewise, so that it may be said that investigators are in accord in observing a dissolution of the muscles and other tissues in atrophying organs prior to the advent of leucocytes.

It is to be expected that if phagocytosis plays any important rôle in the inception of the process of absorption of tissue in the larva of the frog, the blood would show an increase in the total number of leucocytes during the stages of metamorphosis and moreover there would be an increase in polymorphonuclear leucocytes during these stages, to compensate for the drainage of these cells into the muscles. In order to examine this point, smears were made of the blood from the larvz of the bull-frog, Rana catesbiana and from the western pickerel frog, Rana areolata. Thirty specimens were used, the blood being permitted to flow into a capillary tube from a lesion in the heart and then blown upon a slide, dried in the air and stained with Wright's stain.

In a previous communication (Journal of Exper. Medicine, XIV, 274, 1911), I stated that after intravenous injections of large amounts of dextrose the intestine contained on the average 1.2 per cent. of the injected amount of destrose and in nephrectomized animals the dextrose content of the intestine reached the average of 2.2 per cent. The question arose then whether in the presence of a strong hyperglycemia, and especially after double nephrectomy, all mucous membranes are slightly permeable to this substance. The presence of measurable quantities of dextrose in the gastro-intestinal canal is no evidence for the permeability of the mucosa as such, since the mucous membrane of this organ is studded with numerous glandular structures which may readily be instrumental in the removal from the blood of substances present there in excess. In the experiments I wish to communicate here, I selected, therefore, the bladder for the study of this problem since the mucous lining of this organ is practically the only membrane which possesses no glands. The rabbit was the experimental animal used.

After performing a double nephrectomy and tying a cannula in the bladder, seven to ten grams of dextrose per kilogram were injected intravenously in from twenty to fifty minutes, washing the bladder several times during and after the injection. The washings were then analyzed. The results are briefly as follows. In nine experiments either no sugar or else indeterminable amounts were found, and in four experiments only six to eleven milligrams, i. e., 0.05 to 0.08 per cent. of the amount injected. In addition, in some of the experiments uranine, the sodium salt of fluorescei'n, was dissolved in the injection fluid. In the experiments in which twenty milligrams were introduced, no trace of uranine, a very diffusible dye, could be detected in the washings of the bladder.

It has been stated by Rosenfeld that the administration of phlorhizin to dogs with Eck fistula does not induce glucosuria. This observation would have so much bearing upon our ideas of the mode of action of phlorhizin that the subject required confirmatory evidence.

One gram of phlorhizin in olive oil emulsion was given to a dog on which an Eck fistula had been done. Glucose was found in the urine in considerable quantity for nine days subsequent to the phlorhizin adminstration.

The influence of pancreatic transplants in the spleen on the prevention of diabetes was studied. The spleen was selected because it has been shown that transplanted bits of tissue are especially well nourished in the pulp of this organ. In nine animals, bits of the pancreas of various size were buried in the pulp of the spleen. The method used by Payr in transplanting thyroid tissue was followed. In four animals (2 dogs and 2 cats) pancreatic cells were demonstrable in the spleen. The interval between the operation and the death of the animal varied from 18 hours to 13 days. In five animals (all dogs) no pancreatic remains were found. Autolysis was rapid. In one experiment all the pancreatic cells had disappeared 21 hours after the transplantation. The bit of tissue found 13 days after the transplant measured only 1 mm. by 0.1 mm. It consisted of normal appearing acini surrounded by connective tissue. No islands of Langerhans were demonstrable.

In one dog a large pancreatic graft was placed in the spleen with the blood supply preserved by means of a mesenteric stalk. Three weeks later the original blood supply was cut off and all the pancreatic tissue except the graft extirpated. Diabetes did not develop, but the tolerance for glucose fell within a few weeks to a low point. At the autopsy 187 days after the second operation a large abscess was found in the lower part of the spleen. Projecting into the spleen from the wall of the abscess was a coneshaped mass of fibrous tissue. In this were the remains of the pancreatic transplant measuring less than I cm. in length. The pancreatic tissue consisted of acini separated by connective tissue.

In a series of experiments, which were published with Kessel and Mason, 1 on the excised perfused hearts of dogs, we were able to show that “excision of the sinus node results in an immedate cessation of the contractions of the whole heart,” that “after excision of the sinus node and the ensuing stoppage, the rate of the whole heart falls and does not again reach the original rate,” and that “the function of the secondary pacemaker devolves on no special portion of the heart.” The results of experiments performed by others and differing from ours were mentioned at that time. I have continued the experiments on dogs, anesthetized with ether, the hearts of which were left in situ and normally nourished. After sufficient controls were registered electrocardiographically, the sinus node was clamped with a specially constructed T clamp, 5 cm. long and 1 mm. wide. In the perfused hearts the rate fell 10 to 70 beats, usually 30 to 50 beats, after the node was excised. In the present series the fall in rate was 10 to 90, usually about 50. Stoppage, which is a usual phenomenon in excised hearts, occurred also in these, but very rarely. The site of the secondary pacemaker varied in these, as it did in the perfused hearts. Photographs of three hearts showing the areas of the auricular surfaces which had been included in the jaws of the clamp, and a series of curves from each of the corresponding experiments and of one in addition were demonstrated. The relation of the disturbed function of the heart to the area clamped will be investigated histologically and reported in detail later. The curves indicate that in each of these cases, there was a dislocation of the pacemaking function from the site responsible for it, and that, after clamping had taken place, this function devolved upon the ventricles, the junctional tissues or another portion of the auricle.

In a recent paper (Journ. Biol. Chem., Sept., 1912) we have shown by direct determinations (nitrous acid method) of the amino acid nitrogen of the blood, that during the digestion of protein amino acids pass from the alimentary canal into the blood. The amino acid nitrogen reaches only a relatively low concentration in the blood (3-5 mg. per 100 c.c. after 24 hours fast, 10-12 mg. during digestion), because, as was found after intravenous injection of alanine, the amino acids are either destroyed or removed with great rapidity by the tissues. The object of the present work was to obtain direct proof, by analysis of the tissues themselves with the nitrous acid method, as to whether the amino acids disappear from the blood as the result of destruction, of synthesis to more complex compounds, or of mere concentration in the tissues. The last explanation is the correct one. The tissues (muscle, liver, kidney, spleen, pancreas) contain, stated very roughly, about ten times the amount of amino acid nitrogen, per 100 grams, that is found in the blood. The amount varies in the different tissues, the brain being especially low, but about the same order of magnitude in this respect appears to be normally maintained among the different organs. Amino acids injected into the blood, or absorbed during digestion, are immediately taken up by the tissues; but the removal, although rapid, is never complete, equilibrium being reached and maintained between blood and tissues respectively. The means by which the amino acids are held in the tissues appear to be physical rather than chemical; for the acids can be extracted by even cold water or dilute alcohol. The fact that the amino acids are many times more concentrated in the tissues than in the blood excludes osmosis as an explanation.

It has been shown by F. Voit and P. Mayer that glycogen parenterally introduced into the normal body of man or lower animal, is utilized, there being no elimination of glycogen, dextrin or glucose in the urine. Mendel repeated this work with rabbits, cats and dogs and found, on the contrary, that 5 per cent. to 18 per cent. of the glycogen introduced was eliminated as a dextrin over a period of two days.

In human and experimental pancreas diabetes the administration of glycogen gave the following results:

I. Human diabetes, first trial, E. P., age 16 yrs., wgt. I 10 lbs.

Glycogen subcutaneously, 28.0 gm.

Glycogen in urine as gluc. 28.0 gm.

Elimination time, under 24 hours.

II. Human diabetes, second trial (same case).

Glycogen subcutaneously, 21.5 gm.

Glycogen in urine as gluc. 21.5 gm.

Elimination time under 24 hours.

III. Complete pancreas extirpation; bitch, 8.5 kilos.

Glycogen intraperitoneally, 9.5 gm.

Glycogen recovered in urine as glucose 6.0 gm. = 66 percent.

Glycogen recovered in urine as dextrin 3.5 gm. = 34 percent.

Elimination time under 24 hours.

IV. Complete pancreas extirpation; bitch, 14.5 kilos.

Glycogen intraperitoneally, 25 gm.

Glycogen recovered in urine as glucose, 22.0 gm. = 88 percent.

Glycogen recovered in urine as dextrin, 3.0 gm. = 12 percent.

Elimination time under 22 hours.

These observations show definitely that glycogen is not a utilizable form of carbohydrate for the diabetic organism and they indicate that there is a more rapid conversion of glycogen to glucose in the diabetic than in the normal body.

Morphin given in sufficient dose to normal frogs causes tetanus which does not come on, however, until several hours or even days after the injection.

I have found that decerebration hastens the onset of tetanus and also causes a marked reduction in the amount of morphin required to induce tetanus.

The smallest dose with which tetanus can be induced regularly in normal frogs at room temperature is 1/3 of a milligram per gm. (10 milligrams for a 30 gm. frog). This tetanus comes on in about 24 hours.

In decerebrated frogs, at room temperature, tetanus comes on after such a dose in from 1/2 to 6 hours, and may be induced with certainty after 6 to 24 hours by a dose of 1/10 milligram per gram (3 mg. for a 30 gram frog).

When frogs are kept cold tetanus can be induced by much smaller doses, as we have stated in an earlier paper. Thus, intact frogs kept in the cold show tetanus after doses of 1/30 mg. per gm. (1 milligram for a 30 gram frog). Tetanus comes on after such a dose in from 18 to 24 hours. Decerebrated frogs show tetanus after such a dose in from 4 to 12 hours, and it may be induced with certainty by doses of 1/300 milligram per gm. (1/10 mg. for a 30 gm. frog) after an interval of 1 to 3 days.

Frogs with the entire brain including the medulla destroyed, do not respond as well as frogs with the entire brain except the medulla destroyed.

Since making our previous report on this subject, electrocardiograms have been obtained from the same patient in which occasionally the small downward deflection between R and T fails. The form of the ventricular electrogram in these instances is precisely similar to the others with the sole exception of the absence of the small downward deflection. This we regard as confirmatory evidence of the correctness of the view that the downward deflection is caused by the action of the auricles and that the usual curve is not simply an unusual ventricular complex. In the occasional instances cited conduction seems to have failed.

When this case first came under observation it seemed desirable to attempt to reproduce the condition experimentally. We have performed up to the present time seven experiments with one positive result. Rothberger and Winterberg have published curves showing transient reversal resulting from simultaneous depression of the sinus-region of the right auricle with cold and stimulation of the left ganglion stellatum. The apparent permanence of the condition in our patient leads us to think rather of an anatomical lesion than of nervous influences as the cause of the condition. Our experiments have consisted in depression (with anode) crushing, and excision of the region of the sinus node. In some instances nearly the entire anterior wall of the right auricle including a large part of the walls of the venæ cavæ where they fuse with the auricle, have been excised.

Slowing of the heart beat was the usual result and the auricles and ventricles generally beat nearly or quite simultaneously. In the single positive experiment auricular fibrillation resulted from the application of a clamp to the upper part of the sinus region where the greater bulk of the node is usually found.

In two earlier papers 1 I presented evidence in support of my interpretation of intercalated discs as irreversible contraction bands. In a more recent paper the idea was tested by appearances in a natural experiment, namely, extremely hypertrophied heart muscle. 2 The conditions here obtaining were in perfect accord with, and confirmed the plausibility of, my previous interpretation. Since then I have had opportunity to study lesser degrees of hypertrophy, as well as an excellent specimen of atrophy 3 (weight of heart 180 grams). It is the purpose of this note to complete my report of observations on intercalated discs by a record of my findings in atrophied heart muscle, and to reëmphasize the point that all the evidence, including ontogenetic, comparative anatomical and pathological (experimental—including the phenomena of fragmentation and segmentation) data points to the same conclusion, namely, that the so-called “discs” are aggregations of irreversible contraction foci on the myofibrillæ in the form of bands variously modified by a variety of normal mechanical and pathological, both chemical and mechanical, factors.

The first step, and central fact, in the chain of observations leading to the above conclusion respecting the origin and nature of the discs is the close similarity, amounting practically to an identity, between the normal contraction band in contracted muscle fibers and the simplest type of disc. This correspondence of appearance is very striking in the humming-bird's heart. In the relaxed fiber the Z-line is very conspicuous but delicated The Q-disc is wide and pale. In the contracted fibers the appearance is one of an alternation of robust dark (Q) and light (J)bands of approximately equal thickness. These are the same conditions that obtain in striped muscle generally in relaxed and contracted states respectively.

Under the foregoing title the results of an elaborate investigation will shortly appear as Publication No. 173 of the Carnegie Institution of Washington. This research is supplementary and complementary to, and in support of, Publication No. 116 entitled, “The Differentiation and Specificity of Corresponding Proteins and Other Vital Substances in Relation to Biological Classification and Organic Evolution: The Crystallography of Hemoglobins.” A preliminary review of the latter will be found in the PROCEEDINGS of the Society, 1908, V, 66-68.

Studies of the histological, physical, physico-chemical and chemical properties of over 300 starches from various plant-sources were made. Among the most important conclusions reached are the following:

1. Starch is not a unit substance but exists in a vast number of stereoisomeric forms. Starch from any given plant and of every mature starch grain is a mixture of different forms of starch-substance.

2. Starches from different plants exhibit constant and specific characters in relation to genera, species and varieties, by which the plant can be identified.

3. The stereochemic peculiarities of starches and other complex organic metabolites constitute a strictly scientific basis for the classification of plants and animals.

The raccoon is a very intelligent animal and, as one would expect, relatively to the size of its body, it possesses a large brain with a highly convoluted cerebral cortex. Little appears to be known, however, about its cortical topography or the fiber tracts of its central nervous system.

The present note refers to the experimental localization of the motor areas and the subsequent tracing of the pyramid tract by the method of ablation and secondary degeneration. Five full-grown specimens were obtained. The cerebral cortex was exposed on the left side, under ether anesthesia, the motor areas were localized both by the bipolar and unipolar methods and then removed. The wound was closed and the animal allowed to live for about two weeks when the left hemisphere was exposed and explored in the same way before the animal was finally dispatched by an overdose of ether. The brain and spinal cord were removed and treated by the Marchi method.

The region from which muscular responses were obtained is relatively large and well defined. It occupies the whole free surface of what may be termed the post-cruciate convolution, extending from the mesial border of the hemisphere to a little way beyond the lateral extremity of the cruciate sulcus. Unless when the current was comparatively strong, no movements were obtained from the cortex in front of the cruciate sulcus. From the mesial border lateralwards the order of the responsive areas was as follows: anus, tail, hind limb and digits, body, fore-limb and digits, head and eyes, and face, mouth and tongue, the last curving forwards around the lateral extremity of the sulcus. The movements were readily elicited and the areas well defined but there was always some overlapping at the margins.

We used the dried powdered tonsil of the calf. When a filtered infusion of two grains of the powdered tonsil was injected into the cat by the jugular vein in divided doses, it produced a great fall of blood-pressure, lasting about a minute, followed by a rise above normal, with a much slower and stronger heart beat. Increase of this dose suddenly arrested the heart. In the same animal it was also a diuretic, increasing the flow of urine twenty times the original amount.

Rosenbloom and Gies 1 have shown that many ether soluble substances diffuse from ether through rubber membrane into ether, but oddly various phospholipins do not possess this property.

Together with Boas, 2 I was able to show that various cholesterolesters diffuse from ether through rubber membrane into ether. This is very interesting on account of the high molecular weight of these esters. Cholesterol-stearate with a molecular weight of 652.51 diffuses very readily.

For some time I have been trying to find an ether soluble substance of higher molecular weight than cholesterol-stearate and which could be easily detected in the diffusate. The free dye-acid of iodoeosin fulfilled these requirements. This free dye-acid has been employed by Professor Ehrlich as a very delicate reagent for free alkali in the erythrocytes of man. 3

Iodo-eosin is the potassium salt of tetraiodo-fluorescei'n with the following formula,

As a salt it dissolves in dilute alkalies with red color, but is insoluble in ether or any other organic solvent. The free dye-acid, however, is obtained as a yellow precipitate from the alkaline solution of iodo-eosin by adding hydrochloric acid in excess, and it dissolves readily in ether, or in any other organic solvent, but is insoluble in water. This free dye-acid has the following formula, with a molecular weight of 836.

The dye-acid of iodo-eosin may be made by dissolving ten grammes of iodo-eosin (commercial dye) in one per cent. potassium hydroxide and then adding hydrochloric acid in excess. The dyeacid is precipitated at once, it can then be filtered off and the precipitate washed with hot water till the washings are acid-free. The precipitate after drying is easily soluble in ether, forming a beautiful yellow-colored solution.

Of the twenty-eight guinea pigs treated with sublethal doses of carbonate of lead seven (25 per cent.) showed a peculiar condition to which so far attention does not seem to have been directed. There has developed a hemorrhagic, sero-fibrinous inflammation of the pericardium, of the peritoneum in the upper part of the peritoneal cavity and occasionally also of the pleuræ. In the pericardium the lesion commences with a hemorrhagic exudate followed by the formation of fibrinous deposits especially on the parietal layer and ending with organization with marked thickening and occasionally adhesions. In the pleuræ one finds as a rule a simple serous exudate without deposits but occasionally heavy deposits of fibrin also. In the peritoneum also the condition commences with fibrinous deposits and eventually leads to marked peritoneal thickening more especially in the upper part of the peritoneal cavity. The capsules of liver and spleen are much thickened and the contraction of the thickened capsule often causes marked deformity of the liver. The lesions produced in this way are closely analogous to Curschmann's “Zuckergussleber.” The spleen is always more or less enlarged, but this is due to the excessive destruction of erythrocytes by the lead and is found in all animals. 1

It was only natural to suspect the presence of a bacterial infection of the serous membranes in these cases. A very careful bacteriological examination was made in several of the cases, but it was entirely negative.

Whether the hemorrhagic exudation which ushers in the process is due to the lead anemia appears to me a question well worth debating. Smears of the exudate show few leucocytes and often many nucleated red blood corpuscles, even megaloblasts.

In one case (46) adhesions had formed in the upper part of the peritoneal cavity with partial strangulation of duodenum and intestine and extreme dilatation of the stomach.

During the last years I have been repeating the experiments of Charcot and Gombault 1 on chronic lead poisoning in guinea pigs. Twenty-eight guinea pigs were given sublethal doses of carbonate of lead for periods ranging from one month to three years and ten months. Fourteen of these experiments lasted over one year.

The lesions in the kidneys were much less striking than one would expect from the report of Charcot and Gombault. In all cases there was a limited necrosis and desquamation of the epithelium with marked evidences of regeneration especially in the ascending limbs of the loops of Henle. In some cases the epithelium in places was heavily pigmented. Occasionally there were seen a few glomeruli with slightly thickened capsules. There were only two of the twenty-eight experiments in which more advanced lesions were discovered in the kidneys. In one case (69) in which the guinea pig had received over thirty grams of carbonate of lead in three years the kidneys were actually granular and the cortex distinctly narrow. The lesions in this case consisted in collapse of tubules over large areas with marked development of fibrous tissue between them. There were many casts. The glomeruli showed marked fibrous thickening of the capsules and cystic dilatation. The other guinea pig (51) which received about four grams of carbonate of lead in twenty months showed similar conditions but they were not so well marked. Calcareous infarcts were not found in any of the kidneys. No vascular lesions were ever found in the aorta or in the branches of the renal arteries. On the whole the condition produced in these few animals does not resemble human nephritis, but is much more similar to the lesions observed in experimental uranium nephritis.

During the past year a systematic study of the group of encapsulated bacteria (including Bacterium pneumoniæ Friedländer, Bacterium rhinoscleromatis v. Frisch, Bacterium ozenæ Abel-Löwenberg, and Bacterium capsulatus mucosus Fasching) has been carried on, employing for the purpose biometrical methods somewhat similar in character to those suggested by Winslow 1 in his work on the Coccaceæ.

During the course of the investigation, immunological methods have been used. At the outset, the reaction of agglutination was tried. Paltauf was the first (quoted by Beham 2 ) to suggest that the agglutination of encapsulated micro-organisms is inhibited because the bacilli are surrounded by a slimy nucleo-protein capsule. Porges 3 was able to supply experimental proof of this. v. Eisler and Porges 4 then elaborated a method of removing the capsule, after the application of which these bacteria were agglutinable.

My own work done independently of Beham has given results in harmony with his. I have found that Bacterium rhinoscleromatis on injection into rabbits yields a potent agglutinating serum.

Using this serum, agglutination not only of the homologous microorganism has been obtained but a positive result was found to occur also when four other strains of the same species, from widely different sources, were tested. The bacteria agglutinated in a dilution of the serum, ranging from 1-400 to 1-800. Two strains of Bacterium ozane, one strain of Bacterium capsulatus mucosus Fasching, and one strain of Bacterium pneumonia Friedlander, were not agglutinated at all by the same serum. The strains of Bacterium rhinoscleromatis which agglutinated were no longer encapsulated. All the other species still showed a capsule. One of the strains of rhinoscleromatis when originally isolated was not agglutinated by the serum of the patient from whom it was isolated and at that time it was encapsulated (Thro, Proceedings New York Pathological Society, April and May, 1910).

Some recent work of my students has been interesting in connection with the relative frequency of occurrence of the species of the B. coli group in water contaminated with human feces. The B. coli group consists of four species: B. communior (Durham), B. communis (Escherich), B. aerogenes (Escherich), B. acidi lactici (Hueppe). The general characteristics common to all species are: short, Gram-negative, cocco-bacilli, with rounded ends, non-spore bearing, fermentation of dextrose, glucose and lactose with gas production, non-liquefaction of gelatin after 14 days, indol production, and reduction of nitrates to nitrites. In general, also, litmus milk is acidified and coagulated. Further differentiation of the members of the group is possible by means of fermentation tests in the sugars dulcite, saccharose, mannite and raffinose. B. communior and B. communis ferment dulcite; B. aerogenes and acidi lactici do not. B. communior and B. aerogenes ferment saccharose. B. communis and B. acidi lactici do not. MacConkey 1 isolated a large number of strains of these species from feces and found that their relative frequency was as follows:

Graham Smith 2 in the analysis of thirty-five strains isolated from flies found, on the other hand, a predominance of communior (43 per cent.), communis occurring much less frequently (17 per cent.). B. aerogenes (11 per cent.) and B. acidi lactici (29 per cent.) did not differ greatly in their occurrence.

Winslow and Walker 3 in twenty-five strains isolated from feces found 28 per cent. of communior, 60 per cent. communis, 4 per cent. aerogenes, and 8 per cent. acidi lactici.

The results of an analysis of thirty-two strains isolated in my laboratory from human feces were as follows:

The saccharose fermenting species, then, may predominate in water recently contaminated with human feces.

Very little work has been done to determine whether creatinine is produced by bacteria growing on media free from creatin and creatinine. Germán, 1 in a brief review of the literature on this subject, makes mention of only three contributions. In his paper, Germán gives the result of an investigation of thirty-five species of bacteria to determine whether they were able to produce creatinine. In summing up, he says that this characteristic might be of value in the differentiation of closely allied species.

Our work was undertaken to ascertain: (1) Whether creatinine production is of any value in the differentiation of groups of closely allied species; (2) to determine the method best suited for studying creatinine production by bacteria; (3) to ascertain what amounts of creatinine were produced so that certain quantitative studies might be undertaken.

A medium composed of 2 per cent. Witte's peptone and .5 per cent. salt furnishes a creatin and creatinine free medium. Twenty-eight strains of bacteria, belonging to the mucosus capsulatus group of microorganisms, including B. rhinoscleromatis, B. ozana and B. lactis aerogenes, were grown on the above medium for eight days. Ten cubic centimeters were then tested for creatinine by Weyl's method (adding several C.C. of a 10 per cent. NaOH solution and several drops of a freshly prepared sodium nitroprusside solution, a positive reaction being indicated by the immediate appearance of a dark red zone which soon turns to a greenish color) and 50 C.C. were tested by the Folin method (adding 7.5 C.C. saturated picric acid solution and 2.5 C.C. sodium hydrate, a positive reaction being indicated by the production of an orangered color similar to the color produced by a potassium bichromate solution).

In an endeavor to establish the factor that periosteum plays in the regeneration of bone, the following experiments were performed on the ribs of rabbits.

Several experiments consisting merely of a subperiosteal resection of the rib showed after 20 days a cartilaginous-like material filling the entire space; while after a longer time complete filling in with bone.

The next endeavor was to isolate the periosteum so as to prevent bony elements from growing in to the periosteal space.

In one experiment the rib ends were capped with lead—after 12 days no evidence of regeneration. This method was discarded as it was thought better to try and raise the rib from its periosteal covering without severing the ends, and isolate it by sewing muscle beneath the raised rib.

In 10 experiments of this nature the results uniformly showed a tendency for bone to grow in at the angle where the rib was raised from its periosteal bed. In one experiment after 59 days the whole area was filled in by bone, although in another after 8 months there was only a small spicule of bone growing in at the sternal side. In only one experiment after 26 days was an isolated island of bone found, free from connection with bone elements.

Although this does not prove the point sought, it at least emphasizes the tendency of bone to grow in the direction of existing periosteum.

At this stage it was noticed that there was a difference at times in bone regeneration and it was suggested to try the effect of blood clot. The experiment was performed in the same way as the previous one except that blood clot was placed in the periosteal gutter left by the raised rib.

Sixty-one white rats, obtained from local sources and hereafter designated “Local” were inoculated in the axillary region with portions of a Flexner-Jobling carcinoma which was obtained from the Rockefeller Institute for Medical Research through the kindness of Dr. Peyton Rous. The number of successful inoculations, determined after 19 days, was 42, or 69 per cent.

Sixty-four rats obtained from dealers in Chicago and hereafter designated “Chicago” were similarly inoculated with portions of a Flexner-Jobling carcinoma which had been obtained from the Rockefeller Institute for Medical Research early in the year and propagated through four generations of rats in this laboratory. The number of successful inoculations, determined after 18 days, was 55, or 86 per cent.

Both sets of tumors grew rapidly, attaining average diameters of 15.0 and 11.9 millimeters after 19 and 18 days respectively.

Beginning upon the 19th day in the case of the “Local” and upon the 18th day in the case of the “Chicago” animals, each of the two groups of animals was sorted without selection into three batches, of which one (consisting of 12 animals in each case) served as controls, another (10 “Local” and 13 “Chicago”) received 1 c.c. of a 3.9 per cent. suspension of cholesterin in N/10 sodium oleate solution, which was injected hypodermically directly into the tumors every 2d or 3d day (three times a week); the third batch (20 “Local” and 30 “Chicago”) received 1 c.c. of a 2 per cent. aqueous emulsion of lecithin similarly injected into the tumors upon the same days.

It was found that in the pre-metastatic stage cholesterin causes a very notable acceleration of the primary growth, the gain in diameter between the 19th (18th) and 24th (23d) days being 11.6 mm. in the “Local” and I 1.4 mm. in the “Chicago” animals as compared with 4.8 mm. and 2.9 mm. respectively in the controls.

The cock's comb test has become a popular method of determining the activity of ergot preparations. It was introduced because gangrene frequently occurred in epidemics of ergot poisoning, and because bluing of the cock's comb was believed to be due to arterial constriction. This view was supported by von Recklinghausen's interpretation of the microscopical examination of the comb in chronic ergot poisoning of cocks. We were impressed by the fact that in ergot poisoning often the only pathological feature was venous dilatation and we believe that venous dilatation is probably the real cause of the bluing of the comb. It is admitted that the intravenous injection of epinephrin causes a rise in blood pressure, mainly from vaso-constriction and we have found that it will blue the cock's comb, but the bluing only comes on as the blood pressure falls and persists for an hour or two; in other words it occurs at a time when arterial-constriction has subsided. Large doses of paraldehyde given per os, or the inhalation of amyl nitrite, will also cause bluing of the comb.

On subcutaneous injection neither adrenalin nor p. oxyphenylethylamine caused this bluing. Dale claims that much of the activity of ergot preparations is due to p. oxyphenylethylamine. Now while the subcutaneous injection of 25 mg. of p. oxyphenylethylamine caused marked symptoms in cocks, there was no bluing, hence one would argue that the subcutaneous injection of ergot into cocks cannot be considered a quantitative method for testing ergot. Oncometric studies of the cock's comb during acute ergot poisoning are now being made.

Experiments show that the toxic principle of Rhus diversiloba while not volatile can poison at a distance by means of mechanical carriers. It is not destroyed by subjection for one hour to 100° C. and is carried, potent, by the smoke from burning Rhus plants. The dermatitis produced by this plant is a purely local affection and is not spread by the blood or lymph or by the serum of the blebs. The reaction of the sweat has no relation to individual susceptibility.

Absolute immunity in man has been claimed but was not found to exist, on repeated attempts, in any of the individuals tried (6 persons). In spite of the work of Ford, the conclusion is arrived at that experimental immunity in animals to Rhus toxin has not been proved. The author, working with pure toxin (glucoside) produced by the method of Syme, was unable to intoxicate animals with any reasonable amount. Ford's work was done with a commercial fluid extract containing various impurities. A permanent aqueous suspension of the alcoholic solution of the pure toxin can be prepared, and remains toxic for the human skin. As much as 0.025 gm. of this preparation of toxin can be given intravenously to a 2,000-gram rabbit without fatal effect, and as much as 0.03875 gm. can be given to a 280-gm. guinea pig subcutaneously without fatal effect. Pure alcoholic extract when given subcutaneously produces severe necrosis and death owing to the alcohol itself used as a menstruum, but with no lesions attributable to the toxin. No skin lesions were produced in monkeys, rabbits, or guinea pigs as a rule, but a slight dermatitis was produced on the rabbit's ear at times.

Inasmuch as animals can not be killed by the pure toxin, it is impossible to demonstrate antitoxic effect in the serum of animals that have received repeated doses of the toxin.

It has been frequently observed that the serum of rabbits, immunized with foreign protein, may, at certain times, contain not only precipitin, but also unaltered antigen remnants. Such sera not only precipitate the antigen, but also give precipitates when mixed with other antisera prepared with the same antigen (Linosier et Lemoine, Eisenberg, Michaelis, and Fleischmann, Ascoli, Von Dungern). They have been recently studied by Gay and Rusk. It has been difficult to explain why such sera do not spontaneously precipitate since both reacting factors are present.

In the cases of two sera recently studied by us the phenomena observed were as follows: Sera “3” and “4,” obtained by injecting two rabbits with horse serum on three successive days and bleeding eight days after the last injection, were perfectly clear and showed no spontaneous precipitation on standing several days. Serum “3” precipitated horse serum in dilutions of 1 to 1,000, and serum “4” in dilutions of 1 to 500. When mixed with equal quantities of an antihorse serum precipitin, which contained no antigen, both sera were precipitated, “3” more strongly than “4.” Neither of these sera fixed complement. When “3” and “4” were mixed, slow and slight, but distinct precipitates occurred, unmistakable after 12 hours in the ice chest.

On standing for one month in the ice chest sera “3” and “4” were found to have spontaneously precipitated. The precipitate showed slight complement fixation and the supernatant fluid was found on titration to have lost about one half of its precipitating power.

Such spontaneous precipitation of precipitin sera has been repeatedly observed after prolonged conservation.

Von Dungern who has done much work on this question has assumed a multiplicity of antigens and precipitins to account for this failure of union between antibody and antigen in such sera.