Experimental Biology and Medicine

The following studies were designed to show the hourly variations in the normal dog of the total non-protein nitrogen of the blood compared with the output of urine and of the nitrogen in the urine and the effect upon these of various diets and of variations in the water intake.

Eight dogs have been placed on a diet consisting of meat, lard, sugar, sodium chloride, bone ash and water adequate in calories and containing about 0.4 gm. of nitrogen per kilo of body weight. On the sixth day the ration was varied to suit the study; in dogs 2 to 6 being richer in nitrogen, while in dogs 7, 8 and 9 no food was given on the days of the study. The water intake was varied as shown in the table. On the day of the study the animals were catheterized at intervals of two to four hours and at the time of each catheterization 5 c.c. of blood was taken from the jugular vein for estimation of the non-protein nitrogen by Folin's method. The urinary nitrogen was estimated by Kjeldahl. The results are shown in the table.

Conclusions.—The daily variation in the non-protein blood nitrogen of the normal dog receiving a diet containing 0.4 gm. of nitrogen per kilo is about g milligrams, the maximum being reached about two hours after feeding with a return to the original level in about 10 to 14 hours. By feeding excessive quantities of meat the non-protein blood nitrogen may be increased 25 to 40 milligrams in 6 to 8 hours and the original level is usually not reached even at the end of 24 hrs. In one animal (Dog 4) the blood nitrogen did not reach its maximum until 14 hours after feeding; the urinary nitrogen exhibiting a parallel gradual rise; this is probably only to be explained on the basis of slow absorption from the gastrointestinal tract.

Temperament, which determines mood, is generally recognized as having an hereditary basis. It is obvious, however, that the method of inheritance is not a simple one, but what factors are involved have not hitherto been plain. An examination of 150 matings of two persons the mood of both of whom is known as well as that of their children and that of the children's four grandparents has afforded the means to test various hypotheses of which the following seems to satisfy the conditions very closely. There are two hereditary factors involved in temperament, one of which makes for excitation, the absence of this for self-control. The other factor is one that makes for cheerfulness as opposed to depression; the depression being the more easily possible because of the absence of this factor for cheerfulness or normality. The exciting and the normalizing factors and their absence may be combined in the matings in over 50 ways, but for a given mating the disposition of the offspring follows a definite law and the dispositions occur in certain proportions in a fraternity of brothers and sisters. Dominance in the simplex condition is frequently or usually imperfect so that a clear difference between the disposition of children who receive two doses and only one dose of a factor can be recognized. The exciting factor may occur in the child simultaneously with the factor for normality, in which case the person is constantly or periodically elated and when depressed he is not depressed below the normal. The absence of normality, or depression, may occur without elation, in which case the person is constantly or periodically depressed and when not depressed does not rise above the normal.

In 1889, Klemperer and Scheuerlen 1 found that no fat was absorbed from the stomach. They were able to recover 99.5 per cent. of the fat introduced into ligated stomachs of dogs after six hours. Within the last fifteen years, however, a number of investigators have published histological observations, from which they have concluded that gastric fat absorption does occur.

With the hope of obtaining some decisive evidence in relation to this question, two physiological-chemical methods, other than those previously employed, have been applied. Fat was introduced into the ligated stomachs of cats and dogs and a study of blood-fat content made in the one case, and in the second it was attempted to discover whether or not any fat left the lumen of the stomach by the use of fat stained with Sudan III. The animals were kept under anesthesia during the experiments.

In neither case could any fat be shown to leave the stomach, although histological studies of the gastric mucosa of the same animals gave results similar to those reported by Weiss, 2 Lamb, 3 Greene and Skaer 4 and others, showing that histologically demonstrable fat had entered the stomach walls.

Since no fat could be shown to have left the stomach by way of circulating fluids, it is concluded that no absorption occurs. The histological findings are explained by assuming that the fat present in the gastric cells entered by purely physico-chemical processes.

It has been noticed by former observers that the alveolar carbon dioxide tension usually rises after a meal. Two diametrically opposite explanations have been possible: (1) Acid digestion products displace carbon dioxide from the blood, and during the displacement the rate at which carbon dioxide passes from blood to lungs is increased. (2) The blood becomes more alkaline, as the result of secretion of gastric hydrochloric acid, or absorption of alkaline digestion products. Consequently the carbon dioxide capacity of the blood is increased, and in equilibrium with it the carbon dioxide tension of the alveolar air rises.

We have determined on each of a number of subjects, in conditions of approximate digestive rest and of digestive activity, the following data: (1) Alveolar carbon dioxide tension; (2) Alkaline reserve of the plasma as indicated by its ability to maintain its alkalinity after addition of acid; (3) Alkaline reserve of the plasma as indicated by the amount of carbon dioxide with which it can combine. The results show that the reserve alkalinity of the plasma increases during digestion, the alveolar carbon dioxide increasing simultaneously. The second of the above explanations is therefore correct. The cause of the increase in alkaline reserve is being further studied.

We wish to report a result obtained in a series of experiments in which the effect of intravenous injections of pneumococci upon subsequent experimental pneumonia was studied. Forty-nine dogs were used in these experiments. Observations were made upon the extent and morphology of the pneumonic process, the leukocytic reaction, the occurrence of agglutinins and the presence of living organisms in the blood and in the lungs. In one series, the animals were given intravenously each day for five days 0.7 c.c. of a broth culture of pneumococci per kilo of body weight. In another series graduated doses were given, beginning with 0.07 c.c. per kilo of body weight and gradually increased to the fifth and last injection of 0.7 c.c. per kilo. In both series on the sixth day, the dogs were given an intrabronchial injection of pneumococci. With groups of these animals, control dogs were also injected intrabronchially with pneumococci. The animals were killed at intervals from nineteen to forty-nine hours after the intrabronchial injection.

We wish to confine our present report to one significant result namely: the rapid disappearance of the organisms from the pneumonic lungs of the animals previously injected intravenously with pneumococci.

In previous experiments reported from this laboratory living organisms were observed in the lungs of dogs with experimental lobar pneumonia as late as the third day after the beginning of the process. The results of the lung cultures in the present experiments may be seen from the charts.

It is evident from these charts that previous intravenous injections of pneumococci bring about a destructive effect upon the organisms in the pneumonic lung-the organisms disappear from these lungs much sooner than from the lungs of animals not previously treated.

In the investigation of the action of serum ferments on animal tissues chemical methods have been almost exclusively used. Within the last few years the dialysis method developed by Abderhalden has been extensively employed in experimental as well as clinical investigations. While the technique of this method is not as difficult as it is usually assumed to be, at every step of the procedure a not inconsiderable number of errors are likely to creep in. It requires very careful preparation of the substrate, absolutely uniform dialyzing tubes and great care in boiling the dialysate.

The most weighty objection to the dialysis method, however, lies in the fact that it gives only indirect evidence of the enzymatic action of the serum on the tissue. It is due to this fact that recently Bronfenbrenner and others have claimed that a positive Abderhalden test does not show the proteolytic action of the serum on the tissue substrate at all, but that during the incubation of the serum with tissue, the antiferment of the serum becomes adsorbed. The removal of the antiferment exposes the serum proteins to the action of the ferments in the serum and it is their cleavage products which pass into the dialysate and give the positive reaction.

Considerations of this nature led the author more than a year ago to seek for a method of demonstrating the ferment action of serum in the histologic examination of tissues. A priori it was to be expected that early evidence of the chemical changes resulting from ferment action might be shown by alteration in the microscopic appearance and staining reaction of the tissue.

Almost the first specimen examined, that of a boiled guinea-pig placenta exposed to the action of normal and pregnant guinea-pig serum, having fortunately shown a most remarkable difference, the author was encouraged to continue the investigation.

For the purpose of ascertaining to what extent conditions of hypersusceptibility determine the character of pneumococcus lesions of the lung, series of experiments were made with a moderately virulent strain, and with an extremely virulent strain in its virulent state and after artificial attenuation. Rabbits were inoculated intravenously for purposes of sensitization with .1-15 c.c. pneumococcus filtrates or dead cells and then after two weeks injected tracheally with 1. c.c. live cultures. The animals surviving 48 hours were killed. Microscopic sections were made of all lungs.

In these experiments with attempted active sensitization none of the animals developed symptoms resembling anaphylactic shock nor was the lung involvement definitely increased in any series of previously treated rabbits. Tracheal injection of the moderately virulent organisms, however, caused marked lesions in both sensitized and unsensitized control rabbits.

In experiments with attempted passive sensitization, mixtures of 1 c.c. virulent or attenuated live cultures and .1 c.c. or .5 c.c. sera from normal or immunized rabbits when injected tracheally failed to incite uniformly extensive lesions in any series of animals though the proportion developing diffuse involvement was greater than in the previous experiments with active sensitization. Sudden paroxysms, similar to those of fatal anaphylactic shock were observed about twenty-four hours after tracheal injection in practically all the animals of two or three experiments. But these paroxysms were not associated with extensive lesions of the lung because in a number of the rabbits no characteristic exudative pneumonias were found, although the lungs were almost invariably deeply conjested.

While a hypersensitive state probably takes some part in the inception of the infection, these experiments indicate that the subsequent exudative lobar involvement is essentially a progressive and cumulative process.

The importance of knowing the amino acid content of feed-stuffs has led the authors to apply Van Slyke's method 1 for the analysis of proteins direct, without previous isolation of the proteins themselves. The fine-ground and well-mixed material is weighed off in amount equivalent to 2-3 grams protein (estimated from the N-content) and completely hydrolyzed with 20 per cent. hydrochloric acid. Thereafter the method of Van Slyke is followed.

Analyses of corn, corn germ, cottonseed flour, kafir corn, tomato seed (pressed) and peat have been completed, results in duplicate agreeing well. The authors have had difficulty in accounting for the sulphur of the protein, the results for cystin being lower than was expected. This difficulty has been experienced by others and it is generally believed that if the cystin grouping be present, it is decomposed on hydrolysis and the sulphur changed to a form not precipitated by phosphotungstic acid with the cystin fraction. It is also conceded that sulphur exists in protein in other than the cystin grouping.

The results show Kafir corn and tomato seed meal to be lacking in histidin. Qualitative tests for tryptophan are positive for tomato seed, positive but slight for Kafir corn. Osborne and Clapp 2 found tryptophan and lysin, 2.93 per cent. in glutelin extracted from corn by weak alkali. The same investigators found neither lysin or tryptophan in zein of corn. Osbome's feeding experiments with cottonseed globulin show this protein to be satisfactory for maintenance and growth. Lysin and tryptophan are both present. Other experiments by Osborne and Mendel indicate that lysin and tryptophan in the diet are both necessary for growth. If lysin be present without tryptophan maintenance is secured, but not growth. The writers found lysin N in both corn and cotton seed.

In a recent communication we have shown that the essential factor of shock is a disturbance in the normal distribution of the blood. This disturbance is of such a character that the normal quota upon the arterial side of the circulation is diminished and this diminution is maintained so that, as a consequence, even after the original cause of the disturbed distribution of the blood, whether of a mechanical, toxic or inhibitory nature, is removed the abnormal diminution of the blood upon the arterial side of the circulation not only persists but, in fatal cases, progresses from local peripheral causes alone until death occurs.

The reality and nature of these local factors in the production of shock is made clearest in that form of shock which is produced by mechanical means alone, because in shock created in this manner no other factor can enter except the consequences of a primary disturbance of the normal distribution of the blood.

The mechanical means which we adopted for the production of shock was that used for the reduction of blood pressure when testing out the shock-producing effect of trauma to the peripheral sensory nerves. It consisted in partially occluding the inferior vena cava within the chest by passing a thread around the vein and drawing out the two ends through the incision in the wall of the thorax. By the degree of tension exerted on this thread, the amount of blood passing from the veins to the arterial side of the circulation could be accurately controlled and, in consequence, the blood pressure.

It was found that a two-hour period of reduction of the arterial blood pressure to from 30 to 40 mm. of Hg was, with few exceptions, fatal within the next eighteen hours, even though there might be a rise of arterial blood pressure to nearly the normal height for a period of three to four hours after the release of the ligature around the inferior vena cava.

Practically no variation in the nitrogen content of rabbit serum (centrifuged) is produced by the usual variations in temperature at which the blood sample (sealed) is kept during the initial separation of the serum. Sera separated in the incubator, ice chest and at room temperature are identical, within the limits of the experimental error.

An increase of as much as 100 per cent. in nitrogen content may be brought about by increasing the length of time the serum is allowed to stand in contact with the clot. This increase is most rapid and most pronounced in samples kept at incubator temperature.

Variations as great as 40 per cent. may be observed in different samples removed from the same rabbit at the same bleeding. Small consecutive samples drawn at five-minute intervals may show differences as great as 10 per cent.

Variations as great as 40 per cent. above or below the average may be observed in sera of normal rabbits of the same age, size and breed, kept and bled under identical conditions.

Four dogs in nitrogen equilibrium were studied as to total output of nitrogen in urine and feces, output of ammonia, creatin and creatinin in the urine, and iron and fat in the feces. In each a preliminary control period of seven days was followed by like periods at intervals after splenectomy varying from three days to three months.

In three of the four animals the removal of the spleen caused no change in nitrogen metabolism, fat utilization or iron elimination. Two of these animals showed no change in the blood picture and the third only a slight non-progressive anemia. A fourth animal which developed eventually a moderately severe anemia showed slight loss of weight, a disturbance of nitrogen balance and of creatin-creatinin partition and an increased elimination of iron.

The following conclusions are reached: (1) that the removal of the normal spleen in an animal which remains otherwise normal causes no disturbance of metabolism: (2) that when disturbances of metabolism occur they are in all probability to be explained by the anemia which frequently follows splenectomy and not by a disturbance of metabolism consequent upon the absence of the spleen.

As is well known there is a rearrangement of the proteins of the blood in the course of immunization. The most notable change is an increase of the globulins and a decrease of the albumins. Such being the case, it seemed of interest to ascertain whether these modifications affected the rate of coagulability of the blood. Accordingly, the blood of a number of horses which have been immunized for months or years for the purpose of obtaining antisera against diphtheria, tetanus, meningitis, etc., was investigated from this point of view. Oxalated plasma was employed in the coagulation tests. The accompanying table gives the results of this investigation, and shows that the coagulation-time remains fairly constant in spite of marked variations in the proteins, that it is a factor which is not readily disturbed.

The hemolytic quality of soap solutions and particularly of sodium oleate has been studied by a number of investigators (Sachs, Meyer, Moore). Meyerstein later showed that the oleate hemolysis could be inhibited by lipoids, cholesterin, serum and organ extracts. Sodium stearate and palmitate are less active than the oleate. Moore has shown that the laking qualities of the soaps are in proportion to the unsaturated bonds of the fatty acids and that by iodine saturation can be inhibited.

The toxicity of soaps was studied by Munk who found that the quantity equal to 0.11 to 0.13 gram oleic acid per kilo body weight killed rabbits. In his experiments he introduced these quantities very slowly. The heart action was depressed, but continued longer than the respiratory function.

We have given rabbits sodium oleate intravenously. A 5 per cent. solution of anhydrous sodium oleate (Merck) in normal saline was used. The soap solution was injected into the ear veins in a single and rapidly given dose. Doses of 0.1 sodium oleate per kilo appeared very irritating, the muscles of the extremities were thrown into spastic contractions and the animal had convulsive seizures. The ill effects lasted for a minute and a half when the animal would show great depression with rapid breathing. The muscles then became quite limp; within an hour the animal was fully recovered. Doses of 0.13 gram per kilo weight were fatal. These animals would sometimes enter a convulsive seizure before the injection was completely made. The blood of these animals after death was not perceptibly laked and it would appear that in the animal body the quality of laking and that of toxicity were separate. When, however, the solution of sodium oleate was mixed with an equal quantity of cholesterin the toxic qualities were inhibited.

During the past two years we have utilized the method of thyroid transplantation in rabbits in the attempt to get further data concerning certain questions in the physiology and pathology of this tissue. One of these questions is that of the necessity or not of specific secretory nerves to the gland. The observations of Anderson, Berkeley and Rhinehart have shown that in the thyroid both vessels and gland cells are abundantly supplied with nerve fibers. Stewart, Francois Frank and others have demonstrated the richness of the vasoconstrictor nerve supply, and Von Cyon demonstrated the presence of vasodilator fibers, both sets of fibers for the most part reaching the gland through the superior laryngeal nerves. More recently Asher and Flack and Ossokin have published physiological evidence which they think supports the view that the gland is under the control of secretory nerves, and Beebe and his associates have found that prolonged stimulation of the thyroid nerves causes a slight reduction in the iodin content which they interpret as indicating the presence of secretory nerves.

The method of transplantation eliminates many of the physiological and technical difficulties and objections of the acute experiments.

It has been found that under certain conditions thyroid tissue may be readily transplanted in widely separated parts of the body, as for example in the adrenal, ovary, subperitoneal tissues, muscle, subcutaneous fascia of the neck, chest and abdomen, and also, though with more difficulty, in the spleen and bone marrow. By transplanting and removing a sufficient amount of the main gland, care being taken to avoid all contact with iodin, we have always obtained compensatory hyperplasia of the remaining stump, and in addition a simultaneous and similar degree of hyperplasia of any existing transplants independent of their location.

It was shown by Friedberger and Hartoch that guinea-pigs could be protected against anaphylactic shock by an intravenous injection of hypertonic salt solution immediately preceding the toxic dose of antigen. That this protection is not due, as Friedberger and Hartoch claimed, to the inhibitory action of the salt on the alexin was demonstrated by Ritz, who found that an injection of salt also exerted protective action in animals injected with proteotoxins (the anaphylatoxins of Friedberger).

Dale concluded that the protection afforded by sodium chloride against acute anaphylaxis was due to the decreased irritability of the smooth muscle. He found that if the uterus of a sensitized guinea-pig were suspended in hypertonic salt solution, the addition of the antigen no longer caused the usual anaphylactic reaction. Other stimulating substances, like pilocarpine and pituitary extract, also failed to produce their typical stimulation provided that the uterus was bathed in hypertonic salt solution.

It is the object of the present investigation to show that, when a preliminary injection of salt protects against proteotoxin shock, the absence of reaction is due to the lessened irritability of smooth muscle.

In the first place, it was necessary to confirm, if possible, the observations of Ritz. It must be remembered that, as Friedberger and Hartoch pointed out, the protective action of salt is a limited one and if the amount of antigen given at the second injection or if the amount of proteotoxin used is much above the minimal fatal dose, the poisoning is so acute and powerful that the protective effects of the salt are entirely masked. We had considerable difficulty at first in obtaining satisfactory experiments because of our failure to appreciate this relation, but when the toxic dose did not materially exceed the minimal fatal dose there was no difficulty in determining the protective effect of salt in guinea-pigs injected with proteotoxin.

Morgenroth and his collaborators have demonstrated the prophylactic and curative and prophylactic powers of ethylhydrocuprein, a quinin derivative, in pneumococcus septicemia in mice.

The writers have given ethylhydrocuprein hydrochloride in doses of 0.1 to 0.5 G. three times a day by mouth to 7 cases of scarlet fever whose fever then had an average duration of 8.9 days as compared with 7.4 in 7 untreated cases who came under the same conditions in the same epidemic. In eleven unselected cases of measles, however, treated with the same drug, the average duration was 4.3 days as compared with an average duration of 7.9 days in ten untreated cases. One child who received the drug at the onset of symptoms, however, had an illness of 5 days'duration in spite of the early treatment.

The above experience seems to warrant the clinical use of ethylhydrocuprein in the treatment of measles, but not in scarlet fever.

Negative results with ethylhydrocuprein were obtained in experimental rabies and experimental vaccinia, also in trachoma.

In recent articles Keysser 1 and his associate Hegner 2 have reported the successful inoculation of carcinomata and sarcomata from man and from the mouse into the vitreous humor of rats, an outcome which they ascribe to the absence of protective substances in the eye and to the indifferent character of its proteins. The tumors were injected in the form of a fine emulsion, the bulb being entered on its posterior aspect in order to avoid hemorrhage with the consequent entry of protective bodies from the blood stream.

Repetition of the experiment does not substantiate these claims. A few drops of a fine emulsion of Crocker Fund carcinoma No. 180 in Ringer's solution were injected under ether anesthesia into the eyes of rats, and a similar amount of the same emulsion into the eyes of normal mice to serve as controls; no hemorrhage occurred during the operation. The extreme proliferative capacity of this growth is shown by the fact that forty-nine mice out of fifty-four which survived for more than three weeks developed tumors from 0.5 to 1.5 cm. in diameter, while only five were negative. Since this approaches very closely the outcome of subcutaneous inoculation, the eye, as such, is not an unfavorable site for the growth of this neoplasm. Among eighty-three rats, however, none developed tumors, although thirty-five lived more than forty-two days after injection and eight as long as seventy-one days; hence, racial resistance protects the eye in common with the rest of the body. These findings were confirmed by microscopic examination of serial sections of the inoculated eyes.

In studying the excretion of any substance by the kidney there are at first glance three separate things to be considered.

First, the passage from the blood into the kidney cell; second, the transit through the cell substance; and third, the passage from the cell into the lumen of the glomerulus or tubule. It is obvious that if the transfer through either cell boundary is impossible that none of the substance can be excreted, and it is also obvious that if anything prevents the passage through the cell substance itself that the kidney will appear impermeable.

In considering the passage of glucose through the body of the cell there are two well-known facts to be borne in mind that so far as we know have not received sufficient attention in this connection. The first is that the kidney uses a relatively large amount of oxygen and the second is that glucose is one of the main sources of energy for the body in general. It is therefore very probable that the kidney normally oxidizes a certain amount of glucose. Hence, even if both cell boundaries are permeable for glucose not all the sugar that gets into the cell will reach the lumen of the tubule. Normally only the minutest trace of sugar is present in the urine, whereas if the blood sugar rises beyond a certain level some will appear. Although it is true that in certain cases of diabetes there may be only a very slight hyperglycemia yet it is in the main true that glycosuria is dependent on a more or less pronounced rise in the percentage of sugar in the blood. Many hypotheses have been advanced to account for this so-called threshold phenomenon and I hope you will pardon me for dismissing them with the statement that none of them are generally accepted.

The sphincter muscle from the eyes of cattle, sheep and hogs was tested as quickly as possible after death. Two methods were used: (a) A strip, containing the sphincter, was cut from the pupillary border of the iris, connected with a lever and kept in a bath of either saline, Ringer's solution or aqueous humor, at 38 to 40 degrees Centigrade—in some cases with oxygen streaming through. (b) In other cases the sphincter was only partly excised, but connected with the registering apparatus in such a way that contraction or relaxation of the sphincter alone could affect the lever. In these experiments the iris was kept in an enclosed, warmed, air chamber and adrenalin dropped upon it from a pipette at the time of the test. The 1:1,000 Parke Davis solution was used.

Over 50 experiments are included in this report. Adrenalin produced, practically without exception, a relaxation of the sphincter. This was true for each of the three species of animals tested. The relaxation began promptly and at first was rapid so that the lever traced very nearly a vertical line on the drum. Later the rate of relaxation slowly decreased but in most cases continued until it seemed to be maximal. No recovery was seen even when the adrenalin bath was replaced with a fresh non-adrenalin-containing bath. That the sphincter was still irritable, however, was demonstrated by the addition of physostigmin, when a good contraction usually resulted.

This inhibitory effect of adrenalin indicates that the cervical sympathetic supplies inhibitory nerve fibers to the sphincter muscle of the iris.

Normal rabbits.—First generation pneumococci injected intravenously into normal rabbits decrease in number in the circulating blood during the first thirty minutes. The pneumococcic count (plate method) at the end of thirty minutes is usually about 25 per cent. of the initial count. After thirty minutes the number either slowly increases or slowly decreases depending upon the dosage and virulence of the organism injected.

Immune rabbits.—First generation pneumococci injected intravenously into actively immunized rabbits generally disappear with great rapidity from the circulating blood. By the end of ten minutes the blood is usually sterile.

Little or no decrease in the pneumococcic count is observed in samples of blood isolated from the general circulation between ligatures placed about a blood vessel. Even at the end of an hour the count in the isolated blood sample may be nearly as great as the initial count at the time the ligatures were closed.

From this it is evident that the rapid disappearance of the pneumococci from the circulating blood of actively immunized rabbits is not due to a destruction of the pneumococci by the plasma or leucocytes, but to their mechanical removal or destruction by the fixed tissues through which the blood circulates.

Strips of the anaphylactic guinea-pig uterus contract strongly when tested with the foreign proteid toward which the guinea-pig is sensitive. Our attempts thus far to record similar contractions by applying the proteid to the uterus in situ have been unsuccessful.

From this it would seem that the reaction of the anaphylactic uterus while still supplied with normal nerve and blood elements is different from its reaction when isolated from the rest of the body.

Skin transplantations on smaller laboratory animals are usually unsuccessful due to difficulties of bandaging and immobilization. As a preliminary to certain immunity studies an operation has been devised to avoid these difficulties, the animals'own tissues being used as a means of immobilization and surgical dressing.

On guinea-pigs, for example, an incision is made in the mid-dorsal line and the skin and superficial muscles resected. The skin graft is placed in the sub-muscular pocket thus formed and fastened to the periostium of the ribs with silk sutures. The resected tissues are drawn up over the graft and the dorsal incision permanently closed.

About a week later, the superficial tissues are resected over the graft, and the edges of the resection wound sewed to the graft with silk sutures. The exposed graft is protected for a week or two with a light dressing of silver foil and cotton.

The results of the operation are good, so far as the initial union of tissues is concerned.

Riegel (1875) and J. Pal (Die Gefässkrisen, 1906) suggested the use of amylnitrite for the control of pain in lead colic and in the gastric crises of tabes dorsalis, believing that the fall of blood pressure sufficed to bring about diminution in these symptoms.

The present series of experiments was undertaken in order to determine whether the nitrites might not act by causing relaxation of a spastic condition of the intestinal walls. The experiments were carried out on rabbits whose intestinal movements were observed through a large window in the abdominal wall, which was closed from the outside air by inserting a crystallizing dish 10 cm. in diameter. The sides of the crystallizing dish were coated with thick beeswax and the window was held in place by fixing the abdominal walls around it with a purse string suture. The animals were lightly anæsthetized with ether.

Injection of 5 mg. lead acetate per kg. immediately caused the onset of intense peristaltic movements which seemed to be due to stimulation of the preganglionic synapse, since they were abolished by injection of nicotin or by painting nicotin upon the intestine, but were not affected by extirpation of the spinal cord and section of the vagi. As was suspected from the clinical results reported by Riegal and Pal, this peristalsis could be inhibited by inhalation of amylnitrite, by placing two drops of nitroglycerin upon the tongue or by the injection of 80-100 mg. per kg. sodium nitrite. The effects of the two former drugs were most marked, though the effects of the latter were more prolonged. Atropin, either locally or intravenously, also inhibited the peristalsis, as had been shown by Harnack.

The same effects were also obtained with the nitrites and with atropin when increased peristalsis was brought on by the local effect of heat, secured by placing warm water in the crystallizing dish window.

In view of the accepted value of total embryo emulsion as on immunizing agent against propagable tumors first demonstrated by Schöne (Munch. Med. Woch., 1906, LIII, 2517) and later used successfully by many other investigators, the following notes on the influence of special embryonic tissues and of placenta on established rat sarcoma seem of sufficient interest to be reported.

The tumor used in these experiments was a sarcoma kindly supplied the laboratory by Dr. Loeb. During the period covered by the investigations it showed the following biological characteristics:

It was readily inoculated by direct subcutaneous transplantation of small intact fragments, giving by this method successful transplants in 93.2 per cent. of albino rats used in five separate operations. A palpable tumor usually developed in about six days, growing progressively and rapidly and having an average diameter of 30 mm. at the end of seven weeks. It usually invaded the skin and body wall only in the late stages of its growth. The average life of the animals after the inoculation was about ten weeks. No metastases were found on macroscopic post mortem examination. Retrogressions to complete disappearance occurred in six out of the fifty controls used in these experiments (12 per cent.). These retrogressions occurred only in the tumors which remained under 12 mm. in diameter. Recurrences after retrogression were not observed.

The animals used were albino rats about two thirds grown, males and females in about equal numbers.

The treatment was conducted with normal salt emulsions of the thyroid, the spleen and the liver of rat embryos prepared under aseptic precautions and preserved in the ice-chest under toluol, and also with an extract of placenta with the attached uterus washed free of blood and prepared in a similar manner.