Experimental Biology and Medicine

In the isolation of nitrifying bacteria, soil is inoculated into a synthetic solution containing (NH₄)₂ SO₄ as a source of nitrogen for the “nitrite” bacteria and into one containing NaNO₂ for the “nitrate” bacteria. The solutions are incubated and when a test for nitrites is found in one case or a test for nitrates in the other, a small amount of the liquid is transferred to a new synthetic solution. This transferring is continued for some time. Then the material is plated out on silica jelly containing suitable inorganic salts. From the colonies developing, inoculations are made into synthetic solution and if nitrites are formed in the one containing (NH₄)₂ SO₄ a “nitrite”bacteria has been isolated or if nitrates are formed where NaNO₂ is the source of nitrogen a“nitrate”bacteria has been isolated.

Great difficulty was experienced in isolating cultures by the above procedure. At the beginning, in no case where inoculations from the colonies on silica jelly were made into the synthetic solution, were positive tests subsequently obtained. If, however, a solution giving a positive test were centrifuged and the sediment inoculated on silica jelly by means of a loop and from the colonies that developed, inoculations were made into synthetic solutions, positive tests would be obtained. This applied to both the“nitrite”and the“nitrate”organisms. Now if these positive cultures were inoculated on nutrient agar, growth would appear. This growth would not cause nitrification, showing the cultures in question were not pure. In one experiment the synthetic solution for nitrate bacteria which gave a positive test was centrifuged and inoculated on a gypsum block partly immersed in the synthetic solution. From the growth later obtained, dilutions were made and plated on silica jelly.

Since the beneficial effects of B. acidophilus appear to depend upon a transformation of the intestinal flora, it follows that a mass inoculation is desirable. This means a maximum number of viable organisms per c.c. of pabulum. The usual recommendation on commercial preparations of fermented milk is that such milk be kept in a cool place preferably in the ice-box. Consequently, the influence of low temperature on the number of viable B. acidophilus in milk seemed worth investigating.

The number of viable B. acidophilus in milk held in the icebox at about 97° C. was determined daily. The viable organisms were killed as follows: after 1 day—about 50 per cent.; after 2 days—about 75 per cent.; after 3 days about 90 per cent., etc. The obvious importance of having a large number of viable organisms is manifested in the necessity for increasing dosage in severe cases. In fact, it is likely that in cases reported as failures, a sufficient increase in the number of viable organisms administered might have resulted in success. The practice of ice-boxing B. acidophilus has little to recommend it beyond preserving the palatability of the culture and it is therefore more desirable to keep B. acidophilus at room temperature.

A study was made of the influence of time and pressure in autoclaving milk, prior to inoculation with B. acidophilur, with respect to the growth of the organism. Milk was sterilized at 15 and 20 pounds pressure in the autoclave for different periods of time and inoculated with B. acidophilus. Subcultures were made from each set of flasks for 3 days on milk identically sterilized, thus obtaining acclimatization. On the basis of these comparative tests it was found that milk to be used for inoculation with B. acidophilus should be sterilized in the autoclave at 20 pounds pressure for 20 minutes, or at 15 pounds pressure for 20 to 30 minutes.

The character“bent” in D. melanogaster is a member of the“fourth linkage group”in that species. 1 The genes for this and the other two characters now known in this linkage group have been shown by Bridges 2 to be“carried”by the small, dot-like“fourth chromosome.”Bent is characterized mainly by two variable modifications, one affecting the wings, the other the legs. The wings vary from“normal”through a series of modified shapes including spread wings, narrow wings, broad wings, swollen wings and most frequently wings that are bent sharply backward at a point near the base. The legs likewise vary from a“normal”condition through a series of stages in which progressive degrees of shortening and twisting is exhibited, beginning with the basal tarsal joint of the hind legs. The extent of the modification seems to be influenced considerably by environmental conditions.

Some time ago I found in Drosophila virilis a mutant character which bears considerable resemblance to this bent in D. melanogaster both in appearance and behavior. It exhibits much the same series of variable leg and wing modifications, except that those of the wings are less extreme and less frequent—the sharp bend being absent entirely. There are several lines of evidence pointing toward the conclusion that these characters in the two species are homologous, and that the gene for bent is in the small, dot-like chromosome in virilis as it is in melanogaster. Most of this evidence will be considered elsewhere, however. The present paper deals primarily with the effects of temperature on the two characters.

Since the characters show indications of homology and are both influenced by environmental conditions it is of interest to ascertain whether or not they react similarly to similar conditions.

1. Emesis was induced after destruction of the quadrigeminate bodies, after destruction of the cerebellum, and after section of the columns of Goll and Burdach in the cat, and after destruction of the area described as the vomiting center by Thumas 1 in the cat and in the dog.

2. Emesis could not be induced by any drug that we employed after destruction of the sensory nuclei of the vagi in the cat, nor could it be induced in any of three experiments in this animal in which the sensory nucleus of only one side had been destroyed, but vomiting did occur in one experiment in which an attempt to destroy the sensory nucleus of the right vagus may have been only partially successful.

3. Results of these experiments indicate that the sensory nuclei of the vagi are essential for the coordination of the vomiting reflex (that is, for vomiting however induced), and this is in harmony with our conception of the mechanism of emesis because: (a) It is well know that the vagus nerve is essential for emetic action of many drugs. (b) We have been unable to induce vomiting in the cat after destruction of the sensory nuclei of the vagi while taking especial care to avoid injury to the area described by Thumas as the vomiting center. (c) There are no nerve cells concerned so far as known, in the area described by Thumas.

4. We have shown elsewhere 2 that afferent emetic impulses from the heart pass by way of the sympathetic nerve, hence the conclusion is unavoidable that this nerve must make functional communication with the sensory nuclei of the vagi.

Reports have lately appeared that yeast can grow on a medium composed of known ingredients, viz.: 50 grams cane sugar (Domino brand), 2 grams KH₂ PO₄, 2.35 grams (NH₄)₂ SO₄, 0.25 gram CaCl₂ and 0.25 gram MgSO₄, dissolved in a liter of distilled water. It was claimed that on the above medium, which we will call medium O, yeast could be grown in sufficient quantities to serve as a source of vitamine B in animal feeding experiments.

We have attempted to grow yeast on the above medium but have found it difficult to obtain sufficient yeast by this method. It grew very slowly in the incubator and hardly at all at room temperature. The yields were very small; for example, 1800 c.c. of nutritive solution distributed in twelve flasks gave, after growing for a month, 1.5 grams of dried yeast, whereas on addition of 1 c.c. autolyzed yeast to each flask, 900 c.c. in six flasks gave in four days 3 grams of product. Whereas the yeast in the first case developed a brown pigment and presented a spore-like shrunken appearance, in the second case the cells were colorless and in active budding.

In sub-culturing the yeast obtained on medium 0, by introducing 5 c.c. into each of a series of flasks containing fresh media, the yield remained almost constant. This excluded the cause of the growth as being due to vitamine D introduced with the seeding, and suggested the possibility of an impurity in the medium. It therefore became necessary to investigate more carefully the purity of one or more of the three constituents of the medium, namely, salts, water and cane sugar.

Each one of the salts used as well as the cane sugar was dissolved separately in distilled water, shaken out with fullers earth and the filtrate evaporated to dryness.

In our continued investigation of the calcium in the blood we have paid particular attention to furunculosis and to diabetes. The results here published are part of over 225 determinations made on human beings with the method described by Kramer and Tisdall. 1 They state the normal is 9.2—11.1 mg. in 100 c.c. of blood.

In a previous paper 2 we showed that in a few patients with furunculosis the calcium in the blood was low. We can now add to this record the following confirmation of our former results. (The figures represent milligrams in 100 c.c. of blood). 10.8, 10.6, 10.2, 9.9, 8.7, 8.5, (J. H. 7.9), (E. B. 7.1).

In patient E. B. who had had over forty boils and who at the time the blood was taken had a large boil on the arm the calcium content was 7.1. The patient was given 0.1 gram of parathyroid per day and up to the present time an interval of five weeks, there has not been a recurrence. The calcium content was 10.3 after ten days of parathyroid therapy.

In J. H., at the time the blood was taken, there were two acute boils present on thebocly. After about seven days of parathyroid therapy there is an improvement in the condition.

These findings confirm those of Grove and Vine3 who record good results from the use of parathyroid in chronic infections.

We have made determinations of the calcium in the blood of children with pneumonia for Dr. J. D. Lyttle and obtained results varying from 5.8 to 10.9. Dr. Lyttle informed us that the amount of calcium so far as he could see had no relation to the prognosis.

In dogs permanently intubated for the collection of bile, gall stones not infrequently develop despite the absence of infection, stasis and gall bladder activity. The character of the stones has already been discussed. 1 They are always discrete to begin with, scattered upon the glass and rubber wall of the collecting tube. What determines this punctate localization? In certain instances, of almost pure calcium carbonate concretions, the answer is plain. These form in the midst of organic debris, not infrequently around bits of talc from the tube surface. In other cases minute, rounded, pigmented particles from the bile lodge upon the tube wall, and stone formation takes place upon these as nuclei. To trace the source of such particles and their significance we have made day to day studies of the sediment from sterile 24 hour specimens of bile centrifuged on removal from collecting balloons devoid of air. Also we have followed the early stages of calculus formation in the collecting tubes of the same animals.

The bile was found to yield formed elements identical with those later recovered from the tube system and from the interior of stones forming upon the walls of the latter. The nature and the amount of the sediment vary with the condition of the animal. For a day after the operation whereby intubation is effected it may consist merely of mucus, which later is seldom met with. Usually one obtains from the specimen of the second 24 hours after operation and perhaps more abundantly from that of the third, a slight brown deposit, made up, as the microscope shows, of minute, highly ref ractile, translucent, yellow-brown granules. The shape of these tends to be spherical, but is rendered various by the partial merging of the spheres.

The reduction in concentration of plasma proteins observed in a considerable proportion of nephritic patients has been attributed by some authors to a decrease in the amount of plasma proteins in the body, by others to a dilution of the blood with water (hydremic plethora). We have made one or more determinations of the blood volume by Keith, Rowntree, and Geraghty's “Vital Red” method on all but three of the patients above reported with reduced plasma protein concentration, and on some others.

The cause of the reduced protein concentration was found to be not hydremic plethora, but an actual decrease in the amount of plasma proteins in the body. Normal blood volumes were found, even in cases of extreme edema. The amount of total plasma proteins per kilo body weight (the weight being corrected as nearly as possible for estimated edema fluid present) however, varied from 1.5 to 3.0 grams, compared with 3.5 grams found in normal individuals.

Bright and a number of subsequent observers have described a diminution of proteins in the plasma of some persons suffering from albuminuria and edema; and later it has also been found that the decrease occurs chiefly in the plasma albumin (Epstein) the globulin not being diminished.

We have determined the albumin and globulin over varying periods of time in the plasmas of a number of nephritics by the recent method of Howe, with the results indicated below. The cases were classified as glomerular nephritis, nephrosis, and nephrosclerosis according to Volhard and Fahr.

In glomsrular nephritis a return to a normal total protein content has been observed in some cases, but a return to a normal albumin: globulin ratio has not yet been observed in any of our cases.

In one typical nephrosis patient a great loss of edema was observed without any change in plasma proteins, but after the edema had disappeared the total proteins began to rise towards the normal. The case of nephrosis with normal total content at the first examination was already ccrnvalescent.

The demonstration by the writer, in collaboration with MacLeod, Banting, and Best, that active preparations of the internal secretion of the pancreas conferred upon the depancreatized dog the power of glycogen formation led at once to the idea that wherever glycogen occurred in nature, a hormone similar to that produced by the islet cells of the pancreas would probably be found. Three obvious places to look for such a substance were 1 tissues of lower animals rich in glycogen such as the clam, 2 yeast, 3 fungi. The writer 2 was immediately successful in demonstrating the presence of such a hormone in clam tissue. Yeast was also investigated continuously for many months, and on January 26, 1923, after more than a score of failures, an extract of yeast was obtained which produced marked hypoglycemia in a normal rabbit (blood sugar 0.046 per cent). Since that date extracts of yeasts which have similar properties have been prepared by five different methods. The administration of such a potent yeast extract to a depancreatized dog also caused a marked fall in the percentage amount of blood sugar and a great decrease in the hourly excretion of sugar. As the yeast organism is a plant of the least differentiated type in the vegetabile kingdom, the idea occurred to the writer that, as all plants are sugar burners as well as producers, the preparation of the sugar molecule for combustion in the protoplasmic fire of the plant cell might he quite a secondary affair and be dependent, as Winter and Smith 3 have suggested, on the preliminary formation of 7 glucose, the combustion of sugar or the polymerization of the same being primarily dependent upon presence of the 7 form of glucose.

1. The duration of ventricular systole and its ejection phase are influenced at corresponding heart rates by the diastolic filling and the initial pressure of the ventricles, as demonstrated by recent experimental work.

2. The duration of these periods in hearts with auricular fibrillation is shorter (than normal at corresponding heart rates, even in spite of the higher venous pressures present.

3. These intervals increase when the heart reverts to a normal mechanism.

We conclude, therefore, that these observations give probable if indirect evidence of the dynamic importance of auricular systole in the normal heart beat in man.

The precipitation test for syphilis proposed by the author 1 called forth favorable comment from a number of investigators (Keim and Wile, 2 Herrold, 3 Young, 4 Ide and Smith, 5 Holmes 6 and Levin 7 ). Since the publication of the preliminary studies, however, several observations have been made which led to some changes in the original procedure; also, a modified technic has been evolved which is employed as a check on the original one. The improved method discussed in the following resumé embraces both the original and modified procedures.

In the previous communication, the author described a method for preparing antigen for the Kahn Precipitation Test and showed (Procedure II) that the sediment formed on mixing and centrifuging equal quantities of antigen and salt solution may be redissolved in salt solution and employed in the above test for syphilis. The question came up whether the same sediment taken up in salt solution may not be used as an antigen in the Wassermann test and the following experiments were carried out accordingly.

A cholesterinized extract regularly used in the Wasserman test in a dilution of 1: 70 with salt solution was employed. One c.c. of this extract was measured into a small tube and one c.c. of salt solution added to it. This was mixed and centrifuged, after which the supernatant fluid was poured off and the sediment suspended in 70 c.c. salt solution. This mixture was tested for anti-complementary, hæmolytic and antigenic properties side by side with a suspension prepared by slowly adding 70 c.c. saline to 1 c.c. antigen.

It was found that the suspension prepared from the sediment which resulted from mixing equal amounts of salt solution and antigen was considerably less anticomplementary as well as less hzmolytic than the suspension prepared by adding salt solution to antigen in the regular manner. It was further found that the antigenic properties of the suspensions resulting from either of the two modes of mixing with salt solution was about the same. Similar experiments gave the same results.

Clinical studies may ultimately establish that salt solution suspensions of the lipoid sediment possess high specificity and that the occasional non-specific Wassermann reaction given by cholesterinized antigens may be avoided by employing resuspended antigen-salt solution sediments.

A number of heart drugs or poisons belonging to the digitalis group were examined in respect to their toxicity for plant protoplasm. The method used was the same as that followed by the authors in a study of cocaine, alcohols, quinine alkaloids, etc. 1 The effect of the drugs was studied on the growth or elongation of the hypocotyls of three-day-old seedings of lupinus albus. The following drugs were examined, digitoxin, digitalin (German, Merck), digitalin Kiliani, digitonin, strophanthin U. S. P., ouabain or crystalline strophanthin and bufagin. Those principles which were not soluble in water were used in a 1 per cent. solution of ethyl alcohol, the control in such cases containing also the same amount of alcohol.

It was found that most of the above bodies were not very toxic for the plants, with the exception of bufagin. Thus digitoxin 1:100,000 gave a growth of 87 per cent. as compared with the control plants in normal Shive solution. Digitalin, Merck in the same concentration gave a growth of 86 per cent. Digitalin Kiliani in concentration of 1 :500,000 gave a growth of 49 per cent., while solutions from 1 :100,000 gave a growth of 98 per cent. Digitonin solution 1 :100.000 gave a growth of 82 per cent. Strophanthin U. S. P. 1 :100,000 gave a growth of 73 per cent. and ouabain in the same concentration gave a growth of 85 per cent.

The effects of bufagin were very different. Whereas Abel and Macht 2 found this to be an extremely potent drug for animals, its toxicity for the heart being compared to that of digitoxin and strophanthin, it was found to be very little toxic for plant protoplasm. Thus, while ouabin exerted practically no inhibition on the growth of the lupine (97 per cent) and strophanthin U. S. P. in concentrations of 1 :100,00 gave 73 per cent. growth and 1 :50,000 gave 63 per cent. of growth, it was found that bufagin in solutions 1 :100,000 gave only 20 per cent. of growth and in solutions of 1 :50,000 gave only 15 per cent. of growth as compared with the normal controls.

In Drosophila melanogaster when a white-eyed, long-winged female is crossed with an eosin-eyed, miniature-winged male, the daughters are all heterozygous and may be represented by the formula , indicating that one of the × chromosomes carries the determiners for white eye-color (w) and long wings (M) while the other carries the determiners for eosin eye-color (W^e) and miniature wings (m). If such a heterozygous female is bred she will have four kinds of regular sons irrespective of the male with which she is crossed, since the regular sons obtain their × chromosomes from their mother only. In two of the kinds of sons the characters will appear as they entered in the original cross, i. e., one kind will be white-eyed and long-winged and the other eosin-eyed and miniature-winged; these make up the noncrossover classes. In the other two kinds of sons the characters will be interchanged, i. e., one kind will be white-eyed and miniature-winged and the other eosin-eyed and long-winged; these make up the crossover classes. It is usual in work on crossingover to cross the heterozygous females with a recessive male, which in this case would be white-eyed and miniature-winged, in order that the daughters may also show the crossover and noncrossover classes. The experiments to be described were designed primarily to test for nondisjunction and for that reason the females were mated to males with the dominant character red-eyed so that the exceptional sons and daughters could be recognized. This did not of course affect the character of the regular sons and it is from counts of these that the crossover values have been determined. No attempt has been made to correct the data for double crossingover.

The several species of the genus Paramecium naturally fall into two groups: one with the general cell form represented by P. aurelia, P. caudatum, and P. multimicronucleata (aurelia group), and the other by P. bursaria, P. putrinum and P. calkinsi (bursaria group). Within each of these groups the species are distinguished chiefly by micronuclear structure and number. One type of micronuclear structure (caudatum type) occurs in P. caudatum and P. bursaria, and P. putrinum; the other (aurelia type) in P. aurelia, P. multimicronucleata and P. calkinsi. Species with the “caudatum type” possess a single micronucleus while those with the “aurelia type” possess two (P. aurelia, P. calkinsi) or several (P. multimicronucleata). 1

The present paper records the discovery, on November 29, 1922, in some material collected in Louisiana, of a Paramecium characterized by the general body form of the “bursaria group” and the micronuclear structure of the “aurelia type,” but possessing several (three to eight) micronuclei. In brief, the organism is essentially identical in form and structure with P. calkinsi, but has more than two micronuclei. Thus from the standpoint of micronuclear structure and number this animal holds the same position in the “bursaria group” as P. multimicronucleata in the “aurelia group.”

Pedigree cultures from the original animal found have now been under observation and experimentation for nearly four months, and through upward of one hundred and fifty generations. During this time the animals have bred true, exhibiting the characteristic micronuclear number after emerging from the nuclear reorganization involved in endomixis.

In view of all the above data the animals of this culture are designated a new species, Paramecium polycaryum. 2

The existence of a close relationship between the adrenals, especially the medulla, and the pancreas in the regulation of the carbohydrate metabolism has been assumed by various writers. It seemed therefore of interest to determine whether the action of insulin on rabbits which had survived total adrenalectomy differed from its action on normal rabbits. No difference was made out. The adrenals were removed in two operations. The second adrenal was excised 3 weeks to 8 months before the insulin experiment. Six to eight blood specimens were obtained at intervals during 4 to 6 hours after subcutaneous injection of insulin. The blood sugar followed the course described by Macleod and his collaborators in normal rabbits, falling from initial values of 0.10 to 0.1 1 per cent. to a minimum of 0.039 to 0.048 per cent. (Folin-Wu method). When convulsions occurred, or the animal became comatose without going into convulsions subcutaneous injection of dextrose had the same markedly beneficial effect as in the normal control animals. Usually there was some rise in the rectal temperature (as much as 1.5° C. in a case where the animal became comatose without convulsions, and 2° C. in a case where convulsions occurred and dextrose was injected).

If the internal secretion of the pancreas and the internal secretion of adrenal medulla both take a share in the regulation of the blood sugar content (perhaps, as some have supposed, by exerting actions more or less antagnostic) it might appear not unlikely that each would influence the secretion of the other. We have accordingly made some experiments on the influence of insulin upon the output of epinephrin in the cat, in which animal the blood sugar curve after insulin runs much the same course as in the rabbit, Neither with subcutaneous nor with intravenous injection of insulin was any decided effect produced upon the output.

It has been shown by Macleod and his collaborators that insulin prevents the development of the hyperglycemia caused by etherization, asphyxia, carbon monoxide and piqÛre, or counteracts it if already present. We have demonstrated that while these forms of experimental hyperglycemia are not essentially related to the adrenals, since they can be well elicited in the absence of those glands, this is not the case with the hyperglycemia produced by morphine in the development of which the adrenals seem to intervene in some way. We have therefore tested the influence of insulin upon morphine hyperglycemia in rabbits and cats. Whether morphine was given before, after, or at the same time as insulin, the characteristic effect of insulin upon the blood sugar was always observed. Thus, in a rabbit to which morhpine was administered 1 hour before insulin the blood sugar, which was 0.083 per cent. at the beginning of the experiment and 0.093 an hour after morphine, was 0.047 per cent. an hour after insulin had been injected. The morphine hyperglycemia had not had time to develop before insulin was given, nor did it develop in the 6 hours over which blood samples were collected. In another rabbit morphine was given an hour after insulin when the blood sugar had already fallen from 0.115 to 0.068 per cent. It went on falling to 0.052 per cent, and never even regained the initial value.

In a normal cat morphine and insulin were administered at the same time. A marked hyporglycemia developed, the blood sugar sinking from 0.13 per cent. to 0.033 per cent. A similar experiment was performed on a cat whose right adrenal had been excised and the medulla of the left destroyed by a drill, the left gland being then denervated, 19 days before the experiment.

During a systematic investigation of the normal and pathological reactions of the peritoneal mesothelium, certain rather surprising facts have been revealed which are interesting because of their bearing on the question of adhesions.

Elsewhere 1 experiments have been reported in which rats received repeated injections of solutions of glucose, the mesothelium undergoing certain changes of a morphological character which were not followed by adhesions. It seemed important to determine to what extent such changes could be produced without sufficient injury occuring to cause the development of adhesions. Such experiments have been carried out by the repeated introduction into the peritoneal cavity of various types of mild irritants, the best results having been obtained with laked heterogenous blood. Cats were given injections of 10 to 20 c.c. of laked rabbit's blood made isotonic with NaC1, the dose being repeated twice weekly over a period of 3 to 6 months. In such animals when the procedures were carefully guarded to prevent undue injury on puncture of the abdominal wall, and precautions taken to avoid any septic involvement, there resulted very remarkable changes in the entire mesothelial membrane without the formation of a single adhesion.

The peritoneum on section was often made up of two or three layers of cuboidal or columnar cells, attaining at times a thickness of 20 micra, and yet such a peritoneal lining seemed entirely adequate to prevent the formation of adhesions. From such observations the conclusion seems wholly justified that the presence of a complete layer of peritoneal lining cells, no matter how much their morphological appearance may be altered by such irritations as those used, is entirely sufficient to prevent the adherance of the two layers of peritoneum and thus prevent adhesions.

The observations of various anatomists have shown very clearly that there are wide variations in the structure of the placental barrier in such species as the ungulate, the carnivora the rodentia, the chiroptera, and primates. Grosser 1 classifies the placentalia in terms of the layers of tissue which separate the maternal from the fetal blood-streams; but it must also be noted that the same layer of tissue may be of very different morphology in different species. In view of these differences it seems unfortunate that the vast majority of observations which have been made upon the passage of any substance from mother to fetus, or the reverse have been carried out on single species.

Among the few observers who have made comparative experiments on different species are Römer 2 using tetanus antitoxin, and Wislocki 3 using vital dyes. Römer found that the antitoxin passed readily in the human, occasionally in rodents, and never in sheep or cows. Wislocki found that trypan blue passed in small amounts in the rabbit and guinea-pig, but never in the cat and dog. In his discussion he suggests that both his and Romer's results may be dependant upon the varying complexity of the placental barrier in the animals studied.

Experiments have been reported elsewhere 4 in which it was shown that sodium ferrocyanide passed through the placental barrier in the cat from mother to fetus, while iron ammonium citrate was held up by some mechanism located in the fetal ectoderm. Further experiments have demonstrated that in the rabbit both the salts passed from mother to fetus, but it was found that the sodium femocyanide passed the placental barrier somewhat more easily than the iron ammonium citrate; this was shown by the fact that the ferrocyanide appeared in the fetal blood before the citrate, and always remained in greater concentration.

It has been noted by Howell, Collip, Dale and Evans, the author and others that the intravenous injection of sodium bicarbonate may act as either a respiratory or circulatory stimulant, eliciting hypernea or a marked rise in the blood pressure. Such injection obviously increases the hydrogen ion concentration of the blood and inasmuch as it produces a slight dilution, it decreases the amount of carbon dioxide in the blood eliciting the stimulation. The increased respiration is, therefore contrary to the usually accepted laws of respiration. The only apparent change in the blood which might elicit stimulation is the greatly increased number of HCO₃ ions. Collip, therefore, suggests that the HCO₃ ion exerts a specific stimulating action on the respiratory center.

We believe, however, that this anomalous result may be otherwise explained. When the carbon dioxide is dissolved in water it exists primarily in three forms—dissolved CO₂ molecules, dissolved undissociated H₂CO₃—molecules and dissociated H₂CO₃—thus . Addition of sodium bicarbonate to the solution pushes the reaction to the left increasing both the dissolved CO₂ and the undissociated H₂CO₃. If we accept the view that the dissolved CO₂, and the undissociated H₂CO₃, diffuse freely into cells while the ions as such do not penetrate to any appreciable extent it is apparent that the injection of sodium bicarbonate increases the freely diffusable forms of carbon dioxide at the expense of the poorly diffusible ions and in that way increases the acid effects of the blood, at least on the interior of the cells (and possibly in the lymph bathing the cells as will be discussed in a later paper) even though the actual sum total of the original dissolved carbon dioxide in its various forms is not increased and the hydrogen ion concentrate of the blood is actually decreased.

Soon after the discovery of the agglutinins it was observed that some strains of a given microorganism were not agglutinated by a specific immune serum. Such non-agglutinable strains are said to be serum fast. The mechanism of serum fastness is not well understood. Ehrlich's explanation of this phenomenon on the basis of suppressed receptors does enable us to visualize the condition, but a suppression of receptors is probably far from what actually takes place.

In our studies 1 on pellicle formation it has been shown that pellicle forming bacteria are rich in acetone-ether soluble substances. Bacteria which ordinarily do not grow in pellicle were caused to do so by growing them in broth to which had been added glycerine or carbohydrates which they did not ferment. By growing the staphylococcus on glycerine broth for a few generations it began to grow in a pellicle, and finally formed as definite a pellicle as the tubercle bacillus, or other pellicle forming organisms.

The acetone-ether soluble substances of staphylococci grown in this way increased from 7.9 per cent. to 39.9 per cent.

It was concluded that pellicle formation was due to the increased amounts of acetone-ether soluble substances by virtue of which the bacteria resist wetting, and are thus supported on the surface of the medium by surface tension.

The present study concerns the effect of wetting on the agglutination reaction.

A laboratory strain of the staphylococcus aureus was grown in parallel cultures on ordinary broth and three per cent. glycerin broth respectively. Rabblits were immunized with killed cultures of the staphylococcus grown on ordinary broth. In the interest of brevity the strain grown in ordinary broth will be referred to as the “lean” strain and that grown in glycerine broth as the “fat” strain.

The precipitin test has been found to give reliable results in the diagnosis of active tuberculosis. The antigen is prepared by disrupting tubercle bacilli, preferable an old culture, with carbon dioxide by the method described by Larson, Hartzell and Diehl. 1 The disrupted bacteria are filtered through paper in order to remove the shells. The clear filtrate is layered over the serum to be tested, and the tubes incubated for a period up to two hours. A definite cloudy ring at the interface of the two fluids indicates a positive reaction. The cloud often appears within the first five minutes. In the far advanced cases, however, the reaction develops more slowly, but is usually very definite. Upon standing several hours the ring gradually becomes dispersed.

Thus far the blood serum of 190 cases have been examined. Of these, 100 were patients in the University Hospital and Dispensary, but not in the tubercular clinics. Ninety cases, representing all stages of tuberculosis, were from a local sanatorium. From the 100 cases not suspected of having tuberculosis, eleven positive reactions were obtained. Six of these have since been found to have evidence of tuberculosis. One died from hypertension, and at autopsy an active lesion was found in the apex of the right lung.

An interesting case was that of a newborn, whose mother was a far advanced case, giving a negative Von Pirquet but a positive precipitin test. The blood of the newborn gave a heavy precipitin reaction.

Of the 90 sanatorium cases 85 were positive and 5 negative. The 5 negative cases were either “healed” or “arrested.”

We have found that the acid fast actinomycetes, as A. gypsoides and A. asteroides, make a good antigens as the tubercle bacillus.

The organic material for iodin analysis is dried, mixed with and covered with CaO to render it alkaline and reduce the rate of combustion, and burned in pure oxygen in a large combustion tube with the end narrowed and bent downward for about 50 cm. The first third of this narrowed portion is covered with a thin layer of asbestos fibers to protect it from a lead coil through which cold water runs. The middle third is water-jacketed. The lower third dips into an absorbing apparatus filled with NaOH solution. The greater the amount of CaO mixed with the unknown material, the greater amount of iodin remains in the ash. The alkaline solution from the absorption apparatus is evaporated to 10 c.c. after adding to it the rinsings of the combustion tube. The ash is ground with water or the solution from the absorption apparatus in a ball mill and the extract is evaporated to 10 c.c. The solution is acidified until P_H = 1 and placed in a 12 c.c. separatory funnel and 0.1 c.c. of 0.1 N arsenius acid added, and allowed to remain for one-half hour to reduce any iodate to iodide. Then 0.1 c.c. of 5 per cent. sodium nitrite is added to oxidize the iodide to iodin which is extracted with 1 c.c. of CCl₄. The partition coefficient of iodin between CCl₄ and the aqueous solution is about 86, but varies with conditions. The iodin in CCl₄ is run into a 1 C.C. glass stoppered bottle and centrifuged to remove water droplets. The iodin is determined calorimetrically with a Bausch and Lomb Biological calorimeter specially made with cups holding 1 C.C. at 2 cm. depth, against a standard solution of pure iodin in CCl₄ (1 mg. in 10 c.c).

In the quantitative study of the growth of the human body there is often need for simple and accurate formulæ for expressing the relation between body-length and prenatal age. Empirical formulæ for lineal growth in the fetal period (3 lunar months to birth) have been published by Hassee 1 , Henry and Bastien 2 , and Scammorn 3 . No one of these is entirely satisfactory, for the first gives results which are not in accord with modern findings regarding prenatal lineal growth, the second is highly complicated, and the third is primarily for use in estimating the length from the age, whereas one usually desires to determine the age from the length. Practicable formulæ of this kind should fulfill the following conditions: (a) They should express age in terms of body-length for the entire fetal period (3 months to birth). (b) They should give a close fit to reliable data on the subject. (c) They should be in simple form, which will permit their application with ordinary arithmetic without the use of tables of special functions and the like. With these conditions in mind the following formulæ have been developed on the basis of the data of Mall 4 .

When Mall's observations are placed in graphic form they approximate a parabola having the general form: T = a + bL + cL², where T is the age in fetal or lunar months (calculated from the first day of the last menstruation), L is the total or crown-heel length in cm., and a, b and c are constants. The specific formula for expressing this relations is:

This may be simplified to:

For estimating the crown-heel length from the age, the formula may be transformed into :

These formulæ may also be modified for further use. According to Mall, the cohabitation age averages 10 clays less than the menstrual age which is estimated from the first day of the last menstruation, and the individual deviations from this average are quite small. Therefore, the formula for menstrual age may be modified to

to express the cohabitation age.