These data have been plotted as a heavy line in figure 4, in which the ordinates represent rates per minute and the abscissae distances from the pylorus. Unfortunately, these abscissae can be only approximately correct.

It is interesting that in these fifty-three animals there was no instance in which the average of two or three readings on any one segment gave a figure higher than the average for the segment just above. The gradients were not so even, however, in the sickly animals. Samples of these are shown as broken lines in figure 4. These upsets in gradient have been found to be even more marked in the intact bowels of sick animals studied under salt solution (4).

Three animals were starved for three or four days (without muzzles). Their segments beat with a poorer amplitude than normal but no characteristic changes in rhythmicity or in gradient could be made out.

Behavior after twenty-four hours. It is a remarkable fact that after twenty-four hours in Locke's solution between 5° and 10°C., the segments beat faster and the gradient of rhythm was retained. This will be, observed in the following three protocols:

The strength of the contractions suffered and the segments often became fatigued quickly. They were also less sensitive to drugs. although on two occasions they reacted typically with 1 part of adrenalin to 8,000,000 of the solution.

This evidence fits in with a great deal more (5) which it seems to me has proven that the rhythmicity is initiated in the muscle itself and not in the nerve-net. The differences in rate found normally in the different parts of the gut are due probably to differences in some phase of the metabolism in the muscle.

SUMMARY

Five segments excised from different parts of the rabbit's intestine have been studied under identical conditions in warm aerated Locke's solution.

The segments of duodenum and jejunum have greater tone and contract more after cutting than do the ileal segments. The colon also has a high tone.

The duodenal segment is generally the first to begin beating well. The tendency to rhythmic activity is graded from duodenum to ileum. The first few centimeters of duodenum, corresponding to the duodenal cap in man, does not beat well.

The colon is very slow in starting up and it differs greatly from the small intestine in its behavior.

The duodenum suffers more from trauma and from adverse conditions than do the other segments.

Segments from sickly animals beat poorly and become fatigued early. These changes are often more marked in some segments than in others so that the gradation of rhythm is changed.

The gradation of rate of contraction from duodenum to ileum is remarkably constant in normal animals.

In one case the gradation was upset by the presence of an inflamed area in the ileum. The bowel in that region contracted 21.5 to 25 times per minute, or twice as fast as normal.

After twenty-four hours, the segments beat at a faster rate and maintain the gradient. They continue to react normally to adrenalin and atropin.

BIBLIOGRAPHY

(1) ALVAREZ: This Journal, 1914, xxxv, 177.

(2) ALVAREZ: Loc. cit., 187.

(3) ALVAREZ: This Journal 1917, xlii 445.

(4) ALVAREZ: This Journal, 1915, xxxvii, 267.

(5) ALVAREZ: This Journal, 1917, xlii, 422.

QUANTITATIVE STUDIES ON INTRACELLULAR

RESPIRATION

I. RELATION OF OXYGEN CONCENTRATION AND THE RATE OF INTRACELLULAR OXIDATION IN PARAMECIUM CAUDATUM

E. J. LUND

From the Department of Animal Biology, University of Minnesota

Received for publication January 2, 1918

So far as the writer is aware no attempt has ever been made to study with quantitative methods, the oxygen consumption of any unicellular animal. There are only two papers, Vernon (1) and Barratt (2) which attempt to give anything like measurements of CO2 production by a protozoan cell.

The purpose of the present series of papers will be a, to describe methods which yield accurate and reproducible results on O2 consumption and CO2 production, and b, to show as far as may be, what the conditions are and how they affect the magnitude and rate of O2 consumption and CO2 production in these unicellular animals, using for this purpose to begin with, Paramecium caudatum. In this paper results are given to show what relation the concentration of oxygen has to the rate of intracellular oxidation.

METHOD

Preparation of Paramecium for experiment. The material for all the work was a pure line of Paramecium caudatum grown in large mass cultures in boiled hay infusion. The only other organisms present in the cultures were bacteria of various types commonly found in such infusions, which served as food for Paramecium; occasionally small amoebo-flagellates would occur which then served as part of the food supply for Paramecium. When the cultures were in their height of development the clear supernatant liquid containing the Paramecia was siphoned off carefully to avoid introducing any bacterial zoögloea. The organisms were then concentrated by use of the centrifuge at as slow a speed as possible. In this way injury to the animals was

avoided. The removal of the Paramecia to clear tap water, which whenever necessary had been sterilized by boiling, was accomplished gradually by repeatedly diluting the concentrated suspension with sterile or ordinary tap water at room temperature and then centrifuging. This transfer and washing of the Paramecia must ordinarily be gradual in order to allow time for the animals to adjust themselves to the new chemical and osmotic conditions. Transfer to pure tap water and washing was usually extended over a period of fifteen to twenty-five hours. In this way, with care, it is possible to wash Paramecia perfectly free from the native medium and to reduce the bacterial content of a Paramecium suspension to that of ordinary tap water or tap water which has been previously boiled. Hargitt and Fray (3) have shown by a series of careful tests that it is possible to sterilize a Paramecium by washing it five or six times in different portions of five to ten drops of sterile water. For the purposes of the experiments in this and following papers the small bacterial content of tap water and boiled water has no significance since if all the Paramecia remain alive, there will then be no pabulum in which bacteria in sufficient numbers to disturb the results will grow. Controls which rule out all effects from bacteria were always carried out whenever conditions demanded. These conditions will be referred to at the proper time. When Paramecia are washed in this way they are under starvation conditions. The food reserve of the protoplasm is gradually depleted as shown by the gradual decrease over a number of days in cell lipoids and increase in transparency of the protoplasm to light. But Paramecium will often live for as long as ten to fourteen days in tap water without food, which shows that the food reserve of the protoplasm may be sufficient to meet its expenditure of energy over this period of time. Cell division stops in a pure line population after the Paramecia have been removed to starvation conditions in tap water for twenty-four hours. Hence it is readily possible to obtain suspensions of cells which do not divide, that is, where the number of cells remains constant. It is always well to keep the Paramecia in a comparatively large volume of water before using, for high concentration of suspensions leads sooner or later, depending on conditions, to death of some of the organisms which serve as a food supply for bacteria that may happen to be present and also for other Paramecia which when starved become "hungry" feeders.

Just before using, the washed Paramecia were concentrated at a low speed of the centrifuge so that in 1 cc. of suspension there were

from two thousand to one hundred thousand individuals depending upon the concentration desired for the particular type of experiment. Equal volumes of this suspension were gently withdrawn by use of a 1 or 2 cc. volumetric pipette which had a large smooth opening. Stimulation and injury to individuals by the pipette is often brought about by too rapid suction on a pipette with a small opening and sharp edges. The question arises: Is it possible by this method to obtain equal numbers of Paramecia in different 1 cc. volumes drawn successively by the volumetric pipette? The answer is found by (1) a comparison. of the quantities of oxygen absorbed by different 1 cc. samples of Paramecia from the same suspension, and (2) by actual counts of the number of Paramecia in such samples of equal volume. Both of these methods have been used and as will be shown in this and subsequent papers, are trustworthy criteria for determining the number of individuals in unit volume. The error from this source falls within the limits of experimental errors from other sources.

Use of Winkler's method for determining dissolved oxygen. The reagents used for Winkler's method were made up as given by Treadwell and Hall (4) except that 5 cc. instead of 3 cc. of concentrated HCl was used. The thiosulfate solution was standardized at intervals against known weights of freshly resublimed iodine according to Treadwell and Hall (4, p. 645). For simplicity the tables give the oxygen equivalent in cubic centimeters of thiosulfate.

The tap water used for an experiment was kept in carboys and allowed to stand at room temperature. A stream of air was then passed through it for several hours in order that the oxygen in the water might come into equilibrium with that of the air at room temperature. Bottles of equal volumes (137 cc.) were then filled with the tap water from the carboy. The degree of uniformity of oxygen content in such a series of bottles is illustrated by the figures in tables 4, 5 and 6. The variation is on the average less than 0.1 cc. of thiosulfate per 137 cc. volume and where the average of a number of bottles is taken the error in filling of bottles and analysis can be reduced to less than 1 per cent of the oxygen content of 137 cc. of water at atmospheric pressure and a temperature of 20°C.

After filling the bottles 1 or 2 cc. of the concentrated Paramecium suspensions were added and the bottle tightly stoppered. After varying periods of time the water in the bottles containing Paramecia and the blanks without animals were analyzed. A small amount of the liberated iodine is adsorbed by the dead Paramecia. The amount