20 THE COMPOSITION OF EXPIRED AIR, 



iu au atmosphere containing abnormal proportions of these gases, so far as the 

 effects of increase of carbonic acid and diminution of oxygen aie concerned, but it 

 is evident from the results reported in Tables I and J, that death does not occur in 

 atmospheres in which the carbonic acid does not exceed It) per cent, unless the 

 oxygen is I'educed to below 7 per cent, of the mixture. 



A series of experiments was made by injecting into animals the fluid con- 

 densed from the air expired by healthy persons and by a man with a tracheal 

 fistula, from whom it was possible to obtain such fluid without contamination from 

 the exhalations from the mouth. The details of these experiments, and of the 

 results ol)tained, are given in the Ajipendix, VI. The injections were made into 

 the general circulation in rabbits, and into the peritoneal cavities of labbits, guinea- 

 pigs, and white rats, following the methods employed by Brown-Sequard and 

 d'Arsonval (15) and by v. Hofmaiin-Wellenhof (23). The number of animals 

 inoculated with the condensed fluid of respiration was thirteen, in four sets. The 

 fluid was collected with the greatest care in a stei'ilized apparatus; subsequent cul- 

 tures made from it indicating that it was sterile. It was waiined to about 35° C, 

 before injection. The proportion injected, as compared with the body weight of 

 the animals, Avas, iu some instances, less than that used by Browu-Sequard and 

 d'Arsonval, in others greater than the smallest quantities nsed by them with fatal 

 effects. The results obtained, with the amount of fluid injected in each case, are 

 shown iu Table K, given in the Appendix. 



In most of the animals no observable disturbance of health was produced, nor 

 did this condition alter iu the course of several mouths during which they were 

 kept under observation. One rabbit died thirty-two days after having received an 

 injection into its peritoneal cavity of 5 c.c. of fluid condensed from the breath of a 

 man with tracheal fistula. The results of postmortem examination showed focal 

 necrosis in the liver, but no ecchymoses and hemorihages in the lungs and intes- 

 tines, such as are reported as a characteristic result of such injections by Bi'own- 

 Sequard and d'Arsonval. Three other rabbits which had received injections of 

 the condensed fluid, and had remained apparently perfectly well fj'om six weeks to 

 seven months, were killed and careful ^os^mo;•^»i examinations made. The results 

 of these examinations showed that there was no special disease or degeneration in 

 the organs of these animals. 



The results of this series of experiments are, thei-efore, in accord with those 

 reported by v. Ilofmann-Welleidiof (23), and indicate that fluid condensed from 

 the pulmonaiy exhalations of man has no toxic or specially injurious effect when 

 injected into animals, and that there is no evidence that such fluid contains an 

 organic poison. 



