AND ITS EFFECTS UPON ANIMAL LIFE. 45 



From the determinations of the proportions of C0» and of O. in the air of the jar, after deatii 

 of the animal, in the Hammond experiments, it is evident that two factors were operative in killing 

 it. These were the low percentage of O. ])resent and the iiigh percentage of COj, which the 

 arrangements instituted for the absorption of this gas had failed to remove. In a short time the 

 exterior of the sponges became coated with BaCOj while the Ba(HO)o in the interior became 

 inoiierative. This can be demonstrated by determining the alkalinity of the fluid expressed from 

 the s])onges, at the end of the experiment, with solution of oxalic acid. Another fact which 

 substantiates such a conclusion is that of the clouding of the Ba(H0)2 in the Geissler potash bulbs 

 quite early in the experiment from the C'Oo in the air aspirated from the jar in supplying fresh air. 

 While the solution of Bn(HO)., used in the sponges was twice the strength of that usually employed 

 in C0„ determinations in the Pettenkofer flask method, the amount of solution which can be taken 

 up by the sponges of the size used (about lo c. c. each) is entirely too small to absorb more than a 

 fractional part of the C'0» generated by an animal during the time of an experiment. 



The mode of death in these experiments presented such a close similarity to that noted in 

 cases of COo poisoning, under other circumstances, that it was impossible to distinguish it from 

 death produced by that gas. Judging from the air analyses at death of the animals, from the con- 

 stancy of the symptoms and the close similarity of the gaseous contents of the jars at death of the 

 animals, and, besides these, the absence of any positive indications of the presence and action of 

 other poisonous expiratory products as manifested either by the action of the animals or the mode 

 in which death took place, it is safe to conclude that the low percentage of O, together with the 

 high percentage of COj, in the atmosphere of the jars, were the principal causes of death. The 

 mode of death differed in no particular from that noted in the case of animals dying in the closed 

 vessels, in the " Brown-Sequard " experiments, or in those made with artificial gaseous mixtures 

 where sufficient oxygen was present to support life for several hours. Another fact, observed like- 

 wise in all the other forms of experiment reported on, was the prompt revival of the animals when 

 removed from the jars and supplied with fresh air. In exceptional cases, where the animal was not 

 removed until death was certain to take place in a very short time, the revival of the animal did 

 not follow on removal from the jar, but death supervened at a shorter or longer period after 

 removal. The failure of these animals to revive might be attributed to the presence of ante-7nortcm 

 clots within the heart cavities produced by the long-continued respiration of such high percentages 

 of Coo as existed in the atmosphere of the jars in this and the other experiments. The prompt 

 revival of the animals removed from the jars a little earlier appears to be an additional indication 

 that the symptoms produced in these ex])eriments had been due to the relative proportions of O 

 and COo present in the atmosphere which the animals breathed. The effects of an organic volatile 

 poison would not allow such rapid recovery, and would most probably manifest itself by continued 

 ill-health on the part of the animals subjected to it. 



Some animals vitiated the contained air more rapidly than others, so that, while there is a close 

 relation between the composition of the atmosphere at the end of the experiments, it is evident that 

 the degree of respiratory interchange determined the duration of Hfe for each individual. The 

 room temperature for these experiments was very nearly constant — 18° to 25° C. 



A further attempt was made by modifying the apparatus. This modification is shown in Fig. 

 7. Here the COj is absorbed by passing the air, issuing from the bell-jar containing the animal, 

 through five Pettenkofer absorption tubes, each containing 100 c. c. of a strong solution of Ba(H0)2 

 [to g. Ba(H0)2 -f 8H2O to I L.]. In addition to this, the air is passed through two Pettenkofer 

 tubes, each containing 100 c. c. of Buchner's alkaline pyrogallate solution, to remove some of the 

 O from the air. The moisture is absorbed by CaCU placed in a shallow vessel, covered with a 

 perforated porcelain plate, in the bottom of the bell-jar. 



DESCRIPTION OF THE APP.\RATUS USED IN THE MODIFIED " HAMMOND" EXPERIMENT, FIG. 7. 



a represents a one-litre bell-jar resting on a ground-glass plate, and contains a shallow vessel with 

 CaClj. The vessel containing the CaCU is covered with a perforated porcelain plate on which 

 the mouse under experiment is placed. 





