354 KEPOET OF THE COMMISSIONER OF FISHERIES. 



The time required for a fatal result depends primarily on the degree 

 of excess. Death has been observed within three hours after intro- 

 duction of a healthy fish into the abnormal water, but in this case the 

 exact excess is unknown, and there was no autops3\ At 10^ C. and 

 an excess of 6 c. c. of nitrogen and 2 c. c. of oxygen per liter, a hake 

 was killed at the end of 8 hours, and embolic gas under pressure in 

 the heart was obseryed immediatel}^ after death. Ten hours fre- 

 quently suffices for this result. Species differ in susceptibility. 



Identity of gas in the Mood vessels, external vesicles, and water. — 

 Some relation of identity or source between the several gases within 

 the blood vessels, in the external blebs, and that which separates 

 directly from the water upon the fishes is at once inferred as prob- 

 able, and the gas of all the lesions would seem to be derived from the 

 water. The following four samples were determined by the United 

 States Bureau of Chemistry, the first three collected in ^ebruar3^ 

 The merel}^ adherent bubbles which formed on the exterior of the 

 fishes had the following composition: 



Per cent. 



Carbon dioxid . ; 1. 03 



Oxygen 1 7. 58 



Nitrogen 81 . 39 



A sample precipitated upon blocks of wood, no fishes being in the 

 water, consisted of: ^ 



Per cent. 



Carbon dioxid 0. 58 



Oxygen 22. 87 



Nitrogen 76.55 



The difference between the carl)on dioxid and oxygen in these two 

 samples should be referred to the respiration of the fishes, present in 

 the first case and absent in the second. The gas from the large ves- 

 icles on the belh' of the small sculpin {Myoxocephalus aeneiiK) was as 

 follows: 



^ Per cent. 



Carbon dioxid 3. 78 



Oxygen 1 8. 09 



Nitrogen 78. 13 



In this the oxygen is diminished and the carbon dioxid increased by 

 oxidation of organic matter in the tissues. Methane, hydrogen, or 

 carbon monoxid were not present in any of these samples. 



A sample collected in September from the water alone showed : 



Per cent. 



Carbon dioxid - 0. 4 



Oxygen 20.6 



Nitrogen ^- 79. 



The gas is evidently nothing more than the constituents of air, the 

 proportions varying more or less from those of the atmosphere. 



