236 BOTANICAL GAZETTE [September 



air. This nitrogen procured from the Linde Air Products Com- 

 pany contained a trace of oxygen, about 6-8 parts per thousand 

 by volume, but the results obtained with it were the same as with 

 the nitrogen prepared in other ways. 



Similar experiments were carried out in which hydrogen was 

 substituted for the kelp gas. The 15 determinations made showed 

 no formation of carbon monoxide within 5-7 days, either in the 

 light or in the dark. Here, as in the case of the nitrogen-filled 

 kelp, a small percentage (1-9 'per cent) of carbon dioxide was 

 formed. It should be remarked that there was always a marked 

 reduction in pressure for hydrogen-filled kelp. This amounts to 

 an absorption or diffusion out of hydrogen. The whole relation 

 of hydrogen in this connection deserves more exhaustive study. 



The hydrogen used was from two sources: (1) the action of 

 dilute sulphuric acid on the so-called arsenic free zinc; (2) a com- 

 mercial product prepared by electrolysis. 



A number of sections of kelp stipe were filled with a mixture of 

 nitrogen and oxygen, both chemically prepared. The initial com- 

 position was 15.2 per cent oxygen and the remainder nitrogen. 

 After 6 days' exposure carbon monoxide had been formed in all 

 cases, the quantities ranging from o . 8 to 2 . 1 per cent. The oxygen 

 content decreased and some carbon dioxide was formed just as in 

 the case of the specimens filled with air. Kelp filled with a mix- 

 ture containing 26.2 per cent of oxygen and the rest hydrogen 

 showed in the same time a similar formation of carbon monoxide 

 and a corresponding decrease in oxygen accompanied by the 

 appearance of carbon dioxide. 



The evidence so far presented seems to point to the inevitable 

 conclusion that the carbon monoxide is a product of respiration, 

 since it is not formed by decay or autolysis, and is formed only 

 when there is oxygen present in the pneumatocyst. 5 



A similar supplementary series of experiments was made 

 which supported the conclusion that the carbon monoxide was 

 formed by a katabolic process. The gas-filled corked sections of 



5 That the CO was a respiration product was early suggested by Rigg. This 

 conclusion was based on a special knowledge of the physiology of the plant. It was 

 his belief that the effect was intimately related to the sieve tubes. 



