238 Puget Sound Marine Sta. Pub. Vol. 1, No. 23 



out clearly in the quantitative data to be given later, which shows that 

 carbon dioxide was present only occasionally in quantities large enough 

 to measure with the apparatus used for technical gas analysis, and that 

 when it was found, the quantities were minute. The oxygen content was 

 found to run higher than reported by Zeller and Neikirk (average, about 

 18%), and the range of variation among individual specimens was so great 

 that generalizations with reference to daily changes with light scarcely 

 seem justified. 



For the more complete gas analyses the Morehead apparatus was 

 used, for which the degree of accuracy is 2 to 3 parts per thousand. De- 

 terminations were made for carbon dioxide, illuminants, (that is, unsat- 

 urated hydrocarbon gases), oxygen, carbon monoxide, saturated hydrocar- 

 bon gases (calculated as methane), and hydrogen. The remainder of the 

 gas not absorbed was assumed to be nitrogen and water vapor, the rare 

 inert gases of the atmosphere not being considered. This covers pretty 

 well all of the possible substances which might occur in the kelp gas. 

 Such gases as hydrogen sulfide, phosphene, ammonia, etc., which certainly 

 are not to be expected, would be easily detected by odor and by simple 

 chemical tests even if present in traces. 



Within the limit of accuracy of the analytical methods used, neither 

 hydrogen nor hydrocarbons were found in the kelp gas, although the gas 

 has an odor which is probably due to minute traces of one or more organic 

 substances. A more complete investigation of the possible occurrence of 

 hydrogen or hydrocarbon gases is to be made as soon as suitably delicate 

 apparatus can be constructed. 



Zeller and Neikirk assumed that the gas remaining after the absorp- 

 tion of carbon dioxide and oxygen was nitrogen. The author was sur- 

 prised to find in addition carbon monoxide in quantities ranging from one 

 to twelve per cent, with an average of about 4%. This fact was con- 

 firmed by the analysis of gas from a little less than one thousand kelp 

 plants. Since carbon monoxide has never before been found in the free 

 state in a living plant, so far as the writer could find, the following qual- 

 itative tests were made to demonstrate its presence, and in every case the 

 reactions were pronounced and positive in their character. 



White paper moistened with palladium chloride turned black in the 

 gas from the kelp but was unchanged in the air. A check with known 

 carbon monoxide also turned the palladium chloride paper black. 



Human blood diluted with 200 times its volume of distilled water 

 gave a yellowish-red solution which changed to a pink on contact with 

 the gas. 



Vogel's spectroscopic method of analysis for carbon monoxide haemo- 

 globin, which is the final toxicologic test, gave positive results, as is clearly 



