144 PROCEEDINGS OF THE ACADEMY OF [1897. 



filled, and the rubber stopper, with its penetrating quill-tube, is 

 inserted, the last bubble of air is forced out by pressure, and the 

 tube suspended as shown. The remainder of the solution is acidified 

 with carbon dioxide from the lungs, blown into it through a glass 

 tube. The brownish-yellow tint having developed, tubes B and C 

 are filled with the solution, and into C some clean, living diatoms 

 are put. Both are then corked and hung as figured, the quill-tubes 

 dipping below the surface of the liquid in the dish. These quill- 

 tubes, which allow the pressure within the larger tubes, due to gas or 

 to expansion from heat, to relieve itself into the dish, are drawn 

 down to a very small opening in order to lessen diffusion of liquid 

 up or down, and to confirm the diatoms. The apparatus is now 

 exposed to bright light — if to direct sunlight so much the better, 

 since the action is then more rapid. Gas arises from the diatoms in 

 tube C, and simultaneously the color of the liquid, which is at first 

 like that in B, begins to change. Within fifteen minutes, under 

 proper conditions, the color has again become almost or quite as red 

 as that in tube A. The carbon dioxide has now in large measure 

 disappeared from the solution. The action continues, and the color 

 in tube C deepens rapidly, showing oxidation ; and this action con- 

 tinues until the color is quite blood-red, or even, in case much lime 

 is in the water, until bluish lakes are formed in clouds. The ceas- 

 ing of the action may, conceivably, be determined by exhaustion of 

 every trace of carbon dioxide, but data on this head are wanting as 

 yet. At all events the evolution of gas goes on long after the color 

 reaction of carbon dioxide has disappeared. 



The experiment may be varied in the following manner : All of 

 the tubes are filled with the normal, non-acid, reddish solution of 

 hematoxylin. Into A is put a living snail, into B live diatoms, and 

 C is allowed to remain for comparison. The whole apparatus being 

 exposed to sunshine, A pales rapidly under the influence of the carbon 

 dioxide from the snail, while B as rapidly darkens and reddens 

 compared with C, owing to the oxygen from the diatoms. This 

 result, so significant, is obtainable in a very few minutes. 



The diatoms selected for the above experiments were the long, 

 broad filamentous forms of Eunotia {E. major of Rabenhorst), which 

 are peculiarly applicable because it is easy to procure them in suffi- 

 cient abundance, and to free them, under a dissecting microscope, 

 from any accompanying alga? that might, by their presence, tend to 

 acts doubt upon the conclusiveness of the results. 



