June 1900.] BOTANICAL SOCIETY OF EDINBURGH 325 



The following methods were, however, employed to deter- 

 mine this point for Hydrocharis buds : — Test tubes, con- 

 taining ordinary tap water and drawn to a narrow neck, were 

 boiled in a steam jacket for several hours to expel any 

 dissolved gases, they were then connected, while still hot, 

 with a series of potash bulbs by means of a rubber stopper, 

 and allowed to cool slowly, so as to prevent too rapid an 

 inrush of air. When quite cold, buds were dropped into 

 the upper parts of the tubes, a current of air freed from 

 carbon dioxide was drawn through the apparatus by means 

 of an aspirator, the buds shaken down into the body of the 

 tube, and the whole sealed at the blowpipe. 



As only tubes of a considerable size (10 in. long by 

 1\ in. diameter) were employed, and these were only 

 partially filled with fluid, a sufficient supply of oxygen was 

 always left for the respiration of the plants, and in every 

 case germination proceeded as under normal conditions. 



Cultures in larger vessels treated in a similar manner to 

 get rid of the dissolved gases, but left freely open to the 

 atmosphere through long tubes filled with soda lime, which, 

 while permitting the free passage of oxygen, prevented the 

 entrance of carbon dioxide, gave exactly similar results. 



That, as might be expected, germination does not take 

 place in the absence of oxygen may easily be proved 

 by treating a liask, as above described, to expel all the 

 dissolved gases, and then allowing it to cool iu an atmosphere 

 of carefully washed hydrogen. When cold, buds are 

 introduced, and any air which may have entered at the same 

 time is expelled by means of a rapid current of gas, con- 

 tinued for a few minutes, and the flask then sealed — no trace 

 of germination is observable either in light or darkness, — 

 and after the elapse of some days the buds will be found 

 to be dead, showing that in this, as in other cases, respira- 

 tion is essential to germination, with the necessary result 

 that a certain amount of carbon dioxide must be set free 

 during the process. With the object of obtaining some 

 insight into the influence of illumination on the respiratory 

 process going on in these buds before and during germina- 

 tion, a metliod was sought for by which some indication 

 might be obtained regarding the excretion of carbon dioxide 

 during the periods in question, and after some preliminary 



