THE PLANT WORLD 103 



loosely in a large flask, which was then filled with water that had 

 been previously saturated with carbon dioxide by allowing it to bubble 

 through for half an hour. The flask was completely filled and then 

 closed with a stopper with two holes. Through one hole went a delivery 

 tube to conduct the gas under the mouth of a tube closed at one end and 

 inverted in a strong solution of caustic soda or caustic potash, to absorb 

 any carbon dioxide that might be carried over by the oxygen. This 

 tube should not project into the flask, but should end at the inner side of 

 the stopper. More or less of the water in the flask is driven out by the 

 escaping gases, and a considerable amount of oxygen may be left behind. 

 In order to collect this a funnel is inserted through the second hole in the 

 stopper when the apparatus is set up. At the end of the experiment, 

 that is to say, when the flask has stood in bright sunlight for a day or 

 two, the gas remaining in the flask is forced into the inverted tube by 

 pouring water into the funnel. By the arrangement just described the 

 gas which is collected in the tube is fairly pure oxygen. Without further 

 purification it will cause a stick with a spark on the end to glow brightly. 

 The caustic potash can be removed easily before placing the thumb over 

 the end of the tube, in order to invert it, by allowing water from the tap 

 to flow into the dish for a few moments. The heavier solution soon runs 

 out of the tube and is replaced by water without the loss of gas. 



C. E. Waters. 



GROWTH OF DANDELION FLOWER STALKS. 



A CAREFUL study of the behavior of the flower stalks of the dandelion 

 has been made by K. Miyake, * who divides the period of development into 

 three stages : 1 . From the appearance of the peduncle to the middle of the 

 flowering time. This occupies from seven to ten days, according to con- 

 ditions. At the close of this stage the peduncle has attained one-third to 

 one-half its ultimate length. 2. From the middle of flowering to the 

 beginning of seed development. This occupies six to eight days, during 

 which time the peduncle grows only about one-tenth of its whole ultimate 

 length or even less. 3. From the beginning of seed development until 

 the fruit is ripe. This occupies seven to ten days, during which the 

 peduncle lengthens rapidly to two to three times its length at flowering 

 time, and takes a vertical position, whereas during the second stage it is 

 more or less bent. The region of maximum elongation lies near to the 

 receptacle, and becomes narrower as the limit of growth is reached. 



It will be seen that there are two periods of rapid growth separated by 

 one of very slow growth. Miyake concludes as follows : 



* Ueber das Wachstum des Blutenschaftes von Taraxacum. Beihefte z. Bot. Centralbl. i6 : 403. 

 1904. 



