ELIMINATION 173 



Botanical Gazette, 45, 1908, 254. She shows that the most vigorous tran- 

 spirers, in order of excellence, are Sunflower, Tomato, Lady Washington 

 Geranium, Marguerite, and White Lupine. Some of these, however, have 

 to be raised on purpose or have other practical drawbacks, so that for edu- 

 cational purposes the most available may be considered to be, in order of 

 general excellence, Marguerite, Garden Nasturtium, Lady Washington 

 Geranium, Fuchsia, and Senecio Pelasitis. This paper also gives full numer- 

 ical data upon these and the other common greenhouse plants. 



ENWRAPPING OF Pox AND SOIL. Many and diverse methods of accom- 

 plishing this have been described by different students as follows- (a) The 

 pot is placed in a glass jar roofed by bored-and-split sheet lead or glass 

 plates, all joints being made tight by wax or cement, or by rubber sheeting. 

 (b) The porous pot is replaced by a glazed vessel or by one of metal, roofed 

 over as before, (c) The pot is painted completely by melted paraffin, or 

 by a thin coat of modelling-wax, and roofed over by the same materials. 

 (d) The pot and soil are enwrapped completely by thin rubber sheeting 

 gathered around the stem of the plant. In all these arrangements the roots 

 are usually watered through a thistle -tube permanently inserted for the 

 purpose, and kept corked. More recently a new method, having high value 

 for some special purposes, has been introduced by WHITNEY and CAMERON, 

 as described in Bulletin No. 23, 1904, of the Bureau of Soils of the United 

 States Department of Agriculture, and by LIVINGSTON in the Plant World, 

 9, 1906, 62; it consists in the use of wire baskets covered by paraffin, which 

 have also the advantage, for many purposes, of ensuring an even distribu- 

 tion of roots in the soil. A method of covering the soil, by use of cement, 

 where the plants are in the ground and to be studied under a bell jar, is 

 described by CANNON in Bulletin of the Torrey Botanical Club, 32, 1905, 

 515. A method of avoiding enwrapping the pot at all was introduced by 

 MASTJRE (BURGERSTEIN, 5), and is employed in the RICHARD registering 

 evaporimeter later described; a pot containing the plant is balanced 

 with another exactly like it, but containing no plant, the evaporation 

 from the two being assumed to be equal; but obviously the method may 

 lead to much error. An ideal method should permit of rapid and neat 

 enclosure of the pot, and should allow of ready daily renewal of the vitiated 

 air of the soil; and these conditions are very well met in the use of the alumi- 

 num shells roofed with rubber, which I have described in the Botanical 

 Gazette, 41, 1906, 212; they are now obtainable among my normal appara- 

 tus (page 46) and are figured herewith (Fig. 44). Flower-pots are now 

 made so nearly in standard sizes that it is possible to make the shells to fit 

 them closely, and shells are now made for the 3-inch, 3^-inch, 4-inch, and 

 5-inch sizes. To hold the rubber roof closely to the shell, a narrow band 

 or strap of aluminum, resting in a groove just below the strengthened top 

 of the shell, may be drawn to any desired tightness by a convenient screw- 

 nut, shown on the right in the figure. The rubber roof may be attached 

 to the plant in any of the ways ordinarily used, but I find that, upon the 

 whole, the best method is the following- In the middle of a suitable -sized 

 piece of medium-thick rubber sheeting, a hole a little smaller than the stem 



