1907] HIBBARD—FORMATION OF MECHANICAL TISSUE 367 
has been described by BAtt (p. 309). When the plants had reached 
a more advanced stage of development, this device was unnecessary, 
and was therefore discarded. To guard further against any possible 
injury that the cord might produce in the stem when heavy weights 
were used, the cord was frequently changed to a place above or below 
the original attachment. 
The seeds were germinated in 20°™ pots containing garden loam 
and clay. The loam and clay were thoroughly mixed in the pro- 
portion of one of garden loam to three of clay. This gave the earth © 
a firm cohesion, and for this reason few seedlings were uprooted. 
When necessary, a plaster of Paris cap was. cast over the earth to 
the edge of the pot and securely fastened with cord. The plaster 
was kept from the stems by means of surrounding paper cylinders. 
When the seedlings had reached a height of 8.to 10°", the experi- 
ments were usually begun, although in some cases they were allowed 
to grow stronger and taller before being subjected to tension. 
There are two ways to determine whether or not tension has 
increased mechanical tissue. First, by a measurement of the break- 
ing strength, and second, by the observation of the size of the various 
tissues and the thickness of the cell walls. The first method has 
been followed admirably on various stems by BALL. Beyond a few 
experiments performed by the writer, the need of further observations 
by this method seems quite unnecessary. 
To determine by microscopic means the differences between 
tension and normal plants in the mechanical tissues, several methods 
were used. One method was to determine the thickness of tissues 
and walls by an ocular micrometer. The areas of the woody cylinders 
were computed by making camera-lucida drawings of the xylem, and 
the areas of these irregular drawings were determined by means of a 
polar planimeter. A third method was devised of making camera 
drawings of the woody cylinder on Bristol board, which were then 
carefully cut out and weighed on a chemical balance. A fourth 
method consisted in comparing the phloem bundles of the tension 
and normal plants. The number of bast fibers in cross-section of a 
tension plant was compared with that in the bundles of the normal 
plant. The number of bundles in one was directly compared with 
the number in the other. A fifth method consisted in comparing 
