270 
Mac Dougal, 
divisions were irregularly oval in cross section, and had an epidermis 
of flattened elements against which the underlying parenchymatous 
tissue three or four layers in thickness was crowded. These thin-walled 
cells were irregularly cylindrical and arranged without intercellular 
spacing, with their longest axes parallel to that of the filament of which 
they formed a part (Fig. 9). Whatever the relation may be it was 
noted that whenever regenerating cuttings floating in the water were 
anchored, the influence of the anchorage was to cause the next leaves 
formed to take on some of the characters of terrestrials although these 
Organs were some distance below the surface. Tlius in 1913 cuttings 
which had progressed so far as to be forming aquatic leaves with 
thread-like Segments, began to make strap-shaped divisions when the 
plantlets were fastened in the soil at the bottom of the dish. 
Some of the experiences with this plant in various climates show 
the manner in which developmental proceedure might take place in 
dissimilar complexes of external conditions. 
Six clumps ol plants grown as terrestrials at the New York 
Botanical Garden since 1902 were received at the Desert Laboratory 
March 17, 1906, and were kept in pots until they were taken to the 
Montane plantation just being established in the Santa Catalina moun- 
tains at 8,000 ft. in May. The history of these plants shows that they 
underwent changes by which broadly laminar nepionic leaves would be 
produced from the buds to be followed by others of the finely dissected 
types. These alternations were not seasonally regulär however. In some 
instances nepionic leaves were formed in midsummer with the advent 
of the rainy season to be followed by a series running to aerial 
dissected types. 
Some material was taken to the Cinchona Station of the New York 
Botanical Garden in the Blue mountains of Jamaica (7,345 ft.) by Dr. 
Forrest Shreve and the notes communicated by him read as follows: 
„The four plants of Roripa americana taken to Cinchona were 
set out in the pots about October 23 rd 1905 at different points in 
the beds or borders such as gave them different amounts of light or 
shade. The pots were merely sunk in the ground until the tops were 
flu sh with the level of the beds and the water received by them was 
only the natural precipitation. Düring November a number of leaves 
appeared which were entire and bluntly dentate, being rather spatulate 
in outline. These were quickly followed by leaves which were similar 
in the upper part to those just described but were lyrate at the basal 
portion. These were in turn followed by leaves which were rather finely cut, 
