Osborn.— Some Observations on the Tuber of Phylloglossum. 491 
in the ground, threatened by burial or desiccation, as a result of theedaphic 
and climatic conditions under which it grows. Development is by no 
means always a steady progress in which, through successive seasons, the 
tuber gradually attains a maximum size. In cases where considerable 
depth adjustment has to be made, or where the early onset of the dry 
season brings vegetation to a standstill, 1 the new tuber formed may be 
smaller in size than that from which the plant producing it was derived. 
Production of Tubers by Leaves. 
Field Observations 1917. 
In 1917, when dissecting plants of Phylloglossum from the soil for the 
statistical investigation above, a specimen was found which at first could 
not be interpreted. A small green leaf-like structure was found lying in the 
soil (Text-fig. 8), pointed at one end, the other terminating bluntly, smooth, 
and rounded. When removed from the soil it was found that this blunt 
end had a colourless cylindrical prolongation, nearly as long as the aerial 
portion of the structure, descending vertically into the ground. Its apex 
showed a minute white spot, and resembled the apices of young tuber 
stalks of Phylloglossum . In the angle made by the two portions a few 
colourless rhizoids were developed. 
When searching for similar structures the leaf shown in Text-fig. 9 A was 
found. This was obviously a detached leaf of Phylloglossum, but near its 
proximal end a slight swelling of adventitious tissue had developed on the 
side next the soil. Such a pad of adventitious tissue, which in the follow- 
ing pages will be termed the ‘ cell mass ’ for convenience of reference, has 
been invariably observed as a preliminary development in growth from 
detached leaves of Phylloglossum . Further search yielded two similar leaves, 
Text-figs. 10 and 3 1 A, also a third, in which, however, the cell mass was 
better developed (Text-figs. 12 A, B, and C). This last leaf was somewhat 
curved and lay upon the soil with its proximal end slightly elevated. It 
had produced laterally, from the convex side, a cell mass about 2 mm. long ; 
this was swollen irregularly at the farther end and was partly embedded in 
the soil. The cell mass was green where exposed to the light, while from 
the lower side of the swollen end rhizoids were produced. From the under 
surface of this swelling a colourless, almost cylindrical process descended 
into the soil. 
It now became apparent that detached leaves of Phylloglossum were 
capable of producing, after more or less adventitious growth resulting in 
a cell mass, a structure resembling the stalk of a tuber, such as the plant 
normally forms each growing season. The structure shown in Text-fig. 8 
1 See p. 508, for rainfall record at Belair, S.A., near which township Phylloglossum has been 
studied. In 1915 only 1-37 in. of rain fell during the last three months of the year. 
