46 MISSOURI BOTANICAL GARDEN. 
examined has had cellulose very abundantly and in thick 
layers such as are shown to occur in the wood of Populus 
tremuloides in plate 1 figure 1, and of Populus balsamifera 
in plate 2, figure 2. Consequently there seem to be 
good indications that the genus Populus,is characterized 
by the occurrence of cellulose in an extra layer. 
A number of species of Salix from widely scattered 
localities have also been examined. Salix alba was found 
to have an inner lining layer of cellulose in some of the - 
fibers of the wood. Salix fragilis had a very pronounced 
layer in stems about four years old. Salix sp. from Cali- 
fornia had cellulose in nearly all of the wood fibers. 
Salix fluviatilis from South Dakota had a very pronounced 
cellulose lining in many of the fibers. Salix viminalis has 
been found to have a thin layer of cellulose in the fibers. 
Beside the above species of Populus and Salix which 
have been personally found to have unlignified cellulose 
there are a number of cases which other workers have 
noted. Sanio®* found it to occur inwood of Populus pyr- 
amidalis. Sablon ¢ found it in woodof Salix sp. 
The sections of Populus and Salix were found to have 
the cellulose in the form of a thick and very distinct in- 
ner layer of the fiber wall which seemed to be somewhat 
loosely attached to the secondary lignified one. In some 
cases it filled the entire cell cavity, but this was not com-_ 
monly true. The cellulose layer was quite sharply delim- 
ited from the lignified one. Cellulose was never present 
in the cells of the oldest wood of the annual rings, but 
was always situated in the more open early wood. It 
was found only in the wood fibers and only in those 
fibers which do not join or abut upon the vessels and 
tracheids. 
In the Larix sections cellulose was present only in cer- 
tain of the.annual rings and also only in the wood fibers. 
* Sanio. l. c. + Sablon. Rev. gén. d. Bot. 16: 360-363. (1904). 
