530 
Reginald Ruggles Gates 
cells are, on the average, considerablv larger, in all tlie tissues com- 
pared. The slides front wliicli fliese measurements were made were 
front material collected from typical specimens of the two forms 
compared. Whether these size ditferences will hold for all indivi- 
duals of gigas and Lamarckiana will have to be detennined later. 
0. gigas in particular shows a very wide ränge of variability in its 
leaf characters, and it remains to be seen whether this is correlated 
with variations in the size or shape of the leaf cells. 
The method of measurement found niost satisfactory was to draw 
carefully the outlines of the cells to be measured, bv nteans of a 
cantera lucida, and then to ntake the measurements front these pro- 
jections. A Zeiss 2 mm apochromat lens was used, a no. 6 ocular 
and a Bausch & Lomb cantera, giving a magnification of about 1380 
diameters. The figures thus projected were of convenient size for 
measurenient and the numbers in the table refer to the dintensions 
of the these figures in mm. 
In every case except the pollen mother cells, by length of cell 
is meant the measurenient along the long axis of the organ (anther, 
Stigma or petal) bearing it. This is not necessarily the longest di- 
mension of the individual cell concerned. Similarly, by width is 
meant the measurenient of the cell in the direction wliicli measures 
the thickness of the organ bearing it. This was the most satisfac- 
tory method because in every case except the pollen mother cells 
the cells of the tissue measured are regularly arranged in rows with 
their axes respectivelv parallel to and at right angles to the surface 
of the organ. The pollen mother cells are more looselv and irre- 
gularly arranged during the later stages of reduction, and in this 
case the greatest diameter in median section was called the length 
and the diameter at right angles to this, which was usually the 
shortest, was called the width. 
The epidermal tissues were chosen for measurement because of 
their regulär arrangement in series, which eliminates as far as pos- 
sible the error from oblique cuts, and because of their peculiarly 
regulär rectangular shape. Tltey frequently show a great ränge of 
variability in ‘‘length” (as defined) but comparatively little in “width”. 
This is due to the fact tliat epidermal cells divide and grow onlv 
in a plane at right angles to the surface and are therefore in all 
stages of elongation after division, wliile their “width'' undergoes 
little if any change. For this reason the ränge of variability in 
“length” is always greater than in “width”. 
