812 > REPORT—1904. 
same plot, the lowest percentage of dry matter was only 10°7, and the highest 
19:7, the ratio being 100: 184. In three other varieties the ratios came out 
100 : 183; 100: 184; 100 : 179. 
The sugar and nitrogen vary still more widely. In 100 individual roots of one 
variety, grown side by side on the same plot, the following were the limits :— 
Ratio. 
Highest percentage of sugar eee BILL OV 100 : 190 
Lowest percentage of sugar So Lid | ’ y 5 
Highest percentage of nitrogen . "25 | 100 : 280 
Lowest percentage of nitrogen. 094 ° i ‘ : 
Throughout the examination of the individual roots careful records have been 
kept of the shape, size and colour of each root which has been sampled and exa- 
mined. Shape and colour do not appear to be in any way correlated with any 
peculiarity of chemical composition, As regards size, a mixed sample from fifty 
large roots is certain to contain a lower percentage of dry matter and sugar than 
a mixed sample from fifty small roots of the same variety grown under identical 
conditions, but there is nothing like inverse proportionality between size of root 
and percentage of dry matter. Among the 1,000 roots examined many large ones 
have been found containing high percentages of dry matter, and vice versa. By 
saving such Jarge roots with high percentages of dry matter for seed-mothers it 
should be possible to improve the race. 
Again, there appears to be no definite correlation between percentages of dry 
matter, sugar, and proteid and non-proteid nitrogen. Hach appears to vary inde- 
pendently of the rest. It should therefore be possible, by continuously selecting 
as seed-mothers roots of definite composition, to change the composition of the 
race in any desired direction. Experiments of this kind are already in progress on 
the University Farm. 
MONDAY, AUGUST 22. 
The following Papers were read :— 
1. On the Forms of Stems of Plants. By Lorp Avesury, D.C.L., FBS. 
Some plants have round stems, some square, some triangular, some pentagonal. 
No doubt there are reasons for these and other forms, but the author found no 
explanation in botanical works, 
It is, of course, important for plants, as for architects, to obtain the greatest 
strength with the least expenditure of material. To do this it is necessary that 
the plant should be equally liable to rupture at every point when the strain is 
equal. If not, it is obvious that a certain amount of material may be removed 
from the strongest part without increasing the danger of rupture. If the stem of 
a plant, or any other pillar, is affected by pressure—say of wind—one side will be 
extended and the other compressed, while between them will be a neutral axis, 
and both extension and compression will be greatest along the surface farthest 
from the neutral axis. It follows, therefore, that the strongest form is where 
the material is collected as far as possible from the neutral axis. The two bars 
cannot, however, be entirely separate, and must therefore be connected by a bar or 
bars. This is the origin of the well-known girder (fig. 1). 
If the forces to be resisted act in two directions at right angles to one another, 
two girders must be combined, one at right angles to the other. 
It the forces act in all directions, a circular series of girders will be required, 
as Schwendener and others have pointed out. This is the case in the stems of 
trees, where the woody fibres form a ring, only separated in places by what are 
aoe as the ‘medullary’ rays. ‘This is the reason for the prevalent round form 
of stems. 
The question then arises, Why is this form not universal? As regards plants 
having quadrangular stems, it may be pointed out that when the leaves were 
