236 Browne.—A Second Contribution to our Knowledge of the 
the xylem of Cone B, PL XIII). One expression of reduction common 
in the cone of E. limosum was not observed in the cone of E. maximum ; 
in no case did two meshes originating independently become confluent 
owing to the dying out of the trace-bearing strand between them. 
From what has been said above, it is evident that the reduction of the 
xylem leads to the formation of fewer meshes relatively to the size of 
the cone, but of meshes of higher orders. For instance, if the xylem 
forming a little island between two of the strands of the sixth whorl 
of Cone A (cf. PI. XII) had been considerably more developed, and had 
fused with one or other of the strands between which it lies, the mesh 
between the strands giving rise to the fourteenth and fifteenth traces of this 
whorl would have been markedly constricted at this level; had the increase 
of xylem been sufficient to link the island up with both the neighbouring 
strands, this mesh, one of the third order arising above the whorl below, 
would have been converted into two meshes, the lower of the first and the 
upper of the second order. Thus the statistics giving the relative frequency 
of the meshes and the proportion among them of meshes of higher and 
lower orders are interesting. In the following table these statistics are 
given for two complete cones in each of the four species studied ; under 
each species the cone that has the best developed xylem relatively to its size 
is put first. The table is constructed on a comparative basis. Thus Cone A 
of Equisetum avvense , the cone with the most developed system of xylem 
relatively to its size, has been taken as the unit, and the others compared to it. 
The first column, then, contains the actual number of parenchymatous 
meshes originating within the limits of each cone ; the second column gives 
the number of meshes we should find in each cone, did these bear the same 
proportion to the sporangiophores as they do in Cone A of E. avvense. 
Columns 3, 4, and 5 refer respectively to the proportion of meshes of the 
first and second orders, of the third, fourth, fifth, and sixth orders, and of 
orders higher than the sixth. In all cases the traces at the extreme apex 
of the cone, belonging to the terminal group of incompletely differentiated 
sporangiophores, have been left out of account. 
Species. 
E. 
E. palnstre 
E. limosum 
E. maximum 
Cone. 
Actual 
number 
meshes. 
Number of meshes 
necessary to main¬ 
tain the same pro¬ 
portion as in Cone 
A of E. arvense. 
Proportion of 
meshes of 
first and 
second orders. 
Proportion of 
meshes of the 
third, fourth, 
fifth, and 
sixth orders. 
Proportion of 
meshes of orders 
higher than 
the sixth. 
A 
90 
90 
ft 
E 
-: • 
B 
37 
45 
ft 
A 
— 
A 
34 
6i 
2 7 
E 
— 
B 
22 
64 
1 3 
8 
W 
1 
ITS" 
A 
34 
70 
« 
a 
— 
B 
22 
93 
E 
u 
5 
T2 
A 
88 
200 
n 
w 
3 
IT'S' 
1 B 
167 
467 
tV? 
rW 
XTfV 
