Anatomy of the Cone and Fertile Stem of Equisetum . 237 
Thus the proportion that the meshes bear to the number of sporangio- 
phores is by far largest in Equisetum arveiise ; the species which comes next 
in this matter, and therefore shows a lesser degree of reduction than the 
remaining species, is E. palustre ; then follows E. maximum , and lastly 
E. limosum. But taking the average of the two cones, the proportion of 
meshes to sporangiophores differs but little in the last two species. A further 
test of reduction of xylem is, as already explained, afforded by the analysis 
of the proportion of meshes of relatively low and high orders. This test gives 
the same order of reduction of the species, viz. E. arvense , E. palustre , 
E. maximum , and E. limosum , and again E. arvense shows by far the least 
reduction, while there is very little difference between E. maximum and 
E. limosum . 
Mention has already been made, in the description of the other cones 
studied, of the occasional occurrence (except in that of E. limosum) of 
considerable bands of woody tissues extending uninterruptedly above 
part of one whorl to the level of the next. Cones showing this character 
are regarded as displaying locally the greatest relative development 
of xylem observed in any internode. If this continuous tract of xylem 
above a series of traces were to involve the whole whorl instead of but 
one or two strands (the largest number I have observed to be involved 
except at the extreme apex of the cone of E. pahistre ), we should have 
a complete woody cylinder, devoid of parenchymatous meshes even in the 
internodes. Such internodal sweeps of xylem occur several times in 
Cone A of E. maximum ; in one case, between whorls 3 and 4, such 
a sweep forms a very conspicuous feature in the reconstruction ; relatively 
large sweeps of xylem above two or more traces that have departed are 
not found in Cone B of E. maximum —perhaps not a surprising fact, as the 
xylem of this cone is less well developed than that of Cone A. In both 
cones, however, we notice another feature bearing a superficial resemblance 
to this character, and leading to an increase of xylem in the internode. 
This character consists in the linking up of two or three strands of xylem by 
the development of additional tracheides at a considerable distance below the 
departure of the traces ; sometimes this fusion of strands occurs but little 
above the level of the whorl below the one at which the connected strands 
give off a series of traces. These internodal sweeps of xylem must not be 
confused with those occurring above median traces ; the latter may perhaps 
be a primitive character retained in places, for they occur chiefly in the 
cones in which the xylem of the stele has undergone less reduction. The 
internodal sweeps of xylem that originate at varying distances below 
a whorl, but never extend uninterruptedly and vertically above the traces 
of one whorl to the level of those of the next, seem, on comparative grounds, 
to constitute a fresh character in the phylogeny, leading to the increase of 
xylem by a method different from that found in the more primitive types 
