Anatomy of the Cone and Fertile Stem of Equisetum . 675 
one of its two fused constituents a case of premature division of the trace. 
In some cases, however, it is not easy to decide if we are dealing with 
a rather large sporangiophore whose trace has divided prematurely, or with 
an exceptionally small £ double ’ sporangiophore. The position of the 
meshes in the internode above and the number of members, when constant, 
in the whorls above and below may afford the means of judging if we are 
dealing with a single or a double sporangiophore. In normal cones a 
bifascicular trace is more often due to premature division. 
Cone of E. limosum. 
If we pass to a consideration of the cone of E. limosum we see at once 
that the xylem is much less developed than in E. pahistre. Though this 
poorer development of the xylem is very striking it follows the same lines 
as the reduction of xylem in E. palustre. Meshes arising above traces tend 
to persist through more numerous internodes ; they are widened laterally 
owing to the dying out of the xylem above lateral or slightly internal traces ; 
they tend to { decur ’ below and to one side of the traces above which they lie ; 
meshes originating separately also become confluent by the dying out 
of the strand separating them. This feature, only observed once in E. pa - 
lustre and unknown in if. arvense, is very common in Cone B of E. limosum. 
This cone has relatively to its size less xylem than Cone A ; in the latter 
only two pairs of meshes originally separate become thus confluent ; in both 
cases this confluence took place above the seventh whorl. In Cone A 
31 meshes arose, and were closed within the cone ; of these 13 were of the 
first, 7 of the second, 7 of the third, 3 of the fourth, and 1 of the fifth order ; 
two meshes persisted unclosed to the apex of the cone, and eventually fuse 
above the last whorl round the terminal sporangiophore. In Cone B 
12 meshes arose and were closed within the cone; 4 of the first, 5 of the 
second, 2 of the third, 1 of the fourth order. Ten more meshes arose above 
sporangiophores, but these remained unclosed ; of these, one arose above the 
lowest whorl, and therefore persisted through all the (nine) internodes of 
the cone ; one unclosed mesh persisted through eight internodes, two 
through seven, one through six, two through five, and three through four 
internodes. Yet another mesh was unclosed ; this originated above the 
insertion of the annulus (cf. the account of the transitional region), and 
therefore persisted through the whole of the cone. These unclosed meshes 
become confluent with one another in a very irregular manner as we approach 
the apex of the cone ; above the tenth and last whorl the number of 
independent meshes is reduced to three. But if traced downwards these 
three meshes have a very different history ; one of them, that lying between 
the second and third traces of the tenth whorl of the diagram, consists only 
of two fused meshes extending a very long way downwards ; they become 
confluent by the dying out of a strand that pursued an isolated course 
Y y 2 
