674 Browne . — Contributions to our Knowledge of the 
one is exceptional in that the two bundles of the trace are not closely 
approximated at their point of origin, but are separated by a parenchyma- 
tous mesh ; the two bundles destined to the sporangiophore do not even 
depart from the edges of the two strands bordering the mesh, but from 
points slightly internal to the edges of the axial strands. The middle line 
of the parenchymatous mesh when followed upwards passes midway 
between the two xylem-strands entering the sporangiophore ; below them 
it only extends downwards through one-third of the internode, and its 
origin is clearly quite unconnected with the departure of the traces of the 
whorl below. The conclusion to be drawn would seem to be that here 
a single mesh, such as commonly appears above the two halves of a bifasci- 
cular trace, has, phylogenetically speaking, slipped down between them for 
a certain distance. 
Quite apart from the diminution in number of the sporangiophores 
of a whorl, due to the narrowing of a cone at its apex, there is a certain 
irregularity in the number of members in successive whorls of the cone 
of Eqnisetum (e. g. the seventh and eighth whorls of Cone A of E.palustre). 
Moreover, traces receiving two bundles are by no means rare. Two forms 
may usually be distinguished. In the first case the sporangiophore receiv- 
ing the two bundles is very much larger than the other sporangiophores, 
and obviously represents two approximated and concrescent primordia. 
Every stage between a complete concrescence and a merely basal fusion of 
the stalks can be observed with the naked eye if sufficient material is 
examined, especially of E. arvense and E. maximum , in which species con- 
crescence is unusually frequent. In this form the two traces, though they 
usually arise markedly nearer to one another than to the neighbouring 
traces, are yet not closely approximated at their point of origin. In the 
second kind of sporangiophore that receives two traces these arise very 
close to one another, and the sporangiophore itself is hardly, if at all, larger 
than the other sporangiophores; generally either of the two traces con- 
sidered by itself is markedly smaller than the other traces of the whorl. In 
this case we would seem to be dealing with a case in which the division of 
the trace is so premature that the latter actually originates as two separate 
strands. This interpretation is supported by the fact that a certain number 
of traces, though not completely divided into two bundles at their origin, 
were more or less deeply constricted in the middle ; in this case the division 
of the trace, though still premature, was less markedly so. Another case 
intermediate between the traces that originate as two bundles and those 
that only divide in the sporangiophore was found in the upper part of 
Cone D of E. arvense. Here a trace originated as an already markedly 
constricted bundle and forked in the axial cortex before entering the 
sporangiophore. The latter received not only these two bundles, but 
another whole trace ; it was thus a 5 double ’ sporangiophore showing in 
