430 BOTANICAL GAZETTE [DECEMBER 
such disks during all the field work, and that after reading Professor 
NatuHorst’s letter this watch was made closer still while the various 
specimens were being unpacked, placed in order, and further 
developed. But not until all this work had been done did I finally, 
in an idle moment, uncover a staminate disk on a slab from near 
Mina Consuelo. Then I recognized that a form I had suspected at 
one time was a disk had been truly such, though poorly preserved. 
It seems I had not chanced on quite the best disk locality. Better 
localities will yet be found. . 
The El Consuelo Williamsonia staminate disk is a reduced cam- 
panulate form of the size and general structure indicated in jig. 2. 
As there shown, however, the number of fronds is arbitrarily taken 
for purposes of interpretation as five, instead of the true number of 
eight or ten. One cannot be quite sure of the exact number in the 
specimen. The important structural feature, however, is that instead 
of a bipinnate frond as in Cycadeoidea,' there is a small strictly 
once-pinnate form,. the rachises bearing only two lateral rows of 
synangia. These, however, are still of the marattiaceous or cycade- 
oidean size and structure. The component fronds project beyond 
the disk only to a height of about r.5°™, The basal region appears 
to be rather thin in texture ; but the state of preservation as well as 
the association with other fruits and leaves all go to indicate that 
these disks and doubtless those of other species and genera should 
yet be found in abundance. : 
There can be no gainsaying the supreme importance of this fossil 
flower in enabling us to form more exact and adequate conceptions 
of the course of evolution leading up to and resulting in the present 
diversity of gamopetalous plants. In describing the flowers of 
Cycadeoidea ingens and C, dacotensis, I pointed out that the disk 
type of cycadophytean flower clearly indicated previous stages with 
the staminate organs spirally inserted beneath an apical cone wie 
series of likewise spirally inserted megasporophylls, and that readily 
conceivable reductions and changes in such a primitively via 
inflorescence or fertile apex or branch fully indicated the mode st 
origin of Liriodendron. Whence it followed that the Magnoliaceae 
: ita s Sell 
‘I followed the old usage of calling these fronds pinnate in my descrip ae a 
ARBER has called them bipinnate forms. It seems rather more correct so to a0. 
