is. The prezygapophysis of each vertebra articulates
with the preceding postzygapophysis by a smooth dorsal surface. One
nearly complete neural arch shows (Fig. 7 B) a pit above the neural
canal, clearly corresponding to the canal for a dorsal ligament shown
by Jarvik in _Ichthyostega_. Indeed this view of the neural arch and
intercentrum together brings out the striking resemblance between the
vertebrae of _Hesperoherpeton_ and those of the Ichthyostegids. The
rounded intercentrum in both is an incomplete ring enclosing the
notochordal canal.

[Illustration: FIG. 7. _Hesperoherpeton garnettense_ Peabody. A, End
view of incomplete vertebra, probably near anterior end of column. B,
Neural arch and intercentrum in end view, showing probable association.
C, Left lateral view of trunk vertebra. All figures: KU 10295, x 4.]

TABLE 1.--AVERAGE MEASUREMENTS OF THE TRUNK VERTEBRAE (in mm.).
NUMBERS IN PARENTHESES INDICATE THE NUMBER OF PIECES AVAILABLE FOR
MEASURING

----------------------------+------------+-------------+---------------
PARTS | Ant.-post. | Dors.-vent. | Transv. width
----------------------------+------------+-------------+---------------
Neural spine | 1.5 (3) | 3.0 (3) | --
----------------------------+------------+-------------+---------------
Neural spine and arch | 2.0 (4) | 4.5? (4) | --
----------------------------+------------+-------------+---------------
Neural canal | 2.0 (4) | 2.0 (1) | 1.0 (1)
----------------------------+------------+-------------+---------------
Intercentrum | 1.5 (5) | 3.5 (4) | 3.0 (1)
----------------------------+------------+-------------+---------------
Pleurocentrum | 1.5 (3) | 3.0 (2) | --
----------------------------+------------+-------------+---------------

The shape, in end view, of a partly preserved neural arch (Fig. 7 A)
seems to account for the incompleteness of the intercentrum just
mentioned; the ventral edge of the arch is emarginate in such a way as
to fit the dorsal surface of the notochord. The dorsal portion of this
neural arch is not present (either broken or not yet ossified), but
the opening of the neural canal is comparable in width to the foramen
magnum. Hence this vertebra may be one of the most anterior in the
column. In comparison with the trunk vertebrae seen farther posteriorly
it appears that there may be