Benson.—Mazocarp on or the Structural Sigillariostrobus . 585 
as microsporangia, are split away from the cone-axis, but are still in the 
same relative position in which they were when attached. The sporange 
seen laterally ( d ') is very like the microsporange of Mazocarpon. As to the 
dimensions, the large bodies ( b , c, and e in the Fig. 1) given by Kidston 1 agree 
in every particular with that of the microsporange of Mazocarpon , which 
was nearly 5 mm. across (Figs. 10 and 11) and formed a lateral lamella. 
The lower cone-scales which show unusual spore contents are possibly 
abortive megasporangia at the lower part of the cone and are analogous 
with the abortive (?) microsporangia at the apex of the megasporangial cone 
in Mazocarpon (H. Cn. 527, 8 and 9). 
This remarkable specimen so happily interpreted by Kidston goes far 
to indicate that the fertile mega- and microsporangia of Sigillaria were at 
least sometimes borne on different cones, but showed vestiges, respectively 
above or below, of the cone-scales of the other type (cf. Text-fig. 4). This 
agreement with a very difficult specimen strengthens the evidence that 
Mazocarpon is Sigillaria , since the exceptionally bulky microsporange of 
Mazocarpon agrees in both form and size with specimens already regarded 
as microsporangia of Sigillariostrobus by Dr. Kidston. 
Part III. General Discussion of Results. 
a. Biological aspects of the new structure. 
b. Possible bearing of the work on the affinity of Pleuromoia and 
Isoetes and of genera of Lepidodendraceae. 
c. General conclusions. 
Section XII. 
a. Biological aspects of the new structure . 
A few words might be devoted to the biological aspect of the Mazo¬ 
carpon type of megasporange. The presence in some sporangia of ruminat¬ 
ing ‘ spores 5 which lie in a relatively large mass of sporogenous tissue 
(Fig. 15) would seem to indicate that the sporange is derived from an 
ancestral form with a larger number of functional spores. The decrease in 
the number appears to have been correlated with a limited special increase 
in the size of the prothallus, for when only one is developed in a sporange it 
has only attained the normal size (Fig. 5) found in an eight-spored sporange. 
The seed habit is approached in two ways: (a) the megaspores germinate 
within the sporange and (b) the sporange undergoes a certain amount of 
vegetative development. Fertilization is, however, impossible until frag¬ 
mentation of the sporange has taken place, owing to the centroscopically 
directed archegonia. Thus a new factor is introduced and the lamella 
. 1 Kidston, loc, cit., Plate II, Fig, i. 
