« 
598 THE AMERICAN NATURALIST. [VoL. XXXII. 
has already given us a very accurate account of the development of 
the sporangia in the lycopods, Equisetineæ, and Ophioglossacee, 
together with most important conclusions as to the origin of the 
different sporangial types and the relations of these to one another. 
In the present paper he has taken up the second order of the 
Eusporangiate, the Marattiacee, and has given us by far the 
most complete account of the sporangia of these interesting ferns 
that has ever been published. All of the existing genera are treated 
at length, and in addition there is a most valuable discussion of nS 
relation of these to the different fossil types. 
The Marattiacez comprise at present four genera, two of them 
monotypic, of tropical ferns of very characteristic structure. Of 
these, Marattia is represented in both the Old and New World, but the 
others are more restricted in range. Danza is peculiarly American; 
Angiopteris and Kaulfussia, each with but a single species, belong 
to the Old World. 
The Marattiacezee show many primitive structural characters, and 
it is now known that most of the palæozoic ferns were closely related 
to existing marattiaceous types. Owing to the difficulties in procur- 
ing suitable material for studying the development of the sporangia, 
the earlier studies on these were mostly fragmentary, and entirely 
confined to the two genera Marattia and Angiopteris. This makes 
the careful study here given of the sporangia of Danza and Kaul- 
fussia of more than common interest. 
In all the Marattiaceez except Angiopteris the individual sporangia 
are imperfectly delimited, and the sorus is often spoken of as a 
“‘synangium,” although it is much more probable that this is the 
primitive condition than a case of cohesion of originally free 
sporangia. Bower very properly considers each group of sporogenous 
cells as a single sporangium, and speaks of it as a sporangium. 
The development of the sporangium is much alike in all of the 
genera. The sporogenous cells arise, as a rule, from a single hypo- 
dermal cell, whose sister-cell forms part of the wall of the ripe 
sporangium. Exceptions occur and it is not always possible to refer 
the sporogenous complex to the division of a single mother-cell. 
In Danza the sori are much elongated, and almost completely 
cover the lower surface of the sporophyll, extending from the midrib 
to the margin, and almost or quite touching each other laterally. In 
this genus the sporophylls have the segments decidedly smaller than 
the sterile leaves, and in this respect Danza recalls many leptospo- 
rangiate ferns, or the Ophioglossaceew. The occasional presence of 
