EVOLUTION OF THE FERN FAMILY OSMUNDACEAE BASED ON ANATOMICAL STUDIES

68 Pages·2006·3.85 MB·English

CONTRIBUTIONS FROM THE MUSEUM OF PALEONTOLOGY
THE UNIVERSITY OF MICHIGAN
VOL. 23, NO.8 , p. 105-169 (2 pls., 10 text-figs.) FEBRUARY22 , 1971
EVOLUTION OF THE FERN FAMILY OSMUNDACEAE
BASED ON ANATOMICAL STUDIES
BY
CHARLES N. MILLER, JR.
Department of Botany, University of Montana, Missoula 59801
MUSEUM OF PALEONTOLOGY
THE UNIVERSITY OF MICHIGAN
ANN ARBOR
CONTRIBUTIONS FROM THE MUSEUM OF PALEONTOLOGY
Director: ROBERTV . KESLING
The series of contributions from the Museum of Paleontology is a medium for the
publication of papers based chiefly upon the collection in the Museum. When the number
of pages issued is sufficient to make a volume, a title page and a table of contents will be
sent to libraries on the mailing list, and to individuals upon request. A list of the separate
papers may also be obtained. Correspondence should be directed to the Museum of
Paleontology, The University of Michigan, Ann Arbor, Michigan 48104.
VOLS.2 -22. Parts of volumes may be obtained if available. Price lists available upon inquiry.
1. The rodents from the Hagerman local fauna, Upper Pliocene of Idaho, by Richard J.
Zakrzewski. Pages 1-36, with 13 text-figures.
2. A new brittle-star from the Middle Devonian Arkona Shale of Ontario, by Robert V.
Kesling. Pages 37-51, with 6 plates and 2 text-figures.
3. Phyllocarid crustaceans from the Middle Devonian Silica Shale of northwestern Ohio
and southeastern Michigan, by Erwin C. Stumm and Ruth B. Chilman. Pages 53-71,
with 7 plates and 4 text-figures.
4. Drepanarter wrighti, a new species of brittle-star from the Middle Devonian Arkona
Shale of Ontario, by Robert V. Kesling. Pages 73-79, with 2 plates.
5. Corals of the Traverse Grou-p of Michigan. Part 13, Hexagonaria, by Erwin C. Stumm.
Pages 81-91; with 4 plates.
6. The Pliocene rodent Microtoscoptes disjunctus (Wilson) from Idaho and Wyoming, by
Claude W. Hibbard. Pages 95-98, with 2 text-figures.
7. A new microtine rodent from the Upper Pliocene of Kansas, by Claude W. Hibbard.
Pages 99-103, with 1 plate and 1 text-figure.
EVOLUTION OF THE FERN FAMILY OSMUNDACEAE
BASED ON ANATOMICAL STUDIES
CHARLES N. MILLER, JR.
Department of Botany, University of Montana, Missoula 59801
ABSTRACT-The fern family Osmundaceae has only 16 living species, but nearly 100 fossil
species have been named on the basis of various detached organs. Most important of these are
29 species based on petrified axes which provide a partial but substantial record of the family
from the Late Permian to the present. These axes and some of the Recent species have been
important in the formulation of concepts of stelar evolution and in our understanding of the
evolutionary relationships of living ferns, but an intensive investigation of the evolution of
the species of the Osmundaceae has never been made. Thus, the purpose of the present work
is to investigate the phylogenetic interrelationships of the species of the family as interpreted
from anatomical and morphological characters of their stems, leaf-bases, and roots.
Types or representative materials of 14 Recent and 21 fossil species were examined for
variable characters. Information about nine additional species of petrified axes came from
descriptions in the literature. These species were then analyzed by the ground plan method of
multiple character correlation in three samples: (1) analysis of 43 species based on six
characters, (2) analysis of 31 species based on 13 characters, and (3) analysis of 14 Recent
species based on 18 characters. By this means the species were arranged according to lines of
specialization, and when these lines are considered with respect to geologic time, nine groups
of phylogenetically related species are resolved. Five of the groups are represented by the
present-day genera and subgenera and their fossil members. Three groups are subdivisions of
Osmundacaulis, and the remaining group includes the extinct protostelic forms.
The results show large gaps between these groups. This indicates that the fossil record
of the family contains only a small number of those species that existed in the past. It is
evident, however, that the modern species evolved from protostelic ancestors by intrastelar
origin of a pith and subsequent (or concurrent) dissection of the xylem cylinder by leaf-gaps.
The living species and most of the fossil forms are products of slight modifications of the
dictyoxylic-siphonostele plan of organization. However, two of the groups of species became
highly specialized in the Mesozoic but died out near the close of that era. The Recent genera
Todea and Leptopteris are as distinct from one another as each is from the subgenera Osmun-
da, Osmundastrum, or Plenasium, but their fossil records are not sufficient to indicate their
origin. Osmunda, Osmundastrum, and Plenasium were distinct lines of evolution in the Paleo-
cene, and Osmunda and Plenasium are represented by foliage remains in the Cretaceous. No
definite source for Osmunda and Osmundastrum can be determined, but Plenasium shares a
number of characters with Osmundacaulis atherstonei and 0. natalensis, both of which are
from the Lower Cretaceous, and they may have evolved from forms similar to these species.
The Osmundaceae probably arose during Early Permian or Pennsylvanian time from
forms similar to Grammatopteris baldaufi or G. rigolloti, but neither of these species can be
considered a precursor of the family. Further, no ancestors of other fern taxa can be found
amonq the known Osmundaceae.
Taxonomically, the protostelic forms are treated in the subfamily Thamnopteroideae
with the remaining species forming the subfamily Osmundoideae.
CONTENTS
Introduction ............................................................................................................................................................................. 107
Ackn.o wled.g ments ...................................................................................................................................................................1 07
Classlficatlon ............................................................................................................................................................................. 108
History of investigation ....................................................................................................................................................... 110
Materials and methods ........................................................................................................................................................... 111
Anatomy and morphology ................................................................................................................................................... 112
. -.
Stem .................................................................................................................................................................................. 112
Stele ........................................................................................................................................................................... 112
Cortex ......................................................................................................................................................................... 118
Leaf-trace ................................................................................................................................................................. 119
Petiole base ....................................................................................................................................................................... 122
Root .................................................................................................................................................................................. 124
Stem branching ............................................................................................................................................................... 124
Growth habit ................................................................................................................................................................... 125
106 CHARLES N . MILLER. JR .
Systematic descriptions based on stem. root. and leaf-base characters ...................................................................... 125
Order Filicales ........................................................................................................................................................................ 125
Family Osmundaceae ............................................................................................................................................................ 125
Key to the subdivisions based on anatomical characters ........................................................................................ 125
Subfamily Osmundoideae ....................................................................................................................................................... 127
Genus Osmunda ............................................................................................................................................................... 127
Subgenus Osmunda ......................................................................................................................................................... 127
0. r.e galis. ................................................................................................................................................................... 127
0. japontca ............................................................................................................................................................... 128
0 . h.nc.ea. .... ............................................................................................................................................................... 128
0 .. tlzaenszs ................................................................................................................................................................. 128
0 claytoniana ......................................................................................................................................................... 129
0. nathorstii ............................................................................................................................................................ 129
0 . oregonensis ........................................................................................................................................................... 129
0 . pluma ................................................................................................................................................................... 130
Subgenus Osmundastrum ............................................................................................................................................... 130
0 . cinnamomea ....................................................................................................................................................... 130
0. precinnamomea ................................................................................................................................................... 130
Subgenus Plenasium ....................................................................................................................................................... 131
0 . banksiaefolia. ........................................................................................................................................................ 131
0 .. b. rome.l iaefolia ..................................................................................................................................................... 131
0 javanzca ............................................................................................................................................................... 131
0 . vachellii ............................................................................................................................................................... 132
0 . arnoldii ................................................................................................................................................................. 132
0 . dowkeri ................................................................................................................................................................. 132
Genus Todea ................................................................................................................................................................... 132
T . barbara ................................................................................................................................................................. 133
Genus Leptopteris ........................................................................................................................................................... 133
L . fraseri ................................................................................................................................................................... 133
L .. hymenophylloides ............................................................................................................................................... 133
L su.p erb.a .................................................................................................................................................................1 34
L . wtlkeszana ............................................................................................................................................................. 134
Genus Osmundacaulis ..................................................................................................................................................... 134
Osmundacaulis herbstii group ..................................................................................................................................... 134
0 . herbstii ................................................................................................................................................................. 134
0 . sahnii ................................................................................................................................................................... 135
00 .. dgzu.b nblto.a pnia .................................................................................................................................................................................................................................................................................................................... ....... 113356
0 . patagonica ........................................................................................................................................................... 136
0 . kolbei ................................................................................................................................................................... 136
0 . kidstoni ................................................................................................................................................................. 137
Osmundacau.li .s braziliensis group ................................................................................................................................. 137
00 .. cbararnzzz.lez rez.n s.i..s. ................................................................................................................................................................................................................................................................................................................... ..... 113378
Osmundacaulis skidegatensis group ............................................................................................................................. 138
0 . skidegatensis ....................................................................................................................................................... 138
0 . atherstonei ..........................................................................................................................................................1. 39
0 . natalensis .............................................................................................................................................................1 39
Subfamily Thamnopteroideae ............................................................................................................................................... 139
Genus Anomorrhoea ......................................................................................................................................................1. 39
A . fischeri ................................................................................................................................................................. 139
Genus Bathypteris ........................................................................................................................................................... 140
B . rhomboidea ......................................................................................................................................................... 140
Genus . Cha.sm.ato.pter is ................................................................................................................................................... 140
C pnnczpalzs ........................................................................................................................................................... 140
Genus Zegosigopteris ....................................................................................................................................................... 140
Z. javorskti ................................................................................................................................................................. 141
Genus Petcheropteris ...................................................................................................................................................... 141
P . splendida ............................................................................................................................................................... 141
Genus Thamnopteris ....................................................................................................................................................... 141
T . schlechtendalii ..................................................................................................................................................... 141
T . gwynne-vaughani ............................................................................................................................................... 142
T . kidstoni ............................................................................................................................................................... 142
Genus Zaless.k .y a ............................................................................................................................................................... 142
Z . graczlzs ........................................................................................................................................................... 142
Z . diploxylon ........................................................................................................................................................... 143
Z . uralica ................................................................................................................................................................... 143
Zncertae sedis ..................................................................................................................................................................... 143
Ztopsidema vancleavei ............................................................................................................................................. 143
Taxonomic considerations ............................................................................................................................................. 144
Other osmundaceous fossils ................................................................................................................................................... 144
Rhizomes and trunks ..................................................................................................................................................... 144
Fructifications ................................................................................................................................................................. 146
FERN FAMILY OSMUNDACEAE 107
Isolated spores ................................................................................................................................................................. 146
Foliage ............................................................................................................................................................................... 147
Species relationships ............................................................................................................................................................... 148
Principles of the analysis ................................................................................................................................................. 148
The analyses ................................................................................................................................................................... 150
Analysis of all species ..................................................................................................................................................... 152
Analysis of the Osmundacaulis kolbei line .............................................................................................................. 152
Analysis of extant species ........................................................................................................................................... 155
Discussion 157
Phylogenetic relationships ....................................................................................................................................................1. 58
Phylogeny of the Osmundaceae ................................................................................................................................... 158
Thamnopteris group ...............................................................................................................................................1 58
Osmundacaulis herbstii group ...............................................................................................................................1 58
Osmundacaulis braziliensis group .........................................................................................................................1 60
Osmundacaulis skidegatensis group ....................................................................................................................... 161
Plenasium group ....................................................................................................................................................... 161
Osmunda group ....................................................................................................................................................... 161
Osmundastrum group ............................................................................................................................................. 162
Leptopteris group ..................................................................................................................................................... 163
Todea group .............................................................................................................................................................1 63
Ori.g i.n of other fern taxa from the Osmundaceae ....................................................................................................... 163
Ongin of the Osmundaceae ..........................................................................................................................................1. 64
Summary of phylogenetic relationships ..................................................................................................................... 166
Su.m .m ary ...................................................................................................................................................................................1 66
Blbllography ............................................................................................................................................................................. 167
INTRODUCTION ceae, Schizaeaceae, and the Coenopteridales
occur in the Late '~aleozoic,a nd the Coenop-
THE OSMUNDACEAisE a primitive family of
teridales apparently did not survive into the
ferns that is assigned by most authorities to the
Mesozoic. The Gleicheniaceae and Schizaeaceae
Filicales. The family is important because it is
are known in the Paleozoic only by sporangia
intermediate between the Eusporangiatae and
and some questionable foliage remains (Arnold,
Leptosporangiatae in many respects and be-
1964). Early marattiaceous ferns already show
cause it has the most extensive fossil record of
the specialized synangia that are characteristic
any fern family (Arnold, 1964). Only sixteen
of many of the Recent species, and the poly-
living species are generally recognized, but
cyclic stele of the Psaroniaceae suggests that
nearly one hundred fossil forms have been
members of this group were too specialized to
named (Jongmans & Dijkstra, 1962). About
have served as a probable source of other fern
fifty of the fossil species are based on com-
taxa. Thus, the Paleozoic Osmundaceae, which
pressed foliage, about twenty species are repre-
have a very simple protostelic organization,
sented only by isolated spores and sporangia,
seem likely as a possible ancestral stock for
and thirty species are represented by petrified
certain other fern groups.
stems. The last show that the Osmundaceae - .
Clarification of the phylogenetic relation-
was a distinct line of evolution by Late Perm-
ships within the Osmundaceae is basic to
ian time, though sporangia that may be os-
understanding the role of the family with re-
mundaceous occur in the Coal Measures of
gard to the possible derivation of other fern
Great Britain, France, and Asia Minor (Seward,
groups from it, and this treatment is an attempt
1910).
to place the structurally preserved remains of
While the Osmundaceae is most often in-
the Osmundaceae in phylogenetic series with
cluded in the Filicales, it has many features in
the present-day species. While relationships are
common with the Eusporangiatae, and other
interpreted mostly from comparative characters
characters that are intermediate between com-
of rhizome anatomy, information from osmun-
parable eusporangiate and leptosporangiate
daceous foliage, spores, and sporangia is used
conditions. Furthermore, several attributes are
as additional source of evidence where avail-
completely unique to the Osmundaceae and
suggest long isolation of the family (table 1). able.
This peculiar mixing of characters and the
ACKNOWLEDGMENTS
great antiquity of the family has led to specu-
lation that certain groups of "higher ferns" This work represents a portion of a disser-
may have evolved from the Osmundaceae rather tation submitted to The University of Michigan
than from other ancient ferns. Besides the Os- in partial fulfillment of the requirements of the
mundaceae, only the Marattiales, Gleichenia- Ph.D. degree in Botany. The Department of
CHARLES N. MILLER, JR.
Botany and the Botanical Gardens of The Uni- Throughout the study Dr. C. A. Arnold has
versity of Michigan are gratefully acknowl- been a reliable source of aid and encouragement
edged for the financial support, supplies, and and I also wish to thank Dr. W. H. Wagner for
facilities they provided during the investiga- allowing me to draw upon his knowledge of
tion. In addition, officials of the British Mu- living ferns and for help in applying multiple
seum (Nat. Hist.), London, and the Depart- character correlation. I am deeply indebted to
ment of Botany, University of Glasgow, Scot- my wife Susan for the many sacrifices she made
land, allowed me to examine fossil material in to facilitate my work.
their collections. Dr. W. H. Hewitson, College
of the Pacific, Stockton, California, contributed CLASSIFICATION
a large collection of rhizomes of Recent species The living species of the Osmundaceae are
remaining from his investigation (1962), and placed in three genera: Leptopteris, with six
numerous other people supplied material of species; Osmunda, with nine species; and
living species (table 2). Todea, which is monotypic. Leptopteris was
My special thanks are extended to the mem- originally treated as a subgenus of Todea, but
bers of my doctoral committee for many con- Presl (1845) separated the two. Although some
structive suggestions regarding my work. workers favor the original grouping (Seward &
Other Taxa Having
Osmundaceous Character
Character
1) Massive photosynthetic thallus with thick midrib Marattiaceae and primitive leptosporangiate
(Bower, 1926). ferns.
2) Gametangia superficial (Bower, 1926). Leptosporangiatae.
3) Archegonia borne in two parallel rows, one to either None
side of the midrib (Atkinson & Stokey, 1964, p. 66).
4) Archegonia with six tiers of neck cells Intermediate between eusporangiate and lepto-
(Sporne, 1962, p. 144). sporangiate situations.
5) Large number of spermatocytes produced per Intermediate between eusporangiate and lepto-
antheridium (Bower, 1926). sporangiate situations.
6) Spores chlorophyllous. Various Grammitidaceae and Hymenophylla-
ceae (also Equisetum) .
7) Spores trilete. Various fern taxa, primitive and otherwise.
8) Dehiscence of antheridium by forcible ejection of Primitive Leptosporangiatae.
cap cell (Atkinson & Stokey, 1964).
9) Spore germination bipolar (Atkinson & Stokey, 1964). None.
10) Number of spores per sporangium 128-512 Intermediate between eusporangiate and lepto-
(Bower, 1926). sporangiate situations.
11) Embryo prone (Bower, 1926). Leptosporangiatae.
12) Quadrants of embryo situated with respect to thallus Intermediate between eusporangiate and lepto-
but not to archegonium (Bower, 1926). sporangiate situations.
13) Embryo persisting in globular stage (Bower, 1926). Eusporangiatae.
14) Shoot apex, foot, root, and leaf not referable to Eusporangiatae.
distinct portions of embryo (Bower, 1926).
15) Persistence of endodermis into mature parts of Leptosporangiatae.
sporophyte, rather than disappearance after
juvenile stage (Bower, 1926).
16) Stipular expansions on petiole base (Bower, 1926). Eusporangiatae, Plagiooriaceae (also
.
Mattezrccia pensylvanica)
17) Transverse commissure on stipule in Leptopteris Eusporangiatae (also Cycadaceae) .
and Todea (Bower, 1926).
18) Indefinite segmentation of leaf-blade from single Marattiaceae.
row of marginal initials (Bower, 1926).
19) Root initials varying betweer.:
a) tetrahedral, and Leptosporangiatae.
b) four, truncated prismatic cells, (Bower, 1926). Marattiaceae.
20) Sporogenous cells:
a) cubical, or Eusporangiatae.
b) tetrahedral (Bower, 1926). Leptosporangiatae.
21) Sporangium not always referable to single Intermediate between eusporangiate and lepto-
initial (Bower, 1926). sporangiate situations.
FERN FAMILY OSMUNDACEAE 109
Ford, 1903 ; Tansley, 1907), Presl's classifica- ences in venation (Hewitson, 1962). Recent
tion is preferred here. Leptopteris has filmy evidence from the fossil record indicates that
fronds, while the lamina of Todea barbara is the subgenera of Osmunda have been separate
coriaceous. Sporangia are borne on the abaxial lines of specialization since the Paleocene (Mil-
surface of typical vegetative fronds in both ler, 1967).
genera. Osmunda differs in this respect, the Fossil forms have always been assigned to
sporangia being attached to specialized pinnae Recent taxa where possible, but because of lack
that lack a lamina. The degree of fertile and of knowledge of the structure of many of the
sterile frond dimorphism varies from complete living species, most of the petrified remains of
-the entire fertile frond consisting of speci- the Osmundaceae were originally assigned to
alized pinnaeto incomplete-the fertile frond form genera. Recent work has permitted the
bearing mainly typical vegetative pinnae with a transfer of a number of fossil species to living
few fertile pinnae inserted in apical, medial, or genera (Hewitson, 1962; Chandler, 1965; Mil-
basal position on the rachis. Osmunda is tradi- ler, 1967), but most of the petrified organs
tionally divided into the subgenera Osmunda, belonging to the family remain in form genera.
Osmundastrum, and Plenasium, on the basis of Osmundacaulis, a nomen novum replacing Os-
frond morphology. Species of the subgenus mundites Unger, which is invalid (Miller, 1967),
Osmunda have bipinnate fronds while the fronds includes rhizomes or trunks or their parts which
are once-pinnate in Plenasium and Osmunda- bear structural resemblance to the present-day
strum. Pinnules in Osmundastrum are pinnati- species but which cannot be referred to any of
fid while the margins of Plenasium pinnules are the Recent genera. Certain ancient species differ
entire to coarsely serrate. There are also differ- significantly from the more recent forms in hav-
Rhizomes:
Leptopteris fraseri M. Tindale Sydney, N.S.W.
L. hymenophylloides G. Baylis Dunedin, New Zealand
C. Quinn Auckland, New Zeaiand
L. superba G. Baylis Dunedin, New Zealand
C. Quinn Auckland, New Zealand
L. wilkesiuna J. Parham Suva, Fiji
Osmunda regalis C. Miller Florida and Michigan, U.S.A.
0.j aponica T. Chung Seoul, Korea
S. Kokawa Osaka, Japan
0.l ancea S. Kokawa Osaka, Japan
0.c laytoniana C. Miller Michigan, U.S.A.
0.c innumomea C. Miller Florida and Michigan, U.S.A.
0.b anksiaefolia K. Iwatsuki Kyoto, Japan
0.j avanica N. Seneviratne Gurutalawa, Ceylon
Todea barbara M. Tindale Sydney, N.S.W.
Spores (my number in parentheses) :
Osmunda regalis, Indiana, Gates #61 (1-17-63-2)
Osmzrnda regalis, Texas, Thorp in 1929, (1-17-63-3)
Osmunda regalis, Michigan, Ehlers #45 (1-8-63-8)
Osmunda regalis, Brazil, Mexia #5169 (1-8-63-6)
Osmunda lancea, Japan, Copeland #I318 (1-17-63-1)
Osmunda claytonianu, location unknown, Koelz #I987 (1-8-63-2)
Osmunda claytonianu, Phunkia, India, Flemming #87 (1-8-63-7)
Osmunda claytonianu, Michigan, Voss #6895, (1-15-63-4)
Osmunda claytonianu, Wisconsin, Copeland #I315 (1-15-63-3)
Osmunda claytonianu, Virginia, Tidestrom #4584 (1-15-63-2)
Osmunda claytoniana, Quebec, Canada, Storer #268 (1-15-63-1)
Osmunda cinnamomea, Florida, McFarlin #3313 (1-17-63-6)
Osmundo cinnamomea, Indiana, Gates #I058 (1-17-63-5)
Osmunda cinnumomea, Michigan, Davis in 1891 (1-8-63-1)
Osmunda cinnamomea, Michigan, Voss #%I31 (1-17-63-4)
Osmunda banksiaefolia, Luzon, Philippines, Copeland #I2298 (1-17-63-8)
Osmunda banksiaefolia, Luzon, Philippines, Copeland #22472 (1-17-63-1)
Osmunda javanica, Kwangtung, China, Tsang #21186 (1-8-63-3)
Osmunda vachellii, Canton, Chiia (introduced), McClure #I8580 (1-8-63-5)
CHARLES N. MILLER, JR.
TABL3E-A SSOCIATIONO F Osmundacaulis WITH Ckzdophbbis
Osmumdacaulis Cladobhlebis Horizon Location
0,k olbeil C. denticulata Lower Cretaceous South Africa
0. kolbei C. browniana Lower Cretaceous South Africa
0. patagonica2 C. patagonica Upper Jurassic Argentina
0. dunlopi3 C. denticulata Jurassic New Zealand
0. gibbiana3 C. denticulata Jurassic New Zealand
0. herbstii4 Cladophlebis sp. Upper Triassic Argentina
1) Kidston & Gwynne-Vaughan, 1910.
2) Archangelsky & de la Sota, 1962.
3) Kidston & Gwynne-Vaughan, 1907.
4) Archangelsky & de la Sota, 1963.
ing a protostele, and these species have been the reticulate nature of the xylem cylinder in
placed in Bathypteris, Chasrnatopteris, Zegosi- Osmunda regalis and speculated upon the evo-
gopteris, Petcheropteris, Thamnopteris, and lution of such a stele from a "medullated mono-
Zalesskya. In addition, one Permian species, stele" by dissection of the xylem cylinder by
which is known only by its leaf-bases, is classi- leaf-gaps. Jeffrey (1899) took exception to Van
fied in the genus Anomorrhoea. Itopsidema in- Tieghem & Douliot's (1886) postulated onto-
cludes one siphonostelic species from the Tri- genetic intrastelar origin of the pith and later
assic, but its relationship to the Osmundaceae (1902) extended his opinion to phylogenetic
is questionable. origin as well. He believed that the "medul-
In addition to these taxa based on petrified axes lated monostele" of Van Tieghem was derived
and leaf-bases, over forty species represented from an amphiphloic siphonostele by degenera-
only by foliage have been referred to Osmunda tion of the internal phloem and endodermis.
(table 4). Todites includes fronds from Meso- Accordingly, Jeffrey ( 1902) postulated the evo-
zoic sediments that are similar to those of lution of the living Osmundaceae from a dictyo-
Todea and bear osmundaceous sporangia. Fur- stelic form, such as Osmundacaulis skidegaten-
thermore, some species of Cladophlebis prob- sis, and interpreted Osmunda cinnamomea,
ably also belong to the Osmundaceae. This which has an internal endodermis, as represent-
genus includes sterile foliage from the Meso- ing an intermediate stage. Faull's (1901) study
zoic; and certain species frequently occur in of Osmunda regalis, 0 . claytoniana, 0. cinna-
association with Osmundacaulis of that time trzomea, Leptopteris superba, and Todea bar-
(table 3), but no organic connection between bara, and his (1910) investigation of young
them has been found. sporophytes of Osmunda cinmomea brought
Fossil spores and sporangia that are more forth evidence supporting Jeffrey's concept and
or less similar to those of the living species supplemented it by noting the occurrence of
have been referred to the Recent genera or to internal phloem near the branching region of
numerous form genera. certain plants of 0. cinnammea. Seward &
Ford ( 1903 ) , who examined stems of Leptop-
HISTORY OF INVESTIGATION teris hymenophylloides, L. superba, and Todea
Early investigations of the anatomy of the Os- barbara, thought that these species were primi-
mundaceae dealt with the arrangement, ontogeny tive among the living forms and that the stele
and phylogeny of the pith and vascular tissues, of Osmunda cinnammea represented a spe-
but little attention was given to structural fea- cialization of an ectophloic condition. Material
tures useful in distinguishing between the of their species displayed somewhat confluent
species. The first author to deal with the anat- xylem strands suggestive of Recent origin from
omy of osmundaceous stems was DeBary an ancestor having an ectophloic siphonostele
(1884) who envisoned a series of sympodial with the xylem cylinder uninterrupted by leaf-
"fibro-vascular" strands arranged in a ring gaps.
around the pith as in the conifers, each strand At this time structural details of only three
branching to form leaf-traces but otherwise fossil stems of the Osmundaceae were known.
running independently through the stem. Van These were Osmunda dowkeri, 0. iliaensis
Tieghem & Douliot (1886) described the os- [=Osmundites schemnicensis (Pettko) Unger 1,
mundaceous stele as a "medullated monostele," and Osmundacau2is skidegatensis. The first two
one in which the pith formed ontogenetically are similar to the living species and occur in
by "expansion" of a single, central, "fibro- the Eocene and Miocene-Pliocene respectively;
vascular" strand. In 1895, Zenetti demonstrated but 0. skidegetensis, a dictyostelic form from
FERN FAMILY OSMUNDACEAE 111
the Lower Cretaceous, offered overwhelming sclerenchyma scattered in the inner cortex of
support for the "extrastelar" school of thought. the petiole base, while such strands are absent
Between 1907 and 1914, Kidston & Gwynne- in Leptopteris. Within the genus Osmunda,
Vaughan published a series of five memoirs on species of the subgenus Plenasium differ from
petrified osmundaceous axes, naming two new those of the subgenera Osmunda and Osmun-
species and reinvestigating several other previ- dastrum in having leaf-traces that are deeply
ously described forms whose internal structure, curved at their point of separation from the
and thus their relationship to the Osmunda- stele and have two distinct groups of protoxy-
ceae, was unknown. Five of these were from lem elements at this location. In species of the
the Upper Permian of Russia. The Russian two latter subgenera the xylem of the leaf-trace
forms are the most ancient and the least spe- diverges from the stem xylem as an oblong
cialized known in the family; and when they strand with a single protoxylem group. Osmun-
are considered with the other species treated da and Osmundastrum are most easily distin-
by Kidston & Gwynne-Vaughan, they offer guished from each other on the basis of the pat-
strong evidence for the derivation of the tern formed by thick-walled fibers in the outer
present-day species from protostelic ancestors cortex of the petiole base. In the subgenus Os-
by intrastelar origin of the pith and subsequent munda at some level in the stipular region these
dissection of the xylem cylinder of an ecto- cells are organized in two lateral clusters while
phloic siphonostele by leaf-gaps, a view that is in Osmundastrum two lateral clusters and one
widely accepted today (Posthumus, 1924b; abaxial mass occur at all levels of section. Fur-
Bower, 1926). thermore, Osmunda cinnamomea typically has
Lack of adequate information about the an internal endodermis which is lacking in all
comparative anatomy of the Recent species has Recent species of the subgenus Osmunda.
hindered investigation of the relationships be- These comparative characters and others
tween the fossil species and the living ones. Until to be mentioned later have enabled investiga-
recently, more was known about the internal tors to distinguish between the living genera
structure of the fossil material than the extant of the Osmundaceae, and in 1965 Chandler
species, and comparisons of one with the other transferred Osmundacaulis (Osmundites) dow-
were made only in the broadest sense. (Arnold, keri to Osmunda. Since then seven other species
1964). While subsequent discoveries of petri- of petrified rhizomes have been assigned to
fied stems have doubled the number of species Osmunda, Osmundastrum, and Plenasium show-
known to Kidston & Gwynne-Vaughan, the ing that these subgenera existed as independent
only addition to our knowledge of the living lines of evolution as early as the Paleocene
species was the description of the stele of (Miller, 1967).
Osmunda javanica by Posthumus (1924b).
Hewitson's (1962) survey of thirteen of the MATERIALS AND METHODS
Recent species marked the first significant at- Type material of 22 of the 32 previously
tempt to find characters that would be useful described species of petrified stems and (or)
in comparing the extant and extinct forms. He petiole bases were re-examined during the
concluded that the arrangement of scleren- course of this investigation. Whenever the
chyma tissues in the stipular petiole bases is original or later preparations were not avail-
the most diagnostic feature and that such char- able, special preparations were made for me
acters as the number of strands of stem xylem by curators of the various collections. The 13
and the number of leaf-traces in the cortex, as species in the Kidston Collection were examined
viewed in transverse stem sections, are not in in the Department of Botany, The University
themselves diagnostic but are useful as cor- of Glasgow, Scotland, while six other species
relative evidence. Hewitson (1962) also noted were studied in the British Museum (Nat.
that the outer cortex in rhizomes of Osmunda Hist.), London, England. The remaining spe-
is constructed entirely of very long, thick- cies were studied from specimens and prepara-
walled fibers while in Leptopteris and Todea tions in the Museum of Paleontology, The Uni-
these cells occur only around departing leaf- versity of Michigan, Ann Arbor, Michigan, or
traces with the bulk of the tissue consisting of from loans made to the Museum.
relatively short sclereids. Bower (1926) and Materials of 14 of the 16 Recent species
Chandler (1965) also point out that in Todea were obtained for reference and comparison
and Leptopteris the margins of the stipular with the fossils. Much of this was generously
expansions are joined across the adaxial face contributed by Hewitson at the completion of
of the petiole base by a transverse commissure, his (1962) investigation. These specimens were
and this structure is lacking in Osmunda. Todea already embedded in celloidin. Additional ma-
differs from Leptopteris in having strands of terial (table 2) was acquired by personal col-
112 CHARLES N. MILLER, JR.
lection of the three North American species and inner parenchymatous layer and an outer
from foreign collectors. Living plants were sclerotic layer, but one species has a homo-
maintained in the greenhouses of The Univer- geneous cortex. Petiole bases of most species
sity of Michigan Botanical Gardens until they have pronounced stipular expansions, but they
could be prepared for sectioning. are lacking in one species. No tissue has the
Axes of the living plants were treated ac- same construction in all members of the family;
cording to the method of Hewitson (1962). every tissue has undergone some evolutionary
Petiole bases were removed from the stems, change.
and they and the stems were cut into sections
Stem
0.5-1.0 cm long. The pieces were then softened
-
for two weeks in a 1 : 1 solution of 48% hydro- Stele. Tissue construction. Protostelic
fluoric acid and 98% ethyl alcohol, washed in species, all of which are extinct, have stems
tap water for 24 hours, dehydrated in an ethyl that contain a central column of relatively
alcohol series, and embedded in celloidin (Meth- short tracheids surrounded by a cylinder of
od 2 of Johansen, 1940). Sections were cut on longer tracheids that are more typical of fern
a sliding microtome at 20 to 40 microns de- metaxylem. Since the central column is well
pending on the friability of the material. They differentiated from the outer cylinder, and since
were stained in 1% aqueous solution of Sa- the central xylem shows evidence of evolutionary
franin "0" and 0.5% Fast Green in absolute change toward pith tissue, it is treated here
alcohol and mounted in Canada Balsam. descriptively as a tissue distinct from the outer
Spores of certain Recent species (table 2) cylinder metaxylem.
were obtained from specimens in The Univer- The cells of the central metaxylem tissue
sity of Michigan Herbarium, and voucher labels are large-diameter tracheids whose walls bear
were attached to the herbarium sheets of the elliptical or oval pits. In Zalesskya: grucilis thk
specimens used. The sporangia were washed in tracheids are ten or more times longer than
running water, softened and wetted in 5% broad (pl. 2, fig. 10). Several vertical series of
KOH for two minutes at 90°C, and strained elliptical pits occur on each wall; but the files
through a wire screen to separate spores from are not in strict vertical alignment, so the ar-
sporangia. The spores were then acetolyzed for rangement is termed "irregular scalariform."
three minutes at 90°C, mounted unstained in In contrast, the central xylem tracheids of
glycerine jelly, and the cover slips sealed with Thamnopteris and Zdesskya: diploxylon are
diaphane. A slide of each specimen has been two to five times as long as broad (pl. 2, fig. 9),
placed in the Palynological Collection of the their pits are oval, and the arrangement of pits
Department of Botany, The University of results in a "reticulate" pattern.
Michigan. It is of particular interest that in Thamnop-
teris kidstoni parenchyma cells occur singly or
ANATOMY AND MORPHOLOGY in clusters near the periphery of the central
Structurally, the Osmundaceae is a highly xylem tissue (Zalessky, 1924). They are the
variable group. The majority of the species same size and shape as the tracheids but lack
have a parenchymatous pith, but at least four pits. The association of these cells with the
have a column of tracheids occupying the cen- short central xylem tracheids is important evi-
ter of the stele. Most species have external dence of the intrastelar origin of a pith in the
phloem only, but one has internal as well as Osmundaceae.
external phloem. Likewise, the endodermis in Central xylem is preserved completely to
many species is external to the stele; but several the center of the stem in only four species. In
species have external and internal endodermal other cases a zone of decay occurs in this tissue;
layers, and in at least four species these cylin- and while typical central xylem tracheids border
ders connect through leaf-gaps. The cortex of the vacant zone, other cell types may have oc-
osmundaceous stems generally consists of an curred within. The fact that the central region
FIGS. I, 2--0smunda pluma Miller. I, transverse section of rhizome, X 4%, holotype, UMMP 52157. 2, trans-
verse section of stele, X 24, holotype, UMMP 52157.
3, 4--0smzcnda precinnamomea Miller. 3, transverse section of rhizome, x 5410, holotype, UMMP 52159.
4, transverse section of stele, x 30, holotype, UMMP 52159.
5, 6-Osmunda arnoldii Miller. 5, transverse section of rhizome, X 4%, holotype, UMMP 52161. 6, trans
verse section of stele, X 20, holotype, UMMP 52161.