Gymnosperms

 

by Aljos Farjon, RBG Kew updated January 2003

PLANTS WITH NAKED SEEDS BORNE IN CONES OR ON SPIKE-LIKE AXES, RARELY FULLY EXPOSED ON SHORT SHOOTS OR ON FOLIAR APPENDAGES. STEMS WITH SECONDARY XYLEM. MANY EXTINCT FORMS; AT PRESENT INCLUDING CYCADS, GINKGO, CONIFERS AND GNETALES

The gymnosperms are an ancient, heterogeneous, polyphyletic assembly of plants which can be defined as seed plants lacking an ovary. Whereas in angiosperms, the ovules are enclosed by an ovary constructed of carpels, which develops into a fruit containing the seeds, in gymnosperms the ovules are not so enclosed and the ovules and seeds are consequently borne directly on short shoots, terminal axes, scales or foliar structures (megasporophylls). The ‘seed habit’ first developed in plants with fern-like leaves but it is likely that this happened more than once in different groups of pteridophytes, leading to different ancestors of present groups of gymnosperms. During the long history of gymnosperms, many groups evolved and became extinct and only four major groups, Cycads, Ginkgo, Conifers and Gnetales, survive. The following synopsis illustrates the heterogeneity of gymnosperms, both fossil and extant.

The gymnosperms as defined above include the earliest seed plants, which appeared in the Upper Devonian. Presumably, the first evolutionary steps taken involved both vegetative and reproductive modifications and both were adaptations to life on land away from permanent resources of water. Some tree fern-like plants developed ‘gymnospermous’ secondary wood but retained heterosporous (i.e. with male microspores and female megaspores) free sporing reproductive organs (sporangia) and a gametophyte that developed free from the parent plant (sporophyte). An example is the extinct plant Archaeopteris. In some other phylogenetic lineages, these sporangia further diversified, whereby the male sporangia remained dehiscent and free sporing, but the female sporangia became indehiscent and fertilization took place within the female sporangium. In one well studied evolutionary modification series, sterile telome trusses (groups of bifurcated appendages) united gradually to form a cupule around the megasporangium; this complex evolved further to form the integument of the seed. By the Lower Carboniferous, fern-like plants (some were climbers) with fully evolved seeds borne on fertile pinnae (hydrasperman seed ferns) had appeared. Later in the Carboniferous, there was a radiation of pteridosperms (seed ferns), among which medullosan seed ferns (Medullosaceae) were a dominant group. ‘Higher’ seed ferns evolved along with other forms, e.g. Peltaspermales and some lasted until the Lower Cretaceous. The seed ferns are a polyphyletic group representing several evolutionary lineages which evolved independently, especially since the separation of Gondwana and Laurasia.

One order (or class) of present-day gymnosperms, the cycads or Cycadales (Cycadatae),

is distantly related to one of the (unknown) seed ferns. They have an ancient history beginning in the Permian and flourished in the age of Dinosaurs, the Mesozoic. Some forms looked strikingly different from extant cycads, e.g. with simple, lobed leaves or fertile leaves with numerous seeds. Three extant families: Cycadaceae, Zamiaceae and Stangeriaceae remain, the first with only a single genus, and with a total of over 200 species. They are still widespread, but scattered and nowadays often endangered, and largely confined to the tropics and subtropics. Most are large plants with unbranched or sparsely branched trunks (sometimes subterranean) which contain a large pith, wood elements and an outer cover of leaf scars. The leaves are spirally arranged, large, pinnate fronds that give most species a palm-like appearance. All species are dioecious and the reproductive organs are leaf homologues (sporophylls) arranged in cones (strobilus, plural strobili) except in female Cycas (Cycadaceae, the most primitive of the cycads), where they are organised in subapical whorls like the leaves. The female sporophylls of Cycas resemble reduced leaves, those of other genera and all male sporophylls (microsporophylls) are scale-like or peltate structures but were likewise derived from leaves. Microsporophylls bear numerous pollen sacs; female or macrosporophylls bear 2-8 ovules that develop into large, often brightly coloured seeds. Pollination of many, perhaps of most, species involves transfer of pollen by weevils or other beetles. A primitive trait of cycads is their motile sperm, which they share only with Ginkgo among extant gymnosperms. Most cycads are poisonous and/or have developed ‘mechanical’ deterrents such as spiny pinnae to defend themselves against being eaten.

Cyacadaceae Cycas revoluta megasporophylls (cultivated specimen)
Cycas revoluta Cycas revoluta megasporophylls Cycas revoluta seed

The Ginkgoales (Ginkgoatae)

are only represented today by Ginkgo biloba, the Maidenhair tree, almost extinct in the wild in China, but widely cultivated, in Europe since the 18th century, in China and Japan much longer. They are large, much branched trees with woody stems and flat, fan-shaped leaves, consisting of two halves with separate, dichotomous venation, with long petioles that are borne on short shoots. The species is dioecious, with catkin-like pollen cones with spirally arranged microsporophylls and wind dispersed pollen that produce motile spermatozoids. The ovules are paired and originate on what are perhaps rudimentary scales or leaves subterminal to long peduncles, the often single seeds are shaped like cherries and resemble fruits. The earliest fossils are known from the Permian, perhaps they evolved from now extinct early gymnosperms. Ginkgoales were more diverse and widespread in the Mesozoic, while the present species may already have existed in the Cretaceous. Their phylogenetic position in relation to other (extant) gymnosperms is enigmatic.

The Bennetitales (Bennetitatae) or cycadeoids

are an extinct group of gymnosperms (Triassic to Cretaceous) with hermaphrodite (bisporangiate or bisexual), often compound reproductive structures in which the pedunculate ovules, surrounded by interseminal scales, formed a globose cone, which was surrounded by microsporophylls. This assemblage was in turn often subtended by spirally arranged bracts and the whole compound structure was placed axillary to a subtending pinnate or compound leaf, often broken off and only leaving a leaf scar. Stems were short, unbranched and squat, or slender and branched in other groups. The leaves often resembled those of cycads, but the two groups are not closely related. Botanists have long looked at the Bennetitales for possible ancestors of angiosperms due to their hermaphrodite ‘flowers’ and there may indeed be a phylogenetic relationship that is closer than with other groups so far discovered (Crane, Friis & Pedersen, 1995).

Bennetidales: Diagrammatic reconstruction of the bisporangiate, or bisexual, strobilus of Williamsoniella coronata from the Jurassic (Raven, P. et al ' Biology of Plants' 6th ed. 2003. p473)

The Gnetales (Gnetatae)

are currently represented by three distinct families (sometimes themselves classified as orders): Ephedraceae, Gnetaceae and Welwitschiaceae, of which the first two have each one genus with some 40 species and the latter is monotypic. They are distinct from other gymnosperms in having wood vessels instead of only tracheids, but the vessels are unlike those of angiosperms. They are dioecious; the reproductive organs are arranged in catkin- or spike-like or cone-like, axillary whorls or clusters. The male microsporangiophore bears 1-8 pollen sacs around the apex and is surrounded at the base by an enveloping bract, several of these are aggregated in small cones. The ovules are single, orthotropous, with an inner integument that is elongated into an exserted micropylar tube and is surrounded by 1-2 additional envelopes. Ephedra has much reduced scale leaves, which are soon shed, on numerous, slender branches forming shrubs; the genus is widespread in semi-deserts of Eurasia, North Africa and North America. Gnetum forms mostly lianas, sometimes shrubs or small trees, and has opposite, petiolate, large and broad leaves with pinnate-reticulate venation much resembling dicotyledonous angiosperm leaves. These traits are most likely due to convergent evolution. Of several ovules in the initially compound female ‘cone’ only one large seed develops. Gnetum is exclusively tropical and occurs in forests. Welwitschia mirabilis has a mostly underground stem and two, continuously growing (from basal meristems), strap-like, parallel-veined leaves. Male and female cones occur on branched stalks on separate plants that often grow far apart, insect pollination is very likely involved and still being researched. This strange plant is endemic to the Namib Desert in SW Africa. Gnetales share several features with each other despite apparent great differences in habit and vegetative organs. They are evidently relicts of a more diverse group of plants, but the fossil record is very incomplete, although characteristic pollen can be traced back as far as the Triassic. Phylogenetic relationships are uncertain, with most of the molecular evidence supporting a relationship with conifers and the morphological evidence indicating a relationship as sister group to angiosperms. They probably are not closely related to either.

pollen cones pollen cone
Ephedra equisetina seed cones Ephedra equisetina seed cones Ephedra equisetina pollen cones Ephedra equisetina pollen cones
pollen sacs seed cone uppermost seed cone scales tubular micropyle
Ephedra equisetina pollen sacs Ephedra likiangensis seed cone Ephedra likiangensis uppermost seed cone scales Ephedra likiangensis tubular micropyle

Coniferales (Pinales, Pinatae) or conifers

are at present the most diverse and largest group of gymnosperms with eight families and ca. 630 species. Conifers first appeared in the Carboniferous, with an array of primitive types in which the pedunculate ovules were subtended by sterile leaflets or scales (cataphylls) and were growing from the axil of spirally arranged bracts on terminate axes. They were all medium-sized to large, branched and woody trees. Perhaps one of these ancestral groups were the Cordaitales, with sometimes enormously long, strap-like leaves and very loosely arranged, paniculate reproductive structures, but some palaeobotanists have considered them a ‘dead-end’ lineage not related to conifers. The problem is that conifers, while as extant group ‘monophyletic’ in relation to other extant gymnosperms, may not have been similarly coherent in the distant past. The ‘walchian’ conifers (e.g. Voltziaceae, Majonicaceae) were either directly ancestral to some of the modern conifer families, e.g. Araucariaceae or Pinaceae, or they represent a more distant and long extinct clade from which now also extinct ancestors of modern conifers arose. The palaeobotanist Rudolf Florin hypothesised an evolutionary development of the seed cone involving these Permian and Triassic conifers that led to present-day cone types as seen in Araucariaceae and Pinaceae. He thereby solved a problem that had puzzled botanists for more than a century: the homologies of the female conifer cone. His insights were later refined, both by palaeobotanists who discovered new fossils and by botanists who used refined techniques such as electron microscopes (not available to Florin). Essentially, the evolution involved reduction of the ovuliferous peduncles or dwarf shoots, reduction and fusion of subtending scales and enlargement of a single remaining axillary scale, which is homologous to the leafy dwarf shoot that bore the ovule(s) in the ancestral conifers. Extreme further reduction and specialization has led to single-seeded ‘cones’ with fleshy ‘receptacles’ as in the Podocarpaceae, or to the disappearance of the ovuliferous scale altogether, as in Cupressaceae (now including Taxodiaceae). In Taxaceae, which Florin considered to be a separate order from conifers, the ovuliferous scale has been replaced by an arillous envelope or soft outer layer that partly or entirely covers the single seed. In all conifers, the male cones or pollen cones are simple, catkin-like structures that have remained remarkably constant. Air-borne dispersal of pollen is universal in conifers and has given little selective pressure to evolve differently in the many lineages leading to the extant conifers. In contrast, the seed cones are more specialised and vary considerable in size, shape and structure, from tiny single-seeded ‘berries’ to woody compound cones the size of a football. A variety of seed dispersal strategies have evolved, several are using animal vectors. Conifers were probably most diverse and ubiquitous in the Mesozoic, from the Triassic through to the Late Cretaceous, when angiosperms became the dominant plants in most parts of the world. Yet, conifers, unlike most other gymnosperms, are still very successful as a group, even though many species are relicts of a more abundant occurrence in the geologic past. They occupy many diverse habitats on most continents (in Antarctica they are only represented as fossils) and among them we count the largest, the tallest and the oldest plants on the planet.

Araucariaceae Araucaria araucana in Chile Araucariaceae Araucaria araucana in Chile Araucariaceae Araucaria araucana in Chile Araucariaceae Araucaria araucana in Chile
Podocarpaceae Dacrycarpus dacrydioides in New Zealand Podocarpaceae Dacrydium cf. bidwillii in New Zealand Podocarpaceae Dacrydium cf. bidwillii in New Zealand Podocarpaceae Lepidothamnus laxifolius in New Zealand
Podocarpaceae Phyllocladus alpinus in New Zealand Podocarpaceae Phyllocladus alpinus in New Zealand Podocarpaceae Podocarpus alpinus in New Zealand Podocarpaceae Podocarpus alpinus in New Zealand

Modern palaeobotanic, systematic and phylogenetic analyses emphasize the complexity of gymnosperm evolution. These ‘primitive’ seed plants have existed for nearly 350 million years and many evolutionary lineages became extinct, sometimes leaving very little evidence of their existence. The remainder, highly diverse, may not at all be closely related and is unlikely to form a coherent, ‘monophyletic’ group that can be traced to a single nearest ancestor. They seem to share little else than the absence of an ovary, something Theophrastus (ca. 300 BC) may already have understood when he observed:

Some seeds again are enclosed in a pod,
some in a husk, some in a vessel,
and some are completely naked

 

Aljos Farjon, FLS
Taxonomist of gymnosperms
Chairman of IUCN/SSC Conifer Specialist Group
Herbarium, Royal Botanic Gardens, Kew
Richmond, Surrey, TW9 3AB
a.farjon@rbgkew.org.uk

Pictures sven Landrein

Bibliography

Beck, C. B. (ed.) (1988). Origin and evolution of gymnosperms. Columbia University Press, New York.
Crane, P. R., E. M. Friis & K. R. Pedersen (1995). The origin and early diversification of angiosperms. Nature 374: 27-33.
Enright, N. J. & R. S. Hill (eds.) (1995). Ecology of the Southern Conifers. Melbourne University Press, Melbourne.
Farjon, A. (1990). Pinaceae. Drawings and descriptions of the genera Abies, Cedrus, Pseudolarix, Keteleeria, Nothotsuga, Tsuga, Cathaya, Pseudotsuga, Larix and Picea. Regnum Vegetabile Vol. 121. Koeltz Scientific Books, Koenigstein, Germany.
Farjon, A. (2001). World Checklist and Bibliography of Conifers. Second edition, Royal Botanic Gardens, Kew.
Farjon, A. (to be published 2003/04). A monograph of Cupressaceae and Sciadopitys. Royal Botanic Gardens, Kew & I.A.P.T., Vienna.
Florin, R. (1951). Evolution in Cordaites and Conifers. Acta Horti Bergiani 15 (2): 285-388.

Jones, D. L. (2002). Cycads of the world. Second edition, Reed Publishers, Chatswood, New South Wales and Smithsonian Institution Press, Washington, D.C.
Norstog, K. J. & T. J. Nicholls (1997). The biology of the cycads. Cornell University Press, Ithaca, New York.
Richardson, D. M. (ed.) (1998). Ecology and biogeography of Pinus. Cambridge University Press, Cambridge, U.K.
Sporne, K. R. (1965). The morphology of gymnosperms. Hutchinson University Library, London.
Stewart, N. S. & G. W. Rothwell (1993). Palaeobotany and the evolution of plants. Second edition, Cambridge University Press, Cambridge.

 

Home page

Cycadales

Ginkgoales

Bennetitales

Gnetales

Coniferales

Practical

Bibliography

Aljos Farjon presentation (pdf)

Practical

Pinaceae Larix decidua

 

     

-

Order

Family

Species Number

Gymnosperms

Cycadales

Cycadaceae

105

Gymnosperms

Cycadales

Zamiaceae

200

Gymnosperms

Gingkoales

Gingkoaceae

1

Gymnosperms

Pinales

210

Gymnosperms

Pinales

Araucariaceae

33

Gymnosperms

Pinales

Phyllocladaceae

5

Gymnosperms

Pinales

Podocarpaceae

125

Gymnosperms

Pinales

Sciadopityaceae

1

Gymnosperms

Pinales

Cupressaceae

133

Gymnosperms

Pinales

Taxaceae

30

Gymnosperms

Gnetales

Ephedraceae

65

Gymnosperms

Gnetales

Welwitschiaceae

1

Gymnosperms

Gnetales

Gnetaceae

30


PINACEAE Larix decidua

Range: Europe
European Mountains


male cone

Dissect a female and male cone. Observe the ovules, ovuliferous scales and reduced bracts and how they develop in seed.

male cone cone section Bracts and ovuliferous scale in the seeding stage
cone seed
Bract and ovuliferous scale ovuliferous scale
sections of the female cone
 

Practical

Pinaceae Larix decidua