from extant embryophytes and in situ from early land plant megafossils (see review in
Gray 1985; Wellman and Gray 2000).
Somewhat later, in a ground-breaking insight, Gray and Boucot (1971) realized that
spores with rather peculiar morphology (united in permanent tetrads) recovered from
early Silurian deposits were derived from land plants. The first occurrence of such
tetrads, and hence the earliest evidence for land plants, was subsequently extended back
into the Caradoc (Ordovician) (Gray et al. 1982) and then the Llanvirn (Ordovician)
(Vavrdova 1984; McClure 1988; Gray 1993; Strother et al. 1996). Clearly such an
observation suggests a significant disparity between the first reported microfossil and
megafossil evidence for early land plants. It is likely that this is a reflection of
preservational bias. It is not until lignified tissues with high preservation potential evolved
(possibly coincident with the evolution of vascular plants) that the fossil record of
megafossils begins.
Subsequent research has demonstrated that there are a number of distinct non-trilete
spore morphotypes that dominate the early land plant dispersed spore record. These have
been collectively termed cryptospores (see Richardson et al. 1984; Richardson 1988,
1996; Strother 1991; Steemans 2000). They include monads, dyads, and tetrads, that are
either naked or enclosed within an envelope, and the separated products of dyads (hilate
cryptospores). Initially the affinities of cryptospores was controversial, with many authors
disputing a land plant origin. Subsequent research, however, has provided compelling
support for land plant affinities (recently reviewed in Wellman and Gray 2000). More
precise relationships have been suggested for some of the morphotypes, with Gray (1985,
1991) arguing persuasively that permanent tetrads derived from bryophyte-like, or more
precisely, liverwort-like, plants.
At this juncture it is helpful to consider the type of spore record one might expect from
the earliest land plants. There is strong evidence that bryophytes are the most basal land
plants. Consequently an examination of the spores of extant bryophytes might provide a
search image for the type of spores expected of the earliest land plants. Gray and
subsequent workers (Gray 1985; Wellman and Gray 2000) have compared the spores
comprising the early dispersed spore fossil record with those of extant bryophytes. While
there are some obvious similarities regarding morphology and ultrastructure, it should be
borne in mind that many extant bryophytes produce rather non-descript spores that occur
singly and are small and thin walled. The latter appear to have no counterpart in the early
dispersed spore fossil record, even though taxon composition has been described in its
entirety for certain Ordovician and Silurian spore assemblages from non-marine deposits
(i.e. if such spores were a component of the spore assemblage they would have been
observed). In addition, dispersed spore assemblages from the Llanvirn to Llandovery (an
interval of some 40 myr duration) are very similar in composition, exhibiting virtual stasis
in terms of morphological/ taxonomic composition (Wellman 1996), and the spore
morphologies possessed by a great many extant bryophytes simply are not present.
Interestingly there are very few dispersed spores recognized as bryophyte in origin
throughout the entire dispersed spore fossil record (reviewed in Traverse 1988). Either
such spores are not preserved (perhaps they are too thin walled) and/or are a derived
feature of bryophytes. I consider the latter explanation a distinct possibility, and contend
that the Ordovician-Lower Silurian spore record reflects ‘ancestral’ spore morphologies.
130 CHARLES H.WELLMAN