aL
0 Dh
974

a

COMPOSITAE &@
* NEWSLETTER

Number 25 December 1994

Scientific Editor: Bertil Nordenstam

Technical Editor: Gunnel Wirénius Nohlin

Published and distributed by The Swedish Museum of Natural History,
Department of Phanerogamic Botany,

P.O. Box 50007,

S-104 05 Stockholm, Sweden.

(Director: Prof. Bertil Nordenstam)

ISSN 0284-8422

CONTENTS
Christopher F. Puttock: Re-analysis of Anderberg’s Gnaphalieae data matrix

M.S. Ayodele: Studies on the reproductive biology of Vernonia Schreb.
(Asteraceae). I. Types of inflorescence among different growth habits

M.S. Ayodele: Studies on the reproductive biology of Vernonia Schreb.
(Asteraceae). II. Flowering and post-pollination developments in the
capitulum

F.M. D. Ogbe, L.S. Gill & E.O.0. Iserhien: Effect of aqueous extracts of
Chromolaena odorata (L.) K. & R. on raadicle and plumule growth and
seedling height of maize, Zea mays L.

YR. Ling: The genera Artemisia L. and Seriphidium (Bess.) Poljak. in the
World

Bertil Nordenstam: New combinations in the Calenduleae

Request for material

Comp. Newsl. 25, 1994 1

RE-ANALYSIS OF ANDERBERG’S
GNAPHALIEAE DATA MATRIX

Christopher F. Puttock
Australian National Herbarium Centre for
Plant Biodiversity Research
GPO Box 1600

Canberra ACT, Australia

The cladistic analysis of the Gnaphalieae by Anderberg (1991) and its subsequent
endorsement in Bremer’s monograph "Asteraceae. Cladistics and classification"
(Anderberg 1994), presents an apparently rigorous foundation for future research
on the systematics within this tribe. Authors who use cladistic methodologies to
reconstruct phylogeny will be tempted to select sister taxa as outgroups according
to the published ’phylogenetic’ position of these in relation to their ingroup taxa.
This paper presents a note of caution on accepting the phylogenetic arrangement
of the Gnaphalieae as proposed by Anderberg (1991) and highlights some pitfalls
of analysing large data sets.

Introduction

At the Asteraceae conference at Kew (Compositae: Systematics Biology
Utilization, 24 July-5 August 1994) I proposed that, on the basis of the data
matrix presented by Anderberg (1991, 1992), a significantly different phylogeny
could be obtained. My doubts about the rigour of Anderberg’s original cladistic
analyses were stimulated by my attempts to place a monotypic new genus
Cremnothamnus (Puttock 1994) within this proposed phylogeny. The taxon had
been tentatively placed within the Lawrencella complex (sensu Anderberg 1991)
in the Angianthinae and within Ozothamnus (sensu Wilson in Wilson et al. 1992)
in the Cassiniinae. Commencing with an analysis using Anderberg’s data of the
two subtribes, I found that the introduction of this one new taxon produced many
trees. The strict consensus of these trees indicated that the two subtribes were
polyphyletic. Further experimentation with additional ingroup and outgroup taxa
continued to confound the published phylogeny. Stimulated by the measure of
disbelief that I found at Kew I have now returned to this Gnaphalieae data matrix.

to

Comp. Newsl. 25, 1994

The original analyses

The aim of the Anderberg’s analyses was to produce a phylogenetic arrangement
for the 167 taxa of the Gnaphalieae (sensu Anderberg 1989). According to
Anderberg (1991) he ran five analyses, one tribal and four subtribal, using the
Hennig86 version 1.5 parsimony program. From 143 Gnaphalieae taxa scored,
the data matrix was reduced to 72 taxa for the tribal analysis. These 72 taxa were
"a large selection of taxa representing various characters and character
combinations" ostensibly "To get a clear picture of the phylogeny of the basal
groups, and of the major monophyletic groups that could be identified within the
tribe". Anderberg stated that the excluded taxa "were hypothesised to be close
relatives of taxa included ... based on the notion that they were linked to other
(included) taxa by several characters" (Anderberg 1991:11). The first analysis
resulted in 120 minimal length trees. The chosen tree, rooted by the ’ingroup’
taxon Philyrophyllum, had a ladder of five minor clades (ungrouped basal taxa),
followed by five subtribal clades. Anderberg then performed four further
analyses after re-associating the excluded taxa of the first analysis with their
"close relatives". The outgroups for these subclade (subtribal) analyses were the
basal ingroup taxon, Lawrencella for analyses 3 and 4, and the subclade’s sister
taxon, /xiolaena for analyses 2 and 5.

Review

In Anderberg’s analyses, approximately half of the scored Gnaphalieae taxa were
excluded from the first analysis, then partitioned into the various clades resolved
by that analysis, for subsequent detailed analyses of those clades. This strategy is
sensible, since it reduces the numbers of taxa in each analysis, allowing heuristic
algorithms a better chance of finding a minimal solution. However, the
assumptions on which this exclusion and partitioning process are based need to
be explicit and tested: in Anderberg’s analyses, they were neither.

The most important assumption made by Anderberg is that the clades discovered
in the first analysis remained monophyletic after the extra taxa had been
partitioned into them. That is, the extra taxa should share the synapomorphies on
which the clade is based. However, Anderberg’s first analysis produced low
consistency and retention indices (0.28 and 0.68 respectively). This is evidence
that considerable homoplasy exists in the data, a result that is to be expected for
large morphological data sets (Sanderson & Donoghue 1989). However, it casts
doubt on the ability to find suitable homes for the excluded taxa - few if any
unequivocal synapomorphies are available for defining the main clades, so few
unequivocal placements of the excluded taxa can be made on the basis of
synapomorphy. Instead, Anderberg relied on intuitive, non-explicit assessments
of relationships of the excluded to included taxa. He may have been right, of
course, but the large advantages of cladistic methodology are its rigour and
explicitness, and these were severely compromised in his analysis by this step.

Comp. Newsl. 25, 1994 3

Nevertheless, there are two ways in which his intuitive assessments may be
rigorously tested: 1) the subclade analyses can be set up in such a way that they
allow the assumptions to be falsified; or 2) a final, global analysis, with all taxa
and using a starting tree derived from a composite of trees from the first analyses,
can search for equally minimal or shorter topologies.

1. Any group of taxa, whether monophyletic or not, will form some topology
under a cladistic analysis. If a single outgroup taxon is included in the analysis
and the tree rooted on the node between the outgroup and the remaining taxa (the
ingroup), then the ingroup will appear monophyletic, since it cannot appear any
other way. Assembling a group of taxa into an ‘ingroup’ constitutes an
assumption of monophyly, and if only a single outgroup taxon is used that
assumption of monophyly can never be tested. If several outgroup taxa are used,
however, the possibility exists that a topology will be found in which the ingroup
cannot be monophyletic (e.g. see Fig. 1), thus testing and falsifying the initial
assumption. Anderberg used single outgroup taxa (either a sister taxon chosen
from the initial analysis or the basal member of the clade from the first analysis)
in all his subclade analyses.

2. If the results obtained from the first round of analyses are each minimal
solutions for their respective clades, they should be combinable into a full tree
which will itself be a globally minimal solution for the entire data set. This can
be tested by searching for solutions that are shorter than the full tree, either by
conventional ’top-down’ heuristic searches, or by using the full tree as the
starting point for a round of branch-swapping. If the assumptions of monophyly
used to construct the several subclade ingroups were correct, these global
analyses will be unable to find any shorter solutions which violate the integrity of
any of the subclades. If, however, they are incorrect, alternative or better
placements may be found for some taxa, within other subclades in which they
were never tried. Anderberg did not attempt such global tests of any of his
subtribe topologies.

Methodology

Both methods for testing Anderberg’s assumptions of monophyly have been
applied here.

1. Global analysis

A composite Gnaphalieae matrix was assembled from the published data
matrices: the first 10 taxa from Table 1 (Philyrophyllum to Millotia), all taxa of
Tables 2, 3 and 5 (Anderberg 1991) and all Table 4 (Anderberg 1992), in the
sequence in which they were published. Characters 83 and 84, absent from Table
1, were scored as unknown. Character 21 was coded for three states throughout
with the first 10 taxa scored as 0. As in Anderberg’s subtribal analyses multistate
characters were left unordered and unweighted. Analyses were performed on a
Macintosh LC II using PAUP version 3.1.1 (Swofford 1993) and on a DX486 PC
using Hennig86 (Anderberg used Hennig86 throughout). Anderberg’s first

4 Comp. Newsl. 25, 1994

analysis computed the data with 16 multistate characters coded as ordered. All
other characters were coded as unordered. In the four subsequent analyses all
characters were treated as unordered. The treelength of Anderberg’s complete
phylogenetic hypothesis of Gnaphalieae was reconstructed from his Figures
11-14 using MacClade version 3.01 (Maddison & Maddison 1992). This
treelength was used as the reference for comparison with the new analyses.

All analyses used heuristic algorithms, since exhaustive searches were precluded
by the size of the data matrix. Five ’dumb’ analyses used the simple addition
sequence algorithm, each starting with a different reference taxon from among
the basal members of each of the five subtribes. Each analysis was allowed to
find 100 minimal trees (i.e. MAXTREES = 100) and then these 100 trees
searched for shorter trees by the TBR branch-swapping algorithm to completion.
These analyses are termed ‘dumb’ because they make minimal a priori
assumptions as to the structure of the data or the ’real’ tree topology. Two
smart’ analyses used reasonable guesses as to the true tree topology as starting
points for branch swapping. The first of these (analysis 6) started with a tree
reconstructed from Anderberg’s Figures 11-14. The second (analysis 7) was run
incorporating consistent arrangements found within the original five analyses on
the consensus tree of the first ’smart’ analysis. These two analyses were allowed
to find 200 minimal trees each and search for shorter trees by the TBR
branch-swapping algorithm to completion. In all seven analyses Athrixia,
Printzia, Macowania, Arrowsmithia and Phagnalon, in addition to
Philyrophyllum (Anderberg 1991), were hypothesised as sister taxa to the rest of
the Gnaphalieae (see note below: Outgroup clade).

2. The monophyly of the Cassiniinae

The monophyly of one of Anderberg’s subclades, the Cassiniinae, was tested by
four outgroup taxa: Philyrophyllum, Ixiolaena and Gnaphalium were used in
addition to Lawrencella used by Anderberg. These taxa were chosen on the basis
that Philyrophyllum was the outgroup taxon for Anderberg’s tribal analysis,
Ixiolaena was used as the outgroup for the second and fifth analyses, Lawrencella
was used for the third and fourth analyses, and finally Gnaphalium because it was
a member of a sister subtribe to the Cassiniinae clade. The full compliment of 21
taxa assigned by Anderberg to the Cassiniinae comprised the ingroup taxa.
Heuristic analyses were run using PAUP and Hennig86 with all characters states
unordered. The results of these analyses are presented first.

Results

1. The monophyly of the Cassiniinae

The Cassiniinae cladogram for 21 ingroup taxa and four outgroup taxa
reconstructed in the sequence depicted in Anderberg’s Figures 6 and 12 had a
length of 150 steps.

Comp. Newsl. 25, 1994 5

Both PAUP and Hennig86 analyses of the Cassiniinae found the same 24 trees
with a minimum length of 140 steps (ci = 0.43, ri = 0.59). In none of these trees
were the Cassiniinae monophyletic with respect to all four outgroup taxa. To
make the cladogram convex (Estabrook 1978) required a tree with an additional
ten steps. The illustrated 90% majority rule consensus cladogram (Fig. 1) is
rooted by Philyrophyllum and Ixiolaena as the outgroup. The three nodes
marked with an asterisk collapse in the strict consensus at 92%. With
Philyrophyllum and Ixiolaena as the outgroup, Apaloclamys and Acanthocladium
are sister taxa to the clade of Lawrencella and Argyroglottis in 22 trees. In the
remaining 2 trees Acanthocladium and Raoulia were the basal taxa of the
Lawrencella-Argyroglottis clade. These taxa and the clade of Basedowia and
Ammobium are sister taxa to the rest of the Cassiniinae including Gnaphalium.
Gnaphalium was always the basal taxon of the internal subclade of Anaphalis and
Anaphaloides. In Fig. 1 the polytomy, from Gnaphaliothamnus-Chionolaena to
the Gnaphalium subclade, forms a ladder in 75% of trees in the upwards
sequence shown.

2. Global analysis

The reconstructed cladogram for 143 taxa based on the published phylogeny of
the Gnaphalieae (Anderberg 1991) had a length of 598 steps excluding 29
collapsed branches forming polytomies (i.e. the analysis started with a treelength
of 627). ;

The five analyses using Lawrencella (Angianthinae), Raoulia (Cassiniinae),
Leucogenes (Gnaphaliinae), Psychrophyton (Loricarinae) and Antithrixia
(Relhaniinae) as starting taxa, each taking between 30 and 64 hours computer
time, produced minimal treelengths 587, 587, 590, 590 and 590 steps
respectively. Although none of these analyses produced the shortest trees
eventually found, the results were informative because each subtribe is likely to
better resolve the clade incorporating starting taxon.

Analysis 6, which branch-swapped on Anderberg’s original (598-step), produced
200 minimal trees of length 583 steps. Within several clades of the consensus of
this analysis there were ladders of taxa. In an attempt to test these areas several
taxa were rearranged according to the resolutions found in the first five analyses.
This rearranged tree was used as the starting point for a second round of
branch-swapping (analysis 7) which found trees of length 583 steps but no
shorter. To determine if this analysis was sampling trees from the same place in
the cladistic-treescape as in the previous analysis, these 200 trees were compared
with the 200 trees sampled from analysis 6. The 50% majority rule consensus
cladograms of both sets of 200 trees differed very little in topology although all
400 trees were unique. The resultant consensus cladogram of 400 trees is
presented in Figures 2 to 4. Nodes of 90% or better are indicated as resolved.
Those nodes of lower frequency are shown as polytomies.

6 Comp. Newsl. 25, 1994

In all analyses the outgroup was monophyletic (Fig. 2) and Anderberg’s sequence
of “basal taxa" reading up from Philyrophyllum to Phagnalon is reproduced.
From this point the results presented here diverge from the Gnaphalieae
cladograms of Anderberg (1991). Five main clades were resolved, the Lucilia
group and the subtribes Relhaniinae, Loricariinae, Angianthinae and
Gnaphaliinae (including the Cassiniinae pro parte). The remaining basal taxa
from Anderberg’s analysis were dispersed elsewhere. Jxiolaena, Millotia,
Denekia and Artemisiopsis generally clustered near Apalochlamys and Calomeria
at the base of the Angianthinae.

Outgroup clade

Anderberg’s tribal analysis was not strictly by outgroup comparison. He
considered all taxa in the analysis, including Philyrophyllum, as part of the
Gnaphalieae (ingroup). Should the analysis be rooted by an ancestor at node A
(Fig. 2) these five basal taxa would form a discrete clade as shown here. The
Gnaphalieae clade proper is defined by entirely separated stigmatic lines.
Phagnalon and Philyrophyllum have the plesiomorphic state (also found in the
Inuleae sens. strict. and the Plucheeae) of stigmatic lines apically confluent,
whilst the remaining four basal taxa have stigmatic lines apically converging. In
the analysis of the Inuleae sens. lat. Anderberg (1989) used only two of these
basal taxa (Phagnalon and Athrixia) which occur in the reverse order in his later
analysis (Anderberg 1991). Therefore, Philyrophyllum appears to be untested to
warrant its position as the single outgroup taxon. There are no unique
synapomorphies for Philyrophyllum within the matrix since the pollen sexine
character has a reversal in Printzia (cf. Anderberg 1991: Fig. 6) and the stigmatic
morphology has a state reversal in Phagnalon.

Angianthinae clade

This clade (Fig. 3) consisted of all taxa included in Anderberg’s Angianthinae
with the exception of Bracteantha which was always excluded. In the upper part
of the Angianthinae clade Ammobium and Basedowia (ex Cassiniinae) formed a
sister clade to Thiseltonia and Hyalosperma in all trees. The Waitzia and
Angianthus clades were subdivided and interposed in a small number of trees.
The ungrouped Angianthinae taxa, forming a basal ladder in Anderberg’s
analysis, were completcly reversed to become a distinct clade, with the common
ancestor of Craspedia and Neotysonia in the basal position and Lawrencella at
the top. Jdolaena, Millotia (ungrouped) and Apaloclamys (ex Cassiniinae)
formed a clade in the majority of the trees (54%), sometimes also incorporating
Denekia and Artemisiopsis. A reduced analysis of the Angianthinae clade taxa
was mostly well resolved and had a similar topology to the 90% consensus
cladogram (Fig. 3).

Comp. Newsl. 25, 1994 7

Denekia and Artemisiopsis clade

Denekia and Artemisiopsis formed a sister clade to the Angianthinae and the
Gnaphaliinae or a sister clade to the Angianthinae alone (62%, shown as a
polytomy in Fig. 2). Denekia and Artemisiopsis also formed part of the Ixiolaena
clade in a low percentage of trees.

Gnaphaliinae (including Cassiniinae) clade

This clade (Fig. 4) contains the majority of Anderberg’s Gnaphaliinae and
Cassininae. Excluded from the Gnaphaliinae proper is the Lucilia clade and
from the Cassiniinae are Apalochlamys, Argyroglottis and Raoulia. The three
main differences in the Gnaphaliinae clade topology are the result of
redistribution of the Cassiniinae. Acanthocladium is always the sister taxon to
the rest of the Gnaphaliinae. The subclade of Anaphalis and Anaphaloides in the
majority of trees are placed immediately above the Helichrysum subclade. The
remainder of the Cassiniinae form a polytomy a small clade in the upper part of
the Gnaphaliinae. The taxa from Tenrhynea to Langebergia form a ladder in
60-75% of tree (the reverse of Fig. 12 in Anderberg 1991), leading to the 100%
resolved remainder of the Cassiniinae (Petalacte to Ixodia). A reduced analysis
of only the Gnaphaliinae clade sensu amplo also had poor resolution and a
similar topology to the clade illustrated (Fig. 4).

Loricariinae clade

All members of the Loricariinae were supported by the analysis and maintained
the same laddered arrangement (Fig. 2) illustrated by Anderberg (1991: Fig. 11).

Relhaniinae clade

All members of the Relhaniinae were supported by the analysis (Fig. 2).
Anderberg recognised two groups, the Relhania and Metalasia groups, although
the inclusion of Antithrixia in the Relhania group is unfounded on this data as it
was in Anderberg (1991: Fig. 11, 1994: Fig. 16-2). Antithrixia, always basal to
the Metalasia group, was placed as the basal taxon of the Relhaniinae in a small
percentage of trees.

Lucilia group clade

Six members of this clade were placed in the Gnaphaliinae in Anderberg’s
analysis. The Lucilia clade, now including Bracteantha, Argyroglottis and
Raoulia, is consistently placed outside the clade containing the Angianthiinae,
Denekia and Artemisiopsis, and remainder of the Gnaphaliinae (Fig. 2). In the

8 Comp. Newsl. 25, 1994

analysis of the Inuleae sens. lat., Anderberg (1989) placed Lucilia in a separate
clade from the Gnaphaliinae and noted a possible affinity with Argyroglottis, a
result supported in the analysis. The results using Gamochaeta as the outgroup
for the analysis of the Lucilia group (Anderberg & Freire 1991) should be
re-examined using Raoulia, Argyroglottis and Phagnalon as the outgroup.
However, the topology from this current analysis indicates that the hypothesised
phylogeny may remain unchanged.

The ’infratribal’ polytomy

An analysis of representative taxa from all of the major clades did not provide
any further resolution to the basal polytomy.

Conclusions

Those of us who are currently revising various generic groups within the
Gnaphalieae do not want to detract from the phenomenal amount of work
encapsulated in the treatment of the Gnaphalieae by Anderberg (1991). It has
produced a starting point to a modern systematic study of the Gnaphalieae that
will be referred to for many years to come. The re-analysis presented may not be
more correct than the original one. Anderberg may have been able to intuitively
select taxa for the original analysis on information not carried by the data matrix.
Nevertheless, it has been shown that the same data can produce a different
phylogenetic hypothesis than that already published, and with a markedly shorter
treelength.

The results presented here have some important implications in the choice of
outgroups for generic analyses and the selection of taxa for reduced analyses.
Care must be taken to choose outgroup taxa that are truly sister taxa to the
ingroup being investigated. In this paper I have defaulted to use the ingroup’s
Inuleae-like taxa (sens. strict.) as the outgroup - following assumptions based on
Anderberg (1989). The use of these taxa as the outgroup, three of which were not
included in that analysis, has not been tested within the whole Inuleae sens. lat.
Such an analysis would necessarily introduce new data which goes beyond my
intention of this paper - a re-analysis of the Gnaphalieae data matrix (Anderberg
1991, 1992). Finally, when analysing large data sets, previously excluded taxa
must be tested to ascertain their relationship to the ingroup. Without such testing
new phylogenetic hypotheses, with their taxonomic and nomenclatural
implications, will not stand for very long.

Comp. Newsl. 25, 1994 9

Acknowledgements

I thank Dr Kevin Thiele for helpful discussions on the cladistic approaches and
constructive criticism of the manuscript, and Dr Ilse Breitwieser and Ms Joy
Everett for their reviews. This research, which forms the background to revisions
of genera of the Australian Cassiniinae, was supported by an Australian
Biological Resource Study grant.

References

Anderberg, A.A. 1989. Phylogeny and reclassification of the tribe Inuleae
(Asteraceae). Canadian Journal of Botany 67: 2277-2296.

Anderberg, A.A. 1991. Taxonomy and phylogeny of the tribe Gnaphalieae
(Asteraceae). Opera Botanica 104: 1-195.

Anderberg, A.A. 1992. Cladistics of the Gnaphalieae, additional data.
Compositae Newsletter 20/21: 35.

Anderberg, A.A. 1994. Tribe Gnaphalieae. /n: Bremer, K., Asteraceae.
Cladistics and classification, pp. 304-364. Timber Press, Portland.

Anderberg, A.A. & S.E. Freire 1991. A cladistic and biogeographic analysis of
the Lucilia group (Asteraceae, Gnaphalieae). Botanical Journal of the
Linnaean Society 106: 173-196.

Estabrook, G.F. 1978. Some concepts for the estimation of evolutionary
relationships in systematic botany. Systematic Botany 3: 146-158.

Maddison, W.P. & D.R. Maddison 1992. MacClade. Analysis of phylogeny
and character evolution Computer software program. Sinauer Associates,
Sunderland.

Puttock, C.F. 1994. Anatomy and morphology of Cremnothamnus (Asteraceae),
a new genus for Helichrysum thomsonii F. Muell. Australian Systematic
Botany 7:569-583.

Sanderson, M.J. & M.J. Donoghue 1989. Patterns of variation in levels of
homoplasy. Evolution 43: 1781-1795.

Swofford, D.L 1993. PAUP: Phylogenetic analysis using parsimony, Version
3.1 Computer software program. [Illinois Natural History Survey:
Champaign.

Wilson, P.G., Short P. & A. Orchard 1992. Some nomenclatural changes in
the Angianthinae and Cassiniinae (Asteraceae: Gnaphalieae). Muelleria 7:
519-524.

10 Comp. Newsl. 25, 1994

FIGURES

Fig. 1. 90% majority rule cladogram of 21 Cassiniinae taxa and four outgroup
taxa: Philyrophyllum [OUT] = tribal outgroup, /xiolaena [UNG] = Gnaphalieae
ungrouped, Lawrencella [ANG] = Angianthinae, and Gnaphalium [GNA] =
Gnaphaliinae. Nodes present in only 92% of trees marked with an asterisk (*).

Fig. 2. 90% majority rule cladogram of the Gnaphalieae - excluding the
Angianthinae clade (Fig. 3) and the Gnaphaliinae clade (Fig. 4). (LOR =
Loricariinae; REL = Relhaniinae; LUC = Lucilia group; = Angianthinae; @ =
Cassiniinae).

Fig. 3. 90% majority rule cladogram of the Gnaphalieae continued - the
Angianthinae clade (@= Cassiniinae).

Fig. 4. 90% majority rule cladogram of the Gnaphalieae continued - the
Gnaphaliinae clade (@ = Cassiniinae).

Comp. Newsl. 25, 1994

Ixodia
Odixia
Haeckeria
Ozothamnus
Cassinia
Anaphaloides
Anaphalis
Gnaphalium [GNA]
Antennaria
Ewartia
Ewartiothamnus
Langebergia

* Anaxeton
Petalacte

*p— Gnaphaliothamnus

Chionolaena
Ammobium
Basedowia
Raoulia
Lawrencella [ANG]
Argyroglottis
Acanthocladium
Apalochlamys
Ixiolaena [UNG]
Philyrophyllum [OUT]

Fig. 1

12

—~™ Fig.3 ANG

m~ Fig.4

REL

LUC

OUT

LOR

GNA

Fig . 2

Comp. Newsl. 25, 1994

Angianthus @
Calomeria®@
Apalochlamys®@
Ixiolaena
Millotia
Denekia
Artemisiopsis
Gnaphalium
Acanthocladiume
Raouliopsis
Mniodes
Sinoleontopodium
Loricaria
Pterygopappus
Psychrophyton
Planea
Metalasia
Atrichantha
Calotesta
Hydroidea
Phaenocoma
Dolichothrix
Elytropappus
Stoebe
Disparago
Amphiglossa
Lachnospermum
Bryomorphe
Antithrixia
Relhania
Oedera

Rosenia
Leysera
Oreoleysera
Lucilia

Belloa

Facelis

Berroa
Chevreulia
Cuatrecasasiella
Raoulia®
Argyroglottis®
Bracteantha @
Phagnalon
Arrowsmithia
Macowania
Athrixia
Printzia
Philyrophyllum

Comp. Newsl. 25, 1994

Fig. 3

Pogonolep1s
Hyalochlamys
Gnephosis
Eriochlamys
Calocephalus
Angianthus
Blennospora
Pleuropappus
Cephalosorus
Epitriche
Leptotriche
Chthonocephalus
Siloxerus
Decazesia
Hyalosperma
Thiseltonia
Ammobium ®@
Basedowia®
Gilruthia
Leucophyta
Actinobole
Leptorhynchos
Podolepis
Asteridia
Gratwickia
Chrysocephalum
Waitzia
Triptilodiscus
Acomis
Rutidosis
Bellida
Isoetopsis
Quinetia
Lawrencella
Schoenia
Erymophy1lum
Cephalipterum
Rhodanthe
Pithocarpa
Quinqueremulus
Gilberta
Craspedia
Neotysonia
Dithyrostegia
Polycalymma
Myriocephalus
Calomeria
Apalochlamys @
Ixiolaena
Millotia
Denekia
Artemisiopsis

13

14

Fig. 4

Comp. Newsl. 25, 1994

Psilocarphus
Stylocline
Ancistrocarphus
Bombycilaena
Micropus
Cymbolaena
Evacidium
Logfia

Filago
Micropsis
Troglophyton
Vellerophyton
Gnaphalium
Euchiton
Stuartina
Gamochaeta
Stuckertiella
Lasiopogon
Trichogyne
Ifloga
Edmondia
Syncarpha
Helichrysopsis
Ixodia®
Odixia®
Haeckeria®
Ozothamnus @
Cassinia®
Chionolaena®
Anaxeton @
Petalacte®
Langebergia®
Ewartia®
Antennaria®
Ewartiothamnus @
Tenrhynea
Gnaphaliothamnus ®
Plecostachys
Anaphalis®
Anaphaloides®
Pseudognaphalium
Achyrocline
Helichrysum
Stenophalium
Leontopodium
Leucogenes
Galeomma
Stenocline
Catatia
Syncephalum
Acanthocladium@

Comp. Newsl. 25, 1994 15

STUDIES ON THE REPRODUCTIVE
BIOLOGY OF VERNONIA SCHREB.
(ASTERACEAE)

I. Types of inflorescence among different
growth habits

M.S. Ayodele*
Botany Department
Obafemi Awolowo University
Ile-Ife, Nigeria

Abstract

Eight distinct types of capitula arrangements were identified among plants from
15 species of Vernonia collected from different ecological habitats.

Inflorescence types in arboreous species enhanced terminally clustered small-size
capitula. The large-size capitula of the shrubby species and some herbs, showed
weak but noticeable clustering.

The herbaceous species were more variable in their pattern of inflorescence. The
variability was associated with the type of habitat of the species. In all cases,
adequate presentation of flowers to would-be pollinators was ensured.

Introduction

The genus Vernonia Schreb. of the tribe Vernonieae (Asteraceae) is present in a
wide array of ecological habitats (Hutchinson & Dalziel 1963, Faust 1972, Jones
1979 & 1981, Ayodele 1987).

Jones (1979), citing Gleason (1923) and Cabrera (1944), emphasised the possible
value of inflorescence types as a character in Vernonia. Heywood et al. (1977)
called attention to the significance of morphological variations in relation to
adaptation to habitats and pollinators (i.e. the adaptive significance of
morphological features, related to floral mechanism and fruit dispersal).

* Current mailing address: P.O.Box 586, Ilesa, Nigeria

16 Comp. Newsl. 25, 1994

The purpose of this paper is to present observed variations in inflorescence
among the different growth habits of species of Vernonia collected and studied in
Nigeria. This will be done with a view to elucidating the possible reproductive
strategies associated with the different types of inflorescence.

Materials and Methods

Observations on field populations and collection of Vernonia material were done
during series of field trips covering the different ecological locations in Nigeria.

Close examinations of flowering shoots of Vernonia plants were made on field
populations, garden and screen-house plants. This was in order to observe and
record the arrangement of the capitula on branches and branchlets in each of the
species studied. A special note was taken on the effectiveness of the display of
Vernonia flowers to a prospective external pollinator.

The arrangement of capitula on the panicle was noted and diagrammatically
illustrated. Where possible, photographs of specimens were taken.

Observations

The capitulum inflorescence is strictly an aggregation of flowers (florets) on a
common receptacle. The arrangements of these capitula are similar to the basic
patterns of single-flower arrangements found outside the Asteraceae. However,
no single type of arrangement of capitula in the species studied, absolutely fitted
into the previous descriptions of inflorescence types by Hutchinson & Dalziel
(1963), Olorode (1984) and Gill (1988). The variations in Vernonia cut across
orthodox types of descriptions.

For convenience, an orthodox description or a combination of such descriptions
best suitable for recording observations from 15 species of Vernonia (Fig.1 ) are
given below:

1. Single (solitary) capitulum arrangement: - A single capitulum terminates the
main flowering axis of a plant (Fig. 1:13, V. perrottetii). Sometimes there is a
short lateral branch carrying one other capitulum (Fig.1: 15, V. purpurea).

2. Simple cymose arrangement: - A collection of a few (2-3) large capitula on a
panicle. An initially short peduncle (carrying the older capitulum) developed
from the axil of a terminal leaf. The younger capitulum is bore by a branch from
the peduncle (now elongated) and is subtended by the ’stalk’ of the older
capitulum. The elongating peduncle becomes extended in such a way as to hoist
the younger capitulum fairly higher up than the older capitulum (Fig. 1:4, 6 & 11;
Fig. 2: A,C & D; Fig. 3:8, V. tenoreana, V. galamensis and V. kotschyana).

Comp. Newsl. 25, 1994 17

3. Compound cymose arrangement: - Many simple cymes on an axis (Fig. 1:5 &
12, V. stenostegia and V. ambigua).

4. Racemose arrangement: - Capitula with short simple peduncles or almost
sessile. Both types of capitula are arranged alternately along an elongated axis
like a spike. This constitutes the type of branches formed along a main
inflorescence rachis. On each branch the older capitula are terminally located (an
inverted raceme condition). Similarly, the older branches are terminally located
on the inflorescence rachis (Fig. 1:1, Fig. 3:D, V. conferta). This was a unique
arrangement among the species studied.

5. Racemose arrangement of cymes: - A branch of inflorescence bears two other
branchlets with elongated peduncles. Both subsidiary branches terminate in 2-3
capitula which are younger than the terminal capitula of the mother branch. The
terminating capitula on both of the subsidiary branches are usually of equal size
and age (Fig. 1:7, V. glaberrima).

6. Raceme of umbels arrangement: - The main inflorescence axis is divided at its
apex into three short branchlets, each terminating in a cluster of capitula with
equal (or about equal) length of ’pedicels’. An overall treeshape inflorescence is
produced (Fig. 1:8, Fig. 3:A, V. biafrae). There are usually many inflorescence
branches carrying hundreds of capitula clusters on a branch of the plant. These
capitula are usually of the same size and age.

7. Corymbose arrangement of cymes: - Capitula development occurs at the apex
of the inflorescence axis. This is followed by the development of another axis
from the axils subtending the older (mother) axis. The ’pedicels’ of the capitulum
are of varying lengths, such that the clustering of the capitula is ensured. A
moderately flat-topped inflorescence is thus formed. The capitula in a cluster
were usually found to be at or about the same stages of development (Fig. 1:2,
Fig. 2:B, V. amygdalina and Fig. 1:9, V. migeodi, the latter species bearing fewer
clustering capitula).

8. Umbellate arrangement of helicoid cymes: - Each branchlet of the
inflorescence is a helicoid cyme, with all succeeding axes developing from bracts
on the same side of the preceeding axes. The branchlets are arranged on the main
peduncle in an umbellate form. A cluster of capitula thus terminates the shoot of
the plant. Capitula are usually numerous on each plant and at different stages of
development (Fig. 1:10 & 3, Fig 3:C, V. cinerea & V. colorata). A simplified
form with fewer capitula clusters is represented by V. nestor (Fig.1:14).

Discussion

Aboreous species (Fig. 1:1-3) generally had the type of inflorescence which
enhanced the clustering of capitula at the terminal part of the entire shoot system,
well above the foliage. The flower-heads clustered to form a rather flat or conical
shape (in some cases) and readily presented their showy flowers to visiting

18 Comp. Newsl. 25, 1994

pollinators (Fig. 2:8). This arrangement highly favours reproductive activities in
such plants commonly found in closed communites of the forest zone habitat.

The shrubby species produced their flower-heads amidst leafy stem branches.
The big size of the capitula of V. tenoreana and V. kotschyana made them
noticeable to pollinators. This is further augmented by a weak, but not readily
discerned clustering, resulting from the elongation of the pedicels and bringing
together of capitula on different branches of the same plant. This is also true of
the herbs with large size capitula (V. galamensis, Fig. 1:11, Fig. 2:C, Fig. 3:8).

The herbaceous species were more variable than the arboreous or shrubby species
in their pattern of inflorescence. The herbs were represented in the majority of the
classes of inflorescence pattern described above. This ranged from the solitary
capitulum type as in V. perrottetii to the multiple capitula cluster as in V. cinerea
(Fig. 1:13 and 10, respectively).

Clustering of floral parts in Vernonia was reported to be associated with
smallness of capitulum (Ayodele 1992). The degree of clustering of capitula
observed among the species varied, and reduced as the size of capitula increased.
The advantage in the strategy of capitula clustering is related to the effectiveness
of aggregation of floral parts of plants, for the synchronization of all activities
connected with flower and fruit production. Delevoryas (1966) saw this as a
highly successful feature among flowering plants.

The various types of arrangement of capitula in an inflorescence of Vernonia
(Fig.1) are primarily adaptations of floral aggregation for effective attraction of
pollinators. They are also for efficient and adequate utilization of the mass of
pollen grains produced by the plant. Four of the observed eight types of capitula
arrangement in Vernonia distinctly qualify for this role. They are the raceme of
umbels (Fig.1:8), the corymbose arrangement of cymes (Fig. 1:2 & 3), the
racemose arrangement of cymes (Fig.1:7), and the umbellate arrangement of
helicoid cymes (Fig. 1:10).

These four types of capitula arrangement are germane for those species which
produce the small-size capitula. They also ensure easy access for pollinators;
such that a visiting single pollinator, i.e. an insect, can reach and pollinate
numerous florets within a short time (e.g. as in V. amygdalina, Fig. 1:2, Fig. 2:8).
The synchronization of flowering and fruiting seasons observed in the species
which display any of these four types of capitula arrangement, enhance the
chances of effective pollination by the different visiting pollinators.

The other types of capitula arrangement beside those four highlighted above,
were common in species which flower and fruit intermittently. Such species,
therefore, require more frequent visits by pollinators and at regular intervals as
the growth of the plants progresses (e.g. as in V. tenoreana).

This situation certainly demands some pollination enhancement factors, e.g. large
capitula size, attractive colours of the florets (as in V. kotschyana, V. stenostegia
and V. ambigua) and the production of petaloid phyllaries (as in V. tenoreana and

Comp. Newsl. 25, 1994 19

V. kotschyana, cf. Ayodele 1992). These are adaptive strategies which ensure the
differential identification of the capitula (by pollinators) from among the broad
green leaves on the stem of the plants of the species of Vernonia concerned.

References

Ayodele, M.S. 1987. Cytological and morphological studies on some species of
Vernonia Schreb. in Nigeria. M.Sc. Thesis, Obafemi Awolowo University,
Ile-Ife, Nigeria. 98 pp.

Ayodele, M.S. 1992. Cytogenetic and reproductive studies on some species of
Vernonia Schreb. (Asteraceae) in Nigeria. Ph.D. Thesis, Obafemi Awolowo
University, Ile-Ife, Nigeria. 290 pp.

Delevoryas, T. 1966. Plant Diversification. Yale University, Holt, Rinehart and
Winston Inc. USA. 145 pp.

Faust, W.Z. 1972. A biosystematic study of the /nteriores species group of genus
Vernonia (Compositae). Brittonia 24: 363-378.

Gill, L.S. 1988. Taxonomy of Flowering Plants. FEP Publishers Ltd., Onitha,
Nigeria. 338 pp.

Heywood, V.H., Harborne J.B. & B.L. Turner 1977. An overture to the
Compositae. Jn: Heywood, V.H., Harborne, J.R. & B.L. Turner (eds.), The
Biology and Chemistry of the Compositae 1:1-20. Academic Press, London
& New York.

Hutchinson, J. & J.M. Dalziel 1963. Flora of West Tropical Africa, 2nd ed.
revised by F.N. Hepper Vol. 2: 271-283. Crown Agents, London.

Jones, S.B.Jr. 1979. Synopsis and pollen morphology of Vernonia (Compositae:
Vernonieae) in the New World. Rhodora 81:425-447.

Jones, S.B.Jr. 1981. Synoptic classification and pollen morphology of Vernonia
(Compositae: Vernoniaeae) in the Old World. Rhodora 83:59-75.

Olorode, O. 1984. Taxonomy of West African Flowering Plants. Longman Group
Ltd., London & New York. 158 pp.

20 Comp. Newsl. 25, 1994

FIGURE LEGENDS

Fig. 1. Schematic diagrams of a flowering branch showing the display of capitula
in species of Vernonia. (Broken line depicts possible route of visiting insect
pollinator).

1 - 3: - Tree forms.
4 - 8: - Shrubby forms.
9 - 15: - Herbaceous forms.

Fig. 2. Flowering in Vernonia I.
A. V. tenoreana (4).
B. V. amygdalina (2).
C. V. galamensis (= pauciflora) (11).
D. V. galamensis var. ethiopica (11).

(Note: Numbers in parenthesis depict inflorescence type shown in Fig. 1).

Fig. 3. Flowering in Vernonia II.
A. V. biafrae (8).
B. V. kotschyana (6).
C. V. cinerea (10).
D. V. conferta (1).

(Note the younger capitula buds on lower branches of inflorescence).

Comp. Newsl. 25, 1994

V. colorata
1. Vv. conferta 2. V. amygdalino +
- ——_—_—_ — —_—

4. Vv tenoreona 5. V_ stenostegia €. V. kotschyana

9. V. migeodi 10.

DEGAS .

=
g

———¥°
_4

Fig . 1

7. V. glaberrima

V. cinerea

21

Comp. Newsl. 25, 1994

23

Comp. Newsl. 25, 1994

Fig. 3

24 Comp. Newsl. 25, 1994

STUDIES ON THE REPRODUCTIVE
BIOLOGY OF VERNONIA SCHREB.
(ASTERACEAE)

II. Flowering and post-pollination developments in
the capitulum

M.S. Ayodele *
Botany Department
Obafemi Awolowo University
lle-Ife, Nigeria

Abstract

The protectional role of floral parts is highlighted (e.g. phyllaries; while flower is
in bud and at fruit development phase).

The co-operative features of floral parts (phyllaries, florets and receptacle) in the
display of flowers for pollination and fruits for dispersal are elaborated.

Introduction

Although there are exceptional cases of reproduction by vegetative (asexual)
means in the genus Vernonia, sexual reproduction is apparently common to most
of the species (Olorode 1983, Ayodele 1987).

In all plants the demands of flowering and fruiting phases interact. The structure
and organization of the capitulum of the Compositae (or Asteraceae) must,
therefore, meet the demands of both phases (Burtt 1975). Burtt (1975) was of the
opinion that the study of the co-evolution of the flowering and fruiting phases of
the life history of plants has been rather neglected.

Burtt (1977) articulated a detailed consideration of aspects of diversification in
the capitula of the Compositae. He reported that the requirements of the two
phases (i.e. flowering and fruiting) interact, such that a capitulum gives an
immediate opportunity for the development of certain co-operative features of
floral biology.

*Current mailing address: P.O.Box 586, Ilesa, Nigeria

Comp. Newsl. 25, 1994 25

The objective of this study is to correlate the various observations made on the
stages of development of floral parts (i.e. the capitulum and its contents) in some
species of Vernonia.

Materials and Methods

Regular observations of field, garden and screen-house plants were made on
randomly selected plants. Some 2-5 plants of each species in the garden and
screen house carried labels on which regular entries were recorded. Data gathered
on capitulum development, included: the period (in days) to flowering, flower
bud at anthesis; fertilization of florets; appearance of pappus of ripe fruits
(achenes) on the capitulum, and the mode of achene dispersal.

Special observation on the post-fertilization events were made with respect to
changes in the phyllaries and the receptacle of the capitulum. Photographs of
important occurrences were recorded during the developmental stages of the
capitulum.

A bar diagram, illustrating the incidence of receptacle reflexing among 16 species
of Vernonia, was prepared.

Observations/Results

The Vernonia capitulum in bud was encapsulated by a wall of phyllaries in all the
species. The peripheral florets on a capitulum opened and got pollinated before
the central florets. The post-pollination sign in all the species was the withering
of the corolla of pollinated florets within 24 hours after pollination (Fig. 1:1). The
withered florets subsequently dropped off. This was followed by the gradual
reclosure of the involucre, the diameter of the capitulum becoming smaller (Fig.
1:2). The phyllaries remained green, as the fruits from pollinated florets matured.

The capitulum thereafter reopens gradually (diameter increasing), exhibiting first
the pappus of the peripheral matured achenes (Fig. 1:3). The phyllaries become
dried up, tuming brown. The receptacle of the capitulum in some species,
gradually, is centrifugally reflexed. More pappus emerges from matured achenes
(Fig. 1:4).

Capitulum receptacle may be completely reflexed in some cases, becoming
turned inside-out when dry, like a closed umbrella (Fig.2:3). All pappus radiates
from dry, matured achenes (Fig. 1:5) as they are ready for dispersal. The colour
of pappus varied among the species from cream through shades of brown to
black.

The incidence of capitulum receptacle reflexing varied among and within the
species. Nine of the 16 species studied did not manifest any proportion of

%6 Comp. Newsl. 25, 1994

receptacle reflexing. A species may manifest different proportions of the changes
in receptacle of matured heads (Fig. 3).

Discussion

Heywood et al. (1977) highlighted the adaptive value of characters associated
with the capitulum floral features. They endorsed the use of more quantitative
methods in the field of population ecology, which will provide more meaningful
studies as regards adaptive strategies (i.e. "morphological accommodations") for
reproductive features like pollination, dispersal and survival of propagules.

The structure of the capitulum in Vernonia and the organization of the constituent
parts, namely the phyllaries, florets and receptacle, show meaningful adaptive
values which meet the demands of both the flowering and fruiting phases. For
instance, the phyllaries which encapsulate the capitulum buds, serve as protection
for the internal structures. This protectional role recurs after pollination, as the
involucre recloses (Fig. 1:2), ensuring the safety of the developing fruits
(achenes).

The phyllaries remain green while the fruits are maturing; becoming dry only in
matured flower heads. A participatory role of the phyllaries in the net provision
of photosynthetic products in the fruits cannot be ruled out. Just as the phyllaries
open initially at anthesis to expose the florets, so they are reopened when the
fruits are matured for dispersal (Fig. 1:3, 4 & 5).

Receptacle reflexing was observed as an essential phenomenon aiding efficient
fruit dispersal (Ayodele 1992). This can be compared with the explosive
mechanism in some Euphorbiaceae. A combination of light-weight fruits and
fully reflexed receptacle, resulted in a longer distance of fruit dispersal, even in
the absence of wind (Ayodele 1992).

In spite of the observed morphological diversity in phyllaries, their functions
were similar in all the species investigated. This corroborates an earlier report by
Jones (1979). The presence of petaloid phyllaries in some species is believed to
be strategic for the attraction of pollinators. The varying colours of the pappus on
matured fruits of the species has some value in species grouping and
identification.

Burtt (1977) itemized four major factors necessary to meet demands of both the
flowering and fruiting phases. Two of these factors are related to the flowering
phase, namely the efficiency of pollination and the balance of the breeding
system (i.e. the ratio of inbreeding to outbreeding). The other two factors have to
do with the fruiting phase, namely the protection of the maturing achenes and
their adequate dispersal.

The gradual opening of the flower buds at anthesis ensures a systematic
pollination of the sometimes numerous florets in a head. The protectional

Comp. Newsl. 25, 1994 Dil

arrangement for achenes and their efficient dispersal have already been
highlighted above. An investigation of the breeding system in Vernonia would be
worthwhile for a thorough understanding of the ability of members of the genus
to the flowering and fruiting demands.

References

Ayodele, M.S. 1987. Cytological and morphological studies on some species of
Vernonia Schreb. in Nigeria. M.Sc. Thesis, Obafemi Awolowo University,
Ile-Ife, Nigeria. 98 pp.

Ayodele, M.S. 1992. Cytogenetic and reproductive studies on some species of
Vernonia Schreb. (Asteraceae) in Nigeria. Ph.D. Thesis, Obafemi Awolowo
University, Ile-Ife, Nigeria. 290 pp.

Burtt, B.L. 1975. Patterns of structural change in the flowering plants. Trans.
Bot. Soc. Edinb. 42:133-142.

Burtt, B.L. 1977. Aspects of diversification in the capitulum. Jn: Heywood,
V.H., Harborne, J.B. & B. L. Turner (eds.)., The Biology and Chemistry of
the Compositae 1: 41-59. Academic Press, London & New York.

Heywood, V.H., Harborne, J.B. & B.L. Turner 1977. An overture to the
Compositae. Jn: Heywood, V.H., Harborne, J.B. & B.L. Tuer (eds.), The
Biology and Chemistry of the Compositae 1: 1-20. Academic Press, London
& New York.

Jones, S.B. Jr. 1979. Synopsis and pollen morphology of Vernonia (Compositae:
Vernonieae) in the New World. Rhodora 81: 425-447.

Olorode, O. 1984. Taxonomy of West African Flowering Plants. Longman Group
Ltd., London & New York. 158 pp.

28 Comp. Newsl. 25, 1994

FIGURE LEGENDS

Fig. 1. Post-pollination development in the capitulum of Vernonia tenoreana
Oliv.

1. Withered florets on capitulum (after pollination).

2. Capitulum reclosure post florets shedding.

3. Capitulum reopening post fruit maturation (onset).

4. Advanced capitulum reopening (note appearance of pappus).
5. Capitulum with reflexed receptacle, ready for fruit dispersal.

Fig. 2. Variation in receptacle reflexing in Vernonia.

1. No reflexing.
2. Partial reflexing.
3. Total reflexing.

Fig. 3. Incidence of receptacle reflexing in ripe capitula on plants of some species
of Vernonia.

Key to species numerals:

1. V. conferta Benth.

2. V. amygdalina Del.

3. V. colorata (Willd.) Drake

4. V. tenoreana Oliv.

5. V.

6. V. kotschyana Sch. Bip.

7. V. glaberrima Welw. ex O. Hoffm.

8. V. biafrae Oliv. & Hiern

9. V. migeodi S. Moore

10. V. cinerea (Linn.) Less.

11. V. galamensis (= pauciflora) (Cass.) Less.
12. V. ambigua Kotschy & Peyr.

13. V. perrottetii Sch. Bip.

14. V. nestor S. Moore

15. V. purpurea Sch. Bip.

16. V. galamensis var. ethiopica (Cass.) Less. & Gilb.

. Stenostegia (Stapf) Hutch.& Dalz.

29

Comp. Newsl. 25, 1994

=
eb

F

Fig. 2

Comp. Newsl. 25, 1994

30

le Flexi er plant, (°/o)
cle Flexing per p S
(S}

e Recepta
re)

K
oO

9g

Mean Percenta
ro)

6

7 8

)

[_]*/ abs ence of receptacle reflexing

[][]*/- partially reflexed receptacles
EEE %e fully reflexed receptacle

'

ifcrch a

oa . '
a

oe .

er

10 11 12 te a 15 6

SPECIES OF VERNONIA

rig.

D

Fig. 3

Comp. Newsl. 25, 1994 31

EFFECT OF AQUEOUS EXTRACTS OF
CHROMOLAENA ODORATA (L.) K. & R. ON
RADICLE AND PLUMULE GROWTH AND

SEEDLING HEIGHT OF MAIZE, ZEA MAYS L.

Foluso M. Dania Ogbe, L.S. Gill and E.O.O. Iserhien
Department of Botany
University of Benin
Benin City, Nigeria

Abstract

Effects of 12, 24, 36, 48 hr extracts of leaf, shoot and root of Chromolaena
odorata on radicle and plumule growth and seedling height of Zea mays were
observed. Leaf extracts affected radicle, plumule and seedling growth
irrespective of length of soaking time, while differential inhibitory effects were
noticed with stem and root extracts.

Introduction

De Candolle (1832) was the first to suggest that roots of some weeds like Cirsium
(Asteraceae) inhibit the growth of some crop plants. Since then, many
investigations like Grummer & Burger (1964), Kohlmuenzer (1965), Dzubenko
& Petrenko (1971), Bell & Koeppe (1972), have reported the allelochemical
effects of many weeds on the rate of germination and growth of crop plants. Rice
(1984) put this aspect of chemical ecology on a scientific footing. However, most
of the work done on allelochemical effect of weeds have been reported from
Europe and North America. From Nigeria, Tijani-Eniola & Fawusi (1989) and
Gill et al. (1993) are the only investigators to report the allelochemical effect of
Chromolaena odorata on crop plants.

The present study was undertaken to determine the allelopathic effect of C.
odorata on germination growth and seedling height of maize.

32 Comp. Newsl. 25, 1994

Materials and Methods

In June, 1993, whole plants of C. odorata were collected from Benin City,
Nigeria (latitude 6,5° N, longitude 6.0° W). Portions of 500 g each of leaves, stem
and roots were soaked in 1 litre of distilled water for 12, 24, 36 and 48 hours. The
extracts were filtered and refrigerated. Petri dishes were lined above and below
with Whatman No.1 filter paper each with 10 maize seeds, purchased from the
market, and moistened daily with 12, 24, 36, 48 hour aqueous extracts of leaf,
stem and root of C. odorata. Each treatment was in triplicate. The control was
moistened with distilled water. Petri dishes were kept at room temperature (28°C
+ 2°C) in a growth chamber. Elongation measurements were taken at 24 hr
intervals.

Polybags (30 cm x 30 cm) were filled with loamy soil, pH 6,5. Three maize seeds
soaked in 12, 24, 36, 48 hour aqueous extracts of leaf, stem and root of C.
odorata for 24 hours were sown in the polybags. Each treatment was replicated
five times. The polybags were watered with the corresponding extracts. Seedlings
were thinned down to one per polybag after establishment. Measurement of
seedling height was taken at the end of 28 days.

Results

The results of the different aqueous extracts of leaf, stem and root of the weed on
radicle and plumule growth of maize are shown in Tables 1-6. Extracts from
leaves soaked for 12, 24, 36 and 48 hours retarded radicle growth when
compared with the control, but there was no significant difference in the degree
of inhibition between the different leaf extracts (Table 1). Stem extracts at 12, 24,
36 hours did not retard radicle growth. However, at 48 hours, significant slowing
down of radicle growth occurred (Table 2). Root extracts did not inhibit radicle
growth (Table 3).

Plumule growth was affected by leaf extracts, but the retardation was more
pronounced with 36 and 48 hours extracts (Table 4). Stem extracts at 12, 24, 36
hours did not retard plumule growth initially. After 96 hours, some retardation
occurred. However, with the 48 hour extract, there was significant retardation
with growth occurring only after 120 hours (Table 5). Root extracts retarded
plumule growth significantly after 72 hours of treatment (Table 6). The effect of
the different aqueous extracts on seedling height is shown in Table 2. All
seedlings continuously treated with extracts were shorter than control plants at
the end of 28 days. There was no marked difference between leaf, stem and root
extracts, neither was there any significant difference in height due to length of
soaking time of the different types of extracts (Table 7).

Comp. Newsl. 25, 1994 33

Discussion

The results show that there are inhibitory factors in the tissues of C. odorata that
affect the development of maize seeds when germinated. The most significant
differences in growth occurred when seeds were treated with extracts from the
leaf of the weed. The radicle, plumule and seedling growth were all adversely
affected by extracts from leaf irrespective of length of soaking time before
extraction. However, differential inhibitory effects were noticed with stem and
root extracts. Radicle growth was not adversely affected by root extracts, but only
48 hour stem extract reduced their growth.

With plumule growth, all the extracts from leaf, stem and root inhibited it, but to
varying degrees when compared with control. Retardation from leaf extract
treatments was markedly greater than from the other treatments, especially the 36
and 48 hour leaf extracts.

According to Ambika & Jayachandra (1980), the leaves of C. odorata contain a
large amount of allelochemicals, which could be responsible for reducing the
growth of radicle, plumule and seedlings. Bell & Koeppe (1972) reported the
inhibitory effect of Setaria faberii leachate on the growth in height and
accumulation in dry weight of plant. Recently, Adams & Azimi (1991) reported
the allelopathic effect of Cyperus rotundus leaf extract in seedling growth of
wheat.

The different degree of inhibitory effect of leaves, stem and root of C. odorata at

the various extraction times (12, 24, 36, 48 hr) show that these three components
have either different quantities of allelochemicals, or the nature of allelochemical
may be different. From the present study it is apparent that the leaves extract
irrespective of the duration of extraction has the most inhibiting effect on the
growth of the seedlings of maize. The present authors are of the opinion that in
nature, allelopathy might have played a significant role in the spatial distribution
of plants.

References

Adams, S.M. & R. Azimi 1991. Effect of purple nutsedge (Cyperus rotundus)
leaf extract on germination and seedling growth of wheat (W. pavon).
Pakistan Journal of Weed Science 4: 59-61.

Ambika, S.R. & Jayachandra 1980. Influence of light on seed germination in
Eupatorium odoratum L. Indian Forester 106: 637-640.

Bell, T.D. & D.E. Koeppe 1972. Non competitive effects of giant foxtail on the
growth of corn. Agronomy Journal 64: 321-325.

Candolle, A.P. de 1832. Physiologie Vegetale 3. Béchet Jeune, Paris.

34 Comp. Newsl. 25, 1994

Dzubenko, N.N. & N.I. Petrenko 1971. On biochemical interaction of cultivated
plants and weeds. /n: Gredurisky, A.M. (ed.), Physiological, Biochemical
Basis of Plant Interaction in Phytocenosae Vol. 2: 60-66. Nankova Dermka,
Kiev.

Gill, L.S., Anoliefo, G.O. & U.V. Iduoze 1993. Allelopathic effect of aqueous
extract of Siam weed on growth of cowpea. Chromolaena Newslettter 8:
1-4.

Grummer, G. & H. Burger 1964. The influence exerted by species of Cameline
on flax by means of toxic substances. /n: Harper, J.L. (ed.), The Biology of
Weeds, pp. 153-157. Blackwell, Oxford.

Kohlmuenzer, S. 1965. Effects of extracts and some other chemical components
of Galium mollugo on the germination of seeds and growth of selected
plants. Dissertation Pharmacy 17: 369.

Rice, E.L. 1984. Allelopathy. 2nd Ed. Academy Press, New York.

Tijani-Eniola, H.A. & O.A. Fawusi 1989. Allelopathic activities of crude
methanol extract of Siam weed and wild poinsettia on seed germination and
seedling growth of tomato. Nigerian Journal of Weed Science 2: 15-20.

Comp. Newsl. 25, 1994 35

Table l. Effect of different aqueous leaf extracts of Chromolaena odorata on
radicle length (cm) of Zea mays

Experimental time (hr)

Extract time
72 96

0.32+0.06 1.92 +0.32 3.87 + 0.69 5.97 + 0.79 7.47 + 1.32

0.27+40.10 1.49+0.37 2.34 + 0.52 3.30 + 0.64 4.24+0.51
0.88 + 0.22 1.47 + 0.34 1.97 + 0.33

1.00 + 0.29 1.81 + 0.39 2.79 + 0.40

E52 0335 1.73 + 0.36 2.02 + 0.37

Table 2. Effect of different aqueous stem extracts of C. odorata on radicle length (cm)
of Zea mays

Experimental time (hr)

Extract time
(hr) 72 96

0.32+0.06 1.92 + 0.32 3.87 + 0.69 5.97+0.79 _7.47+1.32
0.314011 1.8140.33 4.09 + 0.73 5.97+0.93 7.89 + 1.06
0.30+0.07 1.82 + 0.43 3.81+0.91 5.78+0.91 10.03 + 1.07

24 :
0.30+0.11 1.90 +0.42 4.23 + 0.90 6.58+0.89 8.87 +0.99

1.22 + 0.21 2.02 + 0.39 3.05 + 0.37

36

Table 3.

of Zea mays

Table 4.

0.32 + 0.06
0.30 + 0.09
0.34 + 0.05
0.32 + 0.04
0.30 + 0.11

of Zea mays

Extract time
(hr)

1.92 + 0.42
1.76 + 0.50
1.64 + 0.55
1.44 + 0.48
2.09 + 0.56

0.61 + 0.14
0.69 + 0.23

Experimental time (hr)

72 96

3.87 + 0.69
4.83 + 1.09
4.33 + 0.82
4.00 + 1.13
5.03 + 1.39

Experimental time (hr)

72 96

2.09 + 0.49
1.55 + 0.45
0.87 + 0.34

0.24 + 0.03

0.61 + 0.18

Comp. Newsl. 25, 1994

OOS
7.78 + 1.84
7.00 + 0.81
6.58 + 1.09
7.49 + 1.32

3.56 + 0.67
2.44 + 0.55
1.97 + 0.42
0.96 + 0.22
1.42 + 0.38

Effect of different aqueous root extracts of C. odorata on radicle length (cm)

7.47 + 1.32
11.09 + 1.90
9.88 + 0.78
9.94 + 1.06
11.06 + 1.35

Effect of different aqueous root extracts of C. odorata on radicle length (cm)

5.03 + 0.87
3.63 + 0.73
3.66 + 0.62
1.81 + 0.35

2.11 + 0.43

Comp. Newsl. 25, 1994 37

Table 5. Effect of different aqueous root extracts of C. odorata on radicle length (cm)
of Zea mays

Experimental time (hr)
Extract time
(hr) 24 96

Control 0.61 + 0.14 3.56 + 0.67 5.03 + 0.87

0.65 + 0.20 1.81 + 0.58 2.58 + 0.56 3.66 + 0.52
0.71 + 0.19 2.11 + 0.95 3.49 + 0.93 4.80 + 0.94
0.79 + 0.24 1.95 + 0.53 3.11 + 0.52 8.87 + 0.99

0.59 + 0.15 1.33 + 0.53

Table 6. Effect of different aqueous root extracts of C. odorata on plumule length (cm)
of Zea mays

Experimental time (hr)
Extract time

(hr) 72 96
0.61 + 0.14 2.09 + 0.49 3.56 + 0.67 5.03 + 0.87
0.79 + 0.16 1.81 + 0.43 2.89 + 0.43 4.37 + 0.37
0.82 + 0.31 1.77 + 0.22 2.79 + 0.29 4.08 + 0.48
0.46 + 0.14 1.40 + 0.32 2.34 + 0.34 3.57 + 0.34

0.76 + 0.41 1.65 + 0.40 2.43 + 0.33 3.72 + 0.32

38 Comp. Newsl. 25, 1994

Table 7. Effect of different aqueous root extracts of
C. odorata on height (cm) of 28 days Zea mays seedlings

Extract treatment Leaf Stem Roo

Control 29.89 + 2.58 29.89 + 2.58 29.89 + 2.58
12 hr 20.50 + 3.77 20.77 + 2.21 18.60 + 2.78
24 hr 19.50 + 1.83 21.10 + 3.92 20.43 + 2.84

36 hr 22.37 + 1.98 18.63 + 2.31 19.33 + 2.34

48 hr 20.77 + 2.22 20.26 + 2.05 21.07 + 3.12

Comp. Newsl. 25, 1994 39

THE GENERA ARTEMISIA L. AND
SERIPHIDIUM (BESS.) POLJAK.
IN THE WORLD

Ling Yeou-ruenn
South China Institute of Botany
Academia Sinica
510656, Guangzhou
Peoples Republic China

Introduction

Artemisia and Seriphidium are two important genera, not only in phylogeny and
floristics, but also in the economic uses in Compositae. In China many species of
both genera are used for medicine, especially for antiphlogistic, detoxifying,
hemostatic, diuretic and anthelmintic drugs, or for anticancer treatment or curing
gynaecological diseases, or for moxibustion. Besides, some species are good for
herbages in pastures, and some shrubs in desert areas are useful for windbreaking
and sand fixation.

Artemisia and Seriphidium in the World

There are 346 species and 69 varieties of Artemisia, sensu stricto, and 130
species and 32 varieties of Seriphidium in the World. The former genus can be
divided into three subgenera and nine sections:

1. Subgen. Artemisia
Sect. 1) Absinthium DC.
Sect. 2) Abrotanum Bess.
Sect. 3) Artemisia
Sect. 4) Viscidipubes Y.R. Ling
Sect. 5) Albibractea Y.R. Ling
2. Subgen. Dracunculus (Bess.) Peterm.
Sect. 6) Dracunculus Bess.
Sect. 7) Latilobus Y. R. Ling
Sect. 8) Turaniphytum (Poljak.) Y.R.Ling

40 Comp. Newsl. 25, 1994

3. Subgen. Artanacetum (Rzazad.) Y.R. Ling & L. Jiang
Sect. 9) Artanacetum (Rzazad.) Y.R. Ling

Seriphidium is divided into
Sect. 1) Seriphidium
Sect. 2) Minchunensa Y.R. Ling
Sect. 3) Juncea (Poljak.) Y.R. Ling & C.J. Humphries

Of all taxa, perhaps, Sect. Absinthium is the most primitive group. Its ancestor
"Pro-Absinthium”, should be derived from "Pro-Anthemideae" in Upper or
earlier Tertiary Era, in the speciation center of North Asia. Sect. Abrotanum is
also primitive, and phylogenetically important. From its ancestor other ancestral
species in more advanced or specialized sections of Artemisia and Seriphidium
were derived. This outline of phylogeny has got support from the research on
pollen grains, chemotaxonomy and cladistics.

After "Pro-Absinthium"” had evolved, it diverged and migrated from the
speciation centre. Probably it migrated,

1. eastwards, i.e. throughout Siberia to the west of North America, and

2. westwards, i.e. across the Ural Mountains and the European plain to the east of
North America, and

3. southwards, i.e. to Central East & South Asia, or South-West Asia, South
Europe and North Africa, or Central America, northern South America and
Oceania, especially during the Ice Age of the Quaternary Era.

Many new species of both genera evolved during the migration. In the southward
migration the genera obtained a special distribution, forming three oblique areas
of species concentration, from northeast (W. Hsingan Mountains) running
obliquely to southwest (E. or W. Hengduang Mountains) or to the west (W.
Kunlun Mountains) of China. Most species of Sect. Absinthium and Sect.
Abrotanun are inside the first zone, i.e. north, northwest & southwest China. The
third zone, in north and northwest China reflects the distribution of most species
of Sect. Dracunculus of Artemisia and the whole of Seriphidium. Other sections
of Artemisia are in the second zone, i.e. east, central, south and southwest China.
The centres of divergence and present distribution for both genera are in Eurasia
and western North America. The species concentration centre is in Central and
East Asia, incl. Kazakhstan, Russia and north, northwest and northeast China.
Here there are 248 species of Artemisia_and 67 species and 18 varieties of
Seriphidium within an area of 6 million km. In western North America there is a
small centre with 66 species and 4 varieties of Artemisia and 13 species and 5
varieties of Seriphidium. Besides, there is another centre in so called Hengduan

Comp. Newsl. 25, 1994 41

—Himalayan Mountains subkingdom by the view-point of plant floristics. This is
either the refuge for Artemisia in the Ice Age or the centre of secondary
speciation, divergence and present distribution after the Tertiary Era. Its special
climate and environment was caused by the contact of the Indian continent of
Gondwana with the South Eurasian continent of Laurasia in the Tertiary Era,
accompanied by the Himalayan Mountains orogenic movement including the
uplifting and folding of the Hengduang Mountains. Now ere. are 188 species
and 65 varieties of Artemisia in that area of 750 thousand km. Artemisia and
Seriphidium in Eurasia and North America underwent parallel development after
the lower Tertiary Era, resulting in many regional endemic species in different
areas. Only a few Artemisia species are worldwide in distribution or
discontinuously distributed in different continents.

The floristic geography of both genera can be divided into 5 kingdoms and 19
subkingdoms and 37 regions in the World, but most of Artemisia is distributed in
temperate and subtropical areas of the northern hemisphere, and a minor portion
in the tropics of both hemispheres. Seriphidium occurs only in the areas of
Central and West Asia, Central North America and North Africa. The
geographical areas are:

I. Holarctic Kingdom.
1) Eurasia Division.
1. Subholarctic subkingdom, incl.
1) Circumboreal (subholarctic) region,
2) North Asia subholarctic region,
3) Northeast Asia—Northwest America subholarctic region and
4) Eurasia subholarctic region,
2. East Siberia—Far East subkingdom, incl.
5) East Siberia region and
6) Far East—North Japan region,
3. Eastern Asia forest subkingdom, incl.
7) Northeast China—North Korea—North Japan region,
8) East and Central China—Korea—Japan region,
9) North-Central China oblique exchange-convergence region,

10) Taiwan mountain regional endemic region, and

11) South and South-east China—Southeast Asia region,

42 Comp. Newsl. 25, 1994

4. Hengduang—Himalayan Mountains forest subkingdom, incl.
12) West Sichuan—Northwest Yunnan subalpine or alpine region,
13) East Xizang region,
14) The islet region of the valleys of Nujiang, Lancangjiang and Yuanjiang
(Rivers) and
15) The South Himalayan Mountains region,
5. Eurasia steppe or semi desert-steppe subkingdom, incl.
16) Central Asia (east)-Mongolia (west) steppe or semidesert-steppe region,
17) West Siberia—East Europe steppe region and
18) Altai mountain steppe region,
6. West and Central European steppe and forest-steppe subkingdom incl.
19) West and Central European plain steppe and forest-steppe region, and
20) Central European mountain steppe and forest-steppe region.

II) Ancient Mediterranean Division.

7. Asia desert (Central Asia—West Asia) subkingdom, incl.

21) Central Asia—West Asia temperate desert region,
22) Pamir, Alai-West Kunlun, West Himalaya region, and
23) Caucasus exchange-convergence region,

8. Qinghai—Xizang plateau subkingdom, incl.
24) Tangut region and
25) Chiangtang Plateau-Gangdis mountains region,

9. Southwest Asia—Arabia—A frica tropical desert and savanna subkingdom, incl.
26) Southwest Asia—Arabia—Northeast Africa desert and savanna region, and
27) East and South Africa savanna region,

10. South Europe-Mediterranean-West and Northwest Africa savanna _ sub-
kingdom, incl.

28) South Europe—Mediterranean—West and Northwest African savanna region.

Comp. Newsl. 25, 1994 43

III) North America Division.
11. Northwest and West of North America subkingdom,
12. East and Central of North America subkingdom.

II. Paleotropical Kingdom.

13. North Indo-China-South China—North of South Asia tropical forest and
savanna subkingdom, and

14. North India—North Pakistan subkingdom.

III. Neotropical Kingdoms.
15. Mexico subkingdom and

16. The Greater Antilles subkingdom.

IV. South America Kingdom.
17. North of South America subkingdom.

V. Australian or Oceanian Kingdom.
18. East Australian subkingdom and

19. Oceanian subkingdom.

It should be noted that subkingdom 11 to 19 each contains only one region.

References
(main sources of information)
Abrams, L. & R. S. Ferris 1960. /llustrated flora of the Pacific States
(Washington, Oregon, and California) 4: 403-415. Stanford.

Anderson, J. P. 1952. Flora of Alaska and adjacent parts of Canada: 427-431.
Ames, Iowa.

44 Comp. Newsl. 25, 1994

Besser, W. S. 1829. Monographie des Artemisia. Bull. Soc. Imp. Nat. Moscou
1(8): 219-265.

Besser, W. S. 1834. Tentamen de Abrotanis seu de sectione II da Artemisiarum
Linnaei. Nouv. Mém. Soc. Imp. Natural. Moscou 3: 1-92.

Besser, W. S. 1835 a. Dracunculi seu sectione IVto et ultima Artemisiarum
Lianaei. Bull. Soc. Imp. Nat. Moscou 8: 3-97.

Besser, W. S. 1835 b. Enumeratio Artemisiarum illarum, quas non vidi et ideo iis
locum in mea divisione hujus generis assignare nequivi. Bull. Soc. Imp. Nat.
Moscou 8: 177-180.

Besser, W. S. 1842. Monographiae Artemisiarum Section I. Dracunculi. Mém.
Acad. Imp. Sci. Petersb. 4: 445-488.

Britton, N. L. & H. A. Brown 1898. An illustrated flora of the northern United
States, Canada and the British possessions... 3: 461-469. New York.

Candolle, A. P. de 1837. Prodromus systematis naturalis regni vegetabilis 6.
Paris, Strasbourg, London.

Gray, A. 1884. Synoptical flora of North America 1(2): 367-375. New York,
London, Leipzig.

Hall, H. M. & F. E. Clements 1923. The phylogenetic method in taxonomy (The
North American Artemisia): 1-156. Washington, D. C.

Hitchcock, C. L., Cronquist, A., Ownbey, M. & J. W. Thompson 1955.
Vascular plants of the Pacific Northwest 5: 54-71. Seattle.

Hoffmann, O. 1892. Compositae. Jn: Engler, A. & K. Prantl (eds.), Die
natiirlichen Pflanzenfamilien 4(5): 273-304. Leipzig.

Hooker, W. J. 1833. Flora boreali-americana 1: 321-327. London, Paris,
Strasbourg.

Hultén, E. 1937. Flora of the Aleutian Islands: 325-328. Stockholm.

Hultén, E. 1968. Flora of Alaska and neighbouring territories: 896-911.
Stanford.

Krascheninnikoy, I. M. 1958. The role and meaning of the Angara Floristic
Center in the phylogenetic development of the basic Europe-Asiatic groups
of Artemisia Linn. Mat. Hist. Fl. Veg. USSR 3.

Ling, Y. & Y. R. Ling 1991. Flora reipublice popularis sinicae 76(2). Science
Press.

Ling, Y. R. 1980. Taxa nova generum Artemisiae et Seriphidii xizangensis. Act.
Phytotax. Sin. 16(1): 61-65.

Comp. Newsl. 25, 1994 45

Ling, Y. R. 1982. On the system of the genus Artemisia Linn. and the
relationship with its allies. Bull. Bot. Res. 2(2): 1-60.

Ling, Y. R. 1988. The Chinese Artemisia Linn. (The classification, distribution
and application of Artemisia Linn. in China). Bull. Bot. Res. 8(4): 1-61.

Ling, Y. R. 1989. The genus Artemisia Linn. (Compositae) in the south-east
Asian countries. Act. Bot. Austr. Sin. 4: 29-37.

Ling, Y. R. 1991. A review of the classification, distribution and application of
Artemisia L. and Seriphidium (Bess.) Poljak. (Compositae) in China.
Guihaia 11(1): 19-35.

Ling, Y. R. 1992. The Old World Artemisia Linn. (Compositae). Bull. Bot. Res.
12(1): 1-108.

Linnaeus, C. 1753. Species plantarum 2. Holmiae.

Poljakov, P. 1961 a. Artemisia L. In: Shishkin, B. K. & E. G. Bobrov (eds.),
Flora SSSR 26: 425-631. Moscou, Leningrad.

Poljakov, P. 1961 b. Materia on the system of Artemisia L. Mat. Fl. Veg.
Kazakh. 11.

Rydberg, P. A. 1916. North American flora 34(3): 244-285. Lancaster, PA.

Tutin, T. G. & W. Gutermann 1976. Artemisia L. In: Tutin, T. G. et al. (eds.),
Flora Europaea 4: 178-186. Cambridge.

Wulff, E. V. 1950. An introduction to historical plant geography. Waltham,
Mass.

46 Comp. Newsl. 25, 1994

NEW COMBINATIONS IN THE
CALENDULEAE

Bertil Nordenstam
Department of Phanerogamic Botany
Swedish Museum of Natural History

S-104 05 Stockholm

Sweden

Abstract

Twelve new combinations in Dimorphotheca Moench and Tripteris Less. are
published, resulting from recent generic revision in the tribe Calenduleae.

Introduction

A revision of generic limits in the small tribe Calenduleae has been initiated, and
some generic re-arrangement has recently been proposed (Nordenstam 1994 & in
press). The largest genus Osteospermum has been divided into three separate
genera, viz. Osteospermum Ss. str., Tripteris and Oligocarpus. Further, the section
Blaxium of Osteospermum has been transferred to the genus Dimorphotheca, and
the small genus Castalis has also been sunk in Dimorphotheca. Most of the taxa
now transferred to Dimorphotheca already have valid names under that genus,
and the same goes for Tripteris, which has previously been recognized as a
genus. However, some new combinations need to be validated and are published
here.

Dimorphotheca Moench (nom. cons.)

After inclusion of Castalis and sect. Blaxium the genus Dimorphotheca now
comprises 18 species. Only two of these, plus one variety, need new
combinations under Dimorphotheca.

Comp. Newsl. 25, 1994 47

D. dregei DC. var. reticulata (T. Norl.) B. Nord., comb. nov.

Basionym: Osteospermum Dregei (DC.) T. Norl. var. reticulatum T. Norlindh
1943: 249,

Type: South Africa, Riversdale Div., hills near Riversdale, ca. 200 m, 1923, Muir
2905 (PRE holotype).

D. tragus (Ait.) B. Nord., comb. nov.
Basionym: Calendula Tragus Aiton 1789: 271.

Type: No type was designated by Norlindh (1943), who mentioned that Aiton
had the species in cultivation at Kew since 1774, when it was introduced from the
Cape of Good Hope by Masson.

D. walliana (T. Norl.) B. Nord., comb. nov.
Basionym: Osteospermum Wallianum T. Norlindh 1943: 246.

Type: South Africa, Stellenbosch Div., at Gordons Bay, 1.XII.1938, Wall s.n. (S
holotype).

Tripteris Less.

For more than one century after its publication Tripteris was generally accepted
as a genus, and accordingly most of the species have a valid name under
Tripteris. Norlindh (1943) reduced Tripteris to a subgenus of Osteospermum.
Because of its recent reinstatement as a genus (Nordenstam 1994), the following
new combinations are needed.

T. afromontana (T. Norl.) B. Nord., comb. nov.
Basionym: Osteospermum afromontanum T. Norlindh 1954: 149.

Type: Tanzania, Mt. Hanang, NE slope, S$. Mbulu distr., ca. 3300 m, 8.1I.1946,
Greenway 7665 (EA holotype; K, PRE, isotypes).

T. angolensis (T. Norl.) B. Nord., comb. nov.

Basionym: Osteospermum angolense T. Norlindh 1954: 148.

48 Comp. Newsl. 25, 1994

Type: Angola, Huila, Quilemba, Serra da Chela, 1900 m, 1937, Gossweiler
10816 (COI holotype).

T. breviradiata (T. Norl.) B. Nord., comb. nov.
Basionym: Osteospermum breviradiatum T. Norlindh 1943: 345.

Type: South Africa, Clanwilliam, 150 m, 1896, Schlechter 8415 (S holotype; B,
BOL, G, GRA, K, P, PRE, W, Z, isotypes).

T. microcarpa Harv. subsp. septentrionalis (T. Norl.) B. Nord., comb. nov.
Basionym: Osteospermum microcarpum (Harv.) T. Norl. var. septentrionale T.
Norlindh 1943: 300. - O. microcarpum subsp. septentrionale (T. Norl.) T.
Norlindh 1960: 398.

Type: Angola, distr. Mossamedes, inter 14° et 16° lat. austr., 1859, Welwitsch
3542 (K lectotype; B, C,G ).

T. nyikensis (T. Norl.) B. Nord., comb. nov.
Basionym: Osteospermum nyikense T. Norlindh 1976: 171.

Type: Malawi, Chitipa distr., Kawozya Hill, 2140 m, 11.VIII.1972, Brummitt &
Synge WC 211 (K holotype; MAL, S, SRGH, isotypes).

T. oppositifolia (Ait.) B. Nord., comb. nov.
Basionym: Calendula oppositifolia Aiton 1789: 272.
Type: In herb. Brit. Mus. (BM lectotype; cf Norlindh 1960: 393).

Syn.: Calendula glabrata Thunberg 1800: 163. - Osteospermum glabratum
(Thunb.) Lessing 1832: 89.

T. pinnatilobata (T. Norl.) B. Nord., comb. nov.
Basionym: Osteospermum pinnatilobatum T. Norlindh 1943: 302.

Type: South Africa, L. Namaqualand Div., Concordia, in collibus, 1897, ca. 1000
m, Schlechter 11331 (B lectotype; LD, Z).

Comp. Newsl. 25, 1994 49

T. rosulata (T. Norl. ) B. Nord., comb. nov.
Basionym: Osteospermum rosulatum T. Norlindh 1943: 325.

Type: South Africa, Griqualand East, Mount Currie, in apertis circa Kokstad,
1400-1500 m, 1883, Tyson 1796 (BOL holotype; K, PRE, SAM, isotypes).

T. sinuata DC. var. linearis (Harv.) B. Nord., comb. nov.
Basionym: Tripteris linearis Harvey 1865: 426.

Type: "Tripteris linearis Harv. n. sp." (Harvey scripsit), herb. Sonder (S
holotype).

References
Aiton, W. 1789. Hortus kewensis 3. London.

Harvey, W. H. 1865. Compositae. Jn: Harvey, W. H. & O. W. Sonder, Flora
capensis 3: 44-530. Dublin, Cape Town.

Lessing, C. F. 1832. Synopsis generum Compositarum. Berolini.

Nordenstam, B. 1994. Tribe Calenduleae. Jn: Bremer, K. (ed.), Asteraceae,
Cladistics & classification: 365-375. Portland, Oregon.

Nordenstam, B. in press. Recent revision of Senecioneae and Calenduleae
systematics. Proceedings of International Compositae Conference, Royal
Botanic Gardens, Kew 1994.

Norlindh, T. 1943. Studies in the Calenduleae 1. Monograph of the genera
Dimorphotheca, Castalis, Osteospermum, Gibbaria and Chrysanthemoides.
Lund.

Norlindh, T. 1954. Further contributions to the genus Osteospermum. Svensk
Bot. Tidskr. 48(1): 148-154.

Norlindh, T. 1960. Additions to the monograph on Osteospermum. Bot. Notiser
113(4): 385-399.

Norlindh, T. 1976. A new species of Osteospermum (Compositae). In: R. K.
Brummitt, Notes arising from the Wye College expedition to Malawi 8. Kew
Bull. 311): 171-172.

Thunberg, C. P. 1800. Prodromus plantarum capensium 2. Upsaliae.

50 Comp. Newsl. 25, 1994

REQUEST FOR MATERIAL

Ms. Gisela Sancho has started a taxonomic revision of Gochnatia sect.
Moquiniastrum for her doctoral dissertation. She is requesting fresh leaf material
and/or seeds of plants belonging to this genus, or information on how to obtain
such material.

Please send plant material or information to

Lic. Gisela Sancho

Dep. Cientifico de Plantas Vasculares
Facultad de Ciencias Naturales y Museo
Paseo del Bosque s.n.

LA PLATA

1900 ARGENTINA

—_ J — — —- i - a a a, | Oe

Sa Gurcit Cri wes rPsdeb,y.

ae®<%

LIBRARY

JAN 11 1995

NEW YORK
PO TANICAL GARDEN