Botanical

classification and
nomenclature
an introduction —
Marc S.M. Sosef
Jérôme Degreef
Henry Engledow
Pierre Meerts
Botanical
classification and
nomenclature
an introduction —
Marc S.M. Sosef
Jérôme Degreef
Henry Engledow
Pierre Meerts
by Marc S.M. Sosef1, Jérôme Degreef1,2, Henry Engledow1 & Pierre Meerts3
1
Meise Botanic Garden, Nieuwelaan 38, B-1860 Meise, Belgium
2
Service Général de l’Enseignement supérieur et de la Recherche scientifique,
Fédération Wallonie-Bruxelles, Rue A. Lavallée 1, B-1080 Brussels, Belgium
3
Herbarium et bibliothèque de botanique africaine, Université Libre de
Bruxelles, Av. F.D. Roosevelt 50, CP 265, B-1050 Brussels, Belgium

Copyright © 2020 Meise Botanic Garden, Nieuwelaan 38, 1860 Meise, Belgium.
Printed in Belgium by Gewadrupo, Arendonk.

This publication is published and distributed in Open Access under the Creative
Commons Attribution 4.0 International license (CC-BY 4.0), which permits
use, distribution, and reproduction in any medium, provided the original work
is properly cited. A PDF file of this publication can be ordered, free of charge
(send an email to webshop@plantentuinmeise.be), or downloaded from the
webshop of Meise Botanic Garden at http://shopbotanicgarden.weezbe.com.

DOI: 10.5281/zenodo.3706707

CIP Royal Library Albert I, Brussels

Botanical classification and nomenclature, an introduction. Marc S.M.

Sosef, Jérôme Degreef, Henry Engledow & Pierre Meerts - Meise, Meise
Botanic Garden, 2020. - 72 p.; ill.; 22 x 15 cm.

ISBN 9789492663207
Subject: Botany
D/2020/0325/002
 Content

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

1. The history of classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

1.1 Theophrastus to the Middle Ages . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.2 Renaissance, Pre-Linnean period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.3 Linnaeus and the Linnaeans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
1.4 Evolutionary thinking enters classification theory . . . . . . . . . . . . . . . 18
1.5 Phenetics, cladistics and phylogenetics . . . . . . . . . . . . . . . . . . . . . . . . 19
1.6 On natural groups, monophyly, paraphyly and polyphyly . . . . . . . . 22

2. Species concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.1 What is a species? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.2 Speciation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.3 Infraspecific taxa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

3. Rules of botanical nomenclature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

3.1 ICN; the book of law . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2 From Kingdom to subforma, mandatory categories . . . . . . . . . . . . . 33
3.3 The type concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.4 Valid and effective publication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.5 Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.6 Author names, new taxon names, new combinations . . . . . . . . . . . . 39
3.7 Accepted names and synonyms: the priority rule . . . . . . . . . . . . . . . 40
3.8 Hybrids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.9 Cultivated plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

4. The art of identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

4.1 Identification keys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.2 Multi-entry keys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.3 DNA barcoding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
4.4 Identification of herbarium specimens . . . . . . . . . . . . . . . . . . . . . . . . . 51
5. The art of preparing a taxonomic revision . . . . . . . . . . . . . . . . . . . . . . 53
A. Taxon names and literature study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
B. Herbarium observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
C. Data basing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
D. Geographical and ecological observations . . . . . . . . . . . . . . . . . . . . . . 61
E. Taxonomic and nomenclatural decisions . . . . . . . . . . . . . . . . . . . . . . . . 62
F. Preparing taxon descriptions, treatments, illustrations and keys . . . . . 63
G. Producing the manuscript and publishing . . . . . . . . . . . . . . . . . . . . . . . 66

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Introduction

5
6
Biology is the science that explores the living world around us. To
communicate the wonders of nature, names are needed to describe the
variety of forms we encounter. This wildly diverse nature may be represented
through a hierarchical structure where names are used to indicate groups
of organisms at different levels. The classification and naming of organisms is
an essential tool for scientific communication. It forms the foundation upon
which biological research is based and the discipline is called “Taxonomy”.
Taxonomists explore, describe, name, and classify all living organisms on Earth.

Correctly classifying and naming organisms is crucial to a wide range of

biological research fields. Such a framework is also essential to address the
topics of sustainable usage and biodiversity management and conservation,
including their legal context. This booklet provides an overview of the most
important elements and processes of the classification and naming of plants
and fungi, hence the field called “Plant Taxonomy”. We will commence with
a historical overview of this discipline. Although we do not deal exhaustively
with the methods to reconstruct evolutionary pathways (phylogenetics), we
do provide the historical context of its development and some elements
that directly influence taxonomic decisions. The aim of this booklet is to
provide an introduction and practical guide to this research field. As such, it
can be used by those having a specific interest in the classification and naming
of plants, but also by those teaching this subject in secondary and tertiary
education institutions.

Although the information is general, most examples are drawn from tropical
African plant and fungal diversity. Each chapter is followed by an overview of
literature and web-based sources related to the subjects dealt with. This is by
no means exhaustive, and again focusses on the taxonomy of African plants
and fungi.

The authors hope that this publication will contribute to the development 7
of taxonomic expertise, notably in the Central African region. The booklet is
produced in English and French, and will be available free of charge (under the
CC-BY license) to high schools and universities (both teachers and students),
courtesy of Meise Botanic Garden.
8
 1.

The history of
classification

9
Any biological study starts with the simple question “What is this?”. Whether
it concerns a manager of a nature reserve who needs to know which species
grow within the park’s boundaries for setting up a management plan; a
primatologist studying the food eaten by chimpanzees; or a plant breeder
studying the close wild relatives of the potato in search of a disease resistant
gene, all need to be able to identify and name their material. It is preferable
that these names are uniformly used and accepted throughout the world.

The need for a uniform system of naming the living world was already
recognized by the ancient Greeks and Romans. Names were given to ‘entities’
we now call species that had specific morphological characters and uses.
Some produced, for example, edible fruits, or a yellow dye, others had
medicinal properties, or were useful for making musical instruments, etc.

In this chapter (largely based on Magnin-Gonze 2009 and Rouhan & Gaudeul
2014), we highlight the major historical phases in the development of plant
naming and classification. The name for this field of science was coined for
the first time in 1813 by the Swiss botanist Augustin Pyramus De Candolle
(1778–1841) in his book “Théorie élémentaire de la Botanique”. He created
the neologism “taxonomy” by combining the words Greek ταξις (order) and
νόμος (law, rule).

1.1 From Theophrastus to the Middle Ages

Even before the invention of written language, c. 5600 years ago, it is likely
that an oral plant classification system existed. Initially, names and organisms
were not placed into a hierarchical system since the plants were all named
following their use such as food, medicines, poisons, or materials (Raven 2004).

The Greeks probably did not just consider plants as being only useful but also
as beautiful; the murals in Knossos (1900 BC) not only have useful plants like
barley, fig, and olive, but also narcissus, roses, and lilies.The Greek Theophrastus
(372–287 BC; figure 1), successor of the great philosopher Aristotle, is espe-
cially well known as the first true botanist. Interested in naming plants and
finding an order in the diversity of plants, he is the first one to provide us
with a philosophical overview of plants. He pointed out some of the impor-
tant questions that would later define taxonomy, such as “What have we
got?” or “How do we differentiate between these things?” Moreover, he was
the first to discuss relationships among plant species and to suggest ways to
group them. Theophrastus described ca. 500 plants — probably representing
all known plants at that time – and classified them as trees, shrubs, subshrubs,
and herbs. He also made a distinction between flowering and non-flowering
plants, deciduous and evergreen trees, and be-
 Figure 1. Statue of tween terrestrial and aquatic plants. Even if 80% 11
Theophrastus in the of the plants included in his works were cultivated,
botanic garden at he had realized that “most of the wild kinds have
Palermo, Italy. no names, and few know about them,” highlighting
the need to recognize, describe, and name plants
  Figure 2.
Plinius the Elder.

 Figure 3.
Dioscorides.

growing in the wild (Pavord 2005). He soon discarded

his trees, shrubs, subshrubs and herbs classes in favour
of floral morphology that was better suited to cluster
plants into more natural groups. Theophrastus was way
ahead of his time, so much so that his botanical ideas and
concepts became lost in Europe for many centuries. His
works survived in Persia and Arabia, and were translated back into Greek and
Latin when rediscovered in Europe in the 15th century. During this long Dark
Age for botany, like for all other natural sciences in Europe, the Roman Plinius
the Elder (23–79 AD; figure  2) and the Greek Dioscorides (~40–90 AD; fig-
ure 3) were two important figures. Although they did not improve the exist-
ing knowledge and methods about the description, naming, or classifications
of plants, they did compile the available knowledge and their written works
were renowned and widely used. For many centuries, the Naturalis Historia
of Plinius and the De Materia Medica of Dioscorides (figure 4) were the only
12 source of information on plants throughout Europe and their works were
repeatedly copied. ’Herbalists’ tried to link plants they found in France or Brit-
ain to those described from the Mediterranean by Plinius and Dioscorides,
and one can imagine the problems they encountered.Throughout the Middle
Ages, hardly any new knowledge was added to the old works.
1.2 Renaissance, Pre-Linnean period

The Renaissance (late 14th to 17th century) marked a new era for science.
Europeans were exploring and discovering America, Africa, Asia and Australia,
bringing back many unknown plants to Europe. These were housed in a
rapidly increasing number of gardens, the first being created in the early
16th century in Italy. At first, these were called medicinal gardens, later, when
the interest shifted towards the study of plants themselves rather than their
useful properties, they transformed into the botanical gardens that we
know today. Moreover, with the invention of the
 Figure 4. Page from printing press (1450-1455), information was more
Dioscorides’ Materia easily shared and distributed, boosting exchange
Medica showing and discussions in scientific knowledge. People
Cassia fistula. became curious about the world surrounding
them. Around 1530, in the botanic garden of Pisa,

13
14
the Italian Luca Ghini (1490–1556) invented a revolutionary method for
preserving plants by drying and pressing them so they could be studied at any
time of the year. The resulting plant specimens were stored in books known
as a “hortus siccus” (dried garden), later the term “herbarium” was adopted,
and were valuable possessions afforded only by the wealthy (Ghorbanie et
al. 2018; figure 5).

This was followed by the era of the great western European herbals, or
books describing the plants and their uses. These works were no longer
produced only in Latin (the scientific language of the time), but also in the
common languages German, English, Dutch and French. This opened up the
information on plants to an even wider public. From this period, the herbals
of Dodoneus (Cruydeboeck, 1554; figure 6), Fuchs
 Figure 5. Frontpiece of (New Kreüterbuch, 1543) and Gerard (Herball, or
the Rauwolf herbarium Generall Historie of Plantes, 1597) are the most
1573-1575 kept at famous. Advances in art led to numerous new
Naturalis Biodiversity plant illustrations. These were way superior to
Center, Leiden. those copied over and over from the books of
Dioscorides and Plinius, from which the actual
 Figure 6. Dodonaeus
species could often hardly be deduced.
and two pages from his
famous illustrated herbal A student of Ghini, Andrea Cesalpino (1519–
(Cruydeboeck) printed 1603), was the first to discuss the work of
in 1554. Theophrastus since the Ancient Greeks. He
pointed out that plants should be classified in a

15
 more natural and rational way. His De Plantis  Figure 7.

Libri XVI (1583) describes 1,500 plants which Carolus Linnaeus.

he organized into 32 groups including the
 Figure 8. Linnaeus’
Umbelliferae and Compositae. The science of
naming plants quickly developed, and in general sexual system to classify
the plant names comprised a combination of plants.
several characteristic features. For example, the
passion flower (Passiflora edulis) was called Flos passionis major (large passion
flower). However, with the rapidly increasing number of species arriving from
around the world, more characters were needed to distinguish one species
from the other, resulting in ever longer names. In an early catalogue of the
Hortus Botanicus at Leiden (The Netherlands) founded in 1592, the same
Passiflora was called Cucumis Flos Passionis dictus triphyllos flore roseo clavato
(Cucumber or Passion flower, three-leaved, pink flowered and clavate; the
latter possibly pointing to the shape of the styles). In short, the name of
a plant also served as a diagnostic summary. The science of botany slowly
diversified from the science of medicine to a broader study of the increasing
wealth of plants arriving in Europe from all over the world. In 1623, the Swiss
Gaspard Bauhin published his Pinax theatri botanici describing no less than
5,640 different plants, wild species but also many cultivated forms. Later, the
British botanist John Ray published his 3-volume Historia plantarum species
(1686, 1688, 1704) containing more than 17,000 different ‘species’ (he also
described a very high number of cultivars, monstrosities and other forms).
This innovative work was the first to distinguish Monocotyledons from
Dicotyledons, and to use text-based dichotomous keys to classify plants. In
16 1694, the Frenchman Joseph Pitton de Tournefort, developed the concept
of genera, which contributed markedly to a better structuration of the
classification.
1.3 Linnaeus and the Linnaeans

In the first half of the 18th century, the bright young Swedish botanist, Carl
von Linné (figure 7), or Linnaeus in Latin, brought order where there was
chaos. While working in The Netherlands, he met famous professors such as
Hermann Boerhaave, Adriaan van Royen and Johannes Burmann with whom
he discussed several of his new ideas.

First, he created a clear classification system of plants based on the number

of stamens and styles in each flower and called it his ‘sexual system’ (figure 8).
He recognized five taxonomic levels, the variety, the species, the genus, the
order (± equal to our present-day family) and the class. This simple system,
although it had a few flaws, worked remarkably well in creating structure.

 Figure 9. Page from Second, he suggested to dissociate the name of

Linnaeus’ famous Species a plant from its description. In his famous Species
plantarum, showing his plantarum, published in 1753, he wrote what he
“trivial names” in the called ‘trivial names’, a single word, in the margin
margin. of each species treatment (figure 9). Preceded by
the genus name, that would form a species name

17
 composed of only two words. It was the start of the binomial (two words)
nomenclature we still use, a system where a species name is composed of
the genus name followed by a word indicating the species, called the epithet.
Soon, other botanists appreciated the simplicity and genius of the new
naming system and adopted it in their own work. Shortly after the success of
his Species plantarum, Linnaeus, being a keen zoologist as well, introduced the
same system for animals in his famous Systema naturae (1758).

Linnaeus travelled to England to meet Sir Hans Sloane and Johann Jacob
Dillenius, who were both sceptical about his new naming and classification
ideas at first, but they came around after several years. In Paris, he met
Bernard de Jussieu, who would together with his nephew Antoine Laurent de
Jussieu publish their Genera plantarum. In that work, they stated that a species,
genus, or any other class in the hierarchical classification, which we now call a
taxon (plural taxa), should group plants showing character constancy within
the given taxon, as opposed to the character variability observed among taxa.
Since not all characters are useful at the same level of the classification, their
principle of subordination led to a character hierarchy: characters displaying
high variability should be given less weight than more conserved ones in plant
classifications.

In this period, classification and the study of nature also had a religious
implication. Biologists were seen as scientists studying the living things which
God has created and placed on Earth. Linnaeus, in his Introduction to his
Species plantarum, wrote: “In his omnipotent omniscience, God created
the theatre of all living beings on earth, and it is our divine task to explore
that great creation, served to us as tasty treats, unworthy as we are, and to
recognize His hand in it” [freely translated from the Latin]. One can imagine
that in this context Darwin’s introduction of the novel idea that species were
not created by God Almighty, but had evolved from others over a very long
period of time, had a tremendous impact on broader society.

1.4 Evolutionary thinking enters classification theory

At the start of the 19th century, new questions arose in the mind of taxonomists.
They were not only interested in naming, describing, and classifying organisms,
but also in the origin of the observed diversity. In 1809, in his Philosophie
zoologique, the zoologist Jean-Baptiste de Lamarck proposed a theory where
species could evolve and change through time.

It took another 50 years before Charles Darwin (1809–1882; figure 10) pub-
lished his famous theory of evolution and survival of the fittest in On the
Origin of Species (1859). Independently, Alfred Russel Wallace (1823–1913)
18 had come to the same conclusion while working in Asia. [In fact, the theory
was published already in 1858, in a paper authored by Darwin and Wallace
in the Journal of the Proceedings of the Linnean Society: Zoology.] Darwin
introduced the central concept of descent with modification that later re-
ceived extensive support and is still generally accepted today. The concept
 Figure 10.
Charles Darwin.

of evolution had a major impact on the development of the

theory behind classifying nature, hence the science of taxono-
my. Biologists understood that since the history of life is unique,
the only classification that is natural is the one reflecting that
unique tree of life, the phylogeny. The latter word was not
coined by Darwin but by Ernst Haeckel (1834–1919) in 1866
in Generelle Morphologie der Organismen. Darwin predicted that “our classifi-
cations will come to be, as far as they can be so made, genealogies” (Darwin
1859, p. 486). This new theory also implied that characters useful to taxono-
my are those inherited from a common ancestor. However, Darwin did not
provide any new techniques or approaches to reconstruct the phylogenetic
tree for a certain group of taxa or assist practicing taxonomists in their work.

1.5 Phenetics, cladistics and phylogenetics

 Figure 11. Example In the early 1960s, a new technique called ‘nu-
of a phenogram, with merical taxonomy’ arose to produce a tree-like
a measure of similarity output, or phenogram (figure 11), on which one
(Manhattan Distance) could subsequently base a classification. Notably
along the x-axis (from the work of Sokal & Sneath (1963, and later edi-
Pometti et al. 2007). tions), Principles of numerical taxonomy, laid the

19
 foundation. The technique, also called phenetics, was based on a quantitative
cluster analysis of overall similarities between taxa, using a characters-by-taxa
data matrix - with a mixture of binary characters (stipules present, yes/no),
multi-state characters (for example flower colour, with states 1=white, 2=yel-
low, 3=blue), or continuous characters (for example calyx length in mm)
- and resulting in pairwise distances among the individuals or taxa, called
OTUs (Operational Taxonomic Units). However, it was soon realized that
overall similarity does not necessarily indicate
an evolutionary relationship. For example, spe-
cies may have developed similar features be-
cause they adapted to the same environmental
stress. As this method was not based on the
evolution theory, it could not interpret the
observed variation in an evolutionary sense
with respect to ancestors and descendants or
observed character state changes. Despite the
fact it produces tree-like phenograms, these do
not represent a natural and evolutionary clas-
sification. Nevertheless, this theory flourished
for a while, greatly benefiting from rapid ad-
 Figure 12.
vances in informatics.
Willi Hennig.
It was the German zoologist Willi Hennig
 Figure 13. Example
(1913–1976; figure 12) who fundamentally
of a cladogram, showing
changed the way biologists reconstruct the
numbered characters
evolutionary pathway of a taxonomic group.
where one may recognize
In 1960 he published his cladistic theory in
apomorphies (black dots),
Grundzüge einer Theorie der Phylogenetischen
parallellisms (open dots)
Systematik, but it remained relatively unknown
and reversals (hatched
until the English translation Phylogenetic Sys-
dots) as well as their
tematics was published in 1966. The primary
character state changes
principle is not to use the overall similarity
(below each dot).

20
among taxa to reconstruct the phylogeny, but to make a distinction between
primitive character states and those that are derived from them. Only de-
rived character states, called apomorphies, shared by several taxa then indi-
cate a shared common ancestry, while primitive ones, called plesiomorphies,
do not. A group derived from a single common ancestor is called a clade and
the theory behind it cladistics. The result of a cladistic analysis is a tree-like
figure called a cladogram (figure 13), where the branches actually represent
one or more character state changes. When, for example, in a group of plants
with red flowers an evolutionary change gave rise to blue flowers, ‘blue’ then
is the derived state of the character ‘flower colour’ and any species having
that state is likely to have evolved from the same common ancestor. Having a
red flower does not indicate such common ancestry and hence cannot serve
as a criterion on which to base a taxonomic group. When, further down the
evolutionary history it turns out that a red colour was derived from white
flowers, red can still be regarded as a derived character state but at a differ-
ent level in the phylogeny. And going from white to blue then requires two
evolutionary steps rather than one. When a derived state evolves back to the
primitive state again, this is called a reversal, while the independent evolve-
ment of the same character state in two or more different branches of the
evolutionary tree is called a parallelism.

Furthermore, Hennig argued that every taxonomic decision, from a

species definition to a system of higher classification, was to be treated as a
provisional hypothesis to be tested by new data or applying other methods.
Various algorithms were developed with which a cladogram could be built
from a character state/taxon matrix (see also figure 18) and the method
benefited from the rapid increase in computational capacity of computers
and the development of bioinformatics. New research fields like cytology and
chemotaxonomy provided additional character sets. The algorithms aimed to
find the cladogram that needed the smallest number of evolutionary changes
(or steps). The argument being that those needing the smallest number of
changes (or hypotheses) represent the most likely phylogeny.This ‘lowest cost’
idea was called the parsimony principle. The shortest tree then is the most
parsimonious one.

In this new setting, it was felt that a new definition of this biological research
field was needed and the term ‘systematic biology’ or simply ‘systematics’ was
coined (Michener et al. 1970). It embraced the entire field from describing,
naming, classifying, studying the distribution patterns (biogeography),
evolutionary relationships, character evolution and adaptations. The term
‘taxonomy’ was then restricted to describing, naming and classification. Some,
however, treat these two words as synonymous.

The discovery of the double helical structure of the DNA molecule in 1953,
by James Watson and Francis Crick, greatly improved our understanding of
the evolutionary processes. But it was only after it became possible to tar-
get specific fragments of the genome (nuclear, mitochondrial or chloroplast 21
DNA) by selectively amplifying the DNA through the polymerase chain reac-
tion (PCR) (Karry Mullis 1986) that it started to have a dramatic impact on
taxonomy and classification. The introduction of DNA sequence data (Meier
2008) offered access to numerous new characters and statistical approaches.
 Thus, at the turn of the 21st century, the use of molecular data and new tree
building algorithms such as Maximum Likelihood and Bayesian statistics led
to a distinct improvement of our abilities to produce phylogenetic hypothe-
ses. The ‘strength’ or reliability of each branch in a cladogram can be assessed
by using other techniques such as bootstrapping (a statistical resampling
technique; Holmes 2003) and, again, Bayesian statistics. All these develop-
ments led to improved insight into the delimitation of orders and families of
flowering plants (Angiosperm Phylogeny Group 2016), as well as a greater
understanding of the classification based on evolutionary relationships.

1.6 On natural groups, monophyly, paraphyly and polyphyly

It logically follows from the above that classifying the natural world into spe-
cies, genera and higher groups has become a search for our best hypothesis
on the structure of the evolutionary tree in order to be able to distinguish
natural groups. In other words, the cladogram produced by one of various
analyses needs to be split into natural parts. However, there are many ways
to do this, and requires informed choices.

Firstly, we need to understand that a

cladogram is not a phylogenetic tree but
a schematic representation of the data,
showing character state changes (mor-
phologic, chemical or genetic) on its
branches (figure 13). Since in systemat-
ics we do not aim to classify characters,
but rather species (or taxa), we need to
transform such a cladogram into a true
phylogenetic tree showing the ances-
tor-descendant relationships between
individuals, or populations, or species. In
a phylogenetic tree, the branches rep-
resent the evolutionary relationships
between the units involved in the evo-
lutionary process (see figure  14) and
hence it can indeed be used to derive a
classification from it.

 Figure 14. Nature of the

branches of a phylogenetic
tree, with ancestor-
22
descendant relationships
between individual
organisms.
Secondly, we need to understand that in our system of nomenclature, rules
are such that some categories are mandatory. All species belong to at least
a genus, and every genus belongs to a family. When we create a subgenus to
accommodate some of the species within a genus, we are forced to make
one or more other subgenera to accommodate the remaining species (see
also Chapter 3). We have to keep this in mind when applying rules to divide a
phylogenetic tree into taxonomic groups that carry a formal name.

The majority of taxonomists will tell you that a classification can only be
natural when it is composed exclusively of so-called monophyletic units;
that is a group of species that includes an ancestral species and all members
derived from that one ancestral species (figure 15), also named the most
recent common ancestor (MRCA). [Note: a single species can represent a
‘group’ comprising a single element.] When some, but not all, of the species
derived from the most recent common ancestor are included, that group is
called paraphyletic (figure 15). The problem is that while mathematically a
cladogram, with taxa represented only at its tips (see figure 13), might be fully
chopped up into monophyletic groups (the nodes of the cladogram are said

 Figure 15. Phylogenetic to represent the character distribution of the po-

tree illustrating the tential ancestors), it is impossible to do so when
meaning of monophyly, using a phylogenetic tree. Every time a new spe-
paraphyly and polyphyly cies splits off from its ancestral one, it may start
(further explanation see a new monophyletic group but will always leave
text). behind a paraphyletic remainder (Brummitt 2002,
Sosef 1997, Horandl 2006, Podani 2010). Many
prefer to only distinguish the ‘nicer looking’ monophyletic groups, but few
seem to realize they subconsciously make the choice to base their classifica-
tion on a cladogram rather than on a phylogenetic tree. Consequently, such a
strict monophyletic classification is not only less natural than one that allows
paraphyly, but is often unable to cope with extant ancestral species, fossils,
or recently extinct species. An example of the latter is the sabretooth tiger
or the mammoth where these species represent the paraphyletic remainder
of extant species and thus inevitably lead to non-monophyly. A single extant 23
species can either be monophyletic (when it contains all descendants of a
single ancestral origin) or paraphyletic (when it gave rise to a new species)
and building a strictly monophyletic classification with such components is
mathematically impossible. Some tried to circumvent this ‘unwanted’ situa-
 tion by proposing a convention in which we would agree that all species
are monophyletic by definition, clearly a theoretical horror. At one point, a
revolutionary new concept of providing names to taxa was developed, called
the PhyloCode (de Queiroz 2006), which involved throwing out the idea
of any mandatory taxon levels such as genus or family (apart from species).
This means that some species may belong to a genus, but others do not and
for example only belong to a family. Theoretically, it is probably a better no-
menclatural system that would also allow a strict monophyletic classification,
but pragmatically systematists did not want to throw the Linnaean binomial
system away and adopt such a rigorous new one.

Finally, a polyphyletic group is a group of species where the most recent

common ancestor belongs to a different group, or where the members arose
from more than a single MRCA (see figure 15). When in the past such groups
have been recognized as taxonomic entities, it was probably due to species
that shared one or more plesiomorphic characters, or one or more char-
acteristics that were not inherited from a common ancestor. For example,
non-related species living in a desert might independently develop scale hairs
to protect them from dehydration. Such results of parallel or convergent
evolution are called homoplasies, i.e. a homoplasious character looks the
same but has a different evolutionary origin. All agree that such groups are
non-natural and should be eliminated from a classification.

After having decided which rules one wants to follow to chop up a phyloge-
netic tree (or a cladogram) into taxa, there are still many choices one can
make that renders the process of classification and naming of taxa partly
subjective. “Which part of the tree will I recognize as a genus?” “Or, would it
be better to call it a subgenus?” etc. are true questions one needs to answer.
Making choices that will cause least perturbation in the existing system is also
a valid argument thereby promoting name stability.

General literature on systematics

• Spichiger R-E., Figeat M., Jeanmonod D. (2016) Botanique systématique

avec une introduction aux grands groupes de champignons. 4ème
édition. Lausanne, Presses Polytechniques et Universitaires Romandes.
ISBN 978.2889151349

• Stace C.A. (1991) Plant Taxonomy and Biosystematics, 2nd ed.. London,
Edward Arnold. ISBN 071.3129557

• Stuessy T.F. (2002) Plant taxonomy, the systematic evaluation of

comparative data. M/s Bishen Singh Mahendra Pal Singh.
24 ISBN 978-8121102841

• Stuessy T.F., Crawford D.J., Soltis D.E., Soltis P.L. (2015) Plant
Systematics. The Origin, Interpretation, and Ordering of Plant
Biodiversity. ISBN 978-3874294522
 2.

Species
concepts

25
2.1 What is a species?

Biologists generally agree that the species is a fundamental natural unit. How-
ever, it has proven incredibly difficult to define what exactly a species is! This
controversy occurs notably on a theoretical rather than a practical level and
has come to be known as ‘the species problem’.
One of the most fundamental aspects of the problem is variation. Most, if
not all, animal and plant species show variation, every individual often be-
ing demonstrably unique. Within a population variation can be continuous
(e.g. height or weight) or discontinuous (e.g. sex; having right- or left-spiralled
corolla lobes), environmental in origin (e.g. flower colour influenced by the
composition of the soil) or genetic (e.g. blood type). Variation can also oc-
cur in space between populations (geographical variation). Even when two
individuals share exactly the same DNA (clones or twins) they may develop
morphological differences under the influence of environment factors; this
is called phenotypic plasticity. The species problem is, in part, a history of
how biologists have tried to address variation. Often, species are thought to
represent a natural unit. The most extreme opposing view to this idea, states
that only individuals exist in nature. Taxonomic groups, including species, are
then seen as man-made abstractions that allow us to conveniently group
large numbers of individuals. Few scientists accept this nominalist approach
with respect to species, but many believe it does apply to higher taxa (World
Conservation Monitoring Centre 1992).
Multiple definitions and species concepts have been proposed. These usually
follow the discipline of the author: the taxonomic species concept, the evolu-
tionary species concept, the ecological species concept, the historical species
concept, and many more. Species concepts can be divided into two main
groups, those concerned with process (evolution, interbreeding) and those
concerned with pattern (morphology, ecological preferences). Below are the
three most widely known ones:
The biological species concept. This concept defines species in terms
of interbreeding. Its biggest advocate was undoubtedly Ernst Mayr, an
ornithologist. He defined species as “groups of interbreeding natural populations
that are reproductively isolated from other such groups”. Later, it was refined
to “a population or group of populations whose members have the potential to
interbreed in nature and produce viable, fertile offspring, but do not produce viable,
fertile offspring with members of other such groups”. It remains the most widely
accepted species concept today. It explains why the members of a species
resemble one another, and differ from other species. The members exchange
genetic material and pass it on to their offspring, but not to other species.
Thus, the evolutionary process involves random mutations that remain inside
a gene pool that has acquired some form of isolation. Over time these
novelties will start to differentiate these populations from other similar gene
pools (or populations). Finally, these differences may lead to reproductive
27
isolation, where the isolated gene pools will start acting as species.
In general, zoologists embrace this species concept, however it poses some
complications for plants. While hybrids in animals are rare, in plants many spe-
cies are known to hybridize and produce fertile offspring (Grant 1981, Stace
 et al. 2015). Only if such events are rare and the offspring are less viable, can
the parent species maintain their unique genetic identity and hence be recog-
nized as distinct species. Furthermore, the concept does not apply to asexual
organisms, and in plants for example the occurrence of apomixis would not
allow defining a species according to the biological species concept.
The morphological species concept. This concept characterizes a species by
morphological distinctiveness and is applied to both asexual and sexual or-
ganisms. It can be applied when information on gene flow is unknown e.g.
when only herbarium specimens are available. Researchers may disagree on
which characters to use to differentiate species which leads to subjectivity.
The evolutionary species concept. This concept stresses the importance of a
species being an evolutionary unit. It defines them as “a lineage of interbreed-
ing organisms, reproductively isolated from other lineages, that has a beginning,
an end, and a distinct evolutionary trajectory and historical fate” (Wiley 1978). It
is definitely the least practical concept, but does include time as an essential
element.
Whatever concept a scientist uses to distinguish a species, the delimitation
actually represents a hypothesis about the relationships among the individual
organisms belonging to the species. Such a hypothesis about which group of
individuals forms a species may be tested using morphological, genetic, behav-
ioural or other types of evidence.

2.2 Speciation

In an evolutionary context, based on progressive change, species are variable

in space and will change over time. Such changes may eventually result in the
formation of one or several new species. This process generally involves two
processes: isolation, where one or more individuals of an existing species
are no longer able to interbreed and therefore no longer exchange genetic
material with the other individuals of the same species, leading to divergence.
The latter process involves the accumulation of random mutations, gradually
or instantly, where the acquisition of new features may cause two isolated
entities to become substantially different and be considered distinct species.
Both processes may influence each other. Partial isolation, where on rare oc-
casions genetic material is still exchanged, may reduce the speed with which
two entities may diverge. Similarly, divergence itself may increase the isolation
of a population.

Figure 16 shows three possible processes leading to speciation. The easiest

to understand is ‘cladogenetic speciation’ where part of an existing species
(sometimes comprising a single individual) is separated and becomes isolated.
Think of a single seed that is blown across an ocean to a remote island. After
28 arrival, it will be the founder of a new population that will gradually accumu-
late random mutations and hence diverge from its ancestral populations on
the mainland. Note that such a process, also referred to as ‘budding’, does not
alter the nature of the parental species which may continue to exist while
the new species differentiates. The second process is ‘anagenetic speciation’,
 Figure 16. Speciation where a species slowly accumulates random mu-
processes (further tations over time and becomes substantially differ-
explanation see text). ent from its ancestral populations thereby being
recognized as something different. In this case, the
‘isolation’ is through a separation in time. Notably palaeontologists, working
with fossils from different time frames, will want to define such groups of
individuals as different species. Finally, species may arise instantly through hy-
bridization, especially when followed by duplication of the genome resulting
in polyploid organisms unable to cross with members of the parental popula-
tions. The latter speciation mechanism is rare in animals but fairly common in
plants (Grant 1981, Soltis & Soltis 2009).

The speciation process is closely linked to the presence of reproductive

isolation mechanisms preventing interbreeding. Below is an overview of such
mechanisms, divided into two groups: pre-mating (in plants before pollination)
and post-mating (after pollination) isolation mechanisms.

1) Pre-mating isolation mechanisms (in plants):

a) Geographical isolation. Individuals occur in different geographical areas
separated by a barrier that cannot be crossed by pollen, seed or spores.
b) Temporal isolation. Pollen is not exchanged between species because
they flower at different times of the day or in different seasons.
c) Ecological isolation. Individuals occupy different habitats, and therefore
pollen is not transferred to other species with different ecological
preferences.
d) Behavioural isolation. Related species may attract different animals as
pollinators.
e) Mechanical isolation. Physical non-correspondence of flower parts
prevents the transfer of pollen to the style, such as in heterostylous flowers.

2) Post-mating isolation mechanisms (in plants):

a) Gametic incompatibility. Pollen does reach the style but does not
germinate or the pollen tube does not reach the egg cells.
29
b) Zygotic mortality. The pollen nucleus does reach the egg cell, but the
zygote does not develop.
c) Hybrid inviability. A hybrid embryo or plant is formed, but has a reduced
viability.
 d) Hybrid sterility. The hybrid plant is viable, but is sterile and does not
produce seeds.
e) Hybrid breakdown. First generation (F1) hybrids are viable and fertile,
but further hybrid generations (F2 and backcrosses) may be less viable
or sterile.

2.3 Infraspecific taxa

Evolution generally is a fairly slow process (apart from some situations

involving hybridization). It may take thousands of years before a population
that has become isolated will eventually develop into a distinct species. Some
DNA mutation may appear, but then disappear again, while others remain
although they do not necessarily lead to a phenotypical differentiation. As a
result, when we observe the living world, we will see variation at many levels
which arises through various processes. We may well witness a time slice in
the process of speciation and observe a species in the making. In some cases,
when the pattern is discontinuous, we may want to capture such variation
in distinct infraspecific taxa. While zoologists only recognize the infraspecific
level of subspecies, plants and fungi use subspecies, varieties and formae.

A subspecies is defined as a part of a species (one or more populations) that

is morphologically or genetically distinct and generally also occurs in a distinct
geographical region.
A variety is defined as a part of a species (one or more populations) that
is morphologically or genetically distinct, but which is generally encountered
within the distribution area of the species as a whole. It often occupies a
different habitat and is thus ecologically distinct.
A forma is defined as a part of a species that is morphologically or genetically
distinct, but which represents a mutation that occurs sporadically within a
population.

Species concepts

• Mayr E. (1982) The Growth of Biological Thought. Cambridge (MA),

Harvard University Press. ISBN 978-0-674-36445-5.
• Pavlinov I., editor (2013) The Species Problem. Ongoing Issues. DOI:
10.5772/3313. ISBN 978-953-51-0957-0. https://www.intechopen.com/
books/the-species-problem-ongoing-issues
• Reydon T.A.C., Kunz W. (2019) Species as natural entities, instrumental
units and ranked taxa: new perspectives on the grouping and ranking
30
problems. Biological Journal of the Linnean Society 126: 623–636.
https://doi.org/10.1093/biolinnean/blz013
• Rieseberg L.H., Wood T.E., Baack E.J. (2006) The nature of plant species.
Nature 440: 524–527. doi:10.1038/nature04402.
 3.

Rules of botanical
nomenclature

31
32
Once the character variation within a group has been studied and conclusions
have been drawn about which entities, or taxa, need to be distinguished, the
question arises as to what the correct names for these entities should be.
That is where one enters the realm of botanical nomenclature.

3.1 ICN: the book of law

After 1753, when Linnaeus introduced his binomial system, only some ele-
mentary rules for naming plants were developed. Later, in 1813, Augustin de
Candolle in his Théorie élémentaire de la Botanique provided a detailed set of
rules regarding plant nomenclature. However, over time it became apparent
that an internationally recognized and accepted system and rules for nam-
ing plants was necessary. It was Alphonse de Candolle, son of Augustin de
Candolle, who convened an assembly of botanists from several countries
to present a new set of nomenclatural rules. In 1867, he organized the First
International Botanical Congress (IBC) in Paris, which led to the publication of
the so-called Paris Code. Subsequent meetings of the IBC were held in 1892
(Rochester Code), 1905 (Vienna Code), 1907 (American Code) and 1912
(Brussels Code). A general agreement regarding internationally acceptable
rules for plant nomenclature was however only reached in 1930 at the IBC
meeting in Cambridge. Here, for the first time in botanical history, a Code
of nomenclature came into being that was both international in function and
name: the International Code of Botanical Nomenclature (ICBN). Today,
it is composed of a number of Principles, Rules and Recommendations laid
out in 61 Articles, as well as the Provisions for governance of the Code. It
looks quite similar to a book of law. Since 1930 many updates of the ICBN
have been produced. In 2011, the name changed to “International Code of
Nomenclature for algae, fungi, and plants (ICN)”. It also covers the fossils of
these groups (see Turland et al. 2018).

Proposals to amend the Code are published in the journal Taxon. Every 6
years, at the start of the International Botanical Congress, during the ‘Nomen-
clatural Session’ that may take a full week, taxonomists from all over the world
meet to discuss all proposals published during the interim period between
conferences. Each institute has a number of votes, depending on their number
of research staff. Basically, changes to the rules of botanical nomenclature are
decided through a democratic process.

The most important nomenclatorial rules are outlined below. It should be

noted that for more detail or more complex cases one should consult the
latest version of the Code. Although it may have been translated into several
other languages, the English version counts as the only official one.

3.2 From Kingdom to subforma, mandatory categories 33

Any taxonomic group, be it a family, species or variety, is referred to as a

‘taxon’ (plural ‘taxa’). Names of taxa above the level of species are composed
of a single word and those from Subtribe and higher have a specific ending.
 Only some of the taxon ranks are mandatory. Below is a list of the most
commonly used taxa in plants, algae and fungi, with their specific ending.
Mandatory ranks are given in bold.

Rank Plants Algae Fungi

Kingdom/Regnum -tae
Division/Phylum -phyta -mycota
Subdivision/Subphylum -phytina -mycotina
Class -opsida -phyceae -mycetes
Subclass -idae -phycidae -mycetidae
Superorder -anae
Order -ales
Suborder -ineae
Superfamily -acea
Family -aceae
Subfamily -oideae
Tribe -eae
Subtribe -inae

Below the level of Subtribe, taxon names do not have a specific ending. The
most important ones are (mandatory ones in bold):

Supergenus
Genus
Subgenus
Section
Species
Subspecies
Variety
Subvariety
Forma
34 Subforma

The names of non-mandatory taxa are composed of a single word that fol-
lows the nearest mandatory name above it. The name of a species is com-
posed of the genus name plus the species indication, hence a second word,
called the epithet. The name of an infraspecific rank is also composed of a
single word.The species epithet and all infraspecific names are always starting
with a lowercase letter, those of ranks above species start with a capital. Some
examples:

Amanita subgen. Amanitopsis

Begonia sect. Scutobegonia
Poaceae tribu Andropogoneae
Monotes rubriglans subsp. upembensis
Chlorophytum gallabatense var. micranthum

3.3 The type concept

The application of names of taxa at ranks above the family may be deter-
mined by the name of an included genus (e.g. Order Asparagales is derived
from the generic name Asparagus), or may be descriptive names (such as
Division Spermatophyta). The application of names of taxa at the rank of
family or below however is determined by means of nomenclatural types. A
nomenclatural type is the element to which a name is permanently attached,
whether that name is the accepted name or a synonym of another name
does not matter.

The type of a name of a species or infraspecific taxon is either a single spec-

imen conserved in a herbarium, or an illustration. The type of a name of a
genus (or of any subdivision of a genus) is the type specimen (or illustration)
of the name of the type species (the first species ever described in the genus
or designated as such by the author of the genus). The type of a name of a
family (or of any subdivision of a family) is the same as that of the generic
name from which it is formed. Note that the nomenclatural type is not nec-
essarily the most typical or representative element of a taxon. More on types
can be found in paragraph 3.5 below.

3.4 Valid and effective publication

The original (first) publication of a name is called the protologue. In order to

be formally accepted as a new name under the rules of the Code, the proto-
logue must fulfil several conditions. When it does not, the name is not accept-
ed by the ICN and is discarded from further processing. The most important
condition is that the new name must be both effectively and validly published.
35
In order to be effectively published the name must appear in print and be
available in (at least 2) publicly accessible places (like libraries) (Art. 29). From
1 January 2012 onwards, a publication in electronic format (PDF) is also ac-
cepted when it has an ISSN or ISBN number.
 In order to be validly published, a new name (of the rank of genus or below)
must be:

- effectively published

- associated with the taxonomic rank it represents (Art. 37), as of 1 Jan-

uary 1953 onwards;

- accompanied by a description or diagnosis indicating how it differs from

related taxa. Between 1 January 1935 and 31 December 2011, descrip-
tions or diagnoses must be in Latin, thereafter it may also be in English.
(Art. 39)

- accompanied by the indication of the type specimen, as of 1 January

1958 (Art. 40.1). After 1 January 1990, the Herbarium where the type
has been deposited must also be indicated (Art. 40.7). Herbaria are
generally cited by their standard acronym which can be found in Thiers
(continuously updated).

In botany, the name of the species epithet cannot be the same as the genus
name. In zoology this is allowed (Bufo bufo for the Common toad, or Giraffa
giraffa for the Southern giraffe). Such a name is called a tautonym and is inva-
lid under the botanical Code.

Occasionally, it happens that someone publishes a name that is exactly the

same as one that was published earlier. Both names are then called homo-
nyms and the more recent one is illegitimate under the botanical Code.

3.5 Types

Having the correct type specimen linked to a name is essential to plant (and
animal) nomenclature. A number of rules have been put in place to deal with
situations where the type may be uncertain.

In botany, a plant collection or gathering, is generally indicated by citing the

collector and their unique collecting number, for example Lebrun 1234.
[When a specimen has an associated barcode this can be cited additionally.]
In the field, a collector often takes several samples or specimens from the
same plant or population where these are given the same collection number
(Lebrun 1234). Hence, a single collection may comprise several duplicates,
which are often sent to various Herbaria in exchange for their duplicate
material. The type of a plant name, however, can only be a single specimen,
indicated as the holotype. Any existing duplicates of the holotype are called
the isotypes. Although the isotypes can be very useful for research, when it
comes to applying the rules of nomenclature only the holotype is considered.
Next to a dried plant or fungus, an illustration may also serve as the holotype.
36
When the protologue does not specifically mention the existence of one or
more duplicates, the specimen present in the herbarium where the author
worked or one that he or she definitely had access to while preparing the
description of the new taxon, may be regarded as the holotype.
All other collections cited in the protologue, but not belonging to the type
collection, are called the paratypes.

Before 1958, one was not obliged to indicate the type specimen for a new
name. As a consequence publications preceding this date regularly have pro-
tologues that do not mention a type but rather cite several collections con-
sulted by the author for the new taxon.These are then all regarded as ‘original
material’ and are called syntypes. As a name may only have a single type, one
must choose a type from amongst this original material (the collections cited
and all their duplicates). Such a chosen type is then called the lectotype.
Duplicates from the lectotype are then called isolectotypes. When someone
publishes a lectotypification, adding the phrase “designated here” is obligatory.
Once a type collection is assigned to the name, the remaining syntype mate-
rial automatically become paratypes.

In a situation where all original material, including any relevant illustrations, has
been lost (after proof of an exhaustive search), one is allowed to select a new
type that is then called the neotype. Duplicates of the neotype then become
isoneotypes. When creating a neotype, one often tries to select material that
was collected at or close to the original type locality, but this is not obligatory.
In general, one aims to select a neotype where nomenclatural stability is guar-
anteed, hence which does not lead to necessary name changes.

Finally, the holotype material may be too scanty to adequately diagnose a

taxon (note that the type can also be an illustration where some details may
not be apparent). In such a case, one is allowed to select a ‘supportive type’
known as an epitype, so as to leave no doubt about the identity of the taxon
concerned. Again, it is important to choose the epitype wisely, so as to guar-
antee nomenclatural stability.

It is important to note that the Code defines a ‘specimen’ as being a gathering

of a single species or infraspecific taxon that may comprise a single organism,
parts of one or several organisms, or multiple small organisms. A specimen is
usually mounted on a single herbarium sheet or in an equivalent preparation,
such as a box, packet, jar, or microscope slide.

Examples with types

Holotype and isotype citation:

Solanum aculeastrum Dunal (1852: 366). – Type: Afrique du Sud, Cape
of Good Hope, eastern part near Morleg, 1500 ft, 1838, Drège s.n.
(holo-: G-DC; iso-: AD, BM, K, P).
37
Explanation: The protologue of the species name Solanum aculeastrum
has been published by Dunal in 1852. The protologue mentions
Dunal saw a single specimen collected by Drège, without a collecting
 number (s.n. = sine numero), and states he saw this in the herbarium
of De Candolle, which is kept at Geneva. Hence, that specimen (in
G-DC) is to be regarded as the holotype. Later, duplicates of this
collection have been identified at Adelaide State Herbarium (AD),
the British Museum (BM), the Royal Botanic Gardens, Kew (K) and
the Muséum national d’Histoire naturelle, Paris (P).

Lectotype designation:
Anthephora elegans Schreb. var. africana Pilg. (Pilger 1901: 119). – Type:
D.R. Congo, Stanley-Pool, June 1899, Schlechter 12508 (lectotype: B
[B 10 0168252], designated here; isolectotypes: B [B 10 0168251], BR
[BR0000013591571], K [K000281098], P).
Explanation: Antephora elegans var. africana was published citing
four specimens, Buchholz 1875, Dinklage 464, Dewèvre 120 and
Schlechter 12508 which are to be regarded as syntypes and
comprise the original material. Since the author worked at Berlin
(B), the lectotype should preferably be located there. All except the
Schlechter specimen are not present at B and were presumably lost
during the 1943 fire. At B, there are two sheets of Schlechter 12508,
one of which has no spikelets left, the other with a few spikelets in an
envelope glued onto the sheet. The latter is selected as the lectotype,
with duplicates located at Meise Botanic Garden, Belgium (BR), Royal
Botanic Gardens, Kew (K) and Muséum national d’Histoire naturelle,
Paris (P). The barcodes are added to the specimens for which these
were available.

Neotype designation (represented by an illustration, also indicated

as iconotype, figure 17):
Dracaena sanderiana Sander ex Mast. (Masters 1892: 731). —
Neotype (designated here): Gard. Chron., ser. 3, 13: 445 (1893), f. 65
(iconotype).
Explanation: Dracaena sanderiana was first exhibited by the
horticulturalist Sander at the international exhibition in Earl’s Court
(1892), and published the same year by Masters with a description,
but without illustration. Original material of the plant exhibited
has not been traced and has probably not been conserved. One
year later, D. sanderiana was exhibited by Sander in Ghent and an
illustration was published in Gard. Chron., ser. 3, vol. 13 (1893).
This illustration most likely represents the same plant as originally
presented in 1892 and is hence chosen here as the neotype.

38 3.6 Author names, new taxon names, new combinations

The person publishing a new taxon name is the author of that name, and in
formal or official documentation is placed after the taxon name concerned.
The author name is often abbreviated, for which a standard abbreviation was
published by Brummitt & Powell (1992) and for which an
online database is now maintained by IPNI (at http://www.
ipni.org).

Sometimes, an author publishes a taxon name within the

publication of someone else (either as a chapter in a book,
or simply a part of an article accredited to them). In that
case, one may cite both authors using the connecting word
‘in’. For example, Verrucaria aethiobola Wahlenb. in Acharius,
Methodus, Suppl.: 17. 1803. The ICN regards the part after
Wahlenb. as a bibliographical reference and hence it is not
a part of the name.

In other situations, an author may validly publish a taxon

name but ascribe it to another person, for example where
the person suggested the name (on a herbarium label, or
even verbally) but failed to publish it. In that case, the name
of the latter is given but is followed by ‘ex’ and the name of  Figure 17. The neotype
the author who validly published it, e.g. Acalypha racemosa (iconotype) of Dracaena
Wall. ex Baill. Here, Baillon validly published the name Aca- sanderiana Sander ex
lypha racemosa that was already coined for that species by Mast. in Gard. Chron., ser.
Wallich. It is also accepted to omit the name of the first au- 3, 13: 445 (1893), f. 65.
thor and simply cite this species as Acalypha racemosa Baill.

The fact that an author intends to publish a new taxon name is often indi-
cated by adding the abbreviation spec. nov. or genus nov. or subsp. nov., etc.,
behind the name.

When an author moves a species from one genus to another, the epithet
is transferred to the new genus while the original author is placed between
brackets after it, followed by the transferring author e.g. Cenchrus purpureus
(Schumach.) Morrone. This species was originally named Pennisetum purpure-
um by Schumacher (1827) and transferred to the genus Cenchrus by Morrone
(2010). Note that the gender of the epithet has changed in accordance with
Latin grammar. The name Cenchrus purpureus (Schumach.) Morrone is called
a new combination (often abbreviated as comb. nov.) since it combines the
original (protologue) epithet with the name of another genus. The name that
provided the epithet for the new combination is called the basionym; in this
case Pennisetum purpureum Schumach.

The same happens when an author downgrades or upgrades a name to a

different taxonomic rank. For example, in Cenchrus polystachios (L.) Morrone
subsp. atrichus (Stapf & C.E.Hubb.) Morrone, the name Pennisetum atrichum
Stapf & C.E.Hubb., the basionym, was transferred to a subspecies of Cenchrus
polystachios by Morrone. The name Cenchrus polystachios (L.) Morrone subsp. 39
atrichus (Stapf & C.E.Hubb.) Morrone is not only a new combination (comb.
nov., since the basionym was moved to a different genus) but has also given
the taxon a new taxonomic status or rank, often indicated by adding stat.
nov. after the new name.
 3.7 Accepted names and synonyms: the priority rule

The science of taxonomy is dynamic where changes occur often as improve-

ments to the natural classification. This means that a publication may provide
new research data that underpin a novel view on the variation present within
a species, or on the delimitation of genera, families etc. It is important to
understand that such a view represents a new hypothesis, a new opinion,
supported by logical arguments. Through this process, the classification, or
the taxonomic framework, is ideally improving and evolving towards a stable
conclusion. However, some may weigh the supportive data for alternative
hypotheses in favour of a different classification. It is then difficult to say which
one is ‘correct’ as we will never be able to fully reconstruct the evolutionary
pathways.

When studying a certain group of taxa, an author may consider two or more
names to represent the same taxonomic unit. Following the type concept,
this basically means that this author is of the opinion that the type specimens
of both names belong to the same taxon. For example, Clayton & Renvoize
(1982) considered the following species names, in alphabetical order, to rep-
resent a single, variable species of grass:

Pennisetum angolense Rendle (Rendle 1899: 189).

Pennisetum giganteum A.Rich. (Richard 1850: 382).
Pennisetum macrourum Trin. (Trinius 1826: 64).
Pennisetum scaettae Robyns (Robyns 1934: 3).
Pennisetum stenorrhachis Stapf & C.E.Hubb. (Stapf & Hubbard 1933: 270).

This implies that all five names are synonyms, but the nomenclatural rules
stipulate that only a single name can be the accepted name; so, which one do
we choose? Here we should apply the priority rule (Principle III of the Code),
which tells us that the oldest synonym has priority over the others. In this
case, the correct and accepted name for this species is Pennisetum macrourum
Trin., since it was published in 1826.

The priority rule applies to all taxonomic levels. For example, in 2010, Mor-
rone published an article in which he merged the genus Pennisetum Rich.
(Richard in Persoon 1805: 72) with Cenchrus L. (Linnaeus 1753: 1049). The
priority rule shows the latter genus has priority over the first, and so the
genus in its new circumscription should be called Cenchrus.

When we follow the view of Morrone (2010), the accepted name for the
species Pennisetum macrourum Trin. becomes Cenchrus macrourus (Trin.) Mor-
rone. Note, that when another author does not agree with this hypothesis
and advocates to maintain the genus Pennisetum, there are two accepted
40 names for the same species, depending on the scientific point of view.

Further, it is important to know that the priority rule only applies to names
of the same taxonomic rank! In the previous example, if the name Pennisetum
polystachion (L.) Schult. var. africana Thunb. (Thunberg 1794: 101) would have
been a synonym of the five other names in Pennisetum mentioned above,
then it would have been the oldest name available. However, as it is a name at
the rank of variety, it has no priority over names at species level. When author
Xxx would want to upgrade that variety to the species level (as Pennisetum
africanum (Thunb.) Xxx) the publication date of that name would be the date
this new combination was published. It logically follows that when the taxon
Ixora aneimenodesma K.Schum. subsp. kizuensis De Block has no synonyms
and an author Xxx wants to raise it to species level, they have two options:
1) to publish the name Ixora kizuensis (De Block) Xxx, or 2) to publish a new
species name (e.g. Ixora congoensis Xxx), with the subspecies name as a syn-
onym. Option 2 is not considered ‘polite’ since it removes the original author
from the name. However, the name Ixora kizuensis may already exist for a dif-
ferent species. In that case, the necessary new combination is ‘occupied’ and
one has to opt for a new name, such as Ixora deblockiae Xxx to honour the
original author. The necessity to create such a new name for a taxon already
described before is often indicated by the addition nom. nov.

There are two exceptions to the priority rule. The first is that there are eight
family names and one subfamily name where one is allowed to choose be-
tween two alternatives (ICN Art. 18.5, 19.8). Such names are called nomina
alternativa (or nom. alt.). Below is a list of these allowed alternative family
and subfamily names. In a single publication it is advised to use the names of
only one of the columns.

Apiaceae Umbelliferae
Arecaceae Palmae
Asteraceae Compositae
Brassicaceae Cruciferae
Clusiaceae Guttiferae
Fabaceae Leguminosae
incl. subfam. Faboideae incl. subfam. Papilionoideae
Lamiaceae Labiatae
Poaceae Gramineae

Secondly, the strict application of the rules laid down in the ICN may lead to
‘unwanted’ changes and major instability of the nomenclature within a par-
ticular taxonomic group. In that case, one can make a proposal to conserve
or reject a specific name. In case of ambiguity related to the correct type
specimen, a similar proposal to conserve a specific type can me formulated. 41
Such proposals need to be published in the journal Taxon and are then vot-
ed on at the next International Botanical Congress. Conserved or rejected
names or types are generally followed by the indication ‘nom. cons.’, ‘nom.
rej.’ or ‘type cons.’.
 3.8 Hybrids

In the Code, names of hybrid taxa are dealt with in a separate chapter. They
can be recognized by the use of the multiplication sign × or by the addition of
the prefix “notho-” to the term denoting the rank of the taxon. A nothospe-
cies name, composed of a genus name (or a nothogenus name, see below)
and an epithet, indicates a hybrid between two individuals of different species.
A nothogenus name, a single word, is used when a hybrid has been formed
between individuals of species belonging to different genera. It is often com-
posed of parts of the names of the two genera involved.

For example, the hybrid between Oenothera biennis L. and Oenothera villosa
Thunb. can be either indicated by the hybrid formula Oenothera biennis L. ×
Oenothera villosa Thunb., or by the nothospecies Oenothera ×drawertii Renner
ex Rostański.

The nothogenus ×Festulolium Asch. & Graebn. groups individuals originated

from a hybridization event between species of the genera Festuca L. and Loli-
um L. The nothospecies ×Festulolium loliaceum (Huds.) P.Fourn. indicates the
hybrid between Festuca pratensis Huds. and Lolium perenne L., which can also
be indicated by the hybrid formula Festuca pratensis Huds. × Lolium perenne L.

3.9 Cultivated plants

Names of cultivated plants are not regulated by the ICN, but rather by the
International Code of Nomenclature for Cultivated Plants (ICNCP).

Cultivated forms can be indicated by only three categories, the Cultivar, the
Group or Cultivar group, and the grex. The latter is used solely in orchid
cultivation and indicates the combined hybrid offspring of any cross between
the same two entities (taxa or cultivars). A cultivar, abbreviated as cv., is a very
specific form derived from any type of selection process and may even have
been taken directly from the wild. It is a non-Latin name added after the name
of the taxon from which it was derived e.g. Solanum tuberosum L. cv. Gogu
valley, also written as Solanum tuberosum ‘Gogu valley’. When it is unclear to
which species a cultivar belongs, the cultivar name can follow directly after
the genus e.g. Rosa cv. Penelope. A new cultivar name can be formally regis-
tered by an International Cultivar Registration Authority that needs to be ap-
proved by the ISHS Commission for Nomenclature and Cultivar Registration.
Each Authority is assigned a specific taxonomic group. After the Authority
has formally approved the registration of a new cultivar name, the person
who provided the information ‘owns’ the rights to this name. They can then
42 market both the name and the plants, quite similar to a patent. A Cultivar
group, abbreviated as cv. gr., comprises a number of cultivars having a distinct
characteristic. One could, for example, create a Cultivar group for all yellow
roses. Here it becomes clear that names of cultivated plants are not part of a
natural classification, since they need not indicate or reflect common ancestry.
In literature on cultivated plants, one may regularly come across ‘variety’ or
‘form’. Note that these terms should actually not have been used for cultivat-
ed plants, as they erroneously refer to the ICN which does not deal with cul-
tivated plants. When possible, such uses of ‘variety’ or ‘form’ should be treated
as informal descriptions of the variation observed without the intention to
create a new taxon name under the ICN.

International Code of Nomenclature for algae,

fungi, and plants

• http://www.iapt-taxon.org/nomen/main.php?

Scientific names and types

• International Plant Name Index: https://www.ipni.org

• Tropicos: http://www.tropicos.org
• World Flora Online: http://www.worldfloraonline.org
• African Plant Database: http://www.ville-ge.ch/musinfo/bd/cjb/africa/
index.php
43
• Linnaean Typification Project: http://www.nhm.ac.uk/our-science/data/
linnaean-typification
• Global Plants: https://plants.jstor.org
 4.

The art
of identification

44
In the preceding chapters, we mentioned that the taxonomic science aims
to organize the immense diversity of living organisms on earth into discrete
units. As such, it provides the essential tools for scientific communication in
the form of names and a classification. To be able to perform biological re-
search, protect nature, use plants for medicinal purposes, etc., it is crucial to
have access to the wealth of information accumulated over several centuries.
Online resources are growing exponentially. However, before one can tap
into the information available on certain species or genera, for example, one
has to know the name of the particular taxon concerned. As specialists who
can identify living organisms by heart are rare, especially in the tropics where
diversity is high, taxonomists have developed tools to reliably identify their
material.

4.1 Identification keys

An identification key is a practical tool used by both specialists and non-spe-

cialists to identify plants, fungi or animals, to the level of family, tribe, genus,
species or other. It is often the most extensively used part of a taxonomic
publication, and hence deserves the utmost attention of the researcher who
creates it!

In order to use an identification key, one generally has to have at least a basic
understanding of plant or fungal morphology and terminology. Having a good
glossary may be useful. Various good and extensive botanical glossaries exist
(see the text box at the end of this chapter).

How to use a key?

An identification key is generally a kind of question-answer ‘game’ where the

user is asked to (carefully!) observe specific characters and report on their
state. An identification key may for example use the character ‘flower colour’,
and the user can choose between the states “yellow” or “white”. When yel-
low, continue to question number 2, when white go to question number 10.
The first part of the identification key will then look as follows:

1. - Flowers yellow ...................................................................................... 2

- Flowers white ..................................................................................... 10
2. - …
- …

In the example above, each question is called a couplet of the key, having
two leads. It is obvious that the two options need to be mutually exclusive,
not showing overlap. After having correctly answered a certain number of
questions, the user will end up with the name of the plant (or fungus, or 45
animal) at hand.

Generally, the user is given the choice between two options. In such a case,
the key is dichotomous. Some keys allow the choice between three or even
 more options (in the example above, one could add ‘Flowers blue’, ‘Flowers
red’, for example to arrive at four leads. These are called polytomous keys.
In general, this structure is deemed less practical and more prone to identifi-
cation errors. One can easily avoid such choices by combining several states
into one lead, such as:

1. - Flowers yellow, blue or red ............................................................ 2

- Flowers white ..................................................................................... 10
2. - Flowers yellow ...................................................................................... 3
- Flowers blue or red ............................................................................ 6
3. - Twigs spiny ................................................................... Rosa banksiae
- Twigs spineless........................................................................................ 4
…
6(2) - Stamens …
- …
…
10(1) - Leaves ….
- …

Note that question #3 provides the name of the plant, and that questions
6 and 10 show which previous question pointed to it. The latter is simply
assisting the users in keeping track of where they came from, and is generally
added only when one has arrived after having made a comparatively large
‘leap’ in the key.

Basically, there are two forms of dichotomous keys. The form shown above
where both leads directly follow each other is called a bracketed key. The
second form is called an indented key and separates the two leads in space.
Here is an example of an indented key (adjusted from a key to the species of
Solanum in Africa, Vorontsova & Knapp 2016):

1. Flowers with stamens of different lengths

2. Leaves orbicular to reniform, 1.2-2.5 cm long, wider than long; petioles
longer than leaves. Rare in northeastern Somalia .......... S. cymbalariifolium
2. Leaves ovate to elliptic or lanceolate, 2-14 cm long, longer than wide;
petioles shorter than leaves. Arid eastern and northeastern Africa.
3. Stem prickles dense, acicular, less than 0.5 mm wide at base, pale
yellow; fruit fully concealed by the accrescent calyx ........... S. coagulans
3. Stem prickles absent or sparse, if present wider than 1 mm at base,
yellow to orange or brown; fruit at least partly exposed ...........................
.................................................................................................................. S. melastomoides

1. Flowers with all stamens equal in length. Widespread.

46
4. Flower one per inflorescence, peduncle and rachis absent; corolla pen-
tagonal, lobed for ¼-⅓ of the way to the base, 0.9-1.3 cm in diameter;
Southern Africa ................................................................................................... S. supinum
4. Flower usually more than one per inflorescence, peduncle and/or rachis
present in at least some inflorescences; corolla usually stellate, lobed for
 more than ⅓ of the way to the base, or if lobed for ¼-⅓ of the way
to the base then corolla of long-styled flowers broader than 1.3 cm in
diameter; widespread. ................................................................................. S. tuberosum

When the first lead of couplet 1 corresponds with the plant to be identified,
one should carry on to the next question formulated immediately below it
(number 2). However, when the second lead of couplet 1 is correct, one con-
tinues with the question below, which is number 4. As can be seen, there are
no numbers to the right-hand side of the key pointing to the next question.
The advantage of an indented key is that the user may more easily derive the
way the species are grouped from the structure of the key. A disadvantage
is that one has to search for the second lead of the question/couplet con-
cerned. This can be quite far down in larger groups, and in longer keys there
will be a lot of unused space on the left side of the page, resulting in a key
that needs more printed pages.

Also, note that geographical information may be used in the key. Despite not
being a morphological character, it may be deemed helpful. Similarly, ecologi-
cal or phenological information (flowering/fruiting period) may be added. It is
however advised that these types of characters be used only as supplemen-
tary data to morphological features.

How to make a key?

Start by choosing which type of key to construct (see above). Then, think of
several clear subgroups that can be recognized within the group concerned.
Next, select those groups that can be defined by character states that are
well distinct and can be easily observed with the naked eye or when using a
10× hand lens. If the first question of a key is about pollen grains, for example,
a fair number of users will be immediately stuck and unable to continue. It is
also important that any character mentioned in one lead of a couplet, should
also be present in the other lead(s)! A couplet like the one below is not to
be recommended:

1. - Flowers yellow; leaves longer than 10 cm ............................. 2

- Flowers white ..................................................................................... 10

A user having a plant that carries white flowers but also has leaves longer
than 10 cm will become uncertain as to what to choose. This brings us to
another practical issue. When making a key, one should always try to imagine
what a user may have on hand! This is often a single plant, so the key needs
to provide exact information. If a couplet states “Flowers big” against “Flow-
ers small”, this is a relative concept and the user is unable to judge whether
flowers of 1 cm in diameter may be considered ‘big’ or ‘small’.
47
A single taxon may key out more than once. That may occur when a specific
taxon is variable for a certain character. For example, a species may have
white as well as occasionally yellow flowers, while most species are constant
in that respect.
  Figure 18. Example of Creating a key for a larger group of species
a species/character data (or other taxa) is often greatly facilitated by
matrix (with fictional taxa the preparation of a taxon/character data ma-
Aus a, Bus x, etc.). trix (see also figure 18). It often helps to get a
better overview of the distribution of charac-
ters and their correlation. (See also the next
paragraph.)

Finally, some advice on key construction:

1. Be practical! Use 100% clear language. Try to avoid characters that

need a lot of explanation or are otherwise difficult to understand or even
difficult to observe.

2. In a fully dichotomous key, when all taxa key out only once, the number
of couplets will always be equal to the number of taxa minus 1. Hence,
one cannot influence the number of couplets. However, one can influ-
ence the number of questions to be answered before a species is keyed
out. The best strategy is to strive for questions/couplets that divide the
remaining group of taxa into more or less equal parts.

3. If the species in a group generally do not have flowers and fruits at the
same time, it may be wise to present two different keys, one for flowering
material and one for fruiting material.

4.2 Multi-entry keys

The keys discussed above, even when prepared with the utmost care, have
a serious flaw. The user cannot choose the sequence in which characters are
observed. It could be that having red fruits would already greatly reduce the
number of potential species, but that fruit colour is only asked in question
number 5. In other words, there should be easier ways to identify a plant!

48 When using a taxon/character data matrix for a selected group of species

(see figure 18 for an example), the user could randomly choose a character
from the list and fill out the character state observed in the specimen to be
identified. Then, this process can be repeated until the specific combination
of character states corresponds to a single species.
Before the computer era, taxonomists already tried various multi-entry keys
based on similar matrix systems. One would work with a large number of
numbered cards, where each card would either represent a character state,
or a taxon. Those interested in such punch card keys could have a look at
figure 19.

Nowadays, such a taxon/character data matrix with its resulting multi-entry

key can be processed by several user-friendly software packages e.g. Xper3
(http://www.xper3.fr/), DELTA-IntKey (https://www.delta-intkey.com) and Lin-
naeus NG (http://linnaeus.naturalis.nl/).The Xper3 and DELTA-IntKey packag-
es even allow you to create a dichotomous key from your data matrix, which
can be used in publications. Some have statistical software included that will
advise you on the best next character(s) to use in order to render the iden-
tification process most efficient.

A different kind of multi-entry key is the so-called diagnostic key (also called
a synoptic key). This comprises a list of diagnostic or spot characters (typical,
remarkable) occurring within a certain group of taxa. Each character is then

 Figure 19. Example of

a punch card showing the
characters (open holes
along the edge) for a
species of Eucalyptus.

49
 followed by a list of all taxa within the group that possess that character. Be-
low is an example of part of a diagnostic key to taxa of Rubiaceae in Central
Africa. Note that the taxa indicated are mostly genera, but also tribes and
even species.

Example of a (part of a) diagnostic key:

- Leaf

blade linear: Amphiasma, Anthospermum usambarense, Cordylostigma,

Galium, Knoxia, Kohautia, Manostachya, Oldenlandia, Spermacoce
blade heart-shaped or kidney-shaped: Geophila, Hymenocoleus,
Pentanisia renifolia, Rubia

- Flower

unisexual: Anthospermum
heterostylous: Colletoecema, Craterispermum, Gaertnera, Knoxieae,
Lasianthus, Morinda, Mussaendeae, Pauridiantha, Psychotrieae,
Sabicea, Schizocolea, Spermacoceae, Tricalysia
4-merous: Anthospermum, Corynanthe, Eumachia, Galium, Heinsia, Ixora,
Keetia, Knoxia, Lasianthus, Nauclea, Otiophora, Paraknoxia, Pavetta,
Polysphaeria, Pouchetia, Psychotria, Rutidea, Spermacoceae, Tricalysia
pleiomerous (with more elements than usual): Coffeeae, Gardenia,
Rothmannia octomera, Schumanniophyton
calyx tube long (> 1 cm): Adenorandia, Gardenia, Rothmannia,
Schumanniophyton hirsutum
calyx tube with a lateral slit: Calycosiphonia, Gardenia, Polysphaeria,
Rothmannia, Sericanthe, Tricalysia
calyx asymmetrical, with highly unequal lobes, or only 1 lateral lobe:
Knoxieae

4.3 DNA barcoding

The most modern method of identification is the use of a unique DNA

profile for individual taxa. The idea is that when this unique DNA sequence,
or more popularly called the ‘DNA barcode’ is known for all species and we
have the sequence of an unknown organism, we can match it to our data
50 bank to determine the name of the species. Simple in theory; a lot more
complicated in practice! Firstly, we need a data bank for all c. 400 000 plant
species, c. 10 000 000 animal species, c. 5 000 000 fungi species, etc., which is
far from evident. Moreover, as sequencing (obtaining the base pair sequence)
the entire DNA content of an organism is still very time consuming and
expensive, we need to find a piece of the genome, a specific part of the
DNA, which provides sufficient variation at the desired level (often species).
In many animal groups, the cytochrome c oxidase I (CO1 or COX1) gene
is used comprising c. 1500 base pairs. For plants, however it is a lot more
difficult. The combination of two chloroplast genes, rbcL and matK, have been
proposed as a suitable candidate. Adding the nuclear non-coding internal
transcribed spacer 2 (ITS2) region was proposed to improve ‘resolution’. For
fungi, the spacer 1 (ITS1) region would be more suitable. Other, possibly bet-
ter, suggestions are still being debated. Furthermore, it appears that it is not
uncommon that these standard barcoding markers show variation within a
species. This means that a single sample is insufficient to represent a species
in the DNA barcode data bank, as the within-species variation needs to be
mapped first before a reliable identification can be made. One cannot, for
example, simply state that: “Since my sample differs in 2 sequence positions
from the other, it represents a different species”, before the variation in the
sequence data has been mapped for both species. As a consequence, for each
species several samples (minimum 10, preferably more) are needed to build
a reliable barcode data bank. Furthermore, each specimen sampled needs
to be vouchered; an organism needs to be preserved in order to be able to
verify its identity if doubts arise. The data bank will also need to be regularly
updated to incorporate changes in taxonomic concepts.

The DNA barcode database needs to be based on a sound and stable tax-
onomic framework of genera and species. Even for a comparatively well-
known group like plants, this framework still has many weak spots. In their
turn, the results of DNA barcoding efforts may well assist in making better
taxonomic decisions, thereby strengthening the framework.

Despite these challenges, major efforts are done to create a world-wide

DNA barcoding database. The activities are coordinated by the International
Barcode of Life Consortium (iBOL) together with many regional centres.
At present, obtaining a DNA sequence from an organism often takes sev-
eral days of lab-work. One needs to be patient when using this method of
identification. New processes are rapidly evolving, with the development of
ever more sophisticated nano-techniques that create the possibility to put
together portable minilabs that can be used in the field.

4.4 Identification of herbarium specimens

While identifying living plants in the field can be difficult, trying to correctly
identify dried and flattened herbarium specimens is often challenging. Not all
characters needed may be directly visible, even with the help of a good 10×
hand lens or a stereo microscope (see also paragraph 5.B). The simple ques-
tion whether the flowers are white or yellow may remain unanswered when
the collector did not record this information in the field. Similarly, one may 51
wonder whether this twig with few leaves and nice flowers or fruits comes
from a big tree, a liana, a shrub or even a perennial herb? The 3-dimension-
al shape (notably of flowers and fruits) could be important, but impossible
to reconstruct, as is information about underground tubers, rhizomes, smell,
 taste, etc.These kinds of information should be noted down in the field by the
collector so that the information can be transferred to the label accompany-
ing the specimen. Various publications (Fish 1999, Victor et al. 2004, Bridson
& Forman 2010) offer good advice on how to collect plants and correctly
prepare valuable dried plant specimens.

When a herbarium specimen has been identified, whether this is at the level
of family, genus, species or other, the taxon name is written on a small piece
of paper, called an identification or determination slip. This also includes the
name of the researcher (and affiliation if possible) and the date. It is glued
onto the herbarium sheet (when not gummed, use special glue provided by
the herbarium curator), preferably somewhere in the lower right corner and
always above any previous identifications slips. Make sure only a small part of
the slip is glued, so that the remainder can be flipped back to examine any
material or written text it may cover. Some herbaria will only allow the use of
needles to attach labels and slips to the sheet.

Any uncertainty about the correctness of the identification can be added

as well. Preferably use the abbreviations cf. or aff. The first is short for “con-
fer” meaning “compare with” and is used when a specimen is very close to
something else, and may even be the same. The second is short for “affinis”
meaning “similar to” and is used when a specimen is similar something else,
but is probably different.

Glossaries

• Beentje H. (2015) The Kew plant glossary. 2nd edition. Richmond, Royal
Botanic Gardens, Kew. EAN: 9781842466049
• Josserand M. (1983). La description des champignons supérieurs. 2e éd.
Paris, Lechevalier.
• Jouy A., Foucault B. de (2016) Dictionnaire illustré de botanique. Mèze,
Biotope.
• Missouri Botanical Garden Glossary:
http://www.mobot.org/mobot/glossary
• Wikipedia: https://en.wikipedia.org/wiki/Glossary_of_botanical_terms
[in French: https://fr.wikipedia.org/wiki/Glossaire_de_botanique]

DNA barcoding

• IBOL (International Barcode Of Life): https://ibol.org

52
• Hebert P.D.N., Cywinska A., Ball S.L., deWaard J.R. (2003) Biological
identifications through DNA barcodes. Proceedings of the Royal
Society of London. Series B: Biological Sciences. 270 (1512): 313–321.
doi:10.1098/rspb.2002.2218.
 5.

The art of preparing a

taxonomic revision

53
54
Here, we focus on a taxonomic revision largely based on herbarium speci-
mens. When studying the taxonomy of a plant or fungal group for which field
observations are not easily obtained, e.g. tropical species, or simply those
from remote places, herbarium specimens are often the only source of infor-
mation available. Adding field observations to a revision based on herbarium
specimens is a big advantage, but it is not a necessity.

Let’s assume that it has been decided to prepare a taxonomic revision for
a specific group. This is generally because specialists have indicated that the
taxonomic framework for a particular group is considered ‘weak’. This could
be the result of the available identification keys being of poor quality, of un-
clear boundaries between taxa/species, or of doubts about the correctness
of the names applied. It is not uncommon that part of the “weakness” may
be due to the presence of undescribed species. Each year, well over 2000
new species of vascular plants are described, as well as some 75 to 100 new
genera. This trend has not decreased over the past 15 years (numbers have
even risen slightly over the past four years), indicating there is still plenty to
be discovered!

A taxonomic revision is often focused on a particular genus, and in the text

and examples below we assume it is. When a genus has many species and a
large distribution, the study is not seldom restricted to a country, a phytoge-
ographic region, or a continent.

Referring to the depth, thoroughness and practicality of the revision, one can
identify four categories:

- Synoptic revision, or Synopsis: a brief update of the taxonomy of a group

(generally provides an identification key, an overview of all accepted
species and their synonyms, sometimes including brief morphologic de-
scriptions and some information on distribution);

- Taxonomic revision: a standard update of the taxonomy of a group (with

identification key, full synonymy, type information, full morphological de-
scriptions, distributional data, often also citing the specimens used);

- Monograph, or Monographic revision: a very thorough, all-inclusive update

of the taxonomy of a group (as detailed as a Taxonomic revision, but
often extended with the results of additional anatomical, molecular, eco-
logical or ethnobotanical studies).

- Flora treatment: by definition a regional account of a group, generally a

critical revision or compilation of existing (published) information. Here,
relatively simple problems are addressed, leaving more complicated mat-
ters to more detailed future studies. A Flora’s primary aim is to provide
tools (identification keys, descriptions, illustrations, etc.) to users who
want to identify plants.
55
For all four categories, the scientific process can be divided into seven phases
(A–G, see below) which will be dealt with in more detail in the following par-
agraphs. For the purpose of convenience, we will use the term ‘taxonomic re-
vision’ in the broader sense, comprising all four categories mentioned above.
 The seven phases of a taxonomic revision:

A. Taxon names and literature study

B. Herbarium observations
C. Data basing
D. Geographical and ecological observations
E. Taxonomic and nomenclatural decisions
F. Preparing taxon treatments, descriptions, illustrations and keys
G. Producing the manuscript and publishing

A. Taxon names and literature study

All scientific studies start with gathering information and data. For a taxonom-
ic revision, it is necessary to gather all the protologues (original publications)
for the names concerned.This is a crucial step, especially in phase E when type
specimens need to be identified, selected and assigned. The service provided
by IPNI (International Plant Name Index; http://www.ipni.org) for vascular
plants and Index Fungorum (http://www.indexfungorum.org) for fungi can
generally provide a list of all names within the relevant genus, although some
further filtering will be needed for regional studies. Note that IPNI did not
collect data on infraspecific taxa until 1971! Some may have been entered
now, but in general these can only be found through extensive searching
(internet and libraries).

From the protologue information, one can now track down the relevant pub-
lications by:

- consulting a good, specialized library;

- using web-based services uncovering old taxonomic literature, such as

Botanicus (http://www.botanicus.org), or the Biodiversity Heritage Li-
brary (https://www.biodiversitylibrary.org);

- using other web-based resources that link names to their protologues;

one of the best is Tropicos (http://www.tropicos.org) from Missouri Bo-
tanical Garden, but IPNI (see above) also provides this service for many
names. Certain families or groups have active communities that maintain
specialist websites (e.g. http://solanaceaesource.org, http://www.palm-
web.org or http://caryophyllales.org).

Scan or download the relevant pages and file these in an easily traceable sys-
tem. Also, be sure to note the full reference (see below)! This will be needed
when publishing the study.
56
Next, gather all relevant books and articles that deal with the systematics
of the chosen genus. It is also useful to study relevant papers dealing with
phylogenetics, biogeography, ecology, etc., as this will give greater insight into
the genus and its relatives, especially the evolutionary importance of some of
the characters. Use available internet search engines as well as portals of spe-
cialized libraries using the name of your taxon as keyword, along with others
such as ‘taxonomy’, ‘revision’, ‘systematics’ etc. Start with the most recent pub-
lications, and study the publications cited in their reference lists. Scroll through
their paragraph on the taxonomic and systematic history of the genus when
available. Also, consult Floras relevant to the study area.

Studying relevant literature should give you a fairly good idea of the position
of the genus within the family, which genera are closely related, and which
characters are deemed distinctive and informative for species delimitation.
Going through identification keys to see which characters have been used to
distinguish the species will probably prove to be extremely worthwhile. Nev-
ertheless, it is always important to develop a personal view on the variation
of the group studied.

A fair amount of unknown terminology will be encountered when reading

the publications, especially the morphological descriptions. A good glossary,
explaining these terms, is crucial in this phase. Various good and extensive
botanical glossaries exist, see the text box at the end of the previous chapter.

Do not forget to record the full reference when taking notes from publi-
cations! It can be very frustrating and time consuming when one cannot
remember where one read something.

Here are examples of references for publication types that you are most likely
to encounter:

1. Article in a journal:
Soreng R.J., Peterson P.M., Davidse G., Zuloaga F.O., Judziewicz E.J., Filgue-
iras T.S., Davis J.I., Morrone O. (2015) A worldwide phylogenetic classifi-
cation of the Poaceae (Gramineae). Journal of Systematics and Evolution
53(2): 117--137. http://dx.doi.org/10.1111/jse.12150

2. Book:
Patil J.V. (2016) Millets and Sorghum: Biology and Genetic Improvement.
Chichester, John Wiley & Sons Ltd. 504 pp.

3. Chapter in a book or series:

Clayton W.D. (1989) Gramineae. XXIV. Paniceae. In: Launert E., Pope G.V.
(eds) Flora Zambesiaca 10(3): 1--192. London, Flora Zambesiaca Manag-
ing Committee.

Finally, set up a documentation system or data base, where you indicate for
each taxonomic name, where the protologue can be found and what its type
specimen is. Other authors may have already indicated which specimen is the
type, however this should always be checked. The protologue will generally
provide the necessary information concerning the type(s), but certainly not 57
in all cases. Searching for type specimens can be a time-consuming activity!
This is particularly true for older literature, as it was not obliged to indi-
cate a type. The majority of major Herbaria have scanned and digitized their
type specimens. These images are available at the JSTOR Global Plants portal:
 https://plants.jstor.org, and generally also through the web-portals of the in-
dividual institutes. Note that this facility only shows known type material for
the various Herbaria. A fair number of types still have to be identified as such
by taxonomists. In the course of a revision, it is common to encounter type
specimens that had previously gone unnoticed.

B. Herbarium observations

It should be noted that herbarium specimens are valuable and often irre-
placeable scientific objects that are brittle and easily damaged! They must be
handled with the utmost care. Always ask the appropriate Herbarium Cura-
tor about the specific handling instructions.

Practicalities on how to observe plant morphological

characters from herbarium material

Plant morphological observations can be performed with a simple hand

lens (10×) or with a binocular usually magnifying up to 20(–50)×. The
highly brittle material can be made flexible again simply by putting it in
boiling water for a short period. Depending on the toughness of the
material, it can take 20 seconds to 3 minutes. Always ask the Curator
if you are allowed to break off small parts from the herbarium sheet
for this purpose. For very tough material add a drop of dishwashing
liquid to the water to soften the tissue. An electric laboratory heater, as
shown in figure 20, is reasonably safe to use, but a normal, good quality
electric kitchen heater and a small steel pot will do as well. You will
need a few needles, forceps and petri dishes to handle the material e.g.
for dissection under a binocular microscope. Thus allowing the often
hidden inside of flowers to be studied, ovaries cut open, and even leaf
or wood anatomical features revealed.

After having studied material taken from a herbarium specimen, all

elements, also the dissected ones, must be returned to the specimen.
Boiled material should be dried again (use absorbent paper to dry it
quickly). All parts are then placed in a small bag, which can be glued or
pinned to the herbarium sheet of the specimen.

When material, like pollen or a piece of leaf for DNA extraction, is not
returned to the sheet (called destructive sampling), the researcher is
supposed to add a label to the specimen stating what was removed,
for which purpose, by whom and when. Again, ask permission from the
58 Curator before you do!

For your own observations, it is advisable to set up a table where you

note all observations and measurements for each specimen (an Excel
 table is fine, see below). It is better not to summarize them for a taxon,
as specimens may shift between taxa at this stage of your work.

The preparation of fungal specimens for microscope observation is

quite specific and detailed in Eyi et al. (2011).

One will need to develop a strategy on how to obtain or consult the majority
of the herbarium specimens available elsewhere. Phase A (taxonomic names
and literature) has provided you with a fairly good idea of the distribution
of the genus, and where most of the species occur. One might consult col-
leagues to find out which Herbaria contain the vast majority of the relevant
material. One can either visit these institutes or have the material to be on
loan to your home institute. Herbaria addresses and contact persons can
be found on Index Herbariorum (http://sweetgum.nybg.org/science/ih). Note
that it may take several months for a loan shipment to arrive. Herbaria will
generally not send more than several hundred specimens as a loan, and they
may have restrictions related to the countries requesting material. An option
will then be to visit the institution where the material is housed.

When starting your study, you will need a fair amount of herbarium spec-
imens, so ask for loans as soon as possible. In some cases (i.e. when one
already has a fair knowledge of the group concerned), loans can be requested
months before the actual start of the revision. Herbaria that house a lot of
your material are best visited physically, but this is often expensive. It will gen-
erally be far more efficient to visit them once you already have a good knowl-
edge of the species/taxa within your group. So, a good work plan is essential.

Some herbaria have their collections available on-line (see textbox at the end
of this chapter). These services often provide some label data and high-res-
olution scans of available material. These on-line services are extremely
helpful, but experience
shows that a fair num-
ber of sheets need to be
loaned to perform more
detailed observations on
the actual specimens.

 Figure 20.
Example of a simple
laboratory device to boil
water in a small cup.

59
 C. Data basing

For taxonomic revisions that concern several hundred specimens or more, it

is worthwhile to place the data relating to the specimens in a data base. This
can be a simple table in a spreadsheet (e.g. Excel) or a multiple table relational
data base (e.g. Access), but many institutes will have their own data base sys-
tem. These systems may often have many fields one does not need, so con-
sider which fields you need before starting. A minimal set is provided below:

- Barcode
- Main collector (preferably surname and initials in separate fields)
- Additional collector(s)
- Prefix (some collectors add a code or number before the collection
number which refers to a research mission, the collecting year or desig-
nates a project by its acronym)
- Collection number
- Suffix (any code given after the collection number, see prefix)
- Collecting date
- Country
- Locality
- Latitude
- Longitude
- Habitat
- Altitude
- Uses
- Vernacular names
- Family
- Genus
- Species
- Author(s)
- Infraspecific level (subspecies, variety, forma)
- Infraspecific name
- Infraspecific author(s)
- Identified by
- Identification date
- Herbarium code
- Type of
- Notes

Often, Herbaria that have their specimens digitized will be willing to send
their data in a digital format, which can be uploaded into your own data base
after some necessary adjustments.

60 D. Geographical and ecological observations

A species distribution area can vary from small to large, this is generally re-
lated to the species having a narrow or broad ecological tolerance. A species
restricted to a defined region is said to be endemic to that region. A species
can be endemic to a mountain, national park, province, country, continent, etc.
(and all species are endemic to the planet Earth!). So, simply stating that a
species is endemic without reference to the region is meaningless.

D.1 Mapping

As indicated in Chapter 2 (the paragraph on speciation), information about

the geographical distribution and/or ecological preferences may be helpful in
making taxonomic decisions. Investigating the distribution of closely related
taxa by plotting their recorded localities on a map is a useful exercise.

To be able to plot specimens on a map, one requires the geographical coor-

dinates (latitude, longitude) of the collecting locality. If this is not present on
the label, you will need to obtain the coordinates using web-based services
(see textbox at the end of this chapter), topographical maps (often historical
ones), or even track down field expedition reports. This process is called
georeferencing. Most contemporary collectors use a GPS (Global Positioning
System) in the field, while others may have obtained the lat./long. information
from a topographic map. It can be important to add the precision of the data.
When a collector indicates “15 km West of Nairobi”, and knowing that city
has a diameter of some 10 km, one may wonder if one should count from
the city edge, or from the city centre, and whether one should measure as the
crow flies, or along the main road leaving Nairobi in a western direction? Not
to mention the fact that some 50 years ago Nairobi occupied a much smaller
surface area than today. A coordinate can be accurate at several meters to
several kilometres or more.

Note that there are various coordinate systems! These are related to different
‘projections’, or ways in which the globe has been transposed onto a map.
There are also different ways of recording coordinates. The most commonly
used coordinate format used by taxonomists is Degrees, Minutes and Seconds,
but in some regions the UTM (Universal Transverse Mercator) coordinate
system is preferred. Also note that some may use a normal DMS (Degrees,
Minutes, Seconds) format, e.g. 15°12’55”N 30°21’32”E, while others prefer
DD (Decimal Degrees), e.g. 1.247°N 25.873°E, or DM (Decimal Minutes), e.g.
11°34.75’N 25°21.30’E. A number of online tools (for example http://www.
synnatschke.de/geo-tools/coordinate-converter.php) allow to easily convert
coordinates from the different systems.

One should always be careful when taking coordinates directly from

herbarium labels. These are frequently poorly recorded, with no information
regarding the projection system used and with wrong coordinate formats,
e.g. minutes and seconds values greater than 60. Also, a common error is the
inversion of the direction indicators North/South or East/West. These values
can be checked against the collecting locality description.

Google Earth is also a useful tool to find the collecting locality, but older col- 61
lections may have place names no longer in use. For some regions, a published
index of plant collecting localities is available, and some websites provide
historical maps (see textbox at the end of this chapter). The libraries of many
natural history institutes will often have a good collection of historical maps,
 collecting registers (collecting notebooks of the collector) or Gazetteers (a
book, usually per country, with all place names, including rivers, mountains,
etc., with their lat./long. information). On-line gazetteers, like GeoNames (see
textbox at the end of this chapter), are also a good way to search for loca-
tions as they often have historic names of places as well as allowing “fuzzy”
search options. Furthermore, a collector will often have collected several
specimens at the same locality and so the specific locality may already have
been georeferenced by someone else. Check web-portals providing such
data on-line, notably from an institute where you know duplicates from a par-
ticular collector were deposited. Some specimen data bases offer the option
to create an itinerary for a specific number range of a collector which may
assist one in finding the right information.

Plotting your specimen data on a map can be done using Google Earth, but
when you want to prepare a high-quality distribution map for publication, you
should look for other software (e.g. DivaGis, QGIS or ArcView).

D.2 Ecological observations

Herbarium specimens will often contain ecological and altitudinal information

on the label. However, these are usually fairly coarse habitat descriptions. This
data can, at best, be supportive of a taxonomic decision. A person skilled
in Ecological Niche Modelling may be able to calculate the environmental
niche envelope of a species based on its distribution, and even establish if it
is significantly different from that of another species. Such analyses may pro-
vide additional support for a taxonomic decision, but are complex and often
require additional skills and expertise.

E. Taxonomic and nomenclatural decisions

After having performed all morphological, geographical, ecological and possi-

bly other observations, one can group the herbarium specimens into a num-
ber of sufficiently homogeneous sets. Each pile of material represents a taxon
that differs from the other piles based on your observations. Now, each pile
needs to be assigned to a species, subspecies, variety or form using the cri-
teria described in Chapter 2. The resulting list of taxa is your hypothesis for
the correct taxonomic framework of the group studied. When this differs
from previously published hypotheses, one can and should discuss these dif-
ferences.

It is often stated that this way of taxon delimitation applies the morphological
species concept (see Chapter 2). However, when we think about what mod-
ern taxonomists actually do, we may conclude that they are in fact trying to
interpret morphological, geographical and ecological data, in terms of a bio-
62 logical species concept based on non-interbreeding populations. When a ‘pile’
of specimens has several distinct morphological characters, one may assume
they have arisen from a distinct genetic basis. This can only remain distinct if
there is no interbreeding. The same applies to geographical and ecological
information. A taxonomist will generally interpret such data within the light of
potential interbreeding. In conclusion, one might state that herbarium-based
plant and fungal taxonomy (but also natural history specimen-based animal
taxonomy) attempts to apply a biological species concept through interpreta-
tion of morphological and other patterns observed. Incorporating the results
of molecular studies to determine whether the various species are reproduc-
tively isolated is encouraged when available, but that is beyond the scope of
this chapter.

It is quite common to find specimens with intermediate characters between

taxa. Such intermediates may indicate the existence of gene flow between
the morphological groups. In plants, hybrids are not uncommon and therefore
do not necessarily disrupt the proposed taxonomic framework. However,
when such intermediates are more frequent, this will cast doubt on the cor-
rectness of the taxonomic hypothesis and may lead to a re-evaluation. When
interpreting patterns of morphological variation, one should keep in mind
that we observe only a small time-slice of a larger evolutionary process.

After one has finalized the taxonomic framework, the next step is to establish
the correct scientific name for each pile. This is where nomenclatorial deci-
sions have to be made.

Locate all type specimens of all names, species as well as infraspecific taxa,
and verify in which pile they are (even those that may not be physically
present!). The type specimens in each pile represent the potential names to
be considered for that particular taxon. Following the rules of the ICN, as
described in Chapter 3, should then lead to identifying the accepted name
and its synonyms. Any pile that has no type specimen is a new taxon that will
need to be formally described.

F. Preparing taxon treatments, descriptions, illustrations and keys

In a taxonomic revision, the formal treatment of a taxon starts with the

nomenclatural part. Firstly, we have the accepted name, followed by its homo-
typic synonyms in chronological order starting with the oldest name, the ba-
sionym, and the data related to the type specimen. Secondly, the heterotypic
synonyms follow, if there are any, with their subsequent homotypic synonyms,
in chronological order with their associated type specimen data (see example
in the text box).

When creating a new taxon, one must meet all requirements of the ICN in
order to validly publish the name (see Chapter 3).

The nomenclatural part is generally followed by the morphological descrip-

tion. Any description of a taxon should be clear, precise and sufficiently de-
tailed. It contains the data on which your taxonomic hypotheses are built. 63
The structure of the description is generally uniform for all taxa dealt with in
order to facilitate comparison. This means that any character mentioned in
one taxon description should also be present in the others. Also, check that all
characters used in the identification key (see below) are incorporated as well.
 Example of the nomenclatural part of a taxon
treatment:

Urochloa dictyoneura (Fig. & De Not.) Veldkamp (Veldkamp 1996:

418). -- Panicum dictyoneurum Fig. & De Not. (Figari & De Notaris
1854: 329). -- Brachiaria dictyoneura (Fig. & De Not.) Stapf (Stapf
1919: 512). -- Type: Soudan, Kordofan, Fazogl, Figari s.n. (holo-: FI).

Panicum golae Chiov. (Chiovenda 1914: 43). -- Type: DRC, Catanga,

Kayoyo, 20-XII-1911, Bovone 87 (holo-: FI).

Panicum humidicola Rendle (Rendle 1899: 169). -- Brachiaria

humidicola (Rendle) Schweick. (Hubbard & al. 1936: 297). --
Urochloa humidicola (Rendle) Morrone & Zuloaga (Morrone &
Zuloaga 1992: 80). -- Brachiaria dictyoneura (Fig. & De Not.) Stapf
subsp. humidicola (Rendle) Catasús (Catasús Guerra 2001: 16). --
Type: Angola, Monino riv., Welwitsch 2678 (holo-: LISU, iso-: K).

One should avoid vague or relative terminology like “rather densely hairy”
or “fairly long”. For 2-dimensional shapes, it is advisable to use the stand-
ardized set of terms provided by the Systematics Association Committee
for Descriptive Biological Terminology (1962). Descriptions follow a logical
format, with the various elements being dealt with in a specific sequence:
plant – root/stem – leaves – inflorescence – flower – fruit – seed. Within
these elements, the sequence is from bottom to the top and from outside to
inside. For any organ, a good sequence for descriptive characters is: number
of elements – position – overall shape – overall dimensions – base/top/mar-
gins – texture – colour and lustre – surface (smooth, rough) – indumentum
and/or appendages. For more detail, see the text box below.

For fungi, a description guideline, similar to the one presented in the text
box, has been recently published in French (Eyi et al. 2011), and can be freely
downloaded from http://www.abctaxa.be/volumes/volume-10-les-champi-
gnons-comestibles-de-l-afrique-centrale

In a taxonomic treatment, the description is then followed by several para-

graphs dealing with distribution, ecology, vernacular names, uses, dispersal etc.
Relevant specimens can be cited here or in a separate table as an Appendix
or as Supplementary Material. The specific format of these specimen lists
varies between authors and journals (see next paragraph). A distribution map
can also be added.

Then, notes may be added providing the arguments for the taxonomic deci-
sions and/or the choices made related to the typification of names, etc.
64
It is advisable to add botanical illustrations to the revision. These are inval-
uable in the identification process and assist the user in understanding the
diagnostic elements of the taxa. One can make these illustrations oneself,
but it is usually better to ask a skilled botanical illustrator. Becoming a skilled
Guide to a well-structured plant description

Plant habit, height, distribution of sexes, exudate, other characters

shared between different elements, indumentum; bole and
branches: diameter, form and structure, indumentum, bark or
surface; twigs and/or nodes: as for bole and branches.
Stipules: presence/absence, position, shape, dimensions, base, apex,
indumentum, colour.
Leaves: position, simple or compound; sheath: position, shape,
dimensions; stipellae: cf stipules; petiole: shape, length, indumentum;
when leaf compound: rachis: length, articulations; petiolules: cf.
petioles; leaflets: number, sessile or not, shape, dimensions; leaf
blade (simple or compound): shape, dimensions, base, apex,
margin, indumentum, texture, colour, upper and lower surface;
venation type; primary vein: sunken or elevated; secondary veins:
number; tertiary venation.
Inflorescence: position, structure, form and/or dimension; peduncle:
dimension, indumentum; axes: position, indumentum; bracts:
position, form, dimension, indumentum; number of flowers, bi- or
unisexual.
Flower: position, symmetry, odour; when flowers unisexual describe
male ones first, then female ones; pedicel: dimension, pubescence;
bracteoles: position, shape, dimension, indumentum; flower buds:
shape, dimensions; hypanthium: shape, dimensions; perianth:
number of distinct whorls; sepals: free or number of calyx
lobes, position, shape, dimensions, colour, texture, apex, margin,
indumentum; petals/tepals: cf. sepals; disk: nectaries or glands,
position, shape, dimension, colour; androecium: type, position;
stamens: number, position (insertion); filaments: length, colour,
indumentum; anthers: insertion, dehiscence, shape, dimensions,
colour; connective: shape, dimension; staminodes: cf. stamens;
gynoecium: position, number, pubescence; ovary: number, position,
shape, dimension, indumentum, number of locules, placentation;
ovule: insertion, number; style: position, number, shape, dimension,
colour, pubescence; stigma: position, number, shape, dimension,
colour.
Infructescence: cf. inflorescence.
Fruit: type, dehiscence type, shape, dimension, colour, surface,
indumentum, number of seeds; peri-, exo- endocarp: structure,
dimension, colour.
Seed: shape, dimension, colour, surface; arilloid/testa: structure,
65
dimension, colour; endosperm, cotyledons, embryo, radicle.
Characters provided within a genus or family description do not have
to be repeated in a species description.
 illustrator does not only require talent, but also a considerable investment
in time. One may follow specific courses or study relevant books on the
subject. In some countries botanical artists have created associations which
one may join (e.g. https://www.botanicalartandartists.com for England, https://
www.botanischkunstenaarsnederland.nl for The Netherlands and the Société
Française d’Illustration Botanique, http://www.sfib.fr for France).

Finally, one will need to prepare an identification key to the taxa studied.
The key should be practical and assist the user in the identification process.
Preferably use characters that are easily observed. For other suggestions, see
Chapter 4.

G. Producing the manuscript and publishing the revision

When the taxonomic part of the study (Genus treatment, Key to the spe-
cies, Species treatments, maps, illustrations, etc.) is finished, one will need to
prepare the manuscript for publication. When the revision is meant to be
incorporated in a Flora, the taxonomic part is often all that is needed. When,
however, one wants to publish the results as an article in a scientific journal,
various other paragraphs will need to be added. While some are general, e.g.
Introduction, Materials and Methods, others are typical for a taxonomic revi-
sion, e.g. the History of the Genus, providing a historical overview of previous
studies and their contributions to the taxonomic framework of the genus.

Important factors to consider when choosing a suitable journal to submit the

manuscript to are: Impact Factor, regional impact, whether it provides Open
Access or not, and publication fees. For taxonomic revisions it is important to
check whether the journal accepts specimen citations (and if so in what for-
mat). As taxonomic revisions may be quite lengthy, page limitations may apply.

Finally, it is customary to acknowledge all herbaria (and their curators) who

have provided access to their collections or have sent specimens on loan.
They generally greatly appreciate receiving a copy of the published revision.

66
Plant collecting

• Fish L. (2004) La préparation des échantillons d’herbier. Scripta

Botanica Belgica 31: 92 p. ISBN 9072619633
• Bridson D., Forman L. (2000) The herbarium handbook, 3rd edition.
Richmond, Royal Botanic Gardens, Kew. EAN: 9781900347433

Historical literature (most useful to retrieve protologues)

• Biodiversity Heritage Library: https://www.biodiversitylibrary.org

• Botanicus: http://www.botanicus.org
• Taxonomic Literature: http://www.sil.si.edu/DigitalCollections/tl-2
• B-P-H: Botanico-Periodicum-Huntianum (Journal titles and their
standard abbreviations): http://fmhibd.library.cmu.edu/fmi/iwp/cgi?-
db=BPH_2015&-loadframes

Index Herbariorum

• http://sweetgum.nybg.org/science/ih

Plant use information

• Plant Resources of Tropical Africa: https://www.prota4u.org/database

• PlantUse: https://uses.plantnet-project.org/fr/Accueil

Digital herbaria and specimen information

• Global Biodiversity Information Facility (GBIF): https://www.gbif.org

• Meise Botanic Garden, Belgium (BR):
http://www.botanicalcollections.be
• MNHN Paris herbarium (P):
https://science.mnhn.fr/institution/mnhn/collection/p/item/search
• Naturalis herbarium, The Netherlands (L, U, WAG, AMS): 67
https://bioportal.naturalis.nl
• Tropicos, Missouri Botanical Garden, St. Louis (MO):
http://www.tropicos.org
 Georeferencing

• AFRITERRA (historical maps of Africa): http://catalog.afriterra.org

• Cartesius (historical maps of Belgium and Central Africa):
http://www.cartesius.be/CartesiusPortal
• GEOLocate (A platform for georeferencing natural history collections
data): https://www.geo-locate.org
• GeoNames (finding place names, also historical ones):
https://www.geonames.org
• Google Earth (free software to view the globe):
http://www.google.co.uk/earth/download/gep/agree.html

68
References

69
 Angiosperm Phylogeny Group (2016) An update of the Angiosperm
Phylogeny Group classification for the orders and families of flowering
plants: APG IV. Botanical Journal of the Linnean Society 181: 1–20.
Bridson D., Forman L. (eds) (2010) The herbarium handbook, 3rd ed.
Richmond, Royal Botanic Gardens, Kew.
Brummitt R.K. (2002) How to chop up a tree. Taxon 51: 31–41.
Brummitt R.K., Powell C.E. (eds) (1992) Authors of plant names. A list of
authors of scientific names of plants, with recommended standard form of
their names including abbreviations. London, Royal Botanic Gardens, Kew.
Darwin C. (1859) On the origin of species. London, J. Murray.
Eyi Ndong H., Degreef J., De Kesel A. (2011) Champignons comestibles des
forêts denses d’Afrique centrale, Taxonomie et identification. In: Degreef
J. (éd.) Abc Taxa, volume 10. Brussels, Belgian National Focal Point to the
Global Taxonomy Initiative.
Fish L. (1999) Preparing herbarium specimens. Strelitzia 7: 4–44.
Fish L. (2004) La préparation des échantillons d’herbier. Scripta Botanica
Belgica 31. Meise, Jardin botanique national de Belgique.
Ghorbani A., Wieringa J.J., Boer H.J. de, Porck H., Kardinaal A., Andel T. van
(2018) Botanical and floristic composition of the historical herbarium of
Leonhard Rauwolf collected in the Near East (1573–1575). Taxon 67(3):
565–580.
Grant V. (1981) Plant speciation, 2nd ed. New York, Columbia University
Press.
Holmes S. (2003) Bootstrapping phylogenetic trees: Theory and methods.
Statistical Science 18(2): 241–255.
Hörandl E. (2006) Paraphyletic versus monophyletic taxa—evolutionary
versus cladistic classifications. Taxon 55: 564–570.
Magnin-Gonze J. (2009) Histoire de la botanique. Paris, Delachaux & Niestlé.
Meier R. (2008) DNA sequences in taxonomy: Opportunities and
challenges. In: Wheeler Q.D. (ed.) The new taxonomy: 95–127. Boca
Raton, CRC Press.
Michener C.D., Corliss J.O., Cowan R.S., Raven P.H., Sabrosky C.W., Squires
D.S., Wharton G.W. (1970). Systematics in support of biological research.
Washington D.C., National Research Council, Division of Biology and
Agriculture.
Pavord A. (2005) The naming of names: the search for order in the world of
plants. New York, Bloomsbury.
Podani J. (2010) Monophyly and paraphyly: A discourse without end? Taxon
59(4): 1011–1015.
70
Pometti C.L., Cialdella A.M., Vilardi J.C., Saidman B.O. (2007) Morphometric
analysis of varieties of Acacia caven: (Leguminosae, Mimosoideae):
Taxonomic inferences in the context of other Argentinean species. Plant
Systematics and Evolution 264: 239–249.
Queiroz K. de (2006) The PhyloCode and the distinction between
taxonomy and nomenclature. Systematic Biology 55(1): 160–162.
Raven P. (2004) Taxonomy: where are we now? Philosophical Transactions of
the Royal Society of London B, Biological Sciences 359: 729–730.
Rouhan G., Gaudeul M. (2014) Plant taxonomy: A historical perspective,
current challenges, and perspectives. In: Besse P. (ed.) Molecular plant
taxonomy: Methods and protocols, vol. 1115: 1–37. New York, Springer
Science & Business Media.
Sokal R.R., Sneath P.H.A. (1963) Principles of numerical taxonomy. New York,
W.H. Freeman & Co.
Soltis P.S., Soltis D.E. (2009) The role of hybridization in plant speciation.
Annual Review of Plant Biology 60(1): 561–588.
Sosef M.S.M. (1997) Hierarchical models, reticulate evolution and the
inevitability of paraphyletic taxa. Taxon 46: 75–85.
Stace C.A., Preston C.D., Pearman D.A. (2015) Hybrid flora of the British
Isles. Durham, Botanical Society of Britain & Ireland.
Stuessy T. (2009) Plant taxonomy, the systematic evaluation of comparative
data. New York, Columbia University Press.
Systematic Association Committee for Descriptive Biological Terminology
(1962) II. Terminology of simple symmetrical plane shapes, Chart 1. Taxon
11: 145–156.
Thiers B. (continuously updated). Index Herbariorum: A global directory of
public herbaria and associated staff. New York Botanical Garden’s Virtual
Herbarium. http://sweetgum.nybg.org/ih/
Turland N.J., Wiersema J.H., Barrie F.R., Greuter W., Hawksworth D.L.,
Herendeen P.S., Knapp S., Kusber W.-H., Li D.-Z., Marhold K., May T.W.,
McNeill J., Monro A.M., Prado J., Price M.J., Smith G.F. (eds) (2018)
International Code of Nomenclature for algae, fungi, and plants
(Shenzhen Code) adopted by the Nineteenth International Botanical
Congress Shenzhen, China, July 2017. Regnum Vegetabile 159. Glashütten,
Koeltz Botanical Books.
Victor J., Koekemoer M., Fish L., Smithies S., Mössmer M. (2004) Herbarium
essentials: the southern African herbarium user manual. SABONET
Report No. 25. Pretoria, SABONET.
Vorontsova S.M., Knapp S. (2016) A revision of the spiny solanums, Solanum
subgenus Leptostemonum (Solanacae), in Africa and Madagascar.
Systematic Botany Monographs 99. London, The Natural History
Museum.
Wiley E.O. (1978) The evolutionary species concept reconsidered.
Systematic Biology 27(1): 17–26.
71
World Conservation Monitoring Centre (1992) Global biodiversity: Status
of the earth’s living resources. London, Chapmann & Hall.
Photo credits:

Figure 1. Statue of Theophrastus in the botanic garden at Palermo, Italy

(photo by tato grasso - Own work (personal work), CC BY-SA 2.5, https://
commons.wikimedia.org/w/index.php?curid=3170845).
Figure 2. Plinius the Elder (from: https://www.britannica.com/biography/
Pliny-the-Elder/images-videos/media/1/464822/234312, accessed August
16th, 2019).
Figure 3. Dioscorides (from: De Desconocido - http://huntbot.andrew.
cmu.edu, Dominio público, https://commons.wikimedia.org/w/index.
php?curid=3607187).
Figure 10. Charles Darwin (from http://www.charlesdarwin.net/biography.
jsp; accessed august 16 th, 2019).
9 789492 663207