20 Phylogenetic Relationships and Evolution of Insects

Back

Rainer Willmann

330

More than 1.2 million recent insect species have been described,

but recent estimates suggest several million additional

species are to be expected (see the early discussion in

Weber 1933). About 25,000 fossil species are known, but

more than one billion insect species must have existed in the

past, of which only a small minority have left any trace in

sediments, and again, of these, only a small fraction will ever

be found (Willmann 2002). Systematically, insects are one

of the most studied groups. Modern biosystematics was developed

with insects as one of its main targets, because Willi

Hennig (1913–1976), the founder of phylogenetic systematics,

was an entomologist dealing mainly with Diptera.

However, the branching sequence of the insect tree is difficult

to reconstruct for several reasons. First, insects are a fastevolving,

enormously diverse group, and synapomorphies of

subordinate groups may have become veiled by more recent

evolutionary changes. Second, some taxa have preserved

ancient characters, and it appears that structures once lost

may reappear from time to time, leading to confusion among

phylogenetists. Third, the amount of homoplasy has been

underestimated. Fourth, although a huge number of fossil

insects have been assembled and described, the gaps in the

fossil record are considerable. Fossils have repeatedly shown

that some phylogenetic conclusions based on extant taxa are

untenable. Last but not least, there are far too few entomological

morphologists, and therefore, there is an enormous

lack of knowledge about the details of their structural disparity,

constructional morphology, relationships of taxa of

all hierarchical levels, and ground patterns of recognized

monophyla.

Many authors prefer the name “Hexapoda” over “Insecta.”

Snodgrass (1952), for example, pointed out that “Insecta” has

been used in very different ways, and indeed Linnй (1758)

included with them crustaceans, myriapods, and chelicerates.

Some recent authors such as Jamieson et al. (1999),

Kristensen (e.g., 1975, 1991), Ross et al. (1982), and Wheeler

et al. (2001) considered “Insecta” a synonym of “Ectognatha,”

whereas Whiting et al. (1997: fig. 5) equated “Insecta” with

“Pterygota” (but did not use the latter term). This recalls

Mayer (1876), who had argued that Thysanura and Collembola

are primarily wingless and also used the term Insecta

to refer only to winged and secondarily wingless kinds. Thirty

years later, Handlirsch (1908), uncertain of the phylogenetic

relationships of Collembola, Diplura, Archaeognatha, and

Zygentoma (Protura had only been introduced in 1907),

distinguished four hexapod classes, calling Pterygota “Insecta

s. str.” In this chapter, “Insecta” is used as synonym for “Hexapoda,”

which is in accordance with the works of Hennig (e.g.,

1953, 1969, 1981), many widespread textbooks both old and

recent (e.g., Naumann 1991, Kaestner 1973, Richards and

Davies 1977, Arnett 2000), and common use (e.g., Snodgrass

1935).

In the following discussion, I offer insights into the arguments

for and against different relationships. Many assumed

nodes are supported by very few characters. For

example, Wheeler et al. (2001) have, on average, seven morPhylogenetic

Relationships and Evolution of Insects 331

phological characters per node (not counting autapomorphies

of terminal taxa) and two to six characters in about

50% of the nodes; Beutel and Gorb (2001) have even fewer

(111 characters for 35 nodes). Exclusion or inclusion of just

a few characters would easily produce new phylogenetic

hypotheses. This emphasizes the need for more morphological

data as well as for more thorough morphological studies.

Many of the taxa treated in the following have been given

categorical ranks, and most of the widely known taxa appear

as “orders” in traditional classifications. In the following, no

use is made of categorical ranks, because different authors

differ in their opinion as to the rank of a particular taxon.

Many systematists state that sister groups must be assigned

the same categorical rank, which implies that usually a particular

rank (e.g., order) can only be used twice along a particular

evolutionary lineage. For example, if Lepidoptera

(moths and butterflies) is ranked as an order, then their sister

group (Trichoptera, caddis flies) would be an order as

well, whereas the immediately superordinate taxon of the two

(Amphiesmenoptera) would deserve a higher rank. Even

higher ranks would be attributed to Mecopteria (which includes

Amphiesmenoptera), Holometabola (which includes,

e.g., Mecopteria), Eumetabola (which includes Holometabola

and Acercaria), Neoptera (which includes Eumetabola) and

Polyneoptera (if this is a monophylum). Odonates (damselflies

and dragonflies) as the sister group of Neoptera would

require the same categorical rank as the latter, which is certainly

not the rank “order”. Thus, although cladograms can

be matched up with systematizations using Linnaean categorical

ranks because both are hierarchies, categorical formal

ranks are impractical in a phylogenetic system. However,

the main issue is that Linnaean categorical formal ranks like

family, order, class, and so forth, were not introduced to indicate

sister group relationships. They were not coined in a

phylogenetic context but to serve classifications on the basis

of Aristotelian logic instead (Griffiths 1974, 1976, Willmann

1987, 1997), a major step in systematics being the transformation

of a purely logical attempt into a phylogenetical science

(Burckhardt 1903). As Artois (2001:10) put it, “[T]he

nested hierarchical structures of the Linnean system and the

nested hierarchy found by phylogenetic analysis are based

on completely different premises and only superficially resemble

one another.” The debate on the issue is ongoing in

various directions (e.g., Artois 2001, Nixon and Carpenter

2000, Papavero et al. 2001).

Origin and Sister Group of the Insects

Insects are primarily wingless. Along with the possession of

three tagmata (head, thorax, and abdomen), two corneagene

cells of the ocelli developed as primary pigment cells (Paulus

1979), posterior tentorial apodemes that are elements of the

head skeleton (Koch 2000), six leg segments (Kristensen

1981, Willmann 1998; possibly a plesiomorphy, because the

number is also present in Symphyla and Pauropoda, but

homology of the podomeres among Tracheata is not clear),

14 body segments (possibly a plesiomorphy shared with

progoneates), and lack of appendages of any kind on abdominal

segment 10. Hexapody is considered to be one of the

apomorphies of insects. Contrasting with this view, Manton

(1977) believed that hexapody has developed within Hexapoda

independently five times, in Diplura, Protura, Collembola,

Thysanura, and Pterygota, because leg mechanics

are different in the groups and because the mechanism in one

of them cannot be ancestral to that in one of the others. What

Manton had overlooked, however, is the fact that the different

mechanisms may have had their origin in a common

ancestor with a leg mechanism not found in any of the recent

taxa. Although insect monophyly is doubted by some

(see Dohle 1998 and discussion in Klass and Kristensen

2001), most molecular sequence studies have supported the

view that insects are a natural taxon (Wheeler et al. 2001),

which is accepted here.

It has been suggested that insects are the sister group of

or derive directly from crustaceans [Crustacea + Insecta =

Tetraconata (Dohle 2001), because they possess a crystal

cone consisting of four pieces in the ommatidia of their compound

eyes). But according to most morphological data, the

Tetraconata hypothesis is in all probability not true, because

insects share an impressive number of derived features with

Chilopoda (centipedes) and Progoneata (millipedes and

relatives), together constituting Tracheata. Possible synapomorphies

are the loss of the first post-antennal head segment

(intercalary segment), loss of the mandibular palp, loss of the

pretarsal levator muscle with only one muscle remaining,

possession of ectodermal Malpighian tubules, and tracheae

(all discussed by Snodgrass 1938, but also many subsequent

authors), as well as the organ of Tцmцsvary (which is a

receptor on the head described under various names in

Chilopoda, Diplopoda, Symphyla, Pauropoda, Collembola,

and Protura), possession of anterior tentorial arms (Kristensen

1989), addition to the head of a sixth segment bearing

the second maxillae, and the centriole adjunct in the sperm

(Jamieson et al. 1999). [Hilken’s (1998) conclusion as to the

multiple origins of tracheae within Tracheata is contradicted

here.] According to this phylogenetic hypothesis, the crystal

cone in the compound eye is assumed to be lost in centipedes

and progoneates. Moreover, the progoneates (Symphyla +

Dignatha) share a number of derived characters only with

hexapods.

The progoneates are characterized by several derived

characters, for example, the position of the genital opening

that is in the anterior part of the body, the development of

the midgut within the yolk (the midgut lumen is therefore

free from the yolk), the formation of a fat body out of vittelophages,

trichobothria with basal bulb, and loss of the

palps of the first maxilla (Dohle 1998, Kraus 2001). Therefore,

progoneates may well be monophyletic and the sister

group of the insects. The group Insecta + Progoneata has been

332 The Relationships of Animals: Ecdysozoans

called Labiata (Snodgrass 1938) because insects and one

progoneate taxon, the symphylans, have their posterior

mouth parts fused into the so-called labium that is often

considered to be a derived labiatan ground pattern character

(but see below).

Some derived characters are possessed by insects and

symphylans only. These include styli on the underside of the

body, vesicles in a ventral position (vs. their presence at the

basal podomere in some diplopodans), and appendages at

the labium (glossa and paraglossa). Furthermore, the labium,

a mouth part consisting of the united second maxillae that

has been used as an argument for the Labiata hypothesis (see

above), is in fact present only in Symphyla and Insecta: it

cannot be traced in Dignatha because it has the second maxillae

reduced. Therefore, it is difficult to decide between the

Progoneata + Insecta hypothesis and the Symphyla + Insecta

hypothesis (see Willmann 2003 for details). The central problem

with the latter grouping is the anterior genital openings

possessed by the progoneates only. Indeed, some authors

claim that the anterior genital opening is not a synapomorphy

of the progoneates but that the positions of the genital openings

in insects somewhere in the posterior body area derive

from a progoneate situation. A fourth hypothesis suggests

that insects constitute the sister group to the myriapods

(Myriapoda = Chilopoda + Progoneata). This hypothesis is

not discussed here at length, because the assumption that

myriapods are monophyletic is not supported by morphological

evidence (Dohle 1998), although Baccetti (1979), Jamieson

(1987), and Jamieson et al. (1999) point to a derived similarity

in sperm structure in chilopods and progoneates. A striated

cylinder surrounding the 9 + 2 axoneme is considered to

be an autapomorphy of Myriapoda, although it has been demonstrated

only in chilopods and pauropods. The cylinder is

possibly the homologue of the coarse fibers (intersinglet or

intertubular material) of the insectan sperm. A phylogenetic

tree based on DNA data for eight loci and morphological

characters produced by Giribet et al. (2001) show myriapods

as monophyletic and as the nearest relatives of Pancrustacea

(= Crustacea + Insecta). These results are difficult to interpret,

however, because insects appear scattered among Crustacea,

results that are “confusing,” in the words of the authors

(e.g., Diplura–Campodeidae as sister group to Protura/Hexapoda,

Diplura–Japygidae as sister group to barnacles/crustaceans,

whereas Giribet et al. accept that japygids are basal

hexapods).

To summarize: (1) current morphological knowledge

offers no clear support for the existence of a sister-group

relationship between insects and a nontracheate group, and

(2) under the tracheate hypothesis it is not clear whether

Progoneata or Symphyla is the nearest relative of the insects.

Only a few molecular studies support this latter view, such

as the 18S ribosomal DNA (rDNA) tree published by Wheeler

et al. (2001). It must be noted in this context, however, that

very few molecular studies have been undertaken to address

this question.

Insect Phylogeny and Evolution

Insect evolution began more than 400 Myr (million years)

ago, as deduced from fossil springtails of Lower Devonian

Age, 395 Myr old (Rhyniella praecursor), and the discovery

of possible ectognathan mandibles from the same locality and

of the same age (Rhyniognatha hirsti). Springtails, which include

about 6000 described species, have unique derived

characters such as specialized appendages at the end of their

abdomen that are united to form a furca or spring used for

jumping. When at rest, the spring is held under the abdomen

and fixed by the retinaculum, another organ produced

by abdominal legs. The springtails also have a so-called ventral

tube, developed from fused abdominal vesicles, and many

other derived characters. These structures are known from

the Devonian springtails as well, and because it is unlikely

that they evolved in a short period of time, it is probable that

insects had a long history before that epoch.

Today, no insect has abdominal legs. Yet, because the

jumping organ of the springtails consists of three segments,

the last ancestor of insects must have had abdominal legs with

at least three podomeres. Legs of that kind must have been

lost independently in the two basal lineages of insects: in

Ellipura, on the one hand, and in Euentomata (= Diplura +

Ectognatha), on the other. A few fossils seem to fill either the

gap between the insects and their multilegged sister group

or the gap between the last stem species of Insecta and the

first split within Euentomata (for figures see Haas in Bechly

2001, Willmann 2003).

The springtails and Protura (telsontails, including about

500 described species) are subgroups of Ellipura. Their close

relationship is well substantiated: they have no abdominal

tracheal stigmata and no styli, they possess cranial folds covering

the mouth parts in a unique way, and they possess a

longitudinal fold on the underside of the head and neck not

found in any other insect, the so-called linea ventralis. On

the anterior part of the abdomen (first segment in Collembola,

segments 1–3 in the Protura) are appendages consisting

of an eversible vesicula that are paired in telsontails but

fused in springtails.

Diplura (800 species) poses a major problem in insect

phylogeny. “Diplurans” means doubletails, and the name

refers to their cerci, which may either be long and multisegmented

or short and used for grasping. The latter character

state is derived. Until a few years ago almost all

entomologists agreed that the diplurans are most closely related

to the ellipurans (fig. 20.1), because their mouth parts

are also hidden by cranial folds (Diplura + Ellipura = Entognatha).

Other possible autapomorphies of Entognatha are

the reduction of the Malpighian tubules and the absence of

a centriole adjunct in the sperm. However, a structure superficially

resembling an adjunct is present in telsontails

(Acerentulus; Jamieson et al. 1999), and according to Koch

(1997, 2000, 2001) it is probable that the cranial folds have

developed independently. Now it seems likely that the dipPhylogenetic

Relationships and Evolution of Insects 333

Figure 20.1. Previous hypotheses of

relationships among insects based on

morphology, illustrating advancements

in insect phylogenetics over 30

years, beginning with the first

“Stammbaumentwurf” of Hennig (for

Holometabola, see Whiting 2002).

(A) Hennig (1953). (B) Hennig

(1969). (C) Kristensen (1981; all

redrawn from sources). The taxon

names are those used by the

respective author. Dashed lines

indicate uncertainty in relationships.

Hennig (1969) inadvertently united

Protura and Diplura in his figure as

“Ellipura,” which is corrected here. It

should be noted that these authors

have favored a particular view, but

they have always discussed alternative

ideas.

Isopteria Psocodea Homoptera

Orthopteria

Neoptera

Metapterygota

Pterygota

Dicondylia

Eumetabola

Saltatoria

Collembola

Protura

Diplura

Machilidae

Lepismatidae

Ephemeroptera

Odonata

Notoptera

Phasmida

Caelifera

Ensifera

Dermaptera

Blattodea

Isoptera

Mantodea

Plecoptera

Embioptera

Zoraptera

Psocoptera

Phthiraptera

Thysanoptera

Auchenorrhyncha

Sternorrhyncha

Heteroptera

Holometabola

Parametabola

Paurometabola

A

Neoptera

Pterygota

Dicondylia

Eumetabola

Ectognatha

Ellipura

Entognatha

Palaeoptera

Paurometabola

Blattopteriformia

Blattopteroidea

Orthopteroidea

Hemiptera

Condylognatha

Acercaria

Psocodea

Paraneoptera

Collembola

Protura

Diplura

Archaeognatha

Zygentoma

Ephemeroptera

Odonata

Embioptera

Notoptera

Dermaptera

Mantodea

Blattariae

Isoptera

Ensifera

Caelifera

Phasmatodea

Plecoptera

Zoraptera

Psocoptera

Phthiraptera

Thysanoptera

Sternorryncha

Auchenorryncha

Heteropteroidea

Holometabola

B

Orthopteromorpha

Collembola

Protura

Diplura

Archaeognatha

Zygentoma

Ephemeroptera

Odonata

Plecoptera

Embioptera

Phasmida

Orthoptera

Grylloblattaria

Dermaptera

Dictyoptera

Zoraptera

Psocoptera

Phthiraptera

Hemiptera

Thysanoptera

Holometabola

Psocodea

Condylognatha

Acercaria

Paraneoptera

Entognatha Insecta (=Ectognatha)

Dicondylia

Pterygota

Neoptera

C

334 The Relationships of Animals: Ecdysozoans

lurans are the sister group of the rest of the insects, whose

mouth parts are externally visible. These are called Ectognatha.

Few characters support the hypothesis that Diplura are the

sister group of Ellipura, whereas a number of characters appear

to be synapomorphies of the Diplurans and the ectognathous

insects: for example, the lack of Tцmцsvary’s organs on

the head, the lack of abdominal legs, a new mode of molting,

the possession of long filamentous appendages of the 11th

abdominal segment (cerci), the structure of the tail of the

sperm, superficial cleavage (character state uncertain as the

type of cleavage is unknown in Protura), and epimorphosis

(the young hatch with the full number of abdominal segments,

whereas the plesiomorphous state is a hatchling that adds several

abdominal segments after having left the egg). Some authors

have suggested that a movable appendage of the

mandible is further evidence for the monophyly of the clade

Diplura + Ectognatha (Richter et al. 2002).

Ectognatha

Ectognatha consists of Archaeognatha (bristletails, about 390

described species), Zygentoma (silverfish, 400 species), and

Pterygota (winged insects). Monophyly of Ectognatha has

never been doubted, because its members have a large number

of derived characters in common (figs. 20.1, 20.2, 20.3).

The more obvious ones are an antenna with a long flagellum

that lacks muscles and possesses Johnston’s organ in its second

segment, and females with an ovipositor whose elements

are contributed by ventral sclerites of the 8th and 9th abdominal

segments.

The characters used for reconstructing insect phylogeny

are sometimes very complex and no doubt determined by a

large number of genes. This makes morphological structures

a powerful tool in reconstructing phylogeny. To give an example:

the monophyly of the ectognathous insects is also

supported by the structure of blood vessels. First, members

of Tracheata have vessels extending from the head into the

antennae. Primitively, the antennal blood vessels are connected

to the large dorsal vessel, but in the ectognathous

insects these vessels are separate and thus there are several

circulatory systems. And second, in Pterygota (winged insects),

each antennal vessel has a pulsatile ampulla that functions

as a pump or as an “antennal heart” (Pass 1998).

Dicondylia

The first two branching events in the phylogeny of Ectognatha

were the central focus of a classic controversy in systematics.

Because bristletails and silverfish are superficially similar, they

were often united in one group. It has long been known, how-

Figure 20.2. Cladogram from a combined molecular analysis of insects minimizing character incongruence between molecular data

sets (18S rDNA and 28S rDNA data); redrawn and condensed from Wheeler et al. (2001). Each insertion:deletion event was weighted

1, as were transitions and transversions. Arrows point to non-holometabolan taxa within the clade next to Dermaptera, which consists

mainly of Holometabola. Holometabola are generally accepted as monophyletic. Numbers identify taxa that are split into several units

and so appear at different places in the cladogram.

Protura

Diplura

Collembola

Archaeognatha

Zygentoma

Ephemeroptera

Odonata

Phasmida 2

Embiidina

Orthoptera

Hemiptera 1

Thysanoptera

Psocodea

Hemiptera 2

Dermaptera

Coleoptera

Hymenoptera 1

Hymenoptera2

Megaloptera1

Mecoptera+

Siphonaptera

Megaloptera 2

Plecoptera 1

Notoptera

Planipennia 1

Zoraptera

Trichoptera

Lepidoptera

Planipennia 2

Raphidiodea

Planipennia 3

Strepsiptera

Planipennia 4

Plecoptera 2

Diptera

Neoptera

Pterygota

Dicondylia

Phasmida 1

Isoptera 1

Mantodea

Blattaria 1

Blattaria 2

Isoptera 2

Phylogenetic Relationships and Evolution of Insects 335

ever, that bristletails are more plesiomorphous than are silverfish

and that silverfish are more closely related to the winged

insects than to bristletails. In 1953 Hennig introduced the

name “Dicondylia” for silverfish and winged insects, reflecting

their phylogenetic relationships (the two taxa share, e.g.,

a mandible with two articulations, or condyli, with the head

capsule). Yet it was only 25 years ago that Hennig’s systematic

framework of basal insects was generally taken into textbooks,

and in fact, there are still a number of textbooks that present

the old classification [e.g., Ross et al. (1982: 284), which unites

Archaeognatha and Thysanura into Apterygota; Richards and

Davies (1977), which subsumes under Apterygota all primarily

winged insects; Borroret al. (1992), which also considered

Archaeognatha and Thysanura as “apterygote insects”].

It may be, however, that things are not that simple. In

California, the species Tricholepidion gertschi (Wygodzinsky

1961), the bristlefish, is the only surviving representative of a

taxon originally described from Baltic amber, Lepidotrichidae.

This species is usually regarded as belonging to Zygentoma

(e.g., Boudreaux 1979, Kristensen 1998, Wygodzinsky 1961),

but it may also be the sister group of Zygentoma + Pterygota

(Kristensen 1991, Klass 1998, Staniczek 2000; for a summary

of the evidence, see Willmann 2003).

Pterygota

The oldest known winged insects come from the uppermost

Mississippian, or middle Carboniferous (Delitzschala bitterfeldensis,

Ampeliptera limburgica, Stygne roemeri, Brodioptera

stricklani; age between 317 and a little more than 320 Myr).

They were already advanced because they were fully winged

and capable of flight, and it is unknown what the first

pterygotes looked like. It is also unknown what the function

of the first winglike structures was. They were certainly very

small, and they cannot have served as flight organs but may

have supported thermoregulation.

For decades it has been debated whether mayflies or

odonates are the first side branch of the Pterygota or whether

the two combined form a clade of their own (Palaeoptera,

fig. 20.1B). Although the Palaeoptera hypothesis persists

(Hovmцller et al. 2002), several characters used to support

it are hardly tenable [aquatic larvae, possibly a convergence;

fusion of galea and lacinia, but the fused parts of the maxilla

in Odonata may not represent galea and lacinia (Staniczek

2000); short antennal flagellum, apparently a convergence

(Soldбn 1997)], whereas the character state of other structures

(wing: anterior media fused to the radial sector, intercalary

veins) is uncertain (Willmann 1999). More convincing

is another hypothesis, favored by Kristensen (e.g., 1981, 1989,

1991; see also Hennig 1953, 1986; fig. 20.1A,C). Based on

evidence from head morphology, the mayflies are the sister

group of the remaining pterygotes (Staniczek 2000). Another

indication that this may be so is the subimago, a flying stage

followed by the final flying stage, the reproducing imago. The

subimaginal stage is considered to be an ancient character, and

it is not retained in any other recent insect group. Grimaldi

(2001) has stressed that loss of an imaginal molt in odonates

and Neoptera cannot simply be attributed to convergence (I

know of no evidence that some Paleozoic neopterans molted

as flying life stages; see Kristensen, 1989, 1991). Again, odonates,

Neoptera, and also Palaeodictyopteroidea have lost their

paracercus (the median terminal filament).

With respect to their copulatory apparatus, mayflies resemble

derived winged insects, the Neoptera. The males have

long styli on their 9th abdominal segment that serve to grasp

the female’s abdomen. It is difficult to tell whether or not this

is a synapomorphy, because the grasping organ in mayflies

and neopterans might be a convergent similarity. As an argument

supporting this view, one could point to Odonata

and Palaeodictyopteroidea. Odonata have an indirect mode

of sperm transfer, and it is unlikely that it derives from a

gonopore-to-gonopore transfer (Bechly et al. 2000). Palaeodictyopteroidea

were minute to huge insects (wing span up

to 55 cm) of impressive diversity and species richness, and

they are known from the Carboniferous to the Triassic. Some

of them had very small copulatory organs, which may not

have served to hold the female (they may have instead been

tactile organs), and palaeodictyopteroids are usually considered

to be more closely related to neopterans than are mayflies

and odonates. In fact, however, it is unclear where they

belong. It is generally stated that they are monophyletic because

they have elongate mouthparts forming a beak, but

mouthparts are well known in only a few specimens and do

not always form a distinctive proboscis (Novokshonov and

Figure 20.3. Basal phylogenetic relationships of insects as

favored in this chapter (but see text for alternative hypotheses).

For the higher winged Insecta (Neoptera), with the exclusion of

Holometabola, see figures 20.4 and 20.5.

Diplura

Ellipura

Metapterygota

Pterygota

Dicondylia

Ectognatha

Euentomata

Neoptera

Palaeodictyopteroidea

Odonata

Ephemeroptera

Zygentoma

Archaeognatha

Protura

Collembola

336 The Relationships of Animals: Ecdysozoans

Willmann 1999). Therefore, palaeodictyopteroids may not

be a natural group. That the transfer of sperm from gonopore

to gonopore may have evolved independently in mayflies

and Neoptera has already been discussed by Kristensen

(1981).

Odonates and the other winged insects have mandibles

and mandibular muscles quite different from those in mayflies

and primarily wingless insects. For this reason, Bцrner

(1908) united Odonata and Neoptera (higher winged insects)

under the name Metapterygota. Odonata consist of Zygoptera

(damselflies) and Epiprocta (= Epiproctophora), which are

in turn composed of Epiophlebioptera and Anisoptera (dragonflies).

Zygoptera, often considered to be a paraphyletic

group, are certainly monophyletic, as evidenced by numerous

autapomorphies, among them the distinctly stalked

wings, extremely broadened hammer-shaped head capsule

with widely separated eyes, extreme obliqueness of the

pterothorax, and an ovipositor pouch formed by the enlarged

outer gonapophyses (valvulae 3) of the 9th abdominal

segment of the female (Bechly 1996, Lohmann 1996).

Members of Epiprocta have enlarged eyes. The wing nodus

lies almost in the middle of the fore margin. The larvae have

rectal folds containing tracheal gills in a rectal gill chamber.

Today, Epiophlebioptera consists only of Epiophlebia. The

widely used name “Anisozygoptera” is no longer in use

among phylogenetists because it denoted a paraphyletic

group. Lohmann (1996) has attempted to reconstruct in

detail the phylogeny of Anisoptera, but the relationships

within the group are disputed (see Bechly 1996) and monophyly

of many odonatan taxa (e.g., the zygopteran “families”)

has yet to be demonstrated (Jarzembowski et al. 1998).

Neoptera

The wings of the Neoptera—all recent pterygotes not belonging

in the mayflies or odonates—are probably more advanced

than those of mayflies and odonates. In particular, neopterans

have sclerites at the wing base, thus allowing the wings to

be folded back over the abdomen (Martynov 1925, Hennig

1969, Hцrnschemeyer 1998, 2002).

Basal relationships among neopteran groups have been

under dispute because of the uncertain position of Plecoptera

(stoneflies). This is a cosmopolitan group of common insects

with about 2300 species. Their nymphs are virtually ubiquitous

in rivers and brooks. The phylogenetic relationships

within the group were the topic of one of the classical studies

in phylogenetic systematics (Zwick 1973), and the results

it revealed are still considered valid (Zwick 2002). In

Plecoptera, two sister taxa with very different distributional

patterns have been recognized. Arctoperlaria (Systellognatha

+ Euholognatha, >1500 species) occurs mainly in

the Northern Hemisphere, whereas Notonemouridae and

representatives of Perlidae live in the Southern Hemisphere.

Antarctoperlaria (Eusthenioidea + Gripopterygoidea, ~300

species) are strictly confined to Australia, South America, and

New Zealand (Zwick 1973, 1980). Eusthenioids are commonly

colored, which is unusual in Plecoptera, because most

other species are grayish brown.

Because plecopterans appear to be more plesiomorphous

than other neopterans in the segmental arrangement of their

testes, and because they have a transversal muscle in the stipes,

otherwise known only from Archaeognatha and Zygentoma,

Zwick (1980) hypothesized that they represent the sister

group of the remaining neopterans. Beutel and Gorb (2001)

believed the aquatic larvae of plecopterans to be another plesiomorphy

compared with the terrestrial larvae found in other

neopterans, but this is unlikely because neither Archaeognatha

nor Zygentoma has an aquatic early life stage, and the larvae

of Palaeodictyopteroidea were, as far as known, terrestrial as

well. Stys and Bilinski (1990) and Bьning (1998) assume that

Plecoptera are the sister taxon to a monophyletic group consisting

of Dermaptera and Eumetabola. Disturbingly, according

to the most recent molecular analyses (Wheeler et al. 2001),

Plecoptera appear in Holometabola in a combined molecular

analysis (18S rDNA and 28S rDNA) that minimized character

incongruence between the molecular data sets. 18S rDNA

analyses put Plecoptera as a sister group to Psocodeans +

Zoraptera + Thysanoptera, whereas 28S rDNA data place them

as the sister group to thrips. These, as well as other acercarians,

form the sister group to Hymenoptera. In a total evidence cladogram

including morphological data (Wheeler et al. 2001),

Plecoptera appeared as a sister group to Embiida. In this case,

the trees resulting from molecular data were modified according

to one interpretation of morphological data, but characters

are interpretations. Thus, structural similarities between

Embiida and Plecoptera were considered to be plesiomorphies

by Rдhle (1970) and Zwick (1980).

Based on evidence from wing structures, it appears likely

that Plecoptera are part of a species-rich group named Polyneoptera

by Martynov (1925). The remaining species possibly

fall into two other neopteran groups, Acercaria and

Holometabola.

Polyneoptera

Polyneopterans (fig. 20.4) are characterized by a number of

probably derived hind wing structures. Two veins, the second

cubitus and the first anal vein, are almost straight and

run parallel to one another. The remaining anal veins form

a fan, and the second anal vein splits into two or more

branches, whereas the others do not. Some polyneopterans,

such as the rock crawlers (Notoptera, Grylloblattaria), are

wingless, whereas others have small wings with reduced venation

like the Embiida (web spinners), but their male genitalia

and molecular data support the view that they belong

in the group. However, as in many other cases, the aforementioned

hind wing structures have been accorded differing

significance, being convergences in Kristensen’s (1991) opinPhylogenetic

Relationships and Evolution of Insects 337

ion. Dermapteran hind wings resemble those of the (other)

polyneopterans only superficially.

The relationships within Polyneoptera are only partially

clear. It has long been suggested that the praying mantids,

roaches, and termites (Mantodea, Blattodea, Isoptera) form

a systematic unity called Dictyoptera, and morphological

work on the gut structure and female genitalia has supported

this view (Klass 1995, 1998: fig. 4), in accordance with several

(Wheeler et al. 2001; fig. 20.2) but not all molecular

analyses. One major step in the evolution of Dictyoptera was

the development of sociality, when some cockroaches became

termites. According to this evolutionary scenario, which was

proposed by Wheeler (1904, 1928) and Handlirsch (1908),

termites are a highly evolved subgroup of the roaches. It

appears that the roach Cryptocercus is the closest relative of

Isoptera. Isopterans and Cryptocercus share a rich diversity

of hindgut symbionts belonging to Oxymonadida/Metamonada

and Hypermastigida/Parabasalia. Klass (2001b) believes

that it is unlikely that the associations are due to lateral

transfer, as suggested by Grandcolas and Deleporte (1996).

Grandcolas (1994, 1996, 1997) assumed that xylophagy and

intestinal symbiosis of Cryptocercus and Isoptera is a matter

of convergence because he was of the opinion that Cryptocercus

has a subordinate position within the Polyphaginae/

Blattaria. Klass (2000, 2001b, and previous publications)

presented evidence that Cryptocercus is only distantly related

to Polyphaginae. He showed that most of the autapomorphies

indicated in the cladogram of Grandcolas had to be rejected

as supporting the respective clades largely because of erroneous

homologies, but he also stated (Klass 2001b:263) that

blattarian phylogeny itself is not finally settled. Lo et al.

(2000) found strong support for the clade Cryptocercus +

Isoptera based on the combined analysis of several gene sequences.

The oldest known termites come from Cretaceous

sediments, and today they are important modifiers of soil

structure in tropical environments, with thousands of billions

of individuals. The queen of Bellicositermes natalensis

lays one egg every two seconds, which gives a total of 43,000

eggs per day.

The closest allies of the Dictyoptera are possibly the dermapterans

(earwigs, about 1900 species), whose cerci are

usually transformed into a forceps. It is sometimes believed

that the very short ovipositor is a synapomorphy of Dictyoptera

and Dermaptera, but Jurassic earwigs with a long

ovipositor show that this is not correct. Indicators of a close

relationship between Dermaptera and Dictyoptera are the

pterothoracic musculature and similarities in wing venation

(Klass 1998, Willmann 2003; but see below).

Grimaldi (2001) lists four apomorphies in favor of a Dermaptera

+ (Zoraptera + Embiida) relationship, but none of

them (three-segmented tarsi, ovipositor highly reduced, loss

of ocelli, cerci reduced to a one- or two-segmented appendage)

were developed in Jurassic earwigs (Vishniakova 1980)

and thus do not pertain to the dermapteran ground plan

(Willmann 1990, 2003). The view that Embioptera are most

closely related to Dermaptera receives weak support from a

spermatozoal similarity (shared oblique implantation fossa),

but this is in conflict with a spermatozoal apomorphy shared

by Phasmatodea and Dermaptera (double anterior axonemal

cylinder; Jamieson et al. 1999).

The situation with regard to the Dermaptera is even

more complicated than indicated above. The hind wing

similarities between Dermaptera (including its stem-group

representatives that are usually united under the term “Protelytroptera”)

and other polyneopterans are only superficial

because of the apomorphic structure of the former. This is

certainly not evidence of a position outside Polyneoptera, but

Bьning (1998) assumes that the earwigs are the sister group

of Eumetabola (Dermaptera + Eumetabola = Meroista) based

on similarities in the ovarioles. Interestingly, Mesozoic male

dermapterans had well-developed gonobases and gonostyli

(Vishniakova 1980) that do not occur in other Polyneoptera.

This demonstrates that reduction of the structures has occurred

independently.

Dermaptera include one taxon, Hemimerus (~10 species),

that has no forceps but segmented cerci instead. Popham

(1985) believed the cercal structure of Hemimerus to be plesiomorphous,

and the earwigs were therefore subdivided into two

subordinate taxa, Hemimerina and Forficulina. As some Jurassic

Dermaptera had unsegmented cerci (e.g., Turanoderma)

but were plesiomophous in many other respects, and because

Hemimerus shares several apomorphies only with recent earwigs,

Willmann (1990) concluded that Hemimerus has secondarily

segmented cerci due to pedomorphosis and that

subdividing Dermaptera into Hemimerina and Forficulina is

unfounded. This corresponds to the view of Giles (1974), who

regarded the forceps as an autapomorphy of Dermaptera, later

lost in Hemimerus. This view has gained strong support from

detailed morphological studies (Klass 2001a).

Figure 20.4. Relationships among Polyneoptera. Polyneopteran

monophyly is not generally accepted, and the positions of most

taxa, especially Embioptera, Dermaptera, and Grylloblattaria,

are controversial.

Saltatoria

Phasmida

Mantodea

Blattodea

+ Isoptera

Dermaptera

Grylloblattaria

Plecoptera

Embioptera

Polyneoptera

Dictyoptera

Mantophasmatodea

338 The Relationships of Animals: Ecdysozoans

Stick insects and leaf insects (phasmids) include one

species-poor taxon (Timema, in California), and the higher

phasmids or Euphasmatodea, composed of about 3000 species.

Their classification has been typological and was based

on work by Gьnther (1953) until Bradler (1999) began a

phylogenetic analysis using morphological data, soon followed

by molecular sequence studies. Within euphasmatodeans,

wingless Agathemera (10 species in South America)

appears to be the sister group of Neophasmatidae, which includes

Phyllinae, Heteropteryginae, Eurycanthinae, Lanceocercata

(200 species; Australia, southern Asia, Madagascar),

and various taxa commonly called stick insects (Bradler 2000,

2002).

Phasmida (or Phasmida + Embiida) are probably the sister

group of Orthoptera or Saltatoria, that is, grasshoppers,

crickets, and allies (fig. 20.4). Earliest saltatorians are known

from the Pennsylvanian or upper Carboniferous period. The

earliest certain fossil stick and leaf insects are known from

the Mesozoic, but the group must be as old as the saltatorians,

if they are their closest relatives. Saltatoria and Phasmatodea

share a large precostal area in the wing that is derived but is

lost in all extant and some of the Mesozoic phasmatodeans

(Sharov 1968, Willmann 2003). Among Recent phasmids,

Heteropteryx exhibits the most plesiomorphic wing structure.

The forewings are elongated, the longitudinal veins radius,

radial sector and media are parallel to one another, and the

cubitus consists of two branches. The venation is very similar

to that of the Cretaceous Coniphasma, differing only in

the fusions in Heteropteryx and the shortage of wing in

Coniphasma (Willmann 2003). This is in conflict with the

results of Whiting et al. (2003) based on DNA sequence data,

where Agathemera + Heteropteryx + Haaniella appear as one

of the most derived phasmatodean subgroups.

Saltatoria are composed of more than 20,000 species, belonging

to two monophyletic groups, Ensifera and Caelifera.

The hypothesis of saltatorian monophyly is founded on the

enlarged hind femora containing the extensor muscle of the

tibia that enables the animals to jump, the presence of prothoracic

cryptopleury (which means that the saddle-shaped

pronotum covers the prothoracic sides), the fusion of the 1st

and 2nd tarsal segments, and other characters interpretable

as autapomorphies of the group. The ensiferans are plesiomorphous

with respect to their long antennae, but they have

lost their arolium, which is an adhesive structure of the tarsus,

and exhibit numerous derived wing characters. Caeliferans,

by contrast, have short antennae that are not longer

than the combined head and prothorax (a derived state). In

their digestive tract, caeliferans have lost the proventriculus

(which means they have to use their mouth parts intensively),

and their tarsi consist of only three tarsomeres at most. Both

Ensifera and Caelifera came out as monophyletic in an analysis

of molecular sequences by Rowell and Flook (1998), who

also proposed a division of Caelifera into subunits based on

an investigation of about 150 species.

The position of Notoptera (rock crawlers, 16 species),

which are confined to East Asia and North America, is unclear.

Almost every group within Polyneoptera has been contemplated

as their sister group, which in turn implies that

the sister group of any of the polyneopteran taxa is uncertain

as well. Rowell and Flook (1998) grouped them along

with Dermaptera and Plecoptera in one clade based on analysis

of genome sequences. In 2002, a new insect taxon was

described, Mantophasmatodea, a name suggesting relationship

to praying mantises and phasmids, but which has no

close affinities to the former, whereas its proventriculus and

midgut structure is similar to that in Notoptera (Klass et al.

2002a, 2002b). Like notopterans, mantophasmatodeans are

wingless and live on other arthropods. Members of the group

had already been described from Baltic amber five years before

(Arillo et al. 1997), although without assignment to any

insect taxon of higher rank.

The phylogenetic position of the Embiida (web spinners,

>1500 species, with many remaining undescribed;

Ross 2000) is unclear (figs. 20.1, 20.2, 20.4). Engel and

Grimaldi (2000) and Grimaldi (2001) regard them as the

closest relatives of Zoraptera. The two groups have in common

the reduction of the cerci (two-segmented in the ground

pattern of Zoraptera), the enlargement of the hind femora,

the presence of at least some wingless morphs, the shedding

of wings along a basal fraction zone, brood care, and the reduction

in the number of tarsomeres (three in Embiida, two

in Zoraptera). Some of these similarities are not convincingly

interpretable as synapomorphies (Rasnitsyn 1998), and Rдhle

(1970) has pointed to several derived similarities shared by

Embiida and Phasmida, among them a gula or gulalike structure

(although a gula does not pertain to the ground pattern

of the phasmids; Bradler 1999, Kristensen 1975), the structure

of the propleura, and the possession of both a ventral

and a dorsal flexor of the paraglossae and two furcafurcasternal

muscles inserting at the profurcal sternite. Molecular

sequence studies (Rowell and Flook 1998) have also

supported a close relationship between phasmids and web

spinners, and these two combined constitute the sister group

to Saltatoria.

Acercaria

The second species-rich branch of neopteran insects is

Acercaria (fig. 20.5), so-called because this group lacks cerci.

Additional derived characters include a reduction in the

number of the Malpighian tubules (four at most), loss of the

first abdominal sternum, possession of a single ganglionic

complex in the abdomen, and the loss of the perforatorium

of the sperm. Within Acercaria, two main branches are distinguishable.

The first branch is Hemiptera, which consist

of Heteropterida (Heteroptera, bugs; and Coleorrhyncha,

a group of fewer than 30 described species occurring in the

Phylogenetic Relationships and Evolution of Insects 339

Southern Hemisphere), Auchenorrhyncha (cicadas), and

Sternorrhyncha (plant lice), one of the most successful lineages

of insects. The relationships among the three have not

been worked out. The assumption of the monophyly of

Auchenorrhyncha, based for example on a pair of sound

producing organs in the first abdominal segment, is sometimes

doubted (Mahner 1993) but has gained support from

examinations of the forewing base (Yoshizawa and Saigusa

2001). The second acercarian clade is Micracercaria, or small

Acercaria. Their wings have an easily recognizable area

formed by the first cubitus, the areola postica, and their tarsus

consists of only three segments. One group belonging to

them is Psocodea, which contain the wingless sucking and

biting lice (the certainly monophyletic Phthiraptera with

>3000 species; Kцnigsmann 1960 and many subsequent

authors) and book lice (Psocoptera, >3000 species). They

share, for example, a unique sclerotization of the esophagus

and therefore possess a so-called cibarial sclerite, and they

are equipped with a modification of the basal part of their

antennal flagellomeres to facilicate rupture, which is interpreted

as an escape device (Kцnigsmann 1960, Seeger 1975).

Monophyly of Psocoptera has been doubted, but Seeger

(1979) found embryological and egg structural evidence that

it is a natural taxon. Lyal (1985) pointed to similarities of

Phthiraptera and Liposcelidae/Psocoptera that might indicate

a sister-group relationship between the two but concluded

that they are most probably convergences. The other micracercarian

group is Thysanoptera (thrips, >4500 species).

They range in body length from less than 1 mm to 15 mm;

their name refers to their fringed wings, which, however, also

occur in small members of other insect taxa. The thrips have

usually been considered to be the closest allies of Hemiptera,

but according to the fossil record they are linked to Posocodea

instead. In the Mesozoic and the Permian there were

the psocodean-like lophioneurids, which share two striking

apomorphies with thrips: a tarsus with only two segments

and a bladderlike structure at its tip (Vishniakova

1981). The Jurassic Karataothrips is already similar to

recent thrips, but its venation is more primitive. The view

that thrips are the nearest relatives of Psocodea is also supported

by the total evidence cladogram of Wheeler et al.

(2001).

Many have accepted the view that Hemiptera and Thysanoptera

constitute a taxon called Condylognatha, which

Bцrner (1904) had erected based on a study of head structures.

However, the interpretation of decisive similarities as possible

synapomorphies has been doubted by several authors, among

them Kцnigsmann (1960). There appears to be no spermatozal

apomorphy supporting the monophyly of the Condylognatha

(Jamieson et al. 1999), but Yoshizawa and Saigusa (2001) have

found two possible synapomorphies of Thysanoptera and

Hemiptera in the sclerites of the forewing base (fusion of

basisubcostale and second axillary sclerite; distal median plate

placed next to the second axillary sclerite).

Zoraptera

The Zoraptera (fig. 20.5) are a little-known insect group, for

which no popular name exists. In German they are called

Bodenlдuse (i.e., groundlice). They are up to 3 mm long, and

fewer than 30 species have been described. Their systematic

position is unclear. In the literature, they appear as the sister

group of Isoptera (which is untenable because isopterans

share derived internal head sclerite structures with cockroaches

and mantises that zorapterans do not), or as the sister

group of Dictyoptera, Embiida (see above), Dermaptera

+ Dictyoptera, Dermaptera, Acercaria, Holometabola, and

others. Similarities with some groups are due to reductions

or losses (e.g., the gonostyli, appendages of the male genital

apparatus, are lacking). A sister group relationship with

Acercaria, for example, has been postulated because of a reduction

in number of the Malpighian tubules, an abdominal

ventral nerve cord that consists of two ganglia only (reduced

to one in Acercaria; Hennig 1969, 1986, Kristensen 1981,

Kцnigsmann 1960, Seeger 1979), and the shared presence

in the wings of some groups (the micracercarians) of a socalled

areola postica formed by the first cubitus that is one

Figure 20.5. Phylogenetic relationships of

Acercaria. Monophyly of Paraneoptera is doubtful

because of the uncertain position of Zoraptera,

which may be closely related to polyneopterans.

Zoraptera

Psocoptera

Phthiraptera

Thysanoptera

Hemiptera

Acercaria

?

Paraneoptera

Heteropterida

Auchenorrhyncha

Sternorrhyncha

Micracercaria

340 The Relationships of Animals: Ecdysozoans

of the posterior veins (significance unclear). Kristensen

(1991), however, feels that zorapterans, generally simplified

because of their minute size, might well have had their origin

among the polyneopterans.

Eumetabola (= Acercaria + Holometabola)

The Acercaria are possibly the closest ally of Holometabola, as

evidenced by the development of the male genital structures

(fig. 20.6). So far, however, none of the cladograms based on

molecular sequence data alone supports the Eumetabola hypothesis

(Whiting et al. 1997, Wheeler et al. 2001). In fact,

acercarians appear scattered within Polyneoptera and Holometabola

in the consensus cladogram for the 18S rDNA data

(Wheeler et al. 2001) in which hemipterans are the nearest

relatives to a group consisting mainly of Holometabola, but also

of Metajapyx (Diplura) and Grylloblatta (Notoptera), whereas

thrips and psocodeans are grouped with Zoraptera among some

of the polyneopterans. According to 28S rDNA data, Acercaria

seems to be part of Holometabola, which also includes the

stoneflies [(((Hemiptera + Psocodea) + (Thysanoptera + Plecoptera))

+ Hymenoptera); Wheeler et al. 2001].

It has been estimated that more than 75% of all organisms

belong in the insects, and of these, more than 75% belong

in Holometabola. The insects discussed to this point

have young that gradually become more and more similar to

the adult, but holometabolans have a larval stage that is very

different from the adult and a pupal stage between the larva

and adult. Sometimes, the pupa is described as a stage of rest,

and in fact it is almost motionless and usually does not take

up food. But it is actually that life stage during which the most

fundamental changes in ontogeny occur, because the larval

body is entirely restructured to become equipped with adult

characters. In the last five or so decades, holometabolan

monophyly has not been doubted by morphologists (contra

numerous earlier publications), but none of the more detailed

molecular sequence studies has produced a cladogram with

a monophyletic Holometabola (Chalwatzis et al. 1996, Whiting

et al. 1997, Wheeler et al. 2001). (For more detail about

the phylogenetic relationships with the Holometabola, see

Whiting, ch. 21 in this vol.).

What Is Really Known?

It may appear that nothing in insect phylogeny and systematics

is well established, and indeed morphological characters

considered to be useful for phylogeny reconstruction

have consistently been interpreted in different ways. However,

the significance of many structures has been clarified,

and a major reason for this is that phylogenetic thinking has

contributed much to an entirely different approach to analytic

examination of characters. Although some authors in

the middle of the 20th century held the view that insect wings

may have developed independently twice, because there are

two different types this assertion is no longer considered to

be tenable, because similarities in wing structure outweigh the

probability of convergence. The same applies to many other

structures, but in many cases—and this has been underestimated

by morphologists—even apparently complex body

parts seem to have evolved in different evolutionary lineages.

This dilemma has not been solved yet. It is certain that in many

cases, structures appear to be superficially similar until more

detailed investigations often unveil differences (and nonhomology).

Sometimes, a name appears to be all that structures

share (e.g., “sperm pump” in Mecoptera and Diptera).

This has also practical aspects: not only is a new generation of

skilled morphologists needed, but such studies are also timeconsuming.

Yet, the reward of years of hard comparative work

is deep insight not only into structural complexity as well as

constructional morphology, functions, ecology and behavior;

most important, a deeper understanding of the organism and

its evolutionary context will ultimately emerge.

Different possible interpretations of similarities limit the

value of any cladogram, and in fact, phylogeneticists used to

discuss the meaning and significance of every single structure

that appeared to relate different taxa. Consequently, computergenerated

cladograms of all of Insecta based on morphological

evidence, or combined molecular sequence and morphological

data, have not, with rare exceptions, led to entirely new and

convincing hypotheses of relationship because it is not char-

Figure 20.6. Summary cladogram of insects as favored in this

chapter.

Zoraptera

Psocoptera

Phthiraptera

Thysanoptera

Hemiptera

Saltatoria

Phasmida

Mantodea

Blattodea + Isoptera

Dermaptera

Grylloblattaria

Plecoptera

Embioptera

Odonata

Ephemeroptera

Zygentoma

Archaeognatha

Diplura

Protura

Collembola

Ellipura

Polyneoptera

Acercaria

Eumetabola

(= Phalloneoptera)

Neoptera

Metapterygota

Pterygota

Dicondylia

Ectognatha

Euentomata

Holometabola

Mantophasmatodea

Phylogenetic Relationships and Evolution of Insects 341

acters that are being coded, but rather character interpretations.

Unveiling relationships of groups of closely related insect species

seems to be much less problematic.

So, what do we know? Insects are probably monophyletic,

as supported by most molecular studies. Almost all easily distinguishable

major taxa are monophyletic, namely, Collembola,

Protura, Diplura, Archaeognatha, Ephemeroptera,

Odonata, Plecoptera, Notoptera, Mantophasmatodea, Dermaptera,

Embioptera, Saltatoria, Phasmida, Mantodea,

Isoptera, Zoraptera, Phthiraptera, Psocoptera, Thysanoptera,

Heteroptera, Coleorhyncha, Auchenorrhyncha, and Sternorrhyncha

(see fig. 20.6); and among Holometabola, the Coleoptera,

Planipennia, Raphidiodea, Megaloptera, Strepsiptera,

Hymenoptera, Lepidoptera, Trichoptera, Diptera, and Siphonaptera

are also monophyletic. However, Blattodea are probably

paraphyletic in terms of Isoptera, serious doubts as to the

monophyly of Mecoptera exist, and Zygentoma may be paraphyletic.

Until recently, the monophyly of several more taxa

had been uncertain, for example, Diplura, Dermaptera, and

Megaloptera. Collembola, Protura, and Diplura are basal insect

lineages and do not belong in the entity composed of

Archaeognatha, Zygentoma, and pterygotes. Archaeognatha

are the sister taxon to Dicondylia, which are composed of

Zygentoma (monophyly not certain) and Pterygota. Odonata

and Ephemeroptera are closely related (but possibly not sister

taxa), and most probably Neoptera forms a clade (fig. 20.6).

The Holometabola appear to be a natural taxon, and probably

Acercaria (Hemiptera, Thysanoptera, Psocodea) are also monophyletic,

being the sister group to holometabolans. The

Zoraptera are often thought to be the nearest relatives of

Acercaria (Zoraptera + Acercaria = Paraneoptera; fig. 20.5),

but this needs confirmation. The positions of the remaining

groups are also uncertain. They may constitute a natural

group (“Polyneoptera,” figs. 20.4, 20.6) or form a series

of taxa between the root of Neoptera and acercarian- (or

paraneopteran-) holometabolan node. Among them are

Mantodea and Blattodea (inclusive of termites), which have

long been known to be a natural unit (Dictyoptera). Almost

certainly, Phasmida and Saltatoria are more closely related to

each other than either of them is to any other neopteran group,

with the possible exception of Embioptera.

Acknowledgments

I thank the organizers of the Tree of Life Symposium for having

invited me to speak. The comments of an anonymous referee

are greatly appreciated. This work was in part supported by

grants from the Deutsche Forschungsgemeinschaft.

Literature Cited

Arillo, A., V. M. Ortuсo, and A. Nel. 1997. Description of an

enigmatic insect from Baltic amber. Bull. Soc. Entomol. Fr.

102:11–14.

Arnett, R. 2000. American insects. A handbook of the insects of

America north of Mexico. 2nd ed. CRC Press, Boca Raton,

FL.

Artois, T. 2001. Cladistic analysis of the Polycystididae

(Platyhelminthes Kalyptorhynchia), with application of

phylogenetic nomenclature. Proefschrift voorgelegd tot het

behalen van de graad van Doctor in de Wetenschappen,

richting Biologie, te verdedigen door. Limburgs Universitair

Centrum, Diepenbeek, Belgium.

Baccetti, B. 1979. Ultrastructure of sperm and its bearing on

arthropod phylogeny. Pp. 609–644 in Arthropod phylogeny

(A. P. Gupta, ed.). Van Norstrand Peinhold, New York.

Bechly, G. 1996. Morphologische Untersuchungen am

Flьgelgeдder der rezenten Libellen und deren Stammgruppenvertreter

(Insecta; Pterygota; Odonata, unter

besonderer Berьcksichtigung der Phylogenetischen

Systematik und des Grundplanes der *Odonaten. Petalura

2(spec vol):1–402.

Bechly, G. 2001. Ur-Geziefer. Die faszinierende Evolution der

Insekten. Stutt. Beitr. Nat. C 49:1–94.

Bechly, G., C. Brauckmann, W. Zessin, and E. Grцning. 2000.

New results concerning the morphology of the most ancient

dragonflies (Insecta: Odonatoptera) from the Namurian of

Hagen-Vorhalle (Germany). J. Zool. Syst. Evol. Res.

39:209–226.

Beutel, R., and S. Gorb. 2001. Ultrastructure of attachment

specialisations of hexapods (Arthropoda): evolutionary

patterns inferred from a revised ordinal phylogeny. J. Zool.

Syst. Evol. Res. 39:177–207.

Bцrner, C. 1904. Zur Systematik der Hexapoden. Zoolog. Anz.

27:511–533.

Bцrner, C. 1908. Neue Homologien zwischen Crustaceen und

Hexapoden. Die BeiЯmandibel der Insekten und ihre

phylogenetische Bedeutung. Archi. Metapterygota. Zool.

Anz. 34:100–125.

Borror, D., C. Triplehorn, and N. Johnson. 1992. An introduction

to the study of insects. 6th ed. Saunders College

Publishing, San Diego.

Boudreaux, H. B. 1979. Arthropod phylogeny with special

reference to insects. John Wiley and Sons, New York.

Bradler, S. 1999. The vomer of Timema Scudder, 1865 (Insecta:

Phasmatodea) and its significance for phasmatodean

phylogeny. Cour. Forsch. Inst. Senckenb. 215:43–47.

Bradler, S. 2000. On the systematic position of Agathemera Stеl,

1875 within the Phasmatodea (Insecta). Zoology 103(suppl.

3):99.

Bradler, S. 2002. Phasmatodea, Gespenstschrecken. Pp. 251–

261 in Lehrbuch der Speziellen Zoologie I: Insecta (H. H.

Dathe, ed.). G. Fischer, Heidelberg, Germany.

Bьning, J. 1998. The ovariole: structure, type and phylogeny.

Pp. 897–932 in Microscopic anatomy of invertebrates 11C

(Insecta) (M. Locke and F. W. Harrison, eds.). John Wiley

and Sons, New York.

Burckhardt, R. 1903. Zur Geschichte der biologischen

Systematik. Verh. Naturforsch. Ges. Basel. 16:388–440.

Chalwatzis, N., J. Hauf, Y. van de Peer, R. Kinzelbach, and F. K.

Zimmermann. 1996. 18S ribosomal RNA genes of insects:

primary structure of the genes and molecular phylogeny of

the Holometabola. Ann. Entomol. Soc. Am. 89:788–803.

Dohle, W. 1998. Myriapod-insect relationships as opposed to

342 The Relationships of Animals: Ecdysozoans

an insect-crustacean sister group Relationship. Pp. 305–315

in Arthropod relationships (R. Fortey and R. Thomas, eds.).

Chapman and Hall, London.

Dohle, W. 2001. Are the insects terrestrial crustaceans? A

discussion of some new facts and arguments and the

proposal of the proper name “Tetraconata” for the monophyletic

unit Crustacea + Hexapoda. Ann. Soc. Entomol. Fr.

(n.s.) 37:85–103.

Engel, M., and D. Grimaldi. 2000. A winged Zorotypus in

Miocene amber from the Dominican Republic, with

discussion on relationships of and within the order. Acta

Geol. Hisp. 35:149–164.

Giles, E. T. 1974. Relationships between the Hemimerina and

other Dermaptera—a case for reinstating the Hemimerina in

the Dermaptera based on numerical procedure. Trans.

R. Entomol. Soc. Lond. 126:189–206.

Giribet, G., G. D. Edgecombe, and W. C. Wheeler. 2001.

Arthropod phylogeny based on eight molecular loci and

morphology. Nature 413:157–161.

Grandcolas, P. 1994. Phylogenetic systematics of the subfamily

Polyphaginae, with the assignment of Cryptocercus Scudder,

1862 to this taxon (Blattaria, Blaberoidea, Polyphagidae).

Syst. Entomol. 19:145–158.

Grandcolas, P. 1996. The phylogeny of cockroach families: a

cladistic appraisal of morpho-anatomical data. Can. J. Zool.

74:508–527.

Grandcolas, P. 1997. What did the ancestors of the woodroach

Cryptocercus look like? A phylogenetic study of the origin of

subsociality in the subfamily Polyphaginae (Dictyoptera,

Blattaria). Mem. Mus. Natl. Hist. Nat. 173:231–252.

Grandcolas, P., and P. Deleporte. 1996. The origin of protistan

symbionts in termites and cockroaches: a phylogenetic

perspective. Cladistics 12:93–98.

Griffiths, G. 1974. On the foundations of biological systematics.

Acta Biotheor. 23:85–131.

Griffiths, G. 1976. The future of Linnean nomenclature. Syst.

Zool. 25:168–173.

Grimaldi, D. 2001. Insect evolutionary history from Handlirsch

to Hennig, and beyond. J. Paleont. 75:1152–1160.

Gьnther, K. 1953. Ьber die taxonomische Gliederung und die

geographische Verbreitung der Insektenordnung der

Phasmatodea. Beitr. Entomol. 3:541–563.

Handlirsch, A. (1906–1908). Die fossilen Insekten und die

Phylogenie der rezenten Formen. Verlag von Wilhelm

Engelmann, Leipzig.

Hennig, W. 1953. Kritische Bemerkungen zum phylogenetischen

System der Insekten. Beitr. Entomol. 3:1–85.

Hennig, W. 1969. Die Stammesgeschichte der Insekten.

W. Kramer Verlag, Frankfurt am Main.

Hennig, W. 1981. Insect phylogeny. Academic Press, New York.

Hennig, W. 1986. Taschenbuch der speziellen Zoologie,

Wirbellose 2: Gliedertiere. Harri Deutsch, Frankfurt.

Hilken, G. 1998. Vergleich von Tracheensystemen unter

phylogenetischem Aspekt. Verh. Naturwiss. Verh. Hamb.

37:5–94.

Hцrnschemeyer, T. 1998. Morphologie und Evolution des

Flьgelgelenks der Coleoptera und Neuropterida. Bonner

Zool. Monogr. 43:1–126.

Hцrnschemeyer, T. 2002. Phylogenetic significance of the wingbase

of the Holometabola (Insecta). Zool. Scr. 31:17–30.

Hovmцller, R., T. Pape, and M. Kдllersjц. 2002. The Palaeoptera

problem: basal pterygote phylogeny inferred from 18S and

28R DNA sequences. Cladistics 18:313–323.

Jamieson, B. 1987. The ultrastructure and phylogeny of insect

spermatozoa. Cambridge University Press, Cambridge.

Jamieson, B. G. M., R. Dallai, and B. A. Afzelius. 1999. Insects.

Their spermatozoa and phylogeny. Science Publishers,

Enfield, NH.

Jarzembowski, E. A., X. Martнnez-Delclтs, G. Bechly, A. Nel,

R. Coram, and F. Escuilliй. 1998. The Mesozoic noncalopterygoid

Zygoptera: description of new genera and

species from the Lower Cretaceous of England and Brazil

and their phylogenetic significance (Odonata, Zygoptera,

Coenagrionoidea, Hemiphlebioidea, Lestoidea). Cretaceous

Res. 19:403–444.

Kaestner, A. 1973. Lehrbuch der Speziellen Zoologie, Bd. I, 3.

Insecta: B. Spezieller Teil. Gustav Fischer Verlag, Jena.

Klass, K.-D. 1995. Die Phylogenie der Dictyoptera. Cuvillier

Verlag, Gцttingen.

Klass, K.-D. 1998. The proventriculus of the Dicondylia, with

comments on evolution and phylogeny in Dictyoptera and

Odonata (Insecta). Zool. Anz. 237:15–42.

Klass, K.-D. 2000. The male abdomen of the relic termite

Mastotermes darwiniensis (Insecta: Isoptera: Mastotermitidae).

Zool. Anz. 239:231–262.

Klass, K.-D. 2001a. The female abdomen of the viviparous

earwig Hemimerus vosseleri (Insecta: Dermaptera:

Hemimeridae), with a discussion of the postgenital

abdomen of Insecta. Zool. J. Linn. Soc. 131:251–307.

Klass, K.-D. 2001b. Morphological evidence on blattarian

phylogeny: “phylogenetic histories and stories” (Insecta,

Dictyoptera). Mitt. Mus. Nat. Berl. Dtsch. Entomol. Z.

48:223–265.

Klass, K.-D., and N. P. Kristensen. 2001. The ground plan and

affinities of hexapods: recent progress and open problems.

Ann. Soc. Entomol. Fr. (n.s.) 37:265–298.

Klass, K.-D., O. Zompro, and J. Adis. 2002b. Ordnung Mantophasmatodea.

Pp. 161–166 in Lehrbuch der Speziellen

Zoologie I: Insecta (H. H. Dathe, ed.). G. Fischer, Heidelberg,

Germany.

Klass, K.-D., O. Zompro, N. P. Kristensen, and J. Adis. 2002a.

Mantophasmatodea: a new insect order with extant

members in the Afrotropics. Science 296:1456–1459.

Koch, M. 1997. Monophyly and phylogenetic position of the

Diplura (Hexapoda). Pedobiologia 41:9–12.

Koch, M. 2000. The cuticular cephalic endoskeleton of

primarily wingless hexapods: ancestral state and evolutionary

changes. Pedobiologia 44:374–385.

Koch, M. 2001. Mandibular mechanisms and the evolution of

hexapods. Ann. Soc. Entomol. Fr. (n.s.) 37:129–174.

Kцnigsmann, E. 1960. Zur Phylogenie der Parametabola unter

besonderer Berьcksichtigung der Phtiraptera. Beitr.

Entomol. 10:705–744.

Kraus, O. 2001. Myriapoda” and the ancestry of the Hexapoda.

Ann. Soc. Entomol. Fr. (n.s.) 37:105–127.

Kristensen, N. P. 1975. The phylogeny of hexapod “orders.” A

critical review of recent accounts. Z. Zool. Syst. Evolut.

Forsch. 13:1–44.

Kristensen, N. P. 1981. Phylogeny of insect orders. Annu. Rev.

Entomol. 26:135–157.

Phylogenetic Relationships and Evolution of Insects 343

Kristensen, N. P. 1989. Insect phylogeny based on morphological

evidence. Pp. 295–306 in The hierarchy of life.

Molecules and morphology in phylogenetic analysis

(B. Fernholm, K. Bremer, L. Brundin, H. Jцrnvall, L.

Rutberg, and H.-E. Wanntorp, eds.). Excerpta Medica,

Amsterdam.

Kristensen, N. P. 1991. Phylogeny of extant hexapods. Pp. 125–

140 in The insects of Australia (I. Naumann, ed.), vol. 1.

Cornell University Press, Ithaca, NY.

Kristensen, N. P. 1998. The groundplan and basal diversification

of the hexapods. Pp. 280–293 in Arthropod relationships

(R. Fortey and R. Thomas, eds.). Chapman and Hall,

London.

Linnй, C. 1758. Systema naturae. Editio Decima. Holmiae.

Lo, N., G. Tokuda, H. Watanabe, H. Rose, M. Slaytor, K.

Maekawa, C. Bandi, and H. Noda. 2000. Evidence from

multiple gene sequence indicates that termites evolved from

wood-feeding cockroaches. Curr. Biol. 10:801–804.

Lohmann, H. 1996. Das phylogenetische System der Anisoptera

(Odonata). Entomol. Z. 106:209–296.

Lyal, C. 1985. Phylogeny and classification of the Psocodea,

with particular reference to the lice (Psocodea: Phthiraptera).

Syst. Entomol. 10:145–165.

Mahner, M. 1993. Systema Cryptoceratorum Phylogeneticum

(Insecta, Heteroptera). Zoologica 48:1–302.

Manton, S. 1977. The Arthropoda. Habits, functional morphology

and evolution. Clarendon Press, Oxford.

Martynov, A. B. 1925. Ьber zwei Grundtypen der Flьgel bei den

Insecten und ihre Evolution. Z. Morphol. Okol. Tiere

4:465–501.

Mayer, P. 1876. Ueber Ontogenie und Phylogenie der Insekten.

Jen. Z. Naturwiss. 10:125–221.

Naumann, I. 1991. The insects of Australia. 2 vols. Cornell

University Press, Ithaca, NY.

Nixon, K. C., and J. M. Carpenter. 2000. On the other “phylogenetic

systematics.” Cladistics 16:298–318.

Novokshonov, V. G., and R. Willmann. 1999. On the morphology

of Asthenohymen uralicum (Insecta; Diaphanopterida:

Asthenohymenidae) from the lower Permian of the Urals.

Paleontol. J. 33:539–545.

Papavero, N., J. Llorente-Bousquets, and J. Abe. 2001. Proposal

of a new system of nomenclature for phylogenetic systematics.

Arq. Zool. 36:1–145.

Pass, G. 1998. Accessory pulsatile organs. Microsc. Anat. Invert.

11B:621–640.

Paulus, H. F. 1979. Eye structure and the monophyly of the

Arthropoda. Pp. 299–383 in Arthropod phylogeny (A. P.

Gupta, ed.). Van Nostrand Reinhold, New York.

Popham, E. 1985. The mutual affinities of the major earwing

taxa (Insecta, Dermaptera). Z. Zool. Syst. Evol. Forsch.

23:199–214.

Rдhle, W. 1970. Untersuchungen an Kopf und Prothorax von

Embia ramburi Rimsky-Korsakoff 1906 (Embioptera,

Embiidae). Zool. Jb. Anat. 87:248–330.

Rasnitsyn, A. 1998. On the taxonomic position of the insect

order Zorotypida = Zoraptera. Zool. Anz. 237:185–194.

Richards, O. W., and R. G. Davies. 1977. Imm’s general

textbook of entomology, Vol. 2: Classification and biology.

Chapman and Hall, London.

Richter, S., G. D. Edgecombe, and G. D. F. Wilson. 2002. The

lacinia mobilis and similar structures—a valuable character

in arthropod phylogenetics? Zool. Anz. 241:339–361.

Ross, E. S. 2000. Contributions to the biosystematics of the

insect order Embiidina. 1. Origin, relationships and

integumental anatomy. Occ. Pap. Calif. Acad. Sci. 149:1–

53.

Ross, H. H., C. A. Ross, and J. R. P. Ross. 1982. A textbook of

entomology. 4th ed. John Wiley and Sons, New York.

Rowell, C. H. F., and P. K. Flook. 1998. Phylogeny of the

Caelifera and the Orthoptera as derived from ribosomal

gene sequences. J. Orthopt. Res. 7:147–156.

Seeger, W. 1975. Funktionsmorphologie an Spezialbildungen

der FьhlergeiЯel von Psocoptera und anderen Paraneoptera

(Insecta); Psocodea als monophyletische Gruppe. Z.

Morphol. Tiere 81:137–159.

Seeger, W. 1979. Spezialmerkmale an Eihьllen und Embryonen

von Psocoptera im Vergleich zu anderen Paraneoptera

(Insecta); Psocoptera als monophyletische Gruppe. Stutt.

Beitr. Nat. 329:1–57.

Sharov, A. G. 1968. Phylogeny of the orthopteroidea. Trudy

Paleontol. Inst. Akad. Nauk. SSSR 118 (repr.: 1971, Keter

Press, Jerusalem).

Snodgrass, R. E. 1935. Principles of insect morphology. Cornell

University Press, Ithaca, NY.

Snodgrass, R. 1938. Evolution of the Annelida, Onychophora,

and Arthropoda. Smithsonian Misc. Coll. 97(6):1–159.

Snodgrass, R. E. 1952. A textbook of arthropod anatomy.

Comstock Publ. Assoc., Ithaca, NY.

Soldбn, T. 1997. The Ephemeroptera: whose sister-group are

they? Pp. 514–519 in Ephemeroptera and Plecoptera:

biology–ecology–systematics (P. Landolt and M. Sartori,

eds.). Mauron + Tinguely and Lachat SA, Fribourg,

Switzerland.

Staniczek, A. H. 2000. The mandible of silverfish (Insecta:

Zygentoma) and mayflies (Ephemeroptera): its morphology

and phylogenetic significance. Zool. Anz. 239:147–178.

Stys, P., and S. Bilinski. 1990. Ovariole types and the phylogeny

of hexapods. Biol. Rev. 65:401–429.

Vishniakova, V. 1980. Earwigs (Insecta, Forficulida) from

the Upper Jurassic of the Karatau Range. Paleont. J.

(no vol.):63–79.

Vishniakova, V. 1981. Novye paleozoyskie i mezozoyskie

lophioneuridy (Thripida, Lophioneuridae) (in Russian).

[New paleozoic and mesozoic lophioneurids (Thripida,

Lophioneuridae).] Trudy Paleontol. Inst. Akad. Nauk, SSSR

183:43–63.

Weber, H. 1933. Lehrbuch der Entomologie. G. Fischer Verlag,

Jena.

Wheeler, W. C., M. Whiting, Q. D. Wheeler, and J. M. Carpenter.

2001. the phylogeny of the extant hexapod orders.

Cladistics 17:113–169.

Wheeler, W. M. 1904. The phylogeny of termites. Biol. Bull.

8:29–37.

Wheeler, W. M. 1928. The social insects. Their origin and

evolution. Paul, Trench, Trubner & Co., London, and

Harcourt, Brace and Company, New York.

Whiting, M. F., S. Bradler, and T. Maxwell. 2003. Loss and

recovery of wings in stick insects. Nature 421:264–267.

Whiting, M. F., J. C. Carpenter, Q. D. Wheeler, and W. C.

Wheeler. 1997. The Strepsiptera problem: phylogeny of the

344 The Relationships of Animals: Ecdysozoans

holometabolous insect orders inferred from 18S and 28S

ribosomal DNA sequences and morphology. Syst. Biol.

46:1–69.

Willmann, R. 1987. Phylogenetic systematics, classification and

the plesion concept. Verh. Naturwiss. Ver. Hamb. 29:221–

233.

Willmann, R. 1990. Die Bedeutung palдontologischer Daten fьr

die zoologische Systematik. Verh. Dtsch. Zool. Ges.

83:277–289.

Willmann, R. 1997. Phylogeny and the consequences of

phylogenetic systematics. Pp. 499–510 in Ephemeroptera

and Plecoptera: biology–ecology–systematics (P. Landolt

and M. Sartori, eds.). Mauran and Tinguely and Lachat SA,

Fribourg, Switzerland.

Willmann, R. 1998. Advances and problems in insect phylogeny.

Pp. 269–279 in Arthropod relationships (R. Fortey and

R. Thomas, eds.). Chapman and Hall, London.

Willmann, R. 1999. The upper carboniferous Lithoneura lameerei

(Insecta, Ephemeroptera ?). Palaontol. Z. 73:289–302.

Willmann, R. 2002. Phylogenese und System der Insecta. Pp. 1–

64 in Lehrbuch der Speziellen Zoologie: Insecta (H. H.

Dathe, ed.). G. Fischer, Heidelberg, Germany.

Willmann, R. 2003. Die phylogenetischen Beziehungen der

Insekten: Offene Fragen und Probleme. Verh. Westdeutsch.

Entomol. Tag. 2003:1–64.

Wygodzinsky, P. 1961. On a surving representative of the

Lepidotrichidae (Thysanura). Ann. Entomol. Soc. Am.

54:621–627.

Yoshizawa, K., and T. Saigusa. 2001. Phylogenetic analysis of

paraneopteran orders (Insecta: Neoptera) based on forewing

base structure, with comments on monophyly of Auchenorrhyncha

(Hemiptera). Syst. Entomol. 26:1–13.

Zwick, P. 1973. Insecta: Plecoptera. Phylogenetisches System

und Katalog. Das Tier. 94:1–465.

Zwick, P. 1980. Plecoptera (Sternfliegen) [sic]. Handb. Zool. IV

Liefg. 26:1–121. Eine Naturgeschichte der Stдmme des

Tierreichs. Walter de Gruyter, Berlin.

Zwick, P. 2002. Plecoptera. Steinfliegen. Pp. 144–154 in

Lehrbuch der speziellen Zoologie I: Insecta (H. H. Dathe,

ed.). G. Fischer, Heidelberg, Germany.

Навигация

• Главная
• Книги
• Статьи
• Обратная связь
• Поиск